United States                              December 2005
Environmental Protection                        EPA/600/R-05/144aA
Agency   	
     Air Quality Criteria for Lead
     (First External Review Draft)

     Volume I  of II

-------
                                               EPA/600/R-05/144aA
                                                  December 2005
Air Quality  Criteria for Lead
              Volume I
National Center for Environmental Assessment-RTF Office
        Office of Research and Development
       U.S. Environmental Protection Agency
           Research Triangle Park, NC

-------
                                   DISCLAIMER

      This document is a first external review draft being released for review purposes only and
does not constitute U.S. Environmental Protection Agency policy.  Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.

                                     PREFACE

      National Ambient Air Quality Standards (NAAQS) are promulgated by the United States
Environmental Protection Agency (EPA) to meet requirements set forth in Sections 108 and 109
of the U.S. Clean Air Act. Sections 108 and 109 of the Clean Air Act require the EPA
Administrator (1) to list widespread air pollutants that reasonably may be expected to endanger
public health or welfare;  (2) to issue air quality criteria for them that assess the latest available
scientific information on  nature and effects of ambient exposure to them; (3) to set "primary"
NAAQS to protect human health with adequate margin of safety and to set "secondary" NAAQS
to protect against welfare effects  (e.g.,  effects on vegetation, ecosystems, visibility, climate,
manmade materials, etc); and (5) to periodically review and revise, as appropriate, the criteria
and NAAQS for a given listed pollutant or class of pollutants.
      Lead was first listed in the mid-1970's as a "criteria  air pollutant" requiring NAAQS
regulation.  The scientific information pertinent to Lead NAAQS development that was available
at the time was assessed in the EPA document Air Quality Criteria for Lead; published in 1977.
Based on the scientific assessments contained in that 1977 lead air quality criteria document
(1977 Lead AQCD), EPA established a 1.5 |ig/m3 (90-day average) Lead NAAQS in 1978.
      To meet Clean Air Act requirements noted above for periodic review of criteria and
NAAQS, newly available scientific information published since the 1977 Lead AQCD was
assessed and discussed in a revised Lead AQCD and Addendum published in 1986 and in a
Supplement to the 1986 AQCD/Addendum published by EPA in 1990. A 1990 Lead Staff
Paper, prepared by EPA's Office of Air Quality Planning and Standards (OPQPS), drew upon
key findings and conclusions from the  1986 Lead AQCD/Addendum and 1990 Supplement
(as well as other OAQPS-sponsored lead exposure/risk analyses) in posing options for the EPA
                                          I-ii

-------
Administrator to consider with regard to possible revision of the Lead NAAQS. However,
EPA decided not to revise the lead NAAQS at that time.
      The purpose of this revised Lead AQCD is to critically evaluate and assess the latest
scientific information that has become available since the literature assessed in the above 1986
Lead AQCD/Addendum and 1990 Supplement, with the main focus being on pertinent new
information useful in evaluating health and environmental effects of ambient air lead exposures.
This includes discussion in this document of information regarding: the nature, sources,
distribution,  measurement, and concentrations of lead in the environment; multimedia lead
exposure (via air, food, water, etc.) and biokinetic modeling of contributions of such exposures
to concentrations of lead in brain, kidney, and other tissues (e.g.,  blood and bone concentrations,
as key indices of lead exposure).;  characterization of lead health effects and associated exposure-
response relationships; and delineation of environmental (ecological) effects of lead. This First
External Review Draft of the revised Lead AQCD mainly assesses pertinent literature  published
or accepted for publication through June, 2004.
      The present First External Review Draft (dated December 2005) of the revised Lead
AQCD is being released for public comment and review by the Clean Air Scientific Advisory
Committee (CAS AC) to obtain comments  on the organization and structure of the document, the
issues addressed, the approaches employed in assessing and interpreting the newly available
information on lead exposures and effects, and the key findings and conclusions arrived at as a
consequence of this assessment. Public comments and CAS AC recommendations will be taken
into account in making appropriate further revisions to this document for incorporation into a
Second External Review Draft of the document to be released in  early 2006 for further public
comment and CASAC review.  Public comments and CASAC advice received on the Second
External Review Draft materials will then be taken into account in incorporating further
revisions into the final version of this Lead AQCD, which is to be completed and issued by
October 1, 2006. Evaluations contained in the present document will be drawn on to provide
inputs to associated Lead Staff Paper prepared by EPA's Office of Air Quality Planning and
Standards (OAQPS), which will pose options for consideration by the EPA Administrator with
regard to proposal and, ultimately, promulgation of decisions on potential retention or revision,
as appropriate, of the current Lead NAAQS.
                                          I-iii

-------
      Preparation of this document was coordinated by staff of EPA's National Center for
Environmental Assessment in Research Triangle Park (NCEA-RTP). NCEA-RTP scientific
staff, together with experts from and academia, contributed to writing of document chapters.
Earlier drafts of document materials were reviewed by scientists from other EPA/units
and by non-EPA experts in several public peer consultation workshops held by EPA in
July/August 2005.
      NCEA acknowledges the valuable contributions provided by authors, contributors, and
reviewers and the diligence of its staff and contractors in the preparation of this draft document.
                                          I-iv

-------
                     Air Quality Criteria for Lead

                     (First External Review Draft)

                             VOLUME I
EXECUTIVE SUMMARY	E-l
(To be prepared and included in future Second External Review Draft)

1.    INTRODUCTION	1-1

2.    CHEMISTRY, SOURCES, AND TRANSPORT OF LEAD	2-1

3.    ROUTES OF HUMAN EXPOSURE TO LEAD AND OBSERVED
     ENVIRONMENTAL CONCENTRATIONS	3-1

4.    MODELS OF HUMAN EXPOSURE THAT PREDICT TISSUE
     DISTRIBUTION OF LEAD	4-1

5.    TOXICOLOGICAL EFFECTS OF LEAD IN LABORATORY ANIMALS,
     HUMANS, AND IN VITRO TEST SYSTEMS	5-1

6.    EPIDEMIOLOGIC STUDIES OF HUMAN HEALTH EFFECTS
     ASSOCIATED WITH LEAD EXPOSURE	6-1

7.    INTEGRATIVE SYNTHESIS
     (To be prepared and included in future Second External Review Draft)	7-1

8.    ENVIRONMENTAL EFFECTS OF LEAD	8-1
                             VOLUME II
CHAPTER 5 ANNEX (TOXICOLOGICAL EFFECTS OF LEAD IN
LABORATORY ANIMALS, HUMANS, AND IN VITRO TEST SYSTEMS)	AX5-1

CHAPTER 6 ANNEX (EPIDEMIOLOGIC STUDIES OF HUMAN HEALTH
EFFECTS ASSOCIATED WITH LEAD EXPOSURE)	AX6-1
                                 I-v

-------
                               Table of Contents
List of Tables	I-xxiv
List of Figures	I-xxx
Authors, Contributors, and Reviewers	I-xxxviii
U.S. Environmental Protection Agency Project Team for Development
      of Air Quality Criteria for Lead	I-xlvii
U.S. Environmental Protection Agency Science Advisory Board (SAB)
      Staff Office Clean Air Scientific Advisory Committee (CASAC) 	1-1
Abbreviations and Acronyms	I-li

EXECUTIVE SUMMARY	E-l
(To be prepared and included in future Second External Review Draft)

1.     INTRODUCTION	1-1
      1.1      LEGAL AND HISTORICAL BACKGROUND	1-1
              1.1.1      Legislative Requirements	1-1
              1.1.2     Criteria andNAAQS Review Process	1-3
              1.1.3      Regulatory Chronology	1-4
      1.2      CURRENT LEAD CRITERIA ANDNAAQS RE VIEW	1-7
              1.2.1      Procedures and Key Milestones for Document Preparation	1-7
      1.3      ORGANIZATIONAL STRUCTURE AND CONTENT OF
              THE DOCUMENT	1-9
              1.3.1      Ascertainment of Literature and General Document Format	1-9
              1.3.2     Organization and Content of the Document	1-10
      1.4      REFERENCES	1-13

2.     CHEMISTRY, SOURCES, AND TRANSPORT OF LEAD	2-1
      2.1      PHYSICAL AND CHEMICAL PROPERTIES OF LEAD	2-1
      2.2      SOURCES OF LEAD	2-11
              2.2.1      Natural Sources	2-11
              2.2.2     Stationary Sources	2-15
              2.2.3      Mobile Sources	2-37
      2.3      TRANSPORT WITHIN THE ENVIRONMENT	2-45
              2.3.1      Atmospheric Transport of Lead Particles	2-45
              2.3.2     Deposition of Airborne Particles	2-47
              2.3.3      Resuspension of Lead-Containing Soil and Dust Particles	2-52
              2.3.4     Runoff from Impervious Surfaces	2-57
              2.3.5      Leaching  of Soil Lead	2-60
              2.3.6     Transport in Aquatic Systems	2-64
              2.3.7     Plant Uptake	2-67
              2.3.8     Routes of Exposure for Livestock and Wildlife	2-68
      2.4      METHODS FOR MEASURING ENVIRONMENTAL LEAD	2-69
      2.5      SUMMARY	2-71
      2.6      REFERENCES	2-75
                                       I-vi

-------
                              Table of Contents
                                    (cont'd)
3.     ROUTES OF HUMAN EXPOSURE TO LEAD AND OBSERVED
      ENVIRONMENTAL CONCENTRATIONS	3-1
      3.1     EXPOSURE: AIR	3-1
             3.1.1     Observed Concentrations- Indoor	3-1
             3.1.2     Observed Concentrations- Outdoor	3-5
             3.1.3     Observed Concentrations - Occupational	3-10
      3.2     EXPOSURE: SOIL AND ROAD DUST	3-12
             3.2.1     Urban Background Concentrations of Soil Lead	3-13
             3.2.2     Soil Concentrations Near Stationary Sources	3-15
             3.2.3     Concentrations of Lead in Road Dust	3-20
      3.3     EXPOSURE: DRINKING WATER	3-22
      3.4     EXPOSURE: FOOD INGESTION	3-27
      3.5     OTHER ROUTES OF EXPOSURE	3-33
             3.5.1     Lead-Based Paint	3-33
             3.5.2     Calcium Supplements	3-33
             3.5.3     Glazes	3-34
             3.5.4     Miniblinds	3-34
             3.5.5     Hair Dye	3-34
             3.5.6     Other Potential Sources of Lead Exposure	3-34
      3.6     MEASUREMENT METHODS	3-35
      3.7     SUMMARY	3-35
      3.8     REFERENCES	3-37

4.     MODELS OF HUMAN EXPOSURE THAT PREDICT TISSUE
      DISTRIBUTION OF LEAD	4-1
      4.1     OBJECTIVES IN MODELING LEAD EXPOSURE AND TISSUE
             DISTRIBUTION OF LEAD	4-1
      4.2     HISTORIC OVERVIEW OF LEAD MODELS	4-4
             4.2.1     Rabinowitz Model	4-4
             4.2.2     Marcus Model(s)	4-6
             4.2.3     Bert Model	4-7
             4.2.4     Contemporary Models	4-9
      4.3     INTEGRATED EXPOSURE UPTAKE BIOKINETIC (IEUBK)
             MODEL FOR LEAD IN CHILDREN	4-10
             4.3.1     Model Structure	4-10
             4.3.2     Model Calibration and Evaluation	4-16
             4.3.3     Model Applications	4-17
             4.3.4     Validation/Verification of IEUBK	4-18
      4.4     LEGGETT MODEL	4-19
             4.4.1     Model Structure	4-19
             4.4.2     Model Calibration and Evaluation	4-22
                                     I-vii

-------
                                Table of Contents
                                      (cont'd)
              4.4.3     Model Applications	4-23
              4.4.4     Implementation Code	4-24
      4.5     O'FLAHERTY MODEL	4-24
              4.5.1     Model Calibration and Evaluation	4-28
              4.5.2     Model Applications	4-28
              4.5.3     Verification/Validation of O'Flaherty Model	4-29
      4.6     EPA ALL AGES LEAD MODEL	4-29
              4.6.1     Model Structure	4-29
              4.6.2     Model Calibration and Evaluation	4-32
              4.6.3     Model Applications	4-33
              4.6.4     Validation and Verification of AALM Implementation Code	4-33
      4.7     SLOPE FACTOR MODELS	4-33
      4.8     MODEL COMPARISONS	4-34
      4.9     CONCLUSIONS AND FUTURE DIRECTIONS	4-42
      4.10    REFERENCES	4-45

5.     TOXICOLOGICAL EFFECTS OF LEAD IN LABORATORY ANIMALS,
      HUMANS, AND IN VITRO TEST SYSTEMS	5-1
      5.1     INTRODUCTION	5-1
      5.2     EFFECTS OF LEAD ON HEME SYNTHESIS	5-2
              5.2.1     Effects of Lead on Erythrocyte Biology and Function	5-2
              5.2.2     Effects of Lead on Erythrocyte Functions	5-3
              5.2.3     Effect of Lead on Erythrocyte Heme Metabolism	5-8
              5.2.4     Effect of Lead on Other Hematological Parameters	5-11
              5.2.5     Effects of Lead on Erythrocyte Enzymes	5-12
              5.2.6     Erythrocyte Lipid Peroxidation and Antioxidant Defense	5-15
              5.2.7     Summary	5-17
      5.3     NEUROLOGICAL/NEUROBEHAVIORAL EFFECTS OF LEAD	5-18
              5.3.1     Neurotoxicological/Neurobehavioral Effects of Lead
                       in Animals	5-18
                       5.3.1.1    Introduction	5-18
                       5.3.1.2    Neurochemical Alterations Resulting from
                                 Lead Exposure	5-20
                       5.3.1.3    Actions of Lead Exposure Defined by
                                 Neurophysiological Approaches	5-26
                       5.3.1.4    Lead Exposure and Sensory Organ Function	5-30
                       5.3.1.5    Neurobehavioral Toxicity Resulting from
                                 Lead Exposure	5-32
                       5.3.1.6    Lead-Induced Changes in Cellular Development
                                 and Disposition of the Metal	5-39
                       5.3.1.7    Integration of Research Findings	5-42
                                       I-viii

-------
                          Table of Contents
                                (cont'd)
        5.3.2     Neurotoxi col ogical/Neurobehavi oral Effects of Lead
                 in Humans	5-43
                 5.3.2.1     Effects of Lead in Young Children to
                            Mid-Adolescence	5-44
                 5.3.2.2     Clinical Manifestations in Adults with
                            Childhood Lead Poisoning	5-70
                 5.3.2.3     Adults with Ambient Exposures to Lead	5-73
5.4      REPRODUCTIVE AND DEVELOPMENTAL EFFECTS OF LEAD	5-79
        5.4.1     Summary of Key Findings on the Developmental and
                 Reproductive Effects of Lead in Animals from the 1986
                 LeadAQCD	5-79
        5.4.2     Effects on Male Reproductive Function	5-81
                 5.4.2.1     Effects on Male Sexual Development
                            and Maturation	5-81
                 5.4.2.2     Effects on Male Fertility: Effects on Sperm
                            Production and Function	5-82
                 5.4.2.3     Effects on Male Sex Endocrine System	5-87
                 5.4.2.4     Effects on Morphology and Histology of Male
                            Sex Organs	5-88
        5.4.3     Effects on Female Reproductive Function	5-89
                 5.4.3.1     Effects on Female Sexual Development
                            and Maturation	5-89
                 5.4.3.2     Effects on Female Fertility	5-90
                 5.4.3.3     Effects on the Female Sex Endocrine System
                            and Menstrual Cycle	5-90
                 5.4.3.4     Effects on Morphol ogy and Hi stol ogy of
                            Female Sex Organs and the Placenta	5-94
        5.4.4     Effects on Embryogenesis	5-94
                 5.4.4.1     Embryo/Fetal Mortality	5-94
                 5.4.4.2     Effects on Embryo/Fetal Morphology	5-95
        5.4.5     Effects on Growth and Endocrine Regulation of Growth	5-101
        5.4.6     Effects on Other Endocrine Systems during Development	5-102
        5.4.7     Effects on Other Organ Systems during Development	5-102
                 5.4.7.1     Developmental Effects on Blood and Liver	5-102
                 5.4.7.2     Developmental Effects on Skin	5-103
                 5.4.7.3     Developmental Effects on the Retina	5-104
        5.4.8     Conclusions	5-104
5.5      CARDIOVASCULAR EFFECTS OF LEAD	5-109
        5.5.1     Introduction	5-109
        5.5.2     Lead Exposure and Arterial Pressure in Experimental
                 Animals	5-109
                                  I-ix

-------
                           Table of Contents
                                 (cont'd)
                  5.5.2.1     Effect of Lead on Production of Reactive
                             Oxygen Species and Nitric Oxide Metabolism	5-110
                  5.5.2.2     Protein Kinase C, Inflammation, NFicB
                             Activation and Apoptosis	5-116
                  5.5.2.3     Effect of Lead Exposure on the Adrenergic
                             System	5-118
                  5.5.2.4     Effects of Lead on the Renin-Angiotensin-
                             Aldosterone (RAAS) and Kininergic Systems	5-120
        5.5.3      Effects of Lead Exposure on Vasomodulators	5-121
        5.5.4      Effects of Lead on Vascular Reactivity	5-122
        5.5.5      Lead-Calcium Interactions in Vascular Tissue	5-124
        5.5.6      Cardiotoxicity and Atherogenesis	5-125
        5.5.7      Effects of Lead on Endothelial Cells	5-125
        5.5.8      Effects of Lead on Vascular Smooth Muscle Cells	5-129
        5.5.9      Summary/Conclusion	5-130
5.6     GENOTOXIC AND CARCINOGENIC EFFECTS OF LEAD	5-131
        5.6.1      Introduction	5-131
        5.6.2      Carcinogenesis Studies	5-132
                  5.6.2.1     Human Studies	5-132
                  5.6.2.2     Laboratory Animal Studies	5-133
                  5.6.2.3     Cell Culture Studies	5-136
                  5.6.2.4     Organ-Specific Studies	5-138
                  5.6.2.5     Carcinogenesis Summary	5-138
        5.6.3      Genotoxicity Studies	5-139
                  5.6.3.1     Human Studies	5-139
                  5.6.3.2     Laboratory Animal Studies	5-141
                  5.6.3.3     Cell Culture Studies	5-143
                  5.6.3.4     Animal Cell Cultures	5-145
                  5.6.3.5     Cell-Free Studies	5-147
                  5.6.3.6     Organ-Specific Studies	5-147
                  5.6.3.7     Genotoxicity Section Summary	5-147
        5.6.4      Genotoxicity as it Pertains to Potential Developmental
                  Effects	5-147
        5.6.5      Epigenetic Effects and Mixture Interactions	5-148
                  5.6.5.1     Gene Expression	5-149
                  5.6.5.2     DNA Repair	5-150
                  5.6.5.3     Mitogenesis	5-150
                  5.6.5.4     Epigenetic Mechanisms Summary	5-151
        5.6.6      Overall Conclusions	5-152
                                    I-x

-------
                          Table of Contents
                                 (cont'd)
5.7      LEAD AND THE KIDNEY	5-152
        5.7.1     Review of Earlier Work	5-152
        5.7.2     Markers of Renal  Toxicity	5-154
        5.7.3     Biochemical Mechanisms of Lead Toxicity	5-155
        5.7.4     Animal Studies	5-157
                 5.7.4.1     Lead  Toxicokinetics	5-157
                 5.7.4.2     Pathology, Ultrastructural, and Functional
                            Studies	5-158
                 5.7.4.3     Biochemical Mechanisms of Lead Toxicity	5-165
                 5.7.4.4     Effect of Age on Lead Toxicity	5-183
        5.7.5     Summary	5-185
5.8      EFFECTS ON BONE AND TEETH	5-186
        5.8.1     Biology of Bone and Bone Cells	5-186
        5.8.2     Summary of Information Presented in the 1986 Lead AQCD	5-187
        5.8.3     Bone Growth in Lead-Exposed Animals	5-188
        5.8.4     Regulation of Bone Cell Function in Animals - Systemic
                 Effects of Lead	5-191
                 5.8.4.1     Hypercalcemia/Hyperphosphatemia	5-191
                 5.8.4.2     Vitamin D[1,25-(OH2)D3]	5-191
                 5.8.4.3     Parathyroid Hormone	5-192
                 5.8.4.4     Growth Hormone	5-193
        5.8.5     Bone Cell Cultures Utilized  to Test the Effects of Lead	5-193
                 5.8.5.1     Bone Organ Culture	5-193
                 5.8.5.2     Primary Cultures of Osteoclasts and  Osteoblasts .... 5-194
                 5.8.5.3     Rat Osteosarcoma Cell Line (ROS 17/2.8)	5-194
                 5.8.5.4     Human Osteosarcoma Cells  (HOS TE 85)	5-198
                 5.8.5.5     Chick Chondrocytes	5-198
        5.8.6     Bone Lead as a Potential Source of Toxicity in Altered
                 Metabolic Conditions	5-199
                 5.8.6.1     Pregnancy and Lactation	5-200
                 5.8.6.2     Age/Osteoporosis	5-203
                 5.8.6.3     Weight Loss	5-205
        5.8.7     Bone and Lead Summary	5-205
        5.8.8     Teeth - Introduction	5-207
        5.8.9     Uptake of Lead by Teeth	5-208
        5.8.10    Effects of Lead on Enamel and Dentine Formation	5-209
        5.8.11    Effects of Lead on Dental Pulp Cells	5-211
        5.8.12    Adverse Effects of Lead on Teeth—Dental Caries	5-211
        5.8.13    Lead from Teeth as a Potential Source of Toxicity	5-213
        5.8.14    Teeth and Lead Summary	5-213
                                  I-xi

-------
                          Table of Contents
                                 (cont'd)
5.9      EFFECTS OF LEAD ON THE IMMUNE SYSTEM	5-214
        5.9.1     Introduction	5-215
                 5.9.2      Host Resistance	5-218
                 5.9.2.1     Viral Diseases	5-219
                 5.9.2.2     Bacterial Diseases	5-219
                 5.9.2.3     Parasitic Diseases	5-220
                 5.9.2.4     Tumors	5-221
        5.9.3     Humoral Immunity	5-221
                 5.9.3.1     General Effects on B lymphocytes and
                             Immunoglobulins	5-222
                 5.9.3.2     IgE Alterations	5-223
        5.9.4     Cell-Mediated Immunity	5-225
                 5.9.4.1     General Effects on Thymocytes and
                            T lymphocytes	5-226
                 5.9.4.2     Delayed Type Hypersensitivity	5-228
                 5.9.4.3     Other T-Dependent Cell-Mediated Immune
                            Changes	5-231
        5.9.5     Lymphocyte Activation and Responses	5-232
                 5.9.5.1     Activation by Mitogens	5-232
                 5.9.5.2     Activation via Other Receptors	5-233
                 5.9.5.3     Cytokine Production	5-234
        5.9.6     Macrophage Function	5-237
                 5.9.6.1     Nitric Oxide (NO) Production	5-239
                 5.9.6.2     Other Functional Alterations	5-239
        5.9.7     Granulocytes and Natural Killer (NK) Cells	5-246
        5.9.8     Hypersensitivity and Autoimmunity	5-247
        5.9.9     Mechanism of Lead-Based Immunomodulation	5-249
        5.9.10    Age-Based Differences in Sensitivity	5-251
        5.9.11    Summary and Conclusions	5-254
5.10    EFFECTS OF LEAD ON OTHER ORGAN SYSTEMS	5-257
        5.10.1    Effects of Lead on the Hepatic System	5-257
                 5.10.1.1    Hepatic Drug Metabolism	5-258
                 5.10.1.2    Biochemical and Molecular Perturbations in
                            Lead-Induced Liver Tissue Injury	5-262
                 5.10.1.3    Effects of Lead Exposure on Hepatic
                            Cholesterol Metabolism	5-264
                 5.10.1.4    Effect of Chel ati on Therapy on Lead-Induced
                            Hepatic Oxidative Stress	5-266
                 5.10.1.5    Lead-Induced Liver Hyperplasia: Mediators and
                            Molecular Mechanisms	5-269
                 5.10.1.6    Effects of Lead on Liver Heme Synthesis	5-274
                 5.10.1.7    Summary	5-275
                                  I-xii

-------
                                 Table of Contents
                                       (cont'd)
              5.10.2     Gastrointestinal System and Lead Absorption	5-277
                        5.10.2.1    Lead and In vitro Cytotoxicity in Intestinal Cells.... 5-278
                        5.10.2.2    Alterations in Intestinal Physiology and
                                   Ultrastructure	5-278
                        5.10.2.3    Intestinal Uptake and Transport	5-279
                        5.10.2.4    Alterations in Gastrointestinal Motility/
                                   Gastrointestinal Transit and Function	5-280
                        5.10.2.5    Lead, Calcium, and Vitamin D Interactions in
                                   the Intestine	5-280
                        5.10.2.6    Lead and Intestinal Enzymes	5-282
                        5.10.2.7    Summary	5-282
       5.11    LEAD-BINDING PROTEINS	5-283
              5.11.1     Lead-Binding Proteins  Within Intranuclear Inclusion
                        Bodies in Kidney	5-284
              5.11.2     Cytoplasmic Lead-Binding Proteins in Kidney and Brain	5-285
              5.11.3     Lead-Binding Proteins  in Erythrocytes	5-287
              5.11.4     Lead-Binding Proteins  in Rat Liver	5-290
              5.11.5     Lead-Binding Proteins  in Intestine	5-290
              5.11.6     Relationship of Lead-Binding Protein to Metallothionein	5-293
              5.11.7     Is ALAD an Inducible Enzyme and is it the Principal
                        Lead-Binding Protein in the Erythrocyte?	5-294
              5.11.8     Summary	5-295
       5.12    REFERENCES	5-297

6.      EPIDEMIC-LOGIC STUDIES OF HUMAN HEALTH EFFECTS ASSOCIATED
       WITH LEAD EXPOSURE	6-1
       6.1     INTRODUCTION	6-1
              6.1.1      Approach to Identifying Lead Epidemiologic Studies	6-2
              6.1.2      Approach to Assessing Epidemiologic Evidence	6-2
              6.1.3      Considerations in the Interpretation of Epidemiologic Studies
                        of Lead Health Effects	6-4
              6.1.4      Approach to Presenting Lead Epidemiologic Evidence	6-6
       6.2     BIOLOGICAL MARKERS OF LEAD BODY BURDEN
              AND EXPOSURE	6-7
              6.2.1      Lead in Blood	6-7
                        6.2.1.1     Summary of Key Findings from the 1986
                                   LeadAQCD	6-7
                        6.2.1.2     Analytical  Methods for Measuring Lead in Blood	6-8
                        6.2.1.3     Levels of Lead in Blood	6-9
                        6.2.1.4     Blood Lead as aBiomarker of Lead Body Burden	6-9
                        6.2.1.5     Blood Lead as aBiomarker of Lead Exposure	6-16
                                        I-xiii

-------
                  Table of Contents
                        (cont'd)
          6.2.1.6     Summary of Blood Lead as a Biomarker of
                    Lead Body Burden and Exposure	6-18
6.2.2      Lead in Bone	6-20
          6.2.2.1     Summary of Key Findings from the 1986
                    LeadAQCD	6-20
          6.2.2.2     Methodology of Bone Lead Analysis	6-21
          6.2.2.3     Bone Lead as a Biomarker of Lead Body Burden	6-23
          6.2.2.4     Distribution of Lead from Bone into Blood
                    and Plasma	6-26
          6.2.2.5     Mobilization of Lead From Bone	6-29
          6.2.2.6     Summary of Bone Lead as a Biomarker of
                    Lead Body Burden and Exposure	6-34
6.2.3      Lead in Teeth	6-34
          6.2.3.1     Summary of Key Findings from the 1986
                    LeadAQCD	6-34
          6.2.3.2     Analytical Methods for Measuring Lead in Teeth	6-35
          6.2.3.3     Tooth Lead as a Biomarker of Lead Body Burden.... 6-36
          6.2.3.4     Relationship between Tooth Lead and
                    Blood Lead	6-37
          6.2.3.5     Mobilization of Lead from Teeth	6-37
          6.2.3.6     Summary of Tooth Lead as a Biomarker of
                    Lead Body Burden and Exposure	6-38
6.2.4      Lead in Urine	6-38
          6.2.4.1     Summary of Key Findings from the 1986
                    LeadAQCD	6-38
          6.2.4.2     Analytical Methods for Measuring Lead in Urine	6-38
          6.2.4.3     Levels of Lead in Urine	6-39
          6.2.4.4     Urine Lead as a Biomarker of Lead  Body Burden .... 6-40
          6.2.4.5     Urine Lead as a Biomarker of Lead  Exposure	6-42
          6.2.4.6     Summary of Urine Lead as a Biomarker of Lead
                    Body Burden and Exposure	6-44
6.2.5      Lead in Hair	6-45
          6.2.5.1     Summary of Key Findings from the 1986
                    LeadAQCD	6-45
          6.2.5.2     Analytical Methods for Measuring Lead in Hair	6-45
          6.2.5.3     Levels of Lead in Hair	6-46
          6.2.5.4     Hair Lead as a Biomarker of Lead Body Burden	6-46
          6.2.5.5     Hair Lead as a Biomarker of Lead Exposure	6-47
          6.2.5.6     Summary of Hair Lead as a Biomarker of Lead
                    Body Burden and Exposure	6-47
                          I-xiv

-------
                           Table of Contents
                                 (cont'd)
6.3      NEUROTOXIC EFFECTS OF LEAD	6-47
        6.3.1      Summary of Key Findings on Neurotoxic Effects of Lead
                  in Children from 1986 Lead AQCD and Addendum, and
                  1990 Supplement	6-47
        6.3.2      Neurotoxic Effects of Lead in Children	6-49
                  6.3.2.1      Neurocognitive Ability	6-51
                  6.3.2.2      Measures of Academic Achievement	6-76
                  6.3.2.3      Measures of Specific Cognitive Abilities	6-80
                  6.3.2.4      Disturbances in Behavior, Mood, and
                             Social Conduct	6-83
                  6.3.2.5      Sensory Acuities	6-89
                  6.3.2.6      Neuromotor Function	6-90
                  6.3.2.7      Brain Anatomical Development and Activity	6-92
                  6.3.2.8      Gene-Environment Interactions in the Expression
                             of Lead-Associated Neurodevelopmental Deficits.... 6-94
                  6.3.2.9      Reversibility of Lead-related Neurodevelopmental
                             Deficits Associated with Prenatal and Postnatal
                             Exposure	6-95
                  6.3.2.10    Periods of Enhanced Developmental
                             Susceptibility to Central Nervous System
                             Effects of Environmental Lead	6-99
                  6.3.2.11    Effect of Environmental Lead Exposure on
                             Neurodevelopment at the Lower
                             Concentration Range	6-103
                  6.3.2.12    Selection and Validity of Neuropsychological
                             Outcomes in  Children	6-105
                  6.3.2.13    Confounding, Causal Inference, and Effect
                             Modification of the Neurotoxic Effect of
                             Lead in Children	6-107
        6.3.3      Summary of the Epi demi ol ogi c Evi dence for the
                  Neurotoxic Effects of Lead in Children	6-109
        6.3.4      Summary of Key Findings on the Neurotoxic Effects of
                  Lead in Adults from the 1986 Lead AQCD	6-110
        6.3.5      Neurotoxic Effects of Lead in Adults	6-111
                  6.3.5.1      Overview of Cognitive and Psychomotor Tests
                             Associated with Adult Lead Exposure	6-111
                  6.3.5.2      Neurobehavioral Effects Associated with
                             Environmental Lead Exposure	6-112
                  6.3.5.3      Neurological Symptoms Associated with
                             Occupational Lead Exposure	6-115
                  6.3.5.4      Neurobehavioral Effects Associated with
                             Occupational Lead Exposure	6-116
                                  I-xv

-------
                           Table of Contents
                                 (cont'd)
                  6.3.5.5     Neurophysiological Function and Occupational
                             Lead Exposure	6-120
                  6.3.5.6     Evoked Potentials and Occupational
                             Lead Exposure	6-122
                  6.3.5.7     Postural Stability, Autonomic Testing,
                             and Electroencephalogram (EEG) and
                             Occupational Lead Exposure	6-123
                  6.3.5.8     Other Neurological Outcomes Associated with
                             Lead in Adults	6-125
                  6.3.5.9     Occupational Exposure to Organolead and
                             Inorganic Lead	6-127
        6.3.6      Summary of the Epidemiologic Evidence for the Neurotoxic
                  Effects of Lead in Adults	6-128
6.4     RENAL EFFECTS OF LEAD	6-129
        6.4.1      Summary of Key Findings on the Renal Effects of Lead
                  from the 1986 Lead AQCD	6-129
        6.4.2      Renal  Outcome Definitions	6-130
        6.4.3      Lead Exposure Measure Definitions	6-132
        6.4.4      Lead Nephrotoxicity in Adults	6-132
                  6.4.4.1     General Population Studies	6-132
                  6.4.4.2     Occupational Studies	6-138
                  6.4.4.3     Patient Population Studies	6-140
                  6.4.4.4     Mortality Studies	6-147
        6.4.5      Lead Nephrotoxicity in Children	6-148
                  6.4.5.1     Studies in Adults Following  Childhood
                             Lead Poisoning	6-148
                  6.4.5.2     Lead Body Burden in Children with Chronic
                             Renal Disease	6-150
                  6.4.5.3     General Population Studies in Children	6-150
        6.4.6      Mechanisms for Lead Nephrotoxicity	6-153
        6.4.7      Susceptible Populations for Lead Nephrotoxicity	6-154
                  6.4.7.1     Chronic Medical Diseases	6-154
                  6.4.7.2     Age	6-155
                  6.4.7.3     Genetic Polymorphisms	6-155
        6.4.8      Confounding of the Renal Effects of Lead by Other
                  Potential Risk Factors	6-159
                  6.4.8.1     Cadmium	6-159
        6.4.9      Summary of the Epidemiologic Evidence for the Renal
                  Effects of Lead	6-162
                                  I-xvi

-------
                          Table of Contents
                                (cont'd)
6.5      CARDIOVASCULAR EFFECTS OF LEAD	6-163
        6.5.1     Summary of Key Findings of the Cardiovascular Effects
                 of Lead from the 1985 Lead AQCD and Addendum,
                 and 1990 Supplement	6-163
        6.5.2     Effects of Lead on Blood Pressure and Hypertension	6-165
                 6.5.2.1     Introduction	6-165
                 6.5.2.2     Blood Pressure and Hypertension Studies
                            Using Blood Lead as Exposure Index	6-167
                 6.5.2.3     Blood Pressure and Hypertension Studies
                            Using Bone Lead as Exposure Index	6-181
        6.5.3     Other Cardiovascular Outcomes	6-195
                 6.5.3.1     Ischemic Heart Disease	6-195
                 6.5.3.2     Stroke	6-197
                 6.5.3.3     Cardiovascular/Circulatory Mortality	6-197
                 6.5.3.4     Other Cardiovascular Effects	6-199
        6.5.4     Potential Confounding of the Cardiovascular Effects of Lead.... 6-200
                 6.5.4.1     Confounding by Copollutants	6-200
                 6.5.4.2     Confounding by Smoking Status	6-201
                 6.5.4.3     Confounding by Alcohol Consumption	6-202
                 6.5.4.4     Confounding by Dietary Calcium Intake	6-203
                 6.5.4.5     Summary of Potential Confounding of the
                            Lead Effect on Cardiovascular Health	6-204
        6.5.5     Gene-Lead Interactions	6-204
        6.5.6     Summary of the Epi demi ol ogi c Evi dence for the
                 Cardiovascular Effects of Lead	6-207
6.6      REPRODUCTIVE AND DEVELOPMENTAL EFFECTS OF LEAD	6-208
        6.6.1     Summary of Key Findings of the Reproductive and
                 Developmental Effects of Lead from the 1986 Lead AQCD	6-208
        6.6.2     Placental Transfer of Lead	6-209
        6.6.3     Effects of Lead on Reproductive Function	6-211
                 6.6.3.1     Effects on Male Reproductive Function	6-211
                 6.6.3.2     Genotoxicity and Chromosomal Aberrations	6-216
                 6.6.3.3     Effects on Female Reproductive Function	6-219
        6.6.4     Spontaneous Abortion	6-220
                 6.6.4.1     Spontaneous Abortion and Maternal Exposure
                            to Lead	6-220
                 6.6.4.2     Spontaneous Abortion and Paternal Exposure
                            to Lead	6-223
        6.6.5     Fetal Growth	6-223
        6.6.6     Preterm Delivery	6-228
        6.6.7     Congenital Abnormalities	6-230
                                 I-xvii

-------
                          Table of Contents
                                (cont'd)
       6.6.8     Summary of the Epi demi ol ogi c Evi dence for the
                 Reproductive and Developmental Effects of Lead	6-232
6.7     GENOTOXIC AND CARCINOGENIC EFFECTS OF LEAD	6-233
       6.7.1     Summary of Key Findings from the 1986 Lead AQCD	6-233
       6.7.2     Summary of Key Findings by the International Agency for
                 Research on Cancer and the National Toxicology Program	6-234
       6.7.3     Meta-Analyses of Lead and Cancer	6-236
       6.7.4     Genotoxicity of Lead	6-238
       6.7.5     Review of Specific Studies on the Carcinogenicity of Lead
                 Since the 1986 Lead AQCD	6-240
                 6.7.5.1      Introduction	6-240
                 6.7.5.2      Key Studies of Occupational Populations in
                            theU.S	6-240
                 6.7.5.3      Key Studies of the General Population	6-247
                 6.7.5.4      Other Lead Studies	6-249
       6.7.6     Confounding of Occupational Lead Studies Due to Other
                 Occupational Exposures:  Arsenic, Cadmium	6-254
       6.7.7     Confounding of Lead Studies: Smoking and Other Factors	6-255
       6.7.8     Summary of Epidemiologic Evidence for the Genotoxic
                 and Carcinogenic Effects of Lead	6-256
6.8     EFFECTS OF LEAD ON THE IMMUNE SYSTEM	6-258
       6.8.1     Summary of Key Findings of the Effects of Lead on the
                 Immune System from the  1986 Lead AQCD	6-258
       6.8.2     Host Resistance	6-259
       6.8.3     Humoral Immunity	6-259
       6.8.4     Cell-mediated Immunity	6-267
       6.8.5     Lymphocyte Function	6-272
       6.8.6     Phagocyte (Macrophage and Neutrophil) Function	6-274
       6.8.7     Summary of the Epidemiologic Evidence for the Effects
                 of Lead on the Immune System	6-275
6.9     EFFECTS OF LEAD ON OTHER ORGAN SYSTEMS	6-276
       6.9.1     Biochemical Effects of Lead	6-276
                 6.9.1.1      Summary of Key Findings of the Biochemical
                            Effects of Lead from the 1986 Lead AQCD	6-276
                 6.9.1.2      Heme Biosynthesis	6-280
                 6.9.1.3      Effects on Blood Lipids	6-281
       6.9.2     Effects of Lead on the Hematopoietic System	6-283
                 6.9.2.1      Summary of Key Findings of the Effects of
                            Lead on the Hematopoietic System from the
                            1986 Lead AQCD	6-283
                 6.9.2.2      Blood Hemoglobin Levels	6-283
                 6.9.2.3      Erythrocyte Volume and Number	6-286
                                I-xviii

-------
                   Table of Contents
                         (cont'd)
          6.9.2.4     Erythropoiesis	6-289
          6.9.2.5     Other Effects on Erythrocyte Metabolism
                     and Physiology	6-293
6.9.3      Effects of Lead on the Endocrine System	6-294
          6.9.3.1     Summary of Key Findings of the Effects of
                     Lead on the Endocrine System from the 1986
                     LeadAQCD	6-294
          6.9.3.2     Thyroid Endocrine Function	6-295
          6.9.3.3     Reproductive Endocrine Function	6-299
          6.9.3.4     Pituitary and Adrenal Endocrine Function	6-302
          6.9.3.5     Calcitropic Endocrine Function	6-303
6.9.4      Effects of Lead on the Hepatic System	6-306
          6.9.4.1     Summary of Key Findings of the Effects of
                     Lead on the Hepatic System from the 1986
                     LeadAQCD	6-306
          6.9.4.2     Non-specific Hepatic Injury	6-307
          6.9.4.3     Hepatic Cytochrome P-450 Function	6-307
6.9.5      Effects of Lead on the Gastrointestinal System	6-308
          6.9.5.1     Summary of Key Findings of the Effects of
                     Lead on the Gastrointestinal System from the
                     1986 LeadAQCD	6-308
          6.9.5.2     Gastrointestinal Colic	6-309
6.9.6      Effects of Lead on the Respiratory System	6-310
          6.9.6.1     Summary of Key Findings of the Effects of
                     Lead on the Respiratory System  from the
                     1986 LeadAQCD	6-310
          6.9.6.2     Pulmonary Function	6-310
6.9.7      Effects of Lead on Bone and Teeth	6-311
          6.9.7.1     Summary of Key Findings of the Effects
                     of Lead on Bone and Teeth from the 1986
                     LeadAQCD	6-311
          6.9.7.2     BoneToxicity	6-311
          6.9.7.3     Dental Health	6-313
6.9.8      Effects of Lead on Ocular Health	6-316
          6.9.8.1     Summary of Key Findings of the Effects
                     of Lead on Ocular Health from the 1986
                     LeadAQCD	6-316
          6.9.8.2     Ocular Effects	6-316
6.9.9      Summary of the Epidemiologic Evidence for the Effects
          of Lead on Other Organ Systems	6-317
                          I-xix

-------
                                Table of Contents
                                      (cont'd)
      6.10    INTERPRETIVE ASSESSMENT OF THE EVIDENCE IN
              EPIDEMIOLOGIC STUDIES OF LEAD HEALTH EFFECTS	6-321
              6.10.1     Introduction	6-321
              6.10.2     Exposure and Outcome Assessment in Lead
                        Epidemiologic Studies	6-321
                        6.10.2.1    Assessment of Lead Exposure and Body
                                  Burdens Using Biomarkers	6-321
                        6.10.2.2   Assessment of Health Outcomes	6-325
              6.10.3     Concentration-Response Relationship of Lead Health
                        Effects	6-326
              6.10.4     Interindividual Variability in Susceptibility to Lead Toxicity	6-329
                        6.10.4.1    Influence of Genetic Polymorphisms on Risk	6-330
                        6.10.4.2   Influence of Nutritional Status on Risk	6-330
                        6.10.4.3    Influence of Health Status on Risk	6-331
                        6.10.4.4   Influence of Co-Exposures on Risk	6-331
                        6.10.4.5    Influence of Timing of Exposure on Risk	6-332
              6.10.5     Reversibility of Lead Health Effects	6-334
                        6.10.5.1    Natural History of Effects	6-334
                        6.10.5.2   Medical Interventions	6-335
              6.10.6     Confounding of Lead Health Effects	6-336
                        6.10.6.1    Adjustment for Confounding in Epidemiologic
                                  Studies of Lead	6-336
                        6.10.6.2   Confounding Adjustment on Lead Health
                                  Effect Estimates	6-337
              6.10.7     Inferences of Causality	6-339
              6.10.8     Effects on the Individual Versus Effects on the Population	6-340
                        6.10.8.1    Effects of Lead on Intelligence	6-340
                        6.10.8.2   Cardiovascular Effects of Lead	6-345
              6.10.9     Summary of Key Findings and Conclusions Derived from
                        Lead Epidemiology Studies	6-349
      6.11    REFERENCES	6-355

7.     INTEGRATIVE SYNTHESIS
      (To be prepared and included in future Second External Review Draft)	7-1

8.     ENVIRONMENTAL EFFECTS OF LEAD	8-1
      8.1     TERRESTRIAL ECOSYSTEMS	8-1
              8.1.1      Introduction	8-1
                        8.1.1.1     Methodologies in Terrestrial Ecosystem Research	8-2
                        8.1.1.2     Distribution of Atmospherically Delivered
                                  Lead in Terrestrial Ecosystems	8-4
                        8.1.1.3     Species Response/Mode of Action	8-7
                                        I-xx

-------
                  Table of Contents
                        (cont'd)
         8.1.1.4     Exposure/Response of Terrestrial Species	8-9
         8.1.1.5     Effects of Lead on Natural Terrestrial Ecosystems ... 8-11
.1.2      Methodologies Used in Terrestrial Ecosystems Research	8-14
         8.1.2.1     Introduction	8-14
         8.1.2.2     Lead Isotopes and Apportionment	8-14
         8.1.2.3     Speciation in Assessing Lead Bioavailability
                    in the Terrestrial Environment	8-16
         8.1.2.4     Tools for Bulk Lead Quantification and
                    Speciation	8-22
         8.1.2.5     Biotic Ligand Model	8-31
         8.1.2.6     Soil Amendments	8-32
         8.1.2.7     Future Needs	8-35
.1.3      Distribution of Atmospherically Delivered Lead in
         Terrestrial Ecosystems	8-35
         8.1.3.1     Introduction	8-35
         8.1.3.2     Speciation of Atmospherically-Delivered Lead
                    in Terrestrial Ecosystems	8-38
         8.1.3.3     Tracing the Fate of Atmospherically Delivered
                    Lead in Terrestrial Ecosystems	8-44
         8.1.3.4     Inputs/Outputs of Atmospherically Delivered
                    Lead in Terrestrial Ecosystems	8-46
.1.4      Species Response/Mode of Action	8-49
         8.1.4.1     Introduction	8-49
         8.1.4.2     Lead Uptake	8-50
         8.1.4.3     Resistance Mechanisms	8-56
         8.1.4.4     Physiological Effects of Lead	8-57
         8.1.4.5     Factors that Modify Organism Response	8-60
         8.1.4.6     Summary	8-66
.1.5      Exposure-Response of Terrestrial Species	8-69
         8.1.5.1     Introduction	8-69
         8.1.5.2     Summary of Conclusions from the 1986 Lead
                    Criteria Document	8-70
         8.1.5.3     Recent Studies on the Effects of Lead on
                    Primary Producers	8-72
         8.1.5.4     Recent Studies on the Effects of Lead
                    on Consumers	8-73
         8.1.5.5     Recent Studies on the Effects of Lead
                    on Decomposers	8-92
         8.1.5.6     Summary	8-96
.1.6      Effects of Lead on Natural Terrestrial Ecosystems	8-98
         8.1.6.1     Introduction	8-98
                          I-xxi

-------
                          Table of Contents
                                (cont'd)
                 8.1.6.2     Effects of Terrestrial Ecosystem Stresses on
                            Lead Cycling	8-99
                 8.1.6.3     Effects of Lead Exposure on Natural Ecosystem
                            Structure and Function	8-104
                 8.1.6.4     Effects of Lead on Energy Flows and
                            Biogeochemical Cycling	8-109
                 8.1.6.5     Summary	8-115
.2      AQUATIC ECOSYSTEMS	8-116
       8.2.1      Introduction	8-116
                 8.2.1.1     Methodologies in Aquatic Ecosystem Research	8-117
                 8.2.1.2     Distribution of Lead in Aquatic Ecosystems	8-119
                 8.2.1.3     Species Response/Mode of Action	8-121
                 8.2.1.4     Exposure/Response of Aquatic Species	8-125
                 8.2.1.5     Effects of Lead on Natural Aquatic Ecosystems	8-127
       8.2.2      Methodologies in Aquatic Ecosystem Research	8-128
                 8.2.2.1     Introduction	8-128
                 8.2.2.2     Analytical Methods	8-129
                 8.2.2.3     Ambient Water Quality Criteria: Development	8-130
                 8.2.2.4     Ambient Water Quality Criteria:
                            Bioavailability Issues	8-131
                 8.2.2.5     Sediment Quality Criteria:  Development and
                            Bioavailability Issues	8-133
                 8.2.2.6     Metal Mixtures	8-135
                 8.2.2.7     Background Lead	8-136
       8.2.3      Distribution of Lead in Aquatic Ecosystems	8-137
                 8.2.3.1     Introduction	8-137
                 8.2.3.2     Speciation of Lead in Aquatic Ecosystems	8-137
                 8.2.3.3     Spatial Distribution of Lead in Aquatic
                            Ecosystems	8-142
                 8.2.3.4     Tracing the Fate and Transport of Lead in
                            Aquatic Ecosystems	8-157
                 8.2.3.5     Summary	8-162
       8.2.4      Species Response/Mode of Action	8-163
                 8.2.4.1     Introduction	8-163
                 8.2.4.2     Lead Uptake	8-163
                 8.2.4.3     Resistance Mechanisms	8-169
                 8.2.4.4     Physiological Effects of Lead	8-176
                 8.2.4.5     Factors That Modify Organism Response
                            to Lead	8-180
                 8.2.4.6     Factors Associated with Global Climate Change .... 8-192
                 8.2.4.7     Summary	8-193
                                 I-xxii

-------
                         Table of Contents
                                (cont'd)
       8.2.5     Exposure/Response of Aquatic Species	8-193
                8.2.5.1     Introduction	8-193
                8.2.5.2     Summary of Conclusions From the Previous
                           Criteria Document	8-194
                8.2.5.3     Recent Studies on Effects of Lead on
                           Primary Producers	8-195
                8.2.5.4     Recent Studies on Effects of Lead on Consumers... 8-201
                8.2.5.5     Recent Studies on Effects of Lead
                           on Decomposers	8-212
                8.2.5.6     Summary	8-212
       8.2.6     Effects of Lead on Natural Aquatic Ecosystems	8-213
                8.2.6.1     Introduction	8-213
                8.2.6.2     Case Study: Coeur d'Alene River Watershed	8-214
                8.2.6.3     Biotic Condition	8-216
                8.2.6.4     Summary	8-228
.3      CRITICAL LOADS FOR LEAD IN TERRESTRIAL AND
       AQUATIC ECOSYSTEMS	8-228
       8.3.1     Introduction	8-228
                8.3.1.1     Definitions	8-229
                8.3.1.2     Historical Perspective	8-229
       8.3.2     Application of Critical Loads to Terrestrial and
                Aquatic Ecosystems	8-231
       8.3.3     Calculation of Critical Loads	8-231
                8.3.3.1     Critical Limits	8-232
                8.3.3.2     Models	8-233
       8.3.4     Critical Loads in Terrestrial Ecosystems	8-237
       8.3.5     Critical Loads in Aquatic Ecosystems	8-239
       8.3.6     Limitations and Uncertainties	8-240
       8.3.7     Conclusions	8-241
.4      REFERENCES	8-242
                                I-xxiii

-------
                                  List of Tables

Number                                                                         Page

1-1       Key Milestones and Projected Schedule for Development of Revised
          Lead Air Quality Criteria Document (Lead AQCD)	1-8

2-1       Lead Alloys and Their Industrial Applications	2-3

2-2       Physical Properties of Elemental Lead	2-4

2-3       Lead Salts: Names, Formulae, Physical Characteristics, and Uses	2-6

2-4       Lead Oxides: Names, Formulae, Physical Characteristics, and Uses	2-7

2-5       Lead Compounds Observed in the Environment	2-8

2-6       Annual, Worldwide Emissions of Lead from Natural Sources	2-12

2-7       Naturally Occurring Lead Concentrations in Maj or Rock Types	2-13

2-8       The Mass-median Aerodynamic Diameters for Particles During Various
          Processes at Primary Lead Smelters	2-17

2-9       The Emissions of Lead from Non-Lead Metallurgical Processes	2-19

2-10      The Range of Lead Concentrations in Coal Lithotypes	2-22

2-11      Emission Factors of Lead for Coal Combustion in Three Different Furnaces	2-24

2-12      The Emissions of Lead from Industrial, Commercial,  and Residential
          Coal Combustion	2-25

2-13      The Concentrations of Lead in Biomass, Char, and Ash Samples from
          Spruce, Beech, Oak, Pine, and of Ailanthus Trees	2-28

2-14      Emission Factors of Lead From Processes Used in Cement Manufacture
          by Control  Device	2-34

2-15      Rate of Lead Compound Emissions from Glass-Melting Furnaces	2-36

2-16      Emission Factors of Lead for Automobiles with Model Years Between
          1971 and 1996	2-39
                                       I-xxiv

-------
                                     List of Tables
                                       (cont'd)
Number
2-17      Emission Factors of Lead for Automobiles with Model Years Between
          1971 and 1996	2-40

2-18      The Concentration of Lead in Particulate Matter Emissions and Emissions
          Factors for Lead from Buses and Trucks Fueled with Diesel No. 2 and
          Jet A Fuel	2-41

2-19      Deposition Velocities for Lead Particles	2-49

2-20      Concentrations of Lead in Rainwater in the United States	2-51

2-21      The Percentage  of Lead in Resuspended Particulate Matter	2-55

2-22      The Concentrations of Lead in Runoff From Building Surfaces	2-59

2-23      Soil/Water Partition Coefficients for Several Different Soils and Conditions	2-63

3-1       Concentrations of Lead in Indoor Dust	3-3

3-2       Airborne Concentrations  of Lead	3-7

3-3       Airborne Concentrations  Surrounding Residential Lead-Based
          Paint Abatement	3-12

3-4       Concentration of Soil Lead in Urban Areas	3-15

3-5       Concentrations of Soil Lead with Distance from Lead Smelters	3-16

3-6       Soil Lead Concentration Profile Measured Near a Lead Smelter in
          Northern France	3-17

3-7       Soil Concentrations Measured Near Mining Sites	3-19

3-8       Concentrations of Lead in Soils Grouped by Soil Grain Size	3-20

3-9       The Concentration of Lead in Road Dusts	3-22

3-10      Tap Water Concentrations of Lead	3-27

3-11      The Concentration of Lead in Food Products	3-29
                                         I-xxv

-------
                                   List of Tables
                                       (cont'd)

Number

4-1       Comparison of Slope Factors in Selected Slope Factor Models	4-15

4-2       Summary of Models of Human Exposure that Predict Tissue Distribution
          of Lead	4-35

5-3.1      Chronic Lead Exposure and LTP	5-27

5-3.2      Mechanisms of Pb-Induced Impairment of Retinal Function	5-33

5-4.1      Selected Studies Showing the Effects of Lead on Reproductive Function
          in Males	5-83

5-4.2      Selected Studies Showing the Effects of Lead on Reproductive Function
          in Females	5-91

5-4.3      Selected Studies Showing the Effects of Lead on Mammalian
          Embryogenesis and Development	5-96

5-9.1      Recent Studies Reporting Lead-Induced Increase in IgE	5-224

5-9.2      Studies Reporting Lead-Induced Shifts in Thl vs. Th2 Cytokines	5-235

5-9.3      Suggested Mechanisms of Lead-Induced Immunotoxicity	5-250

5-9.4      Immunomodulation Associated with Low Blood Lead Levels in Animals	5-252

5-9.5      Comparisons of Age-Based Sensitivity to Lead-Induced Immunotoxicity	5-254

6-2.1      Blood Lead Concentrations in U.S. by Age, NHANES IV (1999-2002)	6-10

6-2.2      Blood Lead Concentrations in U.S. by Gender, NHANES IV (1999-2002)	6-10

6-2.3      Blood Lead Concentrations by Occupation, NHANES III (1988-1994)	6-12

6-2.4      Urine Lead Concentrations in U.S. by Age, NHANES IV (1999-2002)	6-39

6-3.1      Covariate-Adjusted Changes in IQ for Each 1 |ig/dL Increase in Blood
          Lead Concentration	6-66

6-4.1      Summary of Key Studies on the Renal Effects of Environmental
          Lead Exposure	6-134
                                        I-xxvi

-------
                                   List of Tables
                                       (cont'd)
Number
6-5.1      Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Linear
          Lead (Coefficients Represent Effect of Doubling Blood Lead Calculated
          from Mean Blood Lead or Mid-point of Range)	6-182

6-5.2      Systolic and Diastolic Blood Pressure and Blood Lead Modeled with
          Logarithmic Lead  (Coefficients Represent Effect of Doubling Blood Lead)	6-187

6-7.1      Results of Meta-Analyses Addressing the Association Between Lead
          Exposure and Cancer	6-237

6-7.2      Results of Epidemiologic Studies on the Genotoxicity of Lead Exposure	6-239

6-7.3      Epidemiologic Studies of Lead Exposure and Cancer in Specific
          Populations, by Geographic Region and Study Design	6-241

6-8.1      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Serum Immunoglobulin Levels	6-260

6-8.2      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Serum Lymphocyte Abundances	6-268

6-9.1      Blood Lead-Response Relationships for Heme Synthesis Biomarkers in
          Adults and Children	6-278

6-9.2      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Blood Hemoglobin Levels	6-284

6-9.3      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Serum Erythropoietin	6-290

6-9.4      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Thyroid Hormone Levels	6-297

6-9.5      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Male Sex Hormone Levels in Adults	6-300

6-9.6      Summary of Results of Selected Studies of Associations Between Lead
          Exposure and Calcitropic Hormones	6-304

6-10.1     Summary of Studies with Quantitative Relationships for IQ and Blood Lead	6-341
                                        I-xxvii

-------
                                    List of Tables
                                        (cont'd)
Number                                                                           Page

8-1.2.1    Relative Standard Deviation (RSD) for Lead Isotope Ratios on Selected
          Mass Spectrometers	8-15

8-1.2.2    National Institute of Standards and Technology Lead SRMs	8-23

8-1.2.3    Characteristics for Direct Speciation Techniques	8-30

8-1.2.4    Affinity Constants for Lead	8-32

8-1.4.1    Tissue Lead Levels in Birds Causing Effects	8-54

8-1.5.1    Plant Toxicity Data Used to Develop the Eco-SSL	8-73

8-1.5.2    Plant Toxicity Data Not Used to Develop the Eco-SSL	8-74

8-1.5.3    Avian Toxicity Data Used to Develop the Eco-SSL	8-77

8-1.5.4    Mammalian Toxicity  Data used to Develop the Eco-SSL	8-82

8-1.5.5    Invertebrate Toxicity  Data used to Develop the Eco-SSL	8-93

8-1.5.6    Invertebrate Toxicity  Data Not used to Develop the Eco-SSL	8-94

8-2.1.1    Summary of Lead Ambient Water Quality Criteria for Freshwater
          Organisms at  Different Hardness Levels	8-117

8-2.1.2    Summary of Sediment Quality Benchmarks and Guidelines for Lead	8-118

8-2.1.3    Summary of Lead Concentrations in United States Surface Water,
          Sediment, and Fish Tissue	8-120

8-2.2.1    Common Analytical Methods for Measuring Lead in Water,
          Sediment, and Tissue	8-129

8-2.2.2    Development of Current Acute Freshwater Criteria for Lead	8-131

8-2.2.3    Recommended Sediment Quality Guidelines for Lead	8-135

8-2.3.1    NAWQA Land Use Categories and Natural/Ambient Classification	8-144

8-2.3.2    Summary Statistics of Ambient and Natural Levels of Dissolved
          Lead in Surface Water	8-145
                                        I-xxviii

-------
                                    List of Tables
                                        (cont'd)
Number                                                                            Page

8-2.3.3    Summary Statistics of Ambient and Natural Levels of Total Lead
          in<63 |im Bulk Sediment	8-145

8-2.3.4    Summary Statistics of Ambient and Natural Levels of Lead in
          Whole Organism and Liver Tissues	8-155

8-2.3.5    Comparison of NCBP and NAWQA Ambient Lead Levels in
          Whole Organism Tissues	8-156

8-2.4.1    Bioconcentration Factors for Aquatic Plants	8-168

8-2.4.2    Bioconcentration Factors for Aquatic Invertebrates	8-168

8-2.5.1    Effects of Lead to Freshwater and Marine Invertebrates	8-203

8-2.5.2    Effects of Pb to Freshwater and Marine Fish	8-208

8-2.5.3    Nonlethal Effects in Amphibians	8-211

8-2.6.1    Ecological Attributed Studies by Maret et al. (2003) in the
          Coeur d'Alene Watershed	8-215

8-2.6.2    Essential Ecological Attributes for Natural Aquatic Ecosystems
          Affected by Lead	8-220
                                         I-xxix

-------
                                   List of Figures

Number                                                                            Page

2-1       Percentage volatility of Pb during combustion of plastics at
          four temperatures	2-30

2-2       The deposition velocity plotted against the geometric mean Stokes
          diameter for particles with a density of 6 g/cm"3 (i.e., lead)	2-50

2-3       The modeled soil concentrations of lead in the South Coast Air Basin
          of California based on three resuspension rates	2-56

2-4       The modeled and measured airborne concentrations of lead in the South
          Coast Air Basin of California based on two resuspension rates	2-57

2-5       U.S. consumption of lead since 1910	2-72

2-6       Trends in U.S. air lead emissions, 1982-2002	2-73

2-7       Transport pathways for lead in the environment	2-74

3-1       Concentrations of lead throughout the United States	3-6

3-2       Airborne concentrations of lead,  averaged across the U.S., shown in relation
          to the current NAAQS, for the years 1983 through 2002	3-6

3-3       Concentrations of lead measured in 1995-1998 as recorded by the
          IMPROVE network	3-9

3-4       The changes in lead concentration with depth in  two peat cores	3-21

3-5       The change in lead concentration vs. stagnation time	3-25

3-6       The change in lead concentration vs. stagnation time	3-26

4-1       Lead biokinetics based on Rabinowitz et al. (1976)	4-5

4-2       Lead biokinectics based on Marcus (1985b)	4-6

4-3       Lead biokinetics based on Marcus (1985a)	4-7

4-4       Lead biokinetics based on Marcus (1985c)	4-8

4-5       Lead biokinetics based on Bert etal. (1989)	4-8
                                         I-xxx

-------
                                    List of Figures
                                        (cont'd)


Number                                                                            Page

4-6       Structure of the integrated exposure uptake biokinetics model for lead
          in children	4-11

4-7       Age-dependency of absorption fraction for ingested lead in the IEUBK
          model for lead in children	4-14

4-8       Structure of the Leggett Lead Biokinetic Model	4-20

4-9       Age-dependency of absorption fraction for ingested lead in the Leggett
          and O'Flaherty models	4-23

4-10      Structure of the O'Flaherty Lead Exposure Biokinetics Model	4-25

4-11      Bone growth as simulated by the O'Flaherty Lead Exposure
          Biokinetics Model	4-27

4-12      Structure of the All Ages Lead Model	4-30

4-13      Model comparison of predicted lead uptake-blood lead concentration
          relationship in children	4-37

4-14      Model comparison of predicted lead uptake-blood lead concentration
          relationships in adults	4-39

4-15      Model comparison of predicted of lead uptake-bone and soft tissue lead
          burden relationship in adults	4-40

4-16      Comparison of model predictions for childhood lead exposure	4-41

4-17      Comparison of model predictions for adult lead exposure	4-41

5-2.1      Schematic presentation of heme synthesis pathway	5-9

5-3.1      Time course of extracellular GLU concentration and GLU concentration
          in response to lead exposure	5-21

5-3.2      PKC activity as a function of Ca2+ and Pb2+ concentrations	5-24

5-3.3      I/O function difference score-PS amplitude	5-28
                                         I-xxxi

-------
                                   List of Figures
                                        (cont'd)


Number                                                                           Page

5-4.1      Data from male and female experimental animals suggests that Pb has
          multiple targets in the hypothalmic-pituitary-gonadal axis	5-80

5-5-1      This illustration depicts some of the potential mechanisms by which
          oxidative stress may participate in the pathogenesis of Pb-induced
          HTN and cardiovascular complications	5-117

5-7.1      Changes in GFR of experimental high-dose lead and control animals with
          duration of exposure to lead	5-159

5-7.2      Correlation between GFR and blood lead during the first 6 months of
          high-dose lead exposure	5-159

5-7.3      GFR in high-lead and low-lead experimental discontinuous (ED6) and
          DMSA-treated rats (DMSA) as compared to controls (C12)	5-161

5-7.4      Changes in GFR in experimental and control rats, at various time periods	5-162

5-7.5      Urinary NAG concentration in experimental and control rats  at various
          time periods	5-162

5-7.6      Kidney, liver, brain, and bone Pb levels in 56 Pb-exposed rats	5-164

5-7.7      Percentage of moderate and severe hypertrophy and vacuolization lesions
          in small and medium sized arteries in the kidney of lead-exposed rats	5-164

5-7.8      Percentage of moderate and severe muscular hypertrophy lesions in
          arterioles of the kidney in lead-exposed rats	5-165

5-9.1      Windows during prenatal development (days postconception for rat)
          or embryonic development (days postincubation initiation for chicken)
          during which sensitivity  of DTH to lead emerges	5-230

5-9.2      This figure shows the fundamental alterations to the immune system and to
          immunological response and recognition induced by exposure to lead	5-255

5-10.1    Flow diagram indicating the Pb  effects on the cholesterol  synthesis pathway	5-265

5-10.2    Schematic diagram illustrating the  mode of Pb-induced lipid  peroxidation	5-266
                                        I-xxxii

-------
                                    List of Figures
                                        (cont'd)


Number                                                                             Page

5-10.3    Hypothesis of chemical-induced liver injury generated primarily on the
          basis of different types of inhibitors	5-273

5-11.1    Sephadex G-75 gel filtration of RBC hemolysate from
          lead-exposed individual	5-288

5-11.2    SDS-polyacrylamide gel electrophoresis of RBC hemolysates from normal
          control (A) and lead-exposed individuals (B), and of low-mol-wt. lead-
          binding protein (C).  Stained with coomassie blue	5-288

5-11.3    Chromatographic profiles of protein, ALAD activity and Pb in human
          erythrocytes incubated with 5% glucose solution containing Pb acetate	5-291

5-11.4    Chromatic profiles of protein, ALAD activity, Pb, and Se in the
          erythrocytes of lead-exposed workers	5-292

6-2.1      Blood lead concentrations in U.S. children, 1-5 years of age	6-11

6-2.2      Simulation of relationship between blood lead concentration and body burden
          in adults	6-13

6-2.3      Simulation of relationship between blood lead concentration and body burden
          in children	6-15

6-2.4      Simulation of temporal relationships between lead exposure and blood lead
          concentration in children	6-17

6-2.5      Simulation of relationships between lead intake and blood lead concentration in
          adults and children	6-19

6-2.6      Cortical lead to blood leads ratios for occupationally-exposed subjects
          (both active and retired) and referents	6-27

6-2.7      Tibia leads to blood lead ratios for environmentally-exposed pregnancy-related
          subj ects, middle-aged to elderly subj ects, and younger subj ects	6-29

6-2.8      Simulation of relationship between urinary lead excretion and body burden
          in adults	6-41

6-2.9      Simulation of relationship between lead intake and urinary lead excretion
          in adults and children	6-43
                                         I-xxxiii

-------
                                    List of Figures
                                        (cont'd)


Number                                                                             Page

6-3.1      Unadjusted and adjusted relationships between average lifetime blood lead
          concentrations and Wechsler Scale performance IQ	6-56

6-3.2      Log-linear model (95% CI shaded) for concurrent blood lead concentration
          adjusted for HOME score, maternal education, maternal IQ, and birth weight	6-69

6-3.3      Log-linear model for concurrent blood lead concentration along with linear
          models for concurrent blood lead levels among children with peak blood
          lead levels above and below 10 |ig/dL	6-69

6-3.4      Golgi-stained section of human cerebral cortex taken from equivalent areas
          of the anterior portion of the middle frontal gyrus at different ages	6-100

6-3.5      Full scale IQ test scores by previous or concurrent blood lead concentration	6-103

6-4.1      Effect on renal function evaluation using age as the effect modifier	6-141

6-4.2      Estimated mean (+2 SE) glomerular filtration rate according to time in
          the chelation group (n = 31) and the control group (n = 30) during the
          observation and intervention periods	6-146

6-5.1      Change in the systolic pressure (effect estimate in mm Hg) associated
          with a doubling of the blood lead concentration	6-179

6-5.2      Change in the diastolic pressure (effect estimate  in mm Hg) associated
          with a doubling of the blood lead concentration	6-180

6-5.3      Effect of doubling mean blood lead on estimate of blood pressure change
          with95%CIs	6-191

6-8.1      Relationship  between blood lead concentration (PbB), age, and serum
          IgE level in children	6-262

6-8.2      Relationship  between blood lead concentration and serum IgE level
          in children	6-263

6-8.3      Relationship  between blood lead concentration (lead) and serum IgE
          level in lead children	6-264
                                         I-xxxiv

-------
                                    List of Figures
                                        (cont'd)


Number                                                                            Page

6-8.4     Relationship between blood lead concentration and serum immunoglobulin
          (Ig) levels in children	6-265

6-8.5     Relationship between blood lead concentration and serum IgE level in
          lead workers  	6-266

6-8.6     Relationship between blood lead concentration and T- and B-cell
          abundances in children	6-270

6-8.7     Relationship between lead exposure and T- and B-cell abundances in
          firearms instructors	6-272

6-9.1     Effects of lead on heme biosynthesis	6-277

6-9.2     Relationship between blood lead and hematocrit in children	6-287

6-9.3     Relationship between blood lead and serum erythropoietin in children	6-291

6-9.4     Association between blood lead concentration and serum erythropoietin
          in pregnant women	6-293

6-10.1    Comparison of a linear and log-linear model to describe the relationship
          between exposure and response	6-327

6-10.2    Concentration-response relationships of IQ to blood lead for the individual
          studies and the pooled analysis by Lanphear et al. (2005)	6-342

6-10.3    Mean blood lead levels adjusted for HOME Score, maternal  education,
          maternal IQ, and birth weight from the pooled analysis of seven studies
          by Lanphear et al. (2005)	6-343

6-10.4    Effect of blood lead on fraction of population with IQ level <70 or
          <50 points	6-345

6-10.5    Distribution of systolic blood pressure in women and men aged 35 to
          64 years from the Framingham Heart Study (Kannel, 2000a)	6-347
                                         I-xxxv

-------
                                     List of Figures
                                         (cont'd)
Number
6-10.6    Relationship of cardiovascular events (coronary disease, stroke, peripheral
          artery disease, cardiac failure) to systolic blood pressure in women and men
          aged 35 to 64 years from the Framingham Heart Study (Kannel, 2000a)	6-348

6-10.7    Effect of blood lead on expected annual risk of cardiovascular events
          per 1,000 person years	6-349

8-1.2.1      Relationship of bioaccessibility versus speciation	8-18

8-1.2.2      Variation of bioavailability with particle size	8-19

8-1.2.3      Illustration of particle lability and bioavailability	8-20

8-1.2.4      Scanning electron micrograph of a large native Pb particle	8-21

8-1.2.5      Bulk lead versus single species modality	8-24

8-1.5.1      Avian toxicity data considered in development of the Eco-SSL	8-80

8-1.5.2      Mammalian toxicity data considered in development of the Eco-SSL	8-91

8-2.3.1      Distribution of lead aqueous species as a function of pH based on
            a concentration of 1  |ig/L lead	8-139

8-2.3.2      Lead speciation versus chloride content	8-140

8-2.3.3      Spatial distribution of natural and ambient surface water/sediment sites	8-147

8-2.3.4      Spatial distribution of natural and ambient liver tissue sample sites	8-148

8-2.3.5      Spatial distribution of natural and ambient whole organism
            tissue sample sites	8-149

8-2.3.6      Frequency distribution of ambient and natural levels of surface water
            dissolved lead (|ig/L)	8-150

8-2.3.7      Spatial distribution of dissolved lead in surface water	8-151

8-2.3.8      Frequency distribution of ambient and natural levels of bulk sediment
            <63 |im total Pb (|ig/g)	8-152
                                          I-xxxvi

-------
                                    List of Figures
                                         (cont'd)

Number

8-2.3.9      Spatial distribution of total lead in bulk sediment <63 jim	8-153

8-2.3.10     Frequency distribution of ambient and natural levels of lead in
            liver tissue (|ig/g dry weight)	8-155

8-2.3.11     Frequency distribution of ambient and natural levels of lead in whole
            organism tissue (|ig/g dry weight)	8-156

8-2.3.12     Spatial distribution of lead in liver tissues (N= 559)	8-158

8-2.3.13     Spatial distribution of lead in whole organism tissues (N = 332)	8-159

8-2.3.14     Lead cycle in an aquatic ecosystem	8-160

8-3.1       The predicted development of metal concentrations in ecosystems for
            four cases of exceedance or non-exceedance of critical limits and of
            critical loads of heavy metals, respectively	8-232
                                         I-xxxvii

-------
                     Authors, Contributors, and Reviewers
                           CHAPTER 1 - INTRODUCTION
Principal Author

Dr. Lester D. Grant—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
            CHAPTER 2 - CHEMISTRY, SOURCES, TRANSPORT OF LEAD
Coordinating Author

Dr. Brooke L. Hemming—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Principle Authors

Ms. Allison Harris—Carnegie-Mellon University, Department of Civil and Environmental
Engineering, Pittsburgh, PA

Dr. Brooke L. Hemming—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Professor Cliff Davidson—Carnegie-Mellon University, Department of Civil and Environmental
Engineering, Pittsburgh, PA

Contributors and Reviewers

Professor Brian Gulson—Macquarie University, Graduate School of the Environment, Sydney,
NSW, Australia

Professor John W. Winchester (Emeritus)—Florida State University, Department of
Oceanography, Tallahassee, FL

Ms. Rosemary Mattuck—Gradient Corporation, Cambridge, MA

Professor Russell Flegal—University of California, Santa Cruz, Department of Environmental
Toxicology, CA
                                      I-xxxviii

-------
                     Authors, Contributors, and Reviewers
                                      (cont'd)
Contributors and Reviewers
(cont'd)

Dr. Beth Hassett-Sipple—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Zachary Pekar—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Joseph Touma—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
          CHAPTER 3 - ROUTES OF HUMAN EXPOSURE AND OBSERVED
                      ENVIRONMENTAL CONCENTRATIONS
Coordinating Author

Dr. Brooke L. Hemming—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Principal Authors

Allison Harris—Carnegie-Mellon University, Pittsburgh, PA

Professor Cliff Davidson—Carnegie-Mellon University, Pittsburgh, PA

Contributors and Reviewers

Dr. Brian Gulson—Macquarie University, Graduate School of the Environment,
Sydney, NSW, Australia

Professor John W. Winchester (Emeritus)—Florida State University, Department of
Oceanography, Tallahassee, FL

Ms. Rosemary Mattuck—Gradient Corporation, Cambridge, MA
                                      I-xxxix

-------
                     Authors, Contributors, and Reviewers
                                      (cont'd)
Contributors and Reviewers
(cont'd)

Dr. Russell Flegal—University of California, Santa Cruz, Department of Environmental
Toxicology, CA

Dr. Sharon Harper—National Exposure Research Laboratory (D205-05), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Beth Hassett-Sipple—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Zachary Pekar—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
       CHAPTER 4 - MODELS OF HUMAN EXPOSURE THAT PREDICT TISSUE
                             DISTRIBUTION OF LEAD
Coordinating Authors

Dr. Robert Elias—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC  27711

Dr. James Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Principal Authors

Dr. Gary Diamond—Syracuse Research Corporation, Syracuse, NY (?)

Dr. Robert Elias—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC  27711
                                        I-xl

-------
                     Authors, Contributors, and Reviewers
                                      (cont'd)
Contributors and Reviewers

Dr. Lester D. Grant—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Brian Gulson—Macquarie University, Graduate School of the Environment,
Sydney, NSW, Australia

Ms. Rosemary Mattuck—Gradient Corporation, Cambridge, MA

Dr. Russell Flegal—University of California, Santa Cruz, Department of Environmental
Toxicology, CA

Dr. Beth Hassett-Sipple—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Zachary Pekar—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
        CHAPTER 5 - TOXICOLOGICAL EFFECTS OF LEAD IN HUMANS AND
                             LABORA TOR Y ANIMALS
Coordinating Authors

Dr. Anu Mudipalli—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Srikanth Nadadur—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Lori White—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
                                        I-xli

-------
                      Authors, Contributors, and Reviewers
                                       (cont'd)
Principal Authors

Dr. Harvey Gonick—University of California, Los Angeles, CA

Dr. Rodney Dietert—Cornell University, Ithaca, NY

Dr. John Rosen—Montefiore Medical Center, Bronx, NY

Dr. Stephen Lasley—University of Illinois, Peoria, IL

Dr. Gene Watson—University of Rochester, Rochester, NY

Dr. John Pierce Wise—University of Southern Maine

Dr. N.D. Vasiri—University of California - Irvine, Irvine, CA

Dr. Gary Diamond—Syracuse Research Corporation, Syracuse, NY

Contributors and Reviewers

Dr. Lester D. Grant—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Paul Reinhart—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. David A. Lawrence—Dept of Environmental and Clinical Immunology, Albany, NY

Dr Michael J. McCabe, Jr.—University of Rochester, Rochester, NY

Dr. Theodore I. Lidsky—N.Y.S. Inst. for Basic Research in Developmental Disabilities,
Staten Island, NY

Dr. Beth Hassett-Sipple—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Zachary Pekar—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
                                         I-xlii

-------
                     Authors, Contributors, and Reviewers
                                      (cont'd)
          CHAPTER 6 - EPIDEMIOLOGICAL STUDIES OF AMBIENT LEAD
                               EXPOSURE EFFECTS
Coordinating Authors

Dr. Jee-Young Kim—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Dennis Kotchmar—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Principal Authors

Dr. David Bellinger—Children's Hospital, Boston, MA

Dr. Margit Bleeker—Center for Occupational and Environmental Neurology, Baltimore, MD

Dr. Gary Diamond—Syracuse Research Corporation, Syracuse, NY

Dr. Kim Dietrich—University of Cincinnati, Cincinnati, OH

Dr. Pam Factor-Litvak—Columbia University, NY

Dr. Brian Gulson—Macquarie University, Sydney, Australia

Dr. Vic Hasselblad—Duke University, Durham, NC

Dr. Steve Rothenberg—Centre de Investigation y de Estudios Avanzados, Merida, Mexico

Dr. Neal Simonsen—Louisiana State University Health Sciences Center, New Orleans, LA

Dr. Kyle Steenland—Emory University, Atlanta, GA

Dr. David Svendsgaard—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Virginia Weaver—Johns Hopkins University, Baltimore, MD
                                       I-xliii

-------
                     Authors, Contributors, and Reviewers
                                      (cont'd)
Contributors and Reviewers

Dr. J. Michael Davis—National Center for Environmental Assessment (B243-01),
U.S.  Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Lester D. Grant—National Center for Environmental Assessment (B243-01),
U.S.  Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Beth Hassett-Sipple—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Kaz Ito—New York University, Tuxedo, NY

Dr. Kathryn Mahaffey—Office of Prevention, Pesticides and Toxic Substances,
U.S.  Environmental Protection Agency, Washington, DC 20460

Dr. Karen Martin—Office of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711

Dr. Zachary Pekar—Office of Air Quality Planning and Standards, U.S.  Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Mary Ross—Office of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711
    CHAPTER 7- INTEGRATIVE SYNTHESIS OF HUMAN LEAD EXPOSURE AND
                                  HEALTH RISKS
Coordinating Authors
(To be included in Second External Review Draft)

Principal Authors
(To be included in Second External Review Draft)

Contributors and Reviewers
(To be included in Second External Review Draft)
                                        I-xliv

-------
                     Authors, Contributors, and Reviewers
                                      (cont'd)
                CHAPTER 8 - ENVIRONMENTAL EFFECTS OF LEAD


Coordinating Author

Dr. Timothy Lewis—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Principle Authors

Dr. Ruth Hull—Cantox Environmental Inc., Mississauga, Ontario, Canada

Dr. James Kaste—Dartmouth College, Hanover, NH

Dr. John Drexler—University of Colorado, Boulder, CO

Dr. Chris Johnson—Syracuse University, Syracuse, NY

Dr. Linda Chappell—U.S. EPA, OAQPS, RTF, NC

Dr. Bill Stubblefield—Parametrix, Inc. Albany, OR

Dr. Dwayne Moore—Cantox Environmental, Inc., Ottawa, Ontario, Canada

Dr. David Mayfield—Parametrix, Inc., Bellevue, WA

Dr. Barbara Southworth—Menzie-Cura & Associates, Inc., Winchester, MA

Dr. Katherine Von Stackleberg—Menzie-Cura & Associates, Inc., Winchester, MA

Contributors and Reviewers

Dr. Jerome Nriagu—University of Michigan, Ann Arbor, MI

Dr. Judith Weis—Rutgers University, Newark, NJ

Dr. Sharon Harper—National Exposure Research Laboratory (D205-05), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Dr. Karen Bradham—National Research Exposure Laboratory (D205-05), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
                                        I-xlv

-------
                      Authors, Contributors, and Reviewers
                                       (cont'd)
Contributors and Reviewers
(cont'd)

Dr. Ginger Tennant—Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711

Ms. Gail Lacey—Office of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711
[Note: Any inadvertently omitted names of authors/reviewers will be inserted in the Second
External Review Draft and final version of this LeadAQCD, as will more complete addresses
for all authors/reviewers.]
                                         I-xlvi

-------
              U.S. Environmental Protection Agency Project Team
                for Development of Air Quality Criteria for Lead
Executive Direction

Dr. Lester D. Grant (Director)—National Center for Environmental Assessment-RTF Division,
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Scientific Staff

Dr. Robert Elias (Lead Team Leader)—National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. James S. Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Brooke Hemming—National Center for Environmental Assessment (B243-01), U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Dennis Kotchmar—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Jee-Young Kim—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Timothy Lewis—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Paul Reinhart—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Anu Muldipalli—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Srikanth Nadadur—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. David Svendsgaard—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Dr. Lori White—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
                                       I-xlvii

-------
             U.S. Environmental Protection Agency Project Team
                for Development of Air Quality Criteria for Lead
                                      (cont'd)
Technical Support Staff

Mr. Douglas B. Fennell—Technical Information Specialist, National Center for Environmental
Assessment (B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC
27711

Ms. Emily R. Lee—Management Analyst, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC  27711

Ms. Diane H. Ray—Program Specialist, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC  27711

Ms. Donna Wicker—Administrative Officer, National Center for Environmental  Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC  27711 (retired)

Mr. Richard Wilson—Clerk, National Center for Environmental  Assessment (B243-01), U.S.
Environmental Protection Agency, Research Triangle Park, NC  27711
                                       I-xlviii

-------
              U.S. Environmental Protection Agency Project Team
               for Development of Air Quality Criteria for Lead
                                       (cont'd)
Document Production Staff

Ms. Carolyn T. Perry—Task Order Manager, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560

Mr. John A. Bennett—Technical Information Specialist, Library Associates of Maryland,
11820 Parklawn Drive, Suite 400, Rockville, MD 20852

Mr. William Ellis—Records Management Technician, InfoPro, Inc., 8200 Greensboro Drive,
Suite 1450, McLean, VA 22102

Ms. Sandra L. Hughey—Technical Information Specialist, Library Associates of Maryland,
11820 Parklawn Drive, Suite 400, Rockville, MD 20852

Dr. Barbara Liljequist—Technical Editor, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560

Ms. Rosemary Procko—Senior Word Processor, TekSystems, 1201 Edwards Mill Road,
Suite 201, Raleigh, NC 27607

Ms. Faye Silliman—Word Processor, InfoPro, Inc., 8200 Greensboro Drive, Suite 1450,
McLean, VA  22102

Mr. Carlton Witherspoon—Graphic Artist, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560
                                        I-xlix

-------
                     U.S. Environmental Protection Agency
                   Science Advisory Board (SAB) Staff Office
              Clean Air Scientific Advisory Committee (CASAC)
CHAIR
Dr. Rogene Henderson, Scientist Emeritus, Lovelace Respiratory Research Institute,
Albuquerque, NM
MEMBERS
Dr. Ellis Cowling, University Distinguished Professor-at-Large, North Carolina State
University, Colleges of Natural Resources and Agriculture and Life Sciences, North Carolina
State University, Raleigh, NC

Dr. James D. Crapo, Professor, Department of Medicine, National Jewish Medical and
Research Center, Denver, CO

Dr. Frederick J. Miller, Consultant, Gary, NC

Mr. Richard L. Poirot, Environmental Analyst, Air Pollution Control Division, Department of
Environmental Conservation, Vermont Agency of Natural Resources, Waterbury, VT

Dr. Frank Speizer, Edward Kass Professor of Medicine, Channing Laboratory, Harvard
Medical School, Boston, MA

Dr. Barbara Zielinska, Research Professor, Division of Atmospheric Science, Desert Research
Institute, Reno, NV
SCIENCE ADVISORY BOARD STAFF
Mr. Fred Butterfield, CASAC Designated Federal Officer, 1200 Pennsylvania Avenue, N.W.,
Washington, DC, 20460, Phone: 202-343-9994, Fax: 202-233-0643 (butterfield.fred@,epa.gov)
CASAC External Roster FY 2006.wpd

-------
                          Abbreviations and Acronyms
aFGF
AA
AALM
AAS
ACBP
ACE
ACR
ACSL
ADC
ADHD
ADP
AEA
AF
2-AG
A horizon
AHR
ALA
ALAD
ALAS
ALS
ALT
AMD
AMP
ANF
ANOVA
AP-1
APE
AQCD
ASV
ATP
ATP 142
ATPase
a-fibroblast growth factor
arachidonic acid; atomic absorption
All Ages Lead Model
atomic absorption spectroscopy
Achenbach Child Behavior Profile
angiotensin converting enzyme
acute-chronic ratio
Advanced Continuous Simulation Language
analog-digital converter
Attention Deficit/Hyperactivity Disorder
adenosine dinucleotide phosphate
TV-arachi dony 1 ethanol amine
absorption fraction
2-arachidonylglycerol
uppermost layer of soil (litter and humus)
aryl hydrocarbon receptor
5-aminolevulinic acid
5-aminolevulinic acid dehydratase
aminolevulinic acid synthase
amyotrophic lateral sclerosis
alanine aminotransferase; alanine transferase
activity mean diameter
adenosine monophosphate
atrial natriuretic factor
analysis of variance
activated protein-1
apurinic endonuclease
Air Quality Criteria Document
anode stripping voltammetry
adenosine triphosphate
sodium-potassium adenosine triphosphate a2
adenosine triphosphate synthase
                                        I-li

-------
ATSDR
AVS
AWQC
P
PFGF
6-p-OH-cortisol
Bcell
BAEP
BAER
BAF
BCF
BLL
BLM
BMDM
BMI
BMP-6
BRHS
BTQ
BUN
BW, bw
CA
45Ca
CA1
CAS
CAA
Ca-ATPase
43CaCl2
CaCO3
CaEDTA
CAL
cAMP
CaNa2  EDTA
CANTAB
CASAC
Agency for Toxic Substances and Disease Research
acid volatile sulfide
ambient water quality criteria
beta-coefficient; slope of an equation
P-fibroblast growth factor
6-p-hydroxycortisol
B lymphocyte
brainstem auditory-evoked potentials
brainstem auditory-evoked responses
bioaccumulation factor
bioconcentration factor
blood lead level
biotic ligand model
bone marrow-derived macrophages
body mass index
bone morphogenic protein-6
British Regional Heart Study
Boston Teacher Questionnaire
blood urea nitrogen
body weight
chromosomal aberration
calcium-45 radionuclide
cornu ammonis 1 region of hippocampus
cornu ammonis 3 region of hippocampus
Clean Air Act
calcium-dependent adenosine triphosphatase
calcium-43 radionuclide-labeled calcium chloride
calcium carbonate
calcium disodium ethylenediaminetetraacetic acid
calcitonin
cyclic adenosinemonophosphate
calcium disodium ethylenediaminetetraacetic acid
Cambridge Neuropsychological Testing Automated Battery
Clean Air Scientific Advisory Committee
                                        I-lii

-------
CBCL
CCE
CDC
CEC
CESD, CES-D
cGMP

CI
CLRTAP
CMC
CMI
CNS
ConA
COX-2
CP
CPT
CRAC
CRT
CSF
CSF-1
CTH
CTL
CuZn-SOD
CYP2A6
CYP3A4
CYP450
DET
DPS
dfs
DOT
DL
DMEM
DMFS
DMSA
Achenbach Child Behavior Checklist
Coordination Center for Effects
Centers for Disease Control and Prevention
cation exchange capicity
Center for Epidemiologic Studies Depression (scale)
cyclic guanosine-3',5'-monophosphate; cyclic
guanylylmonophosphate
confidence interval
Convention on Long-range Transboundary of Air Pollution
criterion maximum concentration
cell-mediated immunity
central nervous system
concanavalin A
cyclooxygenase-2
coproporphyrin
current perception threshold
calcium release activated calcium reflux
chronic renal insufficiency
cerebrospinal fluid
colony-stimulating factor-1
cystathionine gamma-lyase
cytotoxic T lymphocyte
copper and zinc-dependent superoxide dismutase
cytochrome P-2A6
cytochrome P-3 A4
cytocrome P-450
diffusive  equilibrium thin films
decayed or filled surfaces, permanent teeth
covariate-adjusted number of caries
diffusive  gradient thin films
detection  limit
Dulbecco's modified eagle medium
decayed,  missing, or filled surfaces, permanent teeth
2,3-dimercaptosuccinic acid
                                       I-liii

-------
DMTU
DNA
DNTC
DOC
DOM
DOS
DPASV
DTK
DTPA
dw
E2
Eb
EBE
eCB

ECF
Eco-SSL
EDRF
EDS
EDTA
EEG
EOF
eNOS
EOD
EP
EPA
EPMA
EPT

EqP
ERG
EROD
ESP
dimethyl thio urea
deoxyribonucleic acid
diffuse neurofibrillary tangles with calcification
dissolved organic carbon
dissolved organic matter
Disc Operating System
differential pulse anode stripping voltammetry
delayed type hypersensitivity
diethylenetriaminepentaacetic acid
dry weight
estradiol
electron binding energies
early biological effect
effect concentration for 50% of test population
endocannabinoid (e.g., 2-arachidonylglycerol [2-AG]
and TV-arachidonylethanolamine [AEA])
extracellular fluid
ecological soil screening level
endothelium-derived relaxing  factor
energy dispersive spectrometers
ethylenediaminetetraacetic acid
el ectroencephal ogram
epidermal growth factor
endothelial nitric oxide synthase
explosive ordnance disposal
erythrocyte protoporphyrin
U.S. Environmental Protection Agency
electron probe microanalysis
macroinvertebrates from the Ephemeroptera (mayflies),
Plecoptera (stoneflies), and  Trichoptera (caddisflies) group
equilibrium partitioning (theory)
electroretinogram
ethoxyresorufin-O-deethylase
electrostatic precipitator
                                         I-liv

-------
ESRD
ET
EXAFS
EXANES
FAV
FDA
FEF
FEVi
FIAM
FMLP
foe
FPLC
FR
FSH
FT3
FT4
FVC
G6PD
GABA
GAG
GCI
GD
GDP
GEE
GFAAS
GFR
GH
GI
GM
GMAV
GMP
GnRH
goc
GP
end-stage renal disease
endothelein; essential tremor
Extended X-ray absorption fine structure
Extended X-ray absorption near edge spectroscopy
final acute value
Food and Drug Administration
forced expiratory flow
forced expiratory volume in one second
free ion activity model
N-formyl-L-methionyl-L-leucyl-L-phenylalanine
fraction organic carbon
fast protein liquid chromatography
federal register
follicle stimulating hormone
free triiodothyronine
free thyroxine
forced vital capacity
glucose-6-phosphatedehydrogenase
gamma aminobutyric acid
glycosaminoglycan
General Cognitive Index
gestational day
guanosine diphosphate
generalized estimating equations
graphite furnace atomic absorption spectroscopy
glomerular filtration rate
growth hormone
gastrointestinal
geometric mean
genus mean acute value
guanosine monophosphate
gonadotropin releasing hormone
grams organic carbon)
gross productivity
                                        I-lv

-------
GPEI
GRP78
GSD
GSD;
GSH
GSIM
GSSG
GST
GTP
H+
H202
Hb
HBEF
Hct
HDL
HFE
HFF
HH
HHC
5-HI A A
HOME
HPLC
HQ
HSI
HSPG
HTN
HVA
IARC
ICP
ICP-AES
ICP-MS
ICRP
IDMS
IEC
glutathione S-transferase P enhancer element
glucose-regulated protein 78
geometric standard deviation
individual geometric standard deviation
glutathione; reduced glutathione
gill surface interaction model
oxidized glutathione
glutathione transferase; glutathione S-transferase
guanosine triphosphate
acidity
hydrogen peroxide
hemoglobin
Hubbard Brook Experimental Forest
hematocrit
high-density lipoprotein (cholesterol)
hemochromatosis gene
human foreskin fibroblasts
hydroxylamine hydrochloride
hereditary hemochromatosis
5-hydroxyindoleacetic
Home Observation for Measurement of Environment
high-pressure liquid chromatography
hazard quotient
habitat suitability index
heparan sulfate proteoglycan
hypertension
homovanillic acid
International Agency for Research on Cancer
inductively coupled plasma
inductively coupled plasma atomic emission spectroscopy
inductively coupled plasma mass spectrometry
International Commission on Radiological Protection
isotope dilution mass spectrometry
intestinal epithelial cells
                                        I-lvi

-------
IEUBK
IFN
Ig
IGFi
IL
IMPROVE

iNOS
i.p., IP
IQ
IQR
IT
i.v., IV
KTEA
K-XRF
LA
LCso
LDL
L-dopa
LH
LMW
L-NAME
LOAEL
LOEC
LPO
LPS
LT50
LTD
LTP
LVH
liPIXE
liSXRF
MAO
MATC
Integrated Exposure Uptake Biokinetic (model)
interferon (e.g., IFN-y)
immunoglobulin (e.g., IgA, IgE, IgG, IgM)
insulin-like growth factor 1
interleukin (e.g., IL-1, IL-lp, IL-4, IL-6, IL-12)
Interagency Monitoring of Protected Visual Environments
(network)
inducible nitric oxide synthase
intraperitoneal
intelligence quotient
interquartile range
intrathecal
intravenous
Kaufman Test of Educational Achievement
K-shell X-ray fluorescence
lipoic acid
lethal concentration (at which 50% of exposed animals die)
low-density lipoprotein (cholesterol)
3,4-dihydroxyphenylalanine (precursor of dopamine)
luteinizing hormone
low molecular weight
L-NG-nitroarginine methyl ester
lowest-observed adverse effect level
lowest-observed-effect concentration
lipid peroxide; lipid peroxidation
lipopolysaccharide
time to reach 50% mortality
long-term depression
long-term potentiation
left ventricular hypertrophy
microfocused particle induced X-ray emission
microfocused synchrotron-based X-ray fluorescence
monoaminoxidase
maximum acceptable threshold  concentration
                                        I-lvii

-------
MCH
MCHC
MCV
MDA
MDA-TBA
MDI
MDRD
meso-DMSA
Mg-ATPase
MHC
miDMSA

MINTEQ
MINTEQA2
MK-801
MLR
MMSE
MN
Mn-SOD
MRI
mRNA
MRS
MSV
MT
MVV
N, n
N/A
NAA
NAAQS
NAC
NAD
NADH

NADP
mean corpuscular hemoglobin
mean corpuscular hemoglobin concentration
mean corpuscular volume
malondialdehyde
malondialdehyde-thiobarbituric acid
Mental Development Index
Modification of Diet in Renal Disease (study)
meso-2,3-dimercaptosuccinic acid
magnesium-dependent adenosine triphosphatase
major histocompatibility complex
mono-3-methylbutane-l-yl (monoisomyl) ester of
meso-2,3-dimercaptosuccinic acid
thermodynamic equilibrium model
equilibrium speciation computer model
NMD A receptor antagonist
mixed lymphocyte response
Mini-Mental State Examination
micronuclei formation
manganese-dependent superoxide dismutase
magnetic resonance imaging
messenger ribonucleic acid
magnetic resonance spectroscopy
Moloney sarcoma virus
metallothionein
maximum voluntary ventilation
number of observations
not available
N-acetylaspartate; neutron activation analysis
National Ambient Air Quality Standards
N-acetyl cysteine
nicotinamide adenine nucleotide
reduced nicotinamide adenine dinucleotide; nicotinamide adenine
dinucleotide dehydrogenase
nicotinamide adenine dinucleotide phosphate
                                       I-lviii

-------
NAD(P)H
NADS
NAG
Na-K-ATPase
NAS
NASCAR
NAWQA
NCBP
NCEA-RTP

ND
NE
NEPSY
NF-KB
NHANES
NIOSH
NIST
NK
NMDA
nNOS
NO
N03
NOAEL
NOD
NOEC
NOM
NOS
NP
NR
NRC
NTP
02
OAQPS
OAR
reduced nicotinamide adenine dinucleotide phosphate
nicotinamide adenine dinucleotide synthase
N-acetyl-p-D-glucosaminidase
sodium-potassium-dependent adenosine triphosphatase
Veterans' Administration Normative Aging Study
National Association for Stock Car Automobile Racing
National Water-Quality Assessment
National Contaminant Biomonitoring Program
National Center for Experimental Assessment Division in Research
Triangle Park, NC
not detected; non-detectable
norepinephrine
Developmental Neuropsychological Assessment
nuclear transcription factor-KB
National Health and Nutrition Examination Survey
National Institute for Occupational Safety and Health
National Institute for Standards and Technology
natural killer
N-methyl-D-aspartate
neuronal nitric oxide synthase
nitric oxide
nitrate
no-observed-adverse-effect level
autoimmune diabetes prone strain of mice
no-observed-effect concentration
natural organic matter
nitric oxide synthase
net productivity
not reported
National Research Council
National Toxicology Program
superoxide ion
Office of Air Quality Planning and Standards
Office of Air and Radiation
                                        I-lix

-------
O/E
OH
1,25-OH-D
25-OH-D
1,25-(OH2)D3
25-OH D3
O horizon
ONOO
OR
ORD
P
Pio

PAD
PAH
PAI-1
Pb
203Pb
204Pb, 206Pb, 207Pb, 208Pb
PbB
PbCO3
PBG-S
Pb(OH)2
Pb(N03)2
PbS
PC12
PFCs
PG
PHA
P,
PIR
PIXE
PKC
PKC-a
observed-expected ratio
hydroxyl
1,25-dihydroxyvitamin D
25-hydroxyvitamin D
vitamin D
25-hydroxycholecalciferol
forest floor
peroxynitrate ion
odds ratio
Office of Research and Development
probability value
probability for the occurrence of a blood lead concentration
exceeding 10 |ig/dL
peripheral arterial disease
polycyclic aromatic hydrocarbon
plasminogen activator inhibitor-1
lead
lead-203 radionuclide
stable isotopes of lead-204, -206, -207, -208 respectively
blood lead; blood lead concentration
lead carbonate
porphobilinogen synthase
lead hydroxide
lead nitrate
galena
pheochromocytoma cell
plaque forming cells
prostaglandin (e.g., PGE2 ,PGF2)
phytohemagglutinin A
inorganic phosphorus
poverty-income ratio
particle induced X-ray emission
protein kinase C
protein kinase C a
                                         I-lx

-------
plasma-ECF
PM
PM2.5
PMN
PMNL
P5N
PND
p.o.,PO
POMS
ppb
ppm
PRL
PTH
PTHrP
PVC
PWM
Q
QA/QC
Qco2
r
r2
RAAS
rac-DMSA
RBA
RBC
RBP
RCPM
RDA
RDW
ROS
ROS 17.2.8
RR
RSD
plasma and extracellular fluid combined
particulate matter
combination of coarse and fine particulate matter
fine particulate matter
polymorphonuclear leukocyte
polymorphonuclear leukocyte
pyrimidine  5'-nucleotidase
postnatal day
per os (oral administration)
Profile of Mood States
parts per billion
parts per million
prolactin
parathyroid hormone
parathyroid hormone-related protein
polyvinyl chloride
pokeweed mitogen
flux of air
quality assurance/quality control
flux of carbon dioxide
Pearson correlation coefficient
correlation  coefficient
renin-angiotensin-aldosterone system
racemic-2,3-meso-2,3-dimercaptosuccinic acid
relative bioavailablity
red blood cell; erythrocyte
retinol binding protein
Ravens Colored Progressive Matrices
recommended daily allowance
red cell distribution
reactive oxygen species
rat osteosarcoma cell line
relative risk
relative standard deviation
                                        I-lxi

-------
ZSEM
ZTU
SAB
SAM
s.c., SC
SCE
SD
SDS-PAGE
SE
SEM
SES
SHBG
SIMS
SIR
SLP
SMAV
SMR
SO2
SOD
SOILCHEM
SRBC
SRC
SRD
SRE
SRM
SRT
STORE!

Ti/2; ti/2
T3
T4
TEA
TEARS
Tc
sum of the molar concentrations of simultaneously extracted metal
summed of toxic units for all metals in a mixture
Science Advisory Board
S-adenosyl methionine
subcutaneous
sister chromatid exchange
standard deviation; Spraque-Dawley (rat)
sodium dodecyl sulfate-polyacrylamide gel electrophoresis
standard error; Staphylococcus aureus enterotoxin
simultaneously extracted metal; standard error of the mean
socioeconomic status
sex hormone binding globulin
secondary ion mass spectrometry
standardized incidence ratio
synthetic leaching procedure
species mean acute value
standardized mortality ratio
sulfur dioxide
superoxide dismutase
chemical species equilibrium  model
sheep red blood cell
Syracuse Research Corporation
Self Report of Delinquent Behavior
sterol regulatory element
standard reference materials
simple reaction time
database for the STOrage and RETrieval of chemical, physical,
and biological  data
half-time
triiodothyronine
thyroxine
thiobarbituric acid
thiobarbituric acid-reactive species
cytotoxic T lymphocyte
                                        I-lxii

-------
Tcell
TCLP
TEL
TES
TGF
TH
ThO
Thl
Th2
232Th
tHct
tHcy
203y, 205yi
TLC
TM
TML
TNF
tPA
TPBS
TRY
TSH
TSP
TT3
TT4
TTR
TU
TWA
TX
235U, 238U
UDP
UNECE
USGS
UV
V
T lymphocyte
toxic characteristic leaching procedure
tetraethyllead; tri ethyl lead
testosterone
transforming growth factor (e.g., TGF-a ,TGF-P, TGF-pl)
T-helper lymphocyte
precursor T lymphocyte
T-derived lymphocyte helper 1
T-derived lymphocyte helper 2
stable isotope of thorium-232
total hematocrit
plasma total homocysteine
stable isotopes  of thallium-203 and -205, respectively
treatment of lead-exposed children
T-memory lymphocyte
tetramethyllead
tumor necrosis  factor (e.g., TNF-a, TNF-pl)
plasminogen activator
Total Problem Behavior Score
toxicity reference value
thyroid stimulating hormone
total suspended particulates
total triiodothyronine
serum total thyroxine
transthyretin
toxic unit
time-weighted  average
tromboxane (e.g., TXB2)
uranium-234 and -238 radionuclides
uridine diphosphate
United Nations Economic Commission for Europe
United States Geological Survey
ultraviolet
volume of culture
                                       I-lxiii

-------
Vd
VDR
VEP
vitamin D
VLDL
VMI
VSMC
w/v
WDS
WHO
WIC
WISC-III
WRAT-R
ww
XANES
XAS
XPS
X-rays
XRD
XRF
deposition velocity
vitamin D receptor
visual-evoked potential
1,25-dihydroxyvitamin D3
very low density lipoprotein (cholesterol)
visual motor integration
vascular smooth muscle cells
weight per volume
wavelength dispersive spectrometers
World Health Organization
Women, Infants, and Children (program)
Wechsler Intelligence Scale for Children-Ill
Wide Range Achievement Test-Revised
wet weight
extended X-ray absorption near edge spectroscopy
X-ray absorption spectroscopy
X-ray photoelectron spectroscopy
synchrotron radiation
X-ray diffraction
X-ray fluorescence
                                        I-lxiv

-------
 i                              1.     INTRODUCTION
 2
 3
 4          The present document critically assesses the latest scientific information concerning
 5    health and welfare effects associated with the presence of various concentrations of lead (Pb)
 6    in ambient air, as pertinent to providing updated scientific bases for EPA's current periodic
 7    review of the National Ambient Air Quality Standards for Lead (Pb NAAQS).  As such, this
 8    document builds upon previous assessments published by the U.S. Environmental Protection
 9    Agency (EPA), including:  (a) the document, Air Quality Criteria for Lead (U.S. Environmental
10    Protection Agency, 1977); (b) an updated revision of that Lead Air Quality Criteria Document
11    (Lead AQCD) and an accompanying Addendum published in 1986 (U.S. Environmental
12    Protection Agency, 1986a,b); as well as (c) an associated 1990 Supplement (U.S. Environmental
13    Protection Agency, 1990). This document focuses on evaluation and integration of information
14    relevant to Pb NAAQS criteria development that has become available mainly since that covered
15    by the 1986 and 1990 criteria assessments.
16          This introductory chapter (Chapter 1) of the revised Lead AQCD presents:
17    (a) background information on pertinent Clean Air Act legislative requirements, the criteria and
18    NAAQS review process, and the history of previous Pb criteria reviews; (b) an overview of the
19    current Pb criteria review process, associated key milestones, and projected schedule; and
20    (c) an orientation to the general organizational structure and content of this revised Lead AQCD.
21
22
23    1.1    LEGAL AND HISTORICAL BACKGROUND
24    1.1.1   Legislative Requirements
25          Two sections of the Clean Air Act (CAA) govern the establishment, review, and revision
26    of NAAQS. Section  108 (42 U.S.C. 7408) directs the Administrator of the U.S. Environmental
27    Protection Agency (EPA) to identify ambient air pollutants that may be reasonably anticipated to
28    endanger public health or welfare and to issue air quality criteria for them (U.S. Code, 2003a).
29    These air quality criteria are to reflect the latest scientific information useful in indicating the
30    kind and extent of all identifiable effects on public health or welfare that may be expected  from
31    the presence of a given pollutant in ambient air.

      December 2005                           1-1       DRAFT-DO NOT QUOTE OR CITE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
1 1
12
13
14
15
16
17
18
19
20
21
       Section 109(a) of the CAA (42 U.S.C. 7409) directs the Administrator of EPA to propose
and promulgate primary and secondary NAAQS for pollutants identified under Section 108 (U.S.
Code, 2003b). Section 109(b)(l) defines a primary standard as one that, in the judgment of the
Administrator, is requisite to protect the public health (see inset below) based on the criteria and
allowing for an adequate margin of safety. The secondary standard, as defined in Section
109(b)(2), must specify a level of air quality that, in the judgment of the Administrator, is
requisite to protect the public welfare (see inset below) from any known or anticipated adverse
effects associated with the presence of the pollutant in ambient air, based on the criteria.
         EXAMPLES OF PUBLIC
           HEALTH EFFECTS
     Effects on the health of the general
     population, or identifiable groups within the
     population, who are exposed to pollutants in
     ambient air
     Effects on mortality
     Effects on morbidity
     Effects on other health conditions including
     indicators of:
          •  pre-morbid processes,
          •  risk factors, and
          •  disease
                                                              EXAMPLES OF PUBLIC
                                                               WELFARE EFFECTS
                                                         Effects on personal comfort and well-being
                                                         Effects on economic values
                                                         Deterioration of property
                                                         Hazards to transportation
                                                         Effects on the environment, including:
                                                              animals
                                                              climate
                                                              crops
                                                              materials
                                                              soils
                    vegetation
                    visibility
                    water
                    weather
                    wildlife
       Section 109(d) of the CAA (42 U.S.C. 7409) requires periodic review and, if appropriate,
revision of existing criteria and standards (U.S. Code, 2003b).  If, in the Administrator's
judgment, the Agency's review and revision of criteria make appropriate the proposal of new or
revised standards, such standards are to be revised and promulgated in accordance with Section
109(b). Alternatively, the Administrator may find that revision of the standards is inappropriate
and conclude the review by leaving the existing standards unchanged. Section 109(d)(2) of the
1977 CAA Amendments also requires that an independent scientific review committee be
established to advise the EPA Administrator on NAAQS matters, including the scientific
soundness of criteria (scientific bases) supporting NAAQS decisions. This role is fulfilled by the
Clean Air Scientific Advisory Committee (CASAC), which is administratively supported by
EPA's Science Advisory Board (SAB).
December 2005
                                                  1-2
DRAFT-DO NOT QUOTE OR CITE

-------
 1    1.1.2   Criteria and NAAQS Review Process
 2          Periodic reviews by EPA of criteria and NAAQS for a given criteria air pollutant progress
 3    through a number of steps, beginning with preparation of an air quality criteria document
 4    (AQCD) by the National Center for Environmental Assessment Division in Research Triangle
 5    Park, NC (NCEA-RTP), a unit within EPA's Office of Research and Development (ORD).  The
 6    AQCD provides  a critical assessment of the latest available scientific information upon which
 7    the NAAQS are to be based.  Drawing upon the AQCD, the Office of Air Quality Planning and
 8    Standards (OAQPS), a unit within EPA's Office of Air and Radiation (OAR), prepares a Staff
 9    Paper that (a) evaluates policy implications of the key studies and scientific information
10    contained in the AQCD; (b) presents relevant exposure and risk analyses; and (c) presents EPA
11    staff conclusions and recommendations for standard-setting options for the EPA Administrator to
12    consider. The Staff Paper is intended to help "bridge the gap" between the scientific assessment
13    contained in the AQCD and the judgments required of the Administrator in determining whether
14    it is appropriate to retain or to revise the NAAQS.
15          Iterative drafts of both the AQCD and the Staff Paper (as well as other analyses, such as
16    associated exposure and/or risk assessments supporting the  Staff Paper) are made available for
17    public comment and CAS AC review. Final versions of the  AQCD and Staff Paper incorporate
18    changes in response to CAS AC review and public comment. Based on the information in these
19    documents, the EPA Administrator proposes decisions on whether to retain or revise the subject
20    NAAQS, taking into account public comments and CASAC advice  and recommendations.  The
21    Administrator's proposed decisions are published in the Federal Register,  with a preamble that
22    delineates the rationale for the decisions and solicits public comment. After considering
23    comments received on the proposed decisions, the Administrator makes a final decision, which is
24    promulgated via  a Federal Register notice that addresses significant comments received on the
25    proposal.
26          Promulgated  NAAQS  decisions involve consideration of the four basic elements of a
27    standard: indicator, averaging time, form, and level. The indicator defines the pollutant to be
28    measured in the ambient air for the purpose of determining compliance with the standard.  The
29    averaging time defines the time period over which air quality measurements are to be obtained
30    and averaged, considering evidence of effects associated with various time periods of exposure.
31    The form of a standard defines the air quality statistic that is to be compared to the level of the

      December 2005                            1 -3       DRAFT-DO NOT QUOTE OR CITE

-------
 1    standard (i.e., an ambient concentration of the indicator pollutant) in determining whether an area
 2    attains the standard. The form of the standard specifies the air quality measurements that are to
 3    be used for compliance purposes (e.g., the 98th percentile of an annual distribution of daily
 4    concentrations; the annual arithmetic average), the monitors from which the measurements are to
 5    be obtained (e.g., one or more population-oriented monitors in an area), and whether the statistic
 6    is to be averaged across multiple years. These basic elements of a standard are the primary focus
 7    of the staff conclusions and recommendations posed in the Staff Paper and are explicitly
 8    specified in the ensuing NAAQS rulemaking, building upon the policy-relevant scientific
 9    information assessed in the AQCD and on the policy analyses contained in the Staff Paper.
10    These four elements taken together determine the degree of public health and welfare protection
11    afforded by the NAAQS.
12
13    1.1.3   Regulatory Chronology
14          In 1971, U.S. EPA promulgated national ambient air standards for several major "criteria"
15    pollutants (see Federal Register, 1971), but did not include lead among them at that time. Later,
16    on October 5, 1978, the EPA promulgated primary and secondary NAAQS for lead, under
17    Section 109 of the CAA (43 FR 46258), as  announced in the Federal Register (1979). The
18    primary standard and the secondary standard are the same: 1.5 |ig/m3 as a quarterly average
19    (maximum arithmetic mean averaged over 90 days). The standards were based on the EPA's
20    1977 Air Quality Criteria for Lead (U.S. Environmental Protection Agency, 1977).
21          In 1986, the EPA published a revised Air Quality Criteria Document for Lead (U.S.
22    Environmental Protection Agency, 1986a).  The 1986 AQCD assessed newly available scientific
23    information on the health and welfare effects associated with exposure to various concentrations
24    of lead in ambient air, based on literature published through 1985. The 1986 document was
25    principally concerned with the health  and welfare effects of lead, but other scientific data were
26    also discussed in order to provide a better understanding of the pollutant in the environment.
27    Thus, the 1986 document included chapters that discussed the atmospheric chemistry and
28    physics of the pollutant; analytical approaches; environmental concentrations; human exposure
29    and dosimetry; physiological, lexicological, clinical, and epidemiological aspects of lead health
30    effects; and lead effects on ecosystems. An Addendum to the 1986 Lead AQCD was also
31    published along with it (U.S. Environmental Protection Agency, 1986b). Subsequently,

      December 2005                             1 -4        DRAFT-DO NOT QUOTE OR CITE

-------
 1    a Supplement to the 1986 Lead AQCD/Addendum was published by EPA in 1990 (U.S.
 2    Environmental Protection Agency, 1990a). That 1990 Supplement evaluated still newer
 3    information emerging in the published literature concerning (a) lead effects on blood pressure
 4    and other cardiovascular endpoints and (b) the effects of lead exposure during pregnancy or
 5    during the early postnatal period on birth outcomes and/or on the neonatal physical and
 6    neuropsychological development of affected infants and children.
 7          The evaluations contained in the 1986 Lead AQCD/Addendum and the 1990 Supplement
 8    provided scientific inputs to support decision-making regarding periodic review and, as
 9    appropriate, revision of the Lead NAAQS; and they were  drawn upon by EPA's Office of Air
10    Quality Planning and Standards in preparation of an associated OAQPS Lead Staff Paper (U.S.
11    Environmental Protection Agency, 1990b). However, after consideration of evaluations
12    contained in these documents, EPA chose not to propose revision of the Lead NAAQS.
13          Changes in relative contributions of various lead sources and exposure pathways to
14    human exposures in the United States, and EPA actions to reduce such exposures,  provide
15    important background for this current lead criteria and NAAQS review.  Since 1978, the amount
16    of lead emitted into the air nationally has markedly declined. For example,  as illustrated in
17    Chapters 2 and 3 of this document, from  1982  to 2002 lead emissions into the air decreased by
18    93% and the average air quality concentration  of lead decreased by 94% from 1983 to 2002
19    (http://www.epa.gov/airtrends/lead2.html). Total lead emissions into the air decreased from
20    about 220,000 tons in 1970 to less than 4,000 in 1999. This decline is mainly attributable to
21    EPA's regulatory efforts to reduce the content  of lead in gasoline (see, for example,
22    50 FR 9386), which substantially altered basic patterns of air lead emissions in the United States
23    (http://www.epa.gov/airtrends/lead2.html). Emissions from stationary sources have also been
24    greatly reduced (http://www.epa.gov/airtrends/lead2.html, Figure 2-11);  but, given the even
25    greater reductions in emissions from transportation sources, industrial processes (including
26    smelters and battery manufacturers) now constitute a larger percentage of remaining lead
27    emissions to the atmosphere (http://www.epa.gov/airtrends/lead2.html, Figure 2-12). In short,
28    lead emissions into the atmosphere decreased greatly in the 1980's and 1990's, a trend that has
29    continued on through to the present.  As a consequence, airborne lead now represents only a
30    relatively small component of total exposure to lead in the United States, such that the principal
31    sources and pathways for U.S. lead exposure among the classically-defined most sensitive

      December 2005                             1 -5        DRAFT-DO NOT QUOTE OR CITE

-------
 1    population group (young children) involve non-inhalation pathways, e.g., ingestion of lead from
 2    deteriorating paint, food, drinking water, dust, and historically contaminated soil.  While these
 3    downward trends in air lead exposures nationwide are encouraging, several important sources of
 4    air lead exposure may still persist in some localities.  Lead emissions from specific stationary
 5    sources and/or reentrainment of lead-contaminated soils (including from past deposition of
 6    airborne lead) may still have significant impacts on a local level. Recognition of the multimedia
 7    nature of lead exposure of the general population has been important historically and sorting out
 8    relative contributions to total lead exposure burdens represents an important input to the current
 9    periodic Lead NAAQS review effort.
10          Since the 1980's, EPA has played a major, effective role in working to reduce the main
11    sources of lead exposure for most children, including deteriorating lead-based paint, lead-
12    contaminated dust, and lead-contaminated residential soil (http://www.epa.gov/lead/).
13    For example, EPA has established standards for lead-based paint hazards and lead dust  cleanup
14    levels in most pre-1978 housing and child-occupied facilities, and is now developing standards
15    for those conducting renovation activities that create lead-based paint hazards and for the
16    management and disposal of lead-based debris (http://www.epa.gov/lead/regulation.htm). Also,
17    EPA has developed standards for management of lead in solid and hazardous waste, continues to
18    oversee the cleanup of lead contamination at Superfund facilities, and has issued regulations to
19    reduce lead in drinking water (http://www.epa.gov/lead/sources.htm). Beyond taking specific
20    regulatory actions, the Agency's Lead Awareness Program also continues to work to
21    protect human health and the environment against the dangers of lead by conducting research
22    and designing educational outreach efforts  and materials (http://www.epa.gov/lead/).
23          Since the 1980's, EPA has also promulgated regulations under section 112 of the Clean
24    Air Act (42 U.S.C. § 7412), to address emissions of lead components and other toxic pollutants
25    from both primary lead smelters and secondary lead smelters (40 CFR Subparts X and TTT).
26    Under section 112(d), these emission standards are to require  "the maximum  degree of  reduction
27    in emissions" that are "achievable."  Thus,  EPA promulgated  section 112(d) standards for
28    secondary lead smelters on June 23,  1995 (60 Fed. Reg. 3587) and revised them on June 13,
29    1997 (62 Fed. Reg. 32209), followed by promulgation of section 112(d) standards for primary
30    lead smelters on June 4, 1999 (64 Fed. Reg. 30194).
      December 2005                            1 -6        DRAFT-DO NOT QUOTE OR CITE

-------
 1    1.2    CURRENT LEAD CRITERIA AND NAAQS REVIEW
 2    1.2.1   Procedures and Key Milestones for Document Preparation
 3         It is important to emphasize at the outset that development of the present document has
 4    and will continue to include substantial external (non-EPA) expert inputs and opportunities for
 5    public input through (a) public workshops involving the general scientific community,
 6    (b) iterative reviews of successive drafts of this document by CASAC, and (c) comments from
 7    the public on successive drafts.  Extensive external inputs received through such reviews will
 8    help to ensure that the review of the Lead NAAQS will be based on critical assessment in this
 9    document of the latest available pertinent science.
10         The procedures for developing this revised Lead AQCD build on experience derived
11    from the other recent criteria document preparation efforts, and include close coordination
12    between NCEA-RTP and OAQPS staff, as well as with others, throughout the document
13    preparation/review process. Briefly, the respective responsibilities for production of the
14    document and meeting key milestones are as follows.  An NCEA-RTP Lead Team has been
15    designated as being responsible for the creation and implementation of a proj ect plan for
16    developing the Lead AQCD, taking into account input from individuals in other ORE) units,
17    OAQPS, and other EPA program/policy offices identified as part of the EPA Lead Work Group.
18    The Lead Team defines critical issues and topics to be addressed by the authors and provides
19    direction in order to focus on evaluation of those  studies most clearly identified as likely being
20    important for U.S. air standard setting purposes.  Criteria document materials are authored in part
21    by NCEA-RTP Lead Team staff with appropriate expertise in particular areas and by non-EPA
22    consultants to EPA who are recognized experts in pertinent specific areas (e.g., lead biokinetic
23    modeling, toxicology, epidemiology, etc.).
24         Key milestones for development of this  Lead AQCD are listed in Table  1-1. As a first
25    step, EPA announced on November 9, 2004 official initiation of the current periodic review of
26    air quality criteria for lead. More specifically,  under processes established in Sections 108 and
27    109 of the Clean Air Act, U.S. EPA began by announcing in the Federal Register (69 FR 64,926)
28    the formal commencement of the current review process with a call for information (see Federal
29    Register, 2004). In addition, EPA prepared a January 2005  draft Lead AQCD Work Plan, which
30    was made available for public comment and was the subject of teleconsultation with CASAC on
31    March 28, 2005 as  a means by which to communicate the process and timeline for development

      December 2005                            1 -7       DRAFT-DO NOT QUOTE OR CITE

-------
      Table 1-1. Key Milestones and Projected Schedule for Development of Revised Lead Air
                             Quality Criteria Document (Lead AQCD)1
      Major Milestones                                                    Target Dates
       1.  Literature Search                                                 Ongoing
       2.  Federal Register Call for Information                                  November 9. 2004
       3.  Prepare Draft Lead AQCD Project Work Plan                           Nov-Dec 2004
       4.  Release Draft Project Plan for Public Comment/CASAC Review            January 2005
       5.  Public Comment Period                                            Jan/Feb 2005
       6.  CASAC Teleconsultation on Project Work Plan                         March 28, 2005
       7.  Workshop Drafts of Lead AQCD Chapters                             May/June 2005
       8.  Peer Consultative-Review Workshop(s)                                July/August 2005
       9.  Release First External Review Draft                                  December 1, 2005
      10.  Public Comment Period                                            Dec 2005-Feb 2006
      11.  CASAC/SAB Public Review Meeting (First Ext. Rev. Draft)               Feb. 28-Mar 1, 2006
      12.  Release Second External Review Draft                                June 2006
      13.  Public Comment Period                                            June-July 2006
      14.  CASAC/SAB Public Review Meeting (Second Ext. Rev. Draft)             August 2006
      15.  Final Lead AQCD                                                 October 1,2006

      1 Schedule may be modified from time to time, as necessary, to reflect actual project requirements and progress,
       but EPA is under court order to produce a final Lead AQCD by October 1, 2006. Missouri Coalition for the
       Environment v. EPA. Civil Action No. 4:04-CV-00660 (ERW) (E.D. Mo. Sept. 14, 2005).  Also, note that
       materials contributed by non-EPA authors, at times, have been and will continue to be modified by EPA staff
       in response to internal and/or external review comments and that EPA is responsible for the ultimate content
       of this Lead AQCD.
1    of a revised Lead AQCD. Next, expert consultants to NCEA-RTP and NCEA-RTP staff
2    (a) carefully evaluated pertinent new studies obtained via the call for information and via
3    ongoing literature searches conducted by NCEA-RTP information retrieval specialists and
4    (b) prepared preliminary draft chapter materials for inclusion in this revised Lead AQCD. Those
5    preliminary draft materials then underwent expert peer discussion at public workshops organized
6    and conducted by NCEA-RTP in July/August, 2005. After consideration of comments received
7    at the workshops, appropriate revisions were made in the draft materials and incorporated into
8    this First External Review Draft of the Lead AQCD, which is now being made available for
9    public comment (the comment period ends February 15, 2006) and CASAC review at a public

     December 2005                              1 -8        DRAFT-DO NOT QUOTE OR CITE

-------
 1   meeting scheduled for February 28-March 1, 2006 (to be announced in the Federal Register).
 2   EPA expects that, after consideration of CAS AC and public comments, it will prepare a Second
 3   External Review Draft of this revised Lead AQCD for further review by the public and CAS AC
 4   before completing the final version of it for issuance by October 1, 2006. Publication of the final
 5   document and its availability to the public will be announced in the Federal Register.
 6         Drawing upon evaluations in the Lead AQCD and other lead exposure/risk analyses, the
 7   EPA's Office of Air Quality Planning and Standards (OAQPS) staff will prepare a draft Lead
 8   Staff Paper that will assess policy implications of key information in the Lead AQCD, report
 9   pertinent exposure and risk analyses, and ultimately pose possible options for the EPA
10   Administrator to consider with regard to whether to  retain or, if appropriate, revise the Lead
11   NAAQS. The draft Lead Staff Paper and analyses will also be made available for review by the
12   public and CAS AC. Taking into account CAS AC and public comments, EPA expects to
13   produce revised exposure and risk analyses as well as a revised draft Lead Staff Paper for public
14   comment and CASAC review before final revisions are made in the Lead Staff Paper, to provide
15   information to inform the decisions to be made by the EPA Administrator regarding possible
16   retention or revision of the Lead NAAQS. The proposed NAAQS decisions will then be made
17   available via the Federal Register for public comment and, following consideration of comments
18   received, the EPA Administrator will promulgate  final Lead NAAQS decisions via their
19   announcement in the Federal Register.
20
21
22   1.3   ORGANIZATIONAL STRUCTURE AND CONTENT OF
23         THE DOCUMENT
24   1.3.1   Ascertainment of Literature and General Document Format
25         Lists of references published since completion of the 1986 Lead AQCD/Addendum and
26   1990 Supplement were made available to the authors.  The references were mainly selected from
27   information data base (e.g., Pub Med) searches conducted by EPA. However, additional
28   references have also been added (e.g., for missed or  recently published papers or "in press"
29   publications) as work has proceeded in creating the present draft document materials.  As an aid
30   in selecting pertinent new literature, the authors were also provided with  a summary of issues
31   that need to be addressed in this revised Lead AQCD. Many such issues have been identified in

     December 2005                           1 -9       DRAFT-DO NOT QUOTE OR CITE

-------
 1    the course of previous lead criteria assessments, through interactions between EPA Lead Team
 2    and Lead Work Group members, and via workshop discussions.
 3          The general format used in this draft document is to open each new chapter (or main
 4    section) for the updated Lead AQCD with concise summary of key findings/conclusions from
 5    the previous lead criteria assessments, especially the 1986 Lead AQCD/Addendum (U.S.
 6    Environmental Protection Agency, 1986a,b) and 1990 Supplement (U.S. Environmental
 7    Protection Agency, 1990). After presentation of such background information, the remainder of
 8    each chapter or section typically provides an updated discussion of newer literature and resulting
 9    key conclusions. In some cases where no new information is available, the summary of key
10    findings and conclusions from the previous lead criteria assessment(s) must suffice as the basis
11    for current key conclusions.  Increased emphasis is placed in the main chapters of this revised
12    Lead AQCD on interpretative evaluation and integration of evidence pertaining to a given topic
13    than was typical of many previous EPA air quality criteria documents, with more detailed
14    descriptions of individual studies or other supportive information being provided in a series of
15    accompanying annexes.
16
17    1.3.2    Organization and Content of the Document
18          This updated Lead AQCD critically assesses scientific information on the health and
19    welfare effects associated with exposure to the concentrations of lead in ambient air.  The
20    document is not intended to be a detailed, exhaustive review of the literature. Rather, the cited
21    references reflect the current state of knowledge on the most relevant issues pertinent to
22    decisions regarding possible revision by EPA of the Lead NAAQS. Although emphasis is placed
23    mainly on the  discussion of health and welfare effects data, other scientific data is also presented
24    and evaluated, in order to provide a better understanding of the nature, sources, distribution, and
25    concentrations of lead in ambient air, as well as the measurement of human exposure to lead.
26          The focus of the selected scientific information in the text is on information published
27    since the previous assessments of air quality criteria for lead contained in the 1986 Lead
28    AQCD/Addendum or 1990  Supplement. Emphasis is placed on studies conducted at or near lead
29    concentrations found in ambient air.  Other studies are included if they contain unique data (e.g.,
30    the documentation of a previously unreported effect or of a mechanism  for an observed effect),
      December 2005                            1-10       DRAFT-DO NOT QUOTE OR CITE

-------
 1    or if they are multiple-concentration studies designed to characterize exposure- or dose-response
 2    relationships.
 3          As noted earlier, key findings and conclusions from the 1986 Lead AQCD/Addendum and
 4    1990 Supplement are typically first briefly summarized at the outset of discussion of a given
 5    topic, with appropriate reference back to the previous criteria assessment materials.  Typically,
 6    important prior studies are more specifically discussed only if they are open to reinterpretation in
 7    light of newer data and/or are judged to be potentially useful in decisions on revision of the
 8    standards for lead. Generally, only information that has undergone scientific peer review and has
 9    been published (or accepted for publication) in the open literature through August, 2005 has been
10    thus far considered in this draft criteria document.  It is expected that, ultimately, the final Lead
11    AQCD will consider new peer-reviewed studies published through December 31, 2005. Certain
12    other unpublished analyses (e.g., de Novo analyses of recently available U.S. lead air quality
13    data) may be considered, depending on the importance of the subject information and its
14    pertinence to criteria development for Lead NAAQS, as determined in consultation with
15    CAS AC.
16          The final AQCD will consist of two volumes. Volume 1 will consist of eight chapters
17    that comprise the main body of the revised Lead AQCD. In the first volume of this draft
18    document, this introductory chapter (Chapter 1): (a) provides brief statements regarding the
19    purpose of the document; (b) presents information on the legislative background and regulatory
20    chronology of lead criteria reviews; and (c) presents an overview of the organization of the
21    document.  Chapter 2 provides information on the physics and chemistry of lead, as well as
22    sources, emissions, transport and deposition/fate.  Chapter 3  discusses environmental
23    concentrations, dispersal patterns, and multimedia exposure pathways. Chapter 4 focuses on the
24    modeling of multimedia exposure impacts on human internal lead burdens, especially as indexed
25    by blood or bone lead concentrations. Then, Chapter 5 discusses toxicologic studies of lead
26    health effects in humans, laboratory animals, and in vitro test systems; whereas Chapter 6
27    assesses lead-related epidemiologic (observational) studies of human population groups.
28    Chapter 7, which will ultimately provide an integrative synthesis of key information drawn from
29    the earlier chapters to delineate human lead exposure and health effect findings and conclusions
30    of most importance for derivation of primary Pb NAAQS, will be prepared after CASAC  review
31    of draft Chapters 1 through 6 and will be included in the Second External Review Draft of this

      December 2005                            1-11       DRAFT-DO NOT QUOTE OR CITE

-------
1   Lead AQCD to be circulated for later public comment and CASAC review. Lastly, Chapter 8
2   deals with ecological and other environmental effects of lead as key types of welfare effects
3   pertinent to the derivation of secondary Pb NAAQS.  Several annexes containing more detailed
4   descriptive materials supporting the interpretative evaluations highlighted in the main chapters
5   dealing with health and vegetation/ecological effects are provided in Volume II of this revised
6   Lead AQCD.
7          An Executive Summary will also be developed after CASAC review of the present First
8   External Review Draft Lead AQCD materials and included in Volume I of the Second External
9   Review Draft to be released for later public comment and CASAC review.
    December 2005                           1-12       DRAFT-DO NOT QUOTE OR CITE

-------
  1     1.4    REFERENCES
 3    Federal Register. (1971) National primary and secondary ambient air quality standards. F. R. (April 30)
 4           36:8186-8201.
 5    Federal Register. (1979) National primary and secondary ambient air quality standards: revisions to the National
 6          Ambient Air Quality Standards for lead. F. R. (February 8) 44: 8202-8237.
 7    Federal Register. (2004) Air quality Criteria Document for Lead: Call for Information. F. R. (November 9)
 8          69:64926-64928.
 9    U.S. Code. (2003a) Clean Air Act, §108, air quality criteria and control techniques.. U.  S. C. 42: §7408.
10    U.S. Code. (2003b) Clean Air Act, § 109, national ambient air quality standards. U. S. C. 42: §7409.
11    U.S. Environmental Protection Agency. (1977) Air quality criteria for lead. Research Triangle Park, NC: Health
12          Effects Research Laboratory, Criteria and Special Studies Office; EPA report no. EPA/600/8-77-017.
13          Available from NTIS, Springfield, VA; PB-280411.
14    U.S. Environmental Protection Agency. (1986a) Air quality criteria for lead. Research Triangle Park, NC: Office of
15          Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
16          EPA/600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.
17    U.S. Environmental Protection Agency. (1986b) Lead effects on cardiovascular function, early development, and
18          stature: an addendum to U.S. EPA Air quality criteria for lead. In: Air quality criteria for lead, v. 1. Research
19          Triangle Park, NC: Office of Health and Environmental Assessment,  Environmental Criteria and Assessment
20          Office; pp. A1-A67; EPA report no. EPA/600/8-83/028aF. Available from: NTIS, Springfield, VA;
21          PB87-142378.
22    U.S. Environmental Protection Agency. (1990a) Summary of selected new information on effects of lead on health
23          and supplement to 1986 air quality criteria for lead.  Research Triangle Park, NC: Office of Health and
24          Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no. EPA/600/8-89.
25          Available from NTIS, Springfield, VA; PB92-235670.
26    U.S. Environmental Protection Agency. (1990b) Review of the national ambient air quality standards for lead:
27          assessment of scientific and technical information: OAQPS staff paper. Research Triangle Park, NC:
28          Office of Air Quality Planning and Standards; report no. EPA-450/2-89/022. Available from: NTIS,
29          Springfield, VA; PB91-206185.
      December 2005                                1-13         DRAFT-DO NOT QUOTE OR CITE

-------
 i          2.    CHEMISTRY, SOURCES, AND TRANSPORT
 2                                       OF LEAD

 3
 4          The purpose of this chapter is to provide background information on the chemical
 5   properties of Pb that are relevant to its transport within the environment, into ecosystems and for
 6   human health considerations; to discuss the known sources of Pb in the environment; and to
 7   outline the mechanisms by which Pb is transported within the atmosphere.  The chapter does not
 8   provide a comprehensive list of all sources of lead, nor does it provide emission rates or emission
 9   factors for all important source categories, since such information is available for only a limited
10   number of sources. Rather, the chapter provides data on the chemistry, sources, and transport of
11   lead where information is available in the peer-reviewed literature. Particle size distribution data
12   for lead are even scarcer than total lead emissions from sources; particle size data are presented
13   where such data are available.
14
15
16   2.1   PHYSICAL AND CHEMICAL PROPERTIES OF LEAD
17   Properties of Elemental Lead
18          Elemental Pb possesses an array of useful physical and chemical properties, making it
19   among the first metals to be extracted and used by humankind. It has a relatively low melting
20   point (327.51), is a soft, malleable, and ductile metal, a poor electrical conductor, and is easily
21   cast, rolled and extruded. While sensitive to environmental acids, after exposure to
22   environmental sulfuric acid (H2SO4),  metallic Pb becomes impervious to corrosion due to
23   weathering and submersion in water.  This effect is due to lead sulfate (PbSO4), the relatively
24   insoluble precipitate produced by reaction of Pb with H^SO/t,  forms a protective barrier against
25   further chemical reactions (Schweitzer, 2003). This aspect of its chemistry made Pb especially
26   convenient for roofing, containment of corrosive liquids, and  until the discovery of its adverse
27   health effects, construction of water supply systems.
28          Lead is readily extracted from galena, a widely available sulfide mineral form of lead
29   (PbS), by froth flotation, followed by roasting in the presence of a limited about of oxygen to
30   form litharge, one of two forms of lead oxide (PbO). Elemental Pb is then isolated by reducing

     December 2005                            2-1       DRAFT-DO NOT QUOTE OR CITE

-------
 1    PbO by way of heating in the presence of elemental carbon (coke, charcoal) (Greenwood and
 2    Earnshaw, 1987).  This and other extraction and recovery processes will be discussed in greater
 3    detail, later in this chapter.
 4          Lead alloys constitute 60% of lead used in industry (Prengaman, 2002).  The major
 5    alloying elements are antimony, calcium, tin, copper, tellurium, arsenic, and silver.  Selenium,
 6    sulfur, bismuth, cadmium, indium, aluminum, and strontium are also sometimes used. Lead
 7    alloys are found primarily in lead acid batteries, solder, ammunition, and cable  sheathing
 8    (Prengaman, 2002).  Table 2-1 provides a list of Pb alloys in use by industry.
 9          Some of the physical  properties of elemental Pb are listed in Table 2-2.  The most
10    important of these properties, when evaluating the transport routes for Pb within the atmosphere,
11    is its boiling point.  As indicated, Pb will only exist in the vapor phase at or above 1750  °C.
12    Therefore, at ambient atmospheric temperatures, elemental Pb will deposit to surfaces or exist in
13    the atmosphere as a component of atmospheric aerosol.
14
15    Oxidation States of Lead
16          Lead is the heaviest congener of carbon, and shares many properties with the other
17    elements found in the same column of the periodic chart (silicon, germanium, and tin).
18    As Group IV elements, these elements have four valence electrons (2p and 2 s), allowing for
19    both divalent and tetravalent compounds.
20          Due to its high atomic number (82), the valence electron orbitals of the Pb  atom exist at a
21    comparatively large distance from its nucleus. As with s and/? orbitals at any quantum level,
22    electrons in the 6s orbital tend to occupy space near the nucleus with greater probability  than
23    those in the 6p orbital. The strong attraction produced by the large Pb nucleus combined with
24    the long distance that the 6s  electrons must travel result in electron accelerations to relativistic
25    speeds.  The Theory of Relativity states that as the velocity of matter approaches the speed of
26    light, its apparent mass increases. In this instance, the electrons in the Pb 6s orbital experience
27    an increase in weight, which increases the attractive effect of the positive nuclear charge, which
28    contracts the diameter of the Pb 6s orbital (Pitzer,  1979).  This "relativistic effect" on valence
29    electrons is proportional to the square of atomic number, and manifests within the Group IV
30    elements as a distinctly increasing trend in the stability of the divalent state from Si down to Pb.
31    In the case of Pb, the two 6s electrons behave as if they were chemically inert, leaving only the

      December 2005                            2-2        DRAFT-DO NOT QUOTE OR CITE

-------
                    Table 2-1. Lead Alloys and Their Industrial Applications
 Lead Alloy
                                  Uses
 Lead-Antimony

 Lead-Calcium

 Lead-Tin


 Lead-Copper

 Lead-Silver

 Lead-Tellurium
 Lead-Bismuth

 Lead-Cadmium

 Lead-Indium

 Lead-Strontium
 Lead-Lithium
 Lead-Antimony-Tin
 Lead-Calcium-Aluminum
 Lead-Calcium-Tin
 Lead-Calcium-Silver
 Lead-Antimony-Silver

 Lead-Silver-Tin

 Lead-Strontium-Tin
 Lead-Lithium-Tin
Grids, posts, and connectors for lead-acid batteries, ammunition, cable
sheathing, anodes, tank linings, pumps, valves, and heating and cooling coils
Automotive, standby power, submarines, and specialty sealed batteries,
electrowinning anodes, cable sheathing, sleeving, specialty boat keels, and lead
alloy tapes
Soldering for electronics, general purposes, automobile radiators, and heat
exchangers, corrosion resistant coatings on steel and copper, cable sheathing,
fuses, sprinkler system alloys, foundry pattern alloys, molds, dies, punches,
cores, mandrels, replication of human body parts, and filters for tube bonding
Lead sheet, pipe, cable sheathing, wire, fabricated products, tank linings, tubes
for acid-mist precipitators, steam heating pipes for sulfuric acid or chromate
plating baths, and lead sheathing for roofs
Anodes, high-temperature solders, insoluble anodes in the electrowinning of
zinc and manganese, and soft solders
Pipes, sheets, shielding for nuclear reactors, and cable sheathing
Fuses, sprinkler system alloys, foundry pattern alloys, molds, dies, punches,
cores, mandrels, solders, replication of human body parts, and filters for tube
bonding
Fuses, sprinkler system alloys, foundry pattern alloys, molds, dies, punches,
cores, mandrels, solders, replication of human body parts, and filters for
tube bonding
Fuses, sprinkler system alloys, foundry pattern alloys, molds, dies, punches,
cores, mandrels, solders, replication of human body parts, filters for tube
bonding, and joining metals to glass
Battery grids
Bearings, lead-acid battery grids
Printing, bearings, solders, slush castings, and specialty castings
Negative battery grids of lead-acid batteries
Positive grids of lead-calcium batteries, and lead anodes for electrowinning
Zinc electrowinning
Anodes used for the production of thin copper foil in electronics, and anodes in
cathodic protection of steel pipes and structures in water
Anodes in cathodic protection of steel pipes and structures in water, and soft
solders
Anodes for copper electrowinning
Lead-acid battery grids
 Source: Prengaman (2002).
December 2005
                   2-3
DRAFT-DO NOT QUOTE OR CITE

-------
                          Table 2-2. Physical Properties of Elemental Lead
        Physical Property
       Atomic number                                                  82
       Atomic weight                                                  207.2
       Valence electrons                                          [Xe]4f145d106s26p
       Melting point                                                  328 °C
       Boiling point                                                  1750 °C
       Density                                                     11.34 g/cm3
       Atomic radius                                                  146 pm
       Standard reduction potential                                      - 0.126V
       Oxidation numbers                                              +2, +4
       lonization Energy                                             715.6 kJ/mol

       Source: Kotz and Purcell (1991).
 1   two 6p electrons available for bonding or oxidation under ordinary conditions. For this reason,
 2   the relativistic effect is also known as the "inert pair effect."  Consequently, Pb(II) is the most
 3   common oxidation state in which Pb is found in the environment (King, 1995; Claudio et al.,
 4   2003).
 5          Lead is distinguished from other elements that are subject to relativistic effects by its
 6   preference for forming tetravalent (Pb(IV) organometallic compounds, however. In fact, it is
 7   only with rare exception that Pb(II) organometallic compounds form (Pelletier, 1995;
 8   Greenwood and Earnshaw, 1984). All simple alkyllead compounds, such as the well-known fuel
 9   additives, tetramethyllead (TML) and tetraethyllead (TEL) are composed of Pb(IV). In contrast,
10   inorganic Pb(IV) compounds, such as PbC>2 are strong oxidants, and unstable with respect to
11   their Pb(II) analogs.  There are, overall, more than 200 known organolead compounds
12   (Harrison, 1985).
13          In relation to the other Group IV metals, however, Pb forms the least stable and most
14   reactive organometallic derivatives. This is largely due to the weak bond between lead and
15   carbon, consistent with its large atomic size, and the influence of the relativistic effect on its
16   valence orbitals. Specifically, the mean bond dissociation energies of the metal-carbon bonds for
17   Group IV elements are 56.7 kcal/mol for germanium, 46.2 kcal/mol for tin, and 30.8 kcal/mol for

     December 2005                            2-4       DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead (Shapiro & Frey, 1968). Organolead compounds are thermally unstable and will decompose
 2    to metallic lead and free radicals at relatively low temperatures (Willemsen and van der Kerk,
 3    1965). For example, TML decomposes at temperatures above 200°C, and TEL decomposes at
 4    temperatures above 110°C (King, 1995).  In solution, organolead compounds decompose in the
 5    presence of UV radiation (1 hr/254 nm) and sunlight. (Gomez Ariza et al., 2000)
 6          Tetralkyllead compounds have atmospheric residence times ranging from a few hours to a
 7    few days (Pelletier, 1995). TML and TEL react with OH in the gas-phase, following pseudo-first
 8    order kinetics, to form a variety of products that include ionic trialkyllead (TriAL), dialkyllead
 9    (DiAL) and metallic Pb. Trialkyllead is slow to react with OH and is quite persistent in the
10    atmosphere (Hewitt and Harrison, 1986; Harrison and Laxen, 1980).
11
12    Lead Oxides, Chalcogenides, and Salts
13          A rich variety of inorganic Pb compounds and complex salts can be prepared in the
14    laboratory under conditions of temperature and pressure not usually seen in the environment.
15    Information on the many possible organic and inorganic Pb compounds can be found in the text
16    by Greenwood and Earnshaw (1984).  Several representative Pb salts and oxides are described in
17    Tables 2-3 and 2-4. Inorganic Pb compounds that can be found in the environment are the  focus
18    of this discussion.
19          As explained earlier, Pb exists preferentially in its +2 oxidation  state in the environment.
20    Under aqueous acidic conditions, Pb readily oxidizes, with a strongly positive electrochemical
21    potential (E° =1.355 V), and a large equilibrium constant (K = 10 916), to form Pb(II) (Singley,
22    1994):
23
24                            2Pb + O2 + 4H+  -»  2Pb2+ + 2H2O                           (2-1)
25
26          Table 2-5 lists the various Pb compounds and salts introduced into the environment by
27    natural processes and anthropogenic activities that will be discussed in Sections 2.2 and 2.3 of
28    this chapter. From this list, it is clear that only a relatively limited number of salts and
29    covalently-bound Pb compounds are of significance in the environment, i.e., sulfates (PbSO4),
30    chlorides (PbCl2), carbonates (PbCO3, Pb(HCO3) 2), hydroxides (Pb(OH) 2), nitrates (Pb(NO3) 2),
31    phosphates (PbPO4, Pb(HPO4) 2), silicates, oxides (PbO, Pb3O4), and PbS. With the exception of

      December 2005                           2-5        DRAFT-DO NOT QUOTE OR  CITE

-------
         Table 2-3. Lead Salts: Names, Formulae, Physical Characteristics, and Uses
 Category
Compound
  Name
Formula
Form
Uses
 Lead Acetates   Anhydrous Lead    Pb(C2H3O2)2
                 Acetate
                 Basic Lead         2Pb(OH)2
                 Acetate            Pb(C2H3O2)2
                 Lead Acetate       Pb(C2H3O2)2
                 Trihydrate

                 Lead Tetraacetate   Pb(C2H3O2)4
 Lead           Lead Carbonate     PbCO3
 Carbonates
                 Basic Lead
                 Carbonate
                2PbC03
 Lead Halides    Lead Fluoride      PbF2
                 Lead Chloride      PbCl,
                 Lead Bromide
                 Lead Iodide
                PbBr,
                PbI2
             White, crystalline   Preparing other lead salts
             solid
             Heavy, white       Sugar analysis
             powder
             White,             Making other lead compounds, mordant for
             monoclinic         cotton dyes, water repellant, processing agent
             crystalline solid     for cosmetics, perfumes, and toiletries
             Colorless,          Oxidizing agent in organic synthesis, cleaving
             monoclinic         of a-hydroxy acids, introducing acetyl groups
             crystalline solid     in organic molecules
             Colorless,          Catalytic polymerization of formaldehyde,
             orthorhombic       improving the bonding of polychloroprene to
             crystals            metals in wire-reinforced hoses, a component
                                of high-pressure lubricating greases, and a
                                lubricant for poly vinyl chloride
             White, hexagonal   Ceramic glazes, a curing agent with peroxides
             crystals            to form polyethylene wire insulation, a color-
                                changing component of temperature-sensitive
                                inks, a component of lubricating grease, and a
                                component of weighted nylon-reinforced fish
                                nets made of poly vinyl chloride libers
             Colorless,          Glass sealing  disks for IR sensors, wear-
             orthorhombic       resistant automotive shock absorbers,
             crystals            electrolytic deposition of lead, flux  for
                                brazing of aluminum and its alloys, optical
                                glass libers for IR transmission, and thin film
                                batteries
             White,             Artist's pigment, precursor of organolead
             orthorhombic       compounds, seawater-activated batteries,
             needles            expanding polymer mortar, flux for soldering
                                cast iron and cast brass, sound-insulating
                                rubber sealants, corrosion inhibitor  for
                                galvanized  steel, and infrared-transmitting
                                glasses for CO2 lasers
             White,             Filler for flame-resistant polypropylene, glass
             orthorhombic       optical waveguides for infrared thermometers
             crystals            and catalysts for producing polyesters
             Powdery, yellow,   Aerosols for cloud seeding, making high-
             hexagonal          contrast photographic images of laser
             crystals            radiation, high capacity cathodes in lithium
                                batteries, and low-temperature thermographic
                                copying materials
Lead Silicates


Lead Sulfates
Lead
Monosilicate
Lead Bisilicate
Tribasic Lead
Silicate
Tribasic Lead
Sulfate
3PbO-
2Si02
PbO
0.03A12O3-
1.95SiO2
3PbO-Si02
3PbO
PbSO4
H20
White, trigonal
crystalline
powder
Pale yellow
powder
Reddish-yellow
powder
Fine, white
powder
Formulating lead-bearing glazes for ceramics,
source of PbO in glass manufacturing
Ceramic glazes
Glass and frit production
Providing long-term heat stability to PVC,
electrical insulation, activation for
azodicarbonamide blowing agents for vinyl
foam
 Source: Carr(2002).
December 2005
                                  2-6
                              DRAFT-DO NOT QUOTE OR CITE

-------
            Table 2-4.  Lead Oxides:  Names, Formulae, Physical Characteristics, and Uses
      Name
                   Formula
        Form
                   Uses
      Lead Monoxide
                     PbO
      Lead Dioxide
                    PbO2
      Lead
      Sesquioxide

      Red Lead
Reddish below 489°C,
yellow at high
temperatures
Brownish-black
crystalline powder of
fine flakes
                    Pb2O3     Amorphous, orange-
                              yellow powder

                    Pb3O4     Brilliant orange-red
                              pigment
Pastes for the grids of lead-acid batteries, optical,
electrical, and electronic glasses, glazes for fine
tableware, vulcanizing agent for rubber, lead soaps
used in driers as varnishes, high-temperature
lubricants, neutralizing agent in organic synthesis,
heat stabilizer in plastics, and starting material in
the production of pigments
Active material of the positive plates in lead-acid
batteries, oxidizing agent in the manufacture of
chemicals, dyes, matches, pyrotechnics, and liquid
polysulfide polymers, antifriction agent for plastic
sliding bearings, ballistic modifiers in high-energy
propellants, electrodes for seawater electrolysis,
filters for desulfurization of waste gases,
vulcanizing agents for butyl-rubber puncture-
sealing layers inside tires
Ballistic modifier for high-energy propellants,
cathode material in lithium batteries, additive to
increase the shattering force of explosives
Pigment in anticorrosion paints for steel surfaces,
lead oxide pastes for tubular lead-acid batteries,
ballistic modifiers for high-energy propellants,
ceramic glazes for porcelain, lubricants for hot
pressing metals, radiation-shielding foam coatings
in clinical x-ray exposures, and rubber adhesives
for roadway joints
      Source:  Carr(2002).
4

5
the covalently-bound sulfide and oxide, these compounds are derived from acids (or the related

anions) that are common in the environment, such as sulfuric acid (H2SO4), nitric acid (HNO3),

carbonic acid (H^COs, an acid that forms when CC>2 dissolves in water), and phosphoric acid

(HaPC^). Lead salts, once formed, tend to be only slightly soluble in neutral solutions, but are

quite soluble in the presence of acid (CRC Handbook, 1988).
6    Lead Coordination Chemistry, and Its Role in Biochemistry

1           The formation of coordinate covalent complexes represents a different class of chemical

8    interaction from the formation of simple covalent compounds and salts.  "Coordinate covalent"

9    bonds form when anions or neutral molecules interact with metal ions in solution that are
     December 2005
                                               2-7
                             DRAFT-DO NOT QUOTE OR CITE

-------
                     Table 2-5. Lead Compounds Observed in the Environment
      Location
              Observed Pb Compounds
      Minerals
      Smelting Aerosols




      Coal Combustion Aerosols


      Coal Combustion Flue Gases



      Wood Combustion

      Waste Incineration Aerosols


      Soils Near Mining Operations
      Motor vehicle exhaust (combustion of leaded fuel)
      Roadside dust

      Other mobile sources:
         Brake wear, wheel weights
         NASCAR vehicle emissions
         Aircraft engine wear
         Lawn mowers
   PbS (Galena)
   PbO (Litharge, Massicot)
   Pb3O4 (Minium or "Red Lead")
   PbCO3 (Cemssite)
   PbSO4 (Anglesite)

   Pb°, PbS
   PbSO4,PbO,PbSO4.PbO
   PbCO3
   Pb silicates

   PbS
   PbSe

   Pb°, PbO, PbO2 (Above 1150K)
   PbCl2 (Low rank coals, above 1150K)
   Pb$O4 (Below 1150 K)

   PbCO3
   PbCl2
   PbO

   PbCO3
   PbSO4
   [PbFe6(S04)4(OH)12]
   [Pb5(P04)3Cl]
   [Pb4S04(C03)2(OH)3]
   PbS-Bi2S3
   Pb oxides, silicates

   PbBrCl
   PbBrCl-2NH4Cl
   PbBrCl-NH4Cl

   PbSO4, Pb°, PbSO4 (NH4)SO4, Pb3O4,
   PbO-PbSO4 and 2PbCO3-Pb(OH)2,PbSO4


   Pb°
   Pb halides
   Pb°
   Pb halides (Battery leakage)
      "Source: Biggins and Harrison (1979, 1980).
1    capable of donating both of the electrons required to form a bond. These molecules (or anions)

2    are called, "ligands," or "electron donors." Ligands possess a filled valence orbital with a

3    geometry that allows it to overlap to a substantial degree with an empty orbital associated with

4    the metal ion. In the case of Pb, its large atomic size is associated with several out-lying empty
     December 2005
2-8
DRAFT-DO NOT QUOTE OR CITE

-------
 1    atomic orbitals leading to a tendency to form a large number of coordinate covalent bonds
 2    (Claudio et al., 2003).  This is suggested by the coordination number (9) of PbCb, in its
 3    crystalline form, which is able to share electrons with 9 adjacent chloride ions (Cl")  (Douglas
 4    etal.,  1983).
 5          Molecules capable of serving as ligands for metal ions in solution take many forms.
 6    "Monodentate" ligands are molecules  capable of providing 2 electrons to form a single
 7    coordinate bond, such as water (H2O), ammonia (NH3); "multidentate" ligands can participate in
 8    more than one coordinate bond. A common term for the binding of a metal ion by a multidentate
 9    ligand is "chelation." The chelating agent, ethylenediaminetetraacetic acid (EDTA), is a well
10    known, hexadentate ligand, containing 6 functional groups capable of forming 6-coordinate
11    bonds with metal ions in aqueous solution. Proteins, particularly the active sites of enzymes,
12    contain functional groups (amino acid side-chains) that can serve as ligands for metal ions.
13    In fact, the zinc finger proteins must form coordinate complexes with Zn2+ ions to stabilize their
14    active conformation (Claudio et al., 2003).
15          Several types of equilibrium constants for ligand-metal interactions can be derived,
16    depending on the property of interest.  One formulation, the "binding constant (Kb)," between the
17    free metal ion and ligands in solution,  with the ligand-metal complex, is derived from the
18    following relationship (for a neutral ligand):
19
                                                      FA /IT  n+l i
20                             Kb = binding constant =	—-                           (2-2)
                                                     [Mn+][LJ
21
22                             Where: K     [MLn+1]
                                         bl   [Mn+][L]
23                                     Kb2 =
                                             [ML2n+]
                                             [MLn+][L]
24                                    Etc.
25
26
27          Binding constants are useful, in particular, for evaluating the strength of interactions
28    between metals and small (monodentate) ligands. The form typically used to evaluate binding
29    between metals and proteins is the "dissociation" constant, K
-------
                               Kd = dissociation constant =- - — — — -                   (2-3)
                                                           [MLxn+]
                              Where:
                                         dl,
                                         dl   [MLn+]
                                           _
                                         d2~
 5                                    Etc.
 6
 7          A variety of quantitative methods are available for establishing binding and dissociation
 8    constants for specific combinations of metals and ligands. Conversely, a simple, qualitative
 9    model for estimating the relative strength of coordinate covalent bonding between metals and
10    ligands is the Pearson's Hard-Soft Acid-Base (HSAB) model (Douglas et al., 1983). Heavier
1 1    metals, such as Pb, which have more electrons and more  spatially diffuse valence orbitals, are
12    described as "soft" (Lewis) acids. Lighter metals, with fewer electrons and more closely-spaced
13    valence orbitals, are  described as "hard" (Lewis) acids. These metals tend to preferentially bond
14    with ligands with similar electronic properties.  Hard acids tend, for example, to prefer oxygen-
15    based ligands, i.e.  "hard bases," and soft acids prefer ligands based on larger atoms, such as
16    sulfur and selenium,  i.e., "soft bases."
17          The HSAB concept is useful for understanding the behavior of Pb in the biological
18    context. Lead readily forms coordinate covalent bonds with sulfur and sulfur-containing
19    compounds, carboxylic acids and imidazoles (Claudio et  al., 2003).  In biological systems, Pb
20    competes very effectively with native or homeostatic metal ions for binding with the
21    sulphahydryl, carboxyl and imidazole side-chains comprising enzyme active sites. This
22    competition leads  to  inhibition of enzyme activity, as well as the replacement of calcium in bone
23    and, ultimately, to a  substantial list of negative human health effects. The relative strength of
24    these different interactions appears to be reasonably well-predicted by the HSAB model.
25          By far, the most effective biological ligands for Pb are amino acid side-chains containing
26    sulfur and selenium.  Smaller electron donors (hard bases), such as carboxylic acids that bind Pb
27    via electrons associated with oxygen, form weaker bonds. These complexes are generally more
28    labile, i.e., bonds form and break rapidly, thus allowing more effective competition at protein
29    binding sites amongst metals available in solution. Example simple ligands in this case are the
      December 2005                            2-10       DRAFT-DO NOT QUOTE OR CITE

-------
 1    amine functional group, -NH, and the thiol functional group, -SH.  The amine group has a Pb
 2    binding constant on the order of 100, while the thiol group binding constant is on the order of
 3    107. Example proteins in this instance are carboxypeptidase A, a zinc-binding protein, with
 4    carboxylate and histidine side-chains, and the four cysteine zinc finger consensus peptide, CP-
 5    CCC.  Carboxypeptidase A has a Pb dissociation constant of approximately 10"4 M, versus that
 6    of the  zinc finger protein, which is 3.9 X 10"14 M. Claudio et al. (2003) concluded, on the basis
 7    of these values, that carboxypeptidase A is unlikely to be a protein associated with Pb poisoning,
 8    while  cysteine-rich proteins, including the zinc enzyme, d-aminolevulinic acid dehydratase
 9    (ALAD), the second enzyme in the heme biosynthetic pathway, are more likely targets. ALAD
10    active site, with its Cyss active site, is known to be inhibited at femtomolar (10"15 M)
11    concentrations of Pb in vitro.
12          Additional information concerning the physical aspects of Pb coordination chemistry  and
13    its role in biological systems can be gotten from the  substantial review by Claudio et al. (2003).
14    A complete discussion of the toxicology associated with exposure to Pb can be found in
15    Chapter 5 of this document.
16
17
18    2.2    SOURCES OF LEAD
19          In this section, we summarize information on a number of major sources of lead,
20    categorized as natural sources, stationary point sources, and mobile sources.  In addition to these
21    categories, fugitive emissions such  as resuspension of lead in soil and dust can be important.
22    Resuspension is considered a transport route and is therefore discussed in Section 2.3.
23
24    2.2.1    Natural Sources
25          The common sources of natural Pb include volcanoes, sea-salt spray, biogenic sources,
26    wild forest fires, and wind-borne soil particles in rural areas with background soil concentrations.
27    Natural sources combined contribute an estimated 19,000 metric tons of Pb to the air each year
28    (Nriagu and Pacyna, 1988).  However, there is significant variability in the Pb emissions from
29    volcanoes and forest fires and considerable uncertainty  in biogenic and sea-salt emissions of Pb
30    (Nriagu, 1989). Table 2-6 shows the median value and the range of annual emissions worldwide
31    for natural sources of airborne Pb.

      December 2005                           2-11       DRAFT-DO NOT QUOTE OR CITE

-------
               Table 2-6. Annual, Worldwide Emissions of Lead from Natural Sources
Amount Emitted: Range
Source (thousands of metric tons/yr)
Wind-borne soil particles
Seasalt Spray
Volcanoes
Wild Forest Fires
Biogenic, continental particulates
Biogenic, continental volatiles
Biogenic marine sources
Total
0.3-7.5
0.02-2.8
0.54-6.0
0.06-3.8
0.02-2.5
0.01-0.038
0.02-0.45
0.97-23
Amount Emitted: Median
(thousands of metric tons/yr)
3.9
1.4
3.3
1.9
1.3
0.20
0.24
12
       Source: Nriagu (1989).
 1          The natural lead emissions worldwide are somewhat greater than an estimated 3800
 2   metric tons/year of lead emitted from anthropogenic stationary and mobile sources in the U.S. in
 3   the year 2000 (U.S. EPA, 2003). However, many countries around the world have much greater
 4   lead emissions than the U.S. from stationary and mobile sources, including several countries that
 5   still use leaded gasoline. Furthermore, the EPA estimate does not account for emissions of lead
 6   in resuspended soil. Harris and Davidson (2005) estimate that stationary and mobile source
 7   emissions account for only about 10% of the total lead emissions in the South Coast Air Basin of
 8   California; the remaining 90% of the emissions are from resuspended soil. The soil contains
 9   elevated lead levels because of the many decades of leaded gasoline usage.  Therefore, on a
10   worldwide basis, the anthropogenic emissions of lead are expected to be much greater than
11   natural emissions.
12          There are four stable isotopes of Pb: 204Pb, 206Pb, 207Pb, and 208Pb. The last three of these
13   isotopes are produced by decay of 238U, 235U, and 232Th respectively. The concentrations of
14   natural vs. anthropogenically derived Pb in environmental media are often determined through
15   isotopic ratios. Most Pb ores exhibit ratios of 206Pb/207Pb between 0.92 and 1.20 (Erel et al.,
16   1997). Rock released or "natural" Pb, however, generally exhibits a higher 206Pb/207Pb ratio.
     December 2005                            2-12       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Deep soil samples converge to ratios of 206Pb/207Pb ~ 1.21 and 208Pb/206Pb ~ 2.05 which
 2    are considerably different than the natural ratios found in adjacent bedrock (Erel et al., 1997).
 3          Natural aerosol Pb tends to have large particle sizes (Reuer and Weiss, 2002). As a result,
 4    it deposits rapidly and has an atmospheric residence time of a few hours to -10 days (Reuer and
 5    Weiss, 2002). The average downward flux is estimated as 0.012 mg nT2 yr"1 for natural Pb in all
 6    forms (Bindler et al.,  1999).
 7          Concentrations of lead in the air and soil have most likely been elevated by anthropogenic
 8    activities at least since the rise of the Greek and Roman societies, both of which used Pb
 9    extensively.  The natural, background concentration of Pb in soil is approximately 10-15 ppm
10    (Bindler et al. 1999; Erel et al., 1997). This is significantly higher than the adjacent bedrock but
11    is approximately equal to concentrations found in bedrock residues such as quartz and clay (Erel
12    et al.,  1997).  An estimated 3.1 x 1014 metric tons of Pb are dispersed within the continental crust
13    (Reuer and Weiss, 2002).  Of this, approximately 9.3 x 107 metric tons of Pb are found in Pb
14    ores. Table 2-7 lists the naturally occurring concentrations of Pb in bedrocks, ocean crusts, and
15    continental crusts.
             Table 2-7. Naturally Occurring Lead Concentrations in Major Rock Types
        Lithology                                        Natural Lead Concentration (ppm)
        Continental Crust                                               15.0
        Oceanic Crust                                                   0.9
        Basalts, Gabbros                                                3.5
        Limestones                                                     5.0
        Granulites                                                      9.8
        Greywackes                                                    14.0
        Gneisses, Mica Schists                                           22.0
        Shales                                                         22.0
        Granites                                                       32.0

        Source: Reuer and Weiss (2002).
     December 2005                            2-13       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Natural Pb in surface water is derived from four different sources: biogenic material,
 2    aeolian particles, fluvial particles, and erosion (Ritson et al., 1994). About 90% of natural Pb in
 3    surface waters is in the dissolved phase (Reuer and Weiss, 2002).  Organic ligands are
 4    complexed with 50-70% of this Pb with the balance found in inorganic compounds (Reuer and
 5    Weiss, 2002). Biological particles in the open ocean scavenge a significant portion of the Pb
 6    complexes, which have an estimated two-year residence time in the surface waters (Reuer and
 7    Weiss, 2002).
 8          A naturally occurring, radioactive isotope of Pb, 210Pb, is commonly studied as a tracer to
 9    determine how particles are transported through the environment.  The source of 210Pb is the 238U
10    decay series. In this process, gaseous 222Rn is produced, which escapes from the soil and enters
11    the atmosphere. As radon decays into 210Pb, the paniculate Pb deposits onto soils and surface
12    waters all over the world.  The surfaces of all soils have been exposed to atmospherically derived
13    Pb particles (Bindler et al., 1999).
14          Particles of 210Pb tend to be submicron with an average size of 0.53 jim AMD (Winkler
15    et al.,  1998). The mean residence time for 210Pb in the air is approximately 4-5 days but has been
16    estimated as long as 8 days with some seasonal variability (Winkler et al., 1998). The downward
17    flux has been estimated as 136 Bq nT2 yr"1 for 210Pb (Joshi et al., 1991).  This results in natural,
18    background concentrations in the soil of <0.1 |ig/g (Bindler et al.,  1999).
19          Atmospheric deposition is likely the largest source of 210Pb to water bodies.  Leaching of
20    Pb naturally contained in host rock is a very small source to water (Toner et al., 2003).  Lead-210
21    in surface waters is primarily in particulate form, while dissolved Pb is transported more readily
22    (Joshi et al., 1991).  Dissolved 210Pb is scavenged by suspended matter (Carvalho, 1997).  The
23    residence time of dissolved 210Pb is approximately 30 days although partial re-dissolution from
24    bottom sediments probably occurs (Carvalho,  1997). One estimate found that -56% of
25    atmospherically derived 210Pb in lakes of the Canadian Shield was retained in the sediment
26    (Joshi etal., 1991).
27          Many authors have measured concentrations of 210Pb in plants (including foodstuffs) and
28    animals (including humans). Holtzman (1978) summarized these measurements. Concentrations
29    in United States vegetation range between 30 pCi/kg and  70,000 pCi/kg for wheat and lichens
30    respectively. The estimated human consumption of 210Pb from vegetation averages 1.4 pCi/day
31    in the United States. Overall the concentrations of 210Pb in animals vary  significantly depending

      December 2005                            2-14        DRAFT-DO NOT QUOTE OR CITE

-------
 1    on the type of tissue or organ measured. However concentrations are generally higher in animals
 2    with higher rates of Pb intake.
 3
 4    2.2.2   Stationary Sources
 5    Primary and Secondary Lead Smelters
 6          Primary Pb smelting is the process by which elemental Pb is recovered from Pb ore. Lead
 7    ore is primarily in the form of galena (PbS) but can also occur as plattnerite (PbO2), cerussite
 8    (PbCO3), and anglesite (PbSO4) (Reuer and Weiss, 2002). Producing elemental Pb from ore
 9    involves three processes - sintering, reduction, and refining - each with its own characteristic
10    emissions.  Primary Pb production in the United States emitted about 565 metric tons of Pb in
11    2000, approximately  14.7% of total anthropogenic Pb emissions in the United States (U.S.
12    Environmental Protection Agency, 2003).
13          Secondary Pb  smelters reclaim scrap Pb.  Both the principal input to and the principal
14    major product market of secondary smelters are lead-acid batteries. Secondary Pb production
15    contributed 82% of total Pb production in 2003 (USGS, 2003). Secondary Pb production in the
16    United States emitted about 422 metric tons of Pb in 2000, approximately  11.0% of total
17    anthropogenic Pb emissions in the United States (U.S. Environmental Protection Agency, 2003).
18          The quantity of Pb emitted from a given facility is highly variable and depends on facility
19    processes and meteorological conditions such as wind speed and  ambient temperature.
20    Emissions estimates are typically performed through direct measurements, mass balances,
21    process models, inverse inferences, or emissions factors (Frey and Small, 2003).
22          Emissions estimates and measurements in the literature are scarce.  The AP-42 program of
23    the U.S.  EPA publishes emission factors for each process in the overall smelting sequence.
24    These emission factors are usually expressed as grams of Pb emitted per kg of Pb processed
25    (U.S. Environmental  Protection Agency, 2005). The reader is referred here for emission factors
26    not listed below. A survey of approximately 50 European Pb smelters had mean emission factors
27    of 0.1 grams and 0.05 grams of Pb emitted per kg of Pb processed for primary and secondary Pb
28    smelters respectively (Baldasano et al., 1997). Measurements of emissions from the blast
29    furnace of a primary  smelter were between 1.2 and 3.8 kg Pb/hr (Bennett and Knapp, 1989). The
30    acid-sinter  at the same plant emitted between 0.4 and 8.5 kg Pb/hr (Bennett and Knapp,  1989).
31    Emissions occur during every stage of the overall smelting process. Because the process

      December 2005                            2-15        DRAFT-DO NOT QUOTE OR CITE

-------
 1    emissions mostly are controlled to conserve raw materials, the largest source of emissions is
 2    likely to be fugitive dust from the transport, grinding, and storage of battery scrap (Kimbrough
 3    and Suffet, 1995), which by definition is uncontrolled.
 4          Much work has been done to determine the species of Pb emitted from the various
 5    smelting processes. The fraction of Pb in particulate matter emissions varies significantly
 6    between processes and depends on the type of furnace used.  However, Pb is often the dominant
 7    element in smelter emissions. Lead can be emitted either in particulate matter or in fumes. Lead
 8    fume emissions are particularly high if Pb blast furnace bullion is transferred in an open ladle
 9    (Wang and Morris, 1995).  Major components of particulate Pb emissions are PbS, PbSC>4,
10    PbSCVPbO, and elemental Pb,  and minor species are PbCOs, PbO, Pb silicates, and PbO litharge
11    (Batonneau et al, 2004; Harrison and Williams, 1983; Ohmsen, 2001; Sobanska et al, 1999;
12    Rieuwarts and Farago, 1995).
13          The distribution of particle sizes varies depending on temperature, process, and the
14    conditions of each facility. Ohmsen (2001) found that Pb emissions from a blast furnace tend to
15    be less than 1 jim in size and have a smaller diameter than particulate emissions from either the
16    sintering process or storage areas. Higher temperatures (>600 °C) in the blast furnace tend to
17    produce emissions with finer particle sizes. Dusts from the raw materials area tend to fall
18    between 10  and 100 |im, while  dusts from the refinery tend to fall between 1 and 30 |im
19    (Ohmsen, 2001).  Sobanska et al (1999) found that just 15% of dust particles by mass emitted
20    from a "water jacket" furnace were smaller than 10 jim and the remaining 85% fell between 10
21    and 100 |im. The measurements of Harrison et al. (1981) at a primary smelter found that
22    particles derived from combustion processes were typically between 0.1 and 2 jim, but particle
23    size measurements showed that these particles could agglomerate to more than 10 jam if they are
24    confined to ventilation ducts. Reported sizes from primary smelting processes are shown in
25    Table 2-8.
26          The concentrations of Pb in stack outlets have been measured in several cases.
27    Measurements taken at the stack of a blast furnace at a primary smelter ranged between 3.67 and
28    7.32 mg/m3 (Bennett and Knapp, 1989). Stack concentrations at the sinter plant of the same
29    facility ranged between 4.48 and 71.0 mg/m3 (Bennett and Knapp, 1989).  Two stacks on a blast
30    furnace at a secondary smelting facility had Pb concentrations of 0.002 and 0.0137 mg/m3
     December 2005                            2-16       DRAFT-DO NOT QUOTE OR CITE

-------
         Table 2-8. The Mass-median Aerodynamic Diameters for Particles During Various
                                Processes at Primary Lead Smelters
Average Particle Size
Primary Smelter Harrison et al.
Process (1981)
Raw Materials —
Sinter 5.1 um
Blast Furnace 3.4 um
Copper Drosser 9.4 um
Refinery —
Ohmsen
(2001)
40 um (range = 10-100 um)
range = 10-300 um
90% of particles were <1 um
range = 10-300 um
range = -1-100 um,
mostly <20 um
Bennett and Knapp
(1989)
—
0.91 um, 80% of
particles <10 um
1.1 um, 88% of
particles <10 um
—
—
      Note:  Where there were multiple data points, geometric means were used. Data for Harrison et al. (1981) were
      occasionally given as >11 um. These values were replaced with 11 um before calculating the geometric mean.
      Thus, these values represent a lower limit.
      Source: Harrison et al. (1981), Ohmsen (2001), Bennett and Knapp (1989).
 1   (Sturges and Harrison, 1986). The average values of approximately 50 European smelters were
 2   2 mg/m3 for both primary and secondary smelters (Baldasano, et al., 1997).
 3          The ambient air concentrations in the immediate vicinity of smelters tend to be elevated to
 4   varying degrees depending on facility operations and meteorological conditions. In the UK, an
 5   increase of 15 |ig/m3 in the local ambient air was attributed to the emissions of a single
 6   secondary Pb smelter (Sturges and Harrison, 1986). Harrison and Williams (1983) measured
 7   concentrations of 15.8 |ig/m3, 0.691-5.1 |ig/m3, and 0.151-4.54 |ig/m3 at sites 500 m, 700 m, and
 8   1200 m from the stacks of a primary smelter respectively. Fenceline measurements at two
 9   secondary smelters located in California ranged between 0.85 and 4.0 |ig/m3 (Kimbrough and
10   Suffet, 1995). Air concentration data measured at 50 m, 500 m, and 800 m from the plant were
11   slightly lower but generally the same order of magnitude as the fenceline values.  Ambient
12   concentrations measured at 12 sites within several hundred meters of three secondary Pb
13   smelters in Manitoba were elevated (Tsai, 1987). The geometric means of these samples, which
14   were taken over three month time spans, ranged between 0.107 and 1.69 |ig/m3. Additionally,

     December 2005                           2-17       DRAFT-DO NOT QUOTE OR CITE

-------
 1    the area was shown to be much less likely to meet the Manitoba guideline of <5 |ig/m3 for a
 2    24-hour average when the smelters were operating than when they were not.
 3
 4    Non-Lead Metallurgical Processes
 5          Emissions of Pb from non-lead smelters can be significant. Emissions from smelters,
 6    metal works, and metal refineries depend on the type of equipment used to process the metals,
 7    the concentrations  of Pb in the initial material (ore, recycled material, or alloy), the type and
 8    effectiveness of pollution controls at the facility, and the temperature of operations (Pacyna,
 9    1986). Little work has been done to speciate Pb emissions from metallurgical facilities, although
10    Pb emissions from a primary copper-nickel smelter are primarily in the form of PbO (Barcan,
11    2002). The emissions of Pb from non-lead metallurgical processes are summarized in Table  2-9.
12
13    Ore Mining and Processing
14          Lead mining occurs in 47 countries, although primary Pb production is on the decline
15    (Dudka and Adriano, 1997). World mine production of Pb is approximately 2.8 million metric
16    tons per year (Wernick and Themelis, 1998).  The reserve base of Pb is estimated to be about
17    120 million metric tons, which will sustain current rates of mine production for 43 years
18    (Wernick and Themelis, 1998).
19          Mines can be a significant source of metal emissions to the atmosphere. Lead and zinc
20    ores, which are often mined together, frequently contain high concentrations of cadmium and
21    arsenic (Pacyna, 1986). An emission factor for Pb mines has been reported as 0.91 grams of Pb
22    emitted to the air per kg of Pb mined (Pacyna, 1986).
23          Since Pb is mined in the form of galena (PbS), emissions from Pb mines tend also to be in
24    the form of galena  (Dudka and Adriano, 1997).  However, other species have been detected.
25    In mine spoils, Pb is typically galena and secondary alternation products such as plumbojarosite
26    [PbFe6(SO4)4(OH)i2] (Rieuwerts and Farago, 1995). Other Pb forms detected in the vicinity  of
27    mines are pyromorphite [Pbs^O^Cl], which has a low bioavailability, PbCOs which is formed
28    from the weathering of galena in the soil, leadhillite [Pb4SO4(CO3)2(OH)2], PbS'Bi2S3, Pb
29    oxides, Pb silicates, and PbSC>4 (Rieuwerts and Farago, 1995).
30          Although mining can be considered a point source to  air, mine wastes can have a major
31    widespread effect on soil and water (Riewerts and Farago, 1995).  Mines produce four different

      December 2005                            2-18       DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
Table 2-9. The Emissions of Lead from Non-Lead Metallurgical Processes
to



VO
fe
H

6
o


o
H

O


O
H
W

O


O
HH
H
W
Metallurgical Plant
Aluminum (secondary)
Aluminum (secondary)
Antimony
Antimony
Brass/Bronze refinery
Brass/Bronze refinery - blast
furnace
Brass/Bronze refinery - crucible
furnace
Brass/Bronze refinery - cupola
furnace
Brass/Bronze refinery -
reverberatory furnace
Brass/Bronze refinery - rotary
furnace
Copper-Nickel
Copper-Nickel
Copper-Nickel (primary)
Copper-Nickel (primary)
Copper-Nickel (primary)
Lead Emissions
0.81±0.014% of PM emissions
0.098±0.03 1% of PM emissions
0. 17±0.04% of PM emissions
0. 1 1±0.02% of PM emissions
0.01-1% of PM emissions

16 g/ton produced
10 g/ton produced
65 g/ton produced
60 g/ton produced
60 g/ton produced
184mt/yr, 21 kg/hr
13.4 mt/year
0.6-1.4% of PM emissions
2.3-3.6 kg/ton produced
3.1 kg/ton produced
Particle Sizes
MMAD = Mass median
aerodynamic diameter
Fine (< 2.5 urn)
Coarse (2.5-10 urn)
Fine (< 2.5 urn)
Coarse (2.5-10 urn)
na

na
na
na
na
na
1.2 urn MMAD
0.9 urn MMAD
na
na
na
Location
Philadelphia, USA
Philadelphia, USA
Philadelphia, USA
Philadelphia, USA
na

na
na
na
na
na
Copper Cliff, Ontario
Falconbridge, Ontario
Monchegorsk, Russia
Poland
na
Source
Olmezetal. (1988)
Olmezetal. (1988)
Olmezetal. (1988)
Olmezetal. (1988)
Lee & VonLehmden (1973)

Pacyna (1986)
Pacyna (1986)
Pacyna (1986)
Pacyna (1986)
Pacyna (1986)
Chan &Lusis (1986)
Chan &Lusis( 1986)
Barcan (2002)
Pacyna (1986)
Pacyna (1986)

-------
December 2005
Table 2-9 (cont'd). The Emissions of Lead from Metallurgical Processes
Metallurgical Plant
Copper Smelter - furnace
Copper Smelter - sinter
Copper Smelter (secondary)
Lead Emissions
0.24-0.52 kg/hr
below detection
54-214 g/ton produced
Particle Sizes
MMAD = Mass median
aerodynamic diameter
0.87 urn MMAD
<0. 10 urn MMAD
na
Location Source
na Bennett & Knapp (1989)
na Bennett & Knapp (1989)
na Pacyna(1986)
to
o
fe
H

6
o


o
H

O


O
H
W

O


O
HH
H
W
          Iron Ore Recovery and Ni

          refinery



          Iron and Steel foundry



          Steel works - electric-arc
6 mt/year



0.01-0.1% of PM emissions
Coarse (2.5-10 urn)



       na
Copper Cliff, Ontario   Chan & Lusis (1986)



        na           Lee & Von Lehmden (1973)
turnace
Zinc-Cadmium (primary)
Zinc Smelter - furnace
Zinc Smelter - sinter
4.1-16.3 g/ton produced
1.2-25 kg/ton produced
0.86-1.5 kg/hr
3. 6-6.0 kg/hr
na
na
1.8-2.2 urn MMAD
0.9-2.1 urn MMAD
na
na
na
na
Pacyna (1986)
Pacyna (1986)
Bennett & Knapp (1989)
Bennett & Knapp (1989)
          Source: Olmez et al. (1988), Lee and Von Lehmden (1973), Pacyna (1986), Chan and Lusis (1986), Barcan (2002), Bennett and Knapp (1989).

-------
 1    types of large-volume waste: mine waste, which consists of overburden and barren rocks,
 2    tailings, dump heap leachate, and mine water (Dudka and Adriano, 1997).  Tailings, especially,
 3    are major sources of metal contamination to soil and water (Bridge, 2004). Acid mine drainage
 4    can contain highly elevated levels of Pb, >3000 |ig/L, and can contaminate vast areas (Bridge,
 5    2004; Kurkjian et al., 2004). Soil contamination is addressed in Chapter 3.
 6          Mining of materials other than Pb can also release Pb to the atmosphere.  Zinc-copper
 7    ores, for example, contain Pb in the range of 100-100,000 ppm (Lee and Von Lehmden, 1973),
 8    and about 6.1% of all Pb in the United States is extracted from "zinc mines" (Dudka and
 9    Adriano, 1997).
10          In an underground gold mine, high lead-particulate concentrations were associated with
11    blasting (Annegarn et al., 1988).  These particles were primarily Pb oxides and submicron in
12    size.  A source apportionment analysis on airborne particulate matter in an underground gold
13    mine found that the significant sources of Pb were rock dust and diesel exhaust (McDonald et al.,
14    2003).  Concentrations of airborne Pb inside the mine were measured at 0.21 |ig/m3.
15
16    Stationary External Combustion: Coal Combustion
17          Coal is commonly burned as a fuel for utilities, industries, and commercial and
18    institutional facilities. Coal combustion can be a significant local source of Pb emissions as well
19    as a considerable regional source of airborne Pb.
20          Coal is pulverized, fluidized, or gasified before combustion. Generally, Pb impurities will
21    volatilize early in the combustion process although the precise rate of vaporization depends on
22    the distribution of Pb particles in the coal and the particle sizes (Lockwood and Yousif, 2000).
23    As Pb vapors cool they will condense, either forming individual particles or condensing on  the
24    surface of ash particles (Lockwood and Yousif, 2000; Furimsky, 2000; Clarke, 1993; Pacyna,
25    1986).  A high surface area to volume ratio makes fine ash particles better candidates for surface
26    sorption than coarse particles. Additionally, recondensed Pb particles  tend to be fine, with an
27    average size of 0.2 jim (Lockwood and Yousif, 2000). The fine fraction of particulate matter
28    from coal combustion has an enrichment factor of approximately 22 (Lockwood and Yousif,
29    2000).
30          The primary contributor of Pb  emissions from coal combustion is the Pb content of the
31    coal itself.  Lead is present in all coal samples in varying amounts depending on the location of

      December 2005                            2-21        DRAFT-DO NOT QUOTE OR CITE

-------
 1   the coalfield and even the location of the coal sample within a coalfield. Generally, Pb is present
 2   in trace amounts in the form of PbS, but can also be present as pyrite and PbSe (Lockwood and
 3   Yousif, 2000; Mukherjee and Srivastava, 2005). The rank of the coal - either bituminous,
 4   subbituminous, or lignite - does not seem to correlate with the quantity of trace elements
 5   (Mukherjee and Srivastava, 2005). The age of the coal also does not seem to impact the
 6   concentration of Pb (Ghosh et al., 1987). The most important factors contributing to Pb content
 7   of uncombusted coal seems to be local environmental conditions at the time the coal formed, and
 8   the relative proportions of organic and inorganic matter (Pacyna, 1986; Ghosh et al., 1987).
 9   Globally, the concentrations of Pb in coal range between 2 and 80 ppm (Mukherjee and
10   Srivastava, 2005). Table 2-10 lists the range of Pb concentrations measured in four different
11   coal components.

                 Table 2-10. The Range of Lead Concentrations in Coal Lithotypes
           Coal Lithotype                            Range of Lead Concentrations (ppm)
           Vitrain                                                0.30-16.17
           Clarain                                                4.84-17.55
           Durain                                                4.10-11.76
           Fusain                                                3.64-15.60
           Source: Ghosh etal. (1987).
12          Coal is often combined with limestone as a way to attenuate sulfur dioxide emissions.
13   However, limestone can contain trace elements and has been shown to increase emissions of Pb
14   by four to six times in a fluidized bed system compared to tests performed without a limestone
15   addition (Clarke, 1993). Other measurements performed on a fluidized bed system found that
16   increasing limestone increased particulate emissions of Pb but decreased gaseous emissions of
17   Pb.  The overall emissions of Pb (gaseous + paniculate) remained relatively constant (Furimsky,
18   2000). Limestone had a negligible effect on pressurized fluidized bed systems although Pb
19   emissions from gasification systems may increase with limestone additions (Clarke, 1993).
20          Emissions from coal combustion depend a great deal on the process conditions at a given
21   facility. In addition to the type of boiler,  conditions such as temperature, heating rate, exposure

     December 2005                          2-22        DRAFT-DO NOT QUOTE OR CITE

-------
 1    time at elevated temperatures, and whether the environment is oxidizing or reducing can affect
 2    emissions (Pacyna, 1986). For Pb, changes in the temperature affect the size of particles, the
 3    amount of Pb in the vaporized fraction, and the species of the emissions.  At combustion
 4    temperatures of 1800 K, about 0.1% of the total ash produced was vaporized (Lockwood and
 5    Yousif, 2000).  At 2800 K the vaporized fraction of the ash was increased to 20%.  Additionally
 6    the ratio of air to coal during combustion can have a major effect on emissions (Furimsky, 2000).
 7    In a fluidized bed system, increasing the air to coal ratio from 1.0 to 1.10 decreased the gas to
 8    solid ratio for Pb emissions from 1.5 to 0.18 (Furimsky, 2000).
 9          Uncontrolled combustion of coal can occur - usually as natural, in-ground coal fires - and
10    such combustion can emit Pb (Finkelman, 2004).  Although these fires have local importance,
11    they will not be discussed in detail here.
12          Controlled combustion is the norm for industries and utilities.  The major pollution
13    control systems are electrostatic precipitators (ESP), wet scrubbers, and baghouses. In general,
14    pollution control systems are most effective at removing large particles and are least effective at
15    removing submicron particles.
16          ESPs are highly efficient and can  remove particulates with >99.9% efficiency depending
17    on particle size, ash resistivity, flue gas temperature, and moisture content (Clarke, 1993). ESPs
18    are used at more than 90% of coal-fired utility boilers in the United States (Senior et al., 2000).
19    Particles that escape EPSs are typically in the range of 0.1-1.0 jim in diameter (Senior et al.,
20    2000).
21          Wet scrubbers are also more than 99% efficient (Pacyna, 1986). The majority of particles
22    that escape are <2  jim in size (Pacyna, 1986). Wet scrubbers are used less commonly than ESPs
23    and baghouses (Senior et al., 2000).
24          Baghouses or fabric filters are frequently used by coal-fired utilities. As with ESPs and
25    wet scrubbers, the  collection efficiency of baghouses is a function of particle size (Senior et al.
26    2000). Baghouses are >99% effective with mass emissions averaging <20 mg/m3 (Clarke,
27    1993).
28          Very little information is published regarding the actual quantity of Pb emitted from coal-
29    fired boilers. The EPA AP-42 program publishes emission factors for typical coal-fired boilers,
30    although using process data specific to a given facility is likely to be more accurate. Clarke
31    (1993) reports emissions from fluidized beds. Of the processes tested, the emissions of Pb were

      December 2005                           2-23        DRAFT-DO NOT QUOTE OR CITE

-------
 1   highest from a 0.5 m bed with a limestone sorbent, second highest with a 1.0 m bed without a
 2   limestone sorbent, and lowest with a 0.5 m bed without a limestone sorbent (Clarke, 1993).
 3   Reducing the depth of the fluidized bed by 50% decreased the emissions of trace elements by
 4   -5-50% probably because deeper beds undergo attrition of ash (Clarke, 1993). Olmezetal.
 5   (1988) report the Pb mass fractions of particulate matter in a stack of a coal-fired power plant.
 6   For fine particles, Pb constituted 0.041 ± 0.004%. For coarse particles, Pb constituted 0.026 ±
 7   0.002%.  Coal combustion products that underwent long-range transport from the coal-fired
 8   power plants of the Midwest contributed an estimated 0.05 |ig/m3 to the ambient air in Boston
 9   (Thurston and Spengler, 1985).  Table 2-11 lists the emission factors for three different types of
10   coal, in three different types of power plants.
       Table 2-11. Emission Factors of Lead for Coal Combustion in Three Different Furnaces.
Rank
Bituminous
Subbituminous
Lignite
Cyclone Furnace
Oig/MJ)
85
103
144
Stoker Furnace
Oig/MJ)
128
156
217
Pulverized Furnace
Oig/MJ)
55
66
92
      Note: All furnaces equipped with a 99% efficient ESP. Combustion of each coal type yielded ash at 10% of
      starting mass.
      Source: Pacyna (1986).
11         The species of Pb emitted from coal depends on process conditions. PbSC>4 was found to
12   be the dominant Pb compound in flue gas up to 1150 K (Lockwood and Yousif, 2000). Above
13   this temperature, Pb and PbO, both in the vapor phase, dominate.  As the temperature increases,
14   the equilibrium shifts toward Pb  (Lockwood and Yousif, 2000). In pulverized coal combustion
15   at 1800K, the Pb species found in the gas phase were PbO, elemental Pb, PbCl, and PbCb
16   (Furimksy, 2000). The solid phase was comprised of PbO, PbO»SiO2, elemental Pb, and PbO2
17   (Furimksy, 2000). As the flue gas cools, the composition of Pb changes. PbCl2 increases and is
18   the main constituent of the gas phase before condensation occurs at 900K.  If low rank low

     December 2005                           2-24        DRAFT-DO NOT QUOTE OR CITE

-------
 1   chlorine coal is used, then PbO and elemental Pb will dominate the gas phase.  At 1500K, PbSC>4
 2   dominates the particulate phase; at 1800K PbC>2 was the predominant Pb compound in the
 3   particulate phase (Furimksy, 2000).
 4          The emissions of Pb from coal combustion in industrial, commercial, and residential
 5   boilers are similar to the values listed above for utility boilers. Table 2-12 lists emission factors
 6   for coal combustion.
                  Table 2-12.  The Emissions of Lead from Industrial, Commercial,
                                 and Residential Coal Combustion
       Coal-fired unit                                        Emission factor (g/metric ton)
        Industrial cyclone boiler                                            1.2
        Industrial stoker boiler                                             7.7
        Industrial pulverized coal boiler                                      4.5
        Commercial/Residential boiler
        (stoker or hand-fired)                                              2.7

       Note: Data for industrial boilers assuming 10% ash fraction and 85% efficient control devices.
       Source: Pacyna (1986).
 7   Stationary External Combustion:  Fuel Oil Combustion
 8          Fuel oil combustion constitutes 15% of fossil fuel energy production in the United States.
 9   (U.S. Environmental Protection Agency, 1998). As with coal, fuel oil is used to generate energy
10   for utilities, industries, and commercial and residential boilers.  The discussion below focuses on
11   electric power utilities, which are the largest users of fuel oil.
12          Fuel oil is generally combusted in tangentially-fired or wall-fired boilers. Emissions of
13   Pb from oil combustion depend on the process conditions, the amount of Pb in the oil, and the
14   amount of sulfur in the oil (Pacyna, 1986).
15          The Pb concentration in the oil is the most important factor for determining the eventual
16   emissions from combustion. The concentration of Pb in crude oil ranges between 0.001 to
17   0.31 ppm (Pacyna, 1986). In general, the heavier the crude, the higher the metal concentration.
18   Refining oil removes about 10% of metals (Pacyna, 1986).

     December 2005                            2-25        DRAFT-DO NOT QUOTE OR CITE

-------
 1          As with coal, process conditions and the presence of pollution control devices greatly
 2    affect the rate and characteristics of emissions from fuel oil combustion. Emissions from oil-
 3    fired boilers depend on the efficiency of combustion and how much deposited material has built
 4    up in the boiler (Pacyna, 1986). Additionally, poor mixing, low flame temperatures, and a short
 5    residence time in the combustion zone cause overall particulate emissions to be greater and
 6    individual particle sizes to be larger (Pacyna, 1986). Oil, which is typically atomized prior to
 7    combustion, will emit larger particles and have a higher particulate loading when atomization is
 8    done at low pressures.  Conversely, high pressure atomization leads to smaller particles and
 9    lower particulate loadings (Pacyna, 1986).  In general, about  90% of particulate matter mass is
10    <2.5 |im in diameter (Olmez et al.,  1988).
11          Many emission factors for fuel oil combustion processes are published in the AP-42
12    guidelines. These are of limited quality, but the reader is directed there for more information.
13    Additional published data follow.  An average emission factor for European oil-fired power
14    plants was reported as 126 jig Pb/MJ for oil containing 1% sulfur (Pacyna, 1986).  Lead
15    emissions are higher for oils with greater sulfur contents.  Olmez et al. (1988) report Pb mass
16    fractions for two oil-fired power plants in Philadelphia. Lead was found to be 1.0% ± 0.2% and
17    1.8% ± 0.6% in the fine fraction in these two plants, respectively, and 0.48% ± 0.2% and
18    3% ± 0.4% in the coarse fraction. Lead in particulate matter  at the Philadelphia plants was
19    enriched by more than a factor of 1000 compared to the Pb concentration in the fuel oil. Lead in
20    particulate matter for seven other oil-fired power plants was enriched by more than a factor of
21    100 (Olmez et al., 1988).  A plant in Boston increased the ambient concentration of fine Pb
22    aerosols by an estimated 0.05 |ig/m3 and the ambient concentration of coarse Pb aerosols by
23    0.003 |ig/m3 (Thurston and Spengler, 1985).
24          The combustion of used oil is also common.  About 75% of used oil, which is generated
25    in the transportation, construction, and industrial sectors, is burned as fuel oil (Boughton and
26    Horvath, 2004).  The Pb concentration of used oils is markedly higher than that of low-sulfur
27    crude-based heavy fuel oils (Boughton and Horvath, 2004). Emissions from used oil  combustion
28    are estimated at approximately 30 mg of Pb from the combustion of 1 L of used oil. This is
29    50-100 times higher than emissions from crude-derived fuel oils.
30
      December 2005                           2-26        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Emission rates for industrial boilers are similar to those of utility boilers.  Industrial oil-
 2    fired boilers are not usually equipped with pollution control devices.  Approximately 6.4 g of Pb
 3    are emitted for 1000 L of fuel oil burned with a sulfur content of 1% (Pacyna, 1986).
 4          Commercial and residential boilers, which are also not typically equipped with pollution
 5    control devices, have emissions of approximately 3.3 g of Pb emitted per 1000 L of fuel oil
 6    (Pacyna, 1986).
 7
 8    Stationary External Combustion: Wood Combustion
 9          Wood-fired boilers are used almost exclusively by industries that produce wood or wood
10    products.  These include pulp and paper mills, lumber production facilities, and furniture
11    manufacturers (U.S. Environmental Protection Agency, 1998).  The materials used as fuel may
12    include bark, slabs, logs, cuttings, shavings, pellets, and sawdust.
13          During combustion, elemental pollutants such as Pb are converted to their oxide forms.
14    These are hydrated and later carbonated under atmospheric conditions (Demirbas, 2003a).
15          As with coal and oil, the largest factor affecting emissions from wood combustion is the
16    concentration of Pb in the fuel. Lead concentrations tend to be very low for virgin wood. The
17    median Pb concentration in 24 pine and spruce samples was 0.069 ppm (Krook et al., 2004).
18    The concentrations of Pb in spruce, beech, oak, pine, and ailanthus are listed in Table 2-13.
19          Waste wood recovered from construction and demolition sites is increasingly used as fuel.
20    Although most of this wood is untreated, some can have elevated levels of metals from surface
21    treatment of the wood or industrial preservatives (Krook et al., 2004).  Additionally, waste wood
22    commonly contains contaminants such as metal pieces, concrete, stone, gravel, glass, and soil,
23    which may increase metal emissions during combustion. Lead has been measured in waste wood
24    at levels -40 times higher than levels found in virgin wood. The median concentration of Pb in
25    recovered waste wood in Sweden was 33 ppm (Krook et al., 2004). Lead in recovered waste
26    wood from Germany and the Netherlands had a median value of 110 ppm.
27          Emissions of metals from wood are affected by process conditions. Good air-fuel mixing
28    and high furnace temperatures keep emissions low (Demirbas, 2003a). Additionally  emissions
29    depend on whether or not the wood was combined with other fuels, the feed rate, the physical
30    state of the wood, the stack temperature, the geometry of the boiler which can act as an inertial
      December 2005                           2-27        DRAFT-DO NOT QUOTE OR CITE

-------
            Table 2-13. The Concentrations of Lead in Biomass, Char, and Ash Samples
                       from Spruce, Beech, Oak, Pine, and of Ailanthus Trees
Wood
Spruce trunk wood
Beech trunk wood
Oak trunk wood
Pine trunk wood
Ailanthus trunk wood
Spruce bark
Beech bark
Oak bark
Pine bark
Ailanthus bark
Biomass (ppm)
0.32a
0.36a
0.27a
n.a.
n.a.
0.38a
0.43a
0.31a
n.a.
n.a.
Char (ppm)
2.5a
2.6a
2.1a
n.a.
n.a.
3.1a
3.3a
2.5a
n.a.
n.a.
Ash (ppm)
oo -->a,b
33.2
35.0a'b
284a,b
34.9b
32.7b
5.2a, 36.2b
3.8a, 40.8b
4.0a, 34.0b
38.7b
35.7b
      "Source: Demirbas (2003a).
      b Source: Demirbas (2003b).
 1   paniculate collector, the draft setting, and the amount of moisture in the fuel (Demirbas, 2003a;
 2   Pels et al., 1990; Pacyna, 1986).
 3         Pollution control devices may be present with large-scale wood-fired boilers. These can
 4   greatly reduce particulate emissions. However, in a wood-burner installation in Ontario, a
 5   cyclone was found to have an efficiency of just 53% for total PM mass (Pels et al., 1990). For
 6   particles <2  jim in diameter, the concentrations downstream of the cyclone were actually greater
 7   than those upstream, probably indicating that larger particles were breaking apart during passage
 8   through the cyclone.  The emissions of Pb from wood combustion are highly variable. The
 9   emission factor for wet fuel at a large-scale wood burner was 0.0006 g Pb/kg fuel (Pels et al.,
10   1990). For dry fuel, emission factors were in the range <0.00035 to 0.0014 g Pb/kg fuel burned
11   with an average of 0.00056 g Pb/kg fuel (Pels et al., 1990). Emissions from a wood stove and a
12   fireplace are estimated as 0.007 g Pb and 0.0047 g Pb per kg of wood burned, respectively
13   (Pacyna, 1986).
14         Emissions from the  combustion  of waste wood are higher than emissions from
15   combustion of virgin wood. Although  emission factors are not available, the concentration of Pb
16   in ash is elevated above that from the combustion of virgin wood (Krook et al., 2004).


     December 2005                           2-28        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Data on particle sizes and species of emitted aerosols from wood combustion are not
 2    readily available.
 3
 4    Stationary Combustion Sources: Solid Waste Incineration
 5          Incineration of municipal waste is on the decline in the United States.  Historically it has
 6    been an important source of Pb emissions and locally it is still a concern in some places (Walsh
 7    et al., 2001).  In New York City in the late 1960s, emissions from refuse incineration were
 8    between 602  and 827 tons per year, an appreciable fraction of the emissions from cars, which
 9    totaled -1752 metric tons (Walsh et al., 2001).
10          Incinerator residue is partitioned into bottom ash, fly ash, and flue gas.  Here we focus on
11    Pb in flue gas, due to its importance in increasing airborne Pb concentrations (Chang et al.,
12    1999). Lead  in incinerator effluents is derived primarily from the noncombustible materials that
13    end up in refuse  (Pacyna, 1986). These incinerators may be equipped with pollution control
14    devices such  as cyclones, baghouses, ESPs, electrified gravel beds, and venturi scrubbers (U.S.
15    Environmental Protection Agency,  1998).
16          Factors that affect the quantity of Pb emitted from incinerators include combustion
17    temperature,  the amount of Pb in the refuse, process conditions, moisture content, the addition of
18    reactive species such as calcium, magnesium, and aluminum, and the addition of sorbents. Of all
19    these factors, temperature seems to have the greatest impact on metal volatility (Chen and Yang,
20    1998). Metal volatilization is fast during the initial stages of combustion but levels off after
21    about  15 minutes (Ho et al., 1993;  Chen and Yang,  1998). When plastics only were burned, Pb
22    volatility was at  18% at 600 °C, 61% at 800 °C, and 91% at 1000 °C (Chen and Yang, 1998).
23    Figure 2-1  shows the percent volatility for Pb at four different combustion temperatures over
24    25 minutes of combustion time. Chang et al. (1999) derived the following  relationship for Pb
25    emissions from a fixed bed refuse incinerator in Taiwan:
26
27                             ln£(wf%) = -3.083J11'257 + 3.659                      (2-4)
28
29    where E is the weight percent of Pb in particulate emissions, and T is the combustion
30    temperature in Kelvin.
      December 2005                            2-29        DRAFT-DO NOT QUOTE OR CITE

-------
                        O
100
 90

 80 -
    A
 70 -

 60 J
 50 -
 40 -
 30 -
 20 -

 10 -
Symbol
•e-
*
A
-*-
Temp.
600°C
800°C
900°C
1000°C
                                          10        15         20
                                                Time (min.)
                                             25
     Figure 2-1. Percentage volatility of Pb during combustion of plastics at four temperatures.
     Source: Chen and Yang (1998).
 1          The amount of Pb emitted is dependent on the quantity of Pb in refuse. Typical sources of
 2   Pb include paper, inks, batteries, cans and other metal scrap, and plastics. Plastics are the most
 3   important source of Pb in municipal solid waste. Lead is used in plastic dyes and stabilizers, and
 4   plastics constitute a large portion of the waste stream (U.S. Environmental Protection Agency,
 5   1998; Wagner and Carabello, 1997). For United States municipal solid waste, Pb concentrations
 6   vary between 110 and 1500 ppm with an average of about 330 ppm (Durlak et al., 1997). Since
 7   other countries have very different waste compositions, Pb concentrations elsewhere can vary
 8   greatly.
 9          Additionally, process conditions can affect Pb emissions.  Increasing the amount  of
10   oxygen accelerates the rate of metal volatilization but does not seem to affect the overall amount
11   of metal volatilized (Ho et al., 1993).  Similarly, Chen and Yang (1998) observed that changing
     December 2005
                   2-30
DRAFT-DO NOT QUOTE OR CITE

-------
 1    the N2:O2 ratio from 4:1 to 1:4 increased Pb volatility. An increase in the gas velocity can also
 2    increase Pb emissions although this is a relatively minor effect (Chang et al., 1999; Chen and
 3    Yang, 1998).
 4          The moisture content in an incinerator can affect the behavior of Pb.  At a typical
 5    temperature of 950 °C, decreasing the moisture level from 37% to 5% increased Pb in the fly ash
 6    from 54% to 58% (Durlak et al., 1997). Similarly, decreasing the relative humidity from 60% to
 7    40% at 900 °C increased the Pb volatility from 67% to 76%, respectively (Chen and Yang,
 8    1998).  In addition to these direct effects, moisture can indirectly affect emissions by altering the
 9    combustion temperature (Durlak et al., 1997).
10          Additives can reduce metal emissions from incinerators. Additives such as calcium,
11    magnesium, and aluminum react with metals and bind them.  This has been  shown to reduce the
12    formation of metal particulates. Adding A1(NO3)3, for example, reduced quantities of PbCb
13    emitted (Ho et al., 1993). The addition of Ca(OH)2 did not affect volatility at 600 °C (lower
14    limit for combustion temperature) or at 1000 °C (upper limit for combustion temperature) (Chen
15    and Yang, 1998). However, Ca(OH)2 did appreciably limit Pb emissions at  intermediate
16    temperatures.
17          Sorbents can also reduce metal emissions. Sorbents function by binding metal vapors
18    through heterogeneous chemical absorption and/or condensation before vaporized metals are
19    able to form particulates (Ho et al., 1993). In a fluidized bed incinerator, the efficiency of metal
20    capture with sorbents varied between 4.9% and 94.5% (Ho et al., 1993).  The efficiency was
21    dependent on temperature. Low efficiencies were observed at high and low temperatures; the
22    optimal efficiency was observed in the intermediate range of-600-800 °C.  Limestone was
23    shown to be a more  effective sorbent than sand.
24          Emissions from refuse incinerators have been reported as 0.018 g of Pb emitted per kg of
25    refuse assuming a control device with 85% efficiency (Pacyna, 1986). A source apportionment
26    study showed that refuse incineration increased the ambient concentration of Pb by an estimated
27    0.008  |ig/m3 (Thurston and  Spengler, 1985). This was observed after incinerators had been
28    banned in the  area probably indicating prohibited, residential refuse combustion. Lead in PM
29    emissions has been reported between 6.9% and 8.9% with an average of 8.1% (Pacyna, 1986).
30    Three United States incinerators had emissions in which Pb constituted 8.2 ± 1.6% of the
31    particulate matter (Olmez et al., 1988).

      December 2005                            2-31         DRAFT-DO NOT QUOTE OR CITE

-------
 1          Chlorine plays a critical role in determining the speciation of Pb emissions. Lead exists
 2    primarily as chlorine species (either PbCl or PbCb) (Durlak et al., 1997). However an increase
 3    in moisture content decreases the levels of free chlorine, which has the subsequent effect of
 4    shifting Pb from gaseous PbCbto PbO in particulate form. PbCbfe) is completely volatilized at
 5    430 °C (Chen and Yang, 1998; Chang et al., 1999). Above 800 °C PbCl2 slowly decomposes
 6    and PbO(g) and PbCl(g) are present in greater concentrations.
 7          The presence of sodium can also affect speciation. Sodium has a greater affinity for
 8    binding with chlorine than Pb  (Durlak et al., 1997). Thus increasing the sodium content
 9    effectively shifts the dominant Pb compound from PbCb to PbO. Decreasing the sodium content
10    from 6560 ppm to 4500 ppm (the average value observed in municipal solid waste) was
11    responsible for increasing Pb in the fly ash from 35% to 60% at average moisture levels (Durlak
12    et al., 1997).  High concentrations of sodium attenuate the influence of moisture on Pb
13    emissions.
14          Lead emissions tend to  concentrate in the submicron size range (Chang et al., 1999;
15    Olmez et al., 1988). Lead in the fine fraction was enriched by a factor of more than 105 at
16    several United States incinerators compared with the concentration of Pb in the solid waste
17    (Olmez et al., 1988).  Lead in the coarse fraction was enriched by a factor of more than 1000.
18
19    Stationary Combustion Sources: Sewage Sludge Combustion
20          Sewage sludge incinerators exist at approximately 200 sites in the United States (U.S.
21    Environmental Protection Agency, 1998). Lead can enter the sewage waste stream through car
22    washes, galvanized material, pipe erosion, pigments, food, processed chemicals, and roofs
23    (Krook et al., 2004).  As in other combustion processes, Pb impurities vaporize during
24    incineration and then condense.
25          The Pb content of dry sludge varies between 80 and 26,000 ppm with an average of
26    1,940 ppm (Pacyna, 1986).  Sludge taken from an industrial wastewater treatment plant in
27    Taiwan had Pb levels of 1,500 ppm (Chang et al., 1999). Prior to combustion, Pb is either bound
28    to organic matter in sludge or is present as a carbonate (Lockwood and Yousif, 2000).
29          In sewage sludge incinerators, higher temperatures are associated with higher Pb
30    emissions (Pacyna, 1986).  Additionally, sewage sludge incinerators tend to be equipped with
     December 2005                            2-32       DRAFT-DO NOT QUOTE OR CITE

-------
 1   venturi scrubbers with efficiencies of 90-99% (Pacyna, 1986). Other pollution control devices
 2   are less common.
 3          Sorbents can be effective pollution controls. Kaolinite, in particular, was shown to reduce
 4   Pb emissions significantly (Lockwood and Yousif, 2000).
 5          Emissions have been estimated as 0.14 g Pb emitted per kg of sludge combusted (Pacyna,
 6   1986). The fine fraction of particulate emissions in an experimental setup was enriched with Pb
 7   by a factor of 2.5 (Lockwood and Yousif, 2000).
 8
 9   Stationary Combustion Sources: Scrap  Tire Combustion
10          Waste tires are increasingly used  as a fuel  although uncontrolled burning as a result of
11   accidents or illegal activity is common (U.S. Environmental Protection Agency, 1998). One
12   analysis showed that uncontrolled combustion resulted in Pb emissions on the order of 0.47  mg
13   Pb/kg tire for tires that had been cut into four to six pieces (Lemieux and Ryan, 1993).
14   Emissions were lower for shredded tires, at 0.10 mg Pb/kg tire, probably because of greater
15   oxygen transport between tire pieces. Another analysis detected trace amounts of Pb in the
16   smoke from the combustion of tire bodies but did not detect Pb emissions when the tread was
17   burned (Wagner and Caraballo, 1997).
18
19   Lead-acid Battery Manufacturing
20          Lead-acid batteries constituted 84% of Pb  consumed in 2003 (USGS, 2003). Lead-acid
21   batteries are manufactured from Pb alloy ingots and Pb oxide. Lead alloy ingots are produced by
22   smelters, the emissions of which are  characterized earlier in this chapter. Lead oxide is either
23   produced on-site or is outsourced (U.S. Environmental Protection Agency, 1998).
24          Lead-acid battery manufacture consists of the following processes: grid casting or
25   stamping, paste mixing, plate stacking, plate burning, and assembly into the battery case (U.S.
26   Environmental Protection Agency, 1998). Each process has its own characteristic emissions of
27   Pb. Emissions from Pb oxide manufacture tend also to be in the form of Pb oxides.  These
28   emissions are usually attenuated through a baghouse. The sites of other processes are usually
29   equipped with baghouses or impingement wet scrubbers (U.S. Environmental Protection
30   Agency,  1998).
     December 2005                            2-33        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Cement Manufacturing
 2          The manufacture of Portland cement emits relatively low quantities of Pb.  Trace amounts
 3   of Pb are present in the raw materials of calcium, silicon, aluminum, and iron (U.S.
 4   Environmental Protection Agency, 1998).  As the raw materials are thermo-treated, most of the
 5   Pb is trapped in the resulting clinker although some is released as particulate matter (U.S.
 6   Environmental Protection Agency, 1998).  Additionally, emissions result from the combustion of
 7   the coal, natural gas, or waste tires used to fire the kiln (Pacyna, 1986; U.S. Environmental
 8   Protection Agency, 1998).
 9          Emissions are reduced significantly through the use of pollution control devices. ESPs
10   and baghouses are  both common although baghouses tend to be more effective. Lead is present
11   in the emitted particulate matter in the range of 100 - 1000 ppm (Lee and von Lehmden, 1973).
12   Emission factors for cement production are listed in Table 2-14.
                   Table 2-14. Emission Factors of Lead From Processes Used in
                              Cement Manufacture by Control Device.

Process
Dry Process (total)
Kiln/cooler
Dryer/grinder
Wet Process (total)
Kiln/cooler
Dryer/grinder

Multi-cyclones
16.0
12.0
4.0
12.0
10.0
2.0
Pollution Control Device
ESP
4.0
3.0
1.0
3.0
2.5
0.5

Baghouse
0.16
0.12
0.04
0.12
0.10
0.02
      Note: Units are g PB/metric ton cement.
      Source: Pacyna (1986).


13   Glass Manufacturing
14         The production of leaded glass emits significant quantities of Pb. Its uses primarily
15   include Pb crystal, cathode ray tubes for televisions, and optical glasses such as binoculars,


     December 2005                            2-34       DRAFT-DO NOT QUOTE OR CITE

-------
 1    microscopes, and telescopes (U.S. Environmental Protection Agency, 1998). Leaded glass is
 2    composed of silica sand and Pb oxide. Lead oxide concentrations in the United States-produced
 3    leaded glass typically range between 12% and 60% but can be as high as 92% (U.S.
 4    Environmental Protection Agency, 1998).
 5          The basic process of glass manufacturing includes blending the raw materials, melting,
 6    and forming and finishing. Lead emissions can occur during all of these processes. During
 7    blending, forming, and finishing, Pb is emitted as part of fugitive dust emissions in minor
 8    quantities (Shapilova and Alimova,  2000; U.S. Environmental Protection Agency, 1998).
 9          The major source of emissions comes from the melting process. Emissions from melting
10    depend mostly on the amount of Pb  oxide in the raw material (Shapilova and Alimova, 2000;
11    U.S. Environmental Protection Agency, 1998). Other factors are the type and efficiency of the
12    furnace, the waste-gas volume, the smoke-flue length, and the efficiency of pollution control
13    devices (Shapilova and Alimova, 2000). Electric furnaces emit significantly less Pb than gas-
14    flame furnaces. One analysis found that the rate of Pb emissions from a gas-flame regenerative
15    furnace was more than seven times higher than the rate of emissions from a deep tank electric
16    furnace (Shapilova and Alimova, 2000). Baghouses are the most efficient pollution control
17    device for glass manufacturing operations (U.S. Environmental Protection Agency, 1998).
18    Wet scrubbers are relatively ineffective, and ESPs are between 80% and 90% effective (U.S.
19    Environmental Protection Agency, 1998).  Rates of Pb emissions from several types of furnaces
20    are listed in Table 2-15.
21
22    Ammunition Production and Shooting Ranges
23          In 2003, 48,800 metric tons of Pb were consumed in the United States for the production
24    of ammunition (USGS, 2003). Additionally, some Pb is used to produce Pb azide or Pb
25    styphnate, which is a detonating agent.  Small arms manufacturing plants are likely emitters of
26    Pb although the actual quantity is unknown.
27          Shooting ranges, both outdoor and indoor, may have a local impact on airborne Pb
28    concentrations. Lead is emitted from cast Pb bullets and lead-based primers (Gulson et al.,
29    2002). The propellants contain <2 ppm and seem to have a negligible effect on air
30    concentrations. A 97% reduction in the air Pb concentrations was observed when Cu-jacketed
31    bullets replaced cast Pb bullets (Gulson et al., 2002). In comparing the Pb exposure of

      December 2005                            2-3 5        DRAFT-DO NOT QUOTE OR CITE

-------
            Table 2-15. Rate of Lead Compound Emissions from Glass-Melting Furnaces
       Equipment                                      Product                Lead Compound
                                                                            Emissions g/sec)
       Electric tank furnace with gas-heated              „,    .,.  , ,n/ „ „               „ ,, .
         ..      a         6                     Glass with 16% PbO               0.134
       working zone
       Electric tank furnace with gas-heated              „,    .,.  , ,n/ „ „               „ „„,
         . .                6                     Glass with 16% PbO               0.002
       working zone
       Gas-flame potter furnace3                       Glass with 16% PbO               0.004
       Slag-lining electric furnace with gas-heated        „,    .,. ^.-n/™~               «™,,
         ,.                                     Glass with 64.5% PbO               0.004
       working zone

       "Fitted with a "cassette pulse filter" designed specifically to capture paniculate emissions from small-sized,
        glass-melting furnaces.
       Source: Shapilova and Alimova (2000).
 1   personnel, there seems to be little difference between indoor and outdoor firing ranges (Gulson
 2   et al., 2002). One study found that soil Pb concentrations at an outdoor firing range were
 3   elevated by up to 2600 times background concentrations, indicating significant atmospheric
 4   deposition (DeShields et al., 1998).
 5          An additional source of Pb emissions may be explosive ordnance disposal (EOD) (U.S.
 6   Environmental Protection Agency, 1998). Emissions from EOD are either from the combustion
 7   or detonation of the propellant and primer material or from nonenergetic wastes such as
 8   containers and other wastes associated with the propellant (U.S. Environmental Protection
 9   Agency, 1998).
10
11   Demolition
12          A study of Pb dust-fall during the demolition and debris removal of urban row houses
13   found that Pb was released in very large quantities (Farfel et al., 2003). Many of the row houses
14   demolished at three sites in Baltimore, MD contained lead-based paint in addition to being near
15   sites with elevated levels of Pb in street dust (-700 ppm), sidewalk dust (-2000 ppm), and
16   residential entryway mat dust (-750 ppm).  The results of the study showed that dust fall within
17   10m of the demolition sites was much higher than baseline measurements and was highly
18   enriched with Pb (Farfel et al., 2003). The geometric mean Pb  dust fall rate increased to 410 jig
     December 2005                            2-36        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Pb/m2/hr during demolition and to 61 jig Pb/m2/hr during debris removal.  The baseline rate is
 2   just 10 jig Pb/m2/hr. The Pb concentration in dust fall was 2600 ppm during demolition,
 3   1500 ppm during debris removal, and 950 ppm at baseline (Farfel et al., 2003).
 4
 5   Other Stationary Sources of Lead Emissions
 6         There are additional stationary sources of Pb emissions that have not been mentioned
 7   above. Each of these sources are relatively small, but may be an important local source.
 8   Previously unmentioned Pb sources include: medical waste incineration, hazardous waste
 9   incineration, drum and barrel reclamation, crematories, pulp and paper mills, pigment
10   production, Pb cable coating production, frit manufacturing, ceramics and glaze production, type
11   metal production, pipe and sheet Pb production, abrasive grain processing, solder manufacturing,
12   electroplating, resin stabilizer production, asphalt concrete production, paint application, and
13   rubb er product! on.
14
15   2.2.3   Mobile Sources
16   Automotive Sources of Lead Emissions
17         Lead is used to manufacture many components in on-road vehicles including the battery,
18   bearings, paint primers, corrosion-resistant gas tanks, and some plastic and ceramic electrical
19   components (U.S. Environmental Protection Agency, 1998).  The major sources of Pb
20   emissions—fuel combustion and vehicle wear—are considered below.
21
22   Emissions from Combustion of Unleaded Gasoline
23         Although its phase out began in 1975, Pb was still added to gasoline in the United States
24   as an anti-knock additive at the time of the last Criteria Document.  The United States completed
25   its phase out of Pb additives in 1990, and airborne concentrations have fallen dramatically
26   nationwide.  This is considered one of the great successes for public and environmental health
27   (Nriagu, 1990). Airborne concentrations in the United States fell an average of 94% between
28   1983 and 2002 and 57% between 1993 and 2002 (U.S. Environmental Protection Agency, 2003).
29         Most countries have made a similar move away from leaded fuel, but a few continue the
30   practice of adding tetraethyl Pb to automotive gasoline.  Worldwide Pb consumption for gasoline
31   peaked in the 1970s at just under 400,000 metric tons, but by 1993, this value fell to about

     December 2005                           2-37        DRAFT-DO NOT QUOTE OR CITE

-------
 1    70,000 metric tons (Socolow and Thomas, 1997). Leaded gasoline was the largest source of air
 2    emissions throughout the 1970s and 1980s (Socolow and Thomas, 1997). In Pakistan, a country
 3    that continues to use leaded fuel, the airborne concentrations in the urban center of Karachi range
 4    between 2.0 and 19 jig Pb/m3 (Parekh et al., 2002).  This is 2 to 3 orders of magnitude higher
 5    than typical urban concentrations in the United States.
 6          In the absence of tetraethyl Pb additives, Pb is emitted from automobiles as a trace
 7    element in particulate matter.  Metals enter the vehicle in trace amounts, naturally occurring in
 8    gasoline.  The amount of particulate matter that is emitted from the car depends on a number of
 9    variables including the ambient temperature, the  cruising speed, the amount of stop-and-go
10    activity, the type of catalyst, the fuel quality, the  phase of driving, and the age, size, maintenance
11    level, and engine type of the vehicle.
12          The amount of Pb that naturally occurs in  gasoline is approximately 0.00005 g/L (Harris
13    and Davidson, 2005). An estimated 30-40% of this Pb deposits in the engine and exhaust
14    system; the balance is emitted (Hutzicker et al., 1975; Loranger and Zayed, 1994).
15          Particulate matter emissions have been shown to be higher in older vehicles than in newer
16    vehicles (Gillies et al., 2001; Cadle et al., 1999).  Gillies et al. (2001) compared emission factors
17    from several  studies, and found that emission factors from car models between the years 1964
18    and 1983 had emission factors for PM that were about an order of magnitude higher than models
19    from the 1990s.  This was true even of catalyst-equipped vehicles. Similarly, Cadle et al. (1999)
20    tested  195 cars with model years between 1971 and 1996. Their results, which are listed in
21    Table 2-16, show an increase in emission rates with automobile age.
22          Vehicles that have visible tailpipe emissions are known as "smokers."  The emissions of
23    almost all pollutants are elevated from smoking vehicles compared to their non-smoking
24    counterparts. Emission rates of Pb  from smokers are an order of magnitude higher than typical
25    cars manufactured in the 1990s, as shown in Table 2-16. Interestingly, another study found that
26    smoking and other high-emitting vehicles emitted more Pb after undergoing repair than before
27    (Cadle et al., 1997).  The emission rate of Pb before repair had an average value of 0.029 mg/mi
28    with a standard deviation of 0.047 mg/mi.  After  repair, the emission rate for Pb increased to
29    0.161 mg/mi with a standard deviation of 0.346 mg/mi. The authors explain this surprising result
30    by suggesting that either changes in combustion conditions caused elemental deposits from the
31

      December 2005                            2-3 8        DRAFT-DO NOT QUOTE OR CITE

-------
              Table 2-16. Emission Factors of Lead for Automobiles with Model Years
                                      Between 1971 and 1996

Vehicle Category
1991-1996
1986-1990
1981-1985
1971-1980
Smokers
Diesel

Summer
0.003
0.027
0.006
0.043
0.035
0.15
Emission Factors in mg/mile
Winter
0.019
0.019
0.103
0.222
0.282
0.142
      Note:  "Diesel" denotes diesel automobiles, "Smokers" denotes automobiles with visible emissions.
      Source: Cadle etal. (1999).
 1   engine and exhaust system to be released, or particulate matter deposited during repair and
 2   testing was not removed before emissions testing (Cadle et al., 1997).
 3          Table 2-16 also shows the effect of the ambient temperature on emission rates of Pb.
 4   Emissions tend to be higher during cold months than during warm months (Cadle et al., 1999).
 5          The rate of emissions is largely dependent on the phase of driving.  The Federal Test
 6   Procedure analyzes three phases: cold start, hot stabilized, and hot start, the results of which are
 7   shown in Table 2-17. Driving cycles that are not included are the highway fuel economy test,
 8   and a high speed, high load cycle known as US06 (Cadle et al., 1999).  Emissions were
 9   significantly higher during cold start than during the hot stabilized and hot start phases.
10          Despite the large variability in Pb emissions, several studies describe average on-road
11   emission factors for a typical fleet. Sternbeck et al. (2002) measured metal concentrations in two
12   tunnels in Gothenburg, Sweden.  The emission factors subsequently derived were 0.036 ±
13   0.0077 mg/km per vehicle and 0.035 ± 0.014 mg/km per vehicle for the two tunnels,
14   respectively. Another tunnel study was performed on a fleet comprised of 97.4% light-duty
15   vehicles and 2.6% heavy-duty vehicles in the Sepulveda Tunnel in California (Gillies et al.,
16   2001). The emission factors for Pb were 0.08 mg/km per vehicle and 0.03 mg/km per vehicle in
17   the PM10 and PM2.5 fractions respectively. Lough et al. (2005) analyzed emissions from on-
18   road vehicles in two tunnels in Milwaukee, Wisconsin.  Trucks constituted between  1.5% and
     December 2005                            2-39        DRAFT-DO NOT QUOTE OR CITE

-------
             Table 2-17. Emission Factors of Lead for Automobiles with Model Years
                                     Between 1971 and 1996
Summer Emission Factors in mg/mile
Vehicle Category
1991-1996
1986-1990
1981-1985
1971-1980
Smokers
Diesel
Cold Start
0.005
0.041
0.016
0.112
0.116
0.190
Hot Stabilized
0.002
0.020
0.002
0.015
0.010
0.048
Hot Start
0.002
0.031
0.006
0.044
0.031
0.313
      Source: Cadle etal. (1999).
 1   9.4% of the vehicles, with the balance comprised of passenger cars. Lead emission rates were on
 2   the order of 0.01 mg/km per vehicle and 0.1 mg/km per vehicle in the summer and winter
 3   respectively.  Cadle et al. (1999) analyzed 195 in-use, light-duty vehicles using two
 4   dynamometers. Their results are shown in Tables 2-15 and 2-16. A test on noncatalyst-
 5   equipped, light-duty vehicles found that Pb constituted about 0.03% of the fine particle mass
 6   emitted from these vehicles (Kleeman et al., 2000).
 7         Vehicle-derived Pb seems to have a bimodal distribution. The submicron mode is likely
 8   the product of combustion or high temperatures, and therefore probably came from the tailpipe
 9   (Lough et al., 2005; Harrison et al., 2003; Abu-Allaban et al., 2003). The coarse mode, with an
10   approximate size range of 1.0 to 18 |im in diameter, is likely a product of physical processes
11   such as road dust resuspension and tire or brake wear (Lough et al., 2005; Abu-Allaban et al.,
12   2003). More than 80% of the airborne Pb particles near a roadway were 
-------
 1          Emissions of particulate matter from diesel vehicles are highly dependent on the mode of
 2   operation (Shah et al., 2004). Emission rates are much higher in simulated congested traffic
 3   situations than at cruise or highway speed conditions (Shah et al., 2004).
 4          Extensive  profiles of diesel emissions were developed by Lowenthal et al. (1994). Their
 5   results for Pb are summarized in Table 2-18.
        Table 2-18.  The Concentration of Lead in Particulate Matter Emissions and Emissions
           Factors for Lead from Buses and Trucks Fueled with Diesel No. 2 and Jet A Fuel
Concentration of
Fuel and Vehicle Type Pb in PM (%)
Truck, Diesel No. 2
Truck and Bus, Diesel
No. 2
Truck and Bus, Jet A
Bus, Jet A and Diesel No. 2
with paniculate trap
Bus, Jet A with paniculate
trap
Phoenix PM10 study
0.0007
0.0006
0.0010
0.0009
0.0028
0.0147
Uncertainty
(%)
0.0028
0.0025
0.0055
0.0052
0.0132
0.0294
Emission Factor
(mg/km)
0.0053
0.0045
0.0050
0.0016
0.0018
n.a.
Uncertainty
(mg/km)
0.0187
0.0188
0.0214
0.0100
0.0085
n.a.
       The results of Chow et al. (1991) on heavy-duty paniculate emissions in Phoenix are listed in the last row
       for comparison.
       Source: Lowenthal etal. (1994).
 6          Particulate matter from diesel vehicles tends to be smaller than PM2 5 (Gillies et al., 2001;
 7   Kleeman et al., 2000).  The peak of the particle mass distribution appears to be around 0.1 jim
 8   (Kleeman et al., 2000).  Although no data were available specifically for Pb, such small particle
 9   sizes would be consistent with expectations from high-temperature processes.
10
11   Emissions from Vehicle Wear
12          Vehicle wear and loss of Pb wheel weights are considered as sources of roadside Pb
13   contamination. Brake wear, in particular may emit significant quantities of Pb in parti culate
14   matter. Harrison et al.  (2003) note that Pb is poorly correlated with emissions of NOX, which is
     December 2005
2-41
DRAFT-DO NOT QUOTE OR CITE

-------
 1    emitted from tailpipes. These authors suggest that brake wear contributes the additional
 2    quantities of Pb observed in ambient air. Sternbeck et al. (2002) compare emission factors
 3    derived in other studies. Estimates of Pb emissions from brake pads in Sweden were just under
 4    200 jig/km per vehicle (Sternbeck et al., 2002).  This is an order of magnitude higher than the
 5    tailpipe emissions measured by Cadle et al. (1999).
 6          Up to 35% of brake pad mass loss is emitted as airborne particulate matter (Garg et al.,
 7    2000). One study that analyzed particulate emissions from seven different brake pad
 8    formulations found that only one type of brake pad described as "potassium titanate, aramid, and
 9    copper fiber" emitted  particulate matter with a measurable Pb fraction (Garg et al., 2000).
10          A joint study in Reno, NV and Durham/Research Triangle Park, NC found that the
11    dominant contributors to particulate matter were resuspended road dust and tailpipe emissions
12    (Abu-Allaban et al., 2003).  However, brake wear was a significant  source of particulate matter
13    in places where strong braking occurred, such as at freeway exits (Abu-Allaban et al., 2003).
14          Particulate matter emissions from brake pads were primarily  in the fine fraction.  Eighty-
15    six percent and 63% of airborne PM was smaller than  10 and 2.5  |im, respectively (Garg et al.,
16    2000). It is expected that Pb particles from mechanical processes such as brake wear would be in
17    the coarse fraction. However, smaller particles  may be observed if Pb is vaporized from hot
18    brake surfaces (Harrison et al., 2003; Lough et al., 2005).
19          Lead weights used to balance vehicle wheels may pose an additional threat to roadside
20    concentrations of Pb.  In Albuquerque, NM deposition of Pb wheel weights was estimated to be
21    between 50 and 70 kg/km per year (Root, 2000). Wheel weights are 95% Pb, 5% antimony, and
22    typically weigh between 7 and 113 grams.  These wheel weights can become dislodged during
23    quick stops. Although deposited pieces of wheel weights are quite large, Pb is very malleable
24    and can be worn away into respirable particles by being run over by vehicles (Root, 2000).
25
26    Emissions from Racing Vehicles
27          Vehicles used in racing (including cars, trucks,  and boats) are not regulated by the EPA
28    according to the Clean Air Act, and can therefore use alkyl-lead additives to boost octane. Data
29    on Pb levels in racing fuel and rates of Pb emissions are scarce.  The U.S. Department of Energy
30    stopped tracking information on the production of leaded gasoline for non-aviation use in  1990
31    (U.S. Environmental Protection Agency, 2002). However, the National Motor Sports Council

      December 2005                            2-42        DRAFT-DO NOT QUOTE OR CITE

-------
 1    reports that approximately 100,000 gallons of leaded gasoline were used by National Association
 2    for Stock Car Automobile Racing (NASCAR) vehicles in 1998 (U.S. Environmental Protection
 3    Agency, 2002).
 4          As was the case with on-road emissions during the time of universal leaded gasoline use,
 5    the combustion of racing fuel likely elevates airborne Pb concentrations in the nearby area. This
 6    may pose a serious health risk to some subpopulations such as residents living in the vicinity of
 7    racetracks, fuel attendants, racing crew and staff, and spectators.
 8          The EPA has formed a voluntary partnership with NASCAR with the goal of permanently
 9    removing alkyl-Pb from racing fuels used in the Busch, Winston Cup, and Craftsman Truck
10    Series (U.S. Environmental Protection Agency, 2002).
11          Emissions from the combustion of leaded fuel are generally in the form of submicron
12    particles of inorganic Pb halides.
13          In addition to racing vehicles and piston engine aircraft, legally permitted uses of leaded
14    fuel include construction machinery, agricultural equipment, logging equipment, industrial and
15    light commercial equipment, airport service equipment, lawn and garden equipment, and
16    recreation equipment including boats, ATVs, jet skis, snowmobiles, etc., (U.S. Environmental
17    Protection Agency, 2000). Given the relative unavailability of leaded fuel, it is unlikely that it is
18    commonly used for any of these purposes other than racing vehicles.
19
20    Aircraft
21          Piston-engine aircraft use leaded fuel.  Aviation fuel or avgas contains between 0.1 and
22    1.0 g of tetraethyllead additives per liter. About 32.7% of general aviation aircraft use avgas, the
23    remainder use jet fuel, which does not contain Pb additives (Harris and Davidson, 2005). The
24    overall fraction of aviation fuel containing Pb additives is unknown.
25          In the South Coast Air Basin of California, emissions of Pb from general aviation aircraft
26    was estimated as 634 ±110 kg/year (Harris and Davidson, 2005). This corresponds to
27    0.54 grams of Pb released per flight. Approximately 267 kg of the total was emitted below the
28    mixing height in 2001, which could be a local source of Pb exposure.
29          Commercial jet aircraft do not use leaded fuel.  However, they are also likely sources of
30    Pb emissions. In-flight sampling of contrails from a DC-8 and a 757 showed that metals
31    constituted more than 11% and 5.2% of particulate matter, respectively (Twohy and Gandrud,

      December 2005                           2-43        DRAFT-DO NOT QUOTE OR CITE

-------
 1    1998). This is a lower limit for the fraction of metals in emissions since almost half of the
 2    particles in contrails are from the ambient air (Twohy and Gandrud, 1998).
 3         No known estimates have been made of the quantity of Pb in commercial aircraft
 4    emissions.  However, the dominant metals seem to be Fe, Cr, and Ni (Karcher, 1999). These are
 5    the primary components of stainless steel and indicate that engine erosion is a significant source
 6    of metal  emissions (Karcher, 1999).
 7         Metal particles in contrails have two modes.  One is submicron with an average diameter
 8    of about 0.36 jim (Karcher,  1999; Twohy and Gandrud, 1998).  The larger mode is ~1 |im in
 9    diameter and has a morphology that suggests mechanical generation (Karcher, 1999).
10
11    Lawn-care Equipment
12         A life cycle assessment used to compare gasoline-, electricity-, and battery-powered lawn
13    mowers found that electricity-powered mowers had the fewest overall emissions over its lifetime
14    (Sivaraman and Lindner, 2004).
15         Battery powered mowers are fitted with a lead-acid battery. The total amount of Pb
16    released to the environment from the battery over its lifetime is approximately 0.052 kg Pb
17    which includes consideration of raw material extraction and refining, energy production, Pb
18    mining and refining, battery manufacture, and battery recycling (Sivaraman and Lindner, 2004).
19         Electricity-powered lawn mowers presumably emit less PM and Pb than gasoline-
20    powered mowers.  This is a  reasonable assumption since utility generation plants tend to be fitted
21    with pollution control devices and internal combustion  engines of gasoline-powered mowers
22    do not.
23
24    Other Mobile Sources of Lead Emissions
25         Lead emissions are associated with the combustion of any fossil fuels.  Thus, any of the
26    following may be additional mobile sources of Pb emissions that are not addressed above:
27    construction equipment, off-road recreational vehicles,  generators, marine vessels, locomotives,
28    agricultural equipment, logging equipment, and lawn and garden equipment. However, detailed
29    data on these sources are not readily available.
30
31

      December 2005                           2-44        DRAFT-DO NOT QUOTE OR CITE

-------
 1   2.3   TRANSPORT WITHIN THE ENVIRONMENT
 2   2.3.1   Atmospheric Transport of Lead Particles
 3   Atmospheric Dispersion
 4          The atmosphere is the major environmental transport pathway for anthropogenic lead
 5   (Reuer and Weiss, 2002).
 6          Airborne lead tends to be in the form of submicron aerosols (Davidson and Rabinowitz,
 7   1992; Davidson and Osborn, 1986; Harrison, 1986; Lin et al., 1993). The mass median diameter
 8   averaged for several studies is 0.55 jim (Milford and Davidson, 1985).  A study performed in
 9   1991, after leaded gasoline was no longer the predominant source of lead in the atmosphere,
10   showed a bimodal distribution for lead particles with the larger peak in the fine fraction (Lin
11   et al., 1993). The mass median diameter for lead samples was 0.38 ± 0.06  jim in the fine fraction
12   and 8.3 ± 0.6 jim in the coarse fraction. Since small particles are much slower to deposit than
13   larger particles, lead can be transported great distances in the atmosphere.  Detectable quantities
14   of lead have been found even in the most remote places on earth.  Because much of the airborne
15   lead is generally associated with fine particles, atmospheric dispersion models used for gaseous
16   pollutants can be applied to estimate atmospheric flows of lead under certain conditions. Use of
17   such dispersion models is more accurate for submicron lead emitted from stacks than it is for
18   larger particles resulting from fugitive emissions, such as resuspended soil particles.
19          The airborne concentration of a species emitted from a point source is frequently
20   described with a Gaussian distribution.  This simple description holds true  only when turbulence
21   is stationary and homogeneous. However, the Gaussian model can be modified to account for
22   more complex atmospheric conditions. For a thorough discussion of assorted Gaussian plume
23   models and parameters, the reader is directed toward the work of Seinfeld and Pandis (1998).
24   Gaussian models are in general reasonably accurate for small-scale work - within approximately
25   100 km of the source.
26          The rate and direction of dispersion are dependent both on  pollutant characteristics and
27   meteorological conditions. Important meteorological factors include windspeed, surface
28   roughness, inversion frequency, inversion duration, and the temperature.
29          A Gaussian dispersion model (EMITEA-AIR) was applied to theoretical primary and
30   secondary lead smelters in Europe (Baldasano et al., 1997). This model accounts for plume rise
31   as well as interactions between the plume and terrain.  Two sites were modeled. Conditions in

     December 2005                           2-45        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Copenhagen, Denmark included flat terrain, dominant strong winds, neutral or stable turbulence,
 2    and an annual mean temperature of 10°C.  Conditions in Catalunya, Spain had a complex terrain,
 3    weak winds, unstable turbulence, and an annual mean temperature of 15°C.
 4          The results of these modeling efforts showed that airborne concentrations of lead were
 5    both lower and more symmetric surrounding the Copenhagen site than surrounding the
 6    Catalunya site (Baldasano et al., 1997). Concentrations at the Copenhagen site had a maximum
 7    value of 0.004 |ig/m3. Concentrations at the Catalunya site ranged between 0.065 and 0.3 |ig/m3.
 8    The prevalence of calm winds and the complex terrain were the most important factors
 9    contributing to high lead concentrations surrounding the Catalunya smelter.
10          Modeling efforts for an abandoned battery recycling facility using the EPA Industrial
11    Source Complex Short Term (ISCST) model, based on Gaussian equations, showed good
12    agreement with measured concentrations (Small et al., 1995). Model predictions at three sites at
13    distances between 240 and 310m from the stack were between 3.8  and 4.4 |ig/m3 while
14    measured concentrations taken when the plant was in full operation had averages between 4.1
15    and 5.2 |ig/m3.
16          For long-range transport modeling,  Lagrangian trajectory or Eulerian grid models are
17    commonly employed. These models determine how a parcel of air moves relative to the moving
18    fluid and a fixed coordinate system, respectively.
19          Two Lagrangian experiments were performed in  the Azores  in the northern Atlantic
20    (Veron and Church, 1997). Retrospective  air mass trajectories based on the hybrid single-
21    particle Lagrangian integrated trajectory (HY-SPLIT) model found that air masses  enriched with
22    lead had been over continental regions ten days prior to  testing.  This is consistent with current
23    understanding that most lead emissions are from sources on continents, not from oceanic
24    sources.  Airborne lead at this remote location was transported from several different countries
25    (Veron and Church, 1997).
26          Similarly, backward air mass trajectories estimated for Greenland showed that the highest
27    air concentrations of metals were in air parcels that had  been over continental regions five days
28    earlier (Davidson et al., 1993).  The model used in this study employed a constant acceleration
29    formulation of the trajectory equations, and encompassed air parcel movements affected by
30    terrain and meteorology.  The air masses with the highest metal  concentrations were traced back
      December 2005                            2-46        DRAFT-DO NOT QUOTE OR CITE

-------
 1    to polluted regions including the Arctic Basin, eastern North America, and Western Europe
 2    (Davidson etal., 1993).
 3          A numerical model that combined weather system modeling with three-dimensional
 4    Lagrangian transport and diffusion modeling was used to determine the foreign contributions of
 5    lead to airborne concentrations in Israel (Erel et al., 2002). These predictions in conjunction with
 6    isotopic measurements indicated that Israel received significant amounts of lead from Egypt,
 7    North Africa, the United Arab Emirates, Jordan, Turkey, and Eastern Europe (Erel et al., 2002).
 8
 9    Historical Records of A tmospheric Lead Transport and Deposition
10          An important field of research involves analyzing natural records of lead deposited from
11    the atmosphere. Lead concentrations are measured in media such as soil, sediments, ocean
12    water, peat bogs, plants, snowpacks, or ice cores. Based on concentrations, ratios to other
13    pollutants, or isotopic compositions, an airborne concentration is back calculated and in some
14    cases the major emitters can be identified. Sediments can provide records dating back several
15    million years, peat bogs can reach back to the late glacial period (-15000 years ago), corals and
16    trees can record up to several hundred years, and lichens and mosses can provide recent
17    deposition data (Weiss et al., 1999).  Additionally, some applications can yield data showing
18    variation with seasons or climate.  These methods have been used to monitor both short and
19    long-range transport.  For a comprehensive look at natural historical records, the reader is
20    referred to the review articles by Weiss et al. (1999), Boutron et al. (1994), and Garty (2001).
21
22    2.3.2  Deposition of Airborne Particles
23          Deposition, both  dry and wet, is the major removal mechanism for atmospheric pollutants.
24    Here we focus on deposition data published specifically for lead aerosols, although the literature
25    on particle deposition is extensive.
26
27    Dry Deposition
28          Dry deposition is the process by which pollutants are removed from the atmosphere in the
29    absence of precipitation. The downward flux,-F, is characterized by:
30
31                                       -F = VdC                                     (2-5)

      December 2005                           2-47        DRAFT-DO NOT QUOTE OR CITE

-------
 1    where C is the airborne concentration in |ig/m3 and Vd is the deposition velocity in m/second.
 2    The deposition velocity is an empirical quantity defined by Equation 2-1 as the ratio of F to C
 3    with units of m/s.  It should be noted that both the airborne concentration and the deposition
 4    velocity are dependent on vertical height.
 5          The physical factors governing dry deposition are often described in a manner analogous
 6    to electronic resistances (Davidson and Wu, 1990). The parameters of aerodynamic resistance,
 7    boundary layer resistance, and surface resistance run in parallel with sedimentation resistance or
 8    gravity. The relative importance of each of these resistances varies with particle size and
 9    meteorological conditions (Wu et al., 1992a).
10          The size of depositing particles is arguably the most important factor affecting deposition
11    rates.  For very small particles, Brownian motion is the dominant mechanism that transports
12    particles through the viscous sublayer that borders surfaces (Nicholson, 1988a). For large
13    particles, sedimentation is the most important process governing particle deposition.
14    For intermediate particles impaction and interception largely determine deposition rates.
15    The deposition velocity has the most uncertainty for these intermediate sized particles
16    (Nicholson, 1988a).  Although most of the airborne lead mass is associated with submicron
17    particles, only about 0.5% of the lead particle mass undergoing dry deposition in Chicago were
18    smaller than 2.5 jim in diameter (Lin et al., 1993). Additionally, more than 90% of lead particle
19    mass that undergoes dry deposition is in an insoluble chemical form (Gatz and Chu, 1986).
20          Deposition velocities for lead are in the range of 0.05-1.3 cm/s. Table 2-19 is a
21    compilation of data from the literature. Figure 2-2 shows the variation of deposition velocity for
22    lead with particle size.
23
24    Wet Deposition
25          Wet deposition is the process by which airborne pollutants are scavenged by precipitation
26    and removed from the atmosphere.  The flux of a depositing species can be defined through the
27    following equation:
28
29                                         F=VPCP                                       (2-6)
30
31    where Vp is the rate of precipitation in cm/s and Cp is the concentration of the chemical species
32    in the precipitation (|ig/L) (Miller and Friedland, 1994).

      December 2005                            2-48        DRAFT-DO NOT QUOTE OR CITE

-------
Decembe
to
o






to
\D




\J

%
H
6
o
o
H
/O
o
w
o
n
H H
w
k-LJ
Table 2-19. Deposition Velocities for Lead Particles. Data taken from Davidson and Rabinowitz (1992), Rojas et al. (1993),
Friedlander et al. (1986), Davidson and Wu (1990), Davidson et al. (1985), and El-Shobokshy (1985).
Vd (cm/s)
0.26
0.56
0.06 ± 0.02

0.06 ± 0.02
0.09 ±0.03
0.26
0.14±0.13
0.15 ±0.07
0.41
0.43
0.19
0.33 ±0.03
0.31 ±0.02

0.37 ±0.04
0.31 ±0.02

0.28 ±0.05
0.34 ±0.05
0.9 ±0.3
1.3 ±0.5
0.05
0.005
0.06 ±0.01
0.46
0.06
0.13
MMAD (jim)
all
all
all

all
all
all
10% >4
0.87
0.68
0.75
0.70








0.79
0.79
0.5
0.5
0.5

0.3
82%<1
Surface
water
orchard grass




all
teflon plates
teflon plates
water
water
land
alfalfa + oil
alfalfa + oil

grass + oil
grass + oil

soil
soil
beech canopy
spruce canopy
polyethylene petri dish
oak
polyethylene petri dish
filter paper
bucket
water
Other


model of Rojas et al.(1993)
model of Slinn & Slinn
(1980)
model of Williams (1982)
mass balance model





stable conditions
unstable conditions

stable conditions
unstable conditions

stable conditions
unstable conditions
throughfall
throughfall

foliar extraction



aerometric mass balance
Author
Davidson & Rabinowitz (1992)
Davidson & Rabinowitz (1992)
Rojas etal. (1993)

Rojas etal. (1993)
Rojas etal. (1993)
Friedlander etal. (1986)
Davidson and Wu (1990)
Davidson etal. (1985)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)
El-Shobokshy (1985)
El-Shobokshy (1985)

El-Shobokshy (1985)
El-Shobokshy (1985)

El-Shobokshy (1985)
El-Shobokshy (1985)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)
Davidson and Wu (1990)










(taken from Dedeurwaerder et al. (1983)
(taken from Dedeurwaerder et al. (1983)
(taken from Dedeurwaerder et al. (1983)








(taken from Hofken et al. (1983)
(taken from Hofken et al. (1983)
(taken from Lindberg and Harriss (1981)
(taken from Lindberg and Harriss (1981)
(taken from Lindberg and Harriss (1981)
(taken from Pattenden et al. (1982)
(taken from Rohbock (1982)
(taken from Sievering et al. (1979)

-------
                                           Deposition Velocity
                          2000
                          1000
                     E
                    '5   0.100
                    .2
                     o
                     c
                    JB   0.010
                    "55
                     o
                     Q.
                     O
                    Q
                         0.001
v\
       V

'   Density = 6 g cm"3
                             0.01          0.10            1.00         8.00
                                     Mean Stokes Diameter (|jm)
     Figure 2-2.   The deposition velocity plotted against the geometric mean Stokes diameter
                  for particles with a density of 6 g/cm 3 (i.e., lead).  Error bars are shown and
                  the arrow indicates a negative value for the lower limit of uncertainty.
     Source: Reprinted from Main and Friedlander (1990).
 1         The size of particles can influence rates of wet deposition. Large particles are scavenged
 2   more efficiently.  Lead, which is primarily in the submicron size range, does not undergo wet
 3   deposition as easily as many of the crustal elements (Davidson and Rabinowitz, 1992).
 4         Conko et al. (2004) note a seasonal trend in wet deposition rates of lead.  The highest
 5   concentrations were observed in the summer months, which the authors attribute to increased
 6   emissions from electric power plants. This contradicts the findings of Gelinas and Schmit (1998)
 7   who found the lowest deposition of lead occurred in the summer. The authors suspect that this is
 8   due to decreased traffic in the summer months.
 9         Regional differences may affect deposition rates. Miller and Freidland (1994) observed
10   wet deposition fluxes at an  elevation of 1200 m that were almost twice as high as fluxes
11   observed at low altitudes. This effect is observed because wet deposition of lead occurs almost
12   exclusively through rainfall at low elevations, but cloudwater interception is an important factor
     December 2005
     2-50
      DRAFT-DO NOT QUOTE OR CITE

-------
 1   at high elevations in addition to rainfall.  Comparing values for urban and rural sites, similar
 2   rates of wet deposition were observed indicating that lead is widely disbursed and is emitted by
 3   area sources (Conko et al., 1994).
 4          Precipitation activity has been linked to variability in wet deposition rates. Intense rain
 5   showers had lower lead concentrations than slow, even rainfalls (Chow, 1978). Thunderstorms
 6   typically did not have detectable quantities of lead but occasionally produced very high levels.
 7          The concentration of lead in rainfall does not appear to be correlated to  the amount of
 8   time between rainfalls but meteorological conditions such as a thermal inversion preceding a
 9   rainfall may affect the lead content (Chow, 1978).
10          Lead in rainwater includes both dissolved and particulate material. Approximately 83%
11   of lead in wet deposition  samples was in a soluble form, compared to less than 10% in dry
12   deposition  samples (Gatz and Chu, 1986).
13          Typical concentrations of lead in precipitation are listed in Table 2-20.  The table shows a
14   pronounced downward trend with time presumably following the phase out of leaded fuel.
                Table 2-20.  Concentrations of Lead in Rainwater in the United States
Precipitation
Dates of concentration
Testing (jig/L)
1966-1967
1971-1972
1975-1976
1977-1978
1982
pre-1982
1988-1989
1998
32.7
31.2
25.2
15.6
17.0
19.7
1.9
0.47
Cloudwater
concentration
(jig/L) Source
98.1
93.6
75.6
46.8
51
n.a.
5.4
n.a.
Miller & Friedland (1994)
Miller & Friedland (1994)
Miller & Friedland (1994)
Miller & Friedland (1994)
Miller & Friedland (1994)
Davidson & Rabinowitz
(1992)
Miller & Friedland (1994)
Conko et al. (2004)
(taken from Lazrus et al.,
1970)
(taken from Schlesinger &
Reiners, 1974)
(taken from Smith and
Siccama, 1981)
(taken from Smith and
Siccama, 1981)
(taken from Scherbatskoy
and Bliss, 1984)
(taken from Galloway
etal., 1982)
(taken from Miller and
Friedland, 1991)

       Source: Miller and Friedland (1994), Davidson and Rabinowitz (1992), and Conko et al. (2004).
     December 2005
2-51
DRAFT-DO NOT QUOTE OR CITE

-------
 1   Bulk Deposition
 2          Bulk deposition is the rate of dry and wet deposition combined. It is typically sampled in
 3   open buckets or other open containers. This is often used to estimate the overall rate of
 4   atmospheric input to soil, surface water, or other terrestrial media. However, it is understood
 5   that dry deposition onto surrogate surfaces may differ greatly from dry deposition onto natural
 6   surfaces.  The ratio of dry to wet deposition is 1.5, 0.4, and 0.25 in marine, rural, and urban areas
 7   respectively (Davidson and Rabinowitz, 1992). The ratio of dry deposition to wet deposition
 8   ranged between 0.1 and 0.5 in arctic regions (Davidson and Rabinowitz,  1992). In a literature
 9   survey, Hicks (1986) found that this ratio varied between 0.4 and 1.8.
10
11   2.3.3    Resuspension of Lead-Containing Soil and Dust Particles
12          The resuspension of soil-bound lead particles and contaminated road dust is a significant
13   source of airborne lead. Here we focus on resuspension by wind and vehicular traffic although
14   resuspension through other mechanical processes such as pedestrian traffic, agricultural
15   operations, construction, and even raindrop impaction is possible.  In general, mechanical
16   stresses are more effective at resuspending particles than wind (Sehmel, 1980; Nicholson,
17   1988b).  We begin with a discussion of resuspension by natural winds, then present information
18   on the effect of vehicular traffic on resuspension.
19          Understanding the physics of resuspension requires analyzing the wind stresses on
20   individual particles including frictional drag, form drag, gravitation, and the Bernoulli effect
21   (Sehmel,  1980). Although this analysis can be accurate on a small scale, predicting resuspension
22   on a large scale generally focuses on empirical data for continual soil movement due to three
23   processes: saltation, surface creep, and suspension (Sehmel, 1980; Nicholson, 1988b).  Saltation
24   is the process by which particles in the 100-500 jim size range bounce or jump close to the
25   surface.  The low angle at which these particles strike  the surface can transfer momentum to
26   smaller particles allowing them to be suspended into the atmosphere (Sehmel, 1980; Nicholson,
27   1988b).  Depending on soil conditions, saltation can be responsible for moving 50-75% of
28   surface particles.  Surface creep is the rolling or sliding motion of particles, which is induced by
29   wind stress or momentum exchanged from other moving particles. This generally applies to
30   large particles 500-1000 jim in diameter and moves 5-25% of soil by weight (Sehmel, 1980;
31   Nicholson,  1988b). Suspension is the process that actually ejects particles into the air.  This

     December 2005                        2-52           DRAFT-DO NOT QUOTE OR CITE

-------
 1    affects particles smaller than 100 jim in diameter and moves 3-40% of soil by weight (Sehmel,
 2    1980; Nicholson, 1988b).
 3          Resuspension is often defined in terms of a resuspension factor, K, with units of m"1, or a
 4    resuspension rate (A), with units of sec"1. The resuspension factor was used in early research on
 5    reentrainment and is defined by:
 7                                    K=  mr^°   Y                                    (2-7)
 8
 9    where Ca;r is the airborne concentration of a chemical species and Csoii is the surface soil
10    concentration of the same species.  K has significant limitations in that it is dependent both on
11    the height at which Cair is measured and the depth to which Csoii is measured.  This factor also
12    assumes that all airborne material is a direct result of resuspended soil-bound material, which is
13    not the case in most situations (Sehmel, 1980; Nicholson, 1988b). Additionally, K cannot be
14    used if soil concentrations are not uniform across the area of interest (Nicholson, 1988b).
15          The resuspension rate, A, is the fraction of a surface contaminant that is released per time
16    and is defined by:
17
18                                    A= R(^g/m s)                                    (2-8)
19
20    where R is the upward resuspension flux, and A has units of s"1.  Although A is also dependent
21    on the depth to which soil concentrations are measured, the resuspension rate has a number of
22    advantages over K. Most notably, it can be applied to non-uniform areas of soil contamination,
23    and it allows for other sources of airborne contaminants. It cannot be determined experimentally
24    and is usually deduced by fitting results to a numerical model of airborne dispersion and
25    deposition for the pollutant of interest (Nicholson, 1988b). Resuspension rates are dependent on
26    many factors including wind speed, soil moisture, particle sizes,  the presence of saltating
27    particles, and the presence of vegetation.  Typical values for A can cover 9 orders of magnitude
28    in the range of 10"12-10"4 s"1 (Sehmel, 1980; Nicholson,  1988b).
29          Nicholson (1993) notes that A increases with increasing particle diameter because larger
30    particles protrude farter into the turbulent air stream and the drag force increases more quickly

      December 2005                        2-53            DRAFT-DO NOT QUOTE OR CITE

-------
 1    than the adhesive force. Furthermore, in a laboratory resuspension chamber, the yields of
 2    resuspended matter decreased approximately linearly with increases in the geometric mean
 3    particle sizes of the bulk soil (Young et al., 2002). Lead is associated with the smaller size
 4    ranges in the distribution of soil particles.  Young et al. (2002) suggest that this is because the
 5    higher specific surface area of small particles means that there are higher contents of organic
 6    matter or Fe/Al oxides that serve as lead binding sites.
 7          Saltation is a particularly important factor in determining resuspension rates.  Saltation
 8    moves large quantities of soil particles and is highly efficient at ejecting particles into the
 9    airstream.  Saltating particles rotate between 200 and 1000 revolutions/second and are ejected
10    almost vertically (Sehmel,  1980). Saltating particles strike the surface at very small angles -
11    almost horizontally - and cause an avalanching effect. In the absence of saltation, very little
12    resuspension would occur at all (Sehmel, 1980; Nicholson, 1993).  Because resuspension is
13    driven by saltation and not the direct pick-up by wind, the size distribution of resuspended
14    particles does not change with windspeed (Young et al., 2002).
15          Vehicular resuspension is the result either of shearing stress of the tires or turbulence
16    generated by the passing vehicle (Nicholson, 1988b; Nicholson et al., 1989). This process can be
17    particularly important since the most contaminated roadways tend to have the most traffic.
18    As with wind resuspension, a number of factors can  affect the rate of resuspension from
19    vehicular motion.  These factors include vehicle size, vehicle speed, moisture, and particle size.
20          Lead in street dust appears to have a bimodal  distribution. The fine mode is likely from
21    vapor phase condensation from combustion engines, while the coarse mode is from either vehicle
22    wear or significant coagulation of smaller particles.  Al-Chalabi and Hawker (1997) observed
23    that in roadways with significant resuspension, lead  concentrations were lower indicating either
24    dispersion from the source or the scavenging of smaller lead particles by coarser particles.
25    Abu-Allaban et al. (2003) similarly observed that lead in road dust tended to be in the coarse
26    mode. Measurements performed in tunnel tests indicated that less than 17% of PMio lead was
27    smaller than 2.5 |im (Lough et al., 2005).
28          Resuspension may occur as a series of events. Short  episodes of high windspeeds, dry
29    conditions, and other factors conducive to  resuspension may dominate annual averages of
30    upward flux (Nicholson, 1988b, 1993).
      December 2005                        2-54            DRAFT-DO NOT QUOTE OR CITE

-------
1
2
3
4
5
6
The concentrations of lead in
soils sampled near
Pb/kg of bulk soil
suspended soil and dust vary significantly. In suspended
industrial emitters of lead, PMio-bound lead varied between 0.012 and 1.2 mg
(Young et al., 2002). Tsai and Wu (1995) measured lead
that was 30 times higher than lead in
other pollutants, which may indicate
contaminants. The
fractions of lead
road dust. This enrichment factor was
in airborne particles
much higher than for
that lead is more easily resuspended than other
in suspended dusts and soils are listed in Table 2-21.
Table 2-21. The Percentage of Lead in Resuspended Particulate Matter

















Source
Paved road dust
Paved road dust
Paved road dust
Paved road dust
Unpaved road dust
Unpaved road dust
Unpaved road dust
Unpaved road dust
Agricultural soil
Agricultural soil
Agricultural soil
Agricultural soil
Agricultural soil
Agricultural soil
Playa dust
Sand & gravel
storage
Construction site
Location
urban
urban
urban
rural
rural
rural
residential
staging area






rural

urban
Pb fraction of Pb fraction of
PM10 mass (%) PM2.5 mass (%)
0.0161 ±0.0031
0.3 ±0.03 0.4
l.E-02
0.0057 ± 0.0028
0.0058 ± 0.0073
0.01
0.0203 ±0.0133
0.0043 ± 0.0008
0.0063 ± 0.0059
0.0031 ±0.0025
0.0062 ± 0.0034
0.0024 ± 0.0082
0.003 ± 0.0025
0.01
l.E-03
0.02
l.E-03
Reference
Chow et al. (2003)
Chow etal. (1994)
Gillies etal. (1999)
Chow et al. (2003)
Chow et al. (2003)
Chow etal. (1994)
Chow et al. (2003)
Chow et al. (2003)
Chow et al. (2003)
Chow et al. (2003)
Chow et al. (2003)
Chow et al. (2003)
Chow et al. (2003)
Chow etal. (1994)
Gillies etal. (1999)
Chow etal. (1994)
Gillies etal. (1999)
 Source:  Chow et al. (1994, 2003) and Gillies et al. (1999).
December 2005
2-55
DRAFT-DO NOT QUOTE OR CITE

-------
 1         The contribution of resuspended soil and dust to the airborne burden may be significant.
 2   A source apportionment study in Boston indicated that soil resuspension increased the airborne
 3   concentration of lead by as much as 0.022 |ig/m3 in the fine mode (Thurston and Spengler,
 4   1985). Isotopic measurements in Yerevan, Armenia credited resuspension of contaminated soil
 5   with 75% of the atmospheric lead in 1998 (Kurkjian et al., 2002). Calculations based on road
 6   dust emissions and lead weight fractions indicate that resuspension was responsible for -40% of
 7   overall lead emissions to the South Coast Air Basin of California in 1989 (Lankey et al., 1998).
 8   Resuspension estimates based on modeling efforts for the same area suggest that resuspension
 9   contributed -90% of overall lead emissions in 2001 (Harris and Davidson, 2005).  Figures 2-3
10   and 2-4 demonstrate how air and soil concentrations are affected by long-term resuspension.
                 ^   0.010
                 1   0.009
                 "B)   0.008
                  c
                  0
                  e
                  (1)
                  o
                  c
                  0
                 o
                  0
                 0
                      0
                      0.000
1970    1975    1980
                                                   1985
                                                   Year
 1990    1995    2000
     Figure 2-3.   The modeled soil concentrations of lead in the South Coast Air Basin of
                  California based on three resuspension rates.
     Source: Reprinted from Harris and Davidson (2005).
     December 2005
                 2-56
DRAFT-DO NOT QUOTE OR CITE

-------
             c
             o
               _
            E  S
            8  *
            C  O)
            o  o
             S  E
Symbol
•*-
•e-
-H
Calc. of Airborne Pb Cone.
A=1x10-10/s
A=1x10'11/s
Measured
                    .014
                      1970    1975     1980    1985     1990     1995     2000
                                                  Year
     Figure 2-4.  The modeled and measured airborne concentrations of lead in the South
                 Coast Air Basin of California based on two resuspension rates.
     Source: reprinted from Harris and Davidson (2005).
 1   2.3.4   Runoff from Impervious Surfaces
 2         The runoff of water from impervious surfaces may be a significant transport route for lead
 3   from urban areas to soil, waterways, and catchment basins.  As water runs off roadways and
 4   buildings, it can become laden with dissolved and suspended matter. Dust on roadways contains
 5   a significant fraction of lead due to vehicle wear, vehicle emissions, road wear, fluid leakage,
 6   and atmospheric deposition. Lead in road dust is discussed in further detail in Sections 2.3.3 and
 7   3.2 of this document.  Additionally, lead-containing paints,  gutters, roofing materials, and other
 8   housing materials may leach with rainfall.
 9         Urban catchments in Fresno, California had highly elevated soil lead concentrations,
10   which probably indicates high concentrations of lead in runoff waters (Nightengale, 1987).
11   Basins in use since  1962, 1965,  and 1969 had surface soil concentrations of 570, 670, and
12   1400 ppm, respectively. Nearby control soils had surface concentrations between 8.3 and
13   107.0 ppm.
     December 2005
2-57
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Urban runoff released into a stream in State College, Pennsylvania caused significant
 2    spikes in lead concentrations (Lieb and Carline, 2000). Concentrations upstream of the release
 3    point were 1.50 |ig/L. Downstream concentrations were 1.80 |ig/L when there was no
 4    precipitation and averaged 14.60 |ig/L during storm events.
 5          The amount of lead that is removed from roadways and buildings by rainwater depends
 6    somewhat on the intensity of the storm.  Experiments performed by Davis and Burns (1999)
 7    indicated that high intensity storms washed away significantly more exterior house paint than
 8    low-intensity storms. A separate set of experiments showed that the amount of lead contained in
 9    roadway runoff increased significantly with the length of the dry period prior to a rain event
10    (Hewitt and Rashed,  1992).
11          Lead in runoff water is primarily in the particulate form with a very small fraction in the
12    dissolved  form (Hewitt and Rashed, 1992; Davis  and Burns, 1999; Roger et al., 1998). Between
13    69% and 93% of lead washed from painted structures was in parti culate form (Davis and Burns,
14    1999). More than 90% of lead in highway runoff from a rural highway in the UK was in the
15    particulate phase (Hewitt and Rashed, 1992). Roger et al.  (1998) observed that lead particles in a
16    motorway catchment in France were typically smaller than 50 jim in diameter. Samples  taken
17    from road water samples also in France showed that most lead was in an inorganic, non-
18    bioavailable form (Flores-Rodriguez et al., 1994).
19          The amount of lead from roadways varies by region, the rainfall intensity, maximum
20    inflow, rainfall duration, and the antecedent dry weather period (Shinya et al., 2000).
21    Measurements taken near a roadway in France  showed that in runoff water, concentrations
22    ranged between 0.46 and 4.57 g Pb/kg of suspended particulate matter (Roger et al., 1998).
23    Another study of French roadways had an average lead content of 2.36 g Pb/kg of dried material
24    (Flores-Rodriguez et al., 1994). Thirteen storm events studied at a heavily trafficked, rural
25    highway in England showed mean lead contents of 181 |ig/L (Hewitt  and Rashed, 1992).  Of this
26    total,  16.2 ± 6.9 jig/L was in the dissolved phase and 165 ± 101  jig/L  was in the particulate
27    phase. An additional 0.36 |ig/L was in an organic form. The mean concentrations of lead during
28    four rain events studied near a roadway in Japan ranged between 17 and 39 jig/L (Shinya et al.,
29    2000). The initial concentrations were higher,  ranging from 130 to 567 |ig/L. This indicates the
30    presence of a first flush effect in which much of the contamination is removed within the initial
31    period of rainfall. Hewitt and Rashed (1992) observed a similar downward trend in the lead

      December 2005                        2-58          DRAFT-DO NOT QUOTE OR CITE

-------
 1   concentration with time. However, no first flush phenomenon was observed in the study
 2   performed by Taebi and Droste (2004), who looked at combined urban runoff transported to a
 3   mixed residential and commercial urban catchment in Iran.  The concentrations of lead for each
 4   of 10 major rainfall events ranged between 0.018 and 0.558 |ig/L. The arithmetic mean for all
 5   10 events was 0.278 |ig/L.
 6         Studies of runoff from building materials showed high lead concentrations from painted
 7   wood and painted brick, particularly if the paint is more than 10 years old (Davis and Burns,
 8   1999; Davis et al., 2001).  The maximum concentrations of lead were 1900 |ig/L and 28000 |ig/L
 9   from wood and brick with painted exterior surfaces, respectively (Davis and Burns, 1999). Lead
10   from paint is released into waters in both particulate and dissolved form. The concentrations in
11   runoff from building surfaces are listed in Table 2-22.
             Table 2-22.  The Concentrations of Lead in Runoff From Building Surfaces
Substance
Block (painted)
Brick (painted)
Wood (painted)
0-5 yr. old paint
5-10 yr. old paint
>10yr. old paint
Roofs
Residential roofs
Commercial roofs
Institutional roofs
Geometric
Mean
(US/L)
9.2
22
43
8.0
18
81
6.0



Median
(US/L)
8.0
16
49
8.1
14
88
5.2
2
12
64
Mean
US/L)
38
580
170
27
120
810
38
1.5
62
64
Range
(ug/L)
<2-590
<2-28000
<2-1900
<2-370
<2-2600
<2-28000
<2-590



Reference
Davis and Burns (1999)
Davis and Burns (1999)
Davis and Burns (1999)
Davis and Burns (1999)
Davis and Burns (1999)
Davis and Burns (1999)
Davis and Burns (1999)
Davis etal. (2001)
Davis etal. (2001)
Davis etal. (2001)
      Source: Davis and Burns (1999) and Davis et al. (2001).
     December 2005
2-59
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Matthes et al. (2002) studied runoff from lead sheet to simulate lead in gutters, roofs,
 2    piping, siding, and sculptures.  Typical concentrations in runoff ranged between 700 and
 3    3700 mg/L. This was attributed to the solubility of cerrusite (lead carbonate) and hydrocerrusite
 4    (lead hydroxy carbonate), which form on the surface of air-exposed lead.
 5          The amount of lead removed by runoff events varies. Hewitt and Rashed (1992) estimate
 6    that approximately 8% of lead and 5% of organic lead emitted from vehicles is removed by
 7    highway drainage waters. Shinya et al. (2000) estimate that total lead loads for a roadway in
 8    Japan prior to four storm events ranged between 0.053 and 0.771 mg Pb/m2. These storm events
 9    removed half of the load in 0.07 to 3.18 hours after the start of the rainfall event.
10          Davis et al. (2001) estimate the total annual loading of lead from all sources to be between
11    0.069 and 0.18 kg Pb/ha. They estimate that 80-90% of this is derived from runoff from
12    buildings.
13          Lead corrosion (cerrusite and hydrocerrusite) dissolution rates from lead sheets were
14    measured at 14.3  - 19.6 millimoles of lead/m2 per year (Matthes et al., 2002).
15
16    2.3.5   Leaching of Soil Lead
17          Soil lead has some capacity to leach through the soil column, potentially contaminating
18    ground water. Lead sorbs strongly to constituents of the soil matrix and is only weakly soluble
19    in pore water, so the leaching of lead is a much  slower process than the leaching of many other
20    contaminants (Marcos et al., 2002; Zhang and Xu,  2003; Unlii, 1998; Pang et al., 2002).  The
21    sorbing capacity of the soil and the solubility of the contaminants can be affected by the
22    hydraulic conductivity of the soil, the composition of the soil solution, the content of the soil
23    organic matter, the content of the soil clay minerals, soil pH, microbial activity, preferential flow
24    through plant root channels and animal holes, and geochemical reactions (Rhue et al., 1992;
25    Elzahabi and Yong, 2001). The experiments of Erel  et al. (1997) on soil columns indicate that
26    anthropogenic lead is more readily available for leaching than lead that naturally occurs in
27    the soil.
28          Lead can bind to many different surfaces in the heterogeneous soil matrix. This
29    adsorption greatly affects mobility and is dependent on the characteristics of the soil and lead
30    compounds.  Lead is partitioned between the soil water solution, precipitated forms, secondary
31    Fe or Mn oxides,  carbonates, organic matter, sulfides, or the surfaces of clay, humus, or silicate

      December 2005                        2-60           DRAFT-DO NOT QUOTE OR CITE

-------
 1    particles (Badawy et al., 2002; Venditti et al., 2000; Cajuste et al., 2000; Erel and Patterson,
 2    1994).  The most labile fraction of lead is adsorbed to the surfaces of colloid soil particles, which
 3    may include organic matter, clay, oxides, or carbonates (Erel et al., 1997). Lead that was leached
 4    from a limestone soil during a sequential fractionation procedure was exclusively in the
 5    iron/manganese oxide form (Hee, 1994). A study of industrially  contaminated soils found that
 6    between approximately 50% and 60% of the lead was not susceptible to leaching during any
 7    phase of a sequential fractionation procedure (Cajuste et al., 2000). The remaining lead was
 8    found primarily in the carbonate and Fe-Mn oxide fractions with  sizeable amounts in the organic
 9    and exchangeable phases. None of the lead was water soluble. Maskall and Thornton (1998)
10    also observed a high fraction of lead in the carbonate form in highly contaminated soil. The
11    unusual presence of carbonate-bound lead is probably due to the formation of cerrusite (PbCO3)
12    in soils contaminated with calcareous slag wastes (Maskall and Thornton, 1998).  Lead migration
13    in this contaminated soil was associated with Fe-Mn oxides. A third contaminated site was
14    tested by Jing et al. (2004). These soils showed 57% of lead in the Fe-Mn oxide form, 29% in
15    the carbonate form, and just 5% in the residual, soil-bound form.
16          A high chlorine content in the soil has been shown to increase lead leaching (Urdu, 1998).
17    Chloride complexation with lead enhances lead solubility.
18          The pore-water velocity is inversely proportional to sorption rates.  At low flow, the
19    longer retention times lead to more complete sorption of lead to soil particles (Pang et al., 2002).
20          In laboratory experiments on soil columns, transport of lead was enhanced by the
21    introduction of soil colloid suspensions (Karathanasis, 2000).  Colloids increased transport of not
22    only colloid-bound lead but also dissolved lead.  Colloid transport was enhanced by increasing
23    the colloid surface charge, increasing the pH, increasing the amount of organic carbon,
24    increasing the soil macroposity, decreasing the colloid size, and decreasing the Al, Fe, and quartz
25    contents (Karathanasis, 2000). Colloid binding and co-transport of lead are important
26    mechanisms for lead migration but colloids also enhance the flow of lead through physical
27    blockage from exchange sites, competitive sorption, and organic  complexation (Karathanasis,
28    2000).  Denaix et al. (2001) observed that most of the lead-transporting colloids in an acidic,
29    loamy soil were biological in nature. The lead concentration in the colloid fraction was not
30    correlated with pH, colloidal organic carbon contents, or colloidal silicon concentrations (Denaix
      December 2005                        2-61           DRAFT-DO NOT QUOTE OR CITE

-------
 1    et al., 2001). Approximately 50% of the total lead transfer in these experiments was attributed to
 2    colloidal transfer.
 3          At low pH, metal species bound to carbonates, hydroxides, and other soil matrix
 4    components are more likely to dissolve into solution (Maskall and Thornton, 1998; Elzahabi and
 5    Yong, 2001; Badawy et al., 2002).  This  increases the rate of lead migration through the soil.
 6    The experiments of Jing et al. (2004), which follow eight different leaching protocols, suggest
 7    that pH is the primary factor in determining the concentration of lead in leached solution.  At pH
 8    >12, lead forms soluble hydroxide anion complexes and leaches out of the soil column. At pH
 9    between 6 and 12, lead leachibility is low due to adsorption and precipitation.  At pH <6 free Pb
10    ions leach into the pore water and are removed from the soil columns. Rhue et al. (1992)
11    observed that organic lead species Me2Pb2+ and Et2Pb2+ were absorbed best at pHs of 6.2 and
12    7.2, respectively.  Sorption decreased at pH <5 and >8.2 (Rhue et al., 1992).
13          A partition coefficient, Kd, is often used to describe the susceptibility of lead to leaching.
14    This value is used to compare the fractionation of a contaminant between liquid and solid forms.
15    Kd is defined by the following equation:
16
17                                       Kd = S/C'                                      (2-9)
18
19    where S is the total concentration of lead adsorbed in the solid phase, and C' is the concentration
20    of lead in pore water solution (Elzahabi and Yong, 2001). Kd increases with increasing pH
21    (up to 7.0) and increasing distance from the leachate source (Elzahabi and Yong, 2001; Sheppard
22    and Sheppard,  1991). Kd decreases with an increase in the influent heavy metal concentration
23    and the degree of saturation (Elzahabi  and Yong, 2001).  The highest value of Kd appears to be
24    near the source of lead contamination.  Values of Kd in the literature cover many orders of
25    magnitude between 1.20 L/kg and "infinity" (when no lead  can be detected in pore water).
26    These values are listed in Table 2-23.
27          The rate of migration through the  soil has been estimated in many different studies. Using
28    lead isotopes, Erel et al. (1997) estimate  the rate of lead migration to be 0.5 cm/year in soils
29    collected from rural locations in Israel. Sheppard and Sheppard (1991) measured the rate of flow
30    through spiked soils, which were highly  acidic and had a low organic matter content. These
31    soils, which were especially susceptible to leaching, exhibited migration rates of 0.3 cm/day
      December 2005                        2-62           DRAFT-DO NOT QUOTE OR CITE

-------
      Table 2-23. Soil/Water Partition Coefficients for Several Different Soils and Conditions
Kd
(L/kg)
12.68-w
3.23-oo
1.20-00
1.36-00
-6000
-3000
-5000
20
9000
92.99
14.25
125.58
95.51
1330±200
pH
4.0
4.0
3.5
3.5
n.a.
n.a.
n.a.
4.9
4.8
4.45
4.45
5.01
5.01
3.0-4.0
Beginning Soil
Water Content
(%)
26.69
28.20
26.29
26.32
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
Soil Type
illitic (spiked)
illitic (spiked)
illitic (spiked)
illitic (spiked)
brown pseudopodzolic
rendzina
gley podzolic
acidic (low-organic-matter sand)
sphagnum peat
mining site
mining site
mining site
mining site
acidic (high-organic-matter peat)
Reference
Elzahabi and Yong (2001)
Elzahabi and Yong (2001)
Elzahabi and Yong (2001)
Elzahabi and Yong (2001)
Alumaa et al. (2002)
Alumaa et al. (2002)
Alumaa et al. (2002)
Sheppard and Sheppard (1991)
Sheppard and Sheppard (1991)
Merrington and Alloway
(1994)
Merrington and Alloway
(1994)
Merrington and Alloway
(1994)
Merrington and Alloway
(1994)
Deiss et al. (2004)
      Source: Elzahabi and Yong (2001), Alumaa et al. (2002), Sheppard and Sheppard (1991), Merrington and
      Alloway (1994), and Deiss et al. (2004).
1   during the first year of experiments.  The migration rate appeared to slow down in subsequent
2   years.  Cores taken at smelting sites used during the Roman era, medieval times, and the
3   18th century underwent sequential extraction (Maskall and Thornton, 1998).  The estimated lead
4   migration rates at the Roman, medieval, and 18th century sites were 0.07-0.54 cm/year,
5   0.31-1.44 cm/year, and 0.11-1.48 cm/year, respectively.
6          Mass balance calculations of Miller and Friedland (1994) suggest migration rates of
7   0.11 cm/year and 0.29 cm/year through the organic horizons of spruce-fir and northern hardwood
8   forests, respectively. Similar calculations by Kaste et al. (2003) at the same site predicted that
9   anthropogenic lead will take -60 and -150 years to be transported through the organic horizon in
    December 2005
2-63
DRAFT-DO NOT QUOTE OR CITE

-------
 1    the deciduous and spruce-fir forests, respectively. The difference in response times for the two
 2    forests may be due to differences in the litter depth and/or in the rate of litter decomposition.
 3    Soil tested from a car battery salvage facility showed a significantly greater lead concentration in
 4    the leached solution than in a reference soil (Jensen et al., 2000). Concentrations in the leached
 5    solution went as high as 8000 |ig/L. Other industrially contaminated soils did not show such
 6    high rates of leaching, but these other soils had nearly neutral pHs.
 7          Isotopic ratios in soil  cores in the Sierra Nevada, California showed that 21% of lead at a
 8    depth of 30 cm had anthropogenic origins and had migrated from the surface (Erel and Patterson,
 9    1994).  The remaining 79% of lead at this depth was naturally occurring.
10          Physical mixing of soils through animal activity may also increase the  rate  of lead
11    migration.  Mace et al. (1997) observed a significant decrease in lead transport time through soil
12    as a result of rodent activity in a southern California location.
13          Vilomet et al. (2003) used isotopes to trace the leaching of lead from a landfill into
14    groundwater in France.  The active landfill has been in use since 1900 and has no  bottom liner.
15    Detectable quantities of leached lead were observed as far as 4600 m downgradient (Vilomet
16    etal., 2003).
17
18    2.3.6   Transport in Aquatic Systems
19          Chemical, biological, and mechanical processes govern the cycling of lead  in aquatic
20    environments. Here we focus on the exchange between sediment and surface water, which is
21    affected by many different factors including salinity, the formation of organic complexes, redox
22    conditions, and pH (Arakel and Hongjun, 1992).
23          Lead enters surface waters  from a number of sources. Atmospheric deposition is the
24    largest source, but urban runoff and industrial discharge are also significant (Peltier et al., 2003;
25    Hagner, 2002; Perkins et al., 2000). As expected, concentrations in surface waters are highest
26    near sources of pollution.
27          The dispersal of lead in waterways is relatively quick. If lead is emitted into waterways as
28    a point source, water concentrations decrease rapidly downstream of the source (Rhoads and
29    Cahill,  1999; Hagner, 2002; Kurkjian et al., 2004; Peltier et al., 2003).  Lead is removed from the
30    water column through flushing, evaporation, or sedimentation (Schell and Barnes, 1986).
      December 2005                        2-64           DRAFT-DO NOT QUOTE OR CITE

-------
 1    Kurkjian et al. (2004) note that first order approximations of concentrations of non-conservative
 2    pollutants (such as lead) can be made by using the exponential decay curve:
 3
 4                                      C =  C0e"kx                                    (2-10)
 5
 6    where C is the pollutant concentration, C0 is the concentration at the source, x is the downstream
 7    distance from the source, and k is the decay rate in km"1.  For the Debed River in Armenia,
 8    Kurkjian et al. (2004) found that a decay rate of 0.57 km"1 provided the best fit to measured lead
 9    concentrations.
10          Metals in waterways are transported primarily as soluble chelates and ions, constituents of
11    particulate matter, or by adsorption onto suspended organic or inorganic colloids (Arakel and
12    Hongjun,  1992).  The last two are the most important for lead. The predominant chemical forms
13    of lead that interact with aqueous ecosystems are PbO  and PbCO3 (Schell and Barnes, 1986).
14    Lead is adsorbed  on colloids that are typically secondary clay minerals, Fe-Mn oxides or
15    hydroxides, or organic compounds (Arakel and Hongjun,  1992).  The concentration of lead
16    appears to increase with increasing salinity (Arakel and Hongjun, 1992).
17          Schell and Barnes (1986) describe water columns as "transient reservoirs" for pollutants.
18    They found mean residence times for lead in two lakes and a reservoir to be between 77 and
19    250 days, although it should be noted that residence times tend to be shorter in turbulent
20    waterways. Lead concentrations in water are attenuated by the presence of A1(OH)3
21    precipitation, which is responsible for an estimated 54% of total lead loss, and by the adsorption
22    of lead onto other particles which settle out of the water column, which makes up the other 46%
23    of lead loss (Kurkjian et al., 2004).  Schell and Barnes (1986) measured sedimentation rates for
24    anthropogenic lead, which ranged between 0.0360 g cm"2 a'1 and 0.0644 g cm"2 a~l.
25          The concentration of lead in sediment roughly follows the  concentration of lead in
26    overlying water (Kurkjian et al., 2004; Rhoads and Cahill, 1999). Thus lead concentrations in
27    sediment are highest near sources and decrease downstream.
28          Lead preferentially sorbs onto small particles rather than large particles. Small grain sizes
29    and the larger surface area per unit weight lead to greater potential for adsorption (Rhoads and
30    Cahill, 1999). Concentrations of metals increase approximately logarithmically with decreasing
31    particle size.

      December 2005                        2-65           DRAFT-DO NOT QUOTE OR CITE

-------
 1          Organic matter in sediment has a high capacity to accumulate trace elements.  High humic
 2    levels may lead to greater lead contamination in sediments (Rhoads and Cahill, 1999; Kiratli and
 3    Ergin, 1996).
 4          Sulfides are another potential source of lead adsorption. This is especially true under
 5    anoxic conditions (Kiratli and Ergin,  1996; Perkins et al., 2000).  An increase in the amount of
 6    sulfide in pore water was shown to decrease the dissolved concentration of lead (Peltier et al.,
 7    2003).
 8          Lead in sediment can also be sequestered on iron or manganese oxides (Peltier et al.,
 9    2003; Gallon et al., 2004; Schintu et al., 1991).  These forms may make lead susceptible to
10    recycling into the overlying water column (Schintu et al., 1991).
11          Lead appears to be relatively stable in sediment.  It has a very long residence time, and
12    many studies suggest that lead is not mobile in the sediment. However, many other studies
13    suggest that lead-containing particles can be remobilized into the water column (Ritson et al.,
14    1999; Steding et al., 2000; Hlavay et al., 2001; Kurkjian et al., 2004; Peltier et al., 2003; Gallon
15    et al., 2004). For example, Steding et al. (2000) observe that isotopic concentrations of lead in
16    the San Francisco Bay match those of leaded gasoline from the 1960s and 1970s,  which may
17    indicate that recontamination by sediment may be a significant source of lead to overlying
18    waters. Ritson et al.  (1999) similarly observed that there was a negligible reduction in  lead
19    concentrations in the San Francisco Bay despite the closing of a nearby lead smelter, the
20    implementation of municipal effluent controls, and the elimination of lead additives to  gasoline.
21    That concentrations have remained high may suggest recycling of sediment lead.  Similarly, in a
22    study of water concentrations in the North Sea, concentrations of lead did not decrease
23    significantly with the elimination of major sources (Hagner, 2002). This also may indicate
24    continued high rates  of atmospheric deposition or cycling of lead stored temporarily in sediment.
25          Modeling efforts of Gallon et al. (2004) indicate that processes that resuspend sediment
26    such as diffusion, bioturbation, and bioirrigation are  small compared to sedimentation of
27    colloidal particles. Kurkjian et al. (2004) suggest a correction factor for equation (9) to account
28    for the contribution of lead from sediment.
29
30                                    C  =  CoC^ + Is                                  (2-11)
31

      December 2005                        2-66           DRAFT-DO NOT QUOTE OR CITE

-------
 1    where Is is the amount of lead that is resuspended into the water column.  Depending on
 2    the region of the river under discussion, the authors extrapolated Is values in the range of
 3    1.3-2.8 jig Pb/L.
 4
 5    2.3.7   Plant Uptake
 6          Plants that take up lead can be a source of lead exposure for wildlife, livestock, and
 7    humans that consume contaminated plants.  A more thorough review of soil lead extraction by
 8    plants and subsequent effects on ecosystem health will be addressed in Chapter 8.
 9          Plants grown in soils contaminated by mine spoils (e.g., Cobb et al., 2000), smelting
10    operations (e.g., Barcan et al., 1998), sludge amendments (e.g., Dudka and Miller, 1999),
11    contaminated irrigation water (e.g., Al-Subu et al., 2003), or lead-containing agrochemicals (e.g.,
12    Azimi et al., 2004) will have higher than natural concentrations of lead. In general, higher
13    concentrations of lead in soils resulting increased lead levels in plants.
14          Although the transfer of soil lead to plants is generally small, all plants accumulate soil
15    lead to some degree (Finster et al., 2004). The rate of uptake is affected by plant species, soil
16    conditions, and lead species.
17          Of all the factors affecting uptake, pH is believed to have the strongest effect (Dudka and
18    Miller, 1999).  Acidic soils are more likely to have lead in solution and therefore available for
19    absorption. This is sometimes attenuated by liming.
20          Most lead in plants is stored in roots and very little is stored in fruits (e.g., Finster et al.,
21    2004; Cobb et al.,  2000). Of 33 edible plants grown in urban gardens, roots had a median
22    concentration that was 12% of the soil lead concentration (Finster et al., 2004).  Shoot lead, when
23    it was detectable, was just 27% of root lead.
24          Root vegetables seem the most prone to lead uptake followed by leafy vegetables (Dudka
25    and Miller, 1999; Finster et al.,  2004). Fruits and grains do not seem as susceptible to lead
26    contamination.
27          Metals that are applied as salts (usually as sulfate, chloride, or nitrate salt) are
28    accumulated more readily than the same quantity of metal added via sewage sludge, flue dust, or
29    fly ash (Dudka and Miller,  1999). This is likely because metal salts lead to the formation of
30    metal chloride  complexes and ion pairs, which can increase metal diffusion and subsequent plant
31    uptake.

      December 2005                        2-67           DRAFT-DO NOT QUOTE OR CITE

-------
 1    2.3.8   Routes of Exposure for Livestock and Wildlife
 2          There are many routes of exposure including food ingestion, drinking water, and
 3    inhalation for terrestrial organisms. For aquatic organisms, the main routes of exposure are food
 4    ingestion and water intake.  Thus, it is often difficult to determine the original source of an
 5    organism's lead burden. A few representative studies are summarized here, which have analyzed
 6    routes of lead exposure for non-human animals. For a discussion of health effects, toxicity, and
 7    lead concentrations in animal tissue, the reader is directed toward Chapters 8 and 9 of this
 8    document.
 9          Lead concentration of plants ingested by animals is primarily a result of atmospheric
10    deposition of lead particles onto plant surfaces rather than uptake of soil lead through plant roots
11    (Steinnes, 2001; Palacios et al., 2002; Dudka and Miller, 1999).  The uptake of lead by the
12    lowest trophic levels - invertebrates, phytoplankton, and krill for example - are some of the most
13    important avenues for introducing lead into food chains (Pilgrim and Hughes, 1994; Sanchez-
14    Hernandez, 2000; Hagner, 2002).
15          Some of the highest levels of lead exposure in animals occur near major sources like
16    smelters. In two studies of horses living near smelters, the estimated ingestion rate was in the
17    range of 2.4 to  99.5 mg Pb/kg body weight per day (Palacios et al., 2002) and 6.0 mg Pb/kg body
18    weight per day (Liu, 2003). Both exposure rates were well above the estimated fatal dose for
19    horses. Sheep grazing near smelters were similarly poisoned (Liu, 2003; Pilgrim and Hughes,
20    1994). Installation of pollution controls at a lead smelter in Slovenia greatly reduced the amount
21    of lead in nearby vegetation and the blood lead levels of cows  grazing on this vegetation
22    (Zadnik, 2004). Lead concentrations in topsoil at this site did  not decline in the 20 years since
23    the pollution controls were implemented.
24          The amount of lead entering the food chain depends highly on the species of the animal,
25    the species of their food, and where the organisms live. A study of sheep living in the
26    southernmost part of Norway, which is the most polluted part  of the country, showed a strong
27    correlation between liver lead concentrations and moss concentrations (Steinnes, 2001). The
28    sheep fed almost exclusively on a grass that picks up atmospherically deposited lead easily.
29    Correspondingly high levels were also observed in hare and black grouse in this region.
30    Similarly, a study of lead concentrations in raccoon tissues showed much higher concentrations
31    in urban raccoons that rural raccoons (Khan et al., 1995).  This may be because urban raccoons

      December 2005                       2-68           DRAFT-DO NOT QUOTE OR CITE

-------
 1    are exposed to higher air concentrations, ingest human refuse, or frequently visit storm sewers.
 2    In general, ruminant animals appear to be more resistant to lead ingestion than monogastric
 3    animals (Humphreys, 1991).
 4         Lead levels are somewhat elevated even in Antarctic animals (Sanchez-Hernandez, 2000).
 5    Antarctic food systems are supported by krill (Euphausia superba), which is the primary food
 6    source for organisms in higher trophic levels. Lead concentrations measured in E. superba were
 7    in the range of 0.17-12.0 ppm by dry weight. This is probably elevated above natural levels due
 8    anthropogenic input (Sanchez-Hernandez, 2000).
 9         Acute lead poisoning observed in Laysan albatross (Phoebastria immutabilis) chicks was
10    traced to the direct ingestion of paint chips by using isotopic analysis (Finkelstein et al., 2003).
11    Blood lead levels in P. immutabilis at the Midway Island National Wildlife Refuge had a
12    geometric mean of 190 |ig/dL. P. immutabilis chicks at a reference site had blood lead levels of
13    4.5 |ig/dL.
14         Contamination in mammals and fish livers was shown to be higher in highly polluted
15    coastal zones than in the open sea (Hagner, 2002). In foraminifers, which are meiobenthic
16    organisms, high sediment concentrations corresponded to high tissue concentrations. Sediment
17    concentrations were 10-20 times higher than foraminifer concentrations. Fish take in lead either
18    in their food or in water through their gills.  The relative importance of these two mechanisms
19    depends largely on the fish species.  A literature survey suggests that there has been no
20    observable decrease in fish muscle and liver concentrations in twenty years in marine or
21    freshwater environments (Hagner, 2002). Lead concentrations in the harbor porpoise (Phocoena
22   phocoend) appear to increase with the age of the animal. This was not true for the common seal
23    (Phoca vitulina) (Hagner, 2002). Shrimp (Palaemonetes varians) were shown to absorb 4-8% of
24    the lead content of its prey (Boisson et al., 2003).  Between 52% and 57% of the lead
25    accumulated from food was irreversibly retained in P. varians tissue. Just 2% of dissolved lead
26    accumulated from water was retained in tissue (Boisson et al., 2003).
27
28
29    2.4   METHODS FOR MEASURING ENVIRONMENTAL LEAD
30         The previous 1986 AQCD (U.S. Environmental Protection Agency,  1986) contained a
31    detailed review of sampling and analytical methods for lead in environmental media. Included in

      December 2005                        2-69           DRAFT-DO NOT QUOTE OR CITE

-------
 1    that document were discussions of site selection criteria, sampling methods, sample preparation,
 2    and analysis techniques. Furthermore, the document included discussion of sampling of lead
 3    emissions from mobile and stationary sources. In this section, we present a brief summary of
 4    sampling and analysis of lead.  For a more comprehensive discussion, the reader is referred to the
 5    1986 Lead AQCD.
 6          Emissions can be estimated from measurements at sources using grab samples, periodic
 7    samples, or continuous monitoring. Determining the rate of emissions requires knowing both the
 8    fluid flow rate and the concentration of lead in the fluid, usually air or water.  Thus it is much
 9    easier to measure emissions from stacks than it is to measure fugitive, diffuse, or nonpoint
10    emissions (Frey and Small, 2003).
11          Much of the recent improvement in measurement of lead emissions from sources is due to
12    better sampling and analytical equipment. For example, better dilution tunnels can provide
13    reliable samples from in-stack sampling, and improved analytical methods such as inductively
14    couple plasma mass  spectrometry  permit determination of lead at lower levels than in years past.
15    This means it is possible to obtain data from short sampling runs, permitting better time
16    resolution.
17          Ambient air sampling for lead generally uses filter media, such as  Teflon, or impactors
18    that fractionate the airborne particles into different size ranges. Collection of samples of water,
19    food, dust, or soil for lead determination can be performed in acid-washed containers. Wet
20    deposition can be collected using precipitation buckets that seal tightly immediately before and
21    after rain.  Dry deposition on land can be sampled using surrogate surfaces such as Teflon plates
22    (Davidson et al., 1985; Davidson and Wu, 1990), or alternatively by leaf-washing (Lindberg and
23    Lovett, 1985) or sampling throughfall precipitation that washes previously deposited lead  off the
24    vegetation and onto the forest floor (Wu et al., 1992b). Dry deposition onto bodies of water is
25    more difficult to estimate, usually  requiring airborne concentrations used with deposition
26    velocity estimates (Zufall and Davidson, 1997). Subsequent analysis of all of these samples can
27    be performed by atomic absorption spectrometry,  neutron activation analysis, x-ray fluorescence,
28    or proton-induced x-ray emission (Koutrakis and Sioutas, 1996), or by inductively-coupled
29    plasma mass spectrometry (ICP-MS) (U.S. Environmental Protection Agency, 1991).
30          Recently developed  single-particle instruments can identify which particles contain lead,
31    and what other elements are present in the same particle.  Information on the size of the particle

      December 2005                        2-70           DRAFT-DO NOT QUOTE OR CITE

-------
 1    is also provided (Pekney et al., 2006; Silva and Prather, 1997).  Although such instruments are
 2    not able to determine the precise mass of lead in each particle, they can provide valuable data on
 3    the characteristics of particles that contain lead from individual sources or source categories.
 4    Such "fingerprinting" methods can be used to identify sources of lead-containing particles in the
 5    environment.
 6
 7
 8    2.5    SUMMARY
 9          For most of the past 50-60 years, the primary use of Pb was as additives for gasoline.
10    Leaded gasoline use peaked in the 1970s, and worldwide consumption has declined since
11    (Nriagu, 1990).
12          Currently, the major use of Pb in the United States is in lead-acid batteries, for which the
13    demand is increasing (Socolow and Thomas, 1997). Other major uses are for glass, paints,
14    pigments, and ammunition. United States consumption of Pb is shown in Figure 2-5. The
15    consumption reached -1.4 million metric tons  per year in the mid 1990s (Socolow and Thomas,
16    1997).  Approximately 910,000 metric tons of this was secondary production, indicating high
17    rates of Pb recycling.
18          The largest source of Pb emissions was leaded gasoline throughout the 1970s and 1980s.
19    The largest emitters are now in the manufacturing sector, which includes lead-acid battery plants,
20    smelters, lead-alloy production facilities, and others (Harris and Davidson, 2005). These
21    emissions are not confined to the air —  approximately 90 facilities nationwide generate 90% of
22    the lead-containing solid hazardous waste (Chadha et al., 1998). Natural sources of Pb are
23    insignificant in comparison to anthropogenic sources.  Nationwide air emissions in 2000 were
24    estimated as 1885 metric tons from metals processing, 758 metric tons from incineration,
25    513 metric tons from transportation, primarily  from avgas-fueled aircraft, 439 metric tons from
26    fuel combustion for utility generation as well as industrial and commercial purposes, 198 metric
27    tons from Pb oxide and pigment production, and 48 metric tons from other processes (U.S.
28    Environmental Protection Agency, 2003).
29          Emission inventories for Pb have significant omissions and discrepancies (Harris et al.,
30    2005; Chadha et al., 1998). Thus, the data  above are probably a lower limit for Pb emissions.
31    Research into constructing detailed and  accurate databases of Pb emissions is needed.

      December 2005                        2-71           DRAFT-DO NOT QUOTE OR CITE

-------
                           1910
                               1990
                            Productivity:
                              CD Primary Production
                              CD Secondary Production
                              • Net Imports
        Sources:
          -•§-- Batteries
           -X-  Gasolines Additives
          -•0-- Oxides for Paints, Glass, etc.
          --0-- Ammunition
          -<>• Other Products
     Figure 2-5.  U.S. consumption of lead since 1910.
     Source: Socolow and Thomas (1997).


 1          The U.S. EPA Trends Report provides analysis of the available data on Pb emissions
 1   through the year 2002 (http://www.epa.gov/airtrends/lead2/html) (U.S. Environmental Protection
 3   Agency, 2003). Figure 2-6 shows the observed decline in estimated Pb emissions.
 4          Measurements conducted in any ecosystem worldwide show some level of lead
 5   contamination. Anthropogenic Pb  reaches these ecosystems through many possible transport
 6   routes, some of which are shown in Figure 2-7.
 7          Air is the major transport route for lead emissions.  Deposition of airborne pollutants to
 8   surfaces has been observed in the most remote places on Earth, including the Arctic and
 9   Antarctic.  Mass balance calculations performed on an agricultural plot in France indicate that
10   atmospheric deposition is the dominant source of lead to soil even when lead-containing
     December 2005
2-72
DRAFT-DO NOT QUOTE OR CITE

-------
             80,000
             60,000
          c
          o
          t:  40,000
          o
             20,000
                                  in 1985, EPA refined its methods for estimating emissions
                      | Transportation
                       ^ Industrial Processes
                      r~l Fuel Combustion
                             85
               1982-02: 93% decrease
               1993-02:  5% decrease
         92 93 94 95 96 97 98 99 00  01 02
       Year
     Figure 2-6.  Trends in U.S. air lead emissions, 1982-2002.
     Source: U.S. Environmental Protection Agency (2003).
 1   fertilizer is applied (Azimi et al., 2004). However, on a local scale solid waste disposal or mine
 2   tailings may be the predominant source of soil lead.
 3          A rigorous comparison of resuspension, leaching, and plant uptake "removal" rates for
 4   soil lead has not been undertaken.  Resuspension of lead-containing particles is likely the
 5   dominant removal mechanism from surface soil when soil pH is high.  Leaching may dominate
 6   when soil pH is low.  Leaching of lead through soil occurs more rapidly than uptake to pea or
 7   wheat crops (Azimi et al., 2004). More research is needed to compare removal rates for other
 8   plants with soil lead migration and resuspension rates.
 9          Surface waters are contaminated through several routes. On a global scale, sediment
10   resuspension and wet and dry deposition are the predominant contributors to lead concentrations
11   in surface water. On a local scale industrial effluent and urban runoff may dominate.
     December 2005
2-73
DRAFT-DO NOT QUOTE OR CITE

-------
                                      Solder
                                              Paints
                                              (weathering
                                              burning
                                              sanding)
          Petrol
                             Oil & Coal
                            Combustion
                 Smelting
                 refining
                       Cement
                       production
                 Load pipes
                Storage tanks
                      Sewage
                      effluent
                                                    Runoff from
                                                    roads and land
       Leaching from
       waste disposal
                                                                         Incineration
                                                                        Sewage sludge
                                                                        Municipal waste
                                                                        Solid waste
                                                                         disposal
                          Industrial
                          effluent
                       Sewage
                       sludge
                Inorganic
               phosphatic
                fertilizer
     Figure 2-7.  Transport pathways for lead in the environment.

     Source: Reprinted from Zabel (1993).
1          The major routes of lead transport into the food chain appear to be ingestion of

2    contaminated plants, ingestion of contaminated water, and inhalation of contaminated air.

3    Research into the relative importance of each of these transport routes  is needed.
     December 2005
2-74
DRAFT-DO NOT QUOTE OR CITE

-------
  1     2.6   REFERENCES

  2     Abu-Allaban, M; Gillies, J. A.; Gertler, A. W.; Clayton, R.; Proffitt, D. (2003) Tailpipe, resuspended road dust, and
  3            brake-wear emission factors from on-road vehicles. Atmos. Environ. 37: 5283-5293.
  4     Al-Chalabi, A. S.; Hawker, D. (1997) Response of vehicular lead to the presence of street dust in the atmospheric
  5            environment of major roads. Sci. Total Environ. 206: 195-202.
  6     Al-Subu, M. M.; Haddad, M.; Mizyed, N.; Mizyed, I. (2003) Impacts of irrigation with water containing heavy
  7            metals on soil and groundwater ~ a simulation study. Water, Air, Soil Pollut. 146: 141-152.
  8     Allen, A. G.; Harrison, R. M.; Nicholson, K. W. (1991) Dry deposition of fine aerosol to a short grass surface.
  9            Atmos. Environ. Part A 25: 2671-2676.
10     Alumaa, P.; Kirso, U.; Petersell, V.; Steinnes, E. (2002) Sorption of toxic heavy metals to soil. Int. J. Hyg. Environ.
11            Health 204(5-6): 375-376.
12     Annegarn, H. J.; Zucchiatti, A.; Sellschop, J. P. F.; Kusko, B. (1988) Composition and size of dust in a gold mine
13            atmosphere. J. Mine Vent. Soc. S. Afr. 41: 1-10.
14     Arakel, A. V.; Hongjun, T. (1992) Heavy-metal geochemistry and dispersion pattern in coastal sediments, soil, and
15            water of Kedron Brook floodplain area, Brisbane, Australia. Environ. Geol. Water Sci. 20: 219-231.
16     Azimi, S.; Cambier, P.; Lecuyer, I.; Thevenot, D.  (2004) Heavy metal determination in atmospheric deposition and
17            otherfluxes in northern France agrosystems. Water Air Soil Pollut. 157: 295-313.
18     Badawy, S. H.; Helal, M. I. D.; Chaudri, A. M.; Lawlor, K.; McGrath, S. P. (2002) Soil solid-phase controls lead
19            activity in soil solution. J. Environ. Qual. 31: 162-167'.
20     Baldasano, J. M.; Calbo, J.; Puig, O.; Guinart, X.  (1997) Climatological modeling of lead particles dispersion from
21            typical primary and secondary lead smelters. In: Power, H.; Tirabassi, T.; Brebbia, C. A., eds. Air pollution
22            modelling, monitoring and management. Boston, MA: Computational Mechanics Publications; pp. 259-267.
23     Barcan, V.  (2002) Nature and origin of multicomponent aerial emissions of the copper-nickel smelter complex.
24            Environ. Int. 28: 451-456.
25     Barcan, V.  S.; Kovnatsky, E. F.; Smetannikova, M. S.  (1998) Absorption of heavy metals in wild berries and edible
26            mushrooms in an area affected by smelter emissions. Water Air Soil Pollut. 103: 173-195.
27     Batonneau, Y.; Bremard, C.; Gengembre, L.; Laureyns, J.; Le Maguer, A.; Le Maguer, D.; Perdrix, E.; Sobanska, S.
28            (2004) Speciation of PM10 sources of airborne nonferrous metals within the 3 -km zone  of lead/zinc
29            smelters. Environ. Sci. Technol. 38: 5281-5289.
30     Bennett, R. L.; Knapp, K. T. (1989) Characterization of paniculate emissions from non-ferrous smelters. JAPCA
31            39:169-174.
32     Biggins, P. D. E.; Harrison, R. M. (1979) Atmospheric chemistry of automotive lead. Environ. Sci. Technol.
33            13:558-565.
34     Biggins, P. D. E.; Harrison, R. M. (1980) Chemical speciation of lead compounds in street dusts. Environ. Sci.
35            Technol. 14: 336-339.
36     Bindler, R.; Brannvall, M.-L.; Renberg, I. (1999) Natural lead concentrations in pristine boreal forest soils and past
37            pollution trends: a reference for critical load models. Environ. Sci. Technol. 33: 3362-3367.
38     Boisson, F.; Cotret, O.; Teyssie, J.-L.; El-Baradei, M.; Fowler, S. W. (2003) Relative importance of dissolved and
39            food pathways for lead contamination in shrimp. Mar. Pollut. Bull. 46: 1549-1557.
40     Boughton, B.; Horvath, A. (2004) Environmental assessment of used oil management methods. Environ. Sci.
41            Technol. 38: 353-358.
42     Boutron, C. F.; Candelone, J.-P.; Hong, S. M. (1994) Past and recent changes in the large-scale tropospheric cycles
43            of lead and other heavy metals as documented in Antarctic and Greenland snow and ice: a review. Geochim.
44            Cosmochim. Acta 58: 3217-3225.
45     Bridge, G. (2004) Contested terrain: mining and the environment. Ann. Rev. Energy Environ. 29: 205-259.
46     Cadle, S. H.; Mulawa, P. A.; Hunsanger, E. C.; Nelson, K.; Ragazzi, R. A.; Barrett, R.; Gallagher, G. L.;
47            Lawson, D. R.; Knapp, K. T.; Snow, R. (1999) Composition of light-duty motor vehicle exhaust paniculate
48            matter in the Denver, Colorado area. Environ.  Sci. Technol. 33: 2328-2339.
49     Cadle, S. H.; Mulawa, P. H.; Ball, J.; Donase,  C.; Weibel, A.; Sagebiel, J. C.; Knapp, K. T.; Snow, R. (1997)
50            Paniculate emission rates from in-use high-emitting vehicles recruited in Orange County, California.
51            Environ. Sci. Technol. 31: 3405-3412.
52     Cajuste, L. J.; Cruz-Diaz, J.; Garcia-Osorio, C. (2000) Extraction of heavy metals from contaminated soils:
53            I. Sequential extraction in surface soils and their relationships to DTPA extractable metals and metal plant
54            uptake. J. Environ. Sci. Health A35(7): 1141-1152.
55     Carr, D. S.  (2002) Lead Compounds. In: Ullman's Encyclopedia of Industrial Chemistry. New York,  NY: Wiley.


       December 2005                            2-75             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Carvalho, F. P. (1997) Distribution, cycling and mean residence time of 226Ra, 210Pb and 210Po in the Tagus
  2            estuary. Sci. Total Environ. 196: 151-161.
  3     Chadha, A.; McKelvey, L. D.; Mangis, J. K. (1998) Targeting lead in the multimedia environment in the continental
  4            United States. J. Air Waste Manage. Assoc. 48: 3-15.
  5     Chan, W. H.; Lusis, M.  A. (1986) Smelting operations and trace metals in air and precipitation in the Sudbury Basin.
  6            In: Nriagu, J. O.; Davidson, C. L, eds. Toxic metals in the atmosphere. New York, NY: John Wiley & Sons,
  7            Inc.; pp.  113-143. (Advances in environmental science and technology: v. 17).
  8     Chang, Y.-M; Chang; T.-C.; Lin, J.-P. (1999) Effect of incineration temperature on lead Emission from a fixed bed
  9            incinerator. J. Chem. Eng. Jpn. 32: 626-634.
10     Chen, C.-N.; Yang, W. F. (1998) Metal volatility during plastic combustion. J. Environ. Sci. Health A33: 783-799.
11     Chow, J. C.; Watson, J.  G.; Ashbaugh, L. L.; Magliano, K. L. (2003) Similarities and differences in PM10 chemical
12            source profiles for geological dust from the San Joaquin Valley, California. Atmos. Environ. 37: 1317-1340.
13     Chow, J. C.; Watson, J.  G.; Houck, J. E.; Pritchett, L. C.; Rogers, C. F.; Frazier, C. A.; Egami, R. T.; Ball, B. M.
14            (1994) A laboratory resuspension chamber to measure fugitive dust size distributions and chemical
15            compositions. Atmos. Environ. 28: 3463-3481.
16     Chow, T. J. (1978) Lead in natural waters. In: Nriagu, J. O., ed. The biogeochemistry of lead in the environment;
17            part A. ecological cycles. Amsterdam, The Netherlands: Elsevier/North-Holland Biomedical Press;
18            pp. 185-218.  (Topics in environemntal health: v. 1A).
19     Chow, J. C.; Watson, J.  G.; Richards, L. W.; Haase, D. L.; McDade, C.; Dietrich, D. L.; Moon, D.; Sloane, C.
20            (1991) The 1989-90 Phoenix PM10 study. Volume II: source apportionment. Final report. Phoenix, AZ:
21            Arizona Department of Environmental Quality; Desert Research Institute document no. 8931.6F1.
22     Clarke, L. B. (1993) The fate of trace elements during coal combustion and gasification: an overview. Fuel
23            72:731-736.
24     Claudio, E. S.; Godwin, H. A.; Magyar, J. S. (2003) Fundamental coordination chemistry, environmental chemistry,
25            and biochemistry of lead(II). In: Karlin, K. D., ed. Progress in inorganic chemistry, volume 51. Hoboken,
26            NY: John Wiley & Sons; pp.  1-144.
27     Cobb, G. P.; Sands, K.;  Waters, M.; Wixson, E.G.; Dorward-King, E. (2000)  Accumulation of heavy metals by
28            vegetables grown in mine wastes. Environ. Toxicol.  Chem. 19: 600-607.
29     Conko, K. M.; Rice, K.  C.; Kennedy, M. M. (2004) Atmospheric wet deposition of trace elements to a suburban
30            environment, Reston, Virginia, USA. Atmos. Environ. 38: 4025-4033.
31     Davidson, C. I.; Goold,  W. D.; Mathison, T. P.; Wiersma, G. B.; Brown, K. W.; Reilly, M. T. (1985) Airborne trace
32            elements in Great Smokey Mountains, Olympic and  Glacier National  Parks. Environ. Sci. Technol.
33            19:27-35.
34     Davidson, C. I.; Jaffrezo, J. L.; Small, M. J.; Summers, P. W.; Olson, M. P.; Borys, R. D. (1993) Trajectory analysis
35            of source regions influencing the south Greenland Ice Sheet during the Dye 3  gas and aerosol sampling
36            program Atmos. Environ. 27A: 2739-2749.
37     Davidson, C. I.; Osborn, J. F. (1986) The sizes of airborne trace metal containing particles. In: Nriagu, J. O.;
38            Davidson, C. I., eds. Toxic metals in the atmosphere. New York, NY: John Wiley & Sons. (Advances in
39            environmental science and technology: v. 17).
40     Davidson, C. I.; Wu, Y.-L. (1990) Dry deposition of particles and vapors. In: Lindberg, S. E.; Page, A. L.; Norton,
41            S. A., eds. Acidic precipitation: v. 3, sources, deposition, and canopy interactions. New York, NY: Springer-
42            Verlag; pp. 103-216.
43     Davidson, C.; Rabonowitz, M. (1992) Lead in the environment: from sources to human receptors. In: Needleman,
44            H. L., Human Lead Exposure. Ann Arbor, Michigan. CRC Press: 1992. pp 65-88.
45     Davis, A. P.; Burns, M.  (1999) Evaluation of lead concentration in runoff from painted structures. Water Res.
46            33:2949-2958.
47     Davis, A. P.; Shokouhian, M.; Ni, S. (2001) Loading estimates of lead, copper, cadmium, and zinc in urban runoff
48            from specific sources. Chemosphere 44: 997-1009.
49     Dedeurwaerder, H. L.; Dehairs, F. A.; Decadt, G. G.; Baeyens, W. F. (1983) In: Pruppacher, H. R.; Semonin, R. G.;
50            Slinn, W. G. N., eds. Precipitation scavenging, dry deposition, and resuspension. New York, NY: Elsevier;
51            pp. 1219-1231.
52     DeShields, B. R.; Meredith, R. W.; Griffin, D.; Laughlin, T.; Collins, W. (1998) The use of field methods to
53            evaluate the toxicity of lead to small plants at a small arms firing range. In: Little, E. E.; Delonay, A. J.;
54            Greenberg, B. M., eds. Environmental toxicology and risk assessment: v.7, ASTM STP  1333. West
55            Conshohocken, PA: American Society of Testing and Materials; pp 166-183.
      December 2005                            2-76             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Deiss, I; Byers, C.; Clover, D.; D'Amore, D.; Love, A.; Menzies, M. A.; Powell, I; Walter, T. M. (2004) Transport
  2            of lead and diesel fuel through a peat soil near Juneau, AK: a pilot study. J. Contam. Hyd. 74: 1-18.
  3     Demirbas, A. (2003a) Toxic air emissions frombiomass combustion. Energy Sources 25: 419-427.
  4     Demirbas, A. (2003b) Trace metal concentrations in ashes from various types of biomass species. Energy Sources
  5            25:743-751.
  6     Denaix, L.; Semlali, R. M.; Douay, F. (2001) Dissolved and colloidal transport of Cd, Pb, and Zn in a silt loam soil
  7            affected by atmospheric industrial deposition. Environ. Pollut. 114: 29-38.
  8     Douglas, B. E.; McDaniel, D. H.;  Alexander, J. J., eds. (1983) Concepts and models of inorganic chemistry. 2nd ed.
  9            New York, NY: John Wiley & Sons, Inc.
10     Dudka, S.; Adriano, D. C. (1997)  Environmental impacts of metal ore mining and processing: a review. J. Environ.
11            Qual. 26: 590-602.
12     Dudka, S.; Miller, W. P. (1999) Accumulation of potentially toxic elements in plants and their transfer to human
13            food chain. J. Environ. Sci. Health B 34(4): 681-708.
14     Durlak, S. K.; Biswas, P.; Shi, J. (1997) Equilibrium analysis of the affect of temperature, moisture and sodium
15            content on heavy metal emissions from municipal solid waste incinerators. J. Hazard. Mat. 56: 1-20.
16     El-Shobokshy, M. S. (1985) The dependence of airborne paniculate deposition on atmospheric stability and surface
17            conditions. Atmos. Environ.  19: 1191-1197.
18     Elzahabi,  M.; Yong, R. N. (200 l)pH influence on sorption characteristics of heavy metal in the vadose zone. Eng.
19            Geol. 60: 61-68.
20     Erel, Y.; Axelrod, T.; Veron, A.; Mahrer, Y.; Katsafados,  P.; Dayan, U. (2002) Transboundary atmospheric lead
21            pollution. Environ. Sci. Technol. 36: 3230-3233.
22     Erel, Y.; Patterson, C. C. (1994) Leakage of industrial lead into the hydrocycle. Geochim. Cosmochim. Acta
23            58: 3289-3296.
24     Erel, Y.; Veron, A.; Halicz, L. (1997) Tracing the transport of anthropogenic lead in the atmosphere and in soils
25            using isotopic ratios. Geochim. Cosmochim. Acta 61: 4495-4505.
26     Farfel, M. R.; Orlova, A. O.; Lees, P. S. J.; Rohde, C.; Ashley, P.  J.; Chisolm, J. J., Jr.  (2003) A study of urban
27            housing demolitions as sources of lead in ambient dust: demolition practices and exterior dust fall. Environ.
28            HealthPerspect. Ill:  1228-1234.
29     Pels, M.; Cooper, D. F.; Patterson, M. N. (1990) An analysis of wood-burning installations from an environmental
30            aspect. Energy Convers. Manage. 30: 235-244.
31     Finkelman, R. B. (2004) Potential health impacts of burning coal beds and waste banks. Int. J. Coal Geol. 59: 19-24.
32     Finkelstein, M. E.; Gwiazda, R. H.; Smith, D. R. (2003) Lead poisoning of seabirds: environmental risks from
33            leaded paint at a decommissioned military base. Environ. Sci. Technol. 37: 3256-3260.
34     Finster, M. E., Gray, K. A.; Binns, H. J. (2004) Lead levels of edibles grown in contaminated residential soils: a
35            field survey. Sci. Total Environ. 320: 245-257.
36     Flores-Rodriguez, J.; Bussy, A. -L.; Thevenot, D. R. (1994) Toxic metals in urban runoff: physico-chemical
37            mobility assessment using speciation schemes. Wat. Sci. Tech. 29: 83-93.
38     Frey, H. C.; Small, M. J. (2003) Integrated environmental assessment, Part I: estimating emissions. J. Ind. Ecol.
39            7:9-11.
40     Friedlander, S. K.; Turner, J. R.; Hering, S. V. (1986) A new method for estimating dry deposition velocities for
41            atmospheric aerosols.  J. Aerosol Sci. 17: 240-244.
42     Furimsky, E. (2000) Characterization of trace element emissions from coal combustion by equilibrium calculations.
43            Fuel Proc. Technol. 63: 29-44.
44     Galloway, J. N.; Thornton, J. D.; Norton, S.  A.; Volchok,  H. L.; McLean, R. A. N. (1982) Trace metals in
45            atmospheric deposition: a review and assessment. Atmos. Environ. 16: 1677-1700.
46     Gallon, C.; Tessier, A.; Gobeil, C.; Alfaro-De La Torre, M. C. (2004) Modeling diagenesis of lead in sediments of a
47            Canadian Shield lake.  Geochim. Cosm. Act. 68: 3531-3545.
48     Garg, B. D.; Cadle, S. H.; Mulawa, P. A.; Groblicki, P. J.  (2000) Brake wear paniculate matter emissions. Environ.
49            Sci. Technol. 34: 4463-4469.
50     Garty, J. (2001) Biomonitoring atmospheric heavy metals with lichens: theory and application. Crit. Rev. Plant Sci.
51            20(4): 309-371.
52     Gatz, D. F.; Chu, L.-C. (1986) Metal  solubility in atmospheric deposition. In: Nriagu, J. O.; Davidson, C. I., eds.
53            Toxic metals in the atmosphere. New York, NY: John Wiley & Sons, Inc.; pp. 391-408. (Advances in
54            environmental science and technology: v. 17).
55     Gelinas, Y.; Schmidt, J. -P. (1998) Estimation of the bult atmospheric deposition of major and trace elements to a
56            rural watershed. Atmos. Environ. 32: 1473-1483.
       December 2005                            2-77             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Ghosh, R.; Majumder, T.; Ghosh, D. N. (1987) A study of trace elements in lithotypes of some selected Indian coals.
  2            Int. J. Coal Geol. 8: 269-278.
  3     Gillies, J. A.; Gertler, A. W.; Sagebiel, J. C.; Dippel, W. A. (2001) On-road paniculate matter (PM2.5 and PM10)
  4            emissions in the Sepulveda Tunnel, Los Angeles, California. Environ. Sci. Technol. 35:  1054-1063.
  5     Gillies, J. A.; O'Connor, C. M; Mamane, Y.; Gertler, A. W. (1999) Chemical profiles for characterizing dust
  6            sources in an urban area, western Nevada, USA. In: Livingstone, I., ed. Aeolian geomorphology: papers
  7            from the 4th international conference on aeolian research;  1998; Oxford, United Kingdom. Z. Geomorphol.
  8            116(suppl.): 19-44.
  9     Gomez Ariza, J. L.; Morales, E.; Sanchez-Rodas, D.; Giraldez, I. (2000) Stability of chemical species in
10            environmental matrices. TrAC Trends Anal. Chem. 19: 200-209.
11     Greenwood, N. N.; Earnshaw, A. (1984) Chemistry of the elements. New York, NY: Pergamon Press; 1984.
12     Gulson, B. L.; Palmer, J. M.; Bryce, A. (2002) Changes in blood lead in a recreational shooter. Sci. Total Environ.
13            293:  143-150.
14     Hagner, C. (2002) Regional and long-term patterns of lead concentrations in riverine, marine and terrestrial systems
15            and humans in northwest Europe. Water Air Soil Pollut. 134: 1-39.
16     Harris, A. R.; Davidson, C. I. (2005) The role of resuspended soil in lead flows in the California South Coast Air
17            Basin. Environ. Sci. Technol. 39: 7410-7415.
18     Harris, A. R.; Fifarek, B. J.; Davidson, C. I.; Blackmon, R. L. (2005) Stationary sources of airborne lead: a
19            comparison of emissions data for southern California. J. Air Waste Manage.  Assoc.: in press.
20     Harrison, P. G., ed. (1985) Organometallic compounds of germanium, tin, and lead. New York, NY:  Chapman and
21            Hall; pp. 41-68.
22     Harrison, R. M. (1986) Chemical speciation and reaction pathways of metals in the atmosphere. In: Davidson, J. O.;
23            Nriagu, C. I., eds. Toxic Metals in the Atmosphere, New York, NY: John Wiley & Sons, Inc.; 1986:
24            pp 319-333. (v.17).
25     Harrison, R. M.; Laxen, D. P. H. (1980) Metals in the environment. 1. Chemistry. Chem. Br.  16: 316-320.
26     Harrison, R. M.; Tilling, R.; Callen Romero, M. S.; Harrad, S.; Jarvis, K. (2003) A study of trace metals and
27            polycyclic aromatic hydrocarbons  in the roadside environment. Atmos. Environ. 37: 2391-2402.
28     Harrison, R. M.; Williams, C. R. (1983) Physico-chemical characterization of atmospheric trace metal emissions
29            from a primary zinc-lead smelter. Sci. Total Environ. 31:  129-140.
30     Harrison, R. M.; Williams, C. R.; O'Neill, I. K. (1981) Characterization of airborne heavy-metals within a primary
31            zinc-lead smelting works. Environ. Sci. Technol. 15: 1197-1204.
32     Hee, S. S. Q. (1994) Availability of elements in leaded/unleaded automobile exhausts, a leaded paint, a soil, and
33            some mixtures. Arch. Environ. Contam. Toxicol. 27: 145-153.
34     Hewitt, C. N.; Harrison, R. M. (1986) Formation and decomposition  of trialkyllead compounds in the atmosphere.
35            Environ. Sci. Technol. 20: 797-802.
36     Hewitt, C. N.; Rashed, M. B. (1992) Removal rates of selected pollutants in the runoff waters from a major rural
37            highway. Water Res. 26:311-319.
38     Hicks, B. B. (1986) Measuring dry deposition: a re-assessment of the state of the art. Water Air Soil Pollut.
39            30:75-90.
40     Hlavay, J.; Polyak, K.; Weisz, M. (2001) Monitoring of the natural environment by chemical speciation of elements
41            in aerosol and sediment samples. J. Environ. Monit. 3: 74-80.
42     Ho, T. C.; Chu, H. W.; Hopper, J. R. (1993) Metal volatilization and  separation during incineration. Waste Manage.
43            13:455-466.
44     Hofken, K. D.; Meixner, F. X.; Ehhalt, D. H. (1983) Deposition of atmospheric trace constituents onto different
45            natural surfaces. In: Pruppacher, H. R.; Semonin, R. G.; Slinn, W. G. N., eds. Precipitation scavenging, dry
46            deposition, and resuspension: v.  2, dry deposition and resuspension: proceedings of the fourth international
47            conference;  November-December  1982; Santa Monica, CA. New York, NY: Elsevier; pp. 825-835.
48     Holtzman, R. B. (1978) Application of radio lead to metabolic studies. In: Nriagu, J.  O., ed. The biogeochemistry of
49            lead in the environment; part B. Biological effects. Amsterdam,  The Netherlands: Elsevier/North-Holland
50            Biomeical Press; pp. 37-96. (Topics in environmental health: v.  IB).
51     Humphreys, D. J. (1991) Effects of exposure to excessive quantities of lead on animals. Br. Vet. J.  147: 18-30.
52     Huntzicker, J. J.; Friedlander, S. K.; Davidson, C. I. (1975) Material balance for automobile-emitted lead in Los
53            Angeles basin. Environ. Sci. Technol. 9: 448-457.
54     Jensen, D. L.; Holm, P. E.; Christensen, T. H. (2000) Leachability  of heavy metals from scrap dirt sampled at two
55            scrap iron and metal recycling facilities. Waste Manage. Res. 18: 367-379.
       December 2005                             2-78             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Jing, C.; Meng, X.; Korfiatis, G. P. (2004) Lead teachability in stabilized/solidified soil samples evaluated with
  2            different leaching tests. J. Hazard. Mat. B 114: 101-110.
  3     Joshi, S. R.; McCrea, R. C.; Shukla, B. S.; Roy, J.-C. (1991) Partitioning and transport of lead-210 in the Ottawa
  4            River watershed. Water Air Soil Pollut. 59:311-320.
  5     Karathanasis, A. D. (2000) Colloid-mediated transport of Pb through soil porous media. Int. J. Environ. Studies
  6            57(5): 579-596.
  7     Karcher, B. (1999) Aviation-produced aerosols and contrails. Surv. Geophys. 20: 113-167.
  8     Kaste, J.; Friedland, A.; Stump, S. (2003) Using stable and radioactive isotopes to trace atmospherically deposited
  9            Pb in montane forest soils. Environ. Sci. Technol. 37: 3560-3567.
10     Khan, A. T.; Thompson, S. J.; Mielke, H. W. (1995) Lead and mercury levels in raccoons from Macon County,
11            Alabama. Bull. Environ. Contam. Toxicol. 54: 812-816.
12     Kimbrough, D. E.; Suffet, I. H.  (1995) Off-site forensic determination of airborne elemental emissions by multi-
13            media analysis: a case study at two secondary lead smelters. Environ. Sci. Technol. 29: 2217-2221.
14     King, R. B. (1995) Silicon, germanium, tin, and lead. In: Inorganic chemistry of main group elements. New York,
15            NY: VCH Publishers Inc.; pp. 43-65.
16     Kiratli, N.; Ergin, M. (1996) Partitioning of heavy metals in surface Black Sea sediments. Appl. Geochem.
17            11:775-788.
18     Kleeman, M. J.; Schauer, J. J.; Cass, G. R. (2000) Size and composition distribution of fine paniculate matter
19            emitted from motor vehicles. Environ. Sci. Technol. 34: 1132-1142.
20     Kotz, J. C.; Purcell, K. F., eds. (1991) Chemistry and Chemical Reactivity. 2nd ed. Philadelphia, PA: Saunders
21            College Publishing; pp. 953-968.
22     Koutrakis,  P.; Sioutas, C. (1996) Physico-chemical properties and measurement of ambient particles. In: Wilson, R.;
23            Spengler, J. D., eds. Particles in our air: concentrations and health effects. Cambridge, MA: Harvard
24            University Press; pp 15-39.
25     Krook, J.; Martensson,  A.; Eklund, M. (2004) Metal contamination in recovered waste wood used as energy source
26            in  Sweden. Resour. Conserv. Recycl. 41: 1-14.
27     Kurkjian, R.; Dunlap, C.; Flegal, A. R. (2002) Lead isotope tracking of atmospheric response to post-industrial
28            conditions in Yerevan, Armenia. Atmos. Environ. 36: 1421-1429.
29     Kurkjian, R.; Dunlap, C.; Flegal, A. R. (2004) Long-range downstream effects of urban runoff and acid mine
30            drainage in the Debed River, Armenia: insights from lead isotope modeling. Appl. Geochem.
31            19:1567-1580.
32     Lankey, R. L.; Davidson, C. I.; McMichael, F. C. (1998) Mass balance for lead in the California south coast air
33            basin: an update. Environ. Res. 78: 86-93.
34     Lazrus, A.  L.; Lorange, E.; Lodge, J. P., Jr. (1970) Lead and other metal ions in United States precipitation.  Environ.
35            Sci. Technol. 4: 55-58.
36     Lee, R. E., Jr.; Von Lehmden, D. J. (1973) Trace metal pollution in the environment. J. Air Pollut. Control Assoc.
37            23: 853-857.
3 8     Lemieux, P. M.; Ryan,  J. V. (1993) Characterization of air pollutants emitted from a simulated scrap tire fire. J. Air
39            Waste Manage. Assoc.  43: 1106-1115.
40     Lieb, D. A.; Carline R.  F. (2000) Effects of urban runoff from a detention pond on water quality, temperature and
41            caged Gammarus minus (Say) (Amphipoda) in a headwater stream. Hydrobiol. 441: 107-116.
42     Lin, J.-M.; Fang, G.-C.; Holsen, T. M.; Noll, K. E.  (1993) A comparison of dry deposition modeled from size
43            distribution data and measured with a smooth surface for total particle mass,  lead and calcium in Chicago.
44            Atmos. Environ. Part A 27: 1131-1138.
45     Lindberg, S. E.; Harriss, R. C. (1981) The role of atmospheric deposition in an eastern U.S. deciduous forest. Water
46            Air Soil Pollut. 16:  13-31.
47     Lindberg, S. E.; Lovett, G. M. (1985) Field measurements of particle dry deposition rates to foliage and inert
48            surfaces in a forest canopy. Environ. Sci. Technol. 19: 238-244.
49     Liu, Z. P. (2003) Lead poisoning combined with cadmium in sheep and horses in the vicinity of non-ferrous metal
50            smelters. Sci. Total Environ. 309: 117-126.
51     Lockwood, F. C.;  Yousif, S. (2000) A model for the paniculate matter enrichment with toxic metals in solid fuel
52            flames. Fuel Process. Technol. 65-66: 439-457.
53     Loranger, S.; Zayed, J.  (1994) Manganese and lead concentrations in ambient air and emission rates from unleaded
54            and leaded gasoline between 1981 and 1992 in Canada: a comparative study. Atmos. Environ.
55            28: 1645-1651.
       December 2005                             2-79             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Lough, G. C.; Schauer, J. I; Park, J.-S.; Shafer, M. M; Deminter, J. T.; Weinstein, J. P. (2005) Emissions of metals
  2            associated with motor vehicle roadways. Environ. Sci. Technol. 39: 826-836.
  3     Lowenthal, D. H.; Zielinska, B.; Chow, J. C.; Watson, J. G.; Gautam, M.; Ferguson, D. H.; Neuroth, G. R.; Stevens,
  4            K. D. (1994) Characterization of heavy-duty diesel vehicle  emissions. Atmos. Environ. 28: 731-743.
  5     Mace, J. E.; Graham, R. C.; Amrhein, C. (1997) Anthropogenic lead distribution in rodent-affected and undisturbed
  6            soils in southern California. Soil Sci. 162: 46-50.
  7     Main, H. H.; Friedlander, S. K. (1990) Dry deposition of atmospheric aerosols by dual tracer method--!, area source.
  8            Atmos. Environ. 24A: 103-108.
  9     Marcos L.; Legret M.; Raimbault G.; Le Cloirec P. (2002) Experimental study and modelling of zinc and lead
10            migration in sandy soils due to stormwater infiltration. Water Sci Technol. 45: 57-64.
11     Maskall, J. E.; Thornton, I. (1998) Chemical partitioning of heavy metals in soils, clays and rocks at historical lead
12            smelting sites. Water Air Soil Pollut. 108: 391-409.
13     Matthes, S. A.; Cramer, S. D.; Covino Jr., B. S.; Bullard, S. J.; Holcomb, G. R. (2002) Precipitation runoff from
14            lead. In: Townsend, H. E., ed. Outdoor Atmospheric Corrosion. West Conshocken, PA: American Society
15            for Testing and materials International, pp. 265-274.  (ASTM STP 1421).
16     McDonald, J. D.; Zielinska, B.; Sagebiel, J. C.; McDaniel, M. R.; Mousset-Jones, P.  (2003) Source apportionment
17            of airborne fine paniculate matter in an underground mine.  J. Air Waste Manage. Assoc. 53: 386-395.
18     Merrington, G.; Alloway, B. J. (1994) The flux of Cd, Cu, Pb and Zn in mining polluted soils. Water Air Soil Pollut.
19            73: 333-344.
20     Milford, J. B.; Davidson, C. I. (1985) The sizes of paniculate trace elements in the atmosphere~a review. J. Air
21            Pollut. Control Assoc. 35: 1249-1260.
22     Miller, E. K.; Friedland, A. J. In Proceedings of the Eighth International Conference  on Heavy Metals in the
23            Environment, Edinburgh, 1991; Farmer, J. G., Ed. CEP Consultants Ltd.: Edinburgh, 1991; pp 86-89.
24     Miller, E. K.; Friedland, A. J. (1994) Lead migration in forest soils: response to changing atmospheric inputs.
25            Environ. Sci. Technol. 28: 662-669.
26     Mukherjee, S.; Srivastava, S. K. (2005) Trace elements in high-sulfur Assam coals from the Makum Coalfield in the
27            northeastern region of India. Energy Fuels 19: 882-891.
28     Nicholson, K. W. (1988a) The dry deposition of small particles: a review of experimental measurements. Atmos.
29            Environ. 22: 2653-2666.
30     Nicholson, K. W. (1988b) A review of particle resuspension. Atmos. Environ. 22:  2639-2651.
31     Nicholson, K. W. (1993) Wind tunnel experiments on the resuspension of particulate matter. Atmos. Environ. 27A:
32            181-188.
33     Nicholson, K. W.; Branson, J. R.; Giess, P.; Cannell, R. J.  (1989) The effects of vehicle activity on particle
34            resuspension. J. Aerosol Sci. 20: 1425-1428.
3 5     Nightengale, H. I. (1987) Accumulation of As, Ni, Cu, and Pb in retention and recharge basins soils from urban
36            runoff. Water Resources Bulletin 23: 663-672.
37     Nriagu, J. O. (1989) A global assessment of natural sources of atmospheric trace metals. Nature (London)
38            338:47-49.
39     Nriagu, J. O. (1990) The rise and fall of leaded gasoline. Sci. Total Environ. 92: 13-28.
40     Nriagu, J. O.; Pacyna, J. M. (1988) Quantitative assessment of worldwide contamination of air, water and soils by
41            trace metals. Nature (London) 333: 134-139.
42     Ohmsen, G. S. (2001) Characterization of fugitive material within a primary lead smelter. J. Air Waste Manage.
43            Assoc. 51: 1443-1451.
44     Olmez, I.; Sheffield, A. E.; Gordon, G. E.;  Houck, J. E.; Pritchett, L. C.; Cooper, J. A.; Dzubay, T. G.; Bennett, R. L.
45            (1988) Compositions of particles from selected sources in Philadelphia for receptor modeling applications.
46            JAPCA 38: 1392-1402.
47     Pacyna, J. M. (1986) Emission factors of atmospheric elements. In:  Nriagu, J. O.; Davidson, C. I., eds. Toxic metals
48            in the atmosphere. New York, NY: John Wiley & Sons, Inc.; pp 1-32. (Advances in environmental science
49            and technology: v. 17).
50     Palacios, H.; Iribarren, I.; Olalla, M. J.; Cala, V. (2002) Lead poisoning of horses in the vicinity of a battery
51            recycling plant. Sci. Total Environ. 290: 81-89.
52     Pang, L.; Close, M.; Schneider, D.; Stanton, G. (2002) Effect of pore-water velocity on chemical nonequilibrium
53            transport of Cd, Zn, and Pb in alluvial gravel columns. J.  Contam. Hydrol. 57: 241-258.
54     Parekh, P. P.; Khwaja, H. A.; Khan, A. R.; Naqvi, R. R.; Malik, A.; Khan, K.; Hussain, G. (2002) Lead content of
55            petrol and diesel and its assessment in an urban environment. Environ. Monitor. Assess. 74: 255-262.
       December 2005                             2-80             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Pattenden, N. I; Branson, J. R.; Fisher, E. M. R. (1982) In: Georgii, H. W.; Pankrath, I, eds. Deposition of
  2            atmospheric pollutants. Dordrecht, The Netherlands: Reidel; pp. 173-184.
  3     Pekney, N. J.; Davidson, C. I. ; Bein, K. J. ; Wexler, A. S. ; Johnston, M. V. (2006) Identification of sources of
  4            atmospheric pm at the Pittsburgh supersite, part I: single particle analysis and filter-based positive matrix
  5            factorization. Atmos. Environ,  in press.
  6     Pelletier, E. (1995) Environmental organometallic chemistry of mercury, tin, and lead: present status and
  7            perspectives. In: Tessier, A.; Turner, D. R., eds. Metal speciation and bioavailability in aquatic systems.
  8            New York, NY: John Wiley &  Sons; pp. 103-148. (Analytical and Physical Chemistry of Environmental
  9            Systems series: v. 3)
10     Peltier, E. F.; Webb, S. M.; Gaillard, J.-F. (2003) Zinc and lead sequestration in an impacted wetland system. Adv.
11            Environ. Res. 8: 103-112.
12     Perkins, S. M.; Filippelli, G. M.; Souch, C. J. (2000) Airborne trace metal contamination of wetland sediments at
13            Indiana Dunes National Lakeshore. Water Air Soil Pollut.  122: 231-260.
14     Pilgrim, W.; Hughes, R. N. (1994) Lead, cadmium, arsenic and zinc in the ecosystem surrounding a lead smelter.
15            Environ. Monit. Assess. 32: 1-20.
16     Pitzer, K. S. (1979) Relativistic effects on chemical properties. Ace. Chem. Res.  12: 271-276.
17     Prengaman, R. D. (2002) Lead alloys. In: Ullman's encyclopedia of industrial chemistry. New York, NY: Wiley -
18            VCH.
19     Reuer, M. K.; Weiss, D. J. (2002) Anthropogenic lead dynamics  in the terrestrial and marine environment. Phil.
20            Trans. Roy. Soc. London A 360: 2889-2904.
21     Rhoads, B. L.; Cahill, R. A. (1999) Geomorphological assessment of sediment contamination in an urban stream
22            system. Appl. Geochem. 14(4): 459-483.
23     Rhue, R. D.; Mansell, R. S.; Ou, L.-T.;  Cox, R.; Tang, S. R.; Ouyang, Y. (1992) The fate and behavior of lead alkyls
24            in the environment: a review. Crit. Rev. Environ.  Control 22: 169-193.
25     Rieuwerts, J. S.; Farago, M. E. (1995) Lead contamination in smelting and mining environments and variations in
26            chemical forms and bioavailability. Chem. Speciation Bioavailability 7:  113-123.
27     Ritson, P. I.; Bouse, R. M.; Flegal, A. R.; Luoma, S. N. (1999) Stable lead isotopic analyses of historic and
28            contemporary lead contamination of San Francisco Bay estuary. Marine  Chem. 64: 71-83.
29     Ritson, P. I.; Esser, B. K.; Niemeyer, S.; Flegal, A. R. (1994) Lead isotopic determination of historical sources of
30            lead to Lake Erie, North America. Geochim. Cosmochim. Acta 58: 3297-3305.
31     Roger, S.; Montrejaud-Vignoles, M.; Andral, M. C.; Herremans,  L.; Fortune, J. P., (1998) Mineral, physical and
32            chemical analysis of the solid matter carried by motorway  runoff water. Wat. Res. 32: 1119-1125.
33     Rohbock, E. (1982) In: Georgii, H. W.; Pankrath, J., eds. Deposition of atmospheric pollutants.  Dordrecht, The
34            Netherlands: Reidel; pp. 159-171.
35     Rojas, C. M.; Van Grieken, R. E.; Laane, R. W. (1993) Comparison of three dry deposition models applied to field
36            measurements in the southern bight of the North Sea. Atmos. Environ. 27A: 363-370.
3 7     Root, R. A. (2000) Lead loading of urban streets by motor vehicle wheel weights. Environ. Health Perspect.
38            108:937-940.
39     Sanchez-Hernandez, J. C. (2000) Trace element contamination in Antarctic ecosystems. Rev. Environ. Contam.
40            Toxicol. 166: 83-127.
41     Schell, W. R.; Barnes, R. S. (1986) Environmental isotope and anthropogenic tracers of recent lake sedimentation.
42            In: Fritz, P.; Fontes, J. C., eds. Handbook of environmental isotope geochemistry, the terrestrial
43            environment, B. Vol. 2. New York, NY: Elsevier Science Publishers; pp. 169-206.
44     Scherbatskoy, T.; Bliss, M. (1984) In: Sampson, P. J., ed.  The meteorology of acid deposition. Pittsburgh, PA: Air
45            Pollution Control Association.
46     Schintu, M.; Kudo, A.; Sarritzu, G.; Contu, A. (1991) Heavy metal distribution and mobilization in sediments from a
47            drinking water reservoir near a mining area. Water Air Soil Pollut. 57: 329-338.
48     Schlesinger, W. H.; Reiners, W. A. (1974) Deposition of water and cations on artificial foliar collectors in Fir
49            krummholz of New England mountains. Ecology  55: 378-386.
50     Schweitzer, P. A. (2003) Lead and lead alloys. In: Schweitzer, P. A. Metallic materials: physical, mechanical, and
51            corrosion properties. New York, NY: Marcel Dekker Inc.;  pp. 695-698.
52     Sehmel, G. A. (1980) Particle resuspension: a review. Environ. Int. 4:  107-127.
53     Seinfeld, J. H.; Pandis, S. N. (1998) Atmospheric chemistry and physics: from air pollution to climate change. New
54            York, NY: John Wiley & Sons, Inc.
55     Senior, C. L.; Helble, J. J.; Sarofim, A.  F. (2000) Emissions of mercury, trace elements, and fine particles from
56            stationary combustion sources.  Fuel Process. Tech. 65-66:  263-288.
       December 2005                            2-81             DRAFT-DO NOT QUOTE OR CITE

-------
  1     Sievering, H.; Dave, M.; McCoy, P.; Sutton, N. (1979) Deposition of sulfate during stable atmospheric transport
  2            over Lake Michigan. Atmos. Environ. 13: 1717-1718.
  3     Shah, S. D.; Cocker, D. R., Ill; Miller, J. W.; Norbeck, J. M. (2004) Emission rates of paniculate matter and
  4            elemental and organic carbon from in-use diesel engines. Environ. Sci. Technol. 38: 2544-2550.
  5     Shapilova, M. V.; Alimova, S. I. (2000) Environmental problems in the production of household and crystal glass.
  6            Glass Ceram. 57: 293-295.
  7     Shapiro, H.; Frey, F. W. (1968) The organic compounds of lead. New York, NY: John Wiley & Sons, Inc.;
  8            pp. 5-8, 15-16.
  9     Sheppard, S. C.; Sheppard, M. I. (1991) Lead in boreal soils and food plants. In: Adriano, D. C., ed. Metals in soils,
10            waters, plants and animals: proceedings of an international conference; April 1990; Orlando, FL. Water Air
11            Soil Pollut. 57-58:79-81.
12     Shinya, M.; Tsuchinaga, T.; Kitano, M.; Yamada, Y.; Ishikawa, M., (2000) Characterization of heavy metals and
13            polycyclic aromatic hydrocarbons in urban highway runoff. Wat. Sci. Tech. 42: 201-208.
14     Silva, P. J.; Prather, K. A. (1997) On-line characterization of individual particles from automobile emissions.
15            Environ. Sci. Technol. 31: 3074-3080.
16     Singley, J. E. (1994) Electrochemical nature of lead contamination. J. Am. Waterworks Assoc. 86: 91-96.
17     Sivaraman, D.; Lindner, A. S. (2004) A comparative life cycle analysis of gasoline-, battery-, and electricity-
18            powered lawn mowers. Environ. Eng. Sci. 21: 768-785.
19     Slinn, S. A.; Slinn, W. G. N. (1980) Predictions for particle deposition on natural waters. Atmos. Environ.
20            14: 1013-1016.
21     Small, M. J.; Nunn, A. B., Ill; Forslund, B. L.; Daily, D. A. (1995) Source attribution of elevated residential soil
22            lead near a battery recycling site. Environ. Sci. Technol. 29:  883-895.
23     Smith, W. H.; Siccama, T. G. (1981) The Hubbard Brook ecosystem study: biogeochemistry of lead in the northern
24            hardwood forest. J. Environ. Qual. 10: 323-333.
25     Sobanska, S.; Ricq, N.; Laboudigue, A.; Guillermo, R.; Bremard, C.; Laureyns, J.; Merlin, J. C.; Wignacourt, J. P.
26            (1999) Microchemical investigations of dust emitted by a lead smelter. Environ. Sci. Technol.
27            33: 1334-1339.
28     Socolow, R.; Thomas, V. (1997) The industrial ecology of lead and electric vehicles. J. Ind. Ecol. 1: 13-36.
29     Steding, D. J.; Dunlap, C. E.; Flegal, A. R. (2000) New isotopic evidence for chronic lead contamination in the San
30            Francisco Bay estuary system: implications for the persistence of past industrial lead emissions in the
31            biosphere. Proc. Natl. Acad. Sci. U. S. A. 97: 11181-11186.
32     Steinnes, E. (2001) Metal contamination of the natural environment in Norway from long range atmospheric
33            transport. Water Air Soil Pollut.  1(3/4): 449-460.
34     Sternbeck, J.; Sjodin, A.; Andreasson, K. (2002) Metal emissions from road traffic and the influence of
35            resuspension-results from two tunnel studies. Atmos. Environ. 36: 4735-4744.
36     Sturges, W. T.; Harrison, R. M. (1986) The use of Br/Pb ratios in atmospheric particles to discriminate between
3 7            vehicular and industrial lead sources in the vicinity of a lead works--!. Thorpe, West Yorkshire. Atmos.
38            Environ. 20: 833-843.
39     Taebi, A.; Droste, R. L. (2004) First flush pollution load of urban stormwater runoff J. Environ. Eng. Sci.
40            3:301-309.
41     Thurston, G. D.; Spengler, J. D. (1985) A quantitative assessment of source contributions to inhalable paniculate
42            matter pollution in metropolitan Boston. Atmos. Environ. 19: 9-25.
43     Toner, R. N.; Frost, C. D.; Chamberlain, K. R. (2003) Isotopic identification of natural vs. anthropogenic sources of
44            Pb in Laramie basin groundwaters, Wyoming, USA. Environ. Geol. 43: 580-591.
45     Tsai, E. C.-E. (1987) Analysis of ambient lead concentrations around three secondary lead smelters. Water Air Soil
46            Pollut. 33:321-329.
47     Tsai, J.-H.; Wu, Y.-L. (1995) Contributions of road dust resuspension to the airborne particle concentrations in
48            Taipei. Part. Sci. Technol. 13:  55-67.
49     Twohy, C. H.; Gandrud, B. W. (1998) Electron microscope analysis of residual particles from aircraft contrails.
50            Geophys. Res. Lett. 25: 1359-1362.
51     U.S. Environmental Protection Agency. (1986) Air quality criteria for lead. Research Triangle Park, NC: Office of
52            Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
53            EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.
54     U.S. Environmental Protection Agency. (1991) Methods for the determination of metals in environmental samples.
55            Washington, DC: U.S. Environmental Protection Agency; EPA/600/4-91-010.
      December 2005                            2-82             DRAFT-DO NOT QUOTE OR CITE

-------
  1     U.S. Environmental Protection Agency. (1998) Locating and estimating air emissions from sources of lead and lead
  2            compounds. Washington, DC: Office of Air Quality; report no. EPA-454/R-98-006. Available:
  3            http://www.epa.gov/ttn/chief/le/lead.pdf [13 October, 2005].
  4     U.S. Environmental Protection Agency. (2000) Great Lakes binational toxics strategy report on alkyl-lead: sources,
  5            regulations and options. Ann Arbor, MI: Great Lakes National Program Office. Available:
  6            http://www.epa.gov/glnpo/bns/lead/Step%20Report/steps.pdf [12 October, 2005].
  7     U.S. Environmental Protection Agency. (2002) PBT national action plan for alkyl-Pb. Washington, DC: Persistent,
  8            Bioaccumulative, and Toxic Pollutants (Pbt) Program. Available:
  9            http://www.epa.gov/opptintr/pbt/cheminfo.htm [13 October, 2005].
10     U.S. Environmental Protection Agency. (2003) National air quality and emissions trends report. 2003 special studies
11            edition. Research Triangle Park, NC: Office of Air Quality Standards; Emissions Monitoring and Analysis
12            Division; report no. EPA 454/R-03-005. Available: http://www.epa.gov/air/airtrends/aqtrnd03/toc.html
13            (27 August, 2004).
14     U.S. Geological Survey (USGS). (2003) Minerals yearbook 2003:  lead. Washington, DC: U. S. Department of the
15            Interior. Available: http://minerals.usgs.gov/minerals/pubs/commodity/lead/ [13 October, 2005].
16     Urdu, K. (1998) Transport of metals leaching from land-disposed oil field wastes. Waste Manage. Res.
17            16(6): 541-554.
18     Venditti, D.; Durecu, S.; Berthelin, J. (2000) A multidisciplinary approach to assess history, environmental risks,
19            and remediation feasibility of soils contaminated by metallurgical activities. Part A: chemical and physical
20            properties of metals and leaching ability. Arch. Environ. Contain. Toxicol. 38: 411-420.
21     Veron, A. J.; Church, T.  M. (1997) Use of stable lead isotopes and trace metals to characterize air mass sources into
22            the eastern North Atlantic. J. Geophys. Res. [Atmos.] 102(D23): 28,049-28,058.
23     Vilomet, J. D.; Veron, A.; Ambrosi, J. P.; Moustier, S.; Bottero, J.  Y.; Chatelet-Snidaro, L. (2003) Isotopic tracing
24            of landfill leachates and pollutant lead mobility in soil and groundwater. Environ. Sci. Technol.
25            37:4586-4591.
26     Wagner, J. P.; Caraballo, S. A. (1997) Toxic species emissions from controlled combustion of selected rubber and
27            plastic consumer products. Polym. Plast. Technol. Eng. 36(2):  189-224.
28     Walsh, D. C.; Chillrud, S. N.; Simpson, H. J.; Bopp, R. F. (2001) Refuse incinerator paniculate emissions and
29            combustion residues for New York City during the 20th century. Environ. Sci. Technol. 35: 2441-2447.
30     Wang, L.; Morris, A. E. (1995) A process engineering approach to remedy an environmental problem of fugitive
31            lead emissions during lead refining. J. Mater. Res. 10(3): 538-544.
32     Weast, R. C.; Astle, M. J.; Beyer, W. H., eds. (1984) CRC handbook of chemistry and physics: a ready-reference
3 3            book of chemical and physical data. 65th ed. Boca Raton,  FL: CRC Press, Inc.
34     Weiss, D.; Shotyk, W.; Kempf, O. (1999) Archives of atmospheric lead pollution. Naturwissenschaften 86: 262-275.
35     Wernick, I. K.; Themelis, N. J. (1998) Recycling metals forthe environment. Annu. Rev. Energy Environ.
36            23:465-497.
37     Willemsen, L. C.; vanderKerk, G. J. M. (1965) Investigations in the field of organolead chemistry. Utrecht, The
38            Netherlands: International Lead Zinc Research Organization, Inc.; pp. 1-13.
39     Williams, R. (1982) A model for the dry deposition of particles to  natural water surfaces. Atmos. Environ.
40            16: 1933-1938.
41     Winkler, R.; Dietl, F.; Frank, G.; Tschiersch, J. (1998) Temporal variation of 7Be and 210Pb size distributions in
42            ambient aerosol. Atmos. Environ.  32: 983-991.
43     Wu, Y.-L.; Davidson, C. I.; Dolske, D. A.; Sherwood, S. I. (1992a) Dry deposition of atmospheric contaminants: the
44            relative importance of aerodynamic, boundary layer, and surface resistances. Aerosol Sci. Technol.
45            16:65-81.
46     Wu, Y.-L.; Davidson, C. I.; Lindberg, S. E.; Russell, A. G. (1992b) Resuspension of paniculate chemical  species at
47            forested sites. Environ. Sci. Technol. 26: 2428-2435.
48     Young, T. M.; Heeraman, D. A.; Sirin, G.; Ashbaugh, L. L. (2002) Resuspension of soil as a source of airborne lead
49            near industrial facilities and highways. Environ. Sci. Technol. 36: 2484-2490.
50     Zabel, T. F. (1993) Diffuse sources  of pollution by heavy metals. J. Inst. Water Environ. Manage. 7: 513-520.
51     Zadnik, T. (2004) Lead in topsoil, hay,  silage and blood of cows from farms near a former lead mine and current
52            smelting plant before and after installation of filters. Vet. Hum. Toxicol. 46: 287-290.
53     Zhang, M. K.; Xu, J. M.  (2003) Difference of lead,  copper and zinc concentrations between interiors and exteriors of
54            peds in some contaminated soils. Chemosphere 50: 733-738.
       December 2005                             2-83              DRAFT-DO NOT QUOTE OR CITE

-------
1    Zufall, M. I; Davidson, C. I. (1997) Dry Deposition of Particles to Water Surfaces In: Atmospheric Deposition of
2           Contaminants to the Great Lakes and Coastal Waters, J.E. Baker, editor, Society of Environmental
3           Toxicology and Chemistry (SETAC) Technical Publication Series, SETAC Press, Pensacola, Florida,
4           pp. 1-15.
5
     December 2005                           2-84            DRAFT-DO NOT QUOTE OR CITE

-------
 i      3.    ROUTES OF HUMAN EXPOSURE TO  LEAD AND
 2      OBSERVED ENVIRONMENTAL CONCENTRATIONS

 3
 4   3.1    EXPOSURE: AIR
 5   3.1.1   Observed Concentrations - Indoor
 6         Given the large amount of time people spend indoors, exposure to lead in dusts and indoor
 7   air can be significant.  For children, dust ingested via hand-to-mouth activity is a more important
 8   source of lead exposure than inhalation (Adgate et al., 1998; Oliver et al., 1999). However, dust
 9   can be resuspended through household activities (e.g., Ferro et al., 2004), thereby posing an
10   inhalation risk as well.  The particle size of "dust" is not well defined, although 50 |im or 75 jim
11   in diameter is sometimes given as an upper limit.
12         Lead in housedust is from a number of different sources. In general  particulate matter can
13   originate inside the home from sources such as smoking and cooking.  However, lead appears to
14   come from sources outside the home (Jones et al., 2000; Adgate et al.,  1998).  A chemical mass
15   balance study in Jersey City, New Jersey observed that crustal sources contributed almost half of
16   the lead in residences, lead-based paint contributed about a third, and deposition of airborne lead
17   contributed the remainder (Adgate et al., 1998). Residential concentrations measured at the
18   Bunker Hill Superfund Site in northern Idaho indicate that the concentration in houses depends
19   primarily on the neighborhood soil concentration (von Lindern et al., 2003). However factors
20   such as household hygiene, the number of adults living in the house, and the number of hours
21   children spend playing outside were also shown to affect concentrations.
22         Living near a smelter or a mine contributes significantly to the lead load in residences
23   (Rieuwerts and Farago, 1995; Rieuwerts et al., 1999).  Homes of mine and  smelter employees
24   tend to have lead levels elevated above those of nearby houses indicating that lead can be
25   transported into homes via workers (Rieuwerts et al., 1999).
26         Renovation and especially old paint removal can greatly increase lead levels inside the
27   home (Mielke et al., 2001; Laxen et al.,  1987; Jacobs,  1998). Removal of exterior paint via
28   power sanding released an estimated 7.4 kg of lead as  dust, causing lead levels inside one house
29   to be well above safe levels (Mielke et al., 2001).
     December 2005                       3-1          DRAFT-DO NOT QUOTE OR CITE

-------
 1          Lead concentrations are likely elevated somewhat in houses of smokers. In a nationwide
 2    (U.S.) study, blood lead levels were 38% higher in children who exhibited high cotinine levels,
 3    which indicate high second hand smoke exposure (Mannino et al., 2003). Lead is present both in
 4    tobacco and tobacco smoke, although lead concentrations in tobacco have fallen with decreases
 5    in the airborne concentration (Mannino et al., 2003). In a study performed in the UK, lead in
 6    house dust tended to be bound to the carbonate or Fe-Mn oxides (Feng and Barratt, 1994).
 7          Concentrations of lead in house dust, school dust, and nursing home dust are shown in
 8    Table 3-1.
 9          Metal-cored candlewicks have posed an additional significant source of indoor lead.
10    The U.S. Consumer Product Safety Commission banned the use of metal-cored candlewicks that
11    contain more than 0.06% lead as of October 15, 2003 (USGS, 2003). However, prior to this
12    time, emissions of lead from metal-core wicks were measured in the range of 0.5-66 jig/hour
13    according to one study (Nriagu and Kim, 2000) and 100-1700 jig/hour according to another
14    study (Wassan et al., 2002). In homes where such candles were burned,  airborne concentrations
15    could have been well above ambient levels.
16          An additional concern is attic dust, or dust found in roof cavities.  Significant deposits of
17    atmospheric lead can build up in these spaces.  This dust can seep into living spaces through
18    ceiling decorative artwork, cracks between the wall and ceiling, electric light fittings, wall vents,
19    or exhaust, roof, and ceiling fans (Davis and Gulson, 2005).  Additionally, renovations, housing
20    additions, ceiling collapses, and storm damage can produce large plumes of attic dust (Davis and
21    Gulson, 2005).
22          Studies comparing lead concentrations in attic dust with house age showed an excellent
23    correlation between lead levels and ambient air concentrations throughout the  lifetime of the
24    house (Chiaradia et al., 1997;  Ilacqua et al., 2003). Attic dust may even  serve  as a proxy for
25    predicting historic ambient concentrations although the resolution on such calculations would be
26    low. Attic dust concentrations measured in Australia were an order of magnitude higher in
27    houses near a copper smelter (Chiaradia et al., 1997). However, isotopic analyses showed that
28    alkyl-lead additives were the dominant source of lead contamination in attic dust overall. The
29    geometric mean concentration of lead measured in attics in Sydney was  1660 ppm near industrial
30    sites, 1173 ppm near semi-industrial sites, 447 ppm in non-industrial sites, and 16 ppm in
31    background, crustal materials (Davis and Gulson,  2005).

      December 2005                         3-2          DRAFT-DO NOT QUOTE OR CITE

-------
                                                Table 3-1.  Concentrations of Lead in Indoor Dust
O

O

to
O
O
Concentration of Lead
       (ppm)
                                Location
Reference
H
6
o
O
H
O
o
H
W
O
O
HH
H
W
503 (mean)
308 (median)
43-13,600
9 (geometric mean)
1.5-48.9
117-362
1598

3025-4140
1283
114-185
1984
348
340
786
1870
1560
726
435

857±91inPM60
1133±119inPM10
                                Edinburgh Scotland
                                Various parts of Denmark
                                UK

                                Helena and Silver Valleys, US
                                (near 2 Pb smelters)
                                Trail Canada (near Pb smelter)
                                Illinois (near Pb smelter)
                                Landskrona, Sweden (near Pb smelter)
                                Pribram, Czech Republic (near Pb smelter)
                                Wales, UK (near a mining site)
                                Halkyn, UK (near a mining site)
                                Shipham, UK (near a mining site)
                                Derbys, UK (near a mining site)
                                Winster, UK (near a mining site)
                                Leadville, US (near a mining site)
                                Pribram, Czech Republic
                                (near a mining site)
                                Jersey City, NJ
                                Jersey City, NJ
Laxenetal. (1987)
Jensen (1992)
Feng and Barratt( 1994)

Schilling and Bain (1988)
Hertzmanetal. (1991)
Kimbrough et al. (1994)
Faragoetal. (1999)
Rieuwerts and Farago (1996)
Gallacher et al. (1984)
Daviesetal. (1985)
Thornton (1988)
Thornton etal. (1990)
Cotter-Howells and Thornton (1991)
Cook etal. (1993)
Rieuwerts and Farago (1996)
Adgate etal. (1998)
Adgate etal. (1998)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)
(cited in Rieuwerts and Farago (1995)

-------
o

O

to
O
O
                                     Table 3-1 (cont'd). Concentrations of Lead in Indoor Dust
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
Concentration of Lead
(ppm)
975inPM53
481inPM250
1693 -6799 in PM53
1407 -4590 in PM250
558±544inTSP
(airborne)
612±518inPM10
(airborne)
547±512inPM2.5
(airborne)
2.29 (ug/ft2)
293.40 (ug/ft2)
20.37 (ug/ft2)
8.30 (ug/ft2)
17.79 (ug/ft2)
2.50 (ug/ft2)
1.92 (ug/ft2)
4.73 (ug/ft2)
8.87 (ug/ft2)
6. 11 (ug/ft2)
3.68 (ug/ft2)
5.91 (ug/ft2)
Location
Public school in Port Pirie, Australia
Public school in Port Pirie, Australia
Houses in Port Pirie, Australia
Houses in Port Pirie, Australia
Nursing homes in Vienna
Nursing homes in Vienna
Nursing homes in Vienna
Boston, MA
Cincinnati, OH
Cincinnati, OH
Rochester, NY
Rochester, NY
Butte, MT
Bingham Creek, UT
Leaville, CO
Magna, UT
Sandy, UT
Midvale, UT
Palmerton, PA
Reference
Oliver etal. (1999)
Oliver etal. (1999)
Oliver etal. (1999)
Oliver etal. (1999)
Komarnicki (2005)
Komarnicki (2005)
Komarnicki (2005)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)
cited in Lanphear et al. (1998)

-------
 1    3.1.2   Observed Concentrations - Outdoor
 2          Widespread emissions from stationary and mobile sources, as well as resuspended soil,
 3    have contributed to elevated airborne lead concentrations as described in Chapter 2. In fact,
 4    airborne lead concentrations in many places throughout the world have been several orders of
 5    magnitude higher than natural background levels for the past seventy years (Miller and
 6    Friedland, 1994). The lowest concentrations measured are at the South Pole, where an average
 7    concentration of 0.076 ng/m3 was recorded (Maenhaut et al., 1979). Even at this remote
 8    location, it is likely that the airborne lead levels are elevated above natural background.  This is
 9    evidenced by lead concentrations in Antarctic snow that have risen from <1 ng/kg in 800 BC to
10    200 ng/kg in the 1960's (Murozumi et al., 1969).
11          Airborne concentrations of lead in the U.S. have fallen dramatically over the last 30 years
12    due largely to the phase out of leaded gasoline additives. Major declines over several  orders of
13    magnitude have been observed not only in urban areas, but also in rural regions and remote
14    locations. Figure 3-1 depicts trends in nationwide data for airborne Pb concentrations for 1985
15    to 2000. The national average was calculated from measurements taken at 228 monitoring sites.
16    Data taken at rural sites throughout the United States since 1979 showed a similar decline
17    (Eldred and Cahill, 1994). Figure 3-2 shows the overall U.S. trend for airborne lead
18    concentrations in relation to the Lead NAAQS, from 1983 to 2002.
19          The United States has not been the only country to see a significant drop in lead
20    concentrations. In the early 1980s, 5% of Europe's urban population was exposed to
21    concentrations above the World Health Organization's recommended limit of 0.5 |ig/m3 for an
22    annual average (Fenger, 1999; WHO, 2000).  By the late 1980s, this value had fallen and there
23    were very few locations reporting concentrations above 0.5 |ig/m3. These areas were primarily
24    near large, uncontrolled metal industries (Fenger, 1999). Measurements made in Bermuda
25    between 1993 and 1994 showed that despite its remote location, airborne lead concentrations had
26    fallen by an order of magnitude since the 1970s and by a factor of four since the 1980s (Huang,
27    1996).  Similarly, measurements taken at the South Pole were routinely below the detection limit
28    in 2000-2001, which indicates a significant improvement in  Antarctic air quality since the 1970s
29    (Arimoto et al., 2004).  Table 3-2 lists literature data on airborne lead concentrations.  It should
30    be noted that concentrations are not directly comparable due to different measurement time
31    scales, sampling equipment, and analytical methods.

      December, 2005                            3-5        DRAFT-DO NOT QUOTE OR CITE

-------
                     1987   1989
1991   1993    1995   1997    1999

    Year
Figure 3-1.  Concentrations of lead throughout the United States. The maximum

            quarterly averages are given for 228 urban, suburban, and rural locations.


Source: U.S. EPA (2003).
           CO
                  - 90% of sites have concentrations below this line
           c
           .2
           *3
           s
           T&eS


           I
           o
           O
                                                                  42

                   1983-02; 94% decrease
                   1993-02: 57% decrease
    10% of sites have concentrations below this line
              1.2


              1,0

              0.8


              0,6


              0.4


              0.2


              0.0
                83  84 85 86 87 88 89 90 91 92  93 94 95  96  97  98 99 00 01 02

                                           Year


Figure 3-2.  Airborne concentrations of lead, averaged across the U.S., shown in relation to

            the current NAAQS, for the years 1983 through 2002.
December, 2005
                 DRAFT-DO NOT QUOTE OR CITE

-------
                           Table 3-2. Airborne Concentrations of Lead
 Airborne Cone
   (ng/m3)
Location
      Reference
 326 ± 15.6 in fine mode
 75.6 ± 5.95 in coarse mode
 330
 52
 31
 64
 30-96270
 12 ±6
 10 ± 8
 64 ±47
 110 ±65
 4-444
 45 ± 16
 49 ±43
 13.1
 15.4-18.9
 6.9
 22 ± 17
 <40
 230-650

 16
 2-1700
 9
 2.5
 0.54-6.34
 28.6

 2.2
 4.6
 15
 0.007-64
 0.04-3.2
 O.032
Urban
Boston, MA
Boston, MA
Clemson, SC
Akron, Oh
Norfolk, VA
Chicago, IL
Range reported in U.S. in lit
Cadiz, Spain
Bari, Italy
Malta, Malta
Eleusis, Greece
Caesarea, Israel
Geneva, Switzerland
Vancouver, BC
Riverside, CA
Los Angeles, CA
San Francisco, CA
Jerusalem, Israel
Yerevan, Armenia
St. Louis, MO
Rural
Packwood, WA
Range reported in U.S. in lit
Whiteface Mountain, NY
IMPROVE network
IMPROVE network
Lake Balaton, Hungary
Remote
Olympic National Park
Glacier National Park
Great Smoky Mt. National Park
Range reported in lit
Bermuda
Antarctica
      Thurston and Spengler (1985)
      Thurston and Spengler (1985)
      Del Delumyea and Kalivretenos (1987)
      Del Delumyea and Kalivretenos (1987)
      Del Delumyea and Kalivretenos (1987)
      Del Delumyea and Kalivretenos (1987)
      Schroeder et al. (1987)
      Torfs and Van Grieken (1997)
      Torfs and Van Grieken (1997)
      Torfs and Van Grieken (1997)
      Torfs and Van Grieken (1997)
      Ereletal. (1997)
      Chiaradia and Cupelin (2000)
      Brewer and Belzer (2001)
      Hm (2002)
      Hm (2002)
      Hm (2002)
      Erel et al. (2002)
      Kurkjianetal. (2002)
      Kim et al. (2005)

      Davidson et al. (1985)
      Schroeder et al (1987)
      Miller and Friedland (1994)
      Eldred and Cahill (1994)
      Malm and Sisler (2000)
      Hlavayetal. (2001)

      Davidson et al. (1985)
      Davidson et al. (1985)
      Davidson et al. (1985)
      Schroeder etal (1987)
      Huang etal. (1996)
      Arimoto et al. (2004)
December, 2005
                      5-7
DRAFT-DO NOT QUOTE OR CITE

-------
                        Table 3-2 (cont'd).  Airborne Concentrations of Lead
       Airborne Cone
        (ng/m3)                 Location                          Reference
                               Near Sources of Lead Emissions
       1700-4000                 Fenceline of a lead smelter, CA,          Kimbrough and Suffett (1995)
                               downwind
       960-1200                 Fenceline of a lead smelter, CA, upwind    Kimbrough and Suffett (1995)
       758                      Jerusalem-Tel Aviv freeway, Israel       Erel et al. (1997)
       400-1000                 Australia roadsides                   Al-Chalabi and Hawker (1997)
       127-173                   Hong Kong roadsides                 Chan et al. (2000)
       46-113                   Gothenburg, Sweden roadsides          Sternbeck et al. (2002)
       27.4                     Birmingham, UK roadside              Harrison et al. (2003)
 1          Concentrations of airborne lead are sometimes several orders of magnitude higher in
 2   urban areas compared to remote regions (Schroeder et al., 1987; Malm and Sisler, 2000). Rural
 3   areas tend to have concentrations falling somewhere between those of urban and remote areas.
 4   Urban populations, which tend to be comprised of low-income peoples and/or minorities, are
 5   exposed to comparatively higher levels of airborne lead.
 6          A substantial amount of data for airborne lead outside of urban areas is available from the
 7   Interagency Monitoring of Protected Visual Environments Network, known as IMPROVE.  This
 8   network started in 1987 and includes 110 sites nationwide (California Air Resources Board,
 9   2000).  The sites are located at rural and remote  sites that are impacted by transport of fine
10   particles which reduce visibility. The network uses a sampling frequency of once every three
11   days, and provides PM2.5 lead concentration data determined by Proton-Induced X-ray Emission.
12          The highest lead concentrations reported by the IMPROVE Network are found in
13   southern California and much of the eastern United States (Malm and Sisler, 2000).  Data
14   indicate that high concentrations of lead in the west are found in the urban corridor along the
15   Cascade Mountains in Washington, Oregon, and Utah at around 5.0 ng/m3 (Malm and Sisler,
16   2000).  Rural concentrations in the eastern U.S. are typically in the range of 6-10 ng/m3.
17   Airborne concentrations throughout the United States outside of cities are shown on an isopleth
18   map in Figure 3-3.
     December, 2005                            3-8       DRAFT-DO NOT QUOTE OR CITE

-------
                             c,
             2.20 Denali Natl. Park, AK    ^
     Figure 3-3.  Concentrations of lead measured in 1995-1998 as recorded by the IMPROVE
                 network. Data are given in ng/m3. (Reprinted from Malm and Sisler, 2000).
 1         The quarterly average airborne concentration of lead cannot exceed 1.5 |ig/m , according
 2   to the 1978 air quality standard. Between September 2001 and September 2002, there were
 3   just four areas in the United States that were not in attainment of this standard: Liberty-Acadia,
 4   MO, Herculaneum, MO, East Helena, MT, and Lame Deer, MT (U.S. Environmental
 5   Protection Agency, 2003).  As of 2004, there were only two areas out of attainment
 6   (www.epa.gov/air/oaqps/greenbk/inte.html).
 7         Some seasonal variability is common for lead concentrations. However, whether seasonal
 8   variability is present depends on precipitation trends, changes in wind direction, and mixing
 9   height variability for a given area. A relative maximum was observed in the winter in the Arctic
10   because of the lack of precipitation during winter months (Heidam, 1986). However, in
11   Bermuda a relative maximum was observed in the summer when winds come predominantly
12   from Africa and Europe (Huang, 1996). Chiaradia and Cupelin (2000) observed no seasonality
13   in lead concentrations in Geneva, Switzerland. Measurements taken at a number of U.S. and
     December, 2005
5-9
DRAFT-DO NOT QUOTE OR CITE

-------
 1   French cities suggest some variation based on seasonal differences in mixing height (Delumyea
 2   and Kalivretenos, 1987).
 3          Measurements made in Riverside, CA after lead was removed from gasoline show diurnal
 4   trends (Singh et al., 2002). Concentrations of lead are high in the morning (6am - 10am) and the
 5   late afternoon (4 pm - 8 pm).  This is probably indicative of heavy traffic, a depressed
 6   atmospheric mixing height in the morning, and advection from Los Angeles traffic. Lead
 7   concentrations in Riverside are significantly lower during midday (10am - 4pm) and night
 8   (8 pm - 6 am).
 9          Concentrations of lead are dependent on height. This is particularly true if lead is emitted
10   at street level from traffic. Measurements performed at roadsides in Hong Kong in 1997 show
11   much higher concentrations at breathing level than at rooftop level (Chan et al., 2000).
12   Similarly, lead concentrations measured at four elevations in Berne, Switzerland show a
13   pronounced decrease with height (Gaelli and Nyffeler,  1987). Some leaded gasoline was still
14   used in Hong Kong and Switzerland during these two studies. Measurements made in an urban
15   street canyon in Lahti, Finland show that concentrations declined by a factor of five between
16   street level (1.5m) and rooftop level (25m) (Vakeva et al.,  1999).
17
18   3.1.3   Observed Concentrations - Occupational
19          Lead concentrations inside work places can also be elevated.  Thus, inhalation of lead
20   during work hours is an additional route of exposure for some subpopulations.
21          Feng and Barratt (1994) measured concentrations of lead in two office buildings in the
22   UK.  In general, concentrations in the office buildings were higher than concentrations in nearby
23   houses. Office dust was concentrated in the organic and residual fractions unlike house dust
24   which was bound to carbonate and Fe-Mn oxides. This indicates that offices and houses may
25   have  different lead sources. Office building lead also tends to be in the coarse mode, unlike
26   house dust that is predominantly in fine particles (Feng and Barratt, 1994).
27          As expected, concentrations of lead tend to be highly elevated within manufacturing
28   facilities for lead-based products (Rieuwerts et al., 1999; Harrison et al.,  1981; Tsai et al., 1997).
29   This is a major exposure route for employees. Measurements taken in a battery manufacturing
30   plant show lead concentrations in floor dust to be 47,700 ppm outside of the assembly plant,
     December, 2005                           3-10      DRAFT-DO NOT QUOTE OR CITE

-------
 1    39,200 ppm inside the assembly plant, and 73,700 ppm in the battery grid storage area
 2    (Rieuwertsetal., 1999).
 3          Airborne concentrations of lead in a battery manufacturing plant, a metallic film capacitor
 4    plant, and a lead powder plant were 140 ± 112 |ig/m3, 281 ± 114  |ig/m3, and 485 ± 245  |ig/m3
 5    respectively (Tsai et al., 1997). Work sites that use mechanical actions such as abrasion, friction,
 6    and cutting typically generate large particles. However, work sites that use high temperature
 7    operations generate small, respirable particles. At the three sites listed above, particle sizes were
 8    predominantly >10 jim in diameter (Tsai et al., 1997).
 9          A Pb-Zn smelter in the UK similarly showed much larger lead particle sizes inside the
10    facility than outside of the facility (Harrison et al., 1981). This may be because concentrations
11    are high enough indoors to coagulate. Floor dusts (<60 jim) taken from each process site in the
12    overall smelting process contained the same lead species as the aerosols emitted from each
13    process, which are discussed in Section 2.2.
14          Residential renovation and paint removal are major sources of lead exposure for both
15    workers and residents.  Dry sanding, abrasive blasting, and burning, welding, or heating surfaces
16    generate highly dangerous levels of lead  (Jacobs,  1998). The geometric mean and maximum
17    concentrations during each of these processes are listed in Table 3-3. Daniels et al. (2001)
18    measured airborne concentrations of lead during exterior paint removal from residences via wet
19    abrasive blasting technology.  The eight-hour, time-weighted average (TWA) exposures
20    measured through personal monitors ranged between 55.1 and 81.5 |ig/m3. Area concentrations
21    were between 20.5 and 26.9 |ig/m3.
22          Lead-based paints were the predominant coating for highway  bridges for many years.
23    Paint removal during renovation projects has also been cited as a major source of lead exposure
24    for workers.  As with residential renovation, lead concentrations during industrial paint removal
25    depend largely on the technology used.  Generally, abrasive blasting techniques are used, which
26    breaks lead coatings into small particles that can be inhaled or ingested if hands are not washed
27    prior to eating or smoking (Chute and Mostaghim, 1991). Vacuum blasting may reduce
28    occupational exposures. Personal monitors worn during vacuum blasting on a bridge registered
29    concentrations between 27 and 76 |ig/m3 with a geometric mean of 55 |ig/m3 (Mickelson and
30    Johnston, 1995).  Concentrations fell to 0.1 and 2 |ig/m3 over an eight-hour TWA eleven meters
31    from the removal processes.

      December, 2005                            3-11       DRAFT-DO NOT QUOTE OR CITE

-------
                   Table 3-3. Airborne Concentrations Surrounding Residential
                                 Lead-Based Paint Abatement
Abatement Technique
Preparation (e.g., carpet removal)
Abrasion
Chemical stripping
Encapsulation
Heat gun
Component replacement
Cleaning
Geometric Mean
Oig/m3)
2
8
3
2
7
o
6
2
Maximum Exposure
Oig/m3)
206
403
476
72
915
121
590
      Source: Jacobs (1998).
 1         Lead concentrations measured in underground gold mines were somewhat elevated, but
 2   comparable to ambient concentrations due to adequate air exchange. Measurements made in a
 3   gold mine in South Africa were highly dependent on the process being undertaken (Annegarn
 4   et al., 1988). Concentrations ranged between 1.4 |ig/m3 and 800 |ig/m3.  A source apportionment
 5   study in a Nevada gold mine measured lead concentrations that averaged 0.21 |ig/m3 (McDonald
 6   et al., 2003).
 7
 8
 9   3.2    EXPOSURE:  SOIL AND ROAD DUST
10         Contaminated soil can be a potential source of lead  exposure for humans. Soil lead can be
11   directly ingested through hand-to-mouth behavior common in children, indirectly ingested
12   through contaminated food, or inhaled when breathing air containing resuspended soil particles.
13         Here we address soil concentrations measured in urban, residential, and industrial areas.
14   Soil lead concentrations in rural and remote areas, agricultural soils, and sediment are addressed
15   in Chapter 8 of this document.
16         The natural background concentration of lead in soil is estimated to be in the range of
17   1-200 ppm with an average of 15 ppm (Zimdahl and Skogerboe, 1977).  It should be noted that
18   soil lead measurements are difficult to compare given the variety of extraction techniques and
19   depths of soil cores analyzed in each study.

     December, 2005                         3-12       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The dominant source of lead to soil is atmospheric deposition both from local sources and
 2    long-range transport.  In general soil in urban and residential areas is contaminated primarily via
 3    atmospheric deposition, direct application of agricultural chemicals, and natural mineral
 4    weathering of parent rock (Paces,  1998). At a local level, soil lead contamination can be derived
 5    from agricultural and food wastes, animal wastes and manure, logging and other wood-cutting
 6    activities, urban refuse, municipal sewage sludge, miscellaneous organic wastes including
 7    excreta, solid wastes from metal manufacturing, coal fly ash and bottom fly ash, peat for
 8    agricultural and fuel uses, wastage of commercial products, mine tailings, and smelter slags and
 9    wastes (Nriagu and Pacyna, 1988). Flaking and peeling of lead-based paint can also be a
10    significant source of soil  lead near old structures (Small et al., 1995; Finkelstein et al., 2003).
11
12    Soil Response Times
13          The retention time for lead in the soil is much longer than it is in the air. The only
14    "removal" mechanisms for  soil lead are resuspension and leaching, the latter of which is known
15    to be a slow process (see  Chapter 2 of this document for details).  The retention time, or the
16    amount of time required to  reduce the soil concentration by half, is estimated to be on the order
17    of hundreds to thousands of years (Dudka and Adriano, 1997). Box model estimates based on
18    data for an agricultural catchment in the Czech Republic predict that steady state concentrations
19    for soil lead will not be achieved for 980 years (Paces, 1998). Modeling efforts by Harris and
20    Davidson (2005) in southern California similarly predict that steady state concentrations of soil
21    lead will not be  achieved for hundreds of years assuming emissions rates stay constant.  The
22    lowest estimates of a response time are given by Miller and Friedland (1994) in the northeastern
23    United States. They estimate that soil lead concentrations in a northern hardwood forest zone
24    will stabilize in just 17 years and soil lead concentrations in a subalpine spruce-fir forest zone
25    will stabilize in  77 years.  A later study in the same region estimated the response times as  60
26    years and 150 years for the  two forests, respectively (Kaste  et al., 2003).
27
28    3.2.1   Urban Background Concentrations of Soil Lead
29          The concentration of soil lead varies significantly throughout urban areas depending on
30    proximity to stationary sources and roadways, and wind direction and speed.
      December, 2005                           3-13       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The major source of lead in urban soils is automotive traffic (Sheets et al., 2001; Mielke,
 2    1993; Sutherland, 2000).  Lead remains in the soil for a long time. Thus, much of the existing
 3    urban soil is likely a remnant of leaded gasoline use.  Soil concentrations decrease both with
 4    depth and distance from roadways.  Furthermore, in several urban areas there was little
 5    correlation between soil lead and the age of nearby houses, which suggests that lead-based paint
 6    is not as significant of a source as automotive lead (Mielke, 1993).
 7          The concentrations of lead in soil depend primarily on the size of the city and the location
 8    within the city (Mielke, 1991, 1993).  Extensive lead studies performed in Baltimore,
 9    New Orleans, and cities throughout Minnesota demonstrate that the highest concentrations of
10    lead are found in the central  sections of the city where traffic and population density are greatest
11    (Mielke, 1991,  1993).  The lowest concentrations are found in the outskirts of these cities and in
12    smaller cities. In all of these studies, the age of housing does not seem to be a major factor,
13    which suggests that the impacts of lead-based paint are dominated by historic emissions of
14    leaded gasoline additives. That the highest concentrations are in the inner city, generally
15    populated by minorities and  the poor, suggests that they are the groups most at risk for lead
16    exposure from contaminated soil.
17          The highest concentrations are observed near major roadways. Surface  soil lead
18    concentrations measured  near a major freeway in Cincinnati, OH, for example, are between
19    59 ppm and 1980 ppm, which is well above background (Turer et al., 2001). These
20    concentrations drop off dramatically with depth. An estimated 40% of lead from exhaust is
21    retained in the nearby soil (Turer et al., 2001).
22          Measurements  of Erel et al. (1997) in Israel show that soil lead concentrations decrease
23    more rapidly with depth near roadways than far from roadways. In a soil profile extracted near a
24    local road, lead concentrations fell by a  factor of 42 between the surface and 3 0-36cm from the
25    surface.  However, far from the roadway, lead concentrations fell by about a factor of 3 between
26    the surface and 30-36cm below the surface.
27          Several authors making measurements during the days of leaded gasoline usage reported
28    elevated lead concentrations in soil that decrease with distance from roadways. For example,
29    Pierson and Brachaczek (1976) report soil lead levels that decrease from >1000 ppm adjacent to
30    the road down to less than 200 ppm at 12.5 m from the roadway edge.  Harris and Davidson
31    (2005) have shown through a mass balance model that elevated lead concentrations in soil are

      December, 2005                           3-14       DRAFT-DO NOT QUOTE OR CITE

-------
 1   likely to remain high for hundreds of years; this is consistent with other studies showing
 2   similarly long residence times in soil (e.g., Dudka and Adriano, 1997).
 3          Soil lead concentrations in urban areas are generally higher than soil lead concentrations
 4   in rural or remote areas. The average concentrations of soil lead in urban areas are shown in
 5   Table 3-4.  In many cases these data are averages over commercial, residential, and public areas,
 6   which cover a wide range of concentrations.

                       Table 3-4. Concentration of Soil Lead in Urban Areas
Location
Springfield, MO
Urban locations throughout Egypt
southern California
central New Orleans, LA
outer New Orleans, LA
suburban New Orleans, LA
Baton Rouge, LA
Monroe, LA
Alexandria, LA
Lafayette, LA
Natchitoches, LA
Reno-Sparks, NV
Manoa, Hawaii
Gainesville, FL
Miami, FL
S oil cone, (ppm)
107 ±8
23-200
65.2, 66.3, 99.4
4-69000
1-24400
2-5650
2-6680
8-1 1600
6-2590
6-8860
6-1430
-10
58 ±27
-16
-93
Depth (cm)
0-15
0-30
0-10
0-2.5
0-2.5
0-2.5
0-2.5
0-2,5
0-2.5
0-2.5
0-2.5
0-1
0-2.5
0-20
0-10
Reference
Sheets etal., 2001
Badawy et al., 2002
Young etal., 2002
Mielke, 1993
Mieike, 1993
Mielke, 1993
Mielke, 1993
Mieike, 1993
Mielke, 1993
Mielke, 1993
Mielke, 1993
Gillies etal., 1999
Sutherland, 2000
Chirenje etal., 2004
Chirenje etal., 2004
 7   3.2.2   Soil Concentrations Near Stationary Sources
 8   Concentrations Near Lead Smelters
 9         Lead in soil is highly elevated near sources of lead emissions. In particular, stationary
10   facilities such as smelters and battery disposal sites can have very high levels of soil lead.
11         Major smelter deposits exist primarily within a 0.5 km radius of the stack (Chatterjee and
12   Banerjee,  1999; Rieuwerts et al., 1999) although some studies observe elevated concentrations of
13   lead as far away as 30  km (Liu, 2003).  Franssens et al. (2004) used isotopic measurements to
14   show that between 50% and 80% of dry depositing lead within a 3-4 km radius of a lead-zinc
15   smelter had an industrial origin.
16         Soil concentrations of lead decrease dramatically with distance from the source and
17   depend greatly on windspeed and direction (Kimbrough and Suffet, 1995; Palacios et al., 2002;
18   Suchara and Sucharova, 2004). Godin et al. (1985) measured soil concentrations that were
     December, 2005
3-15
DRAFT-DO NOT QUOTE OR CITE

-------
1 almost proportional to the inverse of the distance from
2 frequency. Suchara and Sucharova (2004) estimate an
3 with distance from a lead
the source and the square root of the wind
exponential decrease in soil concentration
smelter in the Czech Republic. Data collected within a 14 km radius
4 showed an exponential decrease in soil lead concentration with distance from the source.
5 Exponential decreases in
6 Charter] ee and Banerjee,
soil concentrations have been
1999; Rieuwerts et al., 1999).
suggested elsewhere as well (e.g.,
Examples of data showing decreases in
7 soil concentration with distance from major sources are shown in Table 3-5.
8
9




Table 3-5. Concentrations of Soil Lead with Distance from Lead Smelters
Distance from Smelter
(m)
fenceline
20
30
40
100
123-256
250
400
500
700
1500
3000
5000
10000
20000
30000

23003'4
5657d>1
393741
3253d'1
783d'1
636 ± 522C'8
229dJ
12741
400 ± 20a'5
792e,7
519C'3
242C'3
216.7 ±87.6e'2
110.3±76.4e'2
57.4 ± 24.9e'2
32.9±21.4e'2
Concentration
(ppm, dry weight)
46700 ± 21003'5 12650b'6



312.8±98.7e'2 1800M

20200 ± 11003'5





137C'3



 Note: In cases where multiple transects were sampled, only the downwind transects are shown.

 "Depth sampled was not defined                      ^alacios et al. (2002)
 bSample depth was 0-5 cm                          2Liu (2003)
 "Sample depth was 0-10 cm                         3Godin et al. (1985)
 dSample depth was 0-15 cm                         4Kimbrough and Suffet (1995)
 eSample depth was 0-30 cm                         5Chatterjee and Banerjee (1999)
                                                 6Rieuwerts et al. (1999)
                                                 7Venditti et al. (2000)
                                                 8Young et al. (2002)
December, 2005                              3-16       DRAFT-DO NOT QUOTE OR CITE

-------
 1   As is the case with urban soils, lead concentrations decrease significantly with depth near
 2   industrial sites. Results of Chatterjee and Banerjee (1999) indicate that lead concentrations
 3   remain relatively constant within about 250 meters and decrease with distance after this.
 4   Table 3-6 lists a lead concentration profile measured near a lead smelter in northern France.
 5
 6
                    Table 3-6.  Soil Lead Concentration Profile Measured Near a
                                 Lead Smelter in Northern France
Depth (cm)
0-6
6-9
9-36
36-50
50-70
70-85
85-120
120-165
Soil Horizon
Oi
Oa
Ag
ABg
BAg
Bg
HC2g
HC3g
Soil Cone, (ppm)
2340
4480
383
21.7
18.2
17.1
12.4
10.2
       Source: Denaix et al. (2001).
 7          The species of metals found near smelters vary depending on soil conditions. One study
 8   observed lead in topsoil that was either in the form PbsO^O^Cl or Pb(II) compounds that were
 9   adsorbed onto Fe(II) oxides or associated with clay particles (Batonneau et al., 2004).
10          Other measurements at a site contaminated with automotive battery wastes showed lead
11   species in the soil to be Pb(CO)3, Pb(CO3)2, Pb(OH)2, PbO, and PbSO4 (Pichtel et al., 2000).
12   Additional studies have shown lead contamination bonded to bacteria (Denaix et al., 2001),
13   carbonate (Maskall  and Thornton,  1998; Pichtel et al., 2000; Venditti et al., 2000), sulfide phases
14   (Pichtel et al., 2000; Venditti et al., 2000), organic phases (Pichtel et al., 2000; Venditti et al.,
15   2000) and Fe-Mn oxides (Venditti et al., 2000). The prevalence of carbonate forms in
16   contaminated soil is due to coinciding contamination with calcareous slag wastes (Maskall and
17   Thornton, 1998).
     December, 2005                           3-17       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Lead concentrations do not appear to have decreased surrounding smelters despite the
 2    implementation of pollution controls. A smelter in Slovenia was fitted with protective filters in
 3    1978 (Zadnik, 2004).  Since that time concentrations have fallen dramatically in hay samples and
 4    cow blood within 10 km of the smelter however, soil concentrations did not decrease between
 5    1978 and 2003 (Zadnik, 2004).  Similarly a lead-zinc smelter in British Columbia, Canada was
 6    replaced by a new smelting facility in 1997 (Hilts, 2003). Lead concentrations fell by 50% in
 7    outdoor dustfall, street dust, and indoor dustfall. Airborne concentrations fell by nearly 75%.
 8    However, no statistically significant decline was observed in soil or carpet concentrations
 9    (Hilts, 2003).
10          Lead in soil appears to be associated with relatively small soil particle sizes. Young et al.
11    (2002) observed that the lead  concentration was much higher in the <38 |im size range than in
12    the 300 |im-2mm size range in contaminated soils. This is likely due to the higher specific
13    surface area of smaller soil particles and that lead tends to bond with organic matter and Fe/Al
14    oxides, which can also concentrate in smaller size particles. (Young et al., 2002).
15
16    Concentrations Near Mines
17          Concentrations of lead are highly elevated near mines as well.  Lead and zinc mines in
18    particular have large deposits of lead in nearby soil, but mines used for extracting other metals
19    can also have lead contaminated soil. Mine sites are contaminated by the disposal of mine
20    tailings, acid mine drainage, and atmospheric deposition of airborne emissions (Dudka and
21    Adriano, 1997). Mines in the United States produced an estimated 480 Tg of lead tailings and
22    50 Tg of lead mine wastes between 1910 and 1981 (Dudka and Adriano, 1997).
23          Lead is widely  dispersed surrounding mining sites (Dudka and Adriano, 1997; Rieuwerts
24    and Farago, 1995).  Thus, it is not easy to determine a relationship between distance and soil
25    concentration as is the case with smelting emissions. However, a study of an  abandoned lead-
26    zinc mine in Tyndrum, Scotland located near a river showed that fluvial transport had carried
27    lead contamination at  least as far as 6.5  km although contamination is suspected as far as 25 km
28    downstream (MacKenzie and Pulford, 2002).  Soil concentrations measured near mining sites are
29    shown in Table 3-7.
30
31

      December, 2005                            3-18       DRAFT-DO NOT QUOTE OR CITE

-------
             Table 3-7. Soil Concentrations Measured Near Mining Sites













1
2
3
4
Location
Wales, UK
Halkyn, UK
Shipham, UK
Shipham, UK
Derbys, UK
Winster, UK
Leadville, US
Derbys, UK
Shipham, UK
Pribram, Czech
Republic
Tyndrum,
Scotland
Goldenville,
Canada
Sao Domingos,
Portugal
Dubuque, U.S.
Lead is
Type of
Mine
Pb
Pb-Zn
Zn,Pb
Zn,Pb
Pb
Pb
Pb
Pb
Zn,Pb
Pb
Pb-Zn
Au
Cu
Zn,Pb
Main Period
of Operation
historic,
not specified
1845-1938
1700-1850
1650-1850
18th and
19th cent.
Up to end of
18th cent.
1860s-1960s
18th and
19th cent.
18th and
19th cent.
Depth
(cm)
0-15
0-15
0-15
0-5
0-5
0-5
n.a.
0-15
0-15
18th-20thcent. 0-5
Up to 1862
1869-1927
Pre-Roman-
Roman times
19th century
found in many different
mineral form of galena (Rieuwerts and
in mine spoils,
[Pb5(P04)3Cl],
lead is also
n.a.
n.a.
0-30
0-20
forms near
Mean cone.
(ppm)
1159
1127
7900
2002
5610
7140
1110
1800
7360 (max)
1451
13000
70-120
2694
791
Reference
Gallacher et al. (1984) (taken from
Rieuwerts and Farago, 1995)
Davies et al. (1985) (taken from
Rieuwerts and Farago, 1995)
Mattigod et al. (1986) (taken from
Rieuwerts and Farago, 1995)
Thornton (1988) (taken from
Rieuwerts and Farago, 1995)
Thornton (1990) (taken from
Rieuwerts and Farago, 1995
Cotter-Howells and Thornton
(1991) (taken from Rieuwerts and
Farago, 1995)
Cook et al. (1993) (taken from
Rieuwerts and Farago, 1995
Li and Thornton (1993) (taken from
Rieuwerts and Farago, 1995)
Li and Thornton (1993) (taken from
Rieuwerts and Farago, 1995)
Rieuwerts and Farago (1996)
(taken from Rieuwerts and
Farago, 1995)
MacKenzie and Pulford (2002)
Wong et al. (2002)
Freitas et al. (2004)
Mbila and Thompson (2004)
mining sites. It is commonly found in its
Farago, 1995; Dudka and
found as plumbojarosite [PbFee(SO4)4
lead carbonate [PbCOs]
, leadhillite
[Pb4SO4(CO3)
Adriano, 1997). However,
(OH)i2], pyromorphite
2(OH)2], PbS'Bi2S3, lead
December, 2005
3-19
DRAFT-DO NOT QUOTE OR CITE

-------
 1    oxides, lead silicates, and lead sulfate [PbSO4] (Rieuwerts and Farago, 1995; Mbila and
 2    Thompson, 2004).
 3          Lead tends to be more heavily concentrated in smaller soil grain sizes than in larger grain
 4    sizes (MacKenzie and Pulford, 2002). Results of one study are listed in Table 3-8.  Additionally,
 5    Rieuwerts and Farago (1995) note that soil lead particles are typically larger in mining areas than
 6    in smelting areas.
 7
 8
               Table 3-8. Concentrations of Lead in Soils Grouped by Soil Grain Size
       Size Fraction             Pb cone, of main mine waste       Pb cone, of processing site waste
       >180 urn                          0.91%                            17%
       53-180 nm                          1.5%                             14%
       <53 jun                           4.5%                             18%

       Source: MacKenzie and Pulford (2002).
 9          In addition to soil, lead concentrations in peat have been shown to decrease with depth.
10    Figure 3-4 illustrates two peat profiles sampled near an abandoned lead mine.
11
12    3.2.3   Concentrations of Lead in Road Dust
13          Elevated concentrations of lead in road dust pose an important exposure risk through wind
14    and traffic resuspension as outlined in Chapter 2 of this document.
15          The primary source of lead in road dust is adjacent soil (de Miguel et al., 1997). However
16    traffic emissions, the weathering and corrosion of building materials (de Miguel et al., 1997), and
17    brake pad wear (Garg  et al., 2000) are additional sources.  Between 60-90% of the mass of road
18    dust is comprised of soil particles (Adgate et al., 1998). As mentioned above soil is still an
19    important reservoir for lead emitted from vehicles despite the widespread phase out of leaded
20    gasoline.
21          The concentration of lead in road dust is generally elevated above background. This is
22    particularly true in urban areas.  Additionally, measurements reported in 2003 in the San Joaquin
23    Valley of California show concentrations that are significantly lower than concentrations
24

      December, 2005                           3-20       DRAFT-DO NOT QUOTE OR CITE

-------
         (I)
Pb concentration (mg kg-l)
            10 -
           30 -
           30 -
           40
                   2000
                   	I
          4000
         	I	
6000
 I	
                      I       I
             1.12       1.14       1.16
                   »«Pb/"'Pb «torn rilio
1000  IlOpb
    chronolof y
                           I96t


                           1911
                           111*

                           U76

                           1152
                      1.1!
                                                            Pb concentrtlion (mg kg'1)
                                                         0       100      200      300 »»Pb
                                                       4 —
                                                      10-
                                                      14
                                         1.11
                                                 [
                                                 1.14
                                          I
                                         1 It
                                      Horn ratio
chronology
1*7*
                                                                                   1*31
                                                                                   1)11
                                                                                   1592
                                                                                   1872
                                                                Lit
     Figure 3-4.  The changes in lead concentration with depth in two peat cores.  Core A was
                 taken at a location adjacent to the ore processing area of the abandoned lead
                 mine in Tyndrum, Scotland. Core B was taken 0.5 km from the main mine
                 waste dump at the same site.
     Source:  MacKenzie and Pulford (2002).
1    measured in the same area in 1987 (Chow et al., 2003). Lead data reported in the literature are
2    listed in Table 3-9.
3          Metals in road dust tend to be associated with small size grains.  Measurements of Kuang
4    et al. (2004) show that metals are concentrated in grains smaller than 0.125 mm in diameter.
5          De Miguel et al. (1997) observe a steep gradient in road dust concentrations of lead in the
6    north-south direction in Oslo, Norway.  This indicates that lead concentrations are much higher
7    in the highly urbanized areas and lower in the suburban and residential  areas.  This is consistent
8    with traffic and  building construction, renovation, and weathering of building materials being the
9    dominant source of lead to soil and subsequently road dust (de Miguel et al., 1997).
     December, 2005
                                3-21
                           DRAFT-DO NOT QUOTE OR CITE

-------
                       Table 3-9. The Concentration of Lead in Road Dusts
Cone, of Lead (ppm)
180 ± 14
1927 ±508
536 + 39
57.2 ±27,3
-TOO
1209 ± 170 (PM2.5)
1061 ±155 (PM10)
588 ±688
470 ±524
151 ±124
161 ±31
57 ±28
109 ±74
58 ±73
203 ± 133
43 ±8
101 ±88
Location
Oslo, Norway
Madrid, Spain
Calcutta, India
Beijing, Chinga
Reno-Sparks, NV
Hong Kong
Hong Kong
Honolulu, HI
Honolulu, HI
Honolulu, HI
San Joaquin Valley, CA
San Joaquin Valley, CA
San Joaquin Valley, CA
San Joaquin Valley, CA
San Joaquin Valley, CA
San Joaquin Valley, CA
San Joaquin Valley, CA
Land Use
urban, paved road
urban, paved road
near lead smelter, paved
urban, paved road
urban, paved road
urban, paved road
urban, paved road
urban, paved road
urban, paved road
urban, paved road
urban, paved road
rural, paved road
composite, paved road
agricultural unpaved road
residential unpaved road
staging area soil
unpaved composite
R eference
de Miguel etal., 1997
de Miguel etal., 1997
Chatterjee and Banerjee, 1999
Kuang etal., 2004
Gillies etal., 1999
Ho etal., 2003
Ho etal., 2003
Sutherland etal., 2003
Sutherland etal., 2003
Sutherland etal., 2003
Chow etal, 2003
Chow etal., 2003
Chow etal., 2003
Chow etal,, 2003
Chow etal., 2003
Chow etal., 2003
Chow etal., 2003
 1   3.3    EXPOSURE:  DRINKING WATER
 2         Lead in drinking water is primarily a result of corrosion from lead pipes, lead-based
 3   solder, or brass or bronze fixtures within a residence (Lee et al., 1989; Singley, 1994; Isaac et al.,
 4   1997). Very little lead in drinking water comes from utility supplies. Experiments of Gulson
 5   et al. (1994) confirm this by using isotopic analysis. Tap water analyzed in a public school,
 6   apartments, and free standing houses also indicates that the indoor plumbing is a greater source
 7   of lead in drinking water, even for residences and schools serviced by lead-pipe water mains
 8   (Moir et al., 1996). Ratios of influent lead concentration to tap concentrations in homes in four
 9   municipalities in Massachusetts ranged between 0.17 to 0.69, providing further confirmation that
10   in-home lead  corrosion dominates the trace quantities of lead in municipal water supplies (Isaac
11   etal., 1997).
12         The material of pipes is of great importance when considering how much lead is leached
13   into drinking water.  Copper piping with lead-based solder has largely replaced pure lead piping
14   in the United  States. A survey of 94 water companies nationwide in 1988 revealed that copper
15   pipe was present in 73%  of homes, galvanized pipe was present in 13% of homes, a mixture of
16   galvanized and copper was present in 11% of homes, and plastic pipes were present in 2% of
17   homes (Lee et al., 1989). An analysis of PVC pipes indicated that lead leached from PVC in
18   measurable amounts,(Sadiq et al., 1997).  PVC, which contains -1% lead, increased the tap
19   concentration to an average of 0.017 ± 0.038 mg/L, which is a statistically significant increase
     December, 2005
3-22
DRAFT-DO NOT QUOTE OR CITE

-------
 1    over the influent concentration of 0.011 ± 0.026 mg/L (Sadiq et al., 1997).  Guo et al. (1997)
 2    suggested that lead may be leached from cement-mortar lined pipes in significant quantities if the
 3    cement was made from clinker derived from combusted, hazardous materials.
 4          In addition to piping, lead may leach from faucets. Measurements performed on
 5    12 faucets of different compositions typically found in homes indicated that new cast-brass
 6    faucets leached more lead than any of the other designs (Gardels and Sorg,  1989). Lead levels
 7    were below the detection limit from a plastic faucet.  In houses with copper piping and lead-
 8    based solder, brass fixtures may contribute as much as 50% of lead in drinking water (Lee et al.,
 9    1989).
10          The primary type of solder used in the U.S. is 50-50 tin-lead solder (50% tin, 50% lead).
11    In comparing lead leached from 50-50 tin-lead solder, 95-5 tin-antimony solder, and a liquefied
12    50-50 tin-lead formulation that contained a flux, Birden et al. (1985) showed that the liquefied
13    50-50 formulation leached the most lead into drinking water. The 9-5 tin-antimony solder was
14    the safest with respect to drinking water quality. Measurements of metals leached from four,
15    nonlead-based solders in copper pipes were undertaken by Subramanian et al. (1991,  1994).
16    Of the four solders tested (95-5 Sn-Sb, 96-4 Sn-Ag, 94-6 Sn-Ag, and 95.5-4.0-0.5 Sn-Cu-Ag)
17    all showed that metals (Ag, Cd, Cu, Sb, Sn, and Zn) were leached in small enough quantities to
18    make these solders safe alternatives to lead-based solders.
19          Lead corrosion is essentially an electrochemical process. Electrons may be transferred
20    from the metal (lead) to the solution (drinking water) where the major electron acceptors are
21    dissolved oxygen,  hydrogen ions, or disinfectant residuals (Singley, 1994).  Alternatively, when
22    two different metals are in contact,  there is a difference in potential and the difference in electron
23    demand may increase corrosion (Singley, 1994). In either case, lowering the pH and increasing
24    the dissolved oxygen demand are known to increase rates of corrosion. The corrosion process
25    occurs faster at high temperatures than at low temperatures  (e.g., Thompson and Sosnin, 1985;
26    Lee etal., 1989).
27          The combined pH and alkalinity of water are sometimes described as the aggressiveness
28    of the water and is measured using the Langelier Index. A pH above 8.0 is generally considered
29    safe for lead leaching (e.g., Lee et al., 1989; Frey, 1989).
30          There are conflicting reports on the effect of chlorine in  water.  Chlorine, which is
31    typically used as a disinfectant in municipal supplies, may increase the rate of corrosion by

      December, 2005                            3-23       DRAFT-DO NOT QUOTE OR CITE

-------
 1    providing a source of electron acceptors (Singley, 1994).  However, measurements of Lee et al.
 2    (1989) show an absence of statistically significant change in lead levels with increasing
 3    concentration of free chlorine.  Laboratory tests of Edwards and Dudi (2004) show that chlorine
 4    reacts with soluble Pb2+ to precipitate a red-brown colored lead solid. This solid is highly
 5    insoluble, even at a pH of 1.9 for twelve weeks.  Thus, chlorine may actually lessen the overall
 6    quantity of lead in drinking water. Elevated levels of lead in drinking water in Washington DC
 7    in 2000 were traced to a change from chlorine to chloramine disinfectant.  The red-brown lead
 8    solid does not form in the presence of chloramines, and the data suggest that chloramines
 9    dramatically increase the amount of lead leached from brass (Edwards and Dudi, 2004).
10         Flouridating water does not seem to affect the solubility or reactivity of lead compounds
11    (Urbansky and Schock, 2000).
12         Corrosion inhibitors are sometimes added to water to inhibit scaling or iron precipitation.
13    Zinc orthophosphate in the range of 0.4-0.6 mg/L is an effective inhibitor for lead corrosion
14    (Lee et al., 1989). Results indicate that zinc orthophosphate is more  effective at reducing lead
15    levels than increasing the pH.  Soluble lead release is reduced by up to 70% with the addition of
16    orthophosphate (Edwards and McNeill, 2002). Other proposed corrosion inhibitors such as
17    sodium zinc hexametaphosphate or sodium hexametaphosphate are not effective at reducing lead
18    corrosion (Lee et al., 1989).  In fact, results of McNeill and Edwards (2004) indicate that
19    hexametaphosphate increased the levels of soluble lead in drinking water.  Each milligram per
20    liter of hexametaphosphate increased the lead content by -1.6 mg/L after a 72 hour stagnation
21    period in pure lead pipes (Edwards and McNeill, 2002).
22         The length of time that drinking water remains in a pipe also affects the lead
23    concentration. Thus, a first flush phenomenon is generally observed in the morning after water
24    has stayed in the pipe through the night. An estimated 47% of total leached lead was observed in
25    the first 500 mL of water after  prolonged stagnation (Singh and Manivic, 1991). Gardels and
26    Sorg (1989) demonstrated that 60-75% of total lead leached appeared in the first 125 mL of
27    water after prolonged stagnation. For cold water the peak lead concentrations occurred in the
28    first or second 25 mL sample and decreased exponentially with time  thereafter.  For hot water
29    the peak lead concentration occurred in the second or third 25mL sample before decreasing
30    exponentially (Gardels and Sorg, 1989). In a system where fully flushed water had a lead
31    content of just 1.7 |ig/L, removing just 125 mL of water from the tap every hour kept lead

      December, 2005                           3-24       DRAFT-DO NOT QUOTE OR CITE

-------
1    concentrations elevated (35-52 |ig/L) throughout the day (Gulson et al., 1997). Lytle and Schock
2    (2000) showed a temporary exponential increase in lead concentration with stagnation time
3    before the rate leveled off. After 10 hours of stagnation, approximately 50-70% of the maximum
4    lead concentration had been achieved, although, lead levels continued to increase even after
5    90 hours of stagnation.  Their results are shown in Figures 3-5 and 3-6. It should be noted that
6    the shape of the stagnation-concentration curves was the same for all situations regardless of
7    water quality.
1,00
0.90
0.80
0.70
O.«0
0.50
0.40
0.30
0.20
0.10
0.00
                                  -•- Aged 161-184 d»ys
                              £   -C-Aged 453-162 diyt
                                                                         4  r
                                                                            •«-•
                                                                            I
                                                                            E
                                                                         3  1
                                                                            I
                                                                            1
                                                                            ft
                                                                         2  2
                             0    10   20   30   40   50   60   TO   SO   90
                                                uoc time (h)
                  Impact of stagnation time on lead and dissolved oxygen concentration in lead pipe
                            (13 mm diameter) exposed to softened water in Study A.
     Figure 3-5.  The change in lead concentration vs. stagnation time.
                 (Reprinted from Lytle and Schock, 2000).
     December, 2005
                       3-25
DRAFT-DO NOT QUOTE OR CITE

-------
                       1.00
                       0.90
                       0,80
                       0.70
                    -  0.60
                    §  0.50
                       0.40
                       OJO
                       0.20
                       0.10
                       o.oo H
Aged lol-18-i div
Aged 455-U.2 days
                                     4
                                     J  I
                                        1
                                     2  3
                                        £
                                        c
                                     i
                                10   20   30   40   50   60
                                          Stagnation tune (h>
                       70   80    90
                  Impact of stagnation time on lead and dissolved oxygen concentration in lead pipe
                          (13 mm diameter) exposed to non-softened water in Study A.
     Figure 3-6.  The change in lead concentration vs. stagnation time.
                  (Reprinted from Lytle and Schock, 2000).
 1          Some concentrations of lead in drinking water are shown in Table 3-10. The lead
 2   standard for drinking water was set by the U.S. EPA in 1988 with a maximum allowable limit of
 3   5 ng/L for water entering the distribution system (Frey,  1989).
 4         Lead in drinking water can be either in particulate or soluble form. Lead can be in the
 5   form of aqueous ions or complexes, particularly when pH is low.  Solids are the product of
 6   nonadherent corrosion deposits, eroded pieces of plumbing material, or background
 1   concentrations from the distribution system (Lytle et al., 1993). Lead particles are released when
 8   pH and alkalinity are low, and are typically in the form of hydrocerrusite scales (McNeill and
 9   Edwards,  2004). The lead products of corrosion are CaCO3, PbCO3, Pb3(CO3)2(OH2,
10   Pbio(CO3)6(OH)6O, Pb5(PO4)3OH, and PbO (Lytle et al., 1993; McNeill and Edwards, 2004).
11         Based on the conditions described above, models to  predict drinking water lead
12   concentrations have been proposed (e.g., Clement et al., 2000; Van Der Leer et al., 2002).
     December, 2005
         3-26
DRAFT-DO NOT QUOTE OR CITE

-------
                          Table 3-10.  Tap Water Concentrations of Lead
Water Cone, (ug/L)
20
13
0.70
0.32
16
8
3
2
6
5
1?
7.7
25.0
15.3
11,6
Location
Vancouver, Canada
Vancouver, Canada
Arizona
Mexico/US border
Halifax, Canada
Halifax, Canada
Halifax, Canada
Halifax, Canada
Halifax, Canada
Halifax, Canada
Dharan, Saudi Arabia
Clinton, MA
Gardner, MA
Fall River, MA
New Bedford, MA
Residence Type
Apartments
Houses
Residences
Residences
Houses
Houses
Apartments
Apartments
Public School
Public School
Community sites
Residences
Residences
Residences
Residences
Description
copper or plastic pipes
copper or plastic pipes
-
.
standing water
running water
standing water
running water
standing water
running water
PVC pipes
standing water
standing water
standing water
standing water
Reference
Singh and Manivic, 1991
Singh and Manivic, 1991
Sofuogluetal, 2003
Sofuogluetal., 2003
Moiretal., 1996
Moiretal., 1996
Moiretal., 1996
Moiretal., 1996
Moiretal., 1996
Moiretal., 1996
Sadiqetal., 1997
Isaac eta I., 1997
Isaac eta I., 1997
Isaac eta I., 1997
Isaac eta I., 1997
 1   3.4    EXPOSURE:  FOOD INGESTION
 2         Lead contaminated food is a major route of lead exposure.  In one of the most thorough
 3   studies of lead ingestion in food, Flegal et al. (1990) showed that North Americans ingest an
 4   estimated 50 jig of lead each day through food, beverages, and dust, and approximately 30-50%
 5   of this amount is through food and beverages.  The global average daily intake is about 80
 6   Hg/day from food and 40 |ig/day from drinking water according to estimates made by the UN
 7   Environment Program (Juberg et al., 1997).  In Australia, women between 20 and 39 years of age
 8   ingest between 7.3 and 9.7 ng/day (Gulson et al., 2001b). Infants that are breast-fed take in
 9   approximately 0.73 |ig/day compared to 1.8  |ig/day for formula-fed infants (Gulson et al.,
10   200Ib). Australian children ingest approximately 6.4 |ig/day.  A duplicate diet study shows that
11   most diets contain a large amount of house dust (Manton et al., 2005). Other significant sources
12   of lead in the diet are calcium-supplemented food where calcium is derived from limestone and
13   lead in tin coatings. For U.S. children age 0-12 months,  13-24 months,  2-6 years, and their
14   mothers, the estimated rate of lead ingestion was 1.8 jig/day, 3.3 jig/day, 4.1 |ig/day and
15   7.5  |ig/day respectively (Manton et al., 2005).  This is significantly lower than the value reported
16   above by Flegal et al. (1990), which may reflect the drop in lead emissions since the late 1980s.
17         The primary source of lead in food is atmospheric deposition (Flegal et al., 1990).
18   Overall, anthropogenic aerosols account for  an estimated 40% of lead in food, while the bulk of
     December, 2005
3-27
DRAFT-DO NOT QUOTE OR CITE

-------
 1    the remainder is derived from harvesting, transport, processing, packaging, or preparation (Flegal
 2    et al., 1990; Juberg et al., 1997; Dudka and Miller, 1999). Lead contamination in poultry and
 3    livestock is also primarily atmospheric in origin. Lead deposits on forage or feed, or soil that is
 4    directly ingested (Flegal et al., 1990). Lead concentrations in food increase by a factor of 2 to
 5    12 between harvest and consumption (Flegal et al., 1990). A food production facility in Turkey
 6    was shown to contaminate pasta with lead (Demirozu and Saldamli, 2002). Lead concentrations
 7    in the semolina were between 14.2 and 36.5 ng/g compared with the finished pasta product
 8    where concentrations ranged between 107.1 and 147.6 ng/g (Demirozu and Saldamli, 2002).
 9    A similar increase (from 0.5 ng/g to 230 ng/g) between raw and finished cocoa products has also
10    been observed (Rankin et al., 2005). In this case, contamination seems to occur during shipping
11    and/or processing.
12          Lead concentrations in vegetables may be increased by soil amendments such as mine
13    wastes, slag, or fly ash. Historically, mine tailings were often disposed in streambeds, and this
14    poses an exposure risk when sediments are harvested to boost productivity in gardens (Cobb
15    et al., 2000). Slag is sometimes used for constructing agricultural and forestry roads or for
16    landfill.  This can be an additional source of lead contamination for nearby crops (Bunzl et al.,
17    2001). Fly ash is applied to land infrequently for alkaline adjustment, as cover for landfills,  or to
18    amend agricultural soils.  Elevated lead levels in fly ash can subsequently contaminate crops
19    (Brake et al., 2004).  Although soil contamination may be important on a local scale,
20    atmospheric deposition is, overall, a more significant source of food lead than uptake from soil.
21    For example, more than 52% of the total lead present in citrus fruits was removed  by washing,
22    indicating that surface deposits make up the bulk of lead contamination in unprocessed foods
23    (Caselles, 1998).
24    The concentrations of lead measured in food are shown in Table 3-11. In general, food lead
25    concentrations have decreased as a direct result of the decrease in airborne emissions  of lead
26    from automotive gasoline.  This has been directly shown through measurements performed on
27    vintage wines (Lobinski,  1995; Medina et al., 2000).  The organolead concentration in French,
28    Californian, Australian, and Argentinean wines peaked in 1978 (Lobinski, 1995).  The maximum
29    concentration was -0.5 |ig/L which was 10-100 times higher than lead concentrations in drinking
30    water. Conversely, Medina et al. (2000) observed a peak lead concentration in French wine  in
31

      December, 2005                           3-28       DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
                                     Table 3-11.  The Concentration of Lead in Food Products
          Food
                                   Cone.
                           Location
Description
Reference
to
VO
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
Barley, grain


Barley, grain


Potato tubers, peeled


Potato tubers, peeled


Lettuce


Spinach


Potatoes


Wheat


Rice


Sweet corn


Field corn


Carrots


Onions


Tomatoes


Peanuts


Soybeans


Applesauce, canned


Fruit cocktail, canned


Spinch, fresh


Peaches, canned
   0.4 ppm


   2.0 ppm


  0.21 ppm


  0.89 ppm


  0.19 ppm


  0.53 ppm


  0.03 ppm


  0.02 ppm


  0.01 ppm


  0.01 ppm


  0.01 ppm


  0.05 ppm


  0.04 ppm


  0.03 ppm


  0.01 ppm


  0.04 ppm


8.5 ug/serving


7.1 ug/serving


2.4 ug/serving


6.0 ug/serving
Uncontaminated soil


Zn-Pb smelter contaminated


Uncontaminated soil


Zn-Pb smelter contaminated


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


Edible portion, untreated soil


FDA Total Diet Study


FDA Total Diet Study


FDA Total Diet Study


FDA Total Diet Study
Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Dudka and Miller (1999)


Jubergetal. (1997)


Jubergetal. (1997)


Jubergetal. (1997)


Jubergetal. (1997)

-------
o

O

to
O
O
                                Table 3-11 (cont'd). The Concentration of Lead in Food Products
         Food
                                   Cone.
Location
                                            Description
Reference
H

6
o


o
H

O

o
H
W

O


O
HH
H
W
Pears, canned


Strawberries, fresh


Apple juice, bottled


Wine


Vaccinium vitis-idaea


Vaccinium myrtillus


Rubus chamaemoms


Empetmm hermaphroditum


Leccinum auranticcum


Leccinum sacbrum


Russul vesea


Xerocomus subtomentosus


Suillus luteus


Lactarius trivialis


Lactarius torminosus


Lettuce


Lettuce


Lettuce


Carrots


Carrots
 4.9 ng/serving


 1 . 1 ug/serving


 2.6 ug/serving


 7.7 ug/serving


 0.4-2.3 ppm


 0.7-1.6 ppm


 0.3-4.7 ppm


 0.3-1. 5 ppm


 0.8-2.3 ppm


  1.1-5.2 ppm


  1.1-3.4 ppm


  1.3-3.1 ppm


 2.0-2.3 ppm


  1.1-3.1 ppm


 0.6-3. 5 ppm


 0.65-1.3 ppm


 0.15-0.46 ppm


   0.36 ppm


 0.07-0.28 ppm


<0.02-0.09 ppm
                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Monchegorsk, Russia


                                                            Copenhagen, Denmark


                                                            Copenhagen, Denmark


                                                            Copenhagen, Denmark


                                                            Copenhagen, Denmark


                                                            Copenhagen, Denmark
                  FDA Total Diet Study


                  FDA Total Diet Study


                  FDA Total Diet Study


                  FDA Total Diet Study


                  Berry, near Ni-Cu smelter


                  Berry, near Ni-Cu smelter


                  Berry, near Ni-Cu smelter


                  Berry, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Mushroom, near Ni-Cu smelter


                  Close to lead smelter


                  Removed from lead smelter


                  Background concentration


                  Close to lead smelter


                  Removed from lead smelter
Jubergetal. (1997)


Jubergetal. (1997)


Jubergetal. (1997)


Jubergetal. (1997)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Barcanetal. (1998)


Moseholm et al. (1992))


Moseholm et al. (1992))


Moseholm et al. (1992))


Moseholm etal. (1992))


Moseholm etal. (1992))

-------
o

O

to
O
O
                                Table 3-11 (cont'd).  The Concentration of Lead in Food Products
          Food
                                  Cone.
                           Location
                        Description
                                                                                                                           Reference
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
Carrots


Potatoes


Potatoes


Potatoes


Kale


Kale


Kale


Wine


Breast milk


Infant formula


Baby food


Brassica juncea


Triticum aestivum L.


Basil


Cabbage


Cilantro


Collard greens


Coriander


Ipasote


Lemon balm
 0.02-0.03 ppm


<0.02-0.12ppm


O.02-0.06 ppm


  <0.02 ppm


  1.4-9.3 ppm


 0.58-2.4 ppm


 0.52-0.72 ppm


   65ug/L


  0.55 ug/kg


   1.6 ug/kg


   2.9 ug/kg


  298.3 ppm


   19.2 ppm


   <10 ppm


   <10 ppm


   49 ppm


   12 ppm


   39 ppm


   14 ppm


   20 ppm
Copenhagen, Denmark


Copenhagen, Denmark


Copenhagen, Denmark


Copenhagen, Denmark


Copenhagen, Denmark


Copenhagen, Denmark


Copenhagen, Denmark


       France


      Australia


      Australia


      Australia


    Taihe, China


    Taihe, China


     Chicago, IL


     Chicago, IL


     Chicago, IL


     Chicago, IL


     Chicago, IL


     Chicago, IL


     Chicago, IL
                                                                                     Background concentration


                                                                                     Close to lead smelter


                                                                                     Removed from lead smelter


                                                                                     Background concentration


                                                                                     Close to lead smelter


                                                                                     Removed from lead smelter


                                                                                     Background concentration


                                                                                     Vintage 1990-1995
                                                                                     Indian mustard, near lead smelter


                                                                                     Common wheat, near lead smelter


                                                                                     Edible portion, urban garden


                                                                                     Edible portion, urban garden


                                                                                     Edible portion, urban garden


                                                                                     Edible portion, urban garden


                                                                                     Edible portion, urban garden


                                                                                     Edible portion, urban garden


                                                                                     Edible portion, urban garden
Moseholmetal. (1992))


Moseholmetal. (1992))


Moseholmetal. (1992))


Moseholmetal. (1992))


Moseholmetal. (1992))


Moseholmetal. (1992))


Moseholmetal. (1992))


Medina et al. (2000)


Gulsonetal. (200la)


Gulsonetal. (200la)


Gulsonetal. (200la)


Cui et al. (2003)


Cui et al. (2003)


Finster et al. (2004)


Finster et al. (2004)


Finster et al. (2004)


Finster et al. (2004)


Finster et al. (2004)


Finster et al. (2004)


Finster et al. (2004)

-------
December 2005






OJ
OJ
to


o
H
6
o
0
H
O
O
H
W
O
O
HH
H
W
Table 3-11 (cont'd). The Concentration of Lead in Food Products
Food
Mint
Mustard greens
Parsley
Red chard
Rhubarb
Sage
Swiss chard
Thyme
Carrot
Onion
Radish
Tuna, canned
Sardines, canned
Blue mussel, canned
Balsamic vinegar
Wine vinegar
Tea leaves
Cocoa beans
Cocoa, manufactured
Chocolate products

Cone.
<10 - 60 ppm
<10 ppm
<10 ppm
<10 ppm
<10 - 36 ppm
<10 ppm
22-24 ppm
<10 ppm
10 ppm
21 ppm
12-18 ppm
0. 1 ppm (max.)
0.2 ppm (max.)
0.3 ppm (max.)
15-307 ug/L
36-50 ug/L
0.59-4.49 ppm
0.5 ng/g
230 ng/g
70 ng/g

Location
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL
Chicago, IL


Zhejiang Province, China
Nigeria
Nigeria
Nigeria

Description
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden
Edible portion, urban garden


Commercial tea producing areas


Reference
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Finster et al. (2004)
Lourenco et al. (2004)
Lourenco et al. (2004)
Lourenco et al. (2004)
Ndung'u et al. (2004)
Ndung'u et al. (2004)
Jin et al. (2005)
Rankin et al. (2005)
Rankin et al. (2005)
Rankin et al. (2005)


-------
 1   the early 1950s.  Isotopic analysis indicates that automotive emissions were the dominant source
 2   of lead contamination since 1950.  It is not clear why French wine concentrations decreased
 3   through the late 1970s while automotive emissions were still increasing.
 4
 5
 6   3.5   OTHER ROUTES OF EXPOSURE
 7   3.5.1   Lead-Based Paint
 8          Lead-based paint poses a potential exposure risk due to inhalation during renovation or
 9   demolition projects, or due to ingestion from hand-to-mouth activities and pica, which are
10   common in children. As lead-based paint degrades, it becomes incorporated into house dust,
11   which children ingest at a rate of-100 mg per day (Flegal et al., 1990).  Lead-based paint
12   exposure is one of the most common causes of clinical lead toxicity. Lead-based paint can pose
13   an inhalation risk during renovation and demolition activities. As described in Section 3.1 of this
14   document, renovation projects often involve abrasive blasting techniques to remove old layers of
15   paint.  This breaks the lead into small pieces that are easily inhaled (Chute and Mostaghim, 1991;
16   Mickelson and Johnston, 1995; Jacobs et al., 1998; Mielke et al., 2001). At industrial sites,
17   exposure is limited primarily to workers. However, during residential renovation or abatement
18   projects, residents may be unduly exposed to very high levels of airborne lead.
19
20   3.5.2   Calcium  Supplements
21          Potentially toxic levels  of lead were measured in calcium supplements in studies
22   undertaken in the 1960s through the early 1990s (Scelfo and Flegal, 2000). An analysis of 136
23   different brands of supplements showed that two-thirds of the supplements did not meet the 1999
24   California criteria for acceptable lead levels:  1.5  jig/daily dose of calcium (Scelfo and Flegal,
25   2000). The lowest concentrations were observed  in calcium products that were nonchelated
26   synthesized and/or refined.  These corresponded to antacids and infant formulas.  Antacids and
27   infant formulas had  lead concentrations ranging from below the detection limit to 2.9 jig Pb/g
28   calcium (Scelfo and Flegal, 2000). Natural calcium supplements derived from bonemeal,
29   dolomite,  or oyster shell were  much more likely to be in exceedance of the 1999 standard.
30   Lead levels reported elsewhere showed comparable lead levels in supplements and cow milk
     December 2005                        3-33          DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Juberg et al., 1997). Whole milk, 2% milk, and calcium supplements had lead concentrations in
 2    the range of 1.7-6.7 jig Pb/g calcium, 0.8-9.0 jig Pb/g calcium, and 3.1-6.9 jig Pb/g calcium,
 3    respectively (Juberg et al., 1997).
 4
 5    3.5.3   Glazes
 6          Lead glazes have been commonly used throughout history. Kitchen glassware cannot
 7    have a lead solubility in excess of 2.5-7 |ig/mL according to a 1980 rule by the Food and Drug
 8    Administration (Flegal et al., 1990). However, lead glazes on imported pottery may persist.
 9    Foods with low pH are particularly susceptible to solubilizing lead and contaminating food
10    during storage in lead glazed glassware.
11
12    3.5.4   Miniblinds
13          Some imported vinyl miniblinds form lead dust upon disintegration (Juberg et al., 1997).
14    This exposure route was responsible for several cases of lead poisoning in Arizona and North
15    Carolina in the mid 1990s. Lead stabilizers are not used in vinyl miniblinds manufactured in the
16    U.S. (Juberg etal., 1997).
17
18    3.5.5   Hair Dye
19          The analysis of Mielke et al. (1997) shows that some hair dyes contain lead acetate in the
20    range of 2300-6000 jig Pb/g of product.  This lead can be easily transferred via hand-to-mouth
21    and hand-to-surf ace activity, and an estimated 3-5% of lead acetate can be transferred through
22    the skin. Hair dyes tested in this study contain 3-10 times more lead than is allowable for paint
23    (Mielke etal., 1997).
24
25    3.5.6   Other Potential Sources of Lead Exposure
26          Additional consumer products that may  pose a risk of lead exposure include lead crystal,
27    pool cue chalk (Miller et al., 1996), cosmetics,  and folk  remedies.
28
     December 2005                        3-34          DRAFT-DO NOT QUOTE OR CITE

-------
 1    3.6   MEASUREMENT METHODS
 2          Emissions estimates made through direct measurements can be done either using grab,
 3    periodic, or continuous monitoring. Determining the rate of emissions requires knowing both the
 4    fluid flow rate and the concentration of lead in the air (or water).  Thus it is much easier to
 5    measure emissions from stacks than it is to measure fugitive, diffuse, or nonpoint emissions
 6    (Frey and Small, 2003).
 7          The concentration of lead in air can be measured through several different methods.
 8    Use of filter media is one of the most common methods. Inertial impactors are another method,
 9    and in this case particles are separated by size. An additional method involves mounting a
10    particle separation device in the stack along with gas  flow control and metering equipment.
11    Measurement of the mass collected in the cyclone hopper and the filter follows (Clarke and
12    Bartle, 1998).
13          Sampling of airborne particles to determine quantity and species can be performed via
14    direct-reading instruments, which include optical counters, electrical counters, resonant
15    oscillation aerosol mass monitors, and beta radiation  detection (Koutrakis and Sioutas, 1996).
16    Additionally, particles may be collected in cyclones and denuder systems.
17          Collected particles are  analyzed for lead using x-ray fluorescence analysis (XRF), proton-
18    induced x-ray emission (PIXE), neutron activation analysis (NAA), or atomic absorption (AA)
19    (Koutrakis and Sioutas, 1996).
20          Lead concentrations in soil, food, and other environmental media are determined using
21    similar techniques. Generally substances undergo acid digestion in an HC1 or HNO3 solution
22    before analysis via XRF, PIXE, NAA, or AA.
23
24
25    3.7   SUMMARY
26          Concentrations of lead in all environmental media are highly elevated in urban areas.
27    A comprehensive analysis of multimedia concentrations of lead showed that people in cities,
28    especially in poor and minority-dominated neighborhoods, are the most at risk for lead exposure
29    (Chadhaetal., 1998).
30          The highest air, soil, and road dust concentrations are found near major lead sources such
31    as smelters, mines, and heavily trafficked roadways. Airborne concentrations have declined

      December 2005                        3-35          DRAFT-DO NOT QUOTE OR CITE

-------
 1   dramatically with the phase out of leaded gasoline. Soil concentrations have remained relatively
 2   constant.
 3          Drinking water is susceptible to lead contamination primarily through leaching from
 4   pipes, solder, and faucets. Water that has been stagnant in pipes, been disinfected with
 5   chloramines, has a low pH, or has a low alkalinity are particularly high risk for leaching lead into
 6   drinking water.
 7          Lead-contaminated food is a major exposure route. Deposition of airborne lead and house
 8   dust are the major sources of lead in food. Significant quantities of lead are ingested by certain
 9   populations every day.
10          Other sources of lead exposure vary in their prevalence and potential risk. These include
11   lead-based paint, calcium supplements, lead-based glazes, some kinds of miniblinds, hair dye,
12   and other consumer products.
     December 2005                        3-36          DRAFT-DO NOT QUOTE OR CITE

-------
  1     3.8    REFERENCES

  2     Adgate, J. L.; Willis, R. D.; Buckley, T. I; Chow, J. C; Watson, J. G.; Rhoads, G. G.; Lioy, P. J. (1998) Chemical
  3            mass balance source apportionment of lead in house dust. Environ. Sci. Technol. 32: 108-114.
  4     Al-Chalabi, A. S.; Hawker, D. (1997) Response of vehicular lead to the presence of street dust in the atmospheric
  5            environment of major roads. Sci. Total Environ. 206: 195-202.
  6     Annegarn, H. J.; Zucchiatti, A.; Sellschop, J. P. F.; Kusko, B. (1988) Composition and size of dust in a gold mine
  7            atmosphere. J. Mine Vent. Soc. S. Afr. 41: 1-10.
  8     Arimoto, R.; Schloesslin, C.; Davis, D.; Hogan, A.; Grube, P.; Fitzgerald, W.; Lamborg, C. (2004) Lead and
  9            mercury in aerosol particles collected over the South Pole during ISCAT-2000. Atmos. Environ.
10            38: 5485-5491.
11     Badawy, S. H.; Helal, M. I. D.; Chaudri, A. M; Lawlor, K.; McGrath, S. P. (2002)  Soil solid-phase controls lead
12            activity in soil solution. J. Environ. Qual. 31: 162-167.
13     Barcan, V. S.; Kovnatsky, E. F.; Smetannikova, M. S. (1998) Absorption of heavy metals in wild berries and edible
14            mushrooms in an area affected by smelter emissions. Water Air SoilPollut. 103: 173-195.
15     Batonneau, Y.; Bremard, C.; Gengembre, L.; Laureyns, J.; Le Maguer, A.; Le Maguer, D.; Perdrix, E.; Sobanska, S.
16            (2004) Speciationof PM10 sources of airborne nonferrous metals within the 3-km zone of lead/zinc smelters.
17            Environ. Sci. Technol. 38: 5281-5289.
18     Birden, H. H., Jr.; Calabrese, E. J.; Stoddard, A. (1985) Lead dissolution from soldered joints. J. Am. Water Works
19            Assoc. 77: 66-70.
20     Brake, S. S.; Jensen, R. R.; Mattox, J. M. (2004) Effects of coal fly ash amended soils on trace element uptake in
21            plants. Environ. Geol. 45: 680-689.
22     Brewer, R.; Belzer, W. (2001) Assessment of metal concentrations in atmospheric particles from Burnaby Lake,
23            British Columbia, Canada. Atmos. Environ. 35: 5223-5233.
24     Bunzl, K.; Trautmannsheimer, M.; Schramel, P.; Reifenhauser, W. (2001) Availability of arsenic,  copper, lead,
25            thallium, and zinc to various vegetables grown in slag-contaminated soils. J. Environ. Qual. 30: 934-939.
26     Caselles, J. (1998) Levels of lead and other metals in citrus alongside a motor road. Water Air and Soil Pollut.
27            105:593-602.
28     Chadha, A.; McKelvey, L. D.; Mangis, J. K. (1998) Targeting lead in the multimedia environment in the continental
29            United States. J. Air Waste Manage. Assoc. 48: 3-15.
30     Chan, L. Y.; Kwok, W. S.; Chan, C. Y. (2000) Human exposure to respirable suspended paniculate and airborne
31            lead in different roadside microenvironments. Chemosphere 41: 93-99.
32     Chatterjee, A.; Banerjee, R. N. (1999) Determination of lead and other metals in a residential area of greater
33            Calcutta. Sci. Total Environ. 227: 175-185.
34     Chiaradia, M.; Cupelin, F. (2000)  Behaviour of airborne lead and temporal variations of its source effects in Geneva
35            (Switzerland): comparison of anthropogenic versus natural processes. Atmos. Environ. 34: 959-971.
36     Chiaradia, M.; Chenhall, B. E.; Depers, A. M.; Gulson, B. L.; Jones, B. G. (1997) Identification of historical lead
3 7            sources in roof dusts and recent lake sediments from an industrialized area: indications from lead isotopes.
38            Sci. Total Environ. 205: 107-128.
39     Chirenje, T.; Ma, L. Q.; Reeves, M.; Szulczewski, M. (2004) Lead distribution in near-surface soils of two Florida
40            cities: Gainesville and Miami. Geoderma 119: 113-120.
41     Chow, J. C.; Watson, J. G.; Ashbaugh, L. L.; Magliano, K. L. (2003) Similarities and differences inPM10 chemical
42            source profiles for geological dust from the San Joaquin Valley, California. Atmos. Environ. 37: 1317-1340.
43     Chute, D.  O.; Mostaghim, N. L. (1991) Protecting workers from lead. A review of regulations and practices. J. Prot.
44            Coat. Linings 8(4): 36-43.
45     Clarke, A. G.; Bartle, G. (1998) Paniculate emissions by extractive sampling. In: Clark, A. G., ed. Industrial air
46            pollution monitoring. New York, NY: Chapman & Hall; pp. 33-60. (Environmental management senes:
47            v. 8).
48     Clement, M.; Seux, R.; Rabarot, S. (2000) A practical model for estimating total lead intake from drinking water.
49            Water Res. 34: 1533-1542.
50     Cobb, G. P.; Sands, K.; Waters, M.; Wixson, E.G.; Dorward-King, E. (2000) Accumulation of heavy metals by
51            vegetables grown in mine wastes. Environ. Toxicol. Chem. 19: 600-607.
52     Cook, M.; Chappell, W. R.; Hoffman, R. E.; Mangione, E. J. (1993) Assessment of blood lead levels in children
53            living in a historic mining and smelting community. Am. J. Epidemiol. 137: 447-455.
54     Cotter-Howells, J.; Thornton, I. (1991) Sources and pathways of environmental lead to children in a Derbyshire
55            mining village. Environ. Geochem. Health 13: 127-135.


       December 2005                            3-37            DRAFT-DO NOT QUOTE OR CITE

-------
  1     Cui, Y.; Wang, Q.; Dong, Y.; Li, H. (2003) Elemental sulfur effects on Pb and Zn uptake by Indian mustard and
  2            winter wheat. J. Environ. Sci. (China) 15: 836-841.
  3     Daniels, A. E.; Kominsky, J. R.; Clark, P. J. (2001) Evaluation of two lead-based paint removal and waste
  4            stabilization technology combinations on typical exterior surfaces. J. Hazard. Mater. 87: 117-126.
  5     Davidson, C. I.; Goold, W. D.; Mathison, T. P.; Wiersma, G. B.; Brown, K. W.; Reilly, M. T. (1985) Airborne trace
  6            elements in Great Smokey Mountains, Olympic and Glacier National Parks. Environ. Sci. Technol.
  7            19:27-35.
  8     Davies, B. E.; Elwood, P. C.; Gallacher, J.; Ginnever, R. C. (1985) The relationships between heavy metals in
  9            garden soils and house dusts in an old lead mining area of North Wales, Great Britain. Environ. Pollut. Ser.
10            B 9: 255-266.
11     Davis, J. J.; Gulson, B. L. (2005) Ceiling (attic) dust: a "museum" of contamination and potential hazard. Environ
12            Res. 99: 177-194.
13     De Miguel, E.; Llamas, J. F.; Chacon, E.; Berg, T.; Larssen, S.; Royset, O.;  Vadset, M. (1997) Origin and patterns of
14            distribution of trace elements in street dust: unleaded petrol and urban lead. Atmos. Environ. 31: 2733-2740.
15     Del Delumyea, R.; Kalivretenos, A. (1987) Elemental carbon and lead content of fine particles from American and
16            French cities of comparable size and industry, 1985. Atmos. Environ. 21: 1643-1647.
17     Demirozu, B.; Saldamli, I. (2002) Metallic contamination problem in a pasta production plant. Turk. J.  Eng. Env.
18            Sci. 26: 361-365.
19     Denaix, L.; Semlali, R. M.; Douay, F. (2001) Dissolved and colloidal transport of Cd, Pb, and Zn in a silt loam soil
20            affected by atmospheric industrial deposition. Environ. Pollut. 114: 29-38.
21     Dudka, S.; Adriano, D.  C. (1997) Environmental impacts of metal ore mining and processing: a review. J. Environ.
22            Qual.  26: 590-602.
23     Dudka, S.; Miller, W. P. (1999) Accumulation of potentially toxic elements in plants and their transfer  to human
24            food chain. J. Environ. Sci. Health B 34(4): 681-708.
25     Edwards, M.; Dudi, A. (2004) Role of chlorine and chloramine in corrosion of lead-bearing plumbing materials.
26            J. Am. Water Works Assoc. 96: 69-81.
27     Edwards, M.; McNeill, L. S. (2002) Effect of phosphate inhibitors on lead release from pipes. J. Am. Water Works
28            Assoc. 94: 79-90.
29     Eldred, R. A.; Cahill, T. A. (1994) Trends in elemental concentrations of fine particles at remote sites in the United
30            States of America. Atmos. Environ. 28: 1009-1019.
31     Erel, Y.; Veron, A.; Halicz, L. (1997) Tracing the transport of anthropogenic lead in the atmosphere and in soils
32            using isotopic ratios. Geochim. Cosmochim.  Acta 61: 4495-4505.
33     Erel, Y.; Axelrod, T.; Veron, A.; Mahrer, Y.; Katsafados, P.; Dayan, U. (2002) Transboundary atmospheric lead
34            pollution. Environ. Sci. Technol. 36: 3230-3233.
35     Farago, M. E.; Thornton, I.; White, N. D.; Tell, I.; Martensson, M.-B. (1999) Environmental impacts of a secondary
36            lead smelter in Landskrona, southern Sweden. Environ. Geochem. Health 21: 67-82.
37     Feng, Y.; Barratt, R. S.  (1994) Lead and cadmium composition in indoor dust. Sci. Total Environ. 152: 261-267.
3 8     Fenger, J. (1999) Urban air quality. Atmos. Environ.  33: 4877-4900.
39     Ferro, A. R.; Kopperud, R. J.; Hildemann, L. M. (2004) Source strengths for indoor human activities that resuspend
40            paniculate matter. Environ. Sci. Technol. 38: 1759-1764.
41     Finkelstein, M. E.; Gwiazda, R. H.; Smith, D. R. (2003) Lead poisoning of seabirds: environmental risks from
42            leaded paint at a decommissioned military base. Environ. Sci. Technol. 37: 3256-3260.
43     Finster, M. E., Gray, K. A.; Binns, H. J. (2004) Lead levels of edibles grown in contaminated residential soils:
44            a field survey. Sci. Total Environ.  320: 245-257.
45     Flegal, A. R.;  Smith, D. R.; Elias, R. W. (1990) Lead contamination in food. In: Nriagu, J. O.; Simmons, M. S., eds.
46            Food contamination from environmental sources. New York, NY: John Wiley & Sons, Inc.; pp. 85-120.
47     Franssens, M.; Flament, P.; Deboudt, K.; Weis, D.; Perdrix, E. (2004) Evidencing lead deposition at the urban scale
48            using "short-lived" isotopic signatures of the source term (Pb-Zn refinery). Atmos. Environ. 38: 5157-5168.
49     Freitas, H.; Prasad, M. N. V.; Pratas, J. (2004) Plant community tolerant to trace elements growing on the degraded
50            soils of Sao Domingos mine in the south east of Portugal: environmental implications. Environ. Int.
51            30:65-72.
52     Frey, M. M. (1989) The AWWA lead information survey: a final report. J. Am. Water Works Assoc. 81: 64-68.
53     Frey, H. C.;  Small, M. J. (2003) Integrated environmental assessment, Part I: estimating emissions. J. Ind. Ecol.
54            7:9-11.
55     Gaelli, B.C.; Nyffeler, U. P. (1987) Height dependence of heavy metal size distribution and concentration on
56            aerosols. J. Aerosol Sci. 18: 813-816.
       December 2005                             3-38            DRAFT-DO NOT QUOTE OR CITE

-------
  1     Gallacher, J. E. I; Elwood, P. C.; Phillips, K. M; Davies, B. E.; Jones, D. T. (1984) Relation between pica and
  2            blood lead in areas of differing lead exposure. Arch. Dis. Child. 59: 40-44.
  3     Gardels, M. C.; Sorg, T. J. (1989) A laboratory study of the leaching of lead from water faucets. J.- Am. Water
  4            Works Assoc. 81(7): 101-113.
  5     Garg, B. D.; Cadle, S. H.; Mulawa, P. A.; Groblicki, P. J. (2000) Brake wear paniculate matter emissions. Environ.
  6            Sci. Technol. 34:  4463-4469.
  7     Gillies, J. A.; O'Connor, C. M.; Mamane, Y.; Gertler, A. W. (1999) Chemical profiles for characterizing dust
  8            sources in an urban area, western Nevada, USA. In: Livingstone, I., ed. Aeolian geomorphology: papers
  9            from the 4th international conference on aeolian research;  1998; Oxford, United Kingdom. Z. Geomorphol.
10            116(suppl.): 19-44.
11     Godin, P. M.; Feinberg, M. H.;  Ducauze, C. J. (1985) Modelling of soil contamination by airborne lead and
12            cadmium around  several emission sources. Environ. Pollut. Ser. B 10: 97-114.
13     Gulson, B. L.; Law, A. J.; Korsch, M. J.; Mizon, K. J. (1994) Effect of plumbing systems on lead content of drinking
14            water and contribution  to lead body burden. Sci. Total Environ. 144: 279-284.
15     Gulson, B. L.; James, M.; Giblin, A. M.; Sheehan, A.; Mitchell, P.  (1997) Maintenance of elevated lead levels in
16            drinking water from occasional use and potential impact on blood leads in children. Sci. Total Environ.
17            205:271-275.
18     Gulson, B. L.; Mizon, K. J.; Korsch, M. J.; Mahaffey, K. R.; Taylor, A. J. (2001a) Dietary intakes of selected
19            elements from longitudinal 6-day duplicate diets for pregnant and nonpregnant subjects and elemental
20            concentrations of breast milk and infant formula. Environ.  Res. 87: 160-174.
21     Gulson, B. L.; Mizon, K. J.; Palmer, J. M.; Patison, N.; Law, A. J.; Korsch, M. J.; Mahaffey, K. R.; Donnelly, J. B.
22            (200 Ib) Longitudinal study of daily intake and excretion of lead in newly born infants. Environ. Res.
23            85:232-245.
24     Guo, Q. (1997) Increases  of lead and chromium in drinking water from using cement-mortar-lined pipes: initial
25            modeling and assessment. J. Hazard. Mat. 56: 181-213.
26     Harris, A. R.; Davidson, C. I. (2005) The role of resuspended soil in lead flows in the California South Coast Air
27            Basin. Environ. Sci. Technol. 39: 7410-7415.
28     Harrison, R. M.; Williams, C. R.; O'Neill, I. K. (1981) Characterization of airborne heavy-metals within a primary
29            zinc-lead smelting works. Environ.  Sci. Technol. 15: 1197-1204.
30     Harrison, R. M.; Tilling, R.; Callen Romero, M. S.; Harrad, S.; Jarvis, K. (2003) A study of trace metals and
31            polycyclic  aromatic hydrocarbons in the roadside environment. Atmos. Environ. 37: 2391-2402.
32     Heidam, N. Z. (1986) Trace metals in the Arctic aerosol. In: Nriagu, J. O.; Davidson, C. I., eds. Toxic metals in the
33            atmosphere. New York, NY: John Wiley & Sons; pp. 267-293. (Advances in environmental science and
34            technology: v. 17).
35     Hertzman,  C.; Ward, H.; Ames, N.; Kelly, S.; Yates, C. (1991) Childhood lead exposure in trail revisited.  Can. J.
36            Public Health 82: 385-391.
37     Hilts, S. R. (2003) Effect of smelter emission reductions on children's blood lead levels. Sci. Total Environ.
38            303:51-58.
3 9     Hlavay, J.; Polyak, K.; Weisz, M. (2001) Monitoring of the natural environment by chemical speciation of elements
40            in aerosol and sediment samples. J.  Environ. Monit. 3: 74-80.
41     Ho, K. F.; Lee,  S. C.; Chow, J.  C.; Watson, J. G. (2003) Characterization of PM10 and PM25 source profiles for
42            fugitive dust in Hong Kong. Atmos. Environ.  37: 1023-1032.
43     Huang, S. L.; Arimoto, R.; Rahn, K. A. (1996) Changes in atmospheric lead and other pollution elements at
44            Bermuda. J. Geophys. Res. [Atmos.] 101: 21,033-21,040.
45     Hui, C. A. (2002) Concentrations of chromium, manganese, and lead in air and in avian eggs. Environ. Pollut.
46            120:201-206.
47     Ilacqua, V.; Freeman, N. C. J.; Fagliano, J.;  Lioy, P. J. (2003) The  historical record of air  pollution as defined by
48            attic dust. Atmos. Environ. 37: 2379-2389.
49     Isaac, R. A.; Gil, L.; Cooperman, A. N.; Hulme, K.; Eddy,  B.; Ruiz, M.; Jacobson, K.; Larson, C.; Pancorbo, O. C.
50            (1997) Corrosion in drinking water distribution systems: a major contributor of copper and lead to
51            wastewaters and effluents. Environ. Sci. Technol. 31: 3198-3203.
52     Jacobs, D. E.  (1998) Occupational exposures to lead-based paint in structural steel demolition and residential
53            renovation work.  Int. J. Environ. Pollut. 9: 126-139.
54     Jin, C. W.; Zhang,  S. J.; He, Y.  F.; Zhou, G. D.; Zhou, Z. X. (2005) Lead contamination in tea garden soils and
55            factors affecting its bioavailability. Chemosphere 59: 1151-1159.
       December 2005                             3-39            DRAFT-DO NOT QUOTE OR CITE

-------
  1     Jones, N. C.; Thornton, C. A.; Mark, D.; Harrison, R. M. (2000) Indoor/outdoor relationships of paniculate matter in
  2            domestic homes with roadside, urban and rural locations. Atmos. Environ. 34: 2603-2612.
  3     Juberg, D. R.; Kleiman, C. F.; Kwon, S. C. (1997) Position paper of the American Council on Science and Health:
  4            lead and human health. Ecotoxicol. Environ. Saf. 38: 162-180.
  5     Kaste, I; Friedland, A.; Stump, S. (2003) Using stable and radioactive isotopes to trace atmospherically deposited
  6            Pb in montane forest soils. Environ. Sci. Technol.  37: 3560-3567.
  7     Kim, E.; Hopke, P. K.; Pinto, J. P.; Wilson, W. E. (2005) Spatial variability of fine particle mass, components, and
  8            source contributions during the regional air pollution study in St. Louis. Environ. Sci. Technol.
  9            39:4172-4179.
10     Kimbrough, D. E.; Suffet, I. H. (1995) Off-site forensic determination of airborne elemental emissions by multi-
11            media analysis: a case study at two secondary lead smelters. Environ. Sci. Technol. 29: 2217-2221.
12     Kimbrough, R. D.; LeVois, M.; Webb, D. R.  (1994) Management of children with slightly elevated blood lead
13            levels. Pediatrics 93: 188-191.
14     Komarnicki, G. J. K. (2005) Lead and Cadmium in indoor air and the urban environment. Environ. Pollut.
15            136:47-61.
16     Koutrakis, P.; Sioutas, C. (1996) Physico-chemical properties and measurement of ambient particles. In: Wilson, R.;
17            Spengler, J. D., eds. Particles in our air: concentrations and health effects. Cambridge, MA: Harvard
18            University Press; pp 15-39.
19     Kuang, C.; Min, H.; Neumann, T.; Norra, S.;  Stuben, D. (2004) Chemical composition of urban street sediments and
20            its sources. J. China Univ. Geosci. 15: 75-83.
21     Kurkjian, R.; Dunlap, C.; Flegal, A. R. (2002) Lead isotope tracking of atmospheric response to post-industrial
22            conditions in Yerevan, Armenia. Atmos. Environ.  36: 1421-1429.
23     Lanphear, B. P.; Matte, T. D.; Rogers, J.; Clickner, R. P.; Dietz, B.; Bornschein, R. L.; Succop, P.; Mahaffey, K. R.;
24            Dixon, S.; Galke, W.; Rabinowitz, M.; Farfel, M.; Rohde, C.;  Schwartz, J.; Ashley, P.; Jacobs, D. E. (1998)
25            The contribution of lead-contaminated house dust and residential soil to children's blood lead levels.
26            Environ. Res. 79: 51-68.
27     Laxen, D. P. H.; Raab, G. M.; Fulton, M. (1987) Children's blood lead and exposure to lead in household dust and
28            water - a basis for an environmental standard for lead in dust. Sci. Total Environ. 66: 235-244.
29     Lee, R. G.; Becker, W. C.; Collins, D. W. (1989) Lead at the tap: sources and control. J.  Am. Water Works Assoc.
30            81:52-62.
31     Li, X.; Thornton, I. (1993) Multi-element contamination of soils and plants in old mining areas, UK. Appl.
32            Geochem. Suppl. 2: 51-56.
33     Liu, Z. P. (2003) Lead poisoning combined with cadmium in sheep and horses in the vicinity of non-ferrous metal
34            smelters. Sci. Total Environ. 309: 117-126.
3 5     Lobinski, R. (1995) Organolead compounds in archives of environmental pollution. Analyst (Cambridge, U.K.)
36            120:615-621.
37     Lourenco, H. M.; Afonso, C.; Martins, M. F.; Lino, A. R.;  Nunes, M L. (2004) Levels of toxic metals in canned
38            seafood. J. Aquat. Food Prod. Technol. 13: 117-125.
39     Lytle, D. A.; Schock, M. R.  (2000) Impact of stagnation time on metal dissolution from plumbing materials in
40            drinking water. Aqua 49: 243-257.
41     Lytle, D. A.; Schock, M. R.; Dues, N. R.; Clark, P. J. (1993) Investigating the preferential dissolution of lead from
42            solder particulates. J. Am. Water Works Assoc. 85: 104-110.
43     MacKenzie, A. B.; Pulford, I. D. (2002) Investigation of contaminant metal dispersal from a disused mine site at
44            Tyndrum, Scotland, using concentration gradients and stable Pb isotope ratios. Appl. Geochem.
45            17:  1093-1103.
46     Maenhaut, W.; Zoller, W. H.; Duce, R. A.; Hoffman, G. L. (1979) Concentration and size distribution of paniculate
47            trace elements in the south polar atmosphere. J. Geophys. Res. 84: 2421-2431.
48     Malm, W. C.; Sisler, J. F. (2000) Spatial patterns of major aerosol species and selected heavy metals in the United
49            States. Fuel Process. Technol. 65: 473-501.
50     Mannino, D. M.; Albalak, R.; Grosse,  S.; Repace, J. (2003) Second-hand smoke exposure and blood lead levels in
51            U.S. children. Epidemiology 14: 719-727.
52     Manton, W. I.; Angle, C. R.; Krogstrand, K. L. S. (2005) Ongin of lead in the United States diet. Environ.  Sci.
53            Technol.: 10.1021/es051145e.
54     Maskall, J. E.; Thornton, I. (1998) Chemical partitioning of heavy metals in soils, clays and rocks at historical lead
55            smelting sites. Water Air Soil Pollut. 108: 391-409.
       December 2005                             3-40            DRAFT-DO NOT QUOTE OR CITE

-------
  1     Mattigod, S. V.; Page, A. L.; Thornton, I. (1986) Identification of some trace metal minerals in a mine-waste
  2            contaminated soil. Soil Sci. Soc. Am. J. 50: 254-258.
  3     Mbila, M. O.; Thompson, M. L. (2004) Plant-available zinc and lead in mine spoils and soils at the Mines of Spain,
  4            Iowa. J. Environ. Qual. 33: 553-558.
  5     McDonald, J. D.; Zielinska, B.; Sagebiel, J. C.; McDaniel, M. R.; Mousset-Jones, P. (2003) Source apportionment
  6            of airborne fine paniculate matter in an underground mine. J. Air Waste Manage. Assoc. 53: 386-395.
  7     McNeill, L. S.;  Edwards, M. (2004) Importance of Pb and Cu paniculate species for corrosion control. J. Environ.
  8            Eng.  130: 136-144.
  9     Medina, B.; Augagneur, S.; Barbaste, M.; Grouset, F. E.; Buat-Meard, P. (2000) Influence of atmospheric pollution
10            on the lead content of wines. Food Addit. Contam. 17: 435-445.
11     Mickelson, R. L.; Johnston, O. E. (1995) Lead exposure during removal of lead-based paint using vacuum blasting.
12            J. Prot. Coat. Linings 12(2): 78.
13     Mielke, H. W. (1991) Lead in residential soils: background and preliminary results of New Orleans. In: Adriano,
14            D. C., ed. Metals in soils, waters, plants and animals: proceedings of an international conference; April
15            1990; Orlando, FL. Water Air Soil Pollut. 57-58: 111-119.
16     Mielke, H. W. (1993) Lead dust contaminated USA communities: comparison of Louisiana and Minnesota.
17            Appl. Geochem. Suppl. 2: 257-261.
18     Mielke, H. W.;  Taylor, M.  D.; Gonzales, C. R.; Smith, M. K.; Daniels, P.  V.; Buckner, A. V. (1997) Lead-based hair
19            coloring products: too hazardous for household use. J. Am. Pharm. Assoc. 37: 85-89.
20     Mielke, H. W.;  Powell, E. T.; Shah, A.; Gonzales, C. R.; Mielke, P. W. (2001) Multiple metal contamination from
21            house paints: consequences of power sanding and paint scraping in New Orleans. Environ. Health Perspect.
22            109:973-978.
23     Miller, E. K.; Friedland, A. J. (1994) Lead migration in forest soils: response to changing atmospheric inputs.
24            Environ. Sci. Technol. 28: 662-669.
25     Miller, M. B.; Curry, S. C.; Kunkel, D. B.; Arreola, P.; Arvizu, E.; Schaller, K.; Salmen, D. (1996) Pool cue chalk: a
26            source of environmental lead. Pediatrics 97: 916-917.
27     Moir, C. M.; Freedman, B.; McCurdy, R. (1996) Metal mobilization from water-distribution systems of buldings
28            serviced by lead-pipe mains. Can. Water Resour. J. 21: 45-52.
29     Moseholm, L.; Larsen, E. H.; Andersen, B.; Nielsen, M. M.  (1992) Atmospheric deposition of trace elements around
30            point sources and human health risk assessment. I. Impact zones near a source of lead emissions. Sci. Total
31            Environ. 126: 243-262.
32     Murozumi, M.; Chow, T. J.; Patterson, C. (1969) Chemical concentrations of pollutant lead aerosols, terrestrial dusts
33            and sea salts in Greenland and Antarctic snow strata. Geochim. Cosmochim. Acta 33: 1247-1294.
3 4     Ndung'u, K.; Hibdon, S.; Flegal, A. R. (2004) Determination of lead in vinegar by ICP-MS and GFAAS: evaluation
35            of different sample preparation procedures. Talanta 64: 258-263.
36     Nriagu, J. O.; Kim, M.-J. (2000) Emissions of lead and zinc from candles with metal-core wicks. Sci. Total Environ.
37            250:37-41.
38     Nriagu, J. O.; Pacyna, J. M. (1988) Quantitative assessment of worldwide contamination of air, water and soils by
39            trace metals. Nature (London) 333: 134-139.
40     Oliver, D. P.; McLaughlin, M. J.; Naidu, R.; Smith, L. H.; Maynard, E. J.; Calder,  I. C. (1999) Measuring Pb
41            bioavailability from household dusts using an in vitro model. Environ. Sci. Technol. 33: 4434-4439.
42     Paces, T. (1998) Critical loads of trace metals in soils: a method of calculation. Water Air Soil Pollut. 105: 451-458.
43     Palacios, H.; Iribarren, I.; Olalla, M. J.; Cala, V. (2002) Lead poisoning of horses in the vicinity of a battery
44            recycling plant. Sci. Total Environ. 290: 81-89.
45     Pichtel, J.; Kuroiwa, K.; Sawyer, H. T. (2000) Distribution of Pb, Cd and Ba in soils and plants of two contaminated
46            sites. Environ. Pollut. 110: 171-178.
47     Pierson, W. R.; Brachaczek, W. W. (1976) Paniculate matter associated with vehicles on the road. Presented at:
48            SAE automotive engineering congress and exposition; February; Detroit, MI. Warrendale, PA: Society of
49            Automotive Engineers; SAE technical paper no. 760039.
50     Rankin, C.; Nriagu, J. O.; Aggarwal, J. K.; Arowolo, T. A.; Adebayo, K.; Flegal, A. R. (2005) Lead contamination
51            in cocoa and cocoa products: isotopic evidence of globan contamination. Environ. Health Perspect.
52            113:  1344-1348.
53     Rieuwerts, J.  S.; Farago, M. E. (1995) Lead contamination in smelting and mining environments and variations in
54            chemical forms and bioavailability. Chem. Speciation Bioavailability 7: 113-123.
55     Rieuwerts, J.; Farago, M. (1996) Heavy metal pollution in the vicinity of a secondary lead smelter in the Czech
56            Republic. Appl.  Geochem. 11:  17-23.
       December 2005                             3-41            DRAFT-DO NOT QUOTE OR CITE

-------
  1     Rieuwerts, J. S.; Farago, M.;Bencko, V. (1999) Topsoil and housedust metal concentrations in the vicinity of a lead
  2            battery manufacturing plant. Environ. Monit. Assess. 59: 1-13.
  3     Sadiq, M; Zaidi, T. H.; Al Muhanna, H.; Mian, A. A. (1997) Effect of distribution network pipe material on
  4            drinking water quality. J. Environ. Sci. Health A32: 445-454.
  5     Scelfo, G. M.; Flegal, A. R. (2000) Lead in calcium supplements. Environ. Health Perspect. 108: 309-313.
  6     Schilling, R. J.; Bain, R. P. (1988) Prediction of children's blood lead levels on the basis of household-specific soil
  7            lead levels. Am. J. Epidemiol. 128: 197-205.
  8     Schroeder, W. H.; Dobson, M.; Kane, D. M.; Johnson, N. D. (1987) Toxic trace elements associated with airborne
  9            paniculate matter: a review. JAPCA 37: 1267-1285.
10     Sheets, R. W.; Kyger, J. R.; Biagioni, R. N.; Probst, S.; Boyer, R.; Barke, K. (2001) Relationship between soil lead
11            and airborne lead concentrations at Springfield, Missouri, USA.  Sci. Total Environ. 271: 79-85.
12     Singh, I.; Mavinic, D. S. (1991) Significance of building and plumbing specifics on trace metal concentrations in
13            drinking water. Can. J. Civil Eng. 18: 893 -903.
14     Singh, M.; Jaques, P. A.; Sioutas, C. (2002) Size distribution and diurnal characteristics of particle-bound metals in
15            source and receptor sites of the Los Angeles Basin. Atmos. Environ. 36: 1675-1689.
16     Singley, J. E. (1994) Electrochemical nature of lead contamination. J. Am. Waterworks Assoc. 86: 91-96.
17     Small, M. J.; Nunn, A. B., Ill; Forslund, B. L.; Daily, D. A. (1995) Source attribution of elevated residential soil
18            lead near a battery recycling site. Environ. Sci. Technol. 29: 883-895.
19     Sofuoglu, S. C.; Lebowitz, M. D.; O'Rourke, M. K.; Robertson, G. L.; Dellarco, M.; Moschandreas, D. J. (2003)
20            Exposure and risk estimates for Arizona drinking water. J. Am. Water Works Assoc. 95: 67-79.
21     Sternbeck, J.; Sjodin, A.; Andreasson, K. (2002) Metal emissions from road traffic and the influence of
22            resuspension—results from two tunnel studies. Atmos. Environ. 36: 4735-4744.
23     Subramanian, K. S.; Connor, J. W.; Meranger, J. C. (1991) Leaching of antimony, cadmium, copper, lead, silver,
24            tin and zinc from copper piping with nonlead-based soldered joints. J. Environ. Sci. Health Part A: Environ.
25            Sci. Eng. A26: 911-929.
26     Subramanian, K. S.; Sastri, V. S.; Connor, J. W. (1994) Drinking water quality: impact of nonlead-based plumbing
27            solders. Toxicol. Environ. Chem. 44: 11-20.
28     Suchara, I.; Sucharova, J. (2004) Distribution of 36 element deposition rates in a historic mining and smelting area
29            as determined through find-scale biomonitoring techniques. Part II: relative long-term accumulated
30            atmospheric deposition levels. Water Air Soil Pollut. 153: 229-252.
31     Sutherland, R. A. (2000) Depth variation in copper, lead, and zinc concentrations and mass enrichment ratios in soils
32            of an urban watershed. J. Environ.  Qual. 29: 1414-1422.
33     Sutherland, R. A.; Day, J. P.; Bussen, J. O. (2003) Lead concentrations, isotope ratios, and source apportionment in
34            road deposited sediments, Honolulu,  Oahu, Hawaii. Water Air Soil Pollut. 142: 165-186.
35     Thompson, N. G.;  Sosnin, H. A.  (1985) Corrosion of 50-50 tin-lead solder in household plumbing. Weld. J.
36            (Miami, FL, U.S.) 64(4): 20-24.
37     Thornton, I. (1988) Metal content of soils and dusts. Sci. Total Environ.  75: 21-39.
38     Thornton, I.; Davies, D. J. A.;  Watt, J. M.; Quinn, M. J. (1990) Lead exposure in young children from dust and soil
39            in the United Kingdom. In: Conference on advances in lead research: implications for environmental health;
40            January 1989; Research Triangle Park, NC. Environ. Health Perspect. 89: 55-60.
41     Thurston, G. D.; Spengler, J. D. (1985) A quantitative assessment of source contributions to inhalable  particulate
42            matter pollution in metropolitan Boston. Atmos. Environ. 19: 9-25.
43     Torts, K.; Van Grieken, R. (1997) Chemical relations between atmospheric aerosols, deposition and stone decay
44            layers on historic buildings at the Mediterranean coast. Atmos. Environ. 31: 2179-2192.
45     Tsai, C.-J.; Shih, T.-S.; Sheu, R.-N. (1997) Characteristics of lead aerosols in different work environments.
46            Am. Ind. Hyg. Assoc. J.  58: 650-656.
47     Turer, D.; Maynard, J. B.; Sansalone, J. J. (2001) Heavy metal contamination in soils of urban highways:
48            comparison between runoff and soil concentrations at Cincinnati, Ohio. Water Air Soil Pollut.
49            132:293-314.
50     U.S. Environmental Protection Agency. (2003) National air quality and emissions trends report. 2003 special studies
51            edition. Research Triangle Park, NC: Office of Air Quality Standards; Emissions Monitoring and Analysis
52            Division; report no. EPA 454/R-03-005. Available:
53            http : //www. epa .gov/air/airtrends/aqtrnd03/toc .html (27 August, 2004).
54     U.S. Geological Survey (USGS). (2003) Minerals yearbook 2003: lead. Washington, DC: U. S. Department of the
55            Interior. Available: http://minerals.usgs.gov/minerals/pubs/commodity/lead/ [13 October, 2005].
       December 2005                             3-42            DRAFT-DO NOT QUOTE OR CITE

-------
  1     Urbansky, E. T.; Schock, M. R. (2000) Can flouridation affect lead(II) in potable water? Int. J. Environ. Stud.
  2            57: 597-637.
  3     Vakeva, M.; Hameri, K.; Kulmala, M.; Lahdes, R.; Ruuskanen, J.; Laitinen, T. (1999) Street level versus rooftop
  4            concentrations of submicron aerosol particles and gaseous pollutants in an urban street canyon. Atmos.
  5            Environ. 33: 1385-1397.
  6     Van Der Leer, D.; Weatherill, N. P.; Sharp, R. J.; Hayes, C. R. (2002) Modelling the diffusion of lead into drinking
  7            water. Appl. Math. Modell. 26: 681-699.
  8     Venditti, D.; Durecu, S.; Berthelin, J. (2000) A multidisciplinary approach to assess history, environmental risks,
  9            and remediation feasibility of soils contaminated by metallurgical activities. Part A: chemical and physical
10            properties of metals and leaching ability. Arch. Environ. Contam. Toxicol. 38: 411-420.
11     Von Lindern, I. H.; Spalinger, S. M.; Bero, B. N.; Petrosyan, V.; Von Braun, M. C. (2003) The influence of soil
12            remediation on lead in house dust.  Sci. Total Environ. 303: 59-78.
13     Wassan, S. J.; Guo, Z.; McBrian, J. A.; Beach, L. O. (2002) Lead in candle emissions. Sci. Total Environ.
14            296:  159-174.
15     Wong, H. K.  T.; Gauthier, A.;Beauchamp, S.; Tordon, R. (2002) Impact of toxic metals and metalloids from the
16            Caribou gold-mining areas in Nova Scotia, Canada. Geochem: Explor. Environ. Anal. 2: 235-241.
17     World Health Organization. (2000) Air quality guidelines for Europe. 2nd. ed. Copenhagen, Denmark: Regional
18            Office for Europe. (WHO regional publications, European series no. 91). Available:
19            http://www.euro.who.int/air/activities/20050223_4 [29 November, 2005].
20     Young, T. M.; Heeraman, D. A.; Sirin, G.;  Ashbaugh, L. L. (2002) Resuspension of soil as a source of airborne lead
21            near industrial facilities and highways. Environ. Sci. Technol. 36: 2484-2490.
22     Zadnik, T. (2004) Lead in topsoil, hay, silage and blood of cows from farms near a former lead mine and current
23            smelting plant before and after installation of filters. Vet. Hum. Toxicol. 46: 287-290.
24     Zimdahl, R. L.; Skogerboe, R. K. (1977) Behavior of lead in soil. Environ. Sci. Technol. 11: 1202-1207.
25     Zinati, G. M.; Li, Y.; Bryan, H. H.; Mylavarapu, R. S.;  Codallo, M. (2004) Distribution and fractionation of
26            phosphorus, cadmium, nickel, and  lead in calcareous soils amended with composts. J. Environ. Sci. Health
27            Part 639:209-223.
28
       December 2005                             3-43            DRAFT-DO NOT QUOTE OR CITE

-------
 i    4.     MODELS OF HUMAN EXPOSURE THAT PREDICT
 2                    TISSUE DISTRIBUTION OF LEAD
 4   4.1    OBJECTIVES IN MODELING LEAD EXPOSURE AND TISSUE
 5          DISTRIBUTION OF LEAD
 6         Models are essential for quantifying human health risks that derive from exposures to
 7   lead.  Dose-response relationships for nearly all of the major health effects of lead in humans are
 8   expressed in terms of internal dose (e.g., blood or bone lead concentrations).  Application of this
 9   internal dose-response information to the assessment of risks from environmental exposures to
10   lead requires a way of relating internal dose to levels of lead in the environmental media to
11   which humans come in contact (e.g., air, water, surface dust).  Models provide the only means
12   for accomplishing this objective. Models come in various forms. Multivariate regression
13   models, commonly used in epidemiology, provide estimates of the contribution of variance in the
14   internal dose metric to various determinants or control variables (e.g., surface dust lead
15   concentration, air lead concentration). Structural equation modeling links several regression
16   models together to estimate the influence of determinants on the internal dose metric.
17   Regression models can provide estimates of the rate of change of blood or bone lead
18   concentration in response to an incremental change in exposure level (i.e., slope factor). The
19   strength of regression models is that they can have relatively few parameters, which allows a
20   rigorous quantitative assessment of uncertainty in the slope factor. However, the simplicity of
21   regression models also frequently excludes numerous parameters that are known to influence
22   human lead exposures and the relationship between human exposure  and tissue lead levels,
23   parameters which are expected to vary spatially and temporally.  Thus, extrapolation of
24   regression models to other spatial or temporal contexts, which is often necessary for regulatory
25   applications of the models, can be problematic.
26         An alternative to regression models are mechanistic models, which attempt to specify all
27   parameters needed to describe the mechanisms (or processes) of transfer of lead from the
28   environment to human tissues.  Such mechanistic models are more complex than regression
29   models, which introduces challenges in terms of their mathematical solution.  However, by
30   incorporating parameters that can be expected to vary spatially or temporally, or across

     December 2005                           4-1        DRAFT-DO NOT QUOTE OR CITE

-------
 1    individuals or populations, mechanistic models can be extrapolated to a wide range of exposure
 2    scenarios.  Exposure-biokinetic models, a type of mechanistic models, are highly simplified
 3    mathematical representations of relationships between levels of lead in environmental media and
 4    human lead intakes (e.g., jig lead ingested per day). These models include parameters
 5    representing processes of lead transfer between environmental media (e.g., air to surface dust)
 6    and to humans, including rates of human contact with the media and intakes of the media (e.g.,
 7    g soil ingested per day).  Biokinetic models provide the analogous mathematical representation
 8    of relationships between lead intakes and levels of lead in body tissues (e.g., blood lead
 9    concentration); and they include parameters that represent processes of lead transfer (a) from
10    portals of entry into the body and (b) from blood to tissues and excreta. Exposure-biokinetic
11    models provide an approach for predicting blood lead concentrations (or lead concentrations in
12    other tissues) that corresponds to a specified exposure (medium, level, and duration). Detailed
13    information on exposure and internal dose can be obtained from controlled experiments, but
14    almost never from epidemiological observations or from public health monitoring programs.
15    Exposure-biokinetic models can provide these predictions in the absence of complete
16    information on the exposure history and blood lead concentrations for an individual (or
17    population) of interest.  Therefore, these models are critical to applying epidemiologically-based
18    information on blood lead-response relationships to the quantification and characterization of
19    human health risk.  They are also critical for assessing the potential impacts of public health
20    programs directed at mitigation of lead exposure or of remediation of contaminated sites.
21          Mechanistic models also have several other important features that are useful for risk
22    assessment and for improving our basic understanding of lead exposures and biokinetics. They
23    integrate complex information on lead exposure and biokinetics into a form that provides
24    predictions, rather than just an organized grouping of observations. By analyzing the
25    relationships between model assumptions and predictions (i.e., sensitivity analysis), and by
26    comparing predictions to observations (i.e., model evaluation), such models can contribute to the
27    identification of important gaps in our understanding of lead exposure, biokinetics, and risk.
28    Thus, these models provide a consistent method for making, evaluating and improving
29    predictions that support risk assessment and risk management decisions.
30          Modeling of human lead exposures and biokinetics has advanced considerably during the
31    past several decades.  Among the most important new advances are development, evaluation, and

      December 2005                             4-2        DRAFT-DO NOT QUOTE OR CITE

-------
 1    extensive application of the Integrated Exposure Uptake Biokinetic (IEUBK) Model for Lead in
 2    Children (U.S. Environmental Protection Agency, 1994a) and the development of models that
 3    simulate lead biokinetics in humans from birth through adulthood (Leggett, 1993; O'Flaherty
 4    1993, 1995). While these developments represent important conceptual advances, several
 5    challenges remain for further advancements in modeling and applications to risk assessment.
 6    The greatest challenge derives from the complexity of the models. Human exposure-biokinetics
 7    models include large numbers of parameters, which are required to describe the many processes
 8    that contribute to lead intake, absorption, distribution, and excretion. The large number of
 9    parameters complicates the assessment of confidence in parameter values, many of which cannot
10    be directly measured. Statistical procedures can be used to evaluate the degree to which model
11    outputs conform to "real-world" observations and values of influential parameters can be
12    statistically estimated to achieve good agreement with observations. Still, large uncertainty can
13    be expected to remain about many, or even most, parameters in complex exposure-biokinetic
14    models such as those described below.
15          Given the difficulty in quantitatively assessing uncertainty in values of all of the
16    individual parameters in an exposure-biokinetics model, assurance that the model accurately
17    represents the real-world in all aspects is virtually impossible. As consequence of this, Oreskes
18    (1998) noted,  "... the goals of scientists working in a regulatory context should be not validation
19    but evaluation, and where necessary, modification and even rejection.  Evaluation implies an
20    assessment in which both positive and negative results are possible, and where the grounds on
21    which a model is declared, good enough are clearly articulated. "  In this context, evaluation of
22    confidence in a given exposure-biokinetic model rests largely  on assessment of the degree to
23    which model predictions, based on model inputs appropriate for a situation, conform to
24    observations and/or expectations; and, most importantly, the degree to which this conformity
25    does or does not satisfy requirements of model application to a specific context. Because  of
26    limitations in observations of predicted outcomes, it may be possible to evaluate confidence in
27    some uses of a model, but not others.  Similarly, it is possible  for confidence in a model to be
28    judged acceptable for a given use, but not for others.  The concept of validation of highly
29    complex mechanistic models, outside of the context of a specific use of the model, has little
30    meaning.
      December 2005                            4-3        DRAFT-DO NOT QUOTE OR CITE

-------
 1          In the ensuing discussion of specific models, reported efforts to evaluate the models are
 2    noted. In most cases, however, the relevance of these evaluations to the assessment of
 3    confidence in a specific use of that model (e.g., predicting average blood lead concentrations in
 4    children who live in areas that have certain cross-sectionally measured environmental lead
 5    levels) cannot be ascertained from the reported literature.  Nevertheless, as a framework for
 6    qualitatively comparing the various evaluative procedures that have been applied, the following
 7    general classification of model evaluations has been adopted:
 8       •  Sensitivity analysis has been conducted and most influential parameters identified and
 9          uncertainty characterized.
10       •  Model predictions have been compared qualitatively to observations.
11       •  Predictions have been compared quantitatively to observations (i.e., a statistical model
12          has been applied for estimation of "goodness of fit" and uncertainty).
13       •  Confidence in model predictions for specific uses has been quantitatively evaluated.
14       •  Accuracy of model implementation code has been verified.
15          In the sections that follow, an overview is provided with regard to the evolution of
16    important lead biokinetic modeling aspects that constitute major modeling advances during the
17    past 25 years or so leading to the development of EPA's All Ages Lead Model (AALM)
18    discussed below in Section 4.6.  Descriptions of the individual models are intended to provide
19    only brief snapshots of key features of each model, with particular attention to conceptual
20    features that are unique to each model. Key references are cited in which more complete
21    specifications of model parameters can be found.
22
23
24    4.2   HISTORIC OVERVIEW OF LEAD MODELS
25    4.2.1   Rabinowitz Model
26          Early lead modeling applications presented lead biokinetics in classical pharmacokinetics
27    terms. Compartments represented kinetically homogeneous pools of lead which might be
28    associated with individual organs or groups of organs.  Among the first of such models was one
29    proposed by Rabinowitz et al. (1976) based on a study of the kinetics of ingested stable lead
30    isotope tracers and lead mass balance data in five healthy adult males (Figure 4-1). The

      December 2005                            4-4         DRAFT-DO NOT QUOTE OR CITE

-------
DIET -i- AIR

3
DEEP TISSUE
(BONE)
=200 mg
T1/2=104days


~7 pg/day
~7 pg/day
'\

=48 Mg/day
1
BLOOD
1.9 mg
T1C=25 days
i
UR
=36 M
NE
g/day

==15 pg/day
=2 Mg/day
SHALLOV
(SOFT 1
=0.6
Ti,2=2!
i
BILE,
SWEAT,
=12 M9/
)
V TISSUE
ISSUE)
mg
3 days
HAIR,
NAILS...
day
     Figure 4-1.   Lead biokinetics based on Rabinowitz et al. (1976). Half-times are based on
                   reported residence times for compartments 1, 2, and 3: 25, 28, and 104 days,
                   respectively (half-time ln(2)/(l/residence time).
 1   Rabinowitz model has three compartments: (1) a central compartment representing blood and
 2   other tissues and spaces in rapid equilibrium with blood (e.g., interstitial fluid); (2) a shallow
 3   tissue compartment, representing soft tissues and rapidly exchanging pools within the skeleton;
 4   and (3) a deep tissue compartment, representing, primarily, slowly exchanging pools of lead
 5   within bone.  Excretion pathways include urinary (from the central compartment) and bile,
 6   sweat, hair, and nails (from the shallow tissue compartment). The model predicts pseudo-first
 7   order half-times for lead of approximately 25, 28, and 104 days in the central, shallow tissue, and
 8   deep compartments, respectively (these values were calculated based on reported residence
 9   times, the reciprocal of the sum of the individual elimination rate constants).  The slow kinetics
10   of the deep tissue compartment leads to the prediction that it would contain most of the lead
11   burden following chronic exposures (e.g., for years), consistent with lead measurements made in
12   human autopsy samples (Barry, 1975; Gross et al., 1975; Schroeder and Tipton, 1968).  Note that
13   this model did not simulate the distribution of lead within blood (e.g., erythrocytes and plasma),
14   nor did it simulate subcompartments within bone or physiological processes of bone turnover
15   that might affect kinetics in the deep tissue compartment.
     December 2005
4-5
DRAFT-DO NOT QUOTE OR CITE

-------
 1   4.2.2   Marcus Model(s)
 2         Marcus (1985b) reanalyzed the data from stable isotope tracer studies of Rabinowitz et al.
 3   (1976) and derived an expanded multicompartment kinetic model for lead (Figure 4-2).  The
 4   model included separate compartments with different lead turnover rates for cortical (slow,
 5   ti/2 = 1.2 x 104 to 3.5 x 104 days) and trabecular (fast, ti/2 = 100 to 700 days) bone, an approach
 6   subsequently adopted in several other models (O'Flaherty, 1995; U.S. Environmental Protection
 7   Agency, 1994a,b; Leggett, 1993; O'Flaherty, 1993; Bert et al., 1989).  A more complex
 8   representation of the lead  disposition in bone included explicit simulation of lead diffusion
 9   within the bone volume of the osteon and exchange with blood at the canaliculus (Marcus,
10   1985a; Figure 4-3).  Lead diffusion in bone was based on lead kinetics data from studies
11   conducted in dogs. A similar approach to simulating radial diffusion of lead in bone, expanded
12   to include eight concentric diffusion shells, was implemented by O'Flaherty (1995, 1993).
13   Marcus (1985c) also introduced nonlinear kinetics of exchange of lead between plasma and
14   erythrocytes. The blood kinetics included four blood subcompartments: diffusible lead in
15   plasma, protein-bound lead in plasma, a "shallow" erythrocyte pool, and a "deep" erythrocyte
16   pool (see Figure 4-4). The Marcus (1985c) model predicted the curvilinear relationship between
17   plasma and blood lead concentrations that has been observed in humans (DeSilva, 1981).
18
     Figure 4-2.
Lead biokinectics based on Marcus (1985b).  Bone is represented as a slow
turnover (cortical) compartment and a faster (trabecular) compartment.
     December 2005
                            4-6
DRAFT-DO NOT QUOTE OR CITE

-------
     Figure 4-3.   Lead biokinetics based on Marcus (1985a). Bone is represented as an
                  extended cylindrical canalicular territory. The canalicular territory has
                  a radius b, and surrounds the canaliculus of radius a. Lead diffuses
                  across radius library, between the fluid in the canaliculus (which is in
                  communication with blood in the Haversian canal, not shown) and the
                  bone volume of the canalicular territory.
 1   4.2.3   Bert Model
 2         Bert et al. (1989) adopted the bone model from Marcus (1985b), in which the bone
 3   compartment is subdivided into slow cortical bone and faster trabecular bone compartments
 4   (Figure 4-5). The central compartment (denoted as blood) is assumed to be 1.5 times the volume
 5   of whole blood, with the whole blood volume varying in direct proportion with body weight.
 6   The model includes a discrete pathway for excretion of unabsorbed lead from the gastrointestinal
 7   (GI) tract into feces. Secretion of lead in bile, gastric secretions, and saliva are represented as
 8   transfers from the soft tissue compartment to the GI tract.  Compartment transfer coefficients
 9   were based on average values estimated for four individuals from the Rabinowitz et al. (1976)
10   study. Initial average values for lead in cortical bone for a given age at the start of a simulation
11   were derived from Barry (1975).
     December 2005
4-7
DRAFT-DO NOT QUOTE OR CITE

-------
Figure 4-4.  Lead biokinetics based on Marcus (1985c). Blood is represented with a
            plasma (central exchange) compartment and a red blood cell compartment,
            the latter having shallow and deep pools.
Figure 4-5.  Lead biokinetics based on Bert et al. (1989).
December 2005
4-8
DRAFT-DO NOT QUOTE OR CITE

-------
 1    4.2.4   Contemporary Models
 2          Additional information on lead biokinetics, bone mineral metabolism, and lead exposures
 3    has led to further refinements and expansions of these earlier modeling efforts.  In particular,
 4    three pharmacokinetic models are currently being used or are being considered for broad
 5    application in lead risk assessment: (1) the Integrated Exposure Uptake BioKinetic (IEUBK)
 6    model for lead in children developed by EPA (U.S. Environmental Protection Agency, 1994a,b;
 7    White et al., 1998); (2) the Leggett model, which simulates lead kinetics from birth through
 8    adulthood (Leggett, 1993); and (3) the O'Flaherty model, which simulates lead kinetics from
 9    birth through adulthood (O'Flaherty, 1995, 1993). Of the three approaches, the O'Flaherty
10    model has the fewest lead-specific parameters and relies more extensively on physiologically
11    based parameters to describe volumes, flows, composition, and metabolic activity of blood and
12    bone that determine the disposition of lead in the human body. Both the IEUBK model and the
13    Leggett model are more classical multicompartmental models; that is, the values for the
14    age-specific transfer rate constants for lead are based on kinetics data obtained from studies
15    conducted in animals and humans and may not have precise physiological correlates. Thus, the
16    structure and parameterization of the O'Flaherty model  is distinct from both the IEUBK model
17    and Leggett model. All three models represent the rate  of uptake of lead (i.e., amount of lead
18    absorbed per unit of time) as relatively simple functions (f) of lead intake:
19
20                                  Uptake   = Intake-AF                             (4-1)
21
22
23                                  Uptake  = Intake • f(Intake)                            (4-2)
24
25          Values assigned to absorption factor (AF) or other variables in f(intake) are, in general,
26    age-specific and environmental medium-specific in some models.  However, the models do not
27    modify the representation of uptake as functions of the many other physiologic variables that
28    may affect lead absorption (e.g., nutritional status). While one can view this approach as a
29    limitation of the models, it also represents a limitation of the data available to support more
30    complex representations of lead absorption.
31
     December 2005                            4-9         DRAFT-DO NOT QUOTE OR CITE

-------
 1          The IEUBK model simulates multimedia exposures, uptake, and kinetics of lead in
 2    children ages 0-7 years; the model is not intended for use in predicting lead pharmacokinetics in
 3    adults.  The O'Flaherty and Leggett models are lifetime models, and include parameters that
 4    simulate uptake and kinetics of lead during infancy, childhood, adolescence, and adulthood.
 5    Lead exposure (e.g., residence-specific environmental lead concentrations, childhood activity
 6    patterns) is not readily described by current versions of the O'Flaherty and Leggett models.
 7    By contrast, the IEUBK model includes parameters for simulating exposures and uptake to
 8    estimate average daily uptake of lead (jig/day) among populations of children potentially
 9    exposed via soil and dust ingestion, air inhalation, lead-based paint chip ingestion, tap water
10    ingestion, and/or diet.
11          The above three models have been individually evaluated, to varying degrees, against
12    empirical physiological data on animals and humans and data on blood lead concentrations in
13    individuals and/or populations (U.S. Environmental Protection Agency, 1994a,b; Leggett, 1993;
14    O'Flaherty, 1993). However, applications in risk assessment typically require that the models
15    accurately predict blood lead distributions in real  populations, in particular those values or
16    percentages falling in the "upper tails" (e.g., >95th percentiles of the distributions, when input to
17    the models consists of data that describe site-specific exposure conditions (e.g., environmental
18    lead concentrations, physicochemical properties of soil and dust) (Beck et al., 2001; Griffin
19    et al., 1999a,b). In evaluating models for use in risk assessment, exposure data collected at
20    hazardous waste sites have been used to drive model simulations (Bowers and Mattuck, 2001;
21    Hogan et al.,  1998). The exposure module in the  IEUBK model makes this type of evaluation
22    feasible.
23
24
25    4.3   INTEGRATED EXPOSURE UPTAKE BIOKINETIC (IEUBK)
26          MODEL FOR LEAD IN CHILDREN
27    4.3.1   Model  Structure
28          The IEUBK model for lead in children (see Figure 4-6) is a multicompartmental
29    pharmacokinetics model linked to an exposure and probabilistic model of blood lead
30    concentration distributions in children (U.S. Environmental Protection Agency, 1994a,b; White
31    et al., 1998).  The model simulates exposure and biokinetics of lead from birth to age 7 years

      December 2005                           4-10        DRAFT-DO NOT QUOTE OR CITE

-------
1

Respiratory
tract
Uptake Component
1
Respiratory
tract






Gastrointestinal tract


Plasma extra-cellularfluid


	 J Fe
:es )
     m
     o
     a.
     E
     o
     O
r Plasma extra-cellularfluid
— ' ,

Red blood
cells

i

Kidney



V
Other soft
tissues
               O
              o
Elimination pools of
the body

Intake from
environmental
Media (|jg/lead/day)
Body compartment

Body compartment or
elimination pool required in
more than one component
Figure 4-6.    Structure of the integrated exposure uptake biokinetics model for lead
               in children (U.S. Environmental Protection Agency, 1994a,b; White
               et al., 1998).
December 2005
                       4-11
               DRAFT-DO NOT QUOTE OR CITE

-------
 1    (84 months) and was developed for predicting average quasi-steady state blood lead
 2    concentrations corresponding to daily average exposures, averaged over periods > 1 year.
            The model has four major components or submodels:
 3       •  Exposure model, in which average daily intakes of lead (jig/day) are calculated for each
 4          inputted exposure concentration (or rates) of lead in air, diet, dust, soil, and water;
 5       •  Uptake model, which converts environmental media-specific lead intake rates calculated
 6          from the exposure model into a media-specific time-averaged rates of uptake (jig/day)
 7          of lead to the central compartment (blood plasma);
 8       •  Biokinetic model, which simulates the transfer of absorbed lead between blood and
 9          other body tissues, elimination of lead from the body (via urine, feces, skin, hair, and
10          nails), and predicts an average blood lead concentration for the exposure time period of
11          interest; and
12       •  Blood lead probability model, which simply applies a log-normal distribution (with
13          specific geometric mean and geometric standard deviation parameters) to predict
14          probabilities for the occurrence of a specified blood lead  concentration in a population
15          of similarly exposed children.

16          Exposure Model. The exposure model simulates intake of lead (jig/day) for exposures to
17    lead in air (|ig/m3), drinking water (|ig/L), soil-derived dust (|ig/g), and diet (jig/day). The
18    temporal resolution of the exposure model is 1  year; exposure inputs are intended to represent
19    annual averages for an age-year time step (e.g., ages 1, 2, 3...years). Exposure inputs that
20    represent the average daily value for an age-year will yield corresponding daily average intakes
21    for the same age-year. The spatial resolution of the exposure model was intended to be a child's
22    residence (e.g., the home and yard). The model accepts inputs for media intake rates (e.g., air
23    volume breathing rates, drinking water consumption rate, soil and dust ingestion rate). The air
24    exposure pathway partitions exposure to outdoor air and indoor air; with age-dependent values
25    for time spent outdoors and indoors (hours/day).  Exposure to lead in soil derived dust is also
26    partitioned into outdoor and indoor contributions.  The intakes from all ingested exposure
27    media (diet, drinking water, soil-derived dust) are summed to calculate a total intake to the
28    gastrointestinal tract, for estimating capacity-limited absorption (see description of the
29    Uptake Model).
30          Uptake Model.  The uptake model simulates lead absorption in the gastrointestinal tract as
31    the sum  of a capacity-limited (represented by a Michaelis-Menten type relationship) and
      December 2005                            4-12        DRAFT-DO NOT QUOTE OR CITE

-------
 1    unlimited processes (represented by a first-order, linear relationship). These two terms are
 2    intended to represent two different mechanisms of lead absorption, an approach that is in accord
 3    with limited available data in humans and animals that suggest a capacity limitation for lead
 4    absorption (Mushak, 1991).  One of the parameters for the capacity-limited absorption process
 5    (that represents that maximum rate of absorption) is age-dependent.  The above representation
 6    gives rise to a decrease in the fractional absorption of ingested lead as a function of total lead
 7    intake as well as age.  Absorption fractions are also medium specific (Figure 4-7).  At 30 months
 8    of age, at low intakes (<200  jig/day), below the rates at which capacity-limitation has a
 9    significant impact on absorption, the fraction of ingested lead in food or drinking water that is
10    absorbed is 0.5 and decreases to approximately 0.11 at high intake (>5000 jig/day). For lead
11    ingested in soil or dust, fractional absorption  is 0.35 at low intake (<200 jig/day) and decreases
12    to 0.09 at high intake (>5000 jig/day).
13          The uptake model assumes that 32% of inhaled lead is deposited in the respiratory tract.
14    This value was originally assigned based on a scenario of exposure to active smelter emissions,
15    which assumed the particle size distribution in the vicinity of an active lead smelter; size-specific
16    deposition fractions for the nasopharyngeal, tracheobronchial, and alveolar regions of the
17    respiratory tract; and region-specific absorption fractions (Table 4-1). There are three
18    compartments in the Leggett model for lung absorption, and these are distinguished by the depth
19    to which the air particle penetrates.  Lead deposited in the  alveolar region is assumed to be
20    completely absorbed from the respiratory tract, whereas, lead deposited in the nasopharyngeal
21    and tracheobronchial regions is assumed to be transported  to the gastrointestinal tract where
22    absorption (approximately 30%) occurs.
23          Biokinetics Model.  The biokinetics model includes a central compartment, plasma and
24    extracellular fluid combined (plasma-ECF), six peripheral  body compartments, and three
25    elimination pathways. The temporal resolution of the biokinetics model is 1 month and, as
26    discussed below,  parameter values for bone-plasma-ECF exchanges were assigned with the
27    objective  of simulating the quasi-steady state condition of months, rather than short-term kinetics
28    of days.  The body compartments include kidney, liver, trabecular bone, cortical bone, and other
29    soft tissue. The model simulates growth of the body  and tissues, compartment volumes, and lead
30    masses and concentrations in each compartment.  Blood lead concentration at birth (neonatal)
31    is assumed to  be 0.85 of the  maternal blood lead. Neonatal lead masses and concentrations are

      December 2005                            4-13        DRAFT-DO NOT QUOTE OR CITE

-------
          0.55
          0.50
                                                                     ™*~Age6m
                                                                     -B-Age 18m
                                                                     ~+~Age30m
                                                                     —*-Age 54m
                                                                     -»-Age 66m
                                                                     -*-Age78m
          0.35
          0.30
                    10     20     30     40     50     60      70
                                      Food or Water Lead Intake (ug/day)
                                                                      80
                                                                             90
                                                                                   100
          0.35
          0.30
          0.25
        o 0.20
          0.15
                       -»~Age 6m
                       -B~~Age 18m
                       —+™ Age 30m
                       	*r-Age 42m
                       -5*^ Age 54m
                       —»—Age 66m
                       ^^Age 78m
          0.10
                   1000    2000    3000    4000    5000    6000    7000
                                    Soil or Dust Lead Concentration (ppm)
                                                                    8000
                                                                           9000
                                                                                  10000
 Figure 4-7.    Age-dependency of absorption fraction for ingested lead in the IEUBK
                model for lead in children. Absorption fraction for food and water
                (top panel); soil and dust (bottom panel).
December 2005
4-14
DRAFT-DO NOT QUOTE OR CITE

-------
              Table 4-1.  Comparison of Slope Factors in Selected Slope Factor Models
Model
Bowers etal. (1994)
Carlisle and Wade (1992)
Carisle and Wade (1992)
U.S. Environmental Protection
Agency (1996)
Stern (1994)
Stern (1994)
Receptor
Adult
Child
Adult
Adult

Child
Adult
Slope Factor
Intake
N/A
Soil/dust: 0.07
Water: 0.04
Soil/dust: 0.018
Water: 0.04
ND

Residential:
T(0.056, 0.16, 0.18)
Nonresidential:
U(0.014, 0.034)

Uptake
0.375
N/D
N/D
0.4

N/D
N/D
Absorption
Fraction
0.08
N/D
N/D
0.12

N/D

      N/D = Not determined; T = triangular probability distribution function (PDF); U = uniform PDF
 1   assigned to other compartments based on a weighted distribution of the neonatal blood lead
 2   concentration. Exchanges between the central compartment and tissue compartments are
 3   simulated as first-order processes, which are parameterized with unidirectional, first-order rate
 4   coefficients. Rate coefficients are allometrically scaled as a power function of body weight
 5   (BW033).
 6         Saturable uptake of lead into erythrocytes is simulated, with a maximum erythrocyte lead
 7   concentration of 120 |ig/L.  Excretory routes simulated include urine, from the central
 8   compartment; bile-feces, from the liver; and a lumped excretory pathway representing losses to
 9   skin, hair and nails, from the "other soft tissue" compartment.
10         Bone is simulated as a trabecular bone compartment (20% of bone volume) and a cortical
11   bone compartment (80% of bone volume).  Rate constants for transfer from plasma to the two
12   bone compartments are assigned values that result in a 4:1 cortical lead:trabceular lead mass ratio
13   within one biokinetic time step (one month).  This is achieved by assigning the two bone
14   compartments identical rate coefficients for transfer of lead from bone to plasma-ECF (half-time
15   8.5 days, at age 2 years), and faster (cortical, half-time 0.0083 days) and slower transfer
16   (trabecular, half-time 0.035 days) from the plasma-ECF (cortical:trabecular rate ratio is


     December 2005                           4-15        DRAFT-DO NOT QUOTE OR CITE

-------
 1    approximately 4:1). Note, this approach is different from previous and subsequent modeling
 2    approaches, in which cortical bone-to-plasma (or blood) transfer is assumed to occur slowly,
 3    relative to trabecular bone-to-plasma transfer (Marcus, 1985b; Bert et al., 1989; Leggett, 1993;
 4    O'Flaherty, 1993, 1995). For predictions of quasi-steady state conditions and the intended use of
 5    the IEUBK Model, the two general approaches can be expected to yield similar distributions of
 6    lead between the cortical and trabecular bone compartments.
 7          Blood Lead Probability Model.  Inputs to the IEUBK model are exposure point estimates
 8    that are intended to represent time-averaged central tendency exposures. The output of the
 9    model is a central tendency estimate of blood lead concentration for children who might
10    experience the inputted average exposures. However, within a group of similarly exposed
11    children, blood lead concentrations would be expected to vary among children as a result of
12    inter-individual variability in media intakes (e.g., daily average intakes of soil-derived dust,
13    drinking water, or food), absorption, and biokinetics. The model simulates the combined impact
14    of these sources of variability as a lognormal distribution of blood lead concentration for which
15    the geometric mean (GM) is given by the central tendency blood lead concentration outputted
16    from the biokinetics model, and the geometric standard deviation (GSD) is an input parameter.
17    The resulting lognormal distribution also provides the basis for predicting the probability of
18    occurrence of given blood lead concentration within a population of similarly exposed children:
19
20                 Px = probability of exceeding a blood lead concentration of X |ig/dL
21
22                 PIO = probability of exceeding a blood lead concentration of 10 |ig/dL
23
24          The model can be iterated for varying exposure concentrations (e.g., a series of increasing
25    soil lead concentration) to predict the media concentration that would be associated with a
26    probability of 0.05 for the occurrence of a blood lead concentration exceeding 10 |ig/dL
27    (Pio=0.05).
28
29    4.3.2    Model Calibration and Evaluation
30          An evaluation of the IEUBK model has been carried out by comparison of model
31    predictions of blood lead concentrations in children with observations from epidemiologic
32    studies of hazardous waste sites (Hogan et al., 1998). Data characterizing residential lead

      December 2005                           4-16        DRAFT-DO NOT QUOTE OR CITE

-------
 1    exposures and blood lead concentrations in children living at four Superfund National Priorities
 2    List (NPL) sites were collected in a study designed by the Agency for Toxic Substances and
 3    Disease Registry (ATSDR) and EPA. The residential exposure data were used as inputs to the
 4    IEUBK model and predicted blood lead concentration distributions were compared to the
 5    observed distributions in children living at the same residences. The IEUBK model predictions
 6    of geometric mean blood lead concentrations for children whose exposures were predominantly
 7    from their residence (i.e., no more than 10 hours/week away from home) were within 0.7 |ig/dL
 8    of the observed geometric mean at each site.  The prediction of the percentage of children
 9    expected to have blood lead concentrations exceeding 10 |ig/dL were within 4% of the observed
10    percentage at each site.  This evaluation supports IEUBK model use for estimating blood lead
11    concentrations in children at sites where their residential exposures can be adequately
12    characterized. Similar empirical  comparisons have shown that agreement between IEUBK
13    model predictions and observed blood lead concentrations at specific locations is influenced by
14    numerous factors, including (a) the extent to which the exposure and blood lead measurements
15    are adequately matched  and (b) site-specific factors (e.g., soil characteristics, behavior patterns,
16    bioavailability) that may affect lead intake or uptake in children (Bowers and Mattuck, 2001;
17    TerraGraphics Environmental Engineering, Inc., 2001).
18
19    4.3.3    Model Applications
20          Biomarkers Simulated. The IEUBK model computes masses of lead in bone and various
21    soft tissues, and excretion of lead, which are used in the computation of blood lead
22    concentration. However, the model was not developed for the purpose of predicting lead masses
23    in these tissues or excreta. Blood lead concentration is the only lead biomarker output that is
24    accessible to the user.
25          Exposure Inputs.  The IEUBK model was developed to predict the probability of elevated
26    blood lead concentrations in children exposed to user-specified annual average exposures to lead
27    in air, food, drinking water, soil, and dust. As noted above, the exposure model has an age-year
28    time step (the smallest time interval for a single exposure event) and, therefore, is more suited to
29    applications in which long-term (i.e., > lyear) average exposures and quasi-steady state blood
30    lead concentrations are to be simulated.  Intermittent exposures occur for brief periods of time
31    (e.g., a weekend at the beach), or in cases where significant seasonal  variations are different from

      December 2005                           4-17       DRAFT-DO NOT QUOTE OR CITE

-------
 1    the typical residential or occupational exposure.  In these cases, the IEUBK can accept time-
 2    weighted average exposures using the guidance provided in Syracuse Research Corporation
 3    (SRC) (2003).  Shorter-term dynamics of blood lead concentration, that may result from
 4    exposures that are highly variable on time scales of days or weeks, will not be captured with
 5    this approach (Lorenzana et al., 2005; Khoury and Diamond, 2003).
 6         Modeling Variability and Uncertainty.  As noted above, the IEUBK model uses a
 7    lognormal probability model to simulate inter-individual variability in blood lead concentrations
 8    attributable to variability in media intakes, absorption, and biokinetics. The model uses a generic
 9    default value of 1.6 for the blood lead concentration individual GSD (GSD;).  This value was
10    derived from an analysis of exposure (soil lead)-stratified variability in blood lead concentrations
11    in various cohorts of children (U.S. Environmental Protection Agency, 1994a; White et al.,
12    1998). Griffin et al. (1999b) also explores various statistical methods for estimating for
13    estimating an appropriate GSD; (regression, box modeling, structural equation modeling).
14          A Monte Carlo approach has been used to simulate and propagate variability and
15    uncertainty in exposure and absorption through IEUBK model simulation of blood lead
16    concentrations (Goodrum et al., 1996).  This extension of the model provides an alternative to
17    the generic blood lead probability approach for incorporating explicit estimates of variability
18    (and uncertainty in variability) in exposure and absorption into predictions of an expected
19    probability distribution of blood lead concentrations. A quantitative uncertainty analysis of
20    IEUBK model-based estimates of the PIO for a smelter site in Utah revealed that parameters
21    specifying soil ingestion rate were a dominant contributor to uncertainty in the PIO; however, the
22    contribution of soil  ingestion uncertainty, relative to uncertainty in other model parameters (i.e.,
23    mean soil lead concentration,  absorption fraction) varied across individual locations (Initial
24    Study Zones) at the site (Griffin et al., 1999a).
25
26    4.3.4   Validation/Verification of IEUBK
27          The IEUBK  model was initially released to the public in 1994 as a compiled DOS-based
28    C program (IEUBK v99d).  This version was subjected to an independent validation and
29    verification study which verified that the code accurately implement the model (Mickle, 1998;
30    Zaragoza and Hogan, 1998).  A 32-bit C++ (IEUBKwin32) version of the model is available for
31    download from an EPA website (http://www.epa.gov/superfund/programs/lead/ieubk.htm).

      December 2005                           4-18       DRAFT-DO NOT QUOTE OR CITE

-------
 1    4.4   LEGGETT MODEL
 2    4.4.1    Model Structure
 3          The Leggett model was developed from a biokinetic model originally developed for the
 4    International Commission on Radiological Protection (ICRP), for calculating radiation doses
 5    from environmentally important bone-seeking radionuclides, including radioisotopes of lead
 6    (Leggett, 1985, 1992a,b). The model has been used to develop cancer risk coefficients for
 7    internal radiation exposures to lead and other alkaline earth elements that have biokinetics
 8    similar to those of calcium (ICRP, 1993; U.S. Environmental Protection Agency, 1997). The
 9    model includes a central exchange compartment, 15 peripheral body compartments, and
10    3 elimination pools (Figure 4-8).  The central exchange compartment is the diffusible pool of
11    lead in plasma. The model simulates a bound pool in plasma (i.e., lead bound to plasma
12    proteins); that has an equilibrium ratio (bound:free) of approximately 5.  Transport of lead from
13    plasma to tissues is assumed to follow first-order kinetics. The temporal resolution of the model
14    is 1 day. Transfer rate constants vary with age and blood lead concentration. The latter
15    adjustment accounts for the limited uptake of plasma lead into red blood cells and the resulting
16    shift in distribution of lead from plasma-ECF to other tissues. Above a nonlinear threshold
17    concentration in red blood cells (assumed to be 60 |ig/dL), the rate constant for transfer to red
18    blood cells declines and constants to all other tissues increase proportionally (Leggett, 1993).
19    This replicates the nonlinear relationship between plasma and red blood observed in humans
20    (Smith et al., 2002; Manton et al., 2001; Bergdahl et al., 1999, 1998, 1997). The model
21    simulates blood volume as an age-dependent function, which allows simulation of plasma and
22    blood lead  concentrations.  However, volumes of other tissues are not simulated; therefore, only
23    lead masses in these tissues, and not concentrations are simulated.
24          First-order transfer coefficients (day-1) between compartments were developed for six age
25    groups, and intermediate age-specific values are obtained by linear interpolation.  The total
26    transfer rate from diffusible plasma to all destinations (TPALL) combined  is assumed to be
27    2000 day-1, based on isotope tracer studies in humans receiving lead via injection or inhalation.
28    Values for  transfer coefficients from plasma to tissues and tissue compartments are based on
29    measured deposition fractions (DF) or instantaneous fractional outflows of lead between tissues
30
      December 2005                            4-19        DRAFT-DO NOT QUOTE OR CITE

-------
       Non-    i Exchange
       Exchange i
Figure 4-8.   Structure of the Leggett Lead Biokinetic Model (Leggett, 1993). The central
             exchange compartment is diffusible plasma. Bone is represented as having
             surface (which rapidly exchanges with plasma) and volume compartments;
             the latter stimulates slow exchange with the surface and slow return of lead
             to the plasma from bone resorption.
December 2005
4-20
DRAFT-DO NOT QUOTE OR CITE

-------
 1    compartments (Leggett, 1993), where the transfer coefficient to a specific tissue or compartment
 2    (TP;) is given by:
 3
 4                                     TP t  = DF .  • TPALL                         (4-3)
 5
 6    This approach establishes mass balance with respect to the transfer rates from plasma:
 7
 8                                       ., -TPALL                                 (4J|)
 9
10          The model simulates both rapid exchange of lead with plasma via bone surface and slow
1 1    loss by bone resorption.  Cortical bone volume (80% of bone volume) and trabecular bone
12    volume (20% of bone volume) are simulated as bone surface compartments, which rapidly
13    exchanges with lead in plasma, and bone volume, within which are exchangeable and
14    nonexchangeable pools.  Lead enters the exchangeable pool of bone volume via the bone surface
15    and can return to the bone surface, or move to the nonexchangeable pool, from where it can
16    return to the plasma only when bone is resorbed.  Transfers from plasma to bone surface, return
17    from bone surface to  plasma, and bone surface to exchangeable bone volume are assumed to be
18    relatively fast processes (adult ti/2 = 3.85, 1.4, and 1.4 days, respectively).  Return of lead from
19    the exchangeable bone volume is slower (adult ti/2 = 30 days); however, the dominant transfer
20    process determining long-term accrual of bone lead burden are slow rate coefficients for transfer
21    of lead from the nonexchangeable pools of trabecular and cortical bone to plasma (adult ti/2 = 3.8
22    and 23 years, respectively).  Bone transfer coefficients vary with age (faster in children) to
23    reflect the age-dependence of bone turnover. The slow, nonexchangeable, bone volume
24    compartment is much more  labile in infants and children than in adults (e.g., cortical ti/2 =
25    68 days at birth and 1354 days at age 15 years; trabecular ti/2 = 68  days at birth and 725 days at
26    age 15 years).  Other physiological states (such as pregnancy and menopause) that affect bone
27    turnover and, therefore, bone lead kinetics are not simulated, although such states could
28    conceivably be accommodated with adjustments to tissue (e.g., bone) transfer coefficients.
29          The liver is simulated as two compartments; one compartment has a relatively short
30    removal half-life for transfers to plasma and to the small intestine by biliary secretion (adult
3 1    ti/2 =10 days); a second  compartment simulates a more gradual transfer to plasma of

      December 2005                           4-2 1        DRAFT-DO NOT QUOTE OR CITE

-------
 1    approximately 10% of lead uptake in liver (adult ti/2 = 365 days). The kidney is simulated as two
 2    compartments, one that exchanges slowly with blood plasma and accounts for lead accumulation
 3    kidney tissue (adult ti/2 = 365 days) and a second compartment that receives lead from blood
 4    plasma and rapidly transfers lead to urine (adult ti/2 = 5 days), with essentially no accumulation
 5    (urinary pathway).  Other soft tissues are simulated as three compartments representing rapid,
 6    intermediate, and slow turnover rates, without specific physiologic correlates (adult ti/2 = 0.3,
 7    100, and 1824 days, respectively). Other excretory pathways (hair, nails, and skin) are
 8    represented as a lumped pathway from the intermediate turnover rate of the soft tissue
 9    compartment.
10          The Leggett model simulates  lead intakes from inhalation, ingestion, or intravenous
11    injection. The latter was included to accommodate model evaluations based on intravenous
12    injection studies in humans and animal models. The respiratory tract is simulated as four
13    compartments into which inhaled lead is deposited and absorbed with half-times of 1, 3, 10, and
14    48 hours. Four percent of the inhaled lead is assumed to be transferred to the GI tract. These
15    parameter values reflect the data on which the model was based, which were derived from
16    studies in which human  subjects inhaled submicron lead-bearing particles (Morrow et al., 1980;
17    Chamberlain et al.,  1978; Wells et al., 1975; Hursh and Mercer, 1970; Hursh et al., 1969). These
18    assumptions would not necessarily apply for exposures to larger airborne particles. Absorption
19    of ingested lead is simulated as an age-dependent fraction of the ingestion rate, declining from
20    0.45 at birth to 0.3 at age 1 year (to age 15 years), and to 0.15 after age 25 years (Figure 4-9).
21
22    4.4.2   Model Calibration and Evaluation
23          Leggett (1993) and Pounds and Leggett (1998) describe various  qualitative empirical
24    comparisons of model predictions against observations made on adults (e.g., Skerfving et al.,
25    1985; Campbell et al., 1984; Manton and Cook, 1984; Barry, 1981; DeSilva, 1981; Chamberlain
26    et al., 1978; Rabinowtiz et al., 1976; Barry, 1975; Griffin et al., 1975; Gross et al., 1975; Hursh
27    and Mercer, 1970; Hursh et al.,  1969; Schroeder and Tipton,  1968). Age-specific changes in
28    parameter values that specify the biokinetics of lead in children were assigned values that
29    resulted in agreement between predicted age-specific lead distribution (fraction of body burden)
30    in blood, bone, brain, kidney, liver, and other tissues, and reported postmortem values
31    (Schroeder and Tipton, 1968; Barry, 1975, Gross et al. 1975; Barry, 1981).

      December 2005                            4-22        DRAFT-DO NOT QUOTE OR CITE

-------
                          -IEUBK Food and Drinking Water
        03
            10
             o
              0          5          10          15          20          25          30
                                          Age (years)
    Figure 4-9.   Age-dependency of absorption fraction for ingested lead in the Leggett and
                 O'FIaherty models. The IEUBK model projects absorption only through age
                 seven (84 mo). At intakes below those which approach the limit on "active"
                 absorption of lead, absorption is constant with age, with default valves of
                 50% for diet and drinking water, 30% for soil and dust. Fractional
                 absorption via the active pathway decreases with age and lead intake
                 (see Figure 4-7).
1   4.4.3   Model Applications
2         Biomarkers Simulated.  The Leggett model simulates the concentrations of lead in blood
3   and plasma, lead masses of lead in bone and various soft tissues, and excretion of lead in urine
4   that correspond to lifetime exposures (in terms of daily lead intakes).
5         Exposure Inputs. The model does not contain a detailed exposure module (although it can
6   be linked to an exposure model); lead exposure estimates are incorporated into the simulations as
7   age-specific point estimates of average daily intake (jig/day) from ingestion, inhalation, or
8   injection.  The model operates with a lead intake time step of 1 day, which allows simulation of
9   rapidly changing (i.e., daily) intermittent exposures (Lorenzana et al., 2005; Khoury and
    December 2005                           4-23        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Diamond, 2003).  Assumptions of blood lead concentrations at birth can also be introduced into
 2   the simulations, from which levels in other tissue in the first time step after birth are calculated.
 3          Dose reconstruction is possible with this model, since intakes, and corresponding tissue
 4   lead burdens accrued at any period in the lifetime, prior to an exposure event of interest, can be
 5   simulated.  Pounds and Leggett (1998) illustrate this in a study of a childhood lead poisoning
 6   case, in which the exposure is followed by chelation.  Chelation was simulated as  a short-
 7   duration increase in the plasma lead deposition fraction to urine, with corresponding proportional
 8   decreases in deposition fractions to other tissues.
 9
10   4.4.4   Implementation Code
11          The Leggett model was initially developed as a Fortran code, which can be run, without
12   compiling, from various platforms, including DOS and Windows (see Pounds and Leggett, 1998
13   for a description). A version compiled in Advanced Continuous Simulation Language (ACSL)
14   has also been reported (Lorenzana et al., 2005).  Confirmation of the Leggett model code was
15   carried out by a panel of experts (ICRP,  1989, 1993).
16
17
18   4.5   O'FLAHERTY MODEL
19          The O'Flaherty model simulates lead exposure, uptake, and disposition in humans, from
20   birth through adulthood (O'Flaherty, 1995, 1993). Figure 4-10 shows a conceptualized
21   representation of the model. Important novel features of the O'Flaherty model are the simulation
22   of growth, bone formation, and resorption. A growth  curve is simulated with a logistic
23   expression relating body weight to age in males  or females. The full expression relating weight
24   to age has five parameters (constants), so that it  can readily be adapted to fit a range of
25   standardized growth curves for males and females.  Tissue growth and volumes are linked to
26   body weight; this provides explicit modeling of  lead concentrations in all tissues simulated.
27   Other physiologic functions (e.g., bone formation) are linked to body weight, age, or to both.
28   The model can be implemented with a temporal  resolution of 1 day; however, as originally
29   configured, the rate parameters are expressed in  time units of years.
     December 2005                            4-24       DRAFT-DO NOT QUOTE OR CITE

-------
                        Respiratory tract
                                                      o
                                                      o
                Elimination pools of
                the body

                Intake from environmental
                Media (|jg/lead/year)

                Body compartment
                         Blood plasma
                      Well-perfused tissues
                     Poorly-perfused tissues
                         Cortical bone
                        Trabecular bone
                            Liver
                            Kidney
                                                          ;Diet, Dust,
                                                          Daint, Soil,
                                                           Water
                                                      Gastrointestinal tract
                                                                        Feces
Figure 4-10.  Structure of the O'Flaherty Lead Exposure Biokinetics Model (O'Flaherty,
              1993,1995). The central exchanges compartment is diffusible plasma.
              Lead distribution is represented by flows from blood plasma to liver, kidney,
              richly-perfused tissues, poorly-perfused tissues, and cortical and trabecular
              bone. The model simulates tissue growth with age, including growth and
              resorption of bone mineral.
December 2005
4-25
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Rates of bone formation and resorption are simulated as age-dependent functions
 2    (Figure 4-11).  Uptake and release of lead from trabecular bone and metabolically active cortical
 3    bone are functions of bone formation and resorption rates, respectively; this establishes the age-
 4    dependence to the lead kinetics in and out of bone.  Lead exchange between blood plasma and
 5    bone is simulated as parallel processes occurring in cortical (80% of bone volume) and trabecular
 6    bone (20% of bone volume).  The model simulates an age-related transition from immature bone,
 7    for which bone turnover (formation and resorption) rates are relatively high, to mature bone, for
 8    which turnover is relatively slow. Changes in bone mineral turnover associated with senescence
 9    (e.g., postmenopausal osteoporosis) are not represented in the model. Metabolically active
10    regions of bone, in which lead uptake and loss is dominated by bone formation and loss, a region
11    of slow kinetics in mature cortical bone is also simulated, in which lead uptake and release to
12    blood occur by heteroionic exchange with other minerals (e.g., calcium).  Heteroionic exchange
13    is simulated as a radial diffusion in bone volume of the osteon. All three processes are linked to
14    body weight, or the rate of change of weight with age.  This  approach allows for explicit
15    simulation of the effects of bone formation (e.g., growth) and loss, changes in bone volume,  and
16    bone maturation on lead uptake and release from bone. Exchanges of lead between blood plasma
17    and soft tissues (e.g., kidney and liver) are represented as flow-limited processes. The model
18    simulates saturable binding of lead in erythrocytes (maximum capacity is 2.7 mg Pb/L cell
19    volume); this replicates the curvilinear relationship between plasma and erythrocyte lead
20    concentrations observed in humans (Smith et al., 2002; Manton et al., 2001; Bergdahl et al.,
21    1999,1998,1997).  Excretory routes include kidney to urine and liver to bile. Total excretion
22    (clearance from plasma attributable to bile and urine) is simulated as a function of age-dependent
23    glomerular filtration rate.  Biliary and urinary excretory rates are proportioned as 70 and 30% of
24    the total plasma clearance, respectively.
25          The O'Flaherty model simulates lead intake from inhalation and ingestion. Inhalation
26    rates are age-dependent. Absorption of inhaled lead is simulated as a fraction (0.5) of the
27    amount inhaled and is independent of age.  Gastrointestinal absorption of lead in diet and
28    drinking water is simulated as an age-dependent fraction, declining from 0.58 of the ingestion
29    rate at birth to 0.08 after age 8 years (Figure 4-9).  These values can be factored to account for
30    relative bioavailability when applied to absorption of lead ingested in dust or soil.
      December 2005                            4-26        DRAFT-DO NOT QUOTE OR CITE

-------
             2.5
             2.0
                                     Total bone
                                 10       15
                                        Age (yr)
           20
        25
30
             0.6
             0.5
            • 0.4
                                            Juvenile cortical
                                                 bone
                               Juvenile trabecular
                                    bone
                                 10       15
                                        Age (yr)
Figure 4-11.  Bone growth as simulated by the O'Flaherty Lead Exposure Biokinetics
             Model (O'Flaherty, 1993,1995).  The model simulates an age-related
             transition from juvenile bone, in  which bone turn-over (formation and
             resorption) rates are relatively high, to mature bone, in which turn-over is
             relatively slow. Cortical bone comprises approximately 80% of total
             bone volume.
December 2005
4-27
DRAFT-DO NOT QUOTE OR CITE

-------
 1    4.5.1   Model Calibration and Evaluation
 2          The O'Flaherty model was initially calibrated to predict blood, bone, and tissue lead
 3    concentrations in rats (O'Flaherty, 1991a,b,c), and subsequently modified to reflect anatomical
 4    and physiological characteristics in children (O'Flaherty, 1995), adults (O'Flaherty,  1993), and
 5    Cynomolgus monkeys (M. fasicularis) (O'Flaherty et al., 1998). Model parameters were
 6    modified to correspond with available information on species- and age-specific anatomy and
 7    physiological processes. Empirical comparisons (largely qualitative) of model predictions
 8    against observations made in adults (e.g., Van De Vyver et al., 1988; Kehoe, 1987; Marcus,
 9    1985c; Manton and Malloy,  1983; Sherlock et al., 1982; DeSilva, 1981; Moore et al., 1977;
10    Cools et al., 1976; Rabinowitz et al., 1976; Azar et al., 1975) are provided in O'Flaherty (1993);
11    and comparisons against observations made in children (e.g., Sherlock and Quinn, 1986;
12    Bornschein et al., 1985;  Chisolm et al., 1985; Lacey et al., 1985) are described in O'Flaherty
13    (1995). Additional  discussion of model evaluation can be found in O'Flaherty (1998).
14
15    4.5.2   Model Applications
16          Biomarkers Simulated. The O'Flaherty model simulates lead concentrations in blood and
17    plasma, bone, and various soft tissues, and excretion of lead in urine that correspond to lifetime
18    exposures (in terms of daily lead intakes).  Lead in feces is a mixture of unknown proportions of
19    unabsorbed lead in food, drinking water, ingested dust, a small amount of inhaled lead entering
20    the GI tract by the mucociliary clearance from the respiratory tract, and a small amount of
21    absorbed lead eliminated with the red blood cells passing along the bile duct to the GI tract.
22    In this respect, lead in feces represents a poorly defined measure of lead exposure.
23          Lead in perspiration represents lead in extracellular plasma, but the concentration is low
24    and difficult to measure  in a small volume (1 drop  = 0.05 mL), and is potentially contaminated
25    with lead in dust on the skin surface.
26          The model predicts blood lead concentrations for a broad age range (infants to adults),
27    which allows for simulated dose reconstruction, since intakes and corresponding tissue lead
28    burdens accrued at any period in the lifetime, prior to an exposure event of interest can be
29    simulated.  Physiological states (such as pregnancy and menopause) that affect bone turnover
30    and, therefore, bone lead kinetics are not simulated, although such states could be accommodated
31    with adjustments to the physiological bone formation and resorption rates.

      December 2005                            4-28        DRAFT-DO NOT QUOTE OR CITE

-------
 1         Exposure Inputs.  The O'Flaherty model simulates lead intake by inhalation and ingestion.
 2    The model simulates ingestion exposures from infant formula, soil, dust, and drinking water.
 3    Rates of soil and dust ingestion are age-dependent, increasing to approximately 130 mg/day at
 4    age 2 years, and declining to <1 mg/day after age 10 years. However, the ACSL implementation
 5    code allows constructions of simulations with an exposure time step as small as 1 day, which
 6    would allow simulation of rapidly changing intermittent exposures (e.g., an acute exposure
 7    event).
 8         Modeling Variability and Uncertainty. The O'Flaherty model, as described in O'Flaherty
 9    (1995, 1993), utilizes point estimates for parameter values and yields point estimates as output;
10    however, a subsequent elaboration of the model has been reported that utilized a Monte Carlo
11    approach to simulate variability in exposure, absorption, and  erythrocyte lead binding  capacity
12    (Beck et al., 2001).  This approach could be used to predict the probability that children exposed
13    to lead in environmental media will have blood lead concentrations exceeding a health-based
14    level of concern (e.g., 10 jig/dL).
15
16    4.5.3   Verification/Validation of O'Flaherty Model
17         The O'Flaherty model was developed in ACSL.  A compiled C program has also been
18    developed (personal communication, E. O'Flaherty).  The extent to which code verification and
19    validation studies have been conducted for the O'Flaherty model is unclear at this time.
20    However, analogs of certain components of the O'Flaherty model (e.g., parameters related to
21    bone growth) have been incorporated into the EPA All Ages Lead Model (see Section 4.6) as a
22    potential option for evaluation.
23
24
25    4.6   EPA ALL AGES LEAD MODEL
26    4.6.1   Model Structure
27         The EPA All Ages Lead Model (AALM) (Figure 4-12), currently under development,
28    simulates lifetime lead exposures and biokinetics in humans.  The model can be used to simulate
29    exposure and biokinetics of lead from birth to age 90 years and is expected to incorporate, at
30    some near-future time, a pregnancy module that simulates transplacental transfer of lead from the
31    other to the fetus.

      December 2005                           4-29        DRAFT-DO NOT QUOTE OR CITE

-------


Respiratory
tract
                 Skeleton
        Non-
        Exchange
        Non-
        Exchange i
                                          Gastrointestinal tract
                                        Plasma extra-cellular fluid
1
Volume
;
s Exchange

' Volume
i
i
I


Cortical
Surface


w w Trabecuiar
Surface



«*,







— ».




	 te.
Figure 4-12.  Structure of the All Ages Lead Model. The AALM adds a comprehensive
             exposure component and an uptake component to a revised and recoded
             version of the Leggett model to produce a model with fully selectable
             exposure, uptake, and biokinetic parameters.
December 2005
4-30
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Exposure Module. The exposure component of the AALM incorporates and extends the
 2    exposure component of the IEUBK model. The AALM exposure model defines an individual in
 3    terms of age, sex, date of birth, and activity pattern profile. The age specification establishes up
 4    to nine age ranges (e.g., infant, child, adolescent, adult, etc.) for which various exposure (and
 5    biokinetic) parameter values can be applied. This provides a means for varying parameter values
 6    with age.  The sex specification links the modeled individual to the appropriate growth algorithm
 7    (O'Flaherty 1993, 1995), and the date specification links the individual to historical exposure
 8    levels (e.g., air,  diet) for the selected age range. The activity pattern specification sets the
 9    relative amount of time the individual spends in various exposure settings (e.g., residential,
10    school, recreational, occupational) for which exposure concentrations can be specified.
11    Parameters that can be set include sleep (hours/day); hours/week spent at the residential, school,
12    occupational, and recreational settings; and fraction of time spent indoors at each setting.
13          The exposure module simulates an average daily intake of lead (jig/day) based on user
14    defined exposure levels in air, diet,  dust (ingestion and dermal), drinking water, soil, or paint
15    chips (pica scenario). The air exposure module allows inputs for exposure levels in terms of
16    (a) outdoor air and (b) indoor residential, school, or occupational air, all as a fraction of the
17    outdoor air lead concentration as  specified. Ventilation rates (i.e., m3 air inhaled/day) can be
18    varied for each exposure scenario (e.g., residential, recreational, occupational). The model
19    calculates a weighted average amount of lead inhaled (jig/day) for the combined scenarios and
20    passes this intake rate to the biokinetic model.
21          The diet exposure module  allows input values (current or historical) for lead levels (|ig/g)
22    in market basket fruits, vegetables, meat and fish; recreational- or subsistence-harvested fish and
23    meat; and corresponding food intakes for each food type (jig food/day).  Lead intake from
24    drinking water is calculated from concentrations (|ig/L) in tap water (first draw and/or flushed),
25    fountain water, and/or bottled water; and corresponding source water intake rates (L/day).
26          The dust exposure module accepts input values for dust concentrations (|ig/g) in various
27    settings (e.g., residential, school,  recreational, occupational) or dust loadings (|ig/m2) and
28    corresponding dust ingestion rates (jig dust/day) or contact rates (m2/day), the lead ingestion rate
29    for a given loading being calculated as the product of loading and contact rate. Pica ingestion for
30    soil and/or paint chips can be simulated with input values for lead levels in soil (|ig/g) or paint
31    (jig/cm2) and corresponding pica  ingestion rates (g soil/day, cm2 paint/day).  Dermal exposure to

      December 2005                            4-31         DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead in dust can also be simulated with input values for dust lead level (ng/g), dust loading on the
 2    skin (mg/cm2), and skin exposure rate (cm2/day).
 3         Calculated lead intakes for each exposure pathway are summed to calculate total intakes
 4    (jig/day) to the respiratory tract, gastrointestinal tract, and dermal pathway, respectively.
 5    The exposure model time step is 1 day (the smallest time interval for a single exposure event).
 6         Biokinetics Module. The biokinetics module of the AALM is based on Leggett (1993)
 7    with the following modifications and enhancements (see Figure 4-8 for diagram of the
 8    Leggett model):
 9
10         1.   A simulation of dermal absorption is implemented that calculates transfer
11              of lead from the skin to the central plasma compartment, as a function of
12              rate of dermal  contact with lead (jig/day) and a dermal absorption
13              fraction.

14         2.   Male and female growth algorithms for body weight, soft tissues, and
15              cortical and trabecular bone are implemented, based on O'Flaherty
16              (1995, 1993).  This allows simulation of tissue growth and volumes,
17              as well as lead concentrations in all tissues simulated.

18         3.   A simulation of maternal-fetal transfer is implemented that simulates
19              lead levels in fetal tissues,  and establishes blood and tissue lead levels
20              for a postnatal simulation.  This provides a means for multigeneration
21              simulation of exposure and lead biokinetics.

22
23    4.6.2   Model Calibration and  Evaluation
24         The AALM currently under development incorporates key exposure model features from
25    the IEUBK model (plus age-related extensions)  and key biokinetic model features from the
26    Leggett model.  To the extent that model validation evaluations have indicated reasonably good
27    matches between IEUBK or Leggett model outputs and empirical observations, the same can be
28    reasonably expected for the AALM. However, this remains to  be verified by future AALM
29    model validation evaluations that include comparisons of AALM model run results with other
30    model outputs and with empirical observations.
31
     December 2005                            4-32       DRAFT-DO NOT QUOTE OR CITE

-------
 1    4.6.3   Model Applications
 2         Biomarkers Simulated.  The AALM simulates the concentrations of lead in blood and
 3    plasma, bone, and various soft tissues, and excretion of lead in urine that correspond to lifetime
 4    exposures, in terms of daily lead intakes.  Algorithms for transplacental transfer of lead are also
 5    planned to be incorporated soon, so that concentrations of lead in fetal tissue resulting from
 6    maternal lead exposures can be simulated.
 7         Exposure Inputs. The model simulates daily lead intakes based on inputted current or
 8    historic levels of lead in environmental media (e.g., air, diet, drinking water, dust, soil) in various
 9    exposure settings (e.g., residential, school, recreational, occupational).  The model operates with
10    a lead intake time step of 1 day, which allows simulation of rapidly changing (i.e., daily)
11    intermittent exposures.
12         Multigeneration dose reconstruction should be possible with this model, since intakes, and
13    corresponding lead burdens accrued at any period in the lifetime, prior to an exposure event of
14    interest, can be  simulated, including lead burdens received in utero.
15
16    4.6.4   Validation and Verification of AALM Implementation Code
17         The AALM is implemented as a compiled C++ program. Code verification and validation
18    are performed with each iteration of the model code as part of the model development process,
19    and these are tracked by a formal tracking process.
20
21
22    4.7   SLOPE FACTOR MODELS
23         Slope factor models have been used as simpler alternatives to compartmental models for
24    predicting blood lead concentrations, or the change in blood lead concentration associated with a
25    change in exposure (Maddaloni et al., 2005; SRC, 2003b; Abadin and Wheeler, 1997; Stern,
26    1996; Bowers et al., 1994;  Stern, 1994; Carlisle and Wade, 1992). In slope factor models, lead
27    biokinetics are represented as a linear function between the blood lead concentration and either
28    lead uptake (uptake slope factor, USF) or lead intake (intake slope factor, ISK).  The models take
29    the general mathematical forms:
     December 2005                           4-33        DRAFT-DO NOT QUOTE OR CITE

-------
 1
 2                                       PbB = E-ISF                                   (4-5)
 3
 4
 5                                  PbB = E-AF- USF                                   (4-6)
 6
 7    where PbB is the blood lead concentration, E is an expression for exposure (e.g., soil intake x
 8    soil lead concentration) and AF is the absorption fraction for lead in the specific exposure
 9    medium of interest. Intake slope factors are based on ingested rather than absorbed lead and,
10    therefore, integrate both absorption and biokinetics into a single slope factor, whereas models
11    that utilize an uptake slope factor include a separate absorption parameter. In general, slope
12    factor models predict quasi-steady state blood lead concentrations that correspond to average
13    daily lead intakes (or uptakes) that occur over sufficiently long periods to produce a quasi-steady
14    state (i.e., >75 days, ~3 times the ti/2 for elimination of lead in blood).
15          Slope factors used in various models were presented in Table 4-1.  Of the models
16    presented in Table 4-2, Bowers et al. (1994) and SRC (2003) implement uptake slope factors.
17    The slope factors used in both models (-0.4 ug/dL per ug Pb/day) are similar to biokinetic slope
18    factors predicted from the O'Flaherty model (0.65 ug/dL per ug Pb uptake/day) and Leggett
19    model (0.43 ug/dL per ug Pb uptake/day) for simulations of adult exposures (Maddaloni et al.,
20    2005). A review of reported intake slope factors relating medium-specific exposures and blood
21    lead concentrations derived from epidemiologic studies can be found in the 1986 AQCD and in
22    Abadin and Wheeler (1997).
23
24
25    4.8   MODEL COMPARISONS
26          Table 4-2 summarizes the major features of various models of human exposure that
27    predict tissue lead  burdens. The slope factor models give similar predictions of quasi-steady
28    state blood lead concentration when similar inputs and parameter values were applied to each
29    model (Maddaloni et al., 2005).
30          Lead uptake-blood lead concentration relationships in children, predicted by the IEUBK,
31    Leggett, and O'Flaherty models are shown in Figure 4-13.  In the range of uptakes shown
32    (0.1-100 ug lead absorbed/day), nonlinearity of the relationship is apparent in the Leggett
      December 2005                           4-34        DRAFT-DO NOT QUOTE OR CITE

-------
O
O
i

^
to
O
O
<*s\









^
1
0,




O
§
H
d_.
O
o

0
H
O
o
H
W
O
O
HH
H
W
Table 4-2. Summary of Models of Human Exposure that Predict Tissue


Age Exposure
Model Range Exposure Pathways Time Step Biokinetics Simulation
U.S. Environmental 0-7 yr Air 1 year Multicompartmental
Protection Agency Diet
IEUBK Model Soil/dust
White etal. (1998) Water
Other
U.S. Environmental 0-90 yr Air 1 day Multicompartmental
Protection Agency Diet
AALM (2005) Soil/dust
Water
Other



Leggett (1985) 0-Adult Intakes (inhaled, 1 day Multicompartmental
ingested, injected)




O'Flaherty (1993, 1995) 0-Adult Air 1 year Multicompartmental
Diet (code supports
Soil/dust 1 day)
Water
Other









Distribution


Biomarkers
Predicted
Blood lead


of Lead


Variability and
Uncertainty Simulation
Variability: blood lead
GSD,
Variability/uncertainty :








MCA (Griffin et al., 1999b)

Blood
Bone
Brain
Fetus
Kidney
Liver
Urine

Blood
Bone
Brain
Kidney
Liver
Urine
Blood
Bone
Brain
Kidney
Liver
Urine










Variability and uncertainty
determined by independent
assessment of multiple runs
of the model.




NA





Beck etal. (2001)







































-------
o

O
ft>
cr

O
O
^








OJ
ON

O
H
1
0
o
0
H
O
o
H
W
O
O
HH
H
W
Table 4-2 (cont'd). Summary of Models of Human Exposure that Predict Tissue Distribution of Lead


Age
Model Range Exposure Pathways
U.S. Environmental Adult Soil (supports
Protection Agency ALM other pathways)
Maddaloni et al. (2005)
California Environmental Child Air
Protection Agency, Adult Diet
Carlisle and Wade Soil/dust
(1992) Water
Bowers et al. (1994) Adult Air
Soil/dust
Water
Stern (1994, 1996) Child Dust/soil
Adult

















Biokinetics
Exposure Time Step Simulation
>3 months Uptake slope factor
(quasi-steady state)

>3 months Intake slope factor
(quasi-steady state)


<3 months Uptake slope factor
(quasi-steady state)

>3 months Intake slope factor
(quasi-steady state)

















Biomarkers Variability and
Predicted Uncertainty Simulation
Blood Variability: blood lead
GSD,

Blood Variability: blood lead
GSD,


Blood Variability: blood lead
GSD

Blood Variability: blood lead
GSD,: MCA
















-------
           60

           50
        T3
        ~B>40
 A  Leggett - Child (2-3yr)
 »  O'Flaherty - Child (2-3yr)
	IEUBK - Child (2-3yr)
               0
      20        40        60
             Lead Uptake (ug/day)
       80
100
           40
           35
           30
           25
               0
 A  Leggett - Child (2-3yr)
 «  O'Flaherty - Child (2-3yr)
	IEUBK - Child (2-3yr)
          10           20           30
              Lead Uptake (ug/day)
                  40
Figure 4-13.  Model comparison of predicted lead uptake-blood lead concentration
             relationship in children. In the range of uptakes shown, the nonlinearity of
             the relationship is apparent in the Leggett and O'Flaherty Models
             simulations, reflecting the simulation of the limited capacity of red blood cells
             to take up lead. Regression slopes (ug /dL blood per ug/day uptake) for the
             predictions ^lOug/dL are: Leggett Model, 0.88; IEUBK Model, 0.36;
             O'Flaherty Model, 0.29.
December 2005
                     4-37
DRAFT-DO NOT QUOTE OR CITE

-------
 1    and O'Flaherty models simulations. This reflects assumptions in each model regarding the
 2    limited capacity of red blood cells to take up lead.  Regression slopes (|ig/dL blood per ng/day
 3    uptake) for the predictions < 10 |ig/dL are: Leggett model, 0.88; IEUBK model, 0.36; O'Flaherty
 4    model, 0.29. The models predict an average blood lead concentration of 10 |ig/dL for the age
 5    range 2-3 years, in association with average lead uptakes (jig/day) for the same period of
 6    approximately: Leggett model, 12; IEUBK model, 29; O'Flaherty model, 36.
 7          A similar comparison of uptake-blood lead concentration relationships predicted in adults
 8    is shown in Figure 4-14. Regression slopes for adults predicted by the Leggett and O'Flaherty
 9    models (at blood lead concentrations < 10 |ig/dL) are more similar for adults (Leggett model,
10    0.54; O'Flaherty model, 0.72) than for children (see Figure 4-13 vs. Figure 4-1). The models
11    predict an average blood lead concentration of 10 |ig/dL for the age range 31-32 years, in
12    association with average lead uptakes, for the same period, of approximately 18 and 13 jig/day,
13    Leggett and O'Flaherty models, respectively.  The nonlinearity in both children and adults is due
14    largely to the limited capacity of red blood cells to take up lead at concentrations above
15    15-20 |ig/dL.  Figure 4-12 shows that the Leggett and O'Flaherty models reach this point at
16    about 30 jig Pb/day for children and about 40 ng/day for adults. The IEUBK model (for
17    children)  does not include this nonlinearity feature.
18          Comparisons of predicted bone and soft tissue lead burdens are shown in Figure 4-15.
19    Leggett and O'Flaherty models predict bone lead burdens. Both the Leggett and O'Flaherty
20    models predict a bone lead burden in adults of approximately 90 and 98% of total body burden,
21    respectively. Regression slopes (mg lead in bone per jig uptake/day) are 1.2 for the Leggett
22    model and 2.1 for the O'Flaherty model.
23          Figures 4-16 and 4-17 compare model predictions for blood lead concentration for
24    hypothetical childhood or adult lead exposures. The hypothetical child (Figure 4-16) has a
25    blood lead concentration of 2 |ig/dL at age 2 years and then experiences a 1-year exposure to
26    100 jig Pb/day.  All three models (Leggett, IEUBK, and O'Flaherty) predict a similar temporal
27    pattern of increase in blood lead concentration at the start of exposure, then attainment of a
28    quasi-steady state, followed by a decrease in blood lead concentration, with fast and slower
29    phases of the decline in blood lead concentration after the exposure ceases.  However,
30    differences in the predicted kinetics of the blood lead changes and the predicted quasi-steady
31    state blood lead concentrations are evident.  For this hypothetical scenario,  the Leggett model

      December 2005                            4-3 8       DRAFT-DO NOT QUOTE OR  CITE

-------
   60

   50

TJ 40

"§ 30
CD
            20
         DO
            10
             0
                0
                            Leggett - Adult (30-31 yr)
                           O'Flaherty - Adult (30-31 yr)
                                                                 	
20           40          60
           Lead Uptake (ug/day)
                                                           80
                        100
     Figure 4-14.  Model comparison of predicted lead uptake-blood lead concentration
                  relationships in adults.  The nonlinearity of the relationship is apparent in
                  both the Leggett and O'Flaherty Models. Regression slopes (jig /dL blood
                  per jig/day uptake) for the predictions ^lOug/dL are:  Leggett Model, 0.54;
                  O'Flaherty Model, 0.72.
 1   predicts the highest blood lead concentrations (23 |ig/dL) compared to the O'Flaherty (12 |ig/dL)
 2   and IEUBK (10 |ig/dL) models. These differences are not solely the result of different values for
 3   the absorption fraction in 2-3 year old children (Figure 4-9): Leggett model, 30%; O'Flaherty
 4   model, 45% (descending from 49% at age 2 years to 39% at age 3 years); IEUBK model, 25%
 5   (at a soil lead intake of 100 jig/day). A similar pattern is evident in the simulation of the same
 6   exposure (100 |ig/day for 1 year) in an adult (age 30 years; Figure 4-17).  The Leggett model
 7   predicts a quasi-steady state blood lead concentration of approximately 8.2 |ig/dL and the
 8   O'Flaherty model predicts 5.4 |ig/dL. However, most of this difference can be attributed to the
 9   different absorption fraction values used for adults in the two models;  15% in the Leggett model
10   and 8% in the O'Flaherty model.
     December 2005
                                    4-39
DRAFT-DO NOT QUOTE OR CITE

-------
           240

           200

        | 160

        1 120
        _i
         (!)
         §   80
        CD
              •Leggett-Adult (30-31 yr)

              O'Flaherty - Adult (30-31 yr)
                0
              20        40        60
                     Lead Uptake (ug/day)
       80
100
         CO
         CD
         CO
         O
        U)
14

12

10

 8

 6
                         -Leggett-Adult (30-31 yr)

                         O'Flaherty - Adult (30-31 yr)
               0
             20        40         60
                    Lead Uptake (ug/day)
       80
100
Figure 4-15.  Model comparison of predicted of lead uptake-bone and soft tissue lead
             burden relationship in adults. Both the Leggett and O'Flaherty Models
             predict a bone lead burden of approximately 90% and 98% of total body
             burden, respectively. Soft tissue burdens shown include blood. Regression
             slopes (mg Pb per ug uptake/day) for uptake-bone burden relationship is:
             Leggett, 1.2; O'Flaherty Model, 2.1.
December 2005
                           4-40
DRAFT-DO NOT QUOTE OR CITE

-------
              25
              20
            D>
           3-15
            ca
            0)

           13
            O
           _o
           DO
10
               0
                  0
                                        Base

                                    	Leggett Model

                                     o  IEUBK Model

                                    —— O'Flaherty Model
                          3      4

                         Age (year)
           6
7
Figure 4-16.  Comparison of model predictions for childhood lead exposure. The

             simulations are of a hypothetical child who has a blood lead concentration of

             2 ug/dL at age 2 years, and then experiences a 1-year exposure to 100 ug

             Pb/day. Default bioavailability assumptions were applied in all three models.
              10
               8
           o>
           13

           T3
           CD
           CD


           "O
           O
           o
           DO   2
                                      Leggett Model

                                      O'Flaherty Model
               0
                 27    28    29    30    31     32

                                       Age (year)
                                         33     34
                  35
Figure 4-17.  Comparison of model predictions for adult lead exposure.  The simulations

             are a hypothetical adult who has a blood lead concentration of 2 ug/dL at age

             30 years and then experiences a 1-year exposure to 100 ug Pb/day. Default

             bioavailability assumptions were applied in the Leggett and O'Flaherty

             models.
December 2005
                         4-41
DRAFT-DO NOT QUOTE OR CITE

-------
 1    4.9   CONCLUSIONS AND FUTURE DIRECTIONS
 2          Modeling of relationships between lead exposures and lead levels in tissues has advanced
 3    considerably during the past 25 years or so. Three multicompartmental exposure-biokinetics
 4    models have been developed and evaluated to varying degrees for predicting associations
 5    between exposure and body burden (TEUBK model, Leggett model, O'Flaherty model).  A fourth
 6    model, the All Ages Lead Model (AALM), is  still under development and may resolve some of
 7    the issues regarding minor discrepancies between other models, while at the same time adding
 8    new features directly applicable to risk assessment.
 9          The IEUBK model has had the most extensive application in the regulatory context, as
10    EPA guidance recommends that, where possible, risk estimates for residential exposures to lead
11    at hazardous waste sites be based on IEUBK model predictions of blood lead concentrations in
12    children.  Although, these models are constructed very differently (e.g., the O'Flaherty
13    biokinetics model has only 17 lead parameters, compared to 65 in the Leggett biokinetics model,
14    and 47 in the IEUBK biokinetics model), the three models yield remarkably similar predictions
15    of blood lead concentration for similar hypothetical exposure scenarios. The three models
16    predict similar kinetics of change of blood lead concentrations in association with a change in
17    lead exposure (e.g., Figures 4-16 and 4-17). Both the Leggett and O'Flaherty models predict
18    similar rates of lead accumulation in bone, for the same rates of uptake of lead into the body.
19    Predictions of quasi-steady state blood lead concentrations for the scenarios are simulated in
20    Figures 4-16 and 4-17 and differ across models by a factor of approximately 2. While this
21    magnitude of difference may be substantial in the context of regulatory use of the models (e.g.,
22    for establishing cleanup goals at hazardous waste sites), it represents a remarkable convergence
23    of various approaches taken to reduce the complex biokinetics of lead to tractable, and relatively
24    simple, mathematical expressions.  Given that the AALM incorporates and combines key
25    features from predecessor models (especially exposure components of the IEUBK and the
26    Leggett biokinetics components) it is reasonable to expect likely convergence of its outputs with
27    those of such predecessor models, but this remains to be verified by future validation evaluations
28    for the AALM.
29          Several major challenges remain to be confronted in further developing our ability to
30    simulate lead exposure-tissue level relationships in real individuals or populations.  The three
31    earlier mechanistic models described above do not simulate the kinetics of lead in pregnancy or

      December 2005                            4-42       DRAFT-DO NOT QUOTE OR CITE

-------
 1    in senescence (e.g., menopause). Only one of these three earlier models (Leggett) simulates lead
 2    levels in brain, a potential target organ for lead toxicity.  None of the models have been
 3    rigorously evaluated for accuracy of predictions of bone lead levels in humans, for which there is
 4    a rapidly expanding set of observations of importance to dose-response assessment.  The fourth
 5    multicompartmental model discussed above, the EPA AALM currently under development
 6    simulates lead kinetics out to age 90 years and include features designed to simulate both
 7    maternal and fetal lead biokinetics during pregnancy  (as a future option). In addition to these
 8    useful features of the biokinetics model, the AALM incorporates a life-time exposure module to
 9    simulate complex life-time exposure patterns (i.e., complex temporal patterns of exposure to
10    multiple exposure media and in multiple exposure settings).  These exposure simulations can be
11    used to drive simulations of lead biokinetics, based on the AALM biokinetics model or other
12    lead biokinetics models.
13         While extending the functionality of the models, as noted above, the AALM also provides
14    important insight with regard to desirable future directions in model development. Of great
15    importance for regulatory uses of the models, for example, is the need for more rigorous
16    quantitative assessment of confidence (i.e., uncertainty) in model predictions. To date, such
17    assessments have not been applied uniformly in a manner that allows cross-model comparisons
18    of confidence for specific regulatory uses.
19         The IEUBK Model has undergone the most extensive and thoroughly reported evaluation
20    of a regulatory use of the model, i.e., (a) quantitative  evaluation of predicted distributions of
21    blood lead concentrations in  children who live in areas for which cross-sectional measurements
22    of environmental lead levels  were available and (b) independent verification of the IEUBK
23    model implementation code (Hogan et al., 1998; Zaragoza and Hogan, 1998). However, a
24    similar  level of evaluation of the Leggett and O'Flaherty models has not been reported, although
25    specific predictions of the models have been evaluated against observations (e.g.,
26    experimentally- observed kinetics of change in blood lead following a change in intakes).
27    Nor has the AALM yet undergone a similar level of evaluation as the IEUBK model.
28         To a large extent, the important information gap regarding evaluation of model confidence
29    derives from a lack of observational data and/or public access to observational data on which
30    predictions could be evaluated.  An additional challenge for applications of the models in a
31    regulatory context relates to uncertainties in exposure data from which exposure model inputs

      December 2005                            4-43        DRAFT-DO NOT QUOTE OR CITE

-------
1    are derived. Model development and uncertainty assessment could be substantively advanced by
2    assembling verified (for accuracy) sets of data on lead biokinetics against which models could be
3    uniformly evaluated. Examples of the types of data that would be valuable include data on the
4    kinetics of change in blood or tissue lead concentrations, or stable lead isotope ratios, in response
5    to a change in exposure. Also, access to large data bases that include reported lead exposure
6    measurements for various media that are paired with blood or tissue lead measurements for
7    individuals affected by pertinent exposure scenarios would also be extremely valuable for cross-
8    model evaluations.
    December 2005                            4-44        DRAFT-DO NOT QUOTE OR CITE

-------
  1     4.10   REFERENCES

  2     Abadin, H. G.; Wheeler, J. S. (1997) Guidance for risk assessment of exposure to lead: a site-specific, multi-media
  3            approach. In: In Hazardous Waste and Public Health: International Congress on the Health Effects of
  4            Hazardous Waste, Andrews, J. S.; Frumkin, H.; Johnson, B.L.; Mehlman, M.A.; Xintaras, C.; Bucsela, J.A.,
  5            eds. Princeton Scientific Publishing Co, Princeton: pp. 477-485.Azar, A.; Snee, R. D.; Habibi, K. (1975) An
  6            epidemiologic approach to community air lead exposure using personal samplers. In: Griffin, T. B.;
  7            Knelson, J. H., eds. Lead. Stuttgart, Federal Republic of Germany: Georg Thieme Publishers; pp. 254-290.
  8            (Coulston, F.; Korte, F., eds. Environmental quality and safety: supplement v. 2).
  9     Barry, P. S. I. (1975) A comparison of concentrations of lead in human tissues. Br. J. Ind. Med. 32: 119-139.
10     Barry, P. S. I. (1981) Concentrations of lead in the tissues of children. Br. J. Ind. Med. 38: 61-71.
11     Beck, B. D.; Mattuck, R. L.; Bowers, T. S.; Cohen, J. T.; O'Flaherty, E. (2001) The development of a stochastic
12            physiologically-based pharmacokinetic model for lead. Sci. Total Environ. 274: 15-19.
13     Bergdahl, I. A.; Schutz, A.; Gerhardsson, L.; Jensen, A.; Skerfving, S. (1997) Lead concentrations in human plasma,
14            urine and whole blood. Scand. J. Work Environ. Health 23: 359-363.
15     Bergdahl, I. A.; Sheveleva, M; Schutz, A.; Artamonova, V. G.; Skerfving, S. (1998) Plasma and blood lead in
16            humans: capacity-limited binding to  "delta"-aminolevulinic acid dehydratase and other lead-binding
17            components. Toxicol. Sci. 46: 247-253.
18     Bergdahl, I. A.; Vahter, M.; Counter, S. A.; Schutz, A.; Buchanan, L. H.; Ortega, F.; Laurell, G.; Skerfving, S.
19            (1999) Lead in plasma and whole blood from lead-exposed children. Environ. Res. 80: 25-33.
20     Bert, J. L.; van Dusen, L.  J.; Grace, J. R. (1989) A generalized model for the prediction lead body burdens. Environ.
21            Res. 48: 117-127.
22     Bornschein, R. L.; Hammond, P. B.; Dietrich, K. N.; Succop, P.; Krafft, K.; Clark, S.; Berger, O.; Pearson, D.;
23            Que Hee, S. (1985) The Cincinnati prospective study of low-level lead exposure and its effects on child
24            development: protocol and status report. Environ. Res. 38: 4-18.
25     Bowers, T. S.; Beck, B. D.; Karam, H. S. (1994) Assessing the relationship between environmental lead
26            concentrations and adult blood lead levels. Risk Anal. 14: 183-189.
27     Bowers, T. S.; Mattuck, R. L. (2001) Further comparisons of epidemiological data with predictions of the integrated
28            exposure uptake biokinetic model for lead in children. Hum. Ecol. Risk Assess. 7: 1699-1713.
29     Campbell, B. C.; Meredith, P. A.; Moore, M. R.; Watson, W. S. (1984) Kinetics of lead following intravenous
30            administration in man Toxicol. Lett.  21: 231-235.
31     Carlisle, J. C.; Wade, M. J. (1992) Predicting blood lead concentrations from environmental concentrations.
32            Regul. Toxicol. Pharmacol. 16: 280-289.
33     Chamberlain, A. C.; Heard, M. J.; Little, P.; Newton, D.; Wells, A. C.; Wiffin, R. D. (1978) Investigations into lead
34            from motor vehicles. Harwell, United Kingdom: United Kingdom Atomic Energy Authority; report no.
35            AERE-R9198.
36     Chisolm, J. J., Jr.; Mellits, E. D.; Quaskey, S. A. (1985) The  relationship between the level of lead absorption in
37            children and the age, type, and condition of housing. Environ. Res. 38: 31-45.
38     Cools, A.; Salle, H. J. A.; Verberk, M. M.; Zielhuis, R. L. (1976) Biochemical response of male volunteers ingesting
39            inorganic lead for 49 days. Int. Arch. Occup. Environ. Health 38: 129-139.
40     DeSilva, P. E. (1981) Determination of lead in plasma and studies on its relationship to lead in erythrocytes. Br. J.
41            Ind. Med. 38: 209-217.
42     Goodrum, P. E.; Diamond, G. L.; Hassett, J. M.; Johnson, D. L. (1996) Monte Carlo modeling of childhood lead
43            exposure: development of a probabilistic methodology for use with the U.S. EPA IEUBK model for lead in
44            children. Hum. Ecol. Risk Asses. 2: 681-708.
45     Griffin, T. B.; Coulston, F.; Wills, H.; Russell, J. C.; Knelson, J. H. (1975) Clinical studies on men continuously
46            exposed to airborne paniculate lead.  In: Griffin, T. B.; Knelson, J. H., eds. Lead. Stuttgart, Federal Republic
47            of Germany: Georg Thieme Publishers; pp. 221-240. (Coulston, F.; Korte, F., eds. Environmental quality
48            and safety: supplement v. 2).
49     Griffin, S.; Goodrum, P. E.; Diamond, G. L.; Meylan, W.; Brattin, W. J.; Hassett, J. M. (1999a) Application of a
50            probabilistic risk  assessment methodology to a lead smelter site. Hum. Ecol. Risk Assess. 5: 845-868.
51     Griffin, S.; Marcus, A.; Schulz, T.; Walker, S. (1999b) Calculating the interindividual geometric standard deviation
52            for use in the integrated exposure uptake biokinetic model for lead in children. Environ. Health Perspect.
53            107:481-487.
54     Gross, S. B.; Pfitzer, E. A.; Yeager, D. W.; Kehoe, R. A. (1975) Lead in human tissues. Toxicol. Appl. Pharmacol.
55            32:638-651.
       December 2005                                 4-45          DRAFT-DO NOT QUOTE OR CITE

-------
  1     Hogan, K.; Marcus, A.; Smith, R.; White, P. (1998) Integrated exposure uptake biokinetic model for lead in
  2            children: empirical comparisons with epidemiologic data. Environ. Health Perspect. 106(suppl.
  3            6): 1557-1567.
  4     Hursh, J. B.; Mercer, T. T. (1970) Measurement of 212 Pb loss rate from human lungs. J. Appl. Physiol.
  5            28:268-274.
  6     Hursh, J. B.; Schraub, A.; Saltier, E. L.; Hofmann, H. P. (1969) Fate of 212Pb inhaled by human subjects. Health
  7            Phys. 1969 16: 257-267.
  8     International Commission on Radiological Protection. (1989) Age-dependent doses to members of the public from
  9            intake of radionuclides: part 1. New York, NY: Pergamon Press. (ICRP publication 56; Annals of the ICRP:
10            v. 20, no. 2).
11     International Commission on Radiological Protection. (1993) Age-specific biokinetics for the alkalille earth
12            elements. In: Age-dependent doses to members of the public from intake of radionuclides: part 2. Ingestion
13            dose coefficients. New York, NY: Elsevier  Science, Inc.; pp. 95-120. (ICRP publication no. 67, appendix
14            A).
15     Kehoe, R. A. (1987) Studies of lead administration and elimination in adult volunteers under natural and
16            experimentally induced conditions over extended periods of time. Food Chem. Toxicol. 25: 425-493.
17     Khoury, G. A.; Diamond, G L. (2003) Risks to children from exposure to lead in air during remedial or removal
18            activities at Superfund sites: a case study  of the RSR lead smelter superfund site. J. Expo. Anal. Environ.
19            Epidemiol. 13: 51-65.
20     Lacey, R. F.; Moore, M. R.; Richards, W. N. (1985) Lead in water, infant diet and blood: the Glasgow Duplicate
21            Diet Study. Sci. Total Environ. 41: 235-257.
22     Leggett, R. W. (1985) A model of the retention, translocation and excretion of systemic Pu. Health Phys.
23            49: 1115-1137.
24     Leggett, R. W. (1992a) A retention-excretion model for americium in humans. Health Phys. 62:  288-310.
25     Leggett, R. W. (1992b) A generic age-specific biokinetic model for calcium-like elements. Radiat. Prot. Dosim.
26            41: 183-198.
27     Leggett, R. W. (1993) An age-specific kinetic model of lead metabolism in humans. Environ. Health Perspect.
28            101:598-616.
29     Lorenzana, R. M.;  Troast, R.; Klotzbach, J. M.; Follansbee, M.  H.; Diamond, G. L. (2005) Issues related to time
3 0            averaging of exposure in modeling risks associated with intermittent exposures to lead. Risk Anal.
31            25:169-178.
32     Maddaloni, M.; Ballew, M.; Diamond, G.; Follansbee, M. H.; Gefell, D.; Goodrum, P.; Johnson, M.; Koporec, K.;
33            Khoury, G.; Luey, J.; Odin, M.; Troast, R.;  VanLeeuwen, P.; Zaragoza, L. (2005) Assessing nonresidential
34            lead risks at hazardous waste sites. Hum.  Ecol. Risk Assess. 11: 1-37.
35     Manton, W. I.; Cook, J. D. (1984) High accuracy  (stable isotope dilution) measurements of lead in serum and
36            cerebrospinal fluid. Br. J. Ind. Med. 41: 313-319.
37     Manton, W. I.; Malloy, C. R. (1983) Distribution  of lead in body fluids after  ingestion of soft solder. Br. J. Ind. Med.
38            40:51-57.
39     Manton, W. I.; Rothenberg, S. J.; Manalo, M. (2001) The lead content of blood serum. Environ. Res. 86: 263-273.
40     Marcus, A. H. (1985a) Multicompartment kinetic  models for lead. I. Bone diffusion models for long-term retention.
41            Environ. Res. 36: 441-458.
42     Marcus, A. H. (1985b) Multicompartment kinetic models for lead. II. Linear  kinetics and variable absorption in
43            humans without excessive lead exposures. Environ. Res. 36: 459-472.
44     Marcus, A. H. (1985c) Multicompartment kinetic  model for lead. III. Lead in blood plasma and erythrocytes.
45            Environ. Res. 36: 473-489.
46     Mickle, M. H. (1998) Structure, use and validation of the IEUBK model. Environ. Health Perspect.: in press.
47     Moore, M. R.; Meredith, P. A.; Campbell, B. C.; Goldberg, A.; Pocock, S. 1.  (1977) Contribution of lead in drinking
48            water to blood-lead. Lancet 2(8039): 661-661.
49     Morrow, P. E.; Beiter, H.; Amato, F.; Gibb, F. R.  (1980) Pulmonary retention of lead: an experimental study in man.
50            Environ. Res. 21: 373-384.
51     Mushak, P. (1991) Gastro-intestinal absorption of lead in children and adults: overview of biological and
52            biophysico-chemical aspects. Chem. Speciation Bioavailability 3(3/4): 87-104.
53     O'Flaherty, E. J. (1991a) Physiologically based models for bone-seeking elements. I. Rat skeletal and bone growth.
54            Toxicol. Appl. Pharmacol. Ill:  299-312.
55     O'Flaherty, E. J. (1991b) Physiologically based models for bone-seeking elements: II. kinetics of lead disposition in
5 6            rats. Toxicol. Appl. Pharmacol.  111:313-331.
       December 2005                                 4-46          DRAFT-DO NOT QUOTE OR CITE

-------
  1     O'Flaherty, E. J. (1991c) Physiologically based models for bone-seeking elements. III. Human skeletal and bone
  2            growths. Toxicol. Appl. Pharmacol. Ill: 332-341.
  3     O'Flaherty, E. J. (1993) Physiologically based models for bone-seeking elements. IV. Kinetics of lead disposition in
  4            humans. Toxicol. Appl. Pharmacol. 118: 16-29.
  5     O'Flaherty, E. J. (1995) Physiologically based models for bone-seeking elements: V. Lead absorption and
  6            disposition in childhood. Toxicol. Appl. Pharmacol. 131: 297-308.
  7     O'Flaherty, E. J. (1998) A physiologically-based kinetic model for lead in children and adults. Environ. Health
  8            Perspect.: in press.
  9     O'Flaherty, E. J.; Inskip, M. J.; Franklin, C. A.; Durbin, P. W.; Manton, W. I.; Baccanale, C. L. (1998) Evaluation
10            and modification of a physiologically based model of lead kinetics using data from a sequential isotope
11            study in cynomolgus monkeys. Toxicol. Appl. Pharmacol. 149: 1-16.
12     Oreskes, N. (1998) Evaluation (not validation) of quantitative models. Environ. Health Perspect.: in press.
13     Pounds, J. G.; Leggett, R. W. (1998) The ICRP age-specific biokinetic model for lead: validations, empirical
14            comparisons, and explorations. Environ. Health Perspect.: in press.
15     Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1976) Kinetic analysis of lead metabolism in healthy humans.
16            J. Clin. Invest. 58: 260-270.
17     Schroeder, H. A.; Tipton, I. H. (1968) The human body burden of lead. Arch. Environ. Health 17: 965-978.
18     Sherlock, J. C.; Quinn, M. J. (1986) Relationship between blood and lead concentrations and dietary lead intake in
19            infants: the Glasgow Duplicate Diet Study 1979-1980. Food Addit. Contam. 3: 167-176.
20     Sherlock, J.; Smart, G.; Forbes, G. L; Moore, M. R.; Patterson, W. J.; Richards, W. N.; Wilson, T. S. (1982)
21            Assessment of lead intakes and dose-response for a population in Ayr exposed to a plumbosolvent water
22            supply. Hum. Toxicol.  1:  115-122.
23     Skerfving,  S.; Ahlgren,  L.; Christoffersson, J -O. Haeger-Aronson, B.; Mattsson, S.; Schutz, A; Lindberg, G. (1985)
24            Metabolism of inorganic lead in man. Nutr. Res. (Suppl. 1): 601.
25     Smith, D.; Hernandez-Avila, M.; Tellez-Rojo, M.M.; Mercado, A.; Hu, H. (2002) The relationship between lead in
26            plasma and whole blood in women. Environ. Health Perspect. 110: 263-268.
27     Stern, A. H. (1994) Derivation of a target level of lead in soil at residential sites corresponding to a de minimis
28            contribution to blood lead concentration. Risk Anal. 14(6): 1049-1056.
29     Stern, A. H. (1996) Derivation of a target concentration of Pb in soil based on elevation of adult blood pressure.
30            Risk Anal. 16(2): 201-210.
31     Syracuse Research Corporation (SRC). (2003) Evaluation of the ICRP lead biokinetics model: empirical
32            comparisons with observations of plasma-blood lead concentration relationships in humans [draft final].
33            Washington, DC: U.S. Environmental Protection Agency, Office of Emergency and Remedial Response;
34            contract no. GS-10F-0137K; SRC no. FA332.
3 5     TerraGraphics Environmental Engineering, Inc. (2001) Final human health risk assessment for the Coeur d'Alene
3 6            Basin extending from Harrison to Mullan on the Coeur d'Alene River and tributaries remedial
37            investigation/feasibility study. Washington, DC: U.S. Environmental Protection Agency, prepared for the
3 8            Idaho Department of Health and Welfare, Idaho Department of Environmental Quality.
39     U.S. Environmental Protection Agency. (1986) Air quality criteria for lead. Research Triangle Park, NC: Office of
40            Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
41            EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.
42     U.S. Environmental Protection Agency. (1994a) Guidance manual for the integrated exposure uptake biokinetic
43            model for lead in children. Washington, DC: Office of Emergency and Remedial Response; report no.
44            EPA/540/R-93/081. Available from: NTIS, Springfield, VA; PB93-963510.
45     U.S. Environmental Protection Agency. (1994b) Technical support document: parameters and equations used in
46            integrated exposure uptake biokinetic model for lead in children (v 0.99d). Washington, DC: Office of Solid
47            Waste and Emergency Response; report no. EPA/540/R-94/040. Available from:  NTIS, Springfield, VA;
48            PB94-963505.
49     U.S. Environmental Protection Agency. (1996) Recommendations of the Technical Review Workgroup for Lead for
50            an interim approach to assessing risks associated with adult exposures to lead in soil. Draft report.
51            Washington, DC: Technical Review Workgroup for Lead. Available:
52            www.epa.gov/superfund/programs/lead/products/adultpb.pdf [1999,  November 23].
53     U.S. Environmental Protection Agency. (1997) Health risks from low level environmental exposures to
54            radionuclides. Washington, DC: U.S. Environmental Protection Agency. Federal Guidance Report No. 13.
55            EPA402-R-97-014.
       December 2005                                 4-47         DRAFT-DO NOT QUOTE OR CITE

-------
 1    U. S. Environmental Protection Agency. (2005) All ages lead model [draft version 1.05]. Research Triangle Park,
 2           NC: National Center for Environmental Assessment.
 3    VanDe Vyver, F. L.; D'Haese, P. C.; Visser, W. I; Elseviers, M. M; Knippenberg, L. I; Lamberts, L. V.; Wedeen,
 4           R. P.; De Broe, M. E. (1988) Bone lead in dialysis patients. Kidney Int. 33: 601-607.
 5    Wells, A. C.; Venn, J. B.; Heard, M. J. (1975) Deposition in the lung and uptake to blood of motor exhaust labelled
 6           with 203Pb. Inhaled Part. 4: 175-189.
 7    White, P. D.; Van Leeuwan, P.; Davis, B. D.; Maddaloni, M.; Hogan, K. A.; Marcus, A. H.; Elias, R. W. (1998)
 8           The conceptual structure of the integrated exposure uptake biokinetic model for lead in children. Environ.
 9           Health Perspect. 106(suppl. 6): 1513-1530.
10    Zaragoza, L.; Hogan, K. (1998) The integrated exposure uptake biokinetic model for lead in children: independent
11           validation and verification. Environ. Health Perspect. 106(suppl. 6): 1551-1556.
12
      December 2005                                4-48          DRAFT-DO NOT QUOTE OR CITE

-------
 i           5.    TOXICOLOGICAL EFFECTS OF LEAD IN
 2      LABORATORY ANIMALS, HUMANS, AND IN VITRO
 3                                 TEST SYSTEMS

 4
 5   5.1    INTRODUCTION
 6         As noted in Chapter 1, air quality criteria documents evaluate scientific knowledge of
 7   relationships between pollutant concentrations and their effects on the environment and public
 8   health. Chapters 2 and 3 of this document discussed the chemistry and physical properties of
 9   lead (Pb); sources, emissions, transport, and deposition of Pb; and environmental concentrations
10   and pathways to human exposure. Chapter 4 discussed models of human exposure that predict
11   tissue distribution of lead.  This chapter (Chapter 5) assesses information regarding the
12   toxicological effects of Pb in laboratory animals, humans, and in vitro test systems. Emphasis is
13   placed here on qualitative characterization of various Pb-induced effects, with attempts to define
14   dose-effect relationships for the key health effects that are thought to occur at ambient exposure
15   levels encountered by the general population of the United States. Chapter 6 follows with a
16   discussion of epidemiologic studies of ambient Pb-exposure effects.  Chapter 7 provides an
17   integrative synthesis of information on Pb exposures and health effects. The environmental
18   effects of Pb are discussed in Chapter 8.
19         The framework used here for presenting the toxicologic effects of Pb is subdivided mainly
20   according to organ systems.  As noted in the 1986 Pb AQCD, this facilitates presentation of the
21   information, but it must be stressed that all systems are interdependent, functioning in delicate
22   concert to preserve the physiological integrity of the whole organism.
23         The information discussed in this chapter is derived from a very wide body of literature on
24   studies in humans, laboratory animals, and in vitro test systems of animal cell lines and organ
25   systems that may mimic responses in intact animals. This chapter is not intended to be a
26   compendium of all that is known about lead; rather, it is an update of the reported biological
27   effects from the last previous Pb AQCD (U.S. Environmental Protection Agency, 1986), the
28   Addendum to that document (Lead Effects on Cardiovascular Function, Early Development, and
29   Stature) (U.S. Environmental Protection Agency, 1986), and the Supplement to the 1986
30   Addendum (U.S. Environmental Protection Agency, 1990).  The historical Pb literature is briefly

     December 2005                           5-1        DRAFT-DO NOT QUOTE OR CITE

-------
 1    summarized at the opening of each section or subsection and is intended as a very concise
 2    overview of previous work  The reader should refer to the previous documents listed above for
 3    more detailed discussion of the literature prior to the late 1980s. Each section then continues
 4    with brief discussions of key studies published since 1986.  Longer discussions of the newly
 5    available studies are included where warranted.  Sections are ended with comparisons of data
 6    from the 1986 AQCD with new data, and basic conclusions are drawn. More detailed summaries
 7    of newly available studies and results are provided in tables in Annex AX5.
 8
 9
10    5.2   EFFECTS OF LEAD ON HEME SYNTHESIS
11    5.2.1    Effects of Lead on Erythrocyte Biology and Function
12          Lead poisoning is one of the most common acquired environmental diseases, because of
13    physical properties of the metal and its widespread distribution in the environment. It is a
14    complex disorder affecting several organs in the body, including developing erythrocytes (red
15    blood cells [RBCs]).  Anemia is frequently observed with Pb poisoning and is thought to result
16    from the shortening of erythrocyte life span and is also due to the effects of Pb on hemoglobin
17    synthesis. However, the exact mechanisms by which Pb affects the red blood cell (RBC) life
18    span and heme synthesis are not clear. It is postulated that the mechanisms may be due to the
19    effects of Pb on iron uptake; Pb poisoning also causes an increased urinary excretion of
20    porphyrins and 5-aminolevulinic acid (ALA), the first precursor for heme synthesis. In addition,
21    the striking similarities between Pb poisoning and acute intermittent porphyria (the disease
22    associated with lesions in the heme biosynthetic enzyme, porphobilinogen deaminase) strongly
23    suggests that one of the major sites of Pb intoxication is the heme biosynthetic pathway.
24          The 1986 Pb AQCD presented a concise summary of literature available at that time from
25    both animal and human studies indicating potential effects of Pb intoxication on enzymes and
26    precursors involved in heme synthesis, erythrocyte morphology and function as well as the
27    influence of these perturbations on the nervous system and vitamin D metabolism and associated
28    physiological process. In summary, these studies reported an association between increased Pb
29    exposure and increased ALA-S activity (which is increased in kidney with acute exposure and in
30    spleen with chronic exposure, while it decreased in liver tissue in both the exposure scenarios).
31    The activity of ALA-D appeared to be inversely correlated to blood Pb values and was found to

      December 2005                            5-2        DRAFT-DO NOT QUOTE OR CITE

-------
 1   be inhibited in several tissues. It was also inferred from several animal studies that the effect of
 2   Pb on heme formation involved both ferrochelatase inhibition and impaired mitochondrial
 3   transport of iron.  Human studies indicated that occupational exposure to Pb results in decreased
 4   erythrocyte cell survival and alterations in erythrocyte membrane integrity and energetics.  The
 5   vast scientific literature on the effects of Pb on various aspects of heme metabolism in diverse
 6   organ systems both in human and animals has accumulated over the past two decades.
 7   Recognizing the magnitude of this literature, this chapter is primarily concerned with discussions
 8   of data from animal and in vitro studies, while the human studies are dealt with in Chapter 6.
 9
10   5.2.2  Effects of Lead on Erythrocyte Functions
11         The cellular membrane is one of the main targets for toxic effects of heavy metals,
12   including Pb. Anemia, one of the clinical symptoms of Pb intoxication, can develop because of
13   impairment of hemoglobin synthesis and damage of erythrocyte membranes by Pb ions.
14   Although, erythrocyte membrane is not as specialized as other cell membranes are, it carries out
15   important functions common to other cell membranes, such as active and passive transport and
16   the production of ionic and electric gradients. Changes in erythrocyte membrane lipid and
17   protein profiles can alter the membrane fluidity, potentially affecting enzymatic activity and the
18   functionality of receptors and ion channels present on the plasma membrane and also can
19   influence the ionic and molecular composition of intracellular spaces.
20
21   Lead Uptake. Binding, and Transport
22         Studies by Simons (1986a) indicated that the uptake of Pb into human RBCs is a passive
23   process, i.e., it does not require the use of energy in the form of ATP. In addition, Pb may be
24   able to cross the membrane passively in either direction.  This process involves anion transport
25   mechanisms, as the characteristic anion exchange inhibitors have been found to inhibit the
26   passive uptake of Pb by RBCs (Simons, 1986a,b). It has also been demonstrated that the
27   transport of Pb across the membrane depends on the presence of another anion, the bicarbonate
28   ion, and is transported as Pb-carbonate (Simons, 1986a). When Pb enters the cell, it binds
29   mainly to hemoglobin, and the ratio of bound to free Pb in cytoplasm has been estimated to be
30   6000:1. Simons (1986a,b) carried out studies using citrate buffers, which may cause hemolysis
31   of RBCs.  To avoid the influence of a citrate buffer, Sugawara et al. (1990) measured the uptake

     December 2005                            5-3         DRAFT-DO NOT QUOTE OR CITE

-------
 1    of Pb into human RBCs by adding Pb directly into plasma. These investigators also found that
 2    the transport of Pb across the erythrocyte membrane is energy-independent (passive) and carrier
 3    mediated. Little release of Pb from the cells was observed, suggesting absence of any hemolysis
 4    of the cells in this protocol.  Furthermore, the progressive accumulation of Pb was not observed.
 5    More than 98% of the Pb was found accumulated in the cytoplasm in protein-bound form, while
 6    only 2% was found in the membrane fraction.  Sugawara et al. (1990) also reported finding
 7    45 Pb-binding sites on human hemoglobin. On the other hand, studies reported by Bergdahl
 8    et al.  (1997) using liquid chromatography coupled with inductive plasma mass spectrometry
 9    analysis suggested aminolevulinic acid dehydratase (ALAD), the enzyme involved in the heme
10    synthesis pathway, to be the principle Pb-binding protein, not hemoglobin, as previously thought.
11          Additional studies carried out by Simons (1993a) evaluated the transport of Pb into RBCs
12    for cell Pb contents in the range of 1 to 10 jiM  and reported that 203Pb uptake was mediated by an
13    anion exchanger and the efflux was mediated through a vanadate-sensitive pathway identified
14    with the calcium pump (Simons, 1988). He further concluded that the high ratio of RBC to
15    plasma Pb observed in vivo was due to a labile Pb-binding component within the cytoplasm.
16    Simons (1993a) also observed that exit of Pb ions from the RBC was much lower than expected
17    based on his earlier work with erythrocyte ghosts. Utilizing a group of drugs that modify anion
18    exchange and thiol groups in the cytoplasm,  Lai et al. (1996) showed that anion exchange
19    mechanisms and thiol groups were critical factors in how Pb stimulates calcium-dependent
20    processes in erythrocytes. Once the role of anion exchanger proteins had been implicated in Pb
21    transport in erythrocytes, Bannon et al. (2000) investigated whether similar anion exchange
22    processes are involved in the uptake and transport of Pb in other cells, such as Madin-Darby
23    canine kidney epithelial cells. Based on a comparative in vitro study using human  erythrocytes
24    and canine kidney epithelial cells, these authors reported transport of Pb in kidney epithelial
25    cells, suggesting similar anion exchange involvement.
26
27    Erythrocyte Survival, Mobility, and Membrane Integrity
28          It is well recognized that Pb intoxication interferes with RBC survival by  shortening the
29    life span and altering the mobility of the erythrocytes; however, the molecular mechanisms
30    behind these effects of Pb on erythrocyte functions are not well understood.  The shape and
31    deformability of the human erythrocyte, or RBC is maintained by several factors including low

      December 2005                            5-4        DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentration of free intracellular Ca2+ (<0.1 jiM) and a replenished ATP level. An elevated
 2    interfacial Ca2+ concentration inside the RBC activates the passive ion efflux via a K+ selective
 3    (voltage independent) channel and a concomitant water transport (Gordos effect). Low
 4    concentrations of Pb ions can mimic Ca2+ and activate the same channel in the RBC.
 5          Intraperitonially injected Pb significantly decreases rat erythrocyte membrane mobility
 6    (Terayama et al., 1986), an effect evident to some extent even below blood Pb concentration of
 7    100 jig/100 ml. This decrease in rat erythrocyte mobility was found simultaneous or prior to
 8    changes in hematological parameters such as hemoglobin (Hb) levels and hematocrits (Hct). The
 9    same group (Terayama and Muratsugu, 1988) also reported a significant decrease in erythrocyte
10    membrane sialic acid content at the same levels of blood Pb with exposure to Pb (20 mM
11    Pb-acetate once a week for 5 weeks).  Additional studies by the same group reported that other
12    hematological parameters, such as mean corpuscular volume (MCV), mean  corpuscular
13    hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC), were also
14    significantly  decreased upon Pb exposure, along with decreased mobility, sialic acid content, and
15    deformability of rat RBCs. It was speculated that Pb-induced decreases in sialic acid content and
16    deformability of RBCs shorten RBC survival time and may lead to anemia in Pb poisoning.
17    Jehan and Motlag (1995) reported Pb exposure caused significant change in RBC membrane
18    cholesterol and phospholipid contents along with sialic acid. Coexposure to Zn was found to
19    reduce these  alterations.
20          Pb-induced morphological changes in human RBC were studied by Eriksson and Bering
21    (1993) using electron paramagnetic resonance imaging. These authors reported that Pb ions
22    (a) induced time-dependent changes in MCV and cell shrinkage and (b) inhibited the Gardos
23    effect. Trialkyl-Pb compounds have also been reported to induce hemolytic activity in
24    erythrocytes, with intensity increasing with hydrophobicity of the compounds (Kleszcynska
25    et al., 1997).  Serrani et al. (1997)  reported that Pb ions confer protection against RBC lysis in
26    hypotonic low ionic strength media, presumably be due to interaction of Pb  with certain
27    constituents in the cell membrane. This resistance to erythrocyte lysis was found to significantly
28    increase with Pb (20 to 25 jiM) compared to other metals such as Al, Cd, and Zn (Corchs et al.,
29    2001). The Pb-induced reduction in MCV (RBCs derived from umbilical cord) was found to be
30    reversed when the cells were treated with quinidine, an inhibitor of a potassium channel
31    activator, without any effect on resistance to cell lysis, suggesting changes in cell membrane

      December 2005                            5-5       DRAFT-DO NOT QUOTE OR CITE

-------
 1    structure. This effect may also be involved in membrane deformability (Mojzis and Nistiar,
 2    2001).
 3          Heavy metals, including Cd, Zn, and Pb, have been found to alter RBC membrane
 4    microviscosity and fluidity (Amoruso et al., 1987).  These authors labeled RBC membranes with
 5    fluorescent lipid probe all trans 1, 6-diphenyl-l,3,5-hexatriene (DPH) and demonstrated
 6    increased polarization with increased membrane lipid viscosity on exposure to heavy metals.
 7    They also postulated that such alterations in cell membrane lipid and possibly also protein
 8    fluidity may contribute to abnormal cellular function.  Similar changes in RBC fluidity were
 9    observed in the RBC collected from workers exposed to Pb (Cook et al., 1987). The RBC ghost
10    membranes isolated from Pb- exposed workers exhibited a significant increase in
11    phosphotidylcholine to phosphotidylethanolamine ratio (an established correlate of membrane
12    fluidity) along with an increase in RBC cholesterol levels, as also reported by Jehan and Motlag
13    (1995) discussed above. These authors predict that such alterations in phospholipid composition
14    of the membrane are responsible in biochemical instability of RBC in Pb-exposed workers.
15    Zimmermann et al. (1993) investigated the potential of such membrane lipid alterations to cause
16    resistance to oxidation.  These authors induced hyperlipidemia by treating Pb-exposed Wistar
17    rats with triton.  They observed an increase in erythrocyte choline phospholipid levels together
18    with a significant decrease in membrane lipid resistance to oxidation. These authors postulated
19    that such a decrease in resistance might cause RBC fragility, and ultimate destruction, leading to
20    anemic conditions. It has been also reported that exposure to Pb may also increase the levels of
21    fatty acids, e.g., arachidonic acid, in the RBC membrane in humans exposed to Pb (Osterode and
22    Ulberth, 2000).  Based on the negative correlation between serum calcium and increased
23    arachidonic acid content, these authors postulated that Pb ions might have substituted for calcium
24    in the activation of phospholipase enzymes, leading to increased synthesis of arachidonic acid.
25    Suwalsky et al. (2003) investigated the interaction of Pb with the RBC membrane, utilizing intact
26    as well as isolated unsealed RBC membrane models (representing phospholipids present in the
27    inner and outer layers of the membrane). Electron microscopy,  fluorescence spectroscopy, and
28    X-ray diffraction analyses of these models by the authors indicated that Pb particles adhere to
29    both external and internal surfaces of the membrane. Pb ions also have been found to disturb the
30    lamellar organization by causing considerable molecular disorder within lipid layers.
      December 2005                            5-6        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Recently, it has been shown that osmotic shock, oxidative stress, and/or energy depletion
 2    activate Ca2+-sensitive erythrocyte scramblase, leading to the exposure of phosphotidylserine at
 3    the cell surface.  This exposure of phosphotidylserine had been implicated in the phagocytosis of
 4    RBC by macrophages that can be measured by annexin binding, as determined by fluorescence
 5    activated cell sorting analysis. Kempe et al. (2005) carried out experiments to investigate
 6    whether anemic conditions reported in Pb intoxication  are the result of the decreased life span of
 7    RBCs  due to the above mentioned mechanisms. These authors reported that when human RBCs
 8    were exposed to Pb-nitrate (above 0.3 jiM), it caused a significant increase in Pb annexin
 9    binding, indicative of phosphotidylserine exposure. Using inhibitors for Ca2+- sensitive
10    potassium channels and whole cell patch clamp experiments, these authors concluded that Pb
11    exposure increased activation of potassium  channels, leading to shrinkage of cells and also
12    activation of scramblase, resulting in the exposure of phosphotidylserine on the cell membrane
13    surface. These authors further postulated that this exposure of phosphotidylserine on the
14    membrane might have led to them being engulfed by macrophages and the ultimately decreased
15    life span of RBCs in Pb intoxication.
16
17    Membrane Proteins
18          Earlier studies by Fukumoto et al. (1983) reported the differential profile for RBC-
19    membrane polypeptides determined by SDS-PAGE analysis. These investigators found
20    decreased levels of polypeptides in band 3 and increases in the levels of four other bands (i.e.,
21    bands  2, 4, 6, and 7) in the RBCs of human  workers exposed to Pb. From these observations,
22    they postulated that such Pb-induced alteration in RBC membrane proteins may lead to
23    membrane permeability changes. Apostoli  et al. (1988) also observed similar changes in RBC
24    membrane polypeptides in Pb-exposed workers and suggested that band 3 may represent an
25    anion channel protein; they also found that these changes occurred at blood Pb levels of
26    >50 |ig/100  ml.
27          Lead exposure has been known to increase the amount of membrane-bound protein
28    kinase C in rat brain, endothelial, and glial cells.  Belloni-Olivi et al. (1996) reported an
29    increased phosphorylation of RBC membrane proteins  on Pb exposure.  When human RBCs
30    were incubated with Pb-acetate (>100 nM) for 60 min,  it was found to increase phosphorylation
31    of membrane cytoskeletal proteins (120, 80, 52 and 45  kDa). This increase was accompanied by

      December 2005                             5-7        DRAFT-DO NOT QUOTE OR CITE

-------
 1    increase in protein kinase C activity. Membrane proteins were not phosphorylated when treated
 2    with protein kinase C inhibitors. Calcium and diacylglycerol were found not to be involved in
 3    this process.  The authors suggested that this activation of protein kinase was a direct interaction
 4    of the enzyme protein with Pb. Slobozhanina et al. (2005) reported that incubation of human
 5    RBCs with Pb-acetate (1 to 10 jiM for 3 h) caused differential binding of fluorescent probes to
 6    the membrane, suggesting alterations in the physicochemical state of the membrane proteins and
 7    lipids. Based on these observations, the authors postulated that such alterations in membrane
 8    molecular composition may influence the activity of membrane enzymes and function of
 9    receptors and channels present on the membrane.  These and other related studies are
10    summarized in Annex Table AX5-2.1.
11
12    5.2.3  Effect of Lead on Erythrocyte Heme Metabolism
13         Enzyme studies of the heme pathway have shown that Pb is an inhibitor of several
14    enzymes involved in heme synthesis, including 5-aminolevulinic acid dehydratase (ALAD),
15    coproporphyrinogen oxidase, and ferro chelatase (see Figure 5-2.1 for a schematic representation
16    of heme biosynthesis). ALAD is a cytoplasmic enzyme that catalyzes the second, rate-limiting
17    step of the heme biosynthesis pathway; that is, ALAD catalyzes formation of porphobilinogen
18    through the conjugation of two molecules of 5-aminolevulinic acid. ALAD is a Zn-dependent
19    enzyme, and thiol groups are essential for its activity (Bernard and Lauwerys, 1987). Decreased
20    erythrocyte ALAD is the most sensitive indicator of human Pb exposure, to the extent that
21    measurement of ALAD activity reflects well Pb levels in the blood. Similarly, erythrocyte
22    ALAD activity measurements have been used to assess Pb toxicity in  other species.
23
24    Erythrocyte ALAD
25    Terayama et al. (1986) reported decreased ALAD activity in rat RBCs at blood Pb levels of
26    100 jig/100 mL.  Scheuhammer (1987) studied the usefulness of the ALAD ratio
27    (activated/nonactivated enzyme activity) to study Pb effects in avian RBCs. The ALAD activity
28    ratio is a sensitive, dose responsive measure of Pb exposure regardless of the mode of
29    administration of Pb. For example, dietary Pb concentrations as low as 5 ppm (dry weight) can
30    be estimated through the use of the ALAD enzyme activity ratio method. A highly significant
     December 2005                            5-8        DRAFT-DO NOT QUOTE OR CITE

-------
                                                                        Zn protoporphynn
                                                          7    Ferrocnelatase
                                                       Protoporphyrin IX      I Pb
Succinyl CoA+ Glycine
                                                              Protoporphyrinogen Ix oxidase
                      o-aminolevulinic acid
                                                     Protoporphyrinogen IX
                                                            A
                                                              Coporphyrmogen
                                                               oxidase
                                                      Coporphyrinogen
                      6-aminolevulinic acid
                               ALAdehydratase
                                                                   Uroporphyrinogen
                                                                   cosynthase
               Uroporphyrinogen I
               synthetase
           Cytoplasm
                        Porphobilinogen
                              Uroporphyrinogen I
                              Uroporphyrinogen I
                  Coproporphyrinogen I
                  Coporphyrinogen I
      Figure 5-2.1.  Schematic presentation of heme synthesis pathway. Potential lead (Pb)
                    interacting sites are indicated by curved arrows (t  increased, 1 decreased).
 1    positive correlation was observed between dietary Pb concentration over the 5 to 100 ppm range
 2    and the ALAD activity ratio.  The author concluded that RBC ALAD ratio may be a useful
 3    method for estimating average dietary concentrations of Pb over an environmentally relevant
 4    range, in situations where diet is the major source of exposure to Pb or where accurate
 5    estimations of dietary Pb are not possible. Redig et al. (1991) reported heme synthetic pathway
 6    alterations upon chronic exposure (3 or 11 weeks) to Pb in red-tailed hawks. This treatment
 7    resulted in a severe decrease in RBC ALAD activity, which did not return to normal levels until
 8    5 weeks after termination of Pb treatment. Lead exposure also decreased ALAD activity in the
 9    bone marrow and in the liver but did not alter aminolevulinic acid synthase activity. Dorward
10    and Yagminas (1994), using comparative enzyme kinetic analysis of ALAD in Pb-exposed
11    female cynomolgus monkeys and human erythrocyte ALAD, found similar inhibition profiles
12    and concluded that ALAD could be a useful model for measuring the biological response in
      December 2005
                            5-9
DRAFT-DO NOT QUOTE OR CITE

-------
 1    monkeys. Santos et al. (1999) reported that rat RBC heme biosynthesis was affected by either Pb
 2    treatment alone or Pb in combination with ethanol, due to the inhibition of ALAD activity.
 3          Analysis of blood ALAD activity had been used as a powerful clinical biomarker in
 4    evaluating Pb toxicity in occupational exposure.  Fontanellas et al. (2002) further suggested that
 5    this enzyme assay be used in identifying even subclinical Pb poisoning in chronic renal failure
 6    (see Section 5.7 for details).
 7
 8    Other Heme Metabolism Enzymes
 9          Taketani et al. (1985) studied the heme synthesizing activity of ferric ion using purified
10    ferrochelatase from rat liver mitochondria and reported that Pb reduced NAD(P)H-dependent
11    heme synthesis by 50% at 10~5 M, but that it had no effect when ferrous ion was used as the
12    substrate. Based on these results, the authors concluded that heme synthesis from ferric ion was
13    more susceptible to Pb than the ferrous ion. These studies also revealed that the NAD(P)H
14    oxidizing system reduces ferric ion to ferrous ion, which in turn was used for heme synthesis by
15    ferrochelatase.
16          The effect  of various metals, including Pb, on RBC porphobilinogen synthase (PBG-S)
17    was studied using human RBC hemolysate. Farant and Wigfield (1987) reported that the effect
18    on the enzyme depends on the affinity of the metal for thiol groups at its active sites. Additional
19    studies carried out by the same group utilizing rabbit erythrocyte PBG-S indicated that Pb acts as
20    a potent effector of this enzyme both in vitro and in vivo (Farant and Wigfield, 1990). Human
21    RBC porphobilinogen synthetase activity was found to be inhibited by Pb, while Zn ions
22    activated this enzyme (Simons, 1995).  Another enzyme involved in the heme synthetic pathway,
23    porphobilinogen deaminase, was inhibited in human RBC by Pb-nitrate (100 mM) in in vitro
24    studies, but had no effect in vivo (Tomokuni and Ichiba,  1990).  Rossi et al. (1992) reported  no
25    inhibition of coproporphyrinogen oxidase activity in human lymphocytes on exposure to Pb.
26    Heme synthesis can also be affected in Pb intoxication by interference with Fe transport into
27    reticulocytes.  Using a rabbit reticulocyte model, Qian and Morgan (1990) reported that
28    inhibitory effects  of Pb on transferrin endocytosis and iron transport across the membrane may
29    also contribute to altered heme metabolism in RBCs.  These and other related studies are
30    summarized in Annex Tables  AX5-2.2 and 5-2.3.
31

      December 2005                            5-10        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.2.4   Effect of Lead on Other Hematological Parameters
 2          The RBC pyrimidine 5-nucleotidase (P5N) catalysis of the hydrolytic dephosphorylation
 3    of pyrimidine 5-monophosphates is sensitive to inhibition by Pb.  Tomokuni et al. (1989)
 4    evaluated the activity of RBC and bone marrow 5-nucleotidase (P5N) and RBC ALAD in mice
 5    exposed to drinking water Pb (200 to 500 ppm) for 14 or 30 days. These authors reported that Pb
 6    exposure decreased both P5N and ALAD activities in erythrocytes.  Additional studies from this
 7    group, using a similar exposure regimen, indicated no change in levels of urinary coporphyrins.
 8          Lead exposure (4 mg/kg and 6 mg/Kg body wt/30 days) in splenectomized rats was found
 9    to cause depletion of RBC Hb content, to increase numbers of reticulocytes in peripheral blood,
10    and to increase urinary delta aminolevulinic acid excretion (Gautam and Chowdhury, 1987).
11    These  authors further reported that the increased number of reticulocytes found in the blood may
12    be due to induced acceleration of the erythropoeitic cell series.  Redig et al. (1991) reported
13    biphasic effects of Pb on hematological parameters from their chronic exposure studies in red-
14    tailed hawks over 3 or 11 weeks.  These authors observed a rapid and relatively brief increase in
15    RBC free protoporphyrin and a slower, but more prolonged, increase in its Zn complex with
16    3-week exposure to Pb (0.82 mg/kg body wt).  On the other hand, exposure to a higher dose of
17    Pb (1.64 mg/kg body wt) for a longer duration (11 weeks) resulted in a decrease in the Hct and
18    Hb. Panemangalore and Bebe (1996) reported that Zn deficiency increased the Pb-induced
19    accumulation of porphyrin in RBCs to a lesser extent compared to its accumulation in the liver in
20    weaning rats.
21          The effects of Pb on RBC number and other Hct parameters appear to be dose dependent.
22    lavicoli et al. (2003) investigated these effects by feeding mice with eight different doses of Pb
23    below (0.6 to <2.0 |ig/dL) and above (>2.0 to!3 |ig/dL) normal background levels. These
24    authors reported that mice receiving below normal background levels of dietary Pb displayed
25    enhanced RBC counts and increased Hb and Hct values, whereas a marked decrease in RBC
26    number occurred when blood Pb levels approached 10 |ig/dL. Sivaprasad et al. (2003) also
27    reported significant reductions in RBC Hb content and Hct on Pb exposure (0.02% Pb-acetate in
28    drinking water for 5 weeks). Toplan et al. (2004) observed significant decreases in RBC Hb
29    content and Hct and increases in blood viscosity in Wistar rats after 5-week exposure to Pb.
30    Studies cited above are summarized in Annex Table AX5-2.4.
31

      December 2005                           5-11        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.2.5   Effects of Lead on Erythrocyte Enzymes
 2          The toxic effects of Pb on RBCs result from its complexation with the sulfhydryl,
 3    carboxyl, and imidazole groups of proteins, particularly enzymes, by competitive binding of Pb2+
 4    with Zn2+ or Mg2+ in metalloenzymes. This binding of Pb to enzyme proteins can inhibit
 5    enzymes involved in the glycolytic and pentose phosphate pathway, both of which are sources of
 6    energy compounds and intermediates of purine conversion, thus causing a disruption of energy
 7    metabolism. Along with these changes, Pb-induced changes in the membrane integrity, as
 8    discussed earlier (Section 5.2.1), may also affect the enzymes' associated ion channels and other
 9    transport mechanisms.
10
11    Energy Metabolism
12          Erythrocytes generate high-energy ATP by anerobic glycolysis and cycle oxidized and
13    reduced nicotinamide adenine nucleotide phosphate (NADP) by the aerobic pentose phosphate
14    pathway. Anemic conditions associated with Pb poisoning, along with the inhibitory effects of
15    Pb on heme synthesis, may result in increased RBC destruction due to the inhibitory effects of
16    Pb on the activities of the enzyme, pyramidine 5-nucleotidase (P5N). Deficiency of this enzyme
17    is characterized by intracellular accumulation of pyramidine-containing nucleotides, leading to
18    hemolysis.  Inhibition of this enzyme along with the perturbations in heme metabolism create
19    imbalances in the energy currency of the erythrocyte.  Perturbations in energy metabolism can be
20    followed by changes in the concentration of purine nucleotides.  In erythrocytes, these
21    compounds cannot be synthesized de novo, they can only be reconstructed from preexisting free
22    purine bases on nucleosides through salvage type reactions. The cell energy content can be
23    measured by adenylate (ATP + ADP + AMP) and guanylate (GTP + GDP + GMP) nucleotides,
24    and by their sum total. The concentrations of nucleoside monophosphates increase in cases of
25    cell energy deficit, but they quickly degrade to nucleosides and bases.
26          Cook et al. (1987) compared P5N and deoxypyramidine-5-nucleotidase levels in the RBC
27    of Pb-exposed workers and matched controls and reported significantly lower levels of P5N in
28    Pb-exposed workers. Konantakieti et al. (1986) reported similar observations in neonatal rat
29    RBCs. These authors further indicated that the low levels  of nucleotides were due to inhibition
30    of P5N activity by Pb, as the depression in enzyme activity was correlated with blood Pb levels.
31    This was further validated by in vitro inhibition of P5N in  a dose-dependent manner. Tomokuni

      December 2005                            5-12        DRAFT-DO NOT QUOTE OR CITE

-------
 1    and Ichiba (1987) found similar results with human RBCs both in vitro and in vivo. They
 2    reported activation of Pb-exposed human RBCs.  Antonowicz et al. (1990) observed significantly
 3    higher levels of glycolytic enzymes and increased production of lactic acid and 2,3-diphospho
 4    glycerol, when human RBCs were incubated with Pb. Based on their observations, these authors
 5    suggested that Pb exposure may result in  anaerobic glycolysis activation in human RBCs. In
 6    contrast, Grabowska and Guminska (1996) reported that Pb exposure diminished the ATP levels
 7    in human RBCs by inhibiting aerobic glycolysis.
 8          Erythrocyte energy metabolism in workers exposed to heavy metals, but without clinical
 9    manifestations of toxicity, was found to intensify and become more pronounced when they were
10    occupationally  exposed to Pb. Nikolova and Kavaldzhieva (1991) measured the exposed
11    workers and reported higher ratios of ATP/ADP in Pb-exposed workers. Because the RBC
12    energy pool is perturbed due to Pb exposure, Morita et al.  (1997) evaluated the effect of Pb on
13    NAD synthetase and reported an apparent dose-dependent decrease in NAD synthetase activity
14    in the erythrocytes of Pb exposed workers.
15          Baranowska-Bosiacka and Hlynczak (2003)  evaluated Pb effects on distribution profiles
16    of adenine, guanine nucleotide pools and their degradation products in human umbilical cord
17    RBCs.  In vitro exposure equivalent (Pb-acetate;  100 to 200 |ig/dL) to Pb exposure for 20 h were
18    found to significantly lower the levels of nucleotide pools, including NAD and NADP,
19    accompanied by a significant increase in purine degradation products (adenosine, guanosine,
20    inosine, and hypoxanthine). Associated morphological RBC  alterations were also observed, with
21    marked significant increases in stomatocytes, spherocytes, and echinocytes.  These investigators
22    also observed similar alterations in the nucleotide pools in Wistar rat RBCs with short-term
23    exposure to Pb  (Baranowska-Bosiacka and Hlynczak, 2004).  Based on these observations, the
24    authors postulated that decreases in NAD and NADP concentrations in RBCs may be a good
25    indicator of Pb-induced disturbance in the energy process  and can serve as a useful marker for
26    chronic Pb exposure. If NAD synthetase activity had been measured in these studies, it might
27    have provided experimental support for the observation of inhibition of NAD synthetase reported
28    by Morita etal. (1997).
29
      December 2005                            5-13        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Other Enzymes
 2         Lead-induced efflux of K+ from human RBCs had been recognized as being due to the
 3    ability of Pb to selectively increase the membrane permeability for this cation.  Studying the
 4    efflux of 86Rb using inside-out RBC vesicles, Alvarez et al. (1986) demonstrated that Pb
 5    promoted the selective efflux of K+ ions by altering the sensitivity of Ca2+ binding site on the
 6    membrane either by direct binding or by altering Mg2+-mediated modulation. Fehlau et al.
 7    (1989) indicated that this modulation of the Ca 2+-activated K+ channel in human RBCs
 8    coincides with the activation of RBC membrane-bound oxidoreductase. These authors suggested
 9    that, even though these two are independent events, the oxidoreductase enzyme activity may
10    influence K channel gating.
11         Earlier studies by Mas-Oliva (1989) on the potential effects of Pb on the RBC membrane
12    (using RBC ghosts) indicated that Pb has inhibitory effects on Ca2+-Mg2+-ATPase. Further
13    investigations on the role of calmodulin in the inhibition of Ca2+-Mg2+-ATPase indicated that the
14    inhibitory activity on the enzyme may be due either to the effect of Pb on sulfhydryl groups on
15    the enzyme or by direct binding to calmodulin.
16         Jehan and Motlag (1995) reported that when albino rats were administered Pb i.p (5 or
17    20 mg/kg body wt) for 14 consecutive days either alone or in combination with Cu (2 mg/kg
18    body wt) or zinc (5 mg/kg body wt), there were severe decreases in RBC membrane enzyme,
19    acetylcholine esterase (AchE), NADH dehydrogenase, and Na+-K+ ATPase levels along with
20    decreases in phospholipid content, hexose, and hexosamine.  Of the combined metal treatment
21    exposure regimens, Zn was found to considerably reduce such changes. Grabowska and
22    Guminska (1996) assayed three ATPase activities (i.e., Na+-K+ ATPase, Mg2+-ATPase, and
23    Ca2+-ATPase) in human RBC in vitro and reported RBC Na+-K+ ATPase to be  the only enzyme
24    inhibited by Pb, while Ca2+ or Mg 2+ATPases were not sensitive to Pb.  On the  other hand,
25    Sivaprasad et al. (2003) observed Pb-induced reductions in RBC activities of the three of those
26    ATPase activities.
27         Two reports by Calderon-Salinas et al. (1999a,b) indicated Pb effects on calcium transport
28    in human RBC. Initial studies by this group indicated that Pb and Ca are capable of inhibiting
29    the passive transport of other metals in a noncompetitive way. Inhibition studies using N-ethyl-
30    maleimide indicated that Pb and Ca share the same permeability pathway in human RBCs and
31    that this transport system is electrogenic (Calderon-Salinas et al., 1999a). Additional studies by

      December 2005                            5-14        DRAFT-DO NOT QUOTE OR CITE

-------
 1   the same group reported that Pb is capable of inhibiting Ca efflux by inhibiting Ca-ATPase
 2   (Calderon-Salinas et al., 1999b).  These authors further suggested that under physiological
 3   conditions, Pb, via Ca 2+-ATPase, alters Ca influx, while chronic Pb intoxication inhibits Ca
 4   efflux by altering RBC calcium homeostasis.  Silkin et al. (2001) reported Pb-induced activation
 5   of K channels in the RBCs of the teleost fish S. porcus. Exposure of teleost fish RBCs to 1 to
 6   2  |iM Pb led to a minor loss in cellular K+; but, at 20 to 50 |iM Pb, about 70% of cellular K+ was
 7   lost. Based on their observations of Pb-induced K+ efflux from RBCs under competitive and
 8   inhibitory regimens, these authors suggested that Pb activates RBC K+ channels.
 9         Eder et al. (1990) and Loipfuhrer et al. (1993) investigated activity levels of Ca 2+-ATPase
10   and calcium accumulation, respectively, in Pb-depleted rat RBCs.  No alteration in Ca 2+-ATPase
11   activity or Ca accumulation was observed in the PO generation (Eder et al., 1990).  On the other
12   hand, significant reduction in Ca-ATPase activity was observed in the Fl generation. It was
13   suggested that Pb-induced alterations in the metabolism of phospho- and glycoproteins result
14   from Pb depletion and may be responsible for the reduced enzyme activity.  Both of the groups
15   postulated that the decreased MCV observed in Pb depleted rat RBCs could be due to reduced Ca
16   2+-ATPase activity in the RBCs. These and other related studies are summarized in Annex
17   Tables AX5-2.5 and 5-2.6.
18
19   5.2.6  Erythrocyte Lipid Peroxidation and Antioxidant Defense
20         Although several mechanisms have been proposed to explain Pb toxicity, no mechanisms
21   have been defined explicitly. Recent literature on Pb toxicity suggests oxidative stress as one of
22   the important mechanisms for toxic effects of Pb in various organs. Because RBCs accumulate
23   major amounts of Pb compared to other tissues, oxidative stress may also result in the
24   accentuation  of lipid peroxidation with concomitant inhibition of antioxidant enzymes, such as
25   superoxide dismutase (SOD), catalase, GSH peroxidase, GSH reductase, and simultaneous
26   increases in oxidized GSH (GSSG) and reduced GSH/GSSG ratios. Pb-induced lipid
27   peroxidation  and the mitigating effects of experimental chelation therapy are discussed with
28   relevance to each tissue or organ within this chapter. The discussion focuses on the available
29   literature with reference to studies on erythrocytes.
30         Patra and Swarup (2000) reported significant changes in RBC lipid peroxide levels and
31   anti oxidant defense (SOD and catalase) levels in RBC hemolysates from male calves exposed to

     December 2005                            5-15        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Pb (7.5 mg/kg body wt for 28 days).  These authors suggested the potential role for increased
 2    peroxide levels in Pb-induced alterations in RBCs. Mousa et al. (2002) investigated the levels of
 3    various antioxidant enzymes, thiols, lipid peroxide in erythrocytes, and total thiol status of
 4    plasma in goats exposed to Pb (Pb-acetate, 5.46 mg/kg body wt for 2 weeks).  These authors
 5    reported that all the parameters referred above were significantly increased in RBCs by day 7
 6    and receded to normal levels by day 14, while peroxides remained significantly increased even
 7    by day 14. Based on these observations, it was suggested that Pb-induced lipid peroxide
 8    generation in RBCs appears to be a continuous process and can lead to persistent oxidative stress
 9    in RBCs with chronic exposure.
10          Metal chelator agents have been used clinically to reduce internal Pb body burden. These
11    agents form an insoluble complex with Pb and are excreted.  Though the majority of studies on
12    the clinical potential of various experimental agents, including certain  antioxidants, have been
13    extensively performed mainly in relation to toxicity associated with hepatic and kidney tissues,
14    such studies have also considered their potential effects on heme metabolism and blood Pb levels
15    (see Sections 5.7  and 5.10).  In the following paragraphs, two recent representative studies in
16    experimental animals that specifically assessed the protection conferred to erythrocytes are
17    described.
18          El-Missiry (2000) investigated the protective role of the pineal hormone, melatonin, on
19    Pb-induced suppression of heme synthesis as a consequence of reduced antioxidant status.
20    Intramuscular injection of Pb-acetate (10 mg/kg body wt for 7 days) caused a significant
21    reduction in heme synthesis with decreased blood Hb levels and decreased RBC and liver
22    ALAD. Pretreatment of rats with melatonin (30 mg/kg body wt) intragastrically prevented the
23    suppressive effects of Pb on RBC heme metabolism by conferring protection to the antioxidant
24    capacity of the cells and also by scavenging free radicals generated by Pb intoxication.
25          Sivaprasad et al. (2003) studied the protective effects of dl-alpha-lipoic acid (LA,
26    25 mg/kg body wt) and meso-2,3-dimercaptosuccinic acid (DMSA, 25 mg/kg body wt); and they
27    found such treatments, either alone or in combination for a week, had an effect on alterations in
28    RBC functions induced by Pb-acetate (0.02% in drinking water for 5 weeks).  These authors
29    reported that treatment with LA or DMSA, alone or in combination, reversed Pb-induced
30    increased LPO and reductions in Hb and Hct, along with changes in other biochemical
31    parameters affected by Pb treatment. These authors further concluded that combined treatment

      December 2005                            5-16        DRAFT-DO NOT QUOTE OR CITE

-------
 1   was much more potent and effective. These and other related studies are summarized in Annex
 2   Table AX5-2.7.

 3

 4   5.2.7   Summary

 5      •  The 1986 Pb AQCD reported that the activity of ALAD appeared to be inversely
 6         correlated to blood Pb values and was found inhibited in several tissues. Human studies
 7         reviewed in 1986 Pb AQCD also indicated that occupational exposure to Pb results in
 8         decreased RBC survival along with alterations in RBC membrane integrity and energetics.

 9      •  More recent studies reviewed in this AQCD indicate that the transport of Pb across the
10         RBC membrane is energy-independent, carrier-mediated and that the uptake of Pb is
11         mediated by an anion exchanger through a vanadate-sensitive pathway.

12      •  Lead intoxication interferes with RBC survival and alters RBC mobility. Hematological
13         parameters, such as mean corpuscular volume (MCV), mean corpuscular hemoglobin
14         (MCH), and mean corpuscular hemoglobin concentration (MCHC), are also significantly
15         decreased upon exposure to Pb.  These changes are accompanied by decreased membrane
16         sialic acid content.

17      •  Morphological analyses using electron paramagnetic resonance imaging and spin labeling
18         techniques indicate that changes occur in RBC morphology upon Pb exposure.

19      •  Lead-induced RBC membrane lamellar organization and decreases in membrane lipid
20         resistance to oxidation in rats appear to be mediated by perturbations in RBC membrane
21         lipid profiles. Similarly, Pb-induced altered phosphorylation profiles  of RBC membrane
22         proteins have been reported.

23      •  Erythrocyte ALAD activity ratio (ratio of activated/non activated enzyme activity) has
24         been shown to be a sensitive, dose-responsive measure of Pb exposure, regardless of the
25         mode of administration of Pb. Competitive enzyme kinetic analyses in RBCs from both
26         human and Cynomolgus monkeys indicated similar inhibition profiles by Pb.

27      •  Consistent observation of Pb-mediated inhibition of pyramidine 5'-nucleotidase (P5N)
28         suggests this enzyme as a potential biomarker for Pb exposure.

29      •  Significant reductions in levels of nucleotide pools (e.g., NAD and NADP) accompanied
30         by significant increase in purine degradation products have been implicated in the Pb-
31         induced altered energetics of RBCs.

32      •  Lead-induced increased permeability for K+ in RBCs appears to be due to the selective
33         efflux of K+ ions on the RBC membrane due to altered sensitivity of the Ca2+-binding site
34         on the membrane.  Erythrocyte Na+-K+ ATPase appears to be more  sensitive to Pb-induced
3 5         inhibition than Ca2+-Mg2+ ATPase.
     December 2005                           5-17        DRAFT-DO NOT QUOTE OR CITE

-------
 1      •  Chelation agents and the pineal hormone, melatonin, have been reported to confer
 2         protection against Pb-induced lipid peroxidation and increased antioxidant defense
 3         inRBCs.
 4
 5          The newly available (since 1986) scientific evidence presented in this section
 6    convincingly demonstrates deleterious effects of lead on erythrocyte cell morphology, function,
 7    lead uptake and alterations in certain enzymes involved in heme synthetic pathways.  However,
 8    some of the interesting and important conclusions are derived mainly from in vitro studies, often
 9    using short time incubations. It would be useful to substantiate such findings further by more
10    systematic studies employing meaningful experimental designs for in vivo evaluation of
11    laboratory animal models.
12
13
14    5.3   NEUROLOGICAL/NEUROBEHAVIORAL EFFECTS  OF  LEAD
15    5.3.1   Neurotoxicological/Neurobehavioral Effects of Lead in Animals
16    5.3.1.1 Introduction
17          Since the initial description of Pb encephalopathy in the developing rat in the mid-1960s,
18    (Pentschew and Garro, 1966), a continuing research focus has been on defining the extent of
19    CNS involvement at subencephalopathic, environmentally relevant, levels of exposure.  These
20    efforts have primarily addressed the developing organism, consistent with the primary public
21    health concerns for neurotoxicity from Pb during  this period.  While significant research
22    advances have been made in animal studies over the last four decades, relating these findings to
23    neurotoxicity in children has been challenging and difficult. The barriers to greater progress
24    have primarily been due to Pb's multiple toxic mechanisms of action in brain tissue, which
25    encompasses variable, overlapping, and, at times, opposing dose-effect relationships. One goal
26    of this section is to bring greater clarity to the current state of knowledge.
27          The Pb neurotoxicity evidence available for assessment in the 1986 Lead AQCD was
28    considerably  different in character from current, newly available findings. The literature was
29    dominated by various types of assessments of CNS biogenic amine function in exposed animals,
30    with dopaminergic neuronal systems seeming the most sensitive to the metal and drawing the
31    most attention.  In addition, the prevailing wisdom was that the neuronal actions of Pb were best

      December 2005                           5-18        DRAFT-DO NOT QUOTE OR CITE

-------
 1    elucidated by perturbing neurotransmitter systems with CNS agents of known mechanism of
 2    action and comparing the responses in exposed animals to those in control subjects, an approach
 3    of limited value.  Only some of those studies reported blood and/or brain Pb concentrations along
 4    with the experimental findings, rendering interpretation of results across different laboratories
 5    difficult and somewhat unreliable.
 6          As of 1986, perhaps the most reliable evidence concerned the effects of acute exposure to
 7    Pb2+ in vitro on voltage-sensitive Ca2+ channel function in the nerve cell membrane, developed to
 8    a great extent by Cooper and co-workers (Kober and Cooper,  1976; Cooper and Manalis, 1984;
 9    Suszkiw et al., 1984). Using neuromuscular endplate or synaptosomal preparations, these studies
10    demonstrated that Pb2+ interfered with Ca2+ influx through voltage-sensitive channels. These
11    findings significantly advanced the field, though acute exposure in vitro bore little resemblance
12    to environmentally relevant routes and magnitudes of exposure.
13          In the ensuing two decades, the Pb neurotoxicity literature has reflected an increased
14    focus on cognitive-related mechanisms and the refinement of approaches and methodologies.
15    Exposure-induced alterations at glutamatergic synapses have become a primary substrate of
16    attention. Synaptic plasticity models (e.g., long-term potentiation [LTP]) developed in the 1990s
17    are used in Pb studies in laboratories around the world. Behavioral paradigms, refined to more
18    consistently discriminate Pb effects, aided in identifying optimal testing conditions and
19    developmental periods for exposure.  The cumulative result of these advances has lead to clearer
20    understanding of likely mechanisms underlying Pb-induced cognitive impairments found in
21    Pb-exposed children.
22          The Pb neurotoxicity evidence reviewed in this section is organized largely according to
23    scientific discipline: neurochemical alterations involving glutamatergic, cholinergic, and
24    dopaminergic function;  mechanisms defined by neurophysiological approaches; changes in
25    auditory and visual  function; modifications in behavioral function; induced alterations in cellular
26    morphology; and findings on cellular disposition  of Pb.  This type of organization permits a more
27    focused analysis of a very extensive, broad literature.
28
      December 2005                            5-19       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.3.1.2  Neurochemical Alterations Resulting from Lead Exposure
 2          The following areas of investigation have drawn the most attention in the Pb neurotoxicity
 3    field over the last 20 years. A summary of the key studies evaluating neurochemical alterations
 4    resulting from Pb exposure are listed in Table AX5-3.1.
 5
 6    Lead and Neurotransmitter Release Processes
 1          By the mid-1980s, it was evident that acute exposure to Pb2+ in vitro reduced the
 8    magnitude of depolarization-induced transmitter release, apparently by inhibiting Ca2+ influx
 9    into the nerve ending through voltage-sensitive Ca2+ channels (Kober and Cooper,  1976; Cooper
10    and Manalis, 1984; Suszkiw et al., 1984). Since then, several investigators utilizing various
11    preparations (Shao and Suszkiw, 1991 [cortical synaptosomes];  Tomsig and Suszkiw, 1993
12    [bovine chromaffin cells]; Braga et al.,  1999a,b [cultured hippocampal cells]; Westerink and
13    Vijverberg, 2002 [PC12 cells]), have demonstrated that, in the absence of Ca2+, Pb2+  exhibits
14    Ca2+-mimetic properties in stimulating exocytosis and is substantially more potent  in doing so.
15    That is, in the absence of Ca2+ and depolarization, nM concentrations of Pb2+ alone stimulate
16    transmitter release. Many investigators have proposed that this action, in conjunction with the
17    ability of Pb2+ to suppress evoked release, produces a higher noise level in synaptic transmission
18    in Pb-exposed animals.
19          The ability  of Pb to diminish stimulated transmitter release has been demonstrated in
20    intact chronically exposed animals via the use of intracerebral microdialysis (Kala  and Jadhav,
21    1995; Lasley and Gilbert,  1996; Lasley et al., 1999). More recently, Lasley and Gilbert (2002)
22    used Ca2+-free perfusate containing a Ca2+ channel antagonist for microdialysis to identify the
23    Ca2+-independent component of release. These workers demonstrated that under these
24    conditions high K+-stimulated glutamate and GAB A release were elevated in chronic high level
25    Pb-exposed animals, suggesting a Pb2+-induced enhancement of evoked release. It was
26    concluded that this pattern of results indicated the presence of two actions of Pb on transmitter
27    release in vivo:  (1) a more potent suppression of stimulated release seen at lower exposure levels
28    (associated with blood Pb values of 27-62 |ig/dL) combined with (2) Ca2+-mimetic actions that
29    independently induce the exocytosis seen at higher exposure levels (associated with blood Pb
30    values >62 jig/dL). Together, these two actions produce a biphasic dose-effect relationship (see
      December 2005                            5-20       DRAFT-DO NOT QUOTE OR CITE

-------
1    Figure 5-3.1).  Thus, there is good correspondence between findings in in vitro and in vivo
2    studies with respect to the actions of Pb on transmitter release.
                          5.0
                          4.5
                          4.0
                          3.5
                          3.0
                          2.5
                          2.0
                          1.5
                          1.0
                          0.5
                          0.0
                           150 mM K+
         Total Release
        -O— Control
         •- - • 0.2% Pb
                              -120  -90    -60   -30    0   20  40  60  80
                                                  Time, min
                                              110   140
                      
                      ro
                     ••0
                      
-------
 1   Lead and Glutamatergic NMDA Receptors
 2          Because of the established importance of the NMDA subtype of glutamate receptor in
 3   synaptic plasticity and learning, these receptors have been a focus of intense interest in Pb
 4   neurotoxicity for the last 15 years.  Using whole cell and single channel patch clamp
 5   methodologies, Alkondon et al. (1990) were the first to report that Pb2+ inhibited the function of
 6   the NMDA receptor channel complex. Guilarte and Miceli (1992) reported similar findings
 7   using nominal Pb2+ concentrations  and receptor binding techniques, and drew parallels between
 8   Zn2+ Ca2+2- and Pb2+-induced inhibition of the channel.  However, Lasley and Gilbert (1999),
 9   using free Pb2+ ion concentrations and radioligand binding, demonstrated that, despite the
10   similarities, Pb2+ did not inhibit the NMDA receptor channel complex by binding to the Zn2+
11   allosteric site. Furthermore, they indicated that the Pb2+ ICso of 0.55 jiM for inhibition of the
12   channel complex was likely approximately two orders of magnitude greater than the extracellular
13   fluid concentrations of Pb2+ associated with environmentally relevant exposure.  This does not
14   mean that NMDA receptor function does not change after Pb exposure, but it strongly suggests
15   that the alterations are not based on a direct Pb2+ action.
16          Unfortunately, a consensus on the effects of chronic Pb exposure on NMDA receptor
17   expression and function has not been achieved. Extensive effort has been invested to assess
18   NMDA receptor subunit mRNA and protein expression in exposed animals (Guilarte and
19   McGlothan, 1998; Nihei and Guilarte, 1999; Guilarte et al., 2000; Nihei et al., 2000; Toscano
20   et al., 2002; Guilarte and McGlothan, 2003), but consistent findings have not emerged. An
21   exception was perhaps the work of Nihei et al. (2000) who found decreases in hippocampal NR1
22   subunit mRNA and protein expression deficits in LTP to be associated with impaired spatial
23   learning in PB-exposed  animals. Correlations of this type with functional measures are valuable
24   in validating the biochemical observations.
25          While exposure-induced alterations of NMDA receptor binding have been observed in
26   multiple laboratories, there has not been uniform agreement as to the direction of change.
27   Upregulation of NMDA receptor density has been observed in rats continuously exposed
28   throughout development (Ma et al., 1997; Lasley et al., 2001), but receptor downregulation has
29   also been reported when exposure was begun immediately postweaning (Cory-Slechta et al.,
30   1997a).  The results of behavioral investigations are best explained by increases in NMDA
31   receptor density.  Cohn and Cory-Slechta (1993, 1994b), using a repeated learning component of

     December 2005                            5-22         DRAFT-DO NOT QUOTE OR CITE

-------
 1    a multiple reinforcement schedule, observed enhanced performance sensitivity to exogenous
 2    NMDA administration and diminished sensitivity to MK-801, an NMDA receptor antagonist in
 3    exposed animals. The same findings resulted when the drug discrimination paradigm was
 4    utilized (Cory-Slechta, 1995a; Cory-Slechta et al., 1996b): enhanced sensitivity to NMDA and
 5    reduced sensitivity to MK-801 in Pb-exposed groups.  A decreased sensitivity to MK-801 can
 6    result from either increased numbers of NMDA receptors or a diminished access of the
 7    antagonist to its binding site in the ion channel.  Thus, all these behavioral observations may be
 8    accounted for by Pb-induced increases in NMDA receptor density resulting in increased
 9    sensitivity to agonists coupled with decreased sensitivity to antagonists. That is, the functional
10    measures suggest that an NMDA receptor upregulation occurs.
11
12    Pb2+ and Protein Kinase C
13          Another important focus area for Pb  neurotoxicity research has been the interactions of
14    Pb2+ with protein kinase C  (PKC) activity.  Markovac and Goldstein (1988a) were the first to
15    report that Pb2+ directly stimulated PKC activity at picomolar concentrations, thereby exhibiting
16    greater potency for this action than Ca2+ by  4-5 orders of magnitude. Long et al. (1994) made
17    similar observations using free Pb2+ and Ca2+ ion concentrations and nuclear magnetic resonance
18    spectroscopy, resulting in an ECso of 55 pM for Pb2+ stimulation of PKC.  These workers also
19    presented evidence suggesting that the maximal efficacy of Pb2+ was less than that of Ca2+,
20    despite its greater potency.  Tomsig and Suszkiw (1995) elegantly elucidated multiple
21    interactions of Pb2+ with PKC, identifying both stimulatory (affinity in the pM range) and
22    inhibitory (affinities in the  nM and |iM range) binding sites on the kinase.  They also showed
23    that, on the basis of these interactions, Pb2+  induced a peak efficacy for stimulation of PKC that
24    was only -40% of the maximal efficacy produced by  Ca2+, leading to their terming Pb2+ a partial
25    agonist of the kinase, as reflected in Figure  5-3.2.
26          Subsequent studies have begun to examine the cellular impact of the Pb2+ effects on PKC.
27    Kim et al. (2000) showed that acute Pb2+ exposure in  vitro stimulated immediate early gene
28    expression in cultured cells by a mechanism that requires PKC. Braga et al. (2004) have
29    demonstrated that Pb2+ stimulation of PKC  results in inhibition of nicotinic cholinergic
30    modulation of glutamate  and GABA synaptic transmission in cultured hippocampal cells. It is
31    anticipated that future studies will further develop this line of investigation.

      December 2005                            5-23       DRAFT-DO NOT QUOTE OR CITE

-------
                               14  13  12  11  10   9   8   7   6   5   4   3
                                             -log [free metal]

     Figure 5-3.2.  PKC activity as a function of Ca2+ and Pb2+ concentrations.
     Source: Tomsig and Suszkiw (1995).
 1          The effects of chronic Pb exposure on PKC signaling have been more difficult to
 2   discriminate. Most investigators have utilized broken cell preparations and measures of either
 3   kinase translocation or enzyme activity; however, a broken cell preparation has not been shown
 4   to simulate the intracellular milieu of a chronically exposed intact animal.  In the preparation of a
 5   tissue extract for determination of kinase activity, the unbound Pb2+ is removed or greatly
 6   diluted, so that the resulting activity measure largely reflects changes in total PKC expression
 7   resulting from the exposure. That is, this measure does not identify a synaptic pool of PKC or
 8   necessarily represent the pool of kinase involved in signal transduction.  Alternatively, the
 9   translocation of kinase from a cytosolic to membrane cellular fraction is a somewhat nonspecific
10   measure and observed changes should be independently confirmed. From the effects of acute
11   Pb2+ exposure in vitro it is abundantly clear that PKC is a lexicologically significant intracellular
12   target for Pb2+. However, various investigators have been unable to define how this acute effect
13   translates, if at all, to chronic exposure in the intact animal. Neither is it evident how one could
14   discriminate inhibition of PKC activity (e.g., resulting from decreased efficacy relative to that
15   associated with Ca2+) from downregulation  of the enzyme from prolonged  stimulation.
16   Resolution of these issues awaits the development of more specific methodologies.
     December 2005
5-24
DRAFT-DO NOT QUOTE OR CITE

-------
 1    Pb2+-Ca2+ Interactions
 2          In general Pb2+-Ca2+ interactions have long been proposed as important factors in
 3    manifestations of cellular Pb toxicity and have been under investigation since before the 1986
 4    AQCD was prepared.  The classical effects of Pb2+ mentioned earlier include inhibiting Ca2+
 5    influx through cell membrane voltage-sensitive Ca2+ channels and exhibiting Ca2+-mimetic
 6    properties at multiple intracellular proteins. In addition, Pb2+ is known to disturb intracellular
 7    Ca2+homeostasis (Simons, 1993b).  Ca2+-dependent proteins whose actions have been reported
 8    to be stimulated by Pb2+ include calmodulin and calmodulin-dependent phosphodiesterase
 9    (Goldstein,  1993), calcineurin (Kern and Audesirk, 2000), and Ca2+-ATPase (Ferguson et al.,
10    2000).  These actions of Pb2+ are thought to be the points of initiation of much of the metal's
11    cellular toxicity.
12
13    Lead Exposure and Cholinergic Neuronal Systems
14          The actions of chronic exposure have also been studied with respect to changes in CNS
15    cholinergic  systems, as another substrate thought to underlie cognitive function. Bielarczyk et al.
16    (1996) reported (a) decreased functional cholinergic innervation in the hippocampus and (b)
17    depression of choline acetyltransferase activity in hippocampus and  cortex in young adult rats
18    exposed to Pb only during early development. Similar changes were reported by Bourjeily and
19    Suszkiw (1997), leading to the conclusion that perinatal exposure results in a loss of
20    septohippocampal cholinergic projection neurons that persists until testing in young adulthood.
21    Tian et al. (2000) exposed PC12 cells to Pb2+ for <48 h and found that the downregulation of
22    choline acetyltransferase activity reflected the effects of the metal at the level of gene expression.
23    Consistent with these other findings, Jett et al. (2002) employed a similar perinatal exposure
24    protocol and observed increased nicotinic receptor binding in multiple brain regions.  These
25    reports reinforce the belief that  Pb exposure during early development deleteriously affects
26    cholinergic  function and indicate that these actions are an important component of the cognitive
27    impairment resulting from exposure to the metal.
28
29    Summary
30          In reviewing the Pb neurotoxicity literature of the last 20 years and the research focus
31    areas presented above, it is evident that the effects of Pb exposure on components of

      December 2005                             5-25        DRAFT-DO NOT QUOTE OR CITE

-------
 1    neurotransmitter release and Pb2+-Ca2+ interactions are closely intertwined. Exposure-induced
 2    decreases in glutamatergic, cholinergic, and dopaminergic transmission are most prominent
 3    because of the purported role of these neuronal systems in brain development and cognitive
 4    function.  In contrast, the weight of the data suggest an upregulation of NMD A receptors
 5    resulting from chronic exposure, but a consensus on the effects of Pb on expression and function
 6    remains to be attained, and it is increasingly apparent that this glutamate receptor subtype may
 7    not be a primary target of chronic exposure in the intact animal. While the in vitro interactions
 8    of Pb2+ and PKC have been carefully described and are broadly relevant to cellular signaling
 9    pathways, meaningful and valid observations of the functional effects of these interactions in
10    intact animals have not been achieved.
11
12    5.3.1.3 Actions of Lead Exposure Defined by Neurophysiological Approaches
13          One of the most significant advances in Pb neurotoxicity research over the last two
14    decades is the widespread application of synaptic plasticity models to studies of the effects of
15    exposure. Key studies are listed in Table AX5-3.2. The incorporation of these paradigms into
16    Pb studies could be seen as a natural progression, and one might expect that they would receive
17    greater use in neurotoxicology, as they have in the broader field of neuroscience.
18
19    Chronic Lead and Models of Synaptic Plasticity
20          Throughout the 1990s, the LTP model of synaptic plasticity was utilized in studies of Pb
21    neurotoxicity in laboratories around the world, undoubtedly because it was widely accepted that
22    the model invoked synaptic processes that also were involved in learning and cognitive function.
23    These investigations resulted in large body of evidence that characterized the actions of chronic
24    exposure  across several  experimental parameters (see Table 5-3.1). Furthermore, at least in the
25    hippocampal subregions, CA1 and dentate gyrus, there was uniform agreement as to the
26    alterations that resulted.
27          Chronic developmental Pb exposure decreased the magnitude  of LTP and increased the
28    threshold for induction (Altmann et al., 1993; Gilbert et al., 1996; Gutowski et al., 1998; Ruan
29    etal., 1998). Simultaneous assessments of paired-pulse functions also uncovered reductions in
30    paired-pulse facilitation, indicative of reduced glutamate release (Lasley and Gilbert, 1996;
      December 2005                            5-26       DRAFT-DO NOT QUOTE OR CITE

-------
                             Table 5-3.1. Chronic Lead Exposure and LTP
                            Exposure                                            Effect of Exposure on
      Recording Site          Period1      Blood Pb2   Brain Pb3   Preparation             LTP

      Hippocampal Dentate Gyrus

      Gilbert et al. (1996)     PO - P90-120

      Ruanetal. (1998)      PO-P90-115

      Gilbertetal. (1999a)    G16-P130-210
                          P30-P130-210

      Gilbert etal. (1999b)    G16-P120-180
      Gilbert and Mack
      (1998)

      Hippocampal CA1

      Altmann et al. (1993)
Gl 6-P210-540
GO-P70-210
      Gutowski et al. (1998)   GO - P90-130

      Hippocampal CAS

      Gutowski et al. (1997)   GO - PI 3-140

      Gutowski et al. (1998)   GO - P90-130
37.2

30.1

40.2
38.7

26.84
40.2
61.8

ND
14.3


16.0



28.5

16.0
ND

180

378
350

220
378
670

ND
160


135



180

135
in vivo      elevated induction threshold

in vivo      diminished magnitude

in vivo      elevated induction threshold
           and diminished magnitude

in vivo      elevated induction threshold
           and diminished magnitude


in vivo      accelerated decay
 slices      blocked, required exposure
           during early development

 slices      diminished magnitude
                                           slices      no effect across 4 ages

                                           slices      no effect
      'Exposure duration in terms of gestational (G) or postnatal (P) days of age; PO = day of birth.
      2Values expressed as ug/100 ml.
      3Values expressed as ng/g tissue.
      4Different blood Pb values generated by differing levels of exposure.
1    Ruan et al., 1998).  It was also shown that the potentiation produced in Pb-exposed animals

2    decayed more rapidly than in controls (Gilbert and Mack, 1998).

3           Gilbert et al. (1999a) compared the effects on LTP when exposure occurred during

4    different developmental periods. These workers found that animals whose exposure began

5    shortly after weaning exhibited the same impairments in LTP as animals continuously exposed

6    from late gestation when testing in both groups occurred well into adulthood.  A smaller effect

7    on potentiation was observed when exposure was restricted to the late gestation/weaning period.
     December 2005
                         5-27
                    DRAFT-DO NOT QUOTE OR CITE

-------
 1          Gilbert et al. (1999b) also examined the effects of Pb on LTP as a function of chronic
 2   exposure level, utilizing a range of 0.1-1.0% Pb in the drinking water (corresponding to blood
 3   Pb values of 26 to 117 |ig/dL).  A reduced capacity for LTP was found at all exposure levels
 4   except in the 1.0% groups, indicative of a biphasisc dose-effect relationship (Figure 5-3.3).
 5   The 1.0% Pb-exposure level was clearly less effective than the lower exposure groups in
 6   reducing LTP magnitude and did not differ significantly from control values. Blood Pb values
 7   were elevated as a function of increasing exposure and could not account for the lack of effect in
 8   the 1.0% exposure group.
 9
                                                              L
                                               i      i     I     r
                                          0   0.1   0.2  0.5   1.0
                                         Pb Concentration (%)
     Figure 5-3.3.   I/O function difference score-PS amplitude.
     Source: Gilbert et al. (1999b).
10         Zhao et al. (1999) utilized low frequency electrical stimulation in the paradigm of long-
11   term depression (LTD) and found that chronic Pb exposure depressed the magnitude of this form
12   of synaptic plasticity in both hippocampal CA1 and dentate gyrus subregions. The authors also
13   made the point that in  combination with the reduced magnitude of LTP reported by other
14   workers, the decrease in LTD magnitude results in a reduced range of synaptic plasticity in
15   chronically exposed subjects.
     December 2005
5-28
DRAFT-DO NOT QUOTE OR CITE

-------
 1          While the effects of Pb on synaptic plasticity are quite similar in the CA1 and dentate
 2    gyrus regions, they are not uniformly present throughout this brain area.  Gutowski et al. (1997,
 3    1998) were unable to find any effect of chronic Pb exposure on LTP in hippocampal CAS (i.e.,
 4    mossy fiber LTP), even when the investigation was extended across multiple ages. The bases for
 5    this distinction await future investigation.
 6
 7    Lead Exposure, Glutamatergic Transmission, and Synaptic Plasticity
 8          Investigation  of the synaptic processes underlying LTP has provided insight into the bases
 9    for Pb exposure-related impairment of potentiation and cognitive ability (Lasley and Gilbert,
10    2000).  Biochemical and neurophysiological approaches (Lasley and Gilbert, 1996; Gilbert et al.,
11    1996; Ruan et al., 1998) have found stimulated glutamate release to be diminished in
12    hippocampus at blood Pb values where deficits in LTP have been observed. Multiple actions of
13    Pb may be involved  at this exposure level because animals exposed postweaning exhibited
14    similar decrements in evoked glutamate release to those exposed continuously from conception
15    (Lasley et al.,  1999), similar to the observations for measures of LTP.  A biphasic dose-effect
16    relationship was also found in which stimulated glutamate release in hippocampus was decreased
17    at intermediate exposures, but not at higher levels (Lasley and Gilbert, 2002).  On the basis of
18    these observations, it is apparent that decreases in stimulated glutamate release are a significant
19    contributing factor to the Pb exposure-related changes seen in LTP.
20          In comparison to the concordance across laboratories with regard to effects of chronic Pb
21    exposure on LTP and the notable similarities to its actions on glutamate release,  the effects of
22    exposure on the NMDA receptor are relatively variable. That is, there is not widespread
23    agreement as to the nature of the exposure-induced changes. Alterations in receptor function
24    occur readily in response to externally applied treatments and might be expected to vary in a
25    dynamic fashion as a function of exposure parameters (e.g., Lasley et al., 2001). However, most
26    studies have involved measures of NMDA receptor integrity in adult animals exposed to constant
27    levels of Pb for at least three, and more commonly 6 to 15, months, so that receptor-mediated
28    effects  should have stabilized.  Consequently, the following alternative conclusions could be
29    proposed regarding the actions of Pb exposure on the NMDA receptor that are related to its
30    effects  on LTP. First, changes in NMDA receptor function may depend on specific Pb exposure
31    conditions. For example,  a postweaning exposure protocol  may not necessarily produce similar

      December 2005                            5-29        DRAFT-DO NOT QUOTE OR CITE

-------
 1    effects to an exposure protocol initiated during earlier development. Alternatively, exposure
 2    effects on LTP may be produced at signal transduction or other cellular loci that exert regulatory
 3    influences on the NMDA receptor. This latter conclusion implies that changes in the NMDA
 4    receptor do not mediate the primary action of Pb on LTP. In addition, this indicates that
 5    identification of some site of direct Pb effect that has regulatory influence on the receptor would
 6    produce more consistently observable findings.
 7
 8    Lead and Electrophysiological Changes in Dopaminergic/Cholinergic Systems
 9          Electrophysiological approaches have been employed to delineate other interesting
10    findings in Pb-exposed animals not directly related to synaptic plasticity.  Using standard
11    extracellular recording methods, Tavakoli-Nezhad et al. (2001) identified an exposure-dependent
12    decrease in the number of spontaneously active dopamine cells in the substantia nigra and ventral
13    tegmental area, but they found no evidence that this decrease was related to a physical loss of
14    cells.  In subsequent work, Tavakoli-Nezhad and Pitts (2005) determined that the decrease in the
15    number of active dopamine cells was not based on depolarization inactivation.  However they
16    were able to discern a reduced impulse flow in dopamine neurons and a diminished sensitivity of
17    DI receptors in the nucleus accumbens.
18          The actions of Pb2+ on cholinergic nicotinic receptors have been investigated in acutely
19    dissociated or cultured hippocampal cells using the patch clamp technique in whole cell mode
20    (Ishihara et al., 1995).  These workers found that Pb2+potently inhibits activation of fast-
21    desensitizing nicotinic currents in a noncompetitive and voltage-dependent manner.  The
22    nicotinic receptors affected (methyllycaconitine-sensitive) were more sensitive to Pb2+ than other
23    nicotinic subtypes and are known to be highly permeable to Ca2+. This latter observation likely
24    explains the potency for their inhibition by Pb2+.
25
26    5.3.1.4  Lead Exposure and Sensory Organ Function
27          Another focus area for Pb neurotoxicity research that has generated valuable and relevant
28    scientific findings has been sensory organ function.  Visual and auditory systems have received
29    the most attention, have generated results closely resembling clinical observations, and have
30    been successful in defining some of the mechanisms underlying the exposure-induced
31    alterations. These studies are summarized in Table AX5-3.3.

      December 2005                            5-30        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Sensory Organ Assessments in Nonhuman Primates
 2          Lilienthal and Winneke (1996) tested monkeys continually exposed to Pb from gestation
 3    through 8 to 9 years of age and found increased latencies for waves I, II, and IV in brainstem
 4    auditory evoked potentials.  These effects persisted for at least 18 months after exposure was
 5    terminated and blood Pb values had declined nearly to control levels, leading to the conclusion
 6    that these actions of Pb were not dependent on current exposure. Rice (1997) determined pure
 7    tone detection thresholds in monkeys exposed continually from birth to 13 years of age, and
 8    reported that half of the subjects exhibited thresholds outside of the control range at some
 9    frequencies. These findings are consistent with reported alterations in auditory function in
10    humans exposed to Pb developmentally (Otto and Fox, 1993).
11          Visual function was assessed in monkeys by Reuhl et al. (1989), who exposed a high- and
12    a low-dose group from birth to 6 years of age.  This investigation uncovered a decrease in
13    neuronal volume density in cortical areas VI and V2 in the high-exposure compared to the low-
14    exposure group. These workers also found a decrease in dendritic arborization in pyramidal
15    neurons in these brain areas, leading to the conclusion that chronic developmental Pb exposure
16    produces changes in cytoarchitecture in visual projection areas.
17
18    Retinal Function in Rodents
19          The actions of Pb on retinal cells have been a focus of research for more than two
20    decades. It has long been recognized that Pb2+ exhibits a selective effect on rod cells (Fox and
21    Sillman, 1979) and, more  recently, that the associated loss of rod and bipolar cells was due to
22    exposure-induced apoptotic  changes (e.g., Fox et al., [1997]). These observations have been
23    linked with exposure-related alterations in rod-mediated visual function. In vitro studies
24    utilizing free Pb2+ ion concentrations have done much to elucidate the mechanistic bases of these
25    observations.
26          These latter efforts have established the concentration-dependent inhibition of cyclic
27    GMP (cGMP) hydrolysis by free Pb2+, in addition to increases in retinal cGMP and rod Ca2+
28    levels (e.g., Srivastava et al., [1995]). Kinetic studies utilizing purified rod cGMP
29    phosphodiesterase have shown that pM Pb2+ concentrations competitively inhibit the enzyme
30    relative to mM concentrations that are required for Mg+2 cofactor activity, thus binding with 104-
31    106-fold higher affinity than Mg+2 and preventing cGMP hydrolysis (Srivastava et al., 1995).

      December 2005                            5-31         DRAFT-DO NOT QUOTE OR CITE

-------
 1    When retinas are incubated in Ca2+ and/or Pb2+ in vitro, the rods selectively die by apoptosis
 2    associated with mitochondrial depolarization, release of mitochondrial cytochrome c, and
 3    increased caspase activity (He et al., 2000). He et al. (2003) have proposed that apoptosis is
 4    triggered by Ca2+ and Pb2+ overload due to translocation of cytosolic Bax to the mitochondria,
 5    which likely sensitized the overloaded mitochondria to release cytochrome c.  Subsequent work
 6    found the elevations in free Ca2+ and Pb2+ to be localized to photoreceptors  and determined that
 7    the effects of the two ions were additive and blocked by a mitochondrial permeability transition
 8    pore inhibitor (He et al., 2000).  This suggested that the two ions bind to the internal metal
 9    binding site of this pore and, thereby, initiate the  apoptosis cascade.
10          These mechanisms are entirely consistent with electroretinogram (ERG) changes observed
11    in animals chronically exposed  during early development: decreases in maximal ERG
12    amplitude, decreases in absolute ERG sensitivity, and increases in mean ERG latency that were
13    selective for rod photoreceptors (Fox and Farber, 1988). Also in agreement with these
14    mechanisms are observed elevations in retinal cGMP levels and reductions in light-activated
15    cGMP phosphodiesterase activity.  Moreover, the exposure level-dependent degeneration of rod
16    and bipolar cells exhibited the classical morphological features of apoptotic cell death (Fox et al.,
17    1997).  Other measures of visual function in chronically exposed animals also have been found
18    to be consistent with the mechanistic data. Long-term dose-dependent elevations in response
19    thresholds are present but only at scotopic (i.e., rod-mediated) backgrounds, and dark adaptation
20    is delayed (Fox et al., 1994).  In addition, exposure-induced decreases in rhodopsin content that
21    were proportional to the loss of rod cells have been reported (Fox et al.,  1997) as well as dose-
22    dependent decreases in retinal Na+, K+-ATPase activity (Fox et al., 1991).
23          These studies investigating rod photoreceptors are perhaps the best examples of the ability
24    to correlate data obtained in vitro with findings derived from in vivo exposure and with changes
25    in visual physiology. In multiple instances, the same cellular mechanisms are affected with each
26    approach and are consistent with ERG and rod-mediated functional measures. These
27    relationships are summarized in Table 5-3.2.
28
29    5.3.1.5  Neurobehavioral Toxicity Resulting from Lead Exposure
30          The breadth of research examining Pb neurotoxicity utilizing behavioral approaches is
31    quite diverse with respect to test paradigms, exposure parameters, test species, and

      December 2005                             5-32        DRAFT-DO NOT QUOTE OR CITE

-------
               Table 5-3.2.  Mechanisms of Pb-Induced Impairment of Retinal Function
       In Vitro Evidence            In Vivo Evidence                   Physiological Changes
       Competitive inhibition of       Increased retinal cGMP
       cGMP PDE
       Increased retinal cGMP        Decreased stimulated cGMP PDE       Decreased maximal ERG amplitude
                                 activity                            Decreased absolute ERG sensitivity
                                                                  Increased mean ERG latency
       Increased rod [Ca2+]
       Apoptosis from increased       Morphological features of apoptotic rod,  Increased response thresholds at
       photoreceptor Ca2+/Pb2+ via     bipolar cell death                    scotopic backgrounds
       binding to mitochondrial       Decreased rhodopsin proportional to     Delayed dark adaptation
       permeability transition pore     cell loss
                                 Translocated cytosolic Bax to the
                                 mitochondria, cytochrome c released
       Decreased retinal Na+,         Decreased retinal Na+,K+-ATPase
       K+-ATPase activity           activity
       Abbreviations: PDE, phosphodiesterase; ERG, electroretinogram.
 1    neuropharmacological agents used.  This large literature, summarized in Table AX5-3.4, has
 2    permitted development of insightful generalizations while, at the same time, providing focused
 3    descriptions of specific behaviors.  In addition, the accumulated evidence has supported the
 4    development of more effective and refined methodologies.
 5
 6    Lead-Induced Alterations of Behavior - Nonhuman Primates
 1          In reviewing the results of behavioral investigations of Pb neurotoxicity in nonhuman
 8    primates conducted over the last two decades, it is abundantly clear that the results are shaped by
 9    the nature of the test  paradigm and the developmental exposure periods utilized.  Thus, studies
10    employing nonspatial discrimination reversal (Rice and Gilbert, 1990a), spatial delayed
11    alternation (Rice and Gilbert, 1990b), and spatial discrimination reversal (Rice, 1990) produced
12    observations that are distinctly different.  Experimental groups continually exposed to Pb from
13    birth to testing as adults typically exhibit learning deficits, but groups continually exposed
14    beginning after weaning or whose exposure from birth is terminated during development may or

      December 2005                             5-33        DRAFT-DO NOT QUOTE OR CITE

-------
 1    may not display differences from control animals depending on the sensitivity to exposure of the
 2    test paradigm.
 3          Nonetheless, some characteristics of experimental subjects can be gleaned from
 4    investigations of neurobehavioral toxicity in nonhuman primates. Modifications of experimental
 5    parameters that make task acquisition or retention more challenging (Rice and Gilbert, 1990b;
 6    Rice, 1990) are more likely to elicit exposure-related changes in responding. In test paradigms
 7    based on fixed interval reinforcement or differential reinforcement of low rate responding
 8    schedules, Pb-exposed subjects displayed decreased inter-response times and a greater ratio of
 9    responses per earned reinforcement (Rice,  1992a,b).  Exposed animals also are less sensitive to
10    changing reinforcement contingencies and, therefore, commit more perseverative errors in
11    responding (Rice, 1992c; Newland et al., 1994). Not surprisingly, it has been noted that these
12    experimental behavioral effects correspond reasonably well to epidemiologic observations in
13    Pb-exposed children (Rice, 1996; Lasley and Gilbert, 2000), thus validating the use of this
14    species as an exposure model.
15
16    Lead-Induced A Iterations of Behavior — Rodents
17          The observations of Pb's neurobehavioral effects in rodents in many ways resemble those
18    conclusions attained with nonhuman  primates.  However, the test paradigms utilized for rats
19    have been somewhat more refined, and the behavioral data have been subjected to  more detailed
20    analyses. As a result, valuable insights into the component mechanisms underlying the
21    exposure-related changes have been achieved.
22          An olfactory serial reversal paradigm was utilized to demonstrate Pb-induced impairments
23    in learning reversals (Hilson and Strupp, 1997; Garavan et al., 2000). These workers found that,
24    when presented with altered reinforcement contingencies for the  reversals, rats whose exposure
25    was limited to early development exhibited a shortened initial period of responding to the
26    previously correct cue coupled with a prolonged postperseverative learning phase for the new
27    task. Hilson and Strupp (1997) concluded that the impaired reversal learning was due to a
28    deficiency of learning new contingencies of the task (i.e., an associative deficit), and not based
29    on inflexibility or deficient inhibitory control. Subsequent work  by Garavan et al.  (2000)
30    determined that this associative deficit was based on a response bias and an impaired ability to
31    associate cues and/or actions with their affective consequences.

      December 2005                            5-34       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Employing a visual discrimination task, Morgan et al. (2000) found that as the level of Pb
 2    exposure restricted to early development increased, learning of the task slowed and the number
 3    of defined "impaired" animals increased.  The authors concluded that the deficits were not
 4    limited to attentional function and that an associative deficit had resulted along with a tendency
 5    to respond more rapidly. Subsequent work with visual discrimination vigilance tasks found that
 6    animals exposed only during gestation and/or lactation exhibited impaired  response initiation and
 7    increased omission errors, indicating a lasting deficiency in sustained attention and an increased
 8    reactivity to errors (Morgan et al., 2001).  The authors concluded that the effects of exposure are
 9    determined not only by the paradigm, but also by the timing and intensity of exposure. Cory-
10    Slechta (2003) came to a similar conclusion on the factors underlying the manifestation of Pb
11    effects but suggested that the alterations in attention may be due to impulsivity or aversion to
12    delays.
13          The actions of early Pb exposure on memory appear to be task-dependent, but this issue
14    has not been clearly defined. Alber and Strupp (1996) found that exposed  rats performed more
15    poorly on a series of spatial alternation tasks but that the deficit did not vary across intertrial
16    delays, suggesting that memory was not impaired. Murphy and Regan (1999) used a one-trial,
17    light/dark, passive avoidance paradigm and observed a decrease in recall latency on post-training
18    day 5 in rats whose exposure was restricted to early development. Since there was no exposure
19    effect evident during the first 48 h after training, these authors concluded that the impairment
20    was associated with long-term memory storage.  Further studies are needed to more clearly
21    characterize the effects of chronic Pb on memory function.
22
23    Interactions of Lead Exposure and Responding to Cocaine
24          Behavioral responses to a number of neuropharmacological agents have resulted in
25    important and useful insights into Pb neurotoxicity.  One approach that has been unique and has
26    produced scientifically important results has been investigation of the interactions  of chronic Pb
27    exposure and responses to cocaine.  Chronic exposure of adult male rats has been shown to
28    attenuate cocaine-induced locomotor activation (Grover et al.,  1993) and result in a slower
29    development and reduced magnitude of cocaine-induced sensitization of locomotor activity
30    (Nation et al., 1996).  The latter observations are consistent with other evidence of impaired
31    synaptic plasticity that were presented earlier in this chapter. These actions of exposure are not

      December 2005                            5-35         DRAFT-DO NOT QUOTE OR CITE

-------
 1    specific to cocaine, as a similar exposure regimen attenuated the reinforcing effect of brain
 2    stimulation of the medial forebrain bundle (Burkey and Nation, 1994), the nerve tract conveying
 3    nigrostriatal and mesolimbic dopaminergic neurons to forebrain regions.
 4          Using the drug discrimination paradigm, Miller et al. (2001) restricted Pb exposure to the
 5    gestational and lactational periods of early development and observed decreased sensitivity to
 6    dopamine DI and D2 receptor agonists when the animals were tested as adults.  These findings
 7    may be taken as evidence of receptor downregulation, but, in this behavioral task, subjects
 8    received chronic intermittent doses of the training drug, which in this study was a low dose of
 9    cocaine.  Thus, the actions of exposure on dopaminergic systems may be confounded with the
10    receptor changes induced by chronic drug administration.
11          In contrast to the attenuating effects of chronic Pb administration to adults described
12    above, exposure restricted to the gestational and lactational periods exerts potentiating effects on
13    other types of responses to cocaine when animals are tested long after exposure is terminated.
14    Nation et al. (2003) trained rats to self-administer cocaine intravenously, extinguished the
15    response, and then used a systemically  administered priming dose of the drug to initiate a relapse
16    response. Exposed animals  were found to have an increased sensitivity to cocaine relapse
17    compared to identically treated controls. When multiple cocaine doses were provided,
18    identically exposed animals  were found to self-administer more of a low dose of the drug and
19    less of a high dose than controls, again suggesting an enhanced sensitivity to the actions of
20    cocaine (Nation et al., 2004). Finally, animals exposed to Pb in this manner were found to have
21    an accelerated rate of acquisition of cocaine self-administration behavior (Rocha  et al., 2005).
22
23    Lead Exposure and the Stimulus Properties of Neuropharmacological Agents
24          The drug discrimination paradigm has been utilized more widely in Pb neurotoxicity
25    research to characterize postsynaptic receptor status for multiple neurotransmitter systems and
26    has resulted in some useful findings. Rats chronically exposed beginning at weaning and tested
27    as adults were trained to discriminate either a systemically administered DI or D2 receptor
28    agonist (Cory-Slechta and Widzowski,  1991).  Exposed rats learned the discrimination task more
29    rapidly than controls and exhibited greater levels of response to lower doses of the training drugs
30    and less blockade by a D2 receptor antagonist, consistent with generalized dopaminergic receptor
31    supersensitivity.  In groups of animals exposed only from  birth to weaning and trained to

      December 2005                            5-36        DRAFT-DO NOT QUOTE OR CITE

-------
 1    discriminate the same drugs, the D2-D3 subtype receptor supersensitivity in exposed animals was
 2    again present, but no changes in responding to the DI agonist were apparent (Cory-Slechta et al.,
 3    1992).  Further work with this test paradigm employing the postweaning exposure protocol failed
 4    to demonstrate any Di-D2 receptor interactions in the supersensitivity displayed by Pb animals
 5    (Cory-Slechta et al., 1996a).
 6          To test cholinergic sensitivity in animals chronically exposed after weaning, rats were
 7    trained to discriminate a muscarinic agonist (Cory-Slechta and Pokora, 1995) and were tested in
 8    the added presence of a muscarinic antagonist. The results suggest an increased sensitivity to at
 9    least one subtype of muscarinic receptor in Pb-treated rats.
10          Glutamatergic functioning also has been assessed by use of the drug discrimination
11    paradigm.  Rats chronically exposed beginning at weaning and tested as adults exhibited
12    diminished responsiveness to an NMDA subtype receptor antagonist (Cory-Slechta, 1995b) but
13    enhanced responsiveness to lower doses of NMDA (Cory-Slechta et al., 1996b).  When exposure
14    was limited to the period between birth and weaning, the diminished sensitivity to the NMDA
15    receptor antagonist was less evident, but still present (Cory-Slechta, 1997b).
16          Thus, the drug discrimination paradigm appears to provide useful insights into the status
17    of some neurotransmitter systems in chronically Pb-exposed animals. The reports cited above
18    indicate an upregulation of dopaminergic, cholinergic, and glutamatergic receptors that are
19    generally consistent with findings of diminished presynaptic function described earlier in this
20    section of the current Lead AQCD.  Nonetheless, this paradigm has some limitations. As  all
21    drugs in the cited studies were administered systemically, the results provide no evidence  on
22    brain regional sites of action. In addition, the chronic intermittent administration of the training
23    drug has the potential to induce compensatory neuronal changes by itself, and thusly may  mask
24    or otherwise alter the manifestation  of the effects of Pb exposure.  Future use of this paradigm in
25    Pb neurotoxicity studies must acknowledge this latter consideration.
26
27    Other Effective Behavioral Test Paradigms
28          Another test paradigm effectively utilized at least transiently to distinguish changes in
29    chronically Pb-exposed animals is the repeated acquisition and performance  schedule (Cohn
30    et al., 1993). The purpose of this paradigm was to determine the selectivity of Pb-induced
31    changes in learning, as distinct from nonspecific or performance effects, and to explore the

      December 2005                            5-37        DRAFT-DO NOT QUOTE  OR CITE

-------
 1    nature of the underlying error patterns contributing to any learning deficits. This schedule
 2    required completion of a sequence of three responses for reinforcement, with the correct
 3    sequence for the learning (i.e., repeated acquisition) component changing with each successive
 4    experimental session, while the performance component sequence remained constant across
 5    sessions.
 6          The use of this schedule in animals chronically exposed to Pb beginning at weaning
 7    uncovered significant decrements in accuracy on the learning component, but not on the
 8    performance component, in Pb groups compared to controls (Cohn et al.,  1993).  A detailed
 9    analysis of subjects' behavior indicated that Pb exposure impaired learning by increasing
10    perseverative responding on a  single lever, even though such repetitive responding was not
11    directly reinforced. In a  subsequent study, dose-effect curves for the NMDA receptor antagonist
12    MK-801 were determined in controls and animals tested in this paradigm in which chronic
13    exposure began at weaning (Cohn and Cory-Slechta, 1993). The decline in learning accuracy
14    and the increases in perseverative responding produced by MK-801 were attenuated by Pb
15    exposure, and dose-effect curves relating MK-801 dose  to changes in rates of responding were
16    shifted to the right in exposed rats compared to control animals.  These observations, therefore,
17    demonstrated a sub sensitivity of Pb-exposed animals to  both the accuracy-impairing and
18    response rate-altering properties of the antagonist. An additional investigation utilized the same
19    Pb exposure protocol and administration of doses of NMDA as a receptor agonist to rats
20    undergoing this test paradigm (Cohn and Cory-Slechta,  1994b).  In control animals, NMDA was
21    found to decrease accuracy of response in both the repeated acquisition and performance
22    components of this multiple schedule and to  suppress response rates as well. Lead exposure
23    potentiated the accuracy-impairing effects of NMDA by further increasing the frequencies of
24    errors and likewise potentiated the drug's rate-suppressing effects. Thus, as stated earlier in this
25    section, the Pb-induced potentiation of the agonist effects and reduced sensitivity to the
26    antagonist effects in this  test paradigm are consistent with an increased density or some other
27    upregulation of NMDA receptors in exposed brain tissue. In other work, Cohn and Cory-Slechta
28    (1994a) were unable to distinguish any evidence of dopaminergic modulation of responding in
29    this behavioral paradigm. Thus, the repeated acquisition and performance schedule proved
30    valuable not only in providing a finer dissection of the animal's behavior, but in elucidating
      December 2005                            5-38        DRAFT-DO NOT QUOTE OR CITE

-------
 1    important aspects of Pb neurotoxicity without some of the limitations inherent with drug
 2    discrimination or other behavioral test methods.
 3
 4    Summary
 5          There is general agreement that the important factors in determining behavioral responses
 6    of Pb-exposed animals are (a) the nature of the test paradigm and its sensitivity to exposure and
 7    (b) the timing and intensity of the Pb exposure. Detailed analyses of responding have shown that
 8    Pb-exposed animals are less sensitive to the changing reinforcement contingencies that are
 9    integral to  series of reversal tasks.  They exhibit shortened initial periods of responding to the
10    previously correct cue in combination with prolonged postperseverative learning phases for the
11    new task.  These have been proposed to be associative deficits based on deficiencies in learning
12    new response contingencies.  In addition, the impaired responding of Pb-exposed animals in
13    vigilance tasks has been attributed to deficiencies in sustained attention and an increased
14    reactivity to errors.
15          Other test paradigms such as drug discrimination and repeated acquisition/performance
16    tasks have provided useful assessments of the integrity of CNS neurotransmitter systems in
17    Pb-exposed animals.  Evidence from both paradigms has been in general agreement in indicating
18    up-regulated neurotransmitter receptor systems. The timing of Pb exposure is critically
19    important in determining the response to cocaine, and the potentiating action of perinatal Pb
20    exposure is of potential importance for public health purposes.
21
22    5.3.1.6  Lead-Induced Changes in Cellular Development and Disposition of the Metal
23          Alterations in cellular differentiation and morphology can be important structural
24    neuronal and glial components of the manifestations of Pb neurotoxicity.  While these issues
25    have not been thoroughly addressed by research investigations, there have, nonetheless, been
26    important observations made.  This subsection reviews studies concerned with various aspects of
27    this topic.
28
29    Lead Exposure and Neural/Glial Progenitor Cells
30          Studies of the effects of Pb exposure on neural and glial progenitor cells are recent
31    occurrences in the field of Pb neurotoxicity research.  Chronic exposure in rats begun at postnatal

      December 2005                            5-39        DRAFT-DO NOT QUOTE OR CITE

-------
 1    day 25 was found to significantly decrease proliferation of new cells in the dentate gyrus
 2    compared to the extent of this process in control animals (Schneider et al., 2005). Other workers
 3    initiated Pb exposure at birth and determined that continuous exposure to adulthood reduced the
 4    total number of labeled cells in the hippocampal dentate gyrus at 28 days, but not 24 h, after the
 5    last administration of a DNA synthesis marker (Gilbert et al., 2005). Rats whose exposure was
 6    terminated at weaning exhibited no changes in cellular labeling or survival, indicating that
 7    chronic exposure reduces the capacity for hippocampal neurogenesis.
 8          Studies have also been conducted to investigate the effects of exposure on glial progenitor
 9    cells.  Deng et al. (2001) examined cultured oligodendrocytes and their progenitor cells acutely
10    exposed to Pb2+ in vitro; they observed an exposure-induced delay in the differentiation of the
11    progenitors, and that the progenitor cultures were more sensitive to Pb2+ than the mature
12    oligodendrocytes. These findings suggested interference with the timely developmental
13    maturation of the progenitor cells. A subsequent study found that a low concentration of Pb2+ in
14    vitro inhibited proliferation and differentiation of these progenitors without affecting cell
15    viability (Deng and Poretz, 2002). Proliferative capability was decreased and cell-intrinsic
16    lineage progression was inhibited at a late progenitor stage. Thus, acute Pb2+ suppresses both the
17    proliferation and differentiation of these cells.
18
19    Lead Exposure and Neurite Outgrowth
20          Neurite initiation is known to  be highly sensitive to neurotoxic compounds and has been
21    the focus of studies examining morphological alterations caused by exposure to Pb2+ in vitro.
22    Kern and Audesirk (1995) found that 100 nM Pb2+ inhibited neurite initiation in cultured rat
23    hippocampal neurons and, on the basis of results with kinase inhibitors, concluded that this
24    occurred by inappropriate stimulation of protein phosphorylation by Ca2+-calmodulin-dependent
25    or cyclic AMP-dependent protein kinases, possibly through stimulation of calmodulin.
26    Intracellular free Ca2+ concentrations were not altered by up to 48 h exposure to nominal 100 nM
27    Pb2+, leading these workers to propose that the stimulation of the above kinases or calmodulin
28    were not via increased Ca2+ but, instead, were attributable to intracellular Pb2+ concentrations.
29    Evidence of Pb2+-induced inhibition  of neurite outgrowth is in general agreement with
30    observations made after chronic exposure to Pb employing in vivo models. Cline et al. (1996)
31    employed an exposure protocol of 0.1 nM-100 jiM nominal Pb2+ for 6 weeks localized to the

      December 2005                            5-40        DRAFT-DO NOT QUOTE OR CITE

-------
 1    retinotectal system of frog tadpoles, and observed a severely reduced area and branchtip number
 2    of retinal ganglion cell axon arborizations within the optic tectum at nM Pb2+ concentrations.
 3    Reuhl (1989) exposed primates to 2 mg lead/kg/day from infancy to 6 years of age and found
 4    that neuronal volume density was reduced in primary visual area VI and in visual projection area
 5    V2 compared to a group exposed to 25 jig lead/kg/day. Moreover, a relative decrease in the
 6    number of arborizations among pyramidal neurons in both areas VI and V2 was observed in the
 7    higher dose group.  Thus, there is good correspondence between reports that acute Pb2+ exposure
 8    in vitro and extended exposure in animal models in vivo results in diminished neuronal growth
 9    and differentiation at Pb levels of apparent environmental relevance. Studies employing intact
10    animals have not progressed to investigation of specific cellular mechanisms underlying these
11    effects.
12
13    Lead Exposure and Neural Stem Cells
14          Given considerable contemporary interest in the use of neural stem cells to treat various
15    neurological diseases, the efforts of Huang and Schneider (2004) to examine the actions of
16    exposure to Pb2+ in vitro on these cells is worthy of note.  Lead exposure produced no effect on
17    neurosphere viability, but, it did cause a significant dose-dependent inhibition of proliferation.
18    In addition, the number of neurons differentiated from Pb2+-exposed neurospheres was
19    significantly decreased from control, as were the number of oligodendrocytes obtained.
20    However, Pb exposure increased the number of astrocytes obtained.  These observations suggest
21    an important Pb2+-induced influence on stem cell proliferation and differentiation  that has public
22    health relevance to prenatal metal exposure.
23
24    Accumulation of Lead in Brain
25          Most studies of neurotoxicity involving chronic Pb exposure now report blood and brain
26    Pb concentrations to quantify exposure magnitude  and/or as quality control measures.  Thus, a
27    sizable amount of data is available on general aspects of Pb toxicokinetics. While brain Pb
28    values vary monotonically with blood Pb concentrations and exposure levels, steady state
29    accumulation/washout times are longer in tissue and are dependent on exposure magnitude and
30    duration. The half-time for the decline of Pb in brain tissue when exposure is terminated is on
      December 2005                           5-41        DRAFT-DO NOT QUOTE OR CITE

-------
 1    the order of 10 days to 2 weeks, while the value for blood leads would be a matter of a few days
 2    (Lasley, unpublished observations).
 3          The speciation and distribution of Pb in brain tissue is largely unknown except for indirect
 4    indications that only a small fraction of the divalent cation is present in tissue in extracellular
 5    fluid in the free ion state. The existence of a lead-binding protein in brain cytosol was reported
 6    by Goering et al. (1986) and was invoked to explain the relatively weak inhibition by Pb2+ of
 7    brain 8-aminolevulinic acid dehydratase activity.  But the binding protein was not fully
 8    characterized or identified. Pb2+ is known to bind to various intracellular Ca2+ binding proteins,
 9    such as calmodulin, PKC, and synaptotagmin, particularly those with a C2 domain (Sun et al.,
10    1999), but these are low-volume sources that have been studied for their functional importance
11    and would not serve any kind of tissue metal storage function.
12          Pb2+ appears to be taken up into cultured cells by multiple ion channel-based mechanisms
13    including influx through channels activated by depletion of intracellular Ca2+ stores, non-L-type
14    Ca2+ channels, and NMDA receptor-associated channels (Kerper and Hinkle, 1997; Mazzolini
15    et al., 2001). Astroglia are well known to act as a Pb sink and, in culture, accumulate up to
16    24 times more of the metal than neuronal cells (Lindahl et al., 1999); and there is evidence that
17    glutathione may regulate Pb uptake into these cells. Only recently has one astroglial protein
18    been identified—a molecular chaperone in endoplasmic reticulum (Qian et al., 2000, 2005),
19    glucose-regulated protein (GRP78).  Intracellular levels of this protein are increased in cultured
20    astroglia during one week's exposure to Pb2+, suggesting that this protein is a component of the
21    intracellular tolerance mechanism that handles high intracellular Pb accumulation through a
22    direct interaction. GRP78 depletion significantly increased the sensitivity of cultured glioma
23    cells to Pb2+ as indicated by the generation of reactive oxygen species. Thus, it appears that Pb2+
24    directly targets the protein and induces its compartmentalized redistribution, enabling it to play a
25    protective role in Pb neurotoxicity.
26
27    5.3.1.7  Integration of Research Findings
28          It is evident that the Pb neurotoxicity literature is broad and varied and that many valuable
29    observations have been made over the last 20 years. Nevertheless, a few general conclusions are
30    in order so as to help integrate and concisely summarize evidence in at least a few focused areas.
31

      December 2005                            5-42        DRAFT-DO NOT QUOTE OR CITE

-------
 1       •  Lead-induced impairments in glutamatergic neurotransmission appear to underlie the
 2          deficits in synaptic plasticity and in learning or acquisition behavior.  Cholinergic
 3          neurotransmission also serves an important role in impaired learning in exposed subjects,
 4          while deficits in dopaminergic function are manifested as alterations in rates of
 5          responding or incentive motivation. Exposure-related alterations in structural plasticity
 6          appear to be based on interference with Ca2+ signaling and/or glutamatergic transmission.

 7       •  There is little if any support in the Pb neurotoxicity research community for the notion of
 8          thresholds for any of the toxic mechanisms that have been addressed in this section of the
 9          document.  With the pressure to reduce experimental group sizes to the minimal number
10          necessary and the unspecified notion that rats are somewhat more resistant to Pb than
11          children, most studies performed with in vivo models report blood Pb values in the range
12          of 15 to 35-40 |ig/dL. Moreover, in view of the complex and undefined speciation
13          equilibria and distribution of Pb in  physiological milieus, there is no way to directly
14          relate a blood Pb value to the  levels of free Pb2+ ion or to any other complexed active
15          form of the metal, either in extracellular or intracellular fluids. Generally accepted
16          estimates of the free Pb2+ ion  concentrations produced in brain extracellular fluid by
17          environmentally relevant exposures fall in the low nanomolar range.

18       •  Susceptibility factors for Pb neurotoxicity are poorly defined in laboratory animals, and,
19          thus, have not been studied. A compelling rationale for their investigation has not been
20          provided.
21

22    5.3.2   Neurotoxicological/Neurobehavioral  Effects of Lead in Humans

23          This section is divided into three sub-sections, based upon age and exposure scenarios.

24    The sub-sections include (1) children with blood lead levels above and below 10 |ig/dL, (2) adult

25    manifestations of neurotoxicity and other disease states as a result of excessive exposure to lead

26    as children, and (3) adults who were  exposed to "ambient" levels  of lead. In each of these sub-

27    sections, wherever possible, discussion is focused  on biochemical markers, bioclinical markers,

28    and reversibility of lead's neurotoxic effects.  In addition, for each of the groups cited above,

29    vulnerability to the neurotoxic effects of lead is considered.  Topics in this area include

30    developmental toxicology and growth and  development in children. For children and adults,

31    other aspects of vulnerability are considered, such as socioeconomic status, nutrition, and genetic

32    polymorphisms. Based upon the body of studies discussed in each sub-section, it is reasonable

33    to draw conclusions  relating to dose-response paradigms and clinical extensions of

34    epidemiological data to individual children.
35
      December 2005                            5-43        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.3.2.1  Effects of Lead in Young Children to Mid-Adolescence
 2          Since EPA's publication of the Air Quality Criteria Document and Addenda in 1986-
 3    1990{EPA-600/8-83/028aF(1986); EPA/600/8-89/049F(1990)}, major studies with new and
 4    critical information have substantially extended previous hypotheses expressed by the EPA.
 5    These new data are the primary and major departure from EPA's earlier lead criteria reviews.
 6    It is now recognized that lead has adverse effects on the developing central nervous system of
 7    young children and that these effects  on cognition and behavior persist (at least) into the school-
 8    aged years and beyond into mid-adolescence.  While causal conclusions about effects of lead
 9    exposure on cognitive development are made with caution, collectively, the nature and
10    abundance of the evidence is clear and compelling.  This new information causally links
11    detrimental effects of lead on behavior and cognition at blood lead levels both above and below
12    10 |ig/dL.  Compared to the earlier EPA documents (cited above), there is solid evidence for
13    detrimental effects of lead on neuropsychological functions such as fine motor skills,
14    visual-spatial, and executive functioning and attention in large groups of children.  Apparently,
15    there is no threshold below which lead is without adverse effects on the central nervous system
16    of young children to mid-adolescence.  This conclusion is also related to dose-response
17    paradigms of lead in children, as well as to extending the results of epidemiological studies to
18    individual children.
19
20    Biomarkers
21          There are three generally recognized types of biomarkers (National Research Council
22    [NRC], 1993; Lanphear and Bearer, 2005). The first of these is relevant to quantifying exposure
23    and, thus, the internal dose of lead. This subject is covered in Section 6.1. The second type of
24    biomarker focuses on effects. Biomarkers of effect are biochemical, physiological and/or
25    clinically measurable alterations in normal functioning that reflect an impairment of health or a
26    specific disease.  Effects of this nature include early subclinical  effects of lead that are of value
27    in quantifying human health risks; as a result, such effects can lead to an understanding of
28    mechanisms of lead toxicity at the cellular and organ level. An  example of the latter are the
29    impacts of lead on heme synthesis, which have been extensively studied as discussed in
30    Section 5.2. Other examples include  (in children) effects on the electrophysiology and
31    architecture of the brain. In contrast, biomarkers of susceptibility  or vulnerability are native,

      December 2005                            5-44        DRAFT-DO NOT QUOTE OR CITE

-------
 1    inherent or acquired situational characteristics that alter the responses of children (and adults) to
 2    lead exposure.  Examples of these biomarkers include socioeconomic status, nutrition,
 3    developmental aspects of brain functions, and genetic polymorphisms. Ultimately, these
 4    biomarkers, collectively, as a constellation, are considered to be contributing factors that assist in
 5    determining whether the cognitive effects of lead exposure in children are reversible.
 6
 7    Biochemical Biomarkers
 8          In the AQCD of 1986, reported studies in lead-poisoned children (blood lead levels of
 9    12-120 |ig/dL) revealed an inverse correlation (-0.88) between the entire range of blood lead
10    concentrations and plasma levels of the vitamin D hormone (1, 25-dihydroxyvitamin D) in
11    177 children from 1-16 years of age (Rosen et al., 1980; Mahaffey et al., 1982).  These results
12    suggested that lead impairs the biosynthesis of the Vitamin D hormone; and, as a result, calcium
13    absorption, and possibly that of lead, could be inhibited.  Because of 1,25-dihydroxyvitamin D's
14    roles in multiple cellular functions, including the calcium-messenger cascade,  development and
15    proliferation of multiple  cell types, these clinical  observations have substantial implications
16    (NRC, 1993). These data from clinical studies were supported in experimental animals (Smith
17    et al., 1981). In animals fed a low calcium or phosphate diet, oral administration of 0.82% lead,
18    as the acetate, yielded plasma levels of 1,25-dihydroxyvitamin D that were substantially reduced.
19    This effect of lead on circulating 1,25-dihydroxyvitamin D disappeared when  either a high
20    calcium or phosphate diet, including lead, was administered.  Moreover, intestinal lead appeared
21    to block the absorption of calcium in response to administration of 25-hydroxyvitamin D and
22    1,25-dihydroxyvitamin D, although there was no influence on calcium mobilization neither from
23    bone nor of mineralization of rachitic bone (Smith et al.,  1981).
24          More recently, in  animals and children, this relationship has been examined further.
25    Chicks fed lead concurrently on a low calcium diet replicated the findings in animals and
26    children noted above.  However, chicks fed a calcium sufficient diet coupled to dietary lead,
27    failed to exhibit decreased plasma levels of the vitamin D hormone (Fullmer and Rosen, 1990;
28    Fullmer, 1995, 1997).  Similar findings were reported in lead poisoned children with adequate
29    dietary intakes of calcium as those in experimental studies (Koo et al., 1991).  Compared to the
30    1980s, when dietary intakes  of calcium were marginal (at best) in inner-city children,  recent
31    estimates of dietary calcium intakes in inner city  children meet or exceed recommended daily

      December 2005                            5-45        DRAFT-DO NOT QUOTE OR CITE

-------
 1    requirements of 1000 mg/day or greater (Markowitz et al., 2004).  Thus, in calcium-sufficient
 2    children, plasma concentrations of 1,25-dihydroxyvitamin D are not biomarkers of lead's effects
 3    on the vitamin D endocrine system.  Nonetheless, in at-risk populations of children, whose
 4    dietary intakes of calcium are suboptimal, apparent biosynthesis of the vitamin D hormone,
 5    evidenced by decreased circulating levels of the hormonal form of the vitamin, 1,25-
 6    dihydroxyvitamin D, is expected.
 7          Tang and co-workers (1999) assessed 244 infants on the Brunet-Lezine Scales at
 8    9 months of age to evaluate possible relationships between cord blood lead levels and plasma
 9    concentrations of 5-hydroxyindoleacetic acid (5-HIAA) and homovanillic acid (HVA) at
10    9 months of age. Cord blood lead concentrations were in the range of 2.5-7 |ig/dL. At 9 months
11    of age, 5-HIAA and HVA were negatively correlated with blood lead values and with all the
12    neurodevelopmental functions, except for language, on the Brunet-Lezine Scales.  The negative
13    correlations between the serotonergic system, coupled to blood lead levels, were found in global
14    scores, sociability, and coordination on the Brunet-Lezine Scales.  Although further confirmation
15    of these results is needed, these findings are consistent with the findings in experimental studies
16    linking lead effects to impairments in neurotransmission (Section 5.3).
17
18    Clinically Oriented Biomarkers of Effect
19          Very little information was available in the time frame of 1986-1990 relating to clinically
20    oriented biomarkers of lead's effects on the central nervous system of children. Currently, there
21    is a substantial body of knowledge which focuses on the functional status of the brain in
22    excessively lead-exposed children. This new information includes functional (Bhattacharya
23    et al.,  1993; Rothenberg et al., 1995) and electrophysiological (Otto and Fox, 1993; Burchfiel
24    et al.,  1992; Poblano et al., 2001; Rothenberg et al., 2000) studies. The most relevant studies of
25    functional and electrophysiologic data, relating to spectral analyses of EEGs, brainstem auditory -
26    evoked potentials or responses (B AEP and B AER) are those based upon prospective designs of a
27    well characterized cohort of children.  Another line of important new information is based upon
28    results from assessments of the biochemical and anatomical functions of the central nervous
29    system in lead-exposed children carried out by  magnetic resonance spectroscopy (MRS)(Trope
30    et al., 2001; Meng et al, 2005).  Some of these studies link observed results to  cognitive
31    impairments summarized in  the general introduction to this chapter. Although there may be

      December 2005                           5-46        DRAFT-DO NOT QUOTE OR CITE

-------
 1    some overlap in these reported studies, for the purposes of this discussion, studies are
 2    categorized as functional, electrophysiologic and biochemical-anatomical.
 3          A functional assessment of postural equilibrium was carried out in 109 children (in the
 4    Cincinnati Prospective Study) at 5.8 years of age when the mean blood lead was 5.8 |ig/dL (the
 5    geometric mean blood lead for the first 5 years of life was 11.9 jig/dL). Postural sway was
 6    quantitated with a microprocessor-based platform. A negative correlation was found between
 7    blood lead levels and vestibular/proprioceptive systems, suggesting that lead exposure has
 8    detrimental effects on posture and balance.  Although these data may have potential relevance to
 9    psychomotor deficits observed in lead poisoned  children, this methodology has not been further
10    developed, and its potential implications to psychomotor skills in children have not evolved
11    (Bhattacharya et al., 1993).
12          In the Mexico City Prospective Study, analyses of acoustical cries was carried out in a
13    subset of healthy babies at 2,  15, and 30 days of life. The mean maternal blood lead at 12 weeks
14    of pregnancy was 8.2 |ig/dL,  and the mean cord  blood lead level was 7.8 |ig/dL (range:  1-
15    38 |ig/dL).  The percent nasalization, produced by raising the velum of the velopharynx,
16    decreased progressively over the cord blood lead range of 4-40 |ig/dL, and the number of cries
17    were inversely related to cord blood lead levels over the same range.  In a subset of the babies,
18    decreased nasalization was related to increased BAEP-evoked latencies.  These findings
19    suggested that altered baby cries and auditory functions may be associated with developmental
20    delays affecting early communication. However, validation of infant cries as a predictor of
21    subsequent infant development has not yet evolved.
22          Burchfiel et al., (1992) studied a subset of the Philadelphia children reported on by
23    Needleman et al. (1979). The method used was  brain electrical activity mapping (BEAM)
24    coupled to spectral analysis for each individual electrode. Nineteen children in the uppermost
25    10th percentile for dentine lead (>24 ppm) were  compared to children in the lowest 10th
26    percentile (<6 ppm). The spontaneous resting EEG of the high lead children (n = 19) had higher
27    percentages of low frequency delta activity and reduced percentages of alpha activity compared
28    to the lower lead group of 12 children. Qualitatively, these EEG changes are similar to those
29    observed in acute-severe lead poisoning, and, generally, such findings of diffusely increased
30    slow frequency activity and reduced alpha are commonly found in toxic encephalopathies.
31    These results, which are qualitatively similar to results from the Mexico City Prospective Study,

      December 2005                            5-47        DRAFT-DO NOT QUOTE OR CITE

-------
 1    in which BAER were employed in a different age group, indicate that lead may induce
 2    neuropathological effects in a dose-response manner over a continuum of exposure (see
 3    Section 5.3).
 4          In the Mexico City Prospective Lead Study, 100-113 5 to? year olds underwent testing by
 5    BAER (Rothenberg et al., 2000).  The mean blood lead level at 5 years of age was 8 |ig/dL. The
 6    results indicated that intervals I-V and III-V intervals of BEAR recorded at 5-7 years of age were
 7    related to maternal blood lead at 20 weeks of pregnancy, when the geometric mean blood lead
 8    was 7.7 |ig/dL (range: 1-30.5 |ig/dL). This specificity of the lead effect suggests that the CNS is
 9    exquisitely sensitive to lead when auditory structures are undergoing rapid development, and this
10    effect appears to persist to 5 or more years later. As indicated in this study, lead-related
11    alterations in auditory brain stem function may underlie verbal deficits in lead-exposed children,
12    as well as impair auditory functions observed in lead-exposed animals (Section 5.3.1). Also, it
13    becomes increasingly important to examine functional deficits in hearing and language
14    development that may be associated with postnatal lead exposure.
15          Nine to ten year-olds in the Mexico City Prospective Study were evaluated by
16    determining relative theta activity across the scalp. These results, together with life time blood
17    lead concentrations, were assessed by multiple regression models (Poblano et al., 2001).  The
18    most significant increases in theta power were associated with blood lead levels (geometric
19    mean: 10.3 jig/dL) measured between 54 and  72 months of age.  Spatially weighted regression
20    showed that there was a significant anterior-posterior gradient in the lead-induced increase in
21    relative theta activity associated with postnatal blood leads at 54-72 and 78-96 months. These
22    lead effects occurred at an age during which relative theta power reaches its developmental
23    maximum and then starts to decrease. These data have critical implications in  understanding the
24    neurotoxic and developmental impacts of lead exposure. If theta waves continue throughout
25    childhood as one of the most dominant CNS rhythms, this could qualify as a developmental
26    disorder and an "EEG soft sign."  Stated differently, persistence of theta activity reflects an
27    'immature" EEG pattern and/or brain injury. Neuropsychological testing concurrent with this
28    electrophysiological methodology could add important information connecting direct measures
29    of theta activity to CNS development indexed  by neuropsychological  outcomes (Lidsky, 2003,
30    2005).
      December 2005                           5-48        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Using MRS, two studies (Trope et al., 2001; Meng et al., 2005) have provided new and
 2    important information. Both studies employed MRS with N-acetylaspartate (NAA).  MRS has
 3    the capability to monitor brain metabolism by  detecting NAA, a metabolite that is known to
 4    decrease during processes involving neuronal loss. Thus, this methodology provides both
 5    biochemical and anatomical information directly related to the neurotoxic effects of lead on the
 6    CNS.  Trope et al. (2001) studied 16 lead-poisoned children (mean blood lead: 39.9 |ig/dL;
 7    range: 23-69 |ig/dL) who had a mean age of 8 years, 9 months. All of these children received
 8    medical attention before 5 years of age. The latter group was compared to 5 children (blood lead
 9    levels <10 |ig/dL) who had a mean age of 8 years, 6 months. Both groups of children had
10    normal MRIs. The lead-exposed group of children had significant reductions in NAA/creatine
11    and phosphocreatine ratios in frontal gray matter compared to the nonexposed group. Review of
12    medical records in the lead-exposed group failed to reveal an alternative or contributing etiology
13    that could explain this demonstration of brain damage by MRS. These findings, in the regions of
14    the frontal lobes, which are responsible for attention, executive functions and impulse control,
15    are likely to be relevant to neurotoxic outcomes in lead exposed children, who may exhibit
16    impairments in these areas.
17          Using very similar techniques, Meng et al. (2005) evaluated 6 lead-exposed children
18    (mean blood lead: 37 |ig/dL) who lived near a lead-recycling industry for a period of at least
19    5 years.  These children had never been chelated.  The lead-exposed children were compared to
20    6 non-lead exposed children who had blood lead levels <10 |ig/dL. On the Wechsler scale, the
21    control children had a Full Scale IQ of 101 compared to 81  in the excessively exposed group.
22    MRIs were normal in both groups of children. These data parallel those reported by Trope et al.,
23    (2001). However, Meng et al. (2005) reported decreased levels of NAA in the lead-exposed
24    group in four brain regions: the left and right frontal areas and the left and right hippocampus.
25    This study also found that MRS metabolites in the lead-exposed subjects were significantly
26    reduced as compared with controls, thereby suggesting interference with neuronal functioning
27    after lead exposure.
28          Collectively, these electrophysiological and biochemical-anatomical data can assist in
29    providing an understanding of the neurotoxicity of lead and neurophysiological  outcomes in
30    relatively "low" and "higher" level childhood lead exposure. Collectively, these data also appear
      December 2005                           5-49        DRAFT-DO NOT QUOTE OR CITE

-------
 1    to provide evidence that lead interferes with the hard-wiring and differentiation of the central
 2    nervous system in children.
 3
 4    Vulnerability and Susceptibility
 5          The unique susceptibility of children to the adverse health effects of lead were recognized
 6    previously by EPA in 1986-1990. Some of these aspects included the specific behaviors of
 7    children, including their metabolism of lead, physiological considerations that separate children
 8    from adults, greater potential absorption of lead per square meter of body surface, hand-to-mouth
 9    activity, and prevalence of nutritional factors that can enhance the absorption of lead from the GI
10    tract.
11          Since 1986-1990, an enlarged database is now available to construct a somewhat wider
12    approach to understanding not only new information relating to children's susceptibility, but also
13    furthering characterizing interindividual variability as related to manifestations of lead's adverse
14    health effects in children. This section includes topics of developmental toxicology, growth and
15    development, economic status, nutritional aspects of lead and children, and, finally, genetic
16    considerations of children with possibly a biologically based genetic character interacting with
17    exterior environmental realities.  Considerations in this section also delve into risk assessment
18    focused on some child-specific factors that affect health outcomes in populations as well as
19    individual children.
20          Moreover, as a general principle of toxicology and neurotoxicology, it is recognized that a
21    variety of factors can either enhance or decrease an individual's sensitivity to toxic exposures of
22    lead. Besides individual children, there are factors that modify the selective neurotoxic
23    responses of subgroups of children.  Some of these variables that increase a child's vulnerability
24    are discussed below.
25
26    Developmental Toxicology
27          In addition to child-specific factors detailed above, it was concluded previously in 1986-
28    1990 that the critical window of adverse health effects of lead in children was at less than 3 years
29    of age, as briefly mentioned below and in greater detail in Section 6.3.  This suggestion of age
30    should be extended to children in their school-aged years to  mid-adolescence (Chen et al., 2005;
31    Ris et al., 2004; Dietrich et al., 2004) and into the adult years as well (Rice and Barone, 2000).

      December 2005                             5-50        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Developmental toxicity is usually defined by the occurrence of adverse effects on the
 2    developing organism prior to conception and/or during prenatal and postnatal development.
 3    Manifestations of developmental toxicity include death to the developing organism, structural
 4    abnormalities, impaired growth, and functional deficiencies. Moreover, developmental
 5    exposures can result in adverse health effects prenatally, postnatally, in childhood, into school-
 6    aged years, and into adult age groups to the elderly (Selevan et al, 2000; Weiss, 2000; Rice and
 7    Barone, 2000). An important concept in risk management is to identify, whenever possible,
 8    developmental windows for evaluating dose-response relationships. Moreover, in risk
 9    management, identification of critical windows is aimed at recognizing especially susceptible
10    sub-groups within the general population to provide specific interventions (Selevan et al., 2000).
11    Information on critical windows of development is needed to assess real and potential
12    environmental health risks (Weiss, 2000).
13          To protect children's health, it is necessary to understand their unique sensitivity to
14    environmental toxicants, and, to further this understanding, functions of risk and exposure must
15    be considered (Faustman et al, 2000).  Risk is defined as the probability of an adverse outcome
16    as a function of exposure and toxicity. It is evident that in development of the CNS,
17    unidirectional inhibition at one developmental stage can cause substantial alterations in
18    subsequent processes.  In addition, stages of development occur in temporally distinct time
19    frames across regions of the brain (Rice and Barone, 2000; Weiss, 2000). As a result, the CNS
20    has a very limited capability to compensate for cell loss or other injury (Rice and Barone, 2000).
21    Thus, exposure criteria should be based on information relevant to predicting risks and to
22    accounting for toxicokinetic differences that occur during different stages of development
23    (Faustman et al., 2000).
24          Characterization of critical time frames of development in children are based, in large
25    part, on the results of experimental studies discussed in Section 5.3.1.  Initially, the critical time
26    frame for adverse effects of lead on CNS development was considered to be in children <3 years
27    of age (Bellinger et al.,  1991, 1992) in that blood lead levels at 2 years of age were correlated
28    with cognitive impairments at 57 months and 10 years of age.  However, as indicated in Section
29    6.3, the age range for time windows for lead's adverse effects on the CNS has been significantly
30    extended to school-aged children, into adolescence, and into adulthood (Dietrich et al., 1993;
31    Tong et al., 1996; Wasserman et al., 2000; Canfield et al., 2003, 2004; Chen et al., 2005;

      December 2005                            5-51        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Lanphear et al., 2000, 2005). Moreover, recapitulation of synaptogenesis in the form of synaptic
 2    plasticity is modified by experience and the environment as children become adults and age into
 3    the elderly life phase (Rice and Barone, 2000). This concept provides a toxicological framework
 4    for identifying latent or persistent expressions of childhood lead exposure in adults as "growing
 5    into a lesion" (Ris et al., 2004) or magnification of an earlier insult with aging (Rice and Barone,
 6    2000).  This toxicological recognition of latent or persistent expressions of childhood exposure in
 7    adults forms  the basis of Section 5.3.2.2, discussed below.  Additional areas of concern for
 8    children related to risk assessment include consideration of lead's deleterious effects on somatic
 9    growth, socioeconomic status, nutritional correlates of lead exposure and interactions between
10    biologically inherent genetics and the external environment (discussed below).
11
12    Growth and Development
13          In the Supplement to the 1986 Addendum (EPA/600/8-89/049F), early results from
14    prospective studies in Cincinnati, Boston, Port Pirie, and Yugoslavia were noted in terms of
15    lead's effects on perinatal and postnatal growth and development. However, evidence  regarding
16    physical growth effects related to prenatal or early postnatal exposure were inconsistent.
17    Limitations in these  early  data from prospective studies included definitions of the length of
18    gestation, racial makeup, maternal age, sample sizes, and levels of lead exposure.  It appeared
19    likely that prenatal lead exposure did pose a potential hazard to the developing fetus as related to
20    reduced gestational length and possibly other aspects of fetal growth. It proved difficult,
21    however, to define a definite dose-response relationship for fetal outcomes, although there were
22    some indications that pointed to adverse effects on the fetus at blood lead levels of 10-15 |ig/dL.
23          More  recently reported data have extended  assessments of impacts of lead on early
24    postnatal outcomes (birth weight, early weight gain to 1-month of age, birth length and head
25    circumference) to  measurements of maternal bone lead post-delivery by K-XRF.  Additional
26    studies have  analyzed national data (NHANES III) in terms of the growth patterns of children 1
27    to 7 years of age.  From the same national survey,  other studies have focused on sexual
28    maturation as a function of blood lead concentrations.
29          Gonzalez-Cossio et al. (1997) assessed the possible relationship of blood lead levels in
30    cord blood and maternal bone lead to birth weight in Mexico City.  Two hundred seventy-two
31    mother-infant pairs were studied, and the  cord and maternal blood lead levels were, on average,

      December 2005                            5-52        DRAFT-DO NOT QUOTE OR CITE

-------
 1    7.1 and 8.9 |ig/dL, respectively.  Tibial lead (not patellar lead), measured 1-month postpartum by
 2    K-XRF, was the only marker related to birth weight, such that, at the highest quartile of bone
 3    lead (15.15 |ig/g bone mineral), infants were, on average, 156 grams lighter at birth.  Although
 4    these data appear to extend previously reported information from prospective studies, some
 5    caution in the interpretation of these data is indicated: 10% of tibial lead values and 13% of
 6    patellar lead levels were below the instrument's detection limit (as defined by the authors), but
 7    these negative values were included in the statistical analysis. By necessity, the study design
 8    was cross-sectional, because, for ethical reasons, bone lead measurements are precluded during
 9    pregnancy due to radiation exposure, and, for unexplained reasons, there were no statistically
10    significant relationships with blood lead values in mothers or infants.
11          Sanin et al. (2001) studied a  similar population in Mexico City comprising 329 mother-
12    infant pairs. The umbilical cord blood lead (mean) was 6.8  |ig/dL and the mean value for the
13    infants was 5.6 |ig/dL at 1-month of age. A 10 |ig/dL increase in infant blood lead levels at
14    1-month was associated with a 142  gram decrease in weight gain.  Thus, lead exposure
15    postnatally had adverse effects on early perinatal weight gain.  In addition, maternal patellar lead
16    at 1-month was negatively associated with weight gain as well.  The important finding in this
17    study was the inverse correlation between postnatal blood lead and weight gain.  However, the
18    significance of maternal patellar lead by K-XRF is limited by the large standard deviation in the
19    measurement of patellar lead (15.2 jig/g bone mineral) and a revised statistical method for
20    calculating bone leads, which was not delineated in this article.  Moreover, the failure of
21    maternal blood lead concentration to predict early or subsequent birth weight was unexplained.
22          In the third of this series of reports from Mexico City, 233 mother-infant pairs were
23    evaluated as described above, but the outcome measures were birth length and head
24    circumference. These results showed that bone lead biomarkers (tibia and patella) were
25    associated positively and significantly with maternal and umbilical cord lead. However, only
26    patellar lead was significantly and negatively associated with birth length and head
27    circumference. These associations were independent of maternal venous blood lead levels,
28    umbilical cord lead levels, and other predictors of birth  size, including birth weight. Similar
29    concerns apply here, as those above, relating to K-XRF measurements. Collectively, except for
30    the relationship between postnatal blood lead concentrations and decreases in weight gain (Sanin
      December 2005                            5-53        DRAFT-DO NOT QUOTE OR CITE

-------
 1    et al., 2001), the cumulative findings relating to perinatal and postnatal outcomes, as indexed by
 2    bone lead values, fail to provide a consistent or readily interpretable set of conclusions.
 3          Evaluation of 4391 children 1-7 years old was carried out using the nationally
 4    representative data from NHANES III. This population study included non-Hispanic white, non-
 5    Hispanic African-American, and Mexican-American children, and the outcomes measured were
 6    stature, head circumference, weight and body mass index (Ballew et al., 1999). Blood lead
 7    levels were significantly and negatively related to stature and head circumference, yielding a
 8    predicted decrease of 1.57 cm in stature and a 0.52 cm decrease in head circumference for each
 9    10 |ig/dL increase in blood lead values. There was no statistically relevant association between
10    blood lead and weight and body mass index.  These robust findings are of considerable
11    importance, because the observations are very similar to those reported earlier for NHANES II
12    by Schwartz et al. (1986). Thus, although blood leads declined  substantially in the United States
13    over two decades (NHANES II vs.NHANES III), lead exposure at considerably lower levels
14    continued to affect the growth of some children. Stated differently, there was no attenuation of
15    the negative association between blood lead levels and indices of growth in children despite a
16    substantial decrease in national blood lead values in young children. Collectively, these cross-
17    sectional national surveys (Schwartz et al., 1986; Ballew et al., 1999) indicate the following
18    negative associations with blood lead values: a 1.0-1.5 cm decrease in stature and a 0.50 cm
19    decrease in head circumference coupled to modest increases in blood lead levels within the range
20    of blood lead levels in children in the United States.  Over the past two decades, these data
21    provide the most compelling data sets related to the adverse effects of lead on growth patterns in
22    young American children.
23          Two studies, utilizing NHANES III data, have measured  pubertal development, as related
24    to blood lead concentrations, to determine whether sexual maturation may be affected by current
25    environmental lead exposure (Wu et al., 2003; Selevan et al., 2003). In the study by Wu et al.
26    (2003), pubic hair and breast development were evaluated in 1,706 8-16 year-old girls, and
27    information on menarche was delineated in 1,235 girls 10-16 years of age. The blood lead range
28    was 0.7-21.7 |ig/dL. This population was categorized into three groups, according to blood lead
29    values: 0.7-2.0, 2.1-4.9, and 5.0-21.7  |ig/dL. Sexual maturation markers were self-reported
30    attainment of menarche and physician-determined Tanner stage 2  pubic hair and breast
31    development.  The results indicated that girls who had not yet reached menarche or stage 2 pubic

      December 2005                            5-54        DRAFT-DO NOT QUOTE OR CITE

-------
 1    hair had higher blood leads than those girls who had. Negative relationships were found for
 2    blood leads with attainment of menarche and stage 2 pubic hair after adjusting for covariates; no
 3    relationships were evident for breast development.
 4          Selevan et al. (2003) studied a subset of girls (8-18 years old) from NHANES III that
 5    included 600 non-Hispanic white, 805 non-Hispanic African-American, and 781 Mexican-
 6    American girls, who, collectively, had a geometric mean blood lead level of 3  |ig/dL. For all
 7    girls who had blood lead levels of 3 |ig/dL compared to those whose blood leads were 1 |ig/dL,
 8    the higher lead group had a significant decrease in height after adjustment for confounders.
 9    Also, in the higher lead group of girls, there were significant delays in breast and pubic hair
10    development, especially in non-Hispanic African-American and Mexican-American girls. Of the
11    latter two groups, the most profound delays were  observed in non-Hispanic African-American
12    girls. The delays in reaching Tanner stages 2, 3, 4, and 5 (in all girls) was associated with those
13    whose geometric mean blood lead was 3 |ig/dL (as compared to all girls whose geometric mean
14    blood lead level was 1 |ig/dL) were 3.8, 5.3, 5.8, and 2.1 months, respectively, for breast and
15    pubic hair development. There were no significant delays found in non-Hispanic white girls.
16    These findings, within a narrow blood lead range, indicate that environmental lead exposure can
17    delay growth and pubertal development in girls.  Thus, analyses of these national surveys in
18    children 1-7 years old and in girls 8-18 years old provide strong evidence for adverse effects of
19    lead on the growth of young children and adolescent girls at blood lead concentrations
20    commonly found in the U.S. population today.
21
22    Socioeconomic Status
23          In the U.S. EPA's Supplement to the 1986 addendum (EPA/600/8-89/049F), very little
24    information was discussed relating to socioeconomic status (SES) and the vulnerability of
25    children to lead exposure and resulting deficits in cognitive skills. Primarily as a result of
26    analyses of NHANES III, the importance of SES  has reached its appropriate focus and attention.
27    Additional peer-reviewed articles have also contributed to now well-documented interactions
28    between SES and children's vulnerabilities to the neurotoxic effects of lead.
29          A child's SES clearly has an important influence on the possibility of lead exposure in
30    young children. Disadvantaged children may have an already compromised neuropsychological
31    status that is further impaired by the toxic effects  of lead. Although the exact mechanisms of the

      December 2005                            5-55        DRAFT-DO NOT QUOTE OR CITE

-------
 1    impact of SES on lead's neurotoxic effects on the central nervous system are unknown, poverty,
 2    pre-1960 housing in segregated communities, ethnicity, and nutritional deficiencies, collectively,
 3    can contribute substantially to increased vulnerability of individual children and subgroups of
 4    children.  The peer-reviewed literature, discussed in this section and in the following section,
 5    provides support for these conditions contributing to children's susceptibility to the toxic effects
 6    of lead.
 7          An analysis of the early phase (1988-1991) of NHANES III was carried out by Brody
 8    et al. (1994), in which 13,201 persons from 1 year of age through elderly adults were assessed
 9    via a multiple stage probability design.  It was found that the prevalence of elevated blood lead
10    levels for children in low-income families (16.3%) was about four times higher than the
11    prevalence for children with high family incomes. Non-Hispanic African-American children
12    from low-income families had the highest proportion of elevated blood lead values (28.4%).
13          A comparison of results from NHANES II and NHANES III (Pirkle et al., 1994) extended
14    data reported earlier (Brody et al., 1994).  From 1976 to 1991 (NHANES II vs. early NHANES
15    III) there was an overall decline in all children 1-5 years old from 15.0 to 3.2 |ig/dL (geometric
16    means).  For non-Hispanic white children, the decline was from 13.7 to 3.2 |ig/dL, whereas for
17    non-Hispanic African-American children, the decline was from 20.2 to 5.6 |ig/dL.  Income levels
18    were based upon those previously determined by the U. S.  Census Bureau: income level was
19    defined by the poverty-income ratio (PIR), so that the total family income was  divided by the
20    current poverty threshold. PIRs were divided into three categories: low (0 < PIR < 1.30; mid-
21    range (1.30 < PIR 3.0) and high (PIR > 3.00). Based upon PIRs, it is noteworthy that mean
22    blood lead levels decreased by 60%  (24.0 to 9.7) for African-American children from low-
23    income families living in central cities with populations of 1  million or more.  The latter value
24    for low-income African-American children was about 3 times the mean value for non-Hispanic
25    white children.
26          An analysis of phase 2 of NAHANES III showed that it becomes increasingly evident that
27    SES factors, including sociodemographic factors, are closely related to average blood lead
28    concentrations in young children (Pirkle et al., 1998). By phase 2, the  average blood lead in all
29    children 1-5  years old was 2.7 |ig/dL.  The prevalence of elevated blood lead levels in
30    African-American children living in pre-1946 housing was 21.9%; the prevalence in all children
31    of low-income families living in pre-1946 housing was 16.4% (demographic status was

      December 2005                           5-56        DRAFT-DO NOT QUOTE OR CITE

-------
 1    determined by U.S. Department of Agriculture codes populations according to proximity to
 2    major metropolitan areas). Low incomes among all ethnic groups, defined by PIRs, were
 3    significantly associated with higher blood leads (details of these data were not provided in the
 4    article).  It is reasonable to conclude that U.S. children, based on ethnicity, housing age, and
 5    income,  are disproportionately exposed to excessive levels of lead in their environments.
 6          SES was further considered in NHANES III (1988-1994) based upon age and blood lead
 7    concentrations (Bernard and McGheehin, 2003). Overall, 25.6% of children 1-5 years old had
 8    blood leads equal to or greater than 5 |ig/dL; but most of these children (76%) had blood leads
 9    less than 10 |ig/dL. Of those children who had blood leads greater than 5 |ig/dL, 46.8% were
10    non-Hispanic African-American compared to 18.7% non-Hispanic white children. Housing
11    status also played a significant role: 42.5% of children who had blood leads greater than 5 |ig/dL
12    lived in pre-1946 housing, 38.9% lived in housing built between 1946-1973, and 14.1% of
13    children in this blood lead group lived  in housing built after 1973.  Compared to non-Hispanic
14    white children, African-American children were 3 times more likely to have blood leads greater
15    than 5 |ig/dL, 7 times more likely to have blood leads of 10-20 |ig/dL, and 13.5 times more
16    likely to have blood leads equal to or greater than 20 |ig/dL.  Low-income families, defined by
17    PIRs, were at substantially elevated risk for having children with blood lead levels above 5
18    ng/dL, and the odds ratios in these families were the highest when comparing the 10-20 |ig/dL
19    group to those children with blood leads <5 |ig/dL.
20          Among Native American children, 1 to 6 years old, living near a Superfund site in
21    Oklahoma, strong interactions were observed between blood lead levels and poverty, suggesting
22    that poor children were especially vulnerable to the toxic effects of lead (Malcoe et al., 2002).
23    Moreover, blood lead levels  were significantly higher in 52,407 WIC-enrolled families between
24    1996-2000 compared to non-WIC-enrolled families indicating the vulnerability of children with
25    low incomes and poorer nutritional status (Zierold et al., 2004).
26          Similar findings have been reported from the Port Pirie prospective study (long et al.,
27    2000); 375 children, 11-13 years of age, were assessed by Daniel's Scale of Occupational
28    Prestige, which is a surrogate for SES.  With adjustment for confounders, Wechsler-derived IQ
29    scores were reported in three groups of children according to their lifetime blood lead values of
30    <12,  12.1-17, and >17 |ig/dL, respectively. For the less than 12 |ig/dL group, the IQ in the high
31    SES children was 105.6 vs. 103.1 in the low SES group.  In the mid-range blood lead group

      December 2005                           5-57        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (12.1-17 (ig/dL), the IQ score for the high SES group was 104.4 vs. 100.6 in the low SES group.
 2    For the higher lead group, the largest differential in IQ scores was apparent: high SES, 101.5 vs.
 3    90.9 in the low SES group. Poor children were especially vulnerable to the neurotoxic effects of
 4    lead.
 5          Familial and nonfamilial factors were discerned in 717 children, some of whom lived in
 6    Detroit (urban group-low SES) compared with a group of suburban-based middle class children
 7    (Breslau et al., 2001) who lived outside Detroit. Children were prospectively tested via the
 8    Wechsler at 6 and 11 years of age.  Although blood lead levels were not included in this study,
 9    the results are of interest.  On average, in the urban children, over time, there was a downward
10    shift of 5 IQ points in the disadvantaged children while a negligible change was found in the
11    suburban-middle class group.  Income and demographic data for the two groups of children were
12    defined by the 1990 U.S. Census data. Despite the absence of blood lead concentrations, it is
13    reasonable to suggest that compared to the suburban children (according to the 1990 U. S.
14    Census data utilized), that urban Detroit children were more likely to be exposed to lead based
15    paint in their home environments. Familial determinants of IQ, such as maternal IQ, education,
16    and marital status, exerted stable and uniform  influences on children's IQ scores across age in
17    both communities; none of these variables were associated with change in IQ scores.  Although
18    family factors (maternal IQ specifically) explained about two-thirds of the initial 14 point
19    disparity in IQ scores between urban vs. suburban  children, such factor(s) did not account for
20    any part of the IQ decline of 5  points (on average). These authors concluded that IQ is a "joint"
21    product of "genetics" and the environment.  The authors emphasized that the disadvantages of
22    inner city children, including ethnicity, housing, segregation, and educational opportunities
23    underscores the need to fully examine extrafamilial factors, including a community's economic
24    resources, to understand predictors of children's IQs. Although family and community factors
25    are not completely separable, these observations reflect, in part, the legacy and vulnerability of
26    children growing up is socioeconomically disparate communities.
27          Based upon the studies discussed above, there is conclusive evidence that SES has a
28    profound influence on children's vulnerability and susceptibility to the neurotoxic  effects of lead
29    exposure.
30
      December 2005                            5-58        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Nutrition
 2          There was little discussion of nutritional factors and their impacts on children's
 3    vulnerability to lead in earlier EPA documents, because very little, if any, information was
 4    available at that time. It has become evident that the dangers of lead exposure in children are
 5    substantially enhanced by diet deficient in calcium, iron, zinc and other essential nutrients;
 6    specific dietary deficiencies are not infrequently coupled to increased susceptibility to lead in
 7    low SES children.
 8          In 205 one-year olds, who were low SES infants and who were living in old housing,
 9    blood lead levels were measured and related to nutrient intake (primarily of fat).  This sample
10    was stratified so that excessive exposure to lead could be analyzed as an independent or
11    dependent variable to account for changes in blood lead values (Gallicchio et al., 2002).
12    Exposure to environmental lead was assessed by measurements of lead in household dust. The
13    authors reported a positive association between household dust levels and blood lead
14    concentrations, and positive associations were also found between dietary intake of total calories
15    and fat.  The latter dietary observations were found to be independent of environmental exposure
16    to lead at dust lead values above the 2001 EPA Guideline (Federal Register, 2001). These results
17    implied that dietary control of fat and  total caloric intake could have a beneficial effect on
18    children's blood lead levels independent of environmental exposure, although the authors
19    cautioned that control of the external environment was also a critical factor in modulating
20    children's blood lead levels.  Similar findings were reported by Lucas etal. (1996). The
21    relationships between blood lead levels and nutritional factors were studied in 296 children who
22    ranged in age from 9 to 72 months from low-SES families in Baltimore.  When environmental
23    lead exposure was statistically controlled, dietary fat intake had a  positive association with blood
24    lead levels, particularly in children who had blood leads >15 |ig/dL.
25          Several studies have established a negative association between dietary iron intake
26    (as well as biochemical assessments of iron status). Hammad et al. (1996) evaluated
27    299 children (9 months to 5 years of age) in a cross-sectional study design.  The mean blood lead
28    in this group was 11.4  |ig/dL, and the  mean age of the entire group of children was 26.2 months.
29    After adjustment for confounders, the  authors reported that the highest quartile of dietary iron
30    intake had the lowest blood lead values. Bradman et al. (2001) studied 319 1-5 year-old children
31    in Sacramento, CA in terms of iron status as measured by serum levels of ferritin, and

      December 2005                            5-59        DRAFT-DO NOT QUOTE OR CITE

-------
 1    environmental exposure to lead from soil and lead-based paint.  24% of this sample was iron
 2    deficient defined as a serum ferritin level <12 ng/dL. Blood lead levels were higher for each
 3    tertile of iron-deficient children who were also experiencing excessive exposure to exogenous
 4    lead; and the greatest difference between iron-deficient compared to iron-replete children
 5    (a mean difference of 3 |ig/dL) was in children who had on-going excessive exposure to lead.
 6    It was concluded that improvement of iron status, coupled to control of environmental lead
 7    sources led to a significant decline in blood lead concentrations.
 8          An important study was carried out in an urban (Boston) primary care setting of 3,650
 9    9-48 month-old children, and comparisons were made between iron status and "low-level" lead
10    exposure.  Iron deficiency was defined according to red blood cell indices, including mean
11    corpuscular volume (MCV) and red cell distribution (RDW). During the study period of 1994-
12    1996, 9.9% of the children were iron deficient, defined by cut-offs for MCV and RDW, and
13    9.4% of these children had blood lead concentrations of 10 |ig/dL or greater (Wright et al.,
14    1999).  Among lead-poisoned children, 11.6% were iron deficient.  Blood lead levels ranged
15    from less than 5 to 44 |ig/dL.  More than 50% of the children screened had blood lead
16    concentrations below 5 |ig/dL; and the median blood lead was 5 |ig/dL. Blood lead levels were
17    stratified into 3 categories: less  than 5 |ig/dL, 5-9 |ig/dL, and greater than 10 |ig/dL.  Chi-square
18    analysis showed a significant association between rising blood lead levels and iron deficiency,  as
19    previously defined. In group  comparisons, the mean ages of the patients with blood lead levels
20    less than 5 |ig/dL and 10 |ig/dL or greater differed significantly from each other, as did those of
21    patients with blood lead levels of less than 5  |ig/dL and 5-9 |ig/dL.  In contrast, there was no
22    significant difference in the mean age for patients with blood lead levels of 5-9 |ig/dL and those
23    with blood lead values of 10 |ig/dL or greater. Odds ratios were calculated based upon the
24    postulate of iron deficiency as a predictor of blood lead levels after controlling for age,
25    hemoglobin, and insurance status.  The odds ratios were 1.63 for a blood lead concentration of
26    5 |ig/dL or more and 1.44 for a blood lead concentration of 10 |ig/dL or more among iron
27    deficient children. This study concluded that the combination of increased RDW and decreased
28    MCV (markers of iron deficiency) is  associated with blood lead concentrations of more than 5
29    and more than 10 |ig/dL. Thus, this important study demonstrated that iron  deficiency is
30    associated with even lower blood lead levels than currently found in the United States.
      December 2005                            5-60       DRAFT-DO NOT QUOTE OR CITE

-------
 1    However, it is important to point out that the associations is not as strong as reported in children
 2    with more severe lead poisoning, as discussed above.
 3          In this study (Wright et al., 1999), the combined prevalence of lead poisoning and iron
 4    deficiency was present in 1.1% of the children tested. Therefore, secondary preventive measures
 5    of childhood lead poisoning, such as selective dietary interventions (iron supplementation), to
 6    reduce the intestinal absorption of lead can be simultaneously pursued in tandem with primary
 7    preventive efforts (Rosen and Mushak, 2001).
 8          Kordas et al. (2004) examined whether iron status could account in part or for all of the
 9    negative relationship between cognitive performance and lead exposure in 602 6-8 year old
10    children living near a metal foundry in Torreon, Mexico. The average blood lead level was
11    11.5 |ig/dL with a standard deviation of 6.1, and 50% of this group had blood lead levels above
12    10 |ig/dL.  The results showed that the relation between blood lead and cognition was not
13    strongly affected by nutritional status (iron and zinc), indicating that the association between
14    blood lead and cognition was not explained by the presence of iron deficiency in a relatively
15    intact group of children from the standpoint of nutrition (21.7% were iron deficient). However,
16    low serum ferritin values were more prevalent in children who had blood lead concentrations
17    above 15 |ig/dL than those below 15 |ig/dL (33.0 vs. 18.4%, p<0.001). Furthermore, successive
18    addition of iron status did not attenuate lead's negative association with several cognitive
19    outcomes.
20          Based upon these reports, it is reasonable to conclude that caloric and fat intake have
21    important effects on blood lead levels  in children.  Furthermore, based on the above evidence,
22    the inverse association between iron status and blood lead levels is clearly documented in
23    children with low-level lead exposure to more severely elevated blood lead values.
24          Limited information is available concerning effects of iron supplementation on blood lead
25    levels. Rico et al. (2005) have tested the efficacy of iron (and zinc) supplementation in 515
26    6-8 year-old children living in close proximity to the lead metal foundry in Torreon, Mexico.
27    This was a randomized, double blind, placebo-controlled study with about 125 subjects in each
28    treatment/placebo group. In addition to supplements or placebo paradigms, selective tests of
29    cognitive functioning were also administered at baseline and 6 months later.  The overall
30    prevalence of iron and zinc deficiency was 21.7 and 28.9%, respectively.  Thus, in relative terms,
31    this was a reasonably well off population from a nutritional standpoint, with a group of children

      December 2005                            5-61        DRAFT-DO NOT QUOTE OR CITE

-------
 1    many of whom had been excessively exposed to lead since birth. The mean blood lead level was
 2    11.5 |ig/dL (SD of 6.1). Cognitive improvements were not discerned in any of the 11 measures
 3    employed, and there was a very modest decrease in blood lead levels of only 2.6% (or about
 4    0.30 |ig/dL) for the iron supplemented group.  These negative findings can be explained by the
 5    relatively intact iron status in the majority of children at baseline, and the negative results of
 6    cognitive testing can be attributed to a population of children who had been excessively exposed
 7    to lead for long timeframes.  In contrast, 191 children from a community project in Costa Rica,
 8    divided into five treatment groups, had an average blood lead level of 10.98  |ig/dL (Wolf et al.,
 9    2003). Oral iron supplementation led to a mean decrease in blood lead levels of 1.2 |ig/dL over
10    3 months.  These authors concluded that iron therapy can have a substantial effect on decreasing
11    blood lead levels, particularly in children whose iron status is the most compromised and if
12    treated promptly. Interpretation of the results of this study require some degree of caution,
13    because of the limited sample size in each of the five groups.
14         Mahaffey et al. (1986) analyzed calcium intakes in comparison with blood lead levels in
15    NHANES III. The MEAN and 25th, 50th and 75th percentiles for blood lead concentrations in
16    2,926 children were, respectively, 15.7, 11,  14, and 19 |ig/dL, Corresponding dietary calcium
17    intakes were, respectively, 851, 522, 789, and 1,110 mg/day. Dietary calcium intake was a
18    significant explanatory variable for blood lead, and this relationship was inverse.  Thus, in this
19    national survey, a significant and independent inverse association was observed between dietary
20    calcium intake, assessed by the 24-h recall method, and blood lead levels.
21         One hundred sixty-nine Albany-based mother-infant pairs were  evaluated every 3 months
22    during the first year of life according to calcium intakes measured by 24-h recall (Schell et al.,
23    2004). The geometric mean value for blood leads in infants at birth was 1.6 |ig/dL; this value
24    rose to 5.1 |ig/dL by 12 months of age, when 18% of the sample had elevated blood lead levels
25    (Schell et al., 2004). A significant inverse relationship between calcium intake and blood lead
26    values was found at 6 months;  but only the inverse relationship with iron and blood leads
27    persisted to 12 months.  The majority of infants in this study met the recommended daily
28    allowances for calcium.
29         Recent studies that  have assessed the impact of calcium on blood lead levels have yielded
30    reasonably consistent results. Sargent et al.  (1999) studied 103 children 3.5 to 6 months of age;
31    these infants were followed for 9 months, receiving either no treatment of calcium supplements

      December 2005                            5-62       DRAFT-DO NOT QUOTE OR CITE

-------
 1    or treatment to increase daily calcium intake from about 450 mg/day to a supplemented intake of
 2    about 1700 mg/day. Through 4 months of supplementation, the median increase in blood leads
 3    was 57% compared to the control group.  However, beyond 4 months of treatment, the effect on
 4    calcium was attenuated.  Up to that time, coupled to measurements of household dust lead,
 5    calcium supplementation appeared to impair the absorption of lead from the GI tract; but this
 6    apparent effect was not sustained.
 7          In a randomized, double blind, placebo-controlled study, 67 children (1-6 years old),
 8    whose blood lead values ranged from 10-45 |ig/dL, were given a placebo or a calcium
 9    supplement to reach a daily intake of about 1800 mg/day. The mean blood lead levels at baseline
10    were 21.4 and 20.7 |ig/dL in the placebo and treatment groups, respectively.  All children in this
11    study, from an inner-city group of children in the Bronx, NY, were at or above the daily
12    recommended allowance (RDA) for calcium. Blood lead levels declined similarly in placebo vs.
13    treated groups over the 3-month study period without a  differential in final blood lead
14    concentrations between the two groups. It appears that there is a negative association between
15    blood lead levels and calcium intake, particularly, in children who fall below the RDA for
16    calcium. However, even in inner-city children, current calcium intakes appear to readily meet
17    expected RDAs, most likely accounting for the failure of calcium supplements to have effects on
18    blood lead values.
19
20    Genetic Polymorphisms
21          A paucity of information was previously available in EPA's documents in the time frame
22    of 1986-1990. Since that time, at least three genes have been identified that may affect the
23    accumulation and toxicokinetics of lead in children and adults.  The three genes are ALAD, the
24    vitamin D receptor gene (VDR), and the hemochromatosis gene (HFE). Relatively few studies
25    relating to genetic polymorphisms have been reported in children compared to a substantial body
26    of clinical research studies reported especially in excessively exposed adults. ALAD, VDR and
27    HFE are discussed here in detail to serve as an introduction for clinical research reports in adults.
28          The primary importance of incorporating a discussion of genetic polymorphisms in the
29    field of environmental health is their usefulness in detecting differences in levels of risk within
30    specific populations (Kelada et al, 2003). The range of responses to toxic environmental
31    exposures can vary, and population attributable risk may be substantial. Furthermore,

      December 2005                            5-63        DRAFT-DO NOT QUOTE OR CITE

-------
 1    understanding the possible role of genetic polymorphisms in risk assessment can lead to an
 2    enhanced delineation of mechanisms underlying toxic exposures.
 3          The ALAD gene (chromosome 9q34) encodes for ALAD, which catalyzes the second step
 4    of heme synthesis and is polymorphic.  This polymorphism yields two codominant alleles,
 5    ALAD-1 and ALAD-2, and these have been differentially implicated in some clinical research
 6    studies to lead toxicity (Kelada et al., 2001). It is evident that genotypic frequencies differ by
 7    ethnicity and geography; and these considerations require careful assessment in the interpretation
 8    of research results.  It has been suggested in some studies that ALAD-2 may possibly offer some
 9    level of "resistance" to the toxic effects of lead by generating a protein that avidly binds to lead,
10    perhaps  sequestering lead from its toxic expressions at various tissue sites. Other studies suggest
11    that the rarer ALAD-2  allele has been associated with higher blood lead levels and may, thereby
12    increase the risk of lead toxicity by producing a protein that binds more tightly than the ALAD-1
13    protein.  Some recent studies in adults have reported that individuals homozygous for the
14    ALAD-1 allele have higher cortical bone lead concentrations and may be at higher risk for long-
15    term adverse effects of lead. Occupationally exposed adults have been most frequently studied
16    in terms of the possible interaction of ALAD polymorphism and adverse health outcomes. As
17    discussed below, reports in children concerning ALAD polymorphism and risk assessment are
18    limited.
19          The vitamin D receptor (VDR) is a ligand-activated transcription factor that modulates the
20    genomic effects of the vitamin D hormone,  1,25-dihydroxyvitamin D, in a wide variety of
21    tissues.  The gene encoding for VDR is on chromosome 12q and has common allelic variants
22    (Zmuda  et al., 2000). The allelic variants and their halotypes have been extensively studied with
23    regard to osteoporosis susceptibility. Studies involving other disease states, such as breast and
24    prostate  cancer, diabetes, coronary artery disease, and primary hyperparathyroidism, have also
25    focused  on the role(s) of VDR gene variants.  Consideration of VDR gene variants have also
26    been extended to populations with increased lead exposure, particularly within an occupational
27    setting.  Very little information is available  on these gene variants and lead exposure in the
28    pediatric age group.
29          Hereditary hemochromatosis (HHC) is an autosomal recessive disorder of iron
30    metabolism characterized by an increase in  iron absorption  and deposition in the liver, heart,
31    pancreas, joints, and pituitary gland. HFE, the gene for HHC, has been mapped to the short arm

      December 2005                            5-64         DRAFT-DO NOT QUOTE OR CITE

-------
 1    of chromosome 6 (Hanson et al., 2001).  Two of the 37 allelic variants of HFE, described to date,
 2    C282Y and H63D have been significantly correlated with HHC. Homozygosity for the C282Y
 3    mutation has been found in the majority of patients and their probands diagnosed with HHC.
 4    Implications of HFE polymorphism have been proposed in studies of adults excessively exposed
 5    to lead, particularly in occupational settings.  No studies of HFE have been reported in children
 6    with varying blood lead concentrations.
 7          As yet, studies have failed to evaluate arylsulfatase (ASA) polymorphisms in lead
 8    exposed children and adults.  ASA is recognized as playing a significant role in regions of the
 9    brain known to be affected by lead, and it has been established in experimental studies that lead
10    produces low levels of ASA at sensitive stages of nervous system development (Poretz et al.,
11    2000). Studies of ASA in children and adults may yield important information that may explain
12    some of the neurocognitve effects of lead in pediatric and adult populations.  As yet, no studies
13    of this nature are available.
14          A group of 142 lead-poisoned children (mean blood lead: 27.1 |ig/dL; SD:  15.2) in New
15    York City children who expressed the 2-2 or 1-2 isozyme phenotype were reported to have blood
16    lead levels 9-11  |ig/dL higher than children who were homozygous for the ALAD-1 allele
17    (Wetmur et al., 1991). These authors suggested the possibility that, because the ALAD-2
18    polypeptide binds lead more effectively, these individuals may be more susceptible to lead
19    poisoning. At the time of publication, the lead binding properties of purified ALADl-1 and 2-2
20    proteins and tissue distribution of these alleles were unknown.
21          The relationship was investigated between ALAD isozymes and blood lead levels in
22    229 Chinese children within the age range of 6-10 years old (Shen et al., 2000).  The mean blood
23    lead value was 10.3 |ig/dL (SD:  3.3) and for the 92% of children homozygous for ALAD-1, the
24    mean blood lead was 9.7 |ig/dL compared with the 8% of children who were heterozygous
25    (ALAD-1-2) and who had a mean blood lead level of 11.7 |ig/dL (p<0.05). Using step-wise
26    multiple regression, children  who had the ALAD-2 allele were shown to be more likely to have
27    higher blood leads compared to children who had the ALAD-1 allele.
28          In the only published article to date, environmental samples, blood lead levels,  and
29    nutritional factors were assessed together with determinations of VDR-Fokl genotype (Haynes
30    et al., 2003). A significant interaction was found between dust lead, such that at a 1 |ig/ft2
31    increase in floor dust lead, children with VDR-FF genotype had a 1.1% increase in blood lead;

      December 2005                           5-65       DRAFT-DO NOT QUOTE OR CITE

-------
 1    VDR-Ff, a 0.53% increase; and VDR-ff; a 3.8% increase.  At floor dust levels less than
 2    10 |ig/ft2, children with VDR-ff had the lowest blood lead concentrations. It is noteworthy that
 3    only 17 children in this study were homozygous for the ff allele.  Nonetheless, the authors
 4    suggested that VDR-Fokl is an effect modifier for the relationship of floor dust lead exposure
 5    and blood lead concentrations.
 6          The implications for risk assessment and health significance in these three pediatric
 7    studies are limited. Far more detailed studies in this area of investigation are needed before any
 8    firm conclusions  can be reached.
 9
10    Dose-Response Paradigms
11          The aim of this discussion is to bridge the gap between basic neurotoxicology findings
12    assessed in Section 5.3.1  and the neurobehavioral consequences of lead discussed in Section 6.3.
13          Based upon current neurotoxicological studies in vivo and in vitro (Section 5.3.1) and
14    based upon epidemiological studies of children (Section 6.3), it is biologically implausible that
15    neurotoxic effects of lead do not occur at blood lead concentrations in children above and below
16    10 |ig/dL.  Although it is difficult to extrapolate from experimental studies to investigations in
17    children (Manton et al., 2001), some examples are revealing.
18          In experimental systems, lead dose is typically employed in molar concentrations
19    (Section 5.3.1); 10 |ig/dL of lead in whole blood of children is  equivalent to a molar
20    concentration of 0.48 jiM. In vitro studies have reported effects of lead on cellular regulatory
21    systems in neurons (and other tissues) far below 0.48 jiM. Whereas most or all of the lead used
22    in neurochemistry experimental systems can participate in a reaction, only a small fraction of
23    circulating lead in blood enters specific metabolic pathways. It is recognized that the major
24    portion of lead in whole blood is carried by erythrocytes and that the most accessible fraction of
25    circulating lead to other tissues is in plasma.  It has been estimated that 0.24 to 0.29% of lead in
26    whole blood is in plasma (Smith et al., 2002) and that the concentration in CSF is about half of
27    the plasma concentration (Manton and Cook, 1984). Thus, a calculation of the latter
28    concentrations  in plasma and CSF in 10  |ig/dL of whole blood  yields concentrations in the low
29    nanomolar range.
30          It is very unlikely that the plasma concentration of lead in the low nanomolar range is the
31    "dose" that impacts upon the CNS of children at blood lead levels less than 10 |ig/dL. The

      December 2005                            5-66        DRAFT-DO NOT QUOTE OR CITE

-------
 1    "dose" that perturbs the central nervous system of children at "low" blood lead levels is likely to
 2    be much higher. Lead's half-lives in various tissues is a function of the site of deposition and
 3    degree of on-going exposure.  In blood (absent excessive external exposure), the half-life is
 4    about 30 days and in brain, the half-life is about 2 years (Leggett, 1993).  However, in the
 5    absence of on-going external exposure, blood lead levels can remain elevated for relatively
 6    extended periods of time due to mobilization from internal stores (Roberts et al., 2001; Manton
 7    et al., 2000). As a result, lead, which readily penetrates the blood-brain barrier,  can continuously
 8    enter neural tissue from the blood compartment (Leggett,  1993).
 9          Active transport mechanisms are also important to  consider, and these mechanisms cause
10    differential concentrations of lead in the systemic circulation compared to those in neuronal
11    compartments. Metabolic pumps increase concentrations of ions within intracellular organelles
12    to levels that exceed those in the cytosol.  Mechanisms affecting Ca2+ distribution are the most
13    critical (Section 5.3.1).  Lead's toxic effects in the brain and other tissues are based, in large part,
14    on its ability to "mimic" Ca2+ in intracellular processes, coupled to its actions to perturb the Ca2+
15    messenger system (Schanne et al., 1989; Lidsky and Schneider, 2003).  For example, lead enters
16    neurons and glia by channels that, under physiological conditions, permit the passage of Ca2+
17    (Kerper and Hinkle, 1997; Legare et al., 1998). Lead enters and damages mitochondria via
18    cellular mechanisms that bring calcium into this organelle (Chavez et al.,  1987).  Thus, transport
19    mechanisms bring about variations in local concentrations of Ca2+ and, presumably, lead as well
20    (Schanne etal.,  1989).
21          Based upon these considerations, it is concluded that brain cells in children are likely
22    exposed to concentrations of lead, in the context of "low" blood lead levels, in the mid-
23    nanomolar range and possibly higher, particularly in organelles that depend upon the calcium
24    messenger system for their physiological activities.  The experimental literature clearly
25    demonstrates perturbations in fundamental cellular processes in the nanomolar range and
26    considerably lower (Section 5.3.1).
27          The shape of the dose-response curve(s) of IQ and  blood lead concentrations in children
28    below blood lead levels of 10  |ig/dL may be considered to be unexpected. However, the above
29    considerations provide a different and reasonable explanation. It is accepted that lead achieves
30    its neurotoxic effects on multi-neuronal targets, and the threshold concentrations of lead to
31    perturb multiple CNS targets differ by orders of magnitude (Lidsky and Schneider, 2003).

      December 2005                            5-67        DRAFT-DO NOT QUOTE OR CITE

-------
 1    For instance, second messenger systems are affected at picomolar to nanomolar concentrations
 2    (Schanne et al., 1989; Lidsky and Schneider, 2003).  These perturbations can mediate a variety
 3    of toxic consequences of lead by perturbing the temporal and spatial resolution of Ca2+. As a
 4    result, several loci within the complex Ca2+ messenger system may be impaired, thereby
 5    explaining toxic effects of lead on multiple cellular processes affecting brain functioning at
 6    differential intracellular concentrations.  At somewhat higher lead concentrations in the
 7    circulation, other critical subcellular processes will be affected and impaired (e.g., heme
 8    synthesis and cellular energy metabolism). Thus, based upon various targets in the CNS  affected
 9    by lead at widely different concentrations, the dose-response curve would tend to be steeper at
10    lower lead "doses," as may be seen when inspecting the relationships between blood lead and IQ,
11    particularly at blood lead levels less than 10 |ig/dL.
12
13    Neuro-Epidemiological Studies:  Implications for Individual Children
14          Bellinger (2004) pointed out that the clinically evident cognitive outcomes applicable to
15    attributing specific neurobehavioral outcomes to childhood lead poisoning differ from those that
16    are employed more typically to characterize risk in a population of children. The latter type of
17    epidemiological studies have been applied to setting public health standards in children by the
18    EPA and the U.S. Centers for Disease Control and Prevention. However, the clinical
19    presentation of and ultimately the diagnosis of cognitive outcomes caused by excessive exposure
20    to lead in individual children have received little attention.  Moreover, the clinical presentation of
21    lead exposure in the individual child cannot be clearly recognized or ascertained from
22    epidemiological data (Lidsky and Schneider, 2005).
23          The majority of risk assessment studies have reported the averaged performance of large
24    cohorts of children on a traditional IQ test as the neurobehavioral outcome measure.  Such
25    studies, after adjustment for appropriate  confounders, have consistently reported an inverse
26    correlation between blood lead concentrations and IQ scores (Schwartz,  1994). In addition to IQ
27    as the outcome index, children who have elevated blood lead levels have been shown to lack
28    skills in basic academic subjects (Needleman et al., 1990; Fergusson et al., 1997; Lanphear et al.,
29    2000).  As adolescents, such children are at risk for anti-social behavior  (Dietrich et al., 2001;
30    Needleman et al., 2002). These reports indicate that the outcomes in a lead-exposed child focus
31    on impairments in intellectual achievement, academic performance, and problematic behavior.

      December 2005                            5-68        DRAFT-DO NOT QUOTE OR CITE

-------
 1    These nonspecific outcomes are of extremely limited diagnostic utility for a pediatrician to
 2    understand what measurable outcomes may or may not be attributable to childhood lead
 3    exposure in an individual child.
 4          Because lead has neurotoxic effects on a child's developing brain (Bressler et al., 1999;
 5    Lidsky and Schneider, 2003; Finkelstein et al., 1998), diagnostic methods are necessary to
 6    uncover manifestations of brain  dysfunction.  IQ tests were not designed to evaluate brain
 7    dysfunction; IQ tests are insensitive to the symptoms of brain dysfunction resulting from brain
 8    injury (Lezak,  1995). Manifestations of brain injury are manifested by highly specific aspects
 9    of impaired functions that involve language, memory, and executive skills (Lezak, 1995).
10    In contrast, IQ is an aggregate, based on the summed performance of several sub-tests that assess
11    an array of cognitive functions, and this array fails to tap into focal deficits that are the stigmata
12    of brain injury. Given the lack of sensitivity of IQ scores to assess the presence of brain damage,
13    the generally consistent findings of lead's adverse affects on IQ reflect the robustness of the
14    reported data (Hill, 1965; Chen et al., 2005; Lanphear, 2005). Whereas the mean IQ in a large
15    group of children has often shown a decrease as a result of brain damage, the size of the decrease
16    fails to reflect the failures of a child's abilities to carry out daily living activities, which are
17    typically brought to the attention of a pediatrician for treatment and management.
18          Neuropsychology is an applied science focused on the neurobehavioral manifestations of
19    brain dysfunction (Lezak, 1995). Neuropsychological test batteries focus on testing paradigms
20    that are controlled by specific neural systems to detect functional effects of brain injury. Several
21    studies (noted above) have reported impairments in groups of children that have carried out
22    neuropsychological tests of fine  motor skills,  executive abilities, language, and aspects of
23    learning and memory (Bellinger et al., 1994; Faust and Brown, 1987; Stiles and Bellinger, 1993;
24    Dietrich et al.,  1992; Walkowiak et al., 1998; Wasserman et al.,  2000; Campbell et al., 2000;
25    Winneke and Kramer, 1997; Canfield et al., 2004; Ris et al., 2004).
26          Because diffuse  neurocognitive "dulling" is not a typical outcome of childhood lead
27    exposure and because a specific pattern of cognitive deficits ("signature injury or injuries") is not
28    apparent in individual children, the clinical-pediatric presentation is specific to each individual
29    child.  Thus, a  child's specific deficits evidenced by neuropsychological testing are of little
30    assistance in making a clinical diagnosis of past or present exposure to lead and the lack of a
      December 2005                            5-69        DRAFT-DO NOT QUOTE OR CITE

-------
 1    neurobehavioral "signature" is common to other neurotoxic agents that can cause brain injury
 2    (Hartman, 1995).
 3          Neuropsychological testing within a clinical framework is designed to measure cognitive
 4    and behavioral manifestations of normal and abnormal brain function to arrive at a diagnosis of
 5    brain injury, when present. Decisions arriving at evidence for abnormality are based on a pattern
 6    of test results tapping specific neural systems, with the understanding that some systems will be
 7    affected and diminished as a result of brain injury whereas others will be unaffected (Lezak,
 8    1995; Lidsky and Schneider, 2003, 2005).
 9          From this discussion, it is reasonable to conclude that neuropsychological assessments
10    provide additional and important information to the clinical understanding of an individual child
11    compared to what a pediatrician and neuropsychologist can ascertain from epidemiological data.
12    Clinical neuropsychological evaluations can lead to an etiological conclusion, together with a
13    pediatrician's differential diagnosis, whether a child's cognitive deficits are typical of brain
14    injury and whether, if present, that injury can be diagnostically attributed to lead exposure.
15    When impairments are detected, it is then the task of the pediatrician to carry out a physical
16    examination and to review medical records, radiographs, laboratory data, and environmental -
17    exposure information. Based upon review of all this information and a differential diagnosis to
18    rule out other causes of brain damage, a clinical determination  can be made as to the etiology of
19    an individual child's impairments and whether such deficits can be the result of lead exposure.
20    Once alternative or contributing etiologies have been ruled out as the cause of brain damage, a
21    diagnosis can be made causally linking lead exposure to brain damage in an individual child.
22    This describes the collaborative roles of the neuropsychologist and pediatrician in determining
23    the role of lead as the etiological factor (or not) in producing manifestations of brain damage in
24    the context of different patterns of neuropsychological deficits  in each individual  child.
25
26    5.3.2.2  Clinical Manifestations  in Adults with Childhood Lead Poisoning
27          It is reasonable to conclude from the studies discussed in this section that clinical
28    manifestations become manifest in adults as persistent or latent consequences of earlier
29    childhood lead poisoning.  Specific effects of lead in this section include impairments in
30    cognitive abilities that directly involve the central nervous system (White et al., 1993).  These
31    data have been applied to cognitive outcomes (White et al., 1993) and mortality rates in adults

      December 2005                            5-70        DRAFT-DO NOT QUOTE OR CITE

-------
 1    following severe childhood lead poisoning (McDonald et al., 1996).  Data from these analyses
 2    also indicate the presence of long-term latent and/or persistent effects on blood pressure in adults
 3    several decades after severe childhood lead poisoning (Hu, 1991). These data have been
 4    extended to more recent studies of lead's impacts on adults from early excessive childhood
 5    exposure, in terms of adverse health impacts on the central and peripheral nervous systems.
 6    With current analytical techniques, these data have been applied and connected to bone lead
 7    concentrations, as well as to the development of hypertension (Stokes et al., 1998; Gerr et al.,
 8    2002).
 9          This section includes new concepts of health impacts of lead on adults from lead exposure
10    during childhood, concepts that were not expressed in the previous 1986 EPA Lead
11    AQCD/Addendum and the 1990  Supplement to that Addendum.
12          White et al. (1993) evaluated cognitive functioning in 33 adults (mean age of 54 years),
13    all of whom had been admitted to Boston's Children's Hospital during 1930-1942.  Because
14    blood lead measurements were not available then, criteria for the diagnosis of lead poisoning
15    included: (1) lead paint exposure and pica; (2) signs and symptoms of childhood lead poisoning
16    (i.e.,  abdominal pain, vomiting, constipation, anorexia, irritability.) A latter subgroup of
17    27 adults was considered to have the mildest lead poisoning. The second and third groups had
18    more severe central nervous system symptoms of "nerve palsy" (n = 3) and encephalopathy
19    (n = 3) as well as (3) positive lead lines on skeletal radiographs. The 33 retrieved adults from the
20    Boston area were generally characterized as to the severity of their childhood lead poisoning into
21    three groups (according to the above symptoms) and according to blood lead concentrations
22    estimated as 60-100, 90-120, and greater than 120 |ig/dL, respectively. Each adult underwent a
23    90-m neuropsychological test battery.  Compared to matched controls, the 33 adults evidenced
24    widespread cognitive deficits in attention, memory, reasoning, motor speed, visual-spatial-
25    constructional skills, and coordination; previously leaded subjects were lower (compared to
26    controls) in lifetime occupational status. These observations were consistent with the onset of
27    brain damage as children with persistence 50 years later.  Exposure of the CNS during their adult
28    years could also have occurred from release of lead from bone stores (Tsaih et al., 2001).
29          This is the first retrospective report that systematically addressed cognitive outcomes in
30    adults from childhood lead poisoning.  Nonetheless, these data are limited by their observational
      December 2005                           5-71        DRAFT-DO NOT QUOTE OR CITE

-------
 1    context, lack of blood lead measurements, the long interval between childhood to the point of
 2    study as adults, the limited number of subjects, and the retrospective nature of the design.
 3          McDonald and Potter (1996) assessed ratios of observed (O)/expected (E) deaths in a
 4    cohort of 454 adults admitted as lead-poisoned children to Boston Children's Hospital from 1923
 5    to 1966 and traced through December, 1991. These are the only such data reported. As children,
 6    the criteria for lead poisoning was based upon the following:  (1) a history of "paint pica" or
 7    other sources of exposure; (2) positive bone radiographs for lead lines; and (3) GI, neurologic
 8    and/or hematologic signs and symptoms.  Seventy-six percent of this group met all three criteria
 9    and 24% met at least two out of the three criteria for diagnostic inclusion.  Data were adjusted
10    for confounders such as age, sex, ethnicity, and calendar period but not for socioeconomic status.
11    As noted, observed deaths were compared to expected deaths; and O/E ratios were computed for
12    hematological deaths (O/E = 9.7), for seizure disorder deaths  (O/E = 5.0), for cardiovascular
13    disease deaths (O/E = 2.1), and for cerebrovascular disease deaths (O/E = 5.5).  This unique
14    study also has limitations. It was retrospective in design, deaths may have been underestimated,
15    there was an excess of cases dating back to the 1930s, and blood lead measurements did not
16    begin at Boston Children's Hospital until 1963. Moreover, 153 of the original cohort of 454
17    were lost to follow-up. However, the authors pointed out that (1) blood lead levels measured
18    post-1963 were generally consistent with the classification of the severity of lead poisoning pre-
19    1963; (2) although 153 of the original group were lost to follow-up, the remaining cohort was
20    followed for a total period of 29.5 years; (3) if deaths were missed, this would have artificially
21    lowered the observed O/E ratios; (4) interpretation of these results could be limited by the
22    relatively small number of deaths; but, for each of the mortality outcomes, less than one death
23    was expected. Overall, in this cohort, mortality from all causes was about 70% higher than
24    expected.
25          Collectively, although the studies reported by White et al. (1993) and McDonald and
26    Potter (1996) have limitations, these are the first reported data to indicate that severe lead
27    poisoning causes brain damage and impacts on mortality in adults from childhood lead
28    poisoning.
29          Recent reports, utilizing current methodologies, have extended the above  data relating to
30    cognitive outcomes (Stokes et al., 1998), as well as hypertension (Gerr et al., 2002) in a cohort of
31    257 adults (19-29 years old) who had childhood lead poisoning at 9 months to 9  years of age

      December 2005                            5-72         DRAFT-DO NOT QUOTE OR CITE

-------
 1    from lead smelters in Idaho's Silver Valley. In 1974-1975, the mean blood lead level in young
 2    children at each of five towns near the smelter activities was in the range of 40 to 65 |ig/dL and
 3    the standard deviations of the blood lead levels in the five towns ranged from 13.5 to 28 |ig/dL
 4    (Gerr et al., 2002). Of the 257  adults excessively exposed as children, 43 individual blood lead
 5    values were traced back to 1974-1975, and the mean level was 49 |ig/dL. The referent cohort
 6    was in the Spokane, WA area.  The exposed and nonexposed groups were compared in terms of
 7    electrophysiological and neuropsychological testing, and the latter results were evaluated with
 8    concurrent K-XRF tibial lead measurements.
 9          Fine motor and cognitive outcomes in the exposure group, after adjustment for
10    confounding, were significantly associated with poorer performance on hand-eye coordination,
11    reaction time, trails B, symbol  digit, serial digit learning, Raven progressive matrices, and
12    vocabulary tests.  The estimated effect of being in the exposed group was negative for all 12 of
13    the motor and cognitive outcomes. Among tests of peripheral nerve function, vibrotactile
14    thresholds of the fingers and standing steadiness were significantly different between the
15    exposed and nonexposed groups; sural sensory amplitude and peroneal motor amplitude were
16    significantly related to the exposure group. Tibial bone lead measurements failed to reach a
17    p value <.05 in any of the test paradigms, although there was a trend towards significance in
18    vocabulary and vibrotactile thresholds for fingers and toes in the exposed group.  This apparent
19    insensitivity of bone lead measurements to various outcome measures was probably related to
20    the modest precision of K-XRF determinations.
21          Based on this study,  it is reasonable to conclude that excessive accumulation of lead in
22    childhood has latent and/or persistent adverse health effects on both the peripheral and central
23    nervous systems of adults assessed 19-29 years later. The latter report, using currently available
24    methods, is generally consistent with the earlier study by White et al. (1993).  Information is
25    needed in less severely exposed children followed longitudinally into adolescence and the adult
26    age group.
27
28    5.3.2.3  Adults with Ambient Exposures to Lead
29          In the previous 1986 EPA AQCD/Addendum, the focus was on adverse health effects in
30    adults at blood lead levels in the range of 30-50 |ig/dL. The studies reviewed focused on slowed
31    nerve conduction velocities, altered testicular function, reduced Hg production, and other signs

      December 2005                           5-73        DRAFT-DO NOT QUOTE OR CITE

-------
 1    of impaired heme synthesis evident at somewhat lower blood lead levels. These effects pointed
 2    to a generalized impairment of normal physiological functioning as adult lead levels exceeded
 3    30-40 |ig/dL. The lowest observed effect levels of 15-30 |ig/dL were related to impairments in
 4    heme synthesis.  In contrast, in the 1990 Supplement to the 1986 Addendum, it was concluded
 5    that the relationship between lead and blood pressure held across a wide range of blood lead
 6    values, possibly extending down to 7 |ig/dL for middle-aged men. In brief, except for effects of
 7    lead on heme synthesis down to adult blood lead values of about 15 |ig/dL, EPA's emphasis was
 8    on adverse health effects in the 30-40-50 ng/dL blood lead range (1986-1990).
 9          Since that time, studies have shown lead's effects in terms of biomarkers and indices of
10    vulnerability and susceptibility in adult populations with blood lead concentrations, on average,
11    less than 10 |ig/dL.  The number and strength of these studies are limited (see below).  Several of
12    these recent studies have also included K-XRF measurements of lead in bone which should be
13    cautiously interpreted.
14
15    Biochemical Biomarkers
16         Plasma total homocysteine (tHcy) is recognized as an independent risk factor for
17    atherosclerosis and cardiovascular disease and has both environmental and genetic risk factors.
18    Homocysteine is an intermediate metabolite in the trans-sulfation pathway that converts
19    methionine to cysteine. Moreover, the addition to homocysteine of serine by the pyridoxal
20    phosphate-dependent enzyme cystathionine Peta-synthase produces cystathionine, which in turn
21    is converted to cysteine by cystathionine gamma-lyase (CTH) (Mudd et al., 1995).
22    Homocysteine can either be methylated back to methionine by the enzyme methionine synthase
23    (MTH) or can undergo trans-sulfuration to produce cystathionine (Mudd et al., 1995).  As a
24    result, many enzymes can affect plasma tHcy concentrations, and each enzyme can identify a
25    potential candidate gene for evaluating the genetic determinants of plasma tHcy. Of common
26    gene variants, the thermolabile variant in MTHFR encoding methylenetetrahydrofolate has been
27    associated with elevated plasma tHcy. This and other variants have been connected to the
28    disease states noted above (Weisberg et  al., 2003). In this regard, a study of 496 Caucasian
29    adults found that common variants in CTH can be a determinant of plasma tHct levels (Wang
30    et al., 2004).
     December 2005                            5-74       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In view of these interactions and the associations of plasma tHcy with environmental
 2    factors, cardiovascular disease and cognitive dysfunction, Schafer et al. (2005) evaluated the
 3    possible relationship between blood lead levels, tibia lead (by K-XRF), and tHcy in a
 4    longitudinal study carried out within the context of the Baltimore Memory Study. In this study,
 5    1,140 randomly selected adults were assessed. They had a mean age of 59.3 years, an average
 6    (SD) blood lead level of 3.5 (2.4) |ig/dL, and a mean (SD) tibia lead (jig of Pb/g bone mineral)
 7    concentration of 18.9 (12.5). After adjustment for age, sex, ethnicity, educational level, and
 8    tobacco and alcohol use, plasma tHcy levels were found to have increased 0.35  jimol/L per
 9    1 |ig/dL increase in blood lead concentration. No relationship was found between plasma tHcy
10    and tibia lead levels, perhaps because of the wide standard deviation among the subjects and the
11    modest precision of the K-XRF methodology.  At blood lead levels, on average, under 10 |ig/dL,
12    these results provide some initial evidence suggesting that tHcy could be a mechanism
13    underlying lead effects on the cardiovascular and central nervous systems. Whether lead directly
14    elevates plasma tHcy, whether lead kinetics may be modified by tHcy, and/or whether one of
15    homocysteine's polymorphic variants may have specific binding properties for lead are all open
16    questions for which further investigation is required.
17
18    Vulnerability and Susceptibility
19    Socioeconomic Status
20          There was very little information on socioeconomic status (SES) in ambiently exposed
21    adults  in previous EPA Documents (1986-1990). Although some data have been published since
22    1990, it is limited to investigations of a female population in Mexico City (Farias et al., 1996), of
23    male populations from the Normative Aging Study in Boston (Elreedy et al., 1999), and of a
24    minority group of men in the Boston area (Lin et al., 2004).
25          Determinants of blood lead levels were evaluated in 513 pregnant women in Mexico City:
26    one group of women was enrolled from a public general hospital, and was considered to be low
27    SES. The second group, a high-SES cohort, was enrolled from a private hospital. The geometric
28    mean blood lead values were 6.6 and 11.12 |ig/dL from the high and low SES groups,
29    respectively (Farias et al., 1996). The entire population of pregnant women was enrolled in this
30    study during  January 1994 to August 1995 and, beside different exposure paradigms, seasonality
31    played an important role in differentiating blood lead concentrations between the high and low

      December 2005                            5-75       DRAFT-DO NOT QUOTE OR CITE

-------
 1    SES groups.  The primary determining factor for blood lead levels in the low SES population
 2    was the use of lead-glazed ceramics in women from the public hospital; and seasonality was the
 3    main factor influencing blood lead levels in the women from the private hospital. A predictive
 4    model, fitted to milk consumption, dietary supplements of calcium plus gestational age, was
 5    predictive of a 14 |ig/dL difference between the best and worst scenarios in women from the
 6    public hospital.  Seasonal differences in blood lead concentrations, which ranged, on average,
 7    from 4.7 to 12.7 |ig/dL, from summer to winter, respectively, in the high SES-private hospital-
 8    based women, focused on airborne lead as their primary source of exposure, although
 9    measurements of air lead levels were not reported.
10          Elreedy et al. (1999) investigated various factors related to SES  in 538 white males (ages
11    50-92) in the Normative Aging Study or Boston-based adults.  Questionnaire data were collected
12    regarding educational and occupational status, and these data were further analyzed using 1990
13    Census Block Group Data.  Men who had four years of college, compared to others who did not
14    graduate from high school, had, on  average, lower bone lead levels. These data suggested the
15    possibility of individual SES as having an affect on cumulative lead exposure.  Detailed
16    information on the health status of these two groups of Boston men was not provided.
17          Eighty-four minority individuals living in the Boston area were compared  by bone lead
18    measurements to previously studied Caucasian subjects: the mean values for blood lead (SD),
19    tibia lead (SD) and patella lead (SD)  for the minority group of males were 3.0  |ig/dL, 11.9 jig/g
20    (11),  and 14.9 jig/g (15.3), respectively.  These results suggest disparities in body burdens of
21    lead in the minority group of men, particularly in those older than 55 years of age. However, the
22    high standard deviations in the bone lead data, the modest precision of the utilized K-XRF
23    system, and lack of information on the health status within the minority group of men require a
24    level of caution in evaluating these  outcomes, which, in themselves, based upon the NHANES
25    data from childhood national data, are not surprising.
26
27    Nutrition
28          Studies reported in populations from Mexico City, Boston and Rio de Janeiro provided
29    new information on nutritional parameters in  subjects with mean blood and erythrocyte lead
30    levels less than 10 |ig/dL.  These three studies examined the effects of calcium and vitamin D
31    nutrition in various populations.
      December 2005                           5-76        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Erythrocyte lead concentrations were evaluated in 68 pregnant and 45 lactating Rio de
 2    Janeiro women whose dietary intakes of calcium were low on a chronic basis (400-600 mg/day).
 3    Whole blood lead concentrations were less than 10 |ig/dL in these women, including 33 controls
 4    (Pires et al., 2001).  Lactating women had significantly higher erythrocyte lead values compared
 5    to both pregnant and control subjects.  Indices of bone resorption (urinary d-pyridinoline) and
 6    formation (plasma bone alkaline phosphatase) were significantly higher in pregnant and lactating
 7    women, suggesting that RBC lead was elevated in the ambiently exposed women during
 8    lactation with low dietary intakes of calcium.
 9          A larger group of lactating women (617) in Mexico City were examined from 1994-1995
10    to further understand the potential effects of lowering blood lead levels through dietary calcium
11    supplements (Hernandez-Avilla et al., 2003). The average age was 24 years; the mean blood
12    lead level at baseline was 8.5 |ig/dL. Women were randomly assigned to receive either calcium
13    carbonate (1200 mg/day) or placebo in a double-blind study, and blood lead concentrations were
14    measured at 3 and 6 months into the study. A modest decrease of 1.16 |ig/dL (mean) was
15    observed at 6 months in the calcium-supplemented group. This relatively small decrease in
16    blood lead values may be explained, in part, by relatively high lead burdens in this Mexico City
17    population, although blood lead levels did not exceed 8.5 |ig/dL initially.
18          A cross-sectional assessment was carried out by Cheng et al. (1998) in 747 males in the
19    context of the Boston Aging Study. In 67-year-old men (average age) the mean (SD) blood lead,
20    tibia lead and patella lead were 6.2 (4.1) |ig/dL, 21.9 (13.3) |ig/g, and 32 (19.5) |ig/g,
21    respectively. After adjusting for age, education,  cigarette use, and alcohol consumption, men in
22    the lowest quintile of total dietary intakes of vitamin D (179 lU/day) had higher bone lead
23    content compared to men in the highest quintile for vitamin D intake (IU 589/day).  These data
24    are consistent with those discussed above, in that low dietary intakes of vitamin could be
25    expected to decrease calcium and increase lead absorption from the GI tract.  However, dietary
26    calcium intakes were not measured in this study.
27
28    Genetic Polymorphisms
29          Since 1986-1990, two reports have been published relating to genetic polymorphisms in
30    ambiently exposed adults:  one of these is related to ALAD and the other to HFE.  Both of these
31    studies were carried out in the Boston Normative Aging Study in adult males. Hu et al. (2001)

      December 2005                           5-77        DRAFT-DO NOT QUOTE OR CITE

-------
 1    investigated whether ALAD polymorphism may be associated with blood and bone lead values
 2    in 726 middle-aged and elderly men from the Boston area.  In this group of men, the mean (SD)
 3    of blood lead concentrations, tibia lead and patella lead were 6.2 (4.1) |ig/dL, 22.1 (13.5) jig/g,
 4    and 30.4 (17.2) |ig/g, respectively.  The ALAD  1-1 genotype was associated with an increase of
 5    2.55 |ig/g in cortical bone (tibia), thereby suggesting the possibility that the ALAD 2 allele may
 6    decrease the accumulation of lead in bone. Whether this difference of 2.55 jig/g bone mineral
 7    was above or below the precision and/or the minimum detection limits of the K-XRF method
 8    was not addressed in this report.
 9          Within the same Boston population, Wright et al. (2004) evaluated potential relationships
10    between the HFE gene and bone lead values in 730 men. Of this population, 13 and 25% had the
11    C282Y and H63D variants of HFE, respectively. After adjusting for age, smoking, and
12    education, carriers of the HFE variant allele(s) had lower patella bone lead concentrations
13    compared to all groups by  polymorphism analyses. Caution in interpreting these data are
14    expressed, as in other data reported from the Boston Normative Aging Study.
15
16    Neurotoxicology of Lead
17          One study  has been reported since 1986-1990 that assessed aspects of cognitive
18    functioning in the Normative Aging Study in Boston within a group of 466 males who had low-
19    level or ambient lead exposure. The purpose was to evaluate whether biomarkers of lead were
20    related to cognitive functioning, and the latter was indexed by the Mini-Mental State
21    Examination (MMSE) (Weisskopf et al., 2004a). On two occasions, 3.5 years apart, MMSE
22    scores were obtained during 1993-2002 in men whose age averaged 67.4 years.  Bone lead
23    measurements by K-XRF were assessed on two occasions between 1993-2002. The presented
24    results indicated that a one-interquartile range (20 |ig/g) increase in patella lead was associated
25    with a decline in the MMSE equivalent to that of aging 5 years in relation to baseline MMSE
26    scores. Associations were not observed in values for blood lead or tibia lead levels. The authors
27    suggested that this steeper  decline in MMSE scores was thus related to lead that is mobilizable
28    from skeletal store (patella lead).  These data reflect an important beginning to define effects of
29    ambient lead exposure on cognitive functioning in adults.  Although the MMSE has been
30    employed in epidemiological population-based research, it is evident that a comprehensive
      December 2005                            5-78        DRAFT-DO NOT QUOTE OR CITE

-------
 1   neuropsychological test battery has the potential to provide more definitive information related to
 2   understanding further the impact of ambient lead exposure on cognition in adults.
 3
 4
 5   5.4    REPRODUCTIVE AND DEVELOPMENTAL EFFECTS OF LEAD
 6   5.4.1   Summary of Key Findings on the Developmental and Reproductive
 7           Effects of Lead in Animals from the 1986 Lead AQCD
 8         The 1986 Pb AQCD presented unequivocal evidence for effects of Pb on reproduction and
 9   development in laboratory animals, derived principally from studies of rodents. Fetotoxic effects
10   (spontaneous abortion and fetal death) were reported following chronic exposures to relatively
11   high doses (600 to 800 ppm inorganic lead) in the diet, and more subtle effects (such as changes
12   in ALA-D activity or hematocrit) at lower doses (5 to 10 ppm in drinking water and 10 |ig/m3 in
13   air).  The 1986 Pb AQCD reported that the lowest observed adverse effect level (LOAEL) for
14   reproductive and developmental effects was 64 |ig/kg per day (multiple exposures by gavage).
15         The 1986 Pb AQCD also reported evidence for a variety of sublethal effects on
16   reproduction and development in experimental laboratory animals following Pb exposure.
17   Sublethal effects included changes in levels or function of reproductive hormones as well as
18   effects on the gonads (both male and female) and conception.  The animal data also suggested
19   more subtle effects on hormone metabolism and reproductive cell structure. Stowe and Goyer
20   (1971) classified the reproductive effects of Pb as gametotoxic, whether intrauterine or
21   extrauterine.
22         The data reported in the 1986 Pb AQCD,  and more recent studies  conducted in
23   experimental animal models, provide convincing evidence that Pb induces temporary and long-
24   lasting effects on male and female reproductive and developmental function.  The newer
25   literature supports the earlier conclusions presented in the 1986 Pb AQCD that Pb disrupts
26   endocrine function at multiple points along the hypothalamic-pituitary-gonad axis (Sokol et al.,
27   1985; Stowe and Goyer, 1971; Vermande Van Eck and Meigs, 1960; Junaid et al., 1997;
28   McGivern at al., 1991; Ronis et al., 1996, 1998b,c; Sokol, 1987; Sokol et al., 1985, 1994, 1998;
29   Sokol and Berman,  1991; Kempinas  at al., 1988, 1990, 1994; Tchernitchin et al., 1998b; Sant'
30   Ana et al., 2001; Srivastava et al., 2004). A schematic representation of the hypothalamic-
31   pituitary-gonadal axis is shown in Figure 5-4.1.

     December 2005                           5-79        DRAFT-DO  NOT QUOTE OR CITE

-------
                                                                           or
                                                                    Esfradiol
     Figure 5-4.1.   Data from male and female experimental animals suggests that Pb
                    has multiple targets in the hypothalmic-pituitary-gonadal axis.
 1         The majority of the experimental animal studies on developmental and reproductive
 2   effects of Pb examined effects due to inorganic forms of lead; very little is known about the
 3   reproductive and developmental effects due to organic forms. In general, the few available
 4   studies suggest that effects of organic forms of Pb are similar to those produced by inorganic
 5   forms. Administration of triethyl-Pb-chloride during early gestation reduces pregnancy rates in
 6   mice (Odenbro and Kihlstrom, 1977). Growth retardation following organolead exposure has
 7   been reported (Kennedy et al., 1975; McClain and Becker, 1972).  More recent studies have
 8   demonstrated that exposure of mice to triethyl-Pb-chloride during late gestation reduces perinatal
 9   growth rate (Odenbro et al., 1988).
10         This section summarizes the evidence for effects of Pb exposure in developing organisms
11   exposed during the period from conception to maturity that has been reported since 1986.
12   Effects on neurological, immunological, or renal endpoints in developing organisms are
13   discussed in Sections 5.3, 5.9 and 5.7, respectively.
     December 2005
5-80
DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.4.2   Effects on Male Reproductive Function
 2          The 1986 Pb AQCD reported convincing evidence based on experimental animal studies
 3    that Pb acts as an endocrine disrupter in males. Those studies demonstrated an association
 4    between reduced male fertility and repeat-dose Pb exposure.  Lead exposure had been reported to
 5    alter sperm development and function; however, the mechanism underlying these effects was not
 6    completely understood.  These effects were attributed to either alterations in testicular enzymes
 7    important for hormone production or to changes in the hormone receptors. More recent research
 8    supports the conclusion that the mechanisms for endocrine disruption in males involves Pb
 9    acting at multiple sites along the hypothalamic-pituitary-gonadal axis (see Figure 5-4.1).
10          Reported effects of Pb on male reproduction differ substantially across studies, with some
11    studies finding profoundly adverse effects and other studies finding no or minimal effects. The
12    variable findings have been attributed to the complex mechanisms involved in hormone
13    regulation and the multiple sites of action for lead.  Sokol et al. (2002) suggested that differences
14    in results among studies may be, in part, attributed to an adaptive mechanism in the
15    hypothalamic-pituitary-gonadal axis that may render the expression of some toxic effects
16    dependent on exposure duration.  Sokol and Berman (1991) found that timing of exposure was
17    critical to Pb-induced male reproductive toxicity in rats.  Studies conducted in nonhuman
18    primates supported the importance of timing, finding that the adverse effects of Pb on male
19    reproduction are dependent upon age (i.e., developmental stage at time of exposure) and duration
20    of exposure (Foster et al., 1993;  Singh et al., 1993a).
21          The adverse effects of Pb on male reproduction may be expressed as perturbations in
22    sexual development and maturation, changes in fertility, endocrine disruption, and alterations in
23    structure of reproductive cells or tissue. Each of these effects is discussed in greater detail in the
24    sections that follow.
25
26    5.4.2.1  Effects on Male Sexual Development and Maturation
27          The 1986 Pb AQCD reported adverse effects of Pb on male sexual development and
28    maturation.  Experimental studies conducted in animals demonstrated that high-dose (e.g.,
29    dietary exposure to 0.08 to 1.0% Pb-acetate in mice  and to 100 ppm in dogs) preadolescent Pb
30    exposure can produce long-lasting detrimental effects on male sexual development. Numerous
31    more recent studies conducted in experimental animals support the earlier findings that Pb

      December 2005                            5-81       DRAFT-DO NOT QUOTE OR CITE

-------
 1    exposure during early development can delay the onset of male puberty and alter reproductive
 2    function later in life (McGivern et al., 1991; al-Hakkak et al., 1988; Chowdhuri et al., 2001;
 3    Dearth et al., 2002, 2004; Gandley et al., 1999; McGivern et al., 1991; Ronis et al., 1998a,c;
 4    Sokol et al., 1994; Yu et al., 1996).  Studies that provide the strongest evidence for the dose-
 5    response range for typical effects in rodents are discussed below (Table 5-4.1).
 6          McGivern et al. (1991) found that male rats born to dams that received Pb-acetate in
 7    drinking water beginning on gestation day 14 and through parturition (PbB 73  ng/dL) exhibited
 8    reduced  sperm counts, altered male reproductive behavior, and enlarged prostates later in life.
 9    Prepubertal exposure of male Sprague-Dawley rats (age 24 to 74 days) to Pb-acetate in drinking
10    water (PbB 30 to 60  ng/dL) resulted in significant reduction in testis weight and in the weight of
11    secondary  sex organs; however, these effects were not observed in rats exposed postpubertally
12    (day 60 to  74; Ronis et al.,  1996).  A dose-dependent delay in sexual maturation was found in
13    male rats, following prenatal Pb exposure that continued until adulthood (age 85 days) (Ronis
14    et al.,  1998a,b,c). In these studies, PbBs in the pups between the ages of 21 and 85 days were
15    >100 ng/dL. Additional details concerning these studies  are provided in Table 5-4.1.
16          One possible explanation for the persistent effects  of Pb exposure on the male
17    reproductive system is a disruption in pulsatile release of sex hormones during early
18    development (Ronis et al., 1998c). Lead effects on sex hormones are discussed in
19    Section 5.4.2.3.
20
21    5.4.2.2  Effects on Male Fertility: Effects  on Sperm Production and Function
22          The 1986 Pb AQCD presented evidence that Pb exposure affects male fertility in various
23    animal species, including rabbits (Cole and Bachhuber, 1915), guinea pigs  (Weller, 1915), rats
24    (Ivanova-Chemishanska et  al., 1980), and mice (Schroeder and Mitchener,  1971).
25          Several more recent  studies, conducted in various animal species, have demonstrated Pb-
26    induced  alteration of sperm parameters (e.g.,  count, motility, number of abnormal) (Sokol et al.,
27    1985; and  eight other studies). These effects, however, have not been reproduced in  all studies.
28    For example, Foster et al. (1996a) reported that 15- to 20-year-old cynomolgus monkeys
29    receiving Pb-acetate for their lifetime (mean PbB 56 |ig/dL) showed no significant alterations in
30    sperm parameters (i.e., sperm count, viability, motility, and morphology) or circulating levels of
31    testosterone (see Section 5.4.2.3 for discussion of lead-induced changes in testosterone levels).

      December 2005                            5-82        DRAFT-DO NOT QUOTE OR CITE

-------
o

O

3
to
O
O
                   Table 5-4.1. Selected Studies Showing the Effects of Lead on Reproductive Function in Males
           Citation
             Species/
             Strain
                            Dose/Route/Form/Duration/Group Size      Endpoint/Magnitude of Effect/p-value      Blood Lead Concentration (PbB)
oo
fe
H
6
o
o
H
O
o
HH
H
W
Foster etal.
(1993)
Foster etal.
(1996a)
           Foster et al.
           (1998)
           McGivern
           etal.
           (1991)
Ronis et al.
(1996)
             Monkey/
             Cynomolgus
                         Monkey/
                         Cynomolgus
             Monkey/
             Cynomolgus
             Rat/Sprague-
             Dawley
             Rat/Sprague-
             Dawley
0-1500 |ig Pb-acetate/kg-d in gelatin capsules
p.o. for various durations: 9 control monkeys,
4 monkeys in lifetime group (birth to 9 years),
4 in infancy group (first 400 days of life), 4 in
post-infancy exposure (from 300 days to 9 years)

0-1500 ng Pb-acetate/kg-d in gelatin capsules
p.o. from birth until 9 years of age

8 control monkeys, 4 monkeys in low group
(6-20 ng/dL), 7 monkeys in high group
(22-148 ng/dL)

0-1500 |ig Pb-acetate/kg-d in gelatin capsules
p.o. for various durations: birth to 10 years
(lifetime); PND 300 to 10 years  (post-infancy);
birth to 300 days (infancy);
3 control monkeys, 4 lifetime, 4 infancy,
5 post-infancy
                            0.1% Pb-acetate in drinking water from GD 14 to
                            parturition; 8 control litters; 6 Pb-acetate litters
                            (5 males per litter)
0.6% Pb-acetate in drinking water for various
durations:  PND 24-74 (pubertal exposure); PND
60-74 (post pubertal exposure); 11 males and
females in pubertal exposure group (10 each in
control pubertal group); 6 males and females
post-pubertal exposure and control groups
                                               Suppressed LH response to GnRH stimulation in
                                               the lifetime group (p = 0.0370); Sertoli cell
                                               function (reduction in the inhibin to FSH ratio)
                                               (p = 0.0286) in lifetime and post-infancy groups.
                                                                          Mean PbB of 56 ng/dL showed no significant
                                                                          alterations in parameters of semen quality
                                                                          (count, viability, motility, or morphology).
Circulating concentrations of FSH, LH, and
testosterone were not altered by treatment;
semen characteristics (count, motility,
morphology) were not affected by treatment
possibly because not all Sertoli cells were
injured; degeneration of seminiferous epithelium
in infancy and lifetime groups (no difference in
severity between these groups); ultrastructural
alterations in seminal vesicles, most prominent
in infancy and post-infancy groups.

Decreased sperm count (21% at 70 days and
24% at 165 days, p<0.05); reduced male
behavior (p < 0.05); enlarged prostate (25%
increase in weight; p<0.07); irregular release
patterns of both FSH and LH (p<0.05).

PbB >250 ng/dL reduced circulating
testosterone levels in male rats  40-50%
(p < 0.05); reduction in male secondary sex
organ weight (p < 0.005); delayed vaginal
opening (p<0.0001); disrupted estrous cycle in
females (50% of rats); increased incidence of
stillbirth (2% control vs. 19% Pb) (p < 0.005).
                                              Lifetime group 3-26 ng/dL at 4-5
                                              years
                                              Infancy group 5-36 ng/dL at 100-300
                                              days, 3-3 ng/dL at 4-5 years
                                              Post-infancy group 20-35 ng/dL

                                              PbB10±3or56±49|ig/dL
                                                                                                                        PbB -35 ng/dL
                                                                                            Control PbB <5 ng/dL at birth
                                                                                            Maternal PbB 73 ng/dL at birth
                                                                                            Pup PbB 64 ng/dL at birth
                                                                                                                                    Pubertal PbB 30-60 ng/dL
                                                                                                                                    Post-pubertal PbB 30-60 ng/dL
                                                                                                                                    Mean PbBs in male rats 30-60 ng/dL,
                                                                                                                                    respectively

-------
o

O

3
to
O
O
oo
fe
H
6
o
o
H
O
O
HH
H
W
              Table 5-4.1 (cont'd).  Selected Studies Showing the Effects of Lead on Reproductive Function in Males

             Species/
Citation     Strain        Dose/Route/Form/Duration/Group Size     Endpoint/Magnitude of Effect/p-value      Blood Lead Concentration (PbB)
Ronis et al.
(1998a)










Rat/Sprague- 0.6% Pb-acetate in drinking water ad libitum for
Dawley various durations as follows: GD 5 to PND 1;
GD 5 to weaning; PND 1 to weaning; 3 control
litters, 2 gestation exposure litters, 2 lactation
exposure litters, 2 gestation and lactation
exposure litters, 2 postnatal exposure litters,
2 chronic exposure litters; 4 male and 4 female
pups per litter




Suppression of adult mean serum testosterone
levels was only observed in male pups exposed
to Pb continuously from GD 5 throughout life
(p<0.05).








Group

Naive

Control

Gest

Lact
Gest+Lact
Postnatal
Chronic
Male PbB

5.5±2.0 ug/dL

1.9±0.2 ug/dL

9.1±0.7ug/dL

3.3±0.4 ug/dL
16.1±2.3 ug/dL
226.0±29 ug/dL
316.0±53 ug/dL
           Ronis et al.   Rat/Sprague-   Lead acetate in drinking water (0.05% to 0.45%
           (1998b)      Dawley        w/v); dams exposed until weaning; exposure of
                                      pups which continued until PND 21, 35, 55, or
                                      85; 5 control litters (0%), 10 low-dose litters
                                      (0.05%), 8 mid-dose litters (0.15%), 9 high-dose
                                      litters (0.45%); 4 male and 4 female pups per
                                      litter
Ronis et al.    Rat/Sprague-   Lead acetate 0.05, 0.15, or 0.45% in drinking
(1998c)      Dawley        water beginning GD 5 continuing until PND 21,
                           35, 55, or 85; 5 control litters (0%), 10 low-dose
                           litters (0.05%), 8 mid-dose litters (0.15%), 9
                           high-dose litters (0.45%); 4 male and 4 female
                           pups per litter
Dose-response reduction in birth weight
(p < 0.05), more pronounced in male pups;
decreased growth rates in both sexes (p < 0.05)
were accompanied by a statistically significant
decrease in plasma concentrations of IGF 1
through puberty PND 35 and 55 (p < 0.05);
increase in pituitary  growth hormone during
puberty (p < 0.05).

Dose-responsive decrease in birth weight
(p < 0.05); dose-responsive decrease in crown-
to-rump length (p <  0.05); dose-dependent delay
in sexual maturity (p < 0.05); decrease in
prostate weight (p < 0.05); decrease in plasma
concentration of testosterone during puberty
(p < 0.05); decrease  in plasma LH (p < 0.05);
elevated pituitary LH content (p < 0.05);
decrease in plasma testosterone/LH ratio at high
dose (p < 0.05).
                                                                                                                     Mean PbB in offspring at 0.05% (w/v)
                                                                                                                     49±6 ug/dL

                                                                                                                     Mean PbB in offspring at 0.15% (w/v)
                                                                                                                     126 ±16 ug/dL

                                                                                                                     Mean PbB in offspring at 0.45% (w/v)
                                                                                                                     263 ± 28 ug/dL
                                                                                                                                Dams: 0, 48, 88, or 181 ug/dL
                                                                                                                                Pups PND 1:  <1, 40, 83, or 120 ug/dL
                                                                                                                                Pups PND 21: <1, 46,  196, or
                                                                                                                                236 ug/dL
                                                                                                                                Pups PND 35: <1, 20,  70, or
                                                                                                                                278 ug/dL
                                                                                                                                Pups PND 55: <1, 68,  137, or
                                                                                                                                379 ug/dL
                                                                                                                                Pups PND 85: <1, 59,  129, or
                                                                                                                                214 ug/dL

-------
o

O

3
to
O
O
                       Table 5-4.1 (cont'd).  Selected Studies Showing the Effects of Lead on Reproductive Function in Males
oo
fe
H
6
o

o
H
O
Citation
Singh et al.
(1993a)
Sokol and
Berman
(1991)




Species/
Strain Dose/Route/Form/Duration/Group Size
Monkey/ 0-1500 |ig Pb-acetate/kg-d in gelatin capsules for
Cynomolgus various durations: 3 control monkeys, 4 monkeys
in infancy group (exposure first 400 days), 5 in
post-infancy group (exposure 300 days to 9 years
of age), 4 in lifetime group (exposure from birth
until 9 years)
Rat/Wistar 0, 0. 1, or 0.3% Pb-acetate in drinking water for
30 days beginning at 42, 52, or 70 days old; 8-11
control rats for each age, 8-1 1 rats for each age in
0.1% group, 8-11 rats for each age in 0.3% group




Endpoint/Magnitude of Effect/p-value Blood Lead Concentration (PbB)
Degeneration of seminiferous epithelium in all Chronic PbB
exposed groups (frequency not specified); <40-50 |ig/dL
ultrastructural alterations in seminal vesicles,
most prominent in infancy and post-infancy
groups (frequency not specified).
Dose-related suppression of spermatogenesis Group
(decreased sperm count and sperm production
rate) in the exposed rats of the two highest age 0%
groups (p < 0.05); dose-related suppression of
serum testosterone in 52-day old rats (p = 0.04)
and in 70-day old rats (p < 0.003).
0.1%


0.3%


Age
All
42 d
52 d
70 d
42 d
52 d
70 d

PbB
<7 ng/dL
25 |ig/dL
35 |ig/dL
37 ng/dL
36 ng/dL
60 ng/dL
42 ng/dL
          FSH, follicle stimulating hormone; GD, gestational day; GnRH, gonadotropin releasing hormone; IGF], insulin-like growth factor 1; LH, luteinizing hormone; PbB, blood Pb
          concentration; PND, post-natal day
O
HH
H
W

-------
 1          Sokol et al. (2002) provided evidence of an adaptive mechanism in the hypothalamic-
 2    pituitary-gonadal axis in response to prolonged exposure to lead.  The existence of this adaptive
 3    mechanism would explain the apparent inconsistency in reported effects on circulating
 4    testosterone levels, sperm count, and sperm production following Pb exposure. Because of this
 5    adaptive mechanism, changes in testosterone levels and certain sperm parameters may not
 6    always serve as reliable endpoints for assessing the effects of Pb on male fertility and
 7    reproductive function for all exposure durations.
 8          Although gross changes in sperm parameters were not observed in monkeys in which
 9    chronic PbB was approximately  56 |ig/dL, Foster et al. (1996a) reported that monkey sperm
10    exhibited a statistically significant, dose-related reduction in chromatin structure (as determined
11    by susceptibility to weak acid denaturation). These changes may have adverse impacts on
12    fertility, and they are thought to be related to dominant lethal effects of Pb (similar to the effects
13    reported by al-Hakkak et al. [1988] in mice). Additional details concerning Foster et al.  (1996a)
14    are provided in Table 5-4.1.
15          The data from Foster et al. (1996a), demonstrating a change in monkey sperm chromatin
16    suggestive of a subtle lead-induced reduction in male fertility (in the absence of gross changes
17    sperm parameters), are consistent with observations of reduced in vitro fertilization capacity of
18    sperm collected from other mammalian species.  Sokol et al. (1994) reported that exposure of
19    adult male rats to Pb-acetate in drinking water  for 14 to 60 days (PbB 33 to 46 |ig/dL) resulted in
20    reduced in vitro fertilization of eggs harvested  from unexposed females. No differences  were
21    observed  in sperm ultrastructure or in the DNA histogram of sperm obtained from lead-exposed
22    rats compared to controls.  Consistent with this finding are reports of reduced fertilization
23    capacity of rabbit sperm exposed to high concentrations (25 jiM) of Pb chloride in vitro (Foote,
24    1999) and reduced in vitro  fertilization capacity of sperm from mice exposed to Pb in drinking
25    water at 1 g/L for 4 months (PbB not reported) (Johansson et al., 1987).
26          Two modes of action have been proposed  for lead-induced alterations in sperm capacity
27    for fertilization.  The affinity of Pb for sulfhydryl groups may explain some of the lead-induced
28    alterations in sperm structure and function. Mammalian sperm possess high concentrations of
29    sulfhydryl groups, which are critical for maintaining normal function (Johansson and Pellicciari,
30    1988). Reyes et al. (1976)  demonstrated that binding of Pb to membrane thiols inhibits sperm
31    maturation. In addition, recent experimental data also suggest that lead-induced generation of

      December 2005                            5-86        DRAFT-DO NOT QUOTE OR CITE

-------
 1    reactive oxygen species (ROS) may contribute to the injury of tissues responsible for sperm
 2    formation (see Section 5.4.2.4).
 3
 4    5.4.2.3  Effects on Male Sex Endocrine System
 5          The 1986 Pb AQCD reported that, although the mode of action for the adverse effects of
 6    Pb on the male reproductive system was not understood, effects on hormone production or
 7    hormone receptors were likely contributors. More recent studies provide convincing evidence
 8    that Pb acts as an endocrine disrupter in males at various points along the hypothalamic-
 9    pituitary-gonadal axis (Figure 5-4.1). In rats, Pb exposures that decreased serum testosterone
10    levels increased mRNA levels of GnRH and LH in the hypothalamus and pituitary, respectively,
11    and increased levels of LH in pituitary; these changes can occur in the absence of a change in
12    serum gonadotropin levels (Klein et al., 1994; Ronis et al., 1998c; Sokol et al., 2002).
13    In monkeys, chronic Pb exposures (PbB 20 to 35 |ig/dL) suppressed GnRH-induced secretion of
14    LH and decreased serum testosterone:LH and inhibin:FSH ratios (Foster et al., 1993). The
15    mechanisms underlying the effects on the hypothalamic-pituitary-gonadal axis have not been
16    elucidated but may involve a suppression of GnRH secretion (Bratton et al., 1994; Sokol, 1987;
17    Sokol etal., 1998).
18          Although there is evidence for a common mode of action, consistent effects on circulating
19    testosterone levels are not always observed in lead-exposed animals. Rodamilans et al. (1988)
20    and Kempinas et al. (1994) attributed these inconsistencies to the normal biological variation
21    (circannual and seasonal) of testosterone secretion in rats and monkeys. Observations of
22    lead-induced reductions in testosterone levels in some studies, but not others, may be due to
23    enhanced sensitivity to inhibition of the testosterone secretory system during certain periods of
24    development.  In addition, the hypothalamic-pituitary-gonadal axis exhibits compensatory
25    mechanisms that may attenuate the effects  of Pb during prolonged Pb exposure (Sokol et al.,
26    2002). Taken together, the sensitivity of testosterone secretion during certain periods and
27    potential for modulation of the effects during long-term exposures studies, may explain some of
28    the apparent inconsistencies in the reported effects of Pb exposure on circulating testosterone
29    levels.
30
      December 2005                            5-87        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.4.2.4  Effects on Morphology and Histology of Male Sex Organs
 2          The 1986 Pb AQCD reported evidence for histological changes in the testes or prostate in
 3    rats, in association with relatively high doses of Pb (Chowdhury et al., 1984; Hilderbrand et al.,
 4    1973; Golubovich et al., 1968). More recent studies conducted in animal models provide
 5    persuasive support for testicular damage (i.e., ultrastructural changes in testes and cytotoxicity in
 6    Sertoli cells) following lower level lead exposure (Foster et al., 1998; Singh et al., 1993a; Batra
 7    et al., 2001; Chowdhury et al., 1986, 1987; Corpas et al., 1995; Pinon-Lataillade et al., 1993;
 8    Saxenaetal., 1990).  Studies conducted in nonhuman primates warrant particular attention.
 9    These studies found ultrastructural changes in the testes (Sertoli and other spermatogenic cells)
10    of monkeys at PbB 35 to 40 |ig/dL (Foster et al.,  1998; Singh et al., 1993a).
11          Foster et al. (1998) reported that chronic Pb exposure (PbB -35 |ig/dL), beginning in
12    infancy, resulted in persistent ultrastructural changes in the testes of cynomolgus monkeys.
13    Electron microscopy showed disruption of the general structure of the seminiferous epithelium
14    involving Sertoli cells, basal lamina, and spermatids in the groups exposed for lifetime and
15    during infancy (with no duration difference in severity).  Chronic exposures to Pb beginning
16    after infancy, that achieved similar PbBs, did not produce these effects.
17          Similarly, Singh et al. (1993a) demonstrated ultrastructural changes in testicular basement
18    membrane and Sertoli cell morphology (seminiferous tubules) in cynomolgus monkeys exposed
19    chronically to Pb (PbB <40 to 50  |ig/dL); the effects were most prominent when dosing began in
20    infancy or post-infancy. These results suggest that, in monkeys, Pb exposure during certain
21    periods of development produces persistent testicular alterations. Additional details concerning
22    Foster et al. (1998) and Singh et al. (1993a) are provided in Table 5-4.1.
23          A possible mode of action for lead-induced testicular injury is oxidative stress. Foster
24    et al. (1998) suggested that lead-induced oxygen  free radical generation was a plausible
25    mechanism of testicular injury in primates. This  oxygen radical hypothesis is supported by
26    studies conducted in rodents (Chowdhury et al., 1984; Acharya et al., 2003; Adhikari et al.,
27    2001; Batra et al., 2001; Bizarro et al., 2003; Chowdhury et al., 1984; Gorbel et al., 2002; Mishra
28    and Acharya, 2004). Also supporting the oxidative stress hypothesis are observations of
29    increases in the percentage of apoptotic cells in the testes of rodents in response to Pb exposure
30    (Pace et al., 2005;  Gorbel  et al., 2002; Adhikari et al., 2001).
31

      December 2005                            5-88        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.4.3   Effects on Female Reproductive Function
 2          Lead has been shown to disrupt the hypothalamic-pituitary-gonadal axis and to produce
 3    ovarian atrophy and reproductive dysfunction in females (Figure 5-4.1).  The 1986 Pb AQCD
 4    reported that Pb exposure was associated with inhibition of menstruation, ovulation, and
 5    follicular growth in monkeys (Vermande-Van Eck and Meigs, 1960), and in rodents Pb exposure
 6    delayed vaginal opening, decreased frequency of implantation, and reduced rates of pregnancy
 7    (Kimmel et al., 1980; Odenbro and Kihlstrom, 1977, respectively).
 8          Data from more recent experimental animal studies support these findings. Lead effects
 9    on female reproduction may be classified as alterations in female sexual maturation, effects on
10    fertility and menstrual cycle, endocrine disruption, and changes in morphology or histology or
11    female reproductive organs as well as the placenta.  Recent literature concerning each of these
12    effects is summarized below.
13
14    5.4.3.1  Effects on Female Sexual Development and Maturation
15          The 1986 Pb AQCD reported that Pb exposure in rodents produced delays in sexual
16    maturation. Grant et al.  (1980) reported delayed vaginal opening in female rats exposed in utero
17    and during lactation and maturation (PbB -20 to 40 |ig/dL). More recent studies in experimental
18    animals (primarily rodent studies) provide convincing evidence that Pb exposure before puberty
19    (particularly prenatal and early postnatal exposure) delays the maturation of the female
20    reproductive system (Dearth et al., 2002, 2004; Ronis et al., 1996, 1998b,c).
21          Dearth et al. (2002) is of particular interest, because  it employed a cross-fostering design
22    (to allow comparison of pups exposed during gestation only, lactation only, or both) and because
23    maternal and offspring PbBs were monitored throughout gestation and lactation.  Fisher 344
24    dams were exposed to Pb by gavage beginning 30 days before mating until weaning of the pups
25    at 21  days of age (gavage exposure removes possible confounding of exposure by consumption
26    of Pb in drinking water by pups in those studies where drinking water is the route of exposure for
27    dams). Mean maternal PbB was approximately 40 |ig/dL. Pups exposed during gestation and
28    lactation had the highest PbB (38.5 |ig/dL) on day 10; at this time, the PbBs in pups exposed
29    during gestation only or lactation only were 13.7 and 27.6 |ig/dL, respectively. By postnatal  day
30    (PND) 30,  all three groups had PbB <3 |ig/dL. Dearth et al. (2002) reported a statistically
31    significant delay in the onset of puberty (vaginal opening and days at first diestrus) in rats

      December 2005                           5-89        DRAFT-DO  NOT QUOTE OR CITE

-------
 1    exposed during lactation, gestation, or during lactation and gestation (with no differences among
 2    the groups). In addition, a statistically significant reduction in the circulating levels of insulin-
 3    like growth factor 1 (IGFi), LH, and estradiol (E2) were reported on PND 30 in all three
 4    treatment groups (with no differences among treatment groups). Additional details concerning
 5    Dearth et al. (2002) are provided in Table 5-4.2.
 6          A subsequent study in both Sprague-Dawley and F344 rats (Dearth et al., 2004) showed
 7    that the F344 strain is more sensitive to maternal Pb exposure than Sprague-Dawley rats to lead-
 8    induced delayed puberty, which could, in part, explain the inconsistencies with effect levels
 9    observed in Sprague Dawley rats (e.g., Ronis et al., 1998a,b,c; McGivern et al., 1991). Ronis
10    et al.  (1998c) suggested that the delayed onset of puberty may arise from a lead-induced
11    disruption of pulsatile release of sex hormones (see Section 5.4.3.3).
12
13    5.4.3.2  Effects on Female Fertility
14          The 1986 Pb AQCD reported convincing evidence from experimental animal studies for
15    lead-induced alterations in female fertility, including interference with implantation and
16    pregnancy (Odenbro and Kihlstrom, 1977; Wide and Nilsson, 1977).  More recent studies have
17    confirmed these effects. In general, Pb exposure does not produce total  sterility, although Pb
18    exposure clearly disturbs female fertility (Taupeau et al., 2001). Studies in nonhuman primates
19    and rodents have shown that exposure of gravid females to Pb produces implantation  dysfunction
20    and reduces litter size and newborn survival (Logdberg et al., 1987; Flora and Tandon, 1987;
21    Johansson and Wide, 1986; Pinon-Lataillade et al., 1995; Piasek and Kostial,  1991; Ronis et al.,
22    1996).  See Section 5.4.4.1 for details.
23
24    5.4.3.3  Effects on the Female Sex Endocrine System and Menstrual Cycle
25          The 1986 Pb AQCD described numerous studies that found effects of Pb on the female
26    endocrine system and menstrual cycle in various species, including nonhuman primates, and that
27    supported the conclusion that Pb was an endocrine disrupter in females (Grant et al., 1980;
28    Maker et al., 1975; Vermande-Van Eck and Meigs, 1960). Observations of delayed vaginal
29    opening (see Section 5.4.3.1) were attributed to the endocrine disruption effects of Pb on the
30    hypothalamic-pituitary-gonadal axis (Stowe and  Goyer, 1971; Vermande Van Eck and Meigs,
31    1960).

      December 2005                           5-90        DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
                Table 5-4.2. Selected Studies Showing the Effects of Lead on Reproductive Function in Females
                       Species/       Dose/Route/Form/Duration/
          Citation      Strain        Group Size
                                                                 Endpoint/Magnitude of Effect
                                                                 (% or incidence) /p-value
                                                                                                 Blood Lead Concentration
                                                                                                 (PbB)
H
6
o

o
H
O
O
H
W
O

O
HH
H
W
Dearth
etal.
(2002)
          Foster
          (1992)
          Foster et al.
          (1992)
          Foster et al.
          (1996b)
Rat/Fisher
344
             Monkey/
             Cynomolgus
             Monkey/
             Cynomolgus
             Monkey/
             Cynomolgus
12 mg/mL Pb-acetate gavage from
30 days prior breeding until pups were
weaned 21 day afterbirth; 10-32 litters
per group, control group, gestation and
lactation exposure, gestation only
exposure, lactation only exposure
              Daily dosing for up to 10 years with
              gelatin capsules containing Pb-acetate
              (1.5 mg/kg); 8 control group monkeys,
              8 lifetime exposure (birth-10 years),
              8 childhood exposure (birth-400 days),
              and 8 adolescent exposure
              (PND 300-10 years of age)

              Daily dosing for up to 10 years with
              gelatin capsules containing Pb-acetate
              (1.5 mg/kg); 8 control group monkeys,
              8 childhood (birth-400 days),
              7 adolescent (PND 300-10 years),
              7 lifetime (birth-10 years)

              Chronic exposure to Pb-acetate 50 to
              2000 ug/kg-day p.o. beginning at birth
              for 15-20 years; 20 control monkeys,
              4 monkeys in 50 ug/kg-d group,
              3 monkeys in 100  ug/kg-d, 2 monkeys
              in 500 ug/kg-d group, and 3 monkeys in
              2000 ug/kg-d group
Delay in onset of puberty (p < 0.05); reduced
serum levels of IGFj (p < 0.001), LH
(p< 0.001), and E2 (p < 0.001).
                                      Statistically significant reductions in circulating
                                      levels of LH, (p < 0.042), FSH (p < 0.041), and
                                      E2 (p < 0.0001) during menstrual cycle;
                                      progesterone concentrations were unchanged and
                                      menstrual cycle was not significantly affected.
                                      No effect on endometrial response to gonadal
                                      steroids as determined by ultrasound.
                                      Reduced corpora luteal production of
                                      progesterone (p = 0.04), without alterations in
                                      E2, 20-alpha-hydroxyprogesterone, or menstrual
                                      cyclicity.
                                                                                                                         Maternal PbB -40 ug/dL
                                                                                                                         Pups PbB as follows:
                                                                                                                         Gest+lact -38 ug/dL PND 10
                                                                                                                         Gest+lact -15 ug/dL PND 21
                                                                                                                         Gest+lact -3  ug/dL PND 30
                                                                                                                         Gest -14 ug/dL PND 10
                                                                                                                         Gest~3 ug/dL PND 21
                                                                                                                         Gest~l ug/dL PND 30
                                                                                                                         Lact-28 ug/dL PND 10
                                                                                                                         Lact-15 ug/dL PND 21
                                                                                                                         Lact-3 ug/dL PND 30

                                                                                                                         PbB <40 ug/dL
                                             PbB <40 ug/dL
                                             PbB 10-15 ug/dL in low group
                                             (50 or 100 ug/kg-day)
                                             PbB 25-30 ug/dL in moderate
                                             group (500 or 2000 ug/kg-day)

-------
o

O

to
O
O
                      Table 5-4.2 (cont'd). Selected Studies Showing the Effects of Lead on Reproductive Function in Females
                       Species/       Dose/Route/Form/Duration/
          Citation      Strain        Group Size
                                                                           Endpoint/Magnitude of Effect
                                                                           (% or incidence)/p-value
                                              Blood Lead Concentration
                                              (PbB)
fj\

to
Franks       Monkey/      Lead acetate in drinking water (2-8
etal.         Rhesus        mg/kg-d) for 3 3 months; 7 control
(1989)                     and 10 Pb monkeys
          Laughlin
          et al.
          Logdberg
          et al.
          (1988)
                       Monkey/      Lead acetate in drinking water at 3.6,
                       Rhesus        5.9, or 8. 1 mg/kg-day for 1-2 years

                                     7 control and 10 experimental monkeys
                                     per group

                       Monkey/      Lead acetate (varying concentrations
                       Squirrel       <0. 1% in diet) maternal dosing from
                                     5-8.5 weeks pregnant to PND 1

                                     1 1 control monkeys, 3 low Pb exposure
                                     group (PbB 24 ug/dL), 7 medium Pb
                                     group (PbB 40 ug/dL, 5 high Pb group
                                     (PbB 56 ug/dL)
Reduced circulating concentration of
progesterone (p < 0.05); treatment with Pb did
not prevent ovulation, but produced longer and
more variable menstrual cycles and shorter
menstrual flow.

Reductions in cycle frequency (pO.Ol); fewer
days of flow (p<0.01); longer and more variable
cycle intervals (p<0.025).
Dose-dependent reduction in placental weight
(p<0.0007); various pathological lesions were
seen in the placentas (n = 4), including
hemorrhages, hyalinization of the parenchyma
with destruction of the villi and massive
vacuolization of chorion epithelium.
                                                                                                                        PbB 68.9 ± 6.54 ug/dL
                                                                                                              PbB 44-89 ug/dL
                                                                                                              51.2 ug/dL (low dose)
                                                                                                              80.7 ug/dL (mid dose)
                                                                                                              88.4 ug/dL (high dose)

                                                                                                              Mean maternal PbB 37 ug/dL
                                                                                                              (22-82 ug/dL)
                                                                                                              24 (22-26) ug/dL (low dose)
                                                                                                              40 (35-46) ug/dL (mid dose)
                                                                                                              56 (43-82) ug/dL (high dose)
H
6
O

O
H
O

O
H
W
O

O
HH
H
W
          E2, estradiol; FSH, follicle stimulating hormone; GD, gestational day; IGF!, insulin-like growth factor 1; LH, luteinizing hormone; PbB, blood Pb
          concentration; PND, post-natal day

-------
 1          More recent studies have provided convincing support for endocrine-mediated alterations
 2    of the female reproductive system in rats (Srivastava et al., 2004; Dearth et al., 2002; Ronis
 3    et al., 1998a,b,c; Junaid et al., 1997; Ronis et al., 1996), guinea pigs (Sierra and Tiffany-
 4    Castiglioni, 1992), and nonhuman primates (Foster et al., 1992, 1996b; Foster, 1992; Franks
 5    et al., 1989; Laughlin et al., 1987). The nonhuman primate studies are particularly relevant to
 6    extrapolations to humans and provide dose-response information for effects of Pb on female sex
 7    hormones and menstrual cycle.
 8          Laughlin et al. (1987) found that exposure to Pb (PbB 44 to 89 |ig/dL) alters menstrual
 9    cycles (specifically, causing reductions in cycle frequency, fewer days of menstrual flow, and
10    longer and more variable cycle intervals) in female rhesus monkeys. Consistent with these
11    observations, Franks et al. (1989) found that chronic exposure to Pb in the drinking water (PbB
12    70 |ig/dL) reduced circulating concentrations of progesterone (suggesting impaired luteal
13    function), produced longer and more variable menstrual cycles and temporally shorter menstrual
14    flow in female rhesus monkeys. Additional details concerning these studies are provided in
15    Table 5-4.2.
16          At lower blood Pb levels (PbB <40 jig/dL), female cynomolgus monkeys exhibited
17    statistically significant reductions  in circulating levels of LH, FSH, and E2 during the menstrual
18    cycle; however, serum progesterone concentrations were unchanged and menstrual cycle was not
19    significantly affected (Foster, 1992).  Similar exposures and PbB were shown to have no effect
20    on endometrial response to gonadal steroids in cynomolgus monkeys as determined by
21    ultrasound analysis (Foster et al., 1992).  At lower blood lead concentrations (25 to 30 jig/dL),
22    reduced corpora luteal production  of progesterone occurred in the absence of alterations in £2,
23    20-alpha-hydroxyprogesterone, or menstrual cyclicity (Foster et al., 1996b).  In contrast to Foster
24    et al. (1992), this study (Foster et al., 1996b) found no statistically significant effect of Pb on
25    serum progesterone levels in cynomolgus monkeys that had lower PbB (10 to 15 |ig/dL).
26    Additional details concerning these studies are provided in Table 5-4.2.
27          Several modes of action for lead-induced, endocrine disruption-mediated alterations in
28    female reproduction have been proposed, including changes in hormone synthesis or metabolism
29    at the enzyme level (Wiebe and Barr, 1988; Wiebe et al., 1988) and changes in hormone receptor
30    levels (Wiebe et al., 1988; Wide and D'Argy, 1986). In addition, Pb may alter sex hormone
31    release and imprinting during early development (Ronis et al., 1998c;  Tchernitchin et al.,

      December 2005                           5-93        DRAFT-DO NOT QUOTE OR CITE

-------
 1    1998a,b). The latter effects would be consistent with observations of persistent changes in
 2    estrogen receptor levels in the uterus (Wiebe and Barr, 1988) and LH function in the ovary
 3    (Srivastava et al., 2004) in lead-exposed animals.
 4
 5    5.4.3.4  Effects on Morphology and Histology of Female Sex Organs and the Placenta
 6    Lead-induced changes in morphology or histology in female sex organs and the placenta may
 7    explain reduced fertility and impaired female reproductive success (see Sections 5.4.3.2 and
 8    5.4.4.1.).  Logdberg et al. (1988) reported a dose-dependent reduction in placental weight and an
 9    increase in pathological lesions of the placenta in squirrel monkeys that received oral doses of
10    Pb-acetate (0.001 to 0.1% in diet) during the last three-fourths or two-thirds of pregnancy (mean
11    maternal PbB 37 |ig/dL; range: 22 to  82 |ig/dL). These effects occurred without overt toxicity in
12    the mothers.  Additional details concerning Logdberg et al. (1988) are provided in Table 5-4.2.
13          Similar effects on placental weight and histology were observed in mice (Fuentes et al.,
14    1996; Nayak et al., 1989). These effects on the placenta may explain the reduced birth weight
15    that has been associated with prenatal Pb exposure (see Section 5.4.5). Exposure to Pb in early
16    pregnancy also produces structural changes in the epithelium of the uterus of mice (Nilsson
17    et al., 1991; Wide and Nilsson, 1979). These changes in uterine tissue may impair successful
18    implantation of the blastocysts (see Section 5.4.4.1).
19
20    5.4.4   Effects on  Embryogenesis
21          Lead exposure can increase fetal mortality, produce a variety of sublethal effects, and
22    disrupt the growth and development of the offspring. Many of the lead-induced sublethal
23    developmental effects occur at maternal PbB levels that do not result in clinical  toxicity in the
24    mothers.
25
26    5.4.4.1  Embryo/Fetal Mortality
27          The 1986 Pb AQCD concluded that that acute exposure to high doses of Pb interfered
28    with implantation and pregnancy (Wide, 1985; Odenbro and Kihlstrom, 1977; Wide and Nilsson,
29    1977; Vermande-Van Eck and Meigs, 1960). This conclusion is supported by results of more
30    recent studies (Logdberg et al., 1987; Giavini et al., 1980; Jacquet, 1976, 1977; Jacquet et al.,
      December 2005                           5-94        DRAFT-DO NOT QUOTE OR CITE

-------
 1    1975, 1976; Johansson and Wide 1986; Johansson et al., 1987; Johansson, 1989; Maisin et al.,
 2    1978; Pinon-Lataillade et al., 1995; Wide and Nilsson, 1977, 1979).
 3          Logdberg et al. (1987) reported an increase in pre- and perinatal mortality in squirrel
 4    monkeys that received Pb-acetate orally during the last two-thirds of pregnancy (45% versus 7 to
 5    8% among controls). Mean maternal PbB was 54 |ig/dL (39 to 82 |ig/dL).  These fetotoxic
 6    effects occurred without overt toxicity in the mothers. Additional details concerning Logdberg
 7    et al. (1987) are provided in Table 5-4.3. These effects are consistent with data from rodent
 8    studies, wherein gestational exposure to Pb (PbB 32 to >70 |ig/dL) resulted in smaller litters and
 9    fewer implantation sites (e.g., Pinon-Lataillade et al., 1995;  Singh et al., 1993b; Piasek and
10    Kostial,  1991).
11          Numerous studies have been performed to elucidate the mechanisms by which Pb causes
12    prenatal  death (Maisin at al., 1978; Jacquet, 1977, 1976; Jacquet et al.,  1976, 1975).  The
13    available data suggest that Pb may alter blastocyst development and impair implantation.  Hanna
14    et al. (1997) demonstrated that in vitro exposure of 2- and 4-cell mouse embryos to 200 jiM
15    Pb-acetate resulted in reduced cell proliferation and blastocyst formation. Additional evidence
16    for an effect on blastocysts is provided by data from in vitro fertilization studies (Chowdhuri
17    et al., 2001; Johansson, 1989; Johansson et al., 1987). Johansson and co-workers (1989, 1987)
18    reported that Pb delayed the timing of escape from the zona pellucida and induced a premature
19    acrosome reaction. These effects could disrupt attachment and implantation of the blastocyst if
20    they were to occur in vivo.
21
22    5.4.4.2   Effects on embryo/fetal morphology
23          The 1986 Pb AQCD summarized numerous reports that found associations between
24    prenatal  exposure to high doses of Pb and increased incidences of teratogenic effects
25    (particularly tail stunting) in  rodents (Perm and Carpenter, 1967; Dey et al., 2001; Flora and
26    Tandon, 1987; Ronis et al., 1996; Wide, 1985). More recent studies provide additional support
27    for teratogenic effects of Pb in experimental animals  (Flora and Tandon, 1987). Flora and
28    Tandon (1987) demonstrated a dose-dependent effect on the incidence  of tail malformations at
29    > 10 mg/kg i.v. on days 9 to 11 of gestation (PbB 13 to 45 |ig/dL) that occurred only in those
30    dams exhibiting "observable maternal toxicity" (not otherwise specified in the  report). The few
31

      December 2005                           5-95        DRAFT-DO NOT  QUOTE OR CITE

-------
o

O

to
O
O
          Table 5-4.3.  Selected Studies Showing the Effects of Lead on Mammalian Embryogenesis and Development
           Citation
              Species/
              Strain
              Dose/Route/Form/Duration/Group size    Endpoint/Magnitude of Effect/p-value
                                               Blood Lead Concentration
                                               (PbB)
H
6
o

o
H
O
O
H
W
O
O
HH
H
W
Cory-Slechta
et al. (2004)
Rat/Long-
Evans
           Dearth et al.    Rat/Fisher
           (2002)        344
           Flora and
           Tandon
           (1987)
           Fox et al.
           (1991a)
              Rat/Albino
              (NOS)
              Rat/Long-
              Evans
              hooded
                                        Lead acetate in drinking water (150 ppm);
                                        2 months before breeding until the end of
                                        lactation; 14 rats no maternal stress with Pb
                                        exposure, 15 rats no maternal stress with Pb
                                        exposure, 18 rats maternal stress without Pb
                                        exposure, 23 rats maternal stress and Pb
                                        exposure

                                        12 mg/mL Pb-acetate gavage during gestation
                                        and lactation exposure
                                        4 groups: control group, gestation and
                                        lactation exposure, gestation only exposure,
                                        lactation only exposure
                                        10-32 litters per group (NOS)
              Lead nitrate dissolved in water 2-20 mg/kg-d
              i.v. on day 9, 10, 11 of gestation; 6 rats in
              each group (0, 5, 10 , 20, 40 mg/kg lead)
              Lactation exposure via dams exposed to
              0.02 or 0.2% Pb in drinking water from PND
              1 through weaning (PND 21); 8 female pups
              per litter (number of litter unspecified) control
              pups, 8 pups for litter (number of litter
              unspecified) low-level exposure pups, 8 pups
              per litter (number of litter unspecified)
              moderate level exposure pups
Pb alone (in male) (p<0.05) and Pb plus stress
(in females) (p<0.05) permanently elevated
corticosterene levels in offspring
                                                                       Delayed onset of puberty (p<0.05); suppressed
                                                                       serum levels of IGFb LH, and E2 (p<0.001); Pb
                                                                       altered translation and/or secretion of IGF!
                                                                       (p<0.001).
Dose-dependant increase in external malformations
at all doses (p<0.001), particularly tail defects;
dose-dependant decrease in number of live births
at 20 and 400 mg/kg (p<0.001); dose-dependent
increase in number of resorptions per dam at <10
mg/kg (p<0.01).

Long-term, dose-dependent decreases retinal Na/K
ATPase activity in the female offspring (only
female pups were used) (-11%; -26%) (p<0.05).
PbB 30^10 ng/dL
Maternal PbB -40 |ig/dL
Pups PbB as follows:
Gest+lact-38 |ig/dL PND 10
Gest+lact -15 |ig/dL PND 21
Gest+lact -3 |ig/ PND 30
Gest~14|ig/dLPND10
Gest~3|ig/dLPND21
Gest~l |ig/dLPND30
Lact~28|ig/dLPND10
Lact~15|ig/dLPND21
Lact ~3 |ig/dL PND 30

PbB 4.13±0.61 |ig/dL 0 mg/kg
PbB 10.21+0.61 |ig/dL 5 mg/kg
PbB 13.13+0.27 |ig/dL 10 mg/kg
PbB 29.41+0.41 |ig/dL 20 mg/kg
PbB 45.03+0.31 |ig/dL 40 mg/kg
PbB 18.8 ng/dL (0.02%) or
59.4 ng/dL (0.2%) at weaning

-------
o

O

to
O
O
     Table 5-4.3 (cont'd).  Selected Studies Showing the Effects of Lead on Mammalian Embryogenesis and Development
           Citation
               Species/
               Strain
Dose/Route/Form/Duration/Group Size   Endpoint/Magnitude of Effect/p-value
                                               Blood Lead Concentration
                                               (PbB)
 H
 6
 o

 o
 H
O
 O
 H
 W
 O
 O
 HH
 H
 W
Fox et al.
(1997)
                          Rat/Long-
                          Evans
                          hooded
           lavicoli et al.
           (2003)
              Mouse/Swiss
0.02 or 0.2% Pb-acetate in drinking water
from PND 0-PND 21; 8 female pups per litter
control pups; 8 pups per litter low-level
exposure; 8 pups per litter moderate level
exposure (number of litters per dose
unspecified)
Lead acetate in food (0.02, 0.06, 0.11, 0.2, 2,
4, 20, 40 ppm); exposure began 1 day after
mating until litter was 90 days old; one litter
of mice exposed to each dietary concentration
Developmental and adult Pb exposure for 6 weeks
produced age and dose-dependent retinal
degeneration such that rods and bipolar cells were
selectively lost; at the ultrastructural level, all
dying cells exhibit the classical morphological
features of apoptotic cell death; decrease in the
number of rods was correlated with the loss of
rhodopsin content per eye confirming that rods
were directly affected by Pb (p < 0.05); single-
flash rod ERGs and cone ERGs obtained from
lead-exposed rats demonstrated that there were
age- and dose-dependent decreases in the rod a-
wave and b-wave sensitivity and maximum
amplitudes without any effect on cones; in adult
rats exposed to Pb for three weeks, qualitatively
similar ERG changes occurred in the absence of
cell loss or decrease in rhodopsin content
(p < 0.05); developmental and adult Pb exposure
for three and six weeks produced age- and dose-
dependent decreases in retinal cGMP
phosphodiesterase (PDE) activity resulting in
increased cGMP levels (p<0.05); retinas of
developing and adult rats exposed to Pb exhibit
qualitatively similar rod mediated ERG alterations
as well as rod and bipolar apoptotic cell death
(p<0.05)
Similar biochemical mechanism such as the
inhibition of rod and bipolar cell cGMP PDE,
varying only in degree and duration, underlies both
the lead-induced ERG rod-mediated deficits and
the rod and bipolar apoptotic cell death (p<0.05).
Low-level Pb exposure (PbB 2-13 ug/dL) reduced
red cell synthesis (p<0.05); high-level exposure
(PbB 0.6-2 ug/dL) enhanced red cell synthesis
(p<0.05).
PbB weanlings 19±3 (low exposure)
or 59±8 ug/dL (moderate exposure),
adult 7±2 ug/dL (at PND 90)
PbB 0.6 to <2.0 ug/dL or
>2.0-13 ug/dL

-------
o

O

to
O
O
     Table 5-4.3 (cont'd). Selected Studies Showing the Effects of Lead on Mammalian Embryogenesis and Development
                                         Dose/Route/Form/Duration/Group
           Citation   Species/ Strain   size
                                                                       Endpoint/Magnitude of effect/p-value
                                               Blood Lead Concentration
                                               (PbB)
fj\

200 ng/dL
                                                                                                                                       Naive

                                                                                                                                       Control
pel iiiicij Lac
Gest+Lac
Postnatal
Chronic
-3 ng/dL
-13 ng/dL
-260 ng/dL
-287 ng/dL

-------
o

O

to
O
O
     Table 5-4.3 (cont'd). Selected Studies Showing the Effects of Lead on Mammalian Embryogenesis and Development

           Species/
Citation   Strain
                                     Dose/Route/Form/Duration/Group Size   Endpoint/Magnitude of Effect/p-value
                                                                                                           Blood Lead Concentration
                                                                                                           (PbB)
H
6
o

o
H
O
O
H
W
O
O
HH
H
W
Ronis
etal.
(1998b)
Rat/Sprague-
Dawley
           Ronis
           etal.
           (1998c)
           Rat/Sprague-
           Dawley
           Ronis
           etal.
           (2001)
           Rat/Sprague-
           Dawley
Lead acetate in drinking water (0.05% to
0.45% w/v); dams exposed until weaning;
exposure of pups which continued until PND
21,35, 55, or 85
5 control litters (0%), 10 low-dose litters
(0.05%), 8 mid-dose litters (0.15%), 9 high-
dose litters (0.45%)
(4 male and 4 female pups per litter)

Lead acetate 0.05, 0.15, or 0.45% in drinking
water beginning  GD 5 continuing until PND
21, 35, 55, or 85; 5 control litters (0%),
10 low-dose litters (0.05%), 8 mid-dose litters
(0.15%), 9 high-dose litters (0.45%)
(4 male and 4 female pups per litter)
              Lead acetate in drinking water to 825 or
              2475 ppm ad libitum from GD 4 to GD 55
              postpartum; 1 male and female pup/litter
              (5 litters per group) control group, 1 male and
              female pup/litter (5 litters per group) 825 ppm
              Pb-acetate group, 1 male and female pup/litter
              (5 litters per group) 2475 ppm Pb-acetate
              group
                                                                                Prenatal Pb exposure that continues until adulthood
                                                                                (85 days old) delays sexual maturation in female pups
                                                                                in a dose-related manner (p<0.05); birth weight
                                                                                reduced (p<0.05), more pronounced among male pups;
                                                                                decreased growth rates (p<0.05) in both sexes
                                                                                accompanied by decrease in plasma concentrations of
                                                                                IGF] through puberty (p<0.05) and a significant
                                                                                increase in pituitary growth hormone during puberty
                                                                                (p<0.05).

                                                                                Dose-responsive decrease in birth weight (p<0.05),
                                                                                and crown-to-rump length (p<0.05); dose-responsive
                                                                                delay in sexual maturity in male (p<0.05) and female
                                                                                (p<0.05); neonatal decrease in sex steroids (p<0.05);
                                                                                pubertal decrease in testosterone (male) (p<0.05), and
                                                                                E2 (female) (p<0.05); decrease estrous cyclicity at high
                                                                                dose (p<0.05).
                                           Dose-dependent decrease of the load of failure in male
                                           (p<0.05); no difference in plasma levels of vitamin D
                                           metabolites; reduced somatic growth (p<0.05),
                                           longitudinal bone growth (p<0.05, and bone strength
                                           during the pubertal period (p<0.05); sex steroid
                                           replacement did not restore skeletal parameters in Pb
                                           exposed rats; L-Dopa increased plasma IGF]
                                           concentrations, rates of bone growth, and bone
                                           strength measures in controls while having no effect in
                                           Pb exposed groups; DO gap x-ray density and
                                           proximal new endostreal bone formation were
                                           decreased in the distration gaps of the lead-treated
                                           animals (p < 0.01); distraction initiated at 0.2 mm 30
                                           to 60 days of age.
PbBs in the pups between the ages of
21 and 85 days were >100 ug/dL and
reached up to 388 ug/dL
Dams: 0, 48, 88, or 181 ug/dL
Pups PND 1: <1, -40, -70, or
>120 ug/dL
Pups PND 21: <1, >50, >160, or
-237 ug/dL
Pups PND 35: <1, -22, >70, or
>278 ug/dL
Pups PND 55: <1, >68, >137, or
-380 ug/dL
Pups PND 85: <1, >43, >122, or
>214 ug/dL

PbB at 825 ppm was 67-192  ug/dL
PbB at 2475 ppm was 120-388 ug/dL

-------
o

O

to
O
O
     Table 5-4.3 (cont'd).  Selected Studies Showing the Effects of Lead on Mammalian Embryogenesis and Development

           Species/
Citation   Strain
                                   Dose/Route/Form/Duration/Group Size  Endpoint/Magnitude of Effect/p-value
                                                Blood Lead Concentration
                                                (PbB)
           Rat/Sprague-  Lead in drinking water at 34 ppm from
           Dawley       weaning of mothers through gestation and
                        weaning of offspring until birth; 6 pups
                        control group, 6 pups experimental group
Reduced body weight (p = 0.04); parotid function was
decreased by nearly 30% (p = 0.30); higher mean
caries scores than the control pups (p = 0.005); pre-
and perinatal Pb exposure had significantly increased
susceptibility to dental caries (p = 0.015).
PbB48±13|ig/dL
o
o
           GMP, cyclic guanosine—3',5'-monophosphate; DO, distraction osteogenesis; E2, estradiol; ERG, electroretinographic; GD, gestational day; IGF!, insulin-like growth factor 1;
           LH, luteinizing hormone; NOS, not otherwise specified; PbB, blood Pb concentration; PDE, phosphodiesterase; PND, post-natal day
H
6
o

o
H
O
O
H
W
O
O
HH
H
W

-------
 1    studies (including those described in the 1986 Pb AQCD and more recent reports) that have
 2    demonstrated teratogenic effects of Pb exposure are confounded by maternal toxicity.
 3
 4    5.4.5  Effects on Growth and Endocrine Regulation of Growth
 5    Studies conducted in rodents provide convincing evidence for an association between gestational
 6    Pb exposure and reduced birth weight and postnatal growth at doses that produce no clinical
 7    toxicity in the mothers (Dearth et al., 2002; Hamilton et al., 1994; Logdberg et al., 1987; Piasek
 8    and Kostial, 1991; Pinon-Lataillade et al., 1995; Ronis et al., 1998a,b,c; Singh et al., 1993b;
 9    Watson et al., 1997). In squirrel monkeys, Logdberg et al. (1987) reported a statistically
10    significant reduction in mean birth weight following oral exposure to Pb-acetate during the latter
11    trimesters of pregnancy (mean maternal PbB 54 |ig/dL [39 to 82 |ig/dL]). Additional details
12    concerning Logdberg et al. (1987)  are provided in Table AX5-4.3.
13          In addition, the literature provides convincing support for lead-induced impairment of
14    postnatal growth. Although some early studies (Minnema and Hammond, 1994; Hammond
15    et al., 1993, 1990) ascribed the reduction in postnatal growth to reduced food consumption
16    (suggesting an effect of Pb on the satiety endpoint), more recent studies report impaired growth
17    unrelated to changes in food consumption.  Ronis et al. (1996, 1998a,b,c) reported lead-induced
18    reductions in birth weight and postnatal growth that occurred in the absence of a significant
19    alteration in food consumption. Han et al. (2000) found a reduction in the birth length of pups
20    (pup PbB -16 |ig/dL on PND 1) whose mothers had been exposed to Pb up to 1  month before
21    mating (maternal PbB on GD 9, 16, and 21 <40 |ig/dL). Berry et al.  (2002) reported depressed
22    growth in rats exposed to lead, even though food consumption was higher in the lead-exposed
23    rats.
24          Ronis et al. (2001) showed that in rats, pre- and postnatal (through PND 55) exposure to
25    Pb reduced somatic longitudinal bone growth and bone strength during the pubertal period (PbB
26    >67 |ig/dL). These effects could not be reversed by stimulation of the growth hormone axis by
27    supplemental sex hormone.  These results suggest that Pb exposure may impair growth through a
28    mechanism that involves a suppressed pituitary response to hypothalamic stimulation. The
29    mechanism may be related to a reduction in plasma concentrations of IGFi following Pb
30    exposure (Dearth et al., 2002; Ronis et al., 1998b). Dearth et al. (2002) exposed F344 rats to Pb
31    by gavage beginning 30 days before mating and continuing until weaning of the pups at 21 days

      December 2005                           5-101       DRAFT-DO NOT QUOTE OR CITE

-------
 1    of age.  By PND 30, all three groups had PbB <3 |ig/dL and all lead-exposed groups exhibited
 2    decreased serum levels of IGFi, LH, and E2.  Since liver IGFi mRNA was not affected, it
 3    appeared that Pb altered the translation and/or secretion of IGFi, which in turn decreased LH-
 4    releasing hormone at the hypothalamic level. Additional details concerning Dearth et al. (2002)
 5    are provided in Table AX5-4.3. An effect on IGFi also been demonstrated by Ronis et al.
 6    (1998b).
 7
 8    5.4.6   Effects on Other Endocrine Systems during Development
 9         Recent experimental animal studies provide evidence for an interaction between Pb
10    exposure and stress hormones, including glucocorticoids and catecholamines (Cory-Slechta
11    et al., 2004; Yu et al., 1996; Vyskocil et al.,  1991; Saxena et al., 1990). Lead has been reported
12    to increase stress hormone levels (Vyskocil et al., 1991).
13         Cory-Slechta et al. (2004) reported a persistent effect of combined environmental stress
14    (administered as restraint) and maternal Pb exposure (PbB 30 to 40 |ig/dL) on corticosteroid
15    levels in adult offspring. Female adult offspring born to these dams exhibited elevated
16    corticosteroid levels only when Pb exposure was combined with environmental stress, whereas
17    their adult male siblings exhibited elevated corticosteroid levels from Pb exposure  alone. These
18    data suggest that brief exposures to Pb and stress during development may result in persistent
19    changes in the hypothalamic-pituitary-adrenal axis.  Additional details concerning  Cory-Slechta
20    et al. (2004) are provided in Table AX5-4.3.
21         The interplay between Pb and stress hormones is consistent with the findings of Yu et al.
22    (1996) wherein neonatal exposure to Pb (PbB 70 |ig/dL) decreased cold-water swimming
23    endurance (a standard test for stress endurance). The enhancement of lead-induced toxicity by
24    stress was also reported by Saxena et al. (1990) in adult male rats.  Saxena et al. (1990) reported
25    enhanced testicular injury when rats were exposed to immobilization stress in combination with
26    Pb exposure (PbB >200 (ig/dL).
27
28    5.4.7   Effects on Other Organ Systems during Development
29    5.4.7.1  Developmental Effects on Blood and Liver
30         Recent data provides evidence for lead-induced alterations in developing hematopoietic
31    and hepatic system.  The data concerning the effect of Pb  exposure on the developing

      December 2005                           5-102       DRAFT-DO NOT QUOTE OR CITE

-------
 1    hematopoietic system are limited. The 1986 Pb AQCD proposed that alterations in blood ALAD
 2    activity and erythrocyte protoporphyrin were possible biomarkers for subtle, prenatal effects of
 3    Pb on heme synthesis (Hayashi 1983a,b;  Jacquet et al., 1977; Prigge and Greve, 1977;
 4    Hubermont et al., 1976). A more recent study (lavicoli et al., 2003) of Pb effects on RBC
 5    production, HB concentration, and Hct was not able to clearly establish a dose-response
 6    relationship for these endpoints.  Although limited by small group size (one litter per dose),
 7    subchronic dietary exposure to low levels of Pb (PbB 0.6 to 2 or 2 to 13  |ig/dL) revealed that Pb
 8    exposure reduced red cell synthesis, hemoglobin concentration, and hematocrit at PbB 2 to
 9    13 |ig/dL and increased RBC  synthesis, Hb concentration, and Hct at PbB 0.6 to 2 |ig/dL. More
10    data are needed to clarify the effect of low-dose Pb exposure on blood endpoints.
11          Two rodent studies provide limited suggestive evidence that Pb exposure during
12    development produces changes in hepatic enzymes and other biomarkers of hepatic function.
13    Pillai and Gupta (2005) reported  that long-term exposure of rats (pre-mating, gestation, and
14    lactation) to moderate levels of Pb-acetate (subcutaneous injections of 0.05 mg/kg-day; PbB not
15    reported) resulted in reduced activities of hepatic steroid (£2) metabolizing enzymes (17-B-
16    hydroxy steroid oxidoreductase and UDP glucuronyl transferase) and decreased hepatic CYP450
17    content.  Corpas (2002) reported  that exposure to Pb in drinking water exposure during gestation
18    and lactation (pup PbB -22 |ig/dL at PND 12 and PND 21) resulted in alterations in the hepatic
19    systems of neonates (PND 12) and pups (PND 21). The effects manifested as alterations in
20    several biochemical indicators of hepatic toxicity: reductions in Hb, iron, alkaline and acid
21    phosphatase levels, and hepatic glycogen, and elevated blood glucose. These data suggest that
22    Pb may alter hepatic function during development; however, more data are needed to determine
23    whether these effects are persistent.
24
25    5.4.7.2  Developmental Effects on Skin
26          Recent data provides limited evidence of altered soft tissue development resulting from
27    Pb exposure. The literature includes one report of lead-induced abnormalities in skin
28    development. Dey  et al. (2001) reported that the pups of mice exposed orally to Pb citrate
29    (5 |ig/kg-day) throughout gestation exhibited a variety of skin anomalies, including perforations,
30    cell deformity, and  disordered collagen bundles.  The PbB levels  for mothers and pups were not
31    provided. Although detailed biochemical studies are required to elucidate the mechanism for

      December 2005                            5-103       DRAFT-DO NOT QUOTE OR CITE

-------
 1    structural abnormalities, it appears that covalent binding of Pb ions to the sulfate group of
 2    glycosaminoglycans may be involved.
 3
 4    5.4.7.3  Developmental Effects on the Retina
 5          Several studies have found that Pb exposure during early postnatal development impairs
 6    retinal development in female Long-Evans hooded rats (Fox et al., 1997, 1991a,b; Fox and
 7    Rubenstein, 1989; Fox and Chu, 1988).  Of these, two studies are particularly important.  Fox
 8    et al. (1991a) demonstrated that lactation exposure to Long-Evans hooded rats (PbB 18.8 or
 9    59.4 |ig/dL) resulted in long-term, dose-dependent decreases retinal Na/K ATPase activity in the
10    female offspring (only female pups were used). Fox et al. (1997) subsequently demonstrated that
11    lactation exposure to female Long-Evans hooded rats (PbB 19 ± 3 or 59 ± 8 |ig/dL) or drinking
12    water exposure to adult females (PbB 56 ± 9 |ig/dL) resulted in differential age- and dose-
13    dependent alterations in retinal structure and function following low (PbB <20 |ig/dL) and
14    moderate (PbB <60 |ig/dL) exposures during lactation or long-term (-60 days) exposure during
15    adulthood. The mode of action for the effects of Pb on retinal  development may be related to
16    impaired Na/K ATPase activity (Fox et al., 1991a). The observation of reduced enzyme activity
17    in the retina, but not in the kidney, suggests specificity for the retinal alpha-3 isozyme of Na/K
18    ATPase, rather than the renal alpha-1 isozyme of Na/K ATPase. The authors suggested that this
19    specificity may play a role in the target organ-specific toxicity of Pb (Fox et al., 1991a).
20
21    5.4.8   Conclusions
22          The 1986 Pb AQCD presented unequivocal evidence (derived principally from studies of
23    rodents) for effects of Pb on reproduction and development in laboratory animals. This included
24    evidence for lethal effects in developing organisms exposed to Pb during gestation and in the
25    neonatal period, as well  as a variety of sublethal effects on reproduction and development.
26    Sublethal effects included changes in levels or function of reproductive hormones, effects on
27    maturation of reproductive systems, persistent toxic effects on the gonads (both male and
28    female), and adverse effects on the conceptus.  More subtle effects on hormone metabolism and
29    reproductive cell  structure of developing organisms were also documented.
30          More recent studies support earlier conclusions, presented in the 1986 Pb AQCD, that Pb
31    can produce temporary and persistent effects on male and female reproductive function and

      December 2005                          5-104        DRAFT-DO NOT QUOTE OR CITE

-------
 1    development and that Pb disrupts endocrine function at multiple points along the hypothalamic-
 2    pituitary-gonadal axis.
 3          Effects on Male Reproduction. Studies in experimental animals (presented in the 1986 Pb
 4    AQCD and others published subsequent to the 1986 Pb AQCD) provide convincing evidence
 5    that Pb acts as an endocrine disrupter in males. The majority of studies support the conclusion
 6    that endocrine disruption in males involves Pb acting at multiple sites along the hypothalamic-
 7    pituitary-gonadal axis.  The adverse effects of Pb on male reproduction include perturbations in
 8    sexual development and maturation, changes in fertility, changes in male sex hormone levels,
 9    and alterations in gonad tissues and cell structure.
10          Studies conducted in male experimental animals unequivocally demonstrate that Pb
11    exposure during early development (PbB >30 |ig/dL) can delay the onset of puberty and alter
12    reproductive function later in life. Persistent effects of Pb exposure on the male reproductive
13    system may derive from disruption in pulsatile release of sex hormones during early
14    development (Ronis et al., 1998c).
15          The 1986 Pb AQCD reported evidence that Pb exposure affects male fertility in various
16    animal species, including rabbits (Cole and Bachhuber, 1915), guinea pigs (Weller, 1915), rats
17    (Ivanova-Chemishanska et al., 1980), and mice (Schroeder and Mitchener,  1971). More recent
18    studies, conducted in various animal species, have demonstrated lead-induced alteration of sperm
19    parameters (e.g., count, motility, number of abnormal sperm) (Sokol et al.,  1985; Acharya et al.,
20    2003; Adhikari et al., 2000; Foster et al., 1998; Graca et al., 2004; McGivern et al., 1991;  Mishra
21    and Acharya, 2004; Sokol and Berman, 1991).  These effects, however, have not been observed
22    in all studies; the response may be modified by an adaptive mechanism in the hypothalamic-
23    pituitary-gonadal axis.  Lead has also been shown to alter the stability of sperm chromatin in
24    monkeys (PbB 56 |ig/dL) in the absence of gross changes in sperm parameters, a finding which
25    may contribute to a reduction in male fertility (Foster et al.,  1996a).  These results are consistent
26    with observations of reduced in vitro fertilization capacity of sperm collected from rats, rabbits,
27    or mice previously exposed to Pb (Sokol et al., 1994; Foote, 1999; Johansson et al., 1987,
28    respectively). Two modes of action have been proposed for lead-induced alterations in sperm
29    capacity for fertilization: (1) Pb complexation with sulfhydryl groups in sperm, and (2) lead-
30    induced generation of ROS in testes.
      December 2005                           5-105       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Experimental animal studies provide convincing evidence that Pb acts as an endocrine
 2    disrupter in males at various points along the hypothalamic-pituitary-gonadal axis. Although
 3    there is evidence for a common mode of action, consistent effects on circulating testosterone
 4    levels are not always observed in lead-exposed animals. The inconsistency in the reports of
 5    circulating testosterone levels complicates the derivation of a dose-response relationship for this
 6    endpoint.
 7          The 1986 Pb AQCD reported evidence for histological changes in the testes and prostate
 8    in rats in association with relatively high doses of Pb (Chowdhury et al., 1984; Hilderbrand et al.,
 9    1973; Golubovich et al., 1968). More recent studies in animals provide additional support for
10    testicular damage (i.e., ultrastructural changes in testes and cytotoxicity in Sertoli cells)
11    following exposure to Pb (Foster et al.,  1998; Singh et al.,  1993a; Batra et al., 2001; Chowdhury
12    et al., 1986, 1987; Corpas et al., 1995; Graca et al., 2004; Pinon- Lataillade et al., 1993; Saxena
13    et al., 1990). Foster et al. (1998) and Singh et al.  (1993a) demonstrated ultrastructural changes in
14    testes of monkeys at PbB 35 to 40 |ig/dL. Lead-induced oxygen free radical generation is the
15    plausible mechanism of testicular injury in primates (Foster et al., 1998) and rodents
16    (Chowdhury et al., 1984; Acharya et al., 2003; Adhikari et al., 2001; Batra et al., 2001; Bizarro
17    et al., 2003; Chowdhury et al., 1984; Gorbel et al., 2002; Mishra and Acharya, 2004).
18          Effects on Female Reproduction.  In females, Pb exposure has been consistently shown to
19    disrupt the hypothalamic-pituitary-gonadal axis and to produce reproductive dysfunction.  The
20    1986 Pb AQCD reported that Pb exposure was associated with inhibition of menstruation,
21    ovulation, and follicular growth in monkeys (Vermande-Van Eck and Meigs, 1960) and, in
22    rodents, delayed vaginal opening, decreased frequency of implantation, and  reduced rates of
23    pregnancy (Kimmel et al., 1980; Odenbro and Kihlstrom, 1977). Observations from more recent
24    experimental animal studies support these findings.  The effects of Pb on female reproduction
25    may be classified as alterations in female sexual maturation, effects on fertility and menstrual
26    cycle, alterations in levels of female sex hormones, and changes in morphology or histology of
27    female reproductive organs as well as the placenta.
28          The 1986 Pb AQCD reported that Pb exposure (PbB 20 to 40 (ig/dL) in rodents produced
29    delays in sexual maturation.  More recent studies  in experimental animals (primarily rodent
30    studies) provide convincing evidence that Pb exposure before puberty (prenatal and early
31    postnatal PbB -40 |ig/dL) delays maturation of the female reproductive system (Dearth et al.,

      December 2005                           5-106       DRAFT-DO NOT QUOTE OR CITE

-------
 1    2002, 2004; lavicoli et al., 2004; McGivern et al., 1991; Ronis et al., 1998a,b,c,). Ronis et al.
 2    (1998c) suggested that lead-induced disruption of pulsatile release of sex hormones may result in
 3    delayed onset of puberty.
 4          Numerous studies were described in the 1986 Pb AQCD that supported the conclusion
 5    that Pb was an endocrine disrupter in females. More recent studies in various mammalian
 6    species provide convincing support for endocrine-mediated alterations of the female reproductive
 7    system. The nonhuman primate studies provide dose-response information concerning the
 8    effects of Pb on female sex hormones and menstrual cycle (Foster et al., 1996b; Foster, 1992;
 9    Foster et al., 1992; Franks et al., 1989; Laughlin et al., 1987). Exposures of monkeys to Pb
10    resulting in chronic PbB <20 |ig/dL produce few effects on circulating hormone levels and do
11    not alter the menstrual cycle.  Higher exposures of monkeys to Pb (PbB >40 |ig/dL) alter
12    circulating hormone levels and the menstrual cycle, with more marked changes in these
13    endpoints occurring at higher PbB levels.  Several modes of action for lead-induced alterations in
14    female reproduction have been proposed, including changes in hormone synthesis or metabolism
15    (Wiebe and Barr, 1988; Wiebe et al., 1988) and changes in hormone receptor levels (Wiebe
16    et al., 1988; Wide and D'Argy, 1986).  In addition, Pb may alter sex hormone release and
17    imprinting during early development (Ronis et al., 1998c; Tchernitchin et al.,  1998a,b).
18          The 1986 Pb AQCD presented convincing evidence from experimental animal studies for
19    lead-induced alterations in female fertility, including interference with implantation and
20    pregnancy.  More recent studies have confirmed that Pb exposure disturbs  female fertility;
21    however, Pb exposure does not generally produce total sterility.  Studies in nonhuman primates
22    and rodents have also demonstrated reductions in litter size, implantation dysfunction, and
23    decreased postnatal survival following Pb exposure of gravid female experimental animals (PbB
24    >30 (ig/dL) (Logdberg et al., 1987; al-Hakkak et al., 1988; Flora and Tandon, 1987; Piasek and
25    Kostial, 1991; Pinon-Lataillade et al., 1995; Ronis et al., 1996; Singh et al., 1993b; Wide, 1985).
26          Lead-induced changes in morphology or histology in female  sex organs and placenta may
27    explain reduced fertility and impaired female reproductive success.  Logdberg et al. (1988)
28    reported a dose-dependent reduction in placental weight and an increase in pathological lesions
29    of the placenta in squirrel monkeys that consuming Pb-acetate in their diet during the last three-
30    fourths or two-thirds of pregnancy (maternal PbB 37 |ig/dL). Exposure to Pb  in early pregnancy
31    also produces structural changes in the epithelium of the uterus of mice (Nilsson et al., 1991;

      December 2005                            5-107        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Wide and Nilsson, 1979).  These changes in uterine tissue may impair successful implantation of
 2    the blastocysts. In addition, the histological and morphological effects on the uterus and placenta
 3    may explain the reduced birth weight that has been associated with prenatal Pb exposure
 4    (possibly due to placental insufficiency).
 5          Developmental Effect.  Pre- and postnatal exposure to Pb has been demonstrated to result
 6    in fetal mortality and produce a variety of sublethal effects in the offspring. Many of these lead-
 7    induced  sublethal developmental effects occur at maternal PbB that do not result in clinical
 8    (overt) toxicity in the mothers. The few studies that have reported teratogenic effects resulting
 9    from Pb  exposure are confounded by maternal toxicity.
10          Studies conducted in rodents and primates provide convincing evidence for an association
11    between Pb exposure and reduced birth weight and postnatal growth at doses that produce no
12    clinical toxicity in the mothers (maternal PbB >40 |ig/dL) (Dearth et al., 2002; Logdberg et al.,
13    1987; Berry et al., 2002; Bogden et al., 1995; Camoratto et al., 1993 Hamilton et al., 1994;
14    Hammond et al., 1989, 1990,  1993; Minnema and Hammmond 1994; Han et al., 2000; Ronis
15    et al., 1996, 1998a,b,c; Piasek and Kostial,  1991; Pinon-Lataillade et al., 1995; Sant'Ana et al.,
16    2001; Singh et al., 1993b; Watson et al.,  1997). The available data suggest that the mode of
17    action for lead-induced growth suppression involves a reduction in the plasma concentration
18    oflGFi.
19          Recent experimental animal studies provide evidence for an interaction between Pb
20    exposure during development (PbB 30 to 40 |ig/dL) and stress hormones, including
21    glucocorticoids and catecholamines (Cory-Slechta et al., 2004; Yu et al., 1996; Vyskocil et al.,
22    1991; Saxena et al., 1990).  Lead exposure during early postnatal development (PbB -20 |ig/dL)
23    impairs retinal development in female Long-Evans hooded rats (Fox et al., 1997, 1991a,b; Fox
24    and Rubenstein, 1989;  Fox and Chu, 1988).
25          In addition, recent studies provide limited evidence for lead-induced alterations in
26    developing skin, and hematopoietic and hepatic systems; however, more data are needed to
27    clarify the effect of low-dose Pb exposure on these endpoints.
28
      December 2005                           5-108       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.5   CARDIOVASCULAR EFFECTS OF LEAD
 2    5.5.1   Introduction
 3         Numerous large and small epidemiological studies have attempted to examine the link
 4    between Pb exposure and development of hypertension (HTN) in the general population and
 5    occupationally-exposed individuals. In addition, a number of studies have reported on other
 6    cardiovascular effects of Pb in Pb-exposed humans (U.S. Environmental Protection Agency,
 7    1990). While several studies have demonstrated a positive correlation between blood pressure
 8    and blood Pb concentration, others have failed to show such association when controlling for
 9    confounding factors such as tobacco smoking, exercise, body weight, alcohol consumption, and
10    socioeconomic status. Thus,  the studies that have employed blood Pb level as an index of
11    exposure have shown a relatively weak association with blood pressure. In contrast, the majority
12    of the more recent studies employing bone Pb level have found a strong association between
13    long-term Pb exposure and arterial pressure (Chapter 6). Since the residence time of Pb in the
14    blood is relatively short but very long in the bone, the latter observations have provided
15    compelling evidence for the positive relationship between Pb exposure and a subsequent rise in
16    arterial pressure. This section reviews the published studies pertaining to the cardiovascular
17    effects of Pb exposure in experimental animals, isolated vascular tissues, and cultured vascular
18    cells.
19
20    5.5.2   Lead Exposure and Arterial Pressure in Experimental Animals
21         Numerous studies have shown that exposure to low levels of Pb for extended periods
22    results in a delayed onset of arterial HTN that persists long after  the cessation of Pb exposure in
23    genetically normal animals (see Tables AX5-5.1 to AX5-5.5). In addition, Pb exposure during
24    gestation has been reported to significantly raise arterial pressure in the third trimester of
25    pregnancy in SD rats given a low calcium diet (Bogden et al.,  1995).  Taken together, these
26    observations provide irrefutable evidence that extended exposure to low levels of Pb  can result in
27    the subsequent onset of HTN in experimental animals.
28         Many studies have been conducted to explore the mechanisms by which chronic Pb
29    exposure may cause HTN.  Most of these studies have examined various blood-pressure
30    regulatory and vasoactive systems in animal models of Pb-induced HTN. In addition, several


      December 2005                           5-109       DRAFT-DO NOT QUOTE OR CITE

-------
 1    studies have investigated the direct effect of Pb on vascular tone or the ability of Pb to modify
 2    the response to vasoconstrictor/vasodilator agents in isolated vascular tissues. Finally, a number
 3    of studies have explored the effect of Pb on cultured endothelial and vascular smooth muscle
 4    cells.  An overview of the findings of these studies is provided below:
 5
 6    5.5.2.1  Effect of Lead on Production of Reactive Oxygen Species and Nitric
 7            Oxide Metabolism
 8          Reactive oxygen species (ROS), such as, superoxide (02), hydroxyl radical (OH) and
 9    hydrogen peroxide (H2O2) are normally produced in the course of metabolism and are safely
10    contained by the natural antioxidant defense system.  Excess production and/or diminished
11    containment of ROS can  lead to oxidative stress in which uncontained ROS can attack and
12    denature functional/structural molecules and, thereby, promote tissue damage, cytotoxicity, and
13    dysfunction. In fact, oxidative stress has been implicated in the pathogenesis of HTN,
14    atherosclerosis, neurodegenerative disorders, aging, and neoplasm among other afflictions.
15    During the past decade, several studies have demonstrated that Pb exposure causes oxidative
16    stress, particularly in the  kidney and cardiovascular tissues, as well as in cultured endothelial and
17    vascular smooth muscle cells (VSMC). The in vivo studies have further shown that Pb-induced
18    oxidative stress is, at least in part, responsible for the associated HTN in experimental animals.
19    Relevant published  studies pertaining to this issue are summarized below and listed in Annex
20    Table AX5-5.1.
21          Khalil-Manesh et al. (1994) were among the first to suggest that oxidative stress may be
22    involved in the pathogenesis of Pb-induced HTN. This assumption was based on the observation
23    that chelation therapy with dimethyl succinic acid (DMSA) rapidly ameliorated HTN and raised
24    plasma cGMP level in rats with Pb-induced HTN. They further demonstrated that DMSA
25    possesses strong antioxidant properties in vitro. Accordingly, they theorized (a) that Pb exposure
26    may increase the generation of ROS, which, in turn, elevate arterial pressure by reacting with and
27    inactivating endothelium-derived-relaxing factor (EDRF), and (b) that by scavenging ROS,
28    DMSA rapidly lowers blood pressure prior to significantly affecting body Pb burden.
29          In a subsequent study, Gonick et al.  (1997) showed a marked increase in renal tissue
30    content of lipid peroxidation product malondialdehyde (MDA) coupled with significant
31    upregulations of endothelial (eNOS) and inducible (iNOS) nitric oxide synthases.  Thus, the

      December 2005                           5-110       DRAFT-DO NOT QUOTE OR CITE

-------
 1    study provided evidence for the occurrence of oxidative stress and compensatory upregulation of
 2    NOS isotypes in the kidney of animals with Pb-induced HTN.
 3          In another study, Ding et al. (1998) showed that infusion of NOS substrate, L-Arginine,
 4    lowers blood pressure to a much greater extent in rats with Pb-induced HTN than that seen in
 5    either control  animals or DMSA-treated Pb-exposed animals. The data, therefore, provided
 6    indirect evidence for the role of depressed NO availability in the pathogenesis of Pb-induced
 7    HTN. The study further suggested that oxidative stress may be responsible for diminished NO
 8    availability in this model.  It should be noted that administrating cell-impermeable native SOD
 9    did not lead to a further reduction of blood pressure beyond that seen with L-Arginine alone.
10    As with the previous study (Khalil-Manesh 1994), oral DMSA therapy for 2 weeks significantly
11    lowered blood pressure in the Pb-exposed animals.  This was accompanied by a significant
12    reduction of blood Pb concentration.  In an attempt to explore whether the observed amelioration
13    of Pb-induced HTN was due to the reduction of Pb burden or alleviation of oxidative stress by
14    DMSA, Vaziri et al. (1997) carried out a study in which rats with Pb-induced HTN were treated
15    with a lazaroid compound, a potent, non-chelating antioxidant.  The study revealed marked
16    elevation of blood pressure and oxidative stress (increased lipid peroxidation) and reduced NO
17    availability (depressed urinary NO2 + NOs excretion) in the untreated rats with Pb-induced HTN.
18    Antioxidant therapy with the lazaroid compound resulted in a significant alleviation of oxidative
19    stress, improved NO availability, and a marked attenuation of HTN without affecting blood Pb
20    concentration. Thus, the latter study provided convincing evidence for the role of oxidative
21    stress as a major mediator of Pb-induced HTN. The study further demonstrated that Pb-induced
22    HTN is associated with diminished NO availability and that the latter was mediated by oxidative
23    stress. The reduction in NO availability observed in rats with Pb-induced HTN (Pb-acetate,
24    100 ppm in drinking water for 12 weeks) was recently confirmed by Dursun et al. (2005) in rats
25    treated with daily IP injection of Pb-acetate (8 mg/Kg)  for 2 weeks.  The authors showed that the
26    rise in arterial pressure was accompanied by a significant reduction of urinary NO2 + NOs
27    excretion and a significant fall in renal blood flow (indicating increased renal vascular
28    resistance), mimicking the effect of the NOS inhibitor LNAME.
29          To further explore the cause for the observed reduction of NO availability, Vaziri et al.
30    (1999a) subsequently studied the expression of eNOS and iNOS in the kidney and cardiovascular
31    tissues of rats with Pb-induced HTN. The study showed that the reduction in NO availability is

      December 2005                           5-111       DRAFT-DO NOT QUOTE OR CITE

-------
 1    paradoxically associated with a significant upregulation of NOS isotypes. Moreover, in vitro
 2    incubation experiments revealed no significant change in NOS activity in the presence of lead.
 3    Interestingly, antioxidant therapy with pharmacological doses of vitamin E and ascorbic acid
 4    reversed the upregulation of NOS isotypes and paradoxically raised NO availability in the
 5    subgroup of rats with Pb-induced HTN (Vaziri et al., 1999a).  These observations were
 6    subsequently confirmed by Vaziri and Ding (2001) who showed marked reduction of NO
 7    availability despite significant upregulations of eNOS, nNOS, and iNOS in the aorta, heart,
 8    kidney, and brain of rats with Pb-induced HTN and their normalization with the administration
 9    of superoxide-scavenger tempol (15 mg/Kg IP/day) for 2 weeks.  It is noteworthy that tempol
10    administration had no effect on the measured parameters in the control animals. Taken together,
11    these observations indicated that ROS-mediated NO inactivation and, hence, depressed NO
12    availability, results in a compensatory upregulation of NOS isotypes in animals with Pb-induced
13    HTN. This phenomenon is consistent with other studies from this group, which have
14    demonstrated the presence of a negative-feedback regulation of eNOS by NO (Vaziri and Wang,
15    1999; Vaziri  etal., 2005).
16          The occurrence of compensatory upregulation of NOS by oxidative stress in Pb-exposed
17    intact animals described above was subsequently replicated by Vaziri and Ding (2001) in
18    cultured human endothelial cells incubated in media containing different concentrations of Pb-
19    acetate (versus control media containing sodium acetate). Once again, co-incubation with
20    tempol prevented this phenomenon. This study confirmed the ability of Pb to affect endothelium
21    independently of its effects on humoral or hemodynamic factors, which are operative in vivo.
22    Taken together, these observations  suggest that Pb-induced reduction of biologically-active NO
23    is not due to the reduction of NO-production capacity. Instead, it is linked to oxidative stress. In
24    an attempt to explore this supposition, in a separate study, Vaziri et al. (1999b), tested the
25    hypothesis that avid inactivation and sequestration of NO by ROS may be, in part, responsible
26    for the reduction of NO availability in animals with Pb-induced HTN.  To this end, they tested
27    for the presence of immunodetectable nitrotyrosine in kidney, brain, and cardiovascular tissues
28    harvested from untreated and antioxidant-treated (vitamin E + vitamin C) rats with Pb-induced
29    HTN and normal control rats. Nitrotyrosine was used as a marker of NO oxidation by ROS (NO
30    + 02' —»• ONOO , ONOO + tyrosine —»• nitrotyrosine).  The study showed an overabundance of
31    nitrotyrosine in all plasma and tested tissues in the untreated rats with Pb-induced HTN.

      December 2005                           5-112       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Antioxidant therapy reduced nitrotyrosine abundance, attenuated HTN, and simultaneously
 2    raised NO availability in the subgroup of rats with Pb-induced HTN but had no effect on the
 3    normal control group. These observations provided compelling evidence that Pb-induced HTN
 4    causes oxidative stress, which, in turn, promotes functional NO deficiency via ROS-mediated
 5    NO inactivation.  The latter, in turn, participates in the development and maintenance of HTN
 6    and cardiovascular abnormalities.  In addition, the formation of the highly cytotoxic reactive
 7    nitrogen species, peroxynitrite (ONOO ), from the NO-ROS interaction and the associated
 8    nitrosative stress could potentially contribute to the long-term cardiovascular, renal, and
 9    neurological consequences of Pb exposure.
10          In a series of subsequent studies Vaziri et al. (2003) explored the expression of NAD(P)H
11    oxidase (which is a well-recognized source of ROS in, not only, the immune cells but also in
12    renal, cardiovascular, and neuronal tissues) in animals with Pb-induced HTN. In addition,
13    expression of the main antioxidant enzymes, namely Mn and CuZn-superoxide dismutases
14    (SOD), catalase and glutathione peroxidase were investigated.  The study revealed significant
15    upregulation the gp91phox subunit of NAD(P)H oxidase in the brain as well as a trend for higher
16    levels in the renal cortex and left ventricle of rats with Pb-induced HTN. This was accompanied
17    by a significant compensatory upregulation of CuZn SOD in the kidney and brain, and of Mn
18    SOD in the heart, of rats with Pb-induced HTN.  In contrast, despite the presence  of oxidative
19    stress, catalase and glutathione peroxidase activity levels were unchanged. In a more recent
20    study, Farmand et al. (2005), showed a significant increase in CuZn SOD activity with no change
21    in either catalase or glutathione peroxidase activity in the aorta of rats with Pb-induced HTN
22    compared with control animals. Since the latter enzymes are responsible for the reduction of
23    H2O2 and lipoperoxides, the lack of an appropriate rise in their tissue levels may contribute to the
24    severity of oxidative stress in Pb-exposed animals.
25          The contribution of oxidative stress in the pathogenesis of HTN in this model was
26    confirmed by experiments which demonstrated normalization of arterial pressure with the
27    infusion of superoxide-scavenger, tempol, in  rats with Pb-induced HTN (but no change was
28    observed in the blood pressure in the control rats) (Vaziri et al., 2003). As noted above, the
29    relative reduction of tissue catalase and glutathione peroxidase, which are responsible for the
30    reduction of H2O2 to water and molecular oxygen (2H2O2—CAT  > 2H2O+ O2), can result in
      December 2005                           5-113       DRAFT-DO NOT QUOTE OR CITE

-------
 1    accumulation of H2O2. H2O2 serves as a cellular growth signal, as well as a substrate for
 2    hydroxyl radical (-OH) generation. The former action can potentially contribute to
 3    cardiovascular remodeling, whereas the latter can promote oxidative injury. In a recent study,
 4    Ni et al. (2004) demonstrated a transient rise in O2 production followed by a sustained rise in
 5    H2O2 production by human coronary endothelial and vascular smooth muscle cells cultured in
 6    media containing Pb-acetate versus the control media containing Na-acetate.  This was
 7    accompanied by, and primarily due to, upregulation of NAD(P)H oxidase and SOD together with
 8    reduced or unchanged catalase and glutathione peroxidase levels. Accordingly, the results of this
 9    in vitro study confirmed the findings of the in vivo studies and validated the anticipated
10    accumulation of H2O2.
11          As noted above, H2O2 is the substrate for the Fenton and Haber-Weiss reactions, which
12    culminate in formation of the highly cytotoxic-OH (H2O2 + e  —>• OH + OH"). Thus,
13    accumulation of H2O2 in animals with Pb-induced HTN can facilitate-OH production and,
14    thereby, promote oxidative stress and tissue injury.  This supposition was confirmed in a series
15    of studies by Ding et al. (2001), who showed increased hydroxyl radical production in rats with
16    Pb-induced HTN. Oxidative stress, HTN, and excess hydroxyl radical production were all
17    reversed with IV infusion of the reputed hydroxyl radical scavenger, DMTU, in the Pb-exposed
18    animals. Increased hydroxyl radical production observed in intact animals with Pb-induced HTN
19    was confirmed in lead-treated cultured endothelial cells (Ding et al., 2000). The role of oxidative
20    stress in the pathogenesis of HTN and endothelial dysfunction (depressed NO availability) has
21    been substantiated by a number of other investigators.  For instance, Attri et al. (2003),
22    demonstrated that exposure to Pb for up to 3 months resulted in a significant rise in arterial
23    pressure, which was  substantially ameliorated by coadministration of the antioxidant vitamin
24    ascorbic acid (20 mg/rat) in Wistar-Kyoto rats.  The rise in arterial pressure in lead-treated rats
25    was accompanied by diminished NO availability (low plasma NO2 + NOs) and biochemical
26    evidence of oxidative stress, i.e., elevations  of plasma MDA, a DNA oxidation product
27    (8-hydroxyguanosine), and diminished ferric-reducing antioxidant power, as well as
28    electrophoretic evidence of DNA damage.  Amelioration of HTN by antioxidant therapy was
29    accompanied by improved NO availability (plasma NO2 + NOs), marked attenuation of oxidative
30    stress, and partial reduction of DNA damage in this model. In another study, Malvezzi et al.
31    (2001) showed partial amelioration of HTN in Pb-exposed rats with the administration of either

      December 2005                           5-114        DRAFT-DO NOT QUOTE OR CITE

-------
 1   DMSA or L-arginine and showed a much greater response with the combination thereof. These
 2   observations support the role of interaction of ROS and NO in the pathogenesis of Pb-induced
 3   HTN in the rat.
 4          As cited above, Pb-induced HTN is associated with and is, at least in part, due to ROS-
 5   mediated inactivation and hence, reduced availability of biologically active NO. Many of the
 6   biological actions of NO are mediated by cGMP, which is produced from the substrate GTP by
 7   the cytosolic enzyme soluble guanylate cyclase (sGC). sGC is expressed in VSMC and several
 8   other cell types. The enzyme is activated by NO to produce cGMP, which, in turn, promotes
 9   vasorelaxation by lowering cytosolic Ca2+ concentrations. In an earlier study, Khalil-Manesh
10   et al. (1993) demonstrated a significant reduction of plasma and urinary cGMP in rats with Pb-
11   induced HTN.  These observations prompted a number of studies to evaluate the effect of Pb on
12   sGC expression and cGMP production in vascular tissues obtained from rats with Pb-induced
13   HTN or in normal vascular tissues incubated in Pb-containing media. For instance, Marques
14   et al. (2001) found significant reductions of acetylcholine- and Na-nitroprusside-induced
15   vasorelaxation, despite upregulation of eNOS, in the aorta of rats with Pb-induced HTN. This
16   was associated with marked downregulation of sGC abundance and diminished cGMP
17   production in the aorta. In an attempt to explore the possible role of oxidative stress in Pb-
18   induced downregulation of sGC, they included a group of rats that were co-treated with Pb and
19   the antioxidant vitamin ascorbic acid. Antioxidant therapy ameliorated HTN, restored
20   vasorelaxation response to acetylcholine and Na-nitroprusside, and normalized sGC  expression
21   and cGMP production. The authors,  therefore, identified diminished sGC as another mechanism
22   by which Pb exposure can promote endothelial dysfunction and HTN.  They further  showed that
23   Pb-induced downregulation of sGC is mediated by oxidative stress, as evidenced by  its
24   prevention with antioxidant therapy.  Downregulation of sGC protein abundance in the aorta of
25   Wistar rats with Pb-induced HTN was recently confirmed by Farmand et al. (2005) in the
26   Pb-exposed Sprague-Dawley rats.  In another study, Courtois et al. (2003) showed that 24-h
27   incubation of normal rat aorta in the lead-containing media resulted in a concentration-dependent
28   downregulation of sGC (beta subunit), with the maximum effect observed at 1 ppm
29   concentration.  This was associated with increased O2  production and upregulation of
30   cyclooxygenase-2 (COX-2) expression.  Co-incubation with ascorbic acid reduced COX-2
31   expression and Q^  production and attenuated, but did not fully prevent, the Pb-induced

     December 2005                          5-115       DRAFT-DO NOT QUOTE OR CITE

-------
 1    downregulation of sGC.  Similarly, addition of COX-2 inhibitor Rofecoxib or of protein kinase
 2    A inhibitor (H-89) partially mitigated the Pb-induced downregulation of sGC in vitro. However,
 3    the COX-2 inhibitor failed to reduce C>2  production in Pb-exposed vascular tissues. Based on
 4    these observations, the authors concluded that Pb exposure downregulates vascular tissue sGC
 5    abundance via induction of oxidative stress and upregulation of COX-2.
 6          Oxidative stress and altered NO metabolism can potentially trigger a cascade of events
 7    that work in concert to promote HTN and cardiovascular disease in Pb-exposed organisms.
 8    Some of these potential links are illustrated in Figure 5-5.1.
 9
10    5.5.2.2  Protein Kinase C, Inflammation, NFKB Activation and Apoptosis
11          Protein kinase C (PKC) isoforms belong to a family of serine-threonine kinases, which
12    serve numerous diverse cellular functions. For instance, PKC is involved in regulating vascular
13    contractility, blood flow, permeability, and cell growth. In this regard, the activation of PKC has
14    been shown to cause vascular contraction and Pb exposure has been found to raise PKC activity.
15    For example, Hwang et al. (2002) found increased PKC activity in the erythrocytes of a group of
16    Pb-exposed Korean workers, and Markovac and Goldstein (1988b) showed a significant increase
17    in PKC activity in rat brain micro vessels following exposure to micromolar concentrations of
18    Pb.  Also, Watts et al. (1995) demonstrated that Pb-acetate (10~10 to 10~3 M) caused contraction
19    in an isolated rabbit mesenteric artery  preparation. This Pb-induced vasoconstriction was
20    unaffected by denudation of endothelium, while it was significantly potentiated by PKC  agonists
21    and attenuated by a PKC inhibitor. Calcium channel blockade with verapamil attenuated, but did
22    not abolish, Pb-induced vasoconstriction. These findings were considered to indicate that
23    activation of PKC is, in part, responsible for Pb-induced vasoconstriction, independently of
24    endothelium or extracellular influx of calcium. Taken together, these observations suggest that
25    the activation of PKC in the vascular smooth muscle cells  may, in part, contribute to the
26    pathogenesis of Pb-induced HTN by enhancing vascular contractility. It should be noted,
27    however, that Pb-induced contraction has been shown to be unaffected by a PKC inhibitor in the
28    rat aorta rings (Valencia 2001). Thus, the contribution of PKC activation to the Pb-induced
29    alteration of vascular contractility appears to be both vessel- and species-specific. It is of note,
30    that at high concentrations, Pb can reduce PKC activity in  certain cell types, including mouse
31    macrophages and rat brain cortex (reviewed by Watts et al. [1995]).

      December 2005                           5-116       DRAFT-DO NOT QUOTE OR CITE

-------
                                      Chronic Lead Exposure
Figure 5-5-1.
                  NO inactivation
                  I NO production
                 (eNOS uncoupling,
                  BH4 depletion)
                 J, NO bioavailability
                  Vasoconstriction,
                 Renal Na retention
                f Sympathetic activity
                      RAS
                    Activation
                                         Inflammation
                                                  Oxidation of
                                                arachidonic acid
                                                                 Isoprostanes
                                                Vasoconstriction,
                                               Renal Na retention
                      Endothelial dysfunction
                         Cardiovascular
                          remodeling,
                            Platelet
                           activation
                                             II
                          Hypertension
                         Arteriosclerosis,
                          Thrombosis

This illustration depicts some of the potential mechanisms by which
oxidative stress may participate in the pathogenesis of Pb-induced HTN
and cardiovascular complications. In the presence of oxidative stress,
uncontained reactive oxygen species (ROS) inactivate nitric oxide (NO),
deplete NO synthase cofactor (tetrahydrobiopterin), uncouple eNOS,
promote generation of isoprostanes by oxidizing aracidonic acid, and
activate the redox-sensitive transcription factor NFidJ. Together, these
events can cause Vasoconstriction, salt retention, sympathetic system
activation, renin-angiotensin system stimulation, platelet adhesion, and,
thereby, endothelial dysfunction, hypertension (HTN), inflammation,
arteriosclerosis, and thrombosis.
December 2005
                            5-117
DRAFT-DO NOT QUOTE OR CITE

-------
 1          As noted earlier, Pb exposure results in oxidative stress in cultured VSMC and endothelial
 2    cells, as well as in intact animals.  Oxidative stress can promote the activation of the nuclear
 3    transcription factor Kappa B (NFicB) and, thereby, trigger inflammation and apoptosis.  In this
 4    context, Ramesh et al. (2001) showed that exposure to low Pb levels (50 ppm in drinking water)
 5    for 90 days activates NFicB and capsases in the rat brain. It is of note that several studies have
 6    revealed the presence of renal tubulointerstitial infiltration of activated T cells, macrophages, and
 7    angiotensin II (Ang-II) producing cells in various forms of genetic and acquired HTN in
 8    experimental animals.  Moreover, the associated tubulointerstitial inflammation has been shown
 9    to contribute to the pathogenesis of HTN in these disorders (Rodriguez-Iturbe, 2004). These
10    abnormalities are accompanied by activation of the redox-sensitive NFicB, which can account for
11    the associated inflammation (reviewed by Rodriguez-Iturbe et al. [2004]). The NFicB activation,
12    the accompanying inflammation, and HTN are ameliorated by antioxidant therapy in these
13    models, pointing to the role of oxidative stress in this process. In a recent study, Rodriguez-
14    Iturbe, et al. (2005) observed marked activation of NFicB coupled with tubulointerstitial
15    accumulation of activated T-cells, macrophages, and Ang-II-producing cells, as well  as increased
16    apoptotic cells in the kidneys of Pb-exposed rats (100  ppm Pb-acetate in water for 3 months).
17    This was associated with increased nitrotyrosine staining (a marker of NO/ROS interaction) in
18    the kidney tissue. Since tubulointerstitial inflammation plays a crucial role in the pathogenesis
19    of HTN in various other models of HTN, its presence  in the Pb-exposed animals may contribute
20    to the associated HTN. Inflammation in Pb-induced HTN is not limited to the kidney.  In fact,
21    lymphocyte infiltration is reported in the periaortic tissues in rats with Pb-induced HTN
22    (Carmignani et al 2000).  The inflammatory response to Pb exposure in the renal and vascular
23    tissues outlined above parallels the observations reported with immune system in Section 5.9 of
24    this chapter.
25
26    5.5.2.3  Effect of Lead Exposure on the Adrenergic System
27          The adrenergic system plays an important role in regulating arterial pressure, renal and
28    systemic hemodynamics, and cardiac function in health and disease. For this reason, a number
29    of clinical and animal studies have focused on the sympathetic system as a possible mediator of
30    Pb-induced HTN and cardiovascular abnormalities. For instance, in a study of a group  of Pb-
31    exposed workers, Chang  et al.  (1996), found elevated  plasma norepinephrine (NE), but normal

      December 2005                           5-118       DRAFT-DO NOT QUOTE OR CITE

-------
 1    plasma dopamine and epinephrine, levels.  The constellation of these biochemical abnormalities
 2    points to increased sympathetic nervous system activity in Pb-exposed humans.  The impact of
 3    Pb exposure on the sympathetic nervous system activity has been substantiated in experimental
 4    animals. For example, Chang et al. (1997) showed that administration of Pb (Pb-acetate 0.5% in
 5    drinking H^O) for 2 months resulted in significant rises in arterial pressure and plasma NE (but
 6    not epinephrine) in Wistar rats.  This was coupled with significant reductions of the aorta P
 7    adrenergic receptor density and isoproterenol (P agonist)-stimulated cAMP production. In a
 8    subsequent study Tsao et al. (2000) reported a significant rise in plasma NE coupled with marked
 9    reductions of P receptor density as well as diminished basal and isoproterenol-stimulated cAMP
10    productions in the aorta and heart of Wistar rats with Pb-induced HTN. In contrast to the heart
11    and aorta, P receptor density as well as basal and  P agonist-stimulated cAMP production were
12    increased in the kidneys of Pb-exposed animals.
13          In another study, Carmignani et al. (2000)  found significant elevations of blood pressure,
14    plasma catecholamines, and cardiac contractility (dP/dt), together with reduced carotid blood
15    flow in rats with Pb-induced HTN. The effect of Pb on the sympathetic nervous system activity
16    was examined by Lai et al. (2002) who tested the rapid response to intrathecal (IT) injection of
17    PbCb in vivo and its addition to the thoracic cord slices in vitro in the rats.  They found
18    significant rises in arterial pressure and heart rate with IT injection of Pb-chloride. These effects
19    of Pb were abrogated by the administration of ganglionic blockade using hexomethonium. The
20    in vitro studies revealed a significant rise in excitatory and significant fall in inhibitory post-
21    synaptic potentials with the addition of Pb to the bathing medium and their reversal with saline
22    washout.
23          In a recent study, Chang et al. (2005) showed a gradual decline in blood, kidney, heart,
24    and aorta Pb contents toward the control values within 7 months following cessation of exposure
25    in rats with Pb-induced HTN. This was coupled with a parallel declines in arterial pressure,
26    plasma NE and renal tissue P receptor density as well  as parallel rises in the aorta and heart P
27    receptors densities during the 7-month period following cessation of Pb exposure.  However,
28    while HTN and P receptor abnormalities were significantly improved, they were not completely
29    reversed.  It should be noted that bone Pb contents were not measured in this study and were
30    most likely elevated despite normalization  of blood and soft tissue levels. These findings
      December 2005                           5-119        DRAFT-DO NOT QUOTE OR CITE

-------
 1    provided evidence for the stimulatory effect of Pb on the sympathetic nervous system and for its
 2    contribution to the cardiovascular effects of Pb exposure.
 3
 4    5.5.2.4 Effects of Lead on the Renin-Angiotensin-Aldosterone (RAAS) and
 5           Kininergic Systems
 6          The available data on the effects of Pb exposure on the RAAS are contradictory.  This
 7    appears to be primarily due to variability in the dosage and duration of Pb exposure, as well as
 8    the age at which exposure is initiated or the animals studied.  In addition, when present,
 9    nephropathy can potentially affect the RAAS profile of Pb-exposed animals or humans.  The
10    majority of animal studies of the effects of Pb on RAAS were conducted and published in the
11    late 1970s and 1980s.  In a meta-analysis of the studies published in that period, Vander (1988)
12    found increased plasma renin activity and renal tissue renin content in young rats after several
13    weeks of Pb exposure sufficient to achieve blood Pb concentrations in the range of 30 to
14    40 |ig/dL.  Similar results were found in rats exposed to Pb in utero and for 1 month after birth.
15    In contrast, plasma renin activity and renal renin contents were generally unchanged or even
16    reduced in older rats whose Pb exposure had commenced in utero.
17          In a more recent study, Carmignani et al. (1999) showed a significant increase in plasma
18    angiotensin converting enzyme (ACE) activity in the rats exposed to Pb (60 ppm Pb-acetate in
19    water) for 10 months beginning at an early age (weaning).  This was accompanied by a
20    significant increase in plasma kininase II, kininase I, and kallikrein activities.  In a subsequent
21    study, Sharifi  et al. (2004) examined plasma and tissue ACE activity in young adult rats
22    (weighing 200 g) exposed to Pb (100 ppm Pb-acetate) for 2 to 8 weeks. They found significant
23    rises in plasma, aorta,  heart, and kidney ACE activities, peaking at 2 to 4 weeks.  This was
24    followed by a decline  in plasma and tissue ACE activity to subnormal values by 8 weeks, at
25    which point arterial pressure was markedly elevated. The authors concluded that the elevated
26    ACE activity is involved in the induction of HTN but may not be necessary for maintaining HTN
27    in Pb-exposed animals.  Finally, in a recent study, Rodriguez-Iturbe et al. (2005) demonstrated a
28    marked increase in the number of Ang-II positive cells in the kidneys of rats treated with lead-
29    acetate (100 ppm in water) for 3 months. This observation points to heightened intra-renal Ang-
30    II generation in rats with Pb-induced HTN.
      December 2005                           5-120       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Taken together, the data point to activation of the RAAS at some point in the course of
 2    Pb-induced HTN. Further studies are needed to fully elucidate the effects of Pb exposure on
 3    various other RAAS components.
 4
 5    5.5.3   Effects of Lead Exposure on Vasomodulators
 6          In a study of a group of Pb workers with elevated blood Pb concentration, Cardenas et al
 7    (1993) found a significant increase in urinary excretion of the metabolite of vasoconstrictive
 8    prostaglandin, thromboxan (TXB2), and significant reduction of the vasodilatory prostaglandin,
 9    6-keto-PGFl, when compared with the control workers. Subsequently, Hotter et al. (1995)
10    confirmed the elevation of urinary TXB2 in another group of Pb-exposed workers. Based on
11    these observations, the authors suggested that Pb can alter the balance between vasoconstrictive
12    and vasodilatory prostaglandins in a way which may contribute to HTN and cardiovascular
13    disease.  In an attempt to examine such possible effects of Pb exposure  in experimental animals,
14    Gonick et al. (1998) measured urinary excretion of the above metabolites in the rat model of Pb-
15    induced HTN. The study showed no significant difference in urinary excretion of the given
16    prostaglandin metabolites between the Pb-exposed and control rats.  However, in a recent in vitro
17    study, Dorman and Freeman (2002) demonstrated that Pb promotes the release of arachidonic
18    acid by vascular smooth cells via activation of phospholipase A2. They further showed that, at
19    low concentrations, Pb  augments Ang-II-induced VSMC proliferation, whereas at a high
20    concentration it reduces viability and cell count in unstimulated cells and reduces DNA
21    synthases in Ang-II and Fetal Calf Serum (FCS)-stimulated VSMC.  Thus, Pb can increase the
22    release of arachidonic acid (the substrate for prostaglandins) via activation of phospholipase A2.
23          Given the limited and contradictory nature of the published data, further in-depth studies
24    are needed to clarify the effects of Pb on regulation of arachidonic acid metabolism and the
25    synthesis of various classes of prostaglandins.
26
27    Endothelin
28          Endothelins (ET) represent a family of potent vasoconstrictive peptides that are produced
29    by endothelium and a number of other cell types.  Excess production or increased sensitivity to
30    ET can raise arterial pressure. In an attempt to explore the possible contribution of ET to the
31    pathogenesis of Pb-induced HTN, Khalil-Manesh et al. (1993) studied the effects of exposure to

      December 2005                           5-121        DRAFT-DO NOT QUOTE OR CITE

-------
 1    low and high levels of Pb (100 ppm versus 5000 ppm) in the drinking water for 1 to 12 months in
 2    rats.  Rats exposed to low (but not high) levels of Pb exhibited HTN and a significant increase in
 3    plasma ET-3 concentration.  These findings were confirmed by these investigators in a
 4    subsequent study of rats with Pb-induced HTN (Khalil-Manesh et al., 1994).  Similarly, Gonick
 5    et al. (1997) demonstrated a significant elevation of plasma concentration and urinary excretion
 6    of ET-3 in rats with Pb-induced HTN. In a recent study, Martin et al. (2005)  showed that
 7    incubation in the lead-containing media resulted in the downregulation of soluble guanylate
 8    cyclase and cGMP production in the isolated artery segment of normal rats. They further found
 9    that co-incubation with an ET-A receptor antagonist can partially reverse this effect of lead.
10    These findings suggest that the adverse effect of Pb exposure on cGMP production in the
11    vascular tissue is, in part, mediated by its ability to raise ET activity. It, thus, appears that
12    exposure to low-levels of Pb can raise activity or production of ET, which can, in turn, play a
13    part in the pathogenesis of Pb-induced HTN in the rat. Further studies are required to carefully
14    explore the effects of Pb on various components of the ET system.
15
16    Atrial Natriuretic Factor
17          Atrial natriuretic factor (ANF) is produced and secreted by cardiac myocytes. Plasma
18    concentration of ANF rises with volume expansion and declines with volume contraction.  ANF
19    serves as a vasodilator and a natriuretic agent and, as such, plays a role in regulating blood
20    volume, vascular resistance, and, hence, arterial pressure. Giridhar and Isom  (1990) measured
21    ANF in rats treated with IP injection of Pb-acetate (0.0 to 1.0 mg/kg/twice weekly for 30 days).
22    The Pb-exposed animals exhibited fluid retention, which was coupled with a paradoxical dose-
23    dependent decline in plasma ANF concentration. Based on these findings, they suggested that
24    Pb may interfere with the hormonal regulation of cardiovascular system, which may, in turn,
25    relate to the cardiovascular toxicity of this metal.
26
27    5.5.4  Effects of Lead  on Vascular Reactivity
28          Addition  of Pb-acetate to the bathing medium has been shown to elicit a cumulative
29    concentration-dependent vasoconstriction in isolated rabbit mesenteric artery  (Watts et al.,
30    1995).  This effect was reported to be partly mediated by activation of PKC. In a more recent
31    study, Valencia et al. (2001) found a concentration-dependent vasoconstrictive response to

      December 2005                            5-122       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Pb-acetate (0.1 to 3.1 mM) in Wistar rat thoracic aorta rings. The contractile response was
 2    observed in both intact and endothelium-denuded rings.  Likewise, Pb-induced vasoconstriction
 3    was preserved in calcium-free medium and was unaffected by either a-1 blockade (prazosin),
 4    PKC inhibition (Calphostin) or L-type calcium channel blockade (verapamil).  However, Pb-
 5    induced vasoconstriction was inhibited by lanthanum, which is a general calcium-channel
 6    blocker. These observations suggest that Pb can promote an endothelium-independent
 7    vasoconstriction by a direct effect on the vascular smooth muscle cells.  The data further
 8    suggests that the effect of Pb is Ca-independent and may depend on the entry of Pb to the cell via
 9    a lanthanum-blockable channel. In contrast to the latter studies, addition of Pb-acetate did not
10    cause vasoconstriction in the rat aorta rings used in a study reported by Shelkovnikov and
11    Gonick (2001). Moreover, Pb-acetate at either high (10~4m) or low (10~8m) concentrations did
12    not modify the response to NE, phorbol ester, or isoproterenol. However, at 10~4M, Pb-acetate
13    augmented the contractile response to submaximal concentrations of calcium.  Thus, the rapid
14    action of Pb on vascular reactivity in vitro seems to vary depending on the type of the vessel
15    used, the Pb concentration employed, and the animal species being studied.
16          A number of studies have endeavored to discern possible differences in vascular reactivity
17    to various agonists between animals with Pb-induced HTN and control animals. For instance,
18    Purdy et al. (1997) found no significant difference in vasoconstrictive response to NE and
19    phenylephrine or vasodilatory response to acetylcholine or nitroprusside in the aorta rings
20    obtained from Sprague-Dawley rats with Pb-induced HTN.  In contrast, Marques et al. (2001)
21    showed a significant reduction of vasodilatory response  to both acetylcholine and nitroprusside
22    in Wistar rats with Pb-induced HTN. It should be noted that the Wistar rats employed in the
23    latter study had been treated with 5 ppm Pb-acetate in the drinking water for 1 month, whereas
24    those reported by Purdy et al. (1997) had been given a higher dosage (100 ppm) for a longer
25    period (3 months). Therefore, the magnitude and duration of exposure may account for the
26    differences observed between the two reports. Also, the effect of Pb on vascular reactivity may
27    vary from one tissue to the next, as clearly exemplified by studies (Oishi  et al., 1996) that
28    showed significant endothelium-dependent vasorelaxation of mesenteric artery response to
29    acetylcholine in the presence of the NOS inhibitor L-NAME in tissues from rats exposed to
30    Pb-acetate for 3 months.  These observations suggest that chronic Pb exposure may impair
      December 2005                           5-123       DRAFT-DO NOT QUOTE OR CITE

-------
 1    endothelium-dependent hyperpolarization in the rat mesenteric artery.  However, no such effect
 2    was noted in the aorta obtained from the same animals.
 3
 4    5.5.5   Lead-Calcium Interactions in Vascular Tissue
 5          Changes in cytosolic Ca2+ concentrations are intimately involved in regulating vascular
 6    tone and vascular smooth muscle contraction. Consequently, several studies have focused on the
 7    interaction of Pb with cellular Ca2+ and Ca2+-dependent signaling pathways as a means to gain
 8    insight into the pathogenesis of Pb-induced HTN (Piccini et al  1977; Favalli et al 1977; Webb
 9    et al 1981; Goldstein 1993; Watts et al 1995). Lead can potentially compete with Ca2+ in
10    transport systems (i.e., channels and pumps) involved in physiological movements of ions,
11    particularly Ca2+, into and out of the cell (Simons 1993a,b). Moreover, Pb can alter the
12    intracellular distribution of Ca2+ between cytoplasm, endoplasmic reticulum, and mitochondria,
13    which normally regulates cytosolic Ca2+ concentration, (Simons 1993a,b). In addition, Pb can
14    serve as a substitute for calcium in Ca2+-dependent signaling pathways by interacting with
15    calmodulin, PKC, and calcium-dependent potassium channels (Haberman, 1983; Richardt, 1986;
16    Chai and Webb, 1988; Simons, 1993a,b; Watts, 1995).  Thus, interactions of Pb with cellular
17    Ca2+ via these complex mechanisms in the vascular cells may contribute to alterations of
18    vascular resistance and HTN. For example, Piccini et al. (1997) and Favalli et al. (1977) showed
19    that Pb exposure increases calcium content in the tail artery in rats.  The authors attributed this
20    phenomenon to a possible Pb-induced inhibition of Ca2+ extrusion from the vascular cells.  Using
21    rabbit mesenteric artery preparations, Watts et al. (1995), showed that blockade of either PKC or
22    voltage-gated Ca channels by verapamil substantially attenuated Pb-induced vasoconstriction in
23    both intact and endothelium-denuded preparations. Based on these observations, the authors
24    suggested that Pb promotes a vasoconstrictive response in rabbit mesenteric artery via a Ca2+-
25    dependent activation of PKC. In contrast, Valencia et al. (2001) using rat aorta rings reported a
26    vasoconstrictive response to Pb-acetate in rat aorta rings bathed in either Ca2+-free or Ca2+-
27    containing media and in the presence  or absence of the L-type calcium-channel blocker
28    verapamil or of the PKC inhibitor calphostin.  Moreover, depletion  of intracellular Ca2+ stores by
29    preincubation of rings in EGTA, while diminishing the intensity, did not abrogate Pb-induced
30    vasoconstriction in this system. In contrast, Pb-induced vasoconstriction was prevented by
31    lanthanum (a general blocker of calcium channels) in both Ca2+-containing and Ca2+-free media.

      December 2005                            5-124       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Based on these observations, the authors concluded that Pb can elicit a PKC-independent
 2    contractile response in the rat aorta by entering VSMC via a non-voltage-gated Ca2+ channel and
 3    mimicking the action of Ca2+.  It, thus, appears that Pb exerts its effect by mechanisms that are
 4    species- and vessel-specific.
 5
 6    5.5.6   Cardiotoxicity and Atherogenesis
 7          Acute Pb exposure has been reported to affect cardiac function, and chronic exposure has
 8    been linked to atherosclerosis and increased cardiovascular mortality by some, but not by all
 9    investigators, in humans (See Chapter 6). In an attempt to assess the cardiotoxicity of lead,
10    Prentice and Kopp (1985) carried out the in vitro perfusion of isolated rat heart preparations with
11    a perfusate containing 0.3 and 30 jiM Pb-acetate for up to 60 min. At 30 jiM concentration, Pb
12    prolonged the AV node and His bundle conduction times, reduced coronary blood flow and heart
13    rate, and altered cardiac energy metabolism. Milder, and statistically insignificant, changes were
14    also observed at 0.3 jiM Pb concentration in this model. These observations illustrate the direct
15    cardiotoxicity of Pb independently of its systemic and neuroendocrine actions in acute
16    intoxication. In an attempt to determine whether chronic exposure to Pb or cadmium can cause
17    atherosclerosis, Revis et al. (1981), studied male white pigeons that were exposed to Pb (0.8 ppm
18    in drinking water) for extended periods.  Long-term low-level Pb exposure in this model resulted
19    in a significant rise in arterial pressure and a near doubling of the number of atheromatous
20    plaques in the aorta.  These observations demonstrate the proatherogenic effects of chronic
21    exposure to low levels of Pb in pigeons.
22
23    5.5.7   Effects of Lead on Endothelial Cells
24          Endothelium is an important constituent of the blood vessel wall and regulates
25    macromolecular permeability, vascular smooth muscle tone, tissue perfusion, and blood fluidity.
26    Endothelial damage or dysfunction results in atherosclerosis, thrombosis, and tissue injury.
27    Chronic Pb exposure has been shown to promote atherosclerosis in experimental animals (Revis
28    et al., 1981). Given the central role of endothelial injury/dysfunction in the pathogenesis of
29    atherosclerosis, numerous studies have explored the effect of Pb on cultured endothelial cells.
30    These studies have searched for evidence of Pb-mediated endothelial cell injury and the effects
31    of Pb on endothelial  cell proliferation, tube formation (angiogenesis), monolayer wound repair,

      December 2005                           5-125       DRAFT-DO NOT QUOTE OR CITE

-------
 1    and production of heparansulfate proteoglycans, plasminogen activator (tPA), and plasminogen
 2    activator inhibitor-1 (PAI-1).
 3          Using cultured bovine aorta endothelial cells, Kaji et al. (1995a) showed that incubation
 4    with Pb-nitrate at concentrations equal to or below 50 jiM for 24 h, results in mild de-
 5    endothelialization of endothelial monolayers in vitro. They further showed that adding Pb at
 6    10 jiM concentration markedly increased cadmium-induced endothelial injury.
 7          Proliferation of endothelial cells is a critical step for the repair of injured endothelium.
 8    Failure of the repair process can result in thrombosis, VSM cell migration and proliferation, and
 9    atherosclerosis.  In  this regard, Pb (Pb-nitrate 0.5 to 5 jiM) has been shown to significantly
10    reduce DNA synthesis and cell proliferation in growing cultured bovine aorta endothelial cells
11    (Kaji, 1995a). Similarly, the proliferative response to PFGF and aFGF is significantly attenuated
12    by Pb in this system (Kaji, 1995b).  The reported inhibition of endothelial cell proliferation by Pb
13    can potentially diminish the repair process in response to endothelial injury. This supposition
14    has been confirmed by Fujiwara et al. (1998) who showed that at 5 to 10 jiM concentrations, Pb
15    markedly inhibited  the repair of the wounded endothelial monolayer in vitro. Moreover, Pb
16    severely mitigated the zinc-stimulated endothelial cell proliferation and repopulation of the
17    denuded sections in this system.
18          Endothelial cell proliferation is the primary step in angiogenesis, a phenomenon that is
19    essential for numerous physiological functions such as growth, development, wound repair, and
20    menstrual cycle as well as certain pathological events including diabetic retinopathy and tumor
21    growth.  In view of the demonstrated inhibition of endothelial cell growth by lead, it has been
22    postulated that Pb may impair angiogenesis. This assumption has been confirmed by a number
23    of studies testing the effect of Pb by angiogenesis assay (tube formation) in endothelial cells
24    cultured on matrigel (a laminin-rich basement membrane product) matrix in vitro. For instance,
25    Ueda et al. (1997) and Kishimoto et al. (1995) have shown that Pb-acetate (1 to 100 |iM) results
26    in a concentration-  and time-dependent inhibition of tube formation by human umbilical vein
27    endothelial cells cultured on a matrigel matrix.
28          Endothelial cell migration and proliferation are critical for angiogenesis and repair of the
29    damaged endothelium. PFGF is a powerful mitogen for endothelial cells as well as several other
30    cell types. Endothelial cells synthesize PFGF, which is released following injury or spontaneous
31    death of endothelial cells and acts in an autocrine fashion to facilitate the repair process by

      December 2005                            5-126        DRAFT-DO NOT QUOTE OR CITE

-------
 1    promoting endothelial cell migration and proliferation. Binding of PFGF to its receptor on the
 2    endothelial cell is facilitated by heparan sulfate proteoglycans (HSPGS) that are normally
 3    produced and released by the endothelial cells for attachment to the cell surface as well as
 4    incorporation in the extracellular matrix. As noted above, Pb significantly attenuates PFGF and
 5    oFGF-mediated DNA synthesis and proliferation in cultured endothelial cells (Kaji et al.,
 6    1995b).  In this regard, Pb has been shown to reduce PFGF binding to the cell  surface HSPGs
 7    without changing the biosynthesis or intracellular abundance of PFGF in cultured bovine
 8    endothelial cells (Fujiwara and Kaji, 1999a). Moreover, Pb has been shown to significantly
 9    reduce the synthesis of glycosamino-glycans (GAG, measured by sulfate incorporation into
10    heparan  sulfate) in the growing endothelial cells.
11          The above observations suggest that Pb-induced reduction of pFGF-mediated proliferative
12    response in cultured endothelial cells is largely due to impaired production of HSPGs.  This
13    supposition is further supported by observations that DNA synthesis can be restored by adding
14    heparin in lead-treated growing endothelial cells (Fujiwara et al., 1995). The reduction in the
15    production of GAGs by Pb in the growing endothelial cells (Fujiwara et al., 1995) is also seen in
16    confluent (quiescent) cells. For instance, Kaji et al. (1991) demonstrated a marked reduction of
17    GAG production following incubation with 10 jiM Pb nitrate in confluent endothelial cells in
18    vitro.  The Pb-induced reduction of heparan sulfate production was more severe than that of the
19    other GAGs. Moreover, the reduction in the cell surface-associated GAGs was more severe than
20    that of the newly synthesized GAG found in the incubation media. GAGs combine with a series
21    of specific core proteins to form anionic macromolecular complexes known as proteoglycans,
22    which are widely distributed in the extracellular matrix of the mammalian tissues.  Endothelial
23    cells produce two types of HSPGs, i.e., the high-molecular weight and low-molecular weight
24    classes.  Perlecan is a high-molecular weight heparan-sulfate proteoglycan which is a  component
25    of the basement membrane.  Syndecan, glypican, ryudocan, and fibroglygan are among the low-
26    molecular weight subclass and are primarily associated with the cell surface. Proteoglycans play
27    an important role in regulating vascular function and structure. For instance, by providing a
28    negative electrostatic charge, these molecules constitute a major barrier against extravasations of
29    negatively-charged plasma proteins. In addition, by interacting with antihthrombin-III and tPA,
30    these molecules serve as important endogenous anticoagulants.  Moreover, perlecans facilitate
31    PFGF binding to its receptor on endothelial cells and, thus, contributes to the endothelial growth

      December 2005                           5-127       DRAFT-DO NOT QUOTE OR CITE

-------
 1    and repair processes. In contrast, these molecules tend to inhibit migration and growth of
 2    vascular smooth muscle cells and, thereby, help to prevent athero- and arteriosclerosis.  Another
 3    important function of HSPGs is their role in stabilizing and anchoring lipoprotein lipase and
 4    VLDL receptors on the endothelial surface. Consequently, they play an important indirect part
 5    in the clearance of VLDL and chylomicrons from the circulation, a process which has major
 6    implications for energy metabolism and cardiovascular protection.
 7          In a study of cultured bovine endothelial cells, Kaji et al. (1997) found that Pb-chloride, at
 8    10 jiM concentration, markedly lowers incorporation of precursors (glycosamine and sulfate)
 9    into HSPG in confluent bovine aorta endothelial cells. The effect of Pb was more severe on
10    low-molecular than high-molecular weight HSPGs.  However, Pb did not change the length of
11    heparan  sulfate chains. It is of note that Pb slightly increased the abundance of the HSPG core
12    proteins.  This observation excluded a reduction in core protein synthesis as a cause  of
13    diminished HSPGs in the lead-treated confluent endothelial cells.  In a subsequent study,
14    Fujiwara and Kaji (1999) investigated the effect of Pb-nitrate on production of high- and low-
15    molecular weight subclasses of HSPGs in growing bovine aorta endothelial cells. In contrast to
16    the quiescent cells, lead-treated growing cells exhibited a marked reduction in the high-
17    molecular weight with no change in production of low molecular weight (-50KD) HSPGs. They
18    further showed a significant reduction of the core protein of perlecan, which is a high-molecular
19    weight (400 KD) HSPG. Thus, Pb appears to affect productions of subclasses of HSPGs
20    differently depending on the cells' growth cycle. Accordingly, in the growing endothelial cells
21    (a condition which simulates the response to injury), Pb downregulates perlecan,  which is
22    involved in pFGF-mediated migration and proliferation of endothelial cells and inhibition of
23    migration and proliferation of VSMC. This phenomenon may adversely affect endothelial repair
24    and promote athero- and arteriosclerosis.  On the other hand, Pb-induced reduction of the cell
25    surface-associated low-molecular weight HSPGs (which are predominantly involved with
26    lipolytic, anticoagulant,  and other functions of confluent endothelial cells (simulating intact
27    endothelium) can contribute to hyperlipidemia and thromboembolism, among other  disorders.
28          One of the major properties of normal endothelium is  its ability to prevent coagulation.
29    Several factors contribute to the thromboresistance of the endothelial lining.  These include the
30    surface coating of HSPG (which confers heparin-like properties), nitric oxide (which inhibits
31    platelet adhesion and activation), and tPA (which promotes thrombolysis), thrombomodulin, and

      December 2005                          5-128        DRAFT-DO NOT QUOTE OR CITE

-------
 1    prostacycline. As noted earlier, Pb exposure reduces HSPG-production (Kaji et al., 1995b, 1997)
 2    and diminishes nitric oxide availability via ROS-mediated NO inactivation (Vaziri 1999). In
 3    addition, Kaji et al. (1992) showed that incubation of confluent human umbilical vein endothelial
 4    cells with Pb nitrate, at 0.01 to 1.0 jiM concentrations, significantly reduced basal and thrombin-
 5    stimulated tPA release. It thus, appears that Pb exposure may confer a thrombophilic diathesis.
 6
 7    5.5.8   Effects of Lead on Vascular Smooth Muscle Cells
 8          Lead has been shown to stimulate proliferation of bovine aorta VSMCs in a
 9    concentration-dependent manner (Fujiwara et al., 1995). Moreover, the combination of Pb and
10    PPGF results in an additive effect on VSMC proliferation. As with bovine aorta VSMCs,
11    cultured rat aorta VSMCs exhibit hyperplasia in response to a low concentration of (100 |ig/L) of
12    Pb-citrate (Carsia, 1995). The reported hyperplasia is accompanied by phenotypical
13    transformation of cells from the spindle or ribbon shape to cobblestone shape, simulating the
14    neointimal cell morphology. This was accompanied by a significant reduction in Ang-II receptor
15    but no change in a, P, or ANP receptor densities.  It is of note that, in contrast to the low
16    concentration, a high concentration (500  |iM/L) of Pb resulted in growth arrest in this system.
17    Thus, the effect of low concentration of Pb on VSMC proliferation is opposite of its action on the
18    endothelial cells.
19          Under normal conditions, intact endothelial lining shields the cells residing in the
20    subendothelial tissue, i.e., fibroblasts and VSMCs, from coming into contact with the circulating
21    blood. However, this barrier is lost when the endothelium is injured, an event which can lead to
22    platelet adhesion and fibrin thrombosis formation. Propagation of fibrin thrombus is limited by
23    activation of the fibrinolytic system, which, in turn, depends on the balance between tPA and
24    plasminogen activator inhibitor-1 (PAI-1). In addition to endothelial cells, VSMCs and
25    fibroblasts express tPA and PAI-1.  Using cultured human aorta VSMCs and  fetal lung
26    fibroblasts, Yamamoto et al. (1997) investigated the effect of Pb chloride on the release of tPA
27    and PAI-1 in vitro.  The authors found that Pb causes a significant inhibition of tPA release and a
28    significant increase in PAI-1 release in cultured fibroblasts in a dose-dependent manner.  The
29    lead-treated VSMC exhibited a significant dose-dependent decline in tPA release and to a lesser
30    extent of PAI-1 release.  Taken together, exposure to Pb appears to evoke a negative effect on
31    fibrinolytic process by the cellular constituents  of the subendothelial tissue.

      December 2005                           5-129       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.5.9   Summary/Conclusion
 2          In vivo and in vitro studies published during the past 15 years have considerably
 3    expanded our knowledge of the effects of Pb exposure on the cardiovascular system. However,
 4    many questions remain unanswered and await further investigation.
 5          A number of in vivo and in vitro studies conducted during the review period have
 6    provided compelling evidence for the role of oxidative stress in the pathogenesis of Pb-induced
 7    HTN. Moreover, the  effect of oxidative stress on blood pressure has been shown to be, in part,
 8    mediated by avid inactivation of NO and downregulation of sGC. In addition, a limited number
 9    of in vitro studies have provided indirect evidence that, via activations of PKC and NFicB, Pb
10    may raise vascular tone and promote inflammation.
11          The adrenergic system plays a major role in regulating cardiovascular function and
12    structure and, as such, has been the focus of several studies during the review period. Based on
13    these studies, chronic low level lead exposure appears to increase central sympathetic activity,
14    reduce cardiac  and vascular and raise kidney P adrenergic receptor density. These events can, in
15    turn, increase peripheral vascular resistance and renal renin release/production and, thereby,
16    arterial pressure.  Since sympathetic outflow is inhibited by NO, inactivation of NO by oxidative
17    stress may be, in part, responsible for the increased sympathetic activity in Pb-exposed animals.
18          The renin-angiotensin-aldosterone system (RAAS) plays an important role in regulating
19    blood pressure  and cardiovascular function and structure. The  available data published during
20    the review period suggest that Pb exposure can raise plasma ACE and kininase activities at
21    different points in the course of Pb-induced HTN in experimental animals. This can, in turn,
22    contribute to the genesis and/or maintenance of HTN. Since renin release (which is responsible
23    for production  of ACE substrate, i.e.,  Ang-1) is, in part, driven by P adrenergic activation,
24    upregulation of renal P adrenergic activity may,  in part, account for increased RAAS activity in
25    the Pb-exposed animals.
26          The balance in production of vasodilator and vasoconstrictor prostaglandins plays an
27    important role in regulation of blood pressure and cardiovascular function. Studies of the Pb
28    exposed humans have revealed an imbalance in production of prostaglandins favoring a rise in
29    arterial pressure.  However, the animal and in vitro studies published during the review period
30    have been limited and inconsistent. Further studies are needed to address this issue.
      December 2005                           5-130       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Based on the available studies, Pb exposure appears to increase endothelin production in
 2    experimental animals. This phenomenon can, in part, contribute to the rise in blood pressure in
 3    the Pb-exposed animals. A number of studies have explored the effect of Pb on vascular tone as
 4    well as vascular response to vasoconstrictor and vasodilator agents.  For instance, Pb has been
 5    shown to cause vasoconstriction and to attenuate acetylcholine- and NO-mediated vasodilatation
 6    in some, but not all vascular tissues and in some, but not all, studies. These effects have been
 7    variably attributed to lead-mediated activation of PKC and Ca2+-mimetic action of Pb, among
 8    other possibilities.
 9          Finally, a number of studies have explored the effects of endothelial and vascular smooth
10    muscle cells to explore the possible atherogenic effect of Pb exposure. In this context, Pb has
11    been found to inhibit proliferation of the growing (non-confluent) endothelial cells (mimicking in
12    vivo response to injury), impair tube formation (angiogenesis), and the repair of wounded
13    endothelial monolayer in vitro. Likewise, Pb exposure was shown to reduce production of
14    HSPGs and tPA by confluent endothelial monolayers, events that may favor thrombosis and
15    hyperlipidemia. Lead exposure has been also shown to promote vascular smooth muscle cell and
16    fibroblast proliferation and phenotypic transformation in ways that seem to favor arteriosclerosis
17    and vascular remodeling.
18          Among many questions awaiting clarification, a few are of particular interest. For
19    instance, it is not clear as to why low, but not high, levels of Pb exposure cause HTN in
20    experimental animals. Similarly, it is uncertain as to why HTN occurs long after the onset of Pb
21    exposure in the intact animals, whereas the effects on cultured cells  and isolated tissues are
22    manifested within short periods of time.
23
24
25    5.6   GENOTOXIC AND CARCINOGENIC EFFECTS OF LEAD
26    5.6.1   Introduction
27          The 1986 Pb AQCD (U.S. Environmental Protection Agency, 1986) and its 1990
28    Supplement (U.S. Environmental Protection Agency, 1990) concluded that, at relatively high
29    concentrations, Pb may be carcinogenic to laboratory animals, particularly the rat. Cell culture
30    studies were considered to be supportive of these observations, but also indicated that Pb was not
      December 2005                           5-131        DRAFT-DO NOT QUOTE OR CITE

-------
 1    particularly potent.  Human data were considered to be of concern, but not definitive, and given
 2    the animal data, the prudent choice was to consider Pb to be a possible human carcinogen.
 3          This section reviews reports of Pb-induced carcinogenesis and DNA damage published
 4    since 1986. More than 200 publications were read and considered and those that reported any
 5    effect related to carcinogenesis or genotoxicity that was attributable to Pb are presented below.
 6          This report follows the same format as the previous one (1986) and the explanations for
 7    the relative importance of the various types of studies (e.g. epidemiology, animal and cell
 8    culture) can be found in the original report and are not repeated here.  Carcinogenesis studies are
 9    presented first, followed by genotoxicity studies. Each of these sections is further subdivided
10    into human studies (considering adults and then children), animal studies, and then cell culture
11    studies (considering human, mammalian, and then nonmammalian).  When appropriate, these
12    sections are followed by a section describing acellular (cell-free) model studies.
13          There are some differences with this new report. For one,  each section is more distinctly
14    broken out. The epidemiology has been reviewed in more detail in Chapter 6 (Section 6.7) in
15    this document and, so, only a brief summary is presented here.  Because of more recent concerns
16    about effects on childhood development, this issue was specifically considered in a separate
17    section. Following advances in hypotheses and technology, much more specific sections about
18    the possible epigenetic effects of Pb have also been added.
19
20    5.6.2  Carcinogenesis Studies
21    5.6.2.1 Human  Studies
22          The human carcinogenesis studies are only briefly reviewed in this section; for a more
23    detailed review, see Chapter 6 (Section 6.7) in this document.
24
25    Adults
26          The assessment of the carcinogenicity of Pb through human epidemiological studies
27    remains ambiguous.  Several reports state that occupational exposure to Pb increases the risk of
28    lung, kidney, brain, stomach, and liver cancer (Fu and Boffetta, 1995; Kauppinen et al., 1992;
29    Gerhardsson et al., 1995a; Ades and Kazantzis,  1988; Wicklund et al., 1988; Steenland et al.,
30    1992; Englyst et al., 2001; Gerhardsson et al., 1986; Antitila et al., 1995, 1996; Cocco et al.,
31    1998; Shukla et al.,  1998). However, a full interpretation of the data in these studies is

      December 2005                           5-132        DRAFT-DO NOT QUOTE OR CITE

-------
 1    complicated by the fact that the study participants also incurred coexposure to other known
 2    carcinogens, such as arsenic, cadmium, and hexavalent chromium.  Thus, it is difficult to
 3    determine if the excess cancers observed were due to exposure to Pb, one of these other
 4    carcinogens, or some combination of the various chemicals.  In addition, other reports indicate
 5    that occupational or environmental exposure to Pb did not alter cancer risk (Cocco et al., 1996;
 6    Fanning, 1988; Jemai et al., 2002). Consequently a definitive assessment of the carcinogenicity
 7    of Pb from human studies cannot be made at this time.
 8
 9    Children
10          There have been no recent studies of Pb-induced cancers in children. This lack of data is
11    not unexpected and is largely because Pb has not been considered a likely cause of childhood
12    cancers.  There have, however, been  studies of cancers in children resulting from paternal
13    exposure. Here again, the same confounding problems encountered are as seen in the adult
14    population studies, and it is difficult to draw any definitive conclusions. For example, two
15    studies reported elevated childhood tumors (Wilm's tumor and acute nonlymphocytic leukemia)
16    in children whose fathers worked in Pb-related industries, such as welding, painting, and auto
17    repair (Buckley et al, 1989; Olshan et al.,  1990). However, workers in these occupations also
18    experienced coexposure to arsenic, cadmium, and hexavalent chromium, and so the cancers
19    observed cannot be solely linked to Pb exposure.  In addition, a report from the printing industry
20    in Norway found no link between paternal exposure and childhood cancers and, perhaps, even
21    found a possible reduction in the incidence of childhood cancers with paternal Pb exposure
22    (Kristensen and Andersen, 1992).
23          The possible interaction of paternal occupation and childhood cancer is an important area
24    of concern. However, a definitive assessment of paternal exposure to Pb cannot be made at this
25    time and more research is needed.
26
27    5.6.2.2 Laboratory Animal Studies
28          Lead is a well-established animal carcinogen, as noted in the 1986 Lead AQCD.
29    Consequently, limited tumorigenesis studies have been conducted in animal models and the
30    focus has been more on the mechanism of neoplasia (e.g., the roles of calcium and
     December 2005                           5-133       DRAFT-DO NOT QUOTE OR CITE

-------
 1   metallothionein) and possible immunomodulatory effects of Pb in the promotion of cancer.
 2   These studies are summarized in Table AX5-6.1.
 3         All of the studies exposed animals to Pb-acetate except one, which focused on Pb-
 4   chromate. One study investigated the carcinogenicity of a series of chromate compounds, i.e.,
 5   Pb-chromate and several Pb-chromate-based compounds were included as part of the group of
 6   chromate compounds. The Pb-chromate was administered by implantation into the lung after
 7   being embedded within a cholesterol pellet. The authors indicated that in this design, Pb-
 8   chromate was not carcinogenic, but that 4 of the Pb chromate compounds did induce a very rare
 9   tumor in the mice.  Thus, there is some ambiguity about the carcinogenicity of Pb-chromate in
10   the study, as the statistics calculated an expected tumor level based on any tumor and were not
11   based on the occurrence of this very rare (for rats) tumor. It is likely that had the expected value
12   been adjusted for the rare tumor, a  conclusion would have been reached that either Pb-chromate
13   was tumorigenic or that the study lacked the power to make any determination. The previous
14   EPA report had concluded that Pb-chromate is tumorigenic.  Thus, it is difficult to draw a firm
15   conclusion from this study.
16         The remaining five studies focused on Pb-acetate (Schrauzer,  1987; Blakley, 1987; Teraki
17   and Uchiumi, 1990; Bogden et al.,  1991; Waalkes et al., 2004). In most studies, this compound
18   was administered in drinking water at concentrations from 0.5 to 4000 ppm, but one study
19   considered effects from a subcutaneous (SC) injection both  in mice and in rats. Consistent with
20   the findings in the 1986 Pb AQCD, Pb not only induced renal tumors, but also induced other
21   tumors, although the possible effect on mammary tumors is  difficult to interpret, as important
22   study details were omitted, as discussed below. In a  surprising development, during one lifetime
23   exposure study, Pb suppressed liver tumors (Waalkes et al.,  2004).
24         The key study in this group  of studies was a lifetime  exposure study that investigated
25   mice exposed to drinking water concentrations of 1,000 to 4,000 ppm Pb and also considered the
26   role of metallothionein. In wild-type mice, Pb-acetate induced a low frequency of renal tumors,
27   but hyperplasia was common and exhibited overexpression  of cyclin Dl. Lead inclusion bodies
28   were also common. Lead also suppressed liver tumors in this study.
29         By contrast, in metallothionein-deficient mice, Pb-acetate induced a high frequency of
30   kidney tumors and  severe inflammation. Both the tumors and the regions of inflammation
31   exhibited cyclin Dl overexpression.  Lead also suppressed liver tumors in these animals.  In

     December 2005                          5-134       DRAFT-DO NOT QUOTE OR CITE

-------
 1    contrast to the wild-type mice, Pb inclusion bodies were not seen in these animals.  Thus, the
 2    data convincingly indicate that metallothionein binds Pb as part of an inclusion body and
 3    prevents the formation of tumors.
 4          Another study focused on the ability of Pb to induce tumors in rats after SC injection of
 5    Pb-acetate (Teraki and Uchiumi, 1990). Tumors formed at the site of injection, and Pb
 6    accumulated in the tumors, indicating that Pb is tumorigenic. However, full interpretation of the
 7    data is complicated by the absence of data on control animals and the fact that only a single dose
 8    was considered.
 9          Three studies investigated compounds that might reduce or prevent Pb-induced cancers,
10    specifically selenium and calcium compounds (Schrauzer, 1987; Bogden et al., 1991).  The first
11    study used a rather complex approach to study the possibly protective effects of selenium
12    (Shrauzer, 1987). In this study, mice were infected with the murine mammary tumor virus,
13    because they are known to develop mammary adenocarcinomas when maintained on a low-
14    selenium diet. The data indicated that Pb can induce tumors in these mice even when they  are
15    maintained on a high-selenium diet.  However, the data are difficult to interpret and the impact of
16    the study is uncertain, as the methods are incomplete, the data on control animals are not
17    provided, and the experimental results are stated but not presented in tables or figures.
18          The second study investigated the effect of calcium (Bogden et al., 1991).  The main
19    focus of this study appeared to be blood pressure, but tumorigenesis was also considered.
20    It might be anticipated that calcium might reduce Pb tumorigenesis by competing for its binding
21    sites or blocking its uptake. However, in this study, calcium did not affect Pb levels in tissue and
22    actually exacerbated Pb-induced carcinogenesis. The full impact of this study is also difficult to
23    assess, as the calcium-treated  animals incurred profound nephrocalcinosis.
24          The remaining  study considered Pb-induced immunosuppression as a possible factor
25    contributing to the tumorigenesis induced by other agents, including viruses or chemicals
26    (Blakley, 1987). The  results indicated that Pb may suppress humoral immunity but not cellular
27    immunity. However, this is the only study of its kind and the results need to be repeated in other
28    settings. In addition, it is difficult to determine if these data are specific to the agents used  (e.g.,
29    murine lymphocytic leukemia virus) or if they represent a class of agents (e.g., viruses in
30    general).
      December 2005                           5-135       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Overall, the above studies confirm that Pb is an animal carcinogen and extends our
 2    understanding of mechanisms involved to include a role for metallothionein.  Specifically, the
 3    recent data show that metallothionein may participate in Pb inclusion bodies and, thus, serves to
 4    prevent or reduce Pb-induced tumorigenesis. Much more work is needed to determine the
 5    potential exacerbating or ameliorating roles of calcium and selenium and to determine what role
 6    Pb-induced immunomodulation may play in the promotion of tumors.
 7
 8    5.6.2.3  Cell Culture Studies
 9          Carcinogenesis is measured in cell culture systems through studies of neoplastic
10    transformation, where morphologically transformed cells are injected into athymic mice to see if
11    the cells can form a tumor in the host animal.  Morphological transformation refers to cells that
12    incur a change in morphology, such as formation of a focus (or foci) of cell growth. In addition,
13    for faster study results and as a screening tool, the ability of cells to grow in agar without a
14    surface to attach to (anchorage independence) is often used as a short-term substitute measure for
15    transformation.
16
17    Human Cell Cultures
18          Since the  1986 Pb AQCD, only four studies have used human cell culture systems to
19    study the carcinogenesis of Pb compounds. One found that Pb-acetate induced anchorage
20    independence in primary human foreskin fibroblasts (HFF) (Hwua and Yang, 1998). The full
21    impact of these data is uncertain, as previous studies of known metal carcinogens in primary
22    HFF found that these carcinogens induced anchorage  independence, but those anchorage-
23    independent cells ultimately  senesced.  These studies  are summarized in  Table AX5-6.2. Further
24    study is needed to confirm that Pb can induce anchorage independence and to see if these cells
25    can progress to full neoplastic transformation.
26          In an effort to explore the importance of oxidative metabolism in inducing anchorage
27    independence, Hwua and Yang (1998) also co-treated some cells with 3-aminotriazole, a known
28    catalase inhibitor. This co-treatment had no effect on Pb-acetate-induced anchorage
29    independence, suggesting that catalase was not involved in this effect. It would be premature to
30    conclude that oxidative metabolism is not involved in anchorage independence, as  these are the
      December 2005                           5-136        DRAFT-DO NOT QUOTE OR CITE

-------
 1    only data available and are limited to catalase only. More data are needed to elucidate whether
 2    oxidative metabolism is involved in this lead effect.
 3          The remaining three studies focused on Pb-chromate (Beiderman and Landolph, 1987,
 4    1990; Sidhu et al., 1991). Two used similar HFF cells and found that Pb-chromate induced
 5    anchorage independence (Beiderman and Landolph, 1987, 1990). However, these anchorage-
 6    independent cells ultimately underwent senescence, suggesting that anchorage independence
 7    may not be a suitable short-term marker for neoplastic transformation in primary HFF. It should
 8    be noted that these studies were focused on the chromate component of this compound and the
 9    potential contribution of Pb was not investigated or discussed. By contrast, Sidhu et al. (1991)
10    found that Pb-chromate did not induce anchorage independence in a human osteosarcoma cell
11    line, while it did induce full neoplastic transformation of these cells and the transformed cells did
12    grow in agar. It should be noted that this study was also focused on the chromate component of
13    this compound and that the potential contribution of Pb was not investigated or discussed.
14          The 1986 Pb AQCD did not include any studies of transformation in human cells.  Given
15    that other chromate compounds have been shown to induce anchorage independence, it seems
16    quite possible that the data from Pb-chromate exposures may represent effects from chromate
17    and not from Pb. Thus, the data currently seem to indicate that Pb can induce anchorage
18    independence in human cells, but its ability to induce neoplastic transformation of human cells is
19    uncertain. Further study of different Pb compounds and the full assessment of their neoplastic
20    potential (i.e., including studies of the ability of treated cells to form tumors in experimental
21    animal models) are needed before definitive conclusions can be drawn.
22
23    Animal Cell Cultures
24          The 1986 Pb AQCD presented several studies demonstrating that Pb compounds could
25    induce anchorage independence and morphological and neoplastic transformation in rodent cell
26    culture systems. Since that report, six studies have further considered the ability of Pb
27    compounds to induce these effects.  Three focused on Pb-chromate and three on Pb compounds
28    without the confounding factor of chromate; and these studies are summarized in Table AX5-6.3.
29          Four studies considered Pb-acetate, Pb-chloride, or Pb-nitrate in Syrian hamster embryo
30    and C3H10T1/2 mouse embryo cells (Zelikoff et al., 1988; Patierno et al., 1988; Patierno and
31    Landolph, 1989; Elias et al., 1991).  Three found that Pb compounds did not induce

      December 2005                          5-137       DRAFT-DO NOT QUOTE OR CITE

-------
 1    transformation (Patierno et al., 1988; Patierno and Landolph, 1989; Elias et al., 1991); but the
 2    third study (Zelikoff et al., 1988) indicated that Pb was weakly positive, though no statistics were
 3    performed to validate this conclusion. Zelikoff et al. (1988) indicated that the observations were
 4    repeated several times, but only showed data from one experimental run.  It is unclear why the
 5    studies were not averaged together, as multiple repeats would likely have provided the power to
 6    detect whether the observed weak increase was significant.
 7          Five studies considered Pb-chromate, which induced neoplastic and morphological
 8    transformation of Syrian hamster and mouse C3H10T1/2 embryo cells, as well as enhancing
 9    viral transformation (Patierno et al., 1988; Patierno and Landolph, 1989; Schectman et al., 1986;
10    Elias et al.,  1989, 1991). The focus on Pb-chromate was based largely on concern about
11    chromate; but these studies found that Pb-chromate was more potent than other chromate
12    compounds, suggesting that Pb may enhance or contribute to the carcinogenicity. Indeed, one
13    study found that combining Pb-nitrate with soluble chromate was as potent as Pb-chromate and
14    greater than soluble chromate alone (Elias et al., 1991).
15          Thus, all together, these studies suggest that Pb ions alone cannot transform rodent cells;
16    however, they may be co-carcinogenic or promote the carcinogenicity of other compounds.
17    These data are in contrast to findings described in the 1986 Pb AQCD that included a positive
18    study.  One possible factor may be exposure duration; the study in question indicated that the
19    Pb-transformed cells were exposed for 9 days.  The studies discussed here all exposed cells for
20    7 days or less. Further careful study of a time course of exposure is necessary to determine
21    whether Pb  actually induces transformation in cultured rodent cells.
22
23    Nonmammalian Cell Cultures
24          No carcinogenesis studies were located  that used nonmammalian cell culture models.
25
26    5.6.2.4  Organ-Specific Studies
27          No organ-specific or organ culture studies concerning Pb  carcinogenesis were located.
28
29    5.6.2.5  Carcinogenesis Summary
30          It still remains difficult to conclude whether Pb is a human carcinogen. The assessment
31    of the  carcinogenicity of Pb through human epidemiological studies remains ambiguous.

      December 2005                           5-138       DRAFT-DO NOT QUOTE OR CITE

-------
 1   By contrast, the studies confirm that Pb is an animal carcinogen and further extend our
 2   understanding of the mechanism to include a role for metallothionein.  The cell culture data
 3   suggest that Pb can induce anchorage independence, but whether it can induce full neoplastic
 4   transformation of human cells is uncertain.
 5
 6   5.6.3   Genotoxicity Studies
 7          The human genotoxicity studies are only briefly reviewed in this section.  For a more
 8   detailed review, see Chapter 6 (Section 6.7) in this document.
 9
10   5.6.3.1  Human Studies
11   Adults
12          A number of studies investigating the potential genotoxicity of Pb have been conducted in
13   human populations. Endpoints considered include chromosome aberrations, sister chromatid
14   exchanges (SCE), micronuclei formation, DNA strand breaks, and hypoxanthine guanine
15   phosphoribosyl transferase (HPRT) mutations. In general, these studies were much more
16   specific than the carcinogenesis studies,  as correlations with blood-Pb levels could be made,
17   other confounders could be ruled out, and the endpoints were more short-term.
18          The chromosome damage studies are ambiguous and contained  some methodological
19   flaws. Four studies were positive (Xupei et al., 1988; De at al., 1995; Bilban, 1998; Pinto et al.,
20   2000), while two were negative (Anwar and Kamal, 1988; Rajah and Ahuja, 1996).  Moreover,
21   the four positive studies included two that could not rule out potential contributions from other
22   genotoxic metals and one that found a correlation only at very high blood Pb levels (>52 |ig/dL).
23          By contrast, the studies of micronucleus formation (Bilban, 1998; Vaglenov et al., 1998;
24   Pinto et al., 2000; Palus et al., 2003; Minozzo et al., 2004),  SCE (Xupei et al.,  1988; Bilban,
25   1998; Pinto et al., 2000; Duydu et al., 2001; Palus et al., 2003), DNA strand  breaks (Restrepo
26   et al., 2000; Fracasso et al., 2002; Hengstler et al., 2003; Danadevi et al., 2003; Palus et al.,
27   2003) all consistently found clear correlations between Pb and genotoxicity.  It should be noted
28   that there were two negative studies for SCE (Rajah and Ahuja,  1995, 1996), but both were by
29   the same group and considered the same very small population of workers (only 5 Pb-exposed
30   workers) and,  thus, may not have had enough power to detect potential differences.
     December 2005                           5-139       DRAFT-DO NOT QUOTE OR CITE

-------
 1          It is notable that one study found an interesting correlation of HPRT mutation rates and
 2    blood Pb levels from environmental Pb exposure in Belgian women (Van Larebeke et al., 2004).
 3    This study is the first and only one to consider Pb-induced mutations.  Further research is needed
 4    to assess the validity of these data.
 5          Thus, it appears from these studies that Pb is genotoxic to humans, although it may not
 6    induce substantial amounts of chromosome damage.  This conclusion is consistent with the
 7    laboratory studies discussed below. For more in-depth consideration of the epidemiology studies
 8    see Chapter 6, Section 6.7.

 9    Children
10          Two recent studies of Pb-induced genotoxicity in children have been published. One
11    study of children living in a high Pb contamination area of Czechoslovakia found no increase in
12    chromosome damage in white blood cells compared with children living in an area with lower Pb
13    contamination (Smejkalova, 1990). Comparisons were not done with children living in an area
14    with little or no Pb contamination. Measurements of blood Pb levels indicated a statistical
15    difference in blood levels between the two groups but not necessarily a substantial, or
16    biologically significant, difference between them. (Typically the control group levels were in the
17    high 20's compared to the low 30's |ig/dL in the exposed group). Thus, the possibility that each
18    group was exposed to a Pb level that could induce a baseline level of damage cannot be ruled out
19    and, thus, it cannot be conclusively stated that Pb was not clastogenic in this study.
20          The other study  found an increase in Pb-induced strand breaks in white blood cells from
21    children living in an area of Mexico with high Pb contamination compared to children living in
22    an area with lower Pb contamination (Yanez et al., 2003). Blood Pb levels confirmed a
23    difference in exposure to Pb, but urinary arsenic levels confirmed that these children were
24    exposed to higher levels of arsenic, too; and, thus, it cannot be determined which chemical was
25    responsible for the damage.
26          The possible genotoxicity of Pb for children is an important concern. However, there are
27    simply too few data to draw definitive conclusions, and more research is needed.  See Chapter 6
28    (Section 6.7) for more in-depth discussion of the epidemiology of Pb in human populations.
29
      December 2005                           5-140       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.6.3.2  Laboratory Animal Studies
 2         Fourteen studies evaluated the genotoxicity of Pb compounds in animal models. The
 3    majority of these studies focused on mice, and the Pb was administered by intraperitoneal (IP) or
 4    intravenous (IV) injection. Several endpoints were considered including chromosome
 5    aberrations, SCE, micronucleus formation, and DNA strand breaks. Overall, the results are
 6    ambiguous, due in part to study design and the various endpoints considered. These studies are
 7    summarized in Table AX5-6.4.
 8         Lead compounds appear to be able to damage chromosomes, if only weakly. Two studies
 9    with well-performed analyses were positive (Fahmy, 1999; Aboul-Ela, 2002). The other positive
10    studies observed that Pb could induce karyotypic arrangements, indicating a possible clastogenic
11    response; however, these studies did not analyze very many cells (Chakraborty et al., 1987;
12    Nayak et al., 1989a,b; Dhir et al.,  1990, 1992a,b; Nehez et al., 2000).  Some found chromosome
13    damage, but it did not increase with dose (Chakraborty et al., 1987; Nayak et al., 1989a,b; Dhir
14    et al., 1990). Altogether, the data do suggest some role for Pb in inducing chromosome damage,
15    but it may be a weak effect.
16          Similarly, the data for micronuclei and DNA damage are ambiguous.  One study found
17    that Pb induced micronucleus formation in a dose-associated manner, but only considered two
18    doses (Roy et al., 1992). The other study found that Pb induced micronucleus formation but not
19    in a dose-dependent manner (Jagetia and Aruna,  1998). This difference may reflect the
20    somewhat shorter exposure time in the second study.
21         One DNA damage study found that Pb nitrate could induce DNA strand breaks in the
22    white blood cells of mice (Devi et al., 2000);  however, the damage was not dose-dependent.
23    Another found DNA damage in a number of organs, but only one dose was considered and the
24    authors described the effect as weak (Valverde et al., 2002). In both studies, the highest doses
25    caused less damage than the moderate- to low-doses. These data again suggest that Pb is only
26    weakly causing damage.
27         By contrast, the results for SCE are consistently positive.  The three studies that were
28    positive found that SCEs were induced in a dose-dependent manner (Fahmy, 1999; Nayak et al.,
29    1989a; Dhir etal., 1993).
30         The route of administration complicates the interpretation of all of these genetic studies.
31    All of the studies, except for three chromosome damage studies,  used injection-based exposures.

      December 2005                           5-141        DRAFT-DO NOT QUOTE OR CITE

-------
 1    It is unknown if exposures that reflect more realistic scenarios (e.g., from drinking water) would
 2    cause any of these effects. Only one study of DNA strand breaks used a physiologically relevant
 3    exposure (inhalation).
 4          Four studies exposed animals by gavage, which is still a somewhat artificial exposure.
 5    One was a DNA damage study that found weak activity (Devi et al., 2000). The other three
 6    considered chromosome damage (Aboul-Ela, 2002; Dhir et al., 1992b; Nehez et al., 2000).
 7    Two found a dose-response for a 24 h-exposure to Pb nitrate-induced chromosome aberrations in
 8    mice (Aboul-Ela, 2002; Dhir et al., 1992b). The other found that a 4-week exposure to Pb-
 9    acetate induced aneuploidy, but not chromosome aberrations, in rats (Nehez et al., 2000). It is
10    difficult to reconcile these two studies, as they use different exposure times, chemicals, and
11    species.  More work is needed using relevant doses and exposure conditions to Pb compounds in
12    multiple species to determine if Pb induces chromosome aberrations.
13          Some studies also tried to offset the effects of Pb with a variety of compounds. Potential
14    modulators included fruit extract from Phyllanthus emblica, ascorbic acid, calcium, and iron
15    (Aboul-Ela, 2002; Dhir et al., 1990, 1992a, 1993; Roy et al., 1992). Other studies sought to
16    determine if coexposure to other toxicants would potentiate the effects of Pb (Dhir et al., 1992b;
17    Nehez et al., 2000) and considered both zirconium and cypermethrin.  The data indicated that the
18    fruit extract could block the toxic effects of Pb, an effect which may, in part, be attributable to
19    ascorbic acid, but that other components must also be involved, because ascorbic acid alone
20    produced variable results. Iron also had an effect, but only if given just before, or with,  the Pb
21    compound; post treatments with iron had no effect. Calcium had a strong effect.
22          The effects with zirconium and cypermethrin are less clear. Both were reported to
23    exacerbate the effects of Pb, but the effects for both are complicated by experimental design
24    problems. For example, zirconium only exacerbated Pb's effects when given simultaneously and
25    not when given 2 h before, or after, Pb.  This seems rather unusual as the total exposure to each
26    was 24 h and, thus, simultaneous exposure occurred in every circumstance. Thus, the data would
27    seem to suggest that a 22-h coexposure  had no effect, but that a 24-h exposure did.
28    Alternatively, there may have been some interaction of the two chemicals in the gut during
29    coexposure, creating a more toxic species.
30          Interpretation of the cypermethrin study is complicated by its design and the results. Only
31    20 metaphases were analyzed for each animal, instead of the recommended 100.  In addition, the

      December 2005                           5-142       DRAFT-DO NOT QUOTE OR CITE

-------
 1    statistical analyses were done relative to untreated controls and not to animals treated with Pb or
 2    cypermethrin alone. Careful inspection of the tables reveals that actual exposure to Pb plus
 3    cypermethrin induced less damage than that induced by Pb alone.  Thus, the effects of them
 4    together appear to be less than additive. More work is needed to explore the meaning of these
 5    data and the importance of Pb mixtures.
 6          The previous report found a similar amount of ambiguity; some animal studies were
 7    positive for chromosome damage and others were negative. Other endpoints were not described
 8    after Pb exposure in experimental animals.  These data suggest that Pb can induce SCE but that it
 9    can induce chromosome damage, DNA damage, or micronuclei either weakly or not at all.
10
11    5.6.3.3  Cell Culture Studies
12          Few cell culture studies were reported in the 1986 Pb AQCD.  Since 1986, a great deal of
13    theoretical and technological progress has allowed for a large number of cell culture studies to be
14    performed, as discussed below.
15
16    Human Cell Culture
17    Mutagenicity
18          Two studies considered Pb-acetate-induced mutagenesis in human cells.  Both considered
19    mutations at the HPRT locus, with one using keratinocytes and the other skin fibroblasts (Ye,
20    1993; Hwua and Yang, 1998). These studies are summarized in Table AX5-6.5.
21          One study reported no lead-induced mutagenesis (Hwua and Yang, 1998) but sought to
22    explore the importance of oxidative metabolism in lead-induced mutagenesis by co-treatment
23    with 3-aminotriazole, a known catalase inhibitor. This co-treatment did not increase Pb-acetate-
24    induced mutagenesis, suggesting that either catalase was  not involved in this effect or that Pb is
25    truly not mutagenic. It would be premature to conclude that oxidative metabolism is  not
26    involved in anchorage independence, as these are the only data and are limited to catalase.
27    Further data is needed to elucidate whether oxidative metabolism is involved in this effect of Pb
28    as well as further studies of lead-induced mutagenesis.
29          The other study reported that Pb-acetate induced mutagenesis (Ye,  1993). However,
30    interpretation of this study is hampered by its methodology. The study did not actually measure
31    HPRT mutations or colony formation, but rather it attempted a quicker methodology that

      December 2005                           5-143       DRAFT-DO NOT QUOTE OR CITE

-------
 1    measured tritium incorporation.  Although a shorter assay is highly desirable, the study did not
 2    verify the observed effects with standard methods, and, thus, it is uncertain if the tritium
 3    incorporation actually reflected lead-induced mutations.
 4          One study considered Pb-chromate and found that it was not mutagenic (Biedermann and
 5    Landolph, 1990).
 6          There are insufficient data at this point to conclude whether Pb is mutagenic in human
 7    cells, although the few data that exist are largely negative.
 8
 9    Clastogenicity
10          Ten studies investigated the ability of Pb compounds to induce chromosome damage in
11    cultured human cells.  All but one were essentially from the same research group, and all but two
12    considered Pb-chromate.  All were done using normal, or nearly normal, human cells.  These
13    studies are summarized in Table AX5-6.6.
14          Only two of those studies focused on the clastogenicity of Pb itself (Wise et al., 2004b,
15    2005), the remainder used Pb compounds but focused on either chromate or radioactive particles
16    as the clastogenic species. These studies found that Pb-glutamate was not clastogenic.
17          All of the Pb-chromate studies found that Pb-chromate induced chromosome damage in a
18    concentration-dependent manner. However, the effects were either attributed or demonstrated to
19    be caused by chromate ions.  Lead ions were produced by Pb-chromate, but they were not
20    clastogenic.
21          There was one study of radioactive Pb (Martins et al., 1993). The focus was on the
22    clastogenic activity of alpha particles, and the identity of the specific Pb salt was not provided.
23    The alpha particles were able to induce chromosome damage.
24          Overall, the data appear to indicate that Pb does not induce chromosome damage in
25    human cells, although more investigation of different compounds is needed.
26
27    DNA Damage
28          Studies of DNA damage in cultured human cells have considered DNA strand breaks,
29    Pb-DNA adducts, and DNA-protein crosslinks  for a variety of Pb compounds. The only clear
30    positive damage induced by Pb was Pb-DNA adducts following Pb-chromate exposure, although
31    the authors referred to them as Pb associated with DNA (Singh et al.,  1999). It is uncertain if

      December 2005                           5-144       DRAFT-DO NOT QUOTE OR CITE

-------
 1    these represent actual adducts or some weaker association. Two studies found no DNA strand
 2    breaks induced by Pb (Hartwig et al., 1990; Snyder and Lachmann, 1989), and one study
 3    involving several laboratories found no DNA-protein crosslinks after Pb exposure (Costa et al.,
 4    1996). The other study found DNA double-strand breaks, but these were attributed to chromate
 5    and not Pb (Xie et al., 2005).  These studies are summarized in Table AX5-6.7.
 6          One other study was positive (Wozniak and Blasiak, 2003), but the results were unusual
 7    and their impact uncertain.  Specifically, this study found that Pb-acetate induced DNA single-
 8    strand breaks but that the amount of damage decreased with concentration, and ultimately the
 9    highest concentration had less damage than the control. DNA double-strand breaks were
10    observed, but were lowest at the highest concentration.  DNA-protein crosslinks were seen only
11    at the highest concentration, and the authors attempted to explain the decrease in strand breaks
12    with this effect.  This explanation may partially correct, but it does not entirely explain the
13    decreased amount of damage at the middle concentration. These data need to be repeated by an
14    independent group before they can be fully assessed.
15          Together, these data  suggest that Pb likely does not induce DNA damage; however, the
16    data are still too limited to allow any definitive conclusions.
17
18    Human Cell Genotoxicity Summary
19          The cumulative data suggest that Pb is not mutagenic and does not induce chromosome
20    aberrations or DNA damage in cultured human cells.  It is interesting to note that Pb-induced
21    SCEs have not been considered in human cells.
22
23    5.6.3.4 Animal Cell Cultures
24    Mutagenicity
25          The potential mutagenicity of Pb compounds in rodent cells was considered in six studies.
26    In particular, three mutagenesis systems were considered: mutagenesis at the HPRT  locus, the
27    gpt locus, and mutations in  sodium-potassium ATPase.  The results are highly variable and may
28    be specific to the Pb compound considered in each case. In particular, Pb-chromate and Pb-
29    acetate appear to be nonmutagenic.  Lead acetate was positive but only at highly cytotoxic
30    concentrations. By contrast, Pb-chloride and Pb-sulfate appeared to be mutagenic at relatively
31    nontoxic concentrations.  These studies are summarized in Table AX5-6.8.

      December 2005                            5-145        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Insufficient data exist at this point to conclude whether or not Pb is mutagenic in animal
 2    cells.
 3
 4    Clastogenicity
 5          Seven studies investigated the ability of Pb compounds to induce chromosome aberrations
 6    in cultured mammalian cells (Table AX5-6.9). Four of these studies considered Pb-chromate
 7    and further investigation revealed that chromate was responsible for the clastogenic effect (Wise
 8    et al., 1992, 1993; Blankenship et al., 1997). Three of these studies considered other lead-
 9    containing compounds (Wise et al., 1994; Lin et al., 1994; Cai and Arenaz, 1998). All  but one
10    were negative and that one only found a small response at a single high dose (Wise et al., 1994).
11    Lower doses had no effect.  Considered together, the studies indicate that Pb does not induce
12    chromosomal aberrations in cultured mammalian cells.
13          Only two studies considered Pb-induced micronuclei in cultured mammalian cells. One
14    was negative (Lin et al., 1994) and the other positive (Bonacker et al.,  2005).
15          Four studies considered Pb-induced SCE in cultured mammalian cells. The results were
16    predominately negative (three studies [Hartwig et al, 1990; Lin et al., 1994; Zelikoff et al.,
17    1988]). Interpreting these studies, however, is complicated by the fact that too few metaphase
18    cells (less than 30 per concentration) were analyzed in each study. The one positive study
19    considered 100 metaphases per concentration, making those data more reliable (Cai and Arenaz,
20    1998).
21
22    DNA Damage
23          Several measures of DNA damage in cultured human cells have been investigated,
24    including DNA single-strand breaks and DNA-protein crosslinks. Most Pb compounds did not
25    induce DNA single-strand breaks. The exception was Pb-chromate, which did induce DNA
26    strand breaks, but this effect was  likely a result of the chromate ion. These studies are
27    summarized in Table AX5-6.10.
28          Both Pb-chromate and Pb-nitrate induced DNA-protein crosslinks  in cultured mammalian
29    cells. These data suggest that Pb  is genotoxic in this manner; however, it is thought that the Pb-
30    chromate-induced DNA-protein crosslinks result from the chromate and that the method used for
      December 2005                           5-146       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Pb-nitrate is not sufficiently rigorous. Thus, while the data are certainly suggestive, they are
 2    insufficient to make any definitive conclusion.

 3    Nonmammalian Cell Cultures
 4          Only one study was located considering Pb in a nonmammalian model (Table AX5-6.11).
 5    This study found that Pb-chromate was not mutagenic in a bacterial assay. The compound was
 6    studied because of its chromate content and, given that it is the lone study, no definitive
 7    conclusions can be reached.
 8
 9    5.6.3.5  Cell-Free Studies
10          No cell-free studies concerning Pb carcinogenesis or genotoxicity were located.
11
12    5.6.3.6  Organ-Specific Studies
13          One study (Valverde et al., 2002) considered organ-specific effects (see Table AX5-6.4).
14    That study found a different pattern of DNA strand breaks in mice after inhalational exposure to
15    Pb-acetate. DNA in the brain and lung were damaged the most, kidney and liver next, then nasal
16    epithelia and leukocytes, with no damage in testicle DNA. These data are intriguing, as they
17    suggest organ-specific responses after a physiologically relevant exposure (inhalation).  More
18    research is needed, however, to fully assess the impact of these findings.  Moreover, while the
19    damage was statistically significant, the authors described the effects as weak.
20
21    5.6.3.7  Genotoxicity Section Summary
22          There is some ambiguity in the genotoxicity results, as some endpoints were positive
23    while most were negative. Consistent with the animal study data, Pb can induce SCE in rodent
24    cells, but it is unknown if it can do so in human cells because this has not been tested. Lead also
25    seems to induce DNA-protein crosslinks in rodent cells.
26
27    5.6.4   Genotoxicity as it Pertains to Potential Developmental Effects
28          The human genotoxicity studies are only briefly reviewed in this section. For a more
29    detailed review, see Chapter 6 (Section 6.7). Only limited animal data and no cell culture studies
30    focused on this issue as a concern. The available data are described below.


      December 2005                          5-147       DRAFT-DO NOT QUOTE  OR CITE

-------
 1   Adults
 2          One study was located that considered the effects of Pb on sperm quality and quantity.
 3   This study considered Pb, cadmium, and selenium levels in 56 nonsmoking volunteers (Xu et al.,
 4   2003). No effects on sperm quality were correlated with Pb exposure up to 10 |ig/L.
 5          Two studies were located on the effects of Pb on sperm morphology in animals (Fahmy,
 6   1999; Aboul-Ela, 2002). Both were positive, indicating that Pb may have an effect on sperm.
 7   They also found that Pb induced DNA damage in the sperm (See Table AX5-6.4).  These studies
 8   are summarized in Table AX5-6.12.

 9   Children
10          No studies were analyzed that considered the genotoxic effects of Pb in children as a
11   developmental hazard.  There are two studies that considered  the genotoxic effects of Pb in
12   children. They were discussed in Section 5.6.3.1.
13          Three studies were located on the fetal effects of Pb-nitrate on the fetus (Kristensen et al.,
14   1993; Nayak et al., 1989a,b). Lead induced an increase in resorptions and there were hints of
15   possible fetal chromosome damage, but the methods were poorly described and much more work
16   is needed before conclusions can be drawn. These studies are summarized in Table AX5-6.13.
17
18   5.6.5   Epigenetic Effects and Mixture Interactions
19          Lead has been proposed to be a co-mutagen or possibly a promoter.  Thus a number of
20   epigenetic mechanisms have been proposed.  Epigenetic effects occur when a compound such as
21   Pb induces changes in cellular processes that do not result from changes in DNA sequence. In
22   other words, Pb has been proposed to alter cells in ways that may change the cell without
23   breaking or mutating DNA. There are three possible mechanisms:  (1) alterations of gene
24   expression that can stimulate cells to grow (mitogenesis) and/or can interfere with DNA repair;
25   (this possibility has been investigated in several studies); (2) interaction with other metals; and
26   (3) alteration of oxidative metabolism. Neither of the latter two have been extensively
27   investigated.
28
     December 2005                           5-148       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.6.5.1   Gene Expression
 2         It has been argued that Pb may induce or co-induce carcinogenesis by altering cellular
 3    metabolism or by altering the metabolism of another chemical. Both whole animal and cell
 4    culture studies have been conducted to address this question and are described below.
 5
 6    Animal
 1         Animal studies indicate that Pb can induce the expression of some phase I metabolizing
 8    enzymes, such as cytochrome P4501 Al, and phase II metabolizing enzymes, such as glutathione
 9    and glutathione-S-transferase. These studies are summarized in Table AX5-6.14.
10         Thus, it is plausible that through this mechanism, Pb may act as a co-carcinogen by
11    affecting the metabolism of other chemicals or possibly as a direct carcinogen by enhancing
12    endogenously-induced damage. However, no studies have directly shown that such Pb effects
13    are linked to cancer or alter the potency of another chemical; and, thus, it remains only a
14    plausible hypothesis.
15
16    Human Cell Culture Studies
17         A few human cell culture studies have been done, and these generally confirm the animal
18    studies.  These studies are summarized in Table AX5-6.15.
19         Lead has been shown to affect the induction of some phase I metabolizing enzymes (such
20    as cytochrome P4501A1) and phase II metabolizing enzymes (such as glutathione and
21    glutathione-S-transferase and NAPDH oxidase). These experiments also indicate that Pb can
22    affect the metabolism of other carcinogenic compounds, although they do not show that the
23    genotoxic or carcinogenic effects  change as a result of these effects; and, thus, more work
24    remains to make this more than just a plausible explanation.
25
26    Animal Cell Culture Studies
27         No animal cell culture studies concerning the effects of Pb on the expression of metabolic
28    genes were located.
29
     December 2005                           5-149       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.6.5.2  DNA Repair
 2          It has been argued that Pb may induce or co-induce carcinogenesis by altering the repair
 3    of DNA lesions induced by another agent. The greatest focus has been on damage induced by
 4    ultraviolet (UV) light. Only cell culture and cell-free studies have been conducted to address this
 5    question and are described below.

 6    Human
 1          Only one study considered Pb-induced effects on DNA repair in cultured human cells (see
 8    Table AX5-6.16).  This study found that coexposure to Pb caused persistence of strand breaks
 9    induced by UV light.  This persistence suggests that Pb interfered with the repair of these lesions,
10    but direct evidence of that interference was not provided.  These are the only data in human cells
11    and, thus, it cannot be determined if Pb inhibits DNA repair in human cells.
12
13    Mammalian Cell Culture Models
14          Two studies considered Pb-induced effects on DNA repair in cultured mammalian cells.
15    These studies are summarized in Table AX5-6.17.  Both found that Pb-acetate increased UV-
16    induced DNA damage including SCE, mutagenesis, and cytotoxicity. Lead did not affect strand
17    breaks induced by  UV. These data suggest that Pb may indeed inhibit repair, although direct
18    interactions with repair proteins were not demonstrated.
19
20    Cell Free Systems
21          One study considered the effects of Pb on DNA repair proteins (McNeill  et al., 2004).
22    That study found that Pb  can inhibit APE nuclease in cell-free systems.
23
24    5.6.5.3  Mitogenesis
25          It has been argued that Pb may induce or co-induce carcinogenesis by inducing cells to
26    grow when they should not.  Both animal and cell culture studies have been conducted to address
27    this question and are described below.
28
29    5.6.5.3.1   Animal
30          Several studies have considered Pb-induced mitogenesis in animal models. These studies

      December 2005                            5-150       DRAFT-DO NOT QUOTE OR CITE

-------
 1    are summarized in Table AX5-6.18. These studies found that Pb can stimulate cell growth, but
 2    primarily in the liver. One study did consider TNF-a expression in brain cells, but it was not
 3    demonstrated whether these effects were mitogenic. The interpretation of many of the studies is
 4    complicated by the exposure method (IV injection), which does not reflect human exposure.  In
 5    general, the data indicate that Pb is mitogenic to the liver.
 6
 7    Human Cell Culture Studies
 8          A number of studies have considered the potential growth-stimulatory effects of Pb in
 9    cultured human cells (Table AX5-6.19).  These studies all found that Pb did not stimulate cell
10    growth.  Thus,  mitogenesis is not a likely epigenetic effect for Pb in human cells.
11
12    Mammalian Cell Culture Studies
13          A number of studies have considered the potential growth-stimulatory effects of Pb in
14    cultured mammalian cells  other than the kidney. These studies all found that Pb did not
15    stimulate cell growth. Thus, mitogenesis is not a likely epigenetic effect of Pb in human cells.
16    One study found an increased mitotic index; however, it did not consider possible cell cycle
17    arrest (Lin et al., 1994).  Indeed, another study found that Pb increased the mitotic index, because
18    it induced M-phase arrest (Wise et al., 2005).
19
20    Other
21          Lead-induced oxidative damage has been investigated as a potential cause of genotoxic or
22    carcinogenic effects. Generally, the results suggest that Pb only produces low levels of reactive
23    oxygen species, but that it may inhibit some enzymes involved in oxidative metabolism (Table
24    AX5-6.20).  Thus, Pb may affect oxidative metabolism, but more work is needed to draw
25    meaningful conclusions.
26
27    5.6.5.4  Epigenetic Mechanisms Summary
28          The collective data  support the hypothesis that Pb can induce an epigenetic effect. Lead
29    can alter the expression of metabolic genes in cultured cells and may alter DNA repair, although
30    much more study is needed.  Lead may also  affect oxidative metabolism or interact with other
      December 2005                           5-151       DRAFT-DO NOT QUOTE OR CITE

-------
 1    metals, but again more study is needed. By contrast, it is unclear if Pb is mitogenic. It is
 2    mitogenic to the liver in animals, but it is not mitogenic in cultured cells. More study is needed
 3    to determine if this difference reflects differences between in vivo  and cell culture models or if
 4    this property is specific to only certain organs, e.g., the liver.
 5
 6    5.6.6   Overall Conclusions
 7          The overall conclusions have not changed much from the 1986 Pb AQCD.  Lead remains
 8    an ambiguous carcinogen in humans and a clear carcinogen in animals.  Cell culture studies
 9    support both of these conclusions, as effects in rodent cells were not seen in human cells.  Lead
10    does appear to be genotoxic in human epidemiology studies. By contrast, the laboratory studies
11    are more ambiguous in both animal and cell culture studies. In these systems, the genotoxicity in
12    culture is limited to SCE and, perhaps, to DNA-protein crosslinks.  For other endpoints, it is only
13    weakly active, if at all. Lead has not been evaluated sufficiently as a potential genotoxic hazard,
14    but this probably stems from the fact it appears to be weakly genotoxic.  The available data
15    suggest that Pb can damage sperm and affect fetuses. More work is urgently needed on this
16    topic.  Cell culture studies do support a possible epigenetic mechanism or co-mutagenic effects.
17
18
19    5.7  LEAD AND THE KIDNEY
20    5.7.1   Review of Earlier Work
21          This section summarizes key finding from the 1986 Pb AQCD on the effects of Pb  on the
22    kidney in animals. Human  studies published since 1986 are then reviewed in Section 6.4.
23          Both in vivo and in vitro studies on several different animal species revealed that renal
24    accumulation  of Pb is an efficient process that occurs in both proximal and distal portions  of the
25    nephron and at both luminal and basolateral membranes (Victery et al., 1979a; Vander et al.,
26    1977). The transmembrane movement of Pb appears to be mediated by an uptake process that is
27    subject to inhibition by several metabolic inhibitors and the  acid-base status of the organism.
28    Alkalosis increases Pb  entry into tubule cells via both the luminal and basolateral membranes
29    (Victery et al., 1979b).
30          Goyer et al. (1970a)  were principally responsible for defining the role of renal proximal
31    tubular nuclear inclusion bodies in the response to Pb intoxication. In addition to the early

      December 2005                           5-152        DRAFT-DO NOT QUOTE OR CITE

-------
 1    reports of nuclear inclusion bodies appearing in the proximal tubule following Pb exposure
 2    (Goyer et al., 1970b), biochemical studies on the protein components of isolated rat kidney
 3    intranuclear inclusion bodies have shown that the main component has an approximate molecular
 4    weight of 27 kDa (Moore et al., 1973) or 32 kDa (Shelton and Egle, 1982) and is rich in
 5    glutamate and aspartate. Goyer et al. (1970c) suggested that the intranuclear inclusion body
 6    sequesters Pb, to some degree, away from sensitive renal organelles and metabolic pathways.
 7    Goyer et al. (1975, 1978) also showed that single or repeated administration of CaNa2EDTA
 8    leads to the disruption of the nuclear inclusion bodies and their removal from the nuclei.  Rats
 9    treated for 24 weeks with both Pb and CaNa2EDTA had no inclusion bodies, but showed early
10    interstitial nephropathy.  As an extension of this study, Cramer et al. (1974) examined renal
11    biopsies from 5 Pb workers with 0.5 to 20 years of exposure. The two workers with normal
12    GFRs, and shortest exposure duration, showed intranuclear inclusion bodies, whereas the
13    remaining three workers had no intranuclear inclusions but showed peritubular fibrosis.
14          Formation of intranuclear inclusion bodies was a common pathognomic feature for all
15    species examined. In addition, proximal tubular cytomegaly and swollen mitochondria with
16    increased numbers of cytosomes were also observed (Fowler et al., 1980; Spit et al., 1981).  The
17    morphological changes were principally localized in the straight (S3) segments of the proximal
18    tubule. Goyer (1968) and Goyer et al. (1968) had demonstrated earlier that, after lead exposure,
19    mitochondria were not only swollen but had decreased respiratory control ratios  (RCRs) and
20    inhibited state-3 respiration.
21          Aminoaciduria has been reported in several studies (Studnitz and Haeger-Aronson, 1962;
22    Goyer et al., 1970b; Wapnir et al., 1979). Other studies have reported increased  urinary
23    excretion of electrolytes (e.g., sodium, potassium, calcium, water) following Pb administration
24    (Mouw et al., 1978).  Victery et al. (1981, 1982a,b, 1983) found that zinc excretion was
25    increased following injection of lead.
26          Wapnir et al. (1979) observed that Pb-acetate administration caused a reduction in renal
27    alkaline phosphatase activity and an increase in Mg-ATPase activity, but no significant changes
28    in NaK-ATPase activity.  On the other hand, Suketa et al.  (1979) found marked a decrease in
29    renal NaK-ATPase activity following a single oral administration of Pb-acetate at a dose of
30    200 mg/kg, but no change in Mg-ATPase.
      December 2005                           5-153        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Renal ALAD was found to be inhibited by Pb in both acute and chronic experiments
 2    (Silbergeld et al., 1982). Renal ALAD was similar to control levels when GSH was present but
 3    was significantly reduced in the absence of GSH (Gibson and Goldberg, 1970). Accumulation of
 4    both ALA and porphobilinogen was also observed in kidney tissue of Pb-treated rabbits,
 5    compared to controls. Other studies have not shown a reduction in renal ALAD following Pb
 6    exposure (e.g., Fowler et al., 1980).  Higher levels of Pb may be required to cause the reduction
 7    in ALAD reported by Silbergeld et al. (1982), and it may possibly involve Pb-binding proteins in
 8    the kidney.
 9
10    5.7.2   Markers of Renal Toxicity
11          The establishment and validation of new screening tests for nephrotoxic effects have been
12    principally due to the efforts of the Belgian group (Price et al., 1996; Price, 2000; Lauwerys
13    et al., 1992). They proposed the following battery of tests be used to screen both
14    environmentally exposed and occupationally exposed individuals: (1) measures of glomerular
15    integrity, i.e., urinary high-molecular weight proteins (albumin, IgG, transferrin); (2) measures
16    of tubular absorption and secretion, i.e., low-molecular weight proteins (retinol binding protein,
17    a-1-microglobulin; (3) measures of tubular integrity, i.e., enzymes, lysosomal N-acetyl
18    p-D-glucosaminidase (NAG), brush border alanine aminopeptidase, brush border intestinal
19    alkaline phosphatase, nonspecific alkaline phosphatase, a-glutathione-S-transferase (GST), and
20    brush border antigens (BB50, BBA, HF5); (4) measures of glomerular and distal tubular
21    function, i.e., prostanoids (thromboxane B2, prostaglandin F2 alpha, 6-keto prostaglandin
22    Fl alpha) ; (5) measures of glomerular structural proteins (fibronectin and laminin fragments);
23    and (6) measures of distal tubular function, i.e., Tamm-Horsfall protein and u-GST. Other useful
24    markers include urinary p2-microglobulin, as a marker of proximal tubular integrity; PGE2 and
25    PGF2, distal nephron markers; kallikrein, a marker of the distal tubule; lysozyme, ribonuclease,
26    and y-glutamyl transferrase, enzymes reflecting proximal tubule integrity; and sialic acid, an
27    extracellular matrix marker (Fels et al.,  1994; Pergande et al., 1994; Taylor et al., 1997). One or
28    several of these urinary markers have been used in screening tests for human Pb workers and in
29    animal studies of renal nephrotoxicity.
30          Questions have been raised about the usefulness of urinary NAG due to the absence of
31    light or electron microscopic changes in low-dose Pb-treated animals who showed substantial

      December 2005                            5-154       DRAFT-DO NOT QUOTE OR CITE

-------
 1    increases in NAG (vide infra) (Khalil-Maesh et al., 1993). Furthermore, Chia et al. (1994) found
 2    that urinary NAG in workers exposed to Pb correlated best with recent blood lead changes,
 3    suggesting that the increased urinary NAG activity reflected an acute response to a sharp
 4    increase in the renal Pb burden rather than to exocytosis. Questions have also been raised about
 5    the value of measuring the vasoconstricting prostariod cytokine thromboxane B2 (1X82) and the
 6    vasodilating prostanoid 6-keto prostaglandin Fl alpha (PGF1 alpha). Conflicting results have
 7    been reported in human Pb-exposed workers. Cardenas et al. (1993) reported an elevation in
 8    TXB2 and a diminution in PGF1 alpha in 41  Pb-exposed workers in contrast to 41 controls.
 9    Hotter et al. (1995), on the other hand, reported that both substances were increased in 69 Pb-
10    exposed workers in contrast to 62 controls. Blood Pb levels in the two worker groups were
11    comparable, i.e., 48 |ig/dL in the first group  and 43 |ig/dL in the second. In animal experiments
12    (Gonick et al., 1998), the excretion of both prostanoids was equal in low-Pb (100 ppm)-fed rats
13    as contrasted to normal controls after 3 months, despite an elevation in blood pressure in the Pb-
14    fed rats. Blood Pb in the Pb-fed rats averaged  12.4 |ig/dL compared to  1 |ig/dL in the controls.
15    Thus, measurements of these prostanoids remain of questionable value.
16          Attempts to validate nephrotoxic markers were conducted by Pergande et al. (1994),
17    utilizing Pb-exposed workers as contrasted to normal controls. They found that about 30% of the
18    Pb workers showed an increased excretion of ai-microglobulin, NAG, ribonuclease, and/or
19    Tamm-Horsfall protein, with positive correlations between these tubular indicators and blood Pb
20    concentration.
21
22    5.7.3    Biochemical Mechanisms of Lead  Toxicity
23          Nolan and Shaikh (1992) summarized what was known about biochemical mechanisms
24    underlying Pb-induced toxicity at that time.  A more detailed description based on recent animal
25    studies follows in the next section.
26          The initial accumulation of absorbed Pb occurs primarily in the kidneys.  This takes place
27    mainly through glomerular filtration and subsequent reabsorption, and, to a small extent, through
28    direct absorption from the blood.  Lead may be taken up by the renal tubular epithelial cells from
29    the basolateral side by active transport of the free ion.  Smaller amounts can also cotransport
30    with low molecular weight organic anions. The uptake of Pb through the renal brush border does
31    not appear to occur via any specific carriers. Instead, the process may involve binding of Pb to

      December 2005                          5-155      DRAFT-DO NOT QUOTE OR CITE

-------
 1    nonspecific surface sites on the brush border membrane, followed by internalization via
 2    endocytosis. Acute kidney damage due to Pb manifests primarily in the proximal tubules.  The
 3    ultrastructural changes observed in acute experimental Pb nephropathy include both specific and
 4    nonspecific effects on the proximal tubular epithelium, e.g., dilation of the endoplasmic
 5    epithelium, blebbing of the nuclear membrane, enlargement of the autophagosomes, changes in
 6    mitochondrial structure, formation of inclusion bodies.  Chronic exposure to Pb affects
 7    glomerular filtration, renal  clearance, and tubular reabsorption and can lead to renal failure from
 8    interstitial nephritis.
 9          Kidneys of chronically exposed individuals often show fewer or no nuclear inclusion
10    bodies compared to kidneys of acutely exposed individuals. The specific ultrastructural changes
11    associated with Pb nephropathy are the formation of cytoplasmic and nuclear Pb inclusion bodies
12    (discussed at greater length below). These inclusion bodies are not limited to the proximal
13    tubular epithelium, and have also been observed in peritoneum, astrocytes, neuroblastoma cells,
14    and osteoclasts upon Pb exposure.  The inclusion bodies are roughly spherical and typically
15    consist of an electron-dense core, with a fibrillary network at the periphery. Research has
16    revealed that the formation of the nuclear inclusion bodies is preceded by the synthesis of
17    cytoplasmic inclusion bodies with a very similar structure. A protein unique to these structures
18    is rich in acidic amino  acids and has an isoelectric point of 6.3 and a molecular weight of
19    32 kDa. Two additional proteins with apparent molecular weights of 11.5  kDa and 63 kDa have
20    been identified in kidney extracts.  Both of these proteins have a high affinity, but little capacity,
21    for binding lead. A Pb-binding protein of 12 kDa molecular weight was identified in the
22    supernatant of brain homogenate from Pb-treated rats.  A Pb binding protein of 10 kDa has also
23    been isolated from the  erythrocytes of Pb-exposed workers.
24          Mitochondrial function, in addition to structure, is very sensitive to lead. Changes include
25    the uncoupling of oxidative phosphorylation, decreased substrate oxidation, and modification of
26    ion transport processes. Other effects of Pb on cellular energetics include chelation of ATP and
27    inhibition of microsomal NaK-ATPase. These changes may account for the proximal tubular
28    dysfunction seen with acute Pb poisoning in children.
29          A new area of investigation of the mechanism of Pb toxicity was initially proposed by
30    Quinlan et al. (1988) and Hermes-Lima et al. (1991). Both investigators proposed that free
31    radicals, or ROS, stimulated by lead, may accelerate iron-dependent lipid peroxidation, causing

      December 2005                           5-156        DRAFT-DO NOT QUOTE OR CITE

-------
 1   tissue injury. Hermes-Lima et al. (1991) stated further that ALA, which is formed in large
 2   amounts in Pb toxicity, may undergo enolization and autoxidation, yielding ROS. Autoxidation
 3   of ALA, in the presence or absence of iron complexes, yields superoxide, peroxide, and hydroxyl
 4   radicals. Gurer and Ercal (2000), based on several animal studies to be discussed below, have
 5   proposed that antioxidant supplementation following Pb exposure may provide a partial remedy
 6   by restoring the cell's antioxidant capacity.
 7
 8   5.7.4   Animal Studies
 9          Two excellent review articles have been written about the effects of heavy metals on, and
10   their handling by, the kidney (Barbier et al., 2005) as well as the mechanisms of kidney cell
11   injury from metals (Fowler, 1992). The interested reader is directed to these reviews, although
12   individual effects and mechanisms will be discussed subsequently.
13
14   5.7.4.1   Lead Toxicokinetics
15          deVries et al. (1998) published a model for Pb toxicokinetics to be used in planning
16   treatment.  The model is a four-compartment model with first-order kinetics.  The four
17   compartments of this model are blood, bone, liver, and kidney.  Soft tissues are represented by
18   the kidney and liver compartments. In  addition, intake and excretion are included in the model.
19   Excretion of Pb is mainly via the kidneys (70 to 80%), via bile and feces (15%), via nails, hair,
20   and sweat (8%).  The blood makes up the central compartment from which Pb is distributed after
21   uptake in the body.  The blood compartment contains about 4% of the total body burden of lead,
22   and within this compartment, the Pb is mainly taken up by erythrocytes. The half-life of Pb in
23   blood is about 30 days. From  the blood, Pb is distributed relatively quickly to the soft tissues
24   and bone. The distribution constant from blood to bone is much higher than the one from bone
25   to blood, resulting in the accumulation of Pb in bone.  The half-life in the soft tissues is about
26   30 to 40 days. Most of the body burden of Pb can be found in the bone compartment (-94%),
27   where the half-life of Pb is several decades. Because of the vast amount of Pb in bone, a
28   rebound in blood Pb usually occurs after chelation therapy. This model can be compared with a
29   toxicokinetic model developed by Marcus (1985a,b,c) and further explored by Hogan et al.
30   (1998), as discussed in Chapter 4 of this document.
     December 2005                           5-157       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Dieter et al. (1993) examined the effect of the nature of the Pb salt on the oral intake of Pb
 2    in male F344 rats.  For 30 days, they administered doses of 0, 10, 30, and 100 ppm Pb in the
 3    form of soluble Pb-oxide, Pb-acetate, Pb-sulfide, and Pb-ore. At 100 ppm of Pb-acetate or
 4    soluble Pb-oxide, the rats developed -80 |ig/dL of blood and -200 jig/g of bone Pb levels,
 5    whereas rats fed Pb-sulfide or Pb ore developed -10 |ig/dL of blood Pb and 10  jig/g of bone
 6    lead. In rats fed Pb-acetate or soluble Pb-oxide, blood Pb progressively increased with
 7    increasing dose, while in the other two groups measurable levels of Pb were observed only at the
 8    highest dose (100 ppm).
 9
10    5.7.4.2 Pathology, Ultrastructural, and Functional Studies
11          Two important series of studies contrast the pathological and functional changes in the
12    kidney after prolonged exposure to lead, with and without chelation therapy (i.e., DMSA  or
13    CaNa2EDTA).  In the first series of 3 long-term studies, Khalil-Manesh et al. (1992a,b, 1993a)
14    described the effects of Pb-acetate on renal function and morphology in male Sprague-Dawley
15    rats fed a low-calcium diet. Lead acetate was used in concentrations of 0.5% (high dose)  and
16    0.01% (low dose) in drinking water for periods from 1 to 12 months, and then Pb-exposed
17    animals were compared to pair-fed controls (12 rats in each group). In all studies GFR was
18    measured as 125I-iothalamate clearance by a single injection technique. Urinary markers
19    included NAG, GST, and brush border antigens (BB50, HF5, and CG9) and were expressed as
20    units/g creatinine.  Blood and urine Pb were measured prior to sacrifice in each group of animals.
21    Wet and dry weights of kidneys were determined, then the kidneys were processed for light,
22    electron, and immunofluorescent microscopy.
23          In the first study (Khalil-Manesh et al., 1992a), animals treated with continuous high-dose
24    Pb for 12 months reached a maximum blood Pb of 125.4 ±10.1 |ig/dL after 6 months, at which
25    time the dose of Pb was reduced from 0.5% to 0.1%.  Blood Pb  at the end of 12 months averaged
26    55 |ig/dL. Urine Pb remained above 100 jig/g creatinine at all times, but it was highest at
27    3 months, averaging 340 jig/g creatinine. In the Pb-treated animals, GFR was increased above
28    controls at 3 months (1.00 ± 0.14 vs. 0.83 ± 0.26 mL/min/ 100 g body wt, p = 0.05), then
29    declined after 6 months to 0.78 ±0.16 vs. 0.96 ± 0.08 mL/min/100 g body wt in controls
30    (Figures 5-7.1 and 5-7.2).  As indicated by the ratio of kidney dry/wet weight, increased kidney
31    tissue mass  was  observed during the first 3 months of Pb exposure, but decreased tissue mass

      December 2005                            5-158        DRAFT-DO NOT QUOTE OR CITE

-------
                                       P<0.05
                 P<0.05
                           1.5-1
                       CD
                           0.0
                                  1      3      6      9     12
                                    Duration of exposure, months
Figure 5-7.1.   Changes in GFR of experimental high-dose lead and control animals with
               duration of exposure to lead.  Open and closed bars represent GFR in
               experimental and control rats, respectively.

Source: Khalil-Manesh et al. (1992a), with permission.
                 1.50
             o
             •Q
             O)
             cc
             EL
                 1.00 -
                 0.50 -
                 0.00
                     r = 0.703
                     0    15    30    45    60   75    90    T05   120   135  150

                                          Blood lead,
Figure 5-7.2.   Correlation between GFR and blood lead during the first 6 months of
               high-dose lead exposure.

Source: Khalil-Manesh et al. (1992a), with permission.
December 2005
5-159
DRAFT-DO NOT QUOTE OR CITE

-------
 1    was observed by 12 months. With regard to urinary markers, NAG was elevated above control
 2    levels at 3, 6, and 9 months of Pb exposure; GST was elevated at 3, 6, and 12 months of Pb
 3    exposure; and no significant differences were observed in the brush border antigens.  Proximal
 4    tubular nuclear inclusion bodies were present at all time periods in Pb-treated animals.
 5    Enlargement of proximal tubular cells and nuclei were seen beginning at 3 months.  At 6 months,
 6    focal tubular atrophy and interstitial fibrosis appeared, increasing in extent up to 12 months.
 7    Mitochondrial alterations, consisting of rounding and elongation, appeared by 1 month and were
 8    persistent. Glomeruli were normal through 9 months, but, at 12 months, they showed focal and
 9    segmental sclerosis.  There were no electron-dense deposits and immunofluorescent studies were
10    negative. Renal arteries and arterioles were normal at all time point examined.
11          The second study (Khalil-Manesh et al., 1992b) consisted of the discontinuation of both
12    the high- and low-dose Pb  exposure after 6 months, then treatment with three courses of DMSA
13    or discontinuation of high-dose Pb alone after 1, 6, and 9 months of Pb  feeding.  Controls were
14    pair-fed, exposed to Pb for 6 months,  then removed from exposure for 6 months without
15    receiving DMSA. Low-dose Pb-treated rats showed no significant pathologically with or
16    without DMSA treatment but  exhibited a significant increase in GFR after DMSA treatment
17    (1.09 ± 0.19 vs. 0.88 ±  0.22 mL/min/100 g body weight; P < 0.03) (Figure 5-7.3). Urinary
18    markers  remained unchanged, and there were no structural alterations by light or electron
19    microscopy.  High-dose Pb-treated animals showed no functional  or pathologic changes when Pb
20    exposure was discontinued after 1 month. However, when the duration of exposure was 6 or
21    9 months, GFR was decreased and serum creatinine and urea nitrogen were increased compared
22    to controls.  Tubulointerstitial disease was severe. Administration of DMSA resulted in an
23    improvement in GFR (Figure  5-7.3) and a decrease in albuminuria, together with a reduction in
24    size and  number of nuclear inclusion bodies in proximal tubules. However, tubulointerstitial
25    scarring  was only minimally reduced. In conclusion, except for a brief initial exposure,
26    discontinuation of high-dose Pb exposure failed to reverse Pb-induced renal damage. Treatment
27    with the  chelator, DMSA, improved renal function but had less effect on pathologic alterations.
28    Because GFR improved after DMSA treatment in both low- and high-dose Pb-treated animals,
29    irrespective of the degree of pathologic alterations, it may be concluded that the DMSA effect is
30    most likely mediated by hemodynamic changes.
31

      December 2005                           5-160        DRAFT-DO NOT QUOTE OR CITE

-------
                      1.50
                      1,00-
                      0.50
                                 High Lead
               Low Lead
                            ED6     C12     DMSA       ED6
                                                GROUPS
                 C12
                         DMSA
     Figure 5-7.3.   GFR in high-lead and low-lead experimental discontinuous (ED6) and
                    DMSA-treated rats (DMSA) as compared to controls (C12). All rats
                    were studied at 12 months.
                     *p < 0.01 when compared to ED6 and C12.
                    **p < 0.05 when compared to ED6.
     Source: Khalil-Manesh et al. (1992b), with permission.
 1         The third study (Khalil-Manesh et al., 1993a) examined the course of events over
 2    12 months in continuous low level Pb-exposed animals. Maximum blood Pb levels in
 3    experimental animals were reached at 3 months, averaging 29.4 ±4.1 |ig/dL.  GFR was found
 4    to be significantly increased above pair-fed controls at 1 and 3 months, but it was normal at
 5    other time periods (1 month experimental,  1.18 ± 0.12 vs. control, 0.76 ± 0.15mL/min/100 g;
 6    p < 0.001; 3 month experimental, 1.12 ± 0.16 , vs. control, 0.86 ± 0.10 mL/min/100 g; p < 0.001)
 7    (Figure 5-7.4).  Levels of urinary NAG in Pb-exposed rats exceeded control levels at all time
 8    periods, except at 12 months, when the normal increase with aging obscured differences between
 9    experimental animals and controls (Figure 5-7.5). In contrast, urinary GST, a more specific
10    marker of metal-associated proximal tubular injury, was normal at all time periods. Proximal
11    tubular nuclear inclusion bodies were sparse and were observed only at 1 and 3 months.
12
     December 2005
5-161
DRAFT-DO NOT QUOTE OR CITE

-------
           O)
           o
           o
               2.00
               1.50
           J|   1.00

           i

           oc
           IX
           CD
               0.50
               0.00
                        p<0 001
p<0 001
                                         3             6


                                    Duration of Exposure (Months)
Figure 5-7.4.   Changes in GFR in experimental and control rats, at various time periods.


Source: Khalil-Manesh et al. (1993a).
             600
             500
         (D
         C

         I   400
         as
            300
         "  200
         O
              100
                      p<0 001     p<0 001
       p<0 05
p<0.001
                                   369


                                   Duration of Exposure (Months)
Figure 5-7.5.   Urinary NAG concentration in experimental and control rats at various

               time periods.


Source: Khalil-Manesh et al. (1993a).
December 2005
   5-162        DRAFT-DO NOT QUOTE OR CITE

-------
 1          No other pathologic alterations were found in the kidneys until 12 months of exposure,
 2    when mild tubular atrophy and interstitial fibrosis were seen. The absence of changes in urinary
 3    GST accorded with the relative absence of morphologic changes, whereas the observed increases
 4    in urinary NAG suggest that this enzyme may be an overly sensitive indicator of tubular injury,
 5    more probably reflecting upregulation of the enzyme even in the absence of tubular injury.
 6    It should be noted that both low-dose Pb-treated animals and high-dose Pb-treated animals
 7    showed a "hyperfiltration" phenomenon during the first 3 months of Pb exposure. This
 8    observation could be invoked as a partial explanation for the late changes of glomerulosclerosis
 9    in the high-dose animals, but it cannot explain the lack of glomerular changes in the low-dose
10    animals. Thus, these studies join those of Roels et al. (1994) and Hu  (1991) in humans that
11    indicate that Pb nephropathy should be added to diabetic nephropathy as diseases that lead to
12    early hyperfiltration.
13          The second series of studies were performed by Sanchez-Fructuoso et al.  (2002a,b).
14    Sanchez-Fructuoso et al. (2002a,b) evaluated the effect of CaNa2EDTA on tissue mobilization of
15    Pb in Wistar rats initially treated with 500 ppm Pb-acetate for 90 days, followed by treatment
16    with three courses of CaNa2EDTA 50 mg/kg/day for 5 days, separated by 9 days, or placebo.
17    Lead levels were measured in blood, urine, kidney, liver, brain, and femur. There was no change
18    in bone Pb after CaNa2EDTA compared to placebo, but Pb levels were significantly reduced in
19    all other tissues (Figure 5-7.6). The authors emphasized that there was no redistribution to brain.
20    Cory-Slechta et al. (1987) had originally reported that with CaNa2EDTA chelation in rats Pb is
21    preferentially mobilized from bone  and then redistributed to other organs, including brain. The
22    Sanchez-Fructuoso et al. (2002a,b) findings stand in contrast, explained by the authors as due to
23    a 3-fold higher level of CaNa2EDTA used by Cory-Slechta et al. (1987).
24          Sanchez-Fructuoso et al. (2002b) also evaluated  pathologic changes, as well as the
25    response of ALAD activity before and after CaNa2EDTA treatment in the same rats. In the
26    90-day Pb-treated animals, the  main findings were hypertrophy and vacuolization of medium  and
27    small arteries (Figure 5-7.7); mucoid edema and muscular hypertrophy in arterioles; loss of cell
28    brush borders, cell loss, and intranuclear inclusion bodies in the proximal tubule; and fibrosis  and
29    the presence of infiltrates in the interstitial component.  Treatment with CaNa2EDTA slowed the
30    progression of most alterations (Figure 5-7.8) and resulted in a diminution in nuclear inclusion
31

      December 2005                          5-163        DRAFT-DO NOT QUOTE OR CITE

-------
I
T»

2
             106


              80


              60


              40


              20

              0
                          111  12*   139 *nr»
                                                              JO   111   12$  1SS  days
I *
i 4.
           1  *
                 End F*b * x|»o*y w
                 «0    «0   111  123   139
                                                             Ertd Pb ttidMMtm
                                                          60   90   111   iaS  1$$  iln.vs
Figure 5-7.6.   Kidney, liver, brain, and bone Pb levels in 56 Pb-exposed rats.
               After 90 days of poisoning, animals were administered serum saline
               (solid line) or calcium disodium EDTA (broken line).

Source: Sanchez-Fructuoso et al. (2002a), with permission.
                100



                 60



                 20

                  0-*"
                      60     90"  111    !25    139"  111    125    139 Days
                                     No treatment             EDTA

Figure 5-7.7.   Percentage of moderate and severe hypertrophy and vacuolization lesions in
               small and medium sized arteries in the kidney of lead-exposed rats.

Source: Sanchez-Fructuoso et al. (2002b), with permission.
December 2005
                               5-164
DRAFT-DO NOT QUOTE OR CITE

-------
                     100
                                       111    125    139    11!    125    139 Days
                                         No treatment            EOT A
     Figure 5-7.8.   Percentage of moderate and severe muscular hypertrophy lesions in
                    arterioles of the kidney in lead-exposed rats.
     Source: Sanchez-Fructuoso et al. (2002b), with permission.
 1   bodies.  ALAD activity was reduced from 3.18 ± 0.52 U/mL in controls, to 0.82 ± 0.16 U/mL in
 2   the Pb-exposed rats. In the rats treated with CaNa2EDTA, ALAD returned to near control levels
 3   (2.98 ± 0.41 U/mL) at 137 days. It is surprising that such remarkable vascular changes were
 4   noted in this study, while none were noted in Khalil-Manesh et al. (1992a), even with high-dose
 5   Pb for longer periods of time.  The kidney content of Pb (mean 74.6 |ig/g) was also lower than
 6   the mean kidney content at 12 months (294 |ig/g) in the Khalil-Manesh et al. (1992a) study.
 7   The only explanation for these striking differences that can be offered is that different strains of
 8   rats were employed, i.e., Wistar in the Sanchez-Fructuoso (2002b) study and Sprague-Dawley in
 9   the Khalil-Manesh et al. (1992a) study. The presence or absence of hypertension cannot be
10   invoked as an explanation, because in another Khalil-Manesh et al. (1993b) study the low-dose
11   Pb animals became hypertensive while the high-dose animals did not.  These and other related
12   studies are summarized in Table AX5-7.1.

13   5.7.4.3  Biochemical Mechanisms of Lead Toxicity
14   Role of Free Radicals (Reactive Oxygen Species)
15         Since the early  1990s, it has been appreciated that free radicals, now known as reactive
16   oxygen species (ROS), are involved in the manifestations of Pb poisoning, presumably via their
     December 2005
5-165
DRAFT-DO NOT QUOTE OR CITE

-------
 1    adverse effects on tissue integrity and/or their vasoconstrictive effects on vascular endothelium.
 2    Wolin (2000) produced an extensive review of individual ROS, and their interactions with NO,
 3    the major endogenous vasodilator, which acts via a second messenger, cGMP.  The production
 4    of ROS often begins with a one-electron reduction of molecular oxygen to superoxide anion
 5    (02) by various oxidases. NAD(P)H oxidases are the principal enzymes involved. Superoxide
 6    anion is a negatively charged free radical that can be broken down to hydrogen peroxide (H2O2)
 7    by superoxide dismutase (SOD) or can interact with NO to form the highly reactive peroxynitrite
 8    ion (ONOO ), which, because of its extremely short half-life, is measured  as its reaction product,
 9    tissue nitrotyrosine. Catalase and glutathione (GSH) peroxidase (GSHRx) metabolize H2O2 to
10    Compound I and oxidized glutathione (GSSG), respectively, while myeloperoxidase metabolizes
11    H2O2 to hypochlorous acid (HOC1).  The reaction of H2O2 with ferrous ion results in the
12    formation of hydroxyl ion (*OH). ROS can be scavenged by endogenous thiols (e.g., GSH) or
13    exogenous thiol, e.g., N-acetylcysteine (NAC).  ROS can be measured as the concentration of the
14    lipid peroxidation product, malondialdehyde-thiobarbituric acid (MDA-TBA) or by the more
15    recently introduced F-2 isoprostanes.
16          Kumar and Das (1993) explored the involvement of ROS in the pathobiology of human
17    essential hypertension.  They found that plasma levels of lipid peroxides were higher in subjects
18    with uncontrolled essential hypertension compared to normal controls. Angiotensin II, a potent
19    vasoconstrictor, was found to stimulate free radical generation in normal leukocytes, which was
20    thought to inactivate NO, and possibly prostacyclin, which can  lead to increased peripheral
21    vascular resistance and hypertension.
22          Hermes-Lima et al. (1991) also explored the involvement of ROS in Pb poisoning.  They
23    described the process of autoxidation of ALA in the presence or absence of iron complexes,
24    which yields free radicals. Free radicals are also produced by Pb-stimulated iron-dependent lipid
25    peroxidation, as determined by quantification of thiobarbituric acid-reactive species (TEARS).
26    Pereira et al. (1992) demonstrated that chronically ALA-treated rats (40 mg/kg body weight
27    every 2 days for 15 days) under swimming training reached fatigue significantly earlier than the
28    control group, as well as demonstrating decreased mitochondrial enzymatic activities.  In vivo
29    prooxidant properties of ALA were also suggested by the observed increase of CuZnSOD in
30    brain, muscle, and liver of untrained rats submitted to chronic treatment with ALA.
      December 2005                           5-166       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Ercal et al. (1996) contrasted the effects of treatment with DMSA or NAC in Pb-exposed
 2    C57BL/6 mice. Five weeks of Pb exposure was found to deplete GSH levels, increase GSSG,
 3    and promote MDA production in both liver and brain samples.  Glutathione levels increased and
 4    GSSG and MDA levels decreased in groups of Pb-exposed mice that received 1 mmol/kg DMSA
 5    or 5.5 mM/kg NAC for 7 days prior to  sacrifice.  Treatment with DMSA caused reduction in
 6    blood, liver, and brain Pb levels consistent with its function as a chelating agent, while treatment
 7    with NAC  did not reduce these Pb levels. However, NAC treatment reduced indices of oxidative
 8    stress in both brain and liver samples.  Concentrations of blood Pb in controls were 0.5 ±
 9    0.5 |ig/dL;  in Pb-treated mice, were 36.5 ± 2.4 |ig/dL; in Pb + DMSA-treated mice, were 13.7 ±
10    1.3 |ig/dL;  and in Pb + NAC-treated mice, were 36.0 ±3.5  |ig/dL. Thus both DMSA and NAC
11    acted as antioxidants, presumably via their thiol groups, but only DMSA reduced the
12    concentration of lead.
13          Vaziri and co-workers (Gonick et al., 1997; Ding et  al., 1998, 2000, 2001; Vaziri et al.,
14    1997, 1999a,b, 2000, 2001a,b, 2003; Zhou et al., 2002; Ni et al., 2004) have published a number
15    of articles relating to the production of ROS and alterations in enzymatic activities in Pb-induced
16    hypertension.  These were discussed in detail in Section 5.5 but are described briefly here. In the
17    majority of studies, Pb-induced hypertension was produced by the administration of Pb-acetate,
18    100 ppm in drinking water, for 3 months to male Sprague-Dawley rats. Early studies (Gonick
19    et al., 1997) revealed that hypertension could occur in the absence of changes in NO  or cGMP
20    but with an attendant rise in plasma and kidney MDA-TEA, indicating an increase in ROS. In a
21    second study, Ding et al. (1998) showed that infusion of arginine, the precursor of NO, or
22    DMSA, a thiol Pb chelator and antioxidant, reduced blood pressure to or towards normal, while
23    simultaneously increasing  depressed urinary NO and decreasing an elevated MDA-TBA. Ding
24    et al.  (2000, 2001) further  showed that  the ROS species, *OH, measured as salicylate-trapped
25    2,3 dihydroxybutyric acid, was increased in plasma and cultured rat aortic endothelial cells after
26    exposure to lead, and that dimethylthiourea, a reputed scavenger of *OH, returned blood pressure,
27    MDA-TBA, *OH, and nitrotyrosine to or towards normal.  Ni et al., in 2004, demonstrated in
28    both human coronary endothelial (EC)  and vascular smooth muscle cells (VSMC) that Pb-acetate
29    also increased superoxide (demonstrated by flow cytometry using hydroethidine) and H2O2
30    (demonstrated with dihydrorhodamine) production.  After long-term (60-h) exposure, detectable
31    superoxide levels fell to near normal while H2O2 production remained high.

      December  2005                           5-167        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Vaziri et al. (1997) showed that lazaroids, a class of non-thiol antioxidant, also restored
 2    blood pressure, NO, and MDA-TBA to normal. Vaziri et al. (1999a) studied rats treated for
 3    12 weeks with either Pb-acetate alone or Pb-acetate + vitamin E-fortified food (5000 units/kg rat
 4    chow).  They measured urinary excretions of stable NO metabolites (NOX) and plasma and tissue
 5    abundance of nitrotyrosine, the footprint of NO oxidation by ROS. The Pb-treated group showed
 6    a marked rise in blood pressure; a significant increase in plasma and kidney, heart, liver, and
 7    brain nitrotyrosine abundance; and a substantial fall in urinary NOX excretion. Concomitant
 8    administration of high-dose vitamin E ameliorated hypertension and normalized both urinary
 9    NOX excretion and tissue nitrotyrosine without altering tissue Pb content. Vaziri et al. (1999b)
10    also measured eNOS and iNOS in the aorta and kidney of Pb-treated and Pb + vitamin E-treated
11    rats.  Lead treatment increased both isotypes in aorta and kidney, signifying increased NO
12    production, while Pb + vitamin E lowered aortic, but not kidney, expression of eNOS and iNOS.
13    Vaziri and Ding (2001) tested the effect of lead, 1 ppm, on cultured human EC cells. Lead was
14    tested alone or with either the SOD-mimetic agent, tempol, or a potent antioxidant lazaroid
15    compound (both at 10"8 or 10"7mol/L) on eNOS expression and NO production. Lead-treated
16    cells  showed a significant upregulation of endothelial eNOS, increase in protein abundance, and
17    increase in the production of NO metabolites.  Treatment with either tempol or lazaroids
18    abrogated the Pb-induced upregulation of eNOS protein and NOX production.  Vaziri et al.
19    (2001) also studied increases in NOS isoforms in vivo in Pb-induced hypertension and reversal
20    by tempol. Both eNOS and iNOS were increased in kidney, aorta, and heart, while NOS was
21    increased in cerebral cortex and brain stem, of Pb-treated rats; blood pressure  and NOS isoforms
22    were returned to normal by tempol. Vaziri et al. (2003) determined whether the oxidative stress
23    in animals with Pb-induced hypertension is associated with dysregulation of the main antioxidant
24    enzymes (i.e., SOD, catalase, and GSHPx), or increases in the superoxide-producing enzyme
25    NAD(P)H oxidase.  At the conclusion of the experiment, immunodetectable CuZnSOD,
26    MnSOD, catalase, GSHPx, and the gp91phox subunit of NAD(P)H oxidase were measured by
27    Western analysis in the kidney, brain, and left ventricle of control and Pb-exposed rats.  Lead
28    exposure resulted in a significant increase in kidney and brain CuZnSOD with a significant
29    increase in brain,  and insignificant increase in kidney and heart, gp91phox.  In contrast, MnSOD,
30    catalase, and GSHPx in the kidney, brain, and left ventricle were unchanged.  Incubation with
31    Pb-acetate did not alter SOD activity in vitro. Thus, animals with Pb-induced hypertension

      December 2005                           5-168       DRAFT-DO NOT QUOTE OR CITE

-------
 1    exhibited oxidative stress, which was associated with mild upregulation of the superoxide-
 2    generating enzyme NAD(P)H oxidase, with no evidence of quantitative SOD, catalase, or
 3    GSHPx deficiencies.
 4          Vaziri et al. (2000) demonstrated that induction of oxidative stress in normal animals
 5    (by feeding the GSH synthase inhibitor, buthionine sulfoximine, 30 mmol/L in drinking water
 6    for 2 weeks) led to an increase in blood pressure, a reduction of urinary NOX, a 3-fold decrease in
 7    liver GSH, and an increase in nitrotyrosine in kidney, aorta, heart, liver and plasma.
 8    Administration of vitamin E + ascorbic acid ameliorated hypertension and mitigated
 9    nitrotyrosine accumulation despite persistent GSH depletion.  This experiment demonstrated the
10    importance of GSH in protecting against the adverse effects of ROS accumulation in normal
11    animals. Majority of the studies reported by Vaziri and co-workers indicated that low Pb
12    exposure induced hypertension to be primarily mediated by ROS-induced depletion of NO.
13    NO production, on the other hand, is stimulated, as shown by the increase in eNOS and iNOS.
14    Enzymatic control of ROS levels by low Pb is achieved by upregulation of NAD(P)H oxidase
15    with no decrease in SOD, catalase, or GSHPx, i.e., the enzymes that breakdown ROS.
16    Scavengers of ROS ameliorate the elevated blood pressure, while the depletion of the
17    endogenous methyl scavenger, GSH, increases blood pressure in  normal animals. No studies
18    have been done to date to address the question of why high-dose Pb administration does not lead
19    to hypertension.
20          Farmand et al. (2005)  pursued enzymatic studies by activity measurements and measures
21    of protein abundance in the rat kidney and aorta following the protocol of Gonick et al. (1997)
22    whereby rats are fed Pb-acetate 100 ppm for 12 weeks. They demonstrated that the activities of
23    CuZnSOD and catalase were increased by Pb administration in renal cortex and medulla,
24    whereas GSHPx was unchanged.  In the thoracic aorta, Pb exposure resulted in significant
25    upregulation of CuZnSOD activity, while catalase and GSHPx activities were unchanged,
26    CuZnSOD, MnSOD, and catalase protein abundance were likewise unchanged.  However,
27    guanylate cyclase protein abundance in the thoracic aorta was decreased. The authors suggested
28    that the Pb-induced compensatory upregulation of CuZnSOD and catalase and the decrease in
29    aortic guanylate cyclase may be related to Pb-induced hypertension.
30          Gurer et al. (1999a) evaluated whether captopril, an ACE inhibitor, acted as an
31    antioxidant in Pb-exposed F344 rats. Lead acetate was given in drinking water for 6 weeks.

      December 2005                           5-169       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Group I were the controls; group II received 1100 ppm Pb for 5 weeks and plain water during the
 2    week 6; group III received 1100 ppm Pb for 5 weeks and, during the week 6, received water
 3    containing captopril (10 mg/day).  Blood Pb concentrations in the control group measured
 4    0.8 ng/dL; in the Pb treated group, 24.6 ± 20 ng/dL; in the Pb + captopril group, 23.8 ±
 5    1.6 |ig/dL. MDA concentrations in liver, brain, and kidney were increased by Pb administration
 6    and reduced to or towards normal by the Pb + captopril treatment.  GSH concentrations were
 7    decreased by Pb administration and restored by Pb + captopril treatment, whereas GSSG
 8    concentrations were increased by Pb administration and reduced by Pb + captopril treatment.
 9    Thus, this study showed that captopril was capable of augmenting the reducing capacity of the
10    cells by increasing GSH/GSSG ratios without affecting blood Pb concentrations.
11          McGowan and Donaldson (1987) examined total nonprotein sulfhydryl and GSH
12    concentrations in liver and kidney as well as GSH-related free amino acid concentrations in liver,
13    kidney, and plasma in 3-week-old Pb-treated (2000 ppm dietary lead) chicks.  Cysteine,
14    converted from methionine, is the rate-limiting amino acid in GSH formation. The availability
15    of glutamate, cysteine,  and glycine becomes important in the restoration of depleted GSH.
16    GSH, nonprotein sulfhydryl groups, glycine, and methionine were increased versus controls in
17    the liver, but only nonprotein sulfhydryl,  glycine, cysteine, and cystathionine increased in the
18    kidney. Plasma levels of cysteine, taurine, and cystathione were reduced.  Thus, Pb,  for
19    short periods of time, increases GSH turnover. These and other studies are summarized in
20    Table AX5-7.2.
21
22    Effect of Lead on Selective Renal Enzyme Levels
23    Effects of Lead on Renal NAG
24          Dehpour et al. (1999) studied NAG release by the rat kidney perfused with Pb-acetate at
25    10, 20, and 50 ng/dL for 120 min, or Pb + arginine (the  substrate for NO), or Pb + L-NAME
26    (an inhibitor of NOS).  Lead acetate caused a time and concentration-dependent increase in
27    enzymuria. Addition of arginine decreased, while addition of L-NAME increased, Pb-induced
28    NAG release.  Histologic studies showed damage to some of the proximal tubule epithelial cells
29    in rats treated with 50 |ig/dL Pb-acetate, damage that which  was increased further by the addition
30    of L-NAME.
31
      December 2005                           5-170       DRAFT-DO NOT QUOTE OR CITE

-------
 1   Effect of Lead on Renal GST
 2          Two studies (Moser et al., 1995; Oberley et al., 1995) reported the effects of Pb
 3   administration on GST isoforms in developing rat kidney. In the first study (Moser et al., 1995),
 4   rats were treated either acutely (14- and 50-day old rats given three daily injections of Pb-acetate,
 5   114mg/kg) or chronically (Pb levels of 0, 50, 250, and 500 ppm in drinking water for 1, 2, 3, 4,
 6   and 7 weeks postnatal).  Chronic treatment rats were also given a 0.66% low calcium diet or
 7   standard rat chow. Essentially all kidney cytosolic GSTs (Ybl, Yb2, Yp, Ycl, Yl, Yb3, Yal,
 8   Ya2, Yk) increased in the acute experiment (1.1- to 6.0-fold). In the chronic experiment, all but
 9   one isoform (Yb3) increased, and these results were markedly exacerbated by placing the rats on
10   a low-calcium diet (Ybl and Yp increased >25-fold).  In the second study (Oberley et al., 1995),
11   pregnant rats were given 250 ppm Pb from conception until weaning, then pups received 500
12   ppm from weaning until termination at either 3 or 7 weeks of age. By 7 weeks, proximal tubular
13   cells showed intranuclear inclusions, tubular injury, and interstitial fibrosis. Creatinine
14   clearances were reduced (0.55 + 0.05 versus 1.05 + 0.07 mL/min/lOOg; P< 0.001). Treatment
15   with Pb also caused large increases in the immunoreactive protein of Yc, Yk, Ybl, and Yp GST
16   subunits in proximal tubules but did not increase in the antioxidant enzymes CuZnSOD, catalase,
17   and GSHPx.
18          Another experiment that examined the effect of an acute dose of Pb  as Pb-nitrate
19   (100 |imol/kg IV) on GST levels in rat liver and kidney was reported by Planas-Bohne and
20   Elizade (1992).  Seventy hours after injection, there was a marked increase in GST activity in
21   both organs, accompanied by induction of the isoenzyme GST 7-7 in the liver.
22          The relationship between GST induction by acute exposure to Pb-acetate and oxidative
23   stress was explored by Daggett et al. (1998). Rats in the 72-h and 7-day experimental groups
24   received three consecutive daily injections of 114 mg/kg body weight of Pb-acetate. The level of
25   kidney GST was increased at 3, 6, 12, and 24 h after injection, but MDA levels remained
26   unchanged.  Immunohistochemical  markers of oxidative stress and NO  production (MnSOD,
27   eNOS, iNOS, and 4-hydroxy-2-nonenal) also did not change. The authors  concluded that the
28   GST changes were not the result of oxidative stress.
29          Witzman et al.  (1998) and Kanitz et al. (1999) utilized two-dimensional (2-D) gel
30   electrophoresis to explore protein markers of Pb exposure. Witzman et al.  (1998) gave three
31   consecutive IP injections of Pb-acetate  (114 mg/kg) to Sprague-Dawley rats, sacrificed them on

     December 2005                          5-171       DRAFT-DO NOT QUOTE OR  CITE

-------
 1    the fourth day, and subjected the cytosolic fraction of kidney homogenate to 2-D gel
 2    electrophoresis.  Lead exposure caused detectable inductions in both GSTP1 and GSTM1 and
 3    caused quantifiable charge modifications in GSTP1.  Kanitz et al. (1999) examined kidney
 4    protein expression in male rabbits injected with Pb-acetate (260, 360, or 100 |ig/kg) designed to
 5    produce blood levels of 20, 40, or 80 |ig/dL.  Injections were given during weeks 6 to 10,
 6    followed by maintenance doses during study weeks 11 to 20.  Kidney homogenates were
 7    subjected to 2-D electrophoresis.  Significant quantitative changes occurred in 12 proteins in a
 8    dose-related manner. Four proteins cross-reacted with anti-rat GSTpl (rc-GST). Thus, both
 9    studies confirmed GST induction by lead.
10          Daggett et al. (1997) examined the effects of triethyl Pb administration on the  expression
11    on GST isoenzymes and quinone reductase in rat kidney and liver.  Fischer 344 rats were given
12    one IP injection of triethyl Pb chloride (10 mg/kg body weight) and subsequent changes in
13    enzyme expression were measured.  There was a significant increase in GST activity  in kidney;
14    all GST subunits were significantly elevated, the largest increase being a 3.2-fold increase in
15    GST Ybl.  In the liver, injection of triethyl Pb-chloride resulted in decreased GST activity.
16    The largest decrease in subunits was a 40% reduction in GST Yal.  The activity of quinone
17    reductase was elevated 1.5-fold in kidney and 2.7-fold in liver within 14 days after the injection
18    of triethyl Pb chloride.
19
20    Effects of Lead on Renal Heme Enzymes
21          Vij et al. (1998) explored Pb-induced alterations in male rats in the heme synthesizing
22    enzymes, ALAD, and uroporphyrinogen I synthetase, and the effect of ascorbic acid
23    supplementation in reversing these alterations. Lead-treated rats were injected IP with 20 mg/kg
24    of Pb-acetate for 3 consecutive days and sacrificed 4 days later. A separate group of animals
25    were administered 100 mg/kg ascorbic acid PO for 3 days following Pb administration.  Blood
26    Pb concentration was 4.67 ± 1.49 |ig/dL in control rats, 16.59 ± 4.65 |ig/dL in Pb-treated rats,
27    and 7.83 ± 2.03 |ig/dL in the Pb + ascorbic acid treated rats. Lead content of liver and kidney
28    followed the same pattern. Blood ALAD activity was diminished in the Pb-treated rats but was
29    restored in the Pb + ascorbic acid-treated rats. Uroporphyrinogen I synthetase activity followed
30    the same pattern in blood but was not restored by ascorbic acid in liver. Total and nonprotein
      December 2005                           5-172       DRAFT-DO NOT QUOTE OR CITE

-------
 1    sulfhydryl concentrations in blood were depressed by Pb administration and were not restored by
 2    ascorbic acid. However, levels in liver and kidney were restored by ascorbic acid.
 3          ALAD levels following administration of Pb were also investigated by Rodrigues et al.
 4    (1996) and Peixoto et al. (2004).  The study by Rodrigues et al. (1996) examined rats from Pb-
 5    exposed mothers that were maintained after weaning on either 0.5 or 4.0 mM Pb-acetate in
 6    drinking water for 21 days or 6 months. At sacrifice, ALAD activity was measured in kidney,
 7    forebrain, and cerebellum. Both 6-month-old Pb-exposed groups showed an increase in the
 8    kidney-to-body weight ratio, suggesting Pb-induced cell proliferation in the kidney.  Blood Pb
 9    increased from 6.53 to 7.61 |ig/dL in the 21-day-old exposed rats compared to 6-month-old
10    controls.  In the 0.5 mM Pb-treated group, blood Pb was 9.77 |ig/dL in the 21-day-old and
11    41.63  |ig/dL in 6-month-old rats, while in the 4.0 mM group, blood Pb was 44.35 |ig/dL in the
12    21-day-old and 116.91  |ig/dL in the 6-month-old group. ALAD activity was reduced at
13    6 months in the forebrain of the 4.0 mM Pb-treated group, and in the kidneys at 6 months in both
14    the 0.5 mM and 4.0 mM Pb-treated groups. The study by Peixoto et al. (2004)  examined the in
15    vitro sensitivity (ICso) to Pb of ALAD activity of brain, kidneys, and liver from suckling rats
16    aged between 1 and 5, 8 and 13, or 17 and 21 days. The metal concentrations ranged from 0 to
17    50 jiM for Pb-acetate. Rats in the first age group showed the greatest sensitivity in all three
18    organs. Liver was the least sensitive to ALAD inhibition by lead, while brain was the most
19    sensitive.
20
21    Effects of Lead on NaK-A TPase
22          Fox et al. (1991) explored the effect of in vivo Pb exposure on adult rat retinal and kidney
23    NaK-ATPase. Pups, exposed to Pb through the milk of dams consuming 0, 0.02, or 0.2% Pb
24    solutions, had mean blood Pb concentrations of 1.2, 18.8, and 59.4 |ig/dL at weaning,
25    respectively, and 5 to 7 |ig/dL as 90 to 100-day-old adults.  Prior Pb exposure produced
26    significant dose-dependent decreases in isolated retinal NaK-ATPase activity (-11%; -26%),
27    whereas activity in the kidney was unchanged.  In contrast, NaK-ATPase from  both isolated
28    control tissues was inhibited by Pb in vitro. The half-maximal inhibitory dose of Pb for retinal
29    and  renal NaK-ATPase was 5.21  x 10~7 and 1.25 x 10~5 M, respectively.  Retinal and renal NaK-
30    ATPase were 20-fold and 1.1-fold more sensitive to inhibition by Pb than calcium.  The
      December 2005                           5-173        DRAFT-DO NOT QUOTE OR CITE

-------
 1    increased sensitivity of retinal, compared to renal, NaK-ATPase to inhibition following in vivo
 2    or in vitro Pb exposure may be related to their different a subunit composition.
 3          Kramer et al. (1986) had also explored the half-maximal inhibitory dose for Pb-chloride
 4    on renal cortical homogenate NaK-ATPase, and found it to be 7 x ICf5 M. There was a
 5    competitive inhibition with regard to the substrate, ATP. Of several metals tested, Pb was
 6    second only to Hg in potency as a NaK-ATPase inhibitor.
 7          Weiler et al. (1990) studied the effect of Pb on the kinetics of purified (from hog cerebral
 8    cortex) NaK-ATPase and potassium-stimulated p-nitrophenylphosphatase (K-pNPPase), which is
 9    referred to as the E2 configuration of the NaK-ATPase system. ICso for Pb was found to be
10    8.0 x 10~5 M for NaK-ATPase and 5.0 x 10~6 M for K-pNPPase. Inhibition of NaK-ATPase by
11    Pb was found to be noncompetitive with respect to K, but competitive with respect to Na and
12    MgATP.  Inhibition of K-pNPPase by Pb was competitive with respect to K.
13
14    Effects of Lead on Cardiovascular Hormones
15    Effects of Lead on Endothelin
16          Khalil-Manesh et al. (1993a) examined the role of endothelial factors in Pb-induced
17    hypertension.  They found that low Pb administration (0.01%), but not high Pb administration,
18    (0.5%) resulted in increased blood pressure in rats treated for 12 months.  In the low-Pb-treated
19    rats, measurement of plasma endothelins-1 and -3 revealed that endothelin-3 concentration
20    increased significantly after both 3 months (lead, 92.1 ± 9.7 vs. control, 46.7 ± 12.0 pmol/ml;
21    p < 0.001) and  12 months (lead, 105.0 ± 9.3 vs. control, 94.1 ± 5.0 pmol/ml; p < 0.01), while
22    endothelin-1 was unaffected. Plasma and urinary cyclic GMP concentrations, as a reflection of
23    endothelium-derived relaxing factor (EDRF), decreased significantly at 3 months (plasma lead,
24    1.8 ± 0.9 vs. control, 4.2 ± 1.6 pmol/ml; p < 0.001)  and 12 months (plasma Pb 2.2 ± 0.7
25    vs.control, 4.2 ± 0.9 pmol/ml; p < 0.001). High levels of Pb exposure did not result in
26    hypertension, perhaps related to the fact that plasma concentrations of endothelin-1, endothelin-
27    3, and cyclic GMP were unaltered at 3 months, while their concentrations were significantly
28    decreased at 12 months (plasma cyclic GMP at 12 months,  2.2 ± 0.7, lead, vs. 4.2 ± 0.9 pmol/ml,
29    control; p < 0.001).  Thus, the path to development of hypertension in low-Pb rats was thought to
30    be through an increase in the concentration of the vasoconstrictor, endothelin-3, and a decrease
31    in the vasodilator hormone, endothelium-derived relaxing factor or NO.

      December 2005                          5-174       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Novak and Banks (1995) studied the effects of Pb on the actions of endothelin. They
 2    measured renal clearances and mean arterial pressure in rats in which endothelin-1 was infused at
 3    110 ng/kg/min for 30 min. Lead was infused as Pb-acetate throughout the experiment at 0.48,
 4    4.8, and 24 nmoles/min. At the two higher doses, Pb significantly attenuated the endothelin-
 5    induced increase in mean arterial pressure; Pb infused as 0.48 nmoles/min had no effect.
 6    An endothelin-induced decrease in GFR in control rats was completely blocked at the higher
 7    doses of lead. In additional experiments, calcium chloride was infused at 500 nmoles/min for
 8    105 min, and then calcium + Pb (4.8 nmoles/min) were infused for another 105 min.  In these
 9    experiments, there was no Pb-induced inhibition of the mean arterial pressure response to
10    endothelin.  However, the GFR response to the peptide remained blocked. These data illustrate
11    that Pb inhibits the cardiorenal actions of endothelin and that a calcium-related process is
12    involved in the systemic, but not the renal, component of this inhibition.
13
14    Effects of Lead on the Catecholamine System
15          Carmignani et al. (2000) studied the effects of low Pb exposure (60 ppm of Pb-acetate),
16    given for 10 months, on catecholamine and monoaminoxidase (MAO) levels.  Plasma
17    catecholamines were measured by HPLC and MAO  in aorta, liver,  heart, kidney, and brain by a
18    histochemical technique.  Plasma norepinephrine (NE) was increased by 104% and adrenaline by
19    81% with no changes noted in L-DOPA and dopamine levels.  MAO activity was increased in all
20    organs. These workers ascribed the low Pb-induced hypertension in part to raised
21    catecholamines levels.
22          Tsao et al.  (2000) and Chang et al. (2005) measured changes in the P-adrenergic system in
23    Wistar rats during and following Pb exposure.  In Tsao et al. (2000), rats were chronically fed
24    with 0.01, 0.05, 0.1, 0.5, 1.0, and 2.0% Pb-acetate for 2 months. Plasma catecholamine levels
25    were measured by HPLC; cAMP levels  in heart, kidney,  and aorta by radioimmunoassay; and
26    P-adrenergic receptors in heart, kidney,  and aorta membranes by a radio ligand binding assay.
27    Blood Pb increased from 0.05 ± 0.05 |ig/dL in controls to 85.8 ± 4.1 |ig/dL in the 2.0%
28    Pb-treated group.  Plasma NE, but not E, levels increased with increasing Pb dosage.
29    p-Adrenoreceptor density of heart and kidney decreased progressively with increasing Pb
30    dosage, whereas kidney p-adrenoreceptor density increased up to the 0.5% Pb group and then
31    remained constant. Unstimulated cAMP was constant in all tissues, but cAMP stimulated by

      December 2005                          5-175        DRAFT-DO NOT QUOTE OR CITE

-------
 1    isoprotorenol was lowered progressively in aorta and heart and increased in kidney.  Chang et al.
 2    (2005) continued these measurements in rats fed 2% Pb-acetate for 2 months then withdrawn
 3    from Pb for periods of 1, 2, 3, 4, 5, 6, and 7 months. Blood Pb levels, systolic and diastolic
 4    blood pressure levels, and plasma NE were reduced after cessation of Pb exposure. This
 5    occurred in conjunction with an increase in p-adrenoreceptor density in heart and aorta and a
 6    decrease in P-adrenoreceptor density in kidney. (See Table AX5-5.5 for experimental details on
 7    these studies).
 8
 9    Effects ofChelators (Single or Combined) on Lead Mobilization
10    Effects ofDMSA Alone
11          Cory-Slechta (1988) studied the mobilization of Pb by DMSA, following a 3- to 4-month
12    exposure to 50 ppm of Pb-acetate in rats.  These rats received an IP injection of saline or 25 or
13    50 mg/kg ofDMSA once daily for either 1, 2, 3, 4, or 5 days.  Tissue analyses indicated that
14    DMSA mobilized Pb from blood, brain, kidney, and liver with no loss noted from femur.
15         Pappas et al. (1995) reported on Sprague-Dawley rats exposed to 550 or 1100 ppm Pb-
16    acetate for 35  days and treated either with Pb + DMSA or DMSA alone at varying dosage for
17    21 days. Animals showed a dose-related reduction in Pb content of blood, brain, femur, kidney,
18    and liver whether they received DMSA alone or Pb + DMSA.
19          Smith and Flegal (1992) studied the influence ofDMSA on the mobilization and
20    redistribution  of Pb in skeletal and soft tissue compartments of low-Pb-exposed female rats,
21    using stable Pb isotope tracer techniques.  Rats  reared on a low-Pb diet received 206Pb-enriched
22    drinking water for 1.5 days and then were chelated with  a single IP injection of 0.11 mmol/kg
23    dose ofDMSA.  Blood, kidney, brain, tibia, urine, and feces were collected 24 h after chelation
24    and analyzed for Pb concentrations and for Pb isotope compositions. DMSA chelation
25    significantly increased the diuresis of labile soft tissue Pb but not skeletal Pb. DMSA also
26    appeared to cause the redistribution and input of a comparable amount of Pb to the skeleton and
27    smaller relative amounts of Pb to the soft tissues of the chelated animals.
28         Varnai  et al. (2001) determined whether ongoing  Pb exposure influenced the mobilization
29    of Pb in suckling rats.  Six-day-old Wistar rats were given Pb-acetate in a dose of 2 mg/kg/day
30    for 8 consecutive days.  A treated group received a daily dose of 0.5 mmol/kg ofDMSA PO six
31    times on days  1 to 3 and 6 to 8. DMSA efficiently reduced Pb concentration in carcass, liver,

      December 2005                           5-176        DRAFT-DO NOT QUOTE OR CITE

-------
 1   kidneys, and brain by approximately 50% versus with untreated controls.  The results indicate
 2   that DMSA is an efficient oral chelator, even when challenged with ongoing Pb exposure.
 3
 4   Effects of Combined Chelators
 5         Flora et al. (1994) compared the combined use of CaNa2EDTA with DMSA on the
 6   distribution of Pb and Pb-related biochemical effects with the influence of each chelator used
 7   alone.  Wistar rats were given 1000 ppm Pb as Pb-acetate in drinking water for 4 months. They
 8   were then treated for 5 days with either saline, DMSA, 25 mg/kg PO twice daily; CaNa2EDTA,
 9   75 mg/kg once daily; or DMSA, 25 mg/kg twice daily, all followed by a single daily IP injection
10   of 75 mg/kg of CaNa2EDTA.  Blood ALAD was reduced from 6.54 ±0.18 nmol/min/ml in
11   controls to  0.84 ± 0.10 in Pb-treated animals, with restoration to 3.03  ± 0.29 after combined
12   treatment.  Lead content in blood, liver, kidney, brain, and femur followed the same pattern:
13   controls had 2.11 ± 0.23 |ig/dL; Pb-treated, 46.0 ± 4.1 |ig/dL; combined chelator-treated, 12.8 ±
14   0.3 |ig/dL.  Treatment with either DMSA or CaNa2EDTA alone produced intermediate results.
15   Tandon et al. (1994) reported similar results.
16         Jones et al. (1994) compared the effects of DMSA, CaNa2EDTA, ZnNa2EDTA, and
17   ZnNasDTPA on Pb mobilization in mice.  Mice were given 10 IP injections of Pb-acetate,
18   5.0 mg/kg per injection. Three days after the final Pb injection, mice  received one of the
19   chelators.  Injections were given at a dose of 1 mmol/kg/day IP for either 4 days or 8 days.
20   At 8 days, DMSA was the most effective chelator in removing Pb from kidney and bone.
21   CaNa2EDTA was more effective in removing brain lead.  When animals were loaded with
22   100 mg of Pb per kg body weight, DMSA remained more effective in removing Pb from kidney
23   and bone while CaNa2EDTA was more effective in brain.
24         Kostial et al. (1999) evaluated the efficacy of three chelating agents, administered either
25   as monotherapy or as combined treatments, in suckling rats. Lead acetate (5 mg Pb/kg IP) was
26   administered to 7-day-old rat pups on experimental day 1, and chelating agents was administered
27   on experimental days 2 and 3. The pups were divided into untreated control, EDTA-treated,
28   meso-DMSA-treated, racemic DMSA-treated, EDTA plus meso-DMSA-treated, and EDTA +
29   plus racemic DMSA-treated.  Rats were killed on experimental day 5  and tissue analyses were
30   done for lead, zinc, and copper. Treatment with EDTA did not affect tissue lead, but it reduced
31   zinc in the  carcass and liver.  Meso-DMSA reduced Pb in the kidneys and brain and did not

     December 2005                           5-177       DRAFT-DO NOT QUOTE OR CITE

-------
 1    affect organ essential elements. Racemic DMSA most efficiently reduced Pb concentrations in
 2    the carcass, kidneys, and brain, but it also reduced zinc and copper in the liver and zinc in the
 3    kidneys. Combined treatments with EDTA did not improve the efficiency of either DMSA
 4    isoform in decreasing tissue lead, but they did reduce tissue zinc concentrations. The results
 5    suggest that meso-DMSA may be the treatment of choice in acute Pb poisoning in infants,
 6    reducing Pb without affecting trace elements.
 7          Malvezzi et al. (2001) evaluated the effects of DMSA, L-arginine (a precursor of NO),
 8    and the association  of L-arginine and DMSA on tissue Pb mobilization and blood pressure levels
 9    in Pb-intoxicated rats.  Tissue Pb levels and blood pressure evolution were evaluated in rats
10    exposed to Pb (750 ppm in drinking water for 70 days), Pb + water for 30 more days , Pb +
11    DMSA (50 mg/kg day, PO), L-arginine (0.6% in drinking water), the combination of L-arginine
12    + DMSA for 30 more days, and their respective matching controls. Lead exposure increased Pb
13    levels in the blood,  liver, femur, kidney, and aorta. Lead levels in tissue decreased after
14    cessation of Pb administration, except in the aorta. Blood Pb decreased from 67.8 |ig/dL to
15    11.2 |ig/dL in those subsequently treated with water, to 13.8 |ig/dL in animals treated with
16    Pb + DMSA, to 11.6 |ig/dL in animals treated with Pb + L-arginine, and to 6.1 |ig/dL in animals
17    treated with Pb + L-arginine + DMSA. Lead mobilization from the aorta was only effective with
18    the L-arginine/DMSA treatment. Lead administration increased blood pressure starting from the
19    week 5, while L-arginine and DMSA treatments and the combination of L-arginine + DMSA
20    decreased blood pressure levels of intoxicated rats; but these levels did not reach those of
21    nonintoxicated rats. Treatment with L-arginine + DMSA was more effective than individual
22    treatments in mobilizing Pb from tissues and in reducing the blood pressure of intoxicated rats.
23    This paper lacks measurements of NO, which would have allowed the reader to more properly
24    judge the mechanism of the effects of L-arginine administration. Furthermore, the dose of Pb
25    was higher than in earlier studies that showed that DMSA was effective in lowering blood
26    pressure.  These and other studies are summarized in Tables AX5-7.3 and AX5-7.4.
27
28    Effects of Other Metals on Lead Distribution
29    Lead and Calcium
30          Fullmer (1992) published a review of intestinal interactions of Pb and calcium. High
31    affinity Pb binding  to intracellular calcium receptors and transport proteins, as well as the

      December 2005                           5-178        DRAFT-DO NOT QUOTE OR CITE

-------
 1    involvement of Pb in calcium-activated and calcium-regulating processes, are thought to provide
 2    a partial molecular basis for the cellular and systemic effects of lead.
 3          Maidonado-Vega et al. (1996) examined the intestinal absorption of Pb and bone
 4    mobilization during lactation. All experiments were started with 3-week-old female Wistar rats.
 5    Rats were impregnated at 16 weeks and were fed a 100 ppm solution of Pb-acetate for 158 or
 6    144 days (mid-lactation or before lactation).  Rats were also exposed for only 14 days, from 144
 7    to 158 days (i.e., only during lactation). Nonpregnant rats from the same litter were exposed to
 8    Pb for periods equivalent to each of these groups.  In the nonpregnant rats, blood Pb increased to
 9    27.3 |ig/dL from 5.2 |ig/dL in controls. Similarly, kidney Pb increased to 13.2 nmol/g from
10    0.5 nmol/g, and bone Pb increased to 88.9 nmol/g from 0.9 nmol/g. ALAD activity decreased to
11    410 nmol/h/ml from 1004 nmol/h/ml.  Compared to nonpregnant rats, there was a moderate
12    increase in blood Pb in the lactating animals whether the Pb was given to mid-lactation or up to
13    the period before lactation.  Similarly, when Pb was administered only during lactation, there
14    was a much higher increase in blood Pb in the pregnant rats than  in the nonpregnant rats. Bone
15    Pb concentration increased when Pb was given only during lactation, whereas bone Pb decreased
16    (compared to Pb-treated nonpregnant rats) when the Pb was given either before lactation or
17    before and during lactation.  The authors considered that resorption of Pb from bone was the
18    main additional source of Pb during lactation.  The data indicate that Pb stored in bone as a result
19    of prior maternal exposure should be considered as a major source of self intoxication and of Pb
20    in milk available to suckling pups.
21
22    Lead and Cadmium
23          Skoczynska et al. (1994) compared the effects of the combined exposure to Pb and
24    cadmium to each metal singly on tissue composition of trace metals. Experiments were
25    performed on 5- to 6-week-old male Buffalo rats given Pb-acetate (70 mg lead/kg body weight
26    twice a week) and cadmium chlorate (20 mg Cd/kg body weight once a week) intragastrically for
27    7 weeks either singly or in combination. Blood Pb in the control group was  5.1 |ig/dL,  compared
28    to 29.6 |ig/dL in the Pb-treated group.  In contrast, the Pb + cadmium group  showed a blood Pb
29    of 37.4 |ig/dL. After combined exposure to Pb and cadmium, the level  of these metals in the
30    liver and kidney was lower than after the  single administration of Pb or cadmium. Exposure of
31    the rats to cadmium resulted in an increase of kidney zinc and copper and liver zinc

      December 2005                           5-179       DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentrations; combined exposure to Pb + cadmium did not produce more extensive changes in
 2    tissue zinc and copper concentrations.
 O
 4    Lead and Selenium
 5          Othman and Missiry (1998) examined the effect of selenium against Pb toxicity in male
 6    rats. Male albino rats were given a single dose of Pb-acetate (100 jimol/kg body weight) and
 7    sacrificed 3 or 24 h later.  Another group of animals was pretreated with sodium selenite
 8    (10 jimol/kg body weight) 2 h before receiving Pb-acetate and sacrificed 24 h later. Selenium is
 9    well known as an antioxidant and cofactor for GSHPx. In this experiment, GSH content,
10    GSHPx, SOD activities, and the products of lipid peroxidation (i.e., TEARS) were determined.
11    It was  found that lipid peroxidation was prevented and the reduction in GSH caused by Pb in
12    liver and kidney was diminished by selenium. Lead-induced diminution in SOD activity and
13    GSHPx activity was also returned to normal by selenium.
14          Tandon et al. (1992) studied the effect of selenium supplementation during chelation of
15    Pb with CaNa2EDTA. Rats were given Pb-acetate 10 mg/kg/day by gastric gavage for 6 weeks.
16    This was followed by a 5-day treatment course of CaNa2EDTA, 0.3 mmol/kg IP or of
17    CaNa2EDTA + sodium selenite, 0.5 mg/kg PO. Selenium had marginal effects on Pb removal
18    by CaNa2EDTA in blood, liver, and kidney and similar effects on ALAD activity.
19
20    Lead and Zinc
21          Flora et al. (1989) examined the role of thiamine, zinc, or their combination in the
22    prevention or therapy of Pb intoxication. Albino rats received the following treatments daily
23    through gastric gavage for 6 days each week over a six-week period, 10 mg/kg of Pb as Pb-
24    acetate; or the same dose of Pb-acetate + thiamine (25 mg/kg) zinc sulfate (25 mg/kg) or
25    Pb + thiamine and zinc. Rats that had been exposed to Pb only were additionally divided into
26    four groups treated by gastric gavage daily for 6 days  as follows: group I, water  only; group II,
27    thiamine only; group III, zinc only; and group IV, combined zinc + thiamine. The activities  of
28    blood ALAD, blood ZPP, blood lead, and urine ALA were determined.  Blood Pb concentrations
29    increased from 6.2 to 120.9 |ig/dL, contrasting normal controls with Pb-treated animals. There
30    was a slight reduction in blood Pb in animals treated with either thiamine or zinc and a greater
31    reduction in animals treated with thiamine  + zinc.  In the post-Pb-exposure treatment group,

      December 2005                           5-180       DRAFT-DO NOT QUOTE OR CITE

-------
 1   thiamine + zinc was also the most effective treatment. Liver and kidney Pb levels followed the
 2   same course but brain Pb was not reduced by treatment.  Blood ALAD activity was decreased
 3   from a normal level of 7.63 (imol ALA/min/L to 0.69 in Pb-treated animals and restored to 7.52
 4   in Pb + thiamine + zinc-treated rats. ZPP was increased from 1.78 |ig/g hemoglobin to 4.22 in
 5   Pb-treated animals and reduced to 2.50 in Pb + thiamine + zinc-treated animals.  Urine ALA was
 6   increased from 0.07 to 0.24 mg/dL in Pb-treated animals and decreased to 0.17 in Pb + thiamine
 7   + zinc-treated rats. Prevention was more effective than post-Pb-exposure treatment. This was
 8   thought to be due mainly to the decrease in the absorption of Pb in the GI tract in the presence of
 9   thiamine and/or zinc.
10          Flora et al. (1994) explored the dose-dependent effects of zinc supplementation during
11   chelation of Pb in rats. The chelator employed was CaNa2EDTA, whose toxic effects  are known
12   to be mainly due to the depletion of endogenous zinc and, possibly, copper and manganese.
13   In this experiment, male Wistar rats were started on exposure to Pb-acetate, 10 mg/kg,
14   administered through gastric gavage once daily for 56 days. Twenty-four hours later, the
15   Pb-exposed animals were treated daily for 5 days as indicated: group I, saline  ; group II,
16   CaNa2EDTA 0.3 mmol/kg, IP, once daily for 5  days; group III, CaNa2EDTA + zinc sulfate,
17   10 mg/kg, PO once daily for 5 days; and group IV, CaNa2EDTA + zinc sulfate, 50 mg/kg,
18   PO once daily for 5 days. Blood ALAD decreased from 6.30 to 1.44 nmol/min/mL erythrocyte
19   in Pb-exposed animals, with no change after CaNa2EDTA treatment and partial restoration after
20   the CaNa2EDTA + zinc, 10 mg/kg treatment. There was no improvement following zinc,
21   50 mg/kg.  Lead concentration in blood increased from 4.6 |ig/dL to 43.0 |ig/dL in Pb  exposed
22   animals, decreasing to 22.5 |ig/dL in CaNa2EDTA-treated animals and decreasing further to
23   16.5 |ig/dL in CaNa2EDTA plus zinc-treated animals. Zinc at 50 mg/kg led to an increase in
24   blood Pb to 56.1 |ig/dL.  Changes in the liver follow the same pattern, while in the kidney, zinc
25   increased the Pb levels further, and in the femur, zinc had no influence on Pb content.  Blood
26   zinc decreased from 6.1 to 5.7 |ig/ml in Pb-exposed rats and further to 5.0 |ig/ml in
27   CaNa2EDTA-treated animals.  There was an increase to levels of 6.6 |ig/ml on the 10 mg/kg
28   supplement of zinc and a further increase to 8.1 |ig/ml on the 50 mg/kg zinc supplement.
29
     December 2005                           5-181       DRAFT-DO NOT QUOTE OR CITE

-------
 1   Lead and Iron
 2          Hashmi et al. (1989) examined the influence of dietary iron deficiency, Pb exposure, or
 3   the combination of the two on the accumulation of Pb in vital organs of rats. Animals fed an iron
 4   deficient diet for 2 weeks were also subjected to orbital plexus puncturing twice a week to allow
 5   a Hb levels to decrease to 7 to 8 g/dL.  Animals were thereafter treated for the next 6 weeks with
 6   iron deficient diets or iron-deficient diets + 0.1% Pb-acetate in drinking water.  At the end of
 7   3 and 6 weeks, animals from each group were sacrificed.  Feeding of an iron-deficient diet
 8   during Pb exposure enhanced the accumulation of Pb in soft tissues and flat bones.  For example,
 9   liver Pb content  was 0.75 |ig/g in control animals, 8.43  in Pb treated animals, and 12.93 in iron-
10   deficient and Pb-treated animals. The sequence of events was similar in kidney, spleen, and
11   femur except that the Pb content in femur was reduced in the iron deficient  and Pb-treated group.
12          Singh et al. (1991) conducted a study to ascertain the role of iron deficiency during
13   pregnancy in inducing fetal nephrotoxicity in mothers exposed to lead. Rats were fed either a
14   normal iron diet or an iron free synthetic diet for 15 days, followed by a diet containing half of
15   the daily required iron (47 mg/100 g ferrous ammonium sulfate) for a further 15 days. Female
16   animals were mated with healthy adult males. Lead doses of 250,  500, 1000, and 2000 ppm
17   were given in drinking water during pregnancy and lactation. Fetuses were removed by
18   Caesarean section on the 21st day. Maternal blood Pb levels in rats on an iron deficient diet
19   were higher than those in rats on a normal iron diet at all levels of Pb dosing. Similarly,
20   placental Pb levels were higher in animals on an iron-deficient diet as compared to a normal diet.
21   Lead content in the fetuses were higher on the iron-deficient  diet.  Lead administration resulted
22   in dose-dependent hydropic degeneration of renal proximal tubular cells in the fetuses.  At a dose
23   of 2000 ppm Pb  with iron deficiency, more Pb accumulated in maternal blood, placenta, and
24   fetuses and maximum pathological changes were seen in the  fetal kidney as compared to other
25   doses.
26
27   Lead and Aluminum
28          Shakoor et al. (2000) reported beneficial effects of aluminum on the  progression of Pb-
29   induced nephropathy in rats. Male albino rats were treated with water only  or Pb-acetate
30   (125 mg/kg) and/or aluminum chloride (50 mg/kg or 100 mg/kg) for a period of 90 days.
31   Aluminum was found to prevent the Pb-induced increase in relative kidney  weight in a dose-

     December 2005                            5-182       DRAFT-DO NOT QUOTE OR CITE

-------
 1    dependent manner. Aluminum also prevented Pb-induced increases in plasma creatinine levels
 2    of Pb- treated animals.  The net deposition of Pb in kidneys was lower in animals that were given
 3    both Pb-acetate and aluminum chloride simultaneously. By day 90, plasma creatinine was
 4    1.26 mg/dL in control animals, 1.88 mg/dL in Pb-treated animals, and 1.34 and  1.44 mg/dL in Pb
 5    and aluminum-treated animals. Similarly, kidney Pb increased from 5.4 |ig/g in control animals
 6    to 220.0 |ig/g in Pb-treated animals and decreased to 138.5 and 98.9 |ig/g in Pb  and aluminum
 7    treated animals.  These and other studies are  summarized in Table AX5-7-5.
 8
 9    5.7.4.4  Effect of Age  on Lead Toxicity
10          Han et al. (1997) examined the hypothesis that the high rate of bone remodeling during
11    childhood and the consequent high calcium and Pb turnover would  result in a substantial
12    reduction in bone Pb  stores, so that much of the Pb incorporated in bone during  childhood does
13    not persist into adulthood. They treated female Sprague-Dawley rats with 250 ppm of Pb in
14    drinking water for 5 weeks beginning at 5, 10, or 15 weeks of age.  Organ harvesting occurred
15    4 weeks after the end of Pb exposure for all groups, as well as 8 and 20 weeks after cessation of
16    Pb ingestion in the rats  exposed beginning at 5  weeks of age.  Organs examined were brain,
17    kidney, liver, femur, and spinal column bone. Blood and organ Pb concentrations were
18    significantly higher in the rats exposed beginning at 5 weeks of age than in those exposed
19    beginning at 10 or 15 weeks of age. The results of this experiment rejected the hypothesis and
20    suggested instead that a younger age at Pb exposure is associated with greater Pb retention and
21    toxicity, even in the absence of continued Pb exposure.
22          Garcia and Corredor (2004) examined biochemical  changes in  the kidneys after perinatal
23    intoxication with Pb and/or cadmium. Lead acetate (300 ppm) and/or cadmium acetate (10 ppm)
24    were administered in drinking water to pregnant Wistar rats from day 1 of pregnancy to
25    parturition (day 0) or until weaning (day 21). The following kidney enzyme activities were
26    determined: alkaline and acid phosphatases, Mg-ATPase, and NaK-ATPase.  Blood Pb was
27    measured in control pups as well as in pups exposed to lead at parturition and at weaning.
28    Control pups showed 1.43 |ig/dL of blood Pb compared to 31.5 |ig/dL at day 0 and 22.8 |ig/dL
29    at day 21 in pups exposed to lead. In those rats receiving both cadmium and Pb, the blood Pb
30    concentration was 23.2 |ig/dL at day 0 and 13.2 |ig/dL at day 21. Lead caused a significant
31    inhibition of kidney alkaline phosphatase and kidney acid phosphatase. At parturition, Pb

      December 2005                           5-183        DRAFT-DO NOT QUOTE OR CITE

-------
 1    intoxication produced a strong inhibition of NaK-ATPase (-80%) as well as of Mg-ATPase
 2    activities (-24%); whereas, when Pb was given in combination with cadmium, these inhibitory
 3    effects were attenuated. At weaning, Pb continued to produce a significant inhibition of Mg-
 4    ATPase but had no effect on NaK-ATPase. Thus, simultaneous perinatal administration of both
 5    Pb and  cadmium seemed to protect against the toxicity produced by Pb separately.
 6         Cory-Slechta (1990a,b) published two articles on the effects of old age on the disposition
 7    of lead.  In the first study (1990a) male F344  rats, at the ages of 8 months (adult) and 16 months
 8    (old) were exposed to concentrations of 0, 250, or 500 ppm Pb-acetate in drinking water for
 9    7 months. At these Pb doses, prior studies had indicated that blood Pb levels ranged from 60 to
10    90 |ig/dL. Blood lead, ZPP, and urinary ALA levels were determined after both 3 and 7 months
11    of exposure.  Organ weights, tissue Pb concentrations, and urinary excretion of lead, calcium,
12    copper, and zinc were examined after 7 months of exposure.  Tissue Pb distribution was
13    markedly altered in old rats: in bone and kidney, Pb levels were reduced while liver Pb was
14    substantially increased. Blood Pb levels in adult and old rats were comparable at both
15    measurement intervals, as was urinary Pb  excretion at 7 months. Lead-induced elevation of ZPP
16    exhibited differential changes between 3 and  7 months; values in adults declined while levels in
17    old rats increased or remained unchanged. In the adult group, Pb exposure increased calcium
18    excretion primarily at the 500 ppm  exposure level. In contrast, Pb exposure decreased urinary
19    calcium excretion in old animals at the higher exposure level. No effects of either age or Pb
20    exposure were detected in the comparison of  adult versus old urinary excretion of zinc or copper.
21         In the second study, Cory-Slechta (1990b), young (21 days old), adult (8 months old), and
22    (16 months old) rats exposed to 0, 2, or 10 mg of Pb-acetate/kg per day for a period of
23    9.5 months were evaluated. Differences in the tissue distribution of Pb with age included lower
24    bone levels, but increased concentrations in brain, liver, and kidney. Differences in blood Pb
25    levels over the course of exposure were not remarkable. Thus, these effects did not appear to
26    reflect an enhanced Pb absorption from the GI tract with age. Instead, the bone changes may
27    reflect enhanced bone resorption with a concurrent decline in bone apposition with age,
28    combined with altered patterns of urinary Pb  excretion over time, i.e.,  elevated urinary Pb at 3
29    and 6 months, but comparable Pb excretion at 9.5 months, as compared to young and adult rats.
30
      December 2005                           5-184       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.7.5  Summary

 2          Highlights of the previous 1986 Pb AQCD and of studies done between 1986 and the

 3    present are outlined in this section.

 4          1986 Document

 5       •  In animal studies, nuclear inclusion bodies were found in proximal tubules, identified as
 6          27 kDa or 32 kDa proteins in combination with lead.  Subsequently, a 63 kDa Pb-binding
 7          cytosolic protein was described in kidney.

 8       •  Swollen mitochondria, with diminished mitochondrial function, were found in the
 9          proximal tubules.

10       •  Renal ALAD was the same in Pb-treated animals as in controls when GSH was present,
11          but was reduced when GSH was absent.

12       Newer studies

13       •  Hyperfiltration, when compared to age- and sex-matched normal controls, was found in
14          adults who had suffered from childhood Pb poisoning, in young occupationally exposed
15          Pb workers in Korea, and in both low-Pb-treated rats and high-Pb-treated rats up to
16          3 months of exposure. This is paralleled in animal experiments by an increase in
17          kidney weight.

18       •  Various new urinary markers for Pb toxicity have been described. These include NAG,
19          p2-microglobulin, al-microglobulin, retinol binding protein, GST, lysozyme, y-glutamyl
20          transferase, alanine aminopeptidase, prostanoids, and  brush border antigens. The
21          literature on these markers is voluminous, but, on review, only GST and al-
22          microglobulin  seemed to be appropriate urinary  markers. NAG, which has been most
23          extensively investigated, appears in detailed-animal studies to be overly sensitive,
24          increasing in low-Pb-treated animals, despite an absence of pathological changes on
25          ultrastructural study.  p2-Microglobulin, and possibly retinol binding protein, which are
26          low-molecular weight proteins reabsorbed by the proximal tubule, appeared to be
27          elevated only with high levels of blood Pb (>80  jig/dL).

28       •  Animal studies have implicated free radicals in the pathogenesis of Pb-induced
29          hypertension and renal disease. A sequence of free radicals can be demonstrated in
30          Pb-induced disease, as evidenced by an increase in superoxide radicals, hydroxyl
31          radicals, hydrogen peroxide, and peroxynitrite, together with a diminution in GSH in
32          liver, brain, and aorta. Nitric oxide is most commonly decreased (by free radicals) as is
33          urinary cyclic GMP.  Aortic guanylate cyclase is decreased.  The enzyme responsible for
34          an increase in the production of free radicals, NAD(P)H oxidase, is increased by Pb,
35          whereas eNOS  and iNOS, the enzymes involved in the production of nitric oxide, are also
36          increased, attesting to the importance of free radical destruction of nitric oxide.
37          Antioxidants reverse these changes and diminish blood pressure.
      December 2005                           5-185        DRAFT-DO NOT QUOTE OR CITE

-------
 1       •  Norepinephrine and epinephrine are increased by Pb administration, whereas
 2          p-adrenoreceptor density of heart and kidney are decreased.  In a second study,
 3          norepinephrine, but not epinephrine, was increased by Pb.
 4       •  Various antioxidants have been used in conjunction with chelators, to both remove Pb
 5          from tissue and to diminish free radicals. Taurine, lipoic acid, arginine, ascorbic acid,
 6          vitamin E, thiamine, tempol, and lazaroids have been used in conjunction with DMSA,
 7          all improving free radical diminution.
 8       •  Metal combinations have also been employed to reduce tissue Pb and/or affect free
 9          radicals.  Cadmium increases Pb in blood when both are given, but diminishes Pb in liver
10          and kidney.  Selenium, an antioxidant, improves both parameters, as does thiamine or
11          L-lysine plus zinc.  Iron deficiency increases intestinal  absorption of Pb and the Pb
12          content of soft tissues and bone. Aluminum decreases  kidney Pb content and serum
13          creatinine in Pb-intoxicated animals.
14       •  Age also has an effect on Pb retention.  There is higher Pb retention at a very young age
15          and lower bone and kidney Pb at old age, attributed in part to increased  bone resorption
16          and decreased bone accretion.
17
18
19    5.8   EFFECTS ON BONE AND TEETH
20    5.8.1   Biology of Bone and  Bone Cells
21          By weight, bone is composed of 28% collagen fibers (predominantly type I collagen) and
22    5% noncollagenous proteins (osteocalcin, osteonectin, and other proteoglycans), with crystals of
23    hydroxyapatite [Caio(PO4)e(OH)2] making up the remaining 67%. In addition to providing
24    mechanical support for the body and protection of vital organs, the skeletal system also functions
25    in a metabolic capacity. Historically, bones have been classified as either long  or flat based on
26    their appearance, with long bones including limb bones, e.g., the femur and humerus, and flat
27    bones including the bones of the skull, sternum, pelvis, and scapula. Long and  flat bones
28    originate by distinct methods of formation, endochondral and intramembranous, respectively,
29    with long bones eventually using both processes. In endochondral bone formation,  a
30    mineralized, cartilaginous matrix precedes the transition into true bone, while in
31    intramembranous formation, the bone forming cells create bone directly without the cartilaginous
32    template.
33          Bone cells responsible for producing the bone matrix of collagen and ground substance
34    are called osteoblasts. Several signaling factors including growth factors and hormones

      December 2005                            5-186       DRAFT-DO NOT QUOTE OR CITE

-------
 1    influence pre-osteoblastic cells to differentiate into mature osteoblasts and subsequently
 2    synthesize and mineralize the extracellular matrix to form mature bone. It is during the process
 3    of bone mineralization that the Pb ion (Pb2+) can become incorporated by substituting for the
 4    calcium ion (Ca2+). The bone cells responsible for bone resorption are the osteoclasts.
 5    Osteoclasts, which are large and multicellular (4 to 20 cells), dissolve bone matrix and
 6    hydroxyapatite by synthesizing and releasing lysosomal enzymes and acidifying the extracellular
 7    surroundings. It is during the process of dissolving bone, or demineralization, that Pb stored in
 8    bone can be released locally and into the general system.
 9          Bone cell function may be compromised both directly and indirectly by exposure to Pb.
10    Regulation of bone cells occurs by numerous local and systemic  factors, including growth
11    hormone (GH), epidermal growth factor (EOF), transforming growth factor-beta I(TGF-PI), and
12    parathyroid hormone-related protein (PTHrP).  As discussed further below in this section, the
13    presence of lead can potentially interfere with each of these factors. The bones of the skeleton
14    serve as the primary reservoir for calcium and phosphate in the body and help to maintain
15    homeostasis of these ions in the serum through  bone turnover or  remodeling. Vitamin D
16    [1,25-(OH2)D3] maintains the normal range of calcium in the serum by increasing the efficiency
17    of calcium absorption in the intestines and facilitating  differentiation of stem cells into
18    osteoclasts, which break down bone and mobilize calcium (and lead) stores. Parathyroid
19    hormone (PTH), in turn, regulates the production of vitamin D in the kidney. Lead has been
20    shown to interfere with the action of both of these hormones.  Other substances influenced by
21    lead and discussed in this section are alkaline phosphatase, an enzyme necessary for
22    mineralization of bones and  teeth, and osteocalcin, a noncollagenous protein whose  spatial and
23    temporal pattern of expression suggests a role in bone mineralization. Both substances are also
24    markers for osteoblast activity and, by default, bone formation. Alkaline phosphatase is a
25    potential carrier of ionic calcium and is capable of hydrolyzing inhibitors of mineral deposition
26    such as pyrophosphates.
27
28    5.8.2  Summary of Information Presented in the 1986 Lead AQCD
29          Lead has been  shown to become localized and accumulate in bones and teeth, with
30    accumulation beginning as early as fetal development.  Lead administered to rats as a single dose
31    results in blood lead concentrations that are initially elevated, but rapidly fall as Pb is transferred

      December 2005                            5-187       DRAFT-DO NOT QUOTE OR CITE

-------
 1    to bone or excreted.  The dose of Pb administered does not apparently affect distribution to the
 2    various body compartments; however, the rate-limiting step in the clearance of Pb from rats and
 3    mice involves absorption into/clearance from the skeletal system.  The loss of Pb from various
 4    organs and tissues follows first-order kinetics, except from bone. More absorbed Pb is retained
 5    by young animals compared with adult animals, leading to higher tissue levels. Moreover, once
 6    Pb is incorporated into the young animal's body, the long-term rate of retention is greater than
 7    that of adults. In Pb-exposed animals, Pb is distributed subcellularly, preferentially to the
 8    nucleus and mitochondrial fractions.
 9         During lactation in mice, a redistribution of tissue Pb occurs (mobilization), resulting in
10    the transfer of Pb and calcium from mother to pups via the milk, and subsequent overall loss of
11    Pb in the mothers. Lead transfer to suckling rats via mother's milk has been reported to be
12    approximately 3% of the maternal body burden or more, if Pb exposure continues during
13    lactation.  Eight days after a single injection of Pb, the content of Pb in rabbit's milk was 8-fold
14    higher than the maternal blood level,  suggesting Pb transfer can occur against a concentration
15    gradient.  Transplacental transfer of Pb from mother to fetus also occurs in various animals.
16         In rats, a significant reduction of calcium in the diet leads to enhanced uptake of lead into
17    the bones and other tissues.  In general, an enhanced uptake of Pb into tissues is also seen in rats
18    fed diets deficient in iron, zinc, copper, or phosphorus, and in the presence of low or excess
19    vitamin D.
20
21    5.8.3  Bone Growth in Lead-Exposed Animals
22         Lead is readily taken up and stored in the bone of experimental animals, where it can
23    potentially manifest toxic effects that result in stunted skeletal growth. In experiments reported
24    since the 1986 Pb AQCD, Hac and Krechniak (1996) determined uptake and retention of Pb in
25    bone from rats exposed to plain water or water containing Pb-acetate (41.7 to 166.6 mg/L) for 12
26    to 16 weeks. After 4 weeks, the skeletal Pb  in animals receiving the lowest dose was almost
27    5 times higher than control animals (5.9 versus 1.2 jig Pb/g bone, respectively). Lead levels in
28    bones from animals receiving 83.3 mg/L and 166.6 mg/L were dose-dependently higher at 11.7
29    and 17.0 jig Pb/g bone, respectively, after 4  weeks of exposure. All bone Pb  levels were
30    maintained essentially in a steady state until the completion of exposure, when all animals were
31    placed on control water.  Approximately 64% of Pb remained in the bones of rats in the

      December 2005                           5-188       DRAFT-DO NOT QUOTE OR CITE

-------
 1    83.3 mg/L exposed group at 64 days postexposure. Similarly, airborne Pb can be inhaled and
 2    subsequently incorporated into bone. Grobler and co-workers (1991) exposed 6-week-old rats to
 3    either "clean air" (0.05 jig Pb/m3) or air containing 77 jig Pb/m3 and found significant
 4    differences in the amount of Pb incorporated into the alveolar bones of the animals. After
 5    70 days, a mean of only 0.2 jig Pb/g of bone dry mass was found in bone from control animals,
 6    while 16.9 jig Pb/g was present in bone from the 77 jig Pb/m3 exposure group. Exposure to air
 7    containing 249 jig  Pb/m3 for 28 days or 1,546 jig Pb/m3 for 50 days, resulted in mean values of
 8    15.9 and 158 jig Pb/g dry weight of Pb incorporation into the bone, respectively, highlighting the
 9    fact that dose and length of exposure are determinates of amount of Pb  contained in the bones of
10    these animals.  The uptake of Pb by bone has the potential for immediate toxic effects on the
11    cellular processes occurring during bone growth, development, and maintenance, with the
12    additional potential for delayed toxicity from release of stored Pb during periods of normal or
13    accelerated bone remodeling.
14         Numerous studies have examined growth suppression associated with developmental Pb
15    exposure. Hamilton and O'Flaherty (1994) examined the effects of Pb  on growth in female rats,
16    and subsequently, on growth and skeletal development in their offspring. Administration of
17    drinking water  containing either 250 or  1,000 ppm lead to weaning female rats for 49 days
18    produced no alteration in growth rate in these future dams.  The rats were then bred, with Pb
19    exposure continuing through parturition and lactation. Lead did not affect gestation time nor
20    Day 1 suckling body weight, however, pup body weight and tail length were subsequently
21    decreased in both exposure groups. A 10% increase in tibial growth plate width and disruption
22    of chondrocyte organization were observed in offspring from the high exposure group.
23         In male rats exposed to 100 ppm Pb in drinking water and a low calcium diet for up to one
24    year, bone density  was significantly decreased after 12 months, while rats exposed to 5,000 ppm
25    Pb had significantly decreased bone density after 3 months (Gruber et al., 1997). Pb content of
26    femurs was  significantly elevated over the content of control rats at all time points (1, 3, 6, 9,  12
27    months).  Trabecular bone from the low dose animals was significantly decreased from 3 months
28    forward.  Young female rats exposed to 17 mg of Pb-acetate per kg of feed for 50 days showed
29    no differences in the length of the femurs, but the mean length of the 5th lumbar vertebra was
30    significantly decreased (Gonzalez-Riola et al., 1997; Escribano et al., 1997). The mean length of
31    the femur growth plate cartilage was also significantly decreased in Pb-exposed animals.

      December 2005                           5-189      DRAFT-DO NOT QUOTE OR CITE

-------
 1          In a dose-response study, Ronis et al. (1998a, 1998b) exposed pregnant rats to Pb-acetate
 2    in drinking water (0.05% up to 0.45% w/v) beginning at gestation Day 5 and continuing through
 3    weaning of offspring at Day 21. Early bone growth was significantly depressed in a dose-
 4    dependent fashion in pups of all Pb-exposed groups, with growth suppression in male offspring
 5    considerably greater than in females. Significant decreases in plasma insulin-like growth factor
 6    and plasma sex steroids and increased pituitary growth hormone were also observed.  This is
 7    somewhat in contrast to the findings of Camoratto and coworkers (1993), who reported low
 8    exposure to 0.2% Pb nitrate (125 ppm Pb) did not significantly affect growth, though males
 9    weighed significantly less than females.  Between age 57 and 85 days Ronis et al. (1998b) noted
10    that growth rates were similar in control  and Pb-exposed pups, suggesting exposure at critical
11    growth periods such as puberty and gender may account for differences in growth reported by
12    various investigators.  In a series of follow-up experiments (Ronis et al., 2001) reported a dose-
13    dependent decrease in load to failure in tibia from Pb-exposed (0.15% and 0.45% Pb-acetate in
14    drinking water) male pups only. Hormone treatments (estradiol in females or L-dopa,
15    testosterone or dihydrotestosterone in males) failed to attenuate Pb deficits during the pubertal
16    period. Distraction osteogenesis experiments performed after stabilization of endocrine
17    parameters (at 100 days of age) found decreased new endosteal bone formation and gap x-ray
18    density in the distraction gaps of Pb-exposed animals (Ronis et al., 2001).
19          Hamilton and O'Flaherty (1995) found Pb disrupted mineralization during growth when
20    they implanted demineralized bone matrix subcutaneously into male rats.  In the matrix that
21    contained 200 jig Pb/g of plaque tissue, alkaline phosphatase activity and cartilage
22    mineralization were absent, though calcium deposition was enhanced.  Separate experiments
23    found enhanced calcification and decreased alkaline phosphatase activity in  rats implanted with a
24    control (no Pb) matrix and given 1,000 ppm Pb in drinking water for 26 days.
25          In summary, results from animal studies suggest Pb exposure is capable of adversely
26    affecting bone growth and density, potentially manifesting its action through interference with
27    growth and hormonal  factors as well as toxic effects directly on bone.
28
      December 2005                           5-190       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.4   Regulation of Bone Cell Function in Animals - Systemic Effects
 2           of Lead
 3         Lead may exhibit multiple complex systemic effects that ultimately could influence bone
 4    cell function. As discussed in the animal studies below, Pb can modulate alterations in calcium
 5    binding proteins and in calcium and phosphorus concentration in the blood stream, in addition to
 6    potentially altering bone cell differentiation and function by altering plasma levels of growth
 7    hormone and calciotropic hormones such as vitamin D3 [1,25-(OH2)D3] and parathyroid
 8    hormone.
 9
10    5.8.4.1  Hypercalcemia/Hyperphosphatemia
11         Intravenous injection of Pb has been shown to produce both an acute hypercalcemia and
12    hyperphosphatemia in rats (Kato et al., 1977).  Injection of a relatively high dose of 30 mg/kg Pb
13    resulted in maximum values of calcium (17 mg%) after one  hour and maximum values of
14    phosphorus (13.5 mg%) after 30 minutes.  After  12 hours the levels of both calcium and
15    phosphorus had returned to baseline levels. Histochemical examination demonstrated deposition
16    of Pb into bone and dentin in the rats, suggesting a direct action of Pb on bone and/or teeth,
17    ultimately displacing calcium and phosphorus and thereby producing hypercalcemia and
18    hyperphosphatemia.
19
20    5.8.4.2  Vitamin D [1,25-(OH2)D3]
21         As discussed above, vitamin D [1,25-(OH2)D3] modulates the  normal range of calcium in
22    serum. In rats fed a low calcium or low phosphorus diet, ingestion of 0.82% Pb in the diet
23    reduced plasma levels of 1,25-(OH2)D3; however, this effect is lost when a high calcium or
24    normal phosphorus diet is given (Smith et al., 1981), suggesting a high calcium/phosphorus diet
25    reduces the susceptibility of vitamin D system to the effect of Pb. No mobilization of calcium
26    from bone or elevation of inorganic phosphorus was seen. Ronis et al. (2001) also reported no
27    effects of Pb on plasma concentrations of vitamin D metabolites, 25-OH D3 or 1,25-(OH2)D3, in
28    pubertal male rats exposed to either 0.15% or 0.45% Pb acetate in drinking water and maintained
29    on an adequate diet. Fullmer (1995) found vitamin D function was severely compromised in
30    young growing chicks given a diet low in calcium (0.1% calcium) for two weeks and then
31    exposed to 0.2% or 0.8% Pb in their diet for an additional one or two weeks.  In chicks

      December 2005                          5-191       DRAFT-DO NOT QUOTE OR CITE

-------
 1    maintained on an adequate diet (1.2% calcium), exposure to 0.2% or 0.8% Pb in the diet resulted
 2    in increased plasma levels of 1,25-(OH2)D3 as well as significantly increased intestinal
 3    Calbindin-D protein [a calcium binding protein induced by 1,25-(OH2)D3] and its associated
 4    mRNA, when compared with unexposed control chicks. Levels of intestinal Calbindin-D mRNA
 5    and protein and plasma levels of 1,25-(OH2)D3 were elevated during the first week of Pb
 6    exposure to chicks fed a diet deficient in calcium, but were significantly decreased by the second
 7    week of Pb exposure. The study suggested Pb was mediating its effect through 1,25-(OH2)D3,
 8    rather than via a direct action on the Calbindin-D protein.  Follow up studies by Fullmer et al.
 9    (1996) confirmed dose dependent increases in serum 1,25-(OH2)D3 levels (and Calbindin-D
10    protein and mRNA) with increasing dietary Pb exposure (0.1% to 0.8%) in similar experiments
11    performed on Leghorn cockerel chicks fed an adequate calcium diet.
12
13    5.8.4.3   Parathyroid Hormone
14          At least one animal study has associated experimental Pb exposure with secondary
15    hyperparathyroidism. Szabo et al. (1991) exposed Wistar Kyoto rats to either 1% Pb acetate in
16    water for a short term (10 weeks) or varying concentrations (0.001 to 1% Pb acetate) for a longer
17    term (24 weeks) to assess the influence of Pb on the interaction of the parathyroids with
18    1,25-(OH2)D3. Short term administration of 1% Pb resulted in significant increases in bone Pb;
19    however, total serum calcium and ionized serum calcium were significantly decreased,  as
20    compared to controls.  Circulating levels of 1,25-(OH2)D3  were  also decreased, though  the rats
21    were maintained on a normal calcium diet (0.95%). Parathyroid glands from rats exposed short
22    term to Pb were significantly increased in size over those in control animals (178  jig per gland
23    versus 96 jig per gland) and specific binding of 1,25-(OH2)D3 to parathyroid and intestinal tissue
24    was increased. Likewise, long term administration of 1% Pb resulted in significant increases in
25    bone Pb  and normalized parathyroid gland weights, and a significant decrease in the level of
26    1,25-(OH2)D3. In the long term study, a dose-dependent increase in parathyroid weight occurred
27    with increasing exposure to Pb in drinking water. The authors concluded the secondary
28    hyperparathyroidism was associated with, and/or a result of, the hypocalcemia and decreased
29    1,25-(OH2)D3 levels secondary to Pb exposure.
30
      December 2005                           5-192       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.4.4  Growth Hormone
 2          As discussed in Section 5.8.3, exposure to Pb has been associated with altered bone
 3    metabolism and decreased growth and skeletal development (Hamilton and O'Flaherty, 1994,
 4    1995; Gruber et al., 1997; Gonzalez-Riola et al., 1997; Escribano et al., 1997; Ronis et al.,
 5    1998a,b, 2001; Camoratto et al., 1993), suggesting perturbation of one or more endocrine factors
 6    such as growth hormone. To examine the effect of exposure to low-level Pb on pituitary growth
 7    hormone release, Camoratto et al. (1993) exposed pregnant female  rats to 0.02% Pb nitrate
 8    (125 ppm Pb) beginning on gestational day 5 and continuing in pups through postnatal day 48.
 9    Basal release of growth hormone from control and Pb-exposed pups at age 49 days was not
10    significantly different. Growth hormone releasing factor-stimulated release of growth hormone
11    from pituitaries of Pb-exposed pups was smaller than the stimulated release of growth hormone
12    from pituitaries of control animals (75% increase over baseline vs.  171% increase, respectively),
13    but the difference did not achieve significance (p = 0.08).  Growth hormone content of the
14    pituitary glands was also not influenced by Pb exposure.  Ronis et al. (1998b) reported similar
15    findings in rat pups exposed to 0.05%, 0.15%, or 0.45% Pb acetate  in drinking water from
16    gestation day 5 through postnatal day 85, with the exception being  significantly elevated
17    pituitary growth hormone levels at postnatal day 55. Taken together, these rat studies suggest
18    that differences in growth seen with Pb exposure may not necessarily be the result of alterations
19    in secretion of growth hormone.
20
21    5.8.5    Bone Cell Cultures Utilized to Test the Effects of Lead
22    5.8.5.1  Bone Organ Culture
23          In an early bone organ culture study utilizing incorporated radioactive Pb into fetal radii
24    and ulnae, Rosen and Wexler (1977) reported release of Pb as the concentration of calcium in the
25    media was reduced or with addition of parathyroid hormone, but that calcitonin inhibited the
26    release of Pb as expected, verifying the capacity of this model system.  The bone organ system
27    was subsequently used to evaluate the efficacy of Pb chelating agents,  such as D-Penicillamine
28    and CaNa2EDTA (Rosen and Markokwitz, 1980; Rosen et al.,  1982).
29
     December 2005                           5-193        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.5.2  Primary Cultures of Osteoclasts and Osteoblasts
 2          The ability to isolate primary cultures of osteoclasts and osteoblasts from mouse calvaria
 3    provided an additional experimental model system to study the effects of Pb on specific bone
 4    cells.  Using isolated osteoclasts and osteoblasts, Rosen (1983) reported that uptake of
 5    radioactive Pb by osteoclasts was rapid, almost linear, while osteoblasts showed very little
 6    increase in uptake of Pb at increasing media concentrations. Physiological concentrations of
 7    parathyroid hormone markedly  increased uptake of Pb and calcium by osteoclast cells and, once
 8    loaded with Pb, osteoclasts were capable of releasing Pb slowly into the media.  Further kinetic
 9    analysis of cultured osteoclastic bone cells indicated that cellular Pb is primarily associated with
10    the mitochondrial fraction (-78%) and that this Pb is readily exchangeable with the outside
11    media (Pounds and Rosen, 1986; Rosen and Pounds, 1988).  Experiments conducted to
12    characterize the steady-state kinetic distribution and metabolism of calcium and Pb supported the
13    concept that the two elements are metabolized similarly in the osteoclasts cells (Rosen and
14    Pounds, 1989).
15
16    5.8.5.3  Rat Osteosarcoma Cell Line (ROS 17/2.8)
17          In recent years, the rat osteosarcoma cell line ROS  17/2.8 has been used extensively to
18    investigate the influence of Pb on various cellular processes and kinetics within these osteoblast-
19    like cells. The ROS 17/2.8 cell  model is useful in that the cells are capable of producing
20    osteocalcin (a bone protein important for proper bone mineralization), have high alkaline
21    phosphatase activity (an enzyme normally associated with mineralization of cartilage), possess
22    vitamin D receptors, and respond to parathyroid hormone. In comparisons of cellular lead
23    toxicity and metabolism between  primary cell culture from mouse calvaria and the rat
24    osteosarcoma cell line, Long and coworkers (1990a) reported remarkable similarities in the
25    profile of radiolabeled Pb  kinetics and intracellular Pb distribution. Using this cell line, Schanne
26    and coworkers (1989) simultaneously measured intracellular Pb and calcium concentrations and
27    found 5 and 25 micromolar Pb produced sustained 50% and 120% (respectively) increases in
28    intracellular calcium over a 5 hour period, and that measurable entry of Pb into the cells could be
29    demonstrated at the higher concentration.  These findings  advanced the hypothesis that
30    perturbation of intracellular calcium concentration may be the mechanism of Pb  bone toxicity.
31    Schirrmacher and coworkers (1998) reported that calcium homeostasis is upset within

      December 2005                            5-194        DRAFT-DO NOT QUOTE OR CITE

-------
 1    20 minutes of its addition to calvarial bone cell culture.  Their results suggested that the calcium-
 2    ATPases of intracellular stores were potentially poisoned by Pb entering the cells.  Wiemann et
 3    al. (1999) demonstrated that Pb was also capable of interfering with the calcium release activated
 4    calcium influx (CRAC) in calvarial bone cell cultures.  Pb was found to partially inhibit the
 5    influx of calcium into the bone cells, plus influx of Pb into the cells was greatly enhanced
 6    (2.7 fold) after CRAC had been induced.  These effects of Pb were found to be independent of
 7    any inhibitory effect on calcium-ATPase.
 8          Miyahara et al. (1995) performed a series of experiments in 45Ca-labeled bone organ
 9    culture to determine whether the Pb-induced hypercalcemia was the result of the active process
10    of biological bone resorption or simply physiochemical mineral dissolution. Lead introduced
11    into the culture at concentrations of 50 jiM and above stimulated the release of calcium and
12    hydroxyproline into the medium, however no release was elicited from bones inactivated by
13    freezing and thawing. Pb-stimulated 45Ca release was inhibited by eel calcitonin, bafilomycin
14    AI, and scopadulcic acid B, suggesting the release was secondary to osteoclastic bone resorption.
15    Further evidence to support this conclusion came from experiments examining the  influence of
16    two inhibitors of cyclooxygenase on Pb-induced bone resorption.  Lead was found to stimulate
17    prostaglandin E2 release and in cultures, there was a high correlation between prostaglandin E2
18    released into the media and 45Ca release.  In the presence of cyclooxygenase inhibitors (blocking
19    prostaglandin synthesis), Pb-stimulated 45Ca release was inhibited suggesting the mechanism of
20    bone resorption  in this instance was via a prostaglandin E2-mediated mechanism.
21          Lead has been demonstrated to directly impair production of osteocalcin by  ROS 17/2.8
22    cells by 70% after 24 hours of exposure to 25 micromolar Pb (Long et al., 1990a).  The resulting
23    decrease in cell proliferation is in agreement with similar studies by Sauk et al., 1992).
24    Interestingly, exposure of dental pulp cells, which  also produce osteocalcin, to a similar
25    concentration of Pb reduced osteocalcin production by 55% after 12 hours of exposure
26    (Thaweboon et al., 2002). Vitamin D has been shown to increase osteocalcin production in ROS
27    17/2.8 cells; however, Pb inhibited the vitamin D-stimulated osteocalcin production in a dose-
28    dependent manner from 0 up to 25 micromolar concentrations, plus was shown to be capable of
29    attenuating basal (non-vitamin D-stimulated) osteocalcin production (Long et al., 1990a). Lead
30    (5 to 20 micromolar) inhibition of vitamin D stimulation of osteocalcin in ROS cells was also
31    reported by  Guity and coworkers (2002).  Later studies suggested that Pb acts by inhibiting

      December 2005                            5-195       DRAFT-DO NOT QUOTE OR CITE

-------
 1    vitamin D activation of calcium channels and interferes with regulation of calcium metabolism
 2    (Schanne et al., 1992), though apparently this effect is not mediated via PKC (Guity et al., 2002).
 3    Angle and coworkers (1990) reported that 24 hours of incubation with vitamin D (10 nM) was
 4    capable of evoking a 4 to 5 fold increase in osteocalcin production and a 100% increase in
 5    cellular alkaline phosphatase activity in ROS cells. Osteocalcin production and cellular DNA
 6    contents were increased 100% and 20% respectively by addition of insulin-like growth factor
 7    (92.5 ng/mL). Consistent with a toxic effect of Pb on osteoblast function, the addition of 1 to
 8    10 jiM Pb to the system inhibited both basal and stimulated osteocalcin secretion, alkaline
 9    phosphatase activity and DNA contents (Angle et al., 1990). Dose- and time-dependent
10    reduction in alkaline phosphatase activity with Pb exposure (2 to 200 micromolar) has also been
11    reported in osteosarcoma cells, along with parallel reductions in steady state levels of alkaline
12    phosphatase mRNA levels (Klein and Wiren, 1993).  No effect on cell number or DNA and
13    protein synthesis was seen at these levels of Pb exposure.
14          Though the exact mechanism of Pb toxicity on osteocalcin was unclear, Pb was known to
15    inhibit some of the functional properties of osteocalcin including inhibition of osteocalcin
16    adsorption to hydroxyapatite. An investigation by Dowd and coworkers (1994) utilized the
17    ability of osteocalcin added to a solution of 43CaCl2 to broaden 43Ca resonance, as a method to
18    examine binding of calcium to osteocalcin and the influence of Pb on calcium binding. It was
19    determined that the dissociation constant of calcium for osteocalcin was 7 micromolar, while the
20    dissociation constant for Pb was determined by competitive displacement to be 2 nM, indicating
21    more than three orders of magnitude tighter binding of Pb than calcium to osteocalcin and the
22    likelihood that even submicromolar levels of free Pb would significantly inactivate osteocalcin.
23    Circular dichroism indicated that upon binding, Pb induces a similar structural change in
24    osteocalcin to that found with calcium binding, but the binding with Pb occurs at 2 orders of
25    magnitude lower than with calcium (Dowd et al., 2001).  Similarly, hydroxyapatite binding
26    assays indicated Pb causes an increased absorption to hydroxyapatite that is  similar to calcium,
27    but again at 2 to 3  orders of magnitude lower concentration, potentially leading to low bone
28    formation rates and/or density (Dowd et al., 2001).
29          Besides perturbation of calcium metabolism, Pb has been shown to reduce intracellular
30    free magnesium concentrations by 21% in osteosarcoma cells incubated in 10 micromolar Pb for
31    2 hours (Dowd et al.,  1990). Under these same conditions, the unidirectional rate of ATP

      December 2005                           5-196       DRAFT-DO NOT QUOTE OR CITE

-------
 1    synthesis (i.e. P; to ATP) was reduced by a factor greater than 6 over control cultures.
 2    Impairment of both of these processes by Pb could ultimately influence bone growth and
 3    development.
 4          Lead has also been show to perturb Epidermal Growth Factor's (EGF) control of
 5    intracellular calcium metabolism and collagen production in ROS cells (Long and Rosen, 1992).
 6    EGF is known to activate protein kinase C (PKC), resulting in increased calcium influx and
 7    through this mechanism, decreased collagen synthesis. Incubation of ROS cells with
 8    5 micromolar Pb and 50 ng/mL EGF for 20 hours resulted in a 50% increase in total cell calcium
 9    versus the calcium increase seen in cells treated with EGF alone,  suggesting more than one site
10    of action is involved in calcium messenger perturbation. A similar finding was reported by Long
11    and coworkers (1992) who found that treatment of Pb  (25 micromolar) intoxicated osteosarcoma
12    cells with parathyroid hormone (PTH, 400 mg/mL) resulted in a greater increase in cell calcium
13    than with either treatment alone. Supplementary inhibition of collagen synthesis has also been
14    reported with the addition of 25 micromolar Pb plus 50 ng/mL EGF, suggesting more than one
15    site of action for the effect of Pb on collagen synthesis (Long and Rosen, 1992).  Additional
16    study has since suggested that Pb activates PKC in ROS cells and that PKC mediates the rise in
17    intracellular calcium (Schanne et al., 1997). The observation that calphostin C, an inhibitor of
18    PKC, prevented the Pb-induced elevation of intracellular calcium supported this hypothesis, as
19    did the fact that free Pb at concentrations of 10"11 to 10"7 M directly activated PKC in the absence
20    of activating concentrations of calcium. This would suggest Pb is capable of activating PKC at
21    concentrations approximately 3,000 times lower than calcium.
22          Finally, Pb has been shown to be capable of inhibiting secretion of osteonectin, a bone
23    related protein found in areas of active morphogenesis (Sauk et al.,  1992). Treatment of ROS
24    17/2.8 cells with lead (4.5 X 10"6 M to 4.5 10"7 M) demonstrated that intracellular osteonectin
25    levels were actually enhanced; however, the secretion of osteonectin into the media was delayed
26    or inhibited.  Protein production of collagen and the endoplasmic reticulum protein,  Asp47, were
27    relatively unaffected by  Pb at these concentrations. The intracellular retention of osteonectin
28    coincided with  a decrease in levels of osteonectin mRNA, suggesting the processes associated
29    with translation and secretion of osteonectin are sensitive to Pb.
30
      December 2005                           5-197       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.5.4  Human Osteosarcoma Cells (HOS TE 85)
 2          Evidence exists that Pb is directly osteotoxic to bone cells in culture.  Studies examining
 3    the sensitivity of human osteosarcoma cells (HOS TE 85) to Pb found proliferation of the cells
 4    was inhibited at Pb concentrations of 4 |imol/l, while cytotoxicity occurred at the 20 |imol/l Pb
 5    concentration (Angle et al., 1993).  In parallel experiments, rat osteosarcoma cells (ROS  17/2.8)
 6    were found to be somewhat less sensitive to the effects of Pb with inhibition of proliferation
 7    occurring at 6 |imol/l Pb concentration and cytotoxicity at Pb concentrations over 20 |imol/l.
 8
 9    5.8.5.5  Chick Chondrocytes
10          The effects  of Pb on cartilage biology have been examined in isolated avian chondrocytes
11    obtained from 3 to 5 week old chicks (Hicks et al., 1996).  Exposure to media containing 0.1 to
12    200 jiM Pb acetate or chloride were found to decrease thymidine incorporation, suppress alkaline
13    phosphatase, and suppress both type II and type X collagen expression at the mRNA and protein
14    levels.  Cytotoxicity of the cultures from Pb exposure was dismissed as proteoglycan synthesis
15    was found to be augmented, suggested Pb selectively inhibits specific aspects of the chondrocyte
16    growth plate. Using the avian chondrocyte model, Zuscik et al. (2002) similarly reported Pb
17    exposure (1 to 30 jiM) causing a dose-dependent inhibition of thymidine incorporation into the
18    growth plate, with a 60% reduction in proliferation at the highest concentration. Addition of
19    TGF-P1 and PTHrP, regulators of growth plate, both separately stimulated thymidine
20    incorporation, an effect that was dose-dependently blunted in the presence of Pb. At the highest
21    Pb concentration (30 |iM), inhibition was significantly less in the chondrocytes treated with Pb +
22    TGF-P1 (24%) and Pb + PTHrP (19%) than for Pb alone (60%), suggesting the interaction of Pb
23    with these growth factors may be independent of its primary action on the chondrocyte cells.
24    Support for a direct action of Pb on these growth regulators is supported by the finding that
25    normal TGF-pl and PTHrP suppression of type X collagen expression is significantly reversed
26    in a dose-dependent fashion in the presence of Pb.  This  effect evidently was not mediated by
27    BMP-6 (Bone Morphogenic Protein), an inducer of terminal differentiation known to partially
28    reverse the inhibitory effect of PTHrP, because in the presence of Pb, PTHrP significantly
29    suppressed BMP expression, while combined exposure to Pb and TGF-pl  increased BMP
30    expression approximately 3-fold. Further experiments performed on chick sternal chondrocyte
31    cultures, utilized PTHrP responsive (AP-1) and non-responsive (NF-^B) reporter constructs to

      December 2005                           5-198       DRAFT-DO NOT  QUOTE OR CITE

-------
 1    examine potential effects of Pb on signaling. While having no effect on the basal activity of the
 2    AP-1 reporter, Pb dose-dependently enhanced PTHrP induction of the responsive AP-1 reporter.
 3    Lead dose-dependently inhibited the basal activity of the non-PTHrP responsive, NF-K;B
 4    reporter. Taken together, these studies demonstrate that Pb has an inhibitory effect on the
 5    process of endochondral bone formation and that the effects of Pb are likely from its modulation
 6    of growth factors and second messengers involved in cell signaling responses.
 7
 8    5.8.6   Bone Lead as a Potential Source of Toxicity in Altered
 9            Metabolic Conditions
10          Lead is avidly taken up by bone and incorporated  into bone matrix, where a substantial
11    amount can remain over the lifetime of an organism. The uptake and incorporation of Pb into
12    bone during acute exogenous exposures may be of short term benefit by limiting the exposure of
13    other, more sensitive tissues; however, this does not eliminate Pb from the system. Subsequent
14    release of Pb from this endogenous storage can produce a lifetime of steady, low level Pb
15    exposure during periods of normal bone remodeling, while elevated Pb  release during times of
16    increased bone metabolism and turnover (i.e., pregnancy, lactation, menopause, and
17    osteoporosis) can elevate blood levels of Pb significantly, potentially to toxic concentrations.
18    This is especially relevant when there is concurrent exogenous exposure to Pb, as current blood
19    Pb levels are a composite of current and past Pb exposure. Of greater concern is the mobilization
20    of Pb during pregnancy and subsequent transfer to the developing brain of the fetus across the
21    poorly developed blood:brain barrier.  Maternal Pb also appears in breast milk, providing further
22    exposure of the infant to Pb during lactation. Currently, the majority of animal studies
23    examining mobilization of Pb from bone stores have focused principally on elevation of Pb
24    levels or transfer of Pb, rather than reporting toxic effects associated with these exposures. Note
25    that in most instances the mobilization and elimination of Pb is much faster in laboratory animals
26    than in humans.  For example, as discussed in Section 5.8.3, Hac and Kruchniak (1996) reported
27    approximately 64% of Pb given over a 12 week period remained in the bones of rats 64 days post
28    exposure. Therefore, the caveats of experiments performed in small animals, especially when
29    examining mobilization of Pb stores, must be taken into consideration.
30
      December 2005                           5-199        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.6.1  Pregnancy and Lactation
 2          Pregnancy, and to a much greater extent, lactation, place significant calcium demands on
 3    the mother as she provides all the necessary calcium requirements of the developing fetus/infant.
 4    During these times of metabolic stress, increased demineralization of maternal bone occurs to
 5    supplement demand, unfortunately accompanied by the concurrent mobilization and release of
 6    Pb stored in the maternal skeleton from past exposure.  Studies in several animal models have
 7    shown that maternal bone Pb can be mobilized during pregnancy and lactation, ultimately being
 8    transferred to the fetus during gestation and breast feeding. Keller and Doherty (1980)
 9    administered radiolabeled Pb drinking water (200|ig/mL) to female mice for 105 days prior to
10    mating or 105 days prior to mating and during periods of gestation and lactation (total 160 days
11    of exposure).  The results suggested very little Pb was transferred from mother to fetus during
12    gestation, however, Pb transferred in milk and retained by the pups accounted for 3% of the
13    maternal body burden of those mice exposed to Pb prior to mating only.  The amount of Pb
14    retained in these pups exceeded that retained in the mothers, suggesting lactation effectively
15    transfers Pb burden from mother to suckling offspring. Transfer of Pb from mothers was
16    significantly higher when Pb was supplied continuously in drinking water,  rather than terminated
17    prior to mating. Considerably  higher lactational transfer of Pb from rat dams compared to
18    placental transfer has also been reported (Palminger Hallen et al., 1996). Continuous exposure
19    of rat dams to Pb until day 15 of lactation resulted in milk Pb levels 2.5 times higher than in
20    whole blood, while termination of maternal Pb exposure at parturition yielded equivalent blood
21    and milk levels of Pb, principally from Pb mobilized from maternal bone.
22          Using rats chronically exposed to Pb in drinking water, Maidonado-Vega et al. (1996)
23    studied intestinal absorption of Pb, its mobilization, and redistribution during lactation. In rats
24    exposed to Pb 144 days  prior to lactation, the process of lactation itself elevated blood Pb and
25    decreased bone Pb, indicating mobilization of Pb from bone as there was no external source of
26    Pb during the lactation process. Rats exposed to Pb for 158 days (144 days prior to lactation and
27    14 days during lactation) also experienced elevated BLLs and loss of Pb from bone. Lead
28    exposure only during the 14 days of lactation was found to significantly increase intestinal
29    absorption and deposition (17 fold increase) of Pb  into bone compared to non-pregnant rats,
30    suggesting enhanced absorption of Pb takes place during lactation.  As in other previous studies,
31    the highest concentration of Pb in bone was found in non-pregnant non-lactating control animals,

      December 2005                           5-200       DRAFT-DO NOT QUOTE OR CITE

-------
 1    with significantly decreased bone Pb in lactating rats secondary to bone mobilization and transfer
 2    via milk to suckling offspring.  Follow-up studies examining the influence of dietary calcium
 3    found when calcium was altered from the normal 1% to 0.05%, bone calcium concentration
 4    decreased by 15% and bone Pb concentration decreased by 30% during the first 14 days of
 5    lactation (Maidonado-Vega et al., 2002). In non-lactating rats on the 0.05% calcium diet, there
 6    were also decreases in bone calcium, but neither incremental bone resorption nor Pb efflux from
 7    bone, suggesting the efflux from bone during lactation was related to bone resorption. Of
 8    interest, enhancement of calcium (2.5%) in the diet of lactating rats increased calcium
 9    concentration in bone by 21%, but did not decrease bone resorption, resulting in a 28% decrease
10    in bone Pb concentration and concomitant rise in systemic toxicity.  In both studies, the authors
11    concluded that Pb stored in bone should be considered a major source of self-intoxication and of
12    exposure to suckling offspring.
13          In one of few studies showing a toxic effect, Han et al. (2000) demonstrated adverse
14    effects in rat offspring born to females whose exposure to Pb ended well before pregnancy. Five
15    week-old-female rats had been given Pb-acetate in drinking water (250 mg/mL) for five weeks,
16    followed by a one month period without Pb exposure before mating.  To test the influence of
17    dietary calcium on Pb absorption and accumulation, some pregnant rats were fed diets deficient
18    in calcium (0.1%) while others were maintained  on a normal calcium (0.5%) diet. As expected,
19    all Pb-exposed dams and pups had elevated blood Pb levels; however, pups born to  dams fed the
20    diet deficient in calcium during pregnancy had higher blood and organ Pb concentrations
21    compared to pups from dams fed the normal diet. Significantly, pups born to Pb-exposed dams
22    had lower mean birth weights and birth lengths than pups born to non-Pb-exposed control dams
23    (p < 0.0001), even after confounders such as litter size, pup sex,  and dam weight gain were taken
24    into account. The authors concluded that while increases in dietary calcium during  pregnancy
25    are capable of reducing Pb accumulation in the fetus, they cannot prevent the decreases in birth
26    weight and length associated with pre-maternal Pb exposure and subsequent mobilization. This
27    has relevance in human pregnancy, as many women experience exposure to Pb during their
28    lifetimes (especially during childhood) and mobilization  of the Pb from bone stores during
29    pregnancy could present toxic complications.
30          Within the last decade, an invaluable method to explore the kinetics of Pb transfer from
31    bone to blood has been developed and evaluated (Inskip et al., 1996; O'Flaherty et al., 1998).

      December 2005                            5-201        DRAFT-DO NOT QUOTE OR CITE

-------
 1    The method utilizes recent administration of sequential doses of Pb mixes enriched in stable
 2    isotopes ( 204Pb, 206Pb, and 207Pb) to female cynomolgus monkeys (Macaca fascicularis) that
 3    have been chronically (1,300 to 1,500 jig Pb/kg body weight per day for ten years or greater)
 4    administered a common Pb isotope mix.  The stable isotope mixes serve as a marker of recent,
 5    exogenous Pb exposure, while the chronically administered common Pb serves as a marker of
 6    endogenous (principally bone) Pb. From thermal ionization mass spectrometry analysis of the
 7    Pb isotopic ratios of blood and bone biopsies  collected at each isotope change, and using end-
 8    member unmixing equations, it was determined that administration of the first isotope label
 9    allows measurement of the contribution of historic bone stores to blood Pb.  Exposure to
10    subsequent isotopic labels allowed measurements of the contribution from historic bone Pb
11    stores and the recently administered enriched isotopes that incorporated into bone (Inskip et al.,
12    1996).  In general the contribution from the historic bone Pb (common Pb) to blood lead level
13    was constant (-20%), accentuated with spikes in total blood Pb due to the current administration
14    of the stable isotopes.  After cessation of each sequential administration, the concentration of the
15    signature dose rapidly decreased.  Initial attempts to apply a single-bone physiologically based
16    model of Pb kinetics were unsuccessful until adequate explanation of these rapid drops in stable
17    isotopes in the blood were incorporated (O'Flaherty et al., 1998). Once revisions were added  to
18    account for rapid turnover of the trabecular bone compartment and slower turnover rates of
19    cortical bone compartment, an acceptable model evolved.  From this model it was reported that
20    historic bone Pb from  11 years of continuous  exposure contributes approximately 17% of the
21    blood Pb concentration at Pb concentration over 50 |ig/dL, reinforcing the concept that the length
22    of Pb exposure and the rates of past and current Pb  exposures help determine the fractional
23    contribution of bone Pb to total blood Pb  levels (O'Flaherty et al., 1998). The turnover rate for
24    cortical (-88% of total bone by volume) bone in the adult cynomolgus monkey was estimated by
25    the model to be -4.5% per year, while the turnover rate for trabecular bone was estimated to be
26    33% per year.
27          Using the method of sequential stable isotope administration, Franklin et al. (1997)
28    examined flux of Pb from maternal bone  during pregnancy of 5 female cynomolgus monkeys
29    who had been previously exposed to common Pb (approximately 1,100 tol,300 jig Pb/kg body
30    weight) for about 14 years. In general, lead levels in  maternal blood attributable to Pb from
31    mobilized bone were reported to drop 29  to 56% below prepregnancy baseline levels during the

      December 2005                           5-202       DRAFT-DO NOT QUOTE OR CITE

-------
 1    first trimester of pregnancy.  This was ascribed to the known increase in maternal fluid volume,
 2    specific organ enlargement (e.g., mammary glands, uterus, placenta), and increased metabolic
 3    activity that occurs during pregnancy. During the second and third trimesters, when there is a
 4    rapid growth in the fetal skeleton and compensatory demand for calcium from the maternal
 5    blood, the Pb levels increased up to 44% over pre-pregnancy levels. With the exception of one
 6    monkey, blood Pb concentrations in the fetus corresponded to those found in the mothers, both in
 7    total Pb concentration and proportion of Pb attributable to each isotopic signature dose (common
 8    = 22.1% vs. 23.7%, 204Pb = 6.9% vs. 7.4%, and 206Pb = 71.0% vs. 68.9%, respectively). From 7
 9    to 25% of the Pb found in fetal bone originated from maternal bone, with the balance derived
10    from oral dosing of the mothers with isotope during pregnancy. Of interest, in offspring from a
11    low Pb exposure control monkey (blood Pb <5 jig/100 g) -39% of Pb found in fetal bone was of
12    maternal origin, suggesting enhanced transfer and retention of Pb under low Pb conditions.
13          Clearly, the results of these studies  show that Pb stored in bone is mobilized during
14    pregnancy and lactation, exposing both mother and fetus/nursing infant to blood/milk Pb levels
15    of potential toxicity.  Of equal concern,  a significant proportion of Pb transferred from the
16    mother is incorporated into the developing skeletal system of the offspring, where it can serve as
17    a continuing source of toxic exposure.  The above study by Franklin et al. (1997) illustrates the
18    utility of sequentially administered stable isotopes in pregnancy; however, its use may also be
19    applicable in studies of lactation, menopause, osteoporosis, and other disease states where
20    mobilization of bone and release of Pb stores occurs. Furthermore, given that isotopic ratios of
21    common Pbs vary by location and  source of exposure, when humans migrate from one area and
22    source of exposure to another, it is possible to document changes in mobilized Pb,  especially
23    during times of metabolic stress.
24
25    5.8.6.2  Age/Osteoporosis
26          The age of an animal at the  time of exposure to Pb has been shown to influence the uptake
27    and retention of Pb by bone.  In experiments to determine the influence of age on this process,
28    Han et al. (1997) exposed rats for five weeks to 250 mg/L Pb-acetate in drinking water beginning
29    at 5 weeks of age (young child), 10 weeks of age (mid-adolescence), or 15 weeks of age (young
30    adult), followed by a 4 week period of without Pb exposure. An additional group of rats were
31    exposed to Pb beginning at 5 weeks, but examined following an 8 or 20 week period after

      December 2005                            5-203        DRAFT-DO NOT QUOTE OR CITE

-------
 1    cessation of Pb. Significantly lower blood and bone Pb concentrations were associated with
 2    greater age at the start of Pb exposure and increased interval since the end of exposure.
 3    However, young rats beginning exposure to Pb at 5 weeks and examined 20 weeks after
 4    cessation of exposure, still had bone Pb concentrations higher than those found in older rats only
 5    4 weeks after cessation of exposure. This demonstrated that exposure to Pb at a young age leads
 6    to significant skeletal Pb accumulation and retention, despite the high rate of bone remodeling
 7    that occurs during growth and development at that time.
 8          At the opposite end of the spectrum, Cory-Slechta et al. (1989) studied differences in
 9    tissue distribution of Pb in adult and old rats. Adult (8 months old) and old (16 months old) rats
10    were exposed to 50 ppm Pb-acetate in drinking water for 11 months, at which time the
11    experiment was completed. Bone (femur) Pb levels in older rats were found to be less than those
12    in younger rats; however, blood lead levels were higher in the older rats. Of interest, brain Pb
13    concentrations  in the older rats exposed to Pb were significantly higher, and brain weights were
14    significantly less than the brain Pb concentration and weights of unexposed older control rats or
15    adult rats exposed to Pb, suggesting a potential detrimental effect. The authors suggested that a
16    possibility for the observed differences in tissue concentrations of Pb was due to changes in the
17    capacity of bone to store Pb with advanced age. In a subsequent study, Cory-Slechta (1990b)
18    examined kinetic and biochemical responses of young (21 day old), adult (8 months old),  and old
19    (16 months old) rats exposed to Pb at 0, 2, or 10 mg Pb acetate/kg/day over a 9.5 month
20    experimental period. Results suggested that older rats may have increased vulnerability to Pb
21    due to increased exposure of tissues to Pb and greater sensitivity of the tissues to the effects of
22    Pb. As in the previous study (Cory-Slechta et al., 1989), lower bone levels of Pb were present in
23    older rats with  concomitant elevated levels of Pb in brain and other tissues, supporting the
24    hypothesis that exposure to Pb over a lifetime may contribute to deterioration of health in old
25    age, potentially during times of heightened bone remodeling such as occurs during osteoporosis.
26    In studies of bone Pb metabolism in a geriatric, female nonhuman primates exposed to Pb
27    approximately  10 years previously, McNeill et al. (1997) reported no significant changes in bone
28    Pb level over a 10 month observation period as measured by 109Cd K X-ray fluorescence.  The
29    mean half-life of Pb in bone of these animals was found to be 3.0 ± 1.0 years, consistent with
30    data found in humans, while the endogenous exposure level due to mobilized Pb was 0.09 ±
31    0.02 |ig/dL blood. Results examining Pb accumulation in the bones of aging male mice suggest

      December 2005                           5-204       DRAFT-DO NOT QUOTE OR CITE

-------
 1    low levels of bone Pb contributing to the osteopenia observed normally in C57BL/6J mice
 2    (Massie and Aiello, 1992).  The mice were maintained on regular diet (0.258 ppm Pb) and water
 3    (5.45 ppb Pb) from 76 to 958 days of age.  While the Pb content of femurs increased by 83%, no
 4    significant relationship was found between Pb and bone density, bone collagen, or loss of
 5    calcium from bone.
 6
 7    5.8.6.3 Weight Loss
 8          The relationship between body mass and bone mass is highly correlated and during times
 9    of loss of body weight, such as dietary restriction, a concomitant loss of bone mass also occurs.
10    It is therefore possible that Pb stored in bone from prior exposures could be released into the
11    system as skeletal bone is mobilized and result in Pb toxicity. To examine the influence of
12    weight loss on release of stored Pb, Han et al. (1996) first exposed rats to Pb in drinking water
13    (250 mg/1 of Pb as acetate) for 5 weeks, followed by a 4 week washout period without Pb to
14    allow primarily accumulation in the skeleton. Rats were then randomly assigned to a weight
15    maintenance group, a moderate weight loss group (70% of maintenance diet), or a substantial
16    weight loss group (40% of maintenance diet) for a four week period.  At the end of this
17    experimental period the blood and bone levels of Pb did not differ between groups, however, the
18    amount and concentration of Pb in the liver increased significantly. A follow up study in rats
19    previous exposed to Pb for two weeks was undertaken to determine the  effect of weight loss and
20    exercise on the distribution of Pb (Han et al., 1999). They found weight loss secondary to
21    dietary restriction to be the critical factor elevating  organ Pb levels and, contrary to their first
22    study, elevated blood levels of Pb. No significant difference in organ or blood Pb concentrations
23    were reported between the exercise vs. no  exercise  groups.  These studies suggest Pb toxicity
24    could occur in those previously exposed to Pb during times of dietary restriction.
25
26    5.8.7  Bone and Lead Summary
27          Lead substitutes for calcium and is  readily taken up and stored in the bone of experimental
28    animals, potentially allowing bone cell function to be compromised both directly and indirectly
29    by exposure. In general, relatively short term exposure of mature animals to Pb does not result
30    in significant growth suppression, however, chronic Pb exposure during times of inadequate
31    nutrition have been  shown to adversely influence bone growth, including decreased bone density,

      December 2005                            5-205       DRAFT-DO NOT QUOTE OR CITE

-------
 1    decreased trabecular bone, and growth plates.  Exposure of developing animals to Pb during
 2    gestation and the immediate postnatal period has clearly been shown to significantly depress
 3    early bone growth in a dose-dependent fashion, though this effect is not manifest below a certain
 4    threshold.  Numerous mechanisms for the toxic effect of Pb on bone have been explored using
 5    various animal models.  Systemically, Pb has been shown to disrupt mineralization of bone
 6    during growth, to alter calcium binding proteins, and to increase calcium and phosphorus
 7    concentration in the blood stream, in addition to potentially altering bone cell differentiation and
 8    function by altering plasma levels of growth hormone and calciotropic hormones such as vitamin
 9    D3 [1,25-(OH2)D3].
10          Bone cell cultures of both animal  and human derivation have substantially contributed to
11    the general understanding of the adverse effects of Pb on bone cell metabolism directly and its
12    indirect effect on bone and bone cells by perturbation of numerous local and systemic factors.
13    These in vitro studies have indicated that Pb is primarily taken up by osteoclasts and likely
14    perturbs intracellular calcium homeostasis secondary to osteoclastic bone resorption.  Bone cell
15    proliferation is also inhibited. Exposure of bone cell cultures to Pb has been shown to impair
16    vitamin D-stimulated production of osteocalcin, inhibit secretion of bone-related proteins such as
17    osteonectin and collagen, and suppress bone cell proliferation,  potentially by interference with
18    such factors as GH, EOF, TGF-P1, and PTHrP.
19          Finally, several animal studies have suggested Pb stored in bone can serve as a
20    continuing, endogenous source of exposure for an individual or can be transferred from mother
21    to offspring during pregnancy and/or lactation, with potentially toxic consequences. Periods of
22    extensive bone remodeling, (i.e., during weight loss, advanced age, altered metabolic state, and
23    pregnancy and lactation) are all associated with mobilization of Pb stores from bone of animals.
24    During pregnancy, transfer of Pb from mother to offspring has been documented, however,
25    available evidence suggests a more significant transfer from mother to offspring occurs during
26    lactation when the concentration of Pb in mother's milk can be several times higher than
27    corresponding blood levels.  Despite the extensive remodeling of bone that occurs during growth
28    and development of young animals, a significant amount of Pb can be accumulated and retained
29    during times of exposure.
30
      December 2005                           5-206       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.8   Teeth - Introduction
 2          There was little information in the prior 1986 AQCD relating lead exposure to adverse
 3    outcomes in the teeth of animals.  At that time, the incorporation of Pb into teeth was recognized
 4    as was the fact that tooth Pb increased with age, proportional to the rate of exposure and roughly
 5    proportional to the blood Pb concentration.
 6          Teeth consist of a hard outer layer of enamel, supported by an underlying layer of dentin,
 7    which itself is supported by a connective tissue known as the dental pulp. Enamel is the hardest
 8    substance in the body and the most highly  mineralized, consisting of-96% mineral (calcium
 9    hydroxyapatite substituted with carbonate ions) and 4% other organic materials, while dentin is
10    only -70% mineral.  The formation of enamel (amelogenesis) occurs as a two stage process of
11    organic matrix production with -30%  mineralization,  followed by removal of water and proteins
12    from the matrix with concurrent further mineralization.  As in bone, Pb ions are apparently
13    capable of substituting for calcium ions in  the mineralizing tooth, becoming essentially trapped.
14    However, unlike bone, the tooth, with  subtle exceptions, does not undergo a remodeling process.
15    Dentin formation (dentinogenesis) can be likened to endochondral bone formation, in that an
16    unmineralized matrix (predentin, rather than cartilage) is laid down first, followed by
17    mineralization to mature dentin.  The cells  responsible for amelogenesis  and dentinogenesis,
18    called ameloblasts and odontoblasts respectively, are similar to osteoblasts in that they respond
19    to various signaling factors, secrete matrix proteins, and create an environment favorable to
20    deposition of minerals. After enamel formation on a specific tooth is completed, ameloblasts are
21    lost and no additional enamel is laid down  with the exception of certain teeth in rodents. These
22    teeth, typically incisors on rats, mice, and most other rodents, continuously erupt to offset the
23    attrition that occurs with daily use. Therefore, the process of amelogenesis is ongoing, albeit
24    confined to  a localized area, throughout the life of the animals.  For this reason rodents have
25    been utilized extensively to examine the processes of amelogenesis and the influence of various
26    toxic agents, such as Pb,  on tooth development. Ameloblasts are especially sensitive to toxins
27    and altered metabolic conditions and respond to such insults with disruption of enamel
28    formation. When disruption occurs, defects in the enamel can occur, typically as a band of
29    malformed or altered enamel.  As described below, exposure of animals to various
30    concentrations of Pb during tooth development is not only capable of creating distinctive
31    marking of enamel ("Pb lines"), but may influence the resistance of the enamel to dental decay.

      December 2005                           5-207        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Within the dental pulp, a layer of odontoblasts continue to reside against the inner layer of the
 2    primary dentin for the life of the tooth. During this time the odontoblasts are systematically
 3    slowly putting down thin layers of secondary dentin, slowly decreasing the size of the pulp
 4    chamber with age. Lead present during this process has been shown to be readily taken up by
 5    this dentin layer, providing a potential marker of historic Pb exposure.  Though the enamel is a
 6    non-living substance, it is not entirely inert. The external surface of enamel is more or less in a
 7    continuous state of flux or turnover as it chemically demineralizes from acids consumed or
 8    produced in the mouth by bacteria, followed by remineralization of demineralized enamel when
 9    contact with saliva supersaturated with calcium and phosphate ions occurs. Lead present during
10    this process can easily be released from enamel and/or incorporated initially or back into it
11    depending on the circumstances.
12          In summary, Pb has the potential to disrupt the various processes associated with
13    formation of teeth, plus incorporate itself into all mineralized tooth tissues during formation.
14    Posteruptively, Pb can become incorporated into the secondary dentin, and can be taken up or
15    released from the outer surface layer of enamel during times of remineralization/
16    demineralization.  As described below, exposure of animals to Pb has been associated with
17    adverse dental outcomes.
18
19    5.8.9   Uptake of Lead by Teeth
20          As seen with bone, uptake of Pb into the teeth of animals has been demonstrated in a
21    number of studies and by multiple routes of administration.  Twenty four hours after a single
22    intraperitoneal injection of radioactive Pb-203 (203Pb, 1 |ig/kg) to young (15 day  suckling rats)
23    and old (120 day) female rats, 0.7% of the injected dose was present in the four incisor teeth of
24    the younger animals and 0.6% was present in the same teeth of the older animals (Momcilovic
25    and Kostial, 1974). These percentages jumped to 1.43% and 0.88%, respectively, 192 hours
26    after the injection, suggesting incorporation and retention of Pb by teeth is greater in younger
27    animals than in adults,  as found in bone.  Lead has also been shown to be incorporated into
28    incisors of rats exposed to airborne Pb. Grobler and coworkers (1991) exposed 6 week old rats
29    to either "Clean Air" (0.05 jig Pb/m3) or air containing 77 jig Pb/m3 and found significant
30    differences in the amount of Pb incorporated into the incisors of the animals. After 70 days, a
31    mean of only 0.8 jig Pb/g of incisor dry mass was found in incisors from control  animals, while

      December 2005                            5-208       DRAFT-DO NOT QUOTE OR CITE

-------
 1    11.0 jig Pb/g was present in incisors from the 77 jig Pb/m3 group. Exposure to air containing
 2    249 jig Pb/m3 for 28 days or to 1,546 jig Pb/m3 for 50 days resulted in mean values of 13.8 and
 3    153 jig Pb/g incisor dry weight of Pb incorporation, respectively, highlighting the fact that dose
 4    and length of exposure are determinates of amount of Pb contained in the teeth of these animals.
 5    Lead has also been shown to be taken up into the teeth of weanling rats whose mothers were
 6    exposed to Pb in drinking water.  The offspring of pregnant rats exposed during gestation and
 7    lactation until 21 days post partum to water containing 0, 3, or 10 ppm Pb showed  dose-
 8    dependent, significant increases in the Pb content of incisors, first molars, and second molars
 9    (Grobler et al., 1985). Taken together, these studies confirm the uptake of Pb into  teeth as
10    delivered by various means and suggest that maternal exposure can result in uptake in offspring,
11    during gestation and/or lactation.
12
13    5.8.10  Effects of Lead on Enamel and Dentine Formation
14          Early microscopic studies by Eisenmann and Yaeger (1969) confirmed alterations in rat
15    incisor enamel formation 7 days after a single SC dose of Pb (0.15 or 1.5 mM/lOOg animal
16    weight); however, no effect was seen at the 0.075 mM/lOOg dose. Lead was found to have
17    inhibited mineralization of both enamel and dentin, but only to a "mild to moderate"  extent with
18    the mineralization of dentin more affected. It was speculated at the time that Pb could affect the
19    production of normal, mineralizable organic matrix; affect enzymes specific to enamel or dentin
20    formation; affect crystal structure and/or growth; or affect a combination of these factors.  In
21    studies of dentinogenesis, incubation of fixed rat molar germs with Pb-pyrophosphate has shown
22    localization of Pb to the mineralization front of dentin (i.e., the area of recently formed dentin),
23    to the stratum intermedium, and to subodontoblastic cells, suggesting Pb may react with mineral
24    components located in the mineralization zone or have a high affinity for these incompletely
25    mineralized areas (Larsson and Helander, 1974). Localization of Pb was also seen at the area of
26    the dentino-enamel junction. Similar examination of first molar germs from 3-day-old rats
27    showed that Pb also localized to the periphery of dentinal globules (Larsson, 1974).  A single
28    injection of Pb-acetate (30 mg/kg body weight) produces an immediate (within 6 h) response in
29    the growing dentin of the rat incisor, leading to the formation of a so-called "Pb line" (Appleton,
30    1991). A transient rise in serum calcium and  phosphorus accompanied the injection, leading to
31    speculation that lead may have been replacing these minerals in the apatite structure.  However,

      December 2005                            5-209      DRAFT-DO NOT QUOTE OR CITE

-------
 1    backscattered electron imaging of the Pb line showed it to be composed of continuous
 2    hypomineralized interglobular dentin with some incomplete fusion of calcospherites resulting in
 3    uneven mineralization, but no localized concentration of Pb was detectable. This is consistent
 4    with Featherstone and co-workers (1981) who reported that Pb incorporation during apatite
 5    synthesis was widely dispersed, rather than concentrated in areas of calcium deficiency. Once
 6    synthesis is complete, however, Pb is capable of entering calcium deficient areas in enamel,
 7    substituting for calcium (Featherstone et al., 1979). This is essentially the process that occurs
 8    during demineralization/remineralization of enamel. Appleton (1991, 1992) suggested that Pb
 9    has a direct effect on odontoblasts, creating a local disturbance of calcium metabolism, a process
10    similar to that described in bone (Pounds et al., 1991). Interestingly,  no ultrastructural changes
11    in ameloblasts from rat pups whose mothers had been drinking water containing Pb was
12    observed.
13          During the normal process of amelogenesis, water and proteins contained within the
14    organic matrix are lost, leaving densely mineralized enamel. The removal of enamel proteins
15    during this phase is facilitated by enamel proteinases, which are believed to degrade the proteins
16    into smaller units capable of diffusing from the matrix. Using crude extracts from scrapings of
17    rat incisor teeth, Gerlach and co-workers (2000a) demonstrated that Pb inhibited these
18    proteinases in vitro at micromolar concentrations.  In rats given drinking water containing Pb at
19    either 0, 34, or 170 mg/L as Pb-acetate for 70 days, increased amounts of proteins were found in
20    enamel matrix from animals exposed to Pb (Gerlach et al., 2002). Moreover, enamel
21    microhardness analysis of upper incisors revealed a significant decrease in microhardness in
22    regions of enamel maturation, but not in areas of fully mature enamel, suggesting Pb exposure
23    mediates a delay in enamel mineralization. In adult rats with incisors trimmed to remove
24    occlusal (biting) contact, a single IP dose of Pb-acetate (40 mg/kg) significantly delayed the
25    continuous eruption of the incisor at all time points between 8 and 28 days after dosing,
26    compared with controls (Gerlach et al., 2000b).  It is of interest that delayed eruption of teeth in
27    children living in areas of heavy metal contamination (Pb and zinc) has been reported previously
28    (Curzon and Bibby, 1970).
29
      December 2005                           5-210        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.8.11  Effects of Lead on Dental Pulp Cells
 2          Hampered by a general lack of cell cultures specifically for teeth, there remains a paucity
 3    of information regarding both the cultures themselves and the effect of Pb upon such cultures. In
 4    a single in vitro study using a human dental pulp cell culture obtained from teeth extracted for
 5    orthodontic purposes, Thaweboon and co-workers (2002) examined the effects of three
 6    concentrations (4.5 x 10"5 M,  4.5 x 10"6 M, 4.5 x 10"7 M) of Pb-glutamate on cell proliferation,
 7    protein production, and osteocalcin secretion.  Under serum free conditions (DMEM only) all
 8    concentrations of Pb significantly increased cell proliferation on day 1, day 3 and day 5 of
 9    exposure, as measured indirectly by mitochondrial dehydrogenase enzyme assay.  In the
10    presence of 2% fetal bovine serum only, the higher concentration of Pb significantly increased
11    protein production, suggesting an influence of serum constituents on cell growth or binding of
12    free Pb in the medium.  Similar results were reported when rat osteosarcoma cells (ROS 17/2.8)
13    were exposed to identical concentrations of Pb over 2-, 4-, and 6-day time points (Sauk et al.,
14    1992). Concentrations of Pb less than 4.5 x 10"5 M concentration did not affect osteosarcoma
15    cell proliferation in the presence of serum, but in the absence of serum 4.5 x 10"7 M Pb increased
16    cell proliferation at day 4, while at day 6, 4.5  x 10"6 M Pb inhibited proliferation. Further testing
17    of human dental pulp cells in  serum-free conditions showed that Pb exposure caused dose-
18    dependent decreases in intracellular protein and procollagen type I production over the 5-day
19    period experimental period (Thaweboon et al., 2002).  Short-term exposure of the cells to Pb
20    significantly decreased osteocalcin production in a dose-dependent manner at 8- and 12-h
21    exposure time points.  These results suggest that Pb is capable of exerting multiple toxic effects
22    on cells derived from human dental pulp.
23
24    5.8.12  Adverse Effects of Lead on  Teeth—Dental Caries
25          In a recent review, Bowen (2001) highlighted 12 epidemiological studies that examined
26    the association between Pb exposure and dental caries (decay), reporting that 8 studies supported
27    the concept that Pb is a caries-promoting element.  Unfortunately, the source and actual exposure
28    to Pb and measurement of prevalence of caries varied greatly, providing less than completely
29    satisfactory evidence in the opinion of the author. There is also a paucity of well-controlled
30    animal studies examining this issue.
      December 2005                           5-211       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In an early study examining the effect of drinking solutions containing various metallic
 2    ions on dental caries in hamsters, Wisotzky and Hein (1958) reported post-eruptive ingestion of
 3    drinking water containing 0.5 mEq of Pb significantly increased caries scores in molar teeth of
 4    males after 84 days, but, perplexingly, not in females after 98 days of exposure. It should be
 5    noted that in animal studies such as these it is routine to maintain the animals on cariogenic or
 6    caries-promoting diets high in fermentable sugars. Clear evidence supporting Pb's role in
 7    enhancing susceptibility to dental caries was reported by Watson and co-workers in 1997.  In
 8    their study, female rats were exposed to Pb in drinking water (34 ppm as Pb-acetate) as young
 9    adults, during pregnancy, and during lactation.  Lead exposure of the subsequent offspring from
10    the dams was, therefore, from transfer of endogenous Pb from dam to pup during gestation and
11    lactation, with no further exposure after weaning. This pre- and perinatal exposure to Pb resulted
12    in a significant, almost 40%, increase in the prevalence of dental caries over control animals.
13    The study was significant for other reasons, as it mimicked the conditions found in many inner
14    cities where young females are exposed to Pb in their environment and later transfer this Pb to
15    their own fetuses during the extensive bone remodeling that occurs during pregnancy and
16    lactation. The mean blood Pb level in the dams upon weaning was 48 |ig/dL, which is not unlike
17    upper levels reported in humans.
18          The mechanisms by which Pb enhances  susceptibility to caries remain uncertain, though
19    clearly altered mineralization and/or incorporation of Pb into enamel as described above could
20    enhance its solubility in acid. Lead also appears in the saliva of rats at about 5% of the whole
21    blood level and at about 61% of the plasma filtrate Pb level (Mobarak and P'an, 1984), providing
22    an avenue for post-eruptive interaction with the exposed enamel in the oral cavity.  Notably,
23    decreased salivary flow has been reported in rats exposed to Pb, and decreased salivary function
24    is known to increase caries risk.  Stimulated parotid function was decreased by nearly 30% in the
25    Pb-exposed offspring in the study by Watson and co-workers (1997), an effect that could have
26    been mediated by the salivary gland requirement of intact parasympathetic and  sympathetic
27    nervous systems for normal development (Schneyer and Hall, 1970) and Pb's known adverse
28    effect on neurotransmitters (Bressler and Goldstein, 1991).  Acute infusion of 4 jig of Pb per min
29    has been reported to  significantly reduce pilocarpine-stimulated salivary secretion in rats over a
30    50-min period (Craan et al., 1984), while 24-day administration of 0.05% Pb-acetate
31    significantly reduced the concentration of protein and calcium in pilocarpine-stimulated rat

      December 2005                           5-212       DRAFT-DO NOT QUOTE OR CITE

-------
 1    submandibular saliva (Abdollahi et al., 1997).  Of potential interest, postnatal exposure of rats to
 2    Pb (10 or 25 ppm in drinking water) and a caries-enhancing diet containing fluoride (sucrose
 3    containing 15 ppm fluoride) was not associated with an increased risk of dental caries,
 4    suggesting that Pb does not interfere with the protective effect of fluoride (Tabchoury et al.,
 5    1999). Clearly though, the effect of Pb exposure on salivary gland function and the mechanism
 6    by which Pb exposure enhances caries risk needs to be further explored.
 7
 8    5.8.13  Lead from Teeth as a Potential Source of Toxicity
 9          Although no studies currently document the contribution of Pb incorporated into teeth as a
10    source of endogenous Pb exposure, the potential exists during the process of exfoliation of the
11    primary dentition. As described above (Section 5.8.9) Pb is avidly incorporated into the
12    developing dentin and enamel components of teeth. Like bone, the uptake and incorporation of
13    Pb into teeth during acute exogenous exposures may be of short-term benefit by limiting the
14    exposure of other, more sensitive tissues, but, unlike bone, teeth do not undergo a gross
15    remodeling process (the continuous, superficial demineralization/remineralization of the exposed
16    tooth surfaces, principally enamel, are assumed here to be insignificant). However, during the
17    exfoliative process, the erupting secondary tooth erodes away the root (composed of cementum
18    and dentin) of the overlying primary tooth along with some surrounding alveolar bone. Any Pb
19    incorporated into these portions of bone and primary tooth would be released by the erosive
20    process, with the potential to produce highly elevated local concentrations of Pb in the proximity
21    of remodeling alveolar bone and developing secondary teeth.  A more modest contribution to
22    circulating blood Pb would be predicted. Animal research in  this  area has been hampered, as
23    most common rodents (i.e., rats, mice) are monophyodonts (have  only one set of teeth).
24    Although monkeys are an acceptable model, it is problematic how release of Pb stored in teeth
25    could be differentiated from that of remodeling skeletal bones formed at a  similar time point,
26    plus the disproportionate size of the skeletal mass compared to the dentition may mask any
27    contribution of Pb mobilized by exfoliation.
28
29    5.8.14  Teeth and Lead Summary
30          As found with bone, Pb  substitutes for calcium and is readily taken up and incorporated
31    into the developing teeth of experimental animals.  Unlike bone, teeth do not undergo

      December 2005                           5-213       DRAFT-DO NOT QUOTE OR CITE

-------
 1    remodeling per se and, with few exceptions, most Pb incorporated into tooth structure remains
 2    essentially in a state of permanent storage.  Administration of high doses of Pb to animals has
 3    demonstrated the formation of a Pb line, visible in both the enamel and dentin and localized to
 4    areas of recently formed tooth structure. Within this Pb line, areas of inhibition of mineralization
 5    are evident in enamel and dentin.  Lead has been shown to decrease cell proliferation and
 6    production of intracellular protein, procollagen type I,  and osteocalcin in human dental pulp cells
 7    in culture.  Studies of Pb exposure in adult rats have reported inhibition of post-eruptive enamel
 8    proteinases, delayed teeth eruption times, and decreased microhardness of surface enamel.
 9    During the process of enamel formation, Pb is apparently widely dispersed when first
10    incorporated into the developing apatite crystal; however, post-formation, Pb is capable of
11    entering and concentrating in calcium-deficient areas within the enamel. Whether Pb
12    incorporation into the enamel surface compromises the integrity and resistance of the surface to
13    dissolution, and ultimately increases risk of dental decay, is unclear.  Numerous epidemiologic
14    studies suggest Pb is a caries-promoting element.  Animal studies (both post-eruptive Pb
15    exposure and pre- and perinatal Pb exposure studies) support this concept, although the exact
16    mechanism of action remains elusive. No animal studies  have examined the role exfoliation of
17    the primary dentition in release of Pb previously stored in tooth structure, though it is likely this
18    process could serve as an additional  source of Pb exposure in childhood.
19
20
21    5.9    EFFECTS OF LEAD ON THE IMMUNE SYSTEM
22          The immune system, along with the neurological system, has emerged as one of the more
23    sensitive targets of Pb-induced toxicity. However, because Pb exposure at low to moderate
24    levels does not produce overt cytotoxicity of immune cells, immune-associated health effects
25    result from misregulation and shifts in functional capacity rather than profound lymphoid
26    deficiencies. As a result, the most sensitive biomarkers of Pb-induced immunotoxicity are those
27    associated with specific functional capacities as opposed to measures of cell enumeration and/or
28    lymphoid organ pathology. This distinguishes Pb from some other types of immunotoxicants.
29    The following sections provide a survey of the reported immune effects resulting from exposure
30    to Pb in humans and animal models. In general, the focus is on those studies that have been
31    reported since the 1986 AQCD (U.S. Environmental Protection Agency, 1986) was prepared and

      December 2005                           5-214        DRAFT-DO NOT QUOTE OR CITE

-------
 1   have altered our understanding of lead-induced immunotoxicity and the associated immune-
 2   related health risks.
 3
 4   5.9.1   Introduction
 5         The comparative development of the immune system in humans and animal models used
 6   for immunotoxicology was reviewed in recent years by Payne and Crooks (2002) and Holsapple
 7   et al. (2003). Pluripotent hematopoietic stem cells arise from uncommitted mesenchymal stem
 8   cells located in the spanochnopleure area near the heart and appear in the yolk sac (Holsapple
 9   et al., 2003). During human gestation, these stem cells first migrate at approximately 5 weeks
10   and produce lymphoid and myeloid stem cells. Lymphoid stem cells  can be found in the liver at
11   approximately 7-8 weeks of gestation.  In the mouse, the hematopoietic stem cells migrate to the
12   liver on gestational day (GD) 10.
13         Migration of stem cells to the thymus occurs in humans about the 9th week of gestation
14   and in mice  on GD 11.  The equivalent migration probably happens in the rat at GD  13 or later.
15   Bone marrow lymphopoieis begins in humans about week 12 of gestation and in mice about GD
16   18. Immune development continues postnatally in humans as well as rodents. During
17   embryonic development, immune maintenance of the pregnancy is important, and Th2
18   development is favored over Thl. Among other things, the capacity of dendritic cells to promote
19   Thl activity is dramatically suppressed in the newborn (Langrish et al., 2002). However, Thl
20   cytokines can be stimulated shortly after birth in humans (Malamitsi-Pichner et  al., 2005).
21   However, it  is clear that, at birth, rodents lag behind in immune development compared with
22   humans (Dietert et al.,  2000; Holsapple et al., 2003).
23         Immune maturation continues in the thymus and bone marrow to give rise to the broad
24   spectrum of myeloid and lymphoid cells that contribute to host defense and tissue homeostasis.
25   The thymus-derived (T) lymphocytes provide regulatory cells facilitating a wide range of
26   acquired immune responses and also produce cytotoxic T lymphocytes capable of attacking
27   tumor and virally infected cells.  Among regulatory T lymphocytes are at least two types of
28   helper populations termed T helper 1 (Thl) and T helper 2 (Th2).  The former regulatory cells
29   promote immune responses helpful against intracellular pathogens, while the latter are important
30   in defense against extracellular pathogens.  However, skewing of the  Thl/Th2 balance too far in
     December 2005                           5-215        DRAFT-DO NOT QUOTE OR CITE

-------
 1    either direction is problematic in terms of health risk. Such skewing is a large factor in the
 2    consideration of lead-induced immunotoxicity (see Sections 5.9.2, 5.9.4, and 5.9.8).
 3          B lymphocytes are named for the Bursa of Fabricius,  an organ important in their
 4    development in avian species. They constitute the other major lymphoid cell type important in
 5    acquired immunity. B lymphocytes produce first membrane-bound and then secreted
 6    immunoglobulins (antibodies) that are a significant part of humoral  immunity.  Different classes
 7    of immunoglobulins produced during class switching, and promoted by different T lymphocyte
 8    cytokines, are tailored to be effective against different types of pathogens (e.g., viruses vs. extra-
 9    cellular parasites). The potential of Pb on B lymphocytes is discussed along with humoral
10    immunity in Section 5.9.3.1.
11          Another lymphoid cell type is the natural killer (NK)  cell. These cells function during
12    innate immunity as a front line defense against tumor cells and virally infected cells. NK cells
13    have the capacity to recognize a limited number of receptors on target cell surfaces, including the
14    loss of Class I (major histocompatibility complex) proteins.  Such self-protein identity loss is
15    usually associated with viral infection of host cells. NK cells also produce cytokines capable of
16    regulating macrophage and T cell activity and, in turn, NK cells can be activated by
17    lymphoid-produced cytokines.  Consideration of the effect of Pb on NK cells is presented in
18    Section 5.9.8.
19          Myelomonocytic cells such as macrophages and polymorphonuclear leukocytes
20    (neutrophils) are also important in innate immunity.  Neutrophils are usually short-lived cells that
21    are capable of leaving the circulation and migrating into tissues.  From there they can
22    phagocytize pathogenic targets, utilize phagolysosomes to destroy bacteria, and secrete
23    significant quantities of reactive oxygen intermediates (ROIs) into the local environment. They
24    are also capable of producing nitric oxide (NO) in most species.  The impact of Pb on neutrophils
25    is presented in Section 5.9.7.
26          Macrophages are much longer-lived and can perform many of the same functions  as
27    neutrophils. Unlike neutrophils, macrophages can inactivate much of the ROIs they produce
28    internally. However, they can produce vast quantities of NO and have major functional roles in
29    tissue homeostasis, lymphoid regulation, and antigen presentation.  In fact, macrophages reside
30    in virtually every tissue, although their morphology can vary widely and their function spectrum
31    can be quite distinct among different specialized organs. Kupffer cells in the liver and alveolar

      December 2005                           5-216       DRAFT-DO NOT QUOTE OR CITE

-------
 1    macrophages in the lungs are two examples of highly specialized forms of macrophages.
 2    Misregulated or misdirected macrophage activity is a major cause of immune-inflicted tissue
 3    damage. When the cells are activated within tissues and chronically overproduce
 4    proinflammatory cytokines such as interleukin-1 (IL-1), tumor necrosis factor-alpha (TNF-a),
 5    and interleukin-6 (IL-6) in addition to NO and ROIs, the result is usually tissue damage and loss
 6    of function, if not ultimately cancer.  A classic example of the potentially destructive role of
 7    misregulated macrophages is found in the case of asbestos-induced pathology of the lung (Holian
 8    etal., 1997; Driscoll, 2000).  Misregulation of macrophages is a major consideration in the case
 9    of Pb (discussed in Section 5.9.6).
10          Dendritic and Langerhans cells are related in lineage to macrophages. These cells are
11    vitally important in antigen trafficking and presentation,  particularly in lymph nodes and skin.
12    However, far too little is known about their potential sensitivity to lead, largely because their
13    isolation, complete phenotypic and functional characterizations,  and inclusion in assessment
14    methodologies are relatively recent developments within immunology.
15          The other major cell types important in a consideration of lead-induced immunotoxicity
16    are basophils, eosinophils, and mast cells. Basophils are involved in various inflammatory
17    reactions, but little is known about the direct effect of Pb on this cell population.  Mast cells are
18    fixed tissue cells surrounding the vasculature in many organs. These cells can  secrete preformed
19    highly vasoactive products with inflammatory potential (e.g., kinins and histamine) in response
20    to cell-surface-initiated signals.  Mast cells can be triggered by numerous signals including both
21    substance P and immunoglobulin E (IgE). Because mast cell-induced inflammation is associated
22    with IgE-mediated allergic reactions, these cells are important in the clinical ramification of lead-
23    induced immunotoxicity. Eosinophils are a granulocytic cell type associated with Th2-driven
24    inflammatory reactions.  They frequently appear in association with allergic reactions and are
25    regulated by numerous lymphoid cells as well as by mast cells. Remarkably little is known
26    about the direct effects of Pb on eosinophil function despite the probable role of these cells in
27    certain allergic manifestations following exposure to lead.
28          Life-stage related differences in immunotoxicological risk have been reviewed by several
29    authors (Barnett 1996; Holladay and Smialowitz, 2000; Dietert et al., 2002; Holladay, 2005), and
30    it seems clear that the vulnerability of the developing immune system to immunotoxic insult is
31    significantly greater than that of the fully-matured and dispersed immune system for the vast

      December 2005                            5-217        DRAFT-DO NOT QUOTE OR CITE

-------
 1    majority of immunotoxicants (Luebke et al., 2005). Dietert et al. (2000) and Holsapple et al.
 2    (2003) have considered the likely existence of critical windows during immune development
 3    when the immune system as a target may have increased sensitivity or increased resistance to
 4    xenobiotic-induced immunotoxic alteration. These windows correspond to different dynamic
 5    stages of functional development within the embryonic, fetal, and early neonatal immune system.
 6    One issue of immune development particularly pertinent to Pb is the fact that Thl and Th2
 7    functional capacities do not develop at the same time in either humans or rodents.  The need to
 8    protect against maternal-fetal allogeneic reactions results in Thl function being acquired largely
 9    after birth. Therefore, any environmental exposure that interferes with the rapid development of
10    Thl function might leave the individual with a Th2-biased immune system.  The heavy metal Pb
11    apparently represents one of the xenobiotics that are capable of suppressing Thl capacity,
12    resulting in dysregulated immune balance. This is discussed further in Sections 5.9.4, 5.9.8, and
13    5.9.10.
14
15    5.9.2   Host Resistance
16          Host resistance to disease has been used as an effective measure of the impact of
17    environmental toxicants on immune function.  Because different diseases require different
18    combinations  of immune effector functions for host protection, analysis of environmental
19    modulation of host resistance across a spectrum of diseases can help identify clinically relevant
20    immunotoxicity.
21          The 1986 AQCD presented a range of studies in which exposure to Pb inhibited host
22    resistance to disease.  Since the time of that report, few new infectious diseases have been added
23    to the list of those that Pb is known to influence.  Instead, a much broader understanding of the
24    likely basis for the increased disease susceptibility to these pathogens has become evident.
25    Additionally, recognition of an increased risk for some atopic and autoimmune diseases arising
26    from lead-induced immunotoxicity has occurred in recent years.  This is discussed under Section
27    5.9.8.  Lead-induced alterations of host resistance against infectious and neoplastic diseases are
28    considered in the following sections.
29          To date, there has been either no effect or an increased susceptibility to disease resulting
30    from exposure to lead for virtually every infectious agent examined. Given the capacity of Pb to
31    shift immune responses toward Th2, one might expect that enhanced resistance might occur for

      December 2005                            5-218       DRAFT-DO NOT QUOTE OR CITE

-------
 1    diseases where robust Th2 responses were required. For example, an increased resistance
 2    against helminth parasitic disease might be hypothesized. However, this possible association has
 3    not been widely examined to date.
 4
 5    5.9.2.1  Viral Diseases
 6          In general, exposure to Pb increases the susceptibility to viral infections. Studies include
 7    host resistance directed against the encephalomyocarditis virus (Gainer, 1977; Exon et al, 1979),
 8    Langat virus (Thind and Khan, 1978), and Semliki Forrest virus (Gupta et al., 2002). In the last
 9    example, oral dosing of Swiss mice with Pb-acetate (250 mg/kg for 28  days) significantly
10    increased mortality to sublethal doses of the virus.  Ewers et al. (1982)  reported that occupational
11    exposure to Pb resulted in an increased incidence of influenza cases among workers.  In chickens
12    administered Pb-acetate orally (20 and 40 mg/lOOg body weight) for 56 days, antibody
13    production against Newcastle virus vaccine was reduced, while mortality against viral challenge
14    was increased (Youssef et al., 1996).  It seems likely that the reduced Thl capacity (including
15    effective CTL generation) combined with increased TNF-a, ROI, and prostaglandin E2 (PGE2)
16    production by responding macrophages would contribute to increased tissue pathology but
17    reduce viral clearance for many infections.
18
19    5.9.2.2  Bacterial Diseases
20          Most of the lead-associated host resistance research has been conducted on bacterial
21    diseases.  Hemphill et al. (1971) first described the increased susceptibility of mice exposed to
22    Pb (250 jig given i.p. for 30 days) to Samonella typhimurium, while Selye et al. (1966) reported
23    increased susceptibility of rats to bacteria endotoxins.  Cook et al. (1975) found increased
24    susceptibility of lead-exposed rats (2 mg/lOOg body weight given i.v. once) to both Eschrichia
25    coli and Staphylococcus epidermidis.
26          The vast  majority of studies have been conducted using the intracellular bacterium,
27    Listeria monocytogenes,  in mice. Listeria infection and host resistance to the disease have been
28    well characterized.  Essentially, this infection requires an effective antigen presentation
29    (probably involving toll-like receptor 2 involvement), a robust response by activated
30    macrophages leading to interlukin-12 (IL-12) and interferon-y (IFN-y) production and robust Thl
31    driven host protection (Torres et al., 2004; Lara-Tejero and Pamer, 2004). Ideally, activated

      December 2005                            5-219       DRAFT-DO NOT QUOTE OR CITE

-------
 1    macrophages would produce NO in an effective response against Listeria (Ito et al., 2005).  In
 2    the case of lead-induced immunotoxicity, everything works against this type of response. First,
 3    macrophages have severely suppressed NO production. Yet, overproduction of TNF-a, ROIs
 4    and PGE2 leads to tissue inflammation and damage. The skewing of the response toward Th2
 5    means that both IL-12 and IFN-y are lacking. Excessive production of IL-6 and other pro-
 6    inflammatory cytokines results in what has been termed "sickness behavior" which involves both
 7    the immune and central nervous systems (Dantzer et al., 1998; Dyatlov et al., 1998a,b; Lawrence
 8    and Kim, 2000; Dyatlov and Lawrence, 2002). Lead-induced impairment in host resistance to
 9    listeria was reported by Lawrence (1981). CBA/J mice exposed orally to 80 ppm or greater of
10    Pb-acetate for 4 weeks had 100% mortality (after 10 days)  compared with no mortality for mice
11    exposed to 0 or 16 ppm lead.
12          In an important study concerning individual variation to lead-induced immunotoxicity and
13    host resistance, Kim and Lawrence (2000) demonstrated that neurological circuitry as it pertains
14    to brain lateralized behavior could impact the effect of Pb on immune responses and host
15    resistance to Listeria.  Not surprisingly, this suggests that host genotype and epigenetic factors
16    can be influenced by Pb exposure to the individual. Using  female BALB/c mice, Kishikawa and
17    Lawrence (1997)  demonstrated that exogenously administered recombinant IL-12 (1 jig each for
18    three days i.p.) could enhance production of IFN-y as well as host resistance to Listeria in lead-
19    exposed (2 mM in water for 3 weeks) mice.  However, lead-exposed mice continued to have
20    excess IL-6 production (part of the sickness behavior phenotype). The result with IL-12
21    validates the importance of the Th skewing and macrophage impairment induced by Pb on host
22    resistance to certain diseases.
23          Additional bacterial infections in which Pb exposure has been reported to reduce host
24    resistance include Serratia marcesens (Schlipkopter and Frieler, 1979) and Pasteurella
25    multocida (Bouley et al., 1977).
26
27    5.9.2.3 Parasitic Diseases
28          Few studies have been conducted to date regarding the effects of Pb on host resistance to
29    parasitic diseases.  This is unfortunate as some parasitic disease challenges require effective Th2
30    responses for optimal resistance.  Hence, it is not clear that Pb exposure would depress host
31    resistance in every case (e.g., for helminth parasites). Since the AQCD in 1986, one study was

      December 2005                            5-220       DRAFT-DO NOT QUOTE OR CITE

-------
 1   conducted examining the effect of Pb on the killing ability ofLeishmania enriettii parasites in
 2   vitro by mouse macrophages (Mauel et al., 1989). The authors found that 30-100 mM Pb-
 3   acetate interfered with the killing ability of macrophages without producing macrophage
 4   cytotoxicity.
 5
 6   5.9.2.4  Tumors
 7         The primary study concerning tumor immunity/tumor growth and Pb was already known
 8   at the time of the 1986 AQCD. In this study, male C57B1/6 mice were exposed to Pb-acetate in
 9   the drinking water at concentrations of 0, 13, 130, or 1300 ppm. Moloney sarcoma virus (MSV)-
10   induced tumor formation and growth were compared following the exposure of mice to Pb for
11   10-12 weeks. MSV-induced transplantable tumors were also used in this study.  Primary tumor
12   growth was enhanced in animals that received 130 and 1300 ppm of Pb vs. the control.
13   However, all tumors regressed eventually. Most other studies involving Pb exposure and tumors
14   describe the fact that Pb can exacerbate the ability of other toxins to promote tumor formation
15   (Kobayashi and Okamoto, 1974; Hinton et al., 1979).  Much of the tumor-promoting activity of
16   Pb would seem to involve depressed Thl and macrophage function as well as the promotion of
17   excessive ROI release into tissues.
18
19   5.9.3   Humoral Immunity
20         The irony of Pb as an immunotoxicant is that the overall effects on humoral immunity are
21   reasonably modest compared to those reported for macrophages and T lymphocytes (McCabe
22   1994). McCabe et al. (1991) discussed the fact that Pb is not profoundly cytotoxic for most
23   immune cells yet can cause major functional shifts within the immune system as well as
24   decreased host resistance to disease. In many cases, antibody production can remain robust in
25   lead-exposed animals and humans.  However, the nature and spectrum of the antibodies
26   produced is the more  significant cause for concern.  Lead appears to alter the course of T
27   lymphocyte-driven B cell maturation such that class switching may be skewed in lead-exposed
28   animals and humans.  If Pb dosage and duration of exposure is sufficient, antibody production
29   may be depressed overall. However, with low-level Pb exposure, skewed isotype production is
30   the greater health risk.
31

     December 2005                          5-221       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.9.3.1   General Effects on B lymphocytes and Immunoglobulins
 2         Despite the fact that T lymphocytes and macrophages appear to be the more sensitive
 3    targets of lead, the metal can alter B lymphocyte maturation and shift immunoglobulin
 4    production. The 1986 AQCD describes the fact that some early studies reported no effect of Pb
 5    on antibody production (Reigart and Graber, 1976; Ewers et al., 1982), while others reported a
 6    significant decrease in the humoral immune response (Roller, 1973; Roller and Rovaic, 1974;
 7    Blakley et al., 1980). In retrospect, this apparent discrepancy may have been caused by the
 8    various concentrations of Pb administered as well as variations in the duration of exposure.
 9    Additionally, as mentioned in the 1986 AQCD, the temporal relationship of Pb exposure to
10    antigen challenge may be important.
11         In studies measuring generation of plaque forming cells (PFCs) against sheep red blood
12    cells (SRBCs), Pb incubation with lymphocytes in vitro caused an increased response (Lawrence,
13    1981). In a comprehensive  study using several strains of mice, Mudzinski et al. (1986) reported
14    that Pb-acetate administered in the drinking water (10 mM for 8 weeks) elevated the response in
15    the one strain (BALB/c mice) but failed to alter the humoral response to SRBCs (either PFCs or
16    antibody liters) in all other strains. McCabe et al. (1990) reported that Pb caused an elevation in
17    B cell expression of Class II molecules, thereby influencing B cell differentiation.  Lead seemed
18    to impact Class II molecule density at the cell surface via the levels of mRNA translational
19    and/or the posttranslational  stages of cell surface protein synthesis (McCabe et al.,  1991).
20         Some human epidemiological and occupational studies have reported lead-associated
21    differences in levels of circulating immunoglobulins. However, Tryphonas (2001) discussed the
22    pitfalls of relying on total serum immunoglobulin in assessing immunotoxic effects in humans.
23    Sun et al. (2003) reported that immunoglobulin M (IgM) and immunoglobulin G (IgG) were
24    lower but that IgE was higher among females within their high-Pb group.  Basaran and Undeger
25    (2000) found that IgM, IgG, and some complement proteins were reduced among battery
26    workers with high Pb exposure. Results of Undeger et al. (1996) were similar as well.  In
27    contrast, Sarasua et al. (2000) reported an elevation in immunoglobulin A (IgA), IgG, and IgM
28    associated with environmental Pb exposure.  Pinkerton et al. (1998) found no major effects but
29    reported a significant lead-associated decline in serum IgG and an elevation in B cell percentage.
30    In a human in vitro study, Borella and Giardino (1991) showed that Pb exposure caused an
31    increased IgG production following stimulation of cells with pokeweed mitogen.

      December 2005                           5-222        DRAFT-DO NOT QUOTE OR CITE

-------
 1          In more recent animal studies, Miller et al. (1998) and Chen et al. (1999) reported no
 2    effect on antigen-specific IgG liters against keyhole limpet hemocyanin (KLH) protein in F344
 3    strain rats that had been exposed in utero to Pb (0-500 ppm Pb-acetate in drinking water).
 4          It seems likely that Pb exposure may be capable of reducing serum immunoglobulin levels
 5    given sufficient dose and duration of exposure.  However, the more critical issue pertains to the
 6    distribution of class and subclass of immunoglobulins produced after exposure to lead. Because
 7    Pb can alter the development of T cells involved in specific antigen responses, this can impact
 8    the spectrum of immunoglobulins produced in response to T-dependent antigens.  As discussed
 9    in the following section, production of IgE (a class of immunoglobulin that is poorly represented
10    in serum but of great clinical significance) is a central issue for lead-induced immunotoxicity.
11    One additional health concern is the potential for Pb to  enhance the likelihood of autoantibody
12    production (Lawrence and McCabe, 2002; Hudson et al., 2003). This latter concern is discussed
13    under Section 5.9.8.
14
15    5.9.3.2  IgE Alterations
16          One of the three predominant hallmarks of lead-induced immunotoxicity is an increase in
17    IgE production.  This can occur in the context of antigen-specific responses or as measured by
18    total  serum IgE. For this endpoint, the human and animal findings are very similar.  Virtually all
19    of the information concerning the capacity of Pb to elevate IgE production in humans and
20    animals has been obtained since the 1986 AQCD. As a result, this represents a relatively new
21    biomarker for lead-induced immunomodulation, and one not included in most animal or human
22    studies conducted prior to 1990 (e.g., Wagerova et al., 1986).
23          Table 5-9.1 lists the studies reporting lead-induced elevation of IgE. The disease
24    implications of lead-induced increases in IgE production are potentially significant and may help
25    to address, in part, the allergy epidemic that has occurred in the last several decades (Isolauri
26    et al., 2004). A relationship has been established between relative Th2 cytokine levels, serum
27    IgE levels, and the risk of allergic airway inflammation (Maezawa et al., 2004; Cardinale et al.,
28    2005). In fact, attempts to manage allergic inflammation use IgE as one of the major targets
29    (Stokes and Casale, 2005).  IgE levels are directly related to the production of Th2 cytokines
30    such  as interlukin-4 (IL-4), among others (Tepper et al., 1990; Burstein et al., 1991; Carballido
31    et al., 1995; Takeno et al., 2004; Wood et al., 2004).  The relationship between Th2 cytokines

      December 2005                           5-223       DRAFT-DO NOT QUOTE OR CITE

-------
                Table 5-9.1. Recent Studies Reporting Lead-Induced Increase in IgE
Species
Human
Human
Human
Mouse
Human
Rat
Mouse
Human
Strain/Gender
Both genders
Both genders,
91% males
Females
Balb/c males
and females
Both genders,
56% male
F344 females
Balb/c females
Males
Age
Children
Adult
Children
Fetal
Juvenile
Embryo -
fetal
Adult
Adult
In vivo
Ex vivo
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Lowest
Effective Dose
Not available
Not Available
Not Available
0.1 mM
Not Available
100 ppm
50 |jg 3x per
week s.c.
Not Available
Exposure
Duration
Not Available
Not Available
Not Available
3 days
Not Available
5 weeks to dam (2
and 3 gestational)
3 weeks
Not Available
Reference
Karmous et
al. (2005)
Heo et al.
(2004)
Sun et al.
(2003)
Snyder et al.
(2000)
Lutz et al.
(1999)
Miller et al.
(1998)
Heo et al.
(1996)
Horiguchi et
al. (1992)
 1   (e.g., IL-4), IgE levels, and allergic airway disease is supported through various pharmacological
 2   interventions in both animals and humans that either induce Th2 cytokine and promote allergic
 3   airway disease (Wu et al., 2004) or interfere with Th2 cytokine-driven IgE production and inhibit
 4   allergic inflammation (Holgate et al., 2005; Ban and Hettich, 2005).  The production of IgE is of
 5   importance in terms of potential  inflammation. Not only is the level  of IgE a consideration, but
 6   also the expression of the Fc receptor for the epsilon (e) chain of IgE on mast cells and basophils.
 7         In humans, Karmaus et al. (2005) reported a positive association of blood Pb levels with
 8   serum IgE concentration among  second grade children living near a waste incinerator or other
 9   lead-emitting industries. Sun et al. (2003) also found a positive association of blood lead and
10   serum IgE levels among children in Taiwan.  Lutz  et al. (1999) reported a correlation of blood
11   lead levels and serum IgE levels in children in Missouri from 9 months-6 years of age. This
12   association appears to hold not only for children but also for adults. Heo et al. (2004) recently
13   showed that battery workers with blood leads > 30 |ig/dL differed significantly in serum IgE
14   levels from those with blood leads < 30 |ig/dL. Additionally, serum IgE concentration correlated
15   with blood lead among the populations examined (r = 0.0872).
     December 2005
5-224
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Animal data support this relationship between blood lead concentration and IgE level and
 2    further suggest that even very low-level Pb exposure early in development may produce elevated
 3    IgE production in the juvenile offspring. Miller et al. (1998) found that gestational exposure of
 4    rats to 100 ppm Pb-acetate in the drinking water could produce elevated IgE in the adult
 5    offspring.  Snyder et al. (2000) showed that gestational and/or neonatal exposure of mice to Pb-
 6    acetate produced neonatal blood lead levels not above background (5.0 |ig/dL), but nevertheless,
 7    could result in elevated IgE production in the juvenile mouse. In most cases, Pb exposures
 8    associated with elevated IgE were also associated with increases in IL-4 production by T cells
 9    (Chen et al., 1999; Snyder et al., 2000). This is consistent with the fact that high IL-4 production
10    can predispose B lymphocytes to undergo a specific class switch for the production of IgE.
11          One importance of these findings is that, in each case, the elevation in IgE persisted long
12    after blood Pb levels had returned to normal.  This means that Pb exposure could occur early in
13    life and produce an increased risk of later-life allergic disease with no residual evidence that the
14    individual had ever been exposed to lead. This should provide a cautionary note for future
15    human studies examining Pb body burden and immune function.
16
17    5.9.4   Cell-Mediated Immunity
18          Cell-mediated immunity (CMI) essentially involves all host resistance beyond the soluble
19    components of defense, i.e., antibody and complement.  CMI includes any action of the immune
20    system that is a direct effect of leukocytes on neoplastic or virally-infected cells or against
21    extracellular targets such as bacteria.  Even macrophage functional processes involving
22    antibodies, such as antibody-dependent cellular cytotoxicity (ADCC), are considered to be CMI.
23    One of the hallmarks of CMI is that cellular activation is usually required for the effector cells to
24    attack the target. In the case of macrophages, this is usually activation with the Thl-associated
25    cytokine, IFN-y.
26          For NK cells, activation can occur through various pathogenic components such as  double
27    stranded RNA. However, recently Borg et al. (2004) showed that mature dendritic cells
28    produced a Thl-promoting cytokine, IL-12, and this in turn activates NK cells to produce the
29    further Thl-promoting cytokine, IFN-y. Interleukin-18 (IL-18) produced by macrophages is also
30    an activator of NK cells, facilitating Thl-promoting cytokine release while interleukins-2 and
31    -15 (IL-2, IL-15) are growth factors for NK cells. NK cells would appear to be relatively

      December 2005                           5-225        DRAFT-DO NOT QUOTE OR CITE

-------
 1    resistant to the effects of Pb compared to some T lymphocytes and macrophages.  For a detailed
 2    consideration of the effects of Pb on NK cells, see Section 5.9.7.
 3          Cytotoxic T lymphocytes are generated in response to antigen presentation delivered with
 4    Thl cytokines. These cells are capable of mediating antigen specific destruction of neoplastic
 5    and virally-infected cells via binding and release of cytolytic proteins into the intracellular space.
 6    Frequently, the most effective antigen targets of CTLs are the early viral proteins produced in the
 7    first phase of host  cell infection by viruses. IL-12, produced largely by dendritic cells, appears to
 8    be important in the generation of antigen CTL cells and IFN-y produced by Thl lymphocytes.
 9    NK cells are a potent regulator of CTL activity. Cell signaling via certain toll-like receptors on
10    antigen presenting cells seems to have a role in determining the nature  of the Th activation (Thl
11    vs. Th2) and can, therefore, influence the extent of CTL production.
12          Because T lymphocytes and their regulator and effector functions are so critical in CMI,
13    the maturation of thymocytes within the thymus microenvironment and the selection of
14    repertoire among the maturing T lymphocytes are crucial issues for potential developmental
15    immunotoxicants.  In fact, lead seems to be capable of disrupting several aspects of T cell
16    maturation, activation, and repertoire usage (McCabe and Lawrence, 1991; Heo et al., 1998;
17    Miller et al.,  1998; McCabe et al., 2001, Lee and Dietert,  2003).
18
19    5.9.4.1  General Effects on Thymocytes and T lymphocytes
20          In general, cells of the T cell lineage appear to be relatively sensitive to the toxic effects
21    of Pb compared to other lymphoid populations. At the time of the 1986 AQCD, there was some
22    understanding of this sensitivity. However, there appear to be considerable differences in
23    sensitivity across various T cell subpopulations (McCabe and Lawrence, 1991; Heo et al., 1996;
24    1997;  1998).  This was largely unknown when the prior AQCD was prepared as the partitioning
25    of T helper cells into functionally distinct subpopulations (e.g., ThO, Thl,  and  Th2) was not
26    known until the latter part of the 1980s. The differential impact of Pb on T helper cell
27    populations and on immune balance was established during the 1990s.  This has become one of
28    the four hallmarks of lead-induced immunotoxicity.
29          Original observations of both in vivo and in vitro T-dependent immune responses in the
30    presence of Pb suggest that T helper function, as well as the spectrum of cytokines produced, are
31    skewed toward the Th2. The cytokine skewing is discussed as well in  Section 5.9.5.3. Smith

      December 2005                           5-226       DRAFT-DO NOT QUOTE OR CITE

-------
 1    and Lawrence (1988) have shown that Pb can inhibit antigen presentation and stimulation of a
 2    T cell clone of the Thl phenotype. McCabe and Lawrence (1991) were the first to show that this
 3    was caused by the novel capacity of Pb to inhibit Thl stimulation while promoting presentation
 4    to Th2 clones. Heo et al. (1996) provided both in vitro and in vivo results supporting this
 5    immunomodulation of lead. Cytokine skewing accompanied the differential stimulation of Th
 6    cells.
 7         Using naive splenic CD4+ T cells derived from Dl 1.10 ovalbumin-transgenic mice, Heo
 8    et al. (1998) developed T cell clones in vitro in the presence of lead.  The authors found the
 9    T cells that developed from the naive precursors were significantly skewed toward the Th2
10    helper phenotype and away from the Thl phenotype. If IL-4 was inhibited with the addition of
11    anti-IL-4 to the cultures or if the Thl- promoting cytokine IL-12 was added exogenously to the
12    culture, the effects of Pb could be largely overcome.  This study provided firm evidence that Pb
13    can directly promote Th2 development among precursor Th(0) cells and impair development of
14    Thl cells. Among its effects, Pb enhanced adenyl cyclase activity and increased the levels of
15    cAMP.  The authors suggested that Pb may influence cell signaling in such as manner as to
16    promote the Th2 pathway.
17         Beyond the biasing  of immune responses at the level of the T lymphocyte based on
18    Thl/Th2 balance, Pb has the capacity to bias usage of certain VP genes (V(3 5, V(3 7, and VP 13)
19    among T lymphocyte clones in mice (Heo et al., 1997). This is of concern, as it suggests that
20    exposure to Pb may alter the T cell repertoire and skew its representation. Heo et al. (1997)
21    discussed the fact that many autoimmune diseases are characterized by a disproportionate usage
22    of certain VP genes. Different autoimmune conditions are associated with the differential
23    overabundant usage of a specific VP gene. They suggest that this feature of lead-induced
24    T lymphocyte immunotoxicity may contribute to and enhance the risk of autoimmunity.
25         Lee and Dietert (2003)  exposed the developing thymus of embryonic day 12 (E12)
26    chickens to Pb-acetate (single  injection of 400 jig) and evaluated the capacity of thymocytes (ex
27    vivo) from juvenile chickens to produce IFN-y. They found that embryonic exposure at doses
28    that impair juvenile delayed type hypersensitivity (DTK) also inhibit IFN-y production.
29    Similarly, IFN-y production was decreased when thymocytes from juvenile chickens were
30    exposed to Pb in vitro (0.45 |iM).  However, in vitro exposure of thymic stroma to Pb did not
31    result in suppression of control thymocyte IFN-y production in co-cultures.  There is a suggestion

      December 2005                           5-227       DRAFT-DO NOT QUOTE OR CITE

-------
 1   that the balance of reproductive hormones in early life may influence the impact of Pb on
 2   developing thymocytes (Hussain et al., 2005).
 3
 4   5.9.4.2  Delayed Type Hypersensitivity
 5          The DTH assay is an in vivo assay requiring antigen-specific T lymphocytes to be primed,
 6   expanded, and then recruited to a local site of antigen deposition. The most common application
 7   of the DTH is the tuberculin assay for TB in humans. The assay has a long history of application
 8   in immunotoxicology, and its utility within the national toxicology program assessment in the
 9   mouse has been previously reported (Luster et al., 1992). The assay is known to depend largely
10   on Thl participation and is, therefore, an effective measure of Thl-dependent function.
11   However, there are at least two different portions of the response that are under somewhat
12   separate control. Priming and expansion of the antigen-specific T lymphocytes is largely Thl
13   dependent. However, recruitment of T lymphocytes to the periphery involves a variety of
14   locally-produced chemotactic signals that may not be under the  same regulation. In fact, Chen
15   et al.  (1999) showed that a commonly used chelator for Pb poisoning (succimer,
16   meso-2,3-dimercaptosuccinic acid [DMSA]) fails to restore lead-induced suppression of DTH in
17   rats, because the chelator itself somehow interferes with the production of chemotactic factors
18   necessary for T lymphocyte recruitment.  The DTH assay is also generally useful in questions of
19   possible developmental immunotoxicity, because of the natural  skewing toward Th2 that occurs
20   during gestation through birth and the issue of effective Thl functional acquisition in the
21   newborn.
22          Lead-induced suppression of the DTH response is one  of the four hallmarks of lead-
23   induced immunotoxicity.  At the time of the 1986 AQCD, the capacity of Pb to suppress DTH
24   function was already known from two studies conducted during the late 1970s.  However, the
25   association of the function with Thl help had not been established.  Muller et al. (1977) were
26   among the first to demonstrate lead-induced suppression of DTH. Using mice, these
27   investigators administered Pb-acetate i.p. for 30 days prior to  assessment of primary and
28   secondary DTH responses against  SRBCs. Both primary and secondary responses were severely
29   depressed following exposure to Pb, even at the lowest dose tested (0.025 mg).  Faith et al.
30   (1979) exposed developing Sprague-Dawley rats to Pb-acetate in the drinking water (lowest dose
31   at 25 ppm) first via the dams during gestation and through weaning and then with direct exposure

     December 2005                           5-228        DRAFT-DO NOT QUOTE OR CITE

-------
 1    of the offspring until 6 weeks of age. In this case, the purified protein derivative (PPD) of
 2    tuberculin was used as the antigen compared against the saline injection control. Rats
 3    administered the lowest dose of Pb evaluated (producing a BLL of 29.3  |ig/dL) had a
 4    significantly reduced DTH response. Laschi-Loquerie et al. (1984) measured the contact
 5    hypersensitivity reaction against picryl chloride in mice that had received 0.5 mg/Kg Pb via s.c.
 6    administration.  Lead administration was given from 3-6 days in duration at varying times
 7    relative to the sensitization period. These investigators reported that Pb suppressed the DTH
 8    type of response regardless of the window (before or during sensitization) in which it had been
 9    administered.
10          More recently, Miller et al. (1998) found that female F344 rats gestationally exposed to
11    250 ppm of Pb-acetate in drinking water had a persistently reduced DTH reaction against KLH
12    protein. Chen et al. (1999), Bunn et al. (2001a,b,c)  and Chen et al. (2004) had similar findings in
13    studies that included both the F344 and CD  strains of rats.  In the last study conducted in F344
14    rats, a BLL of 6.75 |ig/dL at 4 weeks of age, postgestational exposure to Pb-acetate (250 ppm in
15    drinking water) was associated with depressed DTH against KLH in the 13-week-old adult
16    female offspring (Chen et al., 2004). McCabe et al. (1999) were among the first to draw
17    attention to the relationship between lead-induced suppression of DTH and the prior observations
18    of lead-induced Th skewing. These authors gave varying doses of Pb-acetate in drinking water
19    (32,128, 512, 2048 ppm) to female BALB/c mice for 3 weeks prior to measuring the DTH
20    against SRBCs.  They found that the 512 ppm dose  producing a BLL of 87 |ig/dL significantly
21    impaired the DTH response. Antigen routes proved to be important as Pb depressed DTH when
22    an i.v.  primed with SRBCs was used, but not when  SRBCs were administered i.p.  Timing of Pb
23    administration was found to be important relative to the capacity to depress the DTH response.
24    Lee et al. (2001) showed that Pb-acetate (200 jig) administered in ovo to chicken embryos at
25    9 days of incubation failed to depress juvenile DTH against bovine serum albumin (BSA), but
26    when the same dose of Pb was administered 3 days  later producing the same BLL, juvenile DTH
27    was severely reduced.  Using the latter model, embryonic administration of exogenous thymulin
28    was found to partially restore juvenile DTH function following embryonic exposure to Pb (Lee
29    and Dietert, 2003).
30          Regarding developmental sensitivity  of the DTH response to lead-induced
31    immunosuppression, parallel findings were obtained in the developing rat (CD strain females)

      December 2005                           5-229       DRAFT-DO NOT QUOTE OR CITE

-------
 1   (Bunn et al., 2001c) in agreement with those found in the chicken. Administration of 500 ppm
 2   Pb-acetate during gestational days 3 to 9 or 15 to 21 produced no DTK effect compared with
 3   DTH suppression in the corresponding adult offspring.  As shown in Figure 5-9.1, the sensitivity
 4   of the DTH response to Pb appears to develop sometime between days 9 and 15 of rat embryonic
 5   development. Apparently, the status of the developing thymus may be a consideration in the
 6   capacity of Pb to impact the subsequent DTH response.  This is discussed further in
 7   Section 5.9.10.
Rat
Conception
0369 !f2- 15 18
13-14 17
t
Migration of
T precursors
to ttoyrmis

Chicken
Start of
1

Appeal' ance
of double

incubation
0 3 6 9- 32 15 18
Day of Embryonic Development
First wave T Second wave T
precursors to precursors to
thymus thymus



Birth

f*
20
11
1 post
t birth
I
Appearance I
of fully 1
mature |
thymocytes i^F






T
Emigration to
periphery

Hatch


Third wave T
precursors to
thymus
f«



11



     Figure 5-9.1.  Windows during prenatal development (days postconception for rat) or
                   embryonic development (days postincubation initiation for chicken) during
                   which sensitivity of DTH to lead emerges.
 8         It should be noted that in several studies, lead-induced suppression of the DTH response
 9   was associated with reduced capacity to produce the Thl cytokine, IFN-y (Chen et al., 1999; Lee
10   etal., 2001).
     December 2005
5-230
DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.9.4.3  Other T-Dependent Cell-Mediated Immune Changes
 2          The in vitro response of T lymphocyte populations to various mitogens (e.g.,
 3    Concavavalin A [ConA], Phytohemagglutinin A [PHA]) has been used as a surrogate measure of
 4    antigen-driven T lymphocyte stimulation.  The impact of Pb on these parameters is presented in
 5    Section 5.9.5. Another T cell response altered by exposure to Pb is the mixed lymphocyte
 6    response (MLR). This in vitro assay is a measure for the responsiveness of T cells to the
 7    presentation of allogeneic major histocompatibility complex (MHC) molecules by antigen
 8    presenting cells. The in vivo correlate of the MLR is usually considered to be the graft vs. host
 9    (GvH) reaction.  Several investigators have reported Pb alteration of the MLR as summarized in
10    Table AX5-9.4.
11          McCabe et al. (2001) demonstrated that Pb at very low physiological concentrations
12    (0.1 jiM or approximately the equivalent of 10 |ig/dL) in vitro significantly enhanced the
13    proliferation and expansion of murine alloreactive CD4+ T lymphocytes in the MLR reaction. In
14    fact, the resulting population was found to have a high density of CD4 molecules on the cell
15    surface-making them phenotypically similar to memory T lymphocytes.  The authors
16    hypothesized that lead-induced creation of an exaggerated pool of memory-type T lymphocytes
17    (possessing a lower threshold required for subsequent activation) would be problematic for the
18    host. In a study using Lewis strain rats, Razani-Boroujerdi et al. (1999) also found evidence for
19    lead-induced stimulation of the in vitro MLR response. In this case, both the alloreactive
20    mixtures of cells as well as syngeneic mixtures were elevated in proliferation when cultured in
21    the presence of Pb-acetate (e.g.,  50 ppm or approximately 131 jiM). When concentrations of Pb
22    were significantly higher (200 ppm or greater), proliferation was inhibited in these cultures.
23          Figure 5-9.1 illustrates the developmental appearance of initial sensitivity for Pb-induced
24    suppression of the  DTH function. The mid-embryonic developmental window is the time during
25    which the capacity of Pb to impair later-life DTH responses first emerges. Earlier pulsed
26    exposure to Pb fails to impair juvenile and/or adult DTH despite the continuing of presence of Pb
27    in the embryo. However, during the second half of embryonic development the embryo becomes
28    remarkably sensitive to lead-induced suppression of DTH.  Both the rat and chicken are similar
29    in this window of emerging Thl-dependent functional sensitivity. Thymus-related
30    developmental events are indicated along with the emergence of DTH functional sensitivity to
     December 2005                           5-231        DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead. Information was derived from Gobel (1996), Vicente et al. (1998), Dunon et al. (1999),
 2    Dietert et al., (2000), Bunn et al. (2001c), Lee et al. (2001) and Holsapple et al. (2003).
 3
 4    5.9.5   Lymphocyte Activation and Responses
 5          Many of the broader functional ramifications of Pb exposure on lymphocytes are
 6    discussed under Sections 5-9.3 and 5-9.4. However, the capacity of Pb to directly alter lymphoid
 7    responses is a significant component of lead-induced immunotoxicity and is summarized within
 8    the present section. Lymphoid responses are usually assessed in terms of proliferation and
 9    activation (functional changes). One of the recent endpoints reflecting functional status is the
10    production of cytokines.  These both autoregulate the producing cells and significantly impact
11    the activity of other immune and nonimmune cells carrying the appropriate receptors.  The
12    spectrum and levels of cytokines produced by a population of immune cells tends to reflect their
13    capacity to regulate the host immune response.
14
15    5.9.5.1  Activation by Mitogens
16          The capacity of certain plant- and bacterially derived products to stimulate lymphoid
17    populations to enter the cell cycle and undergo mitogenesis has been used for decades to assess
18    the potential  capacity of lymphocytes to receive proliferation signals and expand their
19    population.  Among the mitogens  employed within the Pb exposure studies are the T lymphocyte
20    subpopulation mitogens, PHA and Con A; the dual T and B cell mitogen, pokeweed mitogen
21    (PWM; the B lymphocyte mitogen derived from gram-negative bacteria, lipopolysaccharide
22    (LPS), and the B cell mitogen, Staphylococcus aureus enterotoxin (SE). It should be noted that
23    these mitogens do not necessarily  stimulate all T lymphocytes or all lymphocytes but, instead,
24    stimulate selected populations of the cells. The mitogens react with a large array of cell surface
25    molecules producing cross-linking and appropriate signal transduction to initiate mitogenesis.
26    In the case of the plant-derived mitogens, lectins, numerous glycoproteins and glycolipids
27    carrying the correct carbohydrate residues serve as the cell surface binding sites for cross-
28    linking.  Mitogen stimulation in vitro has been used as a surrogate for antigen-driven stimulation
29    and proliferation of antigen-specific T and B cell clones.  However, it should be noted that while
30    the assays have been used for decades, there are now more specific assays utilizing more
31    functionally relevant cell surface receptors to assess lymphoid activation potential.

      December 2005                           5-232       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The 1986 AQCD has an extensive review of mitogenic responses of lymphocytes
 2    following both in vivo and in vitro treatment by lead. The results at that time showed no clear
 3    pattern.  At low to moderate levels, Pb was potentially co-mitogenic for some cells and at very
 4    high concentrations could suppress proliferation. Little has changed in conclusions for this
 5    assessment measure since the 1986 report.  The most significant findings from the mitogenic
 6    studies are that at doses encountered physiologically Pb is not a potent cytotoxic agent for most
 7    immune cells. At low concentrations, it can marginally stimulate lymphoid mitogenesis.
 8    However, as one examines more refined subpopulations of lymphocytes than what were able to
 9    be identified prior to 1986 (e.g., Thl vs. Th2 clone of T lymphocytes), it becomes clear that Pb
10    can promote expansion of some lymphoid populations while suppressing others.
11          Annex Table AX5-9.5 for this section summarizes results of Pb effects on mitogen-
12    stimulated proliferation of lymphoid populations.
13
14    5.9.5.2  Activation via Other Receptors
15          In recent years, lymphoid activation and population expansion has been measured using
16    the triggering of specific T and B cell surface receptors (e.g., CD3 on T cells) as well as antigen-
17    driven proliferation of T cell clones known to be specific for the antigen in question. The latter
18    has provided the opportunity to simulate in vivo lymphoid activation and antigen-driven
19    proliferation by using receptors in vitro, which are more physiologically relevant than those
20    activated by plant lectins. Because Pb does not cause profound population loss across  the entire
21    population of T or B lymphocytes, these more refined and functionally-relevant assay systems
22    have enabled a much clearer picture to emerge concerning lead-induced changes in lymphoid
23    population than was available for the 1986 AQCD report.
24          Smith and Lawrence (1988)  and McCabe and Lawrence (1991) utilized antigen-specific
25    mouse T clones.  They found that Pb directly promoted antigen presentation and stimulation of
26    the T cell clones when these clones were Th2 cells. However, when the Thl clones were used,
27    Pb suppressed the antigen-specific presentation signal.  In the McCabe and Lawrence study,
28    direct comparisons were made between Thl and Th2 clones specific for mouse allogeneic MHC
29    molecules. These studies provided the first clear picture of the differential effects of Pb on Thl
30    vs. Th2 cells.  Several studies since these have verified this major effect of Pb (Heo et  al., 1996;
31    1997, 1998). Many of these later studies utilized the  transgenic mouse strain (DO 11.10 OVA-tg)

      December 2005                            5-233       DRAFT-DO NOT QUOTE OR CITE

-------
 1    that carries T cells specific for a peptide fragment of ovalbumin.  These enabled the same
 2    comparisons to be made with the presentation of a soluble protein antigen as the stimulating
 3    signal.  Heo et al. (1998) showed that Pb not only selectively stimulates Th2 cells and suppresses
 4    Thl cells but that it preferentially causes precursor ThO cells to mature into Th2, rather Thl cells,
 5    as well.  Additionally, the  T cell clones in the presence of Pb are skewed in terms of their usage
 6    of VP genes (as reflected in their cell surface receptors) (Heo et al., 1997).  This is of particular
 7    concern relative to the risk of autoimmunity. More recently, McCabe et al. (2001) examined Pb
 8    exposure in the context of the allogeneic MLR against allogeneic MHC molecules.  In vitro
 9    exposure to Pb (as low as  1.0 jiM) enhanced the primary MLR response, but not the secondary
10    MLR response and not the mitogenic response using PHA.  Significantly, the T cell clones that
11    emerged from the primary MLR were in greater proportion than normal and were of the
12    specialized phenotype CD4-plus high density (CD4+hlgh). Because these fit the phenotype  of
13    memory cells, it is likely that an overabundance of memory cells was produced during the
14    primary response, where the antigen may be of lesser biological significance than in a secondary
15    response. The authors discussed the fact that Pb may cause T cells to respond under conditions
16    of low antigen concentration, which could waste valuable and limited resources by  generating
17    T memory cell clones when they are not needed (against unimportant antigens) or even increase
18    the risk of autoimmune responses by altering the threshold requirements for stimulation.  The
19    putative mechanisms suggested for the differential effects of Pb on Th cells are presented in
20    Section 5.9.9.
21
22    5.9.5.3   Cytokine Production
23         At the time of the 1986 AQCD, immune cytokines were essentially absent from the
24    information available for consideration.  Only the antiviral interferons (a,P) had been examined
25    among studies available for that report.  Therefore, one of the most important effects of Pb  on the
26    immune system, i.e., Pb-induced cytokine production was not known at that time.
27          Most studies since  1986 have shown that Pb exposure at low to moderate levels causes a
28    significant shift in the production of Thl vs. Th2 cytokines with the bias toward the latter.
29    Hence,  production of IFN  is decreased and IL-12 is inadequate for effective host resistance.
30    In contrast, production of IL-4, IL-6, and, frequently, interlukin-10 (IL-10) is elevated.
31    Table 5-9.2 illustrates the  studies reporting shifts in cytokine production induced by lead.

      December 2005                           5-234       DRAFT-DO NOT QUOTE OR CITE

-------
December 2
o
o









to
OJ
^



o
s
n
H
I
o
o
o
^^
H
O
o
H
W
O
^
O
HH
H
in
Table 5-9.2. Studies Reporting Lead-Induced Shifts in Thl vs. Th2 Cytokines
Species
Rat

Human



Chicken


Mouse


Rat

Chicken

Mouse




Mouse


Mouse



Rat


Strain/ Gender
F344 Females

Males



Cornell K females


Balb/c


CD females

Cornell K females

Balb/c male




NOD
Autoimmune strain
adult
C57 Bl/6 females

NOD autoimmune
strain females
F344 females


Age
Embryo-fetal

Adults



Embryonic


Neonatal/
Juvenile

Fetal

Embryonic

Adults




Adult


Adult



Embryo-
fetal

In vivo/
Cytokine Alterations Ex vivo
tIL-4 Yes
4 IFN-y splenic lymphocytes
t IFN-y Yes

PHA stimulated peripheral
blood lymphocytes
4IFN-y Yes

stimulated thymocytes
tJL-6 Yes
serum during infection

tIL-10 Yes

4IFN-y Yes

tIL-6 Yes

serum during infection in
certain groups

4 IFN-y, no change long Yes
term
4TGF-P intestinal levels
No effect on gut balance in Yes
normal mice

4TGF-P in autoimmune mice
4IFN-y Yes
tIL-10

Lead Dose/
Concentration
250 ppm in water
to dams
Not available



400 |^g


0.5 mM in water to
dams and their
pups

550ppminwaterto
dams
50 ng

2mM




Oral 10 mM and
ovalbumin antigen

0.5 mg/kg injection
and oral ovalbumin


250 ppm to dams


Duration of
Exposure
2 weeks prior and 3rd
week of gestation for
dam
Not available



Single injection E12


4 weeks (3 via dams)


6 days via gestation
of dam
Single injection

8 weeks




10 days


6 injections over
2 weeks


2 weeks before and
3rd week of gestation

References
Chen et al.
(2004)
Mishra et al.
(2003)


Lee and
Dietert (2003)

Dyatlov and
Lawrence
(2002)

Bunn et al.
(200 Ic)
Lee et al.
(2001)
Kim and
Lawrence
(2000)


Goebel et al.
(2000)

Goebel et al.
(1999)


Chen et al.
(1999)


-------
o
o
3
cr
^
o
o









u\
1
to
OJ
ON



O
H
I
0
o
0
H
O
o
H
W
O
O
H
W
Table 5-9.2 (cont'd). Studies Reporting


Species Strain/ Gender Age Cytokine Alterations
Mouse DOll.lOova-tg, Adult 4IFN-y
ova mice and RAG
knockouts
Rat F344 females Embryo- 4IFN-y
fetal
Mouse Balb/c By J females Adult 4IFN-y

tIL-6
Mouse Balb/c and Adult 4IFN-y
DOll.lOova-tg
mice 4IFN-y/IL-4
ratio


Mouse Balb/c By J female Adult 4IFN-y
or male |IL_4

Mouse Balb/c ByJ female Adult 4IFN-y
or male ^














Lead-Induced Shifts in Thl vs. Th2 Cytokines

In vivo/ Lead Dose/ Duration of
Ex vivo Concentration Exposure
No 25 |j,M 3 days


Yes 500 ppm to dams 2 weeks before and
3rd week of gestation
Yes 2 mM 3 weeks


Yes 50 |j,g each 2 weeks
injection (s.c.)
3 per week



Yes 50 \ig each 2 weeks
injection (s.c.)
3 per week
No 10 |jM - 50 |jM 2 days


















References
Heo et al.
(1998)

Miller et al.
(1998)
Kishikawa et
al. (1997)

Heo et al.
(1997)



Heo et al.
(1996)

Heo et al.
(1996)















-------
 1    (Please note that TNF-a production is considered in the macrophage section, Section 5.9.6).
 2    These shifts in cytokine production are remarkably consistent, occur even at low levels of
 3    exposure, and are reported following both in vivo and in vitro exposure to lead. Furthermore, the
 4    effects are persistent even when exposure to Pb was restricted to early development and cytokine
 5    assessment was  performed in the subsequent juvenile or adult (Miller et al.,  1998; Bunn et al.,
 6    200Ic; Lee et al., 2001; Chen et al., 2004).
 7          The only  exceptions to lead-induced biasing in favor of Th2 occur in the reports by
 8    Goebel et al. (2000) and Mishra et al. (2003).  In the latter case, the authors  attributed this
 9    difference (in humans) to the very high Pb levels considered in the study. In the prior case,
10    Goebel et al. (2000) saw a local bias to Thl in the intestinal tract of a specialized autoimmune
11    diabetes-prone strain of mice (NOD) but not in normal mice.  Initially, the Pb-induced cytokine
12    skewing favored Th2 (after 1 day), but this shifted to Thl with more prolonged Pb exposure
13    (after 10 days).  Loss of oral tolerance accompanied this long-term shift. These results  suggest
14    that in most cases, lead-induced skewing would favor Th2. But with some genotypes or
15    additional  disease conditions, an imbalance may occur in the direction of a gut-associated Thl
16    environment, increasing risk for loss of oral tolerance and the potential for increased food
17    allergies.
18          One ramification for the capacity of Pb to promote Th2 cells is the impact of elevated
19    IL-4 on IgE.  It seems clear that lead-induced overproduction of IgE (seen in virtually all  animal
20    models examined as well as humans) is directly linked with the overproduction of IL-4.
21    Excessive  IL-4 and the resulting IgE production increases the risk for IgE-mediated atopy and
22    asthma.
23          Additionally, Kishikawa et al (1997) demonstrated that administration of the potent
24    Thl-promoting cytokine, IL-12, to lead-exposed mice can restore the balance of Thl (IFN-y) vs.
25    Th2 cytokines (e.g., IL-6), reduce corticosterone levels, and enhance host resistance in Listeria-
26    infected mice. This observation supports the critical role of Thl/Th2 balance in overall risk to
27    host resistance against disease presented by Pb disruption of that balance.
28
29    5.9.6   Macrophage Function
30          Macrophages represent a diverse population of cells that play critical roles in both  host
31    defense and tissue homeostasis. Macrophage subpopulations provide a front line of defense
      December 2005                            5-237       DRAFT-DO NOT QUOTE OR CITE

-------
 1    against bacteria, parasites, viruses, and tumors via the innate immune response.  Additionally,
 2    they are important in tissue repair and remodeling as well as in the removal of senescent cells.
 3    Some forms of macrophages are efficient in the processing of antigens and the presentation of
 4    antigen fragments to T lymphocytes. Additionally, macrophages can regulate lymphoid activity
 5    through the secretion of a variety of cytokines and through the production of various
 6    immunomodulatory metabolites (e.g., NO, ROIs) and the products of the cyclooxygenase and
 7    lipoxygenase pathways.
 8          Because macrophages can be found residing in most tissues, lead-induced modulation of
 9    macrophage functional capacity has the potential to alter overall organ function.  Macrophages
10    originate in the bone marrow from pluripotent stem cells that give rise to both the monocyte-
11    macrophage lineage as well as polymorphonuclear leukocyte populations. Bone marrow-derived
12    macrophages mature under the influence of various cytokine growth factors to become the full
13    array of mature cell subpopulations. Various investigators have examined the effects of lead on
14    the maturation of macrophages in vitro as well as on the functional capacity on fully mature cells
15    both in vitro and in vivo. Blood monocytes represent a functional, yet not fully  specialized, form
16    of macrophage.  As a result the influence of environmental toxicants on monocytes may not be
17    fully  predictive of the effects of the same toxicants on splenic or alveolar macrophages, glial
18    cells, or Kupffer cells.
19          Because macrophages give rise to several specialized populations, e.g. Kupffer cells in the
20    liver, glial cells in the brain, and various skin macrophage populations, it is important to realize
21    that different specialized macrophage populations are likely to have somewhat different
22    sensitivities to lead, as well as potentially different responses following exposure. Not
23    surprisingly, blood monocytes may not always be an appropriate model to accurately predict the
24    outcome of lead-induced immunotoxicity for alveolar macrophages following an inhalation
25    exposure.
26          The 1986 AQCD identified macrophages as a significant target for lead-induced
27    immunotoxicity. Research since the mid-1980s has  served to underscore this point. The
28    understanding of lead-induced alterations in macrophage function has increased significantly
29    since the prior AQCD report.  The following  sections describe the reported immunotoxic effects
30    of lead on macrophages. It should be noted that for a number of endpoints, such as lead-induced
31    alterations in the production of NO, ROIs and TNF-a, there is a general  consensus among a

      December 2005                           5-238        DRAFT-DO NOT QUOTE OR CITE

-------
 1    majority of immunotoxicology studies and agreement with the effects described for the
 2    cardiovascular system (see Chapter 5.5).
 3
 4    5.9.6.1  Nitric Oxide (NO) Production
 5          Nitric oxide is a short-lived metabolite produced in large quantities by macrophages
 6    during cellular activation. The enzyme responsible is an inducible form of nitric oxide synthase
 7    (iNOS), which, utilizing a bioptrin cofactor, converts the amino acid arginine into NO and
 8    citrulline.  A competing alternative pathway utilizing arginine leads to the production of
 9    polyamines, which themselves are immunomodulatory for lymphocytes.  Nitric oxide is critical
10    in the defense against certain infectious agents, including various bacteria.
11          Among the most sensitive immunomodulatory effects of Pb exposure is the capacity to
12    impair NO production by macrophages (Table AX5-9.6).  Several research groups have shown
13    that in vitro as well as in vivo exposure to Pb results in significantly reduced production of NO
14    (Tian and Lawrence, 1995, 1996; Chen et al., 1997; Lee et al., 2001; Pineda-Zavaleta et al., 2004
15    [also reviewed in Singh et al., 2003]).  Similar results were obtained in human, mouse, rat and
16    chicken. Depression of NO production capacity usually occurs shortly after exposure to lead.
17    However, the long-term effects of Pb on NO production following very early life exposure are
18    less clear (Miller et al., 1998; Chen et al.,  1999; Bunn et al., 2001a).
19          Tian and Lawrence (1996) have hypothesized that because very low Pb concentrations (in
20    vitro equivalents to 10 |ig/dL) can impair NO production, impaired NO production may be
21    responsible for reduced host resistance to Listeria seen among lead-exposed rodents as well as
22    for lead-induced hypertension among humans (Pirkle et al.,  1985). Indeed, impaired NO
23    production by macrophages seems to be one of the more sensitive endpoints for immediate lead-
24    induced immunotoxicity.
25
26    5.9.6.2  Other Functional Alterations
27    TNF-a Production
28          Early studies identified the fact that Pb exposure could predispose animals for a
29    dramatically increased sensitivity to bacterially-derived endotoxin (Trejo et al., 1972; Filkins and
30    Buchanan, 1973;  Schlick and Friedberg, 1981).
      December 2005                           5-239       DRAFT-DO NOT QUOTE OR CITE

-------
 1          It is now known that the increased sensitivity to endotoxin is linked to the capacity of Pb
 2    to increase production of TNF-a among macrophages (Dentener et al., 1989; Zelikoff et al.,
 3    1993; Guo et al., 1996; Miller et al., 1998; Chen et al., 1999; Krocova et al., 2000; Flohe et al.,
 4    2002).  Studies in mouse, rat, rabbit, and human provide a clear indication that one effect of Pb
 5    on macrophages is to boost production of the proinflammatory cytokine TNF-a. While most
 6    studies examined the immediate effects of Pb exposure on TNF-a production, studies by Miller
 7    et al. (1998) and Chen et al. (1999, (2004) showed that the effects of early gestational exposure
 8    to Pb on macrophages could persist well into later life, including adulthood.  Additionally, Chen
 9    et al. (1999) showed that chelation of Pb with succimer in developing female rats in utero could
10    eliminate the persistent effect of elevated TNF-a production in the adult offspring. Flohe et al.
11    (2002) found evidence that lead-induced elevation in TNF-a production is sensitive to both PKC
12    signaling as well as to protein production.  While the production of TNF-a can be elevated
13    following exposure to lead, the expression of the receptor for TNF-a (TNF-R) was also increased
14    during the in vitro exposure of human blood monocytes to Pb-chloride (Guo et al., 1996).
15    Therefore, the combined effect of elevated cytokine production by macrophages as well as
16    increased receptor expression would be expected to contribute to problematic inflammatory
17    responses.
18
19    Production of Other Proinflammatory Cytokines
20          Several studies have indicated that macrophage production of cytokines (or that levels of
21    cytokines known to be produced primarily by macrophage populations) is altered after exposure
22    to lead. These vary somewhat, depending upon the exposure protocol and the source of
23    macrophages examined. In addition to the previously discussed elevation of TNF-a by lead, the
24    most significant and consistent lead-induced effects seem to involve elevated production of the
25    other major proinflammatory cytokines, interleukin-lp (IL-1(3)  and IL-6.  Increased production
26    of IL-6 following exposure to Pb has been reported by Dyatlov  and Lawrence (2002), Flohe et al.
27    (2002), Kim and Lawrence (2000), Krocova et al.  (2000), Kishikawa and Lawrence (1998) and
28    Kishikawa et al. (1997). Because IL-6 is a proinflammatory cytokine, its increased production
29    following Pb exposure has the potential to influence many different tissues.  Dyatlov et al.
30    (1998a,b) provided  evidence that lead, IL-6 and LPS can combine to exert a significant impact
31    on the permeability of the blood brain barrier as well as the properties of brain neurons and
      December 2005                           5-240        DRAFT-DO NOT QUOTE OR CITE

-------
 1    endothelial cells.  Lead-induced elevation of IL-1(3 production has been reported by Dyatlov and
 2    Lawrence (2002).  It is probable that enhanced co-production of IL-1P and IL-6 would increase
 3    the likelihood of local tissue inflammation.
 4
 5    Production of Reactive Oxygen Intermediates (ROIs)
 6          Reactive oxygen intermediates (ROIs) are important metabolites in the capacity of
 7    macrophages and other inflammatory cells to kill invading bacteria and to attack cancer cells.
 8    However, increased overall production or inappropriate triggering of ROI release by
 9    macrophages can be a major contributor to tissue damage and the oxidation of cell surface lipids
10    as well as DNA.  The latter is one mechanism through which improperly regulated macrophages
11    can actually increase the incidence of cancer. Results from many studies suggest that lead-
12    exposure of macrophages can increase the release of superoxide anion and/or hydrogen peroxide
13    at least shortly after exposure.  Key studies are summarized in Table AX5-9.6.
14          In a recent study on environmentally exposed children in Mexico, Pineada-Zavaleta et al.
15    (2004) reported that production of superoxide anion by directly activated (interferon-gamma +
16    LPS) monocytes was directly correlated with blood Pb level.  This was in contrast with the effect
17    of arsenic, which had a negative association.  In other studies involving low levels of exposures,
18    Zelikoff et al. (1993) demonstrated that rabbits exposed to Pb via inhalation had pulmonary
19    macrophages that produced elevated levels of both H2O2 and superoxide anion upon stimulation
20    in vitro.  In an in vitro study, Shabani and Rabani (2000) reported that Pb nitrate exposure
21    produced a dose dependent increase in superoxide anion by rat alveolar macrophages.  Baykov
22    et al. (1996) fed BALB/c mice dietary Pb and found that peritoneal macrophages had an
23    increased spontaneous  release of H2O2.
24          Other studies  have reported no effects of Pb on superoxide anion production when a long
25    recovery period was included following in vivo exposure  (Miller et al., 1998) as well as negative
26    effects of Pb  on oxidative metabolism by certain macrophages or macrophage cell lines
27    (Castranova et al., 1980; Hilbertz et al., 1986; Chen et al., 1997).  These somewhat different
28    results suggest that the subpopulations of macrophages examined (e.g., alveolar vs. splenic vs.
29    peritoneal) and the timeframe of assessment relative to exposure may be important factors in the
30    effect of Pb on ROI production.


      December 2005                           5-241       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The biological importance of increased ROI production by lead-exposed macrophages
 2    should not be underestimated. Fernadez-Cabezudo et al. (2003) demonstrated that the potent
 3    antioxidant, vitamin E could protect TO strain mice against some lead-induced
 4    immunosuppressive alterations.  Hence, macrophage-associated oxidative damage following
 5    exposure to Pb may be a mitigating factor in nonlymphoid organ lead-induced pathologies.
 6
 7    Arachidonic Acid Content and Prostaglandin Production
 8          Archidonic acid (AA) is a major surface component of many cells, including
 9    macrophages, and is the precursor of cyclooxygenase and lipoxygenase metabolites.  As a result,
10    the specific AA content of membranes and the capacity of macrophages to produce
11    immunomodulatory metabolites from AA are important to overall health of the individual. One
12    of the findings since 1986 concerning lead-induced modulation of macrophage function is the
13    impact of Pb on PGE2 production.  One study (Knowles and Donaldson, 1990) reported that diets
14    supplemented with Pb at 500 ppm and fed to chicks produced an increase in the percentage of
15    AA included in cell membranes. Such an increase would be expected to raise the risk of overall
16    inflammation.
17          Several groups have reported that Pb exposure increases macrophage production of the
18    immunosuppressive metabolite PGE2.  Lee and Battles (1994) reported that mouse macrophages
19    exposed to Pb (10 jiM) in vitro had elevated basal PGE2 production, but under some stimulatory
20    conditions, had decreased production of PGE2.  When Knowles and Donaldson (1997) fed Pb to
21    turkey poults in the diet at a level of 100 ppm, macrophage production of prostaglandin F2
22    (PGF2), PGE2 and thromboxane production were all significantly  elevated vs. the control. Flohe
23    et al. (2002) showed that exposure  of mouse bone marrow-derived macrophages to Pb-chloride
24    resulted in increased production of PGE2 that correlated with increased mRNA production for the
25    necessary enzyme, prostaglandin H synthase type-2.
26
27    Tissue Homeostasis
28          In an important observation  reflecting the impact of lead-induced immunotoxicity on
29    nonlymphoid tissues, Pace et al.  (2005) showed that neonatal exposure of mice to Pb-acetate via
30    drinking water (0.1 ppm for 6 weeks, both through maternal nursing and direct) produced a
31    significant reduction in the testicular macrophage population.  This correlated with increased

      December 2005                           5-242       DRAFT-DO NOT QUOTE OR CITE

-------
 1    estradiol levels in the testis and reduced male reproductive performance. The authors
 2    hypothesized that lead-induced alteration among testicular macrophages is linked to an impaired
 3    tissue environment that likely includes increased oxidative stress, apoptotic somatic cells, and
 4    reduced fertility of males.
 5
 6    Colony Formation and Population Distribution
 1          The ability of bone  marrow-derived macrophages (BMDM) to form colonies in response
 8    to certain growth factors (e.g., colony stimulating factor-1 [CSF-1]) is a property related to the
 9    growth and differentiation of subsequent macrophage populations.  Kowelenko et al. (1991)
10    found that exposure to CBA/J female mice to Pb-acetate (0.4 mM in drinking water for 2 weeks)
11    reduced colony formation  of macrophages in response to CSF-1. Infection of the mice with
12    Listeria only exacerbated this effect of lead. The same authors (Kowelenko et al., 1989) had
13    previously demonstrated that when BMDM were cultured in vitro with Pb-chloride (0.1 |iM),
14    colony formation was significantly impaired. These combined results suggest that exposure to
15    Pb can impair the generation of macrophage populations as well as modulate the functional
16    spectrum of fully matured  macrophages. Bunn et al. (200la) reported that gestational exposure
17    of CD rats to 50 ppm Pb-acetate via the drinking water of the dams resulted in female adult
18    offspring with a significantly decreased percentage (58% reduced) of circulating monocytes.
19    A 100 ppm dose of Pb-acetate produced a significant reduction (74% reduced) in the absolute
20    numbers of monocytes as well. The blood lead level at birth associated with  the decreased
21    percentage of macrophages in the adult offspring was 8.2 |ig/dL. In general agreement, Lee
22    et al. (2002) reported a significant decrease in the absolute numbers of circulating monocytes and
23    polymorphonuclear leukocytes (PMNs) in juvenile female chickens exposed  in ovo on
24    embryonic day (E) 12 to 200 jig Pb-acetate.  The corresponding blood lead level at hatching was
25    11.0 |ig/dL. However, in this case, the lead-induced reduction in monocytes  and PMNs was  only
26    seen in concert with an airway viral infection (viral stressor) and not in the resting uninfected
27    animal.
28
29    Antigen Presentation and Lymphoid Stimulation
30         Exposure to Pb influences the interaction between macrophages and T lymphocytes, and
31    as a result, the capacity of macrophages to support T lymphocyte proliferation and activation can

      December 2005                           5-243       DRAFT-DO NOT QUOTE OR CITE

-------
 1    be altered as well.  Kowelenko et al. (1988) found that mouse macrophages exposed to Pb (both
 2    in vivo and in vitro) can induce an increased proliferative response of T lymphocytes in co-
 3    culture but that antigen-specific stimulation of primed T cells is significantly reduced.
 4    Lead-suppressed antigen presentation capabilities of mouse macrophages were also reported by
 5    both Smith and Lawrence (1988) and Blakley and Archer (1981).
 6
 7    Chemotaxis
 8          Chemotactic activity of macrophages is an important function required for the directed
 9    migration of macrophages to sites of infection and tumor growth.  However, it is a functional
10    capacity that has not been systematically examined within the lead-immune literature. Using
11    female Moen-Chase guinea pigs, Kiremidjian-Schumacher et al. (1981) showed that Pb chloride
12    exposure of peritoneal macrophages in vitro (10-6 jiM) inhibited the electrophoretic mobility of
13    the cells.
14
15    Phagocytosis and Clearance of Particles
16          Phagocytosis of targets and removal/clearance of dead cells and particles are major
17    functions of macrophages. However,  phagocytosis can involve a variety of different cell surface
18    receptors on macrophages, depending upon both the nature of the target encountered and the
19    subpopulation of macrophages examined.  In general, phagocytic capacity of macrophages seems
20    to be relatively insensitive to lead-induced immunomodulation compared with the effects on NO
21    and TNF-a production.
22          However, differences in outcome in phagocytosis evaluations are likely to be based on the
23    differences in the source of macrophages used and their relative activation state at the time of
24    assessment.  A few studies have described significant effects on phagocytosis, but these have
25    usually relied upon phagocytosis mediated through the Fc receptor on macrophages.  Because
26    cell adherence to surfaces may be influenced negatively by Pb (Sengupta and Bishali, 2002),
27    impairment of phagocytosis may also involve some lack in efficiency with macrophage
28    anchoring to substrates.  De Guise et al.  (2000) reported no effect on bovine macrophage
29    phagocytosis of latex beads by Pb at in vitro treatment concentration of 104 M. This was in
30    contrast with suppressive effects of both cadmium and mercury. Using Sephadex-elicited
31    peritoneal macrophages derived from  young turkeys fed 100 ppm Pb in the diet, Knowles and

      December 2005                           5-244       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Donaldson (1997) found a 50% reduction in the percentage of phagocytic macrophages using
 2    SRBC targets.  The activity per phagocytic macrophage was also reduced.
 3          Kowolenko et al. (1988) studied the effect of Pb-acetate at 10 mM in the drinking water of
 4    CBA/J mice. They reported no effect on phagocytosis ofListeria monocytogenes targets, yet
 5    they found an overall decreased resistance to Listeria.  When the same investigators exposed
 6    peritoneal and splenic macrophages to Pb in vitro (100 |iM), they also found no significant effect
 7    of Pb on phagocytic activity.  Jian et al. (1985) reported that New Zealand white rabbit-derived
 8    alveolar macrophages exposed to Pb in vitro at 10"5 M concentration were significantly impaired
 9    in the phagocytosis of opsonized chicken erythrocytes (Fc receptor-mediated phagocytosis).
10    Trejo et al. (1972) reported that a single i.v. injection of Pb (5 mg/rat) into male Sprague Dawley
11    (SD) strain rats produced an inhibition in the phagocytic capacity of Kupffer cells.
12          Several studies have reported a decreased clearance capacity of the reticuloendothelial
13    system following in vivo exposure to lead. Filkins and Buchanan (1973) found that injection of
14    5 mg of Pb-acetate i.v.  into male Holtzman strain rats produced reduced carbon clearance.
15    Similarly, Trejo et al. (1972) reported that a single i.v. injection of Pb (2.5 mg) into male SD
16    strain rats significantly reduced clearance of colloidal carbon.
17          In  contrast, Schlick and Friedberg (1981) found that 20 |ig/kg Pb-acetate in a single i.p.
18    injection  of NMRI strain mice significantly increased the clearance of India ink. Ironically,  oral
19    administration of Pb for 10, but not 30, days of 10 |ig/kg resulted in an increase in clearance
20    activity.  Difference in route of Pb administration may be a factor in the different results
21    obtained.
22
23    Induction of Heat Shock Proteins
24          One study (Miller and Qureshi, 1992), using a macrophage cell line, reported that
25    exposure  of macrophages (MQ-NCSU) in culture to Pb-acetate (1000 jiM) induced the same set
26    of four heat shock proteins as when the macrophages were subjected to thermal stress.  This
27    result fits the hypothesis that Pb produces a profound immunomodulatory effect in macrophages
28    that has similarities with the exposure of macrophages to certain pathogens.
29
      December 2005                           5-245        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Apoptosis
 2          Significant differences exist in the literature concerning the potential role of Pb in the
 3   apoptosis of macrophages.  The difference may be based on the exposure methodologies (in vivo
 4   vs. in vitro) as well as the source of macrophages utilized.  De la Fuente et al. (2002) found that
 5   human monocytes exposed to Pb in vitro at high concentrations did not undergo apoptosis. This
 6   was in direct contrast with the apoptosis-promoting effects of cadmium in the same assessment
 7   protocol. In contrast, Shabani and Rabibani (2000) exposed rat alveolar macrophage to Pb
 8   nitrate in vitro and found that 60 jiM concentration produced a significant increase (2x) in DNA
 9   fragmentation after 3 to 24  h in culture.
10
11   5.9.7  Granulocytes and Natural Killer (NK) Cells
12          Other cell types important in innate immunity, as well as in immunoregulation, are the
13   lymphoid population of natural killer cells and granulocytes, including PMNs (i.e., neutrophils).
14   Neither population appears to be a major target for lead-induced immunotoxicity, although both
15   may be influenced indirectly via immune cell-cell interactions as well as by changes in cytokine
16   production. Among the two, neutrophils may be the more sensitive cell type based on assays
17   conducted  to date. For neutrophils, several groups have reported alteration in chemotactic
18   activity following exposure to lead. Queiroz et al. (1993) found impaired migration  ability of
19   neutrophils from battery workers occupationally exposed to lead. Likewise, Valentino et al.
20   (1991) had a similar observation among male occupationally exposed workers. Lead exposure of
21   young SD  strain rats can increase the population of neutrophils (Villagra et al., 1997), although,
22   as the authors indicated, this does not necessarily afford enhanced host protection against
23   disease. Baginski and Grube (1991) reported that human neutrophils exposed to Pb had
24   increased killing capacity, probably via increased release of ROIs despite having reduced
25   phagocytic capacity. This would fit the same general profile as the effects of Pb on
26   macrophages. Therefore, neutrophils may contribute to lead-induced tissue inflammation and
27   damage via increased ROI release.  Yet, their effectiveness in protection against disease
28   challenge may be no greater following exposure to Pb, because some impairment in  chemotaxis
29   and phagocytosis has been reported as well.
30          Yucesoy  et al. (1997) reported that either Pb exposure or simultaneous exposure to Pb and
31   cadmium in human workers did not impair NK cytotoxicity activity. This finding was supported
     December  2005                            5-246       DRAFT-DO NOT QUOTE OR CITE

-------
 1    by studies using in vivo exposure to Pb in rats (Kimber et al., 1986) and mice (Neilan et al.,
 2    1983).  Therefore, it would appear that NK cells are not a prime target associated with lead-
 3    induced immunotoxicity, although more subtle effects may certainly exist within the cell type.
 4          Eosinophils represent an important granulocytic cell type in type 2 associated
 5    inflammatory and allergic reactions. However, few studies have examined Pb exposure and
 6    eosinophil activity.  Villagra et al. (1997) reported that exposure of female juvenile SD rats to Pb
 7    [four alternate-day s.c. injections of 172 mg/g body wt Pb-acetate] increased the degranulation of
 8    eosinophils (in animals given estrogen 1 day later).  Such a response would be expected to
 9    contribute to increased inflammation.
10
11    5.9.8   Hypersensitivity and Autoimmunity
12          At the time of preparation of the 1986 AQCD, little was known about the potential for Pb
13    to influence the risk of allergic and autoimmune diseases.  However, since the early 1990s, a
14    significant number of studies have all pointed toward the fact that Pb causes a profound
15    dysregulation of the immune  system. It skews the balance of responses in directions that reduce
16    certain host defenses against infectious diseases while enhancing the risk of allergic and
17    autoimmune disease.  Lead exposure at low to moderate levels appears to alter T lymphocyte
18    responses in such a way as to increase the risk of atopy, asthma, and some forms of
19    autoimmunity.  Increased IgE production following exposure to Pb is among the most frequently
20    reported immune alterations.  Elevated IgE levels would be an associated risk factor for atopy
21    and allergic disease. Several  investigators have discussed the fact that Pb is a likely risk factor
22    associated with the increased incidence of childhood allergic asthma (Miller et al., 1998; Heo
23    et al., 1998; Snyder et al., 2000; McCabe et al., 2001; Dietert et al., 2004; Transande and
24    Thurston, 2005) as well as later life allergic disease (Heo et al., 2004). Joseph et al. (2005)
25    observed no association for childhood BLL and risk of asthma among an African-American
26    population. However, results on other populations from this  study, including those involving
27    Caucasian children with BLLs above 5 |ig/dL, led the authors to call for further studies into the
28    possible linkage of early life lead exposure and risk of asthma (Joseph et al., 2005).
29          As described by McCabe et al. (1991) and discussed by Dietert et al. (2004), lead-induced
30    immunotoxicity is novel in that profound cellular toxicity is not evident following exposure at
31    low to moderate exposure concentrations. In fact, antibody responses overall are usually

      December 2005                           5-247       DRAFT-DO NOT QUOTE OR CITE

-------
 1    unaffected or may be increased depending upon the class/isotype measured.  However, the
 2    functional responses mounted following Pb exposure do not reflect the normal immune balance
 3    that would otherwise occur.  This dysregulation can alter the risk of certain autoimmune diseases
 4    based on several observations. Holladay (1999) has considered the importance of the timing of
 5    exposure and the fact that early life exposure may establish the immune profile that then
 6    contributes to later disease including autoimmunity.
 7          Hudson et al.  (2003) reported that exposure to Pb can exacerbate systemic lupus
 8    erythmatosus (SLE)  in lupus-prone strains of mice. In contrast with the effect of mercury, these
 9    authors found that for lupus, Pb exposure would not induce this autoimmune condition in
10    genetically resistant mice but would increase severity of the disease in genetically prone animals.
11    The authors noted some gender effects within certain strains (e.g., NZM88).  Using early in ovo
12    exposure to Pb (10 jig/egg), Bunn et al. (2000) found that Pb-acetate-exposed male chicks could
13    be induced to produce autoantibodies against thyroglobulin, which were not present in acetate-
14    exposed controls. No lead-induced alteration was observed in females that were predisposed to
15    mount anti-thyroglobulin responses. The gender effect is intriguing in that autoimmune
16    thyroiditis in genetically predisposed strains is always more severe in females than in males.
17          Two lines of evidence suggest that the capacity of Pb to influence the risk of
18    autoimmunity is not  always associated with simply a strict shift from Thl to Th2 responses.
19    Hudson et al. (2003) discussed the fact that lupus is not purely a Th2-mediated disease, but rather
20    seems to occur under conditions associated with skewing in either direction.  McCabe et al.
21    (2001) found that Pb can increase the stimulation of alloantigen reactive T cells (where
22    macrophage processing  of antigen is required) but not enhancement of T cell clonotypic
23    responses against either mitogens or superantigens (where processing is not required). This
24    suggests that the role of Pb in influencing risk of autoimmune disease goes beyond a simple
25    consideration of Thl/Th2 balance.  In fact, Goebel et al. (2000), studying mucosal immunity,
26    reported that administration of Pb-chloride to NOD strain mice produced a gut cytokine
27    microenvironment that was skewed toward Th2 over the short run, but later was shifted toward
28    Thl with increased production of IFN-y. This shift to Thl was accompanied by a loss of
29    tolerance  and capacity to mount an immune response against a diet-associated protein (chicken
30    ovalbumin).  The authors proposed that reduction of the capacity for oral tolerance would
31    predispose an individual toward autoimmune disease.

      December 2005                            5-248        DRAFT-DO NOT QUOTE OR  CITE

-------
 1          Finally, Waterman et al. (1994) and El-Fawal et al. (1999) have described the production
 2    of autoantibodies against neural proteins in both battery workers and rats exposed to low levels
 3    of Pb via drinking water. These authors have suggested that exposure to Pb may precipitate the
 4    autoimmunity by altering antigen immunogenicity and/or the capacity of the immune system to
 5    respond to certain antigens. This, in turn, may contribute to the eventual lead-associated
 6    neurological disease.
 7
 8    5.9.9  Mechanism of Lead-Based Immunomodulation
 9          In the  1986 AQCD, there was little direct information available about the immune system
10    regarding the molecular mechanism(s) of lead-induced immunotoxicity.  Binding to thiol groups
11    and altering cell surface receptors were indicated as possible factors in altered immune function.
12    Since that time, some additional information has been generated through a variety of studies on
13    human and animal immune cells.  However,  a clear or simple explanation remains to be
14    determined. Table 5-9.3 lists studies on the immune system that have contributed to a better
15    understanding of potential mechanisms or have forwarded potential hypotheses with some
16    supporting data.
17          At the level of cell-cell interactions, it seems clear that Pb alters metabolism and cytokine
18    production by macrophages and antigen presenting cells. It also reduces their capacity to
19    respond to growth factors such as CSF-1 (Kowelenko et al., 1989). Pace et al. (2005) discussed
20    the hypothesis that reduced populations of functionally altered macrophages (because of lead-
21    induced unresponsiveness to CSF-1 and over production of ROIs) in tissues can produce
22    nonimmune problems.  The model they used is the homeostatic presence of testicular
23    macrophages and the likelihood that lead-induced macrophage immunotoxicity contributes
24    directly to lead-associated reduction in male  fertility.
25          Additionally, Pb is known to selectively alter cell signaling to CD4+ T cell
26    subpopulations, promoting proliferation in some but not others. The outcome is enhanced tissue
27    inflammation, reduced CMI, and increased production of atopy-inducing antibodies. Risk of
28    autoimmune reactions is increased in some models of lead-induced immunotoxicity. For
29    example, Heo et al. (1997) reported that lead-exposed murine T lymphocytes are biased in
30    expression of VP genes. This is potentially problematic as this phenotype is common among a
31
      December 2005                           5-249        DRAFT-DO NOT QUOTE OR CITE

-------
         Table 5-9.3. Suggested Mechanisms of Lead-Induced Immunotoxicity
Species
Mouse
Mouse
Chicken

Mouse
Mouse
Rat


Mouse

Mouse

Human

Mouse

Mouse
Strain/Gender
Balb/c
TO strain males
Cornell K Strain

Balb/c females
C57 Bl/6
females
C 57B1/6
PC-12 cells


DO11. 10 ova-
mice

DOll.lOova-tg
mice

-

CBA/J females

Swiss Females
Suggested
Endpoints
CSF-1
Responsiveness of
Macrophages
Vitamin E
protection against
lead-induced
splenomegaly
Thymulin partial
reversal of Th
skewing
Lead disruption of
antigen
processing and
presentation
signals
PKC activation
NF-KB activation
AP-1 induction
C-Jun kinase
induction
Adenylcyclase
activation with
elevated cAMP
levels
V(3 gene usage

NF-KB activation
in CD4+ cells

t Immunogenicity
of neural proteins

tTNF-a
production
Associated
Functional
Alteration
ITesticular
macrophages
^Fertility
tPutative ROI
associated
splenomegaly
ILead-induced
DTH
suppression
tAlloreactive
CD4+ hlgh cells
tRisk of
autoimmunity
tTNF-a, tIL-6
tPGE2
tROI


tTh skewing

tRisk of
autoimmunity

tRisk of
autoimmunity
and
hypersensitivity
t Autoimmune
mediated
neurological
damage

t Sensitivity to
endotoxin
Lowest
Effective
Dose
0.1 ppm
Img/kg
400 jig

0.5 |jM in
vitro
20|jM in
vitro
1 |jMin
vitro


2.5 |jM in
vitro

50 ng
2x/week
s.c.
IpM

Lead-
altered
proteins
used as
antigens
5mg
Duration
6 weeks
2 weeks
Single in
ovo
injection
4 days
4.5 hrs
5-120 min


15 mins-
6 hrs

8 weeks

30 min

3 injections
of lead-
modified
neural
proteins
Single i.p.
injection
References
Pace et al.
(2005)
Fernandez-
Cabezudo
et al. (2003)
Lee and
Dietert,
(2003)
McCabe et
al. (2001)
Flohe et al.
(2002)
Ramesh et
al. (1999)


Heo et al.
(1998)

Heo et al.
(1997)

Pyatt et al.
(1996)

Waterman,
etal. (1994)

Dentener et
al. (1989)
December 2005
5-250
DRAFT-DO NOT QUOTE OR CITE

-------
 1    variety of human and animal model autoimmune conditions.  A variety of exogenous factors
 2    have been reported to partially ameliorate the immunotoxic effects of lead. Chelation of Pb in
 3    lead-exposed dams corrected some lead-induced immunotoxic problems in the rat female
 4    offspring, but it left the animals with some DMSA-induced immune alterations (Chen et al.,
 5    1999). Other exogenously administered factors that have been reported to partially restore
 6    lead-suppressed immune function are vitamin E (Fernandez-Carbezudo et al., 2003) and
 7    thymulin (Lee and Dietert, 2003).
 8          At the subcellular level, the bases for immunotoxic changes remain speculative. McCabe
 9    et al. (2001) suggested that altered antigen processing and subsequent cell signaling to T cells
10    may be an explanation for the capacity of Pb to selectively increase CD4+ (high density) cells.
11    Certainly Pb appears to alter signal transduction. It appears to elevate expression of the nuclear
12    transcription factor NF-KB (Pyatt et al., 1996; Ramesh et al.,  1999) as well as increase
13    expression of AP-1 and cJun (Ramesh et al., 1999).  Flohe et al. (2002) found evidence that Pb
14    can elevate the activation of PKC. The authors speculated that this might be involved in lead-
15    induced increases in TNF-a production. Additionally, Heo et al. (1998) reported that Pb
16    increases adenyl cyclase activity  among T lymphocytes, generating elevated  cAMP levels. The
17    authors hypothesized that this effect, in conjunction with differences in cell signaling pathways
18    for promoting Thl vs. Th2 cells,  may be involved in the capacity of Pb to skew ThO helper cells
19    toward Th2.
20
21    5.9.10  Age-Based Differences  in Sensitivity
22          With the literature available at the time of the 1986 AQCD, it was virtually impossible to
23    evaluate age-based differences in susceptibility to lead-induced immunotoxicity.  However, in
24    recent years, this has become a major topic  of study for many toxicants including lead. Several
25    studies have added to the available data assessing the developmental immunotoxicity of Pb
26    (reviewed in Barnett [1996], Dietert et al. [2000, 2004], Lee and Dietert [2005]).  Several
27    patterns have emerged from exposure data using animals of different ages.
28          First, it seems clear that blood Pb levels at or near birth of below 10 |ig/dL can be
29    associated with juvenile and/or adult immunotoxicity.  Several studies reported effects in the
30    range of 5-8 |ig/dL. These low levels would seem to place the immune system on par with the

      December 2005                            5-251        DRAFT-DO NOT QUOTE OR CITE

-------
 1   neurological system in terms of potential sensitivity to lead. Table 5-9.4 shows examples of
 2   studies in which low blood lead levels were linked with immunotoxicity.
              Table 5-9.4. Immunomodulation Associated with Low Blood Lead Levels
                                            in Animals
Species
Mouse


Rat

Rat


Rat

Chicken

Chicken

Chicken
Blood lead
Oig/dL)
-5.0


8.2

6.75


8.0

8.2

11.0

7.0
Age at
Measurement
1 week


1 day

4 weeks


4 weeks

1 day

1 day

1 day
Immune Parameter(s)
tlgE,
4 Splenic
T Cell Populations
Imonocytes

4DTH,
4IFN-Y,
tIL-4
tTNF-a
tRel. Spleen weight
^circulating lymphocytes post
infection
4DTH and 4TLC, monocytes,
PMNs post infection
tautoantibody production
Age at
Assessment
2 weeks


13 weeks

13 weeks


13 weeks

5 weeks

5 weeks

10 weeks
Reference
Snyder et al.
(2000)

Bunn et al.
(200 la)
Chen et al. (2004)


Lee et al. (2002)

Lee et al. (2002)

Lee etal. (2001)

Bunn et al. (2000)
 3          A second finding is that the immunotoxic effects induced by Pb are persistent long after
 4   blood levels and potential body burdens of Pb are significantly reduced.  Miller et al. (1998),
 5   Chen et al. (1999), Snyder et al. (2000), and Lee et al. (2001) all emphasize this latter point.
 6   In fact, in most of these studies immunotoxic alterations were present when Pb levels in exposed
 7   animals were not distinguishable from control levels. This should provide a cautionary note
 8   regarding studies in humans. Data from adult exposures provides little insight into the potential
 9   persistence following adult exposure to lead. However, rather than the developing immune
10   system being more regenerative postexposure and able to withstand immunotoxic insult, it
11   appears that the non-dispersed  developing immune system is a particularly susceptible target to
12   many immunotoxicants (Dietert et al., 2002).
13          A third, and somewhat surprising, finding concerning early exposure to Pb is that
14   qualitative differences in the spectrum of immune alterations can exist, depending upon the
     December 2005
5-252
DRAFT-DO NOT QUOTE OR CITE

-------
 1    developmental window of exposure. Figure 5-9.1 illustrates this point. Early embryonic
 2    exposure of rats and chickens to Pb failed to alter juvenile DTK responses, despite significant
 3    effects on macrophage function. However, exposure to Pb after the mid-embryonic point of
 4    embryonic development readily suppressed subsequent DTH. As shown in Figure 5-9.1, the
 5    development window in which sensitivity to DTH suppression emerges is quite similar in the
 6    two species.  This observation suggests that both quantitative (LOAELs) and qualitative (range
 7    of immune alterations) differences in sensitivity to Pb can exist across different age groups.
 8          Additionally, some studies in animals have noted gender differences in the effects of Pb
 9    following exposure (Bunn et al., 2000, 2001a,b, c; Hudson et al., 2003).  Gender differences
10    have also extended to results in humans as per lead-induced immune and inflammatory
11    alterations (Karmaus et al., 2005; Fortoul et al., 2005). It seems feasible that, even in the
12    embryo, hormonal differences among females and males may impact some outcomes of low-
13    level Pb exposure.
14          Table  5-9.5 shows comparisons of the lowest reported blood Pb levels at different ages
15    associated with the same immunotoxic endpoint. From these limited  comparisons, it would
16    appear that different ages of rodents (e.g., embryonic vs. adult) differ in dose sensitivity for lead-
17    induced immunotoxicity somewhere in the range of 3 to 12-fold.  Clearly, additional direct
18    comparisons  would help to refine this estimate.
19          A fourth observation from the early exposure studies is that exposure to even very low
20    levels of Pb can predispose the immune system for unanticipated postnatal responses when the
21    system is stressed. This general phenomenon is called latency.  Lee et al. (2002) provided an
22    example of this following the single in ovo exposure of embryonic day 5 chick embryos to low
23    levels of Pb (10 jig; blood lead level 1 day post hatch of 8.2 jig/dL). The leukocyte profiles of
24    the animals appeared to be completely normal. However,  when these animals were exposed to a
25    respiratory virus, their pattern of leukocyte mobilization was completely aberrant from controls.
26    Therefore, some immunotoxic alterations following early exposure to low levels of Pb may only
27    be evident during periods of postnatal stress.
28          Several studies have reported the positive association of blood Pb levels in children with
29    elevated serum IgE (Karmaus et al., 2005; Sun et al., 2003; Lutz et al., 1999). These
30    observations  are supported by the animal data in rats and mice (Miller et al., 1998; Snyder et al.,
31    2000) and suggest that lead-induced risk of atopy and asthma may be a particular health issue.

      December 2005                           5-253       DRAFT-DO NOT QUOTE OR CITE

-------
        Table 5-9.5. Comparisons of Age-Based Sensitivity to Lead-Induced Immunotoxicity
Species
Mouse

Rat
Mouse

Rat
Altered Endpoint Embryo - fetal*
tlgE ~5|^g/dL

iDTH 34 |^g/dL
(persistent effect assessed
13 weeks post-exposure)
4DTH

tTNF - a 8 ng/dL
(persistent effect assessed
13 weeks post-exposure)
Neonatal* Adult*
12 |^g/dL 38 |^g/dL

>112|^g/dL
(measured at birth
for persistent
effect)
29 |^g/dL 87 |^g/dL

>112|^g/dL
(measured at birth
for persistent
effect)
References
Snyder et al.
(2000)
Heoetal. (1996)
Miller et al.
(1998)
Bunn et al.
(200 Ib)
Faith etal. (1979)
McCabe et al.
(1999)
Miller et al.
(1998)
Chen et al. (2004)
      * Lowest blood lead concentration reported with effect


 1         Trasande et al. (2005) recently discussed the fact that, despite progress in reducing the
 2   deposition of Pb in the environment, Pb continues to be a concern relative to asthma and
 3   children's health.
 4
 5   5.9.11  Summary and Conclusions
 6         The immune system appears to be one of the more sensitive systems to the toxic effects of
 7   lead.  The 1986 AQCD provided an excellent summary of the studies that had been conducted
 8   prior to that date. But knowledge of fundamental immunology has progressed greatly during the
 9   past 20 years. Not surprisingly, the large number of studies conducted since the mid-1980s
10   provided a much clearer understanding of the immune-associated problems that can arise from
11   problematic exposure to lead. Studies across humans and a variety of animal models are in
12   general agreement concerning both the nature of the immunotoxicity induced by Pb as well as
13   the exposure conditions that are required to produce immunomodulation. Figure 5-9.2
14   summarizes the basic immunotoxic changes induced by Pb that result in Th skewing, impaired
15   macrophage function, and increased risk of inflammation-associated tissue damage.
     December 2005
5-254
DRAFT-DO NOT QUOTE OR CITE

-------
                      Key Effects of Lead on the Immune System
                   Inflammatory Tis
                             issue
             Macrophages, Macrophages and Antigen
                     Presenting Cells
| Membrane
Araehidonic
Acid
         Activation
                                X
                                Effects on Antigen
                                Processing
                                .Presentation
                                                T cells
                                                    Reduced Cell -
                                                    mediated
                                                    Immunity
                                    Skewed VB gene usage &

                                     Risk of Autoi mm unity
            In creased Tissue Inflammation

           Reduced Cell-mediated Immunity
                                                            Be ells
                                                                Cell Surface
                                                                   Q ass II
                   Host Proteins
                                                                             1
                                                                           Modified Neural Anti
                                                                                        igens
                                                                        Auto-antibodies and
                                                                         Pos siW e Tis sue
                                                                           Damage
                                                       IgE  & Increased Risk of
                                                         Atopy and Asthma
     Figure 5-9.2.   This figure shows the fundamental alterations to the immune system and to
                    immunological response and recognition induced by exposure to lead. The
                    functional shifts are disproportionate compared to the relatively modest
                    changes among leukocytes with low to moderate exposure to lead.
1          Lead is unlike many immunotoxicants in that, at low to moderate levels of exposure, it
2    does not produce overt cellular cytotoxicity or lymphoid organ pathology. However, it can
3    induce profound functional alterations that influence risk of disease.  Lead preferentially targets
4    macrophages and T lymphocytes, although effects have been reported in B cells and neutrophils
5    as well.  There are three major hallmarks of lead-induced immunotoxicity. First, Pb can
6    dramatically suppress the Thl-dependent DTH response, as well as production of associated Thl
     December 2005
                                          5-255
DRAFT-DO NOT QUOTE OR CITE

-------
 1    cytokines. Second, Pb can dramatically elevate production of IgE while increasing production of
 2    Th2 cytokines, such as IL-4. Third, and perhaps most sensitive, is the modulation of
 3    macrophages by Pb into a hyperinflammatory phenotype. After exposure to lead, macrophages
 4    significantly increase production of the proinflammatory cytokines TNF-a and IL-6 (and in some
 5    studies IL-1).  Many studies also reported elevated release of ROIs and prostaglandins.
 6    Ironically, production of one of the most important host defense factors, NO, is consistently and
 7    severely suppressed by exposure to lead. This package of lead-induced changes among
 8    macrophages makes them more prone to promote tissue destruction but actually less capable of
 9    killing bacteria or possibly presenting antigens to T lymphocytes. The Pb-induced shift in
10    phenotype explains the capacity of inhaled Pb to promote bronchial inflammation while bacterial
11    resistance is severely depressed.
12          Lead-induced skewing of Th activity (biasing responses toward Th2) across a population
13    would lead to the expectation of a greater risk of atopy, asthma, and some forms of
14    autoimmunity.  Concomitantly, resistance to some infectious diseases could be reduced.  This
15    predicted change of risk might help explain some recent trends in the incidence of diseases, such
16    as the epidemic rise in allergy and some forms of asthma in the United States.
17          Sensitivity of the immune system to Pb appears to differ across life stages.  Studies in rats
18    and mice suggest that the gestation period is the most sensitive life stage followed by the early
19    neonatal stage. But even during embryonic, fetal, and early neonatal development, critical
20    windows of vulnerability are likely to exist.  Compared to adults, the increased dose sensitivity
21    of the embryo-fetus would appear to fall in the range of 3-1 Ox depending upon the immune
22    endpoint considered.  Some studies have found evidence for gender differences in the impact of
23    Pb  on the immune system particularly with early life exposures. Potential gender differences in
24    immunotoxic outcome may be important in the evaluation of those populations at greatest risk.
25          Recent studies have suggested that exposure of embryos to Pb producing neonatal blood
26    lead concentrations below 10 |ig/dL can also produce later-life immunotoxicity (see
27    Table 5-9.4). Furthermore, immunotoxicity persists long after any evidence of prior embryonic
28    Pb  exposure. This latter observation from several laboratories may have implications for the
29    design of human epidemiological studies.
30
31

      December 2005                           5-256       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.10   EFFECTS OF LEAD ON OTHER ORGAN SYSTEMS
 2          In the 1986 Pb AQCD, the discussion of other organ systems included cardiovascular,
 3    hepatic, gastrointestinal (GI), and endocrine systems. Due to our increased understanding on the
 4    effects of Pb on cardiovascular and renal systems and their contribution to potential health
 5    effects of Pb, separate sections (5.5, 5.7) were dedicated earlier in this chapter to detailed
 6    discussions on these aspects. Similarly, with our increased understanding on the effects of Pb on
 7    endocrine functions and its inherent role with respect to neurotoxicological, reproductive, and
 8    developmental  effects,  literature reviewed for Pb effects on the endocrine system is discussed in
 9    the respective sections.  This section focuses on the discussion of Pb effects on the hepatic and
10    GI systems.
11
12    5.10.1 Effects of Lead on the Hepatic System
13          The liver is a highly active metabolic tissue. Apart from its roles in fatty acid metabolism
14    and limited heme synthesis function, the liver also has a major role in guarding other systems
15    from the toxic effects of xenobiotic compounds using a huge complement of detoxification
16    machinery referred to as phase I and phase II enzyme systems. Limited studies on experimental
17    animals reported in the 1986 Pb AQCD indicated that Pb induced effects in the hepatic system.
18    Laboratory animals, especially rats, exposed to Pb-nitrate have exhibited increased liver cell
19    proliferation, DNA synthesis, cholesterol synthesis, and glucose -6-phosphate dehydrogenase
20    (G6PD) activity indicative of Pb-induced hyperplasia. Further, the literature reviewed in the
21    1986 Pb AQCD reported alterations in the levels of drug metabolizing enzymes in experimental
22    animals given large doses of Pb.  The evidence for such effects in humans was less consistent.
23    The 1986 document also concluded that the  effects on the liver occurred only at high exposure
24    levels. The majority of studies on the effects of Pb on the hepatic system in experimental
25    animals that are reviewed in this document report functional and biochemical changes in the
26    liver, clearly pointing to metabolic perturbations in liver. For ease in understanding and
27    integration of these functional changes, the discussion is divided into the following four
28    subsections: hepatic drug metabolism, lipid and glycogen metabolism and lipid peroxidation,
29    heme synthesis, and toxicity mitigation by chelation and other interventions.
30

      December 2005                           5-257       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.10.1.1 Hepatic Drug Metabolism
 2          Approximately 75% of the hepatic blood comes directly from the gastrointestinal viscera,
 3    with the majority of drugs or xenobiotics absorbed coming directly to the liver in concentrated
 4    form.  The liver is equipped with a huge complement of drug metabolizing enzymes that detoxify
 5    many of the xenobiotics but also activate the toxicity of others. Oxidation and conjugation of
 6    xenobiotics have historically been referred to as phase I and phase II reactions.  The phase I
 7    enzymes include cytochrome P450 (CYP450) heme-containing monoxygenases, flavin-
 8    containing monoxygenases, and epoxide hydrolases. The phase II enzymes include glutathione
 9    (GSH) S-transferases (GST), UDP-glucuronyl transferases (UGT), N-acetyltransferases (NAT),
10    and sulfotransferases (SULT).  Xenobiotic metabolism by these two complements of enzyme
11    systems are essential for catabolizing and eliminating of drugs; however, this process can also
12    produce activated toxicants and carcinogens. A limited number of these CYP450s are involved
13    in the biosynthetic pathways of steroid and bile acid production.  It has been increasingly
14    recognized that, under certain circumstances, CYP P450s can produce ROS that result in
15    oxidative stress and cell death.
16          Liver is an active tissue. In addition to xenobiotic metabolism, it also participates in
17    gluconeogenesis, fatty acid metabolism, and cholesterol biosynthesis. Research concerning the
18    effects of Pb on the hepatic system in the past 15 years has provided some preliminary
19    indications of Pb-induced alterations in many of the hepatic functions described above. The
20    following discussion presents, as much as possible, the effects of Pb on individual enzymes, but
21    due to the multifarious interactions of many of these metabolic enzymes, there may be places
22    such separation was not possible.
23
24    Phase I Enzyme
25          Earlier studies on the toxic effects of Pb on hepatic drug metabolizing enzymes
26    demonstrated that acute exposure to Pb-acetate decreased rat hepatic CYP450s with increased
27    levels  of urinary 5-aminolevulinic acid (ALA).  Co-treatment with phenobarbitol, a CYP450
28    inducer, was shown to reverse the decrease CYP450 levels, suggesting a Pb-acetate-mediated
29    inhibition of heme synthetic enzymes.  Decreased activities of estradiol-17 beta enzyme
30    observed in rat liver treated with triethyl Pb-chloride (Odenbro and Arhenius, 1984) suggest that
31    both Pb and organo-Pb compounds are capable of inhibiting CYP450 activities.  Roomi et al.

      December 2005                          5-258       DRAFT-DO NOT QUOTE OR CITE

-------
 1    (1986) also observed decreased levels of hepatic microsomal CYP450s and decreased
 2    aminopyrene-N-demethylase activity on exposure to a single dose of Pb-nitrate (5-10 mmol/kg
 3    body wt). This decrease in phase I enzymes was followed by increased levels of phase II
 4    components such as GSH, GST, and DT diaphorase, suggesting that Pb-nitrate and Pb
 5    compounds can induce biochemical properties characteristic of hepatocyte nodules. Subchronic
 6    (2-3 months) exposure to Pb-acetate (5-50 mg/kg body wt) had been found to induce CYP450s
 7    and cytochrome b5 in rat liver and kidney (Nehru and Kaushal, 1992). As described earlier,
 8    multiple isoforms of CYP450s exist in the liver.
 9          To identify the inhibitory effect of acute Pb exposure on specific isoform(s), Degawa
10    et al. (1994) exposed male F344 rats to Pb nitrate (20,100 jimol/kg body wt) and evaluated liver
11    CYP450s 24 h postexposure. Lead-nitrate exposure preferentially inhibited cytochrome
12    P4501A2 enzyme activity in liver microsomal preparations as assayed for mutagenic conversion
13    of substrates 2-amino-6-methyl-dipyridol [1,2-a; 3',2-d] imidazole and 3-amino-l-methyl-5H-
14    pyridol[4,3,-b]indole.  Lead-nitrate exposure also inhibited the induction of cytochrome
15    P4501A2 by the inducers 3-methylcholanthrene and 2-methoxy-4-aminoazobenzene at both the
16    protein and mRNA levels. The authors further concluded that the specific inhibition of P4501A2
17    by Pb-nitrate observed may have been due to inhibition of heme synthesis, as Pb-nitrate was not
18    found to inhibit P4501A2 activity in vitro. Additional studies carried out by the same group
19    using various metal ions (e.g., Pb, Ni, Co, and Cd) found that the specific inhibition of P4501A2
20    was unique to Pb-nitrate (Degawa et al., 1994, 1995). Degawa et al. (1996) also investigated the
21    effect of Pb-nitrate-mediated inhibition of CYP1A gene activity in rat liver by specific inducers
22    and reported that Pb-nitrate inhibited the induction of CYP1A mRNA by aromatic amines, but
23    not by aryl hydrocarbons, suggesting the role  of other cellular factors in the transcriptional
24    activation of CYP1A genes. Lead-nitrate has been reported to induce the production of TNF-a
25    in rat liver (Shinozuka et al., 1994), a cytokine implicated in the suppression of constitutive
26    expression of CYP1A2 mRNA in rat hepatocytes.  Based on these findings, Degawa et al. (1996)
27    concluded that the inhibition of constitutive and aromatic amine-induced expression of CYP1A2
28    in rat liver caused by Pb-nitrate may occur at least in part by TNF-a-associated  mechanisms.
29    Lead-nitrate (0.33 mg/kg body wt) pretreatment-mediated protection conferred  against carbon
30    tetrachloride (0.3 mL/kg)-induced hepatotoxicity as reported by Calabrese et al. (1995) may be
31    due to the inhibition of CYP450 activities in liver by Pb.

      December 2005                            5-259       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Jover et al. (1996) investigated the effect of heme deficiency on Pb-induced hepatic P450
 2    function and transcription. These authors concluded that the decrease in hepatic P450 resulting
 3    from Pb intoxication was mediated by two different mechanisms. One mechanism is involved
 4    inhibitory effects on P450 by Pb at the transcriptional level; the second was heme- dependent, as
 5    Pb-mediated inhibition of heme synthesis decreased the heme saturation of P450 and the apo-
 6    P450 ratio.
 7          The effect of heavy metals (Cd, Co, Cu, Ni, Pb, and Zn) on 3-methylcholanthrene-
 8    induction of cytochrome P4501A and the activity of ethoxyresorufin-O-deethylase (EROD) were
 9    investigated in fish hepatoma cells (PLHC-1) by Brucshweiler et al. (1996).  The authors
10    reported that all the heavy metals tested had more pronounced effects on EROD activity
11    compared to controls.  The inhibitory potency of Pb was reported to be very low compared to
12    cadmium or cobalt.  A single treatment of Pb-acetate induced hepatic DT diaphorase activity
13    (Sugiura et al., 1993).  This induction of hepatic DT diaphorase by Pb-acetate has been reported
14    to be decreased with concomitant administration of Dil, a calcium antagonist. Based on these
15    observations, Arizono et al. (1996) suggested that DT diaphorase induction by Pb-acetate may
16    occur de novo via protein synthesis mediated by increased cellular calcium.  The potential
17    interaction of metals, including Pb, on the induction of CYP1A1 and CYP1A2 by polycyclic
18    aromatic hydrocarbons (PAHs) in human hepatocyte cultures was investigated by Vakharia et al.
19    (2001). Lead-nitrate, like other metals such as Cd, Hg, and As, decreased the extent of CYP1A1
20    and CYP1A2 induction by five different PAHs. The authors concluded from these studies that
21    Pb (5 jiM) diminished the induction of CYP1A1 and CYP1A2 in human hepatocytes by
22    ultimately decreasing the levels of CYP1 Al protein that was normally attainable through PAH
23    induction. Korashy and El-Kadi (2004) also investigated similar interactions of metals with aryl
24    hydrocarbon receptor (AHR)-regulated gene expression and enzyme activities in wild-type
25    murine hepatoma cells (Hepa Iclc7) and AHR-deficient cells (C12). These studies indicated
26    that metals alone (including Pb) did not significantly alter CYP1 Al proteins or activity, or
27    change AHR ligand-induced enzyme activity. There was no change in mRNA levels. Lead, in
28    the presence or absence of AHR ligand, increased the activity of NAD(P)H:quinone
29    oxidoreductase and its mRNA levels.
30
     December 2005                           5-260       DRAFT-DO NOT QUOTE OR CITE

-------
 1   Phase II Enzymes
 2          A single injection of Pb-nitrate (5-10 jiM/100 g body wt) was found to increase GST
 3   activity levels (Roomi et al.,  1986). Additional studies by the same group identified induction of
 4   a specific form GST-P by Pb-nitrate in rat liver (Roomi et al., 1987). Because a single injection
 5   of Pb-nitrate decreased phase I and increased phase II hepatic enzymes, these investigators
 6   concluded that Pb-nitrate treatment initiated a biochemical phenotype similar to carcinogen-
 7   induced hepatocyte nodules.  Immunohistochemical analysis by the same group reported that Pb-
 8   nitrate administration resulted in the appearance of GST-P in most of the hepatocytes, an enzyme
 9   that is otherwise undetectable in normal rat liver (Columbano et al., 1988; Roomi et al., 1987).
10   On the other hand, Nakagawa (1991) reported inhibition of GST on acute exposure to Pb and
11   that the inhibition of GST followed a reduction in liver GSH levels. Nakagawa (1991)
12   concluded that the depletion  of GSH was not necessarily a critical factor in inhibiting GST.
13          Planas-Bohne and Elizdale (1992) reported that acute  exposure to Pb-nitrate
14   (100 |imol/kg) caused a significant increase in liver and kidney GST activity. Gel
15   electrophoresis analysis to evaluate the contribution of various GST isoforms indicated that
16   enhancement of liver GST activity was predominantly due to induction of GST isoform 7-7 in
17   liver compared to all isoforms in kidney. Liver GST-P isoform was reported to be induced by
18   both Pb-acetate and Pb-nitrate (Boyce and Mantle, 1993; Koo et al., 1994). This transient
19   induction of GST-P has been regulated at transcription, post-transcription, and post-translational
20   levels.  Suzuki et al. (1996) utilized a transgenic approach to investigate the transcriptional
21   regulation of GST-P induced by Pb and identified glutathione S-transferase P enhancer I (GPEI),
22   an enhancer (whose core consists of two AP-1 site-like sequences) located at the 5' flanking
23   region of this gene. The authors demonstrated that GPEI is an essential element in the activation
24   of the GST-P by Pb and that  the trans activating factor AP-1 is likely to be involved, at least in
25   part, in the transcriptional activation of the GST-P gene by Pb via the GPEI sequence.
26          Daggett et al.  (1997, 1998) investigated the effect of inorganic and organic Pb on liver
27   GST expression and other phase II detoxifying enzymes in rat liver and kidney. Triethyl Pb
28   chloride (TEL) injection (10  mg/kg body wt) decreased liver GST activity, as well as levels of
29   various other GST isoforms (Daggett et al., 1997), in contrast to significant induction of kidney
30   GST activity,  suggesting that a single compound,  TEL, had opposite effects on the expression of
31   GST isozymes and indicated the complexity of GST regulation.  Similarly, this group also

     December 2005                           5-261        DRAFT-DO NOT QUOTE OR CITE

-------
 1    reported that a single injection of Pb-acetate (114 mg/kg body wt) reduced GSH levels, increased
 2    production of malondialdehyde (MDA), and did not change the expression of various GST
 3    isoforms analyzed, except GST-pl on repeated injection (Daggett et al., 1998). Similar to
 4    studies with TEL, Pb-acetate also increased the expression of GST enzyme activity and
 5    expression of various isoforms without changing GSH and MDA levels, suggesting that
 6    oxidative stress may not be mediating the toxicity in kidney. On the other hand, TEL exposure
 7    was found to decrease microsomal estradiol metabolism (Odenbro and Rafter, 1988). The
 8    suppression of GST expression reported by Daggett et al. (1997, 1998) is in contrast to the
 9    induction of GST reported by various other groups discussed earlier. Other GSH-dependent
10    enzymes (i.e., GSH peroxidase, GSH reductase) have been found to be suppressed with a
11    simultaneous increase in oxidized GSH (GSSG) and a reduction in GSH/GSSG ratio (Sandhir
12    and Gill, 1995).  More detailed information on these and related studies is summarized in
13    Table AX5-10.1.
14
15    5.10.1.2 Biochemical and Molecular Perturbations in Lead-Induced Liver Tissue Injury
16          Oskarsson and Hellstrom-Lindahl et al. (1989) studied the cellular transport of Pb (203Pb),
17    in rat hepatocytes using dithiocarbamate (DTC). Cells treated with Pb-acetate and Pb-DTC
18    lipophylic complex demonstrated increased cytosolic Pb levels compared to Pb alone.  This was
19    further evaluated by measuring levels of ALAD. Cells treated with Pb-DTC complex showed
20    rapid and stronger inhibition of ALAD compared to Pb-acetate, suggesting that this inhibition
21    was due to increased mobilization of Pb into cells treated with Pb-DTC complex. Another report
22    by the same group, Hellstrom-Lindahl and Oskarsson (1990), suggested that the increased
23    inhibition of ALAD was due to the release of Pb from the Pb-DTC complex by decomposition.
24    Using  the mouse strain with a duplication of the ALAD gene (DBA), Claudio et al. (1997)
25    reported increased accumulation of Pb in this strain by many fold as compared to mice with a
26    single  copy of the ALAD gene (C57).
27          A single injection of Pb-nitrate was reported to cause hepatic hyperplasia correlating with
28    hepatic de novo  synthesis of cholesterol along with alterations in glucose and lipid metabolism
29    leading to altered serum lipid profiles (Dessi et al., 1984; Pani et al., 1984). Mobilization of
30    hepatic glycogen and altered gluconeogenic enzymes, including differential expression of G6PD,
31    have been reported following Pb exposure (Batetta et  al., 1990; Hacker et al., 1990). Chronic Pb

      December 2005                           5-262       DRAFT-DO NOT QUOTE OR CITE

-------
 1    intoxication has also been reported to inhibit gluconeogenic enzymes, alterations that were
 2    implicated in Pb bio-transformation rather than liver cell proliferation in Wistar rats (Calabrese
 3    and Baldwin,  1992). Increased levels of serum lipid peroxide (LPO) were also observed in rats
 4    given SC injection of Pb-acetate, supporting similar increased levels of serum LPO in humans
 5    exposed to Pb (Ito et al., 1985). These initial studies suggest that alterations in liver intermediary
 6    metabolism occur on exposure to Pb with a role for Pb-induced LPO in hepatotoxicity and
 7    potential involvement of oxidative stress in Pb toxicity.
 8          Dessi et al. (1990 ) investigated the role of fasting on Pb-induced hepatic hyperplasia by
 9    monitoring the activities of enzymes involved in cholesterol synthesis and the hexose
10    monophosphate shunt and reported that stimulation of these enzymes, even in Pb-acetate-treated
11    fasting rats, supported the role of new endogenous synthesis of cholesterol and gluconeogenic
12    mechanisms in Pb-induced hepatic cell proliferation.  Chronic exposure to Pb was found to
13    increase the arachidonate/linoleic acid ratio in liver and serum (Donaldson and Leeming, 1984;
14    Donaldson et al., 1985) along with the GSG concentration (McGowan and Donaldson, 1987).
15    As GSH and arachidonate are precursors for peptido-leukotrienes, Donaldson's group
16    investigated the potential effects of dietary Pb on levels of fatty acids, peptido-leukotrienes, and
17    arachidonate/linoleic ratios in chicken fed with diets low in calcium and methionine.  These
18    investigations found similar increases in arachidonate/linoelic acid ratio and in GSH levels
19    without bearing on peptido-leukotriene levels.  The authors also found the influence of a low
20    calcium  and methionine diet on Pb-induced serum fatty acid profiles (Knowles and Donaldson,
21    1990).
22          Chronic sublethal exposure (5 ppm Pb-nitrate for 30 days) has been found to alter liver
23    lipid profiles in blood and liver tissue of the fresh water fish Anabas testudineus (Tulasi et al.,
24    1992). These authors reported significant increases in liver total lipids, cholesterol, and free fatty
25    acids. Tandon et al. (1994b) reported that iron deficiency enhanced the accumulation of Pb in
26    liver and kidney and also increased liver calcium levels.  Induced expression of metallothionein
27    (MT) in  renal  and intestine was also observed in iron deficiency.  Han et al. (1996) investigated
28    the effect of Pb burden on weight loss using an energy restriction diet regimen on rats with prior
29    Pb exposure.  The authors reported that rats on a substantial weight loss regimen (40% of normal
30    calories) exhibited a significant increase in the quantity and concentration of liver Pb and a
31    decrease in the concentration of other metals (e.g., Ca, Cu, Mg, Zn).  The authors concluded that

      December 2005                           5-263        DRAFT-DO NOT QUOTE OR CITE

-------
 1    weight loss can increase the liver concentration of Pb, even in the absence of continued
 2    exposure.  Combined exposure to Pb (70 mg/kg) and Cd (20 mg/kg) in Buffalo rats for 7 weeks
 3    was found to alter liver levels of Zn and Cu, with less accumulation of Pb and Cd, compared to
 4    individuals exposure to either Pb or Cd alone (Skoczynska et al., 1993). These authors also
 5    reported that a combined exposure regimen interfered with serum lipid profiles (Skoczynska and
 6    Smolik,  1994).
 7          Liu et al. (1997) utilized rat primary hepatocyte cultures to explore the protective effect of
 8    Zn-induced expression of metallothionein (MT) in Pb toxicity. These authors found that, in the
 9    control cells without prior Zn exposure, most of the Pb was found bound to high-molecular
10    weight proteins in the cytosol, while in the Zn pretreated cells, a majority of Pb bound to MT,
11    indicating a MT-mediated protection against Pb toxicity to hepatocytes. More details about these
12    and related studies are summarized in Table AX5-10.2.
13
14    5.10.1.3 Effects of Lead Exposure on Hepatic Cholesterol Metabolism
15          Lead-nitrate-induced hyperplasia or liver cell proliferation involves simultaneous increase
16    in both liver and serum total cholesterol levels.  Recent studies have reported various molecular
17    events associated with this process. Induction of gene expression for CYP51 (Lanosterol
18    14a-demethylase), an essential enzyme for cholesterol biosynthesis, was reported in Pb-nitrate-
19    induced  liver hyperplasia, although other cytochrome P450 enzymes involved in drug
20    metabolism have been reported as being suppressed,  as discussed in earlier sections. This gene
21    has various regulatory elements and its constitutive expression in liver is mediated by sterol
22    regulatory element (SRE) and by the SRE binding proteins-la, 2, and Ic. Kojima et al. (2002)
23    reported that Pb-nitrate induced the expression of CYP51 in the livers of both immature (4-week-
24    old) and mature (7-week-old) rats and that this induction appeared to be mediated by the
25    upregulation of SRE binding protein-2. However, this increased synthesis of cholesterol
26    observed in rat liver was not mediated by endogenous feedback regulation by sterols, as no
27    decrease in serum total cholesterol was observed. To understand the molecular mechanisms
28    involved in the Pb-nitrate-mediated development of hepatic hypercholesterolemia, Kojima et al.
29    (2004) investigated the expression of various enzymes involved in cholesterol homeostasis,
30    including some of the associated transcription factors in male rats exposed to Pb-nitrate
31    (100 jimol/kg body wt). The authors reported that Pb-nitrate exposure caused a significant

      December 2005                            5-264       DRAFT-DO NOT  QUOTE OR CITE

-------
 1   increase in liver and serum total cholesterol levels at 3-72 h and 12-72 h, respectively. The
 2   enzymes involved in cholesterol biosynthesis viz. (i.e., 3-hydroxy-3methyglutaryl-CoA
 3   reductase, farnesyl diphosphate synthase, squalene synthase, CYP51) were all activated (3-24 h),
 4   while the enzymes involved in cholesterol catabolism such as 7a-hydroxylase were remarkably
 5   suppressed 3-72 h. Figure 5-10.1 shows the involvement of Pb at various stages of the
 6   cholesterol synthesis pathway. The induction of the cytokines interleukin-loc and TNF-a in rat
 7   liver prior to the induction of the genes for these synthesis enzymes suggested that Pb-nitrate-
 8   induced cholesterol synthesis is independent of sterol homeostasis regulation. Following
 9   gestational and lactational exposure to Pb-acetate (0.05 mg/kg body wt), Pillai and Gupta (2005)
10   reported that the activities of the hepatic steroid metabolizing enzyme 17-p-hydroxy steroid
11   reductase, UDP glucouronyl transferase, and  CYP450 levels decreased in rat pups on PND21.
                       Acetyl-CoA
HMGR-3-hydroxy-3 methyl glutaryl- I Pb
CoA Reductase
                           I
                        HMG-CoA
                           Jl
                           "^
                       Mevalonate
                           J |       fpps     Farnesyl pyrophosphate synthase T Pb
                Farnesyl pyrophosphate
                           JJ.       SQS      Squalene synthase T Pb
                        Squalene
                            I I       Squaline oxidosterol
                       Lanosterol
                            II       cyp51     Lanosterol 14-a-demethylase T Pb
                       Cholesterol
                            I I       cyp7A1   Cholesterol 7-a-hydroxyiase t  Pb
                        Bile acids

     Figure 5-10.1. Flow diagram indicating the Pb effects on the cholesterol synthesis pathway.
     December 2005
5-265
              DRAFT-DO NOT QUOTE OR CITE

-------
 1          Alterations in the hepatic system of neonates and pups (at PND12 and PND21) after
 2    gestational and lactational exposure to Pb-acetate (300 mg/L) have been reported by Corpas et al.
 3    (2002). The authors found significant reductions in the liver weight of pups and in hepatic
 4    glycogen that correlated with increased blood glucose levels. The authors also reported
 5    reductions in liver protein, lipid levels, and alkaline and acid phosphatase activities but did not
 6    find any gross structural alterations in liver tissue.  These and other studies are summarized in
 7    Table AX5-10.3.
 8
 9    5.10.1.4 Effect of Chelation Therapy on Lead-Induced Hepatic Oxidative Stress
10          Although several mechanisms  have been proposed to explain Pb toxicity, no mechanism
11    has been defined explicitly. Recent literature on Pb toxicity suggests oxidative stress as one of
12    the important mechanisms of toxic effects of Pb in  liver, kidneys, brain, and other organs.
13          Schematic representation of the various mechanisms by which Pb induces lipid
14    peroxidation is shown Figure 5-10.2.  Lead toxicity to the liver has been found to be associated
15    with significant accumulation of Pb in the liver. This results in the accentuation of lipid
16    peroxidation with concomitant inhibition of antioxidant enzymes (i.e., SOD, catalase, GSH
17    peroxidase,  GSH reductase) and a simultaneous increase in GSSG with a reduction in
18    GSH/GSSG ratio (Sandhir and Gill, 1995; Aykin-Burns et al., 2003). However, Furono et al.
19    (1996) studied the potential of various redox-active metals to induce LPO in normal and alpha-
20    linolenic acid-loaded rat hepatocytes and suggested that Pb ions were not capable of inducing
21    lipid peroxidation in such hepatocytes.


                                Pb-Induced Oxidative

               I                               I                              I
       Cell membrane:             Genome:                   Antioxidant system:
       • Interaction with PUFA           * Accumulated ALA acts as      * Inhibition of functional
       • FA chain elongation               alkylating agent and forms       sulfhydryl groups on enzymes:
       «Altered membrane enzymes        DNA adducts                 SOD, GSH peroxidase,
       «Altered solute transport          * Interference with Zn-binding     GSH reductase
       * Altered signal transduction         proteins in transcription        • Depletion of GSH
                                       complex
      Figure 5-10.2.  Schematic diagram illustrating the mode of Pb-induced lipid peroxidation.
      December 2005                           5-266       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The currently approved clinical intervention method is to give chelating agents that form
 2    an insoluble complex with Pb and remove the same from Pb-burdened tissues. The efficacy of
 3    various chelating agents and antioxidants studied in experimental animals on Pb induced liver
 4    toxicity is discussed below.
 5          Chelation therapy with mono-3-methylbutane-l-yl (monoisomyl) ester of meso-2,3-
 6    dimercaptosuccinic acid (Mi-DMSA) and meso-DMSA (meso-2,3-dimercaptosuccinic acid) was
 7    found to offer no protection to suckling rat pups as measured by liver Pb levels (Cory-Slechta,
 8    1988; Blanusa et al., 1995; Pappas et al.,  1995; Smith et al., 2000). Similar studies by Kostial
 9    et al. (1999), using various isoforms of DMSA, EDTA, and combined therapy, did not find a
10    chelator-mediated reduction in liver Pb levels except for meso-DMSA (0.5 mmol/kg), which
11    caused significant reduction in Pb levels in kidney and brain. On the other hand, it decreased
12    liver zinc and  copper levels.  The authors concluded that combined therapy may not be the best
13    choice at this age, because infants are more sensitive to trace metal deficiency.  Flora and Seth
14    (1999) investigated the protective role of S-adenosyl-L-methionine (SAM) on acute Pb and
15    Pb + thanol-induced hepatic toxicity in mice by monitoring hepatic GSH and MDA levels. The
16    authors concluded that bioaccumulation of Pb in liver in both Pb- and Pb + ethanol-exposed
17    groups were significantly decreased by SAM.
18          To identify the efficacy of chelation therapy (mono or combined therapy) for acute Pb
19    poisoning in infants, Kostial  et al. (1999) utilized suckling rat pups and monitored tissue Pb
20    levels. Monotherapy of either EDTA (0.3 mmol/kg), meso-DMSA (0.5 mmol/kg), rac-DMSA
21    (racemic-2,3-meso-2,3-dimercaptosuccinic acid, 0.5 mmol/kg), or a combined therapy of
22    EDTA + meso-DMSA, EDTA + rac-DMSO indicated differential effects on liver tissue Pb and
23    other trace metal levels. The authors concluded that meso-DMSA was the more potent therapy
24    for acute Pb poisoning in infants and suggested that combined therapy may not be the best
25    choice, as at this age the infants  are more sensitive to trace metal deficiency.
26          Supplementation with sodium molybdate (1 mg/kg body wt) during the course  of Pb
27    exposure (0.1% Pb-acetate in water for 4 weeks) was found to provide significant protection
28    from the uptake of Pb by blood,  liver and kidneys and also from hepatic LPO (Flora et al.,  1993).
29    Similarly, supplementation with antioxidants and vitamins were explored to reduce the toxic
30    effects of Pb on liver function and activity. Oral supplementation of vitamin C (100 mg/kg for
31    3 days) has been reported to provide significant protection against Pb-induced declines in liver

      December 2005                           5-267       DRAFT-DO NOT QUOTE OR CITE

-------
 1    heme synthesis, drug metabolism, tissue thiols, and vitamin C levels along with reduction in liver
 2    Pb levels (Vij et al.,  1998). Similarly, simultaneous administration of vitamin E (5 mg/kg body
 3    wt) was reported to confer protection against Pb-induced decline in hepatic type-1 iodothyronine
 4    5'-monodeiodinase activity, inhibition of SOD and catalase activities, and increased lipid
 5    peroxidation (Chaurasia and Kar, 1997). Studies by Tandon et al. (1997) also suggested that Pb
 6    and Pb + ethanol-induced biochemical changes in mouse liver can be prevented by the
 7    simultaneous administration of lysine and zinc.  This regimen was also reported to prevent the
 8    Pb-induced depletion of endogenous calcium and magnesium in liver. Two well-known
 9    antioxidants, N-acetylcysteine and lipoic acid, have been reported to reduce Pb-induced
10    oxidative stress (OS) both in vitro in Chinese hamster ovary  cells and in vivo in F344 rats (Ercal
11    et al., 1996; Gurer et al.,  1998, 1999b). The same group also investigated the protective effects
12    of another  antioxidant, taurine, against Pb-induced OS in the same systems both in vitro and in
13    vivo. These authors reported that taurine was effective by increasing cellular GSH while
14    simultaneously reducing  malondiablehde (MAD) and catalase activity levels, offering protection
15    against Pb-acetate-induced OS, without decreasing the liver or blood Pb levels (Gurer et al.,
16    2001).  Patra et al. (2001) studied the ameliorative effects of antioxidants (i.e., ascorbic acid,
17    vitamin E,  L-methionine) alone and vitamin E + EDTA on Pb-induced OS in liver, kidney,  and
18    brain tissues of rats exposed to Pb-acetate (1 mg/kg body wt, 4 weeks) and found that all the
19    antioxidants used conferred protection against OS without a  significant decline in tissue Pb
20    burden. The level of protection conferred exhibited tissue-specific differences. L-Methionine
21    was also found to offer similar protection in mice exposed to Pb (Xie et al., 2003). Othman and
22    El Missiry (1998) reported that administration of selenium (sodium selinite, 10 |iM/kg body wt)
23    prior to Pb-acetate (100 |iM/kg body wt) produced pronounced prophylactic action against Pb-
24    induced LPO in liver and kidney of male albino rats.
25          In earlier combination chelation therapy using thiamine and Ca2+-EDTA, Kim et al.
26    (1992) reported that regardless of the route of exposure, reduction in liver tissue retention of
27    203Pb occurred, while thiamine alone reduced only the Pb content of kidney. Recent studies used
28    a combination of chelators with antioxidants to reduce Pb-induced OS in liver and other tissues
29    (i.e., kidney and brain). a-Lipoic acid, meso-DMSA, and their combination was found to reduce
30    OS by increasing hepatic GSH levels and reducing GSSG and thiobarbituric acid reactive
31    substances (Pande and Flora, 2002). The same group also studied the protective effect of the

      December  2005                           5-268       DRAFT-DO NOT QUOTE OR CITE

-------
 1    combination of ascorbic acid, vitamin E, meso-DMSA, and miADMSA and found a significant
 2    reduction in hepatic OS by the combination therapy of ascorbic acid and thiol chelators (i.e.,
 3    DMSA, miADMSA) in rat. The combination therapy also produced similar reduction in renal
 4    OS (Flora et al., 2003). Studies reported by Varnai et al., (2003) suggested that ascorbic acid
 5    supplementation did not improve the efficiency of meso-DMSA in reducing Pb-induced OS in
 6    suckling rats.  On the other hand, combined treatment of ascorbic acid (1 mg/100 g body wt) and
 7    silymarian (1 mg/100 g body wt) has been reported (Shalan et al., 2005) to cause marked
 8    improvement of the biochemical, molecular and histopathological changes caused by Pb-acetate
 9    (500 mg/kg body wt).  Similarly, combined treatment with lipoic acid + DMSA has been found
10    to completely  ameliorate Pb-acetate-induced  oxidative damage. However, either lipoic acid or
11    DMSA alone conferred partial protection against Pb-induced hepatic damage (Sivaprasad et al.,
12    2004). These  and related studies are summarized in Table AX5-10.4.
13
14    5.10.1.5   Lead-Induced Liver Hyperplasia: Mediators and Molecular Mechanisms
15          The biochemical and molecular events associated with Pb-induced hyperplasia has been
16    accumulating in the scientific literature.  Lead-nitrate, a  known mitogen, is also considered to be
17    a carcinogen that induces liver cell proliferation in rats without any accompanying liver cell
18    necrosis.  It has been recognized that this proliferation is a transient process  and that apoptosis
19    plays a major role in the regression of Pb-nitrate-induced hepatic hyperplasia (Nakajima et al.,
20    1995). Columbano et al. (1996) studied the cell proliferation and regression phases by apoptosis
21    in Wistar male rat liver by monitoring the incorporation of tritiated thymidine as a marker for
22    increased DNA synthesis.  These studies demonstrated the production of Pb-induced
23    proliferation 3 days after a single injection of Pb-nitrate  with complete regression of hyperplasia
24    seen after 15 days. The authors suggested that the apoptosis process observed in the regression
25    phase also involved newly initiated hepatocytes. On the other hand, Dini et al. (1999) reported
26    the regressive or involutive phase as beginning 5 days post single injection of Pb-nitrate.
27    Apostoli et al. (2000) evaluated the proliferative effects  of various Pb salts (i.e., Pb-acetate, Pb-
28    chloride, Pb-monoxide, Pb-sulfate) using liver-derived REL cells. These authors reported that
29    all the Pb compounds tested showed dose- and time-dependent effects on the proliferation of
30    REL cells. Unlike other tumor promoters, Pb compounds did not exhibit effects on cell
31    junctional coupling. Liver hyperplasia induced by Pb-nitrate has been shown to demonstrate

      December 2005                           5-269       DRAFT-DO NOT QUOTE OR CITE

-------
 1    sexual dimorphism in all phases of the proliferation as well as in apoptosis (Tessitore et al.,
 2    1995).  Biochemical changes associated with liver hyperplasia in the intermediary metabolic
 3    pathways were discussed in earlier sections of this chapter; the present discussion focuses on
 4    other molecular characteristics of this process. As the numerous molecular networks involved in
 5    both the proliferation and apoptosis processes have many common mediators and pathways, it is
 6    very difficult to provide a discussion without an overlap.
 7          DNA hypomethylation has been recognized to play a major role in the proliferation of
 8    cells in regenerating and in hepatic pre-malignant lesions when compared to normal non-dividing
 9    liver cells. A single dose of Pb-nitrate (75 |iM/kg body wt) has been found to cause extensive
10    hypomethylation in rat liver (Kanduc et al., 1991). Additional investigations from the same
11    group reported that this hypomethylation status of liver DNA by Pb-nitrate changed significantly
12    with age and exhibited liver cell specificity (Kanduc and Frisco, 1992).
13          Investigations of cell cycle-dependent expression of proto-oncogenes in Pb-nitrate
14    (10 jiM/100 g body wt)-induced liver cell proliferation by Coni et al. (1989) showed that peak
15    DNA synthesis occurred at 36 h after a single injection of Pb-nitrate. In addition to DNA
16    synthesis, induced expression of c-fos, c-myc, and c-Ha-ras oncogenes was also observed in rat
17    liver tissue. Additional studies by the same group reported that Pb-nitrate-induced liver
18    hyperplasia involved an increased expression of c-jun in the absence of c-fos expression (Coni
19    et al., 1993). The induced expression of c-myc persisted up to 40 h post Pb-nitrate exposure.
20    Pb-nitrate-induced liver proliferation and DNA synthesis, as monitored by 5-bromo-2-
21    deoxyuridine immunohistochemistry, lead to DNA labeling in a few hepatocytes (Rijhsinghani
22    et al., 1993). The observed DNA synthesis appeared to be due to the increased activity and
23    expression of DNA polymerase-a observed at 8 h postexposure to a single injection of Pb-nitrate
24    (Menegazzi et al., 1992). Along with DNA synthesis, poly (ADP-ribose) polymerase was also
25    induced by Pb-nitrate (Menegazzi et al., 1990). Differential activation of various PKC isoforms,
26    downregulation of PKC-a, and marked activation of PKC-e in Pb-nitrate-mediated liver
27    hyperplasia suggested the involvement of these PKC enzymes in DNA synthesis and related
28    signal transduction pathways (Tessitore et al., 1994;  Liu et al., 1997).
29          Coni et al. (1992) reported the proliferation of normal and pre-neoplastic hepatic cells
30    treated  with the plasma derived from male Wistar rats treated with a single injection of Pb-
31    nitrate; this was the first report on the secretion of biological cell proliferation signals in the liver

      December 2005                           5-270        DRAFT-DO NOT QUOTE OR CITE

-------
 1    after Pb-nitrate treatment. These authors reported that DNA synthesis was detected as early as
 2    30 min and persisted up to 5 days after Pb-nitrate exposure.  This observation has opened up the
 3    inquiry into the involvement of various growth factors and other biological mediators in hepatic
 4    hyperplasia.  Shinozuka et al. (1994) investigated the expression of various growth factors (i.e.,
 5    hepatocyte growth factor, TGF-a, TGF-P) in rat liver after a single injection of Pb-nitrate
 6    (100 |iM/kg body wt) and reported the involvement of these growth factors in liver cell
 7    proliferation.  Additional studies by this group to observe LPS sensitivity in rats given Pb nitrate
 8    reported that animals given a single injection of LPS up to 100 jig survived, whereas in the
 9    presence of Pb-nitrate, they tolerated only 6 jig of LPS,  indicating that Pb-nitrate may sensitize
10    the animals for LPS toxicity.
11          Earlier studies by Honchel et al. (1991) reported that coexposure of rats to Pb-acetate
12    (15 mg/kg) and LPS or TNF showed markedly increased serum levels for various liver injury
13    parameters. They concluded that Pb may potentiate liver toxicity by LPS via a TNF-mediated
14    pathway. The role of TNF-a in Pb-nitrate-induced liver cell proliferation was further
15    investigated by (Ledda-Columbano et al., 1994) who demonstrated the inhibition of Pb-nitrate-
16    induced cell proliferation by pretreatment with dexamethasone, an inhibitor of TNF-a
17    expression. Additional studies by the same group evaluated the liver cell specificity in Pb-
18    nitrate-induced cell proliferation (Shinozuka et al., 1996). They monitored the incorporation of
19    5-bromo-2-deoxyuridine by immunohistochemical analysis on rat liver as induced by Pb-nitrate
20    and TNF-a and observed 5-bromo-2-deoxyuridine incorporation in hepatocytes and non-
21    parenchymal cells (i.e., Kupffer cells, endothelial cells, periportal nondescript cells), confirming
22    that Pb-induced liver cell proliferation was mediated by TNF-a.  Kubo et al. (1996) used various
23    TNF-a inhibitors to further confirm the role of TNF-a in Pb-nitrate-induced hepatocyte
24    proliferation.  Menegazzi et al. (1997) reported that Pb-nitrate induced proliferation involved the
25    induction of iNOS along with  TNF-a and that appeared to be mediated by a strong, prolonged
26    activation of NFKB but not activator protein-1 (AP-1). Nemoto et al. (2000) investigated the
27    potential role  of neurotrophins and their receptors in Pb-nitrate-induced hepatic hyperplasia. The
28    expression profile of TNF-a, neurotrophins (i.e., nerve growth factor, brain-derived neurotrophic
29    factor neurotrophin-3 and (their receptors), tyrosine kinase receptor (Trk) and neurotrophin
30    receptor (p75NTR) were investigated in liver tissue after a single injection of Pb-nitrate
31    (100 jiM/kg body wt).  The Pb-nitrate induced increased expression of TNF-a preceded the

      December 2005                           5-271       DRAFT-DO NOT QUOTE OR  CITE

-------
 1    expression of the neurotrophins and their receptors. Based on these results, the author's
 2    suggested that neurotrophins and neurotrophin receptors are involved in mediating mitogenic
 3    signals related to hepatic hyperplasia.
 4          The regression phase of Pb-induced liver hyperplasia appears to be mediated by OS.
 5    As discussed earlier, this process involves LPO and other cytokine mediators, including TNF-a.
 6    Sieg and Billings (1997) reported that Pb potentiated cytokine-induced OS, producing a
 7    significant decline in intracellular ATP concentration in mouse hepatocyte culture studies.  The
 8    authors suggested that cytotoxic interaction between Pb and cytokines (e.g., TNF-a and IFN)
 9    may be mediated by oxidative DNA damage resulting from OS.  The potential role OS along
10    with TNF-a has been implicated in the apoptosis of hepatocytes by Milosevic and Maier (2000).
11    Using freshly isolated cultures of hepatocytes and Kupffer  cells and their co-culture system
12    exposed to Pb-acetate (2-50 |iM) and LPS (0.1-1000 ng/mL), the authors reported that, in the
13    co-culture system, the Pb-LPS-induced release of TNF-a from the Kupffer cells, increased nitric
14    oxide levels by 6-fold and downregulated the acute phase protein, albumin, in hepatocytes.
15    From these observations the authors concluded that Pb-induced Kupffer cell-derived signals
16    promoted the toxicity of Pb in hepatocytes,  resulting in hepatocyte death by proteolysis.  The
17    importance of the Kupffer cells role in Pb-nitrate-induced heptatocyte apoptosis was further
18    demonstrated (Pagliara et al., 2003a,b).  These authors reported that in vivo hepatic apoptosis
19    including oxidative response induced by Pb-nitrate, was prevented by pretreatment with
20    gadolinium chloride, a Kupffer cell toxicant that specifically suppresses Kupffer cell activity.
21    When treated hepatocytes were exposed in vitro to Pb-nitrate, hepatocyte apoptosis was not
22    observed. On the other hand, hepatocyte apoptosis was evident when the hepatocytes were
23    incubated with culture medium derived from Kupffer cells  that had been exposed to Pb-nitrate.
24    Based on these studies, the authors concluded that heptocyte apoptosis was potentiated by
25    soluble factors  secreted by  Pb-exposed Kupffer cells. The  role of activated Kupffer cells,
26    macrophages, and TNF-a in chemical-induced hepatotoxicity is presented schematically in
27    Figure 5-10.3.
28          Dini et al. (1993) investigated the expression of asialoglycoprotein receptors on the
29    surface of hepatocytes and  galactose-specific receptors of non-parenchymal cells during the
30    apoptic phase of Pb-induced hepatic hyperplasia. A significant increase in asialoglycoprotein
      December 2005                           5-272        DRAFT-DO NOT QUOTE OR CITE

-------
                                                Chemical Exposure
                                              c
 Tissue Necrosis
                                               Macrophage Activation
         :ivation j
                                             Secretion of Proinflammatory \  t TNF-a
                                                   Cytokines       )  ' IL-1
                               (  Apoptosis  J
(  Inflammation )
    f Proliferation  J
                                              Reactive Oxygen Species
                                                  Nitric Oxide
      Figure 5-10.3.   Hypothesis of chemical-induced liver injury generated primarily on the
                      basis of different types of inhibitors.
 1    receptor expression in hepatocytes coincided with massive apoptosis. Later studies from this
 2    group demonstrated that sinusoidal liver cells predominantly phagocytosed the Pb-nitrate-
 3    induced apoptic hepatic cells and concluded that this process appeared to be mediated by the cell
 4    surface carbohydrate receptors (i.e., mannose and galactose receptors) (Ruzittu et al., 1999).
 5    Pretreatment of rats with gadolinium chloride, a kupffer cell toxicant, was also found to abolish
 6    the altered expression of galactose receptors (Pagliara et al., 2003b).
 7          The role of glucocorticoid-mediated signal transduction in the hepatotoxicity of Pb was
 8    evaluated by Heiman and Tonner (1995), using H4-IIE-C3 hepatoma cells (HTC). Acute
 9    exposure of cells to Pb (300 nMT1 or 10 jiM) was found to inhibit processes involved in
10    glucocorticoid-mediated enzyme induction (e.g., tyrosine  aminotransferase activity) in a dose-
11    dependent manner both at the transcriptional and translational  level, without altering
12    glucocorticoid receptor binding characteristics.  Tonner and Heiman (1997) also reported
13    Pb-induced hepatotoxicity by glucocorticoid-mediated signaling and its involvement in the
      December 2005
5-273
DRAFT-DO NOT QUOTE OR CITE

-------
 1    interference with calcium-mediated events as well as the differential modulation and
 2    translocation of protein kinase isoforms a and P into the nucleus. More information on these and
 3    other related studies is summarized in Table AX5-10.5.
 4
 5    5.10.1.6 Effects of Lead on Liver Heme Synthesis
 6          Effects of Pb on heme metabolism have been extensively investigated in major target
 7    tissues such as liver and erythrocytes.  Section 5.2 described Pb effects on heme synthesis, with
 8    particular relevance to erythrocytes. The effects of Pb on heme synthesis in the liver and the role
 9    of chelation therapy in this process are discussed in this section.
10          Fifteen percent of heme is produced in the liver. Heme metabolism in the liver is an
11    essential component of various cytochrome P450s that participate in cellular redox reactions and
12    xenobiotic detoxification pathways in the liver tissue and, hence, heme plays a vital role in liver
13    function (Jover et al.,  1996). Due to the important and critical role of heme in liver function,
14    Pb-induced effects on hepatic heme metabolism are discussed below.
15          Initial studies on the effects of Pb-nitrate on hepatic heme biosynthesis were reported by
16    Lake and Gerschenson (1978) using the rat liver cell line (RLC-GAI).  The effects of various
17    organic metal compounds on ALAD activity have been studied by Bondy  (1986). The authors
18    reported that triethyl Pb-chloride has the same potency as Pb-nitrate in inhibiting ALAD both in
19    vitro and in vivo, with liver and blood ALAD exhibiting similar sensitivities to Pb compounds.
20    By measuring the conversion of ALA into heme, these authors showed that heme biosynthesis
21    was inhibited by Pb in a dose dependent manner.  Using a lipophilic complex of Pb-acetate +
22    DTC to increase the cellular uptake of Pb, Osksarsson et al. (1989) demonstrated the inhibition
23    of ALAD activity in primary rat hepatocytes cultures. Lead-acetate has been reported to inhibit
24    ALAD activity in rabbit liver tissue without any effect on delta-aminolevulinic acid synthase
25    (ALA-synthase) activity (Zereba and Chemielnicka, 1992).  Exposure to Pb (500 ppm) in
26    drinking water did not inhibit hepatic ALA-synthase, but did inhibit ALA-dehydratase activity in
27    mice (Tomokuni et al., 1991). Exposure to Pb-acetate (20 mg/kg body wt for 3 days) has been
28    reported to decrease hepatic ALAD and uroporphyrinogen activity (Satija and Vij, 1995). These
29    authors also reported that IP injection of zinc (5 mg/kg body wt for 3 days) conferred protection
30    against Pb-acetate effects in liver tissue.


      December 2005                           5-274       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Effects of Pb on hepatic porphyrins, intermediate metabolites of heme metabolism, were
 2    investigated by few researchers. Quntanilla-Vega et al. (1995) reported that 3T3-hepatocyte
 3    cultures, when incubated with a micromolar concentration of Pb-acetate increased cellular
 4    porphyrin content and excretion. This increased porphyrin production may have been due to an
 5    accumulation of protoporphyrin and coproporphyrin, as in coproporphyrinuria, a well-
 6    characterized sign of Pb intoxication (Ichiba and Tomokuni, 1987; Zereba and Chemielnicka,
 7    1992). Dietary supplementation of selenium and monensin increased Pb-induced accumulation
 8    of prophyrins in chicken liver (Khan and Szarek, 1994).  Species-specific differences in the
 9    effects of Pb on protoporphyrins were reported by Jacobs et al. (1998).  These authors
10    investigated the effect of Pb on zinc protoporphyrin synthesis in cultured chick and rat
11    hepatocytes and observed decreased levels of protoporphyrin in rat hepatocytes, but no effect on
12    chick hepatocytes.  Santos et al. (1999) also reported Pb-induced derangements (including
13    porphyrin metabolism) in rat liver heme metabolism, but these effects were far less severe than
14    those observed in erythrocytes.  Their investigations on the effect of chronic alcoholism on Pb
15    effects in hepatic heme metabolism suggested no potentiation by alcohol.
16          Transferrin (TF) is the major iron-transport protein in serum and other biological fluids.
17    Transferrin can also has the capacity to transport other metals. Lead was found to inhibit TF
18    endocytosis and transport of iron across the cell membrane of rabbit reticulocytes (Qian and
19    Morgan, 1990).  The effect of Pb on TF gene expression was investigated by Adrian et al. (1993)
20    using a transgenic mouse with the human TF gene. They  found that Pb suppressed the
21    expression of TF transgene in mouse liver at the transcriptional level; however, the same dose of
22    Pb did not inhibit mouse endogenous hepatic TF gene expression.  Lead exposure was also found
23    to inhibit recombinant TF expression in human hepatoma hepG2 cells.  Other  studies by the
24    same group found that Pb exposure suppressed the expression of endogenous TF in HepG2 cells
25    (Barnum-Huckins et al., 1997).  These authors further suggested that Pb effects on hepatic TF
26    levels may also interfere with iron metabolism in humans. (See  Annex Table AX5-10.6 for more
27    information on these and related studies.)
28
29    5.10.1.7  Summary
30          Extensive in vivo and in vitro experimental evidence has  accumulated over the past
31    20 years and increased our understanding of the potential  toxic effects of Pb in the hepatic

      December 2005                           5-275       DRAFT-DO NOT QUOTE  OR CITE

-------
 1    system.  These studies ranged from simple biochemical studies to molecular characterizations of

 2    the induction of drug-metabolizing enzymes, liver hyperplasia, and the protective effects of
 3    chelation therapy.

 4       •  Rat liver microsomal cytochrome P-450 levels were found to decrease with a single dose
 5          exposure of Pb nitrate.  Inhibition of both constitutive and induced expression of
 6          microsomal P450 Al and A2 activity occurred. Simultaneous induction of the activities
 7          of phase II drug metabolizing enzymes with decreased phase I enzymes with single
 8          exposure to Pb nitrate suggests biochemical properties similar to hepatic nodules.

 9       •  Newer studies examined the induction of GST-P at both transcript onal and translational
10          levels using in vitro systems and indicated a role for Pb-nitrate and Pb-acetate in the
11          induction process. On the other hand, triethyl Pb compounds have been found to
12          suppress the activity of various GST isoforms.

13       •  Studies on Pb-induced liver hyperplasia demonstrated de novo synthesis of cholesterol,
14          alterations in the gluconeogenic mechanism, as well as DNA hypomethylation and
15          subsequent changes in the expression of protooncogenes.

16       •  Lead-induced alterations in cholesterol metabolism appear to be mediated by the
17          induction of several enzymes related to cholesterol metabolism and the decrease of 7
18          a-hydroxylase, a cholesterol catabolizing enzyme.  This regulation of cholesterol
19          homeostasis is modulated by changes in cytokine expression and related signaling.

20       •  Studies using an inhibitor to block TNF-a have clearly demonstrated TNF-a as one of the
21          major mitogenic signals that mediate Pb-nitrate-induced liver hyperplasia. Lead-induced
22          hyperplasia also appears to be modulated by neurotrophins and their receptors.

23       •   In vitro co-culture systems with Kupffer cells and hepatocytes suggested liver cell
24          apoptosis is mediated by Kupffer cell-derived signals and Pb-induced oxidative stress.

25       •  Newer experimental evidence suggests that Pb-induced alterations in liver heme
26          metabolism involves perturbations in ALAD activity, and porphyrin metabolism,
27          alterations in Transferrin gene expression, and associated changes in iron metabolism.

28       •  Limited experimental evidence on the role of weight loss on liver Pb burden in exposed
29          animals indicate that liver Pb content increases even in the absence of prolonged
30          continued exposure.

31       •  Extensive scientific evidence has accumulated over these two decades on the role of
32          chelation therapy, both individual and combined.  Studies using a combination of therapy
33          regimens with chelators such as DMSA, Mi-DMSA, or DMSA+ EDTA did not prove
34          beneficial in ameliorating the Pb-induced oxidative stress in infant/neonatal rats as the
35          combination therapy in young rats resulted in essential mineral deficiencies.

36       •  Therapeutic intervention with S-adenosyl-L-methionine, L-acetyl cysteine, lipoic acid,
37          and vitamin E conferred protection against Pb accumulation in the liver and Pb-induced
38          lipid peroxidation. Intervention with  ascorbic acid, on the other hand, has been found to
39          confer protection against Pb-induced decrease in hepatic heme synthesis.
40

      December 2005                           5-276       DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.10.2  Gastrointestinal System and Lead Absorption
 2          Lead enters the body by many routes, but primarily via the GI tract. The intestinal
 3    epithelium serves as one of the body's primary interfaces with the outside world.  The
 4    transporting epithelia in the small intestine are characterized by layers of anatomically and
 5    biochemically polarized cells that are connected to each other by tight junctions and resting on a
 6    basement membrane. Classically, the intestinal epithelium is thought of primarily as a barrier,
 7    but it also is a highly reactive barrier.  Even modest perturbations in its functions may lead to
 8    diarrhea, constipation, malnutrition, dehydration, and infectious diseases (i.e., ulcerative colitis,
 9    collectively referred as chronic intestinal inflammatory diseases) (Gewirtz et al., 2002).
10    Abdominal colic and constipation are symptoms of Pb poisoning, but its mechanism is not fully
11    understood. Studies have been carried out in the past decade to increase our understanding of the
12    fundamental mechanism(s) in order to extrapolate the experimental observations to human
13    health effects.
14          The intestinal absorption of Pb  is influenced by a variety of factors, including the
15    chemical and physical forms of the element, age at intake, and various nutritional factors.
16    Gastrointestinal absorption of Pb is thought to occur primarily in the duodenum.  In the  isolated
17    rat intestine, absorption, and, in particular, serosal Pb transfer activity (net transfer of Pb from
18    the small intestine lumen across the epithelium and into the serosal space) is highest in the
19    duodenum. The mechanisms of absorption may involve active transport and/or diffusion through
20    the intestinal epithelial cells.  Both saturable and non-saturable pathways of absorption have been
21    inferred from the studies in different animal models, although the understanding of the former is
22    slightly more robust (Diamond et al., 1998).
23          Transport of Pb as a complex with proteins via endocytosis or as a complex with amino
24    acids are postulated as possible mechanisms.  Direct evidence for transport of an organic Pb
25    complex has not been provided, but it seems possible.
26          In the cell, Pb interacts with a variety of intracellular ligands, including calcium-binding
27    proteins and high-affinity Pb-binding proteins. Transfer across  the cell or basolateral  membrane
28    (or both) involves a mechanism(s) that may be sensitive to  vitamin D and iron status.  Alternate
29    transport mechanisms via a Ca2+-Na+ exchanger, independent of regulation by vitamin D, are
30    also possible.
31

      December 2005                            5-277        DRAFT-DO NOT QUOTE  OR CITE

-------
 1    5.10.2.1 Lead and In vitro Cytotoxicity in Intestinal Cells
 2          In vitro cytotoxicity of metal salts for 48 h was determined in the intestinal epithelial cell
 3    line 1-407 by Keogh et al. (1994). The investigations identified rank order cytotoxicity in terms
 4    of LC50 values:  HgCl2 (32 |iM)>CdC!2 (53 |iM) > CuCl2 (156 |iM) > Ti2SO4 (377 |iM) > Pb
 5    (NC>3)2 (1.99 mM). Further studies using a noncytotoxic concentration of butathione
 6    sulphoxamine pretreatment for GSH depletion revealed that the cytotoxicity of Pb was
 7    unaffected by GSH depletion (see Table AX5-10.7).
 8
 9    5.10.2.2 Alterations in Intestinal Physiology and Ultrastructure
10          Karmakar et al.  (1986) investigated the pathologic alterations that occur in the intestine,
11    liver, and kidney of Pb-intoxicated rats upon short-term exposure to sublethal doses of Pb
12    (44 mg/Kg body wt) and reported degeneration of intestinal mucosal epithelium leading  to
13    potential malabsorption.
14          The effect of low-concentration Pb-acetate (0.1%) on the jejunal ultrastructure was
15    studied by Tomczok et al. (1988) in young male rats.  The studies revealed that the villi of
16    jejunum of rats exposed to Pb for 30 days had a rough appearance on the surface, which could be
17    associated with a distortion of glycocalyx layer. Areas of extensive degenerative lesions  were
18    also observed on the surface of most villi  on the 60th day of exposure. All intestinal epithelial
19    cells exhibited various degrees of glycocalyx disturbance, indicating that pronounced toxic
20    effects of Pb were related to modifications of the biochemical properties of the surface coat of
21    the cells. These authors also reported the appearance of goblet cells and of Pb deposition along
22    the goblet cell membrane in blocks of tissue along the border between duodenum and jejunum.
23    Continued treatment up to 60 days resulted in  mucus droplets in the cytoplasm of goblet cells,
24    along with deposition of silver salts indicative of Pb in these cells.  These results demonstrated
25    the significance of goblet cells in Pb detoxification.
26          In another study on the ultrastructure of rat jejunum exposed to Pb-acetate (100 mg/kg
27    body wt/day), Tomczok et al. (1991) found that 30-day treatment resulted in numerous small,
28    rough-membraned vesicles and dilated golgi complexes in the  cytoplasm. Continued treatment
29    for 60 days resulted in vacuolated cytoplasm associated with the golgi complexes, rough-
30    membraned vesicles, and dilated cisternae.  Also, the surface of the intestinal epithelial cell
31    microvilli showed evidence of Pb deposition, as evidenced by  Timm sulfide silver reaction sites.

      December 2005                           5-278        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.10.2.3 Intestinal Uptake and Transport
 2          Infants are a particularly susceptible population for Pb toxicity, possibly due to the
 3    immaturity of the digestive tract, feeding pattern, or source of Pb. To investigate these aspects,
 4    Henning's group (Beach and Henning, 1988; Henning and Cooper, 1988) carried out a series of
 5    experiments using suckling rat pups and reported that Pb in rat and bovine milk and infant milk
 6    formula was primarily associated with casein micelles. Casein-bound Pb may be the most
 7    common form of Pb presented to the small intestine (Beach and Henning, 1988). Other studies
 8    by this group investigated potential differences in the mechanisms when Pb was presented in
 9    ionic or milk-bound form, using 203Pb as a tracer. These studies clearly showed that when 203Pb
10    was administered intragastrically as a soluble salt, it was primarily accumulated in the
11    duodenum, regardless of dose or vehicle used.  In contrast, substantial accumulation of 203Pb was
12    found in the ileal tissue following Pb administration in milk.  These studies clearly indicated
13    strikingly different patterns in the intestinal accumulation of ionic and milk-bound Pb and
14    suggest a greater toxicity for Pb in drinking water compared  to Pb ingestion via milk (Henning
15    and Cooper, 1988).
16          Dekaney et al.  (1997) investigated the uptake and transport of Pb using intestinal
17    epithelial cells (IEC-6).  The authors observed that Pb accumulation in Pb-exposed (5-10 jiM)
18    IEC-6 cells was time- and dose-dependent up to 1 h and that reduction of the incubation
19    temperature significantly reduced the total cellular Pb content of IEC-6 cells.   Simultaneous
20    exposure to Zn resulted in decreased cellular Pb content compared to cells exposed to Pb only.
21    Exposure of cells to ouabin or  sodium azide has been found to increase Pb accumulation in the
22    cells compared to cells treated  with Pb (5 jiM) alone.  These studies clearly demonstrate that Pb
23    transport in IEC-6 cells is time- and temperature-dependent,  involves the presence of sulfydryl
24    groups, and competes with the uptake of Zn.
25          Lead speciation and transport across intestinal epithelium in artificial human digestive
26    fluid (chyme), both in vivo and in vitro, in Caco-2 cells were evaluated by Oomen et al. (2003).
27    In vivo studies indicated that in chyme, Pb-phosphate and Pb-bile complexes are important
28    fractions. The metal ions dissociated from these complexes can subsequently be transported
29    across the intestinal epithelium or they may traverse the intestinal membrane.  In vitro studies, on
30    the transport of bioaccessible Pb across the intestinal epithelium in Caco-2 cells exposed to
31    diluted artificial chyme for 24 h, indicated that  3% of the Pb  was transported across the cell

      December 2005                           5-279       DRAFT-DO NOT QUOTE OR CITE

-------
 1    monolayer. Lead associated with cells in a linear relationship to the total amount of Pb in the
 2    system.  Bile levels were not found to affect the fraction of Pb associated with the cells. The free
 3    Pb2+ concentration in chyme was negligible.  Extrapolating these results to the in vivo situation,
 4    the authors concluded that Pb species other than the free metal ion may have contributed to the
 5    Pb flux towards the cells, possibly involving the dissociation of labile Pb species, such as Pb-
 6    phosphate and Pb-labile complexes and the subsequent transport of the released free metal ions
 7    toward the intestinal membrane.
 8
 9    5.10.2.4 Alterations in Gastrointestinal Motility/Gastrointestinal Transit and Function
10          The effect of Pb on contractility of rat duodenum was determined in vivo in rats given an
11    oral dose of Pb-acetate (44 mg/kg per day, Pb as 53 mM/L for 4 weeks) to investigate the
12    possible mechanisms associated with Pb-induced abdominal colic and constipation (Karmakar
13    and Anand, 1989).  Deodenal motility and the amplitude of contractility of rat duodenum were
14    decreased significantly in the Pb-exposed rats, leading the authors to conclude that there was a
15    fundamental change in the contractility of the intestinal tract due to Pb intoxication.
16          Chronic Pb ingestion through drinking water (2-5 mg/mL, Pb-acetate for 55 days) caused
17    a 20-fold increase in urinary excretion of D-ALA and an increase in blood Pb level (80 |ig/dL),
18    without any perturbations in propulsive motility of guinea pig colon (Rizzi et al., 1989). On the
19    other hand, Lawrel  et al. (1991) observed no changes in gastric contractions during ingestion in
20    red-tailed hawks exposed to Pb-acetate (0.82 or 1.64 mg/kg body wt for 3 weeks). This low
21    level of exposure has also been found to have no bearing on  the regular passing of pellets of
22    undigested material. Shraideh (1999) studied the effect of triethyl Pb-chloride on the rhythmic
23    and peristalitic contractile activity of ileum isolated from Swiss mice. These authors observed
24    no significant effect below 40 jiM of TEL, while higher concentrations (40-120 jiM) caused
25    changes in contraction rhythm.  These studies also reported that TEL above 120 jiM induced
26    irreversible changes in the ileal contractile activity. These and related studies are summarized in
27    Table AX5-10.8.
28
29    5.10.2.5 Lead, Calcium, and Vitamin D Interactions in the Intestine
30          The complex biological interactions between Pb and calcium have been recognized and
31    demonstrated  in virtually every type of tissue. Studies of high-affinity Pb binding to intracellular

      December 2005                           5-280       DRAFT-DO NOT QUOTE OR CITE

-------
 1    calcium receptors and transport proteins, as well as the involvement of Pb in calcium-activated
 2    and calcium-regulated processes, have added to our understanding of the effects of Pb on
 3    biological processes at the cellular level. The intestinal absorption of Pb is influenced by a
 4    variety of factors, including chemical and physical forms of the element, age at intake, and
 5    various nutritional factors. Work dating back to the 1940s established that the deposition of Pb
 6    in bone and soft tissue significantly increases under conditions of dietary calcium and
 7    phosphorus deprivation or by the administration of vitamin D to rachitic animals.  Later, in the
 8    1970s, it was demonstrated that dietary calcium status was a major contributing factor
 9    determining relative susceptibility to Pb intoxication.
10          Fullmer's group (Fullmer and Rosen, 1990; Fullmer, 1991, 1992, 1997)  carried out a
11    series of studies to investigate the potential interaction between calcium and Pb in the ingestion
12    and intestinal absorption of Pb. Various parameters, such as absorption  kinetics for Ca and Pb,
13    activity of alkaline phosphatase, expression of the clabindin D gene, and the potential role of
14    endocrine function in this interaction (as assessed by cholecalciferol and its active hormonal
15    form, 1, 25-dihydroxycholecalciferol levels) were investigated. Fullmer and Rosen (1990)
16    observed that chicks fed with low (0.5%) and adequate (1.2%) dietary calcium and exposed to Pb
17    (0-0.8%) exhibited differential effects on intestinal Ca absorption depending on their dietary Ca
18    status.  In the chicks fed a low-calcium diet, Pb inhibited intestinal Ca absorption  and calbindin
19    D and alkaline phosphatase synthesis in a dose-dependent fashion.  On the other hand, chicks fed
20    the normal diet, showed no inhibition of Ca absorption.  Based on these results, the authors
21    postulated that Pb-induced alterations in intestinal Ca absorption  may involve cholecalciferol  and
22    the endocrine system. In  an extension of this study using young growing chicks, Fullmer (1991)
23    observed similar results in 2-week Pb-exposed, but not in 1-week exposed, chicks.
24          As dietary Ca deficiency is associated with a marked increase in the body burden of Pb
25    and in the susceptibility to Pb toxicity during chronic ingestion, Fullmer (1992) examined the
26    effects of vitamin D supplementation on intestinal Pb and Ca absorption. When vitamin D-
27    deficient chicks received physiologic amounts of vitamin D (O.lmg/day), intestinal 203Pb and
28    47Ca absorption rates were elevated by 4- and 8-fold, respectively. Along with  this, calbindin D
29    and alkaline phosphatase  activities were also found to be significantly elevated. Ingestion of
30    even the highest level of Pb (0.8 %) during the repletion phase had no effect on intestinal Ca
31    absorption.  To further understand the Pb-Ca interactions and the potential involvement of

      December 2005                           5-281        DRAFT-DO NOT QUOTE OR CITE

-------
 1    vitamin D on intestinal absorption, Fullmer (1997) evaluated serum levels of 1, 25-
 2    dihydroxyvitamin D.  Lead ingestion and Ca deficiency alone, or in combination, generally
 3    increased serum 1, 25-dihydroxyvitamin D levels over most of the ranges of Pb or Ca studied.
 4    However, in severe Ca deficiency, Pb ingestion resulted in marked decreases in serum 1, 25-
 5    dihydroxyvitamin D, intestinal Ca absorption, and calbindin D mRNA. From these studies using
 6    response surface models, Fullmer (1997) concluded that the interactions between Pb and Ca were
 7    mediated via changes in circulating 1, 25-dihydroxy vitamin D hormone, rather than via direct
 8    effects on the intestine.
 9          Similar to Ca deficiency, iron deficiency has also been found to increase intestinal
10    absorption of Pb, as indicated by increased blood and kidney Pb levels in iron-deficient rats
11    exposed to dietary Pb; but the mechanistic details are not known (Crowe and Morgan,  1996).
12    These and other related studies are summarized in Table AX5-10.9.
13
14    5.10.2.6 Lead and Intestinal Enzymes
15          Differential effects of Pb on intestinal brush border enzyme activity profiles were reported
16    by Gupta et al. (1994). Across a concentration range  of 0.5-6.0 mM, Pb-acetate was found to
17    significantly inhibit Ca-Mg-ATpase, g-glutamyl transpeptidase, and acetylcholineesterase
18    activities  in a dose-dependent manner without effects on alkaline phosphatase.
19          Cremin et al. (2001) investigated the effects of oral succimer on the intestinal absorption
20    of Pb in infant rhesus monkeys. These studies indicated that chelation therapy with DMSA for
21    two successive 19-day periods significantly decreased GI absorption of Pb and increased urinary
22    excretion of endogenous lead (see Table AX5-10.9).
23
24    5.10.2.7 Summary
25       •   Gastrointestinal absorption of Pb is influenced by a variety of factors, including chemical
26           and physical forms of the element, age at intake, and various nutritional factors. The
27           degeneration of intestinal mucosal epithelium leading to potential malabsorption and
28           alterations in the jejunal ultrastructure (possibly associated with distortion of glycocalyx
29           layer) have been reported in the intestine of Pb-exposed rats.
30       •   Lead in rat and bovine milk and, also, infant milk formula was demonstrated to be
31           primarily associated with caseine micelles.
32       •   Tracer studies using 203Pb indicated that intragastric administration of Pb as a soluble salt
33           resulted in Pb primarily accumulating in the deuodenum, regardless of dose or vehicle

      December 2005                           5-282        DRAFT-DO NOT QUOTE OR CITE

-------
 1          used, whereas Pb from milk was found to be taken up by ileal tissue.  Studies also
 2          suggested Pb ingestion through water was more toxic than ingestion through milk.
 3       •  Lead induced decreases in duodenal motility and amplitude of contractility of the
 4          intestinal tract has been reported for rats.
 5       •  Nutritional studies using various levels of Pb, Ca, and vitamin D in the diet indicate
 6          competition of Pb with Ca absorption. Supplementation with vitamin D has been
 7          reported to enhance intestinal absorption of Ca and lead. Physiological amounts of
 8          vitamin D administered to vitamin D-deficient rats resulted in elevated Pb and Ca levels.
 9          In the case of severe Ca deficiency, Pb ingestion results in a marked decrease in serum
10          1,25-hydroxy vitamin D.
11
12          Overall, our understanding of Pb effects on hepatic and  gastro intestinal systems using in
13    vitro cell culture models and in vivo animal models has increased greatly compared to the  1986
14    AQCD.  Significant insights have emerged regarding the role of Pb in hepatic cholesterol
15    synthesis, the role of inflammation in Pb-induced hepatotoxicity, and the contribution of newer
16    chelation therapy in the amelioration of Pb-induced oxidative burden.  Similarly, our knowledge
17    has greatly enhanced as to the absorption, transport, and toxicity of Pb in the gastrointestinal
18    tract.
19
20
21    5.11    LEAD-BINDING PROTEINS
22          Lead-binding proteins that are constitutively expressed within the cells and bind Pb can be
23    classified into two types of protein.  The first type  of Pb-binding proteins are inducible, i.e., their
24    concentration increases after exposure to Pb. The  second type  of Pb-binding proteins have
25    binding sites that are saturable by Pb, but no discernible increase in protein content occurs after
26    exposure to Pb.  The second type is, perhaps, most pertinent to enzymes that can be inhibited by
27    Pb.
28          The history of research on Pb-binding proteins dates back to 1936, when the presence of
29    intranuclear inclusion bodies  in the liver and kidney as manifestations of Pb poisoning was first
30    described (Blackman, 1936).  Later, detailed studies of the composition of renal tubular
31    intranuclear Pb inclusion bodies and consequent alterations in mitochondrial structure and
32    function followed.
33

      December 2005                           5-283        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.11.1  Lead-Binding Proteins Within Intranuclear Inclusion Bodies
 2            in Kidney
 3          Goyer (1968) examined the renal tubules of rats fed 1% Pb-acetate for up to 20 weeks,
 4    and found that dense, deeply staining intranuclear inclusions were located in the straight portion
 5    of the proximal tubules, accompanied by swollen, globular or ovoid, closely packed
 6    mitochondria with many marginated, irregular, or vesicular cristae. Accompanying these
 7    mitochondrial changes was the presence of generalized aminoaciduria. Goyer et al. (1968) also
 8    isolated mitochondria from Pb-exposed and control rats and demonstrated that mitochondria
 9    from the Pb-exposed rats showed reduced rates of respiration and oxidative phosphorylation.
10          Lead within the kidneys in Pb-poisoned rats was found to be concentrated in the nuclei
11    and, within nuclei, in the nuclear inclusion body (Goyer et al., 1970a,b).  Choie and Richter
12    (1972) showed that rapid induction of inclusion bodies by injections of Pb salts  in the rat resulted
13    in cytoplasmic inclusions, suggesting that they were precursors to the intranuclear inclusions.
14    Lead-containing nuclear inclusions were also found in organs other than the kidney, including
15    liver and glial cells of the central nervous system (Goyer and Rhyne, 1973).  Moore et al. (1973)
16    dissolved the rat renal intranuclear inclusions in strong denaturing agents and found that the
17    protein in the inclusions is acidic, with high levels of aspartic acid, glutamic acid, glycine, and
18    cystine.  Moore and Goyer (1974) later characterized the protein as a 27.5 kDa protein, which
19    migrates as a single band on acrylamide gel electrophoresis.  Repeated intraperitoneal injections
20    of CaNa2EDTA resulted in the disappearance of the inclusion bodies in Pb-exposed rats, together
21    with  a marked decrease in kidney Pb levels (Goyer et al.,  1978).
22          Shelton and co-workers have also explored the composition of Pb-binding proteins in the
23    nuclear inclusion proteins of Pb-exposed rat kidneys. Shelton and Egle (1982) first described a
24    32 kDa protein with an isoelectric point of 6.3, which was isolated from the kidneys of rats
25    treated with 1% Pb-acetate in rat chow or 0.75% Pb-acetate in drinking water for 13-17 weeks.
26    In contrast to Goyer and co-workers, they used two-dimensional gel electrophoresis to isolate the
27    protein from the nuclear inclusion bodies and demonstrated that it was present in Pb-exposed,
28    but not control, kidneys (hence, inducible).  This protein has been termed p32/6.3.  Inhibitor
29    studies with cycloheximide and actinomycin D (McLachlin et al., 1980; Choie et al., 1975) had
30    indicated earlier that protein synthesis was required for induction of the nuclear and cytoplasmic
31    inclusion bodies.

      December 2005                            5-284       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Egle and Shelton (1986) unexpectedly found that p32/6.3, now characterized by a
 2    monoclonal antibody, was constitutively present in the cerebral cortex, both in neurons and
 3    astrocytes.  The protein was concentrated in the insoluble nuclear protein, findings similar as for
 4    the Pb-exposed kidney. Brain p32/6.3 was detected in rat, mouse, dog, man, and chicken.  In rat
 5    brain, adult levels were achieved in 1 to 2 weeks after birth, whereas only trace amounts were
 6    found at 3 days. Brain p32/6.3 increased between postnatal days 10 to 12 in the guinea pig and
 7    days 15 to 21 in the rat, suggesting that the increase may be related in part to exposure to the
 8    external environment (Shelton et al., 1993). When neuroblastoma cells were cultured after 1-
 9    and 3-day exposure to Pb, the abundance of p32/6.3 increased. Simultaneous incubation with Pb
10    and cycloheximide or actinomycin D increased in p32/6.3, suggesting that Pb selectively retards
11    the degradation of the brain protein (Klann and Shelton, 1989).  The amino acid composition of
12    partially purified  p32/6.3 revealed a high percentage of glycine, aspartic and glutamic acid
13    (Shelton et al., 1990).  Thus, the inducible protein, p32/6.3, can be extracted from nuclear
14    inclusion bodies from the Pb-exposed rat kidney, and a similar or identical protein from adult rat
15    brain. Whether the brain protein is constitutive or inducible by exposure to environmental Pb
16    has yet to be determined.
17          Oskarsson and Fowler (1985) examined the influence of pretreatment with Pb  by a single
18    IP injection of Pb-acetate (50 mg Pb per kg) 1, 3, and 6 days before injecting 203Pb. Rats were
19    sacrificed 24 h later and the kidneys were examined both microscopically and for the distribution
20    of 203Pb.  At 3 days, rat kidneys displayed fibrillar cytoplasmic inclusions, but at 6 days, these
21    inclusions were less prominent and intranuclear inclusions were observed. 203Pb uptake at 6 days
22    was maximal in the purified nuclear fraction and in the nuclear inclusion bodies (7x and 20x
23    control, respectively).
24
25    5.11.2  Cytoplasmic Lead-Binding  Proteins  in Kidney and Brain
26          The remaining studies of non-Pb-stimulated cytoplasmic kidney and brain Pb-binding
27    proteins have been provided by Fowler and associates.
28          The first study (Oskarsson et al., 1982) reported on the Pb-binding proteins in  kidney
29    postmitochondrial cytosolic fractions. Binding of 203Pb was found in two protein fractions of
30    control kidneys with molecular weights of  11.5 and 63 kDa. Binding was markedly decreased
31    after Pb pretreatment.  The  use of cadmium to stimulate metallothionein synthesis did not

      December 2005                           5-285       DRAFT-DO NOT QUOTE OR CITE

-------
 1    increase 203Pb binding to the 11.5 kDa protein. The two binding proteins were also present in
 2    brain, but not in liver or lung. Subsequently, Mistry et al. (1985) demonstrated three Pb-binding
 3    proteins (11.5, 63, and >200 kDa) in rat kidney cytosol, which had binding characteristics of
 4    high affinity, low capacity with respective Kd values of 13, 40, and 123 nM. The 11.5 kDa and,
 5    possibly, the 63 kDa proteins were capable of translocating Pb into the nucleus as shown by
 6    uptake of 203Pb into nuclei incubated with tagged cytosolic proteins. Goering and Fowler (1984)
 7    showed that the 11.5 kDa protein, but not the 63 kDa protein was capable of reversing
 8    Pb-induced ALAD inhibition in liver homogenates.  This effect was mediated both by chelation
 9    of Pb by the Pb-binding protein and by donation of zinc to ALAD (Goering and Fowler, 1985).
10    Various divalent metal ions influence the binding of Pb to the rat kidney cytosolic binding
11    proteins, with an order of displacement of Cd2+>Zn2+>Pb2+. Ca2+ had no effect, while Fe2+ had a
12    cooperative effect (Mistry et al., 1985).  These observations may account for the previously
13    demonstrated effect of concomitant Pb and cadmium administration in reducing total kidney Pb
14    (Mahaffey et al., 1981) and preventing the development of intranuclear inclusion bodies
15    (Mahaffey and Fowler, 1977).
16         Later studies by Fowler and Duval (1991) identified the rat renal Pb-binding protein as a
17    cleavage product of oc2-microglobulin, with a Kd of 10"8 M Pb.  There are two forms  of the
18    protein  in the kidney, differentiated by the cleavage of the first 9-N terminal residues from the
19    higher-molecular weight form. Other studies by Smith et al. (1998) found two Pb-binding
20    proteins in environmentally exposed human kidneys, identified as acyl-CoA binding  protein
21    (ACBP) or diazepam binding inhibitor (molecular weight 9 kDa) and thymosin (34 (molecular
22    weight 5 kDa).  These polypeptides have a high affinity for Pb (Kd~14 nM).
23         In rat brain, Goering et al.  (1986) and Duval and Fowler (1989) explored the effects of
24    environmental Pb on Pb-binding proteins and the ability of rat brain Pb-binding proteins to
25    diminish the inhibition of hepatic ALAD by Pb (liver does not contain the Pb-binding protein).
26    In the first study, a brain protein of 12 kDa was described, in comparison to the kidney
27    Pb-binding protein of 9 kDa. Both competition of Pb binding between the brain  Pb-binding
28    protein  and ALAD and donation of zinc by the brain protein (shown by 65Zn uptake)  were found
29    to account for the decreased ALAD inhibition. In the second study the rat brain Pb-binding
30    protein  was described as having a molecular weight of 23 kDa, with significant levels of
31    glutamic acid, aspartic acid, and cysteine. Polyclonal antibody to rat renal Pb-binding proteins

      December 2005                            5-286       DRAFT-DO NOT QUOTE OR CITE

-------
 1    showed a lack of reactivity with the brain protein, indicating that the proteins are
 2    immunologically distinct.
 3          Fowler et al. (1993) examined monkey kidney and brain from non-Pb-treated animals and
 4    isolated Pb-binding proteins that also had a relatively high content of aspartic and glutamic
 5    amino acid residues and were similar in size to the rat Pb-binding proteins. Polyclonal
 6    antibodies to oc-2 microglobulin and metallothionein did not cross-react with either monkey
 7    kidney or brain proteins. Quintanilla-Vega et al. (1995) isolated a thymosin 04 and a second, as
 8    yet unidentified, protein with a molecular weight of 20 kDa and a pi of 5.9 from brains of
 9    environmentally Pb-exposed humans.
10
11    5.11.3  Lead-Binding Proteins in Erythrocytes
12          Intra-erythrocytic Pb-binding was initially attributed primarily to hemoglobin, molecular
13    weight 64 kDa (Barltrop and Smith, 1972; Raghavan and Gonick, 1977; Ong and Lee, 1980;
14    Lolin and O'Gorman, 1988), but more recent studies have ascribed the major Pb binding to
15    ALAD, molecular weight 240-280 kDa. In contrast to this protein, several studies have focused
16    on an inducible low molecular weight protein in workers chronically exposed to Pb and which
17    seems to have a protective effect.  The first recognition of this protein was by Raghavan and
18    Gonick (1977) who found an approximately 10 kDa protein in Pb workers but not in controls,
19    following Sephadex G-75 fractionation (Figure 5-11.1). Upon subsequent SDS-polyacrylamide
20    gel electrophoresis, the protein split into two bands, only the uppermost of which contained Pb
21    (Figure 5-11.2).
22          Raghavan et al. (1980) then went on to fractionate the erythrocyte Pb into a hemoglobin
23    fraction, a 10 kDa fraction, free Pb, and a "residual Pb" fraction thought to be composed of
24    membrane Pb and a high-molecular weight fraction. Lead workers manifesting toxicity at both
25    high blood Pb and relatively low blood Pb levels showed high levels of residual Pb, attributed in
26    the workers with toxicity at low blood leads to a very low quantity of the 10 kDa fraction. In a
27    follow-up study, Raghavan et al. (1981) reported elevated levels of Pb in the high molecular
28    weight fraction (pre-hemoglobin) and in the membrane fraction in workers with toxicity at both
29    high and low BLLs. Again, those with toxicity at low blood Pb had low levels of the Pb bound
30    to the  10 kDa protein. Membrane Pb was found to correlate inversely with membrane Na-K-
31    ATPase; no correlation was seen with total blood Pb.
      December 2005                          5-287       DRAFT-DO NOT QUOTE OR CITE

-------
                                       43000 UX»
                                         MW
                                                        -i500
                                                        - 200
                                                             *—*
                                                        - 100
                                 80   100   120   140
                                 ELUTION VOLUME (ml)
                                                     160
Figure 5-11.1.  Sephadex G-75 gel filtration of RBC hemolysate from lead-exposed
               individual.  Ultraviolet absorption and radioactivity of 210Pb are plotted
               against elution volume. The column was calibrated with ovalbumin (mol
               wt 45,000) and ribonuclease (mol wt 13,700).  Also indicated is the locus of
               hemoglobin (Hb). Hemolysates from normal control individuals showed no
               UV absorption or radioactivity in the volume eluting between 130 and
               155 mL.

Source: Raghavan and Gonick (1977) with permission.

1
•

1








1
f
1
4
5











r
Molecular
Weight



•+A 0,000
ABC
Figure 5-11.2.  SDS-polyacrylamide gel electrophoresis of RBC hemolysates from normal
               control (A) and lead-exposed individuals (B), and of low-mol-wt. lead-
               binding protein (C). Stained with coomassie blue.
Source: Raghavan and Gonick (1977) with permission.

December 2005                          5-288
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Gonick et al. (1985) partially purified the 10 kDa protein by HPLC using a protein 1-125
 2    column followed by isoelectric focusing on a sucrose gradient column. Three protein peaks
 3    resulted: one of 30 kDa, and two of 10 kDa. Only one of the latter peaks contained Pb.  This
 4    peak had a pi of 5.3 and a molecular weight, determined by SDS-PAGE, of 12 kDa.  The
 5    majority of Pb was found in this peak, which also contained calcium, zinc, and cadmium. Amino
 6    acid analysis showed a very high percentage of glycine (44%) and lower quantities of histidine,
 7    aspartic acid, and leucine.
 8          Ong and Lee (1980) studied the distribution of 203Pb in components of normal human
 9    blood. Ninety-four percent of 203Pb was incorporated into the erythrocyte and 6% remained in
10    the plasma.  SDS-PAGE of plasma showed that 90% was present in the albumin fraction. Within
11    the erythrocyte membrane, the most important binding site was the high molecular weight
12    fraction, about 130-230 kDa.  Within the erythrocytic cytoplasm, the protein band associated
13    with 203Pb had a molecular weight of 67 kDa as shown by the elution characteristics on G-75
14    chromatography. This was thought to be hemoglobin.
15          Lolin and O'Gorman (1988) and Church et al. (1993 a,b), following the same procedure
16    as Raghavan and Gonick (1977), confirmed the findings of a low molecular weight protein in the
17    erythrocytes of Pb workers, but not found in control patients. Lolin and O'Gorman (1988)
18    quantitated the protein, which ranged from 8.2 to 52.2 mg/L RBC in Pb workers but found none
19    in controls, again implying it to be an inducible protein. They found that the low molecular
20    weight protein first appeared when the blood Pb concentration exceeded 39 |ig/dL. A positive
21    correlation was seen between the amount of the intra-erythocytic low molecular weight protein
22    and dithiothreitol-activated ALAD activity but not the non-activated activity.  Church et al.
23    (1993a,b) also confirmed the findings of Raghavan et al. (1977).  In 1993a, they described two
24    patients with high blood Pb levels:  an asymptomatic worker with a blood Pb of 180 |ig/dL, and a
25    symptomatic worker with a blood Pb of 161 |ig/dL. In the first patient, approximately 67% of
26    the erythrocyte Pb was bound to a low molecular weight protein of approximately 6-7 kDa.  In
27    the second patient,  the protein only contained 22% of the total erythrocytic Pb. Church et al.
28    (1993b) found that a sample of the low molecular weight protein purified from Pb workers,
29    which they termed protein M, had characteristics of metallothionein, such as a molecular weight
30    of 6.5 kDa, a pi between 4.7 and 4.9, and a greater UV absorbance at 254 nm than at 280 nm.
31    Amino acid composition showed 33% cysteine but no aromatic amino acids. This composition

      December 2005                          5-289       DRAFT-DO NOT QUOTE OR CITE

-------
 1    differed from that of the low molecular weight protein described by Gonick et al. (1985), which
 2    had a molecule weight of 12 kDa, a pi of 5.3, and amino acid analysis that showed no cysteine.
 3    This discrepancy might be explained by a combined Pb  and cadmium exposure in the Church
 4    et al. (1993b) study, which may have produced a Pb-thionein.
 5          Xie et al. (1998) used a Biogel A column instead of Sephadex G-75 to separate
 6    Pb-binding proteins from erythrocyte hemolysates from a control patient and from Pb-exposed
 7    workers.  They clearly showed that the major Pb-binding was associated with a large molecular
 8    weight protein, consistent with ALAD, in both the controls and Pb workers. When they added
 9    increasing amounts of Pb to the blood of the control patient, a second low molecular weight
10    protein peak occurred, in which Pb binding was larger than the ALAD peak (Figure 5-11.3).
11    This second peak was also seen in a chronically Pb-exposed worker (Figure 5-11.4) and was
12    estimated to be less than 30 kDa in molecular weight. Thus these results are consistent with the
13    aforementioned studies.
14
15    5.11.4  Lead-Binding Proteins in Rat Liver
16          Sabbioni and Marafante (1976) explored the distribution of 203Pb in rat whole tissue as
17    well as in subcellular liver fractions. By far the largest quantity of Pb recovered was in the
18    kidney, with lesser amounts in liver, spleen, and blood.  Upon subcellular fractionation of the
19    liver, the majority of 203Pb was found in the nuclei, and  most of the Pb was detected in the
20    nuclear membrane fraction, bound exclusively to membrane proteins.  The intranuclear Pb was
21    associated with histone fractions. As reported by Oskarsson et al. (1982), Pb binding proteins
22    were found in the cytoplasm of the liver.
23
24    5.11.5  Lead-Binding Proteins in Intestine
25          Fullmer et al. (1985) showed in the chick and cow that although Pb does not directly
26    stimulate Pb-binding proteins in the intestine, Pb can displace calcium from calcium-binding
27    proteins; and, thus, calcium-binding proteins may play a role in intestinal Pb transport.  Purified
28    calcium-binding protein from chick and cow, as well as calmodulin, troponin C, and
29    oncomodulin were dialyzed against added labeled and unlabeled Pb or calcium. Results
30    disclosed high affinity binding sites, with greater affinity for Pb than for calcium.   Similar results

      December 2005                          5-290       DRAFT-DO NOT QUOTE OR CITE

-------
        30
      o
      S 25
      o
= 20

I
^ 15

1
^ 10


f-  5

   0
                         Pb Cone.
                         ALAD Activity
                         Se Cone.
                         Protein
                                            0,8
                                                                                   0.6
                                                                                      Q
                                                                                      O
                                                                                      3
                                                                                   0.2
                              11
16
                                                21
                                                    26
31
36
        30



        25

        -


       . 15


        10
      0)
      CO

      £   5
                              11        16        21

                                       Fraction number
                             31
                                                                       36
                                                                                   0.8
                                            0.6
                                            0.4
                                                                                   0.2
    Figure 5-11.3.  Chromatographic profiles of protein, ALAD activity and Pb in human
                   erythrocytes incubated with 5% glucose solution containing Pb acetate.
                   Blood was incubated (a) without Pb (b) 10 uM Pb (final concentrations).

    Source: Adapted from Xie et al. (1998).
1   were obtained with calmodulin, troponin C, and oncomodulin, all members of the troponin C

2   superfamily of calcium-binding proteins.
    December 2005
      5-291
                                                   DRAFT-DO NOT QUOTE OR CITE

-------
       
                                                                                  Q
                                                                               0.6 O
                                                                                  .•

                                                                               0.4 1
0.2


0



1


0.8
                                                                  in
                                                                  in
                                                                  Q
                                                               0.6


                                                               0.4


                                                               0.2


                                                               0
Figure 5-11.4.   Chromatic profiles of protein, ALAD activity, Pb, and Se in the
                erythrocytes of lead-exposed workers, (a) control, (b) subacute exposure,
                (c) chronic exposure.

Source: Xie et al. (1998) with permission.
December 2005
                           5-292
DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.11.6  Relationship of Lead-Binding Protein to Metallothionein
 2          Similarities of Pb-binding protein to metallothionein have been discussed earlier.  Maitani
 3    et al. (1986) commented that hepatic zinc-metallothionein could be induced by intravenous and
 4    intraperitoneal injections of Pb into mice, but not by subcutaneous injection. Ikebuchi et al.
 5    (1986) found that a sublethal dose of Pb-acetate injected intraperitoneally into rats induced the
 6    synthesis of a Pb-metallothionein in addition to zinc-metallothionein.  The Pb-metallothionein
 7    contained 28% half-cysteine and cross-reacted with an antibody against rat zinc-thionein II.
 8          Goering and Fowler (1987 a,b) demonstrated that pretreatment of rats with zinc 48 and
 9    24 h prior to injection of 203Pb resulted in both zinc and Pb co-eluting with a zinc-thionein
10    fraction on Sephadex G-75 filtration.  In addition, both purified zinc-thionein-I and II bound
11    203Pb in vitro. Gel filtration of incubates containing liver ALAD and 203Pb demonstrated that the
12    presence of zinc-thionein alters the cytosolic binding pattern of Pb, with less binding to ALAD.
13    Zinc-thionein also donates zinc to activate ALAD.  Goering and Fowler (1987b) found that
14    pretreatment of rats with either cadmium or zinc affected liver ALAD activity when incubated
15    with Pb.  Liver and kidney zinc-thioneins, and to a lesser extent, cadmium, zinc-thionein
16    decreased the free pool of Pb available to interact with ALAD, resulting in attenuated ALAD
17    inhibition. Liu et al. (1991) further showed that zinc-induced metallothionein in primary
18    hepatocyte cultures protects against Pb-induced cytotoxicity, as assessed by enzyme leakage and
19    loss of intracellular potassium.
20          Qu et al. (2002) and Waalkes et al. (2004) have shown that metallothionein-null
21    phenotypic mice are more susceptible to Pb injury over a 20-week period than wild type mice.
22    Unlike the wild type mice, Pb-treated  metallothionein-null mice showed nephromegaly and
23    significantly decreased renal function  after exposure to Pb. The metallothionein-null  mice
24    accumulated less renal Pb than wild type and formed no inclusion bodies.  When the
25    observations were extended to 104 weeks, renal proliferative lesions (adenoma and cystic tubular
26    atypical hyperplasia) were more common and severe in metallothionein-null than in wild type
27    mice.  A metastatic renal cell carcinoma occurred in a metallothionein-null mouse, whereas none
28    occurred in  wild type mice.  Such studies lend  credence to the view that metallothinein, or a
29    closely related gene, is involved in the formation of Pb-binding proteins in the kidney.
30


      December 2005                           5-293        DRAFT-DO NOT QUOTE OR CITE

-------
 1    5.11.7 Is ALAD an Inducible Enzyme and is it the Principal Lead-Binding
 2           Protein in the Erythrocyte?
 3          The enzyme ALAD has been found to be the most sensitive indicator of Pb exposure and
 4    toxicity (Granick et al., 1973, Buchet et al., 1976).  In the 1980s, two articles were presented
 5    appearing to show that ALAD is inducible after Pb exposure in humans.  By comparing a
 6    nonexposed control population of Pb workers and assaying ALAD by means of immunoassay or
 7    as 'restored' ALAD activity (i.e., incubation with heat, zinc and dithiothreitol) both articles
 8    indicated that the amount of ALAD, as contrasted to ALAD activity, was increased by Pb
 9    exposure (Fujita et al., 1982; Boudene et al., 1984).  Similar findings were reported for the rat
10    (Fujita et al., 1981). Subsequent studies have focused on the effect of ALAD polymorphism on
11    the susceptibility to Pb intoxication. ALAD is a zinc-containing enzyme, which catalyzes the
12    second step of heme synthesis, i.e., catalyzes the condensation of two delta-aminolevulinic acid
13    molecules into one molecule of porphobilinogen (Boudene et al., 1984).  It is a polymorphic
14    protein with three isoforms:  ALAD-1, ALAD 1-2, and ALAD 2-2. Several studies have shown
15    that, with the same exposure to Pb, individuals with the ALAD-2 gene have higher blood Pb
16    levels (Astrin et al., 1987; Wetmur, 1994; Wetmur et al., 1991; Smith et al., 1995a; Bergdahl
17    et al., 1997; Perez-Bravo et al., 2004; Kim et al., 2004).  Initially it was thought that these
18    individuals might be more susceptible to Pb poisoning (Wetmur et al.,  1991), but it is now
19    appreciated that the ALAD-2 gene offers protection against Pb poisoning by binding Pb more
20    securely (Kelada et al., 2001). In support of this statement, it can be cited that individuals with
21    the ALAD 1-2/2-2  genotypes, in comparison to those with the ALAD 1-1 genotype, have not
22    only higher blood Pb but also decreased plasma levulinic acid (Schwartz et al., 1997), lower zinc
23    protopophyrin (Kim et al., 2004), lower cortical bone Pb (Smith et al.,  1995b), and lower
24    amounts of DMSA-chelatable Pb (Schwartz et al., 1997, 2000).
25          The significance of erythrocyte ALAD binding to Pb was initially confirmed by a study
26    by Bergdahl et al. (1997) in which the authors used a FPLC Superdex 200 HR 10/30
27    chromatographic column coupled to ICP-MS (for determination of Pb) to examine erythrocytes
28    from Pb workers and controls. They found the principal Pb-binding protein peak to be 240 kDa
29    (rather than the  presumed hemoglobin peak reported by Barltrop and Smith (1972) and
30    Raghavan and Gonick (1977), using Sephadex G-75 chromatography). This was shown to be
31    ALAD by binding to specific ALAD antibodies. Two additional smaller Pb-binding peaks of

      December 2005                          5-294       DRAFT-DO NOT QUOTE OR CITE

-------
 1    45 kDA and 10 kDa were also seen, but not identified. Bergdahl et al. (1997) attributed the

 2    discrepancies in the studies to the fact that Sephadex G-75 separates proteins in the range of 3 to
 3    80 kDa, making the separation of hemoglobin (molecular weight 64 kDa) from ALAD

 4    (molecular weight 240-280 kDa) very difficult. In addition, the earlier studies had utilized

 5    binding of 203Pb or 210Pb to identify the binding proteins, a technique which may have skewed

 6    the findings if ALAD were already saturated. ALAD binding capacity for Pb has been  measured
 7    at 85 |ig/dL in erythrocytes or 40 |ig/dL in whole blood (Bergdahl et al., 1998), which would

 8    permit a greater degree of binding to the low molecular weight component when blood  Pb

 9    exceeded 40 |ig/dL. Bergdahl et al. (1998) have speculated that the low molecular weight
10    component might be acyl-CoA-binding protein, identical to the kidney Pb-binding protein

11    described by Smith et al. (1995b). Goering and Fowler (1987) had reported earlier that the

12    presence of low molecular weight high affinity (Kd 10"8M) Pb-binding proteins in kidney and

13    brain served as protection against ALAD inhibition in those organs, whereas the absence of the

14    low molecular weight proteins in liver contributed to the greater sensitivity to ALAD inhibition
15    in that organ.

16          A summary of the findings on Pb-binding protein can be found in Table AX5-11.1.

17

18    5.11.8  Summary

19      •  There appears to be a consensus that the enzyme, ALAD, a 280 kDa protein, is inducible
20         and is the major Pb-binding protein within the erythrocyte.  ALAD polymorphism
21         influences the degree of Pb-binding as the ALAD-2 phenotype binds more Pb in a
22         nontoxic fashion than ALAD-1.  What is more confusing is the nature and importance of
23         the low molecular weight erythrocytic Pb-binding protein. There is no doubt that it
24         appears in Pb-exposed workers but not in controls and  that its molecular weight is
25         approximately 10 kDa.  The in vitro addition of Pb to erythrocytes of controls results in
26         progressively increasing Pb binding to a low molecular weight protein peak migrating in
27         the same position as the low molecular weight protein from Pb workers.  This confirms
28         the fact that once the binding capacity of ALAD is saturated, Pb shifts to the low
29         molecular weight protein. The nature of the low-molecular weight protein is also
30         questionable, it has been variously identified as a 12 kDa protein with a high percentage of
31         glycine plus histidine, aspartic acid, and leucine and as a 6.5 kDa molecule with a large
32         percentage of cysteine and a greater UV absorbance at 254 than 280 nm. The latter
33         findings  suggest that the protein might be a metallothionein.

34      •  Metallothionein is a protein that is mildly inducible by Pb but to a much greater degree by
35         zinc and cadmium. What is more significant is that Pb binds to pre-formed
36         metallothionein,  stimulated by zinc or cadmium, so that under these conditions a

      December 2005                          5-295        DRAFT-DO NOT QUOTE OR CITE

-------
 1         Pb-thionein forms. Thus, concomitant Pb and cadmium exposure occurred in Pb workers
 2         that could account for the finding of a metallothionein-like protein in those workers.

 3      •  Extensive studies of cytoplasmic Pb-binding proteins in non-Pb-treated rats, human, and
 4         monkeys have been reported.  The Pb-binding protein in rat kidney has been identified as
 5         a cleavage product of a-2 microglobulin. The low molecular weight Pb-binding proteins
 6         in human kidney have been identified as thymosin 04 (molecular weight 5 kDa) and acyl-
 7         CoA binding protein (molecular weight 9 kDa). In human brain the Pb-binding proteins
 8         were thymosin (34 and an unidentified protein of 23 kDa.  Antibodies to a-2 microglobulin
 9         and metallothionein did not cross-react with monkey kidney or brain Pb-binding proteins,
10         suggesting species differences. Whether the low molecular weight human kidney and
11         brain Pb-binding proteins are similar or identical to the  low molecular weight Pb-binding
12         proteins in erythrocytes is at present unknown. Perhaps some clarification would be
13         provided were subsequent investigators to contrast normal with Pb-exposed rats and to
14         measure the resting and inducible Pb-binding protein levels in kidney, brain, and
15         erythrocyte.
16      •  The possible role of metallothionein as a renal Pb-binding protein assumes greater
17         importance because of the work showing that metallothionein-null mice failed to respond
18         to Pb exposure by developing intranuclear Pb inclusion bodies or greatly increased Pb
19         content of the ki dney s.
20
      December 2005                           5-296       DRAFT-DO NOT QUOTE OR CITE

-------
  1     5.12  REFERENCES

  2     Abdollahi, M; Dehpour, A. R.; Fooladgar, M. (1997) Alteration of rat submandibulary gland secretion of protein,
  3            calcium and 7V-acetyl-p-D-glucosaminidase activity by lead. Gen. Pharmacol. 29: 675-680.
  4     Aboul-Ela, E. I. (2002) The protective effect of calcium against genotoxicity of lead acetate administration on bone
  5            marrow and spermatocyte cells of mice in vivo. Mutat. Res. 516: 1-9.
  6     Acharya, U. R.; Acharya, S.; Mishra, M. (2003) Lead acetate induced cytotoxicity in male germinal cells of Swiss
  7            mice. Ind. Health 41: 291-294.
  8     Ades, A. E.; Kazantzis, G. (1988) Lung cancer in a non-ferrous smelter: the role of cadmium. Br. J. Ind.  Med. 45:
  9            435-442.
10     Adhikari, N.; Sinha, N.; Saxena, D. K. (2000) Effect of lead on Sertoli-germ cell coculture of rat. Toxicol. Lett. 116:
11            45-49.
12     Adhikari, N.; Sinha, N.; Narayan, R.; Saxena, D. K. (2001) Lead-indued cell death in testes  of young rats. J. Appl.
13            Toxicol. 21:275-277.
14     Adrian, G. S.; Rivera, E. V.; Adrian, E. K.; Lu, Y.; Buchanan, J.; Herbert, D. C.; Weaker, F. J.; Walter, C. A.;
15            Bowman, B.  H. (1993) Lead suppresses chimeric human transferrin gene expression in transgenic mouse
16            liver. Neurotoxicology 14: 273-282.
17     Alber, S. A.; Strupp, B. J. (1996) An in-depth analysis of lead effects in a delayed spatial alternation task:
18            assessment of mnemonic effects, side bias, and proactive interference. Neurotoxicol. Teratol. 18:3-15.
19     Al-Hakkak, Z. S.; Zahid, Z. R.; Ibrahim, D. K.; Al-Jumaily, I. S.; Bazzaz, A. A. (1988) Effects of ingestion of lead
20            monoxide alloy on male mouse reproduction. Arch. Toxicol. 62: 97-100.
21     Alkondon, M.; Costa, A. C. S.; Radhakrishnan, V.; Aronstam, R. S.; Albuquerque, E. X. (1990) Selective blockade
22            of NMDA-activated channel currents may be implicated in learning deficits caused  by lead. FEES Lett. 261:
23            124-130.
24     Altmann, L.; Weinsberg, F.; Sveinsson, K.; Lilienthal, H.; Wiegand, H.; Winneke, G. (1993) Impairment of long-
25            term potentiation and learning following chronic lead exposure. Toxicol. Lett. 66: 105-112.
26     Alvarez, J.; Garcia-Sancho, J.; Herreros, B. (1986) Inhibition of Ca2+-dependent K+ channels by lead in one-step
27            inside-out vesicles from human red cell membranes. Biochim. Biophys. Acta 857: 291-294.
28     Amoruso, M. A.; Witz, G.; Goldstein, B. D. (1987) Alteration of erythrocyte membrane fluidity by heavy metal
29            cations. Toxicol. Ind. Health 3: 135-144.
30     Angle, C. R.; Thomas, D. J.; Swanson, S. A. (1990) Lead inhibits the basal and stimulated responses of a rat
31            osteoblast-like cell line ROS 17/2.8 to 1 "alpha", 25-dihydroxyvitamin D3 and IGF-I. Toxicol Appl.
32            Pharmacol. 103:281-287.
33     Angle, C. R.; Thomas, D. J.; Swanson, S. A. (1993) Osteotoxicity of cadmium and lead inHOS TE 85 and ROS
34            17/2.8 cells: relation to metallothionein induction and mitochondrial binding. BioMetals 6:  179-184.
35     Antonowicz, J.; Andrzejak,  R.; Smolik, R. (1990) Influence of heavy metal mixtures on erythrocyte metabolism. Int.
36            Arch. Occup. Environ. Health 62: 195-198.
37     Anttila, A.; Heikkila, P.; Pukkala, E.; Nykyri, E.; Kauppinen, T.; Hernberg, S; Hemminki, K. (1995) Excess lung
38            cancer among workers exposed to lead. Scand. J. Work Environ. Health. 21: 460-469.
39     Anttila, A.; Heikkila, P.; Nykyri, E.; Kauppinen, T.; Pukkala, E.; Hernberg, S.; Hemminki, K. (1996) Risk of
40            nervous system cancer among workers exposed to lead. J. Occup. Environ. Med. 38: 131-136.
41     Apostoli, P.; Romeo, L.; De Matteis, M. C.; Menegazzi, M.; Faggionato, G.; Vettore, L.  (1988) Effects of lead on
42            red blood cell membrane proteins. Int. Arch. Occup. Environ. Health 61:  71-75.
43     Apostoli, P.; Huard, C.; Chaumontet, C.; Mattel, P.; Allesio, L.; Mazzoleni, G. (2000) Effects of four inorganic lead
44            compounds on the proliferation and junctional coupling of cultured REL liver cells. Am. J. Ind. Med. 38:
45            340-348.
46     Appleton, J. (1991) The effect of lead acetate on dentine formation in the rat. Arch. OralBiol. 36: 377-382.
47     Appleton, J. (1992) Dentinogenesis and the calciotraumatic response to the injection of lead or fluoride ions.
48            Scanning Microsc. 6: 1073-1081.
49     Arizono, K.; Sugiura, S.; Miyazato, S.; Takiguchi, M.; Ariyoshi, T. (1996) DT-diaphorase induction by lead acetate
50            in the liver of rats. Bull. Environ. Contam. Toxicol. 57: 41-46.
51     Astrin, K. H.; Bishop, D. F.; Wetmur, J. G.; Kaul, B.; Davidow, B.; Desnick, R. J. (1987) Delta-aminolevulinic acid
52            dehydratase isozymes and lead toxicity. Ann. N. Y. Acad. Sci. 514:  23-29.
53     Attri, J.; Dhawan, V.; Mahmood, S.; Pandhi, P.; Parwana, H. K.; Nath, R. (2003) Effect of vitamin C
54            supplementation on oxidative DNA damage in an experimental model of lead-induced hypertension. Ann.
55            Nutr. Metab. 47: 294-301.

       December 2005                                5-297        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Aykin-Burns, N.; Laegeler, A.; Kellogg, G.; Ercal, N. (2003) Oxidative effects of lead in young and adult Fisher 344
  2            rats. Arch. Environ. Contam. Toxicol. 44: 417-420.
  3     Baginski, B.; Grube, B. (1991) Einfluss vonblei, zink, und cadmium auf die zelltoxische wirklund humaner
  4            polymorphkerniger leukozyten ambeispiel von hefzellen. Zentralbl. Hyg. Umweltmed. 191: 28-35.
  5     Ballew, C.; Khan, L. K.; Kaufmann, R.; Mokdad, A.; Miller, D. T.; Gunter, E. W. (1999) Blood lead concentration
  6            and children's anthropometric dimensions in the Third National Health and Nutrition Examination Survey
  7            (NHANES III), 1988-1994. J. Pediatr. 134: 623-630.
  8     Ban, M; Hettich, D. (2005) Effect of Th2 cytokine antagonist treatments on chemical-induced allergic response in
  9            mice. J. Appl. Toxicol. 25: 239-247.
10     Bannon, D. L; Olivi, L.; Bressler, J. (2000) The role of anion exchange in the uptake of Pbby human erythrocytes
11            and Madin-Darby canine kidney cells. Toxicology 147: 101-107.
12     Baranowska-Bosiacka, I.; Hlynczak, A. J. (2003) The effect of lead ions on the energy metabolism of human
13            erythrocytes in vitro. Comp. Biochem. Physiol. Part C: Toxicol. Pharmacol. 134: 403-416.
14     Baranowska-Bosiacka, L; Hlynczak, A. J. (2004) Effect of lead ions on rat erythrocyte purine content. Biol. Trace
15            Elem.  Res. 100: 259-273.
16     Barbier, O.; Jacquillet, G.; Tauc, M.; Cougnon, M.; Poujeol, P. (2005) Effect of heavy metals on, and handling by,
17            the kidney. Nephron. Physiol. 99: 105-110.
18     Barltrop, D.; Smith, A. (1972) Lead binding to human haemoglobin. Experientia 28: 76-77.
19     Barnett, J. B. (1996) Developmental immunotoxicology. In. Smialowicz, R. J.; Holsapple, M. P., ed. Experimental
20            immunotoxicology. CRC Press, Inc. Boca Raton. 47-62.
21     Barnum-Huckins, K. M.; Martinez, A. O.; Rivera, E. V.; Adrian, E. K., Jr.; Herbert, D. C.; Weaker, F. J.; Walter, C.
22            A.; Adrian, G. S. (1997) A comparison of the suppression of human transferrin synthesis by lead and
23            lipopolysaccharide. Toxicology 118: 11-22.
24     Basaran, N.; Undeger, U. (2000) Effects of lead on immune parameters in occupationally exposed workers. Am. J.
25            Ind. Med. 38:  349-354.
26     Batetta, B.; Dessi, S.; Pulisci, D.; Carrucciu, A.; Pani, P. (1990) Multiple molecular forms of rat liver glucose-6-
27            phosphate dehydrogenase during liver hyperplasia induced by lead nitrate. Res. Commun. Chem. Pathol.
28            Pharmacol. 67: 279-288.
29     Batra, N.; Nehru, B.; Bansal, M.  P. (2001) Influence of lead and zinc on rat male reproduction at 'biochemical and
30            histopathological levels'. J. Appl. Toxicol.  21: 507-512.
31     Baykov, B.; Gunova, M.; Stoyanov, M.; Neychev, H.; Stefanova, T.; Nicolova, N. (1996) Designing an artificial
32            ecological mesocosm for the study of Cd and Pb impact on the immune system of experimental animals.
33            Toxicol Lett. 89:5-10.
34     Beach, J. R.;Henning, S. J. (1988) The distribution of lead in milk and the fate of milk lead in the gastrointestinal
3 5            tract of suckling rats. Pediatr. Res. 23: 58-62.
36     Bellinger, D. C. (2004) What is an adverse effect?  A possible resolution of clinical and epidemiological
37            perspectives on neurobehavioral toxicity. Environ. Res. 95: 394-405.
38     Bellinger, D.; Sloman, J.; Leviton, A.; Rabinowitz, M.; Needleman, H. L.; Waternaux, C. (1991) Low-level lead
39            exposure and children's cognitive function in the preschool years. Pediatrics 87: 219-227.
40     Bellinger, D. C.; Stiles, K. M.; Needleman, H. L. (1992) Low-level lead exposure, intelligence and academic
41            achievement: a long-term follow-up study. Pediatrics 90: 855-861.
42     Bellinger, D.; Hu, H.;  Titlebaum, L.; Needleman, H. L. (1994) Attentional correlates of dentin and bone lead levels
43            in adolescents. Arch. Environ. Health 49: 98-105.
44     Belloni-Olivi, L.; Annadata, M.;  Goldstein, G. W.; Bressler, J. P. (1996) Phosphorylation of membrane proteins in
45            erythrocytes treated with lead. Biochem. J. 315: 401-406.
46     Bergdahl, I. A.; Grubb, A.; Schutz, A.; Desnick, R. J.; Wetmur, J. G.; Sassa, S.; Skerfving, S. (1997) Lead binding
47            to 5-aminolevulinic acid dehydratase (ALAD) in human erythrocytes. Pharmacol. Toxicol. 81: 153-158.
48     Bergdahl, I. A.; Sheveleva, M.; Schutz, A.; Artamonova, V. G.; Skerfving, S. (1998) Plasma and blood lead in
49            humans: capacity-limited binding to "delta"-aminolevulinic acid dehydratase and other lead-binding
50            components. Toxicol.  Sci. 46: 247-253.
51     Bernard, A.; Lauwerys, R. (1987) Metal-induced alterations of 5 aminolevulinic acid dehydratase. In: Silbergeld,  E.
52            K.; Fowler, B. A., eds. Mechanisms of chemical-induced porphyrinopathies. New York, NY: New York
53            Academy of Sciences; pp. 41-47. [Annals of the New York Academy of Sciences: v. 514].
54     Bernard, S. M.; McGeehin, M. A. (2003) Prevalence of blood lead levels >/= 5 "mu"g/dL among US children 1 to 5
5 5            years of age and socioeconomic and demographic factors associated with blood of lead levels 5 to  10


       December 2005                                5-298        DRAFT-DO NOT QUOTE OR  CITE

-------
  1            "mu"g/dL, Third National Health and Nutrition Examination Survey, 1988-1994. Pediatrics 112: 1308-
  2            1313.
  3     Berry, W. D., Jr.; Moriarty, C. M; Lau, Y. S. (2002) Lead attenuation of episodic growth hormone secretion in male
  4            rats.  Int. J.Toxicol. 21: 93-98.
  5     Bhattacharya, A.; Shukla, R.; Dietrich, K. N.; Miller, I; Bagchee, A.; Bornschein, R. L.; Cox, C.; Mitchell, T.
  6            (1993) Functional implications of postural disequilibrium due to lead exposure. Neurotoxicology 14: 179-
  7            189.
  8     Biedermann, K. A.; Landolph, J. R. (1987) Induction of anchorage independence in human diploid foreskin
  9            fibroblasts by carcinogenic metal salts. Cancer Res. 47: 3815-3823.
10     Biedermann, K. A.; Landolph, J. R. (1990) Role of valence state and solubility of chromium compounds on
11            induction of cytotoxicity, mutagenesis, and anchorage independence in diploid human fibroblasts.  Cancer
12            Res.  50: 7835-7842.
13     Bielarczyk, H.; Tian, X.; Suszkiw, J. B. (1996) Cholinergic denervation-like changes in rat hippocampus following
14            developmental lead exposure. Brain Res. 708: 108-115.
15     Bilban, M. (1998) Influence of the work environment in a Pb-Zn mine on the incidence of cytogenetic damage in
16            miners. Am. J. Ind. Med. 34: 455-463.
17     Bizarro, P.; Acevedok, S.; Nino-Cabrera, G.; Mussali-Galante, P.; Pasos, F.; Avila-Costa, M. R.; Fortoul, T. I.
18            (2003) Ultrastructural modifications in the mitochondrion of mouse Sertoli cells after inhalation of lead,
19            cadmium or lead-cadmium mixture. Reprod. Toxicol. 17:  561-566.
20     Blackman, S. S., Jr. (1936) Intranuclear inclusion bodies in the kidney and liver caused by lead poisoning.  Bull.
21            Johns Hopkins Hosp. 58: 384-402.
22     Blakley, B. R. (1987) The effect of lead on chemical- and viral-induced tumor production in mice. J. Appl. Toxicol.
23            7:  167-172.
24     Blakley, B. R.; Archer, D. L. (1981) The effect of lead acetate on the immune response in mice. Toxicol. Appl.
25            Pharmacol. 61: 18-26.
26     Blakley, B. R.; Sisodia, C. S.; Mukkur, T. K. (1980) The effect of methylmercury, tetraethyl lead, and sodium
27            arsenite on the humoral immune response in mice. Toxicol. Appl. Pharmacol. 52: 245-254.
28     Blankenship, L. J.; Carlisle, D. L.; Wise, J. P., Sr; Orenstein, J.  M.; Dye, L. E., Ill; Patierno,  S. R. (1997) Induction
29            of apoptotic cell death by paniculate lead chromate: differential effects of vitamins C and E on genotoxicity
30            and survival. Toxicol. Appl. Pharmacol. 146: 270-280.
31     Blanusa, M.; Kostial, K.; Piasek, M.; Jones, M. M.; Singh, P. K. (1995) Reduction of lead retention by mono-3-
32            methylbutan-1-yl meso-2,3-dimercaptosuccinate in suckling rats. Analyst (Cambridge, U. K.) 120: 951-953.
33     Bogden, J. D.; Gertner, S. B.; Kemp, F. W.; McLeod, R.; Bruening, K. S.; Chung, H. R. (1991) Dietary lead and
34            calcium:  effects on blood pressure and renal neoplasia in Wistar rats. J. Nutr. 121: 718-728.
35     Bogden, J. D.; Kemp, F. W.; Han, S.; Murphy, M.; Fraiman, M.; Czerniach, D.; Flynn, C. J.;  Banua, M. L.;
36            Scimone, A.; Castrovilly, L.; Gertner, S. B. (1995) Dietary calcium and lead interact to modify maternal
37            blood pressure, erythropoiesis, and fetal and neonatal growth in rats during pregnancy and lactation. J. Nutr.
38            125:990-1002.
39     Bonacker,  D.; Stoiber, T.; Bohm, K. J.; Prots, I.; Wang, M.; Unger E.; Thier, R.; Bolt, H. M.; Degen, G. H. (2005)
40            Genotoxicity of inorganic lead salts  and disturbance of microtubule function. Environ. Mol. Mutagen. 45:
41            346-353.
42     Bondy, S. C. (1986) Effect of triethyl lead chloride on "delta"-aminolevulinic acid dehydratase. J. Toxicol. Environ.
43            Health 18: 639-649.
44     Borella, P.; Giardino, A. (1991) Lead and cadmium at very low  doses affect in vitro immune  response of human
45            lymphocytes. Environ. Res. 55: 165-177.
46     Borg, C.; Abdelali, J.; Laderach, D.; Maruyama, K.; Wakasugi,  H.; Charrier, S.; Ryffel, B.; Vainchenker, W.; Galy,
47            A.; Caignard, A.; Zitvogel, L.; Cambi, A.; Figdor, C. (2004) NK cell activation by dendritic cells (DC)
48            require the formation of a synapse leading to IL-12 polarization in DC. Blood 104: 3267-3275.
49     Boudene, C.; Despaux-Pages, N.; Comoy, E.; Bohuon, C. (1984) Immunological and enzymatic studies of
50            erythrocytic delta-aminolevulinate dehydratase. Comparison of results obtained in normal and lead-exposed
51            subjects.  Int. Arch. Occup.  Environ. Health 55: 87-96.
52     Bouley, G.; Dubreuil, A.; Arsac, F.; Boudene, C. (1977) Effet du plomb microparticulaire, introduit dans 1'appareil
53            respiratoire, sur la sensibilite de la souris a 1'infection par aerosol de Pasteurella multocida [Effect of
54            microparticulate lead, introduced through respiratory apparatus, on the resistance of mice to infection by
55            aerosolized Pasteurella multocida].  C. R. Hebd. Seances Acad. Sci. Ser. D 285: 1553-1556.


       December  2005                                5-299         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Bourjeily, N.; Suszkiw, J. B. (1997) Developmental cholinotoxicity of lead: loss of septal cholinergic neurons and
  2            long-term changes in cholinergic innervation of the hippocampus in perinatally lead-exposed rats. Brain
  3            Res. 771:319-328.
  4     Bowen, W. H. (2001) Exposure to metal ions and susceptibility to dental caries. J. Dent. Ed. 65: 1046-1053.
  5     Boyce, S. I; Mantle, T. J. (1993) Effect of lead acetate and carbon particles on the expression of glutathione S-
  6            transferase YfYf in rat liver. Biochem. J. 294: 301-304.
  7     Bradman, A.; Eskenazi, B.;  Sutton, P.; Athanasoulis, M; Goldman, L. R.  (2001) Iron deficiency associated with
  8            higher blood lead in children living in contaminated environments. Environ. Health Perspect. 109: 1079-
  9            1084.
10     Braga, M. F. M; Pereira, E. F. R.; Albuquerque, E. X.  (1999a) Nanomolar concentrations of lead inhibit
11            glutamatergic and GABAergic transmission in hippocampal neurons. Brain Res. 826: 22-34.
12     Braga, M. F. M.; Pereira, E. F. R.; Marchioro, M.; Albuquerque, E. X. (1999b) Lead increases tetrodotoxin-
13            insensitive spontaneous release of glutamate and GABA from hippocampal neurons. Brain Res. 826: 10-21.
14     Braga, M. F.; Pereira, E. F.; Mike, A.; Albuquerque, E. X.  (2004) Pb2+ via protein kinase C inhibits nicotinic
15            cholinergic modulation of synaptic transmission in the hippocampus. J. Pharmacol. Exp. Ther. 311: 700-
16            710.
17     Bratton, G. R.; Hiney, J. K.; Dees, W. L. (1994) Lead  (Pb) alters the norepinephrine-induced secretion of luteinizing
18            hormone releasing hormone from the medium eminence of adult male rats in vitro. Life Sci. 55: 563-571.
19     Breslau, N.; Chilcoat, H. D.; Susser, E. S.; Matte, T.; Liang, K.-Y.; Peterson, E. L. (2001) Stability and change in
20            children's intelligence quotient scores: a comparison of two socioeconomically disparate communities. Am.
21            J. Epidemiol. 154: 711-717.
22     Bressler, J. P.; Goldstein, G. W. (1991) Mechanisms of lead neurotoxicity. Biochem. Pharmacol. 41: 479-484.
23     Bressler, J.; Kim, K.-A.; Chakraborti, T.; Goldstein, G. (1999) Molecular mechanisms of lead neurotoxicity.
24            Neurochem. Res. 24: 595-600.
25     Brody, D. J.; Pirkle, J. L.; Kramer, R. A.; Flegal, K. M.; Matte, T. D.; Gunter, E. W.; Paschal, D. C. (1994) Blood
26            lead levels in the US population: phase 1 of the third National Health and Nutrition Examination Survey
27            (NHANES III, 1988 to 1991). JAMA J. Am. Med. Assoc. 272: 277-283.
28     Bruschweiler, B. J.; Wurgler, F. E.; Pent, K. (1996) Inhibitory effects of heavy metals on cytochrome P4501A
29            induction in permanent fish hepatoma cells. Arch.  Environ. Contam. Toxicol. 31: 475-482.
30     Buchet, J.-P.; Roels, H. E.; Hubermont, G.; Lauwerys, R. (1976) Effect of lead on some parameters of the heme
31            biosynthetic pathway  in rat tissues in vivo. Toxicology 6: 21-34.
32     Buckley, J. D.; Robison, L. L.; Swotinsky, R.; Garabrant, D. H.; LeBeau,  M.; Manchester, P.; Nesbit, M. E.; Odom,
33            L.; Peters, J. M.; Woods, W. G.; Hammond, G. D.  (1989) Occupational exposures of parents of children
34            with acute nonlymphocytic  leukemia: a report from the children's cancer study group. Cancer Res. 49: 4030-
35            4037.
36     Bunn, T. L.; Marsh, J. A.; Dietert, R. R. (2000) Gender differences in developmental immunotoxicity to lead in a
37            chicken: analysis following a single early low-level exposure in ovo. J. Toxicol. Environ. Health A 61: 677-
38            693.
39     Bunn, T. L.; Parsons, P. J.; Kao, E.;  Dietert, R. R. (2001a)  Gender-based profiles of developmental immunotoxicity
40            to lead in the rat: assessment in juveniles and adults. J. Toxicol. Environ. Health A 64: 223-240.
41     Bunn, T. L.; Ladies, G.  S.; Holsapple, M. P.; Dietert, R. R. (2001b) Developmental immunotoxicology assessment
42            in the rat: age, gender and strain comparisons  after exposure to Pb. Toxicol. Methods 11: 41-58.
43     Bunn, T. L.; Parsons, P. J.; Kao, E.;  Dietert, R. R. (2001c)  Exposure to lead during critical windows of embryonic
44            development: differential immunotoxic outcome based on stage of exposure and gender. Toxicol. Sci. 64:
45            57-66.
46     Burchfiel, J. L.; Durry, F. H.; Bartels, P. H.; Needleman, H. L. (1992) Low-level lead exposure: effect on
47            quantitative electroencephalography and correlation with neuropsychologic measures. In: Needleman, H. L.,
48            ed. Human lead exposure. Boca Raton, FL: CRC Press; pp. 209-222.
49     Burkey, R. T.; Nation, J. R. (1994) Brain stimulation reward following chronic lead exposure in rats. Behav.
50            Neurosci. 108: 532-536.
51     Burstein, H. J.; Tepper, R. I.; Leder, P.; Abbas, A. K. (1991) Humoral immune functions in IL-4 transgenic mice. J.
52            Immunol. 147: 2950-2956.
53     Cai, M. Y.; Arenaz, P. (1998)  Antimutagenic effect of crown ethers on heavy metal-induced  sister chromatid
54            exchanges. Mutagenesis 13: 27-32.
55     Calabrese, E. J.; Baldwin, L. A. (1992) Lead-induced cell proliferation and organ-specific tumorigenicity. Drug
56            Metab. Rev. 24:409-416.

       December 2005                                5-300         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Calabrese, E. J.; Baldwin, L. A.; Leonard, D. A.; Zhao, X. Q. (1995) Decrease in hepatotoxicity by lead exposure is
  2            not explained by its mitogenic response. J. Appl. Toxicol. 15: 129-132.
  3     Calderon-Salinas, J. V.; Quintanar-Escorza, M. A.; Hernandez-Luna, C. E.; Gonzalez-Martinez, M. T. (1999a)
  4            Effect of lead on the calcium transport in human erythrocyte. Hum. Exp. Toxicol. 18: 146-153.
  5     Calderon-Salinas, J. V.; Quintanar-Escorcia, M. A.; Gonzalez-Martinez, M. T.; Hernandez-Luna, C. E. (1999b)
  6            Lead and calcium transport in human erythrocyte. Hum. Exp. Toxicol. 18: 327-332.
  7     Camoratto, A. M.; White, L. M.; Lau, Y.-S.; Ware, G. O.; Berry, W. D.; Moriarty, C. M. (1993) Effect of exposure
  8            to low level lead on growth and growth hormone release in rats. Toxicology 83: 101-114.
  9     Campbell, T. F.; Needleman, H. L.; Riess, J. A.; Tobin, M. J. (2000) Bone lead levels and language processing
10            performance. Dev. Neuropsychol. 18: 171-186.
11     Canfield, R. L.; Henderson, C. R., Jr.; Cory-Slechta, D. A.; Cox,  C.; Jusko, T. A.; Lanphear, B. P. (2003)
12            Intellectual impairment in children with blood lead concentrations below 10 micrograms per deciliter. N.
13            Engl. J. Med. 348: 1517-1526.
14     Canfield, R. L.; Gendle, M. H.; Cory-Slechta, D. A. (2004) Impaired neuropsychological functioning in lead-
15            exposed children. Dev. Neuropsychol. 26: 513-540.
16     Carballido, J. M.; Schols, D.; Namikawa, R.; Zurawski, S.; Zurawski, Z.; Roncarolo, M. G.; de Vries, J. E. (1995)
17            IL-4 induces human B cell maturation and IgE synthesis in SCID-hu mice. Inhibition of ongoing IgE
18            production by in vivo treatment with an IL-4/IL-13 receptor antagonist. J. Immunol. 155: 4162-4170.
19     Cardenas, A.; Roels, H.; Bernard, A. M.; Barbon, R.; Buchet, J. P.; Lauwerys, R. R.; Rosello, J.; Ramis, I.; Mutti,
20            A.; Franchini, I.; Pels, L. M.;  Stolte, H.; De Broe, M. E.; Nuyts, G. D.; Taylor, S. A.; Price, R. G. (1993)
21            Markers of early renal changes induced by industrial pollutants. II. Application to workers exposed to lead.
22            Br. J. Ind. Med. 50: 28-36.
23     Cardinale, F.; de Benedictis, F. M.; Muggeo, V.; Giordana, P.; Loffredo, M. S.; lacoviello, G.; Armenio, L. (2005)
24            Exhaled nitric oxide, total serum IgE and allergic sensitization in childhood asthma and allergic rhinitis.
25            Pediatr. Allergy Immunol. 16: 236-242.
26     Carmignani, M.; Boscolo, P.; Poma, A.; Volpe, A. R. (1999) Kininergic system and arterial hypertension following
27            chronic exposure to inorganic lead.  Immunopharmacology 44:  105-110.
28     Carmignani, M.; Volpe, A. R.; Boscolo, P.;  Qiao, N.; Gioacchino, M. Di; Grilli, A.; Felaco, M. (2000) Catcholamine
29            and nitric oxide systems as targets of chronic lead exposure in inducing selective functional impairment.
30            Life Sci. 68: 401-415.
31     Carsia, R. V.; Forman, D.; Hock, C. E.; Nagele, R. G.; Mcllroy, P. J. (1995) Lead alters growth and reduces
32            angiotensin II receptor density of rat aortic smooth muscle cells. Proc. Soc. Exp. Biol. Med. 210: 180-190.
33     Castranova, V.; Bowman, L.; Reasor,  M. J.; Miles, P. R. (1980) Effects of heavy metal ions on selected oxidative
34            metabolic processes in rat alveolar macrophages. Toxicol. Appl. Pharmacol. 53: 14-23.
3 5     Chai, S.; Webb, R. C. (1988) Effects of lead on vascular reactivity. Environ. Health Perspect. 78: 85-89.
36     Chakraborty, I.; Sharma, A.; Talukder, G. (1987) Antagonistic and synergistic effects of lead and selenium in Rattus
37            norvegicus. Toxicol. Lett. 37: 21-26.
38     Chang, H.-R.; Chen, S.-S.; Chen, T.-J.; Ho,  C.-K.; Chiang, H.-C; Yu, H.-S. (1996) Lymphocyte (32-adrenergic
39            receptors and plasma catecholamine levels in lead-exposed workers. Toxicol. Appl. Pharmacol. 139: 1-5.
40     Chang, H.-R.; Chen, S.-S.; Tsao, D.-A.; Cheng, J.-T.; Ho, C.-K.; Yu, H.-S. (1997) Change of cardiac (3-
41            adrenoceptors in lead-exposed rats.  Toxicology 123: 27-32.
42     Chang, H.-R.; Tsao, D.-A.; Yu, H.-S.; Ho, C.-K. (2005) The change of "beta"-adrenergic system after cessation of
43            lead exposure. Toxicology 207: 73-80.
44     Chaurasia, S. S.; Kar, A. (1997) Protective effects of vitamin E against lead-induced deterioration of membrane
45            associated type-I iodothyronine 5'-monodeiodinase (5'D-I) activity in male mice. Toxicology 124: 203-209.
46     Chavez, E.; Jay, D.; Bravo, C. (1987)  The mechanism of lead-induced mitochondrial Ca2+ efflux. J. Bioenerg.
47            Biomembrane 19: 285-295.
48     Chen, S.; Miller, T. E.; Golemboski, K. A.; Dietert,  R. R. (1997)  Suppression of macrophage metabolite production
49            by lead glutamate in vitro is reversed by meso-2, 3-dimercaptosuccinic acid (DMSA). In Vitro Toxicology.
50            10:351-357.
51     Chen, S.; Golemboski, K. A.; Sanders, F. S.; Dietert, R. R. (1999) Persistent effect of in utero meso-2,3-
52            dimercaptosuccinic acid (DMSA) on immune function and lead-induced immunotoxicity. Toxicology 132:
53            67-79.
54     Chen, S. C.; Golemboski, K. A.; Piepenbrink, M.; Dietert, R. R. (2004) Developmental immunotoxicity of lead in
55            the rat: influence of maternal diet. J. Toxicol. Environ.  Health Part A 67: 495-511.
       December 2005                                5-301         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Chen, A.; Dietrich, K. N.; Ware, J. H.; Radcliffe, I; Rogan, W. J. (2005) IQ and blood lead from 2 to 7 years of age:
  2            are the effects in older children the residual of high blood lead concentrations in 2-year-olds? Environ.
  3            HealthPerspect. 113: 597-601.
  4     Cheng, Y.; Willett, W. C.; Schwartz, J.; Sparrow, D.; Weiss, S.; Hu, H. (1998) Relation of nutrition to bone lead and
  5            blood lead levels in middle-aged to elderly men. The Normative Aging Study. Am. J. Epidemiol. 147:  1162-
  6            1174.
  7     Chia, K. S.; Mutti, A.; Tan, C.; Ong, H. Y.; Jeyaratnam, J.; Ong, C. N.; Lee, E. (1994) Urinary N-acetyl-"beta"-D-
  8            glucosaminidase activity in workers exposed to inorganic lead. Occup. Environ. Med. 51: 125-129.
  9     Choie, D. D.; Richter, G. W. (1972) Lead poisoning: rapid formation of intranuclear inclusions.  Science
10            (Washington, DC) 177:  1194-1195.
11     Choie, D. D.; Richter, G. W.; Young, L. B. (1975) Biogenesis of intranuclear lead-protein inclusions in mouse
12            kidney. Beitr. Pathol. 155: 197-203.
13     Chowdhuri, D. K.; Narayan, R.;  Saxena, D. K. (2001) Effect of lead and chromium on nucleic acid and protein
14            synthesis during sperm-zona binding in mice. Toxicol. In Vitro 15: 605-613.
15     Chowdhury, A. R.; Dewan, A.; Gandhi, D. N. (1984) Toxic effect of lead on the testes of rat. Biomed. Biochim.
16            Acta 43: 95-100.
17     Chowdhury, A. R.; Rao, R. V.; Gautam, A. K. (1986)  Histochemical changes in the testes of lead induced
18            experimental rats. Folia Histochem. Cytobiol. 24: 233-237.
19     Chowdhury, A. R.; Rao, R. V.; Gautam, A. K.; Kashyap, S. K. (1987) Functional changes of testes in lead
20            intoxicated rats. Ind. Health 25: 55-62.
21     Church, H. J.; Day, J. P.; Braithwaite, R. A.; Brown, S. S. (1993a) Binding of lead to a metallothionein-like protein
22            in human erythrocytes. J. Inorg. Biochem. 49: 55-68.
23     Church, H. J.; Day, J. P.; Braithwaite, R. A.; Brown, S. S. (1993b) The speciation of lead in erythrocytes in relation
24            to lead toxicity: case studies of two lead-exposed workers. Neurotoxicology 14: 359-364.
25     Claudio, L.; Lee, T.; Wolff, M. S.; Wetmur, J. G. (1997) A murine model of genetic susceptibility to lead
26            bioaccumulation. Fundam. Appl.  Toxicol. 35: 84-90.
27     Cline, H. T.; Witte, S.; Jones, K. W. (1996) Low lead  levels stunt neuronal growth in a reversible manner. Proc.
28            Natl. Acad. Sci. U. S. A. 93: 9915-9920.
29     Cocco, P.; Carta, P.; Flore, C.; Congia, P.; Manca, M. B.; Saba, G.; Salis, S. (1996) Mortality of lead smelter
30            workers with the glucose-6-phosphate dehydrogenase-deficient phenotype. Cancer Epidemiol. Biomarkers
31            Prev. 5: 223-225.
32     Cocco, P.; Dosemeci, M.; Heineman, E. F. (1998) Brain cancer and occupational exposure to lead. J. Occup.
33            Environ. Med. 40: 937-942.
34     Cohn, J.; Cory-Slechta, D. A. (1993) Subsensitivity of lead-exposed rats to the accuracy-impairing and rate-altering
35            effects of MK-801 on a multiple schedule of repeated learning and performance. Brain Res. 600: 208-218.
36     Cohn, J.; Cory-Slechta, D. A. (1994a) Assessment of the role of dopaminergic systems in lead-induced learning
37            impairments using a repeated acquisition and performance baseline. Neurotoxicology 15: 913-926.
38     Cohn, J.; Cory-Slechta, D. A. (1994b) Lead exposure potentiates the effects of NMD A on repeated learning.
39            Neurotoxicol. Teratol. 16: 455-465.
40     Cohn, J.; Cox, C.; Cory-Slechta, D. A. (1993) The effects of lead exposure on learning in a multiple repeated
41            acquisition and performance schedule. Presented at: Ninth international neurotoxicology conference;
42            October 1991; Little Rock, AR. Neurotoxicology 14(2-3):  329-346.
43     Cole, L. J.; Bachhuber, L. J. (1915) The effect of lead on the germ cells of the male rabbit and fowl as indicated by
44            their progeny. Proc. Soc. Exp. Biol. Med. 12:  24-29.
45     Columbano, A.; Ledda-Columbano, G. M.; Ennas, M. G.; Curto, M.; de Montis, M. G.; Roomi,  M. W.; Pani, P.;
46            Sarma, D. S.  R. (1988) Modulation of the activity of hepatic gamma-glutamyl transpeptidase, adenosine
47            triphosphatase, placental glutathione S-transferase and adenylate cyclase by acute administration of lead
48            nitrate. Basic Appl. Histochem. 32: 501-510.
49     Columbano, A.; Endoh, T.; Denda, A.; Noguchi, O.; Nakae, D.; Hasegawa, K.; Ledda-Columbano, G. M.; Zedda, A.
50            I.; Konishi, Y. (1996) Effects of cell proliferation and cell death (apoptosis and necrosis) on the  early stages
51            of rat hepatocarcinogenesis. Carcinogenesis 17: 395-400.
52     Coni, P.; Bignone, F. A.; Pichiri, G.; Ledda-Columbano, G. M.; Columbano, A.; Rao, P. M.; Rajalakhmi, S.; Sarma,
53            D. S. (1989) Studies on the kinetics of expression of cell cycle dependent proto-oncogenes during mitogen-
54            induced liver cell proliferation. Cancer Lett. (Shannon, Irel.) 47: 115-119.
       December 2005                                5-302         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Coni, P.; Pichiri-Coni, G.; Ledda-Columbano, G. M.; Sample, E.; Rajalakshmi, S.; Rao, P. M.; Sarma, D. S. R.;
  2            Columbano, A. (1992) Stimulation of DNA synthesis by rat plasma following in vivo treatment with three
  3            liver mitogens. Cancer Lett. 61: 233-238.
  4     Coni, P.; Simbula, G.; De Prati, A. C.; Menegazzi, M.;  Suzuki, H.; Sarma, D. S. R.; Ledda-Columbano, G. M.;
  5            Columbano, A. (1993) Differences in the steady-state levels of c-fos, c-jun and c-myc messenger RNA
  6            during mitogen-induced liver growth and compensatory regeneration. Hepatology (Baltimore) 17: 1109-
  7            1116.
  8     Cook, J. A.; Hoffmann, E. O.; Di Luzio, N. R. (1975) Influence of lead and cadmium on the susceptibility of rats to
  9            bacterial challenge. Proc. Soc. Exp. Biol. Med. 150: 741-747.
10     Cook, L. R.; Stohs, S. J.; Angle, C. R.; Hickman, T. I.;  Maxell, R. C. (1987) Erythrocyte membrane microviscosity
11            and phospholipid composition in lead workers. Br. J. Ind. Med. 44: 841-844.
12     Cooper, G. P.; Manalis, R.  S. (1984) Interactions of lead and cadmium on acetylcholine release at the frog
13            neuromuscularjunction. Toxicol. Appl. Pharmacol. 74: 411-416.
14     Corchs, J.; Gioia, I. A.; Serrani, R. E.; Taborda, D. (2001) Lead ions but not other metallic ions increase resistance
15            to hypotonic lysis in prenatal hemopoiesis red blood cells. Biocell 25: 287-289.
16     Corpas, I.; Caspar, I.; Martinez, S.; Codesal, J.; Candelas, S.; Antonio, M. T. (1995) Testicular alterations in rats due
17            to gestational and early lactational administration of lead. Reprod. Toxicol. 9: 307-313.
18     Corpas, I.; Benito, M. J.; Marquina, D.; Castillo, M.; Lopez, N.; Antonio, M. T. (2002) Gestational and lactational
19            lead intoxication produces alterations in the hepatic system of rat pups. Ecotoxicol. Environ. Saf. 51: 35-43.
20     Cory-Slechta, D. A. (1988) Mobilization of lead over the course of DMSA chelation therapy and long-term efficacy.
21            J. Pharmacol. Exp. Ther. 246: 84-91.
22     Cory-Slechta, D. A. (1990a) Alterations in tissue Pb distribution and hematopoietic indices during advanced age.
23            Arch. Toxicol. 64: 31-37.
24     Cory-Slechta, D. A. (1990b) Lead exposure during advanced age: alterations in kinetics and biochemical effects.
25            Toxicol. Appl. Pharmacol. 104: 67-78.
26     Cory-Slechta, D. A. (1995) MK-801 subsensitivity following postweaning lead exposure. Neurotoxicology  16: 83-
27            95.
28     Cory-Slechta, D. A. (1997a) Postnatal lead exposure and MK-801 sensitivity. Neurotoxicology 18: 209-220.
29     Cory-Slechta, D. A. (1997b) Relationships between Pb-induced changes in neurotransmitter system function and
30            behavioral toxicity. Neurotoxicology 18: 673-688.
31     Cory-Slechta, D. A. (2003) Lead-induced impairments  in complex cognitive function: offerings from experimental
3 2            studies. Child Neuropsychol. 9: 54-75.
33     Cory-Slechta, D. A.; Pokora, M. J. (1995) Lead-induced changes in muscarinic cholinergic sensitivity.
34            Neurotoxicology 16: 337-347.
35     Cory-Slechta, D. A.; Widzowski, D. V. (1991) Low level lead exposure increases sensitivity to the stimulus
3 6            properties of dopamine D1 and D2 agonists. Brain Res. 553: 65-74.
37     Cory-Slechta, D. A.; Weiss, B.; Cox, C. (1989) Tissue distribution of Pb in adult vs. old rats: a pilot study.
38            Toxicology 59: 139-149.
39     Cory-Slechta, D. A.; Pokora, M. J.; Widzowski, D. V. (1992) Postnatal lead exposure induces supersensitivity to the
40            stimulus properties of D2-D3 agonist. Brain Res. 598:  162-172.
41     Cory-Slechta, D. A.; Pokora, M. J.; Fox, R. A. V.;  O'Mara, D. J. (1996a) Lead-induced changes in dopamine Dl
42            sensitivity: modulation by drug discrimination  training. Neurotoxicology. 17: 445-457.
43     Cory-Slechta, D. A.; Pokora, M. J.; Johnson, J. L. (1996b) Postweaning lead exposure enhances the stimulus
44            properties of N-methyl-D-aspartate: possible dopaminergic  involvement? Neurotoxicology 17: 509-521.
45     Cory-Slechta, D. A.; McCoy, L.; Richfield, E. K. (1997) Time course and regional basis of Pb-induced changes in
46            MK-801 binding: reversal by chronic treatment with the dopamine agonist apomorphine but not the D1
47            agonist SKF-82958. J.  Neurochem. 68: 2012-2023.
48     Cory-Slechta, D. A.; Virgolini, M. B.; Thiruchelvam, M.; Weston, D.  D.; Bauter, M. R.  (2004) Maternal stress
49            modulates the effects of developmental lead exposure.  Environ. Health Perspect. 112: 717-730.
50     Costa, M.; Zhitkovich, A.; Gargas, M.; Paustenbach, D.; Finley, B.; Kuykendall, J.; Billings, R.; Carlson, T. J.;
51            Wetterhahn, K.; Xu, J.; Patierno, S.;  Bogdanffy, M. (1996) Interlaboratory validation of a new assay for
52            DNA-protein crosslinks. Mutat. Res. 369:  13-21.
53     Courtois, E.; Marques, M.; Barrientos, A.; Casado, S.; Lopez-Farre, A. (2003) Lead-induced downregulation of
54            soluble guanylage cyclase in isolated rat aortic  segments mediated by reactive oxygen species and
55            cyclooxygenase-2. J. Am. Soc. Nephrol. 14: 1464-1470.


       December 2005                                 5-303         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Craan, A. G.; Nadon, G.; P'an, A. Y. (1984) Lead flux through the kidney and salivary glands of rats. Am. J.
  2            Physiol. 247: F773-F783.
  3     Cramer, K.; Goyer, R. A.; Jagenburg, R.; Wilson, M. H. (1974) Renal ultrastructure, renal function, and parameters
  4            of lead toxicity in workers with different periods of lead exposure. Br. J. Ind. Med. 31: 113-127.
  5     Cremin, J. D., Jr.; Luck, M. L.; Laughlin, N. K.; Smith, D. R. (2001) Oral succimer decreases the gastrointestinal
  6            absorption of lead in juvenile monkeys. Environ. Health Perspect.  109: 613-619.
  7     Crowe, A.; Morgan, E. H. (1996) Interactions between tissue uptake of lead and iron in normal and iron-deficient
  8            rats during development. Biol.  Trace Elem. Res. 52: 249-261.
  9     Curzon, M. E. J.; Bibby, B. G. (1970) Effect of heavy metals on dental caries and tooth eruption. J. Dent. Child. 37:
10            463-465.
11     Daggett, D. A.; Nuwaysir, E. F.; Nelson, S. A.; Wright, L. S.; Kornguth, S. E.; Siegel, F. L. (1997) Effects of
12            triethyl lead administration on the  expression of glutathione S-transferase isoenzymes and quinone reductase
13            in rat kidney and liver. Toxicology 117:61-71.
14     Daggett, D. A.; Oberley, T. D.; Nelson, S.  A.; Wright, L. S.; Kornguth, S. E.; Siegel, F. L. (1998) Effects of lead on
15            rat kidney and liver: GST expression and oxidative stress. Toxicology 128: 191-206.
16     Danadevi, K.; Rozati, R.; SalehaBanu, B.; HanumanthR. P.; Grover, P. (2003) DNA damage in workers exposed to
17            lead using comet assay. Toxicology 187: 183-193.
18     Dantzer, R.; Bluthe, R. M.; Gheusi, G.; Cremona, S.; Laye, S.; Parnet, P.; Kelley, K. W. (1998) Molecular basis of
19            sickness behavior. Ann.  N.Y. Acad. Sci. 856: 132-138.
20     De Guise, S.; Bernier, J.; Lapierre, P.; Dufresne, M. M.; Dubreuil, P.; Fornier, M. (2000) Immune function of bovine
21            leukocytes after in vitro  exposure to selected heavy metals. Am. J. Vet. Res. 61: 339-344.
22     De Vries, I.; Spaans, E.; VanDijk, A.; Meulenbelt, J. (1998) Lead toxicokinetics. Development of a biokinetic
23            model to understand and predict the outcome of treatment. Przegl. Lek. 55: 500-504.
24     De la Fuente, H.; Portales-Perez, D.; Diaz-Barriga, F.; Saavedra-Alanis, V.; Layseca, E.; Gonzalez-Amaro, R.
25            (2002) Effect of arsenic, cadmium and lead on the induction of apoptosis of normal human mononuclear
26            cells. Clin. Exp. Immunol. 129: 69-77.
27     De, M.; Ghosh, S.; Palit, S.; Ghosh, A.; Talukder, G.; Sharma, A. (1995) Clastogenic effects in human samples
28            following prolonged exposure in metal industry. Bull. Environ. Contam. Toxicol. 54: 357-362.
29     Dearth, R. K.; Hiney, J. K.; Srivastava, V.; Burdick, S. B.; Bratton, G. R.; Dees, W. L. (2002) Effects of lead (Pb)
30            exposure during gestation and lactation on female pubertal development in the rat. Reprod. Toxicol. 16:
31            343-352.
32     Dearth, R. K.; Hiney, J. K.; Srivastava, V.; Les Dees, W.; Bratton, G. R. (2004) Low level lead (Pb) exposure during
3 3            gestation and lactation: assessment of effects on pubertal development in Fisher 344 and Sprague-Dawley
34            female rats. Life Sci. 74: 1139-1148.
3 5     Degawa, M.; Arai, H.; Kubota, M.; Hashimoto, Y. (1994) Ionic lead, a unique metal ion as an inhibitor for
36            cytochrome P450IA2 (CYP1A2) expression in the rat liver. Biochem. Biophys. Res. Commun. 200: 1086-
37            1092.
38     Degawa, M.; Arai, H.; Kubota, M.; Hashimoto, Y. (1995) Ionic lead, but not other ionic metals (Ni2+,  Co2+ and
39            Cd2+), suppresses 2-methoxy-4-aminoazobenzene-mediated cytochrome P450IA2 (CYP1A2)  induction in
40            rat liver. Biol. Pharm. Bull. 18: 1215-1218.
41     Degawa, M.; Arai, H.; Miura, S.; Hashimoto, Y. (1993) Preferential inhibitions of hepatic P450IA2 expression and
42            induction by lead nitrate in the  rat. Carcinogenesis (London) 14: 1091-1094.
43     Degawa, M.; Matsuda, K.; Arai, H.; Hashimoto, Y. (1996) Lead nitrate inhibits the induction of CYP1A mRNAs by
44            aromatic amines but not by aryl hydrocarbons in the rat liver. J. Biochem.  (Tokyo,  Jpn.) 120: 547-551.
45     Dehpour, A. R.; Essalat, M.; Ala, S.; Ghazi-Khansari, M.; Ghafourifar, P. (1999) Increase by NO synthase inhibitor
46            of lead-induced release of N-acetyl-"beta"-D-glucosaminidase from perfused rat kidney. Toxicology 132:
47            119-125.
48     Dekaney, C. M.; Harris, E. D.; Bratton, G.  R.; Jaeger, L. A. (1997) Lead transport in IEC-6 intestinal epithelial cells.
49            Biol. Trace Elem. Res. 58: 13-24.
50     Delville, Y. (1999) Exposure to lead during development alters aggressive behavior in golden hamsters.
51            Neurotoxicol. Teratol. 21: 445-449.
52     Deng, W.; Poretz, R. D. (2002) Protein kinase C activation is required for the lead-induced inhibition of
53            proliferation and differentiation of cultured oligodendroglial progenitor cells. Brain Res. 929: 87-95.
54     Deng, W.; McKinnon, R. D.; Poretz, R. D. (2001) Lead exposure delays the differentiation of oligodendroglial
55            progenitors in vitro. Toxicol. Appl. Pharmacol. 174: 235-244.


       December 2005                                5-304         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Dentener, M. A.; Greve, J. W.; Maessen, J. G.; Buurman, W. A. (1989) Role of tumour necrosis factor in the
  2            enhanced sensitivity of mice to endotoxin after exposure to lead. Immunopharmacol. Immunotoxicol. 11:
  3            321-334.
  4     Dessi, S.; Batetta, B.; Laconi, E.; Ennas, C.; Pani, P. (1984) Hepatic cholesterol in lead nitrate induced liver
  5            hyperplasia. Chem. Biol. Interact.  48: 271-279.
  6     Dessi, S.; Batetta, B.; Pulisci, D.; Carrucciu, A.; Mura, E.; Ferreli, A.; Pani, P. (1990) Modifying influence of fasting
  7            on liver hyperplasia induced by lead nitrate. Res. Commun.  Chem. Pathol. Pharmacol. 68: 103-116.
  8     Devi KD, Banu BS, Grover P, Jamil K. (2000) Genotoxic effect of lead nitrate on mice using SCGE (comet assay).
  9            Toxicology. 2000 Apr 14;145(2-3):195-201.
10     Dey, S.; Arjun, J.; Das, M.; Bhattacharjee, C. R.; Dkhar, P. S. (2001) Effect  of prenatal lead toxicity on surface
11            ultrastructural features, elemental composition and infrared absorption characteristics of the skin of albino
12            mice. Cytobios 106(suppl. 2): 245-254.
13     Dhir, H.; Roy, A. K.; Sharma, A.; Talukder, G. (1990) Modification of clastogenicity of lead and aluminium in
14            mouse bone marrow cells by dietary ingestion of Phyllanthus emblica fruit extract. Mutat. Res. 241: 305-
15            312.
16     Dhir, H.; Ghosh, S.; Sharma, A.; Talukder, G. (1992a) Interaction between two group IV metals: -lead and
17            zirconium - in bone marrow cells of Mus musculus in vivo. Biometals 5: 81-86.
18     Dhir, H.; Sharma, A.; Talukder, G. (1992b) Modifying effect of iron on lead-induced clastogenicity in mouse bone
19            marrow cells. Biol. Trace Elem. Res. 34: 279-286.
20     Dhir, H.; Roy, A. K.; Sharma, A. (1993) Relative efficiency  of Phyllanthus emblica fruit extract and ascorbic acid in
21            modifying lead and aluminium-induced sister-chromatid exchanges  in mouse bone marrow. Environ. Mol.
22            Mutagen. 21:229-236.
23     Diamond, G. L.; Goodrum, P. E.; Felter, S. P.; Ruoff, W. L.  (1998) Gastrointestinal absorption of metals. Drug
24            Chem.  Toxicol. 21: 223 -251.
25     Dieter, M. P.; Matthews, H. B.; Jeffcoat, R. A.; Moseman, R. F. (1993) Comparison of lead bioavailability inF344
26            rats fed lead acetate, lead oxide, lead sulfide, or lead ore concentrate from Skagway, Alaska. J. Toxicol.
27            Environ. Health 39: 79-93.
28     Dietert, R. R.; Etzel, R. A.; Chen, D.; Halonen, M.; Holladay, S. D.; Jarabek, A. M.; Landreth, K.; Peden, D. B.;
29            Pinkerton, K.; Smialowicz, R. J.; Zoetis, T. (2000) Workshop to  identify critical window of exposure for
30            children's health: immune and respiratory systems work group summary. Environ. Health Perspect. Suppl.
31            108(3): 483-490.
32     Dietert, R. R.; Lee, J.-E.; Bunn, T. L. (2002) Developmental immunotoxicology: emerging issues. Hum. Exp.
33            Toxicol. 21:479-485.
34     Dietert, R. R.; Lee, J.-E.; Hussain, L; Piepenbrink, M. (2004) Developmental immunotoxicology of lead. Toxicol.
35            Appl. Pharmacol. 86-94.
36     Dietrich, K. N.; Succop, P. A.; Berger, O. G.; Keith, R. W. (1992) Lead exposure and the  central auditory processing
37            abilities and cognitive development of urban children:  the Cincinnati lead study cohort at age 5 years.
38            Neurotoxicol. Teratol.  14: 51-56.
39     Dietrich, K. N.; Berger, O. G.;  Succop, P. A.; Hammond, P.  B.; Bornschein, R. L. (1993) The developmental
40            consequences of low to moderate prenatal and postnatal lead exposure: intellectual attainment in the
41            Cincinnati Lead Study Cohort following school entry. Neurotoxicol. Teratol. 15: 37-44.
42     Dietrich, K. N.; Ris, M. D.; Succop, P. A.; Berger, O. G.; Bornschein, R. L.  (2001) Early exposure to lead and
43            juvenile delinquency. Neurotoxicol. Teratol. 23: 511-518.
44     Dietrich, K. N.; Ware, J. H.; Salganik, M.; Radcliffe, J.; Rogan, W. J.; Rhoads, G. G.; Fay, M. E.; Davoli, C. T.;
45            Denckla, M. B.; Bornschein, R. L.; Schwarz, D.; Dockery, D. W.; Adubato, S.; Jones, R. L.; for the
46            Treatment of Lead-Exposed Children Clinical Trial  Group. (2004) Effect of chelation therapy on the
47            neuropsychological and behavioral development of lead-exposed children after school entry. Pediatrics 114:
48            19-26.
49     Ding, Y.; Vaziri, N. D.; Gonick, H.  C. (1998) Lead-induced  hypertension. II. Response to sequential infusions of L-
50            arginine, superoxide dismutase, and nitroprusside. Environ.  Res. 76: 107-113.
51     Ding, Y.; Gonick, H. C.; Vaziri, N. D. (2000) Lead promotes hydroxyl radical generation  and lipid peroxidation in
52            cultured aortic endothelial cells. Am. J. Hypertens. 13: 552-555.
53     Ding, Y.; Gonick, H. C.; Vaziri, N. D.; Liang, K.; Wei, L. (2001) Lead-induced hypertension. III. Increased
54            hydroxyl radical production. Am. J. Hypertens. 14: 169-173.
55     Dini, L.; Falasca, L.; Lentini, A.;  Mattioli, P.;  Piacentini, M.; Piredda, L.; Autuori, F. (1993) Galactose-specific
56            receptor modulation related to the onset of apoptosis in rat liver.  Eur. J. Cell. Biol. 61: 329-337.

       December 2005                                5-305         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Dini, L.; Giudetti, A. M; Ruzittu, M.; Gnoni, G. V.; Zara, V. (1999) Citrate carrier and lipogenic enzyme activities
  2            in lead nitrate-induced proliferative and apoptotic phase in rat liver. Biochem. Mol. Biol. Int. 47: 607-614.
  3     Donaldson, W. E.; Leeming, T. K. (1984) Dietary lead: effects on hepatic fatty acid composition in chicks. Toxicol.
  4            Appl. Pharmacol. 73: 119-123.
  5     Donaldson, I; Hemming, R.; LaBella, F. (1985) Vanadium exposure enhances lipid peroxidation in the kidney of
  6            rats and mice. Can. J. Physiol. Pharmacol. 63: 196-199.
  7     Dorman, R. V.; Freeman, E. J. (2002) Lead-dependent effects on arachidonic acid accumulation and the
  8            proliferation of vascular smooth muscle. J. Biochem. Mol. Toxicol. 16: 245-253.
  9     Dorward, A.; Yagminas, A. P. (1994) Activity of erythrocyte "delta"-aminolevulinic acid dehydratase in the female
10            cynomolgus monkey (Macaca fascicularis): kinetic analysis in control and lead-exposed animals.  Comp.
11            Biochem. Physiol. B: Biochem. Mol. Biol. 108: 241-252.
12     Dowd, T. L.;  Rosen, J. F.; Gupta, R. K.  (1990) 31P NMR and saturation transfer studies of the effect of PB2+ on
13            cultured osteoblastic bone cells. J. Biol. Chem. 265: 20833-20838.
14     Dowd, T. L.;  Rosen, J. F.; Gundberg, C. M.; Gupta, R. K. (1994) The displacement of calcium from osteocalcin at
15            submicromolar concentrations of free lead. Biochim. Biophys. Acta 1226: 131-137.
16     Dowd, T. L.;  Rosen, J. F.; Mints, L.;  Gundberg, C. M. (2001) The effect of Pb2+ on the structure and
17            hydroxyapatite binding properties of osteocalcin. Biochim. Biophys. Acta 1535: 153-163.
18     Driscoll, K. E. (2000) TNF alpha and MTP-2: Role in particle induced inflammation and regulation by oxidative
19            stress. Toxicol. Lett. 112-113: 177-183.
20     Dunon, D.; Allioli, N.; Vainio, O.; Ody, C.; Imhof, B. A. (1998) Renewal of thymocyte progenitors  and emigration
21            of thymocytes during avian development. Dev. Comp. Immunol. 22: 279-287.
22     Dursun, N.; Arifoglu, C.; Suer, C.; Keskinol, L. (2005) Blood pressure relationship to nitric oxide, lipid
23            peroxidation, renal function,  and renal blood flow in rats exposed to low lead levels. Biol. Trace Elem. Res.
24            104:  141-150.
25     DuVal, G.; Fowler, B. A.  (1989) Preliminary purification and characterization studies of a low molecular weight,
26            high  affinity cytosolic lead-binding protein in rat brain. Biochem. Biophys. Res. Commun. 159: 177-184.
27     Duydu, Y.; Suzen, H. S.; Aydin, A.; Gander, O.; Uysal, H.; Isimer, A.; Vural, N. (2001) Correlation between lead
28            exposure indicators and sister chromatid exchange (SCE) frequencies in lymphocytes from inorganic lead
29            exposed workers. Arch. Environ. Contam. Toxicol. 41: 241-246.
30     Dyatlov, V. A.; Lawrence, D. A. (2002) Neonatal lead exposure potentiates sickness behavior induced by Listeria
31            monocytogenes infection of mice. Brain Behav. Immun. 16: 477-492.
32     Dyatlov, V. A.; Dyatlova, O. M.; Parsons, P. J.; Lawrence, D. A.; Carpenter, D. O. (1998a) Lipopolysaccharide and
33            interleukin-6 enhance lead entry into cerebellar neurons: application of a new and sensitive flow cytometric
34            technique to measure intracellular lead and calcium concentrations. Neurotoxicology 19: 293-302.
35     Dyatlov, V. A.; Platoshin, A. V.; Lawrence, D.  A.; Carpenter, D. O. (1998b) Lead potentiates cytokine- and
36            glutamate-mediated increases in permeability of the blood-brain barrier. Neurotoxicology 19: 283-292.
37     Eder, K.; Reichlmayr-Lais, A. M.; Kirchgessner, M. (1990) Activity of Na-K-ATPase and Ca-Mg-ATPase in red
3 8            blood cell membranes of lead-depleted rats. J. Trace Elem. Electrolytes Health Dis. 4: 21-24.
39     Egle, P. M.; Shelton, K. R. (1986) Chronic lead intoxication causes a brain-specific nuclear protein to accumulate in
40            the nuclei of cells lining kidney tubules. J. Biol. Chem. 261: 2294-2298.
41     Eisenmann, D. R.; Yaeger, J. A. (1969) Alterations in the formation of rat dentine and enamel induced by various
42            ions.  Arch. Oral Biol. 14: 1045-1064.
43     El-Fawal, H.  A. N.; Waterman, S. J.;  De Feo, A.; Shamy, M. Y. (1999) Neuroimmunotoxicology: humoral
44            assessment of neurotoxicity and autoimmune mechanisms. Environ. Health Perspect. 107(suppl. 5): 767-
45            775.
46     Elias, Z.; Poirot, O.; Pezerat, H.; Suquet, H.; Schneider, O.; Daniere, M. C.; Terzetti, F.; Baruthio, F.; Fournier, M.;
47            Cavelier, C. (1989) Cytotoxic and neoplastic transforming effects of industrial hexavalent chromium
48            pigments in Syrian hamster embryo cells. Carcinogenesis 10: 2043-2052.
49     Elias, Z.; Poirot, O.; Baruthio, F.; Daniere, M. C. (1991) Role of solubilized chromium in the induction of
50            morphological transformation of Syrian hamster embryo (SHE) cells by paniculate chromium(VI)
51            compounds. Carcinogenesis 12: 1811-1816.
52     El-Missiry, M. A. (2000) Prophylactic effect of melatonin on lead-induced inhibition of heme biosynthesis and
53            deterioration of antioxidant systems in male rats. J. Biochem. Mol. Toxicol. 14: 57-62.
54     Elreedy, S.; Krieger, N.; Ryan, P. B.; Sparrow, D.; Weiss, S. T.; Hu, H. (1999) Relations between individual and
5 5            neighborhood-based measures of socioeconomic position and bone lead concentrations among community -
56            exposed men: the Normative Aging Study. Am. J. Epidemiol. 150: 129-141.

       December 2005                                5-306        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Englyst, V.; Lundstrom, N. G.; Gerhardsson, L.; Rylander, L.; Nordberg, G. (2001) Lung cancer risks among lead
  2            smelter workers also exposed to arsenic. Sci. Total Environ. 273: 77-82.
  3     Ercal, N.; Treeratphan, P.; Hammond, T. C.; Matthews, R. H.; Grannemann, N. H.; Spitz, D. R. (1996) In vivo
  4            indices of oxidative stress in lead-exposed C57BL/6 mice are reduced by treatment with meso-2,3-
  5            dimercaptosuccinic acid or N-acetylcysteine. Free Radical Biol. Med. 21:  157-161.
  6     Eriksson,  L. E. G.; Beving, H. (1993) Calcium- and lead-activated morphological changes in human erythrocytes: a
  7            spin label study of the cytoplasm. Arch. Biochem. Biophys. 303: 296-301.
  8     Escribano, A.; Revilla, M; Hernandez, E. R.; Seco, C.; Gonzalez-Riola, I; Villa, L. F.; Rico, H. (1997) Effect of
  9            lead on bone development and bone mass: a morphometric, densitometric, and histomorphometric study in
10            growing rats. Calcif. Tiss. Int. 60: 200-203.
11     Ewers, U.; Stiller-Winkler, R.; Idel, H.  (1982) Serum immunoglobulin, complement C3, and salivary IgA levels in
12            lead workers. Environ. Res. 29: 351-357.
13     Exon, J. H.; Koller, L. D.; Kerkvliet, N. I. (1979) Lead-cadmium interaction: effects on viral-induced mortality and
14            tissue residues in mice. Arch. Environ. Health 34: 469-475.
15     Fahmy, M. A. (1999) Lead acetate genotoxicity in mice. Cytologia 64: 357-365.
16     Faith, R. E.; Luster, M. I.; Kimmel,  C. A. (1979) Effect of chronic developmental lead exposure on cell-mediated
17            immune functions. Clin. Exp. Immunol. 35: 413-420.
18     Fanning, D. (1988) A mortality study of lead workers, 1926-1985. Arch. Environ. Health 43: 247-251.
19     Farant,  J.-P.; Wigfield, D. C. (1987) Interaction of divalent metal ions with normal and lead-inhibited human
20            erythrocytic porphobilinogen synthase in vitro. Toxicol. Appl. Pharmacol. 89: 9-18.
21     Farant,  J.-P.; Wigfield, D. C. (1990) The effects of copper, zinc, mercury, and cadmium on rabbit erythrocytic
22            porphobilinogen synthase in vivo. J. Anal. Toxicol. 14: 222-226.
23     Farias, P.; Borja-Aburto, V. H.; Rios, C.; Hertz-Picciotto, I.; Rojas-Lopez, M.; Chavez-Ayala, R. (1996) Blood lead
24            levels in pregnant women of high and low socioeconomic status in Mexico City. Environ. Health Perspect.
25            104: 1070-1074.
26     Farmand,  F.; Ehdale, A.; Roberts, C. K.; Sindhu, R. K.  (2005) Lead-induced dysregulation of superoxide
27            dismutases, catalase, glutathione peroxidase, and guanylate cyclase. Environ. Res. 98: 33-39.
28     Faust, D.; Brown, J. (1987) Moderately elevated blood lead levels: effects on neuropsychologic functioning in
29            children. Pediatrics 80: 623-629.
30     Faustman, E. M.; Silbernagel, S. M.; Fenske, R. A.; Burbacher, T. M.; Ponce, R. A. (2000) Mechanisms underlying
31            children's susceptibility to environmental toxicants. Environ. Health Perspect. Suppl.  108(1): 13-21.
32     Favalli, L.; Chiari, M. C.; Piccinini, F.; Rozza, A. (1977) Experimental investigations on the contraction induced by
33            lead in arterial smooth muscle. Acta Pharmacol. Toxicol. 41: 412-420.
34     Featherstone, J. D.; Goodman, P.; McLean,  J. D. (1979) Electron microscope study of defect zones in dental enamel.
35            J. Ultrastruct. Res. 67: 117-123.
36     Featherstone, J. D. B.; Nelson, D. G. A.; McLean,  J. D. (1981) An electron microscope study  of modifications to
37            defect regions in dental enamel and synthetic apatites. Caries Res. 15: 278-288.
3 8     Federal Register. (2001) Lead and lead compounds; lowering of reporting thresholds; community right-to-know
39            toxic chemicals release reporting: delay of effective date. F. R. 66 (February 16): 10585-10586.
40     Fehlau, R.; Grygorczyk, R.; Fuhrmann, G. F.; Schwarz, W. (1989) Modulation of the  calcium or lead-activated
41            potassium-selective channels in human red cells. II. Parallelisms to modulation of the activity of a
42            membrane-bound oxidoreductase. Biochim. Biophys. Acta 978: 37-42.
43     Pels, L. M.; Herbert, C.; Pergande, M.; Jung, K.; Hotter, G.; Rosello, J.; Gelpi, E.; Mutti, A.; De Broe, M.; Stolte, H.
44            (1994) Nephron target sites in chronic exposure to lead. Nephrol. Dial. Transplant. 9: 1740-1746.
45     Ferguson, C.; Kern, M.; Audesirk, G. (2000) Nanomolar concentrations of inorganic lead increase Ca2+ efflux and
46            decrease intracellular free Ca2+ ion concentrations in cultured rat hippocampal neurons by a calmodulin-
47            dependent mechanism. Neurotoxicology 21: 365-378.
48     Fergusson, D.  M.; Horwood, L. J.; Lynskey, M. T. (1997) Early dentine lead levels and educational outcomes at 18
49            years. J. Child Psychol. Psychiatry 38: 471-478.
50     Perm, V. H.; Carpenter, S. J.  (1967) Developmental malformations resulting from the administration of lead salts.
51            Exp. Mol. Pathol. 7:  208-213.
52     Fernandez-Cabezudo, M.  J.; Hassan, M. Y.; Mustafa, N.; El-Sharkawy, R.; Fahim, M. A.; Al-Ramada, K. (2003)
53            Alpha tocopherol protects against immunosuppressive and immunotoxic effects of lead. Free Radical Res.
54            37:437-445.
55     Filkins, J. P.; Buchanan, B. J. (1973) Effects of lead acetate on sensitivity to shock, intravascular carbon and
56            endotoxin clearances, and hepatic endotoxin detoxification. Proc. Soc. Exp. Biol. Med.  142: 471-475.

       December 2005                                5-307         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Finkelstein, Y.; Markowitz, M. E.; Rosen, J. F. (1998) Low-level lead-induced neurotoxicity in children: an update
  2            on central nervous system effects. Brain Res. Rev. 27: 168-176.
  3     Flohe, S. B.; Bruggemann, J.; Herder, C.; Goebel, C.; Kolb, H. (2002) Enhanced proinflammatory response to
  4            endotoxin after priming of macrophages with lead ions. J. Leukoc. Biol. 71:417-424.
  5     Flora, G. J. S.; Seth, P. K. (1999) Beneficial effects of S-adenosyl-L-methionine on aminolevulinic acid dehydratase,
  6            glutathione, and lipid peroxidation during acute lead~ethanol administration in mice. Alcohol 18: 103-108.
  7     Flora, S. J. S.; Tandon, S. K. (1987) Influence of calcium disodium edetate on the toxic effects of lead
  8            administration in pregnant rats. Indian J. Physiol. Pharmacol. 31: 261-212.
  9     Flora, S. J. S.; Singh, S.; Tandon, S. K. (1989) Thiamine and zinc in prevention or therapy of lead intoxication. J.
10            Int. Med. Res. 17: 68-75.
11     Flora, S. J. S.; Jeevaratnam, K.; Kumar, D. (1993) Preventive effects of sodium molybdate in lead intoxication in
12            rats. Ecotoxicol. Environ. Saf. 26: 133-137.
13     Flora, S. J. S.; Bhattacharya, R.; Sachan, S.  R. S. (1994) Dose-dependent effects of zinc supplementation during
14            chelation of lead in rats. Pharmacol. Toxicol. 74: 330-333.
15     Flora, S. J. S.; Pande, M.; Mehta, A. (2003) Beneficial effect of combined administration of some naturally
16            occurring antioxidants  (vitamins) and thiol chelators in the treatment of chronic lead intoxication. Chem.
17            Biol. Interact. 145: 267-280.
18     Fontanellas, A.; Navarro, S.; Moran-Jimenez, M.-J.;  Sanchez-Fructuoso, A. I.; Vegh, I.; Barrientos, A.; De
19            Salamanca, R. E. (2002) Erythrocyte aminolevulinate dehydratase activity as a lead marker in patients with
20            chronic renal failure. Am.  J. Kidney Dis. 40: 43-50.
21     Foote, R. H. (1999) Fertility of rabbit sperm exposed in vitro to cadmium and lead. Reprod. Toxicol. 13: 443-449.
22     Fortoul, T. I.; Moncada-Hernandez, S.; Saldivar-Osorio, L.; Espejel-Maya, G.; Mussali-Galante, P.; Del Carmen
23            Avila-Casando, M.; Colin-Barenque, L.; Hernandez-Serrato, M. L; Avila-Costa, M. R. (2005) Sex
24            differences in brochiolar epithelium response after the inhalation of lead acetate (Pb). Toxicology 207: 323-
25            330.
26     Foster, W. G. (1992) Reproductive toxicity  of chronic lead exposure in the female cynomolgus monkey. Reprod.
27            Toxicol. 6: 123-131.
28     Foster, W. G.; Stals, S. L; McMahon,  A. (1992) An ultrasound study of the effect of chronic lead exposure on
29            endometrial cycle changes in the female cynomolgus monkey. J. Med. Primatol. 21: 353-356.
30     Foster, W. G.; McMahon, A.; YoungLai, E. V.; Hughes, E. G.; Rice, D. C. (1993) Reproductive endocrine effects of
31            chronic lead exposure in the male cynomolgus monkey. Reprod. Toxicol. 7: 203-209.
32     Foster, W. G.; McMahon, A.; Rice, D. C. (1996a) Sperm chromatin structure is altered in cynomolgus monkeys with
33            environmentally relevant blood lead levels. Toxicol. Ind. Health 12:  723-735.
34     Foster, W. G.; McMahon, A.; Rice, D. C. (1996b) Subclinical changes in luteal function in cynomolgus monkeys
35            with moderate blood lead levels. J. Appl. Toxicol. 16:  159-163.
36     Foster, W. G.; Singh, A.; McMahon, A.; Rice, D. C.  (1998) Chronic lead exposure effects in the cynomolgus
37            monkey (Macaca fascicularis) testis. Ultrastruct. Pathol. 22: 63-71.
38     Fowler, B. A. (1992) Mechanisms of kidney cell injury from metals. Environ. Health Perspect. 100: 57-63.
39     Fowler, B. A.; DuVal, G. (1991) Effects of lead on the kidney: roles of high-affinity lead-binding proteins. Environ.
40            Health  Perspect. 91: 77-80.
41     Fowler, B. A.; Kimmel, C. A.; Woods, J. S.; McConnell, E. E.; Grant, L. D. (1980) Chronic low-level lead toxicity
42            in the rat. III. An integrated assessment of long-term toxicity with special reference to the kidney. Toxicol.
43            Appl. Pharmacol. 56: 59-77.
44     Fowler, B. A.; Kahng, M. W.; Smith, D. R.; Conner,  E. A.; Laughlin, N. K. (1993) Implications of lead binding
45            proteins for risk assessment of lead exposure. J. Exposure Anal. Environ. Epidemiol. 3: 441-448.
46     Fox, D. A.; Chu, L. W.-F. (1988) Rods are selectively altered by lead: II. ultrastructure and quantitative histology.
47            Exp. Eye Res. 46: 613-625.
48     Fox, D. A.; Farber, D. B. (1988) Rods are selectively altered by lead: I. electrophysiology and biochemistry. Exp.
49            Eye Res. 46:597-611.
50     Fox, D. A.; Rubinstein, S. D. (1989) Age-related changes in retinal  sensitivity, rhodopsin content and rod outer
51            segment length in hooded rats following low-level lead exposure during development. Exp. Eye Res. 48:
52            237-249.
53     Fox, D. A.; Sillman, A. J. (1979) Heavy metals affect rod, but not cone, photoreceptors. Science (Washington, DC)
54            206:78-80.
55     Fox, D. A.; Katz, L. M.; Farber, D. B. (1991a) Low level developmental lead exposure decreases the sensitivity,
56            amplitude and temporal resolution of rods. Neurotoxicology 12: 641-654.

       December 2005                                5-308         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Fox, D. A.; Rubinstein, S. D.; Hsu, P. (1991b) Developmental lead exposure inhibits adult rat retinal, but not kidney,
  2            Na+,K+-ATPase. Toxicol. Appl. Pharmacol. 109: 482-493.
  3     Fox, D. A.; Srivastava, D.; Hurwitz, R. L. (1994) Lead-induced alterations in rod-mediated visual functions and
  4            cGMP metabolism: new insights. Neurotoxicology 15: 503-512.
  5     Fox, D. A.; Campbell, M. L.; Blocker, Y. S. (1997) Functional alterations and apoptotic cell death in the retina
  6            following developmental or adult lead exposure. Neurotoxicology 18: 645-664.
  7     Fracasso, M. E.; Perbellini, L.; Solda, S.; Talamini, G.; Franceschetti, P. (2002) Lead induced DNA strand breaks in
  8            lymphocytes of exposed workers: role of reactive oxygen species and protein kinase C. Mutat. Res. 515:
  9            159-169.
10     Franklin, C. A.; Inskip, M. I; Baccanale, C. L.; Edwards, C. M.; Manton, W. I.; Edwards, E.; O'Flaherty, E. J.
11            (1997) Use of sequentially administered stable lead isotopes to investigate changes in blood lead during
12            pregnancy in a nonhuman primate (Macaca fascicularis). Fundam. Appl. Toxicol. 39: 109-119.
13     Franks, P. A.; Laughlin, N. K.; Dierschke, D.  J.; Bowman, R. E.; Meller, P. A. (1989) Effects of lead on luteal
14            function in rhesus monkeys. Biol. Reprod. 41: 1055-1062.
15     Fu, H.; Boffetta, P. (1995) Cancer and occupational exposure to inorganic lead compounds: a meta-analysis of
16            published data. Occup. Environ. Med. 52: 73-81.
17     Fuentes, M.; Torregrosa,  A.; Mora, R.; Gotzens, V.; Corbella, J.; Domingo, J.  L. (1996) Placenta! effects of lead in
18            mice. Placenta 17: 371-376.
19     Fujita, H.; Orii, Y.; Sano, S. (1981) Evidence  of increased synthesis of "delta"-aminolevulinic acid dehydratase in
20            experimental lead-poisoned rats. Biochim. Biophys. Acta 678: 39-50.
21     Fujita, H.; Sato, K,; Sano, S. (1982) Increase in the amount of Erythrocyte delta-aminolevulinic acid dehydratase in
22            workers with moderate lead exposure. Int. Arch. Occup. Environ. Health 50: 287-297.
23     Fujiwara, Y.; Kaji, T. (1999) Possible mechanism for lead inhibition of vascular endothelial cell proliferation: a
24            lower response to basic fibroblast growth factor through inhibition of  heparan sulfate synthesis. Toxicology
25            133: 147-157.
26     Fujiwara, Y.; Kaji, T.; Yamamoto, C.; Sakamoto, M.; Kozuka, H. (1995) Stimulatory effect of lead on the
27            proliferation of cultured vascular smooth-muscle cells. Toxicology 98: 105-110.
28     Fujiwara, Y.; Watanabe, S.; Sakamoto, M.; Kaji, T. (1998) Repair of wounded monolayers of cultured vascular
29            endothelial cells after simultaneous exposure to lead and zinc. Toxicol. Lett. 94: 181-188.
30     Fukumoto, K.; Karai, I.; Horiguchi, S. (1983) Effect of lead on erythrocyte membranes. Br. J. Ind. Med. 40: 220-
31            223.
32     Fullmer, C. S. (1991) Intestinal calcium and lead absorption: effects of dietary lead and calcium. Environ. Res. 54:
33            159-169.
34     Fullmer, C. S. (1992) Intestinal interactions of lead and calcium. Neurotoxicology  13: 799-807.
3 5     Fullmer, C. S. (1995) Dietary  calcium levels and treatment interval determine  the effects of lead ingestion on plasma
36            1,25-dihydroxyvitaminD concentration in chicks. J. Nutr. 125: 1328-1333.
37     Fullmer, C. S. (1997) Lead-calcium interactions: involvement of 1,25-dihydroxyvitamin D. Environ.  Res. 72: 45-
38            55.
39     Fullmer, C. S.; Rosen, J. F. (1990) Effect of dietary calcium and lead status on intestinal calcium absorption.
40            Environ. Res. 51:91-99.
41     Fullmer, C. S.; Edelstein, S.; Wasserman, R. H. (1985) Lead-binding properties of intestinal calcium-binding
42            proteins. J. Biol.  Chem. 260: 6816-6819.
43     Fullmer, C. S.; Chandra, S.; Smith, C. A.; Morrison, G. H.; Wasserman, R. H. (1996) Ion microscopic imaging of
44            calcium during 1,25-dihydroxyvitamin D-mediated intestinal absorption. Histochem. Cell Biol. 106: 215-
45            222.
46     Furono, K.; Suetsugu, T.; Sugihara, N. (1996) Effects of metal ions on lipid peroxidation  in cultured rat hepatocytes
47            loaded with alpha-linolenic acid. J. Toxicol. Environ. Health 48: 121-129.
48     Gainer, J. H. (1977) Effects of heavy metals and of deficiency of zinc on mortality rates in mice infected with
49            encephalomyocarditis virus. Am. J. Vet. Res. 38: 869-872.
50     Gallicchio, L.; Scherer, R. W.; Sexton, M. (2002) Influence of nutrient intake  on blood lead levels of young children
51            at risk for lead poisoning. Environ. Health Perspect. 110:  A767-A772.
52     Gandley, R.; Anderson, L.; Silbergeld, E. K. (1999) Lead: male-mediated effects on reproduction and development
53            in the rat. Environ. Res. A 89: 355-363.
54     Caravan, H.; Morgan, R.  E.; Levitsky, D. A.;  Hermer-Vazquez, L.; Strupp, B. J. (2000) Enduring effects  of early
55            lead exposure: evidence for a specific deficit in associative ability. Neurotoxicol.  Teratol. 22:  151-164.


       December  2005                                 5-309         DRAFT-DO  NOT QUOTE  OR CITE

-------
  1     Garcia, T. A.; Corrector, L. (2004) Biochemical changes in the kidneys after perinatal intoxication with lead and/or
  2            cadmium and their antagonistic effects when coadministered. Ecotoxicol. Environ. Saf. 57: 184-189.
  3     Gautam, A. K.; Chowdhury, A. R. (1987) Effect of lead on erythropoietic system of intact and splenectomized rats.
  4            Indian J. Physiol. Pharmacol. 31:117-124.
  5     Gerhardsson, L.; Brune, D.; Nordberg, G. F.; Wester, P. O. (1986) Distribution of cadmium, lead and zinc in lung,
  6            liver and kidney in long-term exposed smelter workers. Sci. Total Environ. 50: 65-85.
  7     Gerhardsson, L.; Hagmar, L.; Rylander, L.; Skerfving, S. (1995a Mortality and cancer incidence among secondary
  8            lead smelter workers. Occup. Environ. Med. 52: 667-672.
  9     Gerlach, R. F.; Souza, A. P.; Cury, J. A.; Line, S. R. P. (2000a) Effect of lead, cadmium and zinc on the activity of
10            enamel matrix proteinases in vitro. Eur. J. Oral Sci. 108: 327-334.
11     Gerlach, R. F.; Toledo, D. B.; Novaes, P. D.; Merzel, J.; Line, S. R. P. (2000b) The effect of lead on the eruption
12            rates of incisor teeth in rats. Arch. Oral Biol. 45: 951-955.
13     Gerlach, R. F.; Cury, J. A.; Krug, F. J.; Line, S. R. P. (2002) Effect of lead on dental enamel formation. Toxicology
14            175:27-34.
15     Gerr, F.; Letz, R.; Stokes, L.; Chettle, D.; McNeill, F.; Kaye,  W. (2002) Association between bone lead
16            concentration and blood pressure among young adults. Am.  J. Ind. Med. 42: 98-106.
17     Gewirtz, A. T.; Liu, Y.;  Sitaraman, S. V.; Madara, J. L. (2002) Intestinal epithelial pathobiology: past, present and
18            future. Best Pract. Res. Clin. Gastroenterol. 16: 851-867.
19     Giavini, E.; Prati, M.; Vismara, C. (1980) Effects of cadmium, lead and copper on rat preimplantation embryos.
20            Bull. Environ. Contam. Toxicol. 25: 702-705.
21     Gibson, S. L. M.; Goldberg, A. (1970) Defects in haem synthesis in mammalian tissues in experimental lead
22            poisoning and experimental porphyria. Clin. Sci. 38:  63-72.
23     Gilbert, M. E.; Mack, C. M. (1998) Chronic lead exposure accelerates decay of long-term potentiation in rat dentate
24            gyms in vivo. Brain Res. 789: 139-149.
25     Gilbert, M. E.; Mack, C. M.; Lasley, S. M. (1996) Chronic developmental lead exposure increases the threshold for
26            long-term potentiation in rat dentate gyrus in vivo. Brain Res. 736: 118-124.
27     Gilbert, M. E.; Mack, C. M.; Lasley, S. M. (1999a) The influence of developmental period of lead exposure on long-
28            term potentiation in the adult rat dentate gyrus in vivo. Neurotoxicology 20: 57-69.
29     Gilbert, M. E.; Mack, C. M.; Lasley, S. M. (1999b) Chronic developmental lead exposure and hippocampal long-
30            term potentiation: biphasic dose-response relationship. Neurotoxicology 20: 71-82.
31     Gilbert, M. E.; Kelly, M. E.; Samsam, T. E.; Goodman, J.  H.  (2005) Chronic developmental lead exposure reduces
32            neurogenesis in adult rat hippocampus but does not impair spatial learning. Toxicol. Sci. 86:
33     Giridhar, J.; Isom, G. E. (1990) Interaction of lead acetate with atrial natriuretic factor in rats. Life Sci. 46: 569-576.
34     Gobel,  T. W. F. (1996) The T-dependent immune system.  In.  Davidson, T. F.; Morris, T. R.; Payne, L. N., ed.
3 5            Poultry immunology. Carfax Pub. Abingdon. 83-114.
36     Goebel, C.; Kirchhoff, K.; Wasmuth, H.; Flohe, S. B.; Elliott, R. B.; Kolb, H. (1999) The gut cytokine balance as a
37            target of lead toxicity. Life Sci. 64: 2207-2214.
38     Goebel, C.; Flohe, S. B.; Kirchhoff, K.; Herder, C.; Kolb,  H.  (2000)  Orally administered lead chloride induces bias
39            of mucosal immunity. Cytokine 12: 1414-1418.
40     Goering, P. L.; Fowler, B. A. (1984) Regulation of lead inhibition of delta-aminolevulinic acid dehydratase by alow
41            molecular weight, high affinity renal lead-binding protein. J. Pharmacol. Exp. Ther. 231: 66-71.
42     Goering, P. L.; Fowler, B. A. (1985) Mechanism of renal lead-binding protein reversal of "delta"-aminolevulinic
43            acid dehydratase inhibition by lead. J. Pharmacol. Exp. Ther. 234: 365-371.
44     Goering, P. L.; Mistry, P.; Fowler, B. A. (1986) A low molecular weight lead-binding protein in brain attenuates
45            lead inhibition of "delta"-aminolevulinic acid dehydratase: comparison with a renal lead-binding protein. J.
46            Pharmacol. Exp. Ther. 237: 220-225.
47     Goering, P. L.; Fowler, B. A. (1987a) Regulatory roles of high-affinity metal-binding proteins in mediating lead
48            effects on 5-aminolevulinic acid dehydratase. Ann. N. Y. Acad. Sci. 514: 235-247.
49     Goering, P. L.; Fowler, B. A. (1987b) Kidney zinc-thionein regulation of delta-aminolevulinic acid dehydratase
50            inhibition by lead. Arch. Bichem. Biophys. 253: 48-55.
51     Goldstein, G. W. (1993) Evidence that lead acts as a calcium  substitute in second messenger metabolism. Presented
52            at: Ninth international neurotoxicology conference; October 1991; Little Rock, AR. Neurotoxicology 14(2-
53            3): 97-101.
54     Golubovich, E. Ya.; Avkhimenko, M. M.; Chirkova, E. M. (1968) Biochemical and morphological changes in the
55            testicles of rats induced by small doses of lead. Toksikol. Nov. Prom. Khim. Veschestv. 10: 63-73.


       December 2005                                5-310        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Gonick, H. C.; Khalil-Manesh, F.; Raghavan, S. R. V. (1985) Characterization of human erythrocyte lead-binding
  2            protein. In: Lekkas, T. D., ed. International conference: heavy metals in the environment; September;
  3            Athens, Greece, v. 1. Edinburgh, United Kingdom: CEP Consultants, Ltd.; pp. 313-316.
  4     Gonick, H. C.; Ding, Y.; Bondy, S. C.; Ni, Z.; Vaziri, N. D. (1997) Lead-induced hypertension: interplay of nitric
  5            oxide and reactive oxygen species. Hypertension 30: 1487-1492.
  6     Gonick, H. C.; Ding, Y.; Vaziri, N. D. (1998) Effect of low lead exposure on eicosanoid excretion in rats.
  7            Prostaglandins Other Lipid Mediators 55: 77-82.
  8     Gonzalez-Cossio, T.; Peterson, K. E.; Sanin, L.-H.; Fishbein, E.; Palazuelos, E.; Aro, A.; Hernandez-Avila, M; Hu,
  9            H. (1997) Decrease in birth weight in relation to maternal bone-lead burden. Pediatrics 100: 856-862.
10     Gonzalez-Riola, I; Hernandez, E. R.; Escribano, A.; Revilla, M.; Villa, C.-S. L. F.; Rico, H. (1997) Effect of lead
11            on bone and cartilage  in sexually mature rats: a morphometric and histomorphometry study. Environ. Res.
12            74:91-93.
13     Corbel, F.; Boujelbene, M.; Makni-Ayadi, F.; Guermazi, F.; Croute, F.; Soleilhavoup, J. P.; El Feki, A. (2002)
14            Exploration des effets cytotoxiques du plomb sur la fonction sexuelle endocrine et exocrine chez le rat
15            pubere male et femelle. Mise en evidence d'une action apoptotique [Impact of lead given in drinking water
16            on the endocrine and exocrine sexual activity in pubescent rats. Determination of an apoptotic process]. C.
17            R. Biol. 325: 927-940.
18     Goyer, R. A. (1968) The renal tubule in lead poisoning. I. Mitochondrial swelling and aminoaciduria. Lab. Invest.
19            19:71-77.
20     Goyer, R. A.; Rhyne, B. C. (1973) Pathological effects of lead. Int. Rev. Exp. Pathol. 12: 1-77.
21     Goyer, R. A.; Wilson, M. H. (1975) Lead-induced inclusion bodies: results of ethylenediaminetetraacetic acid
22            treatment. Lab. Invest. 32: 149-156.
23     Goyer, R. A.; Krall, A.; Kimball,  J. P. (1968) The renal tubule in lead poisoning. II. In vitro studies of mitochondrial
24            structure and function. Lab. Invest. 19: 78-83.
25     Goyer, R. A.; Leonard, D. L.; Bream, P. R.; Irons, T. G. (1970a) Aminoaciduria in experimental lead poisoning.
26            Proc. Soc. Exp. Biol. Med. 135: 767-771.
27     Goyer, R. A.; Leonard, D. L.; Moore, J. F.; Rhyne, B. Krigman, M. R. (1970b) Lead dosage and the role of the
28            intranuclear inclusion body: an experimental study. Arch. Environ. Health 20: 705-711.
29     Goyer, R. A.; May, P.; Gates, M.  M.; Krigman, M. R. (1970c) Lead and protein content of isolated intranuclear
30            inclusion bodies from kidneys of lead-poisoned rats.  Lab. Invest. 22: 245-251.
31     Goyer, R. A.; Cherian, M. G.;  Delaquerriere-Richardson,  L. (1978) Renal effects of repeated administration of
32            calcium disodium ethylenediamnetetraacetate during excessive exposure to lead in rats. J. Environ. Pathol.
33            Toxicol. 1:403-410.
34     Goyer, R. A.; Epstein, S.; Bhattacharyya, M.; Korach, K.  S.; Pounds, J. (1994) Environmental risk factors for
35            osteoporosis. Environ. Health Perspect.  102: 390-394.
36     Grabowska, M.; Guminska, M. (1996) The effect of lead on lactate formation, ATP level and membrane ATPase
37            activities in human erythrocytes in vitro. Int. J. Occup. Med. Environ. Health 9: 265-274.
3 8     Graca, A.; Ramalho-Santos, J.; De Lourdes Pereira, M. (2004) Effect of lead chloride on spermatogenesis and sperm
39            parameters in mice. Asian J.  Androl. 6: 237-241.
40     Granick, J. L.; Sassa, S.; Granick, S.; Levere, R. D.; Kappas,  A. (1973) Studies in lead poisoning. II. correlation
41            between the ratio of activated to inactivated "delta"-aminolevulinic acid dehydratase of whole blood and the
42            blood lead level. Biochem. Med. 8: 149-159.
43     Grant, L. D.; Kimmel, C.  A.; West, G. L.; Martinez-Vargas, C. M.; Howard, J. L. (1980) Chronic low-level lead
44            toxicity in the rat. II. Effects  on postnatal physical and behavioral development. Toxicol. Appl. Pharmacol.
45            56:42-58.
46     Grobler, S. R.; Rossouw, R. J.; Kotze, D. (1985) Lead in teeth of weanling rats received via the maternal drinking
47            water. Arch. Oral Biol. 30: 509-511.
48     Grobler, S. R.; Rossouw, R. J.; Kotze, T. J. V.; Slander, I. A. (1991) The effect of airborne lead on lead levels of
49            blood, incisors and alveolar bone of rats. Arch. Oral Biol. 36: 357-360.
50     Grover, C. A.; Nation, J. R.; Brattom. G. R. (1993) Chronic exposure to lead attenuates cocaine-induced behavioral
51            activation. Pharmacol. Biochem. Behav. 44: 221-225.
52     Gruber, H. E.; Gonick, H. C.; Khalil-Manesh, F.; Sanchez, T. V; Motsinger, S.; Meyer, M.; Sharp, C. F.  (1997)
53            Osteopenia induced by long-term, low- and high-level exposure of the adult rat to lead. Miner. Electrolyte
54            Metab. 23:65-73.
55     Guilarte T. R.; McGlothan, J. L. (1998) Hippocampal NMDA receptor mRNA undergoes subunit specific changes
56            during developmental lead exposure. Brain Res. 790: 98-107.

       December  2005                                5-311         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Guilarte T. R.; McGlothan, J. L. (2003) Selective decrease in NR1 subunit splice variant mRNA in the hippocampus
  2            of Pb2+-exposed rats: implications for synaptic targeting and cell surface expression of NMDAR
  3            complexes. Mol. Brain Res. 113: 37-43.
  4     Guilarte, T. R.; Miceli, R. C. (1992) Age-dependent effects of lead on [3H]MK-801 binding to the NMD A receptor-
  5            gated ionophore: in vitro and in vivo studies. Neurosci. Lett. 148: 27-30.
  6     Guilarte, T. R.; McGlothan, J. L.; Nihei, M. K. (2000) Hippocampal expression of N-methyl-D-aspartate receptor
  7            (NMDAR1) subunit splice variant mRNA is altered by developmental exposure to Pb2+. Mol. Brain Res.
  8            76:299-305.
  9     Guity, P.; McCabe, M. J.; Pitts, D. K.; Santini, R. P.; Pounds, J. G. (2002) Protein kinase C does not mediate the
10            inhibitory action of lead on vitamin D3 -dependent production of osteocalcin in osteoblastic bone cells.
11            Toxicol. Appl. Pharmacol. 178: 109-116.
12     Guo, T. L.; Mudzinski, S. P.; Lawrence, D. A. (1996) The heavy metal lead modulates the expression of both TNF-
13            "alpha" and TNF-"alpha" receptors in lipopolysaccharide-activated human peripheral blood mononuclear
14            cells. J. Leukoc. Biol. 59: 932-939.
15     Gupta, K.; Upreti, R. K.; Kidwai, A. M. (1994) Toxicokinetic study of rat intestinal brush border membrane
16            enzymes following in vitro exposure to lead and vanadium. Bull. Environ. Contam. Toxicol. 52: 919-926.
17     Gupta, P.; Husain, M.  M.; Shankar, R.; Seth, P. K.; Maheshwari, R. K. (2002) Lead exposure enhances virus
18            multiplication and pathogenesis in mice. Vet. Hum. Toxicol. 44: 205-210.
19     Gurer, H.; Ercal, N. (2000) Can antioxidants be beneficial in the treatment of lead poisoning? Free Rad.  Biol. Med.
20            29: 927-945.
21     Gurer, H.; Ozgunes, H.; Neal,  R.; Spitz, D. R.; Ercal, N. (1998) Antioxidant effects of N-acetylcysteine and
22            succimer in red blood cells from lead-exposed rats. Toxicology 128: 181-189.
23     Gurer, H.; Neal, R.; Yang, P.;  Oztezcan, S.;  Ercal, N. (1999a) Captopril as an antioxidant in lead-exposed Fischer
24            344 rats. Hum. Exp. Toxicol. 18: 27-32.
25     Gurer, H.; Ozgunes, H.; Oztezcan, S.; Ercal, N. (1999b) Antioxidant role of a-lipoic acid in lead toxicity. Free
26            Radical Biol. Med. 27: 75-81.
27     Gurer, H.; Ozgunes, H.; Saygin, E.; Ercal, N. (2001) Antioxidant effect of taurine against lead-induced oxidative
28            stress. Arch. Environ. Contam. Toxicol. 41: 397-402.
29     Gutowski, M.; Altmann, L.; Sveinsson, K.; Wiegand, H. (1997) Postnatal development of synaptic plasticity in the
30            CAS hippocampal region of control and lead-exposed Wistar rats. Dev. Brain Res. 98: 82-90.
31     Gutowski, M.; Altmann, L.; Sveinsson, K.; Wiegand, H. (1998) Synaptic plasticity in the CA1 and CAS
32            hippocampal region of pre- and postnatally lead-exposed rats. Toxicol. Lett. 95: 195-203.
33     Habermann, E.; Crowell, K.; Janicki, P. (1983) Lead and other metals can substitute for Ca2+ in calmodulin. Arch.
34            Toxicol. 54: 61-70.
3 5     Hac, E.; Krechniak, J.  (1996) Lead levels in bone and hair of rats treated with lead acetate. Biol. Trace Elem. Res.
36            52:293-301.
37     Hacker, H.-J.; Bannasch, P.; Columbano, A. (1990) Effect of lead nitrate on liver carbohydrate enzymes and
38            glycogen content in the rat. Carcinogenesis  11: 2199-2204.
39     Hamilton, J. D.; O'Flaherty, E. J. (1994) Effects of lead exposure  on skeletal development in rats. Fundam. Appl.
40            Toxicol. 22: 594-604.
41     Hamilton, J. D.; O'Flaherty, E. J. (1995) Influence of lead on mineralization during bone growth. Fundam. Appl.
42            Toxicol. 26: 265-271.
43     Hamilton, J. D.; O'Flaherty, E. J.; Ross, R.; Shukla, R.; Gartside, P. S. (1994) Structural equation modeling and
44            nested ANOVA: effects of lead exposure on maternal and fetal growth in rats. Environ. Res. 64: 53-64.
45     Hammad, T. A.; Sexton, M.; Langenberg, P. (1996) Relationship between blood lead and dietary iron intake in
46            preschool children:  a cross-sectional study. Ann. Epidemiol. 6: 30-33.
47     Hammond, P. B.; Chernausek, S. D.; Succop, P. A.; Shukla, R.; Bornschein, R. L. (1989) Mechanisms by which
48            lead depresses linear and ponderal growth in weanling rats. Toxicol. Appl. Pharmacol. 99: 474-486.
49     Hammond, P. B.; Minnema, D. J.; Shulka, R. (1990) Lead exposure lowers the set point for food consumption and
50            growth in weanling rats. Toxicol. Appl. Pharmacol. 106:  80-87.
51     Hammond, P. B.; Minnema, D. J.; Succop, P. A. (1993) Reversibility of lead-induced depression of growth. Toxicol.
52            Appl. Pharmacol. 123: 9-15.
53     Han, S.; Qiao, X.; Simpson, S.; Ameri, P.; Kemp, F. W.; Bogden, J. D. (1996) Weight loss alters organ
54            concentrations and contents of lead  and some essential divalent metals in rats previously exposed to lead. J.
55            Nutr. 126:  317-323.
       December 2005                                5-312        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Han, S.; Qiao, X.; Kemp, F. W.; Bogden, J. D. (1997) Lead exposure at an early age substantially increases lead
  2            retention in the rat. Environ. Health Perspect. 105: 412-417.
  3     Han, S.; Li, W.; Jamil, U.; Dargan, K.; Orefice, M; Kemp, F. W.; Bogden, J. D. (1999) Effects of weight loss and
  4            exercise on the distribution of lead and essential trace elements in rats with prior lead exposure. Environ.
  5            Health Perspect. 107: 657-662.
  6     Han, S.; Pfizenmaier, D. H.; Garcia, E.; Eguez, M. L.; Ling, M.; Kemp, F. W.; Bogden, J. D. (2000) Effects of lead
  7            exposure before pregnancy and dietary calcium during pregnancy on fetal development and lead
  8            accumulation. Environ. Health Perspect. 108: 527-531.
  9     Hanna, L. A.; Peters, J. M.; Wiley, L. M.; Clegg, M. S.; Keen, C. L. (1997) Comparative effects of essential and
10            non-essential metals on preimplantation mouse embryo development in vitro. Toxicology  116: 123-131.
11     Hanson, E. H.; Imperatore, G.; Burke, W. (2001) HFE gene and hereditary hemochromatosis: a HuGE review. Am.
12            J. Epidemiol. 154: 193-206.
13     Hartman, D. E. (1995) Neuropsychological toxicology. 2nd ed. New York, NY: Plenum Press; pp. 95-125.
14     Hartwig, A.;  Schlepegrell, R.; Beyersmann, D. (1990) Indirect mechanism of lead-induced genotoxicity in cultured
15            mammalian cells. Mutat. Res. 241: 75-82.
16     Hashmi, N. S.; Kachru, D.  N.; Khandelwal, S.; Tandon, S. K. (1989) Interrelationship between iron deficiency and
17            lead intoxication (part 2). Biol. Trace Elem. Res. 22: 299-307.
18     Hayashi, M. (1983a) Lead toxicity in the pregnant rat. I. the effect of high - level lead on "delta"-aminolevulinic acid
19            dehydratase activity in maternal and fetal blood or tissues. Environ. Res. 30: 152-160.Hayes, R. B. (1997)
20            The carcinogenicity of metals in humans. Cancer Causes Control 8: 371-385.
21     Hayashi, M. (1983b) Lead toxicity in the pregnant rat. II. Effects of low-level lead on delta-aminolevulinic acid
22            dehydratase activity in maternal and fetal blood or tissue. Ind. Health 21: 127-135.
23     He, L.; Poblenz, A. T.; Medrano, C. J.; Fox, D. A. (2000) Lead and calcium produce rod photoreceptor cell
24            apoptosis by opening the mitrochondrial permeability transition pore. J. Biol. Chem. 275:  12175-12184.
25     He, L.; Perkins, G. A.; Poblenz, A. T.; Harris, J. B.; Hung, M.; Ellisman, M. H.; Fox, D. A.. (2003) Bcl-xL
26            overexpression blocks bax-mediated mitochondria! contact site formation and apoptosis in rod
27            photoreceptors of lead-exposed mice. Proc. Natl. Acad.  Sci. U. S. A. 100: 1022-1027.
28     Heiman, A. S.; Tonner, L. E. (1995) The acute effect of lead acetate on glucocorticoid regulation of tyrosine
29            aminotransferase inhepatoma cells. Toxicology 100: 57-68.
30     Helleday, T.; Nilsson, R.; Jenssen, D. (2000) Arsenic [III] and heavy metal ions induce intrachromosomal
31            homologous recombination in the hprt gene of V79 Chinese hamster cells. Environ. Mol. Mutagen. 35: 114-
32            122.
33     Hellstrom-Lindahl, E.; Oskarsson, A. (1990) Cellular response after mobilization of metals by
34            diethyldithiocarbamate in rat hepatocyte cultures. Toxicology 65: 23-32.
35     Hemphill, F.  E.;  Kaeberle, M. L.; Buck,  W. B. (1971) Lead suppression of mouse resistance to Salmonella
36            typhimurium. Science (Washington, DC) 172: 1031-1032.
37     Hengstler, J.  G.; Bolm-Audorff, U.; Faldum, A.; Janssen, K.; Reifenrath, M.; Gotte, W.; Jung, D.; Mayer-Popken,
38            O.; Fuchs, J.; Gebhard, S.; Bienfait, H. G.; Schlink, K.; Dietrich, C.; Faust, D.; Epe, B.; Oesch, F. (2003)
3 9            Occupational exposure to heavy metals: DNA damage induction and DNA repair inhibition prove co-
40            exposures to cadmium, cobalt and lead as more dangerous than hitherto expected. Carcinogenesis 24: 63-73.
41     Henning, S. J.; Cooper, L. C. (1988) Intestinal accumulation of lead salts and milk lead by suckling rats (42645).
42            Proc. Soc. Exp. Biol. Med. 187: 110-116.
43     Heo, Y.; Parsons, P. J.; Lawrence, D. A. (1996) Lead differentially  modifies cytokine production in vitro and in
44            vivo. Toxicol. Appl. Pharmacol. 138: 149-157.
45     Heo, Y.; Lee, W. T.; Lawrence, D. A. (1997) In vivo the environmental pollutants lead and mercury induce
46            oligoclonal T cell responses skewed toward type-2 reactivities. Cell. Immunol. 179: 185-195.
47     Heo, Y.; Lee, W. T.; Lawrence, D. A. (1998) Differential effects of lead and cAMP on development and activities of
48            Thl- and Th2-lymphocytes. Toxicol. Sci. 43: 172-185.
49     Heo, Y.; Lee, B.-K.; Ann, K.-D.; Lawrence, D. A. (2004) Serum IgE elevation correlates with blood lead levels in
50            battery manufacturing workers. Hum. Exp. Toxicol. 23: 209-213.
51     Hermes-Lima, M.; Pereira, B.; Bechara, E. J. H. (1991) Are free radicals involved in lead poisoning? Xenobiotica
52            21: 1085-1090.
53     Hernandez-Avila, M.; Gonzalez-Cossio, T.; Hernandez-Avila, J. E.; Romieu, I.; Peterson, K. E.; Aro, A.;
54            Palazuelos, E.; Hu, H. (2003) Dietary calcium supplements to lower blood lead levels in lactating women: a
55            randomized placebo-controlled trial. Epidemiology 14: 206-212.


       December 2005                                5-313         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Hicks, D. G.; O'Keefe, R. I; Reynolds, K. I; Cory-Slechta, D. A.; Puzas, J. E.; Judkins, A.; Rosier, R. N. (1996)
  2            Effects of lead on growth plate chondrocyte phenotype. Toxicol. Appl. Pharmacol. 140: 164-172.
  3     Hilbertz, U.; Kramer, U.; De Ruiter, N.; Baginski, B. (1986) Effects of cadmium and lead on oxidative metabolism
  4            and phagocytosis by mouse peritoneal macrophages. Toxicology 39: 47-57.
  5     Hilderbrand, D. C.; Der, R.; Griffin, W. T.; Fahim, M. S. (1973) Effect of lead acetate on reproduction. Am. J.
  6            Obstet. Gynecol. 115: 1058-1065.
  7     Hill, A. B. (1965) The environment and disease: association or causation? Proc. R.  Soc. Med. 58: 295-300.
  8     Hilson, J. A.; Strupp, B. J. (1997) Analyses of response patterns clarify lead effects in olfactory reversal and
  9            extradimensional shift tasks: assessment of inhibitory control, associative ability, and memory. Behav.
10            Neurosci. 111:532-542.
11     Hinton, D. E.; Lipsky, M. M.; Heatfield, B. M.; Trump, B. F. (1979) Opposite effects of lead on chemical
12            carcinogenesis in kidney and liver of rats. Bull. Environ. Contam. Toxicol. 23: 464-469.
13     Hogan, K.; Marcus, A.; Smith, R.; White, P. (1998) Integrated exposure uptake biokinetic model for lead in
14            children: empirical comparisons with epidemiologic data. Environ. Health Perspect. 106(suppl. 6): 1557-
15            1567.
16     Holgate, S.; Casale, T.; Webzek, S.; Bousquet, J.; Deniz, Y.; Reisner, C. (2005) The anti-inflammatory effects of
17            omalizumab confirm the central role of IgE in allergic inflammation. J. Allergy Clin. Immunol. 115: 459-
18            465.
19     Holian, A.; Uthman, M.  O.; Goltsova, T.; Brown, S. D.; Hamilton, R. F. J. (1997) Asbestos and silica-induced
20            changes in human alveolar macrophage  phenotype. Environ. Health Perspect. 105: 1139-1142.
21     Holladay, S. D. (1999) Prenatal immunotoxicant exposure and postnatal autoimmune disease. Environ. Health
22            Perspect.  107: 687-691.
23     Holladay, S. D. (2005) Developmental immunotoxicology. Boca Raton, FL: CRC Press,  Inc.
24     Holsapple, M. P.; West,  L. J.;  Landreth, K. S.  (2003) Species comparison of anatomocial and functional immune
25            system development. Birth Defects Res. B. 68: 321-334.
26     Honchel, R.; Marsano, L.; Cohen, D.; Shedlofsky, S.; McClain, C. J. (1991) Lead enhances lipopolysaccharide and
27            tumor necrosis factor  liver injury. J. Lab. Clin. Med. 117: 202-208.
28     Hotter, G.; Pels, L. M.; Closa, D.; Rosello, J.;  Stolte, H.; Gelpi, E. (1995) Altered levels of urinary prostanoids in
29            lead-exposed workers. Toxicol. Lett. 77: 309-312.
30     Hu, H. (1991) A 50-year follow-up of childhood plumbism: hypertension, renal function, and hemoglobin levels
31            among survivors. Am. J. Dis. Child. 145: 681-687.
32     Hu, H.; Wu, M.-T.;  Cheng, Y.; Sparrow, D.; Weiss, S.; Kelsey, K. (2001) The "delta"-aminolevulinic acid
3 3            dehydratase (AL AD) polymorphism and bone and blood lead levels in community-exposed men: the
34            Normative Aging Study. Environ. Health Perspect. 109: 827-832.
3 5     Huang, F., Schneider,  J.  S. (2004) Effects of lead exposure on proliferation and differentiation of neural stem cells
36            derived from different regions of embryonic rat brain. Neurotoxicology.25: 1001-1012.
37     Hubermont, G.; Buchet, J.-P.; Roels, H.; Lauwerys, R. (1976) Effect of short-term administration of lead to pregnant
38            rats. Toxicology 5: 379-384.
39     Hudson, C. A.; Cao, L.; Kasten-Jolly, J.; Kirkwood, J. N.; Lawrence, D. A. (2003) Susceptibility of lupus-prone
40            NZM mouse strains to lead exacerbation of systemic lupus  erythematosus symptoms. J. Toxicol. Environ.
41            Health A 66: 895-918.
42     Hussain, I.; Piepenbrink, M. S.; Dietert, R. R.  (2005) Impact of in ovo-administered lead and testosterone on
43            developing female thymocytes. J. Toxicol. Environ. Health A. 68:  1309-1319.
44     Hwang, K.-Y.; Lee, B.-K.; Bressler, J. P.; Bolla, K. I.; Stewart, W. F.; Schwartz, B. S. (2002) Proteinkinase C
45            activity and the relations between blood lead and neurobehavioral function in lead workers. Environ. Health
46            Perspect.  110: 133-138.
47     Hwua, Y. S.; Yang J. L.  (1998) Effect of 3-aminotriazole on anchorage independence and mutagenicity in cadmium-
48            and lead-treated diploid human fibroblasts. Carcinogenesis 19: 881-888.
49     lavicoli, I.; Carelli, G.; Stanek, E. J., Ill; Castellino, N.; Calabrese, E. J. (2003) Effects of low doses of dietary lead
50            on red blood cell production in male and female mice. Toxicol. Lett. 137:  193-199.
51     Ichiba, M.; Tomokuni, K. (1987) Urinary excretion of 5-hydroxyindoleacetic acid, "delta"-aminolevulinic acid and
52            coproporphyrin  isomers in rats and men exposed to lead. Toxicol. Lett. 38: 91-96.
53     Ichiba, M.; Tomokuni, K.; Sugimoto, K. (1987)  Erythrocyte pyrimidine 5'-nucleotidase test for occupational lead
54            exposure. Ind. Health 25: 195-203.
55     Ikebuchi, H.; Teshima, R.; Suzuki, K.; Terao,  T.; Yamane, Y. (1986) Simultaneous induction of Pb-metallothionein-
56            like protein and Zn-thionein in the liver of rats given lead acetate. Biochem. J.  233: 541-546.

       December 2005                                5-314         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Inskip, M. I; Franklin, C. A.; Baccanale, C. L.; Manton, W. I.; O'Flaherty, E. I; Edwards, C. M. H.; Blenkinsop, J.
  2            B.; Edwards, E. B. (1996) Measurement of the flux of lead from bone to blood in a nonhuman primate
  3            (Macaca fascicularis) by sequential administration of stable lead isotopes. Fundam. Appl. Toxicol. 33: 235-
  4            245.
  5     Ishihara, K.; Alkondon, M.; Monies, J. G.; Albuquerque, E. X. (1995) Nicotinic responses in acutely dissociated rat
  6            hippocampal neurons and the selective blockade of fast-desensitizing nicotinic currents by lead. J.
  7            Pharmacol. Exp. Ther. 273: 1471-1482.
  8     Isolauri, E.; Huurre, A.; Salminen, S.; Impivaara, O. (2004) The allergy epidemic extends beyond the past few
  9            decades. Clin. Exp. Allergy. 34: 1007-1010.
10     Ito, Y.; Niiya, Y.; Kurita, H.; Shima, S.; Sarai, S. (1985) Serum lipid peroxide level and blood superoxide dismutase
11            activity in workers with occupational exposure to lead. Int. Arch. Occup. Environ. Health 56: 119-127.
12     Jacobs, J. M.; Sinclair, P. R.; Sinclair, J. F.; Gorman, N.; Walton, H. S.; Wood, S.  G.; Nichols, C. (1998) Formation
13            of zinc protoporphyrin in cultured hepatocytes: effects of ferrochelatase inhibition, iron chelation or lead.
14            Toxicology 125: 95-105.
15     Jacquet,  P. (1976) Effets du plomb administre durant la gestation a des souris C57B1 [Effects of  lead administered
16            during the gestation period of mice C57B1]. C. R. Seances Soc. Biol. SesFil. 170:  1319-1322.
17     Jacquet,  P. (1977) Early embryonic development in lead-intoxicated mice. Arch. Pathol. Lab. Med. 101: 641-643.
18     Jacquet,  P.; Leonard, A.; Gerber, G. B. (1975) Embryonic death in mouse due to lead exposure. Experientia 31: 24-
19            25.
20     Jacquet,  P.; Leonard, A.; Gerber, G. B. (1976) Action of lead on early divisions of the mouse embryo. Toxicology 6:
21            129-132.
22     Jacquet,  P.; Gerber, G. B.; Maes, J. (1977) Biochemical studies in embryos after exposure of pregnant mice to
23            dietary lead. Bull. Environ. Contam.  Toxicol.  18: 271-277.
24     Jagetia, G. C.; Aruna, R. (1998) Effect of various concentrations of lead nitrate on the induction of micronuclei in
25            mouse bone marrow. Mutat. Res. 415: 131-137.
26     Jehan, Z. S.; Motlag, D. B. (1995) Metal induced changes in the erythrocyte membrane of rats. Toxicol. Lett. 78:
27            127-133.
28     Jemal, A.; Graubard, B. I.; Devesa, S. S.; Flegal, K. M. (2002) The association of blood lead level and cancer
29            mortality among whites in the United States. Environ. Health Perspect.  110: 325-329.
30     Jett, D. A.; Beckles, R. A.; Navoa, R. V.; McLemore,  G. L. (2002) Increased high-affinity nicotinic receptor-binding
31            in rats exposed to lead during development. Neurotoxicol. Teratol. 24: 805-811.
32     Jian, Z.;  Ying-han, X.; Hong-fu, C. (1985) The effects of lead ion on immune function of rabbit alveolar
33            macrophages: Quantitation of immune phagocytosis and rosette formation by 51-Crin vitro. Toxicol. Appl.
34            Pharmacol. 78: 484-487.
35     Johansson, L. (1989) Premature acrosome reaction in  spermatozoa from lead-exposed mice. Toxicology 54: 151-
36            162.
37     Johansson, L.; Pellicciari, C. E. (1988) Lead-induced changes in the stabilization of the mouse sperm chromatin.
38            Toxicology 51: 11-24.
39     Johansson, L.; Wide, M. (1986) Long-term exposure of the male mouse to lead: effects on fertility. Environ. Res.
40            41:481-487.
41     Johansson, L.; Sjoblom, P.;  Wide, M.  (1987) Effects of lead on the male mouse as investigated by in vitro
42            fertilization and blastocyst culture. Environ. Res. 42:  140-148.
43     Jones, M. M.; Basinger, M.  A.; Gale, G. R.; Atkins, L. M.; Smith,  A. B.; Stone, A. (1994) Effect  of chelate
44            treatments on kidney, bone and brain levels of lead-intoxicated mice. Toxicology 89: 91-100.
45     Joseph, C. L. M.; Havstad, S.; Ownby, D. R.; Peterson, E. L.;  Maliark, M.; McCabe, J., M.J.; Barone, C.; Johnson,
46            C. C. (2005) Blood lead levels and risk of asthma. Environ. Health Perspect. 113: 900-904.
47     Jover, R.; Lindberg, R. L. P.; Meyer, U.  A. (1996) Role of heme in cytochrome P450 transcription and function in
48            mice treated with lead acetate. Mol. Pharmacol. 50: 474-481.
49     Junaid, M.;  Chowdhuri, D. K.; Narayan, R.; Shanker,  R.;  Saxena, D. K. (1997) Lead-induced changes in ovarian
50            follicular development and maturation in mice. J. Toxicol. Environ. Health 50: 31-40.
51     Kaji, T.; Yamamoto,  C.; Sakamoto, M. (1991) Effect of lead on the glycosaminoglycans metabolism of bovine
52            aortic endothelial cells in culture. Toxicology 68: 249-257.
53     Kaji, T.; Yamamoto, C.; Sakamoto, M.; Kozuka, H. (1992) Inhibitory effect of lead on the release of tissue
54            plasminogen activator from human vascular endothelial cells in culture. Toxicology 73: 219-227.
       December 2005                                5-315         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Kaji, T.; Suzuki, M; Yamamoto, C.; Mishima, A.; Sakamoto, M; Kozuka, H. (1995a) Severe damage of cultured
  2            vascular endothelial cell monolayer after simultaneous exposure to cadmium and lead. Arch. Environ.
  3            Contam. Toxicol. 28:  168-172.
  4     Kaji, T.; Fujiwara, Y.; Hoshino, M.; Yamamoto, C.; Sakamoto, M.; Kozuka, H. (1995b) Inhibitory effect of lead on
  5            the proliferation of cultured vascular endothelial cells. Toxicology 95:  87-92.
  6     Kaji, T.; Ohkawara, S.; Nakajima, M.; Yamamoto, C.; Fujiwara, Y.; Miyajima, S.; Koizumi, F. (1997) Lead-induced
  7            alteration of heparan sulfate proteoglycans in cultured vascular endothelial cells. Toxicology 118: 1-10.
  8     Kamel, F.; Umbach, D. M.; Lehman, T. A.; Park, L. P.; Munsat, T. L.; Shefner, J. M.; Sandier, D. P.; Hu, H.;
  9            Taylor, J. A. (2003) Amyotrophic lateral sclerosis, lead, and genetic susceptibility: polymorphisms in the
10            "delta"-aminolevulinic acid dehydratase and vitamin D receptor genes. Environ. Health Perspect. Ill: 1335-
11            1339.
12     Kanduc, D.; Frisco, M. (1992) Hepatic DNA methylation in young, middle-aged, and senescent rats: the effect of
13            mitogen-induced cell proliferation. Biochem. Med. Metab. Biol. 48: 286-291.
14     Kanduc, D.; Rossiello, M. R.; Aresta, A.; Cavazza, C.; Quagliariello, E.; Farber, E. (1991) Transitory DNA
15            hypomethylation during liver cell proliferation induced by a single dose of lead nitrate. Arch. Biochem.
16            Biophys. 286: 212-216.
17     Kanitz, M. H.; Witzmann, F. A.; Zhu, H.; Fultz, C. D.; Skaggs, S.; Moorman, W. J.;  Savage, R. E., Jr. (1999)
18            Alterations  in rabbit kidney protein expression following lead exposure as analyzed by two-dimensional gel
19            electrophoresis. Electrophoresis 20: 2977-2985.
20     Karmakar, N.; Anand, S. (1989) Study of the inhibitory effect of lead acetate on duodenal contractility in rat. Clin.
21            Exp. Pharmacol. Physiol. 16: 745-750.
22     Karmakar, N.; Saxena, R.;  Anand, S. (1986) Histopathological changes induced in rat tissues by oral intake of lead
23            acetate. Environ. Res. 41: 23-28.
24     Karmaus, W.; Brooks, K. R.; Nebe, T.; Witten, J.; Obi-Osius, N.; Kruse, H. (2005) Immune functionbiomarkers in
25            children exposed to lead and organochlorine compounds: a cross-sectional study. Environ. Health Glob.
26            Access Sci. 4: 1-10.
27     Kato, Y.; Takimoto, S.; Ogura, H. (1977) Mechanism of induction of hypercalcemia and hyperphosphatemia by lead
28            acetate in the rat. Calcif. Tissue Res. 24: 41-46.
29     Kauppinen, T.; Riala, R.; Seitsamo, J.; Hernberg, S. (1992) Primary liver cancer and occupational exposure. Scand.
30            J. Work Environ. Health. 18: 18-25.
31     Kelada, S. N.; Shelton, E.;  Kaufmann, R. B.; Khoury, M. J. (2001) "Delta"-aminolevulinic acid dehydratase
32            genotype and  lead toxicity: a HuGE review. Am. J. Epidemiol. 154:  1-13.
33     Keller, C. A.; Doherty, R. A. (1980) Bone lead mobilization in lactating mice and lead transfer to suckling offspring.
34            Toxicol. Appl. Pharmacol. 55: 220-228.
35     Kempinas, W. G.; Lamano-Carvalho, T. L.; Petenusci, S. O.; Lopes, R. A.; Azoubel, R. (1988) Morphometric and
36            stereological analysis of rat testis and epididymis in an early phase of saturnism. Exp. Biol. 8: 51-56.
37     Kempinas, W. G.; Melo, V. R.; Oliveira-Filho, R. M.; Santos, A. C.; Favaretto, A. L.; Lamano-Carvalho, T. L.
38            (1990) Saturnism in the male rat: endocrine effects. Braz. J. Med. Biol. Res. 23: 1171-1175.
39     Kempinas, W. G.; Favaretto, A. L. V.; Melo, V. R.; Lamano Carvalho, T. L.; Petenusci, S. O.; Oliveira-Filho, R. M.
40            (1994) Time-dependent effects of lead on rat reproductive functions. J. Appl. Toxicol.  14: 427-433.
41     Kennedy, G. L.; Arnold, D. W.; Calandra, J. C. (1975) Teratogenic evaluation of lead compounds in mice and rats.
42            Food Cosmet. Toxicol. 13: 629-632.
43     Keogh, J. P.; Steffen, B.; Siegers, C.-P. (1994) Cytotoxicity of heavy metals in the human small intestinal epithelial
44            cell line 1-407: the role of glutathione. J. Toxicol. Environ. Health 43: 351-359.
45     Kern, M.; Audesirk, G. (2000) Stimulatory and inhibitory effects of inorganic lead on calcineurin. Toxicology 150:
46            171-178.
47     Kerper, L. E.; Hinkle,  P. M. (1997) Cellular uptake of lead is activated by depletion of intracellular calcium stores. J.
48            Biol. Chem. 272: 8346-8352.
49     Khalil-Manesh, F.; Gonick, H. C. Cohen, A. H.; Alinovi, R.; Bergamaschi, E.; Mutti, A.; Rosen, V. J. (1992a)
50            Experimental  model of lead nephropathy. I. Continuous high-dose lead administration. Kidney Int. 41:
51            1192-1203.
52     Khalil-Manesh, F.; Gonick, H. C.; Cohen, A.; Bergamaschi, E.; Mutti, A. (1992b) Experimental model of lead
53            nephropathy. II. Effect of removal from lead exposure and chelation treatment with dimercaptosuccinic acid
54            (DMSA). Environ. Res 58: 35-54.
55     Khalil-Manesh, F.; Gonick, H. C.; Cohen, A. H. (1993a) Experimental model of lead nephropathy. III. Continuous
56            low-level lead administration. Arch. Environ. Health 48: 271-278.

       December 2005                                5-316         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Khalil-Manesh, F.; Gonick, H. C; Weiler, E. W. J.; Prins, B.; Weber, M. A.; Purdy, R. E. (1993b) Lead-induced
  2            hypertension: possible role of endothelial factors. Am. J. Hypertens. 6: 723-729.
  3     Khalil-Manesh, F.; Gonick, H. C.; Weiler, E. W. J.; Prins, B.; Weber, M. A.; Purdy, R.; Ren, Q. (1994) Effect of
  4            chelation treatment with dimercaptosuccinic acid (DMSA) on lead-related blood pressure changes. Environ.
  5            Res. 65: 86-99.
  6     Khan, M. Z.; Szarek, J. (1994) Effects of concurrent oral administration of lead, selenium or monensin on hepatic
  7            porphyrin levels in broiler chickens during sub-acute toxicosis. J. Vet. Med. B. 41: 77-82.
  8     Kim, D.; Lawrence, D.  A. (2000) Immunotoxic effects of inorganic lead on host resistance of mice with different
  9            circling behavior preferences. Brain Behav. Immun. 14: 305-317.
10     Kim, J. S.; Hamilton, D. L.; Blakley, B. R.; Rousseaux, C.  G. (1992) The effects of thiamin on lead metabolism:
11            organ distribution of lead 203. Can. J. Vet. Res. 56: 256-259.
12     Kim, R.; Rotnitsky, A.; Sparrow, D.; Weiss, S. T.; Wager,  C.; Hu, H. (1996) A longitudinal study of low-level lead
13            exposure and impairment of renal function. The Normative Aging Study. JAMA J. Am. Med. Assoc. 275:
14            1177-1181.
15     Kim, H.-S.; Lee, S.-S.; Lee, G.-S.; Hwangbo, Y.; Ahn, K.-D.; Lee, B.-K. (2004) The protective effect of "delta" -
16            aminolevulinic acid dehydratase 1-2 and 2-2 isozymes against blood lead with higher hematologic
17            parameters. Environ. Health Perspect. 112: 538-541.
18     Kimber, I.; Stonard, M. D.; Gidlow, D. A.; Niewola, Z. (1986) Influence of chronic low-level exposure to lead on
19            plasma immunoglobulin concentration and cellular immune function in man. Int. Arch. Occup. Environ.
20            Health 57: 117-125.
21     Kiremidjian-Schumacher, L.; Stotzky, G.; Dickstein, R. A.; Schwartz, J. (1981) Influence of cadmium, lead, and
22            zinc on the ability of guinea pig macrophages to interact with macrophage migration inhibitory factor.
23            Environ. Res. 24: 106-116.
24     Kishikawa, H.; Lawrence, D. A. (1998) Differential production of interleukin-6 in the brain and spleen of mice
25            treated with lipopolysaccharide in the presence and absence of lead. J. Toxicol. Environ. Health A 53: 357-
26            373.
27     Kishikawa, H.; Song, R.; Lawrence, D. A. (1997) Interleukin-12 promotes enhanced resistance to Listeria
28            monocytogenes infection of lead-exposed mice. Toxicol. Appl. Pharmacol. 147: 180-189.
29     Kishimoto, T.; Oguri, T.; Ueda, D.; Tada, M. (1995) Effect of lead on tube formation by cultured human vascular
3 0            endothelial cells. Arch. Toxicol. 69:718-721.
31     Klann, E.; Shelton, K. R. (1989) The effect of lead on the metabolism of a nuclear matrix protein which becomes
32            prominent in lead-induced intranuclear inclusion bodies. J. Biol. Chem. 264: 16,969-16,972.
33     Klein, R. F.; Wiren, K. M.  (1993) Regulation of osteoblastic gene expression by lead. Endocrinology 132: 2531-
34            2537.
3 5     Knowles, S. O.; Donaldson, W. E. (1990) Dietary modification of lead toxicity: Effect on fatty acid and eicosanoid
36            metabolism in chicks. Comp. Biochem. Physiol. C. 95: 99-104.
37     Knowles, S. O.; Donaldson, W. E. (1997) Lead disrupts eicosanoid metabolism, macrophage function, and disease
38            resistance in birds. Biol. Trace Element Res. 60: 13-26.
39     Kobayashi, N.; Okamoto, T. (1974) Effects of lead oxide on the induction of lung tumors in Syrian hamsters. J.  Natl.
40            Cancer Inst. 52: 1605-1610.
41     Kohila, T.; Tahti, H. (2004) Effects of aluminium and lead on ATPase activity of knockout +/- mouse cerebral
42            synaptosomes in vitro. Altern. Lab.  Anim 32: 361-367.
43     Kojima, M.; Nemoto, K.; Murai, U.; Yoshimura, N.; Ayabe, Y.; Degawa, M. (2002) Altered gene expression of
44            hepatic lanosterol 14x-demethylase (CYP51) in lead nitrate-treated rats. Arch. Toxicol. 76: 398-403.
45     Kojima, M.; Masui, T.; Nemoto, K.; Degawa, M. (2004) Lead nitrate-induced development of hypercholesterolemia
46            in rats: sterol-independent gene regulation of hepatic enzymes responsible for cholesterol homeostasis.
47            Toxicol. Lett. 154: 35-44.
48     Koller, L. D.  (1973) Immunosuppression produced by lead, cadmium, and mercury. Am. J. Vet. Res. 34:  1457-1458.
49     Koller, L. D.; Kovacic, S. (1974) Decreased antibody formation in mice exposed to lead. Nature 250: 148-150.
50     Konantakieti, C.; Beuthin, F. C.; Louis-Ferdinand, R. T. (1986) Erythrocyte pyrimidine 5'-nucleotidase inhibition by
51            acute lead exposure in neonatal rats. J. Biochem. Toxicol. 1: 51-59.
52     Koo, P.; Nagai, M. K.; Farber, E. (1994) Multiple sites of control of glutathione S-transferase Pl-1 in rat liver. J.
53            Biol. Chem. 269:  14601-14606.
54     Korashy, H. M.; El-Kadi, A. O. S. (2004) Differential effects of mercury, lead and copper on the constitutive and
55            inducible expression of aryl hydrocarbon receptor  (AHR)-regulated genes in cultured hepatoma Hepa Iclc7
56            cells. Toxicology 201: 153-172.

       December 2005                                5-317        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Korrick, S. A.; Hunter, D. I; Rotnitzky, A.; Hu, H.; Speizer, F. E. (1999) Lead and hypertension in a sample of
  2            middle-aged women. Am. J. Public Health 89: 330-335.
  3     Kostial, K.; Blanusa, M; Piasek, M; Restek-Samarzija, N.; Jones, M. M; Singh, P. K. (1999) Combined chelation
  4            therapy in reducing tissue lead concentrations in suckling rats. J. Appl. Toxicol.  19: 143-147.
  5     Kowolenko, M.; Tracy, L.; Mudzinski, S.; Lawrence, D. A. (1988) Effect of lead on macrophage function. J.
  6            Leukocyte Biol. 43: 357-364.
  7     Kowolenko, M.; Tracy, L.; Lawrence, D. A. (1989) Lead-induced alterations of in vitro bone marrow cell responses
  8            to colony stimulating factor-1. J. Leukocyte Biol. 45: 198-206.
  9     Kowolenko, M.; Tracy, L.; Lawrence, D. (1991) Early effects of lead on bone marrow cell responsiveness in mice
10            challenged with Listeria monocytogenes. Fundam. Appl. Toxicol. 17: 75-82.
11     Kramer, H. J.; Gonick, H. C.; Lu, E. (1986) In vitro inhibition of Na-K-ATPase by trace metals: relation to renal and
12            cardiovascular damage. Nephron 44: 329-336.
13     Kristensen, P.; Andersen, A. (1992) A cohort study on cancer incidence in offspring of male printing workers.
14            Epidemiology 3: 6-10.
15     Kristensen, P.; Eilertsen, E.; Einarsdottir, E.; Ovrebo, S.; Haugen, A. (1993) Effect modification by inorganic lead in
16            the dominant lethal assay. Mutat. Res. 302: 33-38.
17     Krocova, Z.; Macela, A.; Kroca, M.; Hernychova, L. (2000) The immunomodulatory effect(s) of lead and cadmium
18            on the cells of immune system in vitro. Toxicol. In Vitro. 14: 33-40.
19     Kubo, Y.; Yasunaga, M.; Masuhara, M.; Terai, S.; Nakamura, T.; Okita, K.  (1996) Hepatocyte proliferation induced
20            in rats by lead nitrate is suppressed by several tumor necrosis factor "alpha" inhibitors. Hepatology 23: 104-
21            114.
22     Kumar, K. V.; Das, U. N. (1993) Are free radicals involved in the pathobiology of human essential hypertension?
23            Free Radical Res. Commun. 19: 59-66.
24     Lai, C.-C.; Lin, H. H.; Chen, C. W.; Chen, S.-H.; Chiu, T. H. (2002) Excitatory action of lead on rat sympathetic
25            preganglionic neurons in vitro and in vivo. Life Sci. 71: 1035-1045.
26     Lake, L.; Gerschenson, L. E. (1978) Cellular and molecular toxicology of lead. III. Effect of lead on heme synthesis.
27            J. Toxicol. Environ. Health 4: 527-540.
28     Lai, B.; Goldstein, G.; Bressler, J.  P. (1996) Role of anion exchange and thiol groups in the regulation of potassium
29            efflux by lead in human erythrocytes. J. Cell Physiol. 167: 222-228.
30     Langrish, C. L.; Buddie, J. C.; Thrasher, A. J.; Goldblatt, D. (2002) Neonatal dendritic cells are intrinsically biased
31            against Th-1 immune responses.  Clin. Exp. Immunol. 128: 118-123.
32     Lanphear, B. P.; Bearer, C. F. (2005) Biomarkers in paediatric research and practice. Arch. Dis. Child. 90: 594-600.
33     Lanphear, B. P.; Dietrich, K.; Auinger, P.; Cox, C. (2000) Cognitive deficits associated with blood lead
34            concentrations < 10 ug/dL in U.S. children and adolescents. Public Health Rep.  115: 521-529.
35     Lanphear, B. P.; Hornung, R.; Khoury, J.; Yolton, K.; Baghurst, P.; Bellinger, D. C.; Canfield, R. L.; Dietrich, K.
36            N.; Bornschein, R.; Greene, T.; Rothenberg, S. J.; Needleman, H. L.; Schnaas, L.; Wasserman, G.;
37            Graziano, J.; Roberts, R. (2005) Low-level environmental lead exposure and children's intellectual function:
38            an international pooled analysis.  Environ. Health Perspect. 113:  894-899.
39     Lara-Tejero, M.; Pamer, E. G. (2004) T cell responses to Listeria monocytogenes. Curr. Opin. Microbiol. 7: 45-50.
40     Larsson, A. (1974) Studies on dentinogenesis in the rat. The interaction between lead-pyrophosphate solutions and
41            dentinal globules. Calcif. Tiss. Res. 16: 93-107.
42     Larsson, A.; Helander, H. F. (1974) Studies on dentinogenesis in the rat. Light, electron microscopic and
43            histochemical studies on the interaction between lead pyrophosphate solutions and dentin-producing tissues.
44            Calcif. Tiss. Res. 14: 87-104.
45     Laschi-Loquerie, A.; Decotes, J.; Tachon, P.; Evreux, J. C. (1984) Influence of lead acetate on hypersensitivity
46            experimental study. J. Immunopharmacol. 6: 87-93.
47     Lasley, S. M.; Gilbert, M. E. (1996) Presynaptic glutamatergic function in dentate gyrus in vivo is diminished by
48            chronic exposure to inorganic lead. Brain Res. 736: 125-134.
49     Lasley, S. M.; Gilbert, M. E. (1999) Lead inhibits the rat N-methyl-d-aspartate receptor channel by binding to a site
50            distinct from the zinc allosteric site. Toxicol. Appl. Pharmacol.  159: 224-233.
51     Lasley, S. M.; Gilbert, M. E. (2000) Glutamatergic components underlying lead-induced impairments in
52            hippocampal synaptic plasticity. Neurotoxicology 21: 1057-1067.
53     Lasley, S. M.; Gilbert, M. E. (2002) Rat hippocampal glutamate and GABA release exhibit biphasic effects as a
54            function of chronic lead exposure level. Toxicol. Sci. 66: 139-147.
55     Lasley, S. M.; Green, M. C.; Gilbert, M. E. (1999) Influence of exposure period on in vivo hippocampal glutamate
56            and GABA release in rats  chronically exposed to lead. Neurotoxicology 20: 619-629.

       December 2005                                 5-318         DRAFT-DO NOT QUOTE  OR CITE

-------
  1     Lasley, S. M; Green, M. C.; Gilbert, M. E. (2001) Rat hippocampal NMD A receptor binding as a function of
  2            chronic lead exposure level. Neurotoxicol. Teratol. 23: 185-189.
  3     Laughlin, N. K.; Bowman, R. E.; Franks, P. A.; Dierschke, D. J. (1987) Altered menstural cycles in rhesus monkeys
  4            induced by lead. Fundam. Appl. Toxicol. 9: 722-729.
  5     Lauwerys, R.; Bernard, A.; Cardenas, A.  (1992) Monitoring of early nephrotoxic effects of industrial chemicals.
  6            Toxicol. Lett. 64-65: 33-42.
  7     Lawler, E. M.; Duke, G. E.; Redig, P. T. (1991) Effect of sublethal lead exposure on gastric motility of red-tailed
  8            hawks. Arch. Environ. Contam. Toxicol. 21: 78-83.
  9     Lawrence, D. A. (1981) In vivo and in vitro effects of lead on humoral and cell- mediated immunity. Infect. Immun.
10            31:  136-143.
11     Lawrence, D. A.; Kim, D. (2000) Central/peripheral nervous system and immune responses. Toxicology 142: 189-
12            201.
13     Lawrence, D. A.; McCabe, M. J., Jr. (2002) Immunomodulationby metals. Int. Immunopharmacol. 2: 293-302.
14     Ledda-Columbano, G. M.; Columbano, A.; Cannas, A.;  Simbula, G.; Okita, K.; Kayano, K.; Kubo, Y.; Katyal, S. L.;
15            Shinozuka, H. (1994) Dexamethasone inhibits induction of liver tumor necrosis factor-"alpha" mRNA and
16            liver growth induced by lead nitrate and ethylene dibromide. Am. J.  Pathol. 145: 951-958.
17     Lee, J. J.; Battles, A. H. (1994) Lead toxicity via arachidonate signal transduction to growth responses in the splenic
18            macrophages. Environ Res. 67: 209-219.
19     Lee, J.-E.; Dietert, R. R. (2003) Developmental immunotoxicity of lead: impact on thymic function. Birth Defects
20            Res. A Clin. Mol. Teratol. 67: 861-867.
21     Lee, J.-E.; Dietert, R. R. (2005) Toxicity  of lead to the developing immune system. In. Holladay, S. D., ed.
22            Developmental immunotoxicology. CRC Press, Inc. Boca Raton, FL. 169-177.
23     Lee, J.-E.; Chen, S.; Golemboski, K. A.; Parsons, P. J.; Dietert, R. R. (2001) Developmental windows of differential
24            lead-induced immunotoxicity in chickens. Toxicology 156: 161-170.
25     Lee, B.-K.; Lee, G.-S.; Stewart, W.  F.; Ahn, K.-D.; Simon, D.; Kelsey, K. T.; Todd, A. C.;  Schwartz, B. S. (2001)
26            Associations of blood pressure and hypertension with lead dose measures and polymorphisms in the vitamin
27            D receptor and "delta"-aminolevulinic acid dehydratase genes. Environ. Health Perspect. 109: 383-389.
28     Lee, J.-E.; Naqi, S. A.; Kao, E.; Dietert, R. R. Embryonic exposure to lead: Comparison of immune and cellular
29            responses in unchallenged and virally stressed chickens. Arch Toxicol. 75: 717-724. (2002) Embryonic
30            exposure to lead: Comparison of immune and cellular responses in unchallenged and virally stressed
31            chickens. Arch. Toxicol. 75: 717-724.
32     Legare, M. E.; Barhoumi, R.; Herbert, E.; Bratton, G. R.; Burghardt, R. C.; Tiffany-Castiglioni E. (1998) Analysis
33            of Pb2+ entry into cultured astroglia. Toxicol. Sci. 46: 90-100.
34     Leggett, R. W. (1993) An age-specific kinetic model of lead metabolism in humans. Environ. Health Perspect. 101:
35            598-616.
36     Lezak, M. (1995) Neuropsychological assessment. New York, NY: Oxford University Press.
37     Lidsky, T. I.; Schneider, J. S. (2003) Lead neurotoxicity in children: basic mechanisms and clinical correlates. Brain
38            126:5-19.
39     Lidsky, T. I.; Schneider, J. S. (2005) Adverse effects of childhood lead poisoning: the clinical neuropsychological
40            perspective. Environ. Res.:  in press.
41     Lilienthal, H.; Winneke, G. (1996) Lead effects on the brain stem auditory evoked potential in monkeys during and
42            after the treatment phase. Neurotoxicol.  Teratol. 18:17-32.
43     Lin, R. H.; Lee, C. H.; Chen, W. K.; Lin-Shiau, S. Y. (1994) Studies on cytotoxic and genotoxic effects of cadmium
44            nitrate and lead nitrate in Chinese hamster ovary cells. Environ. Mol. Mutagen. 23: 143-149.
45     Lin, C.; Kim, R.; Tsaih, S.-W.; Sparrow,  D.; Hu, H. (2004) Determinants of bone and blood lead levels among
46            minorities living in the Boston area. Environ. Health Perspect. 112: 1147-1151.
47     Lindahl, L. S.; Bird, L.; Legare, M.  E.; Mikeska, G.;  Bratton, G. R.; Tiffany-Castiglioni, E. (1999) Differential
48            ability of astroglia and neuronal cells to accumulate lead: dependence on cell type and on degree of
49            differentiation. Toxicol Sci. 50: 236-243.
50     Liu, J.; Kershaw, W. C.; Klaassen, C. D.  (1991)  The  protective effect of metallothionein on the toxicity of various
51            metals in rat primary hepatocyte  culture. Toxicol. Appl. Pharmacol.  107: 27-34.
52     Liu, J.-Y.; Lin, J.-K.; Liu, C.-C; Chen, W.-K.; Liu, C.-P.; Wang, C.-J.; Yen, C.-C; Hsieh, Y.-S. (1997)
53            Augmentation of protein kinase C activity and liver cell proliferation in lead nitrate-treated rats. Biochem.
54            Mol. Biol. Int. 43:355-364.
55     Logdberg, B.; Berlin, M.; Schutz, A. (1987) Effects of lead exposure on pregnancy outcome and the fetal brain of
56            squirrel monkeys. Scand. J. Work Environ. Health 13: 135-145.

       December 2005                               5-319         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Logdberg, B.; Bran, A.; Berlin, M; Schutz, A. (1988) Congenital lead encephalopathy in monkeys. Acta
  2            Neuropathol. 77: 120-127.
  3     Loipfuhrer, A. M.; Reichlmayr-Lais, A. M.; Kirchgessner, M. (1993) Concentration of free calcium in erythrocytes
  4            of lead-depleted rats. J. Trace Elem. Electrolytes Health Dis. 7: 37-40.
  5     Lolin, Y.; O'Gorman, P. (1988) An intra-erythrocyctic low molecular weight lead-binding protein in acute and
  6            chronic lead exposure and its possible protective role in lead toxicity. Ann. Clin. Biochem. 25: 688-97.
  7     Long, G. J.; Rosen, J. F. (1992) Lead perturbs epidermal growth factor (EOF) modulation of intracellular calcium
  8            metabolism and collagen synthesis in clonal rat osteoblastic (ROS 17/2.8) cells. Toxicol. Appl. Pharmacol.
  9            114:63-70.
10     Long, G. J.; Rosen, J. F.; Pounds, J. G. (1990) Lead impairs the production of osteocalcinby rat osteosarcoma (ROS
11            17/2.8) cells. Toxicol. Appl. Pharmacol. 106: 270-277.
12     Long, G. J.; Pounds, J.  G.; Rosen, J. F. (1992) Lead intoxication alters basal and parathyroid hormone-regulated
13            cellular calcium homeostasis in rat osteosarcoma (ROS 17/2.8) cells. Calcif. Tissue Int. 50: 451-458.
14     Long, G. J.; Rosen, J. F.; Schanne, F. A. X. (1994) Lead activation of protein kinase C from rat brain. Determination
15            of free calcium, lead, and zinc by 19F NMR. J. Biol. Chem. 269: 834-837.
16     Lucas, S. R.; Sexton, M.; Langenberg, P. (1996) Relationship between blood lead and nutritional factors in
17            preschool children: a cross-sectional study. Pediatrics 97: 74-78.
18     Luebke, R.; Chen, D.; Dietert, R. R.; King, M.; Yang, Y.; Luster, M. I. (2005) Increased sensitivity of the
19            developing immune system to xenobiotics: Experimental evidence supporting the concept of developmental
20            immunotoxicity testing guidelines. J. Toxicol. Environ. Health B: in press.
21     Luster, M. I.; Portier, C.; Pait, D. G.; White, K. L. J.; Gennings, C.; Munson, A. E.; Rosenthal, G. J. (1992) Risk
22            assessment in immunotoxicology.  I. Sensitivity and predictability of immune tests. Fund. Appl. Toxicol. 18:
23            200-210.
24     Lutz, P. M.; Wilson, T. J.; Ireland, A. L.; Gorman, J. S.; Gale, N. L.; Johnson, J. C.; Hewett, J. E.  (1999) Elevated
25            immunoglobulin E (IgE) levels in  children with exposure to environmental lead. Toxicology 134: 63-78.
26     Ma, T.; Chen, H. H.; Chang, H. L.; Hume, A.  S.; Ho, I. K. (1997) Effects of chronic lead exposure on  [3H]MK-801
27            binding in the brain of rat. Toxicol. Lett. 92: 59-66.
28     Maezawa, Y.; Nakajima, H.; Seto, Y.; Suto, A.; Kumana, K.; Kubo, S.; Karasuyama, H.; Saito, Y.; Iwamoto, I.
29            (2004) IgE-dependent enhancement of Th2 cell-mediated allergic inflammation in the airways. Clin. Exp.
30            Immunol. 135: 12-18.
31     Mahaffey, K. R.; Capar, S. G.; Gladen, B.  C.; Fowler, B. A. (1981) Concurrent exposure to lead, cadmium, and
32            arsenic. J. Lab. Clin. Med. 98: 463-481.
33     Mahaffey, K. R.; Rosen, J. F.; Chesney, R. W.; Peeler, J. T.; Smith, C. M.; De Luca, H. F. (1982) Association
34            between age, blood lead concentration, and serum 1,25-dihydroxycholecalciferol levels in children. Am. J.
35            Clin. Nutr. 35:  1327-1331.
36     Mahaffey, K. R.; Gartside, P. S.; Glueck, C. J. (1986) Blood lead levels and  dietary calcium intake in 1-11 year-old
37            children: the second national health and nutrition examination survey, 1976-1980. Pediatrics 78: 257-262.
38     Maisin, J. R.; Lambiet-Collier, M.; De Saint-Georges, L. (1978) Toxicite du plomb pour les embryons de la souris
39            [Lead toxicity for mouse embryos]. C. R. Seances Soc. Biol. Ses. Fil. 172: 1041-1043.
40     Maitani, T.; Watahiki, A.;  Suzuki, K. T. (1986) Induction of metallothionein after lead administration by three
41            injection routes in mice. Toxicol. Appl. Pharmacol. 83:211-217.
42     Maker, H. S.; Lehrer, G. M.; Silides, D. J.  (1975) The effect of lead on mouse brain development. Environ. Res. 10:
43            76-91.
44     Malamitsi-Pichner, A.; Protonotariou, E.; Boutsikou, T.; Makrakis, E.; Sarandakou, A.; Creatas, G. (2005) The
45            influence of the mode of delivery on circulating cytokine concentration in the perinatal period. Early Hum.
46            Dev. 81:387-392.
47     Malcoe, L. H.; Lynch, R. A.; Kegler, M. C.; Skaggs, V. J. (2002) Lead sources, behaviors, and socioeconomic
48            factors in relation to blood lead of Native American and white children: a community-based assessment of a
49            former mining  area. Environ. Health Perspect.  Suppl. 110(2): 221-231.
50     Maldonado-Vega, M.; Cerbon-Solorzano, J.; Albores-Medina, A.; Hernandez-Luna, C.; Calderon-Salinas, J. V.
51            (1996) Lead: intestinal absorption and bone mobilization during lactation. Hum. Exp. Toxicol. 15: 872-877.
52     Maldonado-Vega, M.; Cerbon-Solorzano, J.; Calderon-Salinas, J. V. (2002)  The effects of dietary calcium during
53            lactation on lead in bone mobilization: implications for toxicology. Hum. Exp. Toxicol. 21: 409-414.
54     Malvezzi, C. K.; Moreira, E. G.; Vassilieff, L; Vassilieff, V. S.; Cordellini, S. (2001) Effect of L-arginine,
5 5            dimercaptosuccinic acid (DMS A)  and the association of L-arginine and DMS A on tissue lead  mobilization
56            and blood pressure level in plumbism. Braz. J.  Med. Biol. Res. 34: 1341-1346.

       December 2005                                5-320         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Manton, W. I.; Cook, J. D. (1984) High accuracy (stable isotope dilution) measurements of lead in serum and
  2            cerebrospinal fluid. Br. J. Ind. Med. 41: 313-319.
  3     Manton, W. I.; Angle, C. R.; Stanek, K. L.; Reese, Y. R.; Kuehnemann, T. J. (2000) Acquisition and retention of
  4            lead by young children. Environ. Res. 82: 60-80.
  5     Manton, W. L; Rothenberg, S. J.; Manalo, M. (2001) The lead content of blood serum. Environ. Res. 86: 263-273.
  6     Marcus, A. H. (1985a) Multicompartment kinetic models for lead. I. Bone diffusion models for long-term retention.
  7            Environ. Res.  36: 441-458.
  8     Marcus, A. H. (1985b) Multicompartment kinetic models for lead. II. Linear kinetics and variable absorption in
  9            humans without excessive lead exposures. Environ. Res. 36: 459-472.
10     Marcus, A. H. (1985c) Multicompartment kinetic model for lead. III. Lead in blood plasma and erythrocytes.
11            Environ. Res.  36: 473-489.
12     Markovac, J.; Goldstein, G. W. (1988a) Lead activates protein kinase C in immature rat brain microvessels. Toxicol.
13            Appl. Pharmacol. 95: 14-23.
14     Markovac, J.; Goldstein, G. W. (1988b) Picomolar concentrations of lead stimulate brain protein kinase C. Nature
15            (London, U.K.) 334: 71-73.
16     Markowitz, M. E.; Sennett, M.; Rosen, J. F. (2004) A randomized trial of calcium supplementation for childhood
17            lead poisoning. Pediatrics 113: e34-e39.
18     Marques, M.; Millas, I.; Jimenez, A.; Garcia-Colis, E.; Rodriguez-Feo, J. A.; Velasco, S.; Barrientos, A.; Casado, S.;
19            Lopez-Farre, A. (2001) Alteration of the soluble guanylate cyclase system in the vascular wall of lead-
20            induced hypertension in rats. J. Am.  Soc. Nephrol. 12: 2594-2600.
21     Martin, L. M.; Martin, C.  C.; Vidas, M. M.; Vaziri, N. D.; Mateos-Caceres, P. J.; Perez, S. C.; Macaya, C.;
22            Barriento, A.;  Lopez-Farre, A. J. (2005) Involvement of endothelium and endothelin-1 in lead-induced
23            smooth muscle cell dysfunction in rats. Kidney Int.: in press.
24     Martins, M. B.; Sabatier, L.; Ricoul, M.; Pinton, A.; Dutrillaux, B. (1993) Specific chromosome instability induced
25            by heavy ions: a step towards transformation of human fibroblasts? Mutat. Res. 1285: 229-237.
26     Mas-Oliva, J. (1989) Effect of lead on the erythrocyte (Ca2+,Mg2+)-ATPase activity. Calmodulin involvement.
27            Mol. Cell. Biochem. 89: 87-93.
28     Mason, H. J.; Somervaille, L. J.; Wright, A. L.; Chettle, D. R.; Scott, M. C. (1990) Effect of occupational lead
29            exposure on serum 1,25-dihydroxyvitamin D levels. Hum. Exp. Toxicol. 9: 29-34.
30     Massie, H. R.; Aiello, V. R. (1992) Lead accumulation in the bones of aging male mice. Gerontology 38: 13-17.
31     Mauel, J.; Ransijn, A.; Buchmuller-Rouiller, Y. (1989) Lead inhibits intracellular killing of Leishmania parasites
32            and extracellular cytolysis of target cells by macrophages exposed to macrophage activating factor. J.
33            Leukoc. Biol.  45: 401-409.
34     Mazzolini, M.; Traverse, S.; Marchetti, C.  (2001) Multiple pathways of Pb(2+) permeation in rat cerebellar granule
35            neurones. J. Neurochem. 79: 407-416.
36     McCabe, M. J., Jr. (1994) Mechanisms and consequences of immunomodulationby lead. In: Dean, J. H.; Luster, M.
37            I.; Munson, A. E.; Kimber, I., eds. Immunotoxicology and immunopharmacology. 2nd ed. New York, NY:
38            RavenPress, Ltd.; pp. 143-162.
39     McCabe, M. J., Jr.; Lawrence, D. A. (1990) The heavy metal lead exhibits B cell-stimulatory factor activity by
40            enhancing B cell la expression and differentiation. J. Immunol. 145:  671-677.
41     McCabe, M. J.; Lawrence, D. A. (1991) Lead, a major environmental pollutant, is immunomodulatory by its
42            differential effects on CD4+ T cell subsets. Toxicol. Appl. Pharmacol. Ill: 13-23.
43     McCabe, M. J., Jr.; Bias, J. A.; Lawrence, D. A. (1991) Lead influences translational or posttranslational regulation
44            of la expression and increases invariant chain expression in mouse B cells.  J. Biochem. Toxicol. 6: 269-276.
45     McCabe, M. J., Jr.; Singh, K. P.; Reiners, J. J., Jr. (1999) Lead intoxication impairs the generation of a delayed type
46            hypersensitivity response. Toxicology 139: 255-264.
47     McCabe, M. J., Jr.; Singh, K. P.; Reiners, J. J., Jr. (2001) Low level lead exposure in vitro stimulates the
48            proliferation and expansion of alloantigen-reactive CD4high T cells. Toxicol. Appl. Pharmacol. 177: 219-
49            231.
50     McDonald, J. A.; Potter, N. U. (1996) Lead's legacy? Early and late mortality of 454 lead-poisoned children. Arch.
51            Environ. Health 51: 116-121.
52     McGivern, R. F.;  Sokol, R. Z.; Berman, N. G. (1991) Prenatal lead exposure in the rat during the third week of
53            gestation: long-term behavioral, physiological and anatomical effects associated with reproduction. Toxicol.
54            Appl. Pharmacol.  110: 206-215.
55     McGowan, C.; Donaldson, W. E. (1987) Effect of lead toxicity on the organ  concentration of glutathione and
56            glutathione-related free amino acids in the chick. Toxicol. Lett. 38: 265-270.

       December 2005                                5-321         DRAFT-DO NOT QUOTE OR CITE

-------
  1     McLachlin, J. R.; Goyer, R. A.; Cherian, M. G. (1980) Formation of lead-induced inclusion bodies in primary rat
  2            kidney epithelial cell cultures: effect of actinomycin D and cycloheximide. Toxicol. Appl. Pharmacol. 56:
  3            418-431.
  4     McNeill, F. E.; Laughlin, N. K.; Todd, A. C.; Sonawane, B. R.; Van de Wai, K. M.; Fowler, B. A. (1997) Geriatric
  5            bone lead metabolism in a female nonhuman primate population. Environ. Res. 72: 131-139.
  6     McNeill, D. R.; Narayana, A.; Wong, H. K.; Wilson, D. M. III. (2004) Inhibition of Apel nuclease activity by lead,
  7            iron, and cadmium. Environ. Health Perspect. 112: 799-804.
  8     Menegazzi, M.; Carcereri De Prati, A.; Ledda-Columbano, G. M.; Columbano, A.; Uchida, K.; Miwa, M.; Suzuki,
  9            H. (1990) Regulation of poly(ADP-ribose) polymerase mRNA levels during compensatory and mitogen-
10            induced growth of rat liver. Arch. Biochem. Biophys. 279:  232-236.
11     Menegazzi, M.; Carcereri de Prati, A.; Ogura, T.; Columbano, A.; Ledda-Columbano, G. M.; Libonati, M.; Esumi,
12            H.; Suzuki, H. (1992) Involvement of DNA polymerase beta in proliferation of rat liver induced by lead
13            nitrate  or partial hepatectomy. Febs. Lett. 310: 135-138.
14     Menegazzi, M.; Carcereri-De Prati, A.; Suzuki, H.; Shinozuka, H.; Pibiri, M.; Piga, R.; Columbano, A.; Ledda-
15            Columbano, G. M. (1997) Liver cell proliferation induced by nafenopin and cyproterone acetate is not
16            associated with increases in activation of transcription factors NF-"kappa"B and AP-1 or with expression of
17            tumor necrosis factor "alpha". Hepatology 25: 585-592.
18     Meng, X.-M.; Zhu, D.-M.; Ruan, D.-Y.; She, J.-Q.; Luo, L. (2005) Effects of chronic lead exposure on H MRS of
19            hippocampus and frontal lobes in children. Neurology 64: 1644-1647.
20     Miller, L.; Qureshi, M. A.  (1992) Heat-shock protein synthesis in chicken macrophages: Influence of in vivo and in
21            vitro heat shock, lead acetate, and lipopolysaccharide. Poul. Sci. 71: 988-998.
22     Miller, T. E.; Golemboski, K. A.; Ha, R. S.; Bunn, T.; Sanders, F. S.; Dietert, R. R. (1998) Developmental exposure
23            to lead causes persistent immunotoxicity in Fischer 344 rats. Toxicol. Sci. 42: 129-135.
24     Miller, D. K.; Nation, J. R.; Bratton, G. R. (2001) The effects of perinatal exposure to lead on the discriminative
25            stimulus properties of cocaine and related drugs in rats. Psychopharmacology (Berl). 158: 165-174.
26     Milosevic, N.; Maier, P. (2000) Lead stimulates intercellular signalling between hepatocytes and Kupffer cells.  Eur.
27            J. Pharmacol. 401:317-328.
28     Minnema, D. J.; Hammond, P. B. (1994) Effect of lead exposure on patterns of food intake in weanling rats.
29            Neurotoxicol. Teratol. 16: 623-629.
30     Minozzo, R.; Deimling, L. I.; Gigante, L. P.; Santos-Mello, R. (2004) Micronuclei in peripheral blood lymphocytes
31            of workers exposed to lead. Mutat. Res. 565: 53-60.
32     Misra, M.; Acharya, U. R.  (2004) Protective action of vitamins on the spermatogenesis in lead-treated Swiss mice.  J.
33            Trace Elem. Med.  Biol.  18: 173-178.
34     Mishra, K. P.; Singh, V. K.; Rani, R.; Yadav, V. S.; Chandran, V.; Srivastava, S. P.; Seth, P. K.  (2003) Effect of
3 5            lead exposure on the immune response of some occupationally exposed individuals. Toxicology 188: 251-
36            259.
37     Mistry, P.; Lucier, G. W.; Fowler, B. A. (1985) High-affinity lead binding proteins in rat kidney cytosol mediate
38            cell-free nuclear translocation of lead. J. Pharmacol. Exp. Ther. 232: 462-469.
39     Miyahara, T.; Komiyama, H.; Miyanishi, A.; Takata, M.; Nagai, M.; Kozuka, H.; Hayashi, T.; Yamamoto, M.; Ito,
40            Y.; Odake, H.; Koizumi, F. (1995) Stimulative effects of lead on bone resorption in organ culture.
41            Toxicology 97: 191-197.
42     Mobarak, N.; P'an, A. Y. (1984) Lead distribution in the saliva and blood fractions of rats after intraperitoneal
43            injections. Toxicology 32: 67-74.
44     Mojzis, J.; Nistiar, F. (2001) Lead-induced changes of cation-osmotic hemolysis in rats. Gen. Physiol. Biophys. 20:
45            315-319.
46     Momcilovic, B.; Kostial, K. (1974) Kinetics of lead retention and distribution in suckling and adult rats. Environ.
47            Res.  8: 214-220.
48     Moore, J. F.; Goyer, R. A.  (1974) Lead-induced inclusion bodies: composition and probable role in lead metabolism.
49            Environ. Health Perspect. 7:  121-127.
50     Moore, J. F.; Goyer, R. A.; Wilson, M. (1973) Lead-induced inclusion bodies: solubility, amino acid content, and
51            relationship to residual acidic nuclear proteins. Lab. Invest. 29: 488-494.
52     Morgan, R. E.; Levitsky, D. A.; Strupp,  B. J. (2000) Effects of chronic lead exposure on learning and reaction time
53            in a visual discrimination task. Neurotoxicol Teratol. 22: 337-345.
54     Morgan, R. E.; Caravan, H.; Smith, E. G.; Driscoll, L. L.; Levitsky, D. A.; Strupp, B. J. (2001) Early lead exposure
55            produces lasting changes in sustained attention, response initiation, and reactivity to errors. Neurotoxicol.
56            Teratol. 23:519-531.

       December 2005                                5-322         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Morita, Y.; Sakai, T.; Araki, S.; Araki, T.; Masuyama, Y. (1997) Nicotinamide adenine dinucleotide synthetase
  2            activity in erythrocytes as a tool for the biological monitoring of lead exposure. Int. Arch. Occup. Environ.
  3            Health 70: 195-198.
  4     Moser, R.; Oberley, T. D.; Daggett, D. A.; Friedman, A. L.; Johnson, J. A.; Siegel, F. L. (1995) Effects of lead
  5            administration on developing rat kidney: I. Glutathione S-transferase isoenzymes. Toxicol. Appl.
  6            Pharmacol. 131:85-93.
  7     Mousa, H. M; Al-Qarawi, A. A.; Ali, B. H.; Abdel Rahman, H. A.; ElMougy, S. A. (2002) Effect of lead exposure
  8            on the erythrocytic antioxidant levels in goats. J. Vet. Med. A Physiol. Pathol. Clin. Med. 49: 531-534.
  9     Mouw, D. R.; Vander, A. J.; Cox, J.; Fleischer, N.  (1978) Acute effects of lead on renal electrolyte excretion and
10            plasma renin activity. Toxicol. Appl. Pharmacol. 46: 435-447.
11     Mudd, S. H.; Levy, H. L.; Skovby, F. (1995) Disorders of transsulfuration. In: Scriver, C. R.; Beaudet, A. L.; Sly,
12            W. S.; Valle, D., eds. The metabolic and molecular bases of inherited disease. New York, NY: McGraw-
13            Hill Publishing Co.; pp. 1279-1328.
14     Muller, S.; Gillert, K.-E.; Krause, C.; Gross,  U.; L'Age-Stehr, J.; Diamantstein, T. (1977) Suppression of delayed
15            type hypersensitivity of mice by lead. Experientia 33: 667-668.
16     Murphy, K.J.; Regan, C. M. (1999) Low-level lead exposure in the early postnatal period results in persisting
17            neuroplastic deficits associated with memory consolidation.  J. Neurochem. 72: 2099-2104.
18     Nakagawa, K. (1991) Decreased glutathione S-transferase  activity in mice livers by acute treatment with lead,
19            independent of alteration in glutathione content. Toxicol. Lett. 56: 13-17.
20     Nakajima, T.; Deguchi, T.; Kagawa, K.; Hikita, H.; Ueda,  K.; Katagishi, T.; Ohkawara, T.; Kakusui, M.; Kimura,
21            H.; Okanoue, T.; Kashima, K.; Ashihara, T. (1995) Identification of apoptotic hepatocytes in situ in rat liver
22            after lead nitrate administration. J. Gastroenterol. 30: 725-730.
23     Nation, J. R.; Livermore, C. L.; Burkey, R. T. (1996) Chronic lead exposure attenuates sensitization to the
24            locomotor-stimulating effects of cocaine. Drug Alcohol Depend. 41: 143-149.
25     Nation, J. R.; Cardon, A. L.; Heard, H. M.; Valles, R.; Bratton, G. R. (2003) Perinatal lead exposure and relapse to
26            drug-seeking behavior in the rat: a cocaine reinstatement study. Psychopharmacology (Berl). 168:236-243.
27     Nation, J. R.; Smith, K. R.; Bratton, G. R. (2004) Early developmental lead exposure increases sensitivity to cocaine
28            in a self-administration paradigm. Pharmacol. Biochem. Behav. 77: 127-135.
29     National Research Council,  Committee on Measuring Lead in Critical Populations. (1993) Measuring lead exposure
30            in infants, children, and other sensitive populations. Washington, DC: National Academy Press. Available:
31            http://www.nap.edu/openbook/030904927X/html/ [21 July, 2005].
32     Nayak, B. N.; Ray, M.; Persaud, T. V. N. (1989) Maternal and fetal chromosomal aberrations in mice following
33            prenatal exposure to subembryotoxic doses of lead nitrate. Acta Anat. 135: 185-188.
34     Nayak, B. N.; Ray, M.; Persaud, T. V. N.; Nigli, M. (1989) Relationship of embryotoxicity to genotoxicity of lead
35            nitrate in mice. Exp. Pathol.  36: 65-73.
36     Needleman, H. L.; Gunnoe, C.; Leviton, A.; Reed, R.; Peresie, H.; Maher, C.; Barrett, P. (1979) Deficits in
37            psychologic and classroom performance of children with elevated dentine lead levels. N. Engl. J. Med. 300:
38            689-695.
39     Needleman, H. L.; Schell, A.; Bellinger, D.; Leviton, A.; Allred, E. N. (1990) The long-term effects of exposure to
40            low doses of lead in childhood; an 11-year follow-up report. N. Engl. J. Med. 322: 83-88.
41     Needleman, H. L.; McFarland, C.; Ness, R. B.; Fienberg, S. E.; Tobin, M.  J. (2002) Bone lead levels in adjudcated
42            delinquents. A case control study. Neurotoxicol. Teratol. 24: 711-717.
43     Nehez, M.; Lorencz, R.; Desi, I. (2000) Simultaneous action of cypermethrin and two environmental pollutant
44            metals, cadmium and lead, on bone marrow cell chromosomes of rats in subchronic administration.
45            Ecotoxicol. Environ. Saf. 45: 55-60.
46     Nehru, B.; Kaushal, S. (1992) Effect of lead on hepatic microsomal enzyme activity. J. Appl. Toxicol. 12: 401-405.
47     Neilan, B. A.; O'Neill, K.; Handwerger, B. S. (1983) Effect of low-level lead exposure on antibody-dependent and
48            natural killer cell-mediated cytotoxicity. Toxicol. Appl. Pharmacol. 69:  272-275.
49     Nemoto, K.; Miyata, S.; Nemoto, F.; Yasumoto, T.; Murai, U.; Kageyama, H.; Degawa, M. (2000) Gene expression
50            of neutrophins and their receptors in lead nitrate-induced rat liver hyperplasia. Biochem. Biophys. Res.
51            Commun. 275: 472-476.
52     Newland, M. C.; Yezhou, S.; Logdberg, B.; Berlin, M. (1994) Prolonged behavioral effects of inutero exposure to
53            lead or methyl mercury: reduced sensitivity to changes in reinforcement contingencies during behavioral
54            transitions and in steady state. Toxicol. Appl. Pharmacol.  126: 6-15.
55     Ni, Z.; Hou, S.; Barton, C. H.; Vaziri, N. D. (2004) Lead exposure raises superoxide and hydrogen peroxide in
56            human endothelial and vascular smooth muscle cells. Kidney Int. 66: 2329-2336.

       December 2005                                5-323         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Nihei, M. K.; Guilarte, T. R. (1999) NMDAR-2A subunit protein expression is reduced in the hippocampus of rats
  2            exposed to Pb2+ during development. Mol. Brain Res. 66: 42-49.
  3     Nihei, M. K.; Desmond, N. L.; McGlothan, J. L.; Kuhlmann, A. C.; Guilarte, T. R. (2000) N-methyl-D-aspartate
  4            receptor subunit changes are associated with lead-induced deficits of long-term potentiation and spatial
  5            learning. Neuroscience 99: 233-242.
  6     Nikolova, P.; Kavaldzhieva, B. (1991) The effect of certain heavy metals (Mn and Pb) on parameters of erythrocyte
  7            energy metabolism. J. Hyg. Epidemiol. Microbiol. Immunol. 35: 361-365.
  8     Nilsson, B. O.; Ljung, L.; Wide, M. (1991) Electron microscopy and X-ray microanalyses of uterine epithelium
  9            from lead-injected mice in an experimental delay of implantation. Arch. Toxicol. 65: 239-243.
10     Nolan, C. V.; Shaikh, Z. A. (1992) Lead nephrotoxicity and associated disorders: biochemical mechanisms.
11            Toxicology 73: 127-146.
12     Novak, J.; Banks, R. O. (1995) Lead and nickel alter the cardiorenal actions of endothelin in the rat. Proc. Soc. Exp.
13            Biol. Med. 208: 191-198.
14     O'Flaherty, E. J.; Inskip, M. J.; Franklin, C. A.; Durbin, P. W.; Manton, W. I.; Baccanale, C. L.  (1998) Evaluation
15            and modification of a physiologically based model of lead kinetics using data from a sequential isotope
16            study in cynomolgus monkeys. Toxicol. Appl. Pharmacol. 149: 1-16.
17     Oberley, T. D.; Friedman, A. L.; Moser, R.; Siegel, F. L.  (1995) Effects of lead administration on developing rat
18            kidney. II. functional, morphologic, and immunohistochemical studies.  Toxicol. Appl. Pharmacol. 131: 94-
19            107.
20     Odenbro, A.; Arrhenius, E. (1984) Effects of triethyllead chloride on hepatic microsomal N- and C-oxygenation of
21            N,N-dimethylaniline in rats. Toxicol. Appl. Pharmacol. 74: 357-363.
22     Odenbro, A.; Kihlstrom, J. E. (1977) Frequency of pregnancy and ova implantation in triethyl lead-treated mice.
23            Toxicol. Appl. Pharmacol. 39: 359-363.
24     Odenbro, A.; Rafter, J. (1988) Effects of triethyl lead chloride on oestradiol metabolism in the female rat liver
25            microsomal fraction. Pharmacol. Toxicol. (Copenhagen) 63: 248-52.
26     Odenbro, A.; Kihlstrom, I.; Kihlstrom, J. E. (1988) Perinatal growth retardation caused by triethyl lead chloride
27            treatment of mice during late gestation. Pharmacol. Toxicol. (Copenhagen) 63: 253-256.
28     Oishi, H.; Nakashima, M.; Totoki, T.; Tomokuni, K. (1996) Chronic lead exposure may inhibit  endothelium-
29            dependent hyperpolarizing factor in rats. J. Cardiovasc. Pharmacol. 28: 558-563.
30     Olshan, A. F.; Breslow, N. E.; Daling, J. R.; Falletta, J. M.; Grufferman, S.; Robison, L. L.; Waskerwitz, M.,
31            Hammond, G. D. (1990) Wilms' tumor and paternal occupation. Cancer Res. 50: 3212-3217.
32     Ong, C. N.; Lee, W. R. (1980) Distribution of lead-203 in human peripheral blood in vitro. Br. J. Ind. Med. 37: 78-
33            84.
34     Oomen, A. G.; Tolls, J.; Sips, A. J.; Groten, J. P. (2003) In vitro intestinal lead uptake and transport in relation to
35            speciation. Arch. Environ. Contam. Toxicol. 44:  116-124.
36     Oskarsson, A.; Fowler, B. A. (1985) Effects of lead inclusion bodies on subcellular distribution of lead in rat kidney:
37            the relationship to mitochondrial function. Exp. Mol. Pathol. 43: 397-408.
38     Oskarsson, A.; Squibb, K.  S.; Fowler, B. A. (1982) Intracellular binding of lead in the kidney: the partial isolation
39            and characterization of postmitochondrial lead binding components. Biochem. Biophys. Res. Commun. 104:
40            290-298.
41     Oskarsson, A.; Hellstrom-Lindahl, E. (1989) Lead-dithiocarbamate interaction.  Effect on ALAD activity in isolated
42            rat hepatocytes. Biol. Trace Elem. Res. 21: 325-330.
43     Osterode, W.; Ulberth, F. (2000) Increased concentration of arachidonic acid in erythrocyte membranes in
44            chronically lead-exposed men. J. Toxicol. Environ. Health A 59: 87-95.
45     Othman, A.  I.; El Missiry,  M. A. (1998) Role of selenium against lead toxicity in male rats. J. Biochem. Mol.
46            Toxicol. 12:  345-349.
47     Otto, D. A.; Fox, D. A. (1993) Auditory and visual dysfunction following lead exposure. Presented at: Ninth
48            international neurotoxicology conference; October  1991; Little Rock, AR. Neurotoxicology 14(2-3): 191-
49            207.
50     Pace, B. M.; Lawrence, D. A.; Behr, M. J.; Parsons, P. J.; Dias, J. A. (2005) Neonatal lead exposure changes quality
51            of sperm and number of macrophages in testes of BALB/c mice. Toxicology 210: 247-256.
52     Pagliara, P.; Carla, E. C.; Caforio, S.; Chionna, A.; Massa, S.; Abbro, L.; Dini, L.  (2003a) Kupffer cells promote
53            lead nitrate-induced hepatocyte apoptosis via oxidative stress. Comp. Hepatol. 2:8-21.
54     Pagliara, P.; Chionna, A.; Carla, E. C.; Caforio, S.; Dini,  L. (2003b) Lead nitrate and gadolinium chloride
55            administration modify hepatocyte  cell surfaces. Cell Tissue Res. 312: 41-48.


       December 2005                                 5-324         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Palminger Hallen, I.; Jonsson, S.; Karlsson, M. O.; Oskarsson, A. (1996) Kinetic observations in neonatal mice
  2            exposed to lead via milk. Toxicol. Appl. Pharmacol. 140: 13-18.
  3     Palus, I; Rydzynski, K.; Dziubaltowska, E.; Wyszynska, K.; Natarajan, A. T.; Nilsson, R. (2003) Genotoxic effects
  4            of occupational exposure to lead and cadmium. Mutat. Res. 540: 19-28.
  5     Pande, M.; Flora, S. J. (2002) Lead induced oxidative damage and its response to combined administration of
  6            "alpha"-lipoic acid and succimers in rats. Toxicology 177: 187-196.
  7     Panemangalore, M.; Bebe, F. N. (1996) Effects of low oral lead and cadmium exposure and zinc status on heme
  8            metabolites in weanling rats. Int. J. Occup. Med. Environ. Health 9: 141-151.
  9     Pani, P.; Dessi, S.; Rao, K. N.; Batetta, B.; Laconi, E. (1984) Changes in serum and hepatic cholesterol in lead-
10            induced liver hyperplasia. Toxicol. Pathol. 12: 162-167'.
11     Pappas, J. B.; Ahlquist, J. T.; Allen, E. M.; Banner, W., Jr. (1995) Oral dimercaptosuccinic acid and ongoing
12            exposure to lead: effects on heme synthesis and lead distribution in a rat  model. Toxicol. Appl. Pharmacol.
13            133: 121-129.
14     Patierno, S. R.; Landolph, J. R. (1989) Soluble vs insoluble hexavalent chromate. Relationship of mutation to in
15            vitro transformation and particle uptake. Biol. Trace Elem. Res. 21: 469-474.
16     Patierno, S. R.; Banh, D.; Landolph, J. R. (1988) Transformation of C3H/10T1/2 mouse embryo cells to focus
17            formation and anchorage independence by insoluble lead chromate but not soluble calcium chromate:
18            relationship to mutagenesis and internalization of lead chromate particles. Cancer Res. 48: 5280-5288.
19     Patra, R. C.; Swamp, D. (2000) Effect of lead on erythrocytic antioxidant defence, lipid peroxide level and thiol
20            groups in calves. Res. Vet. Sci. 68: 71-74.
21     Patra, R. C.; Swamp, D.; Dwivedi, S.  K. (2001) Antioxidant effects of "alpha" tocopherol, ascorbic acid and L-
22            thenionine on lead induced oxidative stress to the liver, kidney and brain in rats. Toxicology 162: 81-88.
23     Payne, K. J.; Crooks, G. M. (2002) Human hematopoietic lineages. Immunol. Rev. 187: 46-64.
24     Peixoto, N.  C.; Roza, T.; Pereira, M. E. (2004) Sensitivity of "delta"-ALA-D (B.C.  4.2.1.24) of rats to  metals in vitro
25            depends on the stage of postnatal growth and tissue. Toxicol. in Vitro 18: 805-809.
26     Pentschew,  A.; Garro, F. (1966) Lead encephalo-myelopathy of the suckling rat and its implications on the
27            porphyrinopathic nervous diseases, with special reference to the permeability disorders of the  nervous
28            system's capillaries. Acta Neuropathol. 6: 266-278.
29     Pereira, B.;  Curi, R.; Kokubun, E.; Bechara, E. J. H.  (1992) 5-aminolevulinic acid-induced alterations  of oxidative
30            metabolism in sedentary and exercise-trained rats. J. Appl. Physiol. 72: 226-230.
31     Perez-Bravo, F.; Ruz, M; Moran-Jimenez, M. J.; Olivares, M.; Rebolledo, A.; Codoceo, J.; Sepulveda, J.; Jenkin, A.;
32            Santos, J. L.; Fontanellas, A. (2004) Association between aminolevulinate dehydrase genotypes and blood
33            lead levels in children from a lead-contaminated area in Antofagasta, Chile. Arch. Environ. Contam.
34            Toxicol. 47(2): 276-280.
35     Pergande, M.; Jung, K.; Precht, S.; Pels, L. M.; Herbert, C.; Stolte, H. (1994) Changed excretion of urinary proteins
36            and enzymes by chronic exposure to lead. Nephrol. Dial. Transplant. 9: 613-618.
37     Piasek, M.; Kostial, K.  (1991) Reversibility of the effects of lead on the reproductive performance of female rats.
3 8            Reprod. Toxicol. 5:45-51.
39     Piccinini, F.; Favalli, L.; Chiari, M. C. (1977) Experimental investigations on the conctraction induced by lead in
40            arterial smooth muscle. Toxicology 8: 43-51.
41     Pillai, A.; Gupta, S. (2005) Effect of gestational and  lactational exposure to lead and/or cadmium on reproductive
42            performance and hepatic oestradiol metabolising enzymes. Toxicol. Lett. 155: 179-186.
43     Pineda-Zavaleta,  A. P.; Gracia-Vargas, G.; Borja-Aburto, V. H.; Acosta-Saavedea, L. C.; Vera Aguilar, E.; Gomez-
44            Munoz, A.; Cebrian, M. E. Calderon-Aranda, E.  S. (2004) Nitric oxide and superoxide anion production in
45            monocytes from children exposed to arsenic and lead in region Lagunera, Mexico. Toxicol.  Appl.
46            Pharmacol. 198: 283-290.
47     Pinkerton, L. E.; Biagini, R. E.; Ward, E. M.; Hull, R. D.; Deddens, J. A.; Boeniger, M. F.; Schnorr, T. M.;
48            MacKenzie, B. A.; Luster, M. I. (1998) Immunologic findings among lead-exposed workers. Am. J. Ind.
49            Med. 33:400-408.
50     Pinon-Lataillade, G.; Thoreux-Manlay, A.; Coffigny, H.; Monchaux, G.; Masse,  R.; Soufir, J.-C. (1993) Effect of
51            ingestion and inhalation of lead on the reproductive system and fertility of adult male rats and their progeny.
52            Hum.  Exp. Toxicol. 12: 165-172.
53     Pinon-Lataillade, G.; Thoreux-Manlay, A.; Coffigny, H.; Masse, R.;  Soufir, J. C. (1995) Reproductive toxicity of
54            chronic lead exposure in male and female mice. Hum. Exp. Toxicol. 14:  872-878.
55     Pinto, D.; Ceballos, J. M.; Garcia, G.; Guzman, P.; Del Razo, L. M.;  Vera, E.; Gomez, H.; Garcia, A.;  Gonsebatt, M.
56            E. (2000) Increased cytogenetic damage in outdoor painters. Mutat. Res. 467:  105-111.

       December 2005                                 5-325         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Pires, J. B.; Bezerra, F. F.; Laboissiere, F. P.; Miekeley, N.; Donangelo, C. M. (2001) Lead levels in erythrocytes
  2            and biomarkers of bone turnover in pregnant and lactating women with marginal calcium intakes. Nutr. Res.
  3            21:831-841.
  4     Pirkle, J. L.; Schwartz, J.; Landis, J. R.; Harlan, W. R. (1985) The relationship between blood lead levels and blood
  5            pressure and its cardiovascular risk implications. Am. J. Epidemiol. 121: 246-258.
  6     Pirkle, J. L.; Brody, D. J.; Gunter, E. W.; Kramer, R. A.; Paschal, D. C.; Flegal, K. M.; Matte, T. D. (1994) The
  7            decline in blood lead levels in the United States: the National Health and Nutrition Examination Surveys
  8            (NHANES). JAMA J. Am. Med. Assoc. 272: 284-291.
  9     Pirkle, J. L.; Kaufmann, R. B.; Brody, D. J.; Hickman, T.; Gunter, E. W.; Paschal, D. C. (1998) Exposure of the U.S.
10            population to lead, 1991-1994. Environ. Health Perspect. 106: 745-750.
11     Planas-Bohne, F.; Elizalde, M. (1992) Activity of glutathione-S-transferase in rat liver and kidneys after
12            administration of lead or cadmium. Arch. Toxicol. 66:  365-367.
13     Poblano, A.; Rothenberg, S. J.; Schnaas, L.; Elias, Y.;  Cruz, M. L. (2001) Spatial distribution of EEC theta activity
14            as a function of lifetime lead exposure in 9-year-old children. Neurotoxicology 22: 439-446.
15     Poretz, R. D.; Yang, A.; Deng, W.; Manowitz, P. (2000) The interaction of lead exposure and arylsulfatase A
16            genotype affects sulfatide catabolism in human fibroblasts. Neurotoxicology 21: 379-387.
17     Pounds, J. G.; Rosen, J. F. (1986) Cellular metabolism of lead: a kinetic analysis in cultured osteoclastic bone cells.
18            Toxicol. Appl. Pharmacol.  83: 531-545.
19     Pounds, J. G.; Long, G. J.; Rosen, J. F. (1991) Cellular and molecular toxicity of lead in bone. Environ. Health
20            Perspect. 91:  17-32.
21     Prentice, R. C.; Kopp, S. J. (1985) Cardiotoxicity of lead at various perfusate calcium concentrations: functional and
22            metabolic responses of the  perfused rat heart. Toxicol. Appl. Pharmacol. 81: 491-501.
23     Price, R. G. (2000) Urinalysis to exclude and monitor nephrotoxicity. Clin. Chim. Acta 297:  173-182.
24     Price, R. G.; Taylor, S. A.; Olivers, I.; Arce-Tomas, M.; Crutcher, E.; Franchini, I.; Slinovi, R.; Cavazzini, S.;
25            Bergamaschi, E.; Mutti, A.; Vettori, M. V.; Lauwerys, R.; Bernard, A.; kabanda, A.; Roels, H.; Thielemans,
26            N.; Hotz, P.; De Broe, M. E.; Elseviers, M. M.; Nuyts, G. D.; Gelpi, E.; Hotter, G.; Rosello, J.; Ramis, I.;
27            Stolte, H.; Pels, L. M.; Eisenberger, U. (1996) Development and validation of new screening tests for
28            nephrotoxic effects. Hum. Exp. Toxicol. 15(suppl. 1): S10-S19.
29     Prigge, E.; Greve, J. (1977) Effekte einer Bleiinhalation alleinund inKombination mit Kohlenmonoxidbei
3 0            nichttragenden und tragenden Ratten und deren Feten. II. Effekte auf die Aktivitat der 5-Aminolavulinsaure-
31            Dehydratase, den Hematokrit und das Korpergewicht [Effects of lead inhaltion exposures alone and in
32            combination with carbon monoxide in nonpregnant and pregnant rats and fetuses. II. Effects on 5-
33            aminolevulinic acid dehydratase activity, hematocrit and body weight. Zentralbl. Bakteriol. Parasitenkd.
34            Infektionskrankh. Hyg.  16: 294-304.
35     Purdy, R. E.; Smith, J. R.; Ding, Y.; Oveisi, F.; Varizi, N. D.; Gonick, H. C. (1997) Lead-induced hypertension is
36            not associated with altered vascular reactivity  in vitro. Am.  J. Hypertens. 10: 997-1003.
37     Pyatt, D. W.; Zheng, J.-H.; Stillman, W. S.; Irons, R. D. (1996) Inorganic lead activates NF-kB in primary human
38            CD4+ T lymphocytes. Biochem.  Biophys. Res. Commun. 227: 380-385.
39     Qian, Z. M.; Morgan, E. H. (1990)  Effect of lead on the transport of transferrin-free and transferrin-bound iron into
40            rabbit reticulocytes. Biochem. Pharmacol. 40:  1049-1054.
41     Qian, Y.; Harris, E. D.; Zheng, Y.;  Tiffany-Castiglioni, E. (2000) Lead targets GRP78, a molecular chaperone, in C6
42            rat glioma cells. Toxicol. Appl. Pharmacol. 163: 260-266.
43     Qian, Y., Zheng, Y., Ramos, K.  S.; Tiffany-Castiglioni, E. (2005) GRP78 compartmentalized redistribution in Pb-
44            treated glia: role of GRP78 in lead-induced oxidative stress. Neurotoxicology. 26: 267-275.
45     Qu, W.; Diwan, B. A.; Liu, J.; Goyer, R. A.; Dawson,  T.; Horton, J. L.; Cherian, M. G.; Waalkes, M. P. (2002) The
46            metallothionein-null phenotype is associated with heightened sensitivity to lead toxicity and an inability to
47            form inclusion bodies. Am. J. Pathol. 160: 1047-1056.
48     Queiroz, M. L.  S.; Almeida, M.; Gallao, M. I.; Hoehr,  N. F. (1993) Defective neutrophil function in workers
49            occupationally exposed to lead. Pharmacol. Toxicol. 72: 73-77.
50     Quinlan, G. J.; Halliwell, B.; Moorhouse, C. P.; Gutteridge, J. M. (1988) Action of lead(II) and aluminium (III) ions
51            on iron-stimulated lipid peroxidation in liposomes, erythrocytes and rat liver microsomal fractions.
52            Biochim. Biophys. Acta. 962: 196-200.
53     Quintanilla-Vega, B.; Smith, D.  R.; Kahng, M. W.; Hernandez, J. M.; Albores, A.; Fowler, B. A. (1995) Lead-
54            binding proteins in brain tissue of environmentally lead-exposed humans. Chem. Biol. Interact. 98: 193-209.
55     Raghavan, S. R. V.; Culver, B. D.;  Gonick, H. C. (1980) Erythrocyte lead-binding protein after occupational
56            exposure. I. Relationship to lead toxicity. Environ. Res. 22: 264-270.

       December 2005                                 5-326         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Raghavan, S. R. V.; Culver, B. D.; Gonick, H. C. (1981) Erythrocyte lead-binding protein after occupational
  2            exposure. II. influence on lead inhibition of membrane Na+, K+ - adenosinetriphosphatase. J. Toxicol.
  3            Environ. Health 7: 561-568.
  4     Rajah, T. T.; Ahuja, Y. R. (1995) In vivo genotoxic effects of smoking and occupational lead exposure in printing
  5            press workers. Toxicol. Lett. 76: 71-75.
  6     Rajah, T. T.; Ahuja, Y. R. (1996) In vivo genotoxicity of alcohol consumption and lead exposure in printing press
  7            workers. Alcohol 13: 65-68.
  8     Ramesh, G. T.; Manna, S. M; Aggrarwal, B. B.; Jadhav, A. L. (1999) Lead activates nuclear transcription factor
  9            NF-kB, activator protein-1 and amino-terminal c-Jun kinase in pheochromocytoma cells. Toxicol. Appl.
10            Pharmacol. 155: 280-286.
11     Ramesh, G. T.; Manna, S. K.; Aggarwal, B. B.; Jadhav,  A. L. (2001) Lead exposure activates nuclear factor kappa
12            B, activator protein-1, c-Jun N-terminal kinase and caspases in the rat brain. Toxicol. Lett. 123: 195-207
13     Razani-Boroujerdi, S.; Edwards, B.; Sopori, M. L. (1999) Lead stimulates lymphocyte proliferation through
14            enhanced T cell-B cell interaction. J. Pharmacol. Exp. Therap. 288: 714-719.
15     Redig, P. T.; Lawler, E. M.; Schwartz, S.; Dunnette, J. L.; Stephenson, B.; Duke, G. E. (1991) Effects of chronic
16            exposure to sublethal concentrations of lead acetate on herne synthesis and immune function in red-tailed
17            hawks. Arch. Environ. Contam. Toxicol. 21: 72-77.
18     Reigart, J. R.; Graber, C. D. (1976) Evaluation of the humoral immune response of children with low level lead
19            exposure. Bull. Environ. Contam. Toxicol. 16: 112-117.
20     Restrepo, H. G.; Sicard, D.; Torres, M. M. (2000) DNA damage and repair in cells of lead exposed people. Am. J.
21            Ind. Med. 38: 330-334.
22     Reuhl, K. R.; Rice, D. C.; Gilbert, S. G.; Mallett, J. (1989) Effects of chronic developmental lead exposure on
23            monkey neuroanatomy: visual system. Toxicol.  Appl. Pharmacol. 99: 501-509.
24     Revis, N. W.; Zinsmeister, A. R.; Bull, R. (1981) Atherosclerosis and hypertension induction by lead and cadmium
25            ions: an effect prevented by calcium ion. Proc. Natl. Acad. Sci. U. S. A. 78: 6494-6498.
26     Reyes, A.; Mercado, E.; Goicoechea, B.; Rosado, A. (1976) Participation of membrane sulfhydryl groups in the
27            epididymal maturation of human and rabbit spermatozoa. Fertil. Steril. 27: 1452-1458.
28     Rice, D. C. (1990) Lead-induced behavioral impairment on a spatial discrimination reversal task in monkeys
29            exposed during different periods of development. Toxicol.  Appl. Pharmacol. 106: 327-333.
30     Rice, D. C. (1992a) Lead exposure during different developmental periods produces different effects on FI
31            performance in monkeys tested as juveniles and adults. Neurotoxicology 13: 757-770.
32     Rice, D. C. (1992b) Effect of lead during different developmental periods in the monkey on concurrent
33            discrimination performance. Neurotoxicology 13: 583-592.
34     Rice, D. C. (1992c) Behavioral effects of lead in monkeys tested during infancy and adulthood. Neurotoxicol.
35            Teratol. 14: 235-245.
36     Rice, D. C. (1997) Effects of lifetime lead exposure in monkeys on detection of pure tones. Fundam. Appl. Toxicol.
37            36: 112-118.
38     Rice, D.; Barone, S., Jr. (2000) Critical periods of vulnerability for the developing nervous system: evidence from
39            humans and animal models. Environ. Health Perspect. Suppl. 108(3): 511-533.
40     Rice, D. C.; Gilbert, S. G. (1990) Sensitive periods for lead-induced behavioral impairment (nonspatial
41            discrimination reversal) in monkeys. Toxicol. Appl.  Pharmacol. 102: 101-109.
42     Rice, D. C.; Gilbert, S. G. (1990) Lack of sensitive period for lead-induced behavioral impairment on a spatial
43            delayed alternation task in monkeys. Toxicol. Appl.  Pharmacol. 103: 364-373.
44     Richardt, G.; Federolf, G.; Habermann, E. (1986) Affinity of heavy metal ions to intracellular Ca2+-binding
45            proteins. Biochem. Pharmacol. 35: 1331-1335.
46     Rico, J.; Kordas, K.;  et al. (2005) The efficacy of iron and/or zinc supplementation on cognitive performance of
47            lead-exposed mexican school children: a randomized, placebo-controlled trial. Pediatrics: in press.
48     Rijhsinghani, K.; Choi, H.-S. H.; Burton, L. A.; Paronetto, F.; Tavoloni, N. (1993) Immunoelectron microscopy
49            identification of early proliferating cells in rat liver tissue during hyperplasia induced by lead nitrate.
50            Hepatology (Baltimore) 17: 685-692.
51     Ris, M. D.; Dietrich, K. N.; Succop, P. A.; Berger, O.  G.; Bornschein, R. L. (2004) Early exposure to lead and
52            neuropsychological outcome in adolescence. J. Int. Neuropsychol. Soc. 10: 261-270.
53     Rizzi, C. A.; Manzo, L.; Tonini, M.; Minoia, C.; Crema, A. (1989) Propulsive motility of the guinea-pig colon after
54            chronic lead treatment. Pharmacol. Res. 21: 127-128.
55     Roberts, J. R.; Reigart, J. R.; Ebeling, M.; Hulsey, T. C. (2001) Time required for blood lead levels to decline in
5 6            nonchelated children. Clin. Toxicol.  39:153-160.

       December 2005                                5-327         DRAFT-DO NOT  QUOTE OR CITE

-------
  1     Rocha, A.; Valles, R.; Cardon, A. L.; Bratton, G. R.; Nation, J. R. (2005) Enhanced acquisition of cocaine self-
  2            administration in rats developmentally exposed to lead. Neuropsychopharmacology 30: 2058-2064.
  3     Rodamilans, M.; Mtz.-Osaba, M. J.; To-Figueras, J.; Rivera-Fillat, F.; Torra, M.; Perez, P.; Corbella, J. (1988)
  4            Inhibition of intratesticular testosterone synthesis by inorganic lead. Toxicol. Lett. 42: 285-290.
  5     Rodrigues, A. L.; Rocha, J. B.; Pereira, M. E.; Souza, D. O. (1996) Delta-aminolevulinic acid dehydratase activity in
  6            weanling and adult rats exposed to lead acetate. Bull. Environ. Contam. Toxicol. 57: 47-53.
  7     Rodriguez-Iturbe, B.; Vaziri, N.  D.; Herrera-Acosta, J.; Johnson, R. J. (2004) Oxidative stress, renal infiltration of
  8            immune cells, and salt-sensitive hypertension: all for one and one for all. Am. J. Physiol. 286: F606-F616.
  9     Rodriguez-Iturbe, B.; Sindhu, R. K.; Quiroz, Y.; Vaziri, N. D. (2005) Chronic exposure to low doses of lead results
10            in renal infiltration of immune cells, NF-"kappa"B activation, and overexpression of tubulointerstitial
11            angiotensin II. Antioxid. Redox Signaling 7: 1269-1274.
12     Roels, H.; Lauwerys, R.; Konings, J.; Buchet, J.-P.; Bernard, A.; Green, S.; Bradley, D.; Morgan, W.; Chettle, D.
13            (1994)  Renal function and hyperfiltration capacity in lead smelter workers with high bone lead. Occup.
14            Environ. Med. 51: 505-512
15     Rogan, W. J.; Dietrich, K. N.; Ware, J. H.; Dockery, D. W.; Salganik, M.; Radcliffe, J.; Jones, R. L.; Ragan, N. B.;
16            Chisolm, J.  J., Jr.; Rhoads, G. G. (2001) The effect of chelation therapy with succimer on
17            neuropsychological development in children exposed to lead. New Engl. J. Med. 344:  1421-1426.
18     Ronis, M. J. J.;  Badger, T. M.; Shema, S. J.; Roberson, P. K.; Shaikh, F. (1996) Reproductive toxicity and growth
19            effects  in rats exposed to lead at different periods during development. Toxicol. Appl.  Pharmacol. 136: 361-
20            371.
21     Ronis, M. J.; Badger, T.  M.; Shema, S. J.; Roberson, P. K.; Shaikh, F. (1998a) Effects onpubertal growth and
22            reproduction in rats exposed to lead perinatally or continuously throughout development. J. Toxicol.
23            Environ. Health A 53: 327-341.
24     Ronis, M. J. J.;  Gandy, J.; Badger, T. (1998b) Endocrine mechanisms underlying reproductive toxicity in the
25            developing  rat chronically exposed to dietary lead. J. Toxicol. Environ. Health Part A 54: 77-99.
26     Ronis, M. J. J.;  Badger, T. M.; Shema, S. J.; Roberson, P. K.; Templer, L.; Ringer, D.; Thomas, P. E. (1998c)
27            Endocrine mechanisms underlying the growth effects of developmental lead exposure in the rat. J. Toxicol.
28            Environ. Health Part A 54: 101-120.
29     Ronis, M. J. J.;  Aronson, J.; Gao, G. G.; Hogue, W.; Skinner, R. A.; Badger, T. M.; Lumpkin, C. K., Jr. (2001)
30            Skeletal effects of developmental lead exposure in rats. Toxicol.  Sci. 62: 321-329.
31     Roomi, M. W.;  Columbano, A.;  Ledda-Columbano, G. M.; Sarma, D. S. R. (1986) Lead nitrate induces certain
32            biochemical properties characteristic of hepatocyte nodules.  Carcinogenesis  7: 1643-1646.
33     Roomi, M. W.;  Columbano, A.;  Ledda-Columbano, G. M.; Sarma, D. S. R. (1987) Induction of the placenta! form
34            of glutathione S-transferase by lead nitrate administration in rat liver. Toxicol. Pathol.  15: 202-205.
3 5     Rosen, J. F. (1983) The metabolism of lead in isolated bone cell populations: interactions between lead and calcium.
36            Toxicol. Appl. Pharmacol. 71: 101-112.
37     Rosen, J. F.; Markowitz, M. E. (1980) D-Penicillamine: its actions on lead transport  in bone organ culture. Pediatr.
38            Res. 14: 330-335.
39     Rosen, J. F.; Mushak, P. (2001) Primary prevention of childhood lead poisoning ~ the only solution [comment]. N.
40            Engl. J. Med. 344: 1470-1471.
41     Rosen, J. F.; Pounds, J. G. (1988) The cellular metabolism of lead and calcium: a kinetic analysis in cultured
42            osteoclastic bone cells. In: De Broe, M. E.; Van de Vyver, F. L.,  eds. Bone and renal failure: international
43            symposium; November  1986; Antwerp, Belgium.  Basel, Switzerland: S. Karger; pp. 74-82. (Contributions
44            to nephrology: v. 64.)
45     Rosen, J. F.; Pounds, J. G. (1989) Quantitative interactions between Pb2+ and Ca2+  homeostasis in cultured
46            osteoclastic bone cells. Toxicol. Appl. Pharmacol. 98: 530-543.
47     Rosen, J. F.; Wexler, E.  E. (1977) Studies of lead transport in bone organ culture. Biochem. Pharmacol. 26: 650-
48            652.
49     Rosen, J. F.; Chesney, R. W.; Hamstra, A.; DeLuca, H. F.; Mahaffey, K. R. (1980) Reduction in 1,25-
50            dihydroxyvitamin D in children with increased lead absorption. N.  Engl. J. Med. 302:  1128-1131.
51     Rosen, J. F.; Kraner, H. W.; Jones, K. W. (1982) Effects of CaNa2EDTA on lead and trace metal metabolism in
52            bone organ  culture.  Toxicol. Appl. Pharmacol. 64: 230-235.
53     Rossi, E.; Attwood,  P. V.; Garcia-Webb, P. (1992) Inhibition of human lymphocyte coproporphyrinogen oxidase
54            activity by metals, bilirubin and haemin. Biochim. Biophys.  Acta 1135: 262-268.
       December 2005                                5-328         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Rothenberg, S. I; Cansino, S.; Sepkoski, C.; Torres, L. M; Medina, S.; Schnaas, L.; Poblano, A.; Karchmer, S.
  2            (1995) Prenatal and perinatal lead exposures alter acoustic cry parameters of neonate. Neurotoxicol. Teratol.
  3            17: 151-160.
  4     Rothenberg, S. I; Poblano, A.; Schnaas, L. (2000) Brainstem auditory evoked response at five years and prenatal
  5            and postnatal blood lead. Neurotoxicol. Teratol. 22: 503-510.
  6     Roy, A. K.; Dhir, H.; Sharma, A. (1992) Modification of metal-induced micronuclei formation in mouse bone
  7            marrow erythrocytes by Phyllanthus fruit extract and ascorbic acid. Toxicol. Lett. 62: 9-17.
  8     Ruan, D.-Y.; Chen, J.-T.; Zhao, C.; Xu, Y.-Z.; Wang, M.; Zhao, W.-F. (1998) Impairment of long-term potentiation
  9            and paired-pulse facilitation in rat hippocampal dentate gyrus following developmental lead exposure in
10            vivo. BrainRes. 806: 196-201.
11     Ruzittu, M.; Carla, E. C.; Montinari, M. R.; Maietta, G.; Dini, L. (1999) Modulation of cell surface expression of
12            liver carbohydrate receptors during in vivo induction of apoptosis with lead nitrate. Cell Tissue Res. 298:
13            105-112.
14     Sabbioni, E.; Marafante, E. (1976) Identification of lead-binding components in rat liver: in vivo study. Chem. Biol.
15            Interact. 15: 1-20.
16     Sanchez-Fructuoso, A.  L; Blanco, I; Cano, M.; Ortega, L.; Arroyo, M.; Fernandez, C.; Prats, D.; Barrientos, A.
17            (2002a) Experimental lead nephropathy: treatment with calcium disodium ethylenediaminetetraacetate. Am.
18            J. Kidney Dis.  40: 59-67.
19     Sanchez-Fructuoso, A.  L; Cano, M.; Arroyo, M.; Fernandez, C.; Prats, D.; Barrientos, A. (2002b) Lead mobilization
20            during calcium disodium ethylenediaminetetraacetate chelation therapy in treatment  of chronic lead
21            poisoning. Am. J. Kidney Dis. 40: 51-58.
22     Sandhir, R.; Gill, K. D. (1995) Effect of lead on lipid peroxidation in liver of rats. Biol. Trace Elem.  Res. 48: 91-97.
23     Sanin, L. H.; Gonzalez-Cossio, T.; Romieu, I.; Peterson, K. E.; Ruiz, S.; Palazuelos, E.; Hernandez-Avila, M.; Hu,
24            H. (2001) Effect of maternal lead burden on infant weight and weight gain at one month of age among
25            breastfed infants. Pediatrics 107: 1016-1023.
26     Sant'Ana, M. G.; Spinosa, H. S.; Florio, J. C.; Bernardi, M. M.; Oliveira, C. A.; Sarkis, J. E.;  Kakazu, M. H. (2001)
27            Role of early GnRH administration in sexual behavior disorders of rat pups perinatally exposed to lead.
28            Neurotoxicol. Teratol. 23: 203-212.
29     Santos, J. L.; Fontanellas, A.; Moran, M.  J.; Enriquez de Salamanca, R. (1999) Nonsynergic  effect of ethanol and
30            lead on heme metabolism in rats. Ecotoxicol. Environ. Saf. 43: 98-102.
31     Sarasua, S. M.; Vogt, R. F.; Henderson, L. O.; Jones, P. A.; Lybarger, J.  A. (2000) Serum immunoglobulins and
32            lymphocyte subset distributions in children and adults living in communities assessed for lead  and cadmium
33            exposure. J. Toxicol. Environ. Health A. 60: 1-15.
34     Sargent, J. D.; Dalton, M. A.; O'Connor, G. T.; Olmstead, E. M.; Klein, R.  Z. (1999) Randomized trial  of calcium
35            glycerophosphate-supplemented infant formula to prevent lead absorption. Am. J. Clin. Nutr. 69: 1224-
36            1230.
37     Satija, N. K.; Vij, A. G. (1995) Preventive action of zinc against lead toxicity. Indian J. Physiol. Pharmacol. 39: 377-
38            382.
39     Sauk, J. J.; Smith, T.; Silbergeld, E. K.; Fowler, B. A.; Somerman, M. J. (1992) Lead inhibits secretion of
40            osteonectin/SPARC without significantly altering collagen or Hsp47 production in osteoblast-like ROS
41            17/2.8 cells. Toxicol. Appl. Pharmacol. 116: 240-247.
42     Saxena, D. K.; Lai, B.;  Srivastava, R. S.;  Chandra, S. V. (1990) Lead induced testicular hypersensitivity in stressed
43            rats. Exp. Pathol. 39: 103-109.
44     Schafer, J. H.; Glass, T. A.; Bressler, J.; Todd, A. C.; Schwartz, B. S. (2005) Blood lead in a  predictor of
45            homocysteine levels in a population-based study of older adults. Environ. Health Perspect. 113:31-35.
46     Schanne, F. A. X.; Dowd, T. L.; Gupta, R. K.; Rosen, J. F. (1989) Lead increases free  Ca2+ concentration in
47            cultured osteoblastic bone cells: simultaneous detection of intracellular free Pb2+ by 19F NMR. Proc. Natl.
48            Acad. Sci.U. S. A. 86:5133-5135.
49     Schanne, F. A. X.; Gupta, R. K.; Rosen, J. F. (1992) Lead inhibits 1,25-dihydroxyvitamin D-3 regulation of calcium
50            metabolism in  osteoblastic osteosarcoma cells (ROS 17/2.8). Biochim. Biophys. Acta 1180:  187-194.
51     Schanne, F. A. X.; Long, G. J.; Rosen, J. F. (1997) Lead induced rise in intracellular free calcium is mediated
52            through activation of protein kinase C in osteoblastic bone cells. Biochim. Biophys.  Acta 1360: 247-254.
53     Schechtman, L. M.; Hatch, G. G.; Anderson, T. M.; Putman, D. L.; Kouri, R. E.; Cameron, J. W.; Nims, R. W.;
54            Spalding, J. W.; Tennant, R. W.; Lubet, R. A.  (1986) Analysis of the interlaboratory  and intralaboratory
5 5            reproducibility of the enhancement of simian adenovirus S A7 transformation of Syrian hamster embryo
56            cells by model  carcinogenic and noncarcinogenic compounds. Environ. Mutagen. 8:  495-514.Shelkovnikov,

       December 2005                                 5-329        DRAFT-DO NOT QUOTE OR CITE

-------
  1            S. A.; Gonick, H. C. (2001) Influence of lead on rat thoracic aorta contraction and relaxation. Am. J.
  2            Hypertens. 14: 873-878.
  3     Scheuhammer, A. M. (1987) Erythrocyte "delta"-aminolevulinic acid dehydratase in birds. II. The effects of lead
  4            exposure in vivo. Toxicology 45: 165-175.
  5     Schirrmacher, K.; Wiemann, M.; Bingmann, D.; Busselberg, D. (1998) Effects of lead, mercury, and methyl
  6            mercury on gap junctions and [CA2+]i in bone cells. Calcified Tiss. Int. 63: 134-139.
  7     Schlick, E.; Friedberg, K. D. (1981) The influence of low lead doses on the reticulo-endothelial system and
  8            leucocytes of mice. Arch. Toxicol.  47: 197-207.
  9     Schlipkoter, H.-W.; Frieler, L. (1979) Der Einfluss kurzzeitiger Bleiexposition auf die Bakterienclearance der Lunge
10            [The influence of short-term lead exposure on the bacterial clearance of the lung].  Zentralbl. Bakteriol.
11            Parasitenkd. Infektionskr. Hyg. Abt. 1: Orig. Reihe B  168: 256-265.
12     Schneider, J. S.; Anderson, D. W.; Wade, T. V.; Smith, M. G.; Leibrandt, P.; Zuck, L.; Lidsky, T. I. (2005)
13            Inhibition of progenitor cell proliferation in the dentate gyrus of rats following post-weaning lead exposure.
14            Neurotoxicology 26:  141-145.
15     Schneyer, C. A.; Hall, H. D. (1970) Influence of physiological activity on mitosis in immature rat parotid gland.
16            Proc. Soc. Exp. Biol.  Med. 133: 349-352.
17     Schrauzer, G. N. (1987) Effects of selenium antagonists on cancer susceptibility: new aspects of chronic heavy
18            metal toxicity. J. UOEH 9 Suppl: 208-215.
19     Schroeder, H. A.; Mitchener, M. (1971) Toxic effects of trace elements on the reproduction of mice and rats. Arch.
20            Environ. Health 23: 102-106.
21     Schwartz, J. (1994) Low-level lead exposure and children's IQ: a meta-analysis and search for a threshold. Environ.
22            Res. 65: 42-55.
23     Schwartz, J.; Angle, C.; Pitcher, H. (1986) Relationship between childhood blood lead  and stature. Pediatrics  77:
24            281-288.
25     Schwartz, B. S.; Lee, B.-K.; Stewart, W.; Ann, K.-D.; Kelsey,  K.; Bresssler, J. (1997) Associations of subtypes of
26            hemoglobin with delta-aminolevulinic acid dehydratase genotype and dimercaptosuccinic acid-chelatable
27            lead levels. Arch. Environ. Health 52: 97-103.
28     Schwartz, B. S.; Stewart, W. F.; Bolla, K. I.; Simon, M. S.; Bandeen-Roche, K.; Gordon, B.; Links, J. M.; Todd, A.
29            C. (2000) Past adult lead exposure is associated with longitudinal decline in cognitive function. Neurology
30            55: 1144-1150.
31     Selevan, S.  G.; Kimmel, C. A.; Mendola, P. (2000) Identifying critical windows of exposure for children's health.
32            Environ. Health Perspect. Suppl. 108(3): 451-455.
33     Selevan, S.  G.; Rice, D. C.; Hogan, K. A.; Euling, S. Y.; Pfahles-Hutchens, A.; Bethel, J. (2003) Blood lead
34            concentration and delayed puberty in girls. N.  Engl. J. Med. 348: 1527-1536.
3 5     Selye, H.; Tuchweber, B.; Bertok, L. (1966) Effect of lead acetate on the susceptibility of rats to bacterial
36            endotoxins. J. Bacteriol. 91: 884-890.
37     Sengupta, M.; Bishayi, B. (2002) Effect of lead and  arsenic on murine macrophage response. Drug Chem. Toxicol.
38            25:459-472.
39     Serrani, R. E.; Gioia, I. A.; Corchs, J. L. (1997) Lead effects on structural and functional cellular parameters in
40            human red cells from a prenatal hematopoiesis stage. Biometals 10:331-335.
41     Shabani, A.; Rabbani, A. (2000) Lead nitrate induced apoptosis in alveolar macrophages from rat lung. Toxicology
42            149: 109-114.
43     Shakoor,  A.; Gupta, P. K.; Singh, Y. P.; Kataria, M. (2000) Beneficial effects of aluminum on the progression of
44            lead-induced nephropathy in rats. Pharmacol. Toxicol. 87: 258-260.
45     Shalan, M. G.; Mostafa, M. S.; Hassouna, M. M.; El-Nabi, S. E.; El-Refaie, A. (2005) Amelioration of lead toxicity
46            on rat liver with vitamin C and  silymarin supplements. Toxicology 206: 1-15.
47     Shao, Z.;  Suszkiw, J. B. (1991) Ca2+-surrogate action of Pb2+ on acetylcholine release from rat brain
48            synaptosomes. J. Neurochem. 56: 568-574.
49     Sharifi, A. M.; Darabi, R.; Akbarloo, N.; Larijani, B.; Khoshbaten, A. (2004) Investigation of circulatory and  tissue
50            ACE activity during development of lead-induced hypertension. Toxicol. Lett. 153: 233-238.
51     Shelkovnikov, S. A.; Gonick, H. C. (2001) Influence of lead on rat thoracic aorta contraction and relaxation. Am. J.
52            Hypertens. 14: 873-878.
53     Shelton, K.  R.; Egle, P. M. (1982) The proteins of lead-induced intranuclear inclusion bodies. J. Biol. Chem. 257:
54            11802-11807.
       December 2005                                 5-330        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Shelton, K. R.; Cunningham, J. G.; Klann, E.; Merchant, R. E.; Egle, P. M.; Bigbee, J. W. (1990) Low-abundance
  2            32-kilodalton nuclear protein specifically enriched in the central nervous system. J. Neurosci. Res. 25: 287-
  3            294.
  4     Shelton, K. R.; Egle, P. M.; Bigbee, J. W.; Klann, E. (1993) A nuclear matrix protein stabilized by lead exposure:
  5            current knowledge and future prospects. Presented at: Ninth international neurotoxicology conference;
  6            October 1991; Little Rock, AR. Neurotoxicology 14(2-3): 61-67.
  7     Shinozuka, H.; Kubo, Y.; Katyal, S. L.; Coni, P.; Ledda-Columbano, G. M.; Columbano, A.; Nakamura, T. (1994)
  8            Roles of growth factors and of tumor necrosis factor-"alpha" on liver cell proliferation induced in rats by
  9            lead nitrate. Lab. Invest. 71:  35-41.
10     Shinozuka, H.; Ohmura, T.; Katyal, S. L.; Zedda, A. I.; Ledda-Columbano, G. M.; Columbano, A. (1996) Possible
11            roles of nonparenchymal cells in hepatocyte proliferation induced by lead nitrate and by  tumor necrosis
12            factor "alpha". Hepatology 23: 1572-1577.
13     Shraideh, Z. (1999) Effect of triethyl lead on peristaltic contractile activity of the ileumof mice.  Cytobios 99: 97-
14            104.
15     Shukla, V. K.; Prakash, A.;  Tripathi, B. D.; Reddy, D. C.; Singh, S. (1998) Biliary heavy metal concentrations in
16            carcinoma of the gallbladder: case-control study. BMJ 317: 1288-1289.
17     Sidhu, M. K.; Fernandez, C.; Khan, M. Y.; Kumar, S. (1991) Induction of morphological transformation, anchorage-
18            independent growth and plasminogen activators in non-tumorigenic human osteosarcoma cells by lead
19            chromate. AnticancerRes. 11:  1045-1053.
20     Sieg, D. J.; Billings, R. E. (1997) Lead/cytokine-mediated oxidative DNA damage in cultured mouse hepatocytes.
21            Toxicol. Appl. Pharmacol. 142: 106-115.
22     Sierra, E. M.; Tiffany-Castiglioni, E. (1992) Effects of low-level lead exposure on hypothalamic hormones and
23            serum progesterone levels in pregnant guinea pigs. Toxicology 72: 89-97.
24     Silbergeld, E. K.; Hruska, R. E.; Bradley, D. (1982) Neurotoxic aspects of porphyrinopathies: lead and
25            succinylacetone. Environ. Res. 29: 459-471.
26     Silkin, Y. A.; Silkina, E. N.; Sherstobitov, A.  O.; Gusev, G. P. (2001) Activation of potassium channels in
27            erythrocytes of marine teleost Scorpaena porcus. Membr. Cell Biol. 14: 773-782.
28     Simons, T. J. (1986a) Passive transport and binding of lead by human red blood cells. J. Physiol. 378: 267-286.
29     Simons, T. J. (1986b) The role of anion transport in the passive movement of lead across the human red cell
30            membrane. J. Physiol. 378: 287-312.
31     Simons, T. J. B.  (1988) Active transport of lead by the calcium pump in human red cell ghosts. J. Physiol. (London)
32            405: 105-13.
33     Simons, T. J. B.  (1993a) Lead transport and binding by human erythrocytes in vitro. Pflugers Arch. 423: 307-313.
34     Simons, T. J. B.  (1993b) Lead-calcium interactions in cellular lead toxicity. Presented at: Ninth international
35            neurotoxicology conference; October 1991; Little Rock,  AR. Neurotoxicology 14(2-3): 77-86.
36     Simons, T. J. B.  (1995) The affinity of human erythrocyte porphobilinogen synthase forZn2+ and Pb2+. Eur. J.
37            Biochem. 234:  178-183.
38     Singh, U. S.; Saxena, O.K.; Singh, C.; Murthy, R. C.; Chandra, S. V. (1991) Lead-induced fetal nephrotoxicity in
39            iron-deficient rats. Reprod. Toxicol. 5: 211-217.
40     Singh, A.; Cullen, C.; Dykeman, A.;  Rice, D.; Foster, W. (1993)  Chronic lead exposure induces ultrastructural
41            alterations in the monkey testis. J. Submicrosc. Cytol. Pathol. 25: 479-486.
42     Singh, C.; Saxena, D. K.; Murthy, R. C.; Chandra, S. V. (1993) Embryo-fetal development influenced by lead
43            exposure in iron-deficient rats. Hum.  Exp. Toxicol. 12: 25-28.
44     Singh, J.; Parkash, P.; Gupta, G. S. (1999) State of pregnancy modifies lead toxicity in mice. Biol. Trace Elem. Res.
45            67:205-213.
46     Singh, J.; Pritchard, D. E.; Carlisle, D. L.; Mclean, J. A.; Montaser, A.; Orenstein, J. M.; Patierno, S. R. (1999)
47            Internalization of carcinogenic lead chromate particles by cultured normal human lung epithelial cells:
48            formation of intracellular lead-inclusion bodies and induction of apoptosis. Toxicol. Appl. Pharmacol. 161:
49            240-248.
50     Singh, V. K.; Mishra, K. P.; Rani, R.; Yadav, V. S.; Awasthi, S. K.; Garg, S. K. (2003) Immunomodulationby lead.
51            Immunol. Res. 28:  151-165.
52     Silvaprasad, R.; Nagaraj, M.; Varalakshmi, P. (2003) Combined efficacies of lipoic acid and meso-2,3-
53            dimercaptosuccinic acid on lead-induced erythrocyte membrane lipid peroxidation and antioxidant status in
54            rats. Hum. Exp. Toxicol. 22: 183-192.
55     Sivaprasad, R.; Nagaraj, M.; Varalakshmi, P.  (2004) Combined efficacies of lipoic acid and 2,3-dimercaptosuccinic
56            acid against lead-induced lipid peroxidation in rat liver. J. Nutr. Biochem. 15: 18-23

       December 2005                                5-331         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Skoczynska, A.; Smolik, R. (1994) The effect of combined exposure to lead and cadmium on serum lipids and lipid
  2            peroxides level in rats. Int. J. Occup. Med. Environ. Health 7: 263-271.
  3     Skoczynska, A.; Smolik, R.; Men, M. (1993) Lipid abnormalities in rats given small doses of lead. Arch. Toxicol.
  4            67:200-204.
  5     Skoczynska, A.; Smolik, R.; Milian, A. (1994) The effect of combined exposure to lead and cadmium on the
  6            concentration of zinc and copper in rat tissues. Int. J. Occup. Med. Environ. Health 7: 41-49.
  7     Slobozhanina, E. I.; Kozlova, N. M.; Lukyanenko, L. M.; Oleksiuk, O. B.; Gabbianelli, R.; Fedeli, D.; Caulini, G.
  8            C.; Falcioni, G. (2005) Lead-induced changes in human erythrocytes and lymphocytes. J. Appl. Toxicol. 25:
  9            109-114.
10     Smejkalova, J. (1990) The  chromosomal aberrations investigation in children permanently living in the lead polluted
11            area. Sb. Ved. Pr. Lek. Fak. Karlovy Univerzity Hradci Kralove 33: 539-564.
12     Smith, D. R.; Flegal, A. R.  (1992) Stable isotopic tracers of lead mobilized by DMS A chelation in low lead-exposed
13            rats. Toxicol. Appl. Pharmacol. 116: 85-91.
14     Smith, K. L.; Lawrence, D. A. (1988) Immunomodulation of in vitro antigen presentation by cations. Toxicol. Appl.
15            Pharmacol. 96: 476-484.
16     Smith, C.  M.; DeLuca, H. F.; Tanaka, Y.; Mahaffey, K. R. (1981) Effect of lead ingestion on functions of vitamin D
17            and its metabolites. J. Nutr. Ill: 1321-1329.
18     Smith, C.  M.; Hu, H.; Wang, X.; Kelsey, K. T. (1995a)  ALA-D genotype is not associated with HT or HB levels
19            among workers exposed to low levels of lead. Med. Lav. 86: 229-235.
20     Smith, C.  M.; Wang, X.; Hu, H.; Kelsey, K. T. (1995b) A polymorphism in the "delta"-aminolevulinic acid
21            dehydratase gene may modify the pharmacokinetics and toxicity of lead. Environ. Health Perspect. 103:
22            248-253.
23     Smith, D. R.; Kahng, M. W.; Quintanilla-Vega, B.; Fowler, B. A. (1998) High-affinity renal lead-binding proteins
24            ini environmentally-exposed humans. Chem. Biol. Interact. 115: 39-52.
25     Smith, D. R.; Woolard, D.; Luck, M. L.; Laughlin, N. K. (2000) Succimer and the reduction of tissue lead in
26            juvenile monkeys.  Toxicol. Appl. Pharmacol. 166: 230-240.
27     Smith, D.; Hernandez-Avila, M.; Tellez-Rojo, M.M.; Mercado, A.; Hu, H. (2002) The relationship between lead in
28            plasma and whole blood in women. Environ. Health Perspect. 110: 263-268.
29     Snyder, R. D.; Lachmann, P. J. (1989) Thiol involvement in the inhibition of DNA repair by metals in mammalian
30            cells. J. Mol. Toxicol. 2: 117-128.
31     Snyder, J. E.; Filipov, N. M.; Parsons, P. J.; Lawrence, D. A. (2000) The efficiency of maternal transfer of lead and
32            its influence on plasma IgE and splenic cellularity of mice. Toxicol. Sci. 57:  87-94.
33     Sokol, R.  Z. (1987) Hormonal effects of lead acetate in the male rat: mechanism of action. Biol. Reprod. 37: 1135-
34            1138.
35     Sokol, R.  Z.; Berman, N. (1991) The effect of age of exposure on lead-induced testicular toxicity. Toxicology 69:
36            269-278.
37     Sokol, R.  Z.; Berman, N.; Okuda, H.; Raum, W. (1998) Effects of lead exposure on GnRH and LH secretion in male
38            rats: response to castration and "alpha"-methyl-p-tyrosine (AMPT) challenge. Reprod.  Toxicol. 12: 347-
39            355.
40     Sokol, R.  Z.; Madding, C. E.; Swerdloff, R. S. (1985) Lead toxicity and the hypothalamic-pituitary-testicular axis.
41            Biol. Reprod. 33:722-728.
42     Sokol, R.  Z.; Okuda, H.; Nagler, H. M.; Berman, N. (1994) Lead exposure in vivo alters the fertility potential of
43            sperm in vitro. Toxicol. Appl. Pharmacol. 124:  310-316.
44     Sokol, R.  Z.; Wang, S.; Wan, Y.-J. Y.; Stanczyk, F. Z.;  Gentzschein, E.; Chapin, R. E. (2002) Long-term, low-dose
45            lead exposure alters the gonadotropin-releasing hormone system int he male  rat. Environ. Health Perspect.
46            110:871-874.
47     Spit, B. J.; Wibowo, A. A.  E.; Feron, V. J.; Zielhuis, R. L. (1981) Ultrastructural changes in the kidneys of rabbits
48            treated with lead acetate. Arch. Toxicol. 49: 85-91.
49     Srivastava, D.; Hurwitz, R. L.; Fox, D. A. (1995)  Lead- and calcium-mediated inhibition of bovine rod cGMP
50            phosphodiesterase: interactions with magnesium. Toxicol. Appl. Pharmacol.  134: 43-52.
51     Srivastava, V.; Dearth, R. K.; Hiney, J. K.; Ramirez, L. M.; Bratton, G. R.; Dees, W. (2004) The effects of low-level
52            Pb on steroidogenic acute regulatory protein (StAR) in the prepubertal rat ovary. Toxicol. Sci. 77: 35-40.
53     Steenland, K.; Selevan, S.;  Landrigan, P. (1992) The mortality of lead smelter workers: an update. Am. J. Public
54            Health 82: 1641-1644.
55     Stiles, K. M.; Bellinger, D. C. (1993) Neuropsychological correlates of low-level lead exposure in school-age
56            children: a prospective study. Neurotoxicol. Teratol. 15: 27-35.

       December 2005                                5-332         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Stokes, I; Casale, T. B. (2004) Rationale for new treatments aimed at IgE immunomodulation. Ann. Allergy
  2            Asthma Immunol. 93: 111-111.
  3     Stokes, L.; Letz, R.; Gerr, F.; Kolczak, M; McNeill, F. E.; Chettle, D. R.; Kaye, W. E. (1998) Neurotoxicity in
  4            young adults 20 years after childhood exposure to lead: the Bunker Hill experience. Occup. Environ. Med.
  5            55:507-516.
  6     Stowe, H. D.; Goyer, R. A. (1971) Reproductive ability and progeny of F 1 lead-toxic rats. Fertil. Steril. 22: 755-
  7            760.
  8     Studnitz, W. von; Haeger-Aronsen, B. (1962) Urinary excretion of amino acids in lead-poisoned rabbits. Acta
  9            Pharmacol. Toxicol.  19: 36-42.
10     Sugawara, E.; Nakamura, K.; Fukumura, A.;  Seki, Y. (1990) Uptake of lead by human red blood cells and
11            intracellular distribution. Kitasato Arch. Exp. Med. 63: 15-23.
12     Sugiura, S.; Dhar, S. K.; Arizono, K.; Ariyoshi, T. (1993) Induction of DT-diaphorase in the liver of rats treated
13            with various metals. Jpn. J. Toxicol. Environ. Health 39: P-7.
14     Suketa, Y.; Hasegawa, S.; Yamamoto, T. (1979) Changes in sodium and potassium in urine and serum of lead-
15            intoxicated rats. Toxicol. Appl. Pharmacol.  47: 203-207.
16     Sun, X.; Tian, X.; Tomsig, J. L.; Suszkiw, J. B. (1999) Analysis of differential effects of Pb2+ on protein kinase C
17            isozymes. Toxicol. Appl. Pharmacol. 156: 40-45.
18     Sun, L.; Hu, J.; Zhao, Z.; Li, L.;  Cheng, H. (2003) Influence of exposure to environmental lead on serum
19            immunoglobulin in preschool children. Environ. Res. 92: 124-128.
20     Suszkiw, J.; Toth, G.; Murawsky, M.; Cooper, G. P. (1984) Effects of Pb2+ and Cd2+ on acetylcholine release and
21            Ca2+ movements in synaptosomes and subcellular fractions from rat brain and Torpedo electric organ.
22            Brain Res. 323:31-46.
23     Suwalsky, M.; Villena, F.; Norris, B.; Cuevas, F.; Sotomayor, C. P.; Zatta, P. (2003) Effects of lead on the human
24            erythrocyte membrane and molecular models. J. Inorg. Biochem. 97: 308-313.
25     Suzuki, T.; Morimura, S.; Diccianni, M. B.; Yamada, R.; Hochi, S.-L; Hirabayashi, M.; Yuki, A.; Nomura. K.;
26            Kitagawa,  T.; Imagawa, M.; Muramatsu, M. (1996) Activation of glutathione transferase P gene by lead
27            requires glutathione transferase P enhancer  I. J. Biol. Chem. 271: 1626-1632.
28     Szabo, A.; Merke, J.; Hugel, U.; Mall, G.; Stoeppler, M.; Ritz, E. (1991) Hyperparathyroidism and abnormal
29            l,25(OH)2vitamin D3 metabolism in experimental lead intoxication. Eur. J. Clin. Invest. 21:512-520.
30     Tabchoury, C. M.;  Pearson, S. K.; Bowen, W. H. (1999) Influence of lead on the cariostatic effect of fluoride co-
31            crystallized with sucrose in desalivated rats. Oral Dis. 5: 100-103.
32     Takeno, M.; Yoshikawa, H.; Kurokawa, M.; Takeba, Y.; Kashoiwakura, J. L; Sakaguchi, M.; Yasueda, H.; Suzuki,
33            N. (2004) Thl -dominant shift of T cell cytokine production and subsequent reduction of serum
3 4            immunoglobulin E response by administration in vivo of plasmid expressing Txk/Rlk, a member of Tec
35            family tyrosine kinases, in a mouse model. Clin. Exp. Immunol. 34: 965-970.
36     Taketani, S.; Tanaka, A.; Tokunaga, R. (1985) Reconstitution of heme-synthesizing activity from ferric ion and
37            porphyrins, and the effect of lead on the activity. Arch. Biochem. Biophys. 242: 291-296.
38     Tandon, S. K.; Singh, S.; Jain, V. K. (1994a)  Efficacy of combined chelation in lead intoxication. Chem. Res.
39            Toxicol. 7: 585-589.
40     Tandon, S. K.; Khandelwal, S.; Jain, V. K.; Mathur, N. (1994b) Influence of dietary iron deficiency on nickel, lead
41            and cadmium intoxication. Sci. Total Environ.  148: 167-173.
42     Tandon, S. K.; Singh, S.; Prasad, S.; Mathur,  N. (1997) Influence of L-lysine and zinc administration during
43            exposure to lead or lead and ethanol in rats. Biol. Trace Elem. Res. 57: 51-58.
44     Tang, H.-W.; Huel, G.; Campagna, D.; Hellier, G.; Boissinot, C.; Blot, P. (1999) Neurodevelopmental evaluation of
45            9-month-old infants exposed to low levels of lead in utero: involvement of monoamine neurotransmitters. J.
46            Appl. Toxicol. 19: 167-172.
47     Taupeau, C.; Poupon, J.; Nome,  F.; Lefevre, B. (2001) Lead accumulation in the mouse ovary after treatment-
48            induced follicular atresia. Reprod. Toxicol.  15: 385-391.
49     Tavakoli-Nezhad, M.; Pitts, D. K. (2005) Postnatal inorganic lead exposure reduces midbrain dopaminergic impulse
50            flow and decreases dopamine Dl receptor sensitivity in nucleus accumbens neurons. J. Pharmacol. Exp.
51            Ther. 312: 1280-1288.
52     Tavakoli-Nezhad, M.; Barren, A. J.; Pitts, D. K. (2001) Postnatal inorganic lead exposure decreases the number of
53            spontaneously active midbrain dopamine neurons in the rat. Neurotoxicology 22: 259-269.
54     Taylor, S. A.; Olivers, I. D.; Price, R. G.; Arce-Thomas, M.; Milligan, P.; Francini, I.; Alinovi, R.; Cavazzini, S.;
55            Bergamaschi, E.; Vittori, M.; Mutti, A.; Lauwerys, R. R.; Bernard, A. M.; Roels, H. A.; De Broe, M. E.;
56            Nuyts, G. D.; Elseviers, M. M.; Hotter, G.; Ramis, L; Rosello, J.; Gelpi, E.; Stolte, H.; Eisenberger, U.; Pels,

       December 2005                                5-333        DRAFT-DO NOT QUOTE OR CITE

-------
  1            L. M. (1997) The assessment of biomarkers to detect nephrotoxicity using an integrated database. Environ.
  2            Res. 75: 23-33.
  3     Tchernitchin, N. N.; Tchernitchin, A. N.; Mena, M. A.; Villarroel, L.; Guzman, C.; Poloni, P. (1998a) Effect of
  4            subacute exposure to lead on responses to estrogen in the immature rat uterus. Bull. Environ. Contam.
  5            Toxicol. 60: 759-765.
  6     Tchernitchin, N. N.; Villagra, A.; Tchernitchin, A. N. (1998b) Antiestrogenic activity of lead. Environ. Toxicol.
  7            Water Qual. 13:43-53.
  8     Tepper, R. L; Levinson, D. A.; Stanger, B. Z.; Campos-Torres, I; Abbas, A. K.; Leder, P. (1990) IL-4 induces
  9            allergic-like inflammatory disease and alters T cell development in transgenic mice. Cell 62: 457-467.
10     Teraki, Y.; Uchiumi, A. (1990) Inorganic elements in the tooth and bone tissues of rats bearing nickel acetate- and
11            lead acetate-induced tumors. Shigaku. 78: 269-273.
12     Terayama, K.; Muratsugu, M. (1988) Effects of lead on sialic acid content and survival of rat erythrocytes.
13            Toxicology 53: 269-276.
14     Terayama, K.; Maehara, N.; Muratsugu, M.; Makino, M.; Yamamura, K. (1986) Effect of lead on electrophoretic
15            mobility of rat erythrocytes. Toxicology 40: 259-265.
16     Tessitore, L.; Perletti,  G. P.; Sesca, E.;  Pani, P.; Dianzani, M. U.; Piccinini, F. (1994) Proteinkinase C isozyme
17            pattern in liver hyperplasia. Biochem. Biophys. Res. Commun. 205: 208-214.
18     Tessitore, L.; Sesca, E.; Pani, P.; Dianzani, M. U. (1995) Sexual dimorphism in the regulation of cell turnover
19            during liver hyperplasia. Chem. Biol. Interact. 97: 1-10.
20     Thaweboon, S.; Chunhabundit, P.; Surarit, R.; Swasdison, S.; Suppukpatana, P. (2002) Effects of lead on the
21            proliferation, protein production, and osteocalcin secretion of human dental pulp cells in vitro. Southeast
22            Asian J. Trap. Med. Public Health 33: 654-661.
23     Thind, I. S.; Khan, M. Y. (1978) Potentiation of the neurovirulence of Langat virus infection by lead intoxication in
24            mice. Exp. Mol. Pathol. 29: 342-347.
25     Tian, L.; Lawrence, D. A. (1995) Lead inhibits nitric oxide production in vitro by murine splenic macrophages.
26            Toxicol. Appl. Pharmacol. 132: 156-163.
27     Tian, L.; Lawrence, D. A. (1996) Metal-induced modulation of nitric oxide production in vitro by murine
28            macrophages: Lead, nickel, and cobalt utilize different mechanisms. Toxicol. Appl. Pharmacol. 141: 540-
29            547.
30     Tian, X.; Sun, X.; Suszkiw, J. B. (2000) Upregulation of tyrosine hydroxylase and downregulation of choline
31            acetyltransferase in lead-exposed PC12 cells: the role of PKC activation. Toxicol. Appl. Pharmacol.  167:
32            246-252.
33     Tomczok, J.; Grzybek, H.; Sliwa, W.; Panz, B. (1988) Ultrastructural aspects of the small intestinal lead toxicology.
34            Part II. The small intestine goblet cells of rats during lead poisoning. Exp. Pathol. 35: 93-100.
35     Tomokuni, K.; Ichiba, M. (1988) Comparison of inhibition of erythrocyte pyrimidine 5'-nucleotidase and delta-
36            aminolevulinic acid dehydratase by lead. Toxicol. Lett.  40: 159-163.
37     Tomokuni, K.; Ichiba, M.; Hirai, Y. (1989) Effect of lead exposure on some biological indices related to porphyrin
3 8            metabolism and the activity of erythrocyte pyrimidine 5'-nucleotidase in the mice. Arch. Toxicol. 63: 23-28.
3 9     Tomokuni, K.; Ichiba, M. (1990) Effect of lead on the activity of erythrocyte porphobilinogen deaminase in-vivo
40            and in-vitro. Toxicol. Lett. 50:  137-142.
41     Tomokuni, K.; Ichiba, M.; Hirai, Y. (1991) Elevated urinary excretion of "beta"-aminoisobutyric acid and "delta"-
42            aminolevulinic acid (ALA) and the inhibition of ALA-synthase and ALA-dehydratase activities in both liver
43            and kidney in mice exposed to lead. Toxicol. Lett. 59: 169-173.
44     Tomsig, J. L.; Suszkiw, J. B. (1993) Intracellular mechanism of Pb2+-induced norepinephrine release from bovine
45            chromaffin cells. Am. J. Physiol. 265: C1630-C1636.
46     Tomsig, J. L.; Suszkiw, J. B. (1995) Multisite interactions between Pb2+ and protein kinase C and its role in
47            norepinephrine release from bovine adrenal chromaffin cells. J. Neurochem. 64: 2667-2673.
48     Tong, S.; Baghurst, P.; McMichael, A.; Sawyer, M.; Mudge, J. (1996) Lifetime exposure to environmental lead and
49            children's intelligence at 11-13 years: the Port Pirie cohort study. Br. Med. J. 312: 1569-1575.
50     Tong, S.; McMichael, A. J.; Baghurst, P. A. (2000) Interactions between environmental lead exposure and
51            sociodemographic factors on cognitive development. Arch. Environ. Health 55: 330-335.
52     Tonner, L. E.; Heiman, A. S. (1997) Lead may affect glucocorticoid signal transduction in cultured hepatoma cells
53            through inhibition of protein kinase C. Toxicology 119: 155-166.
54     Toplan, S.; Ozcelik, D.; Gulyasar, T.; Akyoleu, M. C. (2004) Changes in hemorheological parameters due to lead
55            exposure in female rats. J. Trace Elem. Med. Biol.  18: 179-182.


       December 2005                                 5-334        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Torres, D.; Barrier, M; Bikl, F.; Quesniaux, V. J.; Maillet, I.; Akira, S.; Ryffel, B.; Erard, F. (2004) Toll-like
  2            receptor 2 is required for optimal control of Listeria monocytogenes infection. Infect. Immun. 72: 2131-
  3            2139.
  4     Toscano, C. D.; Hashemzadeh-Gargari, H.; McGlothan, J. L.; Guilarte, T. R. (2002) Developmental Pb2+ exposure
  5            alters NMDAR subtypes and reduces CREB phosphorylation in the rat brain. Dev. Brain Res. 139: 217-226.
  6     Trasande, L.; Thurston, G. D. (2005) The role of air pollution in asthma and other pediatric morbidities. J. Allergy
  7            Clin. Immunol. 115: 689-699.
  8     Trejo, R. A.; Di Luzio, N. R.; Loose, L. D.; Hoffman, E. (1972) Reticuloendothelial and hepatic functional
  9            alterations following lead acetate administration. Exp. Mol. Pathol. 17: 145-158.
10     Trope, L; Lopez-Villegas, D.; Cecil, K. M.; Lenkinski, R. E. (2001) Exposure to lead appears to selectively alter
11            metabolism of cortical gray matter. Pediatrics 107: 1437-1443.
12     Tryphonas, H. (2001) Approaches to detecting immunotoxic effects of environmental contaminants in humans.
13            Environ. Health Perspect. Suppl. 109(6): 877-884.
14     Tsaih, S.-W.; Korrick, S.; Schwartz, J.; Lee, M.-L. T.; Amarasiriwardena, C.; Aro, A.; Sparrow, D.; Hu, H. (2001)
15            Influence of bone resorption on the mobilization of lead from bone among middle-aged and elderly men: the
16            Normative Aging Study. Environ. Health Perspect. 109: 995-999.
17     Tsao, D.-A.; Yu, H.-S.; Cheng,  J.-T.; Ho, C.-K.; Chang, H.-R. (2000) The change of "beta"-adrenergic system in
18            lead-induced hypertension. Toxicol. Appl. Pharmacol. 164: 127-133.
19     Tulasi, S. J.; Reddy, P. U. M.; RamanaRao, J. V. (1992) Accumulation of lead and effects on total lipids andlipid
20            derivatives in the freshwater fish Anabas testudineus (Bloch).  Ecotoxicol. Environ. Saf. 23: 33-38.
21     U.S. Environmental Protection Agency. (1986) Air quality criteria for lead. Research Triangle Park, NC: Office of
22            Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
23            EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.
24     U.S. Environmental Protection Agency. (1990) Air quality criteria for lead: supplement to the  1986 addendum.
25            Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
26            Assessment Office; report no. EPA/600/8-89/049F. Available from: NTIS, Springfield, VA; PB91-138420.
27     Ueda, D.;Kishimoto, T.; Dekio, S.; Tada, M. (1997) Inhibitory effect of lead on tube formation by cultured human
28            vascular endothelial cells. Hum. Cell 10: 283-291.
29     Undeger, U.; Basaran, N.; Canpinar, H.; Kansu, E. (1996) Immune alterations in lead-exposed workers. Toxicology
30            109: 167-172.
31     Vaglenov, A.; Carbonell, E.; Marcos, R. (1998) Biomonitoring of workers exposed to lead. Genotoxic effects, its
32            modulation by polyvitamin treatment and evaluation of the induced radioresistance. Mutat. Res. 418: 79-92.
33     Vakharia, D. D.; Liu, N.; Pause, R.; Fasco, M.;  Bessette, E.; Zhang, Q.-Y.; Kaminsky, L. S. (2001) Effect of metals
34            on polycyclic aromatic  hydrocarbon induction of CYP1 Al and CYP1A2 in human hepatocyte cultures.
35            Toxicol. Appl. Pharmacol. 170: 93-103.
36     Valencia, I.; Castillo, E. E.; Chamorro, G.; Bobadilla, R. A.; Castillo, C. (2001) Lead induces endothelium- and
37            Ca2+-independent contraction in rat aortic rings. Pharmacol. Toxicol. (Oxford, UK) 89: 177-182.
38     Valentino, M.; Governa, M.; Marchiseppe, I.; Visona, I. (1991) Effects of lead on polymorphonuclear leukocyte
39            (PMN) functions in occupationally exposed workers. Arch. Toxicol. 65: 685-688.
40     Valverde, M.; Fortoul, T. I.; Diaz-Barriga, F.; Majia, J.; Del Castillo, E. R. (2002) Genotoxicity induced in CD-I
41            mice by  inhaled lead: differential organ response. Mutagenesis 17: 55-61.
42     Van Larebeke, N.; Koppen, G.;  Nelen, V.; Schoeters, G.; Van Loon H, Albering H, Riga L, Vlietinck, R.; Kleinjans,
43            J.; Flemish Environment and Health Study Group. (2004) Differences in HPRT mutant frequency among
44            middle-aged Flemish women in association with area of residence and blood lead levels. Biomarkers 9:71-
45            84.
46     Vander, A. J. (1988) Chronic effects of lead on the renin-angiotensin system. In: Victery, W., ed. Symposium on
47            lead-blood pressure relationships; April 1987; Chapel Hill, NC. Environ. Health Perspect. 78: 77-83.
48     Vander, A. J.; Taylor, D. L.; Kalitis, K.; Mouw, D. R.; Victery, W. (1977) Renal handling of lead in dogs: clearance
49            studies. Am. J. Physiol. 2: F532-F538.
50     Varnai, V. M.; Piasek, M.; Blanusa, M.; Juresa, D.; Saric, M.; Kostial, K. (2003) Ascorbic acid supplementation
51            does not improve efficacy of meso-dimercaptosuccinic acid treatment in lead-exposed suckling rats.
52            Pharmacol. Toxicol. (Oxford, UK) 93:  180-185.
53     Varnai, V. M.; Piasek, M.; Blanusa, M.; Saric, M. M.; Kostial, K. (2001) Succimer treatment during ongoing lead
54            exposure reduces tissue lead in suckling rats. J. Appl. Toxicol. 21: 415-416.
55     Vaziri, N.; Ding, Y. (2001) Effect of lead on nitric oxide synthase expression in coronary endothelial cells: role of
56            superoxide. Hypertension 37: 223-226.

       December 2005                                5-335         DRAFT-DO NOT  QUOTE OR CITE

-------
  1     Vaziri, N. D.; Wang, X. Q. (1999) cGMP-mediated negative-feedback regulation of endothelial nitric oxide synthase
  2            expression by nitric oxide. Hypertension 34: 1237-1241.
  3     Vaziri, N. D.; Ding, Y.; Ni, Z.; Gonick, H. C. (1997) Altered nitric oxide metabolism and increased oxygen free
  4            radical activity in lead-induced hypertension: effect of lazaroid therapy. Kidney Int. 52: 1042-1046.
  5     Vaziri, N. D.; Ding, Y.; Ni, Z. (1999a) Nitric oxide synthase expression in the course of lead-induced hypertension.
  6            Hypertension 34: 558-562.
  7     Vaziri, N. D.; Liang, K.; Ding, Y. (1999b) Increased nitric oxide inactivationby reactive oxygen species in lead-
  8            induced hypertension. Kidney Int. 56:  1492-1498.
  9     Vaziri, N. D.; Wang, X. Q.; Oveisi, F.; Rad, B. (2000) Induction of oxidative stress by glutathione depletion causes
10            severe hypertension in normal rats. Hypertension 36: 142-146.
11     Vaziri, N. D.; Ding, Y.; Ni, Z. (2001) Compensatory up-regulation of nitric-oxide synthase isoforms in lead-induced
12            hypertension; reversal by a superoxide dismutase-mimetic drug. J. Pharmacol. Exp. Ther. 298: 679-685.
13     Vaziri, N. D.; Lin, C.-Y.; Farmand, F.; Sindhu, R. K. (2003) Superoxide dismutase, catalase, glutathione peroxidase
14            and NADPH oxidase in lead-induced hypertension.  Kidney Int. 63: 186-194.
15     Vaziri, N. D.; Ding, Y.; Ni, Z.; Barton, C. H. (2005) Bradykinin down-regulates, whereas arginine analogs up-
16            regulates, endothelial nitric-oxide synthase expression in coronary endothelial cells. J. Pharmacol. Exp.
17            Ther. 313: 121-126.
18     Vermande-van Eck, G. J.; Meigs, J. W. (1960) Changes in the ovary of the rhesus monkey after chronic lead
19            intoxication. Fertil. Steril. 11: 223-234.
20     Vicente, A.; Varas, A.; Acedon, R. S.; Jimenez, E.; Munoz,  J. J.; Zapata, A. G. (1998) Appearance and maturation
21            of T-cell subsets during rat thymus development. Dev. Comp. Immunol. 5:  319-331.
22     Victery, W.; Vander, A. J.; Mouw, D. R. (1979a) Effect of acid-base status on renal excretion and accumulation of
23            lead in dogs and rats. Am. J. Physiol. 237: F398-F407.
24     Victery, W.; Vander, A. J.; Mouw, D. R. (1979b) Renal handling of lead in dogs: stop-flow analysis. Am. J. Physiol.
25            237: F408-F414.
26     Victery, W.; Soifer, N. E.; Weiss, J. S.; Vander, A. J. (1981) Acute effects of lead on the renal handling of zinc in
27            dogs. Toxicol. Appl. Pharmacol. 61: 358-367.
28     Victery, W.; Vander, A. J.; Markel, H.; Katzman,  L.; Shulak, J. M; Germain, C. (1982a) Lead exposure, begun in
29            utero, decreases renin and angiotensin II in adult rats. Proc. Soc. Exp. Biol. Med. 170: 63-67.
30     Victery, W.; Vander, A. J.; Shulak, J. M.; Schoeps, P.; Julius, S. (1982b) Lead, hypertension, and the renin-
31            angiotensin system in rats. J. Lab. Clin. Med. 99: 354-362.
32     Victery, W.; Vander, A. J.; Schoeps, P.; Germain, C. (1983) Plasma renin is increased in young rats exposed to lead
33            in utero and during nursing. Proc. Soc. Exp. Biol. Med. 172:  1-7.
34     Vij, A. G.; Satija, N. K.; Flora, S. J. (1998) Lead induced disorders in hematopoietic and drug metabolizing enzyme
3 5            system and their protection by ascorbic acid supplementation. Biomed. Environ. Sci. 11: 7-14.
36     Villagra, R.; Tchernitchin, N. N.; Tchernitchin, A. N. (1997) Effect of subacute exposure to lead and estrogen on
37            immature pre-weaning rat leukocytes.  Bull. Environ. Contam. Toxicol. 58:  190-197.
38     Vyskocil, A.; Fiala, Z.; Salandova, J.; Popler, A.; Ettlerova,  E.; Emminger, S. (1991) The urinary excretion of
39            specific proteins in workers exposed to lead. Arch. Toxicol. Suppl. 14: 218-221.
40     Waalkes, M. P.; Liu, J.; Goyer, R. A.; Diwan, B. A. (2004) Metallothionein-I/II double knockout mice are
41            hypersensitive to lead-induced kidney carcinogenesis: role of inclusion body formation. Cancer Res. 64:
42            7766-7772.
43     Wagerova, M.; Wagner, V.; Madlo, Z.; Zavazal, V.; Wokounva, D.; Kriz, J.; Mohyla, O. (1986) Seasonal variation
44            in the levels of immunoglobulins and serum proteins of children differing by exposure to air-borne lead. J.
45            Hyg. Epidem. Microbiol. 30: 127-139.
46     Walkowiak, J.; Altmann, L.; Kramer, U.; Sveinsson, K.;  Turfeld, M.; Weishoff-Houben, M.; Winneke, G. (1998)
47            Cognitive and sensorimotor functions  in 6-year-old children in relation to lead and mercury levels:
48            adjustment for intelligence and contrast sensitivity in computerized testing. Neurotoxicol. Teratol. 20: 511-
49            521.
50     Wang, J.; Huff, A. M.; Spence, J. D.; Hegele, R. A. (2004) Single nucleotide polymorphism in CTH associated with
51            variation in plasma homocysteine concentration. Clin. Genet. 65: 483-486.
52     Wapnir, R. A.; Moak, S. A.; Lifshitz, F.; Teichberg, S. (1979) Alterations of intestinal and renal functions in rats
53            after intraperitoneal injections of lead  acetate.  J.  Lab. Clin. Med. 94: 144-151.
54     Wasserman, G. A.; Musabegovic, A.; Liu, X.;  Kline, J.; Factor-Litvak, P.; Graziano, J. H. (2000) Lead exposure and
55            motor functioning in 4 1/2-year-old children: the Yugoslavia prospective study. J. Pediatr. 137: 555-561.


       December 2005                                 5-336         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Waterman, S. I; El-Fawal, H. A. N.; Snyder, C. A. (1994) Lead alters the immunogenicity of two neural proteins: A
  2            potential mechanism for the progression of lead-induced neurotoxicity. Environ. Health Perspect. 102:
  3            1052-1056.
  4     Watson, G. E.; Davis, B. A.; Raubertas, R. F.; Pearson, S. K.; Bowen, W. H. (1997) Influence of maternal lead
  5            ingestion on caries in rat pups. Nature Med. 3: 1024-1025.
  6     Watts, S. W.; Chai, S.; Webb, R. C. (1995) Lead acetate-induced contraction in rabbit mesenteric artery: interaction
  7            with calcium and protein kinase C. Toxicology 99: 55-65.
  8     Webb, R. C.; Winquist, R. I; Victery, W.; Vander, A. J. (1981) In vivo and in vitro effects of lead on vascular
  9            reactivity in rats. Am. J. Physiol. 241: H211-H216.
10     Weiler, E.; Khalil-Manesh, F.; Gonick, H. C. (1990) Effects of lead and a low-molecular-weight endogenous plasma
11            inhibitor on the kinetics of sodium - potassium-activated adenosine triphosphatase and potassium-activated
12            para-nitrophenylphosphatase. Clin. Sci. 79: 185-192.
13     Weisberg, I. S.; Park, E.; Ballman, K. V.; Berger, P.; Nunn, M; Suh, D. S.; Breksa, A. P., Ill; Garrow, T. A.; Rozen,
14            R. (2003) Investigation of a common genetic variant in betaine-homocysteine methyltransferase (BHMT) in
15            coronary artery disease. Atherosclerosis 167: 205-214.
16     Weiss, B.; Landrigan, P. J. (2000) The developing brain and the environment:  an introduction. Environ. Health
17            Perspect. 108(Suppl. 3): 373-374.
18     Weisskopf, M. G.; Wright, R. O.; Schwartz,  J.; Spiro, A., Ill;  Sparrow, D.; Aro, A.; Hu, H. (2004) Cumulative lead
19            exposure and prospective change in cognition among elderly men. The VA Normative Aging Study. Am. J.
20            Epidemiol. 160: 1184-1193.
21     Weiler, C. V. (1915) The blastophthoric effect of chronic lead poisoning. J. Med. Res. 33: 271-293.
22     Westerink, R. H.; Vijverberg, H. P. (2002) Ca2+-independent vesicular catecholamine release in PC12 cells by
23            nanomolar concentrations of Pb2+. J. Neurochem. 80: 861-867.
24     Wetmur, J. G. (1994) Influence of the common human delta-aminolevulinate dehydratase polymorphism on lead
25            body burden. Environ. Health Perspect. 102(suppl. 3): 215-219.
26     Wetmur, J. G.; Lehnert, G.; Desnick, R. J. (1991) The 5-aminolevulinate dehydratase polymorphism: higher blood
27            lead levels in lead workers and environmentally exposed children with the 1-2 and 2-2 isozymes. Environ.
28            Res. 56: 109-119.
29     White, R. F.; Diamond, R.; Proctor, S.; Morey, C.; Hu, H. (1993) Residual cognitive deficits 50 years after lead
30            poisoning during childhood. Br. J. Ind. Med. 50: 613-622.
31     Wicklund, K. G.; Daling, J. R.; Allard, J.; Weiss, N. S. (1988) Respiratory cancer among orchardists in Washington
32            State, 1968-1980. J. Occup. Med. 30: 561-564.
33     Wide, M. (1985) Lead exposure on critical days of fetal life affects fertility in the female mouse. Teratology 32:
34            375-380.
35     Wide, M.; D'Argy, R. (1986) Effect of inorganic lead on the primordial germ cells in the mouse embryo. Teratology
36            34:207-212.
37     Wide, M.; Nilsson, O. (1977) Differential susceptibility of the embryo to inorganic lead during periimplantation in
38            the mouse. Teratology 16: 273-276.
39     Wide, M.; Nilsson, B. O. (1979) Interference of lead with implantation in the mouse: a study of the surface
40            ultrastructure of blastocysts and endometrium. Teratology 20: 101-113.
41     Wiebe, J. P.; Barr, K. J. (1988) Effect of prenatal and neonatal exposure to lead on the affinity and number of
42            estradiol receptors in the uterus. J. Toxicol. Environ. Health 24: 451-460.
43     Wiebe, J. P.; Barr, K. J.; Buckingham, K. D. (1988) Effect of prenatal and neonatal exposure to lead on
44            gonadotropin receptors and steroidogenesis in rat ovaries. J. Toxicol. Environ. Health 24: 461-476.
45     Wiemann, M.; Schirrmacher, K.; Busselberg, D. (1999) Interference of lead with the calcium release activated
46            calcium flux of osteoblast-like cells. Calcif. Tissue Int. 65: 479-485.
47     Winneke, G.; Kramer, U. (1997) Neurobehavioral aspects of lead neurotoxicity in children. Cent. Eur. J. Public
48            Health 5: 65-69.
49     Wise, J. P.; Leonard, J. C.; Patierno, S. R. (1992) Clastogenicity of lead chromate particles in hamster and human
50            cells. Mutat. Res. 278: 69-79.
51     Wise, J. P.; Orenstein, J. M.; Patierno, S. R.  (1993) Inhibition of lead chromate clastogenesis by ascorbate:
52            relationship to particle dissolution and uptake. Carcinogenesis (London) 14: 429-434.
53     Wise, J. P., Sr; Stearns, D. M.; Wetterhahn,  K. E.; Patierno, S. R. (1994) Cell-enhanced dissolution of carcinogenic
54            lead chromate particles: the role of individual dissolution products in clastogenesis. Carcinogenesis 15:
55            2249-2254.
       December 2005                                 5-337         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Wise, S. S.; Schuler, J. H.; Holmes, A. L.; Katsifis, S. P.; Ketterer, M. E.; Hartsock, W. J.; Zheng, T.; Wise, J. P., Sr.
  2            (2004a) Comparison of two paniculate hexavalent chromium compounds: Barium chromate is more
  3            genotoxic than lead chromate in human lung cells. Environ. Mol. Mutagen. 44: 156-162.
  4     Wise S. S.; Holmes A. L.; Ketterer, M. E.; Hartsock, W. J.; Fomchenko, E.; Katsifis, S.; Thompson, W. D.; Wise, J.
  5            P. Sr. (2004b) Chromium is the proximate clastogenic species for lead chromate-induced clastogenicity in
  6            human bronchial cells. Mutat. Res. 560: 79-89.
  7     Wise, S. S.; Holmes, A. L.; Moreland, J. A.; Xie, H  .; Sandwick, S. J.; Stackpole, M. M.; Fomchenko, E.; Teufack,
  8            S.; May, A. J., Jr.; Katsfis, S. P.; Wise, J. P., Sr. (2005) Human lung cell growth is not stimulated by lead
  9            ions after lead chromate-induced genotoxicity. Mol. Cell. Biochem. 279: 75-84.
10     Wisotzky, J.;Hein, J. W. (1958) Effects of drinking solutions containing metallic ions above and below hydrogen in
11            the electromotive series on dental caries in the Syrian hamster. J. Am. Dent. Assoc. 57: 796-800.
12     Witzmann, F. A.; Daggett, D. A.; Fultz, C. D.; Nelson, S. A.; Wright, L. S.; Kornguth, S. E.; Siegel, F. L. (1998)
13            Glutathione S-transferases: two-dimensional electrophoretic protein markers of lead exposure.
14            Electrophoresis 19:  1332-1335.
15     Wolf, A. W.; Jimenez, E.; Lozoff, B. (2003) Effects of iron therapy on infant blood lead levels. J. Pediatr. 143: 789-
16            795.
17     Wolin, M. S. (2000) Interactions of oxidants with vascular signaling systems. Arterioscler. Thromb. Vase. Biol. 20:
18            1430-1442.
19     Wood, N.; Bourque, K.; Donaldson, D. D.; Collins, M.; Vercelli, D.; Goldman, S. J.; Kasaian, M. T. (2004) IL-21
20            effects on human IgE production in response to IL-4 or IL-13. Cell. Immunol. 231: 133-145.
21     Wozniak, K.; Blasiak, J. (2003) In vitro genotoxicity of lead acetate: induction of single and double DNA strand
22            breaks and DNA-protein cross-links. Mutat. Res. 535: 127-139.
23     Wright, R. O.; Shannon, M.  W.; Wright, R. J.; Hu, H. (1999) Association between iron deficiency and low-level
24            lead poisoning in an urban primary care clinic. Am. J. Public Health 89: 1049-1053.
25     Wright, R. O.; Silverman, E. K.; Schwartz, J.; Tsaih, S.-W.; Senter, J.; Sparrow, D.; Weiss, S. T.; Aro, A.; Hu, H.
26            (2004) Association between hemochromatosis genotype and lead exposure among elderly men: the
27            Normative Aging Study. Environ. Health Perspect. 112: 746-750.
28     Wu, T.; Buck,  G. M.; Mendola, P. (2003) Blood lead levels and sexual maturation in U.S. girls: the Third National
29            Health and Nutrition Examination Survey, 1988-1994. Environ. Health Perspect. Ill: 737-741.
30     Wu, W.; Rinaldi, L.; Fortner, K. A.; Russell, J. Q.; Tschoop, J.; Irvin, C.; Budd, R.  C. (2004) Cellular FLIP long-
31            transgenic mice manifest a Th2 cytokine bias and enhanced allergic airway inflammation. J. Immunol. 172:
32            4724-4732.
33     Xie, Y.; Chiba, M.; Shinohara, A.; Watanabe, H.; Inaba, Y. (1998) Studies  on lead-binding protein and interaction
34            between lead and selenium in the human erythrocytes. Ind. Health  36: 234-239.
35     Xie, L.; Gao, Q.; Xu, H. (2003) Ameliorative effect  of L-methionine onPb-exposed mice. Biol. Trace Elem. Res.
36            93:227-236.
37     Xie, H.; Wise, S. S.; Holmes, A. L.; Xu, B, Wakeman, T. P.; Pelsue, S. C.;  Singh, N. P.; Wise, J. P., Sr. (2005)
3 8            Carcinogenic lead chromate induces DNA double-strand breaks in human lung cells. Mutat. Res. 586: 160-
39            172.
40     Xu, D. X.; Shen, H. M.; Zhu, Q. X.; Chua, L.;, Wang, Q. N.; Chia, S. E.; Ong, C. N. (2003) The associations among
41            semen quality, oxidative DNA damage in human spermatozoa and concentrations of cadmium, lead and
42            selenium in seminal plasma. Mutat. Res. 534: 155-163.
43     Xupeietal., 1988.
44     Yamamoto, C.; Miyamoto, A.; Sakamoto, M.; Kaji,  T.; Kozuka, H. (1997)  Lead perturbs the regulation of
45            spontaneous release of tissue plasminogen activator and plasminogen activator inhibitor-1 from vascular
46            smooth muscle cells and fibroblasts in culture. Toxicology 117: 153-161.
47     Yanez, L.; Garcia-Nieto, E.; Rojas, E.; Carrizales, L.; Mejia, J.; Calderon, J.;  Razo, I.; Diaz-Barriga, F. (2003) DNA
48            damage in blood cells from children exposed to arsenic and lead in a  mining area. Environ. Res. 93: 231-
49            240.
50     Ye, S.-H. (1993) Hypoxanthine phosphoribosyl transferase assay of lead mutagenicity on keratinocytes. Zhongguo
51            YaoliXuebao 14: 145-147.
52     Youssef, S. A. H. (1996) Effect of subclinical lead toxicity on the immune  response of chickens to Newcastle's
53            disease virus vaccine. Res. Vet. Sci. 60: 13-16.
54     Yu, S. Y.; Mizinga, K. M.; Nonavinakere, V. K.; Soliman, K. F. (1996) Decreased  endurance to cold water
55            swimming and delayed sexual maturity in the rat following neonatal lead exposure. Toxicol. Lett. 85: 135-
56            141.

       December 2005                                5-338         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Yucesoy, B.; Turhan, A.; Ure, M; Imir, T.; Karakaya, A. (1997) Simultaneous effects of lead and cadmium on NK
  2            cell activity and some phenotypic parameters. Immunopharmacol. Immunotoxicol. 19: 339-348.
  3     Zareba, G.; Chmielnicka, J. (1992) Disturbances in heme biosynthesis in rabbits after administration per os of low
  4            doses of tin or lead. Biol. Trace Elem. Res. 34: 115-122.
  5     Zelikoff, J. T.; Li, J. H.; Hartwig, A.; Wang, X. W.; Costa, M.; Rossman, T. G. (1988) Genetic toxicology of lead
  6            compounds. Carcinogenesis 9: 1727-1732.
  7     Zelikoff, J. T.; Parsons, E.; Schlesinger, R. B. (1993) Inhalation of paniculate lead oxide disrupts pulmonary
  8            macrophage-mediated functions important for host defense and tumor surveillance in the lung. Environ.
  9            Res. 62: 207-222.
10     Zhao, W.-F.; Ruan, D.-Y.; Xy, Y.-Z.; Chen, J.-T.; Wang, M.; Ge, S.-Y. (1999) The effects of chronic lead exposure
11            on long-term depression in area CA1 and dentate gyrus of rat hippocampus in vitro. Brain Res. 818: 153-
12            159.
13     Zhou, X. J.; Vaziri, N. D.; Wang, X. Q.; Silva, F. G.; Laszik, Z. (2002) Nitric oxide synthase expression in
14            hypertension induced by inhibition of glutathione synthase. J. Pharmacol. Exp. Ther. 300: 762-767'.
15     Zierold, K. M.; Anderson, H. (2004) Trends in blood lead levels among children enrolled in the Special
16            Supplemental Nutrition Program for Women, Infants and Children from 1996 to 2000. Am. J. Public Health
17            94:  1513-1515.
18     Zimmermann, L.; Pages, N.; Antebi, H.; Hafi, A.; Boudene, C.; Alcindor, L. G. (1993) Lead effect on the oxidation
19            resistance of erythrocyte membrane in rat triton-induced hyperlipidemia. Biol. Trace Elem. Res. 38:311-
20            318.
21     Zmuda, J. M.; Cauley, J. A.; Ferrell, R. E. (2000) Molecular epidemiology of vitamin D receptor gene variants.
22            Epidemiol. Rev. 22: 203-217.
23     Zuscik, M. J.; Pateder, D. B.; Puzas, J. E.; Schwarz, E. M.; Rosier, R. N.; O'Keefe, R. J. (2002) Lead alters
24            parathyroid hormone-related peptide and transforming growth factor-"beta" 1 effects and AP-1 and NF-
25            "kappa"B signaling in chondrocytes. J. Orthop. Res. 20: 811-818.
26
       December 2005                                5-339        DRAFT-DO NOT QUOTE OR CITE

-------
 i     6.    EPIDEMIOLOGIC STUDIES OF HUMAN HEALTH
 2        EFFECTS ASSOCIATED WITH LEAD EXPOSURE
 3
 4
 5   6.1    INTRODUCTION
 6         This chapter assesses information regarding the biological effects of lead exposure,
 7   with emphasis on (1) qualitative characterization of lead-induced effects and (2) delineation of
 8   concentration-response relationships for key health effects of most concern. Epidemiologic
 9   studies linking lead exposure to health effects were assessed in the 1986 Air Quality Criteria for
10   Lead (U.S. Environmental Protection Agency, 1986a), an associated addendum (U.S.
11   Environmental Protection Agency, 1986b), and a 1990 Supplement (U.S. Environmental
12   Protection Agency, 1990). Many earlier studies reported lead effects on child development
13   (psychometric intelligence), blood pressure and related cardiovascular endpoints, heme
14   biosynthesis, kidney, and reproduction and development. Numerous more recent epidemiologic
15   studies discussed in this chapter have further evaluated these relationships to lead exposure,
16   thereby providing an expanded basis for assessment of health effects associated with exposure to
17   lead at concentrations currently encountered by the general U.S. population.
18         Special emphasis is placed here on discussion of the effects of lead exposure in children.
19   Children are particularly at risk due to sources of exposure, mode of entry, rate of absorption and
20   retention, and partitioning of lead in soft and hard tissues. The greater sensitivity of children to
21   lead toxicity, their inability to recognize symptoms, and their dependence on parents and
22   healthcare professionals make them an especially vulnerable population requiring special
23   consideration in developing criteria and  standards for lead.
24         As discussed elsewhere in this document (Chapter 5), extensive experimental evidence
25   links lead exposure with health effects in laboratory animals. Thus, many of the reported
26   epidemiologic associations of lead health effects have considerable biological credibility.
27   Accordingly, the new epidemiologic studies of lead assessed here are best considered in
28   combination with information from the other chapters on lead exposure and on toxicological
29   effects of lead in animals. The epidemiologic studies constitute important information on
30   associations between health effects and exposures of human populations to "real world" lead
31   concentrations and also help to identify susceptible subgroups and associated risk factors.

     December 2005                        6-1            DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.1.1   Approach to Identifying Lead Epidemiologic Studies
 2          Numerous lead epidemiologic papers have been published since completion of the 1986
 3    Lead AQCD/Addendum, and 1990 Supplement.  A systematic approach has been employed to
 4    identify relevant new epidemiologic studies for consideration in this chapter. In general, an
 5    ongoing literature search has been used in conjunction with other strategies to identify lead
 6    epidemiologic literature pertinent to developing criteria for the National Ambient Air Quality
 7    Standards (NAAQS) for lead. A publication base was established using Medline, Pascal,
 8    BIOSIS, and Embase, and a set of search terms aimed at identifying pertinent literature.
 9          While the above search regime accessed much of the pertinent literature, additional
10    approaches augmented such traditional search methods. For example, a Federal Register Notice
11    was issued requesting information and published papers from the public at large.  Also, non-EPA
12    chapter authors, expert in this field, identified literature on their own, and EPA staff also
13    identified publications as part of their assessment and interpretation of the literature. Lastly,
14    additional potentially relevant publications are expected to be identified and included as a result
15    of external review of this draft document by the public and CAS AC.  The principal criteria used
16    for selecting literature for the present assessment is to focus mainly on those identified studies
17    that evaluate relationships between health outcome and lead exposure at concentrations in the
18    range of those currently encountered in the United States.  New studies published or accepted for
19    publication through June 2005, as identified using the approaches above, have been included in
20    this draft lead air quality criteria document (Lead AQCD), and additional efforts are being made
21    to identify and assess more recent studies.
22
23    6.1.2   Approach to Assessing Epidemiologic Evidence
24          Epidemiologic studies have evaluated lead effects on a wide range of health endpoints that
25    include, but are not limited to: neurotoxic effects (e.g., psychometric intelligence, behavioral
26    disturbances, and neurodevelopmental deficits), renal effects, cardiovascular effects,
27    reproductive and  developmental effects, genotoxic and carcinogenic effects, and immune effects.
28    The epidemiologic strategies most commonly used in lead health studies are: (1) cross-sectional
29    studies that examine the exposure and health outcome at a single point in time; and
30    (2) prospective longitudinal cohort studies that follow a group of individuals over time.
31    Both of these are types of observational rather than experimental studies.

      December 2005                         6-2            DRAFT-DO NOT QUOTE OR CITE

-------
 1          An overall approach useful for assessing epidemiologic evidence was stated in the 2004
 2    PM AQCD (U.S. Environmental Protection Agency, 2004), as summarized here.  That is, the
 3    critical assessment of epidemiologic evidence presented in this chapter is conceptually based
 4    upon consideration of salient aspects of the evidence of associations so as to reach fundamental
 5    judgments as to the likely causal significance of the observed associations (see Hill, 1965).  The
 6    general evaluation of the strength of the epidemiologic evidence reflects consideration not only
 7    of the magnitude and precision of reported lead effect estimates and their statistical significance,
 8    but also of the robustness of the effects associations. Statistical significance corresponds to the
 9    allowable rate of error (Type I error)  in the decision framework constructed from assuming that a
10    simple null hypothesis of no association is true. It is a conditional probability; for statistical
11    significance, typically there is a less than 0.05 chance of rejecting the null hypothesis given that
12    it is true. Robustness of the associations is defined as stability in the effect estimates after
13    considering a number of factors, including alternative models and model specifications, potential
14    confounding by copollutants, as well as issues related to the consequences of measurement error.
15          Consideration of the consistency of the effects associations, as discussed in the following
16    sections, involves looking across the  results obtained by various investigators in different
17    locations and times.  Relevant factors are known to exhibit much variation across studies, e.g.,
18    (1) presence and levels of other toxicants or pollutants of concern and (2) relevant demographic
19    factors related to sensitive subpopulations. Thus, consideration of consistency is appropriately
20    understood as an evaluation of the similarity or general  concordance of results, rather than an
21    expectation of finding quantitative results within a very narrow range.
22          Looking beyond the  epidemiologic evidence,  evaluation of the biological plausibility  of
23    the lead-health effects associations observed in epidemiologic studies reflects consideration of
24    both exposure-related factors and dosimetric/toxicologic evidence relevant to identification of
25    potential biological mechanisms underlying the various health outcomes. These broader aspects
26    of the assessment are only touched upon in this chapter but will be more fully integrated in
27    discussions presented in Chapter 7 (Integrative Synthesis).
28          In assessing the relative scientific quality of epidemiologic studies reviewed here and to
29    assist  in interpreting their findings, the following considerations were taken into account:
      December 2005                          6-3            DRAFT-DO NOT QUOTE OR CITE

-------
 1        (1)     To what extent are the biological markers used of adequate quality and sufficiently
 2                representative to serve as credible exposure indicators, well-reflecting geographic
 3                or temporal differences in study population exposures?
 4        (2)     Were the study populations well defined and adequately selected so as to allow
 5                for meaningful comparisons between study groups or meaningful temporal
 6                analyses of health effects results?
 7        (3)     Were the health endpoint measurements meaningful and reliable, including clear
 8                definition of diagnostic criteria utilized and consistency in obtaining dependent
 9                variable measurements?
10        (4)     Were the statistical analyses used appropriate, as well as being properly performed
11                and interpreted?
12        (5)     Were likely important covariates (e.g., potential confounders or effect
13                modifiers) adequately controlled for or taken into account in the study design
14                and statistical  analyses?
15        (6)     Were the reported findings internally consistent, biologically plausible, and
16                coherent in terms of consistency with other known facts?
17          These guidelines provide benchmarks for judging the relative quality of various studies
18    and in assessing the body of epidemiologic evidence. Detailed critical analysis of all
19    epidemiologic studies on lead health effects, especially in relation to all of the above questions,
20    is beyond the scope of this document.
21
22    6.1.3   Considerations in the Interpretation of Epidemiologic Studies  of
23            Lead Health Effects
24          Prior to assessing results from recent lead epidemiologic studies, issues and questions
25    arising from study designs and analysis methods used in the evaluation of lead health effects are
26    briefly discussed.  Study design can restrict the health effect parameters that can be estimated.
27    Separate considerations need to be made for acute versus chronic effect studies, as well as
28    individual versus aggregate-level analyses.  Issues include  measurement error, the functional
29    form of relationships (especially at low exposure levels) and the potential for confounding.
30    Aspects of these issues are briefly noted below, then are considered as various studies  are
31    reviewed in the following sections on specific health effect endpoints. Finally, they are further
32    examined as part of the  interpretive assessment(Section 6.9) at the end of this chapter.
      December 2005                         6-4            DRAFT-DO NOT QUOTE OR CITE

-------
 1          Measurement error is an important factor to consider, both for measurement of the health
 2    effect outcome and the representativeness of the biomarkers of exposure (principally blood and
 3    bone lead) used in most key epidemiologic studies.  For health outcome measures, the reliability
 4    and validity of the measurement need to be assessed. In addition, the appropriateness of the
 5    outcome measure for studying the hypothesis of interest needs to be determined. The critical
 6    issues of outcome measurement and classification are, to some extent, endpoint-specific, and will
 7    thusly be discussed further in the individual sections.
 8          Exposure misclassification can result in a notable reduction of statistical power in studies,
 9    especially in those that focus on the lower end of the exposure range. Limitations of blood lead
10    as an exposure index include the use of a single blood lead concentration to represent lead body
11    burden. Also of concern is the most relevant blood sample collection time point for to use in
12    evaluating possible associations with health outcomes (e.g., at 2 years of age when peak lead
13    exposure is expected versus concurrent blood lead samples). Another consideration is that
14    similar blood lead concentrations in two individuals do not necessarily reflect similar body
15    burdens. An added complication is that the relationship between lead intake and blood lead
16    concentration appears to be curvilinear. Bone lead determinations are typically considered a
17    measure of longer-term lead exposure; but, the X-ray fluorescence (XRF) method typically used
18    to assess lead levels in bone also has limitations, including the relatively high minimum
19    detection limit. The  type of bone measured to determine lead exposure is another important
20    aspect.
21          The relationship between a measurement of a health outcome endpoint and an estimate of
22    lead exposure based  on  a biomarker is an important concept. Modeling this relationship provides
23    a numerical slope that quantifies the relationship between lead exposure and health outcome.
24    These models must address differences in the relationship at different concentration ranges of
25    exposure and present the functional form that best describes such data. Various models, both
26    linear and nonlinear, have been considered to examine lead exposure-health effect relationships.
27    This is especially important at low lead exposures.  For example, a curvilinear relationship has
28    been reported for neurodevelopmental and cardiovascular outcomes at low lead exposure levels.
29          Depending on the subjects being examined for lead exposure effects, various other factors
30    can lead to confounding of the relationship being considered. Potential confounding factors
31    largely depend on the health outcome of interest and the study population.  Some potential

      December 2005                         6-5            DRAFT-DO NOT QUOTE OR CITE

-------
 1    confounding factors in children, for whom the major health concerns include neurological and
 2    developmental deficiencies, include:  socioeconomic status (SES); nutritional status; quality of
 3    home environment (e.g., HOME score); parental education; parental IQ; and birth weight, as a
 4    few examples. For adults, factors that may confound the association between lead and
 5    cardiovascular health outcomes include:  age; diet; alcohol use; smoking; and potential for
 6    copollutant exposures, such as cadmium.  Control for potential confounding factors can be
 7    attempted at the study design phase and/or during statistical analysis.
 8
 9    6.1.4   Approach to Presenting Lead Epidemiologic Evidence
10          In the main body of this chapter, each section starts by concisely highlighting important
11    points derived from the 1986 Lead AQCD/Addendum, and the 1990 Supplement.  Particular
12    emphasis is focused on studies and analyses that provide pertinent information for the critical
13    assessment of health risks from lead exposure.  Not all studies are accorded equal weight in the
14    overall interpretive assessment of evidence regarding lead-associated health effects.  Among
15    well-conducted studies with adequate control for confounding, increasing scientific weight is
16    accorded in proportion to the precision of their effect estimates. To ensure a thorough appraisal
17    of the evidence, more detailed information on key features (including study design, analysis, lead
18    biomarkers of exposure, and health outcome results) of important new studies are summarized in
19    tables in the Annex for this Chapter 6 (Annex AX6).
20          Emphasis is placed on main body text discussion below of (1) new studies employing
21    standardized methodological  analyses for evaluating lead effects across several cities and
22    providing overall effect estimates based on combined analyses of information pooled across
23    multiple cities; (2) studies conducted in the U.S. or Canada; and (3) meta-analyses of individual
24    studies conducted in various cities. Multicity studies are of particular interest and value due to
25    their evaluation of a wider range of lead exposures and large numbers of observations, thus
26    generally providing more precise effect estimates than most smaller  scale studies of single cities.
27    Furthermore, multicity studies have the potential to provide especially valuable evidence
28    regarding relative homogeneity and/or heterogeneity of lead health effects relationships across
29    geographic locations. The potential impacts of the underlying health status of populations and
30    cultural differences in the case of intelligence testing (one of the major health outcomes in
31    children) also need to be accounted for in the assessment; thus, U.S.  studies are emphasized over

      December 2005                         6-6            DRAFT-DO NOT QUOTE OR CITE

-------
 1    non-U.S. studies. In accordance with the emphasis placed on the lead epidemiologic studies in
 2    this chapter, Chapter 6 Annex tables are organized by region, with multicity studies in each
 3    region presented first.
 4          In the ensuing sections, epidemiological studies of biological markers of lead exposure are
 5    discussed first, in Section 6.2. The neurotoxic effects of lead are next discussed in Section 6.3
 6    for children and adults, followed by discussion of the renal and cardiovascular effects of lead in
 7    Sections 6.4 and 6.5.  Section 6.6 then discusses reproductive and developmental effects of lead,
 8    and Section 6.7 discusses genotoxic and carcinogenic effects of lead.  Section 6.8 discusses the
 9    effects of lead on the immune system.  The effects of lead on other organ systems (including the
10    hematopoietic, endocrine, hepatic, gastrointestinal, and respiratory systems) are assessed in
11    Section 6.9. Effects of lead on bone and teeth, as well as on ocular health are also discussed in
12    Section 6.9. Finally, Section 6.10 provides an interpretative assessment of the overall
13    epidemiologic evidence for lead health effects.
14
15
16    6.2   BIOLOGICAL MARKERS OF LEAD BODY BURDEN
17          AND EXPOSURE
18    6.2.1   Lead in Blood
19    6.2.1.1    Summary of Key Findings from the 1986 Lead AQCD
20          The extensive use of blood lead concentration as a dose metric reflects mainly the greater
21    feasibility of incorporating blood lead measurements into clinical or epidemiologic studies,
22    compared to other potential dose indicators, such as lead in kidney, plasma, urine, or bone
23    (Flegal and Smith, 1995; Graziano, 1994; Skerfving, 1988).  However, blood lead measurements
24    have  several limitations as measures of lead body burden and exposure that relate to the kinetics
25    of blood lead in relation to exposure and body burden.  These limitations were noted in Section
26    13.3.2 of the 1986 Lead AQCD, which discusses attributes and limitations of blood lead
27    concentration as an indicator of internal exposure. Relevant developments since the 1986 Lead
28    AQCD was completed include numerous studies of determinants of lead levels in bone (see
29    Section 6.2.2), which provide further support for the importance of this relatively slow kinetic
30    compartment in assessing the blood lead concentration as an  index of lead exposure. The
31    enhanced understanding of lead biokinetics has also been consolidated into exposure-biokinetics

      December 2005                        6-7            DRAFT-DO NOT QUOTE OR CITE

-------
 1    models (see Chapter 4), which not only serve to illustrate exposure-blood-body burden
 2    relationships, but also provide a means for making predictions about these relationships that can
 3    be tested experimentally or in epidemiologic studies. The basic concepts laid out in the 1986
 4    Lead AQCD, that the concentration of lead in blood is largely determined by the relatively recent
 5    exposure history of the individual and that it reflects the level of lead in a relatively mobile and
 6    small compartment, remain valid. Especially in children, who experience a more rapid turnover
 7    of bone mineral, an endogenous source of lead, blood lead concentrations closely parallel
 8    changes in total body burden.
 9
10    6.2.1.2   Analytical Methods for Measuring Lead in Blood
11          Analytical methods for measuring lead in blood include flame atomic absorption
12    spectrometry (AAS), graphite furnace atomic absorption spectrometry (GFAAS), anode stripping
13    voltammetry (ASV), inductively coupled plasma-atomic emission spectroscopy (ICP-AES), and
14    inductively coupled plasma-mass spectrometry (ICP-MS). GFAAS and ASV are generally
15    considered to be the methods of choice (Flegal and Smith, 1995). Background correction, such
16    as Zeeman background correction that minimizes the impact of the absorbance of molecular
17    species, must be applied.  Limits of detection for lead using AAS are on the order of 5-10 |ig/dL
18    for flame AAS measurements, approximately 0.1 |ig/dL for flameless AAS measurements, and
19    0.005 |ig/dL for GFAAS (Flegal and Smith, 1995; National Institute for Occupational Safety and
20    Health, 1994). A summary of standard methods that have been reported for blood lead analysis
21    are provided in Annex Table AX6-2.1. Sample preparation usually consists of wet ashing in
22    heated strong acid (National Institute for Occupational Safety and Health, 1977a,b,c,d,e);
23    however, preparation methods not requiring wet ashing have also been reported (Aguilera de
24    Benzo et al., 1989; Delves and Campbell, 1988; Manton and Cook,  1984; National Institute for
25    Occupational Safety and Health, 1977f; Que Hee et al., 1985; Zhang et al., 1997). The presence
26    of phosphate, ethylenediaminetetraacetic acid (EDTA), or oxalate can sequester lead and cause
27    low readings in flame AAS (National Institute for Occupational Safety and Health, 1984).
28    A comparison of IDMS,  ASV, and GFAAS showed that all three of these methods can be used to
29    quantify lead levels in blood (Que Hee et al., 1985).
30
     December 2005                         6-8           DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.1.3    Levels of Lead in Blood
 2         Blood lead concentrations in the U.S. general population have been monitored in the
 3    National Health and Nutrition Examination Survey (NHANES) conducted by the Centers for
 4    Disease Control and Prevention. Data from the most recent survey (NHANES IV, Centers for
 5    Disease Control, 2005) are shown in Tables 6-2.1 and 6-2.2.  For survey years 2001-2002, the
 6    geometric mean blood lead concentration for ages >1 year (n = 8,945) was 1.45 |ig/dL (95% CI:
 7    1.39, 1.52); with the geometric mean in males (n = 4,339) being 1.78 |ig/dL (95% CI: 1.71,
 8    1.86) and in females (n = 4,606) being 1.19 |ig/dL (95% CI:  1.14, 1.25).  Blood lead
 9    concentrations in the U.S. general population have decreased over the past three decades as
10    regulations regarding lead paint, leaded fuels, and lead-containing plumbing materials have
11    decreased exposure. Changes over time in children are shown in Figure 6-2.1.
12         Yassin et al. (2004) analyzed occupational category strata from NHANES III (1988-1994;
13    Table 6-2.3). The geometric mean for all adults (n = 11,126) included in the analysis was
14    2.42 |ig/dL (GSD 6.93), with the highest means  estimated for vehicle mechanics (n = 169;
15    GM 4.80 |ig/dL [GSD 3.88]) and construction workers (n = 122;  GM 4.44 |ig/dL [GSD  7.84]).
16    See Annex Table AX6-2.2 for a summary of selected measurements of blood lead levels in
17    humans.
18
19    6.2.1.4    Blood Lead as a Biomarker of Lead Body Burden
20         A simple conceptual representation of the lead body burden is that it is comprised of a fast
21    turnover pool, comprised mainly of soft tissue, and a slow pool, comprised mainly of skeletal
22    tissues (Rabinowitz et al., 1976; see Chapter 4 for detailed discussion of this and other more
23    complex models of lead biokinetics). The rapid pool has an elimination half-life of-28 days and
24    comprises <1% of the lead body burden.  The slow pool has an elimination half-life of several
25    decades and comprises >90% of the total lead body burden. Blood, which comprises -1% of
26    body burden, exchanges with both the slow and  fast pools, and exhibits multiphasic elimination
27    kinetics. The dominant phase, exhibited shortly after a change in exposure occurs, has a half-life
28    of-20-30 days.  A slower phase becomes evident with longer observation periods following a
     December 2005                         6-9           DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
                              Table 6-2.1. Blood Lead Concentrations in U.S. by Age, NHANES IV (1999-2002)
Age
Survey Period
N
Blood Lead
(Mg/dL)a
1-5 years
1999-2000 2001-2002
723 898
2.23 1.70
(1.96,2.53) (1.55,1.87)
6-11 years
1999-2000 2001-2002
909 1,044
1.51 1.25
(1.36, 1.66) (1.14, 1.36)
12-19 years
1999-2000 2001-2002
2,135 2,231
1.10 0.94
(1.04,1.17) (0.90,0.99)
^20 years
1999-2000 2001-2002
4,207 4,772
1.75 1.56
(1.68, 1.81) (1.49, 1.62)
         aBlood lead concentrations presented are geometric means (95% CI).
                            Table 6-2.2. Blood Lead Concentrations in U.S. by Gender, NHANES IV (1999-2002)
Gender
Survey Period
n
Blood Lead
(Mg/dL)a
Males
1999-2000
3,913
2.01
(1.93,2.09)
2001-2002
4,339
1.78
(1.71, 1.86)
Females
1999-2000
4,057
1.37
(1.32, 1.43)
2001-2002
4,606
1.19
(1.14, 1.25)
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
         aBlood lead concentrations presented are geometric means (95% CI).

-------
                18
                16-
            ^  14-
            1;  12
            "ra  10
O
_o
CQ
                 8-
                 6-
                 4
                 2
                 0
                      1976-1980   1988-1991    1991-1994    1999-2000
                                            Survey Period
                                                             2001-2002
     Figure 6-2.1.  Blood lead concentrations in U.S. children, 1-5 years of age. Shown are
                   geometric means and 95% confidence intervals as reported from the
                   NHANES II (1976-1980) and NHANES III Phase 1 (1988-1991;
                   Pirkle et al., 1994); NHANES III Phase 2 (1991-1994; Pirkle et al., 1998);
                   and NHANES IV (1999-2000, 2001-2002; Centers for Disease Control, 2005).
 1   decrease in exposure. The half-life of this slow phase has been estimated to be ~3 to 30 years
 2   and appears to correlate with finger bone lead levels. This characterization is supported by
 3   measurements of lead contents of cadaver tissues (Barry, 1975; Schroeder and Tipton, 1968),
 4   lead isotope kinetics in adults (Chamberlain et al., 1978; Rabinowitz et al., 1976; Griffin et al.,
 5   1975), and measurements of blood and bone lead levels in retired lead workers (Schiitz et al.,
 6   1987; Christoffersson et al., 1986).
 7          As a consequence of a relatively large fraction of the body burden having a relatively slow
 8   turnover compared to blood, a constant lead uptake (or constant intake and fractional absorption)
 9   gives rise  to a quasi-steady state blood lead concentration, while the body burden continues to
10   increase, largely as a consequence of retention of lead in bone (Figure 6-2.2). As a result, the
11   contribution of blood lead to body burden decreases over time. An abrupt change in lead uptake
12   gives rise  to a relatively rapid change in blood lead, to a new quasi-steady  state, achieved in
13   -75-100 days (i.e., 3-4 times the blood elimination half-life). In the hypothetical simulation
14   shown in Figure 6-2.2, body burden has approximately doubled (from 5 to 10 mg) as a result of a
     December 2005
                                  6-11
DRAFT-DO NOT QUOTE OR CITE

-------
December 2005






Oi
i
to


DRAFT-DO I
^
0
H
O
O
H
W
O
O
HH
H
W
Table 6-2.3. Blood Lead Concentrations by Occupation, NHANES III (1988-1994)
Blood Lead (ug/dL)
Occupation
Vehicle mechanics
Food service workers
Management, professional, technical, and sales workers
Personal service workers
Agricultural workers
Production workers: machine operators, material movers, etc.
Laborers other than in construction
Transportation workers
Mechanics other than vehicle mechanics
Construction trades people
Construction laborers
Health service workers
All workers
Data from Yassin et al. (2004).






n
169
700
4,768
1,130
498
1,876
137
530
227
470
122
499
11,126







GM
4.80
2.00
2.13
2.48
2.76
2.88
3.47
3.49
3.50
3.66
4.44
1.76
2.42







GSD
3.88
2.69
4.05
4.52
4.02
4.24
3.36
5.19
4.91
4.64
7.84
2.24
6.93







Maximum
28.1
27.0
39.4
25.9
23.4
52.9
21.8
22.3
16.6
16.9
36.0
22.4
52.9








-------
             CO
             CD
             _l
             T3
             O
             _g
             tn
25

20

15

10

 5

 0
                  20
                 30             40
                     Age (year)
                                                                25
                                                                20
           50
                                                                 30
                                                  TWA Blood
                 20
                 30            40
                    Age (year)
           50
Figure 6-2.2.   Simulation of relationship between blood lead concentration and body
              burden in adults. A constant baseline intake gives rise to a quasi-steady
              state blood lead concentration, while the body burden continues to increase,
              largely as a consequence of retention of lead in bone (upper panel). An
              abrupt change in lead uptake gives rise to a relatively rapid change in blood
              lead, to a new quasi-steady state, and a relatively small change in body
              burden.  The long-term average blood lead concentration more closely
              tracks the pattern of change in body burden (lower panel). Simulation
              based on lead biokinetics model of Leggett (1993).
December 2005
                     6-13
DRAFT-DO NOT QUOTE OR CITE

-------
 1    5-year period of increased lead uptake; however, the blood lead concentration prior to and 1 year
 2    following cessation of the increased uptake has not changed (~2 |ig/dL).  Therefore, a single
 3    blood lead concentration measurement, or a series of measurements taken over a short-time span,
 4    can be expected to be a relatively poor index of lead body burden. On the other hand, an average
 5    of individual blood lead concentrations measured over a longer period of time (long-term
 6    average blood lead concentrations) can be expected to be a better index of body burden.  In the
 7    hypothetical simulation shown in Figure 6-2.2, both the long-term average blood lead
 8    concentration and the body burden have approximately doubled.
 9          The disparity in the kinetics of blood lead and body burden has important implications for
10    the interpretation of blood lead concentration measurements in epidemiology studies. Cross-
11    sectional studies, by design, sample blood lead concentration at one time or over relatively
12    narrow windows of time. In these samples, the blood lead concentration may or may not reflect
13    well the body burden; it is more likely to do so if the measured value is a reflection of the long-
14    term average blood lead concentration.  However, in cross-sectional samples, this cannot be
15    ascertained.  Longitudinal sampling provides a means for estimating average blood lead
16    concentrations over time, and such estimates are more likely to be more strongly influenced by
17    differences in body burden, than by differences in short-term variability in exposure. The degree
18    to which repeated sampling will reflect the actual long-term time-weighted average blood lead
19    concentration will depend on the sampling frequency in relation to variability in exposure. High
20    frequency variability in exposures can produce episodic (or periodic) oscillations in blood lead
21    concentration and body burden that may not be captured with low sampling frequencies.
22    The same basic concepts described above regarding lead biokinetics of adults also apply to
23    children.  The empirical basis for the  understanding of the biokinetics of lead in children is much
24    weaker than that for adults. However, based on the understanding of bone mineral kinetics and
25    its importance as a mechanism for uptake and loss of lead from bone (Leggett, 1993; O'Flaherty,
26    1991, 1993, 1995), the slow pool, described above for adults, is thought to be much more labile
27    in children, reflecting a more rapid turnover of bone mineral in children.  As a result, changes in
28    blood lead concentration in children are thought to more closely parallel changes in total body
29    burden (Figure 6-2.3). Nevertheless,  in children, as in adults, the long-term time-weighted
30    average blood lead concentration is more likely to provide a better reflection of lead body burden
31    than a single sample.

      December 2005                        6-14           DRAFT-DO NOT QUOTE OR CITE

-------
            T3
            03
            CD
            _J
            T3
            O
            _O

            CD
25


20


15


10


 5


 0
                                    4         6

                                     Age (year)
                                DO
                                O
                                Q.

                                00

                                EL
                                CD
                               CO
                           10
               10
            CD

            -3-
            T3
            03
            CD
            _l
            T3
            O
            _0

            DO
                  0    1
456

Age (year)
                                                 10
Figure 6-2.3.  Simulation of relationship between blood lead concentration and body
             burden in children. Blood lead concentration is thought to parallel body
             burden more closely in children than in adults, due to more rapid turnover
             of bone and bone-lead stores in children (upper panel).  Nevertheless, the
             long-term average blood lead concentration more closely tracks the pattern
             of change in body burden (lower panel). Simulation based on Leggett (1993)
             lead biokinetics model.
December 2005
   6-15
                                    DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.1.5   Blood Lead as a Biomarker of Lead Exposure
 2          Characterizing quantitative relationships between external lead exposures and blood lead
 3    concentrations has become central to concentration-response analyses for human populations
 4    exposed to lead. The 1986 Lead AQCD summarized the empirical basis for this as it stood at the
 5    time. A summary of empirically-derived regression slope factors relating lead exposures and
 6    blood lead is provided in Abadin and Wheeler (1993). More recent meta-analyses, based on
 7    structure equation modeling, provide further support for quantitative relationships between lead
 8    exposures and blood lead concentrations in children (e.g., U.S. Environmental Protection
 9    Agency, 2001; Lanphear et al., 1998; Succop et al.,  1998).
10          As noted above, the elimination half-life of lead in blood is -25 to 30 days (Chamberlain
11    et al., 1978; Rabinowitz et al., 1976; Griffin et al., 1975); therefore, the blood lead concentration
12    mainly reflects the exposure history for the previous few months. However, a single blood lead
13    measurement cannot distinguish between a history of long-term low level lead exposure or a
14    history that includes higher acute exposures.  This is illustrated in Figure 6-2.4.  Two
15    hypothetical  children are simulated. Child A has a relatively constant lead intake from birth;
16    whereas Child B has the same long-term lead intake as Child A, with a 1-year elevated intake
17    which begins at age 24 months (Figure 6-2.4, upper panel). The absorption fraction is assumed
18    to be the same for both children. Blood lead samples 1 and 5, or 2 and 4, will yield similar blood
19    lead concentrations (~3 or 10 |ig/dL, respectively), yet the exposure contexts for these samples
20    are very different. Two samples (e.g., 1 and 2, or 4 and 5), at a minimum, are needed to
21    ascertain if the blood lead concentration is changing over time. The rate of change can provide
22    information about the magnitude of change in exposure, but not necessarily about the time
23    history of the change (Figure 6-2.4, lower panel). Here again, time-integrated measurements of
24    lead concentration may provide a means for accounting for some of these factors and, thereby,
25    provide a better measure of long-term exposure. The same concepts apply to  estimation of long-
26    term exposure based on blood lead measurements in adults (Gerhardsson et al.,  1992, 1995a;
27    Roelsetal., 1995).
28          An additional complication is that the relationship between lead intake and blood lead
29    concentration is curvilinear; that is, the increment in blood lead concentration per unit of lead
30    intake decreases with increasing blood lead concentration, both in children (Lacey et al., 1985;
31    Ryu et al.,  1983; Sherlock and Quinn, 1986) and in adults (Kehoe,  1987; Laxen et al., 1987;

      December 2005                           6-16        DRAFT-DO NOT QUOTE OR CITE

-------
12    24     36     48    60
             Age (months)
                    72
                                                                  84
                          12     24
36     48
 Age (months)
72
                                        84
Figure 6-2.4.  Simulation of temporal relationships between lead exposure and blood lead
             concentration in children.  Child A and Child B have a relatively constant
             basal lead intake (ug/day/kg body weight) from birth; Child B experiences
             1-year elevated intake which begins at age 24 months (upper panel). Blood
             lead samples 1 and 5, or 2 and 4, will yield similar blood lead concentrations
             (~3 or 10 ug/dL, respectively), yet the exposure scenarios for these samples
             are very different. As shown in the example of Child C and Child D, two
             samples can provide information about the magnitude of change in exposure,
             but not necessarily the temporal history of the change (lower panel).
December 2005
6-17
                        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Pocock et al., 1983; Sherlock et al., 1982, 1984).  The nonlinearity is evident even at blood lead
 2    concentrations below 25 |ig/dL (Figure 6-2.5).  The nonlinearity in the lead intake-blood lead
 3    concentration relationship is derived, at least in part, from a capacity limitation in the
 4    accumulation of lead in erythrocytes (Bergdahl et al., 1997, 1998, 1999; Manton et al., 2001;
 5    Smith et al., 2002). A capacity-limited process may also reside at the level of intestinal
 6    absorption; however, the dose at which absorption becomes appreciably limited in humans is not
 7    known. Lead intake-blood lead relationships also vary (a) with age, as a result of age-
 8    dependency of gastrointestinal absorption of lead, and (b) with diet and nutritional status
 9    (Mushak, 1991).
10          The blood lead concentration is also influenced by lead in bone. Evidence for the
11    exchange of bone lead and soft tissue lead stores comes from analyses of stable lead  isotope
12    signatures of lead in bone  and blood.  As noted earlier, bone lead likely contributes to the slow
13    phase of elimination of lead from blood that has been observed in retired lead workers
14    (Christoffersson et al.,  1986; Schutz et al., 1987). Bone lead stores may contribute 40-70% of
15    the lead in blood (Smith et al., 1996). This contribution increases during pregnancy,  when
16    mobilization of bone lead  increases, apparently as the bone is resorbed to produce the fetal
17    skeleton (Gulson et al., 2003). The mobilization of bone lead during pregnancy may contribute,
18    along with other mechanisms (e.g., increased absorption), to the increase in lead concentration
19    that has been observed during the later stages of pregnancy (Gulson et al., 1997; Lagerkvist
20    et al., 1996; Schuhmacher et al., 1996). In addition to pregnancy, other states of increased bone
21    resorption appear to result in release of bone lead to blood; these include lactation, osteoporosis,
22    and menopause (Gulson et al., 2003). These observations are consistent with epidemiologic
23    studies that have shown increases in blood lead concentration after menopause and in association
24    with decreasing bone density in postmenopausal women (Hernandez-Avila et al., 2000; Nash
25    et al., 2004; Symanski and Hertz-Picciotto, 1995).  The relationship between blood and bone lead
26    is discussed further in Section 6.2.2 on bone lead as a biomarker of lead exposure.
27
28    6.2.1.6   Summary of Blood Lead as a Biomarker of Lead Body Burden and Exposure
29          The blood lead concentration measured in an individual will be determined by the recent
30    exposure history  of the individual, as well as the long-term exposure history that gives rise to
31    accumulated bone lead stores. The contribution of the latter to blood lead may change with the

      December 2005                            6-18       DRAFT-DO NOT QUOTE OR CITE

-------
            "O
            03
            QJ
            _l
            T3
            O
            O
70

60 -

50 -

40 -

30 -
            £ 20 -
            CD
               10 -
                                 •Child
                                 Adult
                               5           10           15
                                  Lead Intake (ug/kg/day)
                                                       20
           T3
            re
            
-------
 1    duration and intensity of the exposure, age, and various physiological variables (e.g., nutritional
 2    status, pregnancy, menopause).  Longitudinal measurements of blood lead can be expected to
 3    provide a more reliable measure of exposure history of an individual (and will more closely
 4    parallel body burden) compared to a single measurement; however, the degree to which this will
 5    apply will depend on the sampling frequency with respect to the temporal pattern of exposure.
 6          In general, higher blood lead concentrations can be interpreted as indicating higher
 7    exposures (or lead uptakes); however, they do not necessarily predict appreciably higher body
 8    burdens.  Similar blood lead concentrations in two individuals (or populations) do not necessarily
 9    translate to similar body burdens or similar exposure histories.
10
11    6.2.2    Lead in Bone
12    6.2.2.1   Summary of Key Findings from the 1986 Lead AQCD
13          In the 1986 Lead AQCD, the discussion on the distribution of lead in bone was fairly
14    limited and mostly based on postmortem studies. The distribution between the two major
15    compartments of cortical and trabecular bone were addressed especially based on the pioneering
16    isotopic work of Rabinowitz et al. (1977). Estimates of the amount of lead in bone were also
17    provided. There was limited discussion of the half-life of lead in bone as being on the order of
18    several decades.
19          One of the major conclusions of the 1986 Lead AQCD regarding bone lead was that the
20    traditional view that the skeletal system was a total sink for body lead was now giving way to the
21    notion that there were at least several bone compartments for lead, with different mobility
22    profiles. The possibility of bone lead serving as a source of long-term internal exposure was also
23    considered.
24          Since 1986, the main focus of lead in bone studies has been on occupationally-exposed
25    subjects, because of concern, until more recent times, about the ability to measure  lower levels of
26    lead in bone from environmentally-exposed subjects. Furthermore most of the focus has been on
27    adult males, with very few studies on females and children. The newly available studies of lead
28    in bone are discussed in the following sections.
      December 2005                           6-20        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.2.2    Methodology of Bone Lead Analysis
 2    6.2.2.2.1   Analytical Methods for Measuring Lead in Bone
 3         Bone is comprised of two main types (cortical and trabecular) that have distinct rates of
 4    turnover and lead release, resulting in potential differences in implications with respect to
 5    toxicity aspects (further discussed in Section 6.2.2.3). The most commonly measured bones are
 6    the tibia, calcaneus, patella, and finger bone. For cortical bone, the midpoint of the tibia is
 7    measured.  For trabecular bone, both the patella and calcaneus are measured. Recent studies
 8    favor measurement of the patella, because it has more bone mass and may afford better
 9    measurement precision than the calcaneus. The advantages and disadvantages of patella and
10    calcaneus sites have not been thoroughly investigated.  Bone lead measurements  in cadavers,
11    environmentally-exposed subjects, and occupationally-exposed subjects are presented in Annex
12    Tables AX6-2.3, AX6-2.4, and AX6-2.5, respectively.
13         Bone analysis methods for in vivo measurements have included AAS, ASV, ICP-AES,
14    ICP-MS, laser ablation inductively coupled plasma mass spectrometry (LAAICP-MS), thermal
15    ionization mass spectrometry (TIMS), synchrotron radiation induced X-ray emission (SRIXE),
16    particle induced X-ray emission (PIXE), and X-ray fluorescence (XRF).  Since the 1986 Lead
17    AQCD, there have been many new papers published on bone lead using XRF.  The upsurge in
18    popularity of the XRF method has paralleled a decline in the use of the other methods.
19         In the past, two main approaches for XRF measurements have been used to measure lead
20    concentrations in bone, the K-shell and L-shell methods.  The K-shell method is now the most
21    widely used, as there have been no further developments in L-shell devices since the early 1990s.
22    The K-shell methods using 57Cd and 109Cd have been described in detail by Somervaille et al.
23    (1989).  Briefly, the K-shell XRF method uses 88.034 keV gamma rays from 109Cd to fluoresce
24    the K-shell X-rays of lead.
25         Since 1986, several investigators have reported refinements to hardware and software to
26    improve the precision and accuracy of XRF measurements and there have been a number of
27    investigations into the precision, accuracy and variability in XRF measurements (e.g., Aro et al.,
28    2000; Todd et al., 2000, 2001, 2002). Todd et al. (2000) provided a detailed discussion of
29    factors that influence the variability and measurement uncertainty, including repositioning,
30    sample measurement duration, overlying tissue, operator expertise, detector resolution, and
31    changes to measurement process over time. Some of these aspects were also discussed by

      December 2005                            6-21       DRAFT-DO NOT  QUOTE OR CITE

-------
 1   Hu et al. (1995). From their cadaver and in vivo measurements, Todd et al. (2000) concluded
 2   that the uncertainty in an individual measurement was an underestimate of the standard deviation
 3   of replicate measurements, suggesting a methodological deficiency probably shared by most
 4   current 109Cd-based K-shell XRF lead measurement systems. In examining the reproducibility of
 5   the bone lead measurements over a 41A> month period, Todd et al. found the average difference
 6   between the XRF results from short term and longer term measurements was 1.2 |ig/g, indicating
 7   only a small amount of variability in the XRF results over a sustained period of time.
 8
 9   6.2.2.2.2   Statistical Methods for Analyzing Bone Lead Concentrations in
10              Epidemiologic Studies
11          In the literature, XRF bone data has typically been reported in two ways:  one involving a
12   methodological approach to assessing the minimum detection limit and the other termed an
13   epidemiologic approach by Rosen and Pounds (1998). In the methodological approach, a
14   minimum detection limit is defined using various methods, including two or three times the
15   square root of the background counts; one, two, or three times the SD of the background; and
16   two times the observed median error. This approach relies upon the minimum detection limit to
17   define a quantitative estimate that is of sufficient precision to be included in the statistical
18   analysis. The following are examples of methodological minimum detection limits for bone lead
19   analyses. Bellinger et al. (1994) observed minimum detection limits, equivalent to the SD, of
20   5.4 |ig/g for tibia and 9.2 jig/g for patella. Using twice the median observed error, Gerhardsson
21   et al. (1993) observed minimum detection limits of 9.8 |ig/g for tibia and 19.1  |ig/g for
22   calcaneus.  For finger bone lead measurements, Chri staffers son et al. (1986) observed a
23   minimum detectable limit of 20 |ig/g, which was equivalent to three times the  square root of the
24   background counts.
25          With the epidemiologic approach, to determine the minimum detection  limit of an
26   instrument all values are used (including negative values), which results in extremely low
27   detection limits.  Rosen and Pounds (1998) noted that this approach yields population bone lead
28   averages that they considered artificially low and inconsistent with observations from many other
29   earlier studies. However, not including values that are negative or below the detection limit, or
30   assigning these values a fixed number for the statistical analysis is also of concern. To examine
31   and compare the two methods used to analyze data at low levels of bone lead concentration,
     December 2005                            6-22        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Kim et al. (1995) performed serial measurements on phantoms containing spiked amounts of
 2    lead.  The results demonstrated that the use of methodological minimum detection limits to
 3    recede low-level observations reduced the efficiency of the analysis and the ability to distinguish
 4    between the phantoms. Using the epidemiologic approach of retaining all point estimates of
 5    measured bone lead concentrations provided less bias and greater efficiency in comparing the
 6    mean or median levels of bone lead of different populations.
 7
 8    6.2.2.3   Bone Lead as a Biomarker of Lead Body Burden
 9    6.2.2.3.1   Uptake of Lead in Bone
10          The dominant compartment for lead in the body is in bones. In human adults, more than
11    90% of the total body burden of lead is found in the bones, whereas bone lead accounts for -70%
12    of the  body burden in children (Barry, 1975). Bone is comprised of two main types, cortical and
13    trabecular. The tibia consists of more than 95% cortical bone, the calcaneus and patella
14    comprise more than 95% trabecular bone, and finger bone is a mixed cortical and trabecular bone
15    although the second phalanx is dominantly cortical.  The cortical and trabecular bones have
16    distinct rates of turnover and lead release, as well as  potentially different associated toxicity
17    implications (Hu et al., 1998). For example, adult tibia has a turnover rate of about 2% per year
18    whereas trabecular bone has a turnover rate of more  than 8% per year (Rabinowitz, 1991). The
19    proportion of cortical to trabecular bone in the human body varies by age, but on average is
20    about  80 to 20 (International Commission on Radiological Protection, 1973). Although not so
21    important for certain types of measurements, the periosteum is of limited dimension and may
22    reflect a bone compartment of more rapid deposition and turnover of lead than the other two
23    types (Skerfving et al., 1993), which would also likely have implications for toxicity, especially
24    for chelation therapy.
25          Much of the understanding of bone structure and metal deposition comes from studies of
26    radioactive elements (e.g., International Commission on Radiological Protection, 1996). Durbin
27    (1992, page 823) suggests that there is "an initial deposition of lead on anatomical bone surfaces
28    with some skewing to the well nourished trabecular surfaces in red marrow, intense deposits at
29    bone growth sites, and later on, a nearly diffuse labeling throughout the bone volume.  For
30    constant intake of lead during growth, it is expected  that lead will be nearly uniformly distributed
31    in the  mineralized bone. Single or irregular intakes during growth are expected to result in

      December 2005                           6-23        DRAFT-DO NOT QUOTE OR CITE

-------
 1    residual buried lines and hotspots superimposed on a relatively uniform diffuse concentration in
 2    bone mineral volume. . .  For example, periosteal and subperiosteal lead deposits in the long
 3    bones, including those of the hands and feet, are likely to be greater than at many other sites,
 4    since bone growth continues at the periosteal surface while the endosteal surface is resorbed."
 5          The importance of bone marrow was also stressed by Salmon et al. (1999), with a key
 6    factor affecting lead uptake into bone being the fraction of bone surface in trabecular and cortical
 7    bone adjacent to active bone marrow. The fraction of total marrow that is red and active
 8    decreases from 100% at birth to about 32% in adulthood (Cristy, 1981). Early lead uptake is
 9    greater in trabecular bone due to its larger surface area and higher metabolic rate.  Of the total
10    bone surface against red marrow, 76% is trabecular and 24% is cortical endosteal (Salmon et al.,
11    1999).  Bone marrow has much lower lead concentrations than bone matrix (Skerfving
12    etal., 1983).
13
14    6.2.2.3.2   Half-Life of Lead in Bone
15          Estimates of the half-life of lead in trabecular bone are partly dependent on the tissue
16    analyzed and the "purity" of the trabecular component (e.g., patella, calcaneus, and phalanx).
17    Earlier estimates of the half-life of lead in trabecular bone ranged from 12 to  19 years (Bergdahl
18    et al., 1998; Gerhardsson et al., 1993). For cortical bone, estimates for the half-life of lead
19    were on the order of 13 to 27 years (Bergdahl et al., 1998; Gerhardsson et al., 1993;
20    Rabinowitz, 1991).
21          With respect to half-lives in bone, recent K-shell XRF bone studies have indicated that
22    earlier concepts of a constant rate of removal of lead from bone throughout adulthood assumed
23    in models of human metabolism (Leggett,  1993; O'Flaherty, 1993) may be incorrect.  In a study
24    of active and retired smelter workers, Brito et al. (2001) suggested that people less than 40 years
25    old had a shorter half-life for the release of lead from the tibia than those older than 40 years,
26    4.9 years (95% CI: 3.6, 7.8) compared to 13.8 years (95% CI:  9.7, 23.8), respectively. Also,
27    they suggested that less intensely exposed subjects with a lifetime averaged blood lead of
28    <25 |ig/dL had a shorter half-life in the tibia (6.2 years [95% CI: 4.7, 9.0]) than those with  a
29    lifetime averaged blood lead >25 |ig/dL (14.7 years [95% CI:  9.7, 29.9]).
30          Even by the end of the sixth decade, -35 to 40%  of skeletal mass consists of
31    unremodelled first generation bone acquired during childhood and adolescence (International

      December 2005                            6-24        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Commission on Radiological Protection, 1973). This statement contrasts with that of O'Flaherty
 2    (1993) who suggested that because of the relatively short half-life of lead in the bones of children
 3    that much of the lead incorporated during active growth would not persist into adulthood.  In a
 4    comparison of lead in tooth dentine and the tibia from young adults who were followed up after a
 5    period of 13 years, Kim et al. (1996) suggested that "pockets" of lead acquired in childhood may
 6    persist into adults. Likewise, McNeill et al. (2000) compared tibia lead levels and cumulative
 7    blood lead indices in a population of 19 to 29 year olds who had been highly exposed to lead in
 8    childhood from the Bunker Hill, Idaho smelter.  They concluded that lead from exposure in early
 9    childhood had persisted in the bone matrix until adulthood.
10
11    6.2.2.3.3   Changes in Bone Lead Concentrations with Age
12          Conventional and XRF analyses of bone have shown significant increases in bone lead
13    with age (Hu et al., 1990, 1996; Kosnett et al., 1994; Morgan et al., 1990).  Kosnett et al. (1994)
14    observed no significant change in bone lead concentrations up to age 20 years, but found an
15    increasing trend with the same slope for men  and women between the ages of 20 to 55 years and
16    an increase to a faster rate in men older than 55 years.  Kosnett et al. reanalyzed earlier cadaver
17    cortical bone data of Drasch et al. (1987) and found that male bone lead values increased
18    significantly after age 40 years, whereas female values slightly declined. A similar analysis of
19    the post-mortem data of Barry (1975) showed an upward inflection for all males after age
20    35 years. Kosnett et al. (1994) found no significant slope to the relationship between age and
21    bone lead for the  10 to 20 year old subjects, in contrast to Barry (1975) and Drasch  et al. (1987).
22          Annual increments of lead to bone vary although no attempt has been made to determine
23    whether the differences are significant.  For example, the annual increment of 0.46 |ig/g bone
24    mineral/year found by Gordon et al. (1993) was slightly lower than that found by Somervaille
25    et al. (1989), but the difference was not significant. After age 20 years, Kosnett et al. (1994)
26    found the annual increment to be 0.38 jig/g bone mineral/year. Hu et al. (1990)  reported a value
27    of 0.31 |ig/g bone mineral/year for subjects ranging in age from 20 to 58 years.
28
      December 2005                           6-25        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.2.4    Distribution of Lead from Bone into Blood and Plasma
 2    6.2.2.4.1   Contribution of Bone Lead to Blood Lead
 3         Although the skeleton was recognized as a potentially significant contributor to blood lead
 4    in the 1986 Lead AQCD, there have been several investigations using both bone lead XRF and
 5    stable lead isotope methods which have helped quantify the contribution. The earlier estimation
 6    of skeletal contribution to blood lead was 70% by Manton (1985) and -65% ranging up to 100%
 7    by Schiitz et al. (1987). The more recent isotope studies confirmed these estimates.  Using
 8    female immigrants to Australia and their children, Gulson et al. (1995, 1997, 1999a) found a
 9    mean value of 50% (range 16-73%) deriving from the skeleton.  Smith et al. (1996) found a
10    range of 40-70% in five patients who underwent total hip or knee joint replacement. Gwiazda
11    et al. (2005) observed a range of 40-65% in two children and >90% in one child.  Studies
12    examining the bone lead contribution to blood lead are presented in Annex Table AX6-2.6.
13         The contribution of skeletal lead to blood lead was further examined in females from
14    varying environments. In middle-aged to elderly subjects (46-74 years), an increase of 19 |ig/g
15    of lead in tibia bone mineral was associated with an increase in blood lead of 1.7  |ig/dL, which
16    corresponds to a 0.09 |ig/dL increase in blood lead per 1 jig/g bone mineral (Korrick et al.,
17    2002). A study of 108 former workers at the Bunker Hill smelter in northern Idaho and
18    99 referents from the Spokane, WA area examined the endogenous bone lead release rate of
19    postmenopausal and premenopausal women (Popovic et al., 2005). The results indicated that the
20    endogenous release rate in postmenopausal women (0.13 |ig/dL per jig/g bone) was greater than
21    the rate found in premenopausal women (0.07 |ig/dL per |ig/g bone).  In a Mexico City study, the
22    endogenous bone lead release rate in postmenopausal women also was observed to be double
23    that in premenopausal women (Garrido-Latorre et al., 2003). A change of 10  |ig/g bone mineral
24    was associated with an increase in blood lead of 1.4 |ig/dL in postmenopausal subjects,
25    compared to an increase of 0.8 |ig/dL in premenopausal women. Lactation was also found to
26    affect the endogenous bone lead release rate. After adjusting for patella lead concentration, an
27    increase in blood lead levels of 12.7% (95% CI: 6.2, 19.6) was observed for women who
28    practiced partial lactation and an increase of 18.6% (95% CI:  7.1, 31.4) for women who
29    practiced exclusive lactation compared to those who stopped lactation (Tellez-Rojo et al., 2002).
30         The mean cortical  leads to current blood lead ratios for occupationally-exposed subjects
31    are shown in Figure 6-2.6. Box  plots were calculated using data from the following studies:

      December 2005                            6-26        DRAFT-DO NOT QUOTE OR CITE

-------
                         6.0
                      TO
                      C£
                      T3
                      S 40-1
                      _i
                      TJ
                      O
                      _O
                      m
                      •o
                      03
                      
-------
 1    reflects mainly ongoing exposure whereas that in retired employees is more dependent on lead
 2    released from the skeleton.
 3          The mean tibia lead to current blood lead ratios for environmentally-exposed subjects is
 4    shown in Figure 6-2.7. The box plot for pregnancy-related subjects was calculated using data
 5    from the following studies:  Brown et al., 2000; Chuang et al., 2001; Ettinger et al., 2004; Gomaa
 6    et al., 2002; Gonzalez-Cossio et al., 1997; Hernandez-Avila et al., 1996, 1998, 2002, 2003;
 7    Hu et al., 1996; Moline et al., 2000; Rothenberg et al., 2000; Sanin et al., 2001; Tellez-Rojo
 8    et al., 2002, 2004. The box  plot for middle-aged and elderly subjects included the following
 9    studies:  Berkowitz et al., 2004; Cheng et al., 1998a; Garrido-Lattore et al., 2003; Hu et al., 1996,
10    2001; Korrick et al., 2002; Kosnett et al., 1994; Oliveira et al., 2002; Schafer et al., 2005; Tsaih
11    et al., 2004; Webber et al., 1995.  The box plot for the younger subjects (age range 1-30 years)
12    included Farias et al., 1998;  Kim et al., 1996; Rosen et al., 1989; Stokes et al., 1998.  The mean
13    tibia lead to blood lead ratio for pregnancy-related subjects (n = 21) is 1.5 (range 1.0-4.2) and is
14    statistically significantly different (p < 0.001) from the mean ratio of 3.4 (range 1.6-5.4)  for
15    middle-aged to elderly subjects (n = 27).  Similar relationships are observed for the patella lead
16    to blood lead ratios for pregnancy-related subjects and middle-aged to elderly subjects.
17          In several other studies of environmentally-exposed subjects, there is a stronger
18    relationship between patella lead and blood lead than tibia lead and  blood lead (e.g., Hernandez-
19    Avila et al., 1996; Hu et al.,  1996,  1998). Hu et al. (1998) suggest that these relationships
20    indicate that trabecular bone is the predominant bone type providing lead back into circulation
21    under steady-state and pathologic conditions.
22
23    6.2.2.4.2   Partitioning of Bone Lead into Plasma
24          Although most of the lead in whole blood is associated with erythrocytes (-99%), it has
25    been suggested that the small fraction of lead in plasma (<0.3%) may be the more biologically
26    labile and lexicologically active fraction of the circulating lead.  Several authors have proposed
27    that lead released from the skeleton was preferentially partitioned into serum compared with red
28    cells (Cake et al., 1996; Hernandez-Avila et al., 1998; Tsaih et al., 1999) with one explanation
29    being that the lead from endogenous sources was in a different form to that from exogenous
30    sources. However, this concept has been withdrawn by its main proponents.  In contrast to using
      December 2005                            6-28        DRAFT-DO NOT QUOTE OR CITE

-------
                         6.0-
                      8.
                      to
                      cu
                      o
                      _o
                      CO
                      o
                      R3
                      CD
                         4.0
                         2.0
                         0.0-
T
                                                                T
                                    Pregnancy   Middle-aged     Young
                                                 to Elderly
     Figure 6-2.7.   Tibia leads to blood lead ratios for environmentally-exposed pregnancy-
                    related subjects, middle-aged to elderly subjects, and younger subjects.
                    Data compiled from several studies.  See text for more details.
 1   urine as a proxy for serum and measuring lead isotopes, Gulson et al. (2000) concluded that there
 2   was no evidence for preferential partitioning of lead into serum compared with whole blood.
 O
 4   6.2.2.5   Mobilization of Lead From Bone
 5          Although earlier investigators such as Brown and Tompsett (1945), Ahlgren et al. (1976)
 6   and Christoffersson et al. (1984) suggested that the skeleton was a potential endogenous source
 7   of lead poisoning, the opposing concept of the skeleton as a "safe" repository for lead persisted
 8   until the mid-1980s and early 1990s. Potential mobilization of lead from the skeleton could
 9   occur at times of physiological stress associated with enhanced bone remodeling such as during
10   pregnancy and lactation (Hertz-Picciotto et al., 2000; Manton, 1985; Silbergeld, 1991),
11   menopause or in the elderly (Silbergeld,  1991; Silbergeld et al., 1988), extended bed rest
12   (Markowitz and Weinberger, 1990), hyperparathyroidism (Kessler et al., 1999), and
13   weightlessness.  The lead deposited in the bone of adults can serve to maintain blood lead levels
     December 2005
        6-29
DRAFT-DO NOT QUOTE OR CITE

-------
 1    long after exposure has ended (Fleming et al., 1997; Gulson et al., 1995; Inskip et al., 1996;
 2    Kehoe, 1987; Manton, 1985; Nilsson et al., 1991; O'Flaherty et al., 1982;  Schutz et al., 1987;
 3    Smith etal.,  1996).
 4          In the 1986 Lead AQCD, there was a comprehensive summary of chelation therapies and
 5    the recognition that there was limited release of lead from bones. The potential role of bone lead
 6    as an endogenous source of lead in blood, resulting in elevated levels for former lead employees,
 7    was mentioned although data to support this hypothesis were limited.
 8
 9    6.2.2.5.1   Mobilization of Lead from Bone during Pregnancy and Lactation
10          Bone  lead studies of pregnant and lactating subjects are summarized in Annex Table
11    AX6-2.7. Most of the bone XRF studies on pregnancy and lactation have focused on subjects
12    from Mexico City  and Latin subjects from Los Angeles, California.  Relationships and/or health
13    outcomes from these investigations include: patella bone as a significant contributor to blood
14    lead (Brown et al., 2000; Hernandez-Avila et al.,  1996);  a positive association between plasma
15    lead and bone lead in the highest bone lead group of pregnant women (Tellez-Rojo et al., 2004);
16    a positive association of tibia and calcaneus lead with prenatal lead concentration, and calcaneus
17    lead with postnatal lead (Rothenberg et al., 2000); a positive association of tibia lead and
18    seasonal variations in blood lead (Rothenberg et al., 2001);  maternal tibia  and patella lead as
19    significant predictors of fetal exposure determined using cord blood (Chuang et al., 2001);
20    a positive association of calcaneus  lead and increased systolic and diastolic blood pressure in the
21    third trimester (Rothenberg et al., 2002); an inverse relationship between maternal tibia and
22    patella lead,  and birth weight (Gonzalez-Cossio et al., 1997; Sanin et al., 2001); an inverse
23    association between tibia lead and birth length, and patella lead and head circumference
24    (Hernandez-Avila  et al., 2002); an inverse association of maternal patella bone and Mental
25    Development Index (Gomaa et al.,  2002); increased bone resorption during lactation (Tellez-
26    Rojo et al., 2002);  increased lead in breast milk with an increase in patella and tibia lead
27    (Ettinger et al., 2004).
28          Lead isotope studies on immigrant women to Australia (Gulson et al.,  1997, 1998a)
29    confirmed the earlier work of Manton (1985) of increased blood lead during pregnancy.  Gulson
30    et al. reported that, during pregnancy, blood lead concentrations in the first immigrant cohort
31    (n = 15) increased  by an average of about 20% compared to non-pregnant migrant controls

      December 2005                            6-30        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (n = 7).  The percentage change in blood lead concentration was significantly greater during the
 2    postpregnancy period than during the second and third trimesters (p < 0.001). Skeletal
 3    contribution to blood lead, based on the isotopic composition for the immigrant subjects,
 4    increased in an approximately linear manner during pregnancy. The mean increases for each
 5    individual during pregnancy varied from 26% to 99%. Skeletal lead contribution to blood lead
 6    was significantly greater during the postpregnancy period than during the second and third
 7    trimesters.  The contribution of skeletal lead to blood lead during the postpregnancy period
 8    remained essentially constant at the increased level of lead mobilization. In a follow-up study
 9    using a different immigrant cohort of 12 women with  calcium supplementation at the
10    recommended level of approximately 1,000 mg/day (National Institutes of Health,  1994), Gulson
11    et al. (2004) found increased mobilization of lead occurred in the third trimester  rather than in
12    the second trimester as observed with first cohort. In  addition, the extra flux released from bone
13    during late pregnancy and postpartum varied from 50  to 380 jig (geometric mean 145  jig)
14    compared with 330 jig in the previous cohort.
15          In an extended monitoring  of 7 subjects for up  to 22 months postpartum, Gulson et  al.
16    (1999a) found that blood lead concentrations in some  of the subjects decreased to about half the
17    earlier levels almost immediately  after cessation of breastfeeding. However, in 4 of the 7 cases
18    there was a rebound in blood lead concentrations that  exceeded the earlier levels in 3 cases. The
19    authors interpreted these results to indicate that there is ongoing increased mobilization of lead
20    from the maternal skeleton for much longer than predicted, probably associated with remodeling
21    processes. Also using lead isotopes, Manton et al. (2003) observed that blood lead
22    concentrations decreased in early  pregnancy and rose  during late pregnancy.  They attributed
23    these results to changes in bone resorption with decoupling of trabecular and cortical bone sites.
24
25    6.2.2.5.2   Transplacental Transfer of Lead and Transfer through Breast Milk
26          Transplacental transfer of lead in humans has been suggested in a number of studies based
27    on cord blood to maternal blood lead ratios ranging from about 0.6 to 1.0 at the time of delivery.
28    Maternal-to-fetal transfer of lead appears to be related partly to the mobilization  of lead from the
29    maternal skeleton. Evidence for transfer of maternal bone lead to the fetus has been provided
30    from stable lead isotope studies in cynomolgus monkeys (Macacafascicularis).  Approximately
31    7 to 39% of the maternal lead burden that is transferred to the fetus appears to derive from the

      December 2005                           6-31       DRAFT-DO NOT QUOTE OR CITE

-------
 1    maternal skeleton (Franklin et al., 1997; O'Flaherty et al., 1998). Further evidence for maternal-
 2    to-fetal transfer of lead in humans can be gained from stable lead isotope measurements. For
 3    example, a 0.99 correlation in lead isotopic ratios for maternal and cord blood (Manton, 1985;
 4    Gulson et al., 1998b) and the similarity of isotopic ratios in maternal blood and in blood and
 5    urine of newly-born infants provide strong evidence for placental transfer of lead to the fetus
 6    (Gulson et al., 1999b).
 7          Breast milk can also be a pathway of maternal excretion of lead. However, given the very
 8    low lead concentrations and analytical difficulties arising from high fat contents in breast milk,
 9    their analyses require careful attention.  Selected studies appear to show a linear relationship
10    between breast milk and maternal whole blood with the percentage of lead  in breast milk
11    compared with whole blood of <3% in subjects for blood lead concentrations ranging from 2 to
12    34 |ig/dL.  Blood lead concentrations in breastfed newborn infants decreased in spite of the
13    maternal blood lead concentrations having risen or remained elevated postpartum compared to
14    lower levels during prepregnancy or in the first trimester (Gulson et al., 1999b).  Similar trends
15    were noted by Manton et al. (2000). However, in a Mexico City study, an association between
16    patella lead and blood lead concentrations was higher for women with partial lactation than for
17    those who stopped lactation, and it was increased among women who breastfed exclusively
18    (Tellez-Rojo et al., 2002). In another Mexico City study, Ettinger et al. (2004) concluded that an
19    interquartile increase in patella lead was associated with a 14% increase in  breast milk lead,
20    whereas for tibial lead the increase was -5%.
21          In conclusion, there is evidence that maternal-to-fetal transfer of lead occurs, likely
22    resulting from the mobilization of lead from the maternal skeleton during pregnancy. Breast-fed
23    infants appear to be at greater risk only if the mother is exposed to high lead concentrations
24    either from exogenous sources or endogenous sources such as the skeleton.
25
26    6.2.2.5.3   Mobilization of Lead in Bone During Menopause and in the Elderly
27          Increases in blood lead for postmenopausal women have been attributed to release of lead
28    from the skeleton associated with increased bone remodeling during menopause. Many of the
29    studies have been based on blood lead concentration. Bone lead studies of menopausal and
30    middle-aged to elderly subjects are summarized in Annex Table AX6-2.8.
      December 2005                            6-32       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Overall, the various studies of bone and blood lead levels, as well as hormone
 2    replacement therapy, have provided conflicting outcomes. Hormone replacement therapy alone
 3    or combined with calcium supplementation prevents bone resorption and increases the bone
 4    mineral density in trabecular and cortical bones of women with or without metabolic bone
 5    disease.  The effect of hormone replacement therapy may result in a decrease of lead
 6    mobilization from bone along with a reduction in blood lead concentration levels. Several
 7    studies have found that tibia bone lead levels were higher in women who used hormone
 8    replacement therapy (Popovic et al., 2005; Webber et al., 1995). In contrast, other investigators
 9    have found no association between bone lead and use of estrogens (Berkowitz et al., 2004;
10    Korrick et al., 2002). In addition, some studies observed a decrease in blood lead concentrations
11    associated with hormone replacement therapy (Garrido-Latorre et al., 2003), whereas others
12    observed no association  (Webber et al., 1995).
13          The endogenous release rate of lead from bone in postmenopausal women was double the
14    rate in premenopausal former smelter employees (Popovic et al., 2005) and environmentally-
15    exposed women from Mexico (Garrido-Latorre et al., 2003). In middle-aged to elderly males
16    from the Normative Aging Study, patella lead accounted for the dominant portion of variance in
17    blood lead (Huetal., 1996).
18
19    6.2.2.5.4  Effect of Nutritional Status on Mobilization of Lead from Bone
20          Most studies that  investigated the effect of nutritional status on the mobilization of lead
21    from the skeleton have examined the effects of calcium supplementation. Several studies have
22    suggested that dietary calcium may have a protective role against lead by decreasing absorption
23    of lead in the gastrointestinal tract and by decreasing the mobilization of lead from bone stores to
24    blood, especially during  periods of high metabolic activity of the bone such as pregnancy,
25    lactation, and menopause.  An inverse association between patella lead and low calcium intake in
26    postpartum women has been found (Hernandez-Avila et al., 1996).  In contrast, Rothenberg et al.
27    (2000) observed that dietary calcium intake had no effect on calcaneus lead in women monitored
28    during the third trimester and 1 to 2 months postpartum. Likewise,  no effect from calcium
29    supplementation on bone lead was found amongst lactating women from Mexico City (Tellez-
30    Rojo et al., 2002), although in a follow-up study, Hernandez-Avila et al. (2003) reported a 16.4%
31    decrease in blood lead concentration among women with the highest patella bone lead levels who

      December 2005                            6-33        DRAFT-DO NOT QUOTE OR CITE

-------
 1   were taking supplements. Gulson et al. (2004) observed that calcium supplementation was found
 2   to delay increased mobilization of lead from bone during pregnancy and halved the flux of lead
 3   release from bone during late pregnancy and postpartum.  In another study, women whose daily
 4   calcium intake was 850 mg per day showed lower amounts of bone resorption during late
 5   pregnancy and postpartum than those whose intake was 560 mg calcium per day (Manton et al.,
 6   2003). Tellez-Rojo et al. (2004) observed that plasma lead levels were inversely related to
 7   dietary calcium intake.  Results for whole blood lead were similar but less pronounced.
 8         Some researchers have noted concerns regarding potential lead toxicity resulting from
 9   calcium supplementation. However, Gulson et al. (2001)  observed that lead in calcium or
10   vitamin supplements did not appear to increase blood lead concentrations. No information was
11   available on the effects  of other nutritional  supplements (e.g., iron or zinc) on lead body burden.
12
13   6.2.2.6    Summary of Bone Lead as a Biomarker of Lead Body Burden and Exposure
14         Bone accounts for more than 90% of the total body burden of lead in adults and 70% in
15   children. In addition, the longer half-life of lead in bone, which largely depends on the bone type
16   but is generally estimated in terms of years compared to days for blood lead, allows a more
17   cumulative measure of lead dose.  The more widespread use of in vivo XRF lead measurements
18   in bone and indirect measurements of bone processes with stable lead isotopes since the 1986
19   Lead AQCD have enhanced the use of bone lead as a biomarker of lead body burden.
20         In addition to considering bone lead as an indicator of cumulative lead exposure, lead in
21   the skeleton can also be regarded as a source of lead. Key studies have examined  the
22   contribution of bone lead to blood lead; the preferential partitioning of bone lead into plasma;
23   mobilization of lead from bones during pregnancy, lactation, and menopause; and the role of
24   nutritional supplementation in bone mobilization.
25
26   6.2.3   Lead in Teeth
27   6.2.3.1    Summary of Key Findings from the 1986 Lead AQCD
28         The importance of dentine as a potential indicator of lead exposure was noted in the 1986
29   Lead AQCD. There was more emphasis and optimism on using dentine to assess  lead exposure
30   in this document as the  bone XRF method was in its infancy. The issue of deciduous tooth type
31   was addressed but there was little information on permanent teeth.  The portion  of the tooth

     December 2005                           6-34        DRAFT-DO NOT QUOTE OR CITE

-------
 1    analyzed (i.e., whole tooth or circumpulpal dentine) was also addressed. In the 1990 Addendum,
 2    the use of tooth lead as an exposure metric was described in a number of the longitudinal and
 3    cross-sectional studies.
 4
 5    6.2.3.2   Analytical Methods for Measuring Lead in Teeth
 6          Analytical methods for tooth analysis vary from the most widely used AAS, to energy-
 7    dispersive XRF, laser ablation  inductively coupled plasma mass spectrometry (LA-ICP-MS), and
 8    high precision lead isotopes.
 9          As a standard analytical method has yet to be established for tooth lead analysis, some of
10    the discrepancies in findings between studies could arise from several factors, including
11    differences in tooth type, part of the tooth analyzed, and tooth location. Any real differences
12    among populations are unlikely to be the result of physiological factors such as blood supply to
13    teeth or mineralization rates. As enamel and dentine in different teeth calcify at overlapping but
14    different times (Orban, 1953), they could retain varying amounts of lead.
15          In a systematic evaluation of the magnitude of random errors associated with dentine lead
16    measurements, Fergusson et al. (1989) measured lead concentrations in two  samples of dentine
17    from 996 New Zealand children. They estimated that  15 to 20% of the variance was
18    unexplained. Tests of differences  of means and variances showed no significant differences
19    between the two samples.
20          Lead measurements in deciduous teeth in individuals from urban and remote
21    environments and from polluted environments are presented in Annex Tables AX6-2.9 and
22    AX6 2.10, respectively. Based on the limited number of studies, it would appear that the range
23    in whole deciduous tooth lead for environmentally exposed subjects is about 1-10 jig/g, but the
24    most likely levels are <5  jig/g and probably even <2 jig/g. Studies of whole deciduous teeth
25    from industrial environments, including those in urban settings, are also commonly much less
26    than 10 |ig/g.
27          The utility of circumpulpal  dentine (Shapiro et al., 1973) as the metric of lead exposure in
28    deciduous teeth has not been enthusiastically received.  This is  likely due to the separation
29    difficulties, as well as the limited amount of circumpulpal dentine that may be present when the
30    teeth are resorbed, prior to exfoliation.
      December 2005                           6-3 5        DRAFT-DO NOT QUOTE OR CITE

-------
 1          In another approach to gain more information about exposure during pregnancy and early
 2    childhood, the teeth may be sectioned into dominantly enamel or dominantly dentine. These
 3    samples can then be analyzed for lead isotopic ratios and lead concentrations (Gulson and
 4    Wilson, 1994). Even for children living in lead mining and smelting communities, levels of lead
 5    in the enamel are generally low (<5 |ig/g) and are consistent with other studies of whole teeth.
 6    However, higher levels are observed in the dentine samples (e.g., 32 |ig/g), which likely reflect
 7    the early childhood exposure. Permanent teeth tend to have up to three times the level of lead
 8    compared with deciduous teeth, but the number of studies is very limited.
 9
10    6.2.3.3   Tooth Lead as a Biomarker of Lead Body Burden
11          Compared with the amount of lead in the skeleton, tooth lead is a minor contributor to the
12    body burden of lead. Most of the tooth lead information is based on analyses of deciduous teeth.
13    There is still controversy over the amounts of lead in different whole teeth but it appears that the
14    highest concentrations are in central incisors, with decreasing amounts in lateral incisors,
15    canines, first molars, and second molars.  Teeth from the upper jaw tend to have higher lead
16    concentrations than those from the lower j aw.
17          As teeth accumulate lead, tooth lead levels are generally considered an estimate of
18    cumulative lead exposure. Rabinowitz et al. (1993) found that tooth lead was a better measure of
19    exposure than current blood lead levels; however, it was not a good measure of the child's
20    cumulative exposure from birth to exfoliation due to the mobilization of lead from dentine.
21          Teeth are composed of several tissues formed over the years. Therefore, if a child's lead
22    exposure during the years of tooth formation varied widely, different amounts of lead would be
23    deposited at different rates (Rabinowitz et al., 1993).  This  may allow investigators to elucidate
24    the history of lead  exposure in a child.
25          Gulson and Wilson (1994) advocated the use of sections of enamel and dentine to obtain
26    additional information compared with analysis of the whole tooth (e.g., Fosse et al., 1995;
27    Tvinnereim et al.,  1997). For example, deciduous teeth lead in the enamel provides information
28    about in utero exposure whereas that in dentine from the same tooth provides information  about
29    postnatal exposure until the tooth exfoliates at about 6 to 7 years of age.
      December 2005                           6-36        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.3.4   Relationship between Tooth Lead and Blood Lead
 2          As with bone lead-blood lead relationships, there is interest in understanding more about
 3    potential relationships between tooth lead and blood lead. The tooth lead-blood lead relationship
 4    is more complex than the bone lead-blood lead relationship because of differences in tooth type,
 5    location, and analytical method.
 6          Rabinowitz (1995) used studies which reported values for dentine, whole shed teeth, or
 7    crowns, but discarded those measuring circumpulpal dentine because of the higher values in this
 8    medium.  The mean tooth lead levels varied from 2.8 to 12.7 jig/g and blood lead levels from
                                                                                  rj
 9    6.5 to 17 |ig/dL. In a plot of blood versus tooth lead, Rabinowitz found a good fit (R = 0.97;
10    p< 0.0001) with the relationship:
11
12              Tooth Lead (|ig/g) = P x [Blood Lead Og/dL)], where P = 0.49 (SE 0.04).
13
14          In an earlier Boston study, Rabinowitz et al. (1989) found that the association between
15    tooth and blood lead increased with age, first achieving statistical significance at 18 months;
16    by 57 months, the correlation  coefficient was 0.56. A correlation of 0.47 was found between
17    current blood lead and incisors amongst 302 German children (Ewers et al., 1982).
18
19    6.2.3.5   Mobilization of Lead from Teeth
20          Although mobilization of lead from bone appears well established, this is not the case for
21    lead in teeth. Conventional wisdom has lead fixed once it enters the tooth. Although that may
22    be the case for the bulk of enamel, it is not true for the surface of the enamel and dentine.
23          In evaluating deciduous teeth data, Rabinowitz et al. (1993) suggested that their data were
24    compatible with a model that  allows lead to be slowly removed from dentine. Such a process
25    may be associated with resorption of the root and dentine that precedes exfoliation, which allows
26    reequilibration of dentine lead with blood lead.
27          In children exposed to  lead sources from mining, paint, or petrol in communities such as
28    the Broken Hill lead mining community, Gulson and Wilson (1994) and Gulson (1996) showed
29    that the source of lead from the incisal (enamel) sections was different from the source of lead in
30    the cervical (dentine) sections of deciduous teeth, reflecting the change in lead from in utero
31    exposure to early childhood. Based on changes in the isotopic composition of enamel and


      December 2005                            6-37        DRAFT-DO NOT QUOTE OR CITE

-------
 1    dentine in deciduous teeth sections from the Broken Hill mining community children, Gulson
 2    (1996) estimated that lead is added to dentine at a rate of approximately 2-3% per year.
 3          Stable lead isotopes and lead concentrations were measured in the enamel and dentine of
 4    permanent (n = 37) and deciduous teeth (n = 14) from 47 European immigrants to Australia to
 5    determine whether lead exchange occurs in teeth and how it relates to lead exchange in bone
 6    (Gulson et al., 1997). The authors concluded that enamel exhibited no exchange of its European-
 7    origin lead with lead from the Australian environment, whereas dentine lead exchanged with
 8    Australian lead to the extent of ~1 ± 0.3% per year.
 9
10    6.2.3.6    Summary of Tooth Lead as a Biomarker of Lead Body Burden and Exposure
11          Tooth lead is a minor contributor to the total body burden of lead. Moderate-to-high
12    correlations have been observed between tooth lead levels and blood lead levels. Differences in
13    tooth type, part of the tooth analyzed, and tooth location may contribute to some of the
14    discrepancies in findings between studies of tooth lead. As teeth are composed of several tissues
15    formed over the years, if a child's lead  exposure during the  years of tooth formation varied
16    widely, different amounts of lead would be deposited at different rates.  Deciduous teeth lead in
17    the enamel provides information about  in utero exposure, whereas that in dentine provides
18    information about postnatal exposure until the tooth exfoliates.
19
20    6.2.4  Lead in Urine
21    6.2.4.1    Summary of Key Findings  from the 1986 Lead AQCD
22          The 1986 Lead AQCD provided an extensive discussion of the physiological basis for
23    "chelatable" urinary lead. Also discussed was lead excretion provoked by EDTA, including the
24    pools of lead in the body that might be  mobilized in the EDTA provocation test, and the
25    relationship between the outcome and blood lead concentration. The 1986 Lead AQCD noted
26    observations that formed the basis for application of the EDTA provocation test for detecting
27    elevated lead body burden.
28
29    6.2.4.2    Analytical Methods for Measuring Lead in Urine
30          Standard methods that have been reported for urine lead analysis are summarized in
31    Annex Table AX6-2.1 and are, in general, the same as those analyses noted for determination of

      December 2005                          6-3 8        DRAFT-DO NOT QUOTE OR CITE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
lead in blood. Reported detection limits are approximately 50 |ig/L for AAS, 5-10 |ig/L for ICP-
AES, and 4 jig/L for ASV for urine lead analyses.  Sample preparation usually consists of wet
ashing; however, chelation and solvent extraction has also been reported (National Institute for
Occupational Safety and Health, 1994, 1977a).

6.2.4.3   Levels of Lead in Urine
       A summary of selected measurements of urine lead levels in humans can be found in
Annex Table AX6-2.11.  Urine lead concentrations in the U.S. general population have been
monitored in NHANES.  Data from the most recent survey (NHANES IV, Centers for Disease
Control, 2005) for subjects >6 years of age are shown in Table 6-2.4. The geometric mean for
the entire sample (n = 2,689) was 0.64 |ig/g creatinine (95% CI: 0.60, 0.68).  The geometric
means  for males (n = 1,334) and females (n = 1,335) were 0.64 |ig/g creatinine (95% CI: 0.61,
0.67) and 0.64 |ig/g creatinine (95% CI: 0.59, 0.69), respectively. These values correspond to
approximately 1-1.3 jig lead/day for an adult, assuming a daily creatinine excretion rate of
approximately 1.5 g/day in adult females, a body weight of 70 kg for males and 58 kg for
females,  and a lean body mass fraction of 0.88 for males and 0.85 for females (Forbes  and
Bruining, 1976; International Commission on Radiological Protection, 1981).
          Table 6-2.4. Urine Lead Concentrations in U.S. by Age, NHANES IV (1999-2002)
Age
Survey Period
n
Urine Lead
6-11 years
1999-2000
340
1.17
(0.98, 1.41)
2001-2002
368
0.92
(0.84, 1.00)
12-19 years
1999-2000
719
0.50
(0.46, 0.54)
2001-2002
762
0.40
(0.38,0.43)
^20 years
1999-2000
1406
0.72
(0.68, 0.76)
2001-2002
1559
0.66
(0.62, 0.70)
       aUrine lead concentrations presented are geometric means (95% CI) of ug lead/g creatinine.
20          Geometric mean urinary lead excretion rates of 7-10 |ig/g creatinine (maximum 43) have
21    been reported in groups of children living in areas impacted by lead smelting operations
22    (Brockhaus et al., 1988).  Daily urinary lead excretion can exceed 200 |ig/day in association with
      December 2005
                                          6-39
DRAFT-DO NOT QUOTE OR CITE

-------
 1    occupational exposures (Biagini et al., 1977; Cramer et al., 1974; Lilis et al., 1968; Lin et al.,
 2    2001; Wedeenetal., 1975).
 3
 4    6.2.4.4   Urine Lead as a Biomarker of Lead Body Burden
 5          Urine is a major route of excretion of absorbed lead (Chamberlain et al., 1978; Griffin
 6    et al., 1975; Kehoe, 1987; Rabinowitz et al., 1976). The kinetics of urinary excretion following a
 7    single dose of lead is similar to that of blood (Chamberlain et al., 1978), likely due to the fact
 8    that lead in urine derives largely from lead in blood plasma.  Evidence for this is the observation
 9    that urinary lead excretion is strongly correlated with the rate of glomerular filtration of lead (i.e.,
10    glomerular filtration rate x plasma lead concentration; Araki et al.,  1986). Estimates of urinary
11    clearance of lead from serum (or plasma) range from 13-22 L/day, with a mean of 18 L/day
12    (Araki et al., 1986; Chamberlain et al., 1978; Manton and Cook, 1984; Manton and Malloy,
13    1983). Estimates of blood-to-urine clearance, on the other hand, range from 0.03-0.3 L/day with
14    a mean of 0.12 L/day (Araki et al., 1990; Berger et al., 1990; Chamberlain et al., 1978; Gulson
15    et al., 2000; Koster et al., 1989; Manton and Malloy, 1983; Rabinowitz et al., 1976, 1973; Ryu
16    et al., 1983; see Diamond, 1992 for an analysis of these data), consistent with a plasma to blood
17    concentration ratio of approximately 0.005-0.01 L/day (U.S. Environmental Protection Agency,
18    2003). Based on the above,  urinary excretion  of lead can be  expected to reflect the concentration
19    of lead in plasma and variables that affect delivery of lead from plasma to urine (e.g., glomerular
20    filtration and other transfer processes in the kidney).
21          Plasma lead makes a  small  contribution (<1%) to the blood lead concentration and a
22    negligible contribution to total lead body burden. Furthermore, the kinetics of elimination of
23    lead from plasma is fast, relative to lead in bone, where most of the lead burden resides.
24    Therefore, the basic concepts described for blood as a biomarker for body burden also apply to
25    urine. A single urine lead measurement, or a series of measurements taken over short-time span,
26    is likely a relatively poor index of lead body burden (Figure 6-2.8). On the other hand, long-term
27    average measurements of urinary  excretion can be  expected to be a better index of body burden.
28    In the hypothetical simulation  shown in Figure 6-2.8, both the long-term  average urinary lead
29    excretion rate and the body burden have approximately doubled.
      December 2005                            6-40        DRAFT-DO NOT QUOTE OR CITE

-------
                 20
 30            40
     Age (year)
                                                                 25
                                                                 20
            50
                                                                 25
                20
30             40
    Age (year)
            50
Figure 6-2.8.  Simulation of relationship between urinary lead excretion and body burden
             in adults. An abrupt change in lead uptake gives rise to a relatively rapid
             change in urinary excretion of lead, to a new quasi-steady state, and a
             relatively small change in body burden (upper panel). The long-term
             average urinary lead excretion more closely tracks the pattern of change in
             body burden (lower panel). Simulation based on Leggett (1993) lead
             biokinetics model.
December 2005
       6-41
DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.4.5   Urine Lead as a Biomarker of Lead Exposure
 2          Assuming first-order kinetics, a plasma-to-urine clearance (UC1P) of 13-22 L/day
 3    corresponds to half-time for transfer of lead from plasma to urine of 0.1-0.16 day for a 70 kg
 4    adult who has a plasma volume (VP) of approximately 3 L:
 5
                                                 _ In(2) • Vp
 6                                           
-------
              140
s; 120 -\
CD
T3

i1 100 -
           |   80

           o

           UJ   60 -
           T3
           CO
           CD
           CD
    40 -



    20 -



     0
0
                                 •Child

                                  Adult
                               5            10           15

                                   Lead Intake (ug/kg/day)
                                                           20
              140
 >; 120 H
 CO
"O

 ™ 100 -
           '•g  80 -

           o

           UJ  60 -
           -D
           TO
           CU
           CD
    40 -


    20 -



     0
0
                                 •Child

                                  Adult
                               10            20           30

                                   Lead Intake (ug/kg/day)
                                                           40
Figure 6-2.9.  Simulation of relationship between lead intake and urinary lead excretion in

              adults and children. Predictions are for a 2-year-old child and 30-year-old

              adult, for a constant lead intake (ug/kg/day). The relationship is linear, for

              intake and plasma lead concentration (not shown). Predictions are based on

              Leggett (1993, upper panel) and O'Flaherty (1993,1995, lower panel).
December 2005
                              6-43
                                  DRAFT-DO NOT QUOTE OR CITE

-------
 1          Urine flow rate can vary by a factor or more than 10, depending on the state of hydration
 2    and other factors that affect glomerular filtration rate and renal tubular reabsorption of the
 3    glomerular filtrate.  All of these factors can be affected by lead exposure at levels that produce
 4    nephrotoxicity (i.e., decreased glomerular filtration rate, impaired renal tubular transport
 5    function; see Section 6.4 for discussion of effects of lead on the renal system). Therefore, urine
 6    lead concentration measurements provide little reliable information about exposure (or lead body
 7    burden), unless they can be adjusted to account for unmeasured variability in urine flow rate
 8    (Arakietal., 1990).
 9          A determination of urinary lead excretion rate requires measurement of two variables,
10    urine lead concentration, and urine flow rate; the later requires collection of a timed urine
11    sample, which is often problematic in epidemiologic studies.  Collection of un-timed ("spot")
12    urine samples, a common alternative to timed samples, requires adjustment of the lead
13    measurement in urine to account for variation in urine flow (Diamond, 1988). Several
14    approaches to this adjustment have been explored, including adjusting the measured urine lead
15    concentration by the urine creatinine concentration, urine osmolality, or specific gravity (Araki
16    etal., 1990).
17          The measurement of lead excreted in urine following an injection (intravenous or
18    intramuscular) of the chelating agent calcium disodium EDTA (EDTA provocation) has been
19    used to detect elevated body burden  of lead in adults (Biagini et al., 1977; Lilis et al., 1968;
20    Wedeen, 1992; Wedeen et al.,  1975) and children (Chisolm et al., 1976; Markowitz and Rosen,
21    1981). EDTA-provoked urinary lead excretion has been shown to correlate with tibia bone lead
22    measurements (Wedeen, 1992). Given the difficulties associated with the parenteral
23    administration of EDTA, XRF measurements of bone lead, offer  a more feasible alternative to
24    the EDTA provocation test for assessment of bone lead stores in epidemiologic studies. More
25    recently, DMSA (DMSA-provocation) has been used as an orally-effective alternative to EDTA
26    and has been applied to epidemiologic studies as dose metric for lead body burden (e.g., Lee
27    et al., 2001; Schwartz et al., 2001a, 2000a, 2000b).
28
29    6.2.4.6   Summary of Urine Lead as a Biomarker of Lead Body Burden and Exposure
30          Similar to blood lead concentration measurements, urinary lead excretion measured in an
31    individual at a single point in time will reflect the recent exposure history of the individual and

      December 2005                           6-44        DRAFT-DO NOT QUOTE OR CITE

-------
 1    physiological variables that determine the plasma lead concentration time profile. Longitudinal
 2    measurements of urinary lead excretion can be expected to provide a more reliable measure of
 3    exposure history of an individual and will more closely parallel body burden than will single
 4    measurements; however, the degree to which this will apply will depend on the sampling
 5    frequency with respect to the exposure temporal pattern.
 6          Although, in general, higher urinary lead excretion can be interpreted as indicating higher
 7    exposures (or lead uptakes), it does not necessarily predict appreciably higher body burdens.
 8    Similar urinary lead excretion rates in two individuals (or populations) do not necessarily
 9    translate to similar body burdens or similar exposure histories.
10          Measurement of the urinary lead excretion rate requires either a timed urine sample, or an
11    approach to adjusting measured urinary lead concentrations for variability in urine flow rate,
12    which by itself may be affected by lead  exposure (i.e., lead-induced nephrotoxicity). Both
13    approaches, timed urine samples or adjustment of concentration, introduce complications into the
14    assessment and uncertainties into the interpretation of urinary lead measurements as biomarkers
15    of lead body burden or exposure. The EDTA-provocation test provides a more reliable indicator
16    of elevated body burden than do measurements of basal lead excretion; however, it is not feasible
17    to apply this test for epidemiologic investigations.  The DMSA-provocation test may provide a
18    more feasible alternative.
19
20    6.2.5   Lead in Hair
21    6.2.5.1   Summary of Key Findings from the 1986 Lead AQCD
22          The 1986 Lead AQCD did not discuss applications of hair lead measurements for
23    assessing lead body burden or exposure.
24
25    6.2.5.2   Analytical Methods for Measuring Lead in Hair
26          Methods used for hair lead analysis are summarized in Annex Table AX6-2.1.  Wilhelm
27    et al. (1989) reported a detection limit of 0.16 |ig/g for GFAAS; use of GFAAS for hair lead
28    measurements has been reported elsewhere (Annesi-Maesano et al., 2003).  Gerhardsson  et al.
29    (1995a) reported a detection limit of 0.5 jig/g for XRF of the hair shaft; but Campbell and
30    Toribara (2001) found XRF to be unreliable for hair root lead determinations. Use of other
      December 2005                           6-45        DRAFT-DO NOT QUOTE OR CITE

-------
 1    methods has been reported, including ICP (Tuthill, 1996), ET/AAS (Drasch et al., 1997), and
 2    AAS (Sharma and Reutergardh, 2000; Esteban et al., 1999).
 3
 4    6.2.5.3   Levels of Lead in Hair
 5          A summary of selected measurements of hair lead levels in humans can be found in
 6    Annex Table AX6-2.12.  Reported hair lead levels vary considerably. Esteban et al. (1999)
 7    reported a geometric mean levels of 5.4 ng/g (range 1-39) for a sample of 189 children (aged
 8    1.9 to 10.6 years) residing in Russian towns impacted by smelter and battery plant operations.
 9    By contrast, Tuthill (1996) reported much higher levels in a sample of Boston, MA children
10    (aged 6.5 to 7.5 years, n = 277). Approximately 41% had levels that ranged from 1 to 1.9 |ig/g.
11    DiPietro et al. (1989) reported a geometric mean hair lead level of 2.42 jig/g (10-90th percentile
12    range <1.0-10.8) in a general population sample  of U.S. adults (aged 20 to 73 years, n = 270).
13    In a post-mortem sample of the general population from Germany (aged 16 to 93 years, n = 150),
14    the median hair lead level was 0.76 jig/g (range 0.026-20.6) (Drasch  et al., 1997). Also,
15    Gerhardsson et al. (1995a) reported median values for postmortem samples of 8.0 jig/g (range
16    1.5-29,000) in active workers (n = 6), 2.6 |ig/g (range 0.6-9.3) in retired workers (n = 23), and
17    2.1 |ig/g (range 0.3-96) in a reference group (n = 10).
18
19    6.2.5.4   Hair Lead as a Biomarker of Lead Body Burden
20          Lead is incorporated into human hair and hair roots (Bos et al., 1985; Rabinowitz et al.,
21    1976) and has been explored as a possibly noninvasive approach for estimating lead body burden
22    (Gerhardsson et al., 1995a; Wilhelm et al., 1989, 2002). Hair lead measurements are subject to
23    error from contamination of the surface with environmental lead and contaminants in artificial
24    hair treatments (i.e., dyeing, bleaching, permanents) and are a relatively poor predictor of blood
25    lead concentrations, particularly at low levels (<12 |ig/dL) (Campbell and Toribara, 2001;
26    Drasch et al., 1997; Esteban et al., 1999).  Studies evaluating quantitative relationships between
27    hair lead and lead body burden have not been reported.  Nevertheless, hair lead levels have been
28    used as a dose metric in some epidemiologic studies (e.g., Annesi-Maesano et al., 2003; Esteban
29    et al., 1999; Gerhardsson et al., 1995a; Powell et al., 1995; Sharma and Reutergardh, 2000;
30    Tuthill, 1996).
31

      December 2005                            6-46        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.2.5.5    Hair Lead as a Biomarker of Lead Exposure
 2         Rabinowitz et al. (1976) measured hair lead levels in two adult males who received
 3    a stable lead isotope supplement to their dietary intake for 124-185 days.  Approximately 1% of
 4    the daily lead intake was recovered in hair. Temporal relationships between exposure levels and
 5    kinetics and hair lead levels, and kinetics of deposition and retention of lead in hair have not
 6    been evaluated. Higher hair lead levels were observed in lead workers than in reference subjects
 7    with lower blood lead levels (Mortada et al., 2001).
 8
 9    6.2.5.6    Summary of Hair Lead as a Biomarker of Lead Body Burden and Exposure
10         Although hair lead measurements have been used in some epidemiologic studies, an
11    empirical basis for interpreting hair lead measurements in terms of body burden or exposure has
12    not been firmly established. Hair lead measurements are subject to error from contamination of
13    the surface with environmental lead and contaminants in artificial hair treatments (i.e., dyeing,
14    bleaching, permanents) and, as such, are relatively poor predictor of blood lead concentration,
15    particularly at low levels (<12 |ig/dL).
16
17
18    6.3    NEUROTOXIC EFFECTS OF LEAD
19         This section assesses epidemiologic evidence for neurotoxic effects of lead exposure in
20    children and adults.  First presented are studies of the neurotoxic effects of lead on children, with
21    a focus on several prospective studies examining neurocognitive ability. Other topics include
22    measures of academic achievement, cognitive abilities, disturbances in behavior, mood, and
23    social conduct, measures of brain anatomical development and activity, gene-environmental
24    interaction, and reversibility of neurodevelopmental deficits.  Then, neurotoxic effects of
25    environmental and occupational lead exposure of adults are discussed.
26
27    6.3.1    Summary of Key Findings on Neurotoxic Effects of Lead in Children
28            from 1986 Lead AQCD and Addendum, and 1990 Supplement
29         The 1986 Lead AQCD stated that children were particularly susceptible to lead-induced
30    neural damage. In particular, human infants and toddlers below that age of 3 years were
31    considered to be at special risk due to their in-utero exposure, increased opportunity for exposure

      December 2005                           6-47        DRAFT-DO NOT QUOTE OR CITE

-------
 1    because of normal mouthing behavior of lead-containing objects, and increased rates of lead
 2    absorption due to factors such as iron and calcium deficiencies.
 3          Effective blood lead levels for producing encephalopathy or death in children were noted
 4    in the 1986 Lead AQCD as starting at 80-100 |ig/dL.  Various types of neural dysfunction were
 5    stated as being evident at lower blood lead levels. Behavioral (e.g., reaction time, psychomotor
 6    performance) and electrophysiological (e.g., altered electrophysiological patterns, evoked
 7    potential measures, and  peripheral nerve conduction velocities) effects were observed at blood
 8    levels as low as  15-30 |ig/dL and possibly lower.  A concentration-response relationship between
 9    blood lead levels and IQ also was observed; a 1-2 point difference in IQ was generally seen with
10    blood lead levels in the  15-30 |ig/dL range.  However, a study by Schroeder and Hawk (1987)
11    reported a highly significant linear relationship between a measure of IQ and blood lead levels
12    over the range of 6 to 47 |ig/dL in a cohort of all African American children of low SES,
13    suggesting that IQ effects might be detected even at these low levels.
14          The 1986 Addendum discussed the newly published results of several prospective cohort
15    studies on the developmental effects of lead in children. These studies improved upon the
16    previous studies with longitudinal study design that followed children from the prenatal stage, a
17    larger number of subjects, and better analytic techniques to more accurately measure blood lead
18    levels. The four prospective studies (conducted in Boston, MA; Cincinnati, OH; Cleveland, OH;
19    and Port Pirie, Australia) reported significant associations between prenatal and postnatal blood
20    lead levels and neurobehavioral deficits,  after adjusting for various potential confounding factors
21    such as maternal IQ and HOME  (Home Observation for Measurement of Environment) scores
22    (Bellinger et al., 1984; Dietrich et al., 1986; Ernhart et al.,  1985, 1986; McMichael et al., 1986;
23    Vimpani et al., 1985; Wolf et al., 1985).  In these studies, the observed maternal and cord blood
24    lead levels were fairly low, with  mean levels of approximately 10 |ig/dL.  These results led the
25    1986 Addendum to conclude that neurobehavioral deficits, including declines in Bayley Mental
26    Development Index (MDI) scores and other assessments of neurobehavioral function, are
27    associated with prenatal blood lead exposure levels on the order of 10 to 15 |ig/dL and possibly
28    even lower, as indexed by maternal or cord blood lead concentrations.
29          The 1990 Supplement updated evidence from the above-mentioned longitudinal cohort
30    studies and summarized results from other more recent prospective cohort studies conducted in
31    Glasgow, Scotland; Kosovo, Yugoslavia; Mexico City; and Sydney, Australia.  Results from

      December 2005                             6-48        DRAFT-DO NOT QUOTE OR CITE

-------
 1    several other international cross-sectional studies also were discussed. The collective evidence
 2    from the various prospective cohort and cross-sectional studies reaffirmed the conclusions from
 3    the 1986 Addendum that neurobehavioral effects were related to blood lead levels of 10 to
 4    15 |ig/dL and possibly lower. Further analyses of the Boston data indicated that deficits in MDI
 5    could be detected in relation to cord blood lead levels of 6-7 |ig/dL in children within the lower
 6    strata for SES (Bellinger et al., 1988). In the Port Pirie study, the relationship between postnatal
 7    blood lead levels and MDI at two years of age provided little evidence of a threshold effect
 8    (Wigg et al., 1988). Restricting the analysis to children with blood lead levels below 25 |ig/dL
 9    yielded an even stronger association between integrated postnatal blood lead  and McCarthy
10    General Cognitive Index (GCI) scores in the Port Pirie study (McMichael et al., 1988).
11          Impaired neurobehavioral development was associated with blood lead measures in
12    pregnant women, umbilical cords, and infants up to at least 2 years of age; thus, no distinction
13    could be made as to whether this level of concern applied to only fetuses or infants or preschool -
14    age children. The issue  of the persistence of the neurobehavioral effects from low-level lead
15    exposure also was considered.  Although the Boston and Cincinnati studies provided limited
16    evidence suggesting that the effects of prenatal lead exposure on neurobehavioral development
17    were not permanent, the evidence available to support this conclusion was inadequate.
18
19    6.3.2    Neurotoxic Effects of Lead in Children
20          Several major developments have occurred in lead research on child neurodevelopment
21    following the 1986 Lead AQCD/Addendum and the 1990 Supplement. First, there has been an
22    attempt to broaden outcome assessments beyond neurocognitive deficits.  The earlier emphasis
23    on neurocognitive measures (e.g., MDI, GCI, IQ) in previous studies is understandable from the
24    perspectives of the strong psychometric properties of most of these rigorously standardized
25    measures as well as the immediate pubic health concerns. Examples of other outcomes used to
26    assess neurodevelopment include the number of errors on tests of visual-motor integration, the
27    time required to complete a task assessing manual dexterity, the number of errors and false
28    alarms on a continuous performance test, and the efficiency of short term memory. Additional
29    neurodevelopment outcomes include those which elucidate brain-behavior relationships or the
30    potential real life consequences of early exposure to lead, such as academic and vocational
31    failure and maladjustment to the daily demands of living in a complex society. Thus,

      December 2005                            6-49        DRAFT-DO NOT QUOTE OR CITE

-------
 1    epidemiologic studies of lead neurotoxicity have been expanded to adopt measures of academic
 2    achievement, specific cognitive abilities, behavior and mood, sensory acuities, neuromotor
 3    function, and direct measures of brain anatomical development and activity. Another
 4    development has been the initiation of nutritional and pharmacological intervention studies to
 5    assess the impact of treatment on reducing blood lead levels and preventing or moderating the
 6    degree of harm to the central nervous systems of young children. Also, in addition to blood and
 7    tooth lead, bone lead has emerged as a reliable biomarker of lead exposure. The technology for
 8    the assessment of lead in cortical (tibial) and trabecular (patellar) bone using K-shell X-ray
 9    fluorescence (XRF) has advanced to the point where it could be applied as a reliable and valid
10    index of cumulative lead dose in neuroepidemiologic studies (Aro et al., 1994).
11          In recent years, more studies have investigated the impact of blood lead levels below
12    10 |ig/dL on the developing brain.  Average blood lead levels in U.S. children ages one to five
13    years decreased from 15 |ig/dL to approximately 3 |ig/dL between 1976-1980 and 1991-1994,
14    allowing newer studies to  examine the effects of low level lead on the neurodevelopment of
15    children (Centers for Disease Control, 2000; Pirkle et al., 1998).
16          At the time of the last previous criteria review, it was recognized that estimating a
17    threshold for toxic effects  of lead on the central nervous system entailed difficulties.
18    As discussed in the 1990 Supplement, insults to the human brain may be irreversible, making it
19    difficult to determine whether any measured insult is the result of current or past exposures.
20    An observed effect concurrent with a measured blood lead concentration may be the result of
21    exposure in the child's earlier life in the womb or infancy. There is also the critical question of
22    reversibility or the persistence of lead effects identified in infants and preschoolers into school
23    age and later.  A given effect  observed at younger ages may not persist due to functional
24    compensation or a return to a normal neuromaturational trajectory (Dietrich et al., 1990).
25    Another problem is that it is sometimes difficult to distinguish between neurobehavioral effects
26    due to lead and effects owing to the many social, economic, urban-ecological, nutritional, and
27    other medical factors that  are known to  have important effects on neurobehavioral development.
28    Equally important is the high probability that the concentration-response relationship and even
29    the neurobehavioral lesion associated with childhood lead exposure may vary as a function of
30    these cofactors (Bellinger, 1995).
      December 2005                            6-50       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In the following sections, prospective cohort studies and cross-sectional studies of
 2    neurocognitive ability published since the 1990 Supplement are presented first. Then, studies
 3    examining the effect of lead on a variety of neurodevelopmental outcomes, including academic
 4    achievement; specific cognitive abilities; disturbances in behavior, mood, and social conduct;
 5    sensory acuities; neuromotor function; and brain anatomical development and acuity, are
 6    discussed. This is followed by a presentation of issues involved in understanding lead
 7    neurotoxicity in children, including gene-environment interactions, reversibility of lead effects,
 8    times of vulnerability, and potential threshold levels for effects.
 9
10    6.3.2.1   Neurocognitive Ability
11    6.3.2.1.1   Prospective Longitudinal Cohort Studies of Neurocognitive Ability
12          Several prospective longitudinal cohort studies were initiated in the 1980s because it
13    became widely recognized that the cross-sectional study design was inadequate to  address a
14    number of research issues (U.S. Environmental Protection Agency, 1986; World Health
15    Organization, 1977). These longitudinal studies were characterized by serial measures of dose
16    (blood lead levels) spanning (in most cases) the prenatal and postnatal periods of central nervous
17    system development, thus helping to clarify the temporal association between exposure and
18    insult. Also, developmental assessments that extended into the school-age period were planned
19    to determine if early lead associated neurobehavioral impairments were permanent or reversible
20    in the fullness of time.  It was also determined that assessment of potential confounding factors
21    should be comprehensive and include measures of perinatal  health, nutrition, maternal
22    consumption of other neurotoxicants during pregnancy, parental intelligence, and direct
23    observations of parenting behavior. These studies were also characterized by very careful
24    attention to biostatistical issues and strategies (Bellinger, 1995; Ernhart,  1995).
25          At the time of the 1990 Supplement, studies were underway or planned in the U.S.,
26    Australia, Scotland, the former Yugoslavia, and Mexico. These cohorts  differed in the source
27    and degree of lead exposure and in other important aspects, notably ethnicity and SES.
28    Nevertheless, the early results from several of these studies have been largely responsible for the
29    emergence of the current perspective that blood lead concentrations as low as 10 |ig/dL, or
30    perhaps even lower, may pose a risk for neurodevelopmental toxicity (Davis and Svendsgaard,
31    1987; U.S. Environmental Protection Agency, 1990). Most  of the prospective studies underway


      December 2005                           6-51       DRAFT-DO NOT QUOTE OR CITE

-------
 1    in 1990 continued to follow their subjects into the later preschool and school age years with age-
 2    appropriate measures of intelligence.  Continued follow-up of these cohorts was important due to
 3    the following: (1) greater reliability and precision of measurements attained with assessments of
 4    older children; (2) high predictability of adult intellectual functioning from measures of IQ in the
 5    older child; and (3) examination of potential effects of lead on important abilities that cannot be
 6    easily tapped during  infancy such as executive functions and higher order reasoning (McCall,
 7    1979).
 8          A unique aspect of this research was that most investigators agreed during the formative
 9    stages of their projects to develop somewhat similar assessment protocols (Bornschein and
10    Rabinowitz, 1985).  This has facilitated comparison of results across studies and allowed for
11    sophisticated meta- and pooled-analyses of these data (e.g., Pocock et al., 1994; Schwartz, 1994;
12    World Health Organization,  1995; Lanphear et al., 2005; Rothenberg and Rothenberg, 2005).
13          In the following sections, further updates on the individual prospective cohort studies are
14    presented in chronological order of study initiation. The prospective cohort studies reviewed are
15    summarized in Annex Table AX6-3.1. Results of the meta- and pooled-analyses are presented
16    later in this section.
17
18    Boston Study
19          In the 1986 Addendum, the most advanced investigation at that time was the Boston
20    Prospective Study (Bellinger et al., 1984). The subjects were 216 middle-to upper-middle-class
21    Boston children, 90% of whom had cord blood lead levels below 16  |ig/dL (maximum
22    25 |ig/dL). Cord blood lead  levels in the "high" group (mean 14.6 |ig/dL) were associated with
23    lower covariate-adjusted scores on the Mental Development Index (MDI) of the Bayley Scales of
24    Infant Development  at 6 months of age. It was concluded that although lower level lead
25    exposure in utero may result in delays in early sensorimotor development, the Boston results
26    did not allow estimation of the persistence of these effects nor the public health significance of
27    the findings.  The association between higher cord blood lead and lower MDI persisted to
28    24 months; however no association was observed between postnatal blood lead levels and MDI
29    (Bellinger et al., 1985,  1986).
30          Particular attention was focused on the Boston study, which was among the more mature
31    in terms  of follow-up, in the  1990 Supplement (Bellinger et al., 1987; Bellinger et al., 1991).

      December 2005                            6-52       DRAFT-DO NOT QUOTE OR CITE

-------
 1    With respect to the effects of cord blood lead concentrations on MDI assessed longitudinally
 2    from 6 to 24 months, the lead associated deficits were evident across the entire range of blood
 3    lead levels starting at 10 |ig/dL, which reinforced the previous designation of 10-15 |ig/dL as a
 4    blood level of concern for early neurodevelopmental deficits. At approximately 5 years of age,
 5    cord blood lead levels were not significantly associated with the McCarthy GCI, but blood lead
 6    level at 2 years of age (mean 6.8 |ig/dL [SD 6.3]) was significantly associated with lower scores.
 7    Although cord blood lead concentrations were not independently associated with deficits in
 8    5-year neurocognitive status, the risk of obtaining lower GCI scores was greater among subjects
 9    with higher prenatal and postnatal blood lead concentrations. Boston investigators also
10    examined the relationship between lead measured in shed deciduous teeth obtained from
11    102 children in their cohort (mean 2.8 ppm [SD 1.7]) and GCI at 5 years of age. Prior to
12    covariate-adjustment, there was a very strong and significant relationship amounting to a loss of
13    more than 10 points in GCI for each log increment in dentine lead. However, in the
14    multivariable analysis the tooth lead coefficient, although negative, was no longer statistically
15    significant. Reduced sample size should be taken into consideration in interpreting this null
16    finding.
17          Since the 1990 Supplement, the Boston investigators reexamined 148 of their subjects at
18    10 years of age with the Wechsler Intelligence Scale for Children-Revised (WISC-R)  and other
19    neurobehavioral assessments (Bellinger et al., 1992). They examined the association of WISC-R
20    scores at  10 years of age with blood lead concentrations in the cord blood and at 6 months,
21    12 months, 18 months, 24 months,  57 months, and 10 years.  Only blood lead levels at
22    24 months were significantly associated with full scale and verbal IQ and marginally associated
23    with performance IQ, after adjusting for HOME score, maternal age, birth weight, and maternal
24    IQ. The integrated average blood lead level in this cohort over the first 2 years was 7.0 |ig/dL
25    (range 4-14 |ig/dL).  An increase of 10 |ig/dL in blood  lead level at age 2 was associated with a
26    decrement of 5.8 points (95% CI: 1.8, 9.9) in full scale IQ. These findings indicated that
27    children's performance was  much more strongly associated with blood lead levels at age 2 than
28    with blood lead levels at other ages. It is unclear whether this reflects a special vulnerability of
29    the nervous system during this period or simply the fact that blood lead level tends to peak in the
30    second year.
      December 2005                            6-53        DRAFT-DO NOT QUOTE OR CITE

-------
 1          A reanalysis involving the total Boston cohort that employed nonparametric smoothing
 2    revealed that the inverse association persisted at blood lead levels below 5 |ig/dL (Schwartz,
 3    1994).  Bellinger and Needleman (2003) reanalyzed data on 48 children whose measured blood
 4    lead concentrations never exceeded 10 |ig/dL. Reduction in full scale IQ at 10 years was
 5    significantly associated with blood lead levels at 2 years of age following covariate adjustment.
 6    A larger deficit of 15.6 points (95% CI not presented) per 10 |ig/dL increase in blood lead levels
 7    was observed in this cohort, compared to the 5.8 point deficit observed in the entire cohort.
 8    These findings indicated that the inverse slope might be steeper at blood lead levels below
 9    10 |ig/dL.
10
11    Cincinnati Study
12          Interim results on a partial sample of 185 subjects from a cohort of 305 were available
13    from the Cincinnati prospective study in the 1986 Addendum and the 1990 Supplement (Dietrich
14    et al., 1986, 1987a).  The Cincinnati study investigators reported an inverse relationship between
15    prenatal maternal blood lead levels (mean 8.3 |ig/dL) and 6 month Bayley MDI. This effect was
16    mediated, in part, through lead-associated reductions in birth weight and gestational maturity.
17    A more complete analysis of the full Cincinnati cohort confirmed these interim findings
18    (Dietrich et al., 1987b).
19          Further updates of the Cincinnati  study appeared after the 1990 Supplement.  The
20    Kaufman Assessment Battery for Children (KABC) was administered to approximately
21    260 children at 4  and 5 years of age (Dietrich et al., 1991; 1992). The principal findings at
22    4 years were that higher neonatal blood lead concentrations were associated with poorer
23    performance on all KABC subscales. However, this relationship was confined to children from
24    the poorer families.  Following full covariate adjustment, few statistically significant
25    relationships remained. At 5 years of age, postnatal blood lead levels were associated with
26    performance on all subscales of the KABC; however,  few statistically significant relationships
27    remained after adjustment for covariates. Nevertheless, it is of interest that at both 4 and 5 years
28    the KABC subscale that assessed visual-spatial skills was among those that remained the most
29    highly associated with various indices of postnatal exposure following covariate adjustment.
30    At the age of approximately 7 years, 253 children in the Cincinnati  cohort were administered the
31    WISC-R (Dietrich et al.,  1993a). In this cohort, approximately 35% had at least one blood lead

      December 2005                             6-54       DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentration >25 |ig/dL while 95% exceeded 10 |ig/dL sometime during the first 5 years of life.
 2    Postnatal blood lead concentrations were inversely associated with full scale and performance
 3    IQ, after adjusting for HOME score, maternal IQ, birth weight, birth length, child gender, and
 4    cigarette consumption during pregnancy.  Figure 6-3.1  presents the unadjusted and adjusted
 5    concentration-response relationship between lifetime average blood lead concentrations and
 6    performance IQ.  Following covariate adjustment, a statistically significant relationship was
 7    observed between postnatal blood lead levels at 5 and 6 years of age and full scale IQ. Postnatal
 8    blood lead levels at nearly all ages (including the integrated average blood lead level) were
 9    inversely associated with performance IQ. Blood lead  levels at 6 years of age were most
10    strongly associated with performance IQ - a 5.2 point [95% CI: 2.3, 8.1] decline was observed
11    for each 10 |ig/dL increase in blood lead level. A 10 |ig/dL increase in lifetime average blood
12    lead concentration was associated with a 2.6 point (95% CI:  0.2, 5.0) decline in performance IQ.
13          At 15-17 years of age, the Cincinnati subjects were administered a comprehensive
14    neuropsychological battery (Ris et al., 2004). Variables derived from the Cincinnati
15    neuropsychological battery were subjected to a principal components factor analysis that yielded
16    five factors, including a learning/IQ factor that had high loadings for the Vocabulary and Block
17    Design subtests from the WISC-III as well as the Reading, Spelling,  and Arithmetic subscales of
18    the Wide Range Achievement Test-Revised (WRAT-R).  Prenatal, Average Childhood, and
19    78 month blood lead levels were used in a series of multiple regression analyses. Following
20    covariate-adjustment, there was a trend towards significance for higher blood lead concentrations
21    in later childhood (e.g., 78 months) to be associated with lower learning/IQ factor scores, but this
22    was largely observed in subjects from the lower end of the socioeconomic scale in the sample.
23    This finding is consistent with previous reports that children in the lower social strata may be
24    more vulnerable to general effects on cognitive development and learning (Bellinger, 2000;
25    Winneke and Kraemer, 1984).
26
27    Cleveland Study
28           Early results of the Cleveland prospective study also were reviewed in the 1986
29    Addendum and 1990 Supplement. By selection, about  half of the mothers had histories of
30    alcohol abuse as measured by the Michigan Alcoholism Screening Test. The other women were
31    matched controls. The initial cohort included 389 infants with a mean cord blood lead level of

      December 2005                           6-55        DRAFT-DO NOT QUOTE OR CITE

-------
                         100
g
o
o
c
CO
                          95
                       o
                       t
                       oi  go
                       o
                       CO
                          85
                       o
                       s
                          80
                                      O  Unadjusted p < 0.004
                                      •  Adjusted p < 0.05
 1
0-10
                      23
                      -1 5  >15-20
                                                             4
                                                           >20
                                Average Lifetime Blood Lead (ng/dL)
    Figure 6-3.1.  Unadjusted and adjusted relationships between average lifetime blood lead
                  concentrations and Wechsler Scale performance IQ. Mean + SD lifetime
                  average blood lead concentrations within each category were as follows:
                  0-10 ug/dL, 7.7 ± 1.4 ug/dL (n = 68); >10-15 ug/dL, 12.3 ± 1.4 ug/dL
                  (n = 89); >15-20 ug/dL, 17.1 ± 1.2 ug/dL (n = 53); and >20 ug/dL,
                  26.3 + 5.0 ug/dL (n = 41).
    Source: Dietrich et al. (1993a).
1   5.84 |ig/dL (maximum 14.7). In addition to size, minor morphological anomalies, and 1- and
2   5-minute Apgar performances, infants were evaluated on the Brazelton Neonatal Behavioral
3   Assessment Scale (NBAS) and part of the Graham-Rosenblith Behavioral Examination for
4   Newborns (G-R). Of the 17 neonatal outcomes examined, the NBAS Abnormal Reflexes scale
5   and neurological soft signs assessed by G-R were associated with cord blood lead levels in the
6   range of 3 to 15 |ig/dL following covariate adjustment (Ernhart et al., 1986). A follow-up study
7   observed a significant effect of the neurological soft signs measure on Bayley MDI scores at
8   12 months; however, prenatal lead exposure was not associated with MDI scores at 6-24 months
9   or Stanford-Binet IQ (S-B IQ) scores at 36 months (Wolf et al., 1985).
    December 2005
                      6-56
                     DRAFT-DO NOT QUOTE OR CITE

-------
 1          In 285 children from the original cohort, maternal and cord blood lead levels, as well as
 2    postnatal blood lead levels at 6 months, 2 years, and 3 years were examined in relation to Bayley
 3    MDI, Psychomotor Index (PDI), and Kent Infant Development Scale (KID) at 6 months, MDI at
 4    1 year and 2 years, and S-B IQ at 3 years of age (Ernhart et al., 1987, 1988).  After covariate
 5    adjustment, only maternal blood lead level at delivery (mean 6.5 |ig/dL [maximum 11.8]) was
 6    inversely associated with MDI, PDI, and KID scores at 6 months. No other indices of prenatal or
 7    postnatal lead exposure were inversely associated with assessments of global intellectual
 8    functioning. Language development also was assessed in the Cleveland cohort at  1, 2, and
 9    3 years of age. Few significant associations remained after covariate adjustment (Morrow-
10    Tlulak and Ernhart, 1987).
11          At 4 years and 10 months, 242 children from the Cleveland cohort were administered the
12    Wechsler Preschool and Primary Scale of Intelligence (WPPSI) test.  Significant negative
13    correlations were observed between full scale, verbal, and performance IQ, and prenatal (both
14    maternal and cord) and postnatal (at 2 years and 3 years) blood lead levels (Ernhart and Morrow-
15    Tlulak, 1987). However, these associations were no longer significant after adjustment for
16    various covariates, including HOME score, maternal IQ, parent education, race, medical
17    problems, maternal alcohol use in pregnancy, Michigan Alcoholism Screening Test score,
18    maternal use of marijuana, and several categories of psychosocial trauma scale.  The authors
19    reported a large  contribution of these covariates to the variance of the WPPSI scores; the smaller
20    effects of lead may have been suppressed by these other social factors.  In particular, the HOME
21    score was found to most strongly contribute to the child's IQ.  In 164 children, shed deciduous
22    incisor were collected between ages 5  and 7. Circumpulpal dentine lead levels were found to be
23    significantly associated with full scale, verbal and performance IQ, assessed using the WPPSI
24    test, at 4 years and 10 months, after adjustment for various covariates except for HOME score.
25    After additional  adjusting for HOME score, the lead effects on all three IQ measures diminished,
26    but remained statistically significant for verbal IQ (p = 0.01) and marginally significant for full
27    scale IQ (p  = 0.06). An increase in dentine lead from the 10th percentile to the 90th percentile
28    level (13.5 jig/g to 129.4 |ig/g) was associated with a 6.0 point (95%  CI:  1.4, 10.6) decrease in
29    verbal IQ and a 4.5 point (95% CI:  -0.2, 9.2) decrease in full  scale IQ.  The estimated lead
30    effect increased  as a function of the level of measurement error in the dentine lead variable.
31    This finding of an adverse effect for dentine lead is not consistent with previous analyses of the

      December 2005                             6-57        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Cleveland study showing that blood lead levels are, generally, not associated with cognitive
 2    outcomes after covariate adjustment.
 3
 4    Port Pirie, Australia Study
 5          Preliminary results from the Port Pirie, Australia study also were described in the 1986
 6    Addendum (Vimpani et al., 1985).  Lower Bayley MDI scores at 2 years from 592 children were
 7    significantly associated with higher integrated postnatal blood lead levels (approximately 20% of
 8    the sample had blood lead levels >30 |ig/dL at the time of assessment), but not with maternal
 9    prenatal, delivery, or cord blood lead levels.  Results of this interim analysis were interpreted
10    with caution since important covariates such as maternal IQ and HOME scores were not
11    available for the entire cohort at the time of the  analyses.
12          The Port Pirie cohort study had reported results out to 4 years when the 1990 Supplement
13    was released (McMichael et al., 1988). Following adjustment for covariates, lead concentrations
14    at most postnatal sampling points as well as an integrated average for the 4-year postnatal period
15    were significantly and inversely associated with scores on the McCarthy Scales of Children's
16    Abilities. The GCI scores declined by approximately 4.5 points (95% CI: 0.2, 8.8) for a
17    doubling in blood lead levels.  Similar deficits occurred in the perceptual-performance and
18    memory scores. The integrated postnatal blood lead levels among the 537 children in this cohort
19    were among the highest of the prospective studies (geometric mean 19 |ig/dL).  However, further
20    analyses indicated that the effects observed did  not depend on children with the more extreme
21    levels of exposure.  The concentration-response relationship between blood lead and GCI was
22    stronger among children with blood lead levels  below 25  |ig/dL than it was overall.
23          Of all of the  prospective studies of lead and child development, the Port Pirie cohort study
24    was probably among the best positioned to reliably detect effects of low level lead exposure into
25    later childhood owing to its wide range of exposure, large sample size, and lack of extremes in
26    terms of sample social  advantage or disadvantage.  The WISC-R was administered to
27    494 children between 7 and 8 years of age (Baghurst et al., 1992). IQ scores were examined in
28    relation to In-transformed blood lead concentration. Following adjustment for covariates there
29    was little association with pre- and perinatal lead exposure assessments. However, significant
30    decrements in full scale and verbal IQ were found to be associated with postnatal blood lead
31    levels.  The estimated effect size was a loss of 3.3 points (95% CI: 0.2, 6.5)  in full scale IQ and

      December 2005                           6-58        DRAFT-DO NOT QUOTE OR CITE

-------
 1    4.0 points (95% CI:  0.7, 7.2) in verbal IQ in association with a doubling of the integrated
 2    postnatal blood lead concentration up to three years. In light of the Cincinnati findings, it is of
 3    interest that the Block Design subtest of the WISC-R (a measure of visual-spatial abilities),
 4    exhibited the strongest association with lead exposure. Port Pirie investigators also collected
 5    deciduous central upper incisors from  262 children in their cohort (McMichael et al., 1994).
 6    After covariate adjustment, a significant inverse association was  observed between tooth lead
 7    concentration and WISC-R full scale IQ at 7 years of age. The adjusted estimated decline in full
 8    scale IQ across the tooth lead range from  3 to 22 jig/g (range for 90% of population) was
 9    5.1 points (90% CI:  0.2, 10.0). Once  again, the Block Design subtest was among the most
10    highly sensitive.
11          Port Pirie children were assessed again at 11-13 years of age to examine the persistent
12    relationship between exposure to environmental lead and intelligence (Tong et al., 1996). At that
13    age, Port Pirie investigators were able  to recall 375 children for IQ assessments. At 11-13 years
14    of age, the geometric mean lifetime average blood lead concentration was 14.1 |ig/dL.  WISC-R
15    scores were significantly and inversely associated with integrated lifetime average blood lead
16    concentrations out to 11-13 years.  Later blood lead concentrations after 3 years of age were
17    more predictive of lower IQ.  Mean full scale IQ declined by 3.0 points (95% CI: 0.1,  5.9)  for a
18    doubling of lifetime average blood lead concentrations. The authors  could find no clear evidence
19    of a threshold level in their data.
20
21    Sydney, Australia Study
22          Unlike Port Pirie, the reports on the Sydney cohort study were consistently negative  with
23    respect to the  effects of exposure on neurodevelopment (Cooney et al.,  1989a,b; McBride et al.,
24    1989).  In the  298 mothers and infants sampled,  geometric mean blood  lead levels at delivery
25    were 9.1 |ig/dL and 8.1 |ig/dL, respectively, with less than 2% in excess of 15 |ig/dL. Mean
26    postnatal blood lead levels peaked at 16.4 |ig/dL when children reached 18 months and then
27    declined to 10.1 |ig/dL at 48 months. No significant, inverse relationships were reported
28    between prenatal or postnatal blood lead concentrations and neurodevelopmental assessments
29    conducted from 6 months through 4 years of age. The McCarthy Scales of Children's Abilities
30    was administered to 207 children at 4 years of age but no associations with blood lead levels
31    were observed prior to or following covariate-adjustment. As in the case of the Cleveland study,

      December 2005                            6-59        DRAFT-DO NOT QUOTE OR CITE

-------
 1    the authors noted that the HOME score was a strong contributor to the neurodevelopmental
 2    assessments at all ages.  As stated in the 1990 Supplement, this raises the questions of whether
 3    lead exposure might have covaried with HOME scores.  If so, adjusting for HOME scores would
 4    reduce the statistical power to examine the effect of postnatal blood lead levels on the
 5    neurocognitive measures.  Also note that the interpretation of the Sydney findings has been
 6    complicated by concerns about possible contamination of capillary blood lead samples collected
 7    during the early phases of the investigation (Cooney et al., 1989b).
 8          The Sydney prospective study further assessed 175 subjects that remained in the study at
 9    7 years of age (Cooney et al., 1991).  Geometric mean blood lead concentrations peaked at
10    2 years of age (15.2 |ig/dL). The geometric mean blood lead level at 7 years of age was
11    7.7 jig/dL. The WISC-R and other neurobehavioral assessments were administered.  The
12    adjusted correlations between postnatal blood lead levels and WISC-R scores were consistently
13    negative but nonsignificant at the p = 0.05 level. The r value (units = SD of IQ per SD of blood
14    lead) for the correlation between full  scale IQ and concurrent blood lead at age 7 years was
15    -0.06 (95% CI: -0.20,  0.09).  The correlation coefficient is not significantly different form
16    Bellinger et al. (1992) for 57-month-old children, -0.07 (95% CI:  -0.23, 0.08), or from
17    Lanphear et al. (2005) for children aged 4.8 to 10 years,  -0.20 (95% CI: -0.28, -0.12).
18    All correlation coefficients are for full scale IQ and concurrent blood lead concentrations.
19          Results from this follow-up study were consistent with their earlier reports of no
20    association between blood lead levels <15 |ig/dL and developmental deficits. However, the
21    authors noted that their study was not designed to examine small deficits associated with blood
22    lead levels at this magnitude. They reported that the size of their cohort did not provide
23    sufficient power to detect effects less than 5%.  Cooney  et al. concluded that results from their
24    study indicate that if developmental deficits do occur at blood lead levels below 25 |ig/dL,  the
25    effect size is likely to be less than 5%.
26
27    Mexico City Study
28          Preliminary results of the Mexico City cohort prospective study were presented in the
29    1990 Supplement (Rothenberg et al.,  1989). Blood lead levels from 42 mother-infant pairs were
30    measured at 36 weeks of pregnancy (mean 15.0 |ig/dL) and delivery (mean 15.4 |ig/dL), and
31    in the cord blood (mean 13.8 |ig/dL). The Brazelton NBAS was administered to infants at

      December 2005                           6-60        DRAFT-DO NOT QUOTE OR CITE

-------
 1    48 hours, 15 days, and 30 days after birth. None of the lead measures were associated with the
 2    NBAS outcomes; however, several differential lead measures (i.e., maternal blood lead at
 3    36 weeks of pregnancy minus cord blood lead) were found to be associated with several outcome
 4    variables. Increases in the blood lead of the mother during the last month of pregnancy or a cord
 5    blood lead level higher than the mother's blood lead level were associated with adverse changes
 6    in Regulation of States, Autonomic Regulation, and Gestation Age.
 7          Schnaas et al. (2000) further examined the effect of postnatal blood lead level on
 8    cognitive development in 112 children with complete data from the Mexico City study.  Lead
 9    was measured in blood every 6 months from 6 to 54 months.  Intellectual status was assessed
10    with the McCarthy GCI. The purpose of the study was to estimate the magnitude of the effect of
11    postnatal blood lead level on the GCI and describe how the effect varies with the time between
12    blood lead measurements and the neurocognitive assessments.  The geometric mean blood lead
13    level between 24-36 months was 9.7 |ig/dL (range 3.0-42.7).  A number of significant
14    interactions were observed between blood lead levels and age of assessment. The greatest effect
15    was found at 48 months, with a decrease of 4.0 points (95% CI  not presented) in adjusted GCI
16    score being observed for a doubling of the 24-36 month blood lead level. The authors concluded
17    that 4 to 5 years of age (when children are entering school) appears to be a critical period for the
18    manifestation of earlier postnatal blood lead level effects.
19          In a related study, Gomaa et al. (2002) examined prenatal and postnatal lead exposure
20    effects on the neurodevelopment of 197 children aged 2 years residing in Mexico City. Lead
21    was measured in the umbilical cord and maternal venous blood samples at delivery. Maternal
22    body burden was measured by obtaining cortical (tibial) and trabecular (patellar) bone lead
23    measurements using K-shell XRF within 4 weeks of delivery.  At 2 years of age, the Bayley
24    MDI and PDI were administered.  The major objective of this study was to compare lead levels
25    in umbilical cord blood and maternal bone as independent predictors of infant mental
26    development.  Mean blood lead concentrations in the cord blood, at 12 months of age, and at
27    24 months at age were 6.7 |ig/dL (SD 3.4), 7.2 |ig/dL (SD 2.8), and 8.4 |ig/dL (SD 4.6),
28    respectively.  Mean maternal patella and tibia bone lead levels were 17.8 jig/g (range 
-------
 1    Bayley MDI. Maternal trabecular bone lead levels predicted poorer sensorimotor functioning in
 2    children 2 years of age independent of the cord blood lead level. The authors concluded that
 3    higher maternal trabecular bone lead concentrations constitute an independent risk factor for
 4    impaired mental development in infants at 2 years of age and that this is likely due to the
 5    mobilization of maternal bone lead stores over the course of gestation.
 6
 7    Kosovo, Yugoslavia Study
 8          The neurodevelopment results of a large birth cohort study of 577 children in two towns
 9    in Kosovo, Yugoslavia were not available at the time of the 1990 Supplement. The study took
10    place in Titova Mitrovica, near the site of a longstanding lead smelter, refinery, and battery plant,
11    and in Pristina, a less exposed community 25 miles to the south. A unique characteristic of this
12    cohort was the high prevalence of anemia secondary to iron deficiency  (34% with hemoglobin
13    concentrations <10.5 |ig/dL at 2 years of age). The investigators began providing iron-fortified
14    multivitamin supplements to the entire cohort when the children were between 18 to 38 months
15    of age (Wasserman et al., 1994).
16          Like Port Pirie, this was one of the more highly exposed cohorts. Blood lead levels were
17    obtained during the second trimester, at delivery, from the umbilical  cord and postnatally at
18    6-month intervals to 90 months. At birth, geometric mean cord blood lead levels were nearly
19    21 |ig/dL in the smelter area (Wasserman et al., 1992).  At age 2 years, geometric mean blood
20    lead concentrations were 35.5 |ig/dL and 8.4  |ig/dL among infants  from Titova Mitrovica and
21    Pristina, respectively.
22          Neurocognitive measures of mental abilities were administered at 2, 4, 7, and 10-13 years
23    of age.  The relationships between these neurocognitive outcomes and log-transformed blood
24    lead levels were assessed.  A doubling of blood lead levels at 2 years of age was associated with
25    a covariate-adjusted decline of 1.6 points (95% CI:  0.2, 3.0) in Bayley MDI.  Statistically
26    nonsignificant decrements in MDI were associated with blood lead levels measured at all other
27    time points. Iron deficiency anemia also was an independent predictor of lower MDI
28    (Wasserman et al., 1992).  When examined at 4 years of age, the geometric mean blood lead
29    concentration of children from the smelter area was 39.9 |ig/dL, while the geometric mean for
30    children in the "unexposed" area was 9.6 |ig/dL (Wasserman et al., 1994).  Children were
31    administered the McCarthy Scales of Children's Abilities. Higher prenatal and cord blood lead

      December 2005                           6-62        DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentrations were associated with lower GCI scores. Following covariate-adjustment, children
 2    of mothers with prenatal blood lead levels greater than 20 |ig/dL scored a full standard deviation
 3    below children in the lowest exposure group (<5 |ig/dL prenatal blood lead).  A statistically
 4    significant association also was observed between nearly every blood lead measurement
 5    (at 6-month intervals since birth) and GCI.  At 4 years of age, a doubling of blood lead levels
 6    was associated with a reduction of 2.8 points (95% CI: 1.4, 4.3) on the GCI.  The Perceptual-
 7    Performance subscale of the McCarthy was found to be most sensitive to lead exposure.
 8          When 301 children were examined at 7 years of age with the WISC-III, significant
 9    associations were observed between postnatal blood lead concentrations and IQ, with
10    consistently stronger associations between performance IQ and later blood lead measures
11    (Factor-Litvak, 1999).  The adjusted intellectual loss associated with a doubling in lifetime
12    average blood lead was 2.7 points (95% CI: 1.7, 3.7) in full  scale IQ, 2.8 points (95% CI:  1.7,
13    4.0) in performance IQ, and 2.1 points (95% CI:  1.1, 3.2) in verbal IQ. By 7 years, measures of
14    iron status were no longer significantly associated with IQ.
15          At age 10-12 years, 290 subjects with complete data on exposure and covariate factors
16    were assessed again with the WISC-III (Wasserman et al., 2003). However, in addition to well-
17    characterized exposure histories based on serial blood lead assessments, tibial bone lead was
18    measured using 109Cd based K-shell XRF (Todd et al., 2001) on a representative subsample of
19    167 subjects from both communities. Blood lead and bone lead measures were highly correlated
20    in Titova Mitrovica, but not in Pristina.  Following covariate-adjustment, average lifetime
21    blood lead level was significantly and negatively related to all components of WISC-III IQ.
22    A doubling of average blood lead concentration was  associated with a decrease in full scale,
23    performance, and verbal IQ of 1.6 points (95% CI: 0.4, 2.8), 1.5 points (95% CI: 0.3, 2.8), and
24    1.5 points (95% CI: 0.3, 2.6), respectively.  The relationships between bone lead and IQ scores
25    were stronger than those for blood lead, at least in the more highly exposed smelter community.
26    For each doubling of tibial bone lead concentrations, full scale, performance, and verbal IQ
27    decreased by an estimated 5.5, 6.2, and 4.1 points, respectively. The authors also reported that
28    significant associations  between tibial lead concentrations and IQ scores persisted despite
29    inclusion of blood lead into the model. The inference drawn from these findings was that
30    associations between bone lead and IQ outcomes may be stronger than those between blood lead
31    measures and IQ.

      December 2005                            6-63        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Shanghai, China Study
 2          A prospective study of low-level prenatal and postnatal exposure was initiated in 1993 by
 3    Shen et al. (1998) in Shanghia, China. Pregnant women were recruited from a maternal and
 4    child health care facility in the community.  Lead levels were determined on 348 cord blood
 5    samples.  The geometric mean cord blood lead level was 9.2 |ig/dL (range 1.6-17.5); 40.8% of
 6    the infants had cord blood lead levels > 10 |ig/dL. Infants were further selected for study on the
 7    basis of their cord blood lead concentrations - the low lead group (n = 64) had levels <30th
 8    percentile while the high lead group (n = 69) had levels >70th percentile. Mean cord blood lead
 9    concentrations in the high lead group and low lead group were 13.4 |ig/dL (SD 2.0) and
10    5.3 |ig/dL (SD 1.4), respectively.  At 3, 6, and 12 months, infants were administered the Chinese
11    version of the Bayley Scales of Infant Development.  Capillary blood samples were collected at
12    each visit to ascertain levels of postnatal exposure. Mean blood lead at 1 year of age was
13    14.9 |ig/dL (SD 8.7) in the high lead group and 14.4 jig/dL (SD 7.7) in the low lead group.
14    Postnatal blood lead levels were not significantly different in the high and low lead groups.
15          At all three ages, the Bayley MDI, but not PDI, was associated with cord blood lead
16    groupings following adjustment for covariates, which included a wide range of perinatal,
17    demographic, social, and environmental factors. Postnatal blood lead concentrations were not
18    associated to any Bayley measures. Differences in mean MDI between cord blood lead groups
19    were 3.4 points at 3 months (p = 0.02), 6.3 points at 6 months (p = 0.03), and 5.2 points at
20    12 months (p = 0.03). The early results of this prospective study are generally in accord with
21    similar investigations in Boston, Cincinnati, and Cleveland.  The authors concluded that the
22    adverse effects of prenatal lead exposure on early neurobehavioral development are readily
23    discernible and stable over the first year of life.
24
25    Rochester Study
26          The Rochester prospective study, initiated in 1994, examined the relationship between
27    blood lead levels and IQ at 3 and 5 years of age in 172, predominantly African-American, lower
28    SES children (Canfield et al., 2003a). Participants were enrolled when children were 5 to
29    7 months of age in what was originally a study of lead dust control methods (Lanphear et al.,
30    1999). Blood lead concentrations were assessed at 6-month intervals until 2 years and annually
31    thereafter. No data were available on prenatal exposure. The measure of IQ was the abbreviated

      December 2005                             6-64        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Stanford-Binet Intelligence Scale-4th Edition (SBIS-4). Potential confounders assessed included
 2    gender, birth weight, iron status, HOME scores, maternal IQ, SES, and tobacco use during
 3    pregnancy.
 4          Blood lead concentrations in the Rochester cohort were quite low for an urban population
 5    as this study was conducted after public health measures to reduce blood lead levels in children
 6    were already having a dramatic impact in the U.S. population. Blood lead levels peaked at
 7    2 years of age (mean 9.7 jig/dL). The mean lifetime average blood lead concentration was
 8    7.7 |ig/dL at the age of 3 years and 7.4 |ig/dL at the age of 5 years. At 5 years of age, 56% of the
 9    children had a peak blood lead concentration below 10 |ig/dL. Following adjustment for
10    covariates, there were significant inverse associations with full scale IQ at both 3 and 5 years of
11    age for all blood lead variables, including lifetime average up to age of behavioral assessment.
12          The effect of lead on IQ was estimated in all children using lifetime average, peak,
13    concurrent, and average in infancy (6-24 months) blood lead levels. Lead effects on IQ for the
14    subgroup of children whose peak lead concentration never exceeded 10 |ig/dL also was
15    estimated. Table 6-3.1 shows the covariate-adjusted changes in IQ for each 1 |ig/dL increase in
16    blood lead concentration for all children and children with peak blood lead concentrations below
17    10 |ig/dL. In all cases, the effect estimates were larger in the subsample of children with peak
18    blood lead concentrations below 10 |ig/dL.  For example, the overall estimate including all
19    children indicated that an increase in the lifetime average blood lead concentration of 1 |ig/dL
20    was associated with a decrease of 0.46 points (95% CI:  0.15, 0.76) in IQ. In comparison, a
21    1 |ig/dL increase in lifetime average lead concentration was associated with a decline of
22    1.37 points (95% CI:  0.17, 2.56) in children with peak blood lead concentrations below
23    10 |ig/dL. In an accompanying editorial of the Canfield et al. (2003a) study, Rogan and Ware
24    (2003) noted that the steepness in the concentration-response relationship below 10 |ig/dL might
25    have been influenced by 10 children with blood lead concentrations at or below 5 |ig/dL and IQs
26    above 115.  However, they added that it was unlikely that the associations reported by Canfield
27    et al. were solely due to these values.  Regression diagnostics performed by Canfield et al.
28    identified only one potential outlier (a child who had a low IQ and low lead concentration);
29    however, this value was retained in all analyses as it did not pass the discordancy test.
30          In the Rochester study, the relationship between children's IQ score and their blood lead
31    level was found to be nonlinear.  A semiparametric analysis indicated a decline of IQ of

      December 2005                            6-65         DRAFT-DO NOT QUOTE OR CITE

-------
o
o
3
(^ '
CD
to
o
o









ON
i
ON
ON



O
!>
H
6
o
o
H
O
o
H
w

o
HH
H
W
Table 6-3.
Type of Blood
Lead Measurement
All Children
Lifetime average

Peak

Concurrent

Average in infancy
(6-24 mo)
1. Covariate- Adjusted Changes in IQ for Each 1 ug/dL Increase in Blood
At 3 Years of Age
n

172

172

171

172

P (95% CI)

-0.35
(-0.69,0.00)
-0.19
(-0.39,0.01)
-0.31
(-0.60, -0.01)
-0.32
(-0.71,0.07)
P

0.05

0.06

0.04

0.10

Lead Concentration3
At 5 Years of Age
P (95% CI)

-0.
(-0.93,
-0.
(-0.47,
-0.
(-0.99,
-0.
(-0.93,

57
-0.20)
26
-0.05)
61
-0.24)
53
-0.13)
P

0.003

0.02

0.001

0.01

Overall
B (95% CI)

-0
(-0.76,

.46
-0.15)
-0.23
(-0.40,
-0.05)
-0.46
(-0.74,
-0
(-0.77,
-0.18)
.43
-0.09)

0.

0

0.

0

P

004

.01

002

.02

Children with Peak Blood Lead Concentrations below 10 ug/dLb

Lifetime average

Peak

Concurrent

Average in infancy
(6-24 mo)

a Estimates were adjusted
weight, and iron status.


101

101

101

105


for maternal IQ,


A total of 7 1 children were found to have
below 10 ug/dL at 3 years of age but at 5
10 ug/dL at 5 years but were not tested at
children the peak blood


-1.22
(-2.53,0.09)
-1.36
(-2.46, -0.27)
-1.36
(-2.37, -0.35)
-0.58
(-1.75,0.59)


0.07

0.002

0.009

0.32


race, level of education, use of tobacco


a peak blood lead
years of age had a
3 years. The total


concentration below

-1.
(-2.94,
-1.
(-2.55,
-1.
(-3.00,
-0.
(-2.09


52
-0.09)
44
- 0.33)
79
-0.60)
92
, 0.25)

during pregnancy,


10 ug/dL


at both a
higher concentration or were not tested
number of children

0.04

0.01

0.004

0.12


household income


ges; an additional


-1.37
(-2.56,
-1
(-2.37,
-1
(-2.50,
-0
(-1.78

-0.17)
.40
-0.44)
.58
-0.65)
.75
, 0.28)

, HOME score, child's


15 children


had a peak

0

0.

0.

0


gender,



.03

005

001

.15


birth


concentration
, and another 15 children had a peak concentration below
in the analysis of the average concentration in infancy is 105
, because in 4
lead concentration occurred after the age of 24 months.










Source: Canfield et al. (2003a).












-------
 1    7.4 points for a lifetime average blood lead concentration of up to 10 |ig/dL, while for levels
 2    between 10 to 30 |ig/dL a more gradual decrease of approximately 2.5 points IQ was estimated.
 3    The authors concluded that the most important aspect of their findings was that effects below
 4    10 |ig/dL that have been observed in previous cross-sectional studies (e.g., Chiodo et al., 2004;
 5    Fulton et al., 1987; Lanphear et al., 2000; see Section 6.3.2.1.2) have been confirmed in this
 6    rigorous prospective longitudinal investigation.
 7
 8    Pooled-Analyses of Prospective Longitudinal Cohort Studies
 9          Investigators have collectively analyzed the results of multiple independent studies using
10    the methods of meta- and pooled data analyses.  A powerful approach involves pooling the raw
11    data from several high quality studies to examine concentration-response relationships in a large
12    sample of children with diverse sociodemographic backgrounds and levels of exposure.  The
13    studies reviewed here are summarized in Annex Table AX6-3.2.
14          Lanphear et al. (2005) reported on a pooled analysis of seven prospective studies that
15    were initiated prior to 1995. The analysis involved 1,333 children with complete data on
16    confounding factors that were essential in the multivariable analyses.  The participating sites
17    included Boston, MA;  Cincinnati, OH; Cleveland,  OH; Rochester, NY; Mexico City; Port Pirie,
18    Australia; and Kosovo, Yugoslavia. A prospective cohort study conducted in Sydney, Australia
19    was not included because the authors were unable to contact the investigators (Cooney et al.,
20    1989b, 1991). The sample size of 175 for children at age 7 years in the Sydney cohort and the
21    wide confidence intervals of the effect estimates, as implied by the lack of significant
22    associations, indicate that the nonavailability of this study is unlikely to influence the results of
23    the pooled analysis by  Lanphear et al.
24          The primary outcome measure was full scale IQ measured at school age (mean age at IQ
25    testing was 6.9 years).  All  children were assessed with an age-appropriate version of the
26    Wechsler scales. Four measures of lead exposure were examined: concurrent blood lead (blood
27    lead level closest in time to the IQ test), maximum blood lead level (peak  blood lead measured at
28    any time prior to the IQ test), average lifetime blood lead (mean blood lead from 6 months to the
29    concurrent blood lead test), and early childhood blood lead (defined as the mean blood lead from
30    6 to 24 months).  A pooled analysis of the relationship between cord blood lead levels and IQ
31    also was conducted in the subsample for which cord blood lead tests were available.

      December  2005                            6-67        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Multivariate regression models were developed adjusting for site as well as ten common
 2    covariates assessing factors likely to be confounders of the relationship between lead and
 3    cognitive development, including HOME scores, birth weight, maternal education and IQ, and
 4    prenatal substance abuse.  A thorough statistical analytic strategy was employed to determine the
 5    linearity or nonlinearity of the relationship between blood lead levels and full-scale IQ.
 6    Regression diagnostics also were performed to ascertain whether lead coefficients were affected
 7    by collinearity or influential observations.  The fit of all four measures of postnatal blood lead
 8    levels was compared using the magnitude of the model R2. The blood lead measure with the
 9    largest R2 (adjusted for the same covariates) was nominated a priori as the preferred blood lead
10    index relating lead exposure to IQ in subsequent inspections of the relationships.  Results were
11    evaluated by applying a random-effects model (with sites random) rather than a fixed-effects
12    model. The authors also examined the impact of any one site on the overall model by calculating
13    the blood lead coefficient in seven identical models, each omitting one of the seven prospective
14    cohort studies. Similar models were fitted for verbal and performance IQ as well.
15          The median lifetime average blood lead concentration was 12.4 |ig/dL (5th-95th
16    percentile 4.1-34.8) with about 18% of the children having peak blood lead levels below
17    10 |ig/dL. The 5th to 95th percentile concurrent blood lead levels ranged from 0.8 to 4.7 |ig/dL
18    in the individual studies. The mean IQ of all children was 93.2 (SD 19.2) but this varied greatly
19    between studies.  All four measures of postnatal exposure were highly correlated.  However, the
                                                                                      rj
20    concurrent blood lead level exhibited the strongest relationship with IQ, as assessed by R .
21    Nevertheless, the results of the regression analyses for all blood lead measures were very similar.
22    Multivariable analysis resulted in a  six-term model including log of concurrent blood lead, study
23    site, maternal IQ, HOME Inventory, birth weight, and maternal education. As illustrated in
24    Figures 6-3.2 and 6-3.3, the shape of the log-linear model and the spline function indicated that
25    the steepest declines in IQ were at blood lead  concentrations below 10 |ig/dL.  The log-linear
26    model estimated a decrement of 1.9 points (95% CI: 1.2, 2.6) in full scale IQ for a doubling of
27    concurrent blood lead. Due to the log-linear relationship, the slope of the lead effect on IQ was
28    greater in the lower ranges of exposure.  The IQ point decrements associated with an increase in
29    blood lead from <1  to  10 |ig/dL compared to 10 to 20 |ig/dL were 6.2 (95% CI: 3.8, 8.6) versus
30    1.9(95%CI:  1.2,2.6).
      December 2005                            6-68        DRAFT-DO NOT QUOTE OR CITE

-------
                105
                100
             !£  95
                 90
                 85
                               10            20            30
                                                     (pg/dL)
                              40
Figure 6-3.2.  Log-linear model (95% CI shaded) for concurrent blood lead concentration
              adjusted for HOME score, maternal education, maternal IQ, and birth
              weight. The mean IQ (95% CI) for the intervals <5, 5-10,10-15,15-20,
              and >20 ug/dL are shown.

Source: Lanphear et al. (2005).
                105
                ioa
                 954
                 90'
                 85
                                                      Log-linear Model
                                                      Peak Blood Lead > 10|jg/dL
                                               	Peak Blood Lead < 10|jg/dL
                               10            20            30
                                                     (pg/dL)
                              40
Figure 6-3.3.  Log-linear model for concurrent blood lead concentration along with linear
              models for concurrent blood lead levels among children with peak blood
              lead levels above and below 10 ug/dL.

Source: Lanphear et al. (2005).
December 2005
6-69
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Rothenberg and Rothenberg (2005) reanalyzed the Lanphear et al. (2005) pooled study to
 2    examine the form of the concentration-response function for the lead exposure effect on child IQ.
 3    This further analysis also focused on concurrent blood lead levels.  Rothenberg and Rothenberg
 4    reported that a log-linear relationship between blood lead and IQ was a significantly better fit
 5    within the ranges of the blood lead levels than was a linear-linear relationship (p = 0.009), with
 6    little evidence of residual confounding from included model variables. However, a segmented
 7    linear model also offers an appropriate alternative since limited data is available at lower levels.
 8          The log-linear model in Lanphear et al. estimated a decline of 6.2 points in full scale IQ
 9    for an increase in concurrent blood lead levels from <1 to 10 |ig/dL. This effect estimate was
10    comparable to the 7.4 point decrement in IQ for an increase in lifetime mean blood lead levels up
11    to 10 |ig/dL observed in the Rochester study (Canfield et al., 2003a), as well as other studies
12    reviewed above.
13
14    6.3.2.1.2  Cross-sectional Studies of Neurocognitive Ability
15          Among the cross-sectional studies reviewed in the 1986 Lead AQCD and the 1990
16    Supplement, the most thorough and methodologically rigorous were those of Needleman et al.
17    (1979) and Fulton et al. (1987). Needleman et al. (1979) measured lead in the dentin of
18    deciduous teeth in elementary school children from two Boston area communities. After
19    statistical adjustment for a number of potential confounding factors, children in the higher tooth
20    lead group performed significantly less well on full scale and verbal IQ. Differences in full scale
21    IQ between the high and low tooth lead groups was on the order of 4.5 points.
22          The general population study by Fulton et al. (1987) studied 501  children aged 6-9 years
23    in Edinburgh, Scotland who were at risk for lead exposure owing to a plumbosolvent water
24    supply and a large number of houses with lead plumbing. Blood lead levels averaged 11.5 |ig/dL
25    (range 3-34).  Following covariate adjustment, there were statistically significant relationships
26    between concurrent blood lead levels and total scores on the British Ability Scale and the
27    Quantitative and Reading subscales. Data showed a clear concentration-response relationship
28    with no evidence of a threshold.
29          Recent cross-sectional studies of neurocognitive ability are summarized in Annex
30    Table AX6-3.3.  Key studies are further discussed in this section.  Lanphear et al. (2000)
31    examined the relationship between blood lead concentrations and cognitive deficits in a

      December 2005                             6-70         DRAFT-DO NOT QUOTE OR CITE

-------
 1    nationally representative sample of 4,853 children aged 6 to 16 years children who participated
 2    in the third National Health and Nutrition Examination Survey (NHANES III).  The purpose of
 3    the study was to examine the relationship between low blood lead concentrations (especially
 4    those below 10 |ig/dL) and two subtests of the WISC-R, Block Design (a measure of visual-
 5    spatial skills)  and Digit Span (a measure of short-term and working memory).  Academic
 6    achievement tests also were administered but are discussed in a later section.  A number of
 7    potential confounders were assessed and included in multivariable analyses including gender,
 8    racial/ethnic background, child's serum ferritin level, serum cotinine level, region of country,
 9    marital status  and education level of primary caregiver, and a poverty index ratio (the ratio of
10    total  family income, as reported by the adult informant, to the federal poverty level for the year
11    of the interview). Other potential confounders such as in utero and postnatal exposure to tobacco
12    smoke, birth weight, and admission to the neonatal intensive care unit were only available for
13    children between 6 and 11 years of age. Therefore, the authors conducted a secondary analysis
14    of the data on these children to verify that inclusion of these potentially important variables did
15    not alter the findings of the main analysis using the larger sample.
16          The geometric mean blood lead concentration for children in the study sample was
17    1.9 |ig/dL (SE 0.1).  Only 2.1% of the NHANES III sample in this analysis had blood lead
18    concentrations greater or equal to 10 |ig/dL. In multivariate analyses, a significant covariate-
19    adjusted relationship was found between blood lead level and scores on both WISC-R subtest for
20    all children as well  as among those children with blood lead levels <10 |ig/dL. Blood lead
21    concentration also was significantly associated with Block Design when the multivariate analysis
22    was restricted to children with blood lead levels <7.5  |ig/dL.  For a 1 |ig/dL increase in blood
23    lead level, Block Design scores declined by 0.10 points (SE 0.04) for all children, 0.13 points
24    (SE 0.06) for children with blood lead levels <10 |ig/dL, and 0.11 points (SE 0.06) for children
25    with  blood lead levels <7.5 |ig/dL. The authors concluded that deficits in intellectual functioning
26    were associated with blood lead levels <10 |ig/dL.  While a large number of potential
27    confounding factors were controlled in these analyses, interpretation of results must be tempered
28    by the fact that no data on maternal IQ or direct observations of caretaking quality in the home
29    were available. Furthermore, it is not clear whether the cognitive deficits observed were due to
30    lead exposure that occurred during early childhood or a function of concurrent exposure.
      December 2005                            6-71        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Chiodo et al. (2004) studied the relationship between blood lead concentrations and IQ,
 2    assessed using WISC-III, in a sample of 237 African-American inner-city children from Detroit,
 3    MI at 7.5 years of age. This cohort was derived from a larger study of the effects of prenatal
 4    alcohol exposure on child development. However, approximately 83% of children for whom
 5    blood lead levels were obtained had either low or no gestational exposure to alcohol. Blood lead
 6    levels were low with a mean of 5.4 |ig/dL (SD 3.3, range 1-25).  Following covariate adjustment,
 7    there was a statistically significant association between blood lead concentrations and full scale,
 8    verbal and performance IQ, with the strongest relationship observed for performance IQ.
 9    Significant effects of lead on full scale and performance IQ were still evident at blood lead
10    concentrations below 7.5 |ig/dL. Nonparametric smoothing analyses confirmed that these effects
11    were linear in nature.
12          Walkowiak et al. (1998) conducted a cross-sectional study examining the relationships of
13    low-level lead and mercury exposure, and various measures of neurocognitive and neuromotor
14    functioning in 384 children aged 6 years in three German cities. Lead was measured in blood at
15    the time of testing and mercury burden was estimated from urine samples. As their measure of
16    IQ, they administered two subtests of the German WISC, Vocabulary and Block Design. These
17    subtests were treated separately as well as a summed index, which served as a surrogate for full
18    scale IQ.  Blood lead concentrations were low (geometric mean 4.3 |ig/dL [95th percentile 8.9]).
19    Following covariate-adjustment, Vocabulary and the combined index, but not Block Design,
20    exhibited negative associations with blood lead of statistical or borderline statistical  significance;
21    no associations were observed for mercury. The authors concluded that these findings roughly
22    correspond with those of other studies that find effects of lead exposure on measures of
23    intelligence at blood lead concentrations below 10 |ig/dL.  However, they also caution that some
24    important covariates and potential confounding variables were not measured, including parental
25    IQ and home environment (e.g., HOME score).
26          Rabinowitz et al. (1991) studied the  relationship between lead measured in shed
27    deciduous teeth (central incisors) and psychometric intelligence in 443 children in grades 1 to 3
28    in Taiwan. Two of the primary schools included in the study were in proximity to primary lead
29    smelters.  The Ravens Colored Progressive Matrices (RCPM), a test of nonverbal reasoning that
30    is widely used in studies of non-western populations because of its more culturally neutral
31    properties, was administered.  Studies on a  subsample of 60 children residing near the lead

      December 2005                           6-72        DRAFT-DO NOT QUOTE OR CITE

-------
 1    smelters revealed mean blood lead level of 13.0 |ig/dL (SD 4.4).  Scores on the RCPM were
 2    negatively correlated with tooth lead concentrations. In multivariate analyses, parental education
 3    was a particularly important predictor of RCPM scores, but tooth lead concentrations still
 4    significantly predicted lower scores  on the RCPM in families occupying the lowest social strata
 5    and among female subjects.
 6          Kordas et al. (2004) examined the relationship between lead exposure and various indices
 7    of psychometric intelligence in a cohort of 602 first grade children attending public schools in
 8    Torreon, a highly industrialized city in northern Mexico. This study investigated whether lead-
 9    associated deficits in intellectual attainment might be explained by correlated nutritional factors
10    such as iron status, anemia, and growth. The mean blood lead concentration was 11.5 |ig/dL
11    (SD 6.1). Approximately half of the children had blood lead concentrations below 10 |ig/dL and
12    only 20% of the subjects had blood lead levels in  excess of 15 |ig/dL. Subjects were
13    administered Spanish or Mexican versions of the Peabody Picture Vocabulary Test-Revised
14    (PPVT-R), the Cognitive Abilities Test (CAT), and subtests of the WISC-R (Coding, Digit Span,
15    and Arithmetic subtests). Letter and Number Sequencing tests (adapted from the Trail Making
16    Test, Trails A) also were administered. Following adjustment for sociodemographic variables,
17    anemia, iron status, and growth, higher blood lead levels were significantly associated with
18    poorer performance on the PPVT, WISC-R Coding, and Number and Letter Sequencing.  The
19    authors concluded that lead's association with iron deficiency anemia or growth retardation
20    could not explain the relationship between lead and cognitive performance.  The authors
21    acknowledged that a major limitation of their study  is the lack of earlier measures of lead
22    exposure and nutritional status, and information on potentially confounding variables such as
23    parental intelligence and  quality  of caretaking in the home.
24          Bellinger et al. (2005) reported on a study of the relationship between blood lead levels
25    and IQ in 55 children aged 4 to 14 years in Chennai, India. This is the first published study that
26    has investigated neurodevelopmental morbidities  associated with undue lead exposure in Indian
27    children.  Children were recruited from a rural primary school on the outskirts of the city. The
28    mean blood lead concentration was 11.1 |ig/dL (SD 5.6, range 2.5-38.3). The Binet-Kamath
29    Intelligence test along with other measures of neurobehavior were administered.  The covariate-
30    adjusted blood lead coefficient was negative but nonsignificant, perhaps due to the small sample
31    size and highly variable performance of subjects with the lowest blood lead  concentrations.

      December 2005                            6-73        DRAFT-DO NOT QUOTE OR CITE

-------
 1    For example, the mean IQ of children in the highest blood lead quartile was 95.6 with a SD of
 2    13.3 compared to 102.0 with a larger SD of 22.5 for children in the lowest blood lead quartile.
 3          The cross-sectional studies examining the effect of lead on neurocognitive abilities varied
 4    widely in study location, population, age of testing, and outcomes measured.  Collectively, they
 5    generally concluded that blood or tooth lead levels were significantly associated with declines in
 6    intelligence and other neurocognitive outcomes. In addition, these associations were consistently
 7    observed in studies with mean blood lead levels <10 |ig/dL.
 8
 9    6.3.2.1.3  Meta-Analyses of Studies of Neurocognitive Abilities
10          The meta-analyses of studies investigating the association between lead and
11    neurocognitive abilities included results from both prospective cohort studies and cross-sectional
12    studies.  The studies reviewed here are summarized in Annex Table AX6-3.2.  Needleman and
13    Gatsonis (1990) conducted a meta-analysis of 12 studies that used multiple regression techniques
14    to assess the relationship between lead levels in tissues (blood or teeth) while adjusting for
15    potentially confounding variables.  Studies were weighted based on sample sizes, which ranged
16    from 75  to 724 children. The authors divided studies into two groups according to the type of
17    tissue analyzed for lead (blood or teeth). Joint p-values and average effect sizes as measured by
18    partial correlation coefficients were calculated using two different methods by Fisher and by
19    Mosteller and Bush (Rosenthal, 1984). The joint p-values for the blood lead studies were
20    <0.0001 for both methods while joint p-values of <0.0006 and <0.004 were obtained for tooth
21    lead studies. The partial correlations ranged from  -0.27 to -0.0003. Sensitivity analyses
22    revealed that no single study was responsible for the significance of the final findings. The
23    authors concluded that the hypothesis that lead lowers children's IQ at relatively low dose was
24    strongly supported by their quantitative analysis.
25          Another meta-analysis conducted by Schwartz (1994) took a different approach. Only
26    studies relating blood lead to IQ were chosen for quantitative review since the concentration of
27    lead in the bloodstream is the only index of exposure that has been used as the basis for public
28    health policy. Three longitudinal and four cross-sectional studies relating blood lead to IQ were
29    examined. Furthermore, while the work of Needleman and Gatsonis (1990) essentially involved
30    combining partial correlations, the measure of effect used in the Schwartz analysis was the
31    predicted change in full scale IQ as blood lead increased from 10 to 20 |ig/dL. For the

      December 2005                            6-74        DRAFT-DO NOT QUOTE OR CITE

-------
 1    prospective longitudinal studies, blood lead levels at 2 years of age or average blood lead levels
 2    up to 3 years of age were selected for the analysis. This approach by Schwartz may be related to
 3    the belief at the time of the analysis that blood lead levels during the first 3 years of life were the
 4    most critical in determining the severity of neurodevelopmental toxicity. The exclusion of blood
 5    lead levels from other time points may be of issue as it appears that later blood lead levels may
 6    be more predictive of mental deficits (Baghurst et al., 1992; Canfield et al., 2003a; Chen et al.,
 7    2005; Dietrich et al., 1993a; Factor-Litvak et al., 1993).  Studies were weighted by the inverse of
 8    the variances using a random-effects modeling procedure.  The estimated decrease in IQ for an
 9    increase in blood lead from 10 to 20 |ig/dL was 2.6 points (95% CI: 1.8,3.4). Sensitivity
10    analyses indicated that the results were not determined by any individual study. Effect estimates
11    were similar for longitudinal and cross-sectional studies. In another analysis, studies with mean
12    blood lead concentrations below 15 |ig/dL and above 15 |ig/dL had estimated effect sizes of
13    -3.23 points (95% CI:  -5.70, -0.76) and -2.32 points (95% CI: -3.10, -1.54), respectively.
14    When the study with the lowest mean blood lead level was examined in greater detail using
15    nonparametric smoothing, no evidence of a threshold was observed down to a blood lead level
16    ofl|ig/dL.
17          Pocock et al. (1994) conducted a review of the epidemiologic evidence for lead effects on
18    IQ that included a meta-analysis.  For the meta-analysis, the fixed-effect method described by
19    Thompson and Pocock (1992) was used. Five prospective and 14 cross-sectional studies  (with
20    both tooth and blood lead measures) were included. For consistency, only blood lead levels at or
21    around 2 years of age were considered for the prospective studies. Their overall conclusion was
22    that a doubling of blood lead levels from 10 to 20 |ig/dL, or tooth lead from 5 to 10 jig/g was
23    associated with an average estimated deficit in IQ of around 1 to 2 points.
24          Other earlier meta-analyses of lead-IQ studies have been published but are not reviewed
25    here, because  later work greatly extended these efforts and included more studies, rendering
26    these analyses outdated (Needleman and Bellinger, 1988; Schwartz, 1985; Thacker et al., 1992).
27    The meta-analyses of studies investigating the effect of lead on neurocognitive ability
28    consistently observed significant associations between blood or tooth lead levels and decrements
29    in IQ. The analysis by Schwartz (1994) observed no evidence of a threshold  at blood lead levels
30    below 10 |ig/dL.
31

      December 2005                            6-75        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.3.2.2   Measures of Academic Achievement
 2          There are relatively little data on the relationship between lead exposure and objective
 3    measures of academic achievement. A few earlier studies reported an inverse relationship
 4    between lead exposure and reading skills (Fergusson et al., 1988a; Fulton et al., 1987; Yule et al.,
 5    1981).  Since the 1990 Supplement, more studies have focused on the practical consequences of
 6    childhood lead exposure by including measures of academic performance in their batteries.
 7    Studies reviewed in this section are summarized in Annex Table AX6-3.4.
 8          Lanphear et al. (2000) examined the relationship between blood lead levels and a
 9    standardized measure of academic achievement in 4,853 children aged 6 to 16 years.  The source
10    of data for this study was the third National Health and Nutrition Examination Survey (NHANES
11    III). This cohort was previously described in Section 6.3.2.1.2.  Subjects were administered the
12    Arithmetic and Reading subtests of the Wide Range Achievement Test-Revised (WRAT-R).
13    The WRAT-R Arithmetic subtest  includes oral and written problems ranging in level from
14    simple addition to calculus, while  the Reading subtest assesses letter recognition and word
15    reading skills.  The geometric mean blood lead concentration was  1.9 |ig/dL.  Only 2.1%  of the
16    subjects had blood lead levels equal to or greater than 10  |ig/dL. Multiple linear regression
17    revealed a 0.70 point (95% CI:  0.37, 1.03) decrement in arithmetic scores and a 0.99 point
18    (95% CI: 0.62, 1.36) decrement in Reading scores for each 1 |ig/dL increase in blood lead
19    concentration (p < 0.001). In the next phase of the analysis, the adjusted relationship between
20    performance on WRAT subtests and blood lead concentration for children with blood lead
21    concentrations <10 |ig/dL, <7.5 |ig/dL, <5 |ig/dL, or <2.5 |ig/dL were carried out.  Statistically
22    significant inverse relationships between blood lead levels and performance for both Reading
23    and Arithmetic subtests were found for children with blood lead concentrations below 5 |ig/dL.
24    Secondary analysis limited to younger children with data on all covariates did not alter findings
25    from the main analysis. The authors concluded that results of these analyses suggest that  deficits
26    in academic skills are associated with blood lead concentrations lower than 5 |ig/dL.  However,
27    although the relationship of blood lead concentration and achievement was adjusted for
28    numerous potential confounders, the study lacked information on at least two covariates that
29    have been shown to be important in other lead studies (HOME scores and parental IQ).  Failure
30    to adjust for these variables may have underestimated or overestimated the deficits in academic
31    skills related to lead. Furthermore, as with all cross-sectional studies utilizing blood lead  as the

      December 2005                            6-76       DRAFT-DO NOT QUOTE OR CITE

-------
 1    index of dose, it is not clear whether the deficits in academic skills were due to lead exposure
 2    that occurred sometime during early childhood or due to concurrent exposure. Nevertheless,
 3    concurrent blood lead levels likely reflect both ongoing exposure and preexisting body burden.
 4          Needleman et al. (1990) reexamined the Chelsea and Somerville, MA cohort of first and
 5    second graders recruited in the 1970s (Needleman et al., 1979).  One hundred and thirty-two of
 6    the original 270 children were recalled.  Neurobehavioral deficits in relationship to the
 7    concentration of lead in shed deciduous teeth had persisted into late adolescence. Subjects with
 8    dentin lead levels >20 ppm were at higher risk of dropping out of high school (adjusted odds
 9    ratio of 7.4, [95% CI: 1.4, 40.7]) and of having a reading disability (adjusted odds ratio of 5.8
10    [95% CI: 1.7, 19.7]). Higher dentin lead levels also were significantly associated with lower
11    class standing, increased absenteeism, and lower vocabulary and grammatical reasoning scores
12    on the Neurobehavioral Evaluation System (NES). The authors concluded that undue exposure
13    to lead had enduring and important effects on objective parameters of success in real life.
14          Bellinger et al. (1992) administered a battery of neuropsychological tests to 148 children
15    in the Boston Lead Study cohort at age 10 years.  The authors  administered the short-form of the
16    Kaufman Test of Educational Achievement (KTEA) in addition to IQ studies. The KTEA
17    assesses reading, math, and spelling skills.  The primary outcome was the Battery Composite
18    Score. As previously indicated, exposures in this cohort were  low with a peak mean blood lead
19    at 18 months of only 7.8 |ig/dL (SD 5.7). The cohort had a high SES that consisted of white
20    intact families with college-educated parents. Average KTEA scores in this cohort were
21    approximately one standard deviation above the population mean.  Nevertheless,  postnatal blood
22    lead levels measured at virtually all ages were significantly associated with lower KTEA Battery
23    Composite Scores. However, after covariate-adjustment, including full scale IQ in the model,
24    only blood lead levels at 24 months of age were significantly predictive of lower academic
25    achievement.  Over the range of approximately 0 to 25 |ig/dL, Battery Composite scores
26    declined by approximately 8.9 points (95% CI:  4.2, 13.6) for each 10 |ig/dL increase in
27    24-month blood lead. The specific subscales of the KTEA that were most significantly
28    associated with lead were Spelling and Math. Within the Math subscale, lead appeared to be
29    more strongly associated with performance on the advanced quantitative Concepts/Applications
30    items than on computation. The associations between these early measures of low level
31    exposure to lead and achievement were significant even after adjustment for IQ, suggesting that

      December 2005                           6-77        DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead-sensitive neuropsychological processing and learning factors not reflected in indices of
 2    global intelligence may contribute to reduced performance on academic tasks.
 3          Leviton et al. (1993) reported on the relationship between pre- and postnatal lead
 4    exposure and academic problems in approximately 2,000 children born in one Boston hospital
 5    between 1979 and 1980 using the Boston Teacher Questionnaire (BTQ). A teacher provided an
 6    assessment of each child's academic functioning when the child reached the age of 8 years.
 7    Mean umbilical cord blood lead was 6.8 |ig/dL and mean tooth (dentin) lead concentration was
 8    2.8 |ig/g. There was limited information on covariate factors. However, following adjustment
 9    for potential  confounding variables, elevated dentin lead concentrations were associated with
10    statistically significant reading and spelling difficulties as assessed by the BTQ among girls. The
11    authors concluded that their findings supported the case for lead-associated learning problems at
12    levels prevalent in the general population.  However, they added that the inability to assess child-
13    rearing quality in this questionnaire study conducted by mail limits the inferences that came be
14    drawn from the findings.
15          Fergusson et al. (1993) examined the relationship between dentin lead levels in shed
16    deciduous teeth at 6-8 years and measures of academic attainment and classroom performance in
17    a birth cohort of over 1,200 New Zealand children enrolled in the Christchurch Health and
18    Development Study when they reached 12-13 years of age. This study was an extension of
19    earlier work  in these children indicating a relationship between low lead levels and deficits in
20    academic skills around the age of 8 years (Fergusson et al., 1988a).  Average dentine lead levels
21    in the cohort were 6.2 jig/g (SD 6.2). Measures of academic performance included word
22    recognition from the Burt Reading Test, reading comprehension from the Progressive
23    Achievement Test, a general measure of scholastic skills based on children's scores on the Test
24    of Scholastic Abilities, and teacher ratings of classroom performance in the areas of reading,
25    written expression, and mathematics. Following adjustment for a wide range of covariates
26    (including  residence in potentially lead-hazardous housing), dentin lead levels were significantly
27    associated  with virtually every formal index of academic skills and teacher ratings of classroom
28    performance. Statistical treatment of the data included a multivariate analysis of all 12
29    regression  equations simultaneously using LISREL modeling methods. This conservative
30    analysis clearly showed that the probability of observing these results under the null hypotheses
31    that lead was unrelated to all covariate-adjusted test outcomes was extremely small. In an

      December  2005                            6-78         DRAFT-DO NOT QUOTE OR CITE

-------
 1    adjunct analysis, Fergusson and Horwood (1993) examined the effects of low-level lead
 2    exposure on the growth of word recognition in this cohort from 8 to 12 years of age. The
 3    New Zealand data were analyzed using growth curve modeling methods.  After adjustment for
 4    potential confounding variables, children with dentin lead levels equal to or greater than 8 jig/g
 5    displayed significantly slower growth in word recognition abilities with no evidence of catch up.
 6    The authors concluded that these results were consistent with their earlier analyses and suggest
 7    that early exposure to very low levels of lead result in small but detectable and enduring deficits
 8    in children's cognitive abilities.
 9          Academic achievement in relationship to lead was reexamined in the New Zealand cohort
10    when subjects reached 18 years of age (Fergusson et al., 1997). The sample at 18 years consisted
11    of 881 subjects, or approximately 70% of the original cohort. Measures of educational
12    achievement included the Burt Reading Test, number of years of secondary education, mean
13    number of School Certificate passes (based on results of national examinations), and leaving
14    school without formal qualifications (analogous to failure to graduate from high school in the
15    U.S.). As in previous analyses, a wide range of potentially confounding sociohereditary factors
16    were measured and controlled for in multivariable analyses, which included both linear and
17    logistic regressions.  Prior to and following covariate adjustment there were statistically
18    significant concentration-response relationships between dentin lead concentrations and lower
19    reading test scores, having a reading level of less than 12 years, failing to complete 3 years of
20    high school, leaving  school without qualifications, and mean number of School Certificates
21    subjects passed. The authors conclude that their results are consistent with the view that there is
22    a relationship between early exposure to low levels of lead and later educational outcomes. The
23    late results of the New Zealand studies confirm the findings of Needleman et al. (1990) in a
24    cohort with lower levels of exposure to environmental lead.
25          Rabinowitz et al. (1992) examined the relationship between tooth lead concentrations and
26    scores on BTQ clusters in 493 Taiwanese children in grades  one through three. Mean lead levels
27    in incisors were 4.6 |ig/g (SD 3.5). Factors associated with lead and the BTQ included 13
28    variables measuring  perinatal, familial, and economic parameters. Prior to adjustment for
29    covariates, girls in this sample with higher exposures to lead evinced a borderline significant
30    trend for reading difficulties while boys displayed significantly increased difficulties with respect
31    to activity levels and task attentiveness.  In multiple logistic regression models, the tooth lead

      December 2005                            6-79        DRAFT-DO NOT QUOTE OR CITE

-------
 1    terms failed to achieve statistical significance.  The authors concluded that lead levels found in
 2    the teeth of children in their Taiwanese sample were not associated with learning problems or
 3    syndromes as assessed by the BTQ.
 4          Wang et al. (2002) examined the relationship between blood lead levels and class ranking
 5    in 934 third graders living in an urban industrial area of Taiwan. The outcome variables were
 6    grades for Chinese (reading and writing), Mathematics, History and Society, and Natural
 7    Science.  To avoid the impact of teacher's bias in grading criteria, the authors converted the
 8    children's grades into class rankings. A limited number of potentially confounding factors were
 9    measured, including maternal education and father's SES. Mean blood lead level was 5.5 |ig/dL
10    (SD 1.89). In multiple regression analyses adjusting for gender, maternal education, and father's
11    SES, blood lead was significantly associated with lower class ranking in all academic subjects.
12    The major shortcoming of this cross-sectional study is the lack of control for potentially
13    important confounding factors such as parental intelligence.  However, the strength and
14    consistency of the reported relationships suggest that relatively low levels of lead may play a role
15    in lowering academic performance.
16          The results of these studies strongly suggest that lead exposure plays a role in the
17    academic performance of children.  The effects of lead on academic achievement appear to
18    include children with blood lead levels that do not exceed 10 |ig/dL.
19
20    6.3.2.3   Measures of Specific Cognitive Abilities
21          Outcomes of specific cognitive abilities, in  particular, the domains of Attention and
22    Executive Functions, Language, Memory and Learning, and Visuospatial Processing have
23    been examined in some detail in recent studies.  These studies are summarized in Annex
24    TableAX6-3.5.
25          In the aggregate, studies suggest that lead exposure impairs a child's ability to regulate
26    attention  and engage several related higher order cognitive processes that have come to  be
27    termed "executive functions." Executive functions refer to strategic planning, control of
28    impulses, organized search, flexibility of thought and action, and self-monitoring of one's own
29    behavior—activities that help the subject maintain an appropriate  mental set in order to  achieve
30    an immediate or future goal (Spreen et al., 1995).  In some earlier studies, increased lead
31    exposure was found to be associated with a higher frequency of negative ratings by teachers

      December 2005                            6-80       DRAFT-DO NOT QUOTE  OR CITE

-------
 1    and/or parents on behaviors such as inattentiveness, impulsivity, distractibility, and
 2    impersistence in assigned tasks, as well as slow psychomotor responses and more errors on
 3    simple, serial, and choice reaction time tasks (e.g., Hatzakis et al., 1989; Hunter et al., 1985;
 4    Needleman et al., 1979; Raab et al., 1990; Winneke et al., 1990). The concept that lead may
 5    impact executive functions in particular is biologically plausible. The prefrontal cortex is highly
 6    innervated by projections of neurons from the midbrain and has the highest concentration of
 7    dopamine of all cortical areas. Dopamine plays a key role in cognitive abilities mediated by the
 8    prefrontal cortex. It has been known for some time that the dopamine system is particularly
 9    sensitive to lead based upon data from studies of rodents and nonhuman primates (Cory-Slechta,
10    1995).
11          Bellinger et al. (1994)  examined a portion of the original Chelsea and Somerville cohorts
12    at 19-20 years of age. The principal neurobehavioral outcomes in the investigation were scores
13    on a battery of attentional measures assembled by Mirsky (1987).  Higher tooth lead
14    concentrations were significantly associated with poorer scores on the Focus-Execute and Shift
15    factors of the battery leading the authors to conclude that early lead exposure may be associated
16    with poorer performance on executive/regulatory functions, which are thought to depend on the
17    frontal or prefrontal regions of the brain.
18          Stiles  and Bellinger (1993) administered a neuropsychological battery of tests to 10-year-
19    olds in the Boston Lead Study cohort.  A large number of assessments were made and, as the
20    authors acknowledge, the number of significant associations was about equal to those that would
21    be expected by chance.  However, as in previous studies, tasks that assess attentional behaviors
22    and executive functions tended to be among those for which lead was a  significant predictor of
23    performance. For example, higher blood lead concentrations at 2 years were significantly
24    associated  with lower scores on the Freedom from Distractibility factor  of the Wechsler scales
25    and an increase in the percentage of preservative errors on the Wisconsin Card Sorting Test and
26    the California Verbal Learning Test.
27          Canfield et al. (2003b) conducted a comprehensive examination of the relationship
28    between low-level lead exposure, executive functioning, and learning in children from the
29    Rochester Lead Study cohort at 48 and 54 months of age. The authors used the Shape School
30    Task (Espy, 1997), which requires only knowing simple shape and primary color names.
31    However, embedded in the tasks are protocols requiring inhibition, attention switching, and a

      December  2005                           6-81        DRAFT-DO NOT QUOTE OR CITE

-------
 1    combination of inhibition and switching mental sets. Following covariate-adjustment, blood lead
 2    level at 48 months was negatively associated with children's focused attention while performing
 3    the tasks, efficiency at naming colors, and inhibition of automatic responding. Children with
 4    higher blood lead concentrations also completed fewer phases of the task and knew fewer color
 5    and shape names.
 6          Canfield et al. (2004) also administered portions of the Cambridge Neuropsychological
 7    Testing Automated Battery (CANTAB) to 174 children in the Rochester cohort at approximately
 8    66 months.  Children were tested with the Working Memory and Planning CANTAB assessment
 9    protocols to assess mnemonic and executive functions. Blood lead levels ranged from
10    0-20 |ig/dL in this cohort.  Following covariate adjustment, children with higher blood lead
11    levels showed impaired performance on tests of spatial working memory, spatial memory span,
12    cognitive and cognitive flexibility, and planning  as indexed by tests of intradimensional and
13    extradimensional shifts and an analog of the Tower of London task.
14          Ris et al. (2004) administered an extensive neuropsychological battery to 16-17 year old
15    subjects from the Cincinnati Lead Study cohort.  In addition to executive functions as assessed
16    by the Wisconsin Card Sorting Test and the Rey-Osterrieth Complex Figure, other domains
17    examined included attention, memory, achievement, verbal abilities, visuoconstructional skills,
18    and fine-motor coordination. A factor analysis of scores selected a priori revealed five factors
19    that included Attention. A strong "executive functions" factor did not emerge.  Following
20    covariate-adjustment, the strongest associations between lead exposure and performance were
21    observed for factor scores derived from the Attention component, which included high loadings
22    on variables from the Conners  Continuous Performance Test. However, this relationship was
23    restricted to males as indicated by a strong lead by gender interaction. This obtained gender
24    interaction suggests that neuromechanisms sub-serving attention were affected by lead in this
25    cohort for boys but not girls. This is not surprising given the heightened vulnerability of males
26    for a wide range of developmental perturbations. A substantial gender difference in the
27    incidence of Attention Deficit/Hyperactivity Disorder (ADHD) is well established, and one could
28    speculate that early exposure to lead exacerbates a latent potential  for such problems.
29          Visual-spatial skills have also been also been explored in some depth by a few studies.
30    When investigations of lead-exposed children have used global IQ measures and conducted
31    subscale analyses, it has been observed that Performance IQ or subtests  contributing to the

      December 2005                            6-82         DRAFT-DO NOT QUOTE OR CITE

-------
 1    performance IQ (i.e., Block Design) are frequently among the most strongly associated with
 2    biological indices of exposure (Baghurst et al., 1992; Chiodo et al., 2004; Dietrich et al., 1993a;
 3    McMichael et al., 1988; Wasserman et al., 1994).  Dietrich et al. (1991, 1992) have also
 4    observed that integrated measures of lead exposure over a child's lifetime are most consistently
 5    associated with simultaneous processing abilities,  cognitive functions closely associated with
 6    visual-spatial integration skills and right cerebral functioning (Kaufman and Kaufman, 1983).
 7    In addition, studies employing specific measures of visual-motor integration skills such as the
 8    Developmental Test of Visual Motor Integration (VMI), the Bender Visual-Motor Gestalt Test
 9    and other have found them to be among the most consistently associated with early exposure to
10    lead (Baghurst et al., 1995; Dietrich et al., 1993b;  Wasserman et al., 2000a; Winneke et al.,
11    1990).  In a follow-up of subjects in the Cincinnati Lead Study cohort at 16 years, Ris et al.
12    (2004)  observed a significant association between prenatal maternal blood lead levels and
13    deficits in visual-spatial and constructional skills as indexed by Visual-Constructional factor
14    scores.  Variables with high loadings on this factor included scores on the WISC-III Block
15    Design subtests and selected variables from the Rey Osterrieth Complex Figure.
16          However, it is still unclear whether the domains of attention/executive functions or visual -
17    motor integration per se are specifically sensitive to lead.  This is because there is rarely a one-
18    to-one correspondence between performance on a focused neuropsychological test and an
19    underlying neuropsychological process. Thus, for example, a low score on the Berry VMI may
20    reflect singular or multiple neurobehavioral deficits, including difficulties with graphomotor
21    control, visual  perception, behavioral monitoring (impulsivity), or planning (executive
22    functions).
23
24    6.3.2.4   Disturbances in Behavior, Mood, and Social Conduct
25          The effects of lead on behavior and mood of children has been an area of recent research.
26    Studies conducted prior to 1990 clearly pointed to behavioral problems as potential sequelae of
27    lower level lead toxicity in children.  Several early case control studies linked lead to
28    hyperactivity (David et al., 1972, 1976, 1979). Low levels of lead in blood and/or teeth have
29    been associated with teacher ratings of hyperactive behavior, aggression,  and attention problems
30    (e.g., Fergusson et al.,  1988b; Hatzakis et al., 1985; Silva et al., 1988; Thomson et al., 1989;
31    Yule et al., 1984). In the seminal study by Needleman et al. (1979), children with higher

      December 2005                            6-83        DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentrations of lead in dentin were more likely to be rated unfavorably by teachers on the
 2    dimensions of hyperactivity, impulsivity, and frustration tolerance. New studies reviewed in this
 3    section are summarized in Annex Table AX6-3.6.
 4          While there is no compelling evidence that lead is directly related to ADHD, elevated
 5    blood or tooth lead levels have been linked to behavioral features of ADHD, including
 6    distractibility, poor organization, lacking persistence in completing tasks, and daydreaming
 7    (Bellinger and Rappaport, 2002).  Bellinger et al. (1994) studied the relationship between early
 8    exposure to lead and problem behaviors in the classroom in a cohort of 1,782 children born at
 9    one hospital in Boston.  Lead levels in umbilical cord blood were low (mean 6.8 |ig/dL [SD 3.1])
10    as were tooth  lead levels (mean 3.4 jig/g [SD 2.4]). Teachers filled out the Achenbach Child
11    Behavior Profile (ACBP) which yields both broad and narrow band scales indexing externalizing
12    and internalizing problems.  Cord blood lead levels were not associated with the prevalence or
13    nature of behavioral problems reported by teachers. However, tooth lead level was significantly
14    associated with ACBP Total Problem Behavior Scores (TPBS).  TPBS scores increased by
15    approximately 2 points for each log unit increase in tooth lead.  Statistically significant tooth
16    lead-associated increases in both externalizing and internalizing scores also were noted. Each
17    log unit increase in tooth lead was associated with a 1.5 point increase in scores for these
18    broadband scales assessing under- and overcontrol of behavior.  Only weak associations were
19    noted between tooth lead concentrations and the tendency to score in the clinically significant
20    range on these scales. As the authors noted, it was somewhat surprising that lead exposure was
21    not more strongly related to externalizing behavior problems than with internalizing behavior
22    problems.  This contradicted several earlier investigations, including one by Sciarillo et al.
23    (1992)  described below. It may be that more attention has been accorded under controlled
24    behaviors, because they are more readily visible and disruptive  in settings such as the classroom.
25    Therefore, internalizing problems may be part of the full spectrum of behaviors in which lead's
26    developmental neurotoxicity is expressed in children. The authors also cautioned that residual
27    confounding could not be ruled out, because of the lack of covariate information on parental
28    psychopathology or direct observations of the family environment—a problem not unique to this
29    particular study. Nevertheless, these data are in accord with other studies that social and
30    emotional dysfunction may be another expression of increased lead exposure during the early
31    postnatal period.

      December 2005                            6-84        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Sciarillo et al. (1992) examined the relationship between early exposure to lead and child
 2    behavior in a cohort of 150 subjects in Baltimore, MD. Children were separated into high
 3    exposure (two consecutive blood lead concentrations greater than or equal to 15 |ig/dL) and low
 4    exposure groups. Blood lead also was treated as a continuous variable in regression analyses.
 5    Mothers of 2-5 year old children were administered the Achenbach Child Behavior Checklist
 6    (CBCL) and given the Center for Epidemiologic Studies Depression scale (CESD) as a control
 7    measure. Mean blood lead concentrations were 28.6 |ig/dL (SD 9.3) and 11.3 |ig/dL (SD 4.3) in
 8    the high and low exposure groups, respectively. When compared to the lower exposed group,
 9    children with higher blood lead levels had a significantly higher mean TBPS, and internalizing
10    and externalizing scores. Using regression procedures to control for maternal symptoms on the
11    CESD, blood lead concentrations were still significantly associated with an increase in the
12    TBPS.  Children in the high exposure group were also nearly 3 times more likely to have a TBPS
13    in the CBCL's clinical range. A significantly higher percentage of these children scored in the
14    clinical range for CBCL subscales measuring aggressive and destructive behavioral tendencies.
15          Fergusson et al. (1993) examined the relationship between tooth lead levels and
16    inattention/restlessness in the large national New Zealand study of over 1,000 children at 12 and
17    13 years of age.  Mothers and teachers were asked to respond to a series  of items derived from
18    the Rutter and Conners parental and teacher questionnaires.  The selected items related to the
19    degree to which the child was restless, inattentive, easily distracted, and lacking in concentration.
20    At each age, an index of the subject's propensity to inattentive and restless behavior was
21    obtained by summing the total reports of attention deficit behaviors made by both teacher and
22    parent respondents. Following  adjustment for a wide  range of sociodemographic and other
23    covariate factors, a  statistically  significant, concentration-response relationship was observed
24    between tooth lead  concentrations (range 1-12+ |ig/g) and the inattention/restlessness variable.
25    The authors concluded that their results were consistent with the view that early mildly elevated
26    lead levels were associated with small but long term deficits in attentional behaviors.
27          Two prospective studies have also examined measures of early exposure to lead and
28    behavioral problems as assessed by the Achenbach system. Wasserman  et al.  (1998) studied the
29    relationship between lead exposure and behavior in the Yugoslavian prospective study. The
30    study survey 379 children at 3 years of age with the parent report form of the Achenbach CBCL.
31    Following covariate adjustment, concurrent blood lead levels were significantly associated with

      December 2005                             6-85       DRAFT-DO NOT QUOTE OR CITE

-------
 1    scores on the Destructive Behaviors CBCL subscale, although the variance accounted for by lead
 2    was small compared to sociodemographic factors. As blood lead increased from 10 to 20 |ig/dL,
 3    CBCL subscale scores increased by approximately 0.5 points.  The authors concluded that while
 4    statistically significant, the contribution of lead to social behavioral problems in this cohort was
 5    small compared to the effects of correlated social factors. Burns et al. (1999) examined the
 6    relationship between lead exposure and children's emotional and behavioral problems at ages
 7    11-13 years in the Port Pirie, Australia cohort study. After adjusting for a number of
 8    confounding variables, including HOME scores, maternal psychopathology and the child's IQ,
 9    regression models showed that for an increase in average lifetime blood lead concentrations from
10    10 to 30 |ig/dL, the externalizing behavior problem score increased by 3.5 points (95% CI:  1.6,
11    5.4) in boys, but only by 1.8 points (95% CI: -0.1,  11.1) in girls.  In contrast, internalizing
12    behavior problems were predicted to increase by 2.1 points (95% CI:  0.0, 4.2) in girls, but by
13    only 0.8 points (95% CI: -0.9, 2.4) in boys.
14          Recently, the question of lead's role in delinquent and criminal behavior has been
15    addressed in several investigations.  Previous studies linking attention deficits, aggressive and
16    disruptive behaviors, and poor self-regulation with lead have raised the prospect that early
17    exposure may result in an increased likelihood of engaging in antisocial behaviors in later life.
18          Denno (1990) surveyed 987 Philadelphia African American youths enrolled in the
19    Collaborative Perinatal Project. Data were available from birth through 22 years of age. The
20    analysis initially considered over  100 predictors of violent and chronic delinquent behavior.
21    Repeat offenders presented consistent features such as low maternal education, prolonged male-
22    provider unemployment, frequent moves, and higher lead intoxication (although Denno does not
23    indicate the level of lead intoxication in her report). In male subjects, a history of lead poisoning
24    was among the most significant predictors of delinquency and adult criminality.
25          Needleman et al. (1996) examined the relationship between lead exposure and several
26    measures of behavioral disturbance and delinquent behavior in subjects from the Pittsburgh
27    Youth Study.  The Pittsburgh Youth Study is a prospective study of the developmental course of
28    delinquency (Loeber et al., 1991). The population consisted of 850 boys who were prescreened
29    with an instrument that measured serious and potentially indictable behaviors extracted from the
30    teachers' and parents' CBCL.  Subjects who  scored above the 30th percentile on the risk score
31    and an approximately equal number of subjects randomly selected from the remainder of the

      December 2005                           6-86        DRAFT-DO NOT QUOTE OR CITE

-------
 1    distribution formed the sample (n = 503). Body burden of lead was measured in the tibia by
 2    K-shell XRF. Measures of antisocial behavior were administered at 7 and 11 years of age and
 3    included the Self Reported Antisocial Behavior scale (SRA), the Self Report of Delinquent
 4    Behavior (SRD), and the parents' and teachers' versions of the CBCL. Outcome data were
 5    adjusted for a number of covariates including mother's IQ, SES, childhood medical problems,
 6    and quality of child rearing. Parents of subjects with higher lead levels in bone reported
 7    significantly more somatic complaints, more delinquent and aggressive behavior, and higher
 8    internalizing and externalizing scores. Teachers reported significant increases  in scores on
 9    somatic complaints, anxious/depressed, social problems, attention problems, delinquent
10    behavior, aggressive behavior, and internalizing and externalizing problems in the higher lead
11    subjects.  At 11 years, subjects SRD scores also were significantly related to bone lead levels.
12    More of the high lead subjects had CBCL scores in the clinical range for the CBCL subscales
13    assessing attention problems, aggression, and delinquency. Odds ratios for these outcomes
14    ranged from 1.5 (95% CI: 0.45, 4.9) for parental reports of aggression to 19.5  (95% CI:  8.9,
15    41.6) for attention problems.  The authors concluded that lead exposure was associated with an
16    increased risk for antisocial and delinquent behavior.
17          Dietrich et al. (2001) reported on the relationship between early exposure to lead and
18    juvenile delinquency in 195 subjects from the Cincinnati Lead Study. Subjects were between
19    16 and 17 years of age when examined.  As previously described, this is an inner-city cohort  of
20    urban children exposed to relatively high levels of lead by virtue of their residence in  older,
21    deteriorated housing units. Relationships between prenatal (maternal) and postnatal exposure to
22    lead (through serial blood lead determinations), and antisocial and delinquent behaviors (self-
23    and parental reports) were examined. Parents were administered a questionnaire developed
24    specifically for the study while CLS subjects were given the SRD.  A wide range of candidate
25    covariates and confounders were examined, but the only ones predicting antisocial or delinquent
26    behavior were birth weight, HOME scores, SES, and parental IQ. In multiple linear regression
27    analyses,  prenatal  exposure was  significantly associated with a covariate-adjusted increase in the
28    frequency of parent-reported delinquent and antisocial acts, while prenatal and postnatal
29    exposure  to lead was significantly associated with a covariate-adjusted increase in frequency  of
30    self-reported delinquent and antisocial behaviors, including marijuana use. To clarify the
31    concentration-response relationships, blood lead indices were transformed to categorical

      December 2005                            6-87        DRAFT-DO NOT  QUOTE OR CITE

-------
 1    variables and least-square means were calculated from an analysis of covariance procedure.
 2    Subjects in the highest prenatal blood lead category (>10 |ig/dL) engaged in 2.3 more delinquent
 3    acts over the preceding 12 months than subjects in the lowest category (<5 |ig/dL).  Using
 4    average childhood blood lead levels, subjects in the medium (16-20 |ig/dL) and highest
 5    (>20 |ig/dL) category engaged in approximately 1.5 more delinquent acts compared to the lowest
 6    category (< 10 |ig/dL). Subjects in the highest 78-month blood lead category (>15 |ig/dL)
 7    engaged in 4.5 more delinquent acts than subjects in the lowest category (<5  |ig/dL). The
 8    authors concluded that lead might play a measurable role in the epigenesis of behavioral
 9    problems in inner-city children independent of other social and biomedical cofactors assessed in
10    the study.
11          Needleman et al. (2002) conducted a case-control study where they examined the levels of
12    lead in bone of 194 adjudicated delinquents and 146 non-delinquent community controls.
13    Subjects were recruited from high schools in the city of Pittsburgh and environs of Allegheny
14    County, PA. Since many delinquents are not arrested or adjudicated, care was taken to ensure
15    that unidentified delinquents did not populate the control group.  Potential control subjects were
16    excluded from the analyses if they were found to have a Juvenile Court record or an SRD score
17    above the 90th percentile. Tibial bone lead was measured by K-shell XRF. Covariates included
18    race, parental  education and occupation, presence of two parental figures in the home, number of
19    children in the home, and neighborhood crime rate. Logistic regression analyses were
20    undertaken to model the association between delinquent status and bone lead concentration.
21    Cases had  significantly higher average concentrations of lead in tibia than controls (11.0 jig/g
22    [SD 32.7] versus 1.5 jig/g [SD 32.1]).  Stratified analyses revealed this was true for both white
23    and African American subjects. Following adjustment for covariates, adjudicated delinquents
24    were four times more likely  to have bone lead concentration greater than 25 |ig/g than controls
25    (odds ratio of 4.0 [95% CI:  1.4, 11.1]).  The effect of lead on delinquency was found to be
26    substantial in this study.  Bone lead level was the second strongest factor in the logistic
27    regression models, exceeded only by race.  In models stratified by race, bone lead was exceeded
28    as a risk factor only by single parent status.  The authors concluded that elevated body lead
29    burdens were  associated with elevated risk for adjudicated delinquency.
30          The extension of lead effects into delinquent and criminal behavior is significant for both
31    the individual and society as a whole.  The particular biological mechanisms that may underlie

      December 2005                             6-88       DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead's effects on aggression, impulsivity, and poor self-regulation are not clearly understood.
 2    However, lead impacts a large number of sites and processes in the brain that are involved in
 3    impulse control (Lidsky and Schneider, 2003).  Needleman et al. (2002) proposed another
 4    pathway.  In addition to lead's direct impact on brain development and neuronal function, lead
 5    exposure may increase risk of delinquency through a separate, indirect route: impaired cognitive
 6    abilities and academic performance. In other words, students who have difficulties in school and
 7    fail to achieve academic goals are more likely to become lawbreakers.
 8
 9    6.3.2.5   Sensory Acuities
10          In comparison to cognitive outcomes, there has been relatively less interest in the effects
11    of lead on sensory functions.  However, there are clear indications that lead exposure during the
12    developmental period has an impact on complex aspects of visual and auditory acuities.  Much of
13    this work has been carried out in animal models (Otto and Fox, 1993). Epidemiologic studies
14    have typically assessed hearing thresholds and features of auditory processing in lead-exposed
15    children.  Studies reviewed in this section are summarized in Annex Table AX6-3.7.
16          Schwartz and Otto (1987) observed significant lead-associated elevations in pure-tone
17    hearing thresholds at various frequencies within the range of human speech among over 4,500
18    4-19 year old subjects in NHANES II. In a later study, this finding was replicated in a sample of
19    over 3,000 6-19 year old subjects in the Hispanic Health and Nutrition Examination Survey
20    (HHANES) (Schwartz and Otto, 1991). An increase in blood lead from 6  to 18 ng/dL was
21    associated with a 2 db loss in hearing at all frequencies, and an additional  15% of children had
22    hearing thresholds that were below the standard at 2,000 Hz.  These relationships continued at
23    blood lead levels less than 10 |ig/dL.
24          Dietrich et al. (1992) assessed the relationship between scores on a test of central auditory
25    processing (SCAN) and prenatal/postnatal blood lead concentrations in 215 children 5 years of
26    age drawn from the Cincinnati Lead Study. Higher prenatal, neonatal, and postnatal (up to
27    concurrent) blood lead concentrations were associated with more incorrect identification of
28    common monosyllabic words presented under conditions of filtering (muffling). Other variables
29    associated with impaired central auditory processing included the results of pure-tone
30    audiometry  testing, social class, HOME scores, birth weight, gestational age, a measure of
31    obstetrical complications, and consumption of alcohol during pregnancy.  Following adjustment

      December 2005                           6-89       DRAFT-DO NOT QUOTE OR CITE

-------
 1    for these covariates, neonatal and postnatal blood lead levels remained significantly associated
 2    with impaired performance on the Filtered Word subtest, more prominently in the right ear.
 3    In the right ear, the Filtered Word subtest score decreased by 0.7 points (p < 0.05; 95% CI not
 4    presented) for a 10 |ig/dL increase in lifetime average blood lead levels.
 5          Osman et al. (1999) examined the relationship between concurrent blood lead levels and
 6    hearing loss in 155 children 4-14 years of age living in an industrial region of Poland. Blood
 7    lead levels ranged from 1.9 to 28 |ig/dL (median 7.2 jig/dL). Hearing thresholds increased
 8    significantly with higher blood lead levels at all  frequencies (500-8,000 Hz).  This relationship
 9    remained statistically significant when restricted to children with blood lead levels below
10    10 |ig/dL.
11          A limited number of epidemiologic studies provide supportive evidence of a relationship
12    between lead exposure and auditory processing.  Lead-related deficits in hearing and auditory
13    processing may be one plausible mechanism by  which an increased lead burden might impede a
14    child's learning (Bellinger, 1995).
15
16    6.3.2.6   Neuromotor Function
17          Relatively few studies have focused on neuromotor deficits as an outcome of early lead
18    exposure.  However, those that have examined motor functions in lead-exposed children often
19    report positive findings.  Studies reviewed in this section are summarized in Annex Table
20    AX6-3.8.
21          In an early study, unsteadiness, clumsiness, and fine-motor dysfunctions were noted in a
22    group of mildly symptomatic lead-poisoned children in Boston, with such effects persisting long
23    after medical treatment (Pueschel et al., 1972).  A study of moderately exposed children living in
24    the vicinity of a longstanding lead smelter in Greece found that children with blood lead levels of
25    35-60 |ig/dL had significantly lower scores on both the Gross and Fine Motor Composite scores
26    from the Oseretsky scales when compared to controls (Benetou-Marantidou et al., 1988).
27          Only two modern prospective  studies of lead have assessed motor development in a
28    comprehensive manner.  Dietrich et al. (1993b) investigated the association between lead
29    exposure and motor developmental status in 245 children 6 years of age in the Cincinnati Lead
30    Study cohort.  Following covariate adjustment, they found that postnatal lead exposure was
31    significantly associated with poorer scores on measures of bilateral coordination, visual-motor

      December 2005                           6-90        DRAFT-DO NOT QUOTE OR CITE

-------
 1    control, upper-limb speed and dexterity, and the fine motor composite from the Bruininks-
 2    Oseretsky scales. Neonatal, but not prenatal, blood lead concentrations also were significantly
 3    associated with poorer scores on upper-limb speed and dexterity and the fine motor composite.
 4    The strongest and most consistent relationships were observed with concurrent blood lead levels
 5    (mean 10.1 |ig/dL [SD 5.6]).  A 10 |ig/dL increase in concurrent blood lead levels was associated
 6    with a 4.6 point (95% CI:  2.1, 7.1) decline in the fine motor composite score.  In the same
 7    Cincinnati cohort, postnatal lead exposure was associated with greater postural instability as
 8    assessed by a microprocessor-based strain gauge platform system (Bhattacharya  et al., 1995).
 9    When assessed at 16 years of age, 78-month postnatal blood lead levels were significantly
10    associated with poorer fine-motor skills as indexed by covariate-adjusted factor scores derived
11    from a factor analysis of a comprehensive neuropsychological battery (Ris et al., 2004).  The
12    variables loading highly on the fine-motor component came from the grooved  pegboard and
13    finger tapping tasks.
14          Some  results of the Cincinnati Lead Study were replicated by Wasserman et al. (2000a)
15    in the Yugoslavian Prospective Study. The authors adapted the Bruininks-Oseretsky Test of
16    Motor Proficiency for use in their population residing in two towns in the province of Kosovo.
17    The measure of exposure was the log of the lifetime average blood lead concentration through
18    54 months of age. Following covariate-adjustment, average  childhood blood lead concentrations
19    were associated with poorer fine motor and visual motor function, but were found to be unrelated
20    to gross motor function.
21          A recent study by Despres et al. (2005) of multiple exposures including lead, mercury,
22    and polychlorinated biphenyls found that  only blood lead concentrations measured at the time of
23    assessment were associated with neuromotor functions in 110 preschool Inuit children residing in
24    Canada.  The mean blood lead level was 5.0 |ig/dL (range 0.8-27.1). Blood lead levels were
25    significantly  associated with increased reaction time, sway oscillations, alternating arm
26    movements, and action tremor.  Ten percent of the children had blood lead levels greater than
27    101 |ig/dL. After eliminating these children from the analyses, results remained  significant for
28    reaction time, sway oscillations, and alternating arm movements. These findings indicated that
29    neuromotor effects of lead occurred at blood lead concentrations below 10 |ig/dL.
      December 2005                           6-91        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.3.2.7   Brain Anatomical Development and Activity
 2          Electrophysiological evaluations have been conducted on lead-exposed children in
 3    attempts to obtain a more direct measure of the toxicant's impact on the nervous system. Much
 4    of this work was conducted by Otto and colleagues during the 1980s (e.g., Otto et al., 1985).
 5    Studies reviewed in this section are summarized in Annex Table AX6-3.9.  These studies have
 6    demonstrated effects of lead on neurosensory functioning (auditory and visual evoked potentials)
 7    within a broad range of exposures (Otto and Fox,  1993).
 8          Rothenberg et al. (1994) reported that higher maternal blood lead levels at 20 weeks of
 9    pregnancy were associated with increased I-V and III-V interpeak intervals in the brainstem
10    auditory evoked response recorded in 1-month-old infants. Mean maternal blood lead level at
11    20 weeks in this subsample from the Mexico City Prospective Study was only 7.7 |ig/dL with a
12    range of 1-30.5 |ig/dL. Rothenberg et al. (2000) repeated these measurements with a larger
13    group of 5-7 year old children (n = 133). In contrast to their previous findings, prenatal blood
14    lead levels at 20 weeks were associated with decreased interpeak intervals. However, after
15    fitting a nonlinear model to their data, they observed that I-V and III-V interpeak intervals
16    decreased as blood lead rose from 1 to 8 |ig/dL and increased as blood lead rose from 8 to
17    30 |ig/dL. The biphasic effect was only observed with maternal blood leads at 20 weeks of
18    pregnancy.  Increasing postnatal blood lead at 12 and 48 months was related to decreased
19    conduction intervals for I-V and III-V interpeak intervals across the entire blood lead range.
20          The methods of Magnetic Resonance Imaging (MRI)  and Magnetic Resonance
21    Spectroscopy (MRS) have recently been applied in studies of lead-exposed children.  Trope et al.
22    (1998) were the first to apply MRI and MRS in an evaluation of a lead-exposed subject.  The
23    subject was a 10 year old boy who had a history of elevated blood lead levels as a toddler (e.g.,
24    51 |ig/dL at 38 months). The subject was compared to his 9-year-old unexposed cousin. The
25    investigation was particularly focused on 7V-acetylaspartate, a metabolite shown to decrease in
26    processes that involve neuronal and axonal loss. Both children presented with normal volumetric
27    MRI, MRS revealed a significant alteration in brain metabolites, with a reduction in N-
28    acetylaspartate:creatine ratio for both gray an white matter compared to the subject's cousin.
29    Trope et al. (2001) performed identical MRI and MRS studies on a sample of 16 subjects with a
30    history of elevated blood lead levels before five years of age (23 to 65 |ig/dL). Average age at
31    time of evaluation was 8 years.  These subjects were compared to age-matched controls

      December 2005                            6-92       DRAFT-DO NOT QUOTE OR CITE

-------
 1    composed of siblings or cousins. Control subjects had blood lead levels that never exceeded
 2    10 |ig/dL. Although all of the participants had normal MRI examinations, the lead-exposed
 3    subjects exhibited a significant reduction in 7V-acetylaspartate:creatine and phosphocreatine
 4    ratios in frontal gray matter compared to controls.
 5          Meng et al. (2005) performed MRI and MRS studies on children with blood lead
 6    concentrations >27 |ig/dL (n = 6) and age- and gender-matched controls with blood lead
 7    concentrations <10 |ig/dL (n = 6).  The average age at time of evaluation was approximately
 8    11 years. Subjects came from the Anhui province in China. Lead-exposed children had an
 9    average blood concentration of 37.7 |ig/dL (SD 5.7) while controls averaged 5.4 |ig/dL (SD 1.5).
10    MRS was used to measure 7V-acetylaspartate, choline-containing compounds, and total creatine
11    in the frontal lobes and hippocampus in cases and controls.  All children presented with normal
12    MRI with no evidence of structural abnormalities. However, peak values of 7V-acetylaspartate,
13    choline, and creatine in all four brain regions were reduced in lead-exposed children relative to
14    controls.  The authors concluded that the reduced brain 7V-acetylaspartate levels they observed in
15    cases may be related to decreased neuronal density or neuronal loss. Furthermore, reduced
16    choline signal may indicate decreased cell membrane turnover or myelin alterations that can lead
17    to central nervous system hypertrophy, while lower creatine may indicate reduced neuronal cell
18    viability.
19          Using functional MRI (fMRI),  Cecil et al.  (2005) examined the influence of childhood
20    lead exposure on language function in a subsample of 48 young adults from the Cincinnati Lead
21    Study.  At age 20-23 years, subjects performed an integrated verb generation/finger tapping
22    paradigm. Higher childhood average blood lead levels were significantly associated with
23    reduced activation in  Broca's area, a recognized region of speech production in the left
24    hemisphere. This association remained statistically significant after adjustment for the subject's
25    latest IQ assessment.  Higher childhood blood lead levels also were associated with increased
26    activation in the right temporal lobe, the homologue of Wernicke's area (an area associated with
27    speech production) in the left hemisphere.  The results of this study suggest elevated childhood
28    lead exposure strongly influences neural substrates of semantic language function  on normal
29    language areas with concomitant recruitment of contra-lateral regions resulting in  a striking,
30    dose-dependent atypical organization  of language function.
31

      December 2005                             6-93        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.3.2.8   Gene-Environment Interactions in the Expression of Lead-Associated
 2             Neurodevelopmental Deficits
 3          The discussion of gene-environment interactions with respect to lead exposure
 4    encompasses differential susceptibilities with respect to race, gender, and genetic polymorphisms
 5    associated with lead metabolism, and neurotransmitter metabolism and function. While the
 6    differential effects of lead on neurodevelopment have been studied to some extent with respect to
 7    race and gender, very little work has been accomplished with respect to specific genetic
 8    polymorphisms.
 9          In the U.S., African-American children are at increased risk for having an elevated blood
10    lead level compared with white children. For example, in the last two NHANES surveys,
11    African-American children were found to have significantly higher blood lead levels than whites,
12    even after adjusting for urban residential status and family income (Brody et al., 1994; Mahaffey
13    et al., 1982). However,  reliable differences with respect to lead's effects on neurodevelopmental
14    morbidity as a function of race have not been reported with consistency.
15          Most surveys find that boys have higher blood lead levels than girls. The data are less
16    clear with respect to gender-related differences in lead-associated neurodevelopmental
17    morbidities. At various assessments from birth to adolescence, a greater male vulnerability has
18    been noted in the Cincinnati Lead Study (e.g., Dietrich et al., 1987b; Ris et al., 2004).  Data from
19    a cross-sectional study in England showed that the lead-IQ deficit association was more
20    pronounced in boys at 6 years of age (Pocock et al., 1987). However, in a study of 764 children
21    in Taiwan, it was found that the relationship between lead exposure and IQ scores was
22    substantially stronger in girls (Rabinowitz et al., 1991). In the Port Pirie cohort study, lead
23    effects on cognition were significantly stronger in girls at ages 2, 4, 7, and 11-13 years
24    (Baghurst et al., 1992; McMichael et al., 1992; Tong et al., 2000).
25          At least two genetic polymorphisms have been identified that can influence the
26    absorption, retention and toxicokinetics of lead in humans (Onalaja and Claudio, 2000). The
27    ALAD gene has been the most studied but, as yet, the consequences of the different alleles for
28    susceptibility to the neurodevelopmental consequences of lead exposure are unclear.  Individuals
29    with the ALAD12 or ALAD22 polymorphism tend to have higher blood lead levels than those
30    with ALAD11. ALAD2 could increase vulnerability by raising blood lead levels or decrease it
31    by maintaining lead in a sequestered state in the bloodstream. Only one pediatric  study has
      December 2005                           6-94        DRAFT-DO NOT QUOTE OR CITE

-------
 1    examined this directly. Bellinger et al. (1994) found that subjects with the ALAD2
 2    polymorphism tended to have lower dentin levels than those with ALAD1.  This is consistent
 3    with the concept that increased affinity of the ALAD2 polymorphism inhibits entry of lead from
 4    the blood stream into other tissues. After adjustment for exposure level, Bellinger et al. found
 5    that adolescents with the ALAD2 polymorphism performed better in the areas of attention and
 6    executive functioning assessed in their study when compared to subjects with the ALAD1
 7    polymorphism. However, as there were only 5 subjects with the ALAD2 form, meaningful
 8    statistical comparisons could not be made.
 9          The other gene that has been studied is the vitamin D receptor or VDR gene. This gene is
10    involved in calcium absorption through the gut. Research on lead workers has shown that
11    variant VDR alleles modify lead concentrations in bone, and the rate of resorption and excretion
12    of lead over time (Schwartz et al.,  2000c).  Haynes et al. (2003) examined the relationship
13    between the VDR Fokl polymorphism and blood lead concentrations in 275 children enrolled in
14    the Rochester Longitudinal Study.  It was hypothesized that children homozygous for the
15    F allele—a marker for increased calcium absorption—would have higher blood lead
16    concentrations than heterozygotes and children homozygous for the/allele, after adjusting for
17    environmental sources of lead (floor dust lead).  A  statistically significant interaction was found
18    between floor dust lead loading and VDR-Fokl genotypes on blood lead concentration, with the
19    FF genotypes having the highest adjusted mean blood lead concentrations at 2 years of age.
20    Consistent with other reports, Haynes et al. (2003)  also found that African American children
21    were significantly more likely to have the VDR-FFthan were non-African American children.
22    The ability of African American children to have increased calcium absorption may partially
23    explain the higher blood lead  concentrations observed in African American children.
24    Unfortunately, there have been no studies to indicate which, if any, of the VDR polymorphisms
25    are associated with increased vulnerability to the neurodevelopmental toxicity of lead.
26
27    6.3.2.9   Reversibility of Lead-related Neurodevelopmental Deficits Associated
28             with Prenatal and Postnatal Exposure
29          The apparent persistence of the neurodevelopmental effects of lead observed into later
30    childhood and adolescence has resulted in a widely held view that the damage to the central
31    nervous system and resulting deficits in neurobehavior are irreversible. The ramifications of the
      December 2005                           6-95        DRAFT-DO NOT QUOTE OR CITE

-------
 1    effects of lead on neurodevelopment depend not only on the extent of the initially observable
 2    effects in early childhood, but also on their enduring consequences for cognition, attainment, and
 3    behavior over the lifetime of the individual. Studies reviewed in this section are summarized in
 4    Annex Table AX6-3.10.
 5          Since 1990, several studies attempted to eliminate or at least reduce lead-associated
 6    neurodevelopmental damage through nutritional and/or pharmacological interventions.
 7    Optimism that such interventions might be effective was raised by a New York study published
 8    in the early 1990s (Ruff et al., 1993). In an observational study, children 13 to 87 months old
 9    with blood lead levels between 25 and 55  |ig/dL were given chelation with EDTA and
10    therapeutic iron when clinically indicated.  The children were then followed for 6 months.  Those
11    whose blood lead levels fell the most had improved cognitive test scores, independent of whether
12    they had been given iron or chelation therapy. Prior to this publication, the National Institute for
13    Environmental Health Sciences (NIEHS) was already in the process of planning a multicenter
14    clinical trial to determine if a recently licensed oral chelating drug (dimercaptosuccinic acid or
15    "succimer") might diminish the neurodevelopmental impact of lead in children with blood lead
16    levels between 20 and 44 ng/dL (Rogan et al., 1998).
17          The Treatment of Lead-Exposed Children (TLC) study was originally designed to test the
18    hypothesis that children with moderate blood lead levels who were given succimer would have
19    better scores than children given placebo on a wide range of tests measuring cognition,
20    neuropsychological functions, and behavior at 36 months of follow-up (Rogan et al., 2001).
21    TLC  enrolled 780 children from four clinical sites into a randomized, placebo-controlled,
22    double-blind trial of up to three 26-day courses of treatment with succimer. Most children
23    lived in deteriorating inner-city housing. Seventy-seven percent of the subjects were African
24    American.  Succimer was effective in lowering the blood lead levels of subjects on active drug
25    during the first 6 months of the trial. However, after 1 year, differences in the blood lead levels
26    of succimer and placebo groups had virtually disappeared. All data analyses were conducted on
27    an intent-to-treat basis. At 36 months of follow-up, the mean IQ score on the WPPSI-R of
28    children given active drug was 1 point lower than that of children administered placebo, and
29    children given succimer evinced more behavioral problems as rated by the primary caregiver on
30    the Conners Parent Rating Scale.  Children given succimer scored marginally better on the
31    Developmental Neuropsychological Assessment (NEPSY), a battery of tests designed to measure

      December 2005                           6-96        DRAFT-DO NOT QUOTE OR CITE

-------
 1    neuropsychological deficits that can interfere with learning.  However, all of these differences
 2    were statistically nonsignificant.
 3          Although results for the first wave of follow-up for TLC were consistently negative for
 4    drug effects on cognition and behavior, they were not necessarily conclusive.  Lead may affect
 5    higher-level neurocognitive processes that are inaccessible, difficult to assess, or absent in the
 6    preschool age child.  In older children, scores on psychometric measures are more precise and
 7    reliable, a wider and more differentiated range of abilities can be examined, and early academic
 8    performance and social functioning outside the home environment can be evaluated. Therefore,
 9    TLC followed the cohort into the first years of elementary education to determine whether these
10    later emerging neurodevelopmental functions were spared the effects of lead in treated children
11    compared to placebo controls (Dietrich et al., 2004). While remaining within  the limits of
12    hypothesis driven inference, a comprehensive battery of tests were administered to TLC subjects
13    at 7 and 7.5 years of age. These included assessments of cognition,  learning, memory, global
14    intellectual attainment, attention/executive functions, psychiatric status, behavioral and academic
15    conduct, neurological functioning, and motor speed. However, treatment with succimer resulted
16    in no benefit in cognitive, behavioral, neurological, and neuromotor endpoints. Indeed, children
17    treated with succimer fared worse than children in the placebo group in several areas, including
18    linear growth, hospitalized and  outpatient injury events in the first 3 years of follow-up, and
19    neuropsychological deficits as assessed by the Attention and Executive Functions core domain
20    score from the NEPSY. The authors concluded that these latest follow-up data confirmed their
21    previous finding that the TLC regimen of chelation therapy is not associated with
22    neurodevelopmental benefits in children with blood lead levels between 20 and 44 |ig/dL.
23    Furthermore, these results emphasize the importance of taking environmental  measures to
24    prevent exposure to lead in light of the apparent irreversibility of lead-associated
25    neurodevelopmental deficits.
26          In addition to pharmacological interventions,  a few studies have attempted to remediate or
27    prevent lead-associated neurodevelopmental deficits through nutritional supplementation.
28    Recent studies attempting to reduce lead absorption through mineral hypersupplementation have
29    been disappointing (Sargent et al.,  1999). However, to date there has been only one controlled
30    clinical trial involving lead-exposed children where central nervous  system outcomes have been
31    the focus of study. Kordas et al. (2005) and Rico et al. (2005) conducted a double-blind

      December 2005                            6-97       DRAFT-DO NOT QUOTE OR CITE

-------
 1    nutritional supplementation trial among 602 first grade children in the city of Torreon in northern
 2    Mexico. The city is located near a metal foundry that has been a source of lead contamination in
 3    the community.  The average blood lead concentration at baseline was 11.5 |ig/dL (SD 6.1).
 4    About half of the children had blood lead concentrations in excess of 10 |ig/dL.  Subjects
 5    received 30 mg ferrous fumarate, 30 mg zinc oxide, both, or placebo daily for 6 months. In their
 6    first report, the principal outcome assessment taken at baseline and at follow-up was the parent
 7    and teacher forms of the Conners Rating Scales. There were no consistently significant
 8    treatment effects and the authors concluded that this regimen of supplementation did not result in
 9    improvements in ratings of behavior in lead-exposed children over 6 months. In addition to
10    behavior, the authors assessed cognitive functioning with 11 tests of memory, attention,  visual -
11    spatial abilities,  and learning. There were no consistent or lasting differences in cognitive
12    performance among treatment groups confirming the earlier conclusion that nutritional
13    supplementation alone is not effective in eliminating or reducing the impact of early lead
14    exposure on functional neurodevelopment.
15          Children's blood lead levels generally decline after they peak at somewhere around
16    2 years of age.  However, the degree of decline is a function of a number of factors including
17    previously acquired body burden and sources of continuing exposure.  Some observational
18    studies have examined the extent to which the rate of decline in blood lead levels is associated
19    with improvements in neurocognitive status.  Tong et al. (1998) assessed the reversibility of the
20    cognitive effects of lead in early childhood in the Port Pirie, Australia cohort study. A total of
21    375 children were followed to the age of 11-13 years. Average blood lead concentrations
22    decreased from 21.2 |ig/dL at 2 years to 7.9 |ig/dL at 11-13 years. However, scores on
23    standardized measures of intellectual attainment administered at 2,  4, 7, and 11-13 years of age
24    in children whose blood lead levels declined the most were not significantly improved over those
25    obtained by children with a more shallow decline in body burden.
26          Liu et al.  (2002) made use of the TLC succimer trial data set (Rogan et al., 2001) to
27    examine the question of reversibility.  As reviewed above, intent-to-treat analyses revealed no
28    benefits of chelation on neurodevelopmental  indices beyond 6 months of treatment. Thus, the
29    scores on the cognitive tests from the two treatment groups could be analyzed either within the
30    treatment groups or as a whole. Data from 741 children were available for analyses. Mean
31    blood lead levels in TLC subjects were 26.2 |ig/dL at baseline, 20.2 |ig/dL at the 6-month

      December 2005                             6-98       DRAFT-DO NOT QUOTE OR CITE

-------
 1    follow-up, and 12.2 |ig/dL at the 36-month follow-up. Mean declines in blood lead levels were
 2    6.0 |ig/dL from baseline to 6-month follow-up, 14.1 |ig/dL from baseline to 36-month follow-up,
 3    and 8.0 |ig/dL from 6- to 36-month follow-ups. Blood lead levels declined more quickly in the
 4    first 6 months in the succimer group than in the placebo group, but the mean blood lead levels
 5    were very similar at baseline and at the 36-month follow-up. Prior to examining changes in
 6    blood lead levels in relationship to changes in cognitive test scores, it was verified that baseline
 7    and later blood lead levels were indeed significantly associated with deficits on measures
 8    administered at specific points in the study after adjustment for sociohereditary factors surveyed
 9    in the study including maternal IQ. Unlike in the New York study by Ruff et al. (1993), Liu
10    et al. (2002) found no overall effect of changing blood lead level on changes in cognitive test
11    score from baseline to 6 months. However, during the follow-up from baseline to 36 months and
12    from 6 to 36 months, falling blood lead levels were significantly associated with increased
13    cognitive test scores, but only because of an association in the placebo group. Cognitive test
14    scores increased by 2  points overall and 4 points in the placebo group when blood lead levels
15    declined by 10 |ig/dL from baseline to 36 months. There is a possibility that the succimer drug
16    regimen blunted the beneficial effect. Due to the inconsistency in the results, the data do not
17    provide strong supportive evidence that lead-induced cognitive impairments are reversible.
18    Therefore, primary prevention and preventing additional increases in blood lead levels among
19    children whose blood lead levels are high remain the only effective means of dealing with
20    lead toxicity.
21
22    6.3.2.10  Periods of Enhanced Developmental Susceptibility to Central Nervous
23             System Effects of Environmental Lead
24          It has been difficult to identify discrete periods of development when the fetus or child is
25    particularly susceptible to lead's effects on neurodevelopment. When the prospective studies of
26    lead and child development were underway, it was hoped that this methodological approach
27    would be revealing. However, these studies observed that age strongly predicted the period of
28    peak exposure  (around 18-27 months when there  is maximum hand-to-mouth activity), making it
29    difficult to distinguish whether greater neurotoxic effects resulted from increased exposure or
30    enhanced susceptibility at a particular age. Furthermore,  children with the highest blood lead
31    levels tended to maintain their rank order relative to their lower exposed peers throughout these
      December 2005                           6-99        DRAFT-DO NOT QUOTE OR CITE

-------
1
2
3
4
5
6
7
studies (e.g., Dietrich et al., 1993a; McMichael et al., 1988), limiting the degree to which
investigators could identify any particular period of development as critical.
      From the perspective of human neurodevelopmental biology, one could argue that the first
3 years of life should represent a particularly vulnerable period.  Maximal ingestion of lead
coincides with the same period of time when major events are occurring in the development of
the central nervous system including some neurogenesis, rapid dendritic and axonal outgrowth,
synaptogenesis, synaptic pruning, and programmed apoptosis (see Figure 6-3.4).
                 Newborn
                   A
                             1 month
                               B
e months
  C
2 years
  D
    Figure 6-3.4.   Golgi-stained section of human cerebral cortex taken from equivalent
                   areas of the anterior portion of the middle frontal gyrus at different ages.
                   Although the packing density of cortical neurons does not appear to change,
                   there is a tremendous increase in the complexity of dendritic arborizations
                   with increasing age with maximal density occurring between two and three
                   years of age.
    Source: Nolle (1993).
    December 2005
                                         6-100
     DRAFT-DO NOT QUOTE OR CITE

-------
 1          This belief that the first 3 years represents a critical window of vulnerability is evident in
 2    the lead literature (Chen et al., 2005).  Two major meta-analyses of the relationships between
 3    childhood lead exposure and IQ focused primarily on the strength of the association between IQ
 4    at school  age and blood lead concentrations at 2 years of age or average blood lead levels up to
 5    3 years of age (Pocock et al, 1994; Schwartz, 1994). Neither meta-analysis considered the
 6    importance of concurrent blood lead associations in older children. The focus on these particular
 7    age groups implied that the interpretation most consistent with the overall results was that peak
 8    blood lead concentration, achieved somewhere between 1 and 3 years of age, was most likely
 9    responsible for the cognitive effects observed years later. These meta-analyses were highly
10    influenced by findings from the Boston prospective study where blood lead concentrations at
11    2 years of age have been exclusively and consistently associated with lower IQ and academic
12    achievement (Bellinger et al., 1992).
13          This particular interpretation of the lead literature has also influenced screening programs
14    (which focus on 1 and 2 year olds), clinical trials that recruit children during the first 3 years of
15    life, and current interpretation of the cross-sectional literature.  For example, the report by
16    Lanphear et al. (2000) that school-age children enrolled in the NHANES III survey displayed a
17    significant inverse relationship between concurrent blood lead  concentrations and measures of
18    IQ and academic achievement at blood lead concentrations below 10 |ig/dL was interpreted by
19    some to reflect the effects of the children's higher blood lead concentrations when they were
20    between 1-3 years of age.
21          However, it is not clear that only the period of peak blood lead concentration matters in
22    terms of the risks for neurodevelopmental morbidity. Other prospective studies of children with
23    both high and low lead exposures  found concurrent or lifetime average blood lead levels to be
24    more strongly associated with school age IQ and other measures of neurodevelopment (Canfield
25    et al., 2003a; Dietrich et al., 1993a,b; Tong et al., 1996; Wasserman et al., 2000b).  One study
26    has recently attempted to address this question  directly.  Chen et al. (2005) sought to clarify the
27    strength of the association between IQ and blood lead at various time points, to examine whether
28    the cross-sectional associations observed in school age children 84-90 months of age represented
29    residual effects from 2 years of age or "new" effects emerging among these children, and how
30    the change in blood lead over time is related to IQ at later ages. Chen et al. (2005) used data on
      December 2005                           6-101       DRAFT-DO NOT QUOTE OR CITE

-------
 1    780 children from the previously described TLC multicenter clinical trial (Dietrich et al., 2004;
 2    Rogan et al., 2001) to examine these relationships. Homogeneity between the two treatment
 3    groups was verified. There were no statistical differences between succimer and placebo groups
 4    in either blood lead concentrations or cognitive scores at the time points under consideration.
 5    At baseline, children were given the Bayley Scales of Infant Development.  The children's full
 6    scale IQ at the 36-month follow-up was measured with the WPPSI-R.  At the 60 month follow-
 7    up, IQ was assessed with the WISC-III. All neurodevelopmental outcomes were adjusted for
 8    clinical center, race, gender, language, parent's education, parent's employment, single parent
 9    family, age at blood lead concentration, and caregiver's IQ.
10          Figure  6-3.5 displays the mean IQ at current and subsequent ages by quartiles of blood
11    lead measured at 2, 5, and 7 years of age.  The concurrent blood lead concentration always had
12    the strongest association with IQ.  As the children aged, the relationship grew stronger. The
13    peak blood lead concentration from baseline to 7 years of age was not associated with IQ at
14    7 years of age. Furthermore, in models including both prior and concurrent blood lead
15    concentrations, concurrent blood lead was always more predictive of IQ. Adjustment for prior
16    IQ did not fundamentally change the  strength of the association with concurrent blood lead
17    concentration. Chen et al. (2005) found a stronger relationship between IQ at 7 years of age and
18    blood lead concentration at 7 years compared with blood  lead at 2 years of age. A similar
19    relationship was observed between IQ and blood lead at 5 years of age. The strength of the
20    cross-sectional associations increase over time, despite lower blood lead concentrations in  older
21    children.  These data support the idea that lead exposure continues to be toxic to  children as they
22    reach school age, and does not lend support to the interpretation that all of the damage is done by
23    the time the child reaches 2 to 3 years of age. These findings also imply that  cross-sectional
24    associations observed in children, such as the study recently conducted by Lanphear et al. (2000)
25    using data from NHANES III should  not be dismissed. Chen et al. (2005) concluded that if
26    concurrent blood lead remains important until school age for optimum cognitive  development,
27    and if 6 and 7 year olds  are as or more sensitive to lead effects as 2 year olds, then the difficulties
28    in preventing  lead exposure are magnified but the potential benefits of prevention are greater.
29
      December 2005                           6-102       DRAFT-DO NOT QUOTE OR CITE

-------
                    £
                    o
                    u
                    V)
                    a
                       95
                       90 -
                   85
                       80 -
                       75
                                       10
                                         15
20
25
 I
30
 I
35
                                                             ((jg/dL)
40
                                    -0- MDI at baseline by baseline blood lead concentration
                                    -Q- IQ at age 5 by baseline blood lead concentration
                                    ~^V IQ at age 7 by baseline blood lead concentration
                                    ~~X~~ IQ at age 5 by blood lead concentration at age 5
                                    •jf  IQ at age 7 by blood lead concentration at age 5
                                       IQ at age 7 by blood lead concentration at age 7
     Figure 6-3.5. Full scale IQ test scores by previous or concurrent blood lead concentration.
                  Each data point shows the mean IQ test scores of children measured at
                  baseline or at two follow-ups, grouped by quartiles of blood lead
                  concentration.  The abscissa of each point is the middle value of each
                  blood lead concentration category.
     Source:  Chen et al. (2005).
4
5
6
6.3.2.11   Effect of Environmental Lead Exposure on Neurodevelopment at the
          Lower Concentration Range
      Over the last three decades, epidemiologic studies of lead and child development have
demonstrated inverse associations between blood lead concentrations and children's IQ and other
outcomes at successively lower levels.  The 1986 Addendum and 1990 Supplement concluded
that neurobehavioral effects were related to blood lead levels of 10 to 15 |ig/dL and possibly
     December 2005
                                          6-103
        DRAFT-DO NOT QUOTE OR CITE

-------
 1    lower. In response to these data, agencies such as the U.S. Centers for Disease Control and
 2    Prevention and the World Health Organization have repeatedly lowered the definition of an
 3    elevated blood lead concentration, which now stands at 10 ng/dL (CDC, 1991; WHO, 1995).
 4    At the time when these policies were put in place, there were too few studies of children with
 5    blood lead levels consistently below 10 |ig/dL on which to base an opinion as to effects at lower
 6    levels of exposure. Since the removal of lead from gasoline, the median blood lead
 7    concentration has dropped dramatically in U.S. children, permitting more studies of this nature to
 8    be done  in recent years. Furthermore, the use of meta- and pooled analytic strategies has
 9    permitted investigators to get a clearer picture of effects below 10 |ig/dL.
10          The Rochester Prospective Study (n = 172) by  Canfield et al. (2003a) is illustrative.  This
11    study extended the relationship between blood lead concentrations and deficits in IQ to levels
12    well below 10 |ig/dL. Over half of the children in this study did not have a recorded blood lead
13    concentration above 10 |ig/dL. Nonlinear semiparametric smoothing revealed a covariate-
14    adjusted decline of more than 7 points up to 10 |ig/dL of childhood average blood lead and a
15    further decline of 2 points associated with an increase from 10 to 20 |ig/dL.  In response to the
16    Rochester findings, Bellinger and Needleman (2003) reanalyzed data from the Boston
17    Prospective Study focusing on children whose blood lead levels never exceeded 10 |ig/dL
18    (n = 48). In their analyses,  10 year IQ was inversely related to blood  lead levels at 24 months
19    following adjustment for covariates.  Nonparametric smoothing analyses indicated that the
20    inverse association persisted at blood lead levels below 5 |ig/dL.
21          Perhaps the most compelling evidence for effects below 10 |ig/dL comes  from an
22    international pooled  analysis of seven prospective cohort studies (n = 1,333) by Lanphear et al.
23    (2005) described earlier.  Although exposures in some cohorts were high, by pooling data from
24    these studies a substantial number (n = 244) of children with blood lead levels that never
25    exceeded 10 |ig/dL could be included in the analyses.  For the entire pooled data set, the
26    observed decline of 6.2 points in IQ for an increase in blood lead levels from 1-10  |ig/dL
27    was comparable to the decrements for an increase in lifetime mean blood lead  levels from
28    <1-10 |ig/dL observed in the Rochester Longitudinal Study (Canfield et al., 2003a). The pooled
29    analysis  of Lanphear et al. also demonstrated that deficits in IQ extended to blood lead levels
30    <7.5  |ig/dL.  Therefore, recent evidence is  suggestive  of effects  of lead on neurocognitive
31    deficits at blood lead levels below 10 |ig/dL, and possibly below 7.5 |ig/dL, in children.

      December 2005                           6-104       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.3.2.12  Selection and Validity of Neuropsychological Outcomes in Children
 2          A fair amount of material has been written about methodologies for neurobehavioral
 3    evaluation in studies of environmental chemicals and child development (Bellinger, 2002, 2003;
 4    Dietrich et al., 2005). Much of the discussion has centered on the ability of neurobehavioral tests
 5    to detect damage to the central nervous system as a result of in utero or early postnatal
 6    exposures. In other words, the sensitivity of these tests to toxicity has been in question. The
 7    sensitivity of a neuropsychological or any other diagnostic test is defined as the proportion with
 8    the abnormality that the test classifies as abnormal (true positives).  In the selection of
 9    neurodevelopmental measures in studies of lead or any other toxicant, it is clearly advantageous
10    to include tests that have the best prognostic value.  This is particularly important in the current
11    context, because the neurobehavioral endpoints reviewed in this document are being
12    incorporated into an assessment of risk (Bellinger, 2002). In addition, it is important to select
13    instruments that tap into neurodevelopmental domains that have shown to be sensitive to
14    particular environmental toxicants.  As evident in this review,  a large number of
15    neuropsychological instruments, tapping a wide range of domains have proven to be sensitive to
16    lower level lead exposure.  Certain domains such as attention,  executive functions, visual-spatial
17    skills, fine-motor abilities, academic achievement (reading  in particular), and externalizing
18    behaviors appear to be affected by lead with some degree of consistency.  However, the
19    identification of behavioral phenotypes for lead has been a largely elusive goal. There are a
20    number of plausible reasons for this. The sample's SES; level, pattern and timing of exposures;
21    nutritional intake; general health; educational opportunities; and the particular instruments that
22    were employed in a given study probably play an important role in between-study differences
23    (Bellinger, 1995; Schantz, 1996). This may be one reason why the broad net provided by global,
24    multiple domain assessments of cognition such as IQ have proven to be the most consistently
25    sensitive across studies of various design and sample characteristics. These measures combine
26    subscales that are representative of a broad number of underlying cognitive functions; thus, they
27    are likely to pick up exposure-related deficits across cohorts that differ in their functional
28    expressions of toxicity (Dietrich et al., 2005).
29          The validity of neuropsychological tests as indices of neurodevelopment in lead studies
30    also is of concern. In psychometrics, there are various types of validity. But the  validity lead
31    researchers are usually most concerned about is "construct validity." If a measure has construct

      December 2005                           6-105        DRAFT-DO NOT QUOTE OR CITE

-------
 1    validity it measures what it purports to measure.  Most lead researchers utilize assessments with
 2    proven construct validity. This means that the instruments utilized by the investigator have
 3    proven that they possess concurrent and predictive "criterion" validity (i.e., it relates to other
 4    manifestations of the construct the instrument is supposed to be measuring and predicts an
 5    individual's performance in the future in specific abilities).  It also means that the instrument
 6    possesses good "convergent validity." This means that the test returns similar results to other
 7    tests that purport to measure the same or related constructs. Finally, the instrument should
 8    demonstrate "discriminant validity." That is, the instrument is not measuring a construct that it
 9    is not supposed to measure,  it discriminates.
10          Bellinger (2003) states that the general literature attests to robust observations between IQ
11    and important measures of life success, such as grades in school, years of education, job success,
12    social status, and income (Neisser et al., 1996; Salkever, 1995).  Testing is difficult depending on
13    examined age, especially for infants who are in a period of rapid developmental change. Also,
14    the way an infant's cognitive function can be probed is restricted. The lack of continuity
15    between their response modalities and ones that can be exploited as a child gets older is also a
16    factor.  Still neurobehavioral tests scores in infancy do possess strong concurrent validity.
17          There many potential sources of invalidity which researchers take steps to avoid.  These
18    include unreliability (an instrument that, all other things being equal, yields scores that are
19    unrepeatable and inconsistent) and bias (e.g.,  due to factors such as culture, gender). Most
20    modern standardized measures of development and cognitive attainment have taken steps to
21    reduce these sources of invalidity and must meet certain minimum requirements such as those
22    formulated by the American Educational Research Association, American Psychological
23    Association, and the National Council on Measurement in Education (American Educational
24    Research Association et al.,  1999).  One reason that global measures of IQ have been used so
25    widely is because of their outstanding psychometric properties.  The Wechsler series has
26    excellent reliability and validity  (Groth-Marnat, 2003). For example, the average internal
27    consistency for the Wechsler children's scales across all age groups is 0.96.  Test-retest
28    reliability is similarly very high.  The  underlying factor structure of these  scales has also been
29    strongly confirmed.  The validity of so-called experimental measures of learning and cognition is
30    sometimes less certain.
      December 2005                            6-106       DRAFT-DO NOT QUOTE OR CITE

-------
 1          All measurement procedures have the potential for error, so the goal of the researcher is to
 2    minimize it.  In elementary psychometric theory, any observed test score is made of the "true"
 3    score plus measurement error.  It is assumed that measurement errors are essentially random (the
 4    child's true score may not be reflected in the observed score because of errors of administration,
 5    inconsistency of administration across examiners, the child's health, or aspects of the testing
 6    environment that are not conducive to performance). This does not mean that lead researchers
 7    cannot take pains to reduce these sources of error.  In fact, most modern lead researchers do
 8    minimize measurement error through attention to training, establishing inter-examiner reliability,
 9    attention to child factors, site factors, and vigilant monitoring of examiner performance
10    throughout the course of a study (Dietrich et al., 2005).
11
12    6.3.2.13  Confounding, Causal Inference, and Effect Modification of the Neurotoxic
13             Effect of Lead in Children
14          The major challenge to observational studies of lead's impact on parameters of child
15    development has been the assessment and control for confounding factors.  By definition, a
16    confounder is associated with both the exposure and the outcome, thus has the potential to
17    influence the association between the exposure and the outcome.  Confounding by various
18    factors can be controlled for in the design phase of the study or in the analytical phase.  In the
19    realm  of lead research, there are a wide range of potential confounders, the foremost of which is
20    SES.  Socioeconomic status is measured rather crudely in most studies with such indices as the
21    Hollingshead Four-Factor Index of Social Position that incorporates education and income of
22    both parents. However, even these so-called blunt measures often account for a great deal of the
23    variance in neurodevelopmental outcomes.  Given the crude nature of these measures, to control
24    for confounding by SES as well as rearing environment of the child, many recent lead studies
25    have incorporated more direct assessments such as the HOME scale, parental intelligence,
26    parental attitude assessments, and measures of parental substance abuse and psychopathology.
27    Given the relatively high correlation between indices of lead exposure and social environmental
28    factors, the consistency among studies in finding effects following adjustment for these
29    confounding factors is remarkable. It is important to consider the enormous experimental animal
30    evidence not compromised by the possibility of confounding in examining lead effects on health
31    (Bellinger, 2004; Davis et al., 1990; U.S. Environmental Protection Agency, 1986a, 1990).
      December 2005                           6-107       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Another problem in the analyses of data on lead and child development is the lack of
 2    critical consideration of which potential confounder in a particular model "owns" the variance in
 3    neurodevelopmental performance.  Thus, for example, in the case of social class it is assumed
 4    that if an effect of lead is reduced to nonsignificance following adjustment for some measure of
 5    socioeconomic standing, the assumption is that all of the variance belongs to the confounder.
 6    However, in some instances this could be seen as an excessively conservative interpretation and
 7    raises the specter of Type II error.  Social class could be seen as either a confounder or a proxy
 8    for exposure.  Lower social class in urban children is closely linked to residence in older housing
 9    in poor condition that, in turn, is associated with higher levels of environmental lead (Clark et al.,
10    1985).  If studies adjust for social class in the usual  manner, the effects of the toxicant will be
11    underestimated (Bellinger, 2004).  One extreme example of overcontrol of this nature can be
12    found in the New Zealand studies where investigators regularly "controlled" for residence in
13    older "weatherboard" housing (e.g., Fergusson et al., 1988a,b).  However, it is worth noting that
14    even in the models including this variable lead remained a significant predictor of intellectual
15    and academic under-attainment in the Christchurch  Health Study.
16          Most of the important confounding factors in lead studies have been identified and efforts
17    have been made to control them in studies conducted since the 1990 Supplement.  Invocation of
18    the poorly measured confounder as an explanation for positive findings is not substantiated in the
19    database as a whole when evaluating the impact of lead on the health of U.S. children
20    (Needleman, 1995).  Of course, it is often the case that following adjustment for factors such as
21    social class, parental neurocognitive function, and child rearing environment using covariates
22    such as parental education, income, and occupation, parental IQ, and HOME scores, the lead
23    coefficients are substantially reduced in size and statistical significance (Dietrich et al., 1991).
24    This has sometimes led investigators to be quite cautious in interpreting their study as positive
25    (Wasserman et al., 1997). This is a reasonable way of appraising any single study, and such
26    extreme caution would certainly be warranted if forced to rely on a single study to confirm the
27    lead effects hypothesis. Fortunately, a large database of high quality studies on which to  base
28    inferences regarding the relationship between lead exposure and neurodevelopment exists.
29    In addition, lead has been extensively studied in animal models at doses that closely approximate
30    the human situation. Experimental animal studies are not compromised by the possibility of
31    confounding by such factors as social class and correlated environmental factors. The enormous

      December 2005                           6-108        DRAFT-DO NOT QUOTE OR CITE

-------
 1    experimental animal literature that proves that lead at low levels causes neurobehavioral deficits
 2    and provides insights into mechanisms is to be considered when drawing causal inferences
 3    (Bellinger, 2004; Davis et al., 1990; U.S. Environmental Protection Agency, 1986a, 1990).
 4          In addition to being a confounder, social class and related variables have been shown to
 5    be effect modifiers in many studies of lead and child development (Bellinger, 2000; Tong et al.,
 6    2000). Effect modification occurs when the magnitude of an association between an exposure
 7    (lead) and an outcome (neurobehavior) varies across strata of some other factor (Last, 2001).
 8    The disadvantages that accompany poor education and underemployment have been found to
 9    exacerbate the effects of lead when carefully examined (Bellinger et al., 1989).  Indeed,
10    evaluating potential effect modifiers should be considered an important part of an overall data
11    analytic plan.
12
13    6.3.3  Summary of the Epidemiologic Evidence for the Neurotoxic Effects
14           of Lead in Children
15          Effects of lead on neurobehavior have been detected with remarkable consistency across
16    numerous studies of various designs, populations studied, and developmental  assessment
17    protocols.  The negative impact of lead on IQ and other neurobehavioral outcomes persist in
18    most  recent studies following adjustment for numerous confounding factors including social
19    class, quality of caregiving, and parental intelligence.
20          Three meta-analyses and one international pooled analysis of seven prospective studies
21    have  confirmed that exposure to lead at low dose has an effect on the intellectual attainment of
22    preschool and school age children. Recent analyses examining the association of lead with
23    intellectual attainment and academic performance in children with low lead exposures have
24    observed effects at blood lead concentrations below 10 |ig/dL.  The pooled analysis by Lanphear
25    et al.  (2005) observed a decline of 6.2 points (95% CI:  3.8, 8.6) in full scale IQ for an increase in
26    concurrent blood lead levels from 1 to 10 |ig/dL.
27          The effects of lead on behavior and mood of children has been an area of recent research.
28    These studies have demonstrated that the impact of lead may extend into increased risk for
29    antisocial and delinquent behavior. This may be a consequence of attentional problems and
30    academic underachievement among children who have suffered higher exposures to lead during
     December 2005                           6-109       DRAFT-DO NOT QUOTE OR CITE

-------
 1   their formative years.  Several studies that have used methods of MRI and MRS to assess direct
 2   measures of brain damage also are suggesting evidence of harm due to lead exposure.
 3          Attempts to eliminate or limit lead-associated neurodevelopmental morbidities with
 4   pharmacological or nutritional intervention strategies have been shown to be ineffective, further
 5   emphasizing the importance of taking environmental measures to reduce and possibly prevent
 6   exposure to lead in children.
 7
 8   6.3.4    Summary of Key Findings on the Neurotoxic Effects of Lead in Adults
 9            from the 1986 Lead AQCD
10          Lead intoxication in adults occurred primarily in occupational settings with historically
11   high exposure levels. In more recent times, occupational lead exposure has been reduced to
12   much lower levels and is often associated with no symptoms.  The symptom constellation
13   associated with high levels of lead exposure include impaired memory and attention span,
14   irritability, headache, muscular tremors, and hallucinations (Cantarow and Trumper, 1944) that
15   may progress to signs of frank encephalopathy (Smith et al., 1938). Symptoms of lead
16   intoxication begin with blood lead >40  |ig/dL (Baker et al., 1979) accompanied by poorer
17   performance on cognitive and visuomotor tasks, reaction time, verbal learning, and reasoning
18   ability that reflect involvement of both the central nervous system and the peripheral nervous
19   system (Arnvig et al., 1980; Campara et al., 1984; Grandjean et al., 1978; Haenninen et al., 1978,
20   1979; Hogstedt et al., 1983; Mantere et al., 1982; Valciukas et al., 1978; Zimmermann-Tansella
21   et al., 1983).  Impaired occulomotor function, measured by saccade accuracy and velocity,
22   depended upon the age group of the lead-exposed worker (Baloh et al., 1979; Glickman et al.,
23   1984; Spiveyetal., 1980).
24          With regard to peripheral nerve function as measured by nerve conduction studies, the
25   28 studies reviewed by the U.S. EPA in the 1986 Lead AQCD found no consistent single nerve
26   involved but, overall, the exposed group had slower conduction velocity at blood lead
27   concentrations as low as 30 |ig/dL.
28          Studies reviewed in 1986 found that amyotrophic lateral sclerosis (ALS) was
29   inconsistently associated with elevated  lead levels in the nervous system. Chelation for 1 year
30   did not did not alter elevated lead levels in the tissue of patients with motor neuron disease.
31

     December 2005                           6-110       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.3.5   Neurotoxic Effects of Lead in Adults
 2    6.3.5.1   Overview of Cognitive and Psychomotor Tests Associated with Adult
 3             Lead Exposure
 4          Examination of lead effects on neurobehavioral performance in adults differs from that in
 5    children, since the neurobehavioral tests in adults focus on loss of abilities previously present
 6    rather than the lack of attainment of those abilities.  Also, there is contribution of cognitive
 7    reserve acquired by years of education, self-education, on-the-job training, avocational, and
 8    non-avocational activities that increases the ability to compensate for the effects of lead exposure
 9    on learning new information. Medical conditions requiring medications, head trauma, and other
10    neuropsychiatric conditions that impact nervous system performance have increased in
11    prevalence in the adult population. These factors may increase the impact of lead exposure or be
12    mistaken for the effects of lead and, therefore, must be handled in the analysis.
13          As alterations in mood may be associated with lead exposure, many neurobehavioral
14    batteries use self-administered questionnaires to screen for mood. The Center for Epidemiologic
15    Studies Depression Scale (CES-D) screens for depression.  The Profile of Mood State (POMS)
16    screens for six subscales, namely anger, confusion, depression, fatigue, anxiety/tension, and
17    vigor. The six mood scales of the POMS were originally validated in a clinical psychiatric
18    population; thus, the factor structure needed to be validated in an occupational population.
19    Factor analysis of the POMS in lead smelter workers found only two relevant factors:
20    (1) "general distress," composed of the subscales anger, confusion, depression, fatigue, and
21    tension; and (2) "psychological adjustment," which contained vigor (Lindgren et al, 1999).
22    This brings into question the use of the six scales as separate outcome variables in the study of
23    lead exposure.
24          Neurobehavioral tests commonly used to demonstrate the effects of lead are listed below
25    (for a more complete description,  see Lezak, 1995). Mini-Mental-State Examination (MMSE),
26    a screening tool for cognitive impairment, is a compilation of many cognitive domains including
27    orientation to time and place, registration, and recall of three words, attention, language, and
28    visual construction with a total possible score of 30 (Folstein et al,  1975). MMSE is sensitive to
29    age and education. In 194 healthy subjects aged 40 to 89 years with 7-21 years of education,
30    only 1% of the subjects obtained an MMSE score of 24/30 and none below (Bleecker et al.,
31    1988).  MMSE errors are sensitive to age effects including delayed recall, spelling "WORLD"
      December 2005                           6-111       DRAFT-DO NOT QUOTE OR CITE

-------
 1    backwards and repetition of "no ifs, ands, or buts." With lead exposure, examination of errors is
 2    important to compare with age-related changes and to determine the biological plausibility of the
 3    effects of exposure especially when performing repeated measures of the test.
 4         Neurobehavioral batteries should always include a benchmark test such as Vocabulary or
 5    a reading test such as the Wide Range Achievement Testing for Reading (WRAT) or the North
 6    American Reading Test (NART) that are considered to be resistant to neurotoxic exposure.
 7    Results from these tests should be adjusted for in the analysis. In blue-collar workers, this may
 8    be a better measure of educational achievement than years of education (Bleecker et al., 2002).
 9         Neuropsychological batteries screening for the effects of lead usually include the
10    following domains (Lezak,  1995): attention/concentration (Digit Span); conceptual and
11    executive  functioning (Stroop, Trails B); visuoperceptive/visuoconstructive (Block Design);
12    visuomotoric (Reaction Time, Pegboard Test, Digit Symbol Substitution, Trails A); verbal
13    memory (Rey Auditory Verbal Learning Test, Logical Memory, Paired Associated Learning);
14    and nonverbal memory (Rey-Osterreith Complex Figure, Benton Visual Retention).  When
15    analyzing the association of lead exposure and test performance, adjusting for potential
16    confounders is critical. Potential confounders are namely age, education (preferably a measure
17    of verbal intelligence), depressive symptoms, alcohol use, and smoking. In some cases, age
18    (Bleecker et al., 1997a) and education (Bleecker et al., 2002) may serve as effect modifiers.
19    The association of lead and poorer neurobehavioral outcome has been found to be present only in
20    older workers or those with less education.
21
22    6.3.5.2    Neurobehavioral Effects Associated with Environmental Lead Exposure
23         Exposure to chronic low levels of environmental lead and its association with effects on
24    the nervous system were  examined in several populations originally followed to study conditions
25    associated with aging: the VA Normative Aging Study (NAS) (Payton et al., 1998; Weiskopf
26    et al., 2004; Wright et al., 2003); the Study of Osteoporotic Fractures (Muldoon et al., 1996);  and
27    the Kungsholmen Project on aging and dementia (Nordberg et al., 2000).  Studies reviewed in
28    this section are summarized in Annex Table AX6-3.11.
29         The VA Normative Aging Study (NAS) is a multidisciplinary longitudinal investigation
30    of the aging process established in 1963 and conducted at the VA Outpatient Clinic in Boston,
31    MA.  The NAS cohort cannot be considered to be exclusively representing the general

      December 2005                           6-112        DRAFT-DO NOT QUOTE OR CITE

-------
 1    population, as bone lead measurements are higher than expected for only environmental
 2    exposure and, thus, suggest the possibility of other sources such as past occupational exposure,
 3    diet, and drinking water (Elmarsafawy et al., 2002, Vijayalakshmi et al., 1999).
 4          The relationship of bone lead and blood lead to psychiatric symptoms in NAS (Rhodes
 5    et al., 2003) found mood symptoms for anxiety and depression potentially associated with bone
 6    lead levels. Education also was inversely related to bone lead; however, high school graduates
 7    had significantly higher general stress that may be related to SES and not lead exposure.
 8          Neuropsychological testing in NAS found response speed sensitive to low levels of lead
 9    but it was not a consistent finding in all tests measuring the same domain upon examination of
10    141 healthy men with a mean age of 67 years, education 14 years.  The mean blood lead level
11    was 6 |ig/dL, patella bone lead was 32 jig/g bone mineral, and tibia bone lead was 23 jig/g bone
12    mineral (Payton et al., 1998). Vocabulary,  a measure of verbal intelligence and predictor of
13    neurobehavioral performance, was used as an outcome variable instead of being adjusted for as a
14    potential confounder.  Education was negatively correlated with bone lead and blood lead,
15    suggesting other factors bedsides lead exposure may have contributed to neuropsychological
16    performance.  The handling of multiple comparisons was not addressed.
17          Another analysis of the NAS (Wright et al.,  2003) examined 736 men, mean age 68 years
18    with education level of 54% high school or less.  The mean blood lead was 5 |ig/dL, and mean
19    patellar and tibia lead levels were 30 and 22 jig/g bone mineral, respectively.  The subjects had a
20    mean MMSE score of 27. Relation of MMSE scores <24 (n = 41) and blood lead by logistic
21    regression estimated an odds ratio of 1.21 (95% CI: 1.07, 1.36). For patella lead and tibia lead,
22    odds ratios of 1.21 (95% CI:  1.00, 1.03) and 1.02 (95% CI:  1.00, 1.04), respectively, were
23    observed. Risk of MMSE <24 (6% of the present population versus 1% of previously described
24    healthy aging study) when comparing the lowest and highest quartiles was 2.1 (95% CI:  1.1,
25    4.1) for patella lead, 2.2 (95% CI: 1.1, 3.8) for tibia lead, and 3.4 (95% CI:  1.6, 7.2) for blood
26    lead.  Interaction of age with patella lead and blood lead in predicting MMSE found steeper
27    decreases in MMSE scores relative to age in the higher quartiles of patella lead and blood lead.
28    Types of errors on the MMSE were not included.  It was not addressed how medical conditions
29    and medications that occurred over the duration of the study and could potentially affect
30    cognitive performance were handled. If the community dwelling population in NAS
31    (Wright et al., 2003) had older individuals with chronic medical conditions and less education

      December 2005                          6-113       DRAFT-DO NOT QUOTE OR CITE

-------
 1    (213 subjects had an education less than high school) living in areas with higher past lead
 2    pollution, the confounding may be impossible to sort out.
 3          Weisskopf et al. (2004) expanded the MMSE study in NAS by examining 466 men, mean
 4    age 70 years who had completed the MMSE twice with an interval of approximately 3.5 years.
 5    Mean blood lead was 4 |ig/dL, and mean patella and tibia bone lead were 23 and 19 jig/g bone
 6    mineral, respectively.  Baseline mean MMSE score was 27 and mean change in MMSE score
 7    over 3.5 years was 0.3 points. Even though MMSE change was significantly associated with
 8    bone lead, a change in MMSE score by a fraction of a point does not constitute a meaningful
 9    change of cognitive performance. To address the biological plausibility of change in the MMSE
10    over 3.5 years, errors by functional domain need to be identified to rule out the possibility of
11    random change with repeat performance.
12          Muldoon et al. (1996) studied participants in the Study of Osteoporotic Fractures for an
13    association of nonoccupational lead exposure and cognitive function.  The Study of Osteoporotic
14    Fractures began in 1986 and included women over age 65 years living in four different
15    communities - Baltimore, MD; Portland, OR; Minneapolis, MN; and the Monongahela Valley
16    outside of Pittsburgh, PA.  A sample of 325 women from rural sites with a mean age of 71 years
17    (mean blood lead 4.5 |ig/dL) and 205 women from urban sites with a mean age of 69 years
18    (mean blood lead 5.4 |ig/dL) were examined. The urban group was more educated and had
19    higher use of cigarettes and alcohol.  Performance examined by blood lead groups adjusting for
20    age, education, smoking, and alcohol use found no significant differences in the urban group.
21    However, in the rural group, individuals with blood lead >7 |ig/dL had significantly poorer
22    performance when compared to those with blood lead <4 |ig/dL for Trails B, Digit  Symbol, and
23    Reaction Time. Response time across blood lead groups increased for the rural group and
24    decreased or remained the same for the urban group. Mean MMSE for the whole population was
25    25, with poorer performance in the rural group—thus, suggesting an increased prevalence of
26    clinical cognitive disorders of another etiology.  Even though  the neuropsychological battery was
27    simple, 9  participants were unable to perform some of the tests including 3 on the MMSE.
28    Such severe impairments were not found with higher occupational exposures, which raises the
29    question as to whether other factors not measured accounted for these differences attributed to
30    blood lead.
     December 2005                           6-114       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In the Kungsholmen Project on aging and dementia in Stockholm, Sweden, no
 2    relationship was found between blood lead and MMSE (Nordberg et al., 2000). The study
 3    population included 762 participants with a mean age of 88 years. The mean blood lead in this
 4    group was 3.7 |ig/dL and the mean MMSE was 25. In contrast to the other populations
 5    examined, this study cohort was more homogenous, comprised entirely of elderly Swedes.
 6    Their likelihood of prior exposure to elevated lead levels was low.
 7          Overall, these studies of environmental lead exposure in adults are difficult to interpret, as
 8    many competing risk factors for neurobehavioral performance in the elderly were not considered.
 9    Also, bone lead levels were higher than expected from environmental exposure suggesting
10    unrecognized previous occupational exposure. The association of bone lead with
11    neurobehavioral performance was unusual, in as much as it was not  demonstrated in studies of
12    occupational exposure (reviewed below). At this time, these studies do not demonstrate that the
13    aging nervous system is at increased risk for poorer neurobehavioral performance related to
14    environmental lead exposure as reported in children.
15
16    6.3.5.3   Neurological Symptoms Associated with Occupational Lead Exposure
17          Studies reviewed in this section are summarized in Annex Table AX6-3.12.  Several
18    occupational studies  found blood lead levels of 29-43 |ig/dL associated with POMS subscales
19    (Hanninen et al., 1998; Maizlish et al.,  1995; Niu et al., 2000). However, other studies with
20    blood lead levels of 27-38 |ig/dL found no relationship with POMS  (Chia et al., 1997; Lucchini
21    et al., 2000; Osterberg et al., 1997; Stollery et al., 1989). A screen for depression, CES-D, was
22    administered to 803 lead-exposed Korean workers. CES-D was significant associated with tibia
23    lead (mean 37 jig/g bone mineral), but not with blood lead (mean 32 |ig/dL), after adjusting for
24    covariates (Schwartz et al., 2001a).
25          Dimercaptosuccinic acid (DMSA)-chelatable lead reflects the mobilizable fraction of lead
26    in the soft tissue. Korean lead-exposed workers (n = 95) with DMSA-chelatable lead (mean
27    289 jig) above the median of 261 jig were 6.2 times more likely to have tingling or numbness in
28    their extremities, 3.3 times more likely to experience muscle pain, and 3.2 times more likely to
29    feel irritable (Lee et al., 2000).  The workers with higher chelatable  lead were 7.8 times more
30    likely to experience neuromuscular symptoms compared to workers with lower chelatable lead.
31    Blood zinc protoporphyrin predicted weakness of ankle and wrist and fatigue while delta-

      December 2005                            6-115       DRAFT-DO NOT QUOTE OR CITE

-------
 1    aminolevulinic acid (ALAD) in urine (mean 3 mg/L) predicted inability to sleep; however, blood
 2    lead (mean 45 |ig/dL) was not significantly associated with any symptoms.
 3          In some studies, difficulty concentrating, irritability, fatigue, muscle pain, and joint pain
 4    were more likely in workers with a mean blood lead of 43 |ig/dL (Maizlish et al., 1995) and
 5    27 |ig/dL (Lucchini et al., 2000), whereas other studies with mean blood lead >30 |ig/dL found
 6    no association with symptoms (Chia et al., 1997; Osterberg et al., 1997). Lucchini et al. (2000)
 7    provided an estimated threshold of blood lead 12 |ig/dL for significant increase of neurological
 8    symptoms.
 9          In summary, even though one study suggested a threshold for neurological symptoms at a
10    blood lead of 12 |ig/dL (Lucchini et al., 2000), other studies with blood lead >30 |ig/dL found no
11    association with lead-related symptoms. The study by Lee et al. (2000) observed that higher
12    levels of DMSA-chelatable lead was associated with irritability, tingling or numbness in their
13    extremities, muscle pain, and neuromuscular symptoms.
14
15    6.3.5.4    Neurobehavioral Effects Associated with Occupational Lead Exposure
16          Studies reviewed in this section are summarized in Annex Table AX6-3.13.  Discriminate
17    analysis of neurobehavioral performance found the group of tests that best differentiates lead-
18    exposed workers (mean blood lead 49 |ig/dL) from nonexposed workers were Simple Reaction
19    Time (SRT), Digit Symbol (WAIS),  and Trail Making Test (Part A) (Boey et al., 1988).  Using a
20    similar battery with 44 lead-exposed workers, mean blood lead 29 |ig/dL, performance was
21    significantly associated with blood lead for SRT, digit symbol and pursuit aiming (Niu et al.,
22    2000).
23          Seventy workers grouped by blood lead (<20, 21-40, and 41-80 |ig/dL) were examined on
24    three occasions each separated by 4 months. Performance on reaction time was stable except in
25    the high lead group where decision time was slowed more than movement time along with
26    concentration difficulties that remained consistently across testing sessions.  Memory testing did
27    not improve with repetition in the high lead group  (Stollery et al., 1991). Decision gaps as
28    opposed to movement gaps were selectively affected by lead exposure in this population
29    (Stollery,  1996).
30          A review of occupational lead exposure in 1995 (Balbus-Kornfeld et al., 1995) concluded
31    that the association of cumulative lead exposure or body burden of lead and neurobehavioral

      December 2005                           6-116       DRAFT-DO NOT QUOTE OR CITE

-------
 1    performance in adults was inadequately covered in the literature.  Studies have addressed these
 2    deficiencies with the use of a working lifetime integrated blood index and bone lead
 3    concentrations. Even though exposure assessment has improved, there is variability based upon
 4    differences in past exposure versus present exposure, duration of exposure, frequency of
 5    monitoring for blood lead, lead exposure from other occupational sources and nonoccupational
 6    activities. Measurement of bone lead addresses some of these problems but the relationship of
 7    bone lead concentration and lead levels in the brain or peripheral nervous system is inconsistent.
 8          Subsequent studies used measures of cumulative lead exposure, namely lifetime
 9    integrated blood index, weighted average blood lead, and bone lead.  More consistent
10    associations occurred with the lifetime integrated blood index and weighted average blood lead
11    for visuomotor/visuoperceptive tasks of Pegboard, Pursuit Aiming, Digit Symbol, Trails, and
12    Block Design (Bleecker et al., 1997a; Chia et al., 1997; Hanninen et al., 1998; Lindgren et al.,
13    1996; Schwartz et al., 2005) while others found no association with these lead exposure
14    measures (Lucchini et al.,  2000; Osterberg et al.,1997; Schwartz et al., 2001a). Age served as an
15    effect modifier for the association of the lifetime integrated blood index with pegboard (Bleecker
16    etal., 1997a).
17          One difficulty with cumulative lead dose is the inability to separate the effect of past high
18    exposure from a lower proximate exposure. To address this issue, workers with similar past high
19    exposure were grouped by those with proximate exposure above blood lead of 40 |ig/dL and
20    those with proximate exposure below blood lead of 40 |ig/dL and were compared on
21    performance of verbal memory (Lindgren et al., 2003).  Use of regression analyses found pattern
22    group contributed significantly to the explanation  of variance in verbal memory after adjusting
23    for current blood lead and lifetime integrated blood index measures.  The relationship between
24    past high exposure and verbal memory no longer existed in the group that maintained proximate
25    blood lead below 40 |ig/dL, suggesting the possibility of reversibility of the effects of lead in
26    adults.
27          The first study to report the effects of cumulative lead exposure on the nervous system
28    examined 467 Canadian lead smelter workers, with a mean of 18 years of employment (Lindgren
29    et al., 1996). Their mean blood lead level was 28  |ig/dL, the weighted average blood lead level
30    was 40 |ig/dL, and the lifetime integrated blood index was 765 jig-year/dL.  Fourteen
31    neuropsychological variables were  examined by MANCOVA using exposure groups of high,

      December 2005                            6-117       DRAFT-DO NOT QUOTE OR CITE

-------
 1    medium, and low. The analysis was adjusted for age, education, years employed, CES-D, and
 2    alcohol use. Exposure groups categorized using lifetime integrated blood index differed
 3    significantly on digit symbol, logical memory, Purdue dominant hand, and Trails A and B.
 4    No concentration-response relationship between blood lead and neuropsychological performance
 5    was found.  From this smelter population, 256 workers currently employed had a median MMSE
 6    score of 29 (range 19-30). A concentration-response relationship between lifetime integrated
 7    blood index and MMSE was found only in the 78 workers with a WRAT-R reading grade level
 8    less than 6.  The absence of a concentration-response relationship in workers with higher reading
 9    grade levels and the same lifetime integrated blood index dose was attributed to increased
10    cognitive reserve (Bleecker et al., 2002). An in-depth  examination of verbal learning and
11    memory in this same population found no association with blood lead. However, increasing
12    lifetime integrated blood index or weighted average blood lead was associated with poorer
13    storage and retrieval of previously learned material. Alterations in the ability to organize
14    materials in long-term memory interfered with retrieval efficiency (Bleecker et al., 2005a).
15    The one test sensitive to blood lead in this smelter population was simple reaction time that had a
16    curvilinear relationship with slowing beginning at a blood lead of approximately 30 |ig/dL
17    (Bleecker et al., 1997b).
18          Fifty-four  lead battery workers were stratified by those whose blood lead never exceeded
19    50 |ig/dL (n = 26) and those who had higher exposure  in the past (n = 28) to examine the
20    neuropsychological effects of current low level blood lead versus those of higher blood lead in
21    the past (Hanninen et al., 1998).  Partial correlations controlling for age, sex, and education in
22    the low group found block design, digit symbol, digit span, similarities, Santa Ana 1, and
23    memory for design associated with recent measures of exposure and embedded figures with
24    maximum blood lead (mean maximum blood lead 40 jig/dL). Embedded figures, digit symbol,
25    block design, and associative learning were associated with the lifetime integrated blood index
26    (mean 823 |ig-year/dL) and maximum blood lead (mean 69 |ig/dL) in the high blood lead group.
27    There was essentially no association with bone lead in either group. A concentration-response
28    relationship existed for digit symbol, embedded figures, and memory for design. Overall past
29    high exposure with blood lead levels >50 |ig/dL had the greatest effect on tests requiring the
30    encoding of complex visually presented stimuli.  The authors concluded that the effect of lead on
31    brain function was better reflected by history of blood  lead than content of lead in bone.

      December 2005                           6-118       DRAFT-DO NOT  QUOTE OR CITE

-------
 1          However, some studies that included measures of cumulative lead and current lead
 2    exposures found the strongest association with current blood lead.  Schwartz et al. (200la)
 3    examined the associations of blood lead, DMSA-chelatable lead, and tibia lead with
 4    neurobehavioral tests in 803 Korean lead-exposed workers from a variety of industries and
 5    135 controls. In lead-exposed workers, the mean blood lead level was 32 |ig/dL, DMSA-
 6    chelatable lead level was 186 jig, and bone lead levels was 37 jig/g bone mineral, compared to
 7    controls with a mean blood lead level of 5 |ig/dL and bone lead level of 6 jig/g bone mineral.
 8    Compared to controls, lead-exposed workers performed significantly worse on SRT, Digit Span,
 9    Benton Visual Retention, Colored Progressive Matrices, Digit Symbol, and Purdue Pegboard
10    after controlling for age, gender, and education. The association of DMSA-chelatable lead with
11    test performance became nonsignificant after the addition of blood lead in the model.  Bone lead
12    was not associated with neurobehavioral performance. Blood lead was the best predictor
13    for significant decrements in neurobehavioral performance on Trails B, Purdue Pegboard
14    (4 measures) and Pursuit Aiming (2 measures). The effect of a 5 |ig/dL increase in blood lead
15    was equivalent to an increase of 1.05 years in age. Use of LOWESS  functions suggested a
16    threshold at blood lead 18 |ig/dL after which there is a decline of performance in Purdue
17    Pegboard (assembly) and Trails B.
18          From the above cohort of Korean lead workers, 212 consecutively enrolled workers were
19    examined for protein kinase C (PKC) activity and the relations between blood lead and
20    neurobehavioral performance (Hwang et al., 2002).  Blood lead range from 5 to 69 |ig/dL was
21    associated significantly with decrements in Trails B, SRT, and Purdue Pegboard (3 measures).
22    PKC activity, as measured by back-phosphorylation of erythrocyte membrane proteins, was not
23    associated with neurobehavioral test scores.  Addition of the interaction term of blood lead with
24    PKC activity dichotomized at the median found significant effect modification with the
25    association of higher blood lead and poorer neurobehavioral performance occurring only among
26    workers with lower PKC activity that corresponds to higher in vivo PKC activity. The authors
27    suggested that PKC activity might identify a subpopulation at increased risk of neurobehavioral
28    effects of lead.
29          Occupational lead exposure and longitudinal decline in neurobehavioral performance was
30    examined in 576 current and former Korean lead workers  who completed testing at three visits at
31    approximately yearly intervals (Schwartz et al., 2005). At baseline, the mean blood lead was

      December 2005                           6-119       DRAFT-DO NOT QUOTE OR CITE

-------
 1    31 |ig/dL and the mean tibia lead was 38 jig/g bone mineral. Blood lead from baseline correlated
 2    with those from visit 2 and 3 and baseline tibial lead correlated with that measured at visit 2.
 3    Cross-sectional associations of blood lead or short-term change occurred with Trails A and B,
 4    Digit Symbol, Purdue Pegboard (4 measures), and Pursuit Aiming after adjusting for potential
 5    confounders. However, longitudinal blood lead was only associated with poorer performance on
 6    Purdue Pegboard (4 measures). Historical tibial bone lead was associated with digit symbol and
 7    Purdue Pegboard (dominant hand). Magnitude of lead  associations was expressed as the number
 8    of years of increased age at baseline that was equivalent to an increase of lead from the 25th to
 9    75th percentile.  The effect of cross-sectional blood lead at baseline was equivalent to 3.8 years
10    of age, 0.9 years of age for historical tibial lead, and 4.8 years of age for longitudinal blood lead.
11          In  summary, performances on visuomotor and verbal memory tasks are consistently
12    associated with occupational lead exposure. In several  studies, cumulative blood lead index was
13    found to be a strong predictor of neurobehavioral  performance. Lead concentrations in bone
14    were a weaker predictor of lead effects on brain function.
15
16    6.3.5.5   Neurophysiological Function and Occupational Lead Exposure
17          A meta-analysis including 32 nerve conduction studies with occupational lead exposure
18    found blood lead to be a weak  predictor of peripheral nerve impairment (Davis and Svendsgaard,
19    1990). Nerve conduction velocities were reduced in lead-exposed subjects, with the greatest
20    sensitivity observed in the median motor nerve. Decreasing effect sizes were observed with
21    increasing duration of exposure.  Meta-analyses of neurobehavioral effects in adults are
22    presented in Annex Table AX6-3.14.
23          Studies reviewed in this section are summarized in Annex Table AX6-3.15. Nerve
24    conduction studies of workers  in a lead battery factory (Kovala et al.,  1997) found sensory
25    amplitudes of the median and sural nerves had a negative correlation with long-term exposure
26    (lifetime integrated blood index and duration of exposure).  Chia et al. (1996b) also found the
27    strongest  concentration-response relationship between median sensory conduction velocity and
28    lifetime integrated blood index. He et al. (1988) found sensory conduction abnormalities related
29    to blood lead levels.
30          Yokoyama et al. (1998) measured the distribution of conduction velocities in large
31    myelinated fibers of the sensory median nerve twice at  a year interval in 17 gunmetal workers.

      December 2005                           6-120       DRAFT-DO NOT QUOTE OR CITE

-------
 1    In workers with a 1-year change in chelatable-lead (mobilized lead) greater than 440 jig/24 h,
 2    conduction velocities of faster fibers were decreased significantly. Measure of body burden
 3    (readily mobilized lead from soft tissue) was a stronger predictor of peripheral nerve impairment
 4    than blood lead.
 5          A group of studies examined vibration threshold in the extremities (Chuang et al., 2000;
 6    Kovala et al., 1997; Schwartz et al., 2001a, 2005). In 60 workers exposed to lead, Kovala et al.
 7    (1997) found vibration threshold at the ankle related to the lifetime integrated blood index and
 8    duration of exposure while the finger vibration threshold was associated with current blood lead
 9    exposure. Overall historical blood lead measures were more closely associated with peripheral
10    nerve function than bone lead in this population.  By contrast, Schwartz et al.  (2001a) also
11    examined vibration thresholds and bone lead in 803 Korean workers and 135 controls and found
12    that tibia lead (mean 37 jig/g bone mineral) but not blood lead (mean 32 |ig/dL) was significantly
13    associated with poorer vibration threshold in the dominant great toe but not the finger.  In a
14    follow-up study of 576 lead workers who completed three visits at yearly intervals, vibration
15    threshold in the toe was associated with current blood lead (mean 31 |ig/dL), longitudinal blood
16    lead,  and tibia lead (38 |ig/g) after adjusting for covariates (Schwartz et al., 2005). Chuang et al.
17    (2000) reported on vibration perception in the foot in 206 lead battery workers.  There was a
18    significant association with blood lead (mean 28 |ig/dL) and weighted average blood lead (mean
19    32 |ig/dL) with vibration perception in the foot after adjustment for covariates including the use
20    of vibrating hand tools. A hockey stick regression analysis of foot vibration threshold versus
21    mean blood lead concentration for the past 5 years found an inflection point around 30 |ig/dL
22    with a positive linear relation above this point, suggesting a potential threshold.
23          Bleecker et al. (2005b) examined peripheral nerve function in 80 smelter workers using
24    Current Perception Threshold (CPT), a neuro-selective test that measures integrity of the large
25    and small myelinated nerve fibers and unmyelinated nerve fibers. CPT was not associated with
26    blood lead (mean 26 |ig/dL) or bone lead (mean 40 jig/g bone mineral). CPT for large
27    myelinated nerve fibers had a curvilinear relationship with weighted average blood lead (mean
28    42 |ig/dL), with a threshold observed at 28 |ig/dL. Results from further regression analyses
29    suggested that even with cumulative lead exposure, intensity is more important than duration of
30    exposure to lead with regard to the peripheral nervous system.  At the highest criterion blood
31    lead level, both large and small myelinated nerve fibers were impaired.  The presence of

      December 2005                            6-121        DRAFT-DO NOT QUOTE OR CITE

-------
 1    activated motor units, equated to ergonomic stressors by job title, enhanced the effect of lead on
 2    the peripheral nervous system.
 3          In summary, occupational lead exposure studies consistently found peripheral sensory
 4    nerve impairment as opposed to the classic motor neuropathy described historically with high
 5    lead exposure. A possible threshold for this effect on the sensory nerves was observed at a blood
 6    leadof28|ig/dL.
 7
 8    6.3.5.6   Evoked Potentials and Occupational Lead Exposure
 9          Visual evoked potentials (VEPs) and brainstem auditory evoked potentials (BAEPs)
10    measure speed of conduction in the visual and auditory pathway. BAEPs have discrete
11    waveforms with wave I arising from the auditory nerve; its latency reflects peripheral
12    transmission time. Wave III is predominantly generated from the caudal pons and wave V from
13    the inferior colliculus. The use of interpeak latencies removes abnormalities in the auditory
14    nerve latency from changes in brainstem transmission in the auditory pathway. Studies reviewed
15    in this section are summarized in Annex Table AX6-3.16.
16          Abbate et al. (1995) performed VEPs on 300 lead-exposed men (aged 30 to 40 years) in
17    good health with no other neurotoxic exposure.  Range of blood lead levels was 17 to 60 |ig/dL.
18    Individuals were stratified into 4 groups with  mean blood lead levels of 23  |ig/dL (n = 39),
19    30 |ig/dL (n = 113), 47 |ig/dL (n = 89), and 56 |ig/dL (n = 59).  P100 latency measured for VEPs
20    were significantly prolonged across the blood lead groups.  Linear regression found the
21    association of blood lead and PI 00 were significant in each group, but the relationship was not
22    proportional. Prolonged VEP began at a blood lead levels of 17-20 |ig/dL. With  age limited to
23    one decade, contribution from age was not a concern. Even though there was no comparison
24    group, careful screening ruled out other medical and eye conditions, and other potential
25    exposures.
26          BAEPs recorded in 49 lead-exposed workers and age and sex matched controls (Discalzi
27    et al., 1992) had mean blood lead levels of 55 |ig/dL and a mean weighted average blood lead
28    level of 54 |ig/dL. Interpeak latencies, I-V, I-III, and III-V were all prolonged in the
29    lead-exposed workers. In those workers with weighted average blood lead >50 |ig/dL, I-V
30    latency was longer. Discalzi et al. (1993) reported identical results in a subsequent publication
31    of 22 battery storage workers with a mean blood lead of 47 |ig/dL and a mean weighted average

      December 2005                            6-122       DRAFT-DO NOT QUOTE OR CITE

-------
 1    blood lead of 48 |ig/dL. Holstein et al. (1986) examined 20 adults accidentally exposed to lead
 2    through food until 1 year prior to the study.  On the day of examination, the mean blood lead
 3    level was 31 |ig/dL, while the mean weighted average blood lead was 43 |ig/dL. Latencies I, III,
 4    and I-III interpeak intervals were longer in the exposed group with a concentration-response
 5    relationship observed for the weighted average blood lead and I-III interpeak interval.
 6          BAEPs were performed in 359 currently-employed smelter workers with a mean of
 7    17 years of employment. The mean blood lead levels was 28 |ig/dL (SD 8.4), mean working-
 8    lifetime weighted average blood lead was |ig/dL 39 (SD  11.9), and working-lifetime integrated
 9    blood lead index was 719 |ig-year/dL (SD 421.0) (Bleecker et al., 2003). After adjusting for the
10    contribution of age, blood lead and weighted average blood lead were significantly associated
11    with Wave I, while lifetime integrated blood index was significantly associated with Wave III
12    and I-III interpeak interval.  Four groups similar in age were created with increasing
13    abnormalities based upon clinical cut-off scores for wave I latency and  I-V interpeak interval.
14    Blood lead, weighted average blood lead, and lifetime integrated blood index were all
15    significantly higher in the group with prolonged Wave I and I-V interpeak interval compared to
16    the group with normal BAEPs.
17          In summary, one detailed study found blood lead associated with prolonged VEPs with a
18    threshold effect at 17-20 |ig/dL. The four studies examining BAEPs and lead exposure
19    consistently found prolonged interpeak latencies in the brainstem auditory pathway more
20    strongly associated with cumulative or average blood lead exposure.
21
22    6.3.5.7   Postural Stability, Autonomic Testing, and Electroencephalogram (EEG)
23             and Occupational Lead Exposure
24          Postural sway measures balance or steadiness on a force platform.  It is a complex task
25    that requires the integration of visual, vestibular, and  peripheral sensory inputs, as well as motor
26    output. No standard protocol is used across studies.  Studies reviewed in this section are
27    summarized in Annex Table AX6-3.17.
28          Postural sway was evaluated in 49 chemical workers exposed to lead stearate, with a mean
29    blood lead of 18 |ig/dL, a mean weighted average blood lead of 24 |ig/dL and a mean cumulative
30    blood lead of 391 jig-year/dL (Yokoyama et al., 1997).  Twenty-three controls were also
31    examined. After adjustment for covariates, a concentration-response relationship was observed
      December 2005                           6-123        DRAFT-DO NOT QUOTE OR CITE

-------
 1    for blood lead and sway in the anterior-posterior direction and for weighted average blood lead
 2    with right to left sway. The authors concluded that change in the vestibulocerebellum was
 3    affected by blood lead while the anterior cerebellar lobe was affected by average lead exposure.
 4          Chia et al. (1994) measured postural sway parameters in 60 lead storage battery workers
 5    (mean blood lead 36 |ig/dL) and 60 controls (mean blood lead 6 jig/dL).  Computerized postural
 6    sway measurements showed that lead workers had poorer postural stability that increased with
 7    eyes closed but no concentration-response association was observed with blood lead. A second
 8    publication examined cumulative blood lead over 10 years and found that lifetime integrated
 9    blood index for the 2 years  prior to testing was associated with all postural sway parameters with
10    eyes closed (Chia et al., 1996c).
11          Postural control measured in 63 lead battery workers (mean past blood lead 38 |ig/dL)
12    indicated significantly increased mean body oscillations with eyes closed and head tilted forward
13    (Ratzon et al., 2000).  Total lead exposure was significantly associated with increased body
14    oscillations with head tilted forward after adjusting for education, coffee consumption, hours of
15    sleep, and estimate of health.  In order to maintain balance, lead-exposed workers required
16    increased oscillations when visual and vestibular inputs were altered.
17          The effects of lead on the cardiac autonomic nervous system, expressed as the decrease of
18    R-R interval variation on an electrocardiogram, was examined in 172 male lead-exposed workers
19    (mean blood lead 36 |ig/dL) (Teruya et al., 1991).  A significant blood lead concentration-related
20    decrease of R-R interval variation during deep breathing was present in 132 workers with stable
21    blood lead over the past year. An approximate threshold effect was found at blood lead
22    >20 |ig/dL. Similar findings were reported by Niu et al. (2000) in 44 lead exposed workers who
23    had a mean blood lead of 29 |ig/dL.
24          One hundred twenty-eight workers in the ceramic painting industry (mean blood lead
25    13 |ig/dL) were monitored for measures of sympathetic nerve function by variations in R-R
26    interval on electrocardiography and changes in finger blood flow with postural changes using
27    Doppler flowmetry (Ishida  et al., 1996).  No significant association was found between blood
28    lead levels and the results of the neurophysiological tests, except for change in finger blood flow.
29    Increased blood lead was associated with decreased changes in finger blood flow.
30          Sixty workers in a lead battery factory examined with quantitative electroencephalographs
31    found alpha and beta frequencies were more abundant in workers with higher long term lead

      December 2005                           6-124       DRAFT-DO NOT QUOTE OR CITE

-------
 1    exposure (Kovala et al., 1997). Biomarkers of long-term lead exposure included tibia bone lead
 2    (mean 26 |ig/g), calcaneal bone lead (mean 88 |ig/g), lifetime integrated blood index (mean
 3    546 jig-year/dL), and weighted average blood lead (mean 32 |ig/dL).  The finding of slow alpha
 4    activity positively correlated with lead exposure may reflect increased episodes of
 5    "microdrowsiness" in workers with higher lead exposure.  In the study by Niu et al. (2000),
 6    quantitative electroencephalographs in 44 lead-exposed workers (mean blood lead 29 |ig/dL)
 7    indicated significantly increased beta activity and diminished amplitudes abnormalities in 81%
 8    of exposed workers.
 9          In summary, postural sway is associated with lead exposure at blood lead levels
10    <40 |ig/dL, and is believed to  be caused by the effect of lead on the cerebellum. A standard
11    protocol was not employed across the studies. Parasympathetic and sympathetic integrity is
12    compromised in lead-exposed workers beginning at blood lead >20 |ig/dL. Quantitative
13    electroencephalographs found increased beta activity associated with lead exposure.
14
15    6.3.5.8   Other Neurological Outcomes Associated with Lead in Adults
16          Studies reviewed in this section are summarized in Annex Table AX6-3.18. The 1986
17    Lead AQCD concluded that the evidence for an association of lead and  ALS or motor neuron
18    disease was inconsistent.  The subsequent publications remain mixed but more studies are
19    reporting an association. Using 109 cases of ALS and 256 controls matched for age, gender, and
20    region of residence, Kamel et  al. (2002) examined the relation of lead and ALS using blood lead
21    and bone lead levels. Ranges  of exposure were <1 to  14 |ig/dL  for blood lead, -4 to 107 jig/g
22    for patella lead, and -7 to 61 jig/g for tibia lead. History of occupational lead exposure
23    increased the risk of ALS (adjusted odds ratio of 1.9 [95% CI:   1.1,3.3]).  Elevations in both
24    blood lead and patella and tibia bone lead were found in ALS cases, though the precision of these
25    measurements was questioned. In summary, this study found lead exposure from historical
26    questionnaire data and biological markers to be associated with  ALS. The same data was used to
27    determine the associations of ALS with polymorphism in ALAD and VDR and the influence of
28    genotype in the previously discussed associations of ALS with lead (Kamel et al., 2003). The
29    ALAD2 allele was associated  with a 2-fold increased risk of ALS after adjustment for age,
30    gender, region, education, and physical activity. Additionally adjusting for blood lead
     December 2005                           6-125        DRAFT-DO NOT QUOTE OR CITE

-------
 1    strengthened the association of ALAD2 and ALS risk. This was not found for bone lead or
 2    occupational history of lead exposure.  VDR was not associated with lead or ALS risk.
 3          A study from the Mayo Clinic examined risk factors for sporadic ALS in 45 male ALS
 4    patient-patient control pairs (Armon et al., 1991).  When lifetime exposure to lead exceeded
 5    200 hours, the relative risk for ALS was 5.5 (95% CI:  1.44, 21.0).  Overall, men with ALS had
 6    worked more at blue-collar jobs with significantly more time welding or soldering than controls
 7    (p < 0.01). The association between lead exposure and development of ALS was supported as
 8    these authors had the same findings in a previous pilot study of another patient population
 9    (Roelofs-Iverson et al., 1984).
10          Another study of risk factors for ALS in 103 patients found increased odds ratio for
11    manual occupation (2.6 [95% CI:  1.1,  6.3]) and occupational exposure to lead (5.7 [95% CI:
12    1.6, 30]) (Chancellor et al.,  1993). A Swedish study of 92 cases of motor neuron disease
13    (includes ALS, progressive bulbar palsy,  and progressive muscular atrophy) found a Mantel -
14    Haenszel odds ratio for welding equal to 3.7 (95% CI: 1.1, 13.0) (Gunnarsson et al., 1992).
15          Guidetti et al. (1996) performed a retrospective incidence, prevalence, and mortality
16    survey in northern Italy. The area studied had documented lead pollution for years. Based upon
17    79 cases, incidence  and prevalence rates of ALS were comparable to the surrounding area.
18    A subsequent publication by this group found that mean blood lead levels in cases of sporadic
19    ALS and controls were not significantly different (mean blood lead of 13 |ig/dL versus
20    11  |ig/dL) (Vinceti et al.,  1997). Blood lead was associated with disability due to ALS but
21    no support was found for involvement of lead in the etiology of sporadic ALS.
22          Louis et al. (2003) examined the relationship between blood lead and essential tremor
23    (ET) in 100 cases with ET (mean blood lead 3 |ig/dL) and 143 controls (mean blood lead
24    2 |ig/dL). Ten cases and 7 controls had bone lead levels measured that were significantly
25    correlated with blood lead suggesting that higher blood lead may have occurred in the past.
26    Logistic regression adjusting for age and  current cigarette smoking found an association between
27    blood lead and ET.  An odds ratio of 1.19 (95% CI:  1.03, 1.37) was estimated. Blood lead was
28    higher in the 39 ET cases with no family history. Both current and lifetime prevalence of
29    occupational lead exposure was the same in ET cases and controls. In a second publication
30    (Louis et al., 2005), 63 ET cases (mean blood lead 4 |ig/dL) and 101 controls (mean blood lead
31    3 |ig/dL) who were  similar in age, education, gender, and ethnicity were examined for interaction

      December 2005                            6-126       DRAFT-DO NOT QUOTE OR CITE

-------
 1    of blood lead and ALAD gene polymorphisms and increased odds of ET. Of the 63 ET cases,
 2    18 (29%) had an ALAD2 allele compared to 17 (17%) of the 101 controls (odds ratio of
 3    1.98 [95% CI: 0.93, 4.21]). When log blood lead was examined by presence of ALAD2 allele in
 4    ET, log blood lead was highest in ET cases with the ALAD2 allele, intermediate in ET cases
 5    without an ALAD2 allele, and lowest in controls (test for trend, P = 0.10; p = 0.001). When the
 6    ALAD2 allele was present, blood lead was significantly associated with odds of ET (80.29
 7    [95% CI: 3.08,2,096.36]).  This increased odds of ET with an ALAD2 allele was 30 times
 8    greater than in individuals with only an ALAD1  alleles.  In the highest log blood lead tertile,
 9    ALAD2 allele was present in 22% of ET cases and 5% of controls. It was proposed that
10    increased blood lead along with the ALAD2 allele could affect the cerebellum and, thereby,
11    increase the risk of tremor.
12          Graves et al. (1991) performed a meta-analysis on 11 case-control studies of Alzheimer's
13    disease for occupational exposure to solvents and lead. Four studies had data for lead exposure
14    with a pooled analysis of relative risks for occupational lead of 0.71 (95% CI: 0.36,1.41).  The
15    exposure frequencies were 16 of 261 (6%) for the cases  and 28 of 337 (8%) for the controls.
16    These nonsignificant results were further confirmed by measuring lead concentration in the brain
17    of cases with diffuse neurofibrillary tangles with calcification (DNTC), Alzheimer's disease, and
18    non-demented controls. The lead concentration was significantly higher in DNTC compared to
19    Alzheimer's disease and non-demented controls  (Haraguchi et al., 2001).
20          In summary, more studies are reporting an association with past exposure to lead, usually
21    in the occupational setting, and the motor neuron disease ALS. There appears to be a 2-fold
22    increased risk for ALS when the ALAD2 allele is present.  The odds of ET in individuals with
23    the ALAD2 allele were 30 times greater compared to those with only ALAD1 alleles.
24
25    6.3.5.9  Occupational Exposure to Organolead and Inorganic Lead
26          Compared to inorganic lead, organolead exposure has a greater impact on the brain and,
27    therefore, is discussed separately.  Direct comparison of trimethyl lead (a metabolite of
28    organolead), tetraethyl lead, and inorganic lead on the in vitro assembly of microtubules from the
29    mammalian brain found no effects with inorganic lead but trimethyl lead produced dramatic
30    impairment  of neurotubular structures and functions (Roderer and Doenges, 1983). Another
31    study examining organic and inorganic lead found differential effects on neurite growth in

      December 2005                           6-127       DRAFT-DO NOT QUOTE OR CITE

-------
 1    neurons in culture, suggesting that the mechanism of action for organic and inorganic lead was
 2    not the same (Audesirk et al., 1989). Studies reviewed in this section are summarized in Annex
 3    Table AX6-3.19.
 4          Two hundred and twenty-two current employees that manufactured tetraethyl lead had
 5    cumulative lead exposure associated with poorer performance in many cognitive domains but
 6    most often in manual dexterity and verbal memory/learning (Schwartz et al., 1993). Simple
 7    visual reaction time and blood lead had a curvilinear relation with an increase in simple visual
 8    reaction time occurring above a blood lead of 30 |ig/dL (Balbus et al., 1997, 1998).
 9          In former organolead workers (n = 543), peak tibial lead was a stronger predictor of
10    poorer cognitive function than current tibial lead (Stewart et al., 1999).  Examination of the
11    peripheral nervous system in this population found no strong association between lead
12    biomarkers and measures of sensory and motor function (Tassler et al., 2001).  Five hundred and
13    thirty-five of these former organolead workers were re-examined over a 4-year period (Schwartz
14    et al., 2000d, 2001b).  Peak tibia lead predicted decline in tests of verbal memory and learning,
15    visual memory, executive ability, and manual dexterity. Effect size for an increase of 15.7 |ig/g
16    bone mineral of peak tibia lead was equivalent to 5 more years of age at baseline. This
17    relationship of neurobehavioral tests with bone lead levels was influenced by the apolipoprotein
18    E (ApoE) genotype (Stewart et al., 2002). The slope of the relation between tibia lead and
19    neurobehavioral outcome was more negative in those individuals with at least one s4 allele than
20    individuals without this allele.  It is suggested that the presence of one Apo-e-4 allele increases
21    the risk of persistent central nervous system effects of lead.
22          Overall when these neurobehavioral outcomes related to organolead exposure are
23    compared to the literature reviewed with inorganic lead exposure, the absence of effects on the
24    peripheral nerves  and the global nature of central nervous system impairment suggests the
25    impact on the brain is greater with organolead exposure.
26
27    6.3.6  Summary of the Epidemiologic Evidence for the Neurotoxic Effects
28           of Lead in Adults
29          There is no consistent evidence that environmental lead exposure is associated with
30    impaired cognitive performance in the elderly if competing risk factors  are considered.  In adults,
31    the effect of lead on the nervous system may not be detected through neurobehavioral testing due

      December 2005                           6-128       DRAFT-DO NOT QUOTE OR CITE

-------
 1    to cognitive reserve, the ability to compensate for brain impairment. Cognitive reserve is related
 2    to pre-morbid cognitive abilities, education, and occupational attainment, and is able to modify
 3    the clinical expression of central nervous system insult from lead exposure.  Therefore, when
 4    chronic lead exposure is the same in two groups of individuals that differ by educational
 5    achievement levels, the concentration-response relationship will only be seen in the group with
 6    low educational achievement, as cognitive reserve allows the high educational achievement
 7    group to compensate for the central nervous system expression of the effects due to lead.
 8          Chronic occupational lead exposure affects the sensory  nerve fibers in the extremities
 9    with a possible threshold at a weighted average blood lead level of 28 |ig/dL. Intensity of lead
10    exposure appears to be more critical than duration of exposure for this outcome. Slowing in the
11    brainstem auditory pathway in the caudal pons was consistently associated with chronic
12    occupational lead exposure.
13          Past occupational exposure to lead increased the risk of developing ALS and motor
14    neuron disease  in 4 studies. This risk was increased 2-fold by the presence of the ALAD2 allele.
15    Essential tremor in two well-done  studies was associated with low blood lead levels (mean
16    3 |ig/dL). The  odds of developing ET with the ALAD2 allele increased 30-fold compared to
17    those individuals with only an ALAD1 allele.
18          Numerous studies of occupational lead exposure also found chronic and current blood
19    lead associated with visuomotor and memory impairment with a threshold effect at blood lead
20    18 |ig/dL. As with ET, postural sway abnormalities associated with blood lead <40 |ig/dL is
21    believed to result from the effects of lead on different parts of the cerebellum.
22
23
24    6.4    RENAL EFFECTS OF LEAD
25    6.4.1   Summary of Key Findings on the Renal Effects  of Lead from the
26            1986  Lead AQCD
27          Chronic lead nephropathy is a disease characterized by tubulointerstitial nephritis, which
28    can ultimately result in  small,  fibrotic kidneys. It occurs in individuals who sustain chronic high-
29    level lead exposure.  In these individuals, lead exposure is the primary cause of renal failure.
30    The pathophysiologic characteristics of lead nephropathy and the populations at increased risk
31    for this diagnosis were the foci of the human research portion of Section 12.5, entitled "Effects

      December 2005                          6-129       DRAFT-DO NOT QUOTE OR CITE

-------
 1    of Lead on the Kidney," in the 1986 Lead AQCD. The 1986 document clearly identified several
 2    high-risk groups for this diagnosis, including children in the Queensland, Australia lead
 3    poisoning epidemic, moonshine alcohol drinkers, and lead workers in poorly controlled settings.
 4    The section concluded that data in the latter group indicated an increased risk for lead
 5    nephropathy associated with blood lead levels ranging from 40 to >100 |ig/dL, with adverse
 6    renal effects possibly  occurring at levels as low as 30 |ig/dL.
 7          The 1986 Lead AQCD noted that research at that time was not sufficient to address some
 8    of the most critical questions relating to the impact of lead exposure on the kidney.  The last
 9    paragraph of the renal section begins with "Among the questions remaining to be answered more
10    definitively about the effects of lead on the kidneys is the lowest blood lead level at which renal
11    effects occurs."  The last sentence reads "Conversely, the most difficult question of all may well
12    be to determine the contribution of low levels of lead exposure to renal disease of non-lead
13    etiologies." Advances in the research conducted since that document was written allow a much
14    more informed discussion of exactly those critical issues. As discussed below, recent research
15    indicates that lead nephropathy is merely the tip of the iceberg in terms of the contribution that
16    lead makes to renal dysfunction overall.  Research increasingly indicates that lead, at much lower
17    doses than those causing lead nephropathy, acts as a  cofactor with other more established renal
18    risks to increase the risk for renal dysfunction and the rate of subsequent decline. The
19    populations at risk for renal dysfunction  (diabetics and hypertensives) are increasing worldwide,
20    particularly in countries where obesity is epidemic.  Lead exposure is declining in many
21    industrialized countries, although less so among high-risk minority populations.  The extent of
22    the public health impact of lead on the kidney depends on the balance of these two factors.
23
24    6.4.2   Renal Outcome Definitions
25          The renal literature can be confusing since several of the clinical renal measures are
26    inversely related. Therefore, the pertinent outcomes are briefly reviewed below.  The glomerular
27    filtration rate (GFR) is considered to be the best measure of renal function.  GFR is assessed by
28    urinary clearance of exogenous (e.g., 125I-iothalamate) or endogenous (e.g., blood urea nitrogen
29    [BUN] and serum creatinine) compounds.  Creatinine is used most commonly.  Therefore,
30    increases in BUN or serum creatinine or decreases in renal  clearance of creatinine or other
31    markers are all consistent with decreased renal function. Serum creatinine and its reciprocal

      December 2005                            6-130        DRAFT-DO NOT QUOTE OR CITE

-------
 1    have been the most frequently used measures of renal function in the lead-kidney literature.
 2    However, creatinine is not an ideal GFR marker, because it is influenced by factors such as
 3    muscle mass, diet, gender, age, and tubular secretion. Measurement or calculation of creatinine
 4    clearance takes some of these variables into account.  Measured creatinine clearance utilizes
 5    timed urine collections, traditionally over a 24-h period, making compliance difficult.  Therefore,
 6    equations to estimate creatinine clearance have gained popularity.  The Cockcroft-Gault equation
 7    (Cockcroft and Gault, 1976) has been used most commonly. Recently, several equations to
 8    estimate actual GFR were studied in the Modification of Diet in Renal Disease (MDRD) Study
 9    (Levey et al., 1999). The abbreviated MDRD equation (GFR in mL/min/1.73m2 = 186 x
10    creatinine L154 x age °'203 x (0.742 if female) x (1.212 if African American); Stevens and Levey
11    [2005a]) estimates GFR more  accurately than the Cockcroft-Gault equation in patients with renal
12    insufficiency (Levey et al., 2003).  Despite their promise, however, the MDRD equations are
13    relatively new and their use in the literature on the renal effects of lead exposure has been limited
14    to date.
15           Cystatin C is another recent addition  to the tools used to assess GFR (Stevens and Levey,
16    2005b). This is a 13,000 Dalton, non-glycosylated basic protein, which is generated by all
17    nucleated cells and filtered, reabsorbed, and  catabolized, but not secreted, in the kidney.
18    Very little appears in the urine. The majority of studies done to date indicate that serum cystatin
19    C is a better marker for GFR than serum creatinine (Stevens and Levey, 2005b).
20          Most of the renal outcome measures discussed above were developed for use in the
21    clinical setting. Unfortunately, they are insensitive for early renal damage, as evidenced by the
22    fact that serum creatinine remains normal after kidney donation. Therefore, in the last two
23    decades, the utility of renal early biological effect (EBE) markers as indicators of preclinical
24    renal damage has been of interest.  These can be categorized as markers of function (i.e., low
25    molecular weight proteins that should be reabsorbed in the proximal tubules such as (32-
26    microglobulin and retinol-binding protein [RBP]); biochemical alteration (i.e., urinary
27    eicosanoids such as prostaglandin E2, prostaglandin F2 aipha, 6-keto-prostaglandin FI aipha, and
28    thromboxane B2); and cytotoxicity (e.g., N-acetyl-(3-D-glucosaminidase [NAG]) (Cardenas et al.,
29    1993).  Elevated levels may indicate an increased risk for subsequent renal dysfunction.
30    However, with the exception of microalbuminuria in diabetes and (32-microglobulin in  cadmium
31    exposure, most are research tools only and their prognostic value remains controversial.

      December 2005                           6-131       DRAFT-DO NOT QUOTE OR CITE

-------
 1    European and Asian nephrotoxicant researchers have utilized them more frequently than have
 2    renal researchers in the United States. Prospective studies of most of these markers in
 3    nephrotoxicant-exposed populations are quite limited to date.
 4
 5    6.4.3   Lead Exposure Measure Definitions
 6          Although these definitions are reviewed in detail elsewhere in this Lead AQCD, a brief
 7    discussion is included here due to the number of key studies in this section that measured bone or
 8    chelatable lead dose. Inorganic lead is a cumulative toxicant that is stored in bone. Blood lead is
 9    a relatively short-term measure (half-life of 30 days [Hu et al., 1998]) that reflects exposure from
10    current exogenous sources and the release of lead from internal lead stores.  Bone is a source of
11    lead as well as a repository (Hu et al., 1998).  As such, bone lead measures provide information
12    on the potential for ongoing internal exposure as well as cumulative exposure. Lead in
13    trabecular bone (commonly measured in the patella or calcaneus) is more bioavailable than lead
14    in cortical bone (measured in the mid-tibia) and has a shorter half-life (Gerhardsson, et al., 1993;
15    Hu et al., 1998). An additional lead measure, chelatable lead, is thought to represent a
16    bioavailable pool  of lead from blood, soft tissue, and bone. Either calcium disodium
17    ethylenediaminetetraacetic acid (EDTA) or dimercaptosuccinic acid (DMSA; succimer) may be
18    used for this purpose although DMSA is newer and, thus, has been used less frequently to date.
19
20    6.4.4   Lead Nephrotoxicity in Adults
21    6.4.4.1   General Population Studies
22          Over the past two decades, several studies have examined the effect of lead exposure on
23    renal function in environmentally exposed general populations. This is a new category of lead-
24    renal research with no high quality examples (by current standards) having been available for
25    review in the 1986 Lead AQCD.  The studies discussed below provide critical evidence that the
26    adverse effects of lead on the kidney occur at much lower doses than previously appreciated.
27    General  population studies of the renal effects of lead are further summarized in Annex
28    Table AX6-4.1.
29
      December 2005                           6-132        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.4.4.1.1    Cadmibel Study
 2          In the first large environmental study that adjusted for multiple renal risk factors, Staessen
 3    et al. (1992) evaluated 965 men and 1,016 women in the Belgian Cadmibel study. Lead dose
 4    was indexed by blood lead and zinc protoporphyrin.  Renal outcome measures included serum
 5    creatinine and p2-microglobulin and 24-h measured and calculated (Cockcroft and Gault, 1976)
 6    creatinine clearances. Mean blood lead was 11.4 |ig/dL (range 2.3-72.5) and 7.5 |ig/dL (range
 7    1.7-60.3) in men and women, respectively. After adjustment, log transformed blood lead and
 8    zinc protoporphyrin, in separate models, were negatively associated with measured creatinine
 9    clearance (effect estimates are presented in Table 6-4.1). A 10-fold increase in blood lead was
10    associated with a decrease in creatinine clearance of 10 and 13 mL/min in men and women,
11    respectively. Both lead measures were also negatively associated with estimated creatinine
12    clearance.  This landmark study raised concern that the lead dose threshold for adverse renal
13    effects in the general population was much lower than previously appreciated based on
14    occupational data.
15
16    6.4.4.1.2    Normative Aging Study
17          Four studies assessing the renal impact of lead exposure in the Normative Aging Study
18    have been published to date. Participants in this study were originally recruited in the 1960s in
19    the Greater Boston area.  Inclusion criteria included male gender, age between 21 and 80 years,
20    and absence of chronic medical conditions. Payton et al. (1994) analyzed data from a periodic
21    follow-up evaluation performed between 1988 and 1991 in 744 participants.  Lead dose was
22    assessed with blood lead; renal outcome measures included serum creatinine  and 24-h measured
23    and calculated (Cockcroft and Gault, 1976) creatinine clearances. Mean blood lead
24    concentration and measured creatinine clearance were 8.1  |ig/dL (SD 3.9) and 88.2 mL/min
25    (SD 22.0),  respectively. After adjustment, In blood lead was negatively associated with In
26    measured creatinine clearance (effect estimates are presented in Table 6-4.1).  Borderline
27    statistically significant associations (p < 0.1) between blood lead and serum creatinine and
28    estimated creatinine clearance were also observed.
29          Kim et al. (1996) studied 459 men whose blood lead levels from past periodic
30    examinations, conducted every 3-5 years during 1979-1994, were measured from stored samples.
31    Participants were randomly selected to be representative of the entire Normative Aging Study

      December 2005                            6-133       DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
                    Table 6-4.1.  Summary of Key Studies on the Renal Effects of Environmental Lead Exposure
Reference
Study location
Study population
Sample size
Mean exposure and
outcome measures
Analysis methods
Covariates adjusted for in analysis
                                                                                                               Major significant findings
 H
 6
 o

 O
 H
O
 o
 H
 W
 O
 O
 HH
 H
 W
           Muntner et al. (2003)
           NHANES III,
           1988-1994
           n= 15,211
           4,813 hypertensives
           Paytonetal. (1994)
           Boston, MA
           Normative Aging Study,
           1988-1941
           n=744
           Kim etal. (1996)
           Boston, MA
           Normative Aging Study,
           1979-1994
           n = 459
                           Blood lead
                           4.21 ug/dL (hypertensives)
                           3.3 ug/dL
                           (normotensives)

                           Renal outcomes = elevated
                           serum creatinine, chronic
                           kidney disease (GFR
                           <60mL/min/1.73m2)
                           Blood lead
                           8.1 ug/dL

                           Measured creatinine
                           clearance
                           88.2 mL/min

                           Blood lead at baseline
                           9.9 ug/dL

                           Serum creatinine at
                           baseline
                           1.2mg/dL
                           Multiple logistic regression

                           Age, race, gender, diabetes, systolic blood
                           pressure, smoking status, history of
                           cardiovascular disease, body mass index,
                           alcohol consumption, household income,
                           marital status, and health insurance
                           Multiple linear regression

                           Age, body mass index, analgesic and diuretic
                           use, alcohol consumption, smoking status,
                           systolic/ diastolic blood pressure


                           Cross-sectional and longitudinal analyses

                           Random-effects modeling

                           Baseline age, time since initial visit and
                           between visits, body mass index, smoking
                           status, alcohol ingestion, education level,
                           hypertension (defined as blood pressure
                           > 160 or 95 mm Hg or antihypertensive
                           medication use), and baseline serum creatinine
                                             Higher odds ratios of both increased serum creatinine and
                                             chronic kidney disease by quartile of blood lead in
                                             hypertensives but not normotensives

                                             Odds ratios for elevated serum creatinine in hypertensives

                                             Blood lead (range, ug/dL)       %     Odds ratio (95% CI)
                                                                                                               Quartile 1  (0.7-2.4)
                                                                                                      7.2
                                                                                  1.00
                                                                                                               Quartile 2  (2.5-3.5
                                                                                                                                 12.1    1.47(1.03,2.10)
                                                                                                               Quartile 3  (3.9-5.9)
                                                                                                                                 12.4    1.80(1.34,2.42)
                                                                                                               Quartile 4  (6.0-56.0)
                                                                                                                                 16.3    2.41(1.46,3.97)
                                             p < 0.001 for chi-squared test for trend

                                             Twofold higher blood lead associated with odds ratio of 1.43
                                             (95% CI:  1.20,1.71)
                                             Log blood lead negatively associated with log measured
                                             creatinine clearance
                                             -0.04 (95% CI: -0.079,-0.001)

                                             10 ug/dL higher blood lead associated with a 10.4 mL/min
                                             lower creatinine clearance

                                             In cross-sectional analyses of associations between log
                                             transformed blood lead and concurrent serum creatinine, the
                                             largest (3 was in the 141 participants whose peak blood lead
                                             < 10 ug/dL:
                                             0.06 (95% CI:  0.023, 0.097)

                                             Positive association between log transformed blood lead and
                                             change in serum creatinine over subsequent follow-up period
                                             in participants  whose peak blood lead was <25 ug/dL
                                             0.027 (95% CI: 0.0, 0.054)

-------
December 2005
Table
Reference
Study location
Study population
Sample size
Wu et al. (2003)
Boston, MA
Normative Aging Study,
1991-1995
n=709
6-4.1 (cont'd). Summary
Mean exposure and
outcome measures
Blood lead
6.2 ug/dL
Patella lead
32.1 ug/g bone
of Key Studies on the Renal Effects
Analysis methods
Covariates adjusted for in analysis
Multiple linear regression
Age, body mass index, hypertension,
smoking status, alcohol ingestion, analgesic
medication use
of Environmental Lead Exposure
Major significant findings
Significant association between patella lead and creatinine
clearance
P = -0.069 (SE not provided)
Tsaih et al. (2004)
Boston, MA
Normative Aging Study
1991—2001
n = 448
Calculated creatinine
clearance
71.3 mL/min

Blood lead at baseline
6.5 ug/dL

Tibia lead at baseline
21.5 ug/g bone mineral

Serum Creatinine at
Baseline
1.3mg/dL
Longitudinal analysis, mean of 6 years
between evaluations

Age, body mass index, diabetes, hypertension,
smoking status, alcohol consumption, analgesic
use, baseline serum creatinine and its square
Lead dose not associated with change in creatinine in all

Significant interaction of blood and tibia lead with diabetes in
predicting annual change in serum creatinine

For natural In baseline blood lead P = 0.076 (95% CI:  0.031,
0.121) compared to P = 0.006 (95% CI:  -0.004, 0.016) for
non-diabetics

For natural In baseline tibia lead P = 0.082 (95% CI:  0.029,
0.135) compared to P = 0.005 (95% CI:  -0.005, 0.015) for
non-diabetics
H
6
o
0
H
O
o
H
W
O
O
H
W
„, . ,,„„,, Blood lead
Staessen et al. (1992) ,, . ,,T , . ,
v ' 1 1 .4 ug/dL (males)
^ j i -i. i en j 7.5 ug/dL (females)
Cadmibel Study rb v '
n= 1,981; 965 males ,, , ,. .
Measured creatinine
clearance
99 mL/min (males)
80 mL/min (females)






Multiple linear regression
Age, age squared, body mass index, log
transformed gamma-glutamyl transpeptidase,
and diuretic use









Log transformed blood lead negatively associated with
measured creatinine clearance
-9.5(95%CI: -18.1, -0.9) males
- 12.6 (95% CI: -20.3, -5.0) females
Tenfold increase in blood lead associated with a decrease in
creatinine clearance of 10 and 13 mL/min in men and women,
respectively







-------
 1    population in terms of age and follow-up.  Renal status was assessed with serum creatinine.
 2    Data from 4-5 evaluations were available for the majority of participants. Relations were
 3    evaluated cross-sectionally (associations between blood lead and concurrent serum creatinine)
 4    as well as longitudinally (associations between blood lead and change in serum creatinine over
 5    the subsequent follow-up period). Mean age, blood lead level, and serum creatinine, at baseline,
 6    were 56.9 years (SD 8.3), 9.9 jig/dL (SD 6.1), and 1.2 mg/dL (SD 0.2), respectively. With
 7    random-effects modeling, a significant positive association between In-transformed blood lead
 8    and concurrent serum creatinine was observed.  This association was stronger when models were
 9    confined to participants with lower peak blood lead levels, i.e., the P coefficient was largest in
10    the 141 participants whose highest blood lead level was < 10 |ig/dL. In longitudinal analysis,
11    In-transformed blood lead was associated (p = 0.05) with change in serum creatinine over the
12    subsequent follow-up period in the 428 participants whose highest blood lead level was
13    <25 |ig/dL. Similar to the cross-sectional analysis, the P coefficient in the participants whose
14    highest blood lead level was < 10 |ig/dL was larger; however, in the longitudinal analysis, the
15    standard error also increased such that the p-value was not significant.
16          Cortical and trabecular bone lead measurements were obtained in evaluations performed
17    between 1991 and 1995 in 709 participants in the Normative Aging Study (Wu et al., 2003).
18    Lead dose was assessed with blood, tibia, and patella lead concentrations.  Renal outcome
19    measures included serum creatinine and estimated creatinine clearance.  Mean blood, tibia and
20    patella lead levels were 6.2 ng/dL (SD 4.1), 22.0 |ig/g bone mineral (SD 13.4), and 32.1  |ig/g
21    bone mineral (SD 19.5), respectively. After adjustment,  analyses in the 670 participants from
22    whom these data were available, revealed a significant inverse association between patella lead
23    and creatinine clearance. A borderline significant (p = 0.08) inverse association between tibia
24    lead and creatinine clearance was also observed. None of the lead measures were significantly
25    associated with serum creatinine.
26          Tsaih et al. (2004) reported associations between baseline lead dose and change in serum
27    creatinine in 448 men. Lead dose was assessed with blood, tibia, and patella lead.  Serum
28    creatinine was measured at baseline and at follow-up, an average of 6 years later. Six percent
29    and 26% of subjects had diabetes and hypertension, at baseline, respectively. Mean blood lead
30    levels and serum creatinine decreased significantly over the follow-up period in the group. Lead
31    dose was not associated with change in creatinine in all participants.  However, the authors found

      December 2005                            6-136        DRAFT-DO NOT QUOTE OR CITE

-------
 1    a significant interaction between lead dose (blood and tibia lead) and diabetes on change in
 2    serum creatinine. Interaction was also observed between tibia lead and hypertension, although it
 3    is possible that many of the 26 diabetics were also included in the hypertensive group and were
 4    influential there  as well.
 5
 6    6.4.4.1.3   NHANESIII
 1          Muntner et al. (2003) analyzed associations between blood lead and renal outcomes in
 8    15,211 adult subjects enrolled in the NHANES III study, conducted from 1988 through 1994.
 9    Dichotomous renal outcome measures analyzed included elevated serum creatinine and chronic
10    kidney disease (GFR < 60mL/min/1.73 m2). Due to interaction between blood lead and
11    hypertension, the population was stratified. Mean blood lead was 4.21 |ig/dL in the 4,813
12    hypertensives and 3.30 |ig/dL in normotensives. The prevalence of elevated serum creatinine in
13    hypertensives and nonhypertensives was 11.5% and 1.8%, respectively; prevalence of chronic
14    kidney disease was similar. The odds ratios for both renal outcomes increased by quartile of
15    blood lead among the hypertensive subjects but not among those without hypertension.  Among
16    those with hypertension, after adjustment for age, race and gender, the odds ratios for elevated
17    creatinine in quartiles 2, 3, and 4 compared to the lowest quartile of blood lead, were 1.56
18    (95% CI:  1.04, 2.35), 1.68 (95% CI: 1.24, 2.26), and 2.07 (95% CI:  1.26, 3.40), respectively.
19    As shown in Table 6-4.1, the odds ratios were the same following additional adjustment. The
20    authors noted that the "associations were strong, dose-dependent and consistent before and after
21    comprehensive adjustment." They  also noted that in nonhypertensives, higher blood lead was
22    associated with a higher prevalence of chronic kidney disease in diabetics. This study is notable
23    for the sample size, for the reported associations being observed at the lowest mean blood lead
24    level in any environmental study to date, for the comprehensive adjustment for other renal risk
25    factors, and for the study population being representative of the U.S. population.
26
27    6.4.4.1.4   Summary of Lead Nephrotoxicity in the General Population
28          Studies of environmentally exposed general populations constitute one of the two most
29    important types of research on the adverse renal effects of lead during the past two decades.
30    Study  designs are generally strong; some have the added strength of analyzing longitudinal data.
31    Populations are large, assessment of lead dose is comprehensive, including the use of bone lead

      December 2005                           6-137        DRAFT-DO NOT QUOTE OR CITE

-------
 1    as a measure of cumulative lead body burden in some studies, and statistical approaches are
 2    advanced, utilizing a range of exposure and outcome measures, while adjusting for numerous
 3    renal risk factors.  Given these strengths, the fact that these studies have reached consistent
 4    conclusions provides strong evidence indicating that lead is a contributor to renal dysfunction in
 5    susceptible populations at much lower levels that those identified in the 1986 Lead AQCD.
 6    Chronic kidney disease has been observed at the lowest lead dose levels studied (category II
 7    from 2.5 to 3.8 |ig/dL in Muntner et al. [2003]). An association between cumulative lead dose
 8    (mean tibia lead of 21.5 |ig/g bone mineral) and longitudinal decline in renal function has been
 9    observed as well, although data on any threshold for this effect were not reported (Tsiah et al.,
10    2004).  Susceptible populations include those with other risk factors for renal disease, including
11    hypertension and diabetes. Populations who are also at increased risk for obesity, diabetes, and
12    hypertension represent groups potentially most impacted by lead exposure.
13
14    6.4.4.2   Occupational Studies
15          The vast majority of studies in the lead-renal literature were conducted in the occupational
16    setting.  This was especially true prior to the 1986 Lead AQCD but is still currently the case.
17    Occupational studies of the renal effects of lead are presented in Annex Table AX6-4.2. Recent
18    studies in the general population, discussed above, and in the patient population, discussed in the
19    next section, provide consistent evidence supporting a role for lead in renal  dysfunction at lower
20    lead concentrations of interest.  However, one phenomenon that has been observed more
21    frequently in occupational rather than environmental studies of lead exposure and kidney
22    function deserves specific comment.
23          Several  studies have reported statistically significant negative associations between higher
24    lead dose and worse renal function, specifically positive associations between higher lead dose
25    and lower BUN, serum creatinine and/or higher creatinine clearance. Roels et al. (1994)
26    observed higher mean creatinine clearance in 76 lead workers compared to 68 controls from the
27    same smelter who were not occupationally exposed to lead (mean of 121.3 versus
28    115.5 mL/min/1.73 m2 in workers and controls, respectively [p < 0.05]). More importantly, in
29    the combined group, tibia lead was positively correlated with measured creatinine clearance.
30    However, no other significant associations between lead dose and the renal  outcomes (which also
31    included serum creatinine, urea nitrogen, and (32-microglobulin, along with urinary NAG, RBP

      December 2005                           6-138        DRAFT-DO NOT QUOTE OR CITE

-------
 1    and (32-microglobulin as well as other early biological effect markers) were observed. Lead
 2    workers had evidence of high past exposure and controls also had high blood lead levels by
 3    current standards (mean blood and tibia lead levels were 43.0 and 14.1  |ig/dL and 66 and 21 jig/g
 4    bone mineral, in workers and controls, respectively).
 5          Weaver et al. (2003a) performed a cross-sectional analysis of first evaluation data from a
 6    longitudinal study of 803 lead workers in South Korea, including 94 former lead workers.  Lead
 7    exposure was assessed with job duration; blood, tibia, and DMSA-chelatable lead; and three
 8    hematologic measures as surrogates for lead dose. Clinical renal function was assessed with
 9    blood urea nitrogen (BUN), serum creatinine, measured creatinine clearance, and calculated
10    creatinine clearance (Cockcroft and Gault, 1976).  Urinary NAG and RBP were also measured.
11    Mean job duration, and blood, tibia, and DMSA-chelatable lead levels were 8.2 years (SD 6.5),
12    32.0 |ig/dL (SD 15.0), 37.2 |ig/g bone mineral (SD 40.4), and 767.8 |ig/g creatinine (SD 862.1),
13    respectively.  Higher lead measures were associated with worse renal function in nine of the
14    42 associations, however, an additional five were in the opposite direction (higher lead measures
15    associated with lower serum creatinine  and higher creatinine clearances). These opposite
16    direction (inverse) associations were observed only for the clinical outcomes whereas the
17    associations between higher lead dose and worse renal function were predominantly with the
18    EBE markers. In three of 16 models (analyses) assessing effect modification by age on
19    associations between lead job duration and the three lead dose biomarkers with the four clinical
20    renal outcomes, positive associations between higher lead measures and worse renal function in
21    participants in the oldest age tertile were significantly different from associations in those in the
22    youngest age  tertile, which were in the opposite (inverse) direction; this pattern was observed at
23    borderline significance (p < 0.1) in three other models. However, this pattern was not observed
24    in the EBE marker models.  Similar inverse associations were observed in this population in the
25    third evaluation, performed a mean of 2.2 years after collection of the data discussed above,
26    but only with DMSA-chelatable lead and not patella, blood, or tibia lead (Weaver et al., 2005b).
27    Hsiao et al. (2001) also reported positive associations between higher blood lead and lower
28    concurrent serum creatinine in an analysis of 8 years of annual medical surveillance data in
29    30 lead battery workers (this study is of note since it is one of the few longitudinal, occupational
30    studies to date; additional findings are described in Annex Table AX6-4.2).
      December 2005                           6-139        DRAFT-DO NOT QUOTE OR CITE

-------
 1          These inverse associations, evidenced by higher lead measures and lower BUN and serum
 2    creatinine and/or higher creatinine clearance, may represent lead-induced hyperfiltration,
 3    a phenomenon initially observed in patients with diabetes but also implicated in other settings,
 4    including hypertension and obesity (Nenov et al., 2000). In this process, initial supranormal
 5    renal function is paradoxically associated with increased risk for subsequent renal dysfunction.
 6    Hu (1991) has  also reported increased mean creatinine clearance in 22 adults who were lead
 7    poisoned as children compared to matched controls. Longitudinal data for lead-exposed rodents
 8    (discussed in Section 5.7) are critical in relating this process to lead.  However, in that work,
 9    despite similar initial hyperfiltration, subsequent renal dysfunction was much more severe in the
10    high-dose lead-exposed rodents compared to the low-dose animals.  This suggests that
11    hyperfiltration  may be one, but not the only, mechanism for the adverse renal effects of lead.
12    Whether hyperfiltration contributes to pathology in humans is  unclear; longitudinal studies are
13    needed.
14          Regardless, the issue for this document is that significant findings could be obscured if
15    opposite direction associations are present in different segments of the study population and
16    interaction models to address this are not performed.  This is a valid concern, since the factors
17    involved in these inverse associations in lead exposed populations are not well defined at
18    present. Work by Weaver and colleagues have used age as the effect modifier; however, other
19    factors, such as lead job duration, may be more important modifiers.
20          Figure 6-4.1 provides an example of the different associations observed depending on
21    whether effect  modification is examined. In the work of Weaver et al. (2003a), in several
22    models, no associations were observed when the entire population was studied; however, when
23    interaction models using age as the effect modifier were evaluated, significant associations in
24    opposite directions were observed.
25
26    6.4.4.3   Patient Population Studies
27          Studies  in various patient populations have also contributed to the body of knowledge
28    concerning adverse renal impacts of lead exposure. Such studies of renal effects of lead in
29    patient populations are presented in Annex Table AX6-4.3. Populations studied include those
30    with chronic renal insufficiency (CRI), end-stage renal disease (ESRD), gout, and hypertension.
31    Patients were selected for study due to the fact that these diseases are thought to be increased by

      December 2005                            6-140        DRAFT-DO NOT QUOTE OR CITE

-------
                                                      ^         Older
                        Worse
                        Renal
AH
                        Function   t                    _^
                                                                  Younger
                                              Lead dose

     Figure 6-4.1.   Effect on renal function evaluation using age as the effect modifier.
     Source: Weaver et al. (2003a).


 1   high-level lead exposure, particularly when two or more coexist in the same patient.  Early work
 2   in this area was discussed in the 1986 Lead AQCD, but some research is discussed again below
 3   in order to allow conclusions to be drawn on the data to date. Early research focused on patients
 4   with potential lead nephropathy; and lead body burdens of interest, assessed with EDTA
 5   chelation, were above 600 to 650 |ig/72 h. More recent work has involved patients with CRI but
 6   not ESRD and lead body burdens below that range.  Of note, the terminology for degree of renal
 7   dysfunction was more variable in the older literature. Therefore, for clarity in the discussion
 8   below, patients with impaired renal function who are not on dialysis are noted to have CRI,
 9   rather than using terms such as "failure" that were used in some of the original reports.
10          Two issues have been a recurring concern in this work, particularly in work with patients
11   on dialysis. The first concern is whether lead body burden is higher in all patients with renal
12   insufficiency or failure due to decreased lead excretion (reverse causality). The second concern
13   is whether EDTA-chelatable lead levels when measured over a 72-h period in patients with CRI
14   can be equated to those in participants with normal renal  function measured over 24 h.  It is
15   possible that, due to decreased excretion of EDTA in renal insufficiency, more lead per dose is
16   ultimately chelated.  These concerns have been addressed in various ways as noted in the
17   research discussed below. Lastly, this work also illustrates the limitations of blood lead levels
18   which often do not reflect the differences in lead body burden noted between populations.
19

     December 2005                          6-141        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.4.4.3.1   Lead Body Burden in Chronic Renal Disease
 2          Batuman et al. (1983) studied 27 hypertensives with CRT (defined as serum creatinine
 3    >1.5 mg/dL) and 21 without associated renal impairment. Blood lead levels were similar in the
 4    two groups. However, mean EDTA-chelatable lead levels were significantly higher in
 5    hypertensives with CRT than those without (860 and 340 |ig/72 h, respectively). Further,
 6    chelatable lead levels in patients with CRT from causes not thought to be related to lead
 7    nephropathy and who had no history of lead exposure were similar to patients with hypertension
 8    but no CRT. This provides some evidence against reverse causality.
 9          Sanchez-Fructuoso et al. (1996) performed a similar study in a much larger number of
10    patients in Spain, none of whom had a known history of lead exposure.  These authors reported
11    that EDTA-chelatable lead levels >600 |ig/72 h were present in none of 30 controls, 16 (15.4%)
12    of  104 patients with essential (primary) hypertension and normal renal function, 74 (56.1%) of
13    132 patients with CRT of unknown etiology along with hypertension (64 of the 132 also had
14    gout), but none of the 30 patients with CRT of known (non-lead related) etiology. Mean blood
15    and EDTA-chelatable lead levels in the patients with CRT of known cause were not statistically
16    different from controls with normal renal function. These researchers also reported significant
17    correlations between bone lead levels (assessed by biopsy) and EDTA-chelatable lead level in
18    12  patients whose chelatable lead levels were >600 |ig/72 h, which provides support for the
19    validity of chelatable lead levels in CRT.
20          In contrast, Osterloh et al.  (1989) reported no significant difference in EDTA-chelatable
21    lead levels between 40 male subjects with hypertensive nephropathy (hypertension preceded
22    renal insufficiency; serum creatinine 1.8-4 mg/dL) and 24 controls with renal  dysfunction from
23    other causes. Lead dose and serum creatinine were not correlated. Chelatable lead levels in this
24    population were much lower than those reported by Wedeen et al. (1983) and Sanchez-Fructuoso
25    et al. (1996). The authors noted that only 17% of their study participants had  a history of
26    possible lead exposure based on questionnaire. In contrast, Batuman et al. (1983) found that
27    89% of patients with hypertension and CRI had a possible history of lead exposure. The
28    inconsistent results in these studies may reflect differences in the patients studied. Batuman
29    et al. (1983) studied Veterans Administration patients, Sanchez-Fructuoso et al. (1996) studied
30    patients from a low-medium income area in Madrid, Spain, and Osterloh et al. (1989) recruited
31    patients from the database of a large health maintenance organization in California.

      December 2005                          6-142       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Van de Vyver et al. (1988) reported lead data from bone biopsies in 153 dialysis patients,
 2    11 cadavers without known excessive lead exposure, 13 patients with renal insufficiency, gout,
 3    and/or hypertension and 22 lead workers. Bone lead levels in 5% of the dialysis population were
 4    in the range observed in lead workers, suggesting lead as a primary cause of their renal failure.
 5    Levels in the 10 patients with analgesic nephropathy were the lowest (all <7 |ig/g). However,
 6    Winterberg et al. (1991) subsequently noted that the bone lead levels in patients with analgesic
 7    nephropathy and cadaver controls in Van de Vyver et al. (1988) were much higher than in
 8    control groups of other researchers.  They reiterated the concern that lead did accumulate due to
 9    decreased renal excretion. In a longitudinal study, Price et al. (1992) reported similar half-lives
10    of lead in bone in eight renal patients compared with age-matched controls who had XRF finger
11    bone lead conducted twice 5 years apart. The small number and inclusion of outliers without
12    formal statistical analysis limits conclusions that can be drawn from these data. The longitudinal
13    studies of Lin and colleagues, discussed below, provide more definitive data in this regard.
14
15    6.4.4.3.2   Impact of Lead Body Burden on Decline in Renal Function in Patients with CRI
16          Lin and colleagues have addressed the issue of low-level lead as a cofactor with other
17    renal risk factors in susceptible populations, including those with CRI and/or gout. They have
18    approached this work in two ways: prospective follow-up of populations with CRI to determine
19    if renal function decline is greater in those with higher lead body burdens and through
20    randomized trials to determine if chelation therapy changes the rate of renal function decline.
21    Importantly, their work is in an EDTA-chelatable lead range well below that considered
22    abnormal as described in Section 6.4.4.3.1.
23          In their most recent publication, Yu et al. (2004) followed 121 patients over a 4-year
24    period. Eligibility required well-controlled CRI.  Importantly, serum creatinine between 1.5 and
25    3.9 mg/dL and EDTA-chelatable lead <600 |ig/72 h were required at baseline. Patients with
26    potentially unstable renal disease were excluded (i.e., due to  systemic diseases such as diabetes).
27    Sixty-three patients had "high-normal" EDTA-chelatable lead levels (> 80 but <600 |ig/72 h);
28    58 had "low-normal" EDTA-chelatable lead levels (<80 jig lead/72 h). The groups were similar
29    in most other baseline risk factors. Borderline statistically significant (p < 0.1) differences
30    included mean older age in the high chelatable lead group and certain renal diagnoses. Fifteen
31    patients in the "high-normal" chelatable lead group reached the primary endpoint (doubling of

      December 2005                            6-143       DRAFT-DO NOT QUOTE OR CITE

-------
 1    serum creatinine over the 4-year study period or need for hemodialysis) compared to only two in
 2    the "low-normal" group (p = 0.001).
 3          In a Cox multivariate regression analysis, chelatable lead was significantly associated
 4    with overall risk for the primary endpoint (hazard ratio for each 1 jig chelatable lead was 1.01
 5    [95% CI: 1.00, 1.01; p = 0.002]). In this model, the only other variable reaching at least
 6    borderline significance (p < 0.1) was baseline serum creatinine. The associations between
 7    baseline chelatable lead or blood lead level and change in GFR (estimated by an MDRD
 8    equation [Levey et al., 1999]) were modeled separately using GEE.  Based on these models, a
 9    10 jig higher chelatable lead level or  a 1 |ig/dL higher blood lead level reduced the GFR by 1.3
10    and 4.0 mL/min, respectively, during the 4-year study period.  Similar to the primary outcome
11    analysis, of the many traditional renal risk factors adjusted for in these models, only diagnosis of
12    chronic interstitial nephritis was significantly associated, in this case with an increase in GFR.
13    Of note, chronic interstitial nephritis was also a more frequent diagnosis in the group with the
14    low-normal chelatable lead levels (p = 0.09).
15
16    6.4.4.3.3   Therapeutic EDTA Chelation in Patients
17          Chelation in lead exposure is controversial due to the potential for it to be used in lieu of
18    exposure reduction. Chelation in lead nephropathy, in particular, is controversial, because cases
19    of acute tubular necrosis were reported following early clinical use of EDTA that involved large
20    doses in the treatment of hypercalcemia and lead poisoning. Adverse renal effects have not been
21    observed in subsequent work using much lower doses (Sanchez-Fructuoso et al., 1996; Wedeen
22    etal.,  1983).
23          Work prior to the 1986 Lead AQCD suggested that chelation might be beneficial in lead
24    nephropathy (Morgan, 1975; Wedeen et al., 1979).  This issue has been addressed more
25    recently by Lin and colleagues in patients with much lower lead doses. Lin et al. (1999) studied
26    43 patients with serum creatinine and EDTA-chelatable lead levels between 1.5-4 mg/dL and
27    150 and 600  |ig/72 h, respectively. Patients were followed for 12 months to determine their
28    baseline rate of renal function decline.  A group of 32 was then randomized; and 16 underwent a
29    2-month treatment period consisting of weekly chelation with 1 g EDTA; whereas the other
30    16 continued their regular care. Traditional renal risk factors, such as blood pressure control,
31    were similar in the two groups.  Prior to therapeutic chelation, the rate of progression of renal

      December 2005                           6-144        DRAFT-DO NOT QUOTE OR CITE

-------
 1    insufficiency was not statistically different. However, actual improvement in renal function was
 2    noted in the treated group during chelation and subsequent renal function decline was slower in
 3    this group. The mean difference in the change in the reciprocal of serum creatinine post therapy
 4    was 0.000042 L/|imol per month (95% CI: 0.00001,0.00007).
 5          In subsequent work, Lin et al. (2003) published results of a randomized chelation trial in a
 6    larger group. This work included a 2-year prospective study of renal function decline prior to
 7    chelation in 202 patients with CRT and EDTA-chelatable lead <600  |ig/72 h. Results of the Cox
 8    proportional-hazards model were similar to those reported in Yu et al. (2004). Associations
 9    between baseline EDTA-chelatable lead level and change in GFR were modeled using GEE.
10    After adjustment, an increase of 10 jig in EDTA-chelatable lead was associated with a GFR
11    decrease of 0.03 mL/min/1.73 m2 of body-surface area during the observation period (p < 0.001).
12    Of note, this effect, although statistically significant, is 40-fold lower than that reported in Yu
13    et al. (2004) over a follow-up period that is only 2-fold shorter.  At 24 months, 64 patients whose
14    EDTA-chelatable lead levels were 80-600 |ig/72 h were randomized; half to a 3-month treatment
15    period consisting of weekly chelation with 1 g EDTA until their excreted lead levels fell below
16    60 |ig/72 h and half to placebo infusion over 5 weeks.  Renal risk factors were similar in the two
17    groups. Mean blood lead levels were 6.1 |ig/dL and 5.9 |ig/dL in treated and control groups,
18    respectively. In the subsequent 24 months, chelation in 19 (59%) participants was repeated due
19    to increases in serum creatinine in association with rebound  increases in EDTA-chelatable lead
20    levels.  Each received one additional chelation series (mean  = 4.1 g EDTA) a mean of 13.7
21    months after the first chelation period. At the end of the study period, mean estimated GFR
22    increased by 2.1 mL/min/1.73 m2 of body-surface area in the chelated group compared to a
23    decline of 6.0 mL/min/1.73  m2 of body-surface area in the controls (p < 0.01) (see Figure 6-4.2).
24    The 95% CI for the difference between the chelated and control groups was -11.0 to - 5.1
25    mL/min/1.73 m2 of body-surface area.
26          Lin and colleagues have also reported chelation results in patients with gout.  Historically,
27    gout was known to be a risk from high-level lead exposure such as in the Queensland, Australia
28    epidemic and in moonshine alcohol drinkers (U.S. Environmental Protection Agency, 1986a).
29    Higher EDTA-chelatable lead levels in patients with both gout and CRT compared to those with
30    CRT or gout alone have also been reported in several studies at lower levels of exposure
      December 2005                          6-145        DRAFT-DO NOT QUOTE OR CITE

-------

CM
E
CO
r:
"c
E
i
0)
4-rf
a

-------
 1          The key studies in patients followed prospectively with and without chelation constitute
 2    the other major advance in research on the adverse renal effects of lead over the past two
 3    decades. This work suggests that lead is nephrotoxic in susceptible populations at lower levels
 4    than currently appreciated. Blood lead levels (mean = 4.2 |ig/dL [range 1.0-13.4]) and body lead
 5    burdens (mean = 99.1 jig/72 h [range 2.5-530]) were associated with decline in GFR over a
 6    4-year follow-up period (Yu et al., 2004).  Chelation was beneficial in a body lead burden range
 7    from 80 or 150 to 600 |ig/72 h, depending on study (Lin et al., 2003; 1999). The published data
 8    do not allow a determination of whether a threshold exists. It is also possible that chelation has a
 9    direct beneficial effect on kidney function, regardless of lead exposure, since DMSA has been
10    reported to prevent renal damage in a non-lead exposed rat model of nephrosclerosis (Gonick
11    et al., 1996). If so, the benefits of chelation do not appear to  occur via reversal of structural
12    damage (Khalil-Manesh et al., 1992b); improved hemodynamics from reduction of reactive
13    oxidant species may be a mechanism (Gonick et al., 1996).
14          Strengths of the work of Lin  and colleagues include prospective study design, lead dose
15    assessment including bioavailable body burden, statistical analysis that includes GEE for
16    longitudinal data, and adjustment for more renal risk factors than any of the other key studies
17    discussed in Section 6.4. Limitations include that fact that, to date, this type of research has been
18    conducted in relatively small  number of participants and in only one center. As noted above, the
19    two reported lead body burden P coefficients in GEE models of decline in renal function vary
20    widely.  Therefore, small study sizes and differences in renal diagnoses between groups may be
21    overly influential in the results. However, if confirmed in large populations, the potential public
22    health benefit could be substantial.  Therapeutic options would be available for high-risk
23    patients, who, despite dramatic reductions in lead exposure in developed countries, are still
24    adversely affected by lead. Lin et al. (2003) noted that, based on their data, chelation could
25    delay the need for hemodialysis by 3 years.  Therefore, this unique line of research is deserving
26    of further study.  Prospective studies consistent with the results of Yu et al. (2004) will likely be
27    needed to justify randomized,  controlled chelation trials.
28
29    6.4.4.4   Mortality Studies
30          As summarized in Steenland et al. (1992), mortality studies have consistently shown
31    excess mortality from chronic kidney disease in lead workers. This increased risk has been most

      December 2005                            6-147        DRAFT-DO NOT QUOTE OR CITE

-------
 1    apparent in workers exposed in earlier time periods, becoming nonsignificant in later calendar
 2    time periods in a number of studies.  Steenland et al. (1992) reported similar results in a study of
 3    1990 former lead smelter workers. This cohort was made up of predominantly white men who
 4    had worked in a lead-exposed department for at least 1 year between 1940 and 1965. Mean (SD)
 5    blood lead, measured in 1976 in 173 members of this cohort, was 56.3 |ig/dL (12.9).  There were
 6    8 deaths from chronic kidney disease. Compared to the U.S. white male population, the
 7    standardized mortality ratio was 1.26 (95% CI:  0.54,2.49). The standardized mortality ratio
 8    increased with duration of exposure from 0.79 in workers exposed  1-5 years to 2.79 in workers
 9    exposed >20 years, although the standardized mortality ratios did not reach  significance (CI not
10    reported). Lead exposure in U.S. industries has declined over the years, and this has been
11    hypothesized as an explanation for the reduction in mortality from renal disease observed in this
12    type of study.  However, that fact that improved treatments for chronic renal disease have led to
13    a decrease in mortality from end-stage renal disease (U.S. Renal Data System, 2004) may also be
14    a factor. The mortality studies by Steenland et al. (1992) and others are further described in
15    Annex Table AX6-4.4.
16
17    6.4.5  Lead Nephrotoxicity in Children
18    6.4.5.1    Studies in Adults Following Childhood Lead Poisoning
19         Henderson clearly established an increased risk for lead nephropathy in adult survivors of
20    untreated childhood lead poisoning (Henderson,  1955). Lead nephropathy was responsible for
21    substantial mortality in the Queensland, Australia population. However, as  noted in the 1986
22    Lead AQCD, other studies of adults who survived childhood lead poisoning have not reported
23    this degree of renal pathology.  Studies published since 1986 have not observed the degree of
24    renal pathology noted in the Queensland work either but have revealed some interesting findings.
25    These studies, along with other studies of renal effects of lead in children, are  presented in
26    Annex Table AX6-4.5. Pertinent results are discussed below.
27         A study  of comparing 21 adults, who had experienced childhood lead poisoning between
28    1930 and 1942, to age, sex, race, and neighborhood-matched controls found no significant
29    differences in blood lead level, serum creatinine, or BUN (Hu, 1991).  Mean measured creatinine
30    clearance was unexpectedly higher in the previously lead-poisoned group compared to controls
31    (112.8 versus 88.8 mL/min/1.73 m2 [p < 0.01]). Mean in the lead-exposed group was also higher

      December 2005                           6-148       DRAFT-DO NOT QUOTE OR CITE

-------
 1    than the predicted value of 94.2 mL/min/1.73 m2 from the nomogram of Rowe et al. (1976).
 2    One survivor, who was identified but not included in the study, had been diagnosed with chronic
 3    interstitial nephritis on renal biopsy. Her blood lead was 30 |ig/dL and her presentation was thus
 4    consistent with actual lead nephropathy. Strengths of this study included clear criteria for lead
 5    poisoning and assessment of clinical renal function that included both measured and estimated
 6    creatinine clearances. However, the study was limited by  small size and the fact that the number
 7    enrolled was a very small subset of the initially identified cohort of 192.  At least 43 (22.4%) of
 8    the 192 were confirmed to be deceased.  That group had evidence of higher initial lead exposure,
 9    which raises concern regarding survivor bias in the study group.  More importantly, the higher
10    mean creatinine clearance in the  lead exposed group provides further evidence for lead-related
11    hyperfiltration.  Again, as discussed in the occupational study section, this may hamper attempts
12    to detect associations between lead dose and adverse renal effects.
13          Another study compared 62 participants who were diagnosed and chelated between 1966
14    and 1972 for initial blood lead levels >100 |ig/dL  to 19 age-matched siblings whose initial blood
15    lead levels were <40 |ig/dL (Moel and Sachs,  1992). Mean initial blood lead level was
16    150.3 |ig/dL (SD 77.1) in the 62  survivors of lead poisoning; levels in the siblings could not be
17    precisely  quantified since values <40 |ig/dL were not reported as exact values.  Mean  age  at
18    diagnosis was 2.2 years; age at follow-up was 22.2 years.  Blood lead and renal function were
19    serially monitored in the population.  In 1983, mean blood lead in the poisoned group was
20    statistically higher than in siblings (means of 14.5  and 11.6 |ig/dL, respectively) but, by 1989,
21    both groups had a mean lead of 7.4 |ig/dL.  Renal outcomes included serum creatinine, uric acid,
22    and (32-microglobulin, fractional  excretion of (32-microglobulin, urinary protein:creatinine  ratio,
23    and tubular reabsorption of phosphate. No statistically significant differences between
24    previously lead-poisoned children and their siblings were observed. The prevalence of abnormal
25    values between the two groups was not different. Initial blood lead level was not associated with
26    serum creatinine, after adjustment for  age, gender, and body mass index. However, since blood
27    lead levels were not quantified in the siblings, their values were entered as 40 |ig/dL in the
28    continuous blood lead model.  Lead was also entered as a dichotomous variable (poisoned versus
29    siblings).  Given the data available, the analysis was limited. No cumulative measure of lead
30    was analyzed nor was the serially obtained data analyzed with longitudinal modeling.  Four of
31    62 participants did experience an increase in serum creatinine such that their levels were

      December 2005                           6-149        DRAFT-DO NOT QUOTE  OR CITE

-------
 1    > 1.4 mg/dL by the end of the study period.  These subjects may reflect lead-related renal
 2    toxicity, especially given the young age of the participants. However, it can be concluded that,
 3    despite lead exposures that are now considered extremely high, the degree of renal pathology
 4    was clearly not to the extent seen in Australia.  The fact that these children were chelated when
 5    lead poisoning was diagnosed but the Queensland children were not may be an important
 6    distinction. Additional follow-up with longitudinal analysis would be of value in these children
 7    in order to evaluate their renal function as they develop other renal risk factors.
 8
 9    6.4.5.2   Lead Body Burden in Children with Chronic Renal Disease
10          Scharer et al. (1991) reported higher lead content in deciduous teeth in 22 German
11    children, age 5-14 years, with varying degrees of renal insufficiency compared to a control group
12    of 20 siblings or neighbors and a group of 16 children without known lead exposure. Mean
13    dental lead content was 2.8, 1.7, and 1.4  jig/g, in the three groups, respectively. Lead levels in
14    teeth were significantly higher in both the patient and sibling/neighbor control groups compared
15    to the unexposed control group. Mean blood lead in the renal patients was only 2.9 |ig/dL
16    (range 1.1-10.1  |ig/dL).  Lead in teeth was not correlated with  duration of renal impairment.
17    However, reflective of the ongoing controversy at the time this article was published regarding
18    whether decreased renal excretion causes increased lead storage, the authors attributed elevated
19    lead levels to both exposure and accumulation from decreased renal excretion.
20
21    6.4.5.3   General Population Studies in Children
22          In the first study of the renal effects of lower level environmental lead exposure in
23    children (as opposed to lead-poisoned children), Bernard et al. (1995)  carried out a cross-
24    sectional study of 195 children in the Czech Republic.  One hundred forty-four children (63 boys
25    and 81 girls) lived in 2 areas close to a lead smelter (designated as exposed groups one and two).
26    A control group of 51 children lived in a rural area. Blood lead levels and urinary renal  early
27    biologic effect markers (RBP, p2-microglobulin, NAG, albumin, and Clara cell protein)  were
28    obtained. Age ranged from 12 to!5 years. In girls, mean blood lead concentrations in controls
29    and exposure groups one and two were 8.4, 9.4, and 12.9 |ig/dL,  respectively. Corresponding
30    values in boys were 8.7, 10.9, and 14.9 |ig/dL, respectively. These levels were significantly
      December 2005                           6-150        DRAFT-DO NOT QUOTE OR CITE

-------
 1    higher in both exposed groups compared to the control group. In contrast, blood cadmium levels
 2    were similar among all groups. After adjustment for age, sex, blood cadmium, and zinc
 3    protoporphyrin, log transformed blood lead was associated with log transformed RBP ((3 = 0.302,
 4    p = 0.005).
 5          Verbeck et al. (1996) studied 151 Romanian children residing at various distances from a
 6    lead smelter. Associations between blood lead levels and renal outcome measures (urinary RBP,
 7    NAG, ai-microglobulin, albumin, and alanine aminopeptidase) were analyzed. Mean age was
 8    4.6 years (range of 3-6 years); gender was evenly divided. Mean blood lead was 34.2 |ig/dL
 9    (SD 22.4), which is much higher than in Bernard et al. (1995).  After adjustment for age  and
10    gender, a  10 |ig/dL increase in blood lead was associated with a 13.5% (90% CI: 10.2, 17)
11    increase in NAG excretion.  No threshold was observed. Blood cadmium levels measured in a
12    subset of the population were all <2 |ig/L; however, this variable was not entered into the
13    regression model.
14          Another study included 200 French children who resided close to smelters along with
15    200 age and gender matched controls recruited from areas believed to be unpolluted by heavy
16    metals (DeBurbure et al., 2003). Blood lead and cadmium were measured.  Renal outcomes
17    included urinary total protein, albumin, transferrin, p2-microglobulin, RBP, brush border
18    antigens, and NAG. Age ranged from 8.5 to 12.3 years. Geometric mean blood leads ranged
19    from 2.7 |ig/dL (SD 0.2) in female controls to 4.2 |ig/dL (SD 0.2) in exposed males.  The highest
20    geometric mean blood cadmium was 0.52 |ig/L. After adjustments for covariates, blood lead
21    was not associated with any renal outcomes; however, blood cadmium was positively associated
22    with NAG.  This association was present in both control and exposed areas.  Notably, the blood
23    lead levels in this study were much lower than in the two studies discussed above.
24          Staessen et al. (2001) studied 200 17-year-old Belgian children. The two exposed groups
25    were recruited from industrialized suburbs, whereas, the control group was recruited from a rural
26    area. Mean blood lead levels were 1.5, 1.8, and 2.7 |ig/dL in controls, and exposed groups one
27    and two, respectively. The renal outcome measures analyzed were urinary p2-microglobulin  and
28    serum cystatin-C. Although blood lead levels were low, after adjustment for sex and smoking
29    status, blood lead was associated with both p2-microglobulin and cystatin-C. Interestingly, blood
30    cadmium  was not associated with either outcome.
     December 2005                           6-151       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The current lack of sensitive markers of early renal damage that have been shown to
 2    predict subsequent renal function decline in longitudinal studies of lead exposed populations is
 3    problematic for research in this field. This is particularly true when studying children who do
 4    not have many of the other renal risk factors, such as hypertension and diabetes that older adults
 5    do.  Coratelli et al. (1988) reported a decline in urinary NAG in association with a 1 month
 6    period of decreased occupational exposure in 20 adult lead battery factory workers followed
 7    over a 1 year period. Clinical renal function measures were not studied however.  Sarasua et al.
 8    (2003) studied 526 adults and children, a mean of 4.5 years after an initial evaluation of renal
 9    function including measurement of urinary albumin, NAG, RBP, and alanine aminopeptidase.
10    These participants were drawn from three populations exposed to volatile organic compounds
11    and explosives via groundwater and controls.  Follow-up was performed to determine if the EBE
12    markers remained elevated and whether the presence of elevated EBE markers at baseline was
13    associated with abnormalities in serum creatinine, serum cystatin C, 24-h creatinine clearance,
14    and urine osmolality at follow-up. Among children who had elevated EBE markers at baseline,
15    renal EBE markers remained elevated in 38%. However, none remained elevated in the 32 who
16    had completed adolescence by the time of the follow-up.  The authors noted the potential for
17    puberty related biomarker changes.  Further, abnormalities in the clinical measures were rare at
18    follow-up. In contrast, elevated EBE markers at baseline in adults with chronic medical
19    conditions of risk to the kidney, such as diabetes and to a lesser extent heart disease and
20    hypertension, had persistent elevation of EBE markers and evidence of worse renal function at
21    follow-up. Limitations of this study include limited data analysis, some loss to follow-up, and
22    limited information on whether the original exposures for which these populations were studied
23    may have influenced these results. The authors stated that no significant differences in renal
24    outcomes between participants from exposed and control communities at follow-up were
25    observed and noted that kidney function was not worse in exposed communities in the initial
26    evaluations, which appeared to refer to the biomarkers as well. However, this report does not
27    indicate whether exposure changed between the initial biomarker collection and the follow-up
28    (i.e., decreased substantially or stopped altogether).  Still the study illustrates the need for further
29    prospective research to validate EBE markers in nephrotoxicant-exposed populations.
      December 2005                           6-152       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.4.6   Mechanisms for Lead Nephrotoxicity
 2          Individuals who have been heavily exposed to lead are at increased risk for both gout and
 3    renal disease (Shadick et al. 2000; Batuman 1993).  Lead is thought to increase serum uric acid
 4    (urate) by decreasing its renal excretion (Emmerson, 1965; Ball and Sorensen, 1969; Emmerson
 5    and Ravenscroft, 1975).  As discussed above, research in the last decade indicates that lead is
 6    nephrotoxic at lower levels than previously recognized. The same is true for uric acid (Johnson
 7    et al., 2003). Therefore, it is possible that one mechanism for lead-related nephrotoxicity, even
 8    at current lower levels of lead exposure, is via increasing serum uric acid.
 9          In order to address this question, Weaver et al. (2005a) analyzed data from 803 current
10    and former lead workers to determine whether lead dose was associated with uric acid and
11    whether previously reported associations between lead dose and renal outcomes (Weaver et al.,
12    2003) were altered after adjustment for uric acid. Outcomes included uric acid, blood urea
13    nitrogen, serum creatinine, measured and calculated creatinine clearances, and urinary NAG and
14    RBP. Mean uric acid, tibia lead, and blood lead levels were 4.8 mg/dL (SD 1.2), 37.2 jig/g bone
15    mineral (SD 40.4), and 32.0 |ig/dL (SD 15.0), respectively. None of the lead measures (tibia,
16    blood, and DMSA-chelatable lead) were associated with uric acid, after adjustment for age,
17    gender, body mass index, and alcohol use.  However, when effect modification by age on these
18    relations was examined, both blood and tibia lead were significantly associated in participants in
19    the oldest age tertile ((3 = 0.0111 [95% CI: 0.003, 0.019] and (3 = 0.0036 [95% CI:  0.0001,
20    0.007]) for blood and tibia lead, respectively). These models were further adjusted for blood
21    pressure and renal function. Hypertension and renal dysfunction are known to increase uric acid.
22    However, they are also risks associated with lead exposure. Therefore, adjustment for these
23    variables in models of associations between lead dose and uric acid likely results in overcontrol.
24    On the other hand, since non-lead-related factors contribute to both renal dysfunction and
25    elevated blood pressure, lack of adjustment likely results in residual confounding. Therefore, as
26    expected, associations between lead dose and uric acid decreased after adjustment for systolic
27    blood pressure and serum creatinine, although blood lead remained borderline significantly
28    associated ((3 = 0.0071 [95% CI: -0.001, 0.015]). However, when the population was restricted
29    to the oldest tertile of workers with serum creatinine greater than the median (0.86 mg/dL), likely
30    the highest risk segment of the population, blood lead remained significantly associated with uric
31    acid even after adjustment for systolic blood pressure and serum creatinine ((3 = 0.0156).

      December 2005                           6-153        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Next, in models of renal function in all workers, uric acid was significantly associated with all
 2    renal outcomes except NAG. Finally, in the oldest tertile of workers, after adjustment for uric
 3    acid, associations between lead dose and NAG were unchanged, but fewer of the previously
 4    significant (p < 0.05) associations noted between lead dose and the clinical renal outcomes in
 5    Weaver et al. (2003a) remained significant.
 6          Data from the Normative Aging Study indicate that lead dose, at levels lower than those
 7    known to increase the risk for gout or in the study of Weaver et al. (2005a), is associated with
 8    increased uric acid (Shadick et al., 2000). Mean blood, patella, and tibia lead levels were
 9    5.9 |ig/dL, 30.2 |ig/g bone mineral, and 20.8 |ig/g bone mineral, respectively, in 777 participants.
10    A significant association between patella lead and uric acid ((3 = 0.007 [[95% CI: 0.001, 0.013];
11    p = 0.02) was found, after adjustment for age, BMI, diastolic  blood pressure, alcohol ingestion,
12    and serum creatinine. Borderline significant associations between tibia (p = 0.06) and blood lead
13    (p = 0.1) and uric acid were also observed. Notably these associations were significant even
14    after adjustment for blood pressure and renal function, providing further evidence that low-level
15    lead increases uric acid.
16          These data suggest that older workers comprise a susceptible population for increased uric
17    acid due to occupational lead exposure. Uric acid may be one mechanism for lead-related
18    nephrotoxicity. However, this is not the only mechanism, since in Weaver et al. (2005a), the
19    association between blood lead and serum creatinine remained significant even after adjustment
20    for uric acid. These mechanistic relations have more than just theoretical importance. Clinically
21    relevant therapies may be possible since, as noted above, EDTA chelation has been reported to
22    improve both renal function and urate clearance in patients with renal insufficiency and gout,
23    even when EDTA-chelatable lead body burdens were low (Lin et  al., 2001).
24
25    6.4.7  Susceptible Populations for Lead  Nephrotoxicity
26    6.4.7.1   Chronic Medical Diseases
27          The general population studies by Tsaih et al. (2004) and Muntner et al. (2003) (discussed
28    in section 6.4.4.1 General Population Studies above) indicate that patient populations with
29    diabetes and hypertension are at increased risk for adverse renal effects of lead.  The work of
30    Lin and colleagues (several articles discussed in section 6.4.43 Patient Population Studies
31    above) indicates that patients with CRI and gout are also at increased risk. In these settings, lead

      December 2005                           6-154       DRAFT-DO NOT QUOTE OR CITE

-------
 1    appears to acts as a cofactor with other renal risk factors to cause early onset of renal
 2    insufficiency and/or a steeper rate of renal function decline.  It is likely that the presence of
 3    larger high risk populations within general populations is an important factor in the lower lead
 4    dose thresholds noted for the adverse effects of lead on the kidney in environmental compared to
 5    occupational research.
 6
 7    6.4.7.2    Age
 8          The work of Weaver and colleagues (discussed in Section 6.4.4.2.3 Korean Lead Workers
 9    above) suggests that older age is a risk factor.  This is consistent with research in general
10    populations (Lindeman et al., 1985) and is biologically plausible, since most renal risk factors
11    increase with age.  Gonick and Behari (2002) have summarized the data regarding the potential
12    contribution of lead exposure to essential  hypertension; similar issues may be involved with the
13    renal dysfunction observed in aging.
14
15    6.4.7.3    Genetic Polymorphisms
16    6.4.7.3.1   S-Aminolevulinic AcidDehydratase (ALAD)
17         Research in the last two decades suggests that several genetic polymorphisms affect lead
18    toxicokinetics (i.e., modify the relation between lead exposure and dose).  Of those that are
19    potentially relevant to the kidney, data on the gene that encodes for 8-aminolevulinic acid
20    dehydratase (ALAD) are the most important in this regard. The ALAD enzyme is a principal
21    lead binding protein; the isozymes in those with the ALAD2 allele are more electronegative and
22    bind a greater proportion of blood lead than does the protein in individuals with the ALAD 11
23    genotype (Bergdahl et al., 1997). Research to date indicates that individuals with the ALAD2
24    allele generally have higher blood lead levels than those with the ALAD 11 genotype, although
25    this may not be the case at lower levels of lead exposure (i.e., mean blood lead levels <10 |ig/dL)
26    (Kelada et al., 2001).  Participants with the ALAD2 allele have been found to have lower bone
27    lead levels in some studies (Hu et al., 2001;  Kamel et al., 2003); other toxicokinetic differences
28    have also been reported (Fleming et al., 1998; Hu et al., 2001;  Schwartz et al.,  1997; Smith et al.,
29    1995).  Overall, these data suggest that tighter binding of lead by the isozymes of the ALAD2
30    allele decreases lead sequestration in bone.
      December 2005                           6-155       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In contrast, data to determine whether the ALAD polymorphism impacts the renal toxicity
 2    of lead are still quite limited.  The only environmentally exposed population in which this has
 3    been addressed is the Normative Aging Study. Wu et al. (2003) (discussed in detail in section
 4    6.4.4.1.2 above) analyzed data to determine whether the ALAD genetic polymorphism modified
 5    associations between lead dose and uric acid, serum creatinine, and estimated creatinine
 6    clearance,  114 (16%) of the study group were either homozygous or heterozygous for the variant
 7    ALAD2 allele. None of the three outcomes were significantly different by genotype. However,
 8    effect  modification by genotype on the association between tibia lead and serum creatinine was
 9    observed; the P coefficient (and slope) was greater in the group with the variant allele ((3 = 0.002
10    [SE not provided]; p = 0.03).  Effect modification of borderline significance (p < 0.1) on
11    relations between of patella and tibia lead with uric acid was observed; this was significant in
12    participants whose patella lead levels were above 15 jig/g bone mineral ((3 = 0.016 [SE not
13    provided]; p = 0.04). Similar to the serum creatinine model, patella lead was associated with
14    higher uric acid in those with the variant allele.  Genotype did not modify lead associations in
15    models of estimated creatinine clearance.
16          The impact of the ALAD polymorphism on renal outcomes has been studied in four
17    occupationally-exposed populations to date. The two that assessed both associations and effect
18    modification by genotype are discussed here. Weaver et al. (2003b) analyzed data from 798 lead
19    workers. Lead and renal function measures, as well as mean lead levels, were described in
20    Weaver et al. (2003a) in Section 6.4.4.2 above.  A total of 79 (9.9%) participants were
21    heterozygous for the ALAD2 allele (none was homozygous). After adjustment, participants with
22    the ALAD2 allele had lower mean serum creatinine and higher calculated creatinine clearance.
23    Effect modification by ALAD on associations between blood lead and/or DMSA-chelatable lead
24    and three of six renal outcomes was observed. Among those with the ALAD12 genotype, higher
25    lead measures were associated with lower BUN and serum creatinine and higher calculated
26    creatinine clearance. Among older workers (age > median of 40.6 years), ALAD genotype
27    modified associations between lead dose and uric acid levels.  Higher lead dose was significantly
28    associated with higher uric acid in workers with the ALAD 11 genotype; associations were in the
29    opposite direction in participants with the variant ALAD12  genotype (Weaver et al., 2005c).
30          Ye and colleagues (2003) assessed effect modification by ALAD on associations between
31    blood  lead with urinary NAG and albumin in a study of 216 lead workers. Geometric mean

      December 2005                           6-156       DRAFT-DO NOT QUOTE OR CITE

-------
 1    blood lead was 37.8 |ig/dL in 14 workers with the ALAD12 genotype and 32.4 |ig/dL in workers
 2    with the ALAD11 genotype.  After adjustment for age, NAG was borderline statistically higher
 3    in those with the variant allele whose blood lead levels were >40 |ig/dL.  In all lead workers,
 4    after adjustment for age, gender, smoking, and alcohol ingestion, a statistically significant
 5    positive association between blood lead and creatinine adjusted NAG was observed in the
 6    workers with the ALAD12 genotype but not in lead workers with the ALAD11 genotype (the
 7    groups were analyzed separately rather than in an interaction model).
 8          Thus, two of the three  studies reported steeper slopes for one or more associations
 9    between lead dose and adverse renal function in participants with the ALAD2 allele compared to
10    those with the ALADl 1 genotype which suggests that the variant ALAD gene confers additional
11    risk for adverse renal outcomes in lead exposed populations. If the associations of Weaver et al.,
12    (2003b) represent lead-induced hyperfiltration their results could be consistent with increased
13    risk from the variant allele as well. Ultimately, analysis of longitudinal data in the Korean lead
14    worker population will be required to understand these complex relations.
15
16    6.4.7.3.2   BsmI Polymorphism of the Vitamin D Receptor (VDR) Gene
17          In contrast to ALAD, relatively few data on the BsmI polymorphism of the gene that
18    encodes for the vitamin D receptor (VDR) are available in lead exposed populations.
19    Polymorphisms of the VDR gene are of interest in these populations due to the role of vitamin D
20    and its receptor in regulating both intestinal calcium absorption and bone mineralization. These
21    pathways impact lead absorption from the gastrointestinal tract and may impact lead storage
22    and/or release from bone (Onalaja and Claudio, 2000). Analysis of data from the first evaluation
23    of the Korean lead worker cohort found that participants with the B  allele had significantly
24    higher blood, DMSA-chelatable, and tibia lead levels than those with the bb  genotype (Schwartz
25    et al., 2000a); significantly higher patella lead in workers with the B allele was reported in data
26    from the third evaluation (Theppeang et al., 2004). A study of 216 lead workers similarly
27    reported higher blood lead levels in workers with the B allele (n = 20), after adjustment for age,
28    gender, smoking, alcohol ingestion, and calcium ingestion, education, ALAD genotype, and
29    ambient lead exposure (Ye et al., 2003). In a study of 504 former organolead manufacturing
30    workers, with an average of almost two decades since last occupational exposure, VDR genotype
31    was not associated with tibia lead concentrations (Schwartz et al., 2000c). However, the slope

      December 2005                           6-157       DRAFT-DO NOT QUOTE OR CITE

-------
 1    of the positive association between age and tibia lead concentration was steeper in participants
 2    with the B allele compared to those with the bb genotype and tibia lead declined with years since
 3    last exposure in participants with the bb genotype, but increased in those with the B allele.
 4    In contrast, Chuang and colleagues (2004) found no difference in current or cumulative blood
 5    lead by Bsml polymorphism in 544 lead workers.
 6          Work in two of the populations described above has also provided information on the
 7    impact of the Bsml VDR polymorphism.  Of the 798 participants in Weaver et al. (2003b),
 8    89 (11.2%) had genotype Bb or BB. No significant differences were seen in renal outcomes by
 9    VDR genotype nor was consistent effect modification observed. However, those authors
10    assessed effect modification by this polymorphism on associations between patella lead and renal
11    outcomes in current and former Korean lead workers in data from the third evaluation where
12    patella lead was measured.  Results were compared to those with three other lead biomarkers.
13    The same six renal outcomes as in Weaver et al. (2003a) were measured. Mean blood, patella,
14    tibia, and DMSA-chelatable lead were 30.9 |ig/dL (SD 16.7), 75.1 jig lead/g bone mineral
15    (SD 101.1), 33.6 |ig lead/g bone mineral (SD 43.4), and 0.63 |ig lead/mg creatinine (SD 0.75),
16    respectively, in 647 lead workers (Weaver et al.  [2005b]). Little evidence of effect modification
17    by genotype on associations between patella lead and renal outcomes was observed. However,
18    the VDR polymorphism did modify associations between the other lead biomarkers and serum
19    creatinine and calculated creatinine clearance. Higher lead dose was associated with worse renal
20    function in participants with the variant B allele. Models in two groups, dichotomized by
21    median age, showed this effect was present in the younger half of the population.  The authors
22    were able to exclude different participant  subsets as an explanation for the difference in VDR
23    findings between the two evaluations.  Longitudinal changes in renal function between
24    evaluations may account for these findings and are currently being evaluated in longitudinal
25    data analysis.
26          Ye and colleagues (2003) reported  higher systolic blood pressure, after adjustment for
27    age, in those with the variant allele whose blood lead levels were >40 |ig/dL.  In all lead
28    workers, after adjustment for age, gender, smoking, and body mass index, a statistically
29    significant positive association between blood lead and systolic blood pressure was observed in
30    the 20 lead workers with the variant B allele, but not in lead workers with the bb genotype.
31    Again, the fact that the genotype groups were analyzed separately, rather than in an interaction

      December 2005                           6-158       DRAFT-DO NOT  QUOTE OR CITE

-------
 1    model, decreased the study's power to detect a difference. This could be an explanatory factor
 2    for the lack of effect modification by VDR genotype on associations between blood lead and
 3    urinary albumin and NAG observed.
 4          In conclusion, an increasing body of literature indicates that both of these polymorphisms
 5    affect lead toxicokinetics.  However, data to determine if these polymorphisms impact renal
 6    function are still quite limited.  Existing data are suggestive of an increased renal risk in lead
 7    exposed populations with the variant alleles of both polymorphisms.
 8
 9    6.4.8  Confounding of the Renal Effects of Lead by Other Potential
10           Risk Factors
11          Studies selected for discussion in Section 6.4 above have generally controlled for at least
12    the most basic risk factors known to affect renal function such as age, gender, and body mass
13    index (or weight and height separately). Some have controlled for many other potentially
14    important risk factors.  In addition, exposure to other nephrotoxicants must be considered.
15    Notably, although these are listed under confounders, some may be effect modifiers as well.
16
17    6.4.8.1    Cadmium
18          Similar to lead, cadmium is an ubiquitous nephrotoxi cant that accumulates in the body.
19    Environmental exposure in the United States occurs primarily through food and smoking
20    (Agency for Toxic Substances and Disease Registry, 1993).  Cadmium in food is a result of soil
21    pollution from a variety of human activities such as phosphate fertilizer use, industrial releases
22    from smelting, and fuel combustion.  An analysis of NHANES III data, collected in a
23    representative sample of the U.S. population from 1988-1994, indicates that mean urinary
24    cadmium is 0.48 |ig/g creatinine and 97.7% of the population has a level <2 |ig/g creatinine
25    (Paschal et al., 2000).  Also similar to lead, cadmium causes proximal tubule pathology and is a
26    risk factor for CRT
27          The existing data indicate that cadmium, at exposure levels common in the U.S.,
28    confounds associations between lead exposure and at least one renal outcome, NAG.  Roels et al.
29    (1994) reported higher mean NAG in their lead-exposed group; however, NAG was correlated
30    with urinary cadmium but not blood or tibia lead, despite the fact that mean urinary cadmium
31    was only 1.04 and 0.53 |ig/g creatinine in workers and controls, respectively. Cardenas et al.

      December 2005                           6-159       DRAFT-DO NOT QUOTE OR CITE

-------
 1    (1993) reported a similar finding. Bernard et al. (1995) found an association between urinary
 2    cadmium and the NAG-B isoenzyme (released with breakdown of proximal tubular cells) in
 3    49 cadmium workers and 20 age-matched controls. In multiple linear regression, urinary
 4    cadmium, but not lead, was associated with NAG-B, after adjustment for age.  The association
 5    was significant even in the 44 participants with levels <2 |ig/g creatinine.  However, NAG-A
 6    (released by exocytosis) was correlated with urinary lead (the only lead measure), but not
 7    cadmium. Roels et al. (1995)  reviewed data pertinent to the potential for cadmium confounding
 8    of associations between lead and NAG. In more recent work, Weaver et al. (2003a) measured
 9    urinary cadmium in a subset of 191 of the 803 workers in their study (mean urinary cadmium
10    was 1.1 |ig/g creatinine). Higher urinary cadmium levels were associated with higher NAG.
11    Of the lead measures obtained, only tibia lead was significantly associated with NAG in the
12    cadmium subset.  When urinary  cadmium and tibia lead were entered as covariates in the same
13    model, both remained associated with NAG (p < 0.05). However, in comparing the effects,
14    a 0.5 |ig/g creatinine increase  in cadmium had the same effect on NAG as a 66.9 jig/g bone
15    mineral increase in tibia lead.  When compared by ranges of exposure in this population,
16    environmental level cadmium dose had a larger impact on NAG than did occupational lead dose.
17          Cadmium exposure may confound relations between lead exposure and other renal
18    outcomes as well, although the data are too limited to draw firm conclusions. Positive
19    associations between urinary cadmium, which is thought to be the best measure of cumulative
20    cadmium exposure in the absence of cadmium-related renal damage, and low molecular weight
21    (LMW) proteinuria are well established in the occupational setting.  LMW proteinuria, most
22    commonly assessed by (32-microglobulin, is generally progressive at levels >1,500 |ig/g
23    creatinine in workers with  substantial body burdens (one or more historical urinary cadmium
24    >20 |ig/g creatinine) but may  also be progressive at lower levels (Roels et al., 1997; Bernard,
25    2004). More importantly, clinical renal function also declines as evidenced by decreasing GFR
26    in cadmium exposed workers  followed longitudinally after removal from exposure due to LMW
27    proteinuria (Roels et al., 1989; 1997).
28          In contrast to the clear  evidence that cadmium is a renal toxicant at occupational levels of
29    exposure, the renal risk from lower level cadmium exposure remains uncertain. Most studies of
30    environmental cadmium exposure are cross-sectional and have assessed EBE markers, rather
31    than clinical renal outcomes (Alfven et al., 2002; Jarup et al., 2000; Noonan et al., 2002; Olsson

      December 2005                           6-160       DRAFT-DO NOT QUOTE OR CITE

-------
 1    et al., 2002). The Cadmibel study, a general population study of exposed residents from both
 2    cadmium polluted and unpolluted areas (discussed in Section 6.4.4.1.1 above), found correlations
 3    between urinary cadmium and several urinary EBE markers (NAG, RBP, p2-microglobulin,
 4    calcium, and amino acids) (Buchet et al., 1990). In those models, after adjustment for urinary
 5    cadmium and other covariates, blood lead was significant in models of (32-microglobulin and
 6    amino acids but not NAG. However, in this same population, blood lead was inversely
 7    associated with creatinine clearance, whereas urinary and blood cadmium were not (Staessen
 8    et al., 1992). A 5 year follow-up was conducted to determine the significance of the EBE
 9    abnormalities (Hotz et al., 1999). In this study,  models of renal function (two dichotomized
10    outcomes:  a 20% decline in creatinine clearance and a 20% increase in albumin excretion) in
11    relation to quartiles of urinary cadmium and the EBE markers at baseline were analyzed by
12    likelihood ratios. Baseline variables did not predict adverse renal outcomes.  However, 25% of
13    the original population was lost to follow-up; available data indicated that their baseline renal
14    function was worse than those who participated in the follow-up study. This may have biased
15    the study towards the null.
16          Two recent publications suggest that low-level cadmium exposure is associated with
17    adverse clinical renal outcomes. Elevated urine cadmium levels were associated with prevalent
18    microalbuminuria and decreased calculated creatinine clearance after adjustment for age, sex,
19    race, smoking, and use of diuretics in an analysis of 16,094 participants in the NHANES III
20    study (Young et al., 2004). Hellstrom et al. (2001) reported increased rates of renal dialysis
21    and transplantation in residents of cadmium-polluted areas in Sweden. Compared to the
22    "no-exposure group" (domicile >10 km from a battery plant), age-standardized rate ratios were
23    1.4 (95% CI: 0.8, 2.0) in the low-exposure group (domicile 2 to 10 km) and 1.9 (95% CI:  1.3,
24    2.5) in the moderate-exposure group (domicile <2 km). Exposure categorization was based on
25    environmental monitoring in the study  areas. Cadmium dose was not directly measured although
26    occupationally  exposed participants were considered in a separate group.  Neither of these
27    studies assessed lead exposure as a covariate, which would be important given the Cadmibel
28    results (Staessen et al., 1992).
29          In conclusion, cadmium clearly  confounds associations between lead dose and
30    NAG.  Given the similarities in both nephrotoxicants, cadmium may confound and/or modify
31    associations between lead and other renal outcomes. However, data on the

      December 2005                          6-161        DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentration-response relation between environmental cadmium and the kidney are too limited
 2    to assess the potential for this at present. Future studies assessing both lead and cadmium are
 3    needed.
 4
 5    6.4.9   Summary of the Epidemiologic Evidence for the Renal Effects of Lead
 6          In the last two decades, the quality of research on the renal impact of lead exposure has
 7    advanced dramatically.  As a result, a much more accurate assessment of the adverse renal
 8    impact of lead exposure can now be made.  Studies of environmentally-exposed general
 9    populations are one of the most important advances in this regard.  The landmark Cadmibel
10    study (Staessen et al., 1992) was the first to raise concern that the lead dose threshold for adverse
11    renal effects in the general population was much lower than previously appreciated based on
12    occupational data.  Research in the Normative Aging Study population reached similar
13    conclusions and suggested that both cumulative and circulating lead are associated with
14    longitudinal decline in renal function. Diabetics were a particularly susceptible risk group in this
15    regard. The NHANES III data analysis (Muntner et al., 2003) are notable for a sample size that
16    is, by far, the largest of the environmental studies, comprehensive adjustment for other renal risk
17    factors and the fact that population is representative of the U.S. population.  Thus, the fact that
18    renal dysfunction was observed in hypertensives at a mean blood lead of only 4.2 |ig/dL and in
19    the 1st quartile compared to the reference group (blood lead range from 2.5 to 3.8 jig/dL),
20    provides strong evidence that the kidney is a target organ for adverse effects from lead  at current
21    U.S. environmental exposure levels.
22          Studies involving the longitudinal assessment of renal function decline in susceptible
23    patient populations in relation to baseline chelatable lead body burden and therapeutic chelation
24    constitute the other major advance in lead-renal research in the last two decades. Chelation was
25    beneficial in an EDTA-chelatable lead level range from 80 or 150 to 600 jig/72 h, depending on
26    the study (Lin et al.,  1999, 2003). These studies suggested that lead body burden, at much lower
27    levels than previously recognized, contributes to renal dysfunction in populations with  CRI from
28    a range of causes.  This work also suggests that renal function in patients with CRI stabilizes
29    and, in some cases, improves after therapeutic EDTA chelation of lead levels well below the
30    level currently thought to require chelation.
      December 2005                           6-162       DRAFT-DO NOT QUOTE OR CITE

-------
 1          A finding of note from the occupational studies is the observation of inverse associations
 2    (higher lead dose with lower BUN, serum creatinine, and/or higher creatinine clearance) in
 3    several studies. This may indicate lead-related hyperfiltration and have mechanistic
 4    implications. Regardless, significant associations could be obscured if opposite direction
 5    associations are present in different segments of the study population and interaction models to
 6    address this are not performed.  This is a valid concern, since the settings in which these inverse
 7    associations are most likely are not well defined.
 8          The renal impact in children from lead exposure at current environmental levels is
 9    difficult to assess, since the studies have involved measurement of EBE markers and their
10    prognostic value is uncertain. Susceptible populations due to chronic medical diseases have
11    clearly been established; risk from genetic polymorphisms may be important, but further study is
12    required.  Studies of potential mechanisms for the adverse renal effects of lead in humans, such
13    as via uric acid have more than just theoretical importance, since EDTA chelation has been
14    reported to improve both renal function and urate clearance in patients with renal insufficiency
15    and gout (Lin et al., 2001). With an improved understanding of mechanisms, clinically relevant
16    therapies may be possible.
17
18
19    6.5   CARDIOVASCULAR EFFECTS OF LEAD
20    6.5.1   Summary of Key Findings of the Cardiovascular Effects of Lead from
21           the 1985 Lead AQCD and Addendum, and 1990 Supplement
22          The greater  part of the evidence reviewed up to 1990 included analyses of the largest
23    datasets available at the time, the National Health and Nutrition Evaluation Survey II (NHANES
24    II), studying the U.S. population between 1976 and 1980, and the British Regional Heart Study
25    (BRHS), studying men aged 40-59 years from 24 British towns.  Analyses of the Welsh Heart
26    Programme, a regional Welsh study, and the Caerphilly Collaborative Heart Disease Study,  a
27    cohort study of men aged 45-59 years living in one town in Wales, as well as smaller population
28    and occupational exposure studies in the U.S., Canada, and Europe provided supporting
29    evidence. These studies set enduring design and analysis standards by example for evaluating
30    cardiovascular effects  associated with blood lead levels in samples from diverse populations.
     December 2005                           6-163       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In general, the reviewed studies used multiple linear regression modeling of blood
 2    pressure and multiple logistic regression modeling of hypertension, cardiovascular mortality, and
 3    other cardiovascular disease, allowing adjustment of the blood lead effect on outcome by other
 4    factors known or suspected to be related to the exposure and outcome under study. The most
 5    commonly considered potential confounding factors were age, body mass index (BMI), alcohol
 6    use, and cigarette smoking.
 7          These studies were almost exclusively cross-sectional studies, measuring cardiovascular
 8    outcome, blood lead, and control variables once, though one Canadian occupational study and
 9    one Danish birth-year cohort study used a longitudinal design. Studies sometimes presented
10    analyses stratified by sex or age, by both sex and age, or by "race." Other analyses only reported
11    results for one particular stratum. Separate analyses of datasets partitioned by stratified variables
12    always reduce sample size available for statistical models, and, thereby, may reduce power to
13    detect real  effects.
14          Evaluated as a whole, the blood pressure studies supported a small but significant
15    association between increasing blood lead concentrations and increasing blood pressure in study
16    groups.  The effect was more consistent across studies in middle-aged men than in other groups,
17    ranging from a 1.5 to 3.0 mm Hg increase in systolic blood pressure for each doubling of blood
18    lead from the mean blood lead level, and from a 1.0 to 2.0 mm Hg increase in diastolic blood
19    pressure for each blood lead doubling, across a wide range of blood lead concentration down to
20    7 |ig/dL. Most studies using multiple regression analyses stratified by sex were unable to find
21    significant associations between blood pressure and blood lead in females, though one reanalysis
22    of the NHANES II dataset did report a statistically significant relationship between diastolic
23    blood pressure and lead in women aged 20 to 74 years. In studies reporting the use of different
24    blood lead-blood pressure concentration-response relationships, log blood lead terms had lower
25    probability values than linear blood lead terms, suggesting that increases in blood pressure with
26    incremental fixed incremental increases in blood lead might be greater at lower blood lead
27    concentrations than at higher concentrations.
28          Three studies of groups with occupational exposure reported mixed results. One study
29    found significant excess mortality due to cardiovascular disease during the period 1946-1965 in a
30    case-control  study in the United Kingdom, but not 1966-1985. A study of U.S. battery and lead
31    production workers from 1947-1980 found significant excess mortality due to "other

      December 2005                           6-164        DRAFT-DO NOT QUOTE OR CITE

-------
 1    hypertensive disease" (codes 444-447 in the ICD 1955 classification system), but not due to
 2    hypertensive diseases outside those classifications.  No excess mortality due to hypertension was
 3    found in a study of U.S. smelter workers between 1940 and 1965.
 4          The BRHS study failed to reveal significant associations between blood lead and ischemic
 5    heart disease and stroke. However, electrocardiogram abnormalities associated with left
 6    ventricular hypertrophy were found related to blood lead in a subset of the NHANES II data,
 7    confirming an earlier study finding significant associations between ischemic changes and blood
 8    lead in lead workers.
 9         Noninvasive measurement of bone lead concentration using XRF techniques was still
10    maturing during the literature review period covered by the 1986 AQCD document and later
11    addendum and supplement. No studies were reported using bone lead as a marker  for lead
12    exposure.
13
14    6.5.2   Effects of Lead on Blood Pressure and Hypertension
15    6.5.2.1    Introduction
16         Blood lead concentration remained the most widely used exposure index in  blood
17    pressure/hypertension epidemiologic studies from 1990 to present. Obtaining the sample is
18    relatively noninvasive and quick, measurement techniques are well standardized and inexpensive
19    with access to external  quality assurance programs worldwide, and existing regulation and
20    medical decision-making are based on blood lead levels.  If exogenous lead exposure were the
21    only determinant for blood lead concentration, it could be fair to state that a single  blood lead
22    measurement represented exposure to lead during the 30-90 day period preceding the
23    measurement. However, blood lead concentration represents a combination of recent exposure
24    to external sources and the influence of internal  sources, principally bone  lead. As detailed in
25    Section 6.2, bone is a long-term storage depot for much of the lead absorbed by the body from
26    external  sources, and by weight can represent over 95% of the total body burden of lead in
27    middle-aged persons, especially if external exposures are currently low. Bone lead has residence
28    times of years to decades.  Bones constantly absorb lead from and release lead to the circulatory
29    system.  Consequently, blood lead concentration is not only determined by current  and recent
30    past external exposure but is also influenced by  existing bone lead concentration to a degree
31    determined by current external exposure, accumulated past exposure stored in bones, and the

      December 2005                           6-165       DRAFT-DO NOT QUOTE OR CITE

-------
 1    physiological state of the bones due to aging, disease, pregnancy, and lactation, among others.
 2    Studies using only blood lead concentration, as an exposure index cannot determine the relative
 3    contributions of current exogenous exposure and endogenous exposure to blood lead.  Thus, they
 4    are unable to assess what part of measured blood lead effect on the circulatory system is due to
 5    possibly higher long duration past exposure and what part is due to the possibly immediate toxic
 6    effects of currently circulating lead. They are, instead, assessing a combined effect of past and
 7    present exposure in a proportion that will differ among subjects according to their past and
 8    present exposure, health history, and age.
 9          Elevated blood pressure can be evaluated as a continuous measure (mm Hg) or as a
10    dichotomized measure (hypertension). The definition of hypertension involves a categorical
11    cut point of mm Hg above which one is hypertensive and below normotensive.  Kannel
12    (2000a,b) notes that this number has dropped over time for systolic/diastolic pressure and further
13    notes a continuous graded influence of blood pressure even within what is regarded as the
14    normotensive range. Some concern for an arbitrary definition as the cut point and  one that has
15    changed over time is a consideration. However defined, even if greater than the cut point at that
16    time used clinically, the separation into these two groups offers  a different perspective than
17    blood pressure per se.  Hypertension has a different clinical relevance than blood pressure
18    changes themselves. The disease condition as an outcome and a change in mm Hg in relation to
19    exposure both offer the opportunity for insight into the clinical relevance of the relationships.
20    Biomarkers like bone lead and blood lead also add input into the acute/chronic nature.
21          The recently developed in vivo technique of XRF measurement of bone lead
22    concentration has been used in a handful of studies to better assess the role of past  exposure to
23    lead on blood pressure and hypertension in essentially cross-sectional studies. Bone lead
24    concentration provides a record of cumulative past exposure due to the long residence times of
25    lead in bones, though the specific temporal pattern of past exposure cannot be readily determined
26    from the measurement.  Primarily cortical bones such as tibia have residence times measured in
27    decades, whereas primarily trabecular bones such as calcaneus and patella have residence times
28    measured in years to decades, reflecting different metabolic rates of the two bone types.  As there
29    is continual interchange of lead in bone and lead in blood, studies  combining the measurement
30    and modeling of both bone lead and blood lead have the best chance of dissecting out the roles of
31    past and present lead exposure on blood pressure and hypertension.

      December 2005                            6-166       DRAFT-DO NOT QUOTE OR CITE

-------
 1          The growing field of toxicogenetics now includes lead exposure epidemiology.  The
 2    several studies combining subject evaluation of polymorphisms of genes thought to play a role in
 3    either the origin of cardiovascular disease, the toxicokinetics of lead or both are also reviewed.
 4    All epidemiologic studies of the cardiovascular effects of lead reviewed in this section as well as
 5    other additional studies are  further summarized in Annex Table 6-5.1.
 6
 7    6.5.2.2   Blood Pressure and Hypertension Studies Using Blood Lead as Exposure Index
 8    6.5.2.2.1   NHANES Studies
 9          NHANES contributed the largest datasets analyzed in this review.  As the  surveys are also
10    representative of the U.S. population, their results may be more readily applied to the general
11    U.S. population than smaller cohort or occupational studies. The several papers using this
12    dataset sometimes come to different conclusions, depending on the statistical techniques used in
13    analysis, including logarithmic or linear specification of the lead variable, stratification of
14    analyses according to sex or ethnic groups or use of interaction terms to define these groups, use
15    of survey-design corrected models, choice of covariates in the models, and different age ranges
16    analyzed.
17
18    NHANES II (1976-1980)
19          In one NHANES Il-based study, males and females (number unreported but less than
20    9,000 combined) aged 20 to 74 years were studied with separate stepwise multiple regression
21    models adjusted for sampling design (Schwartz, 1991).  Mean blood lead levels were not
22    reported.  Covariates common to both male and female models were age and age2, BMI, race,
23    family history, cholesterol, zinc, tricep fold, and natural log lead.  Models for men also included
24    height and cigarette smoking. Natural log blood lead was significantly associated with diastolic
25    blood pressure (systolic not reported) in males, with a 2.03 mm Hg diastolic (95% CI:  0.67,
26    3.39) increase for every doubling of blood lead, and for females a 1.14 mm Hg increase (95% CI:
27    0.13, 2.08). Interactions between blood lead and sex and between blood lead and race in a
28    combined model were insignificant (not shown). The conclusion from these interaction terms is
29    that the association between blood lead and diastolic blood pressure was not significantly
30    different between men and women or between races. There was no mention of consideration of
      December 2005                           6-167       DRAFT-DO NOT QUOTE OR CITE

-------
 1    model diagnostics, and the stepwise modeling may incorrectly include or exclude potentially
 2    confounding variables.
 3          The other NHANES II-based study focused on black-white differences in blood pressure
 4    related to blood lead (Sorel et al., 1991).  There were 473 blacks and 3,627 whites in the study,
 5    each nearly evenly divided by sex, aged 18 to 74 years.  As is usual in U.S.-based studies,
 6    race/ethnicity was based on self-report.  Survey design-adjusted multiple regression models were
 7    stratified on sex and included age, BMI, and linear blood lead as covariates. The effect of race
 8    and poverty index was assessed by including their terms in models with and without blood lead
 9    and determining change in race or poverty coefficients by comparing confidence intervals. Each
10    1 |ig/dL increase in linear blood lead significantly predicted increased systolic blood pressure for
11    both males (2.23 mm Hg [95% CI: 0.69, 3.61]), but not females (0.98 mm Hg [95% CI:  -9.78,
12    3.06]) for each doubling of blood lead. The differences in black and white (race variable) blood
13    pressure coefficients did not significantly change when lead was in or out of the model, either for
14    subjects below the poverty index or above the poverty index.  Race does not appear to
15    significantly modify the relationship between blood lead and systolic blood pressure. The paper
16    reported no model diagnostics.  There were reporting inconsistencies in the female-stratified
17    models, in which the coefficients and 95% CI did not correspond.
18
19    NHANES III (1988-1994)
20          A study using the NHANES III dataset from all adults 20 years of age and up examined
21    the effect of natural log blood lead on systolic and diastolic blood pressure (Den Hond  et al.,
22    2002). Multiple regression analyses for each blood pressure measurement were stratified by sex
23    and race, yielding four models for each blood pressure measurement. The mean blood levels
24    were 3.6 |ig/dL in white males (n = 4,685), 2.1 |ig/dL in white females (n = 5,138), 4.2 |ig/dL in
25    black males (n = 1,761), and 2.3 |ig/dL in black females (n = 2, 197). One group of covariates
26    (age, age-squared, BMI, hematocrit, smoking, alcohol consumption, and an indicator variable for
27    use of antihypertensive medications) were first entered as a block regardless of significance in
28    each model, then another group of variables (coffee consumption, dietary calcium, dietary
29    sodium/potassium ratio, total serum protein, total serum calcium, diabetes, and poverty index)
30    was entered stepwise in the model without lead and the variable retained only if it was
31    statistically significant (p < 0.05).  Then log-transformed blood lead was forced into each model.

      December 2005                           6-168        DRAFT-DO NOT QUOTE OR CITE

-------
 1    The model building procedure resulted in eight distinct models, each with their own unique mix
 2    of covariates. No model diagnostics were reported, nor was adjustment of results by survey
 3    sample weights and design.  Only blacks had significant lead-systolic blood pressure
 4    associations; each doubling in blood lead was associated with a 0.90 mm Hg (95% CI: 0.04, 1.8)
 5    and 1.20 mm Hg (95% CI: 0.4, 2.0) increase in males and females respectively. The association
 6    of lead-diastolic blood pressure was also significant for black females (0.50 mm Hg [95% CI:
 7    0.01,1.1]).  Interestingly, increasing blood lead was associated with significantly decreased
 8    diastolic blood pressure in white males (-0.6 mm Hg [95% CI: -0.9, -0.3]).  The authors did
 9    not comment on their finding that the significant total serum calcium covariate in these two
10    groups had opposite signs too (white male serum calcium P = 6.50 mm Hg/mmol/L, black
11    female serum calcium P = -5.58 mm Hg/mmol/L). Though the authors offered no formal test of
12    the difference between the two serum calcium coefficients, since both were significantly
13    different than the null hypothesis coefficient of 0 and different in sign, it could be concluded that
14    those coefficients were significantly different between the two groups.  As the authors do not
15    present the serum calcium coefficients before forcing lead into the models, it is not certain that
16    blood lead in the model was associated with the significant sign difference of the calcium
17    coefficients or if the calcium coefficients had opposite signs between the two groups without
18    lead in the model. As each model had a different set of covariates, the presence or absence of
19    one of the other covariates could have produced the same results.  Nevertheless, this pattern of
20    results may indicate significant confounding between serum calcium and blood lead associations
21    with blood pressure. Though the study suggested differences between blacks  and whites in
22    response to  lead, no statistical tests were performed of differences in lead coefficients based on
23    race. In addition, the black-white effect differences associated with blood lead may be due to
24    possible confounding in some or all of the models.
25          Limiting the study sample from NHANES III to women aged 40 to 59 years, another
26    group of researchers addressed the relationship between blood lead and both blood pressure
27    (n = 1,786)  and hypertension (n = 2,165) (Nash et al., 2003).  Blood pressure models excluded
28    women who reported being under treatment for hypertension.  Separate blood pressure multiple
29    regression models were presented for diastolic and systolic blood pressure, each with and
30    without stratification for dichotomous premenopausal/postmenopausal status.  One block of
31    covariates was  entered without regard to statistical significance (age, race/ethnicity, BMI, and

      December 2005                           6-169        DRAFT-DO NOT QUOTE OR CITE

-------
 1    serum creatinine). Another block of covariates (education, poverty income ratio, alcohol use,
 2    and cigarette smoking status) was entered second but only retained if variables were significantly
 3    associated with blood pressure. Finally, linear blood lead was forced in last.  Logistic regression
 4    models for hypertension used the same covariate entry scheme with and without stratification on
 5    the menopause variable, but using a blood lead quartile exposure variable. Despite the stated
 6    procedure for covariate selection, all models used the same set of covariates: linear (or quartile)
 7    lead, age, race/ethnicity, alcohol use, cigarette smoking status, BMI, and serum creatinine.
 8    All models were adjusted for survey weights and design.  Linear lead was significantly
 9    associated with systolic blood pressure only in the entire study sample; each 1 |ig/dL increase in
10    blood lead was associated with a 0.32 mm Hg (95% CI:  0.01, 0.63) increase in blood pressure.
11    No associations were observed in the menopause-stratified analyses. Linear lead also was
12    significantly associated with diastolic blood pressure in the entire study sample (0.25 mm Hg
13    [95% CI:  0.07, 0.43]).  Odd ratios of diastolic hypertension (>90 mm Hg) in logistic regression
14    models was significantly related to blood lead with an odds ratio of 4.26 (95% CI:  1.36,  12.99)
15    comparing the 1st quartile blood lead group (0.5-1.6 |ig/dL) to the 4th quartile blood lead group
16    (4.0-31.1 |ig/dL) in all women not taking antihypertensive medications. Further stratification
17    produced occasional significant odds ratios for either diastolic or systolic hypertension.  There
18    were some differences in table and text reporting of results and an inconsistency between the SE
19    and the p-values.
20          Another study using the NHANES III database was notable for its formal testing of race
21    and sex differences in lead effect by interactions terms (Vupputuri et al., 2003). The study used
22    5,360 white men (mean blood lead 4.4 jig/dL), 2,104 black men (mean blood lead 5.4 jig/dL),
23    5,188 white women (mean blood lead 3.0 jig/dL), and 2,300 black women (mean blood lead
24    3.4 |ig/dL). Multiple linear and logistic regression models of blood pressure and hypertension
25    (systolic > 140 mm Hg, diastolic >90 mm Hg, and/or taking antihypertensive medication),
26    respectively, were adjusted for age, high school education, BMI, alcohol, leisure-time physical
27    activity, and dietary intake of sodium, potassium, and total energy.  The models used linear lead,
28    except for one set of hypertension models with a cut point for "high" lead exposure at >5 |ig/dL.
29    Subjects taking antihypertensive medication (n = 2,496) were not included in linear regression
30    models of blood pressure. Neither age nor blood lead range were reported, nor  was the technique
31    of selecting and entering covariates in multiple regression models.  Only coefficients for linear

      December 2005                            6-170        DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead effect for each model were reported.  Significant interactions in multivariate models were
 2    found between lead and race and between lead and sex, though these analyses were not shown.
 3    Only black men and women had significant linear lead-blood pressure effects in adjusted systolic
 4    (0.25 mm Hg [95% CI:  0.06, 0.44] for black men and 0.47 mm Hg [95% CI: 0.14, 0.80] for
 5    black women with each 1 |ig/dL increase in blood lead) and diastolic blood pressure
 6    (0.19mmHg[95%CI:  0.02, 0.36] for black men and 0.32 mm Hg [95% CI: 0.11, 0.54] for
 7    black women). Linear blood lead association with hypertension was significant only in women.
 8    The odds ratios were 1.09 (95% CI:  1.04, 1.13) for white women and 1.10 (95% CI:  1.06, 1.16)
 9    for black women for each 1 |ig/dL increase in blood lead.  The authors presented insufficient
10    detail to evaluate this pattern of results. The use of diagnostic testing was not mentioned.
11
12    European Population Studies
13          The Health Survey for England 1995 examined a representative sample of the English
14    population living in private households and provided up to 2,563 men and 2,763 women with a
15    mean age of 47.6 years in a study of blood lead-blood pressure relationships (Bost et al., 1999).
16    Precise blood lead range was not given but was at least from less than 1.5  |ig/dL to greater than
17    8.5 |ig/dL. The  study used stepwise multiple regressions modeling of diastolic and systolic
18    blood pressure stratified by sex, with and without adjustment for alcohol, and with and without
19    subjects on antihypertensive medications. Candidate covariates,  selected from a larger pool,
20    included age, alcohol use (heavy drinkers versus all other drinkers and nondrinkers), SES
21    (manual classes  versus non-manual classes), location of residence in country (northern resident
22    versus non-northern resident), smoking, and common log blood lead. As nonsignificant
23    variables did not remain in the models, each model contained a unique mix of covariates.
24    A doubling in blood lead in men was associated with an increase in diastolic blood pressure of
25    1.07 mm Hg (95% CI:  0.37, 1.78) when alcohol consumption was not in the model and
26    0.88 mm Hg (95% CI:  0.13, 1.63) when alcohol consumption was in the model. Women had a
27    significant response to lead only for diastolic blood pressure in the model without adjustment for
28    alcohol and with subjects using  antihypertensive medication. There were no significant effects
29    of lead on  systolic blood pressure in any model.  The authors provided no statistical justification
30    for stratified modeling nor did they test for significant differences in lead coefficients as a result
31    of the stratification.

      December 2005                            6-171       DRAFT-DO NOT QUOTE OR CITE

-------
 1    U.S. Cohort Studies
 2          The Boston-based Normative Aging Study, part of a longitudinal study of male veterans,
 3    examined the effects of blood lead on blood pressure in 798 men, aged 45-93 years old, with
 4    blood lead between 0.5 and 35.0 |ig/dL (Proctor et al., 1996). Using multiple regression
 5    modeling with forced entry of natural log lead and other covariates (age, age2, BMI, dietary
 6    calcium, exercise, smoking, alcohol, heart rate, and hematocrit), the authors found a significant
 7    increase of only diastolic blood pressure (0.83 mm Hg [95% CI: 0.08, 1.52]) for each doubling
 8    of blood lead. Though the relationship between blood lead and systolic blood pressure was
 9    positive, it was not significant.  Nearly half the blood lead measures were derived from frozen
10    red blood cells collected previously (up to several years earlier) and corrected for hematocrit
11    determined at the time blood  pressure was measured. Possible errors in correction of these
12    samples and the non-contemporaneous nature of the resulting blood lead concentrations may
13    have compromised the results.
14          Cheng et al. (2001), using the same Normative Aging Study data and stepwise multiple
15    regression, found a near-zero association between systolic blood pressure and linear blood lead
16    (-0.03 mm Hg for each |ig/dL increase in blood lead) in 519 men aged 48 to 93.  The subjects
17    selected for this analysis were all free of hypertension (systolic>160 mm Hg or diastolic>95
18    mm Hg). Differences in subject selection procedures and modeling techniques may have
19    accounted in the different results between Cheng et al. and Proctor et al. They also reported on
20    incidence of hypertension developing between  1991 and 1997 using Cox proportional hazards
21    models.  Controlling for age,  age2, BMI, and family history of hypertension, linear blood lead
22    was not significantly associated with risk of developing hypertension (systolic >140 mm Hg or
23    diastolic >90 mm Hg) in normotensives at the start of the period (rate  ratio of 0.98 [95% CI:
24    0.91, 1.06]) for each 1 |ig/dL increase in blood lead.
25          Gerr et al. (2002) similarly reported near-zero linear blood lead effects on blood pressure
26    on a combined group of 19-29 year old males and females (n =  502), half of whom  had lived
27    around active lead smelters as children, using forced entry of all covariates. Among the
28    covariates forced into the model was tibia lead  concentration, expected to be significantly
29    correlated with blood lead. This may have reduced or confounded the effects of blood lead.
30          Korrick et al. (1999) examined linear and natural log blood lead effect on hypertension,
31    defined as self-reported or physician diagnosis  of hypertension or systolic or diastolic

      December 2005                           6-172       DRAFT-DO NOT QUOTE OR CITE

-------
 1    > 140/90 mm Hg, in 284 middle-aged women from the Nurse Health Study based in Boston.
 2    The association of hypertension and blood lead was nonsignificant.
 3          Rothenberg et al. (1999) tested a group of 1,527 women, aged 15 to 42 years, in their third
 4    trimester of pregnancy, with blood lead ranging from 0.5 to 40.4 |ig/dL.  They stratified testing
 5    into immigrant (n = 1,188) and nonimmigrant (n = 439) groups. They used forced entry of all
 6    covariates in multiple regression models, including natural log lead, age, BMI, coffee, iron
 7    supplement, and job stress, and found lead-related significant increases in systolic (1.18 mm Hg
 8    [95% CI: 0.45, 1.91] for each doubling of blood lead) and diastolic (1.02 mm Hg [95% CI:
 9    0.37, 1.34]) blood pressure only in immigrants. The small size of the nonimmigrant group may
10    have reduced power to detect significant effects. In a follow-up of 668 women returning for
11    postpartum testing (Rothenberg,  et al., 2002), using multiple regression models with forced entry
12    of natural log blood lead, tibia and calcaneus lead, age, BMI, parity, smoking, immigrant status,
13    and education,  the authors found a significant  decreases in systolic (-1.05 mm Hg [95% CI:
14    -1.96, -0.14])  and diastolic  (-1.16 mm Hg [95% CI:  -1.98, -0.35]) blood pressure associated
15    with doubling in blood lead in the postpartum  women.  This subgroup of women had no
16    significant blood lead effects in the third trimester.  Although the covariate pattern was different
17    from the larger prenatal  study (Rothenberg et al., 1999), thorough testing of possible
18    confounding, especially with the bone lead measures, revealed no significant change in blood
19    lead effects.  This study finding is similar to that reported by Den Hond et al. (2002) for
20    white males. No significant effect of blood lead on prenatal or postpartum hypertension
21    (> 140/90 mm Hg) was found.
22          Morris et al. (1990) recruited a group of 105 women and 145 men, aged 18-80 years, from
23    a clinic specializing in non-drug  hypertension  treatment. Blood lead ranged from 5-40.5 |ig/dL.
24    Multiple regression was performed with forced entry of natural log lead,  age, BMI, dietary
25    calcium, "other nutrients," serum ionized calcium, and erythrocyte protoporphyrin.  Only men
26    were found to have lead-related significant increases in systolic (3.17 mm Hg [95% CI:  -2.13,
27    8.48] for each doubling of blood  lead) and diastolic (1.32 mm Hg [95% CI: -2.12, 4.75]) blood
28    pressure. Small study size limits conclusions based on nonsignificant findings in women.
29    Dietary calcium is associated with reduced blood lead in many studies and could be considered a
30    confounder with blood lead.  Erythrocyte protoporphyrin is a biomarker of lead exposure and
31    correlates with  blood lead over at least part of the blood range in study subjects.  There was the

      December 2005                           6-173        DRAFT-DO NOT QUOTE OR CITE

-------
 1    inclusion of at least two collinear variables, a high proportion of covariates to subjects, and
 2    possible subject selection bias.
 O
 4    European Cohort Studies
 5          The Glostrup Population Study (Copenhagen) studied 1,009 men and women (all born in
 6    1936) longitudinally from 1976 to 1987 (M011er and Kristensen, 1992). Blood lead ranged from
 7    2 to 62 |ig/dL, depending on the year and sex stratum studied, with mean concentration dropping
 8    by -40% over the study period. They used multiple regression with forced entry of natural log
 9    lead, BMI, tobacco use, and physical activity.  Strongest associations between a doubling of
10    blood lead and blood pressure were found early in the study period. In 1976, a doubling of blood
11    lead was associated with 3.42 mm Hg (95% CI:  1.25, 5.58) increase in systolic blood pressure
12    and 2.95 mm Hg (95% CI:  1.08, 4.83) increase in diastolic blood pressure in women. For men
13    in 1981, a doubling of blood lead was associated with an increase of 1.89 mm Hg (95% CI:
14    0.00, 3.78) in systolic blood pressure and 1.14 mm Hg (95% CI: -0.37, 2.65) in diastolic blood
15    pressure.  No formal longitudinal analyses were performed, only analyses stratified by year and
16    sex and analyses relating change in lead and other covariates to change in blood pressure from
17    one study period to the next. As the relative risk of mortality was associated with increasing
18    blood lead over the study period (see below), the general reduction in lead-associated blood
19    pressure increase over the study period may have been masked by  lead-associated mortality.
20          The Europe New Risk Factor Project in Rome collected data from 1,319 males aged
21    55-75 years with blood lead between 4.0 and 44.2 |ig/dL (Menditto et al., 1994). They reported
22    significantly increased systolic (4.71 mm Hg [95% CI: 2.81, 6.61]) and diastolic (1.25 mm Hg
23    [95% CI: 0.33, 2.16]) blood pressure associated with a doubling of blood lead.
24          The Cadmibel studies from Belgium specifically selected part of their study group from
25    those living near nonferrous smelters. Staessen et al. (1993) reported on 827 men and
26    821 women, aged 20 to 88 years, with blood lead ranging from 2.7 to 84.9 |ig/dL for men and
27    1.3 to 42.4 |ig/dL for women. They forced natural log blood lead into stepwise multiple
28    regression models stratified by sex.  Covariates available for selection were age, age2, BMI,
29    pulse rate, log gamma-glutamlytranspeptidase, serum total calcium, log serum creatinine, urinary
30    potassium, smoking, alcohol, contraceptive use, and menopause. Near-zero nonsignificant
31    relationships were found between blood lead and  blood pressure for systolic blood pressure for

      December 2005                            6-174       DRAFT-DO NOT QUOTE OR CITE

-------
 1    women and diastolic blood pressure for men and women.  They reported a significant decrease in
 2    men's systolic blood pressure with increasing blood lead (-1.1 mm Hg for a doubling of blood
 3    lead), similar to the relationship found by Den Hond et al. (2002) for white men and by
 4    Rothenberg et al. (2002) for postpartum women. Stepwise regression results in different
 5    covariate patterns for each stratum and capitalizes on chance significance due to repeated testing.
 6          A follow-up of the Cadmibel study, the PheeCad study, evaluated 359 men and
 7    369 women, aged 20 to 82 years (Staessen et al., 1996). Fifty-nine percent of the men had
 8    occupational exposure. They were measured two times, at baseline and at follow-up about
 9    5 years later. Women's mean blood lead at baseline and follow-up was 6.6 |ig/dL (range
10    3.3-24.50 and 4.8 |ig/dL (range 1.7-11.8). Men's mean blood lead at baseline and follow-up was
11    11.4 |ig/dL (range 5.6-28.8) and 7.7 |ig/dL (range 3.7-20.1). Multiple regression models were
12    stratified on sex and in women on menopausal status.  Time-integrated blood pressure
13    measurements were used.  Each doubling of blood log lead was significantly associated with a
14    5.19 mm Hg (95% CI:  1.05, 9.34) increase in diastolic blood pressure in  187 pre- and
15    perimenopausal women. None of the other strata showed significant blood lead-related effects.
16    Using 24-h ambulatory blood pressure readings during the follow-up showed significant
17    associations between natural log blood lead and diastolic blood pressure in the group of all
18    345 women (2.42 mm Hg [95% CI: 0.00, 4.84]).  There were no significant lead effects on
19    systolic blood pressure in women or all blood pressure in men. Change in blood pressure and
20    change in covariates between baseline and follow-up were used to assess the effect of change of
21    blood lead in longitudinal analyses, similar to M011er and Kristensen (1992) above.
22    No significant effects of change in blood lead on change in blood pressure were found. Due to
23    stratification and resulting small groups, there may have been reduced power to  detect significant
24    effects of lead.
25
26    U.S. Occupational Studies
27          Glenn et al. (2003) was one of the few studies to use a prospective design and was the
28    only study using statistical techniques designed for repeated measures.  They studied 496 male
29    workers from New Jersey with former organolead exposure. Using generalized  estimating
30    equations with  baseline linear blood lead, age, BMI, antihypertensive medication, smoking,
31    education, measurement technician, and number of years to follow-up measurement of blood

      December 2005                          6-175       DRAFT-DO NOT QUOTE OR CITE

-------
 1    pressure (range 10 months-3.5 years), they found that every 1 |ig/dL increase in baseline blood
 2    lead was associated with 1.13 mm Hg/year (95% CI:  0.25, 2.02) increase in blood pressure over
 3    the observation period.
 4          Schwartz et al. (2000b) reported significant blood lead associations with 543 male former
 5    organolead workers. Stepwise backward multiple regression showed an increase of 2.3 mm Hg
 6    in systolic blood pressure for each doubling in blood lead. The association with diastolic blood
 7    pressure was not significant.
 8          Sharp et al. (1990) studied 132 black bus drivers (blood lead range 3.1-20.9 |ig/dL) and
 9    117 nonblack bus drivers (blood lead range 2.0-14.7 |ig/dL) in San Francisco, aged 30 to
10    60 years. They used natural log blood lead in multiple regression models and found for each
11    doubling of blood lead an increase of 5.22 mm Hg (95% CI:  0.60, 9.84) in systolic blood
12    pressure among blacks, 3.27 mm Hg (95% CI:  0.10, 6.44) in diastolic blood pressure among
13    blacks, and -3.96 mm Hg (95% CI:  -8.32, 0.42) in systolic blood pressure among nonblacks.
14          Sokas et al. (1997) reported a possible race interaction (p = 0.09) on systolic blood
15    pressure with linear blood lead in 264 construction workers aged 18-79 years. Each 1 |ig/dL
16    increase in blood lead increased systolic blood pressure in blacks by 0.86  mm Hg more than in
17    whites. Neither the black or white lead coefficients were significant.
18
19    European Occupational Studies
20          Maheswaran et al. (1993) reported on 809 male factory workers with blood lead between
21    less than 21 to more than 50 |ig/dL from Birmingham, England. Unfortunately, the inclusion of
22    factors that were related to blood lead, including additional direct measure of lead exposure in
23    addition to linear blood lead, years working in factory, and inclusion of zinc protoporphyrin, may
24    have biased the blood lead effect and resulted in nonsignificant lead effects on blood pressure.
25          Telisman et al. (2004) also reported nonsignificant  effects of natural log blood lead on
26    blood pressure in 115 male industrial workers with blood lead between 9.9 and 69.9 |ig/dL, but
27    included erythrocyte protoporphyrin in models, a variable correlated with blood lead over much
28    of the observed blood lead range. Coefficients were not given, as lead did not enter into stepwise
29    regression models.
30
      December 2005                           6-176        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Asian Occupational Studies
 2          Male and female factory workers (n = 798) from Chonan, Korea (blood lead between
 3    17.8 and 64.8 |ig/dL) were studied principally for the effects of genotype of ALAD and vitamin
 4   D receptor on cardiovascular response to lead (Lee et al., 2001).  These aspects are covered more
 5   thoroughly below. As part of their work, the authors developed multiple regression models
 6   examining the effect of linear blood lead on blood pressure with  forced entry of age and age2,
 7   BMI, sex, antihypertensive medication, lifetime alcohol, and ALAD and vitamin D genotypes.
 8   A marginally significant effect of blood lead on systolic blood pressure (diastolic blood pressure
 9   not modeled) was noted, with a 10 |ig/dL increase in blood lead associated with a 0.7 mm Hg
10   (95% CI: -0.04, 1.4) increase in blood pressure.
11          Nomiyama et al.  (2002) used a combined group of 193 female crystal glass workers and
12   nonexposed controls, aged 16 to 58 years, with blood lead between 3.8 and 99.4 |ig/dL. The
13   authors used a stepwise multiple regression with a novel technique to reduce collinearity among
14   covariates. From a large group of covariates, they selected covariates eligible to enter the
15   regression from a factor analysis.  Although the stepwise entry of these variables resulted in
16   different models for systolic and diastolic blood pressure, both models included linear blood
17   lead, age, urine protein, and plasma triglycerides.  The diastolic model additionally included
18   family hypertension and low density lipoprotein. Each 10 |ig/dL increase in blood lead was
19   significantly associated with a 1.26 mm Hg (95% CI: 0.58, 1.94) increase in systolic blood
20   pressure and a 1.05 mm Hg (95% CI: 0.52, 1.57) in diastolic blood pressure.  In alternative
21   models with ordered categories of blood lead, systolic blood pressure was 7.5  mm Hg (95% CI:
22   3.0, 12.0) and diastolic blood pressure was 6.3 mm Hg (95% CI:  3.4,  9.1) higher in workers with
23   blood lead >60 |ig/dL than in controls with <11.4 |ig/dL. Models did not control for BMI.
24          Wu et al. (1996) examined the effect of ordered blood lead category on blood pressure of
25    112 male (aged 18-67 years) and 110 female (aged 18-71 years) lead battery factory workers in
26   multiple regression models. Blood lead ranged from 8.3 to 95.4  |ig/dL.  Nonsignificant blood
27   lead effects were found possibly due to the inclusion of two additional lead exposure
28   measurements, ambient air lead and work history, likely leading  to substantial collinearity with
29   blood lead.
30
      December 2005                           6-177        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Meta-Analysis of Blood Lead-Blood Pressure Studies
 2          The most recent meta-analysis of the blood lead-blood pressure literature analyzed
 3    31 studies from a large pool of studies published up to 2001 (Nawrot et al., 2002).  Two other
 4    meta-analyses were also published during this reporting period (Schwartz, 1995; Staessen et al.,
 5    1994), covering many of the earlier papers cited in Nawrot et al. (2002), and derived similar
 6    coefficients for the lead effect, so they will not be reviewed here. The meta-analysis authors
 7    selected studies with 50 or more subjects, with subjects 10 years of age and up, with blood
 8    pressure and blood lead measurement techniques presented in sufficient detail to estimate effect
 9    sizes, and with preference given to papers with models adjusting for age, BMI, and "additional
10    factors of proven importance."  Where possible, studies with stratified analyses based on sex and
11    race were entered in the meta-analysis as separate subgroups.  Studies were weighted by the
12    number of subjects to arrive at estimates and CIs for lead effect on diastolic and systolic blood
13    pressure. Nearly half the studies reported lead effects from linear lead terms, the remainder from
14    log-transformed lead.  To include both types of studies in the analyses, the authors reported
15    effect sizes based on doubling the mean blood lead concentration.  For models using logarithmic
16    blood lead, this doubling has the same effect anywhere in the range of blood lead in the study.
17    For models using linear blood lead, the doubling effect was referenced from the mean blood lead
18    reported. Figures 6-5.1 and 6-5.2 depict the effect estimates for systolic and diastolic blood
19    pressure, respectively, included in the meta-analysis from Nawrot et al. (2002).  Ninety-five
20    percent CIs overlapped for males and females and for blacks and whites, indicating no
21    significant differences in lead effect by gender or race. In the group of studies as a whole, the
22    combined meta-analysis coefficients for each doubling of blood lead were highly significant for
23    both systolic (1.0 mm Hg [95% CI: 0.5, 1.4]) and diastolic (0.6 mm Hg [95% CI: 0.4, 0.8])
24    blood pressure. The meta-analysis supports the statistical association between increased blood
25    lead and increased blood pressure over a wide range of populations in many studies. Two major
26    systematic reviews of the lead-blood pressure literature were published during this review period
27    (Hertz-Picciotto and Croft, 1993; Research Triangle Institute,  1999). The findings here,
28    especially noting the significant effects from the meta-analysis (Nawrot  et al., 2002), continue to
29    support the significant association between blood lead and blood pressure/hypertension in
30    diverse segments of the general population.
      December 2005                           6-178        DRAFT-DO NOT QUOTE OR CITE

-------
     Reference
Gender   n
     Reimer and Tittelbach (1989)
     Weiss etal. (1986)
     deKortetal. (1987)
     Morris etal. (1990)
     Lockett and Arbuckle (1987)
     Sharp etal. (W)(1990)
     Sharp etal. (B)(1990)
     Staessen etal. (1990)
     Morris etal. (1990)
     Sokasetal. (W)(1997)
     Apostoli et al. (1990)
     Apostoli et al. (1990)
     Nerietal. (FW)(1988)
     Staessen et al. (P) (1996)
     Staessen et al. (P) (1996)
     Gartside(B){1988)
     Staessen et al. (1990)
     Gartside(B)(1988)
     Parkinson etal. (1987)
     Orssaudetal. (1985)
     Mollerand Kristensen (1992)
     Rothenberg et al. (Nl) (1999)
     Grandjean et al. (1989)
     Schwartz et ai. (2000b)
     Grandjean et al. (1989)
     Kromhout et al. (1985)
     Proctor etal. (1996)
     Maheswaran et al. (1993)
     Elwoodetal. (HP)(1988a,b)
     Elwood et al. (HP) (1988a,b)
     Elwoodetal. (C)(19B8a,b)
     Rothenberg et al. (I) (1999)
     Menditto et al. (1994)
     Chuetal. (1999)
     Chuetal. (1999)
     Henseetal. (1993)
     Henseetal. (1993)
     Den Hond et al. (B) (2002)
     Nerietal. (1988)
     Den Hond et al. (B) (2002)
     Bostetal. (1999)
     Gartside(W)(1998)
     Bostetal. (1999)
     Gartside(W)(1998)
     Rabinowitz et al. (1987)
     Den Hond et al. (W) (2002)
     Den Hond et al. (W) (2002)
     Pococketal. (1984)

     All
                                       Systolic Blood Pressure


Figure 6-5.1.   Change in the systolic pressure (effect estimate in mm Hg) associated with
                  a doubling of the blood lead concentration.  Studies arranged vertically by
                  increasing study size.

         Study key:    C = Caerphilly Study, HP = Welsh Heart Program, P = PheeCad Study, W = whites,
                      B = blacks, NI = nonimmigrants, I = immigrants, FW = foundry workers, CS = civil
                      servants.

Source: Nawrot et al. (2002).
M
M
F
M
M
M
F
M
M
M
F
M
F
M
M
M
F
M
M
M
F
M
M
F
M
M
M
F
M
M
F
M
F
M
F
M
M
M/F
F
M
M
F
F
F
M
F
M
M/F
89
105
106 <
11"7 It
132
145
ICj/j | 	
°T1 I

339 j 	 g

350 i

/\
-------
          Reference
Gender
M
IV!
M
¥
M


M
F
M
M
M

F
M
M
F
M
jyj
M
F
M
M
F
M
M
F
M
M
F
M
F
M
F
M
M
M/F
F
M
M
F
F
F
M
F
M
OO If 	
CiQ t
QJ3 p-™™™™™™™™™™™
105 i
106 i
118 j™»
1 17 i
1 ^^
I O£
1 *"i'"> f Hi
145
•1 *%O
1 DZ
1 °P i mi
254
9T1
£. f I
288

%&K
*2*t«3
350
398 I—
407 I 	
428 i
•131 i
439 I
439 M
504 I 	 •-
543 t
546 I
798
809 I 	 •—
856 1
865 1
1136 h
1188
1319
1329
1471 i
1661
1703 1-
1760 1
2193 h
2187
2563
2705
2763 1
2827
3851 (
4681 m
5126 •
7379 1
M/F 58491
I
-15
*s
r™™™™™™™™™™™™™™™™™™™!
*i

)
4
««


!
H»H
	 |

	 I
t
1 ill I
[[[[i-llfc 	 [[[1

1
«™™™™™™™J
«««««««««««|
-* 	 1
"• 	 1

• I

M
— 1
— * — 1
— • 	 1
W-l
— 1
M
m
N
m
I-«H
1

1_»_|
••— 1
»-i
-» 	 1
•<
| 	 » 	 1
1-W
— • 	 1
f*1
)



M
11 1 1 1 1 1
-10 -5 0 5 10 15 20 25
          Reimer and Titteibach (1989)
          Weiss etal. (1986)
          DeKortetal. (1987)
          Morris etai. (1990)
          Lockett and Arbuckle (1987)
          Sharp etal. (W) (1990)
          Sharp etal. (B) (1990)
          Staessenef a!. (1990)
          Morris etal. (1990)
          Kromhoutetal. (1985)
          Sokasetal. (W)(1997)
          Apostoiietal. (1990)
          Apostoli et ai. (1990)
          Nerietal. (FW)(1988)
          Staessenetal. (P)(1996)
          Staessen et al. (P)(1996)
          Gartside(B)(1988)
          Staessen ei a!. (1990)
          6artside(B)(1988)
          Parkinson et al. (1987)
          Orssaud et al. (1985)
          Mailer and Kristensen (1992)
          Rothenberg et al. (Nl) (1999)
          Grandjean et al. (1989)
          Schwartz et al. (2000b)
          Grandjean et al. (1989)
          Proctor etal. (1998)
          Maheswaran etal. (1993)
          Elwood et al. (HP) (1988a,b)
          Elwood et al. (HP) (1988a,b)
          Elwood et al. (C) (1988a,b)
          Rothenberg et al. (I) (1999)
          Menditto et al. (1994)
          Chu etal. (1999)
          Chuetal. (1999)
          Henseetal. (1993)
          Henseetal. (1993)
          Den Hond et al. (B) (2002)
          Nerietal. (1988)
          Den Hond et al. (B) (2002)
          Bostetal. (1999)
          Gartside(W)(1988)
          Bostetal. (1999)
          Gartside(W)(1988)
          Rabinowitz et al. (1987)
          Den Hond et al. (W) (2002)
          Den Hond et al. (W) (2002)
          Pococketal. (1984)

          Ail
                                        Diastolic Blood Pressure

Figure 6-5.2.   Change in the diastolic pressure (effect estimate in mm Hg) associated with
                  a doubling of the blood lead concentration.  Studies arranged vertically by
                  increasing study size.

         Study key:    C = Caerphilly Study, HP = Welsh Heart Program, P = PheeCad Study, W = whites,
                       B = blacks, NI = nonimmigrants, I = immigrants, FW = foundry workers, CS = civil

-------
 1          Quantitative estimates for the effect of doubling the mean blood lead concentration on
 2    systolic and diastolic blood pressure from the various studies discussed here are summarized in
 3    Tables 6-5.1 and 6-5.2. Results from these individual studies also generally appear to agree with
 4    the results of the meta-analysis by Nawrot et al. that increased blood lead levels are significantly
 5    associated with increased systolic  and diastolic blood pressure.  Figures 6-5.1  and 6-5.2
 6    graphically depict the results of many of the studies discussed.  The effect estimates in the
 7    figures also reflect the effect of doubling blood lead on blood pressure. Most  of the effects are
 8    based on concurrent blood lead. Effects for the entire study population are presented unless only
 9    effects in subsamples are reported. Other selection criteria used in a few instances
10    were subjective.
11          A Bayesian meta-analysis was performed to examine the use of log-linear and linear
12    blood lead in blood pressure regression models (Figure 6-5.3).  A significant blood lead effect on
13    systolic pressure was observed using both log-linear and linear blood lead. Heterogeneity also
14    was significant for both log-linear lead (p = 0.0002) and linear lead (p = 0.05). The source of
15    heterogeneity could be due to several factors, including different methods of statistical analysis,
16    different study protocols, and subject selection differences.  The log-linear and linear lead effects
17    were 0.62 mm Hg (95% CI:  0.12, 1.11) and 0.98 mm Hg (95% CI: 0.51, 1.45), respectively, for
18    systolic blood pressure.  The difference between the effect estimates from using log-linear or
19    linear lead was nonsignificant. This meta-analysis suggests there are significant differences
20    between the studies, but overall there is a combined effect of blood lead on systolic blood
21    pressure. The fact that several individual studies did not detect  a significant effect may be due to
22    small study size or other factors affecting effect measurement precision.
23
24    6.5.2.3   Blood Pressure and Hypertension Studies Using Bone Lead as Exposure Index
25          Korrick et al. (1999) used a case-control design to study  the relationship between
26    hypertension and three measures of lead exposure (blood lead, tibia [cortical bone] lead, and
27    patella [trabecular bone] lead in women. The final study sample consisted of  89 hypertension
28    cases and 195 controls, excluding  those with history of hypertension, cardiovascular disease,
29    renal disease, diabetes, or malignancy, use of antihypertensive medications, BMI >29, and
30    incomplete data, aged from 47 to 74 years. Cases were selected through a randomization
31    procedure that produced approximately equal numbers of cases  for each of three blood pressure
      December 2005                            6-181        DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
oo
to
 H
 b
 o

 O
 H
O
 o
 H
 W
 O
 O
 HH
 H
 W
                       Table 6-5.1. Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Linear Lead
           (Coefficients Represent Effect of Doubling Blood Lead Calculated from Mean Blood Lead or Mid-point of Range)

Reference,
Study Location,
Period
Study
Population,
Sample Size
(n), Age
[years]
POPULATION OR COMMUNITY
Cheng et al.
(2001)
Boston, U.S.
Normative Aging
Study
1991-1997
Males
(519)
[mean age =
66.4]


Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ng/dL
STUDIES
5.9a
(3.9)d
[range not given]





Model Type and
Variables Considered

Multiple linear regression
forced entry of linear lead,
age, age2, BMI, family
hypertension, smoking,
alcohol, dietary calcium.

Systolic Blood Pressure
(mm Hg)


Model Unique
Diagnostic Covariates

Not stated Not
applicable






Lead-Associated
Variables

Alcohol,
smoking, age,
dietary calcium





Coefficient
(95% CI) [SE]

-0.2
(-2.0, 1.6)
[0.92]



Diastolic Blood Pressure
(mm Hg)

Lead- Coefficient
Unique Associated (95% CI)
Covariates Variables [SE]

Not tested Not tested —





          Huetal. (1996)
Gerr et al. (2002)     Females and
          Eastern
          Washington,
          Western Idaho,
          U.S.
          1994
         Nash et al. (2003)
         U.S.-NHANES
         III
         1988-1994
                    Males
                    (502)
                    [19-29]
                   Females
                   (1,786)
                   [40-59]
    —         Almost completely
               overlapping with Cheng
               etal. (2001).

   2.2a         Linear multiple regression
   (1.9)d        with forced entry of linear
  [7]       blood lead, tibia lead, age,
               BMI, sex, education, birth
               control pills, smoking,
               height, hemoglobin, serum
               albumin, childhood
               residence, income, alcohol.

   2.9a         Multiple linear regression
(not stated/      (survey weighted),
 [0.5-31.1]       excluding treated
               hypertensives, with forced
               entry of linear lead, age,
               BMI, race/ethnicity, serum
               creatinine. Education,
               poverty, alcohol, smoking
               among variables for
               stepwise entry.
                                                                          Not stated
                                                                                                   Not
                                                                                                applicable
 Age, tibia lead,
smoking, alcohol,
  hemoglobin
   0.0
(-1.1,1.1)
  [0.57]
                                                                                    Not stated
                                                                                                   Not
                                                                                                applicable
    Alcohol,
  smoking, age
   0.9
 (0.0, 1.8)
  [0.46]
Not
applicable

Not
applicable
Age, tibia
lead,
smoking,
alcohol,
hemoglobin
Alcohol,
smoking,
age
-0.3
(-2.8, 1.6)
[0.48]

0.7
(0.2, 1.2)
[0.26]

-------
o

O

to
O
O
     Table 6-5.1 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Linear Lead

(Coefficients Represent Effect of Doubling Blood Lead Calculated from Mean Blood Lead or Mid-point of Range)
oo
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
Reference,
Study Location,
Period
Study
Population,
Sample Size
(n), Age
[years]
POPULATION OR COMMUNITY
Vupputuri et al.
(2003)
NHANES III,
U.S.









Sorel et al.
(1991)
NHANES II,
U.S.
1976-1980


Black female
(2,300)
[>18-?]

Black male
(2,104)
[>18-?]
White female
(5,188)
[>18-?]
White male
(5,360)
([>18-?]
Female
(2,056)
[18-74]

Male
(2,044)
[18-74]
Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ng/dL
STUDIES (cont'd)
3.4a
(3.3)d
[not stated]

5.4a
(3.3)"
[not stated]
3.0"
(3.3)d
[not stated]
4.4a
(3.3)"
[not stated]
13.2a
(not stated)
[not stated]

20.T
(not stated)
[not stated]
Model Type and
Variables Considered

Multiple linear regression
(not survey corrected) with
forced entry of all variables:
linear lead, age, BMI,
education, alcohol, physical
activity, dietary Na, K,
and calories.






Multiple linear regression
(survey corrected) with
forced entry of linear lead,
age, BMI.



Systolic Blood Pressure
(nun Hg)
Model
„. .. Unique Lead-Associated Coefficient
mgnos ic Covariates Variables (95% CI) [SE]

Not stated Not Age, alcohol 1.6
applicable (0.5, 2.7)
[0.82]

1.3
(0.3, 2.4)
[0.54]
0.3
(-0.4, 1.1)
[0.38]
0.4
(-1.0, 1.8)
[0.72]
Not stated Not Age Errors in
applicable original article


1.7
(-0.6,3.8)
[0.56]
Diastolic Blood Pressure
(in in Hg)
Lead- Coefficient
Unique Associated (95% CI)
Covariates Variables [SE]

Not Age, 1.1
applicable alcohol (0.4, 1.8)
[0.36]

1.0
(0.1,2.0)
[0.48]
0.0
(-0.5,0.4)
[0.22]
0.0
(-0.5,0.5)
[0.27]
Not Age Errors in
applicable original
article

2.6
(0.8, 4.2)
[0.87]

-------
o

O

to
O
O
     Table 6-5.1 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Linear Lead

(Coefficients Represent Effect of Doubling Blood Lead Calculated from Mean Blood Lead or Mid-point of Range)
oo
H

b
o


O
H

O


o
H
W

O


O
HH
H
W
Reference,
Study Location,
Period
Study
Population,
Sample Size
(n), Age
[years]
Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ng/dL
Model Type and Model
Variables Considered Diagnostic
Systolic Blood Pressure
(mm Hg)
Unique Lead-Associated Coefficient
Covariates Variables (95% CI) [SE]
Diastolic Blood Pressure
(mm Hg)
Lead- Coefficient
Unique Associated (95% CI)
Covariates Variables [SE]
POPULATION OR COMMUNITY STUDIES (cont'd)
Hense et al.
(1994)
Augsberg,
Denmark
1987-1988





















Female
(no alcohol)
(701)
[28-67]

Female
(<40 g/day)
(877)
[28-67]
Female
(> 40 g/day)
(83)
[28-67]
Male, urban
(118)
[28-67]

Male, rural
(no alcohol)
(147)
[28-67]
Male, rural
(<40 g/day)
(463)
[28-67]
Male, rural
(> 40 g/day)
(356)
[28-67]
Not stated
(not stated)
[<3->8]


Not stated
(not stated)
[<3->10]

Not stated
(not stated)
[<3->14]

Not stated
(not stated)
[not stated]

Not stated
(not stated)
[<5->ll]

Not stated
(not stated)

Not stated
(not stated)
[<7->15]

Multiple linear regression Not stated
with forced entry of linear
lead, age, BMI, hematocrit,
residence, smoking in
women. In men, analyses
were stratified by place of
residence.






No blood lead values shown;
thus, there are no
coefficients calculated and
shown
Multiple linear regression Not stated
with forced entry of
linear lead, age, BMI,
hematocrit, residence,
smoking in women. In men,
analyses were stratified by
place of residence.





Not Age, smoking, 0.4
applicable hematocrit (-2.6,3.4)
[1.53]


0.7
(-1.5,2.9)
[1.12]

11.0
(3.8, 18.3)
[3.64]

	



Not Age, smoking, 2.8
applicable hematocrit (-2.4, 8.0)
[2.63]

5.8
(1.9,9.8)
[2.01]

5.0
(0.6, 10.9)
[2.87]

Not Age, 1.2
applicable smoking, (-0.8,3.1)
hematocrit [0.99]


1.5
(0.1,2.9)
[0.71]

7.3
(2.8, 11.8)
[1.36]

	



Not Age, 0.2
applicable smoking, (-3.3,3.8)
hematocrit [1.80]

3.5
(0.8, 6.2)
[1.37]

3.3
(0.3,6.3)
[1.53]


-------
o

O

to
O
O
     Table 6-5.1 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Linear Lead

(Coefficients Represent Effect of Doubling Blood Lead Calculated from Mean Blood Lead or Mid-point of Range)
oo
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
Reference,
Study Location,
Period
OCCUPATIONAL
Glenn et al.
(2003)
New Jersey,
U.S.
1994-1998


Glenn et al.
(2001)
New Jersey,
U.S.
1996-1997

Schwartz et al.
(2000b)
New Jersey,
U.S.
1996-1997
Sokas et al.
(1997)
Maryland
U.S.
1989-1990




Study
Population,
Sample Size
(n), Age
[years]
STUDIES
Males, former
organolead
workers




Males, former
organolead
workers
(213)
[mean age =
58.0]
Males, former
organolead
workers
(543)
[41.7-73.3]
Male,
construction
workers
(264)
[18-79]




Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ng/dL

—






5.2a
(3-l)d
[not stated]



4.6"
(2.6)d
[1-20]


8.0"
(not given)
[2-30]






Model Type and Model
Variables Considered Diagnostic

Only measured change in —
blood pressure over a period
ranging from 10 months to
5 years. Design not
commensurate with other
studies in this table.
See Glenn (2001), below.
Subset of Schwartz (2000), —
see below.




Multiple backward Not stated
elimination stepwise linear
regression models. Pool of
available covariates not
specified.
Multiple linear regression, Not stated
presumably with forced
entry (not stated).
Covariates available for
entry not stated. Insufficient
information given for
separate black and white
blood pressure effects on
blood pressure.

Unique
Covariates

—






	





Age, BMI,
smoking,
antihyperte
nsive
medications
Linear lead,
BMI, age,
hematocrit,
erythrocyte
protoporph
yrin, race,
race by lead
interaction

Systolic Blood Pressure
(nun Hg)
Lead-Associated Coefficient
Variables (95% CI) [SE]

— —






	 	





Age, smoking 2.3
(0.2, 4.4)
[1.15]


Age, hematocrit, Not given here,
erythrocyte as coefficient
protoporphyrin was altered by
presence of
race-lead
interaction



Diastolic Blood Pressure
(in in Hg)
Unique
Covariates

—






	





Age, BMI,
smoking,
antihyperte
nsive
medications
Linear lead,
BMI, age,
hematocrit,
erythrocyte
protoporph
yrin, race,
race by lead
interaction

Lead-
Associated
Variables

—






	





Age,
smoking



Age,
hematocrit,
erythrocyte
protoporph
yrin




Coefficient
(95% CI)
[SE]

—






	





1.3
(-0.3,2.9)
[0.86]


Not given
here, as
coefficient
was altered
by the race-
lead
interaction



-------
o

O

to
O
O
     Table 6-5.1 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Linear Lead

(Coefficients Represent Effect of Doubling Blood Lead Calculated from Mean Blood Lead or Mid-point of Range)
oo
Oi
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
Reference,
Study Location,
Period
Study
Population,
Sample Size
(n), Age
[years]
OCCUPATIONAL STUDIES (cont
Maheswaran
et al. (1993)
Birmingham
Europe
Dates not given


Leeetal. (2001)
Chonan, Korea
1997-1999




Lustberg et al.
(2004)
Chonan, Korea
1997-1999
Nomiyama et al.
(2002)
Beijing, China
Dates not given






Wuetal. (1996)
Central Taiwan
Data collection
dates not given
Male factory
workers
(809)
[mean age
43.3]


Male and
female lead-
using factory
workers
(798)
[17.8-64.8]

Overlapping
study with Lee
(2001), above

Female crystal
toy makers and
nonexposed
sewing workers
(193)
[16-58]




—



Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ng/dL
'd)
31. 5b
(5.5)"
[<21->50]




32.0a
(15.0)d
[4-86]




—



37.6a
(9.2)"
[3.8-99.4]







—



Systolic Blood Pressure
(nun Hg)
Model Type and Model Unique Lead-Associated Coefficient
Variables Considered Diagnostic Covariates Variables (95% CI) [SE]

Multiple linear regression, Not stated Not Age, alcohol, 2.2
forced entry of age, BMI, applicable zinc (-0.9,5.4)
alcohol, zinc protoporphyrin, protoporphyrin, [1.60]
years working, smoking, years working,
linear lead. smoking


Multiple linear regression Not stated Not Age, alcohol 2.2
with forced entry of age, applicable (-0.1,5.5)
age2, sex, BMI, [1.18]
antihypertensive medication,
lifetime alcohol, ALAD and
vitamin D receptor
genotypes.
Used deciles as lead — — — —
variable.


Multiple linear regression Not stated Linear lead, Age 4.7
with forward stepwise age, urine (2.0, 7.4 )
addition of variables protein, [1-35]
(p < 0.2). Candidate plasma
variables were selected from triglyceride
a very large group and
narrowed down to ten by
factor analysis. The ten
available for entry were
not stated.
Ordered categories of lead — — — —
used. Coefficients cannot
be calculated.

Diastolic Blood Pressure
(in in Hg)
Unique
Covariates

Not
applicable





Not shown






—



Linear lead,
age, urine
protein,
plasma
triglyceride,
family
hypertensio
n,
low density
lipoprotein
—



Lead-
Associated
Variables

Age,
alcohol,
zinc
protoporph
yrin, years
working,
smoking
Not shown






—



Age









—



Coefficient
(95% CI)
[SE]

-1.3
(-3.5,0.9)
[1.12]




—






—



3.9
(1.9,5.9)
[1.04]







—




-------
o

O

to
O
O
Table 6-5.2. Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Logarithmic Lead
                    (Coefficients Represent Effect of Doubling Blood Lead)
oo
H
6
o

o
H
O
Reference,
Study
Location,
Period
POPULATION
Den Hond
et al. (2002)
NHANES III
U.S.
1988-1994













Morris et al.
(1990)
Oregon, U.S.
1984-1989




Study
Population,
Sample Size
(n), Age
[years]
Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ng/dL
Systolic Blood Pressure
(nun Hg)
Model Type and Model
Variables Considered Diagnostic
Unique
Covariates
Lead-
Associated
Variables
Coefficient
(95% CI)
[SE]
Diastolic Blood Pressure
(in in Hg)
Unique
Covariates
Lead-
Associated
Variables
Coefficient
(95% CI)
[SE]
OR COMMUNITY STUDIES
Black males
(1,761)
[>20]



Black
females
(2,197)
[>20]
White males
(4,685)
[>20]

White
females
(5,138)
[>20]

Females
(106)
[18-80]

Males
(145)
[18-80]

4.2C(2.7, 6.5)e
[<1.2->20.0]




2.3C(1.4, 3.9)e
[<0.8->9.0]


3.6C
(2.3, 5.3)e
[<1.2->15.0]

2.1C
(1.3, 3.4)e
[<0.8->8.0]


6.9a
(3.6)d
[?-39]

8.0a
(4.4)"
[5-40.5]

Multiple linear Not stated
regression (no
adjustment for survey
design): forced entry
of log lead, age, age ,
BMI, hematocrit,
smoking, alcohol,
antihypertensive drugs.
Forward stepwise entry
(p < 0.05) among
coffee, dietary
calcium, dietary
Na/Ca, serum total
protein, serum total
calcium, diabetes.






Linear multiple Not stated
regression with
stepwise entry among
In lead, age, BMI,
dietary calcium, "other
nutrients," serum Ca
erythrocyte
protoporphyrin.
Serum total
protein




Dietary Na/Ca,
serum total
protein

Dietary
calcium, dietary
Na/Ca, serum
total protein,
serum total
calcium
Serum total
protein, diabetes



Age, dietary
calcium, BMI




Age, ionized
calcium
Alcohol,
smoking, age




Alcohol,
smoking, age


Alcohol,
smoking, age,
dietary
calcium

Alcohol,
smoking, age



Age, dietary
calcium




Age

0.9
(0.04, 1.8)
[0.43]



1.2
(0.4, 2.0)
[0.42]

0.3
(-0.2,0.7)
[0.23]

0.1
(-0.4,0.5)
[0.32]


Not stated





3.2
(Not given)
Coffee, dietary
Na, diabetes,
serum total
protein,


Dietary Na/K,
serum total
protein, serum
total calcium
Dietary Na/Ca,
serum total
protein, serum
total calcium,
diabetes

Dietary calcium,
serum total
protein, diabetes


Age, dietary
calcium,
hemoglobin



Age,
hemoglobin,
smoking
Age,
smoking,
alcohol,
dietary
calcium

Age,
smoking,
alcohol

Age,
smoking,
alcohol

Age,
smoking,
alcohol,
dietary
calcium
Age, dietary
calcium




Age,
hemoglobin,
smoking
0.3
(-03, 1.0)
[0.16]



0.5
(0.01, 1.1)
[0.26]

-0.6
(-0.9, -0.3)
[0.15]

-0.2
(-0.5,0.1)
[0.14]


Not stated





1.3
(Not given)
O
HH
H
W

-------
o

O

to
O
O
oo
oo
H
6
o

o
H
O
Table 6-5.2 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Logarithmic Lead

                        (Coefficients Represent Effect of Doubling Blood Lead)
Reference,
Study
Location,
Period
POPULATION
Proctor et al.
(1996)
Normative
Aging Study
Boston, U.S.
1991-1993
Rothenberg
etal. (1999)
Los Angeles,
U.S.
1995-1998




Rothenberg
et al. (2002)
Los Angeles,
U.S.
1995-2001



Study
Population,
Sample Size
(n), Age
[years]
OR COMMUNIT\
Males
(798)
[45-93]



Pregnant
women,
immigrant
(1,188)
[>14-<44]
Pregnant
women,
nonimmigrants
(439)
Females third
trimester
(637)
[15-43]

Females
postpartum
(637)
[15-43]
Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ug/dL
' STUDIES (cont'c
6.5a
(4.0)d
[0.5-35.0]



2.3b
(1.3)d
[0.5-40.4]


1.9b
(+1.3, -0.8)11
[not stated]

1.9b
(+3.6, -I. Of
[not stated]


2.3b
(+4.2, -1.2)d
[0.4-23.7)

Model Type and Model
Variables Considered Diagnostic
I)
Multiple linear regression with Not stated
forced entry of all variables:
In lead, age, age , BMI,
dietary calcium, exercise,
smoking, alcohol, heart rate,
hematocrit.
Multiple linear regression with Residual analyses,
forced entry of all variables: outliers,
In lead, age, BMI, coffee, iron heteroscedasticity
supplement, job stress.





Multiple linear regression with Residual analyses,
forced entry of all variables: outliers,
In lead, age, BMI, education, heteroscedasticity
immigrant status, smoking,
alcohol, parity. Hypertensive
subjects > 140/90 mm Hg
(third trimester or postpartum)
were excluded.

Systolic Blood Pressure
(mm Hg)
Lead- Coefficient
Unique Associated (95% CI)
Covariates Variables [SE]

Not Alcohol, 0.6
applicable smoking, (-0.8, 1.9)
dietary [0.69]
calcium, age


Not Age 1.2
applicable (0.5, 1.9)
[0.37]


Age 0.3
(-1.1,1.6)
[0.67]

Not Alcohol, 0.0
applicable smoking, age (-0.9,0.8)
[0.43]


-1.0
(-2.0, -0.1)
[0.46]

Diastolic Blood Pressure
(mm Hg)
Lead-
Unique Associated
Covariates Variables

Not Alcohol,
applicable smoking,
dietary
calcium,
age

Not Age
applicable



Age



Not Alcohol,
applicable smoking,
age





Coefficient
(95% CI)
[SE]

0.8
(0.1,1.5)
[0.41]



1.0
(0.4, 1.5)
[0.28]


0.1
(-1.3,1.4)
[0.67]

0.1
(-0.5,0.8)
[0.34]


-1.2
(-2.0, -0.3)
[0.41]

O
HH
H
W

-------
o

O

to
O
O
Table 6-5.2 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Logarithmic Lead
                        (Coefficients Represent Effect of Doubling Blood Lead)
oo
VO
H
6
o

o
H
O
Reference,
Study Location,
Period
POPULATION
Schwartz
(1991)
NHANES II
U.S.
1976-1980






Bost et al.
(1999)
Health Survey
for England
1995


Menditto et al.
(1994)
Rome, Europe
New Risk Factor
Project
1989-1990

Study
Population,
Sample Size
(n), Age
[years]
OR COMMUNIT\
Females
(<5,000)
[20-74]


Males
(<5,000)
[20-74]




Females
(2,763)
[16-not
given]
Females
(2,563)
[16-not
given]
Males
(1,319)
[55-75]




Blood Lead
Mean", Geom.
Meanb, or
Median'
(SD)d or (IQR)e
[range] ug/dL
' STUDIES (cont'c
Not stated










2.6"
(not stated)
[<1.5->8.5]

3.7b
(not stated)

11.3a
(not stated)
[4.0-44.2]




Systolic Blood Pressure
(mm Hg)
Model Type and
Variables Considered
I)
Multiple linear
regression with forced
entry of In lead, age,
age , and BMI. Stepwise
entry (p < 0.05) among
race, cigarettes/day,
tricep skinfold, family
hypertension, exercise,
In serum Zn, dietary K
and Na, serum
cholesterol, height, and
In dietary vitamin C.
Multiple linear stepwise
regression (forward or
backward not stated),
with possible entry of
age, log BMI, log lead,
alcohol, social class,
place of residence, and
smoking.

Multiple linear
regression with forward
stepwise entry among In
lead, age, BMI, heart
rate, lipids, triglycerides,
glucose, smoking,
alcohol, skinfold.
Model Unique
Diagnostic Covariates

Not stated No results
presented









Not stated Age, BMI,
residency


Age, BMI,
alcohol, social
class

Not stated BMI, heart rate,
serum lipids, age,
glucose,
smoking,
skinfold,
triglycerides,
skinfold
Lead- Coefficient
Associated (95% CI)
Variables [SE]

No results No results
presented presented









Age No results
presented
(nonsignificant)

Age, No results
alcohol presented
(nonsignificant)

Smoking 3.9
no other
measures given




Diastolic Blood Pressure
(mm Hg)
Unique
Covariates

Race, zinc, family
hypertension,
tricep fold,
cholesterol

Race, zinc, family
hypertension,
tricep fold,
cholesterol, height,
cigarettes


Age, BMI, alcohol



Age, BMI, alcohol


BMI, heart rate,
age, smoking,
lipids, triglycerides




Lead-
Associated
Variables

Age




Age,
cigarettes




Age,
alcohol


Age,
alcohol

Smoking






Coefficient
(95% CI)
[SE]

1.1
(0.2, 2.0)
[0.48]


2.0
(0.6, 3.4)
[0.69]




No results
presented
(p > 0.05)

0.9
(0.1, 1.6)
[0.38]

1.2
no other
measures
given



O
HH
H
W

-------
o

O

to
O
O
                       Table 6-5.2 (cont'd).  Systolic and Diastolic Blood Pressure and Blood Lead Modeled with Logarithmic Lead
                                                     (Coefficients Represent Effect of Doubling Blood Lead)
H
6
o

O
H
O
o
H
W
O
O
HH
H
W


Reference,
Study Location,
Period
Study
Population,
Sample Size
(n), Age
[years]
Blood Lead
Mean", Geom.
- _ b
Mean , or
Median'
(SD)d or (IQR)e
[range] ng/dL



Model Type and
Variables Considered




Model Diagnostic
Systolic Blood Pressure
(mm Hg)

Lead- Coefficient
Unique Associated (95% CI)
Covariates Variables [SE]
Diastolic Blood Pressure
(mm Hg)

Unique Lead- Coefficient
Covariat Associated (95% CI)
es Variables [SE]
OCCUPATIONAL STUDIES
Sharp et al.
(1990)
San Francisco,
U.S.
1986









Telisman et al.
(2004)
Zagreb, Croatia
2000-?

Male bus
drivers,
black
(132)
[30.4-60.7]




Male bus
drivers,
nonblack
(117)
[30.6-58.9]
Male
industrial
workers
(115)
[20-43]
6.5"
(+2.7, -1.9)d
[3.1-20.9]






6.2b
(+2.7, -1.8)d
[2.0-14.7)


36.7b
(not given)
[9.9-69.9]


Linear multiple regression
analysis of unspecified type.
All models presented here
adjusted for In lead, age, age2,
BMI, caffeine, smoking.
Also examined effect of
stratified caffeine modeling
(not shown here as group sizes
were less than 100), and effect
of successive addition of
covariates on lead coefficient,
and including alcohol use
(not shown).

Linear multiple regression
with forward stepwise entry
among In or linear lead, years
of exposure, age, smoking,
alcohol, BMI, ALAD,
Used influence
diagnostics to identify
two influential
subjects. Analyses
stratified by race
showed maximum
10% change in lead
coefficients without
influential subjects.
Data shown here
includes all subjects.



Not stated




Not Age, 5.2
applicable smoking (0.6, 9.8)
[2.3]






-3.9
(-8.3,0.4)
[2.3]


None given




Not Age, 3.3
applicable smoking (0.1,6.4)
[1.6]






0.5
(-2.3,3.4)
[1.4]


None given




                                                  erythrocyte protoporphyrin,
                                                  blood Cd, serum Zn, serum
                                                  Cu. Neither form of lead
                                                  variable was reported as
                                                  significant, but coefficients
                                                  not shown. Even though
                                                  forward stepwise was used
                                                  and insignificant lead
                                                  variables were not shown,
                                                  other nonsignificant variables
                                                  were shown in models.

-------
                Natural Log Lead in Model
               Linear Lead in Model

Sharp stal.(1 990) BM
Rothenberg et al.(1999) nonIMM t—
Mailer etal. (1992) M 1-
Maller etal. (1992) F
Proctor etal.(1 996) y H
Staessenetal.(1996)F I —
Staessen et al. (1 986) M I — • — I
Rothenberg el al. (1999) IMM
DenHondetal.(2002)BM
DenHondelal.(2002)BF
Den Hondetal. (2002) WM i
Den Hondetal. (2002) WM h


*— i Gerr etal. (2002) M/F !-
— « 	 1 Cheng et al. (2001 ) M I — i
i 	 1 	 1 Schwartz et al. (2000) M
« — i Lee etal. (2001) M/F
m Nash etal. (2003) F
m Soreletal. (1991JM I-
i-n Vupputuri etal. (2003) BM
•H Vupputun etal. (2003) BF
n Vupputuri etal. (2003) WF i-
Vupputuri etal. (2003) WM i-

> — i
— i

-n
— i 	 1
i— •— i
i i i
«H
* — 1
                      -10     -505
                 Change  in Systolic Blood Pressure
                                                 10
                 -10     -5      0      5      10
              Change in Systolic Blood Pressure
         Sharp etal. (1990) nonBM
         Sharp etal. (1990) By
         Rothenberg et al. (1999) nor*
         Mailer etal. (1992) M
         Melleretal. (1992) F
         Proctor etal. (1996)M
         Staessen etal. (1996) F
         Staessen etal. (1996) M
         Rothenberg etal.(1999) IMM
         Den Hond et al. (2002) BM
         Den Hondetal.(2002) BF
         Bost etal. (1999)
         Schwartz (1991) F
         Den Hond etal.(2002) WM
         Schwartz (1991) M
         Den Hond etal.(2002) WM
                     -10      -505
                 Change in Diastolic Blood Pressure
                                                 10
      Henseetal. (1994) M
      Nomiyamaetal.(2002) F
      Gerr etal. (2002) M/F
      Schwartz and Stewart (2000) M
      Hense etal.(1994) F
      Maheswaran etal. (1993) y
      Nash etal.(2003) F
      Soreletal.(1991)M
      Vupputuri etal. (2003) BM
      Vupputuri etal (2003) BF
      Vupputuri etal.(2003) WF
      Vupputun et al. (2003) WM
                -10     -5      0      5      10
              Change in Diastolic Blood Pressure
Figure 6-5.3.   Effect of doubling mean blood lead on estimate of blood pressure change
                 with 95% CIs.  Studies arranged vertically by increasing study size.
                 Where multiple models from the same study were presented, such as
                 repeated measures over time or adding a confounding variable, only the
                 effect estimate from the first model is shown. When the same study was
                 multiply published with subsamples, only the effect estimate from largest
                 study is shown.

                 Study key:   B = blacks, W = whites, M = males, F = females, IMM = immigrants,
                             non-IMM = nonimmigrants.
December 2005
6-191
DRAFT-DO NOT QUOTE OR CITE

-------
 1    categories, hypertensive (> 140 mm Hg or 90 mm Hg), high normal (> 121/75 mm Hg up to
 2    hypertension limit), and low normal (<121/75 mm Hg). As many as four controls were matched
 3    to cases by 5 year age grouping.  Though they did not match cases and controls on other
 4    potential confounding variables, they included these variables in their models.  The dependent
 5    variable was constructed by placing blood pressure measurements into the three groups.  The
 6    mean blood lead level was 3.1 |ig/dL; the mean tibia and patella lead levels were 13.3 jig/g and
 7    17.3 |ig/g, respectively. An ordered logistic regression with proportional odds assumptions was
 8    used to asses linear blood lead, patella and tibia bone lead effects on odds of hypertension,
 9    controlling for age, BMI, dietary calcium, alcohol use, dietary sodium, smoking, and family
10    hypertension. They presented results from four models with the same covariates determined
11    a priori, but with each lead variable tested separately. Only patella lead concentration
12    significantly (p = 0.03)  predicted increased odds for hypertension, but the effect was small.
13    Each 10 |ig/g increase in patella lead was associated with an odds ratio of 1.28 (95% CI:  1.03,
14    1.60).  Separate analyses testing interactions of alcohol use, age, and menopausal status showed
15    no significant interaction with patella lead, though the  small sample size had little power to
16    detect significant interaction effects.  Model diagnostics were given for justifying the use of
17    proportional odds ordinal regression but none were given justifying use of a linear blood lead
18    term in the models.
19          Rothenberg et al. (2002) investigated associations between both hypertension and blood
20    pressure with blood lead, tibia lead, and calcaneus lead in 668 women, aged 15-44 years, in the
21    third trimester of pregnancy and during a 3-month postpartum period using a cohort design and
22    multiple logistic and multiple linear regression modeling. Subject exclusion criteria were blood
23    lead > than 5 geometric SDs from the geometric mean, documented renal disease, cardiovascular
24    disease, diabetes, use of stimulant drugs, and extreme postnatal obesity (BMI >40). Geometric
25    mean prenatal and postnatal blood lead levels were 1.9 |ig/dL and 2.3  |ig/dL, respectively. Mean
26    tibia and calcaneus lead levels were 8.0 jig/g and 10.7  jig/g, respectively. Variables in all
27    models were selected a  priori and retained in the models regardless of significance level.  Control
28    variables were education, smoking status, immigrant status, parity, age, and BMI in all models.
29    Prenatal models also controlled for postpartum hypertension in  lieu of family history of
30    hypertension. None of the subjects used antihypertensive medications during the study.
31    All three lead variables  were simultaneously tested in all models. Third trimester blood lead

      December 2005                           6-192       DRAFT-DO NOT QUOTE OR CITE

-------
 1    ranged from 0.4 to 30.0 |ig/dL, postpartum blood lead ranged from 0.2 to 25.4 |ig/dL.  Calcaneus
 2    lead ranged from -30.6 to 49.9 |ig/g and tibia lead ranged from -33.7 to 42.5 |ig/g.  Only
 3    calcaneus lead was significantly  associated with an increase in hypertension (either > 140 mm Hg
 4    systolic or >90 mm Hg diastolic) during pregnancy, with an odds ratio of 1.86 (95% CI:  1.04,
 5    3.32) for each 10 jig/g increase of calcaneus lead.  No association between calcaneus lead and
 6    hypotension was found postpartum. The authors found the same pattern of trabecular lead
 7    concentration association with blood pressure during but not after pregnancy in normotensive
 8    women.  A 10 jig/g increase in calcaneus lead was associated with -0.75 mm Hg (95% CI: 0.04,
 9    1.46) increase in systolic and -0.58 mm Hg (95% CI: 0.01, 1.16) increase in diastolic blood
10    pressure in the third trimester. Only blood lead, tested simultaneously with tibia and calcaneus
11    lead, was significantly associated with postpartum maternal blood pressure, but the relationship
12    was negative, higher blood lead associated with lower postpartum blood pressure. For a
13    doubling in blood lead, the systolic blood pressure increased -1.05 mm Hg (95% CI:  -1.96,
14    -0.14) and diastolic increased -1.16 mm Hg (95% CI: -1.98, -0.30). Though the authors
15    thoroughly explored lead interaction with all other covariates in the models,  they were unable to
16    discover an effect modifier among them to explain the relationship.  Postpartum physiological
17    changes were discussed in relation to this last result. Thorough diagnostic testing was performed
                                                                                       rj
18    for all models. Only linear age terms were used in the models without exploration of age terms.
19    The authors did not use the repeated measures nature  of the design in their analyses, instead they
20    analyzed third trimester pregnancy data and postpartum data separately. They did not
21    statistically test differences in coefficients from the same variables in the two parts of the study.
22          Two studies examined a subset of subjects participating in the Normative Aging Study.
23    Hu et al. (1996) used a cross-sectional design of 590 men with median age in the mid-60s (range
24    48-92 years).  Blood lead ranged from 1 to 28 |ig/dL,  tibia lead from <1 to 96 jig/g, and patella
25    lead from 1 to 142 |ig/g. Logistic regression models were initially constructed by adding age,
26    race, BMI, family history of hypertension, smoking, alcohol use, and dietary sodium and
27    calcium. Testing linear blood lead, tibia lead, and patella lead one by one against hypertension
28    status (systolic >160 mm Hg, diastolic >96 mm Hg, or taking antihypertensive medication), they
29    found no significant relationships with any of the lead variables, each entered separately.  Only
30    when they used backward elimination of nonsignificant variables did they find a significant odds
31    ratio of 1.50 (95% CI:   1.09, 2.10)  for each doubling of tibia lead from the mean (20.8 |ig/g) for

      December 2005                           6-193        DRAFT-DO NOT QUOTE OR CITE

-------
 1    hypertension. Later, Cheng et al. (2001) followed up the same group, constructing a multiple
 2    linear regression model for systolic blood pressure (diastolic blood pressure was not mentioned
 3    in model descriptions) in subjects not hypertensive at baseline measurement. They used a fixed
 4    set of control variables, including age and age terms, BMI, family history of hypertension, and
 5    alcohol and calcium intake, selected by univariate and bivariate testing of a larger set.  After
 6    entering linear blood lead, tibia, and patella bone lead separately into the models, they reported a
 7    significant association only with tibia lead (1.60 mm Hg [95% CI:  0.00, 4.44] increase in
 8    systolic blood pressure for each doubling of tibia lead from the mean).  Several years later (not
 9    specified in methods but no more than 6 years), the group of subjects that was originally not
10    classified as having definite hypertension was retested for presence of definite hypertension
11    (> 160/95  mm Hg). Each lead measure was separately entered into a Cox's proportional hazards
12    model of incident definite hypertension. Only patella lead showed a significant increase in the
13    rate ratio in subjects with no history of definite hypertension, 1.14 (95% CI:  1.02, 1.28) for each
14    10 |ig/g increase in patella lead.  Similar results were obtained when the borderline hypertensive
15    group (>140/90 mm Hg) was combined with the definite hypertension group in patella lead.
16    A rate ratio of 1.23 [95% CI:  1.03, 1.48]) was estimated.  Use of linear lead terms may have
17    affected the ability of the studies to detect significant blood lead effects.
18          A pair of studies using the same group of male workers (age range 42-74 years)
19    previously exposed to organic and inorganic lead at an industrial plant in the U.S. investigated
20    the role of blood lead and bone lead on blood pressure.  Blood lead ranged between 1 and
21    20 |ig/dL and tibia lead ranged from -1.6 to 52 |ig/g. The study by Schwartz et al. (2000b)
22    controlled for age, BMI, current smoking, and current use of antihypertensive medication in
23    backward elimination linear multiple regression models for blood lead, tibia lead, and DMSA-
24    chelatable lead, forcing each lead term into separate models.  Only blood lead was a significant
25    predictor of blood pressure.  In multiple logistic regression models, only blood lead in workers
26    <58 years of age was  significant in predicting hypertension (>160/96 mm Hg). Although this
27    study used linear blood lead in one model, it used another model with both linear and squared-
28    blood lead.  Both lead terms were significant in the respective models.  In a follow-up  study
29    (Glenn et al., 2003) with most of the same subjects of the first study, subsequent measurements
30    of blood pressure occurred at intervals of 4-12 months for 10.2 months to 3.5 years.  The study
31    was notable not only for its prospective nature but in the use of statistical models adjusting for

      December 2005                           6-194        DRAFT-DO NOT QUOTE OR CITE

-------
 1    repeated measurements. Models were constructed by adding to a base model containing age at
 2    start of study, race, BMI, and indicator variables for technician. Lead variables were always
 3    forced in the models, but it is not clear if they were each tested separately. Other potential
 4    confounder variables were added stepwise to the model if they met a probability criterion. Both
 5    increasing linear blood lead and tibia lead were significantly associated with increasing systolic
 6    blood pressure times the number of years of follow-up blood measurement, but not with change
 7    in diastolic blood pressure. Each 10 jig/g increase in tibia lead was associated with a
 8    0.78 mm Hg, year (95% CI:  0.24, 1.31) increase in systolic blood pressure for workers followed
 9    for the longest time. No model diagnostics were reported.
10          Gerr et al.  (2002) tested the effect of blood lead and tibia lead only in young adults (age
11    19-29 years), both males  and females, on blood pressure. Half the subjects had grown up around
12    an active lead smelter.  Multiple linear regression  models always used age, sex, height, BMI,
13    current smoking status, frequency of alcohol consumption, current use of birth-control
14    medication, hemoglobin level, serum albumin, and income, regardless of significance levels.
15    Both blood lead (as a linear term) and bone lead (a four category ordinal variable from <1 jig/g
16    to >10 |ig/g)  were tested together. Tibia lead concentration in the highest group was associated
17    with a significant increase in both systolic (4.26 mm Hg) and diastolic (2.80 mm Hg) blood
18    pressure when compared to the lowest tibia lead group. No model diagnostics were presented.
19
20    6.5.3   Other Cardiovascular Outcomes
21    6.5.3.1    Ischemic Heart Disease
22          A community-based case-referent study taken from  the Stockholm Heart Epidemiology
23    Program compared survivors of first-time myocardial infarction with matched referents based on
24    sex, age, year of study enrollment, and hospital catchment area (Gustavsson et al., 2001). The
25    authors assessed lead exposure by a three category ordinal  scale based on lead levels in airborne
26    dust.  In the comparison of unexposed to >0-0.03 mg/m3 (mean 0.01 mg/3) and unexposed to
27    >0.04 mg/m3 (mean 0.10 mg/m3), the relative risk was 0.88 (95% CI:  0.69, 1.12) and 1.03
28    (95% CI: 0.64, 1.65), respectively.
29          In a reanalysis of the NHANES II dataset, the influence of linear blood lead in the
30    diagnosis of left ventricular hypertrophy (LVH) based on examination of electrocardiograms and
31    body habitus data in less than 9,900 subjects (exact number not given) of age 25-74 years was

      December 2005                           6-195        DRAFT-DO NOT QUOTE OR CITE

-------
 1    tested in a survey-adjusted stepwise logistic regression model (Schwartz, 1991). The final model
 2    adjusted LVH by age, race, and sex.  The odds ratio for LVH was 1.33 (95% CI:  1.20, 1.47) for
 3    each 10 |ig/dL increase in blood lead over an unreported blood lead range. The author reported
 4    no significant interactions between blood lead and race or between blood lead and sex, though
 5    the article noted that the number of cases of LVH was small. The linear lead effect had greater
 6    significance than the natural log lead effect, the reverse of the relationship between the two lead
 7    specifications usually seen when blood pressure is the outcome variable.
 8          In another study of electrocardiograms in 775 men (mean age 68 years, range 48-93) from
 9    the Normative Aging Study, patella and tibia lead concentrations were significantly associated
10    with increased heart rate-corrected QT and QRS intervals in men under 65 years but not over
11    65 years in multiple regression stepwise analysis (Cheng et al., 1998b).  Only tibia lead
12    concentration was significantly associated with an increased odds ratio of intraventricular
13    conduction deficit (2.23 [95% CI:  1.28, 3.90]) for every 10 |ig/g increase in tibia lead), but only
14    in men under 65 years. In contrast, both tibia and patella lead concentration was significantly
15    associated with atrioventricular conduction deficit (odds ratio of 1.22 [95% CI: 1.02, 1.47] and
16    1.14 [95% CI:  1.00, 1.29] for each 10 jig/g increase in tibia and patella lead, respectively), but
17    only for men greater than or equal to 65 years. None of the lead measurements were
18    significantly associated with arrhythmia.  Linear blood lead terms were not significantly
19    associated with any of the above outcomes.  Though the authors reported examining both
20    saturated models (models with all considered control and confounding variables, significant or
21    not) and stepwise models, only  stepwise models were presented or discussed with each lead term
22    forced into separate models.  Thus, each model had an individual mix of control/confounding
23    variables, though age was common to all models. Despite using age as a control/confounding
24    variable in all models,  the article offered no statistical justification for the age-stratified analysis.
25          A  group of male and female battery factory workers (n = 108) working for at least
26    10 years and who were hired from  1960 to 1983 had blood lead levels from 1970 to 1994 ranging
27    from 5 to 93 |ig/dL (Tepper et al., 2001). Using a fixed covariate multiple logistic regression
28    model, including age, BMI, sex, and family history of hypertension, the  authors found a
29    nonsignificant odds ratios for risk of hypertension  (> 165/96 mm Hg or self-reported use of
30    hypertension medications) comparing the first tertile (138-504 jig/dL'year) cumulative blood
31    lead index with the third tertile (747-1447 jig/dL'year) index. Echocardiogram left ventricular

      December 2005                           6-196       DRAFT-DO NOT QUOTE OR CITE

-------
 1    mass was not significantly related to cumulative blood lead index or time-weighted average
 2    blood lead.
 3          The discrepancy in blood lead results between the two electrocardiogram studies by
 4    Schwartz (1991) and Cheng et al. (1998b) could well be explained by population differences.
 5    Though both used large datasets, the age range of the NHANES II subject pool was between
 6    25 and 74 years and used both men and women, whereas the age range for the Normative
 7    Aging study was 48 to 93 years and used only men. Furthermore, the Cheng et al. study had
 8    775 subjects whereas the Schwartz had a much larger, though unspecified number.  The Tepper
 9    et al. (2001) study had the least number of subjects (n = 108), which may have resulted in not
10    detecting significant effects on a different measure of LVH. Nonetheless, the two
11    electrocardiogram studies each reported a significant lead effect, and the study with bone lead
12    (Cheng et al., 1998b) is particularly interesting, not only for its older sample but because the
13    bone lead exposure measure reflected accumulated past exposure, which blood lead only partly
14    reflects. The two studies are in agreement that lead exposure, either past or present, is
15    significantly associated with ischemic heart disease.
16
17    6.5.3.2   Stroke
18          No published articles relating lead specifically to stroke  were uncovered, though some
19    ICD diagnostic codes (the cerebrovascular codes) reported in other parts of this section included
20    stroke.
21
22    6.5.3.3   Cardiovascular/Circulatory Mortality
23          A recent follow-up of the NHANES II cohort provided mortality data used to associate
24    past blood lead concentration  with increased circulatory mortality in the U.S. population
25    (Lustberg and Silbergeld, 2002).  Blood lead concentration as measured during 1976-1980 was
26    divided into three categories (<10  |ig/dL, 10-19 |ig/dL, and 20-29 |ig/dL) after eliminating
27    109 subjects with blood lead >30 |ig/dL, leaving 4,190  subjects 30-74 years of age in the
28    mortality sample followed to the end of 1992.  During the follow-up period, 929 subjects died of
29    all causes. ICD-9 codes 390-459 (circulatory) accounted for 424 deaths. Proportional hazards
30    models using a priori selected potential confounding variables (age, sex, race, education, income,
31    smoking, BMI, exercise, and location) were used to calculate risk ratios of cardiovascular

      December 2005                           6-197       DRAFT-DO NOT QUOTE OR CITE

-------
 1    mortality for the two higher lead categories compared against a <10 |ig/dL reference. The
 2    20-29 ng/dL category showed a significant relative risk of 1.39 (95% CI:  1.01,1.91) for
 3    cardiovascular mortality.
 4          Another longitudinal study combined fatal and nonfatal coronary heart disease (ICD-8
 5    codes 410-414) and cardiovascular disease (ICD-8 codes 410-414 and 430-435) categories from
 6    a Danish 1936 birth cohort (N =  1052) followed from 1976-1990 (M011er and Kristensen, 1992).
 7    During the study period, 54 cases of cardiovascular disease with 19 deaths were reported.
 8    Log-transformed blood lead was used in a Cox proportional hazards model, controlling for a
 9    priori selected variables of tobacco use, cholesterol, physical activity, sex, systolic blood
10    pressure, and alcohol. Two other models were also examined, those leaving out alcohol or both
11    alcohol and systolic blood pressure. None of the adjusted models showed significant risk hazard
12    for combined fatal and nonfatal cardiovascular disease, though blood  lead was significantly
13    associated with outcome in all models except the one containing both alcohol and systolic blood
14    pressure for "total mortality" risk hazard, which presumably counted noncardiovascular
15    mortality as well (not detailed in article). This article is notable for its detailed discussion of
16    using confounding variables, such as hemoglobin and alcohol use, in multivariate models of
17    lead-cardiovascular associations. Small sample size  and low death rate may have contributed to
18    the nonsignificant results.
19          An occupational study, using 1,990 male workers who worked at least 1 day between
20    1940 and 1965 in an active lead smelter in the U.S. (mean length of employment at smelter
21    13.8 years; mean estimated length of lead exposure 9.9 years), failed to show an association with
22    lead and standardized mortality ratios compared to the U.S. population reference group up to
23    1988 (Steenland et al., 1992). Neither mortality from ischemic heart disease (ICD-9 410-414),
24    hypertension with heart disease (ICD-9 402 and 404), hypertension with no heart disease  (ICD-9
25    401, 403, and 405), nor cerebrovascular disease (ICD-9 430-438) were significantly higher in the
26    study group than in the U.S. population when examined in their totality or stratified by "high
27    lead exposure" (>0.2 mg/m3 lead in air, surveyed in 1975)  or "duration of exposure." Imprecise
28    estimation of lead exposure may have contributed to the nonsignificant results.
29          A study of 664 male workers in a Swedish lead smelter from 1942-1987 examined
30    standardized mortality ratios for  cardiovascular disease compared to the county population
31    mortality figures from 1969-1989 (Gerhardsson et al., 1995a). Blood lead measurements  were

      December 2005                            6-198       DRAFT-DO NOT QUOTE OR CITE

-------
 1    available from the workers since 1969 (mean 62.1 |ig/dL) and dropped steadily from that date to
 2    1985 (mean 33.1  |ig/dL).  The consecutive blood lead measurements in the subjects allowed
 3    construction of a  cumulative blood lead index.  Standardized mortality ratios were significantly
 4    elevated in the group for all cardiovascular diseases (ICD-8 390-458) and for ischemic heart
 5    disease (ICD-8 410-414),  1.46 (95% CI: 1.05, 2.02) and 1.72 (95% CI: 1.20, 2.42),
 6    respectively. However, there were no indications of a concentration-response relationship when
 7    analyses were stratified by cumulative blood lead index, peak blood lead, or other exposure
 8    indices.
 9          In a study  of 1,261  male newspaper linotype operators working in 1961 and followed until
10    1984, 38% had died from all causes (Michaels et al., 1991). Compared to the New York City
11    population reference group, there was a marginally significant increased standardized mortality
12    ratio in the printers of 1.35 (95% CI:  0.98, 1.82) for cerebrovascular disease (ICD-8 430-438),
13    which became highly significant in those with 30 or more years exposure (1.68 [95% CI:  1.18,
14    2.31]; 37 of the total 43 deaths due to cerebrovascular disease). Atherosclerotic heart disease
15    (ICD-8 410-414)  mortality in printers was significantly below that expected from the general
16    population, with a standardized mortality ratio of 0.63 (95% CI: 0.59, 0.73).
17          Two studies were longitudinal in nature, following the same cohort for a period from
18    12 to 16 years (Lustberg and Silbergeld, 2002; M011er and Kristensen, 1992). They both used
19    large community  cohorts (NHANES II and a Danish birth cohort, respectively) and they both
20    used multivariate proportional hazards models with blood lead as the principal predictor. Both
21    studies found a significant increase in risk ratio with increased blood lead.
22          Another study examined occupationally-exposed subjects and used population reference
23    groups to assess differences in mortality. Steenland et al. (1992) showed no significant increased
24    mortality from ischemic heart disease (ICD-9 410-414), hypertension with heart disease (ICD-9
25    402 and 404), hypertension with  no heart disease (ICD-9 401, 403, and 405), nor cerebrovascular
26    disease (ICD-9 430-438) in the study group than in the U.S. population, even in the high lead
27    exposure group.
28
29    6.5.3.4   Other  Cardiovascular Effects
30          Peripheral arterial disease (PAD), flow-limiting atherosclerosis in lower limb muscular
31    arteries, was studied using Phase 1 (1999-2000) of the NHANES IV, the most recent NHANES

      December 2005                           6-199       DRAFT-DO NOT QUOTE OR CITE

-------
 1    dataset (Navas-Acien et al., 2004). PAD was categorized as a ratio of brachial artery (arm)
 2    systolic blood pressure to posterior tibial artery (ankle) systolic blood pressure < 0.90. One
 3    hundred thirty-nine subjects were classified as having PAD; there were 1,986 subjects without
 4    the disease. Blood lead was classified by quartile, with the 1st quartile containing subjects with
 5    blood lead <1.4 |ig/dL and the 4th quartile containing subjects with blood lead >2.9 |ig/dL.
 6    Age range was 40 to >70 years.  Three sets of covariates were tested in separate models. The
 7    first set, common to all models, included age, sex, race, and education. The second set included
 8    the first set and added BMI, alcohol intake, hypertension, diabetes, hypercholesterolemia, and
 9    glomerular filtration rate. The third set added self-reported smoking status and serum cotinine.
10    Compared to first quartile blood lead, 4th quartile blood lead subjects had significant odds ratios
11    for PAD of 3.78 (95% CI:  1.08, 13.19) and 4.07 (95% CI:  1.21, 13.73) for the first two models.
12    The odds ratio of 2.88 (0.87, 9.47) for the third model was not statistically significant. However,
13    the increasing odds ratio trend from 1st through 4th quartile was significant for all 3 models
14    (p > 0.02).
15
16    6.5.4  Potential Confounding of the Cardiovascular Effects of Lead
17    6.5.4.1   Confounding by Copollutants
18          High on the list of other metals that might be associated with cardiovascular disease is
19    cadmium, through its known effects on kidney function.  If blood lead and blood cadmium
20    strongly covary in a sample by sharing a common source (e.g., when the study sample is drawn
21    from a population living near a nonferrous smelter emitting both metals), including simultaneous
22    blood lead and cadmium measurements in the same model would likely show a significant
23    reduction in both coefficients when compared to either metal alone. If, however, blood cadmium
24    and lead do not covary  in the sample, their coefficients in the model together would be much the
25    same as when tested separately.  In a study of PAD (Navas-Acien et al., 2004) discussed in
26    Section 6.5.3.4, investigators not only tested both lead and cadmium in separate models but also
27    tested them simultaneously. The correlation coefficient between natural log lead and natural log
28    cadmium was 0.32 (p < 0.001), highly significant, though leaving 90% of the variance between
29    them unexplained. In addition, they tested possible interactions between lead and cadmium, and
30    between the two metals and sex, race-ethnicity, smoking status,  renal function, and C-reactive
31    protein. Although none of the interactions were significant, when blood lead and blood cadmium

      December 2005                           6-200        DRAFT-DO NOT QUOTE OR CITE

-------
 1    were in the same model together they both had significant trends of increasing odds ratios with
 2    increasing quartile of each metal, but the nonsignificant point estimate of the odds ratio for blood
 3    lead comparing 1st and 4th lead quartile tested alone dropped further when tested with cadmium
 4    (odds ratio of 2.88 versus 2.52). Cadmium between 1st and 4th quartile, on the other hand,
 5    showed a similar drop from cadmium tested alone to cadmium tested with  lead (odds ratio of
 6    2.82 versus 2.42), but both point estimates remained significant. Thus, though point estimates of
 7    both lead and cadmium were approximately the same whether tested alone or together, the larger
 8    variance associated with the lead coefficients rendered them nonsignificant. Part of the
 9    difference in variance between the two metals could be explained by noting that the reference
10    group (lowest quartile) for lead contained a little over half the number of subjects (n  = 472; 18
11    cases, 454 noncases) than the reference group for cadmium (n = 856; 27 cases, 829 noncases).
12    The odds ratios for PAD with smoking status dropped from 4.13 (95% CI:  1.87, 9.12) to 3.38
13    (95% CI: 1.56, 7.35) when lead was added to the model, but both odds ratios remained highly
14    significant and the difference was not statistically tested. The failure to find a significant
15    interaction between the two metals and between smoking status and both metals suggests that
16    none of the odds ratio changes discussed above were significant.  The same pattern of results was
17    found when using cotinine blood levels instead of self-reported smoking habit. Adding cadmium
18    alone or cadmium and lead together resulted in nonsignificant odds ratios for both indices of
19    smoking.
20          The Belgian Cadmibel studies also were ideally situated to test possible interactions
21    between blood lead and cadmium, but the technique of stepwise addition of variables to the
22    multiple regression models of blood pressure did not allow retention of both metal variables
23    together in the same model (Staessen et al., 1996).  From the lack of both cadmium and lead in
24    any one model, it can be inferred that, if both variables had been forced into the  model together,
25    they both would have had nonsignificant coefficients.
26
27    6.5.4.2    Confounding by Smoking Status
28         Most studies reviewed in this  section have controlled for tobacco use, where it often
29    appears related to lower blood pressure.  The majority of reviewed studies including  smoking as
30    a covariate never present the coefficients of smoking or related covariates. Only the  Navas-
      December 2005                           6-201        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Acien et al. (2004) study discussed in the previous section systematically addressed the issues
 2    related to possible confounding or effect modification with tobacco use.
 3
 4    6.5.4.3   Confounding by Alcohol Consumption
 5          Possible confounding by alcohol use, generally associated with increased blood pressure,
 6    was thoroughly discussed in the 1990 Supplement (Grandjean et al., 1989). Alcohol, especially
 7    in Europe, contained substantial lead during much of the 20th century.  This can be seen in the
 8    MONICA Augsberg, Germany cohort study (Hense et al., 1994).  The study group was stratified
 9    by sex and then, only in men, by rural-urban location. Within each strata, the blood lead range
10    differed by alcohol use. In women, for example, the 10th and 90th percentile values of blood
11    lead were approximately (as estimated from graphs) 3.5 and 8.5 |ig/dL for self-reported
12    abstainers, 4.5 and 10.5 |ig/dL in those drinking from 1 to 39 g/day, and 6.0 to 14.0 |ig/dL in
13    those drinking 40 plus g/day. Despite the finding that only women in the highest alcohol-use
14    group had a significant lead effect, it cannot be determined if the increase in lead coefficient is
15    significant because the three coefficients associated with use of alcohol strata were not tested for
16    differences among themselves; they were only tested for their significance from the null
17    hypothesis of 0. Another study was based on  subjects from the New Risk Factors Survey from
18    the area around Rome, intended to determine confounding effects of a number of social and
19    biochemical variables on the blood lead-blood pressure relationship (Menditto et al., 1994).
20    Alcohol consumption, as well as BMI, heart rate, non-HDL cholesterol, and HDL cholesterol,
21    triglycerides, cigarettes smoked/day, and skinfold thickness were all examined. A doubling  of
22    blood lead was associated with an increase of 4.71 mm Hg in systolic and  1.25 mm Hg in
23    diastolic blood pressure.  Alcohol as a true confounding variable is likely limited to studies in
24    areas where alcohol contributes significantly to blood lead. In a study of 249 bus drivers in San
25    Francisco, CA, natural log lead coefficients against blood pressure changed less than 10% when
26    alcohol use was included as a covariate (Sharp et al., 1990). Blood lead according to alcohol use
27    was not reported. Another study based on a U.S. population found a significant increase in blood
28    lead of a mixed group of males and females according to alcohol use, ranging from mean blood
29    lead of 7.3 |ig/dL in nonusers to 9.2 |ig/dL in those reporting more than 2 ounces/day over 3 days
30    (Morris et al., 1990), with no report of significant effects of alcohol on blood pressure.
31

      December 2005                           6-202        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.5.4.4   Confounding by Dietary Calcium Intake
 2          The main thrust of the previously reported Morris et al. (1990) study was to examine the
 3    effects of dietary calcium on the effect of lead on blood pressure in 78 males and 64 females
 4    between 18 and 80 years, many of whom were hypertensive (undisclosed number), though those
 5    using medications for hypertension discontinued their use 1 month before testing started.
 6    Subjects were excluded if they had "secondary hypertension." The investigators measured
 7    serum calcium and assessed dietary calcium intake, among other variables.  There were no
 8    changes in blood lead or blood pressure noted as a result of dietary calcium supplementation.
 9          Proctor et al. (1996), using the Normative Aging Study, examined possible modification
10    of the effect of natural log blood lead (blood lead range 0.5-35 |ig/dL) on blood pressure in
11    798 men aged 45-93 years by dietary calcium intake assessed by food questionnaire. The study
12    used multiple regression models with a fixed set of covariates, including age and age2, BMI,
13    adjusted dietary calcium, exercise, smoking, alcohol use, sitting heart rate, and hematocrit.
14    Increased blood lead was significantly associated with diastolic blood pressure and systolic blood
15    pressure.  Only systolic blood pressure significantly decreased with increased dietary calcium
16    (0.004 mm Hg decrease for every 1  mg/day increase of dietary calcium).  The authors formed
17    dichotomized calcium intake  (cut point at 800 mg/day) and blood lead (cut point at  15 |ig/dL)
18    variables to test the interaction between blood lead and calcium on blood pressure.  They did not
19    find a significant interaction.
20          A study of a subset of the Cadmibel Study with 827 males and 821 females, age 20 to
21    88 years, selected from areas  known to represent a wide range of cadmium exposure, specifically
22    studied total serum calcium interactions with blood lead on blood pressure (Staessen et al.,
23    1993). Stepwise regression models, selecting from log blood lead, age and age2, BMI, pulse rate,
24    log serum gamma-glutamyltranspeptidase, serum calcium, log serum creatinine, urinary
25    potassium, smoking, alcohol intake, contraceptive pill use (females only), and a menopause
26    indicator variable (females only),  were stratified by sex for systolic and diastolic blood pressure.
27    The stepwise procedure resulted in models each with a different mix of covariates.  Increased
28    serum calcium was significantly associated with increased systolic blood pressure in both males
29    and females. Every increase of one log unit of blood lead was associated with nonsignificant
30    changes in blood pressure in women but with a significant decrease in systolic blood pressure in
31    men (systolic log blood lead P = -5.2). A separate set of models were constructed with an

      December 2005                           6-203        DRAFT-DO NOT QUOTE OR CITE

-------
 1    interaction term between serum calcium and log blood lead (details not shown). In women only,
 2    both main effects of lead and calcium and the interaction effect were significant (no coefficients
 3    presented).  At the 25th percentile of serum calcium (2.31 |imol/L), a doubling of blood lead was
 4    associated with a 1.0 mm Hg increase in systolic blood pressure.  At  the 75th percentile of serum
 5    calcium (2.42 |imol/L) a doubling of blood lead was associated with  a 1.5 mm Hg increase in
 6    systolic blood pressure. Furthermore, serum calcium may itself be confounded with age in
 7    women, as women showed a sharp rise in serum calcium in their sixth decade of life, coincident
 8    with menopause, whereas the trend for serum calcium in men was steadily downward for each
 9    subsequent decade of age. The authors did not test an interaction term including calcium and age
10    or calcium and menopausal status.  Thus, the significant interaction effect between calcium and
11    lead on blood pressure may be a result of differences due to menopause.
12
13    6.5.4.5   Summary of Potential Confounding of the Lead Effect  on Cardiovascular Health
14          The effects of cadmium exposure, smoking, alcohol use, dietary and serum calcium levels
15    have all been formally tested in a few studies, without significant effects as confounders of the
16    lead effect.  Failure to find a significant confounding effect with lead, however, does not argue to
17    maintain these variables uncritically in models of blood pressure.  If  alcohol contains lead,
18    increased alcohol use will lead to increased blood lead.  In this case,  both variables in the model
19    will be collinear and this tends to distort estimated coefficients and standard errors of their effect
20    on cardiovascular outcome.  Tobacco use may influence lead levels much more in occupational
21    studies than in community exposure studies, especially if smoking in the factory is allowed.
22    Frequent hand to mouth behavior will increase lead exposure and, consequently, raise blood lead
23    concentrations. Serum calcium may statistically modify the lead effect differentially by gender
24    due to menopause in women. Menopause also affects lead turnover.  If serum calcium, blood
25    lead, and blood pressure are all statistically related, serum calcium should not be used in blood
26    lead-blood pressure/hypertension studies.
27
28    6.5.5   Gene-Lead Interactions
29          Study authors characterized sodium-potassium adenosine triphosphatase a2 (ATP1A2)
30    polymorphism in 220 workers  formerly exposed to a mix of organic  and inorganic lead in the
31    U.S., noted above in other references (Glenn et al., 2001).  The ATP1A2 (3') one kilobase probe

      December 2005                           6-204       DRAFT-DO NOT QUOTE OR CITE

-------
 1    produced two homozygous (4.3/4.3 and 10.5/10.5) and one heterozygous (4.3/10.5) genotypes
 2    and two homozygous (8.0/8.0 and 3.3/3.3) and one heterozygous (8.0/3.3) genotypes for the
 3    2.5 kilobase ATP1A2 (5') probe. Of the 209 subjects with data on both polymorphisms, 43.5%
 4    were doubly homozygous for 8.0/8.0 and 4.3/4.3, 34.4% were homozygous for 8.0/8.0 and
 5    heterozygous for 4.3/10.5, 11.5% were heterozygous for 8.0/3.3 and homozygous for 4.3/4.3,
 6    5.3%. Also, 5.3% were doubly homozygous for 8.0/8.0-10.5/10.5, and 4.8% were doubly
 7    heterozygous for the two genotypes. Although only 13 African American workers participated,
 8    prevalence of the 10.5 kilobase allele in the ATP1A2 (3') genotype was statistically higher for
 9    them than for other races.  Prevalence of hypertension (> 160/96 mm Hg or use of hypertension
10    medication) was significantly higher in those with the 10.5/10.5 genotype than in others.
11    Controlling for age, BMI, lifetime number of alcoholic drinks, the  10.5/10.5 genotype was
12    associated with an odds ratio of 7.7 (95% CI: 1.9, 31.4) for hypertension when compared to the
13    4.3/4.3 homozygous genotype, but there were no effects of either blood lead, tibia lead, or their
14    interaction with ATP1A2 (3') genotype. A multiple linear regression model for linear blood lead
15    and systolic blood pressure, controlling for age, use of hypertensive medication, current
16    smoking, quartiles of lifetime alcohol consumption, and season, showed a significant main effect
17    for 10.5/10.5 homozygous contrasted against combined 4.3/4.3 and 4.3/10.5 groups, associated
18    with a 25.5 mm Hg reduction in blood pressure, primarily due to limited blood lead range of the
19    homozygous group (maximum blood lead of the 10.5/10.5 group 9 |ig/dL; maximum blood lead
20    of the contrast group = 20 jig/dL). But the interaction between linear blood lead and the
21    10.5/10.5 condition resulted in a significant increase of the blood lead effect on blood pressure
22    by 5.6 mm Hg for every 1 |ig/dL blood lead compared to the blood lead effect in the other
23    genotypes. The authors stated, but did not show analysis or coefficients, that the ATP1 A3 (3')
24    polymorphism also significantly interacted with tibia lead and systolic blood pressure. There
25    were no significant relationships using the ATP1A2 (5') gene. Thus, the ATP1A2 (3')
26    polymorphism appears to directly influence both prevalence of hypertension and the effect of
27    lead on blood pressure, though the small group (n = 9 with all measures) with the important
28    10.5/10.5 homozygous pattern would argue for enlarging this important study.
29          Another research group focused on polymorphisms of two genes suspected to be involved
30    in lead toxicokinetics, the vitamin D receptor (VDR) and delta-aminolevulinic acid dehydratase
31    (ALAD) (Lee et al., 2001). Polymorphism of both genes is well studied and prevalence appears

      December 2005                          6-205        DRAFT-DO NOT QUOTE OR CITE

-------
 1    associated with race or ethnic background.  Nearly 800 Korean workers aged 18-65 years
 2    (79.4% males) from lead-using businesses were classified according to ALAD polymorphism
 3    (1-1 [homozygous] versus 1-2 [heterozygous]) and VDR polymorphism (bb [predominant
 4    homozygous] versus Bb plus BB [infrequent polymorphisms]). The homozygous 1-1 ALAD
 5    polymorphism was found in 90.1% of the group and the homozygous bb polymorphism was
 6    found in 88.8% of the group.  When compared to a smaller group of non-lead-exposed workers,
 7    blood lead concentration (mean exposed 32.0 |ig/dL [range 4-86] mean nonexposed 5.3 |ig/dL
 8    [range 2-10]  and tibia lead concentration mean exposed 37.2 |ig/g [range -7-338]; and mean
 9    nonexposed 5.8 |ig/dL [range -11-27]) were much higher. The study used stepwise multiple
10    regression models, selecting covariates remaining significant in the models from among a large
11    set of potential control and confounding variables.  They also allowed potential confounders to
12    remain in the models if "there were substantive changes in the coefficients of predictor
13    variables" with their addition. Systolic models controlled for age and age2, sex, BMI,
14    antihypertensive medication use, and cumulative lifetime alcohol use.  Depending on the
15    presence or absence of linear blood lead, tibia lead, and DMSA chelatable lead in the models,
16    and the gene-age interactions tested, blood urea was added to the model. Diastolic models
17    controlled for age, sex, BMI, cumulative alcohol consumption, and linear blood lead.
18    Hypertension (systolic >160 mm Hg or diastolic >96 mm Hg) logistic multiple regression
19    models controlled for age, sex, BMI, tibia lead, and current alcohol use. Among the exposed
20    workers bb VDR genotypes had significantly lower DMSA-chelatable blood lead and lower
21    diastolic and systolic blood pressure than the  combined Bb and BB genotypes.  The only
22    significant interaction reported between predictor variables and gene polymorphism on blood
23    pressure was with the VDR polymorphism bb allele, who had a less pronounced increase in
24    systolic blood pressure with age than subjects with the B allele.  There were only marginally
25    significant associations of systolic blood pressure with tibia lead and linear blood lead. There
26    were no significant associations in models of diastolic blood pressure with linear blood lead,
27    DMSA-chelatable blood lead, or tibia lead. Tibia lead was significantly associated with
28    hypertension (odds ratio of 1.05 [95% CI: 1.00, 1.12] for each 10 |ig/dL increase in tibia lead).
29    Workers with VDR B allele had significantly higher prevalence of hypertension (odds ratio = 2.1
30    [95% CI:  1.0, 4.4]) than workers with the bb genotype, but no other lead variable or interaction
31    with VDR status was reported significant. Though VDR status was significantly related to blood

      December 2005                            6-206       DRAFT-DO NOT QUOTE OR CITE

-------
 1    pressure and prevalence of hypertension, there were no significant effects of ALAD
 2    polymorphism on blood pressure or hypertension or of VDR interactions with any lead exposure
 3    variable.
 4          Lustberg et al. (2004) studied these same Korean lead workers (n = 793) to examine the
 5    relationships between the Q894_x894 polymorphism in the gene regulating endothelial nitric oxide
 6    synthase (eNOS) and blood lead effects on blood pressure and hypertension. Nitric oxide
 7    metabolism has been suggested both as a mechanism for altered blood pressure and for
 8    moderating the effects of lead on blood pressure, though there is experimental support for and
 9    against both hypotheses. After classifying subjects as  homogenous for the GG type (85%),
10    heterogeneous for both types (TG) (14%), or homogenous for TT (1%), the TG and TT types
11    were combined into a single group (TG/TT). Diastolic and systolic multiple regression models
12    were constructed with a fixed set of covariates, including smoking, alcohol consumption, age,
13    sex, BMI, and education. Logistic regression models used blood pressure criteria of either
14    > 140 mm Hg diastolic blood pressure, >90 mm Hg systolic blood pressure, or self-report of
15    using antihypertensive medications.  There was no effect of genotype on diastolic or systolic
16    blood pressure or on hypertension prevalence in multiple regression models, nor any significant
17    interaction of lead exposure indices with gene status.
18
19    6.5.6   Summary of the Epidemiologic Evidence for the Cardiovascular
20            Effects of Lead
21          The combined blood lead studies using blood pressure/hypertension as an  outcome
22    continue to support the conclusions of the 1990 Supplement that there is a positive association
23    between blood lead and increased blood pressure. The occasional finding of significant negative
24    associations of blood lead with blood pressure (e.g., the Cadmibel  study, one NHANES III study,
25    the postpartum phase of the Los Angeles pregnancy study) have not been adequately explained
26    and require further confirmation and study.  The  reported meta-analysis succinctly characterizes
27    the blood pressure findings with blood lead: 1.0 mm Hg systolic pressure increase with each
28    doubling of blood lead; 0.6 mm Hg diastolic pressure increase with each doubling of blood lead.
29    Although females often show lower lead coefficients than males, and blacks higher lead
30    coefficients than whites, where these differences have  been formally tested, they  are usually not
31    statistically significant. The tendencies may well arise in the differential lead exposure in these

      December 2005                           6-207       DRAFT-DO NOT QUOTE OR CITE

-------
 1    strata, lower in women than in men, higher in blacks than in whites.  The same sex and race
 2    differential is found with blood pressure.
 3         The most promising developments in this field since the 1990 Supplement have been the
 4    use of bone lead as a long-term cumulative lead exposure index and the introduction of genetic
 5    analysis into the studies as potential lead effect modifiers.  With one exception, all studies using
 6    bone lead have found a consistently positive and significant effect on blood pressure and/or
 7    hypertension. The ability to estimate past exposure in cross-sectional studies is a significant
 8    advance.  The results of the bone lead studies to date highlight the important role of accumulated
 9    lead exposure in the development of cardiovascular problems.
10         Though the study of genetic polymorphisms is still in its infancy in this field,  it too holds
11    great promise in accounting for individual variability in development of cardiovascular disease
12    and individual response to lead exposure.
13
14
15    6.6    REPRODUCTIVE AND DEVELOPMENTAL EFFECTS OF LEAD
16    6.6.1   Summary of Key Findings of the Reproductive and Developmental
17           Effects of Lead from  the  1986 Lead AQCD
18         Lead has been implicated as a risk factor for reproductive outcomes for over a century
19    Rom, 1976; Oliver, 1911). As early as 1860,  increased rates of stillbirths and spontaneous
20    abortions were found in women with occupational exposure to lead (usually in the ceramics
21    industry) and in women with husbands employed in the lead industry, compared to unexposed
22    worn en (Rom, 1976). Other early investigations found increased rates of physically  and
23    mentally "retarded" offspring among these same groups. In 1910, these findings resulted in the
24    first lead-related occupational regulation; the British Committee on Occupational Health
25    recommended that women not be employed in the lead industry (Oliver, 1911). These
26    observations, however, were based on exposure levels far above those considered acceptable
27    today, and current research now focuses on substantially lower exposure levels.
28         The 1986 Lead AQCD provided  evidence that lead, at high exposure levels, exerted
29    significant adverse health effects on male reproductive functions.  Several studies observed
30    aberrations in both sperm  count and morphology in men occupationally exposed to relatively
31    high levels of lead (blood lead levels of 40-50 jig/dL). However, the effects of lead on female

      December 2005                           6-208        DRAFT-DO NOT QUOTE OR CITE

-------
 1    reproductive function and fetal growth were suggestive but equivocal, perhaps due to the small
 2    sample sizes and inadequate controlling for potential confounding factors.
 3          This section provides a critical review of the literature regarding the associations between
 4    exposure to environmental lead and reproductive outcomes. First, the evidence for the placental
 5    transfer of lead is reviewed; this is key to providing a basis and mechanism for fetal exposure.
 6    Second, the association between exposure and each outcome is reviewed. Outcomes of interest
 7    are reproductive function (fertility), spontaneous abortion, fetal growth, preterm delivery, and
 8    congenital anomalies.  Each section below begins with a summary of the literature up to 1986,
 9    the year of the last EPA Air Quality Criteria Document.  Then, key studies are reviewed and
10    each section ends with a conclusion based on the evidence provided.  The conclusion is based on
11    the generally accepted "Causal Criteria" for bodies of epidemiologic literature (Hill,  1965;
12    Susser, 1991).
13
14    6.6.2   Placental Transfer of Lead
15          In 1968, Barltrop (1968) demonstrated that  lead crosses the placenta beginning as early as
16    gestational week 12. He found that the rate of transfer subsequently increased to term. Lead
17    accumulations were found in the bones, livers, blood, hearts, kidneys, and brains of stillborn and
18    spontaneously aborted fetuses.  These observations were replicated by numerous investigators;
19    for example, Casey and Robinson (1978) found lead accumulations in the livers, kidneys, and
20    brains of stillborn fetuses. Lead accumulations were also found in the livers, brains and kidneys
21    of first trimester abort fetuses (Chaube et al., 1972), suggesting placental transfer earlier than
22    12 weeks of gestation.  Newer findings, published  since 1986,  are reviewed below (also see
23    Section 6.2.2.5.2).
24          Placental transfer of lead is confirmed by correlations of maternal blood lead
25    concentrations, umbilical cord blood lead, and placental lead concentrations in a variety of
26    settings. Umbilical cord blood reflects fetal blood. Early studies, prior to 1986, found
27    correlation coefficients between maternal and umbilical  cord blood lead ranging from 0.5 to 0.8,
28    all of which were highly statistically significant. More recent studies also find significant
29    correlations between maternal and fetal blood lead. For example, a prospective study in Kosovo,
30    Yugoslavia recruited 1,502 women at mid-pregnancy in two towns — one with high exposure
31    due to the presence of a lead smelter, refinery, and battery plant, and one with relatively low

      December 2005                           6-209        DRAFT-DO NOT QUOTE OR CITE

-------
 1    exposure. The correlation between maternal blood lead (either at delivery or at mid-pregnancy)
 2    and cord blood lead ranged from 0.8 to 0.9 (Graziano et al., 1990).  Among women with
 3    substantially lower levels of exposure (e.g., blood lead 1.9 |ig/dL) the correlation between
 4    maternal and cord blood lead was 0.79 (Harville et al., 2005).
 5          Chuang et al. (2001) propose that while maternal and cord whole blood lead are highly
 6    related, fetal exposure may be even more influenced by maternal plasma lead.  Using data from a
 7    cohort of 615 women in Mexico City recruited in 1994-1995, these investigators used structural
 8    equation modeling to estimate the associations between whole blood lead, bone lead (cortical and
 9    trabecular), and the latent variable, plasma lead and cord blood lead. They found the strongest
10    associations between whole blood lead and cord blood lead, even after accounting for plasma
11    lead.  The greatest contributors to plasma lead were bone lead and airborne lead. However, with
12    declining exogenous lead exposure, these investigators note that the measurement of plasma and
13    bone lead may become increasingly important in assessing fetal exposure.
14          These data provide little doubt of fetal exposure to lead via placental transport. Further, it
15    appears that lead crosses the placenta throughout pregnancy, leading to continual exposure of the
16    fetus. Indeed, there is evidence to suggest  that maternal blood leads during the later half of
17    gestation increase (Gulson et al., 2004; Hertz-Picciotto et al., 2000; Rothenberg et al., 1994;
18    Sowers et al., 2002).  The  magnitude  of the increase ranges from 14-40%, possibly due to the
19    different starting blood leads in each study (Bellinger, 2005). The increase in blood lead in the
20    later  half of pregnancy may result from physiologic changes in maternal homeostasis during
21    pregnancy and, in particular, to mobilization of lead stores from other body organs (Bellinger,
22    2005). Indirect evidence for such mobilization comes from the increased rate of bone turnover
23    during the later half of gestation, prompted by the increased fetal need for calcium  (Moline et al.,
24    2000). Thus, both the epidemiological evidence and the biological plausibility of the
25    associations support the role of maternal-fetal transfer of lead.
26          Additionally, in populations with  greater lead burdens, the fetus may be at even greater
27    increased risk for exposure and possible  adverse effects of exposure. Among the variables
28    associated with lead exposure in pregnant (and nonpregnant) women are:  smoking and alcohol
29    consumption (Graziano et al., 1990; Rhainds and Levallois, 1997), pica (Rothenberg et al.,
30    1999), use of ethnic remedies and cosmetics (Al-Ashban et al., 2004; Centers for Disease
31    Control and Prevention, 1993), and food preparation in inappropriately lead-glazed pottery

      December 2005                           6-210        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Azcona-Cruz et al., 2000; Rothenberg et al., 2000). There is some evidence that low calcium
 2    intake is also associated with higher blood lead (Gulson et al., 2004; Hernandez-Avila et al.,
 3    2003; Hertz-Picciotto et al., 2000). Finally, the location where the mother resides (or resided as
 4    a child) may increase blood lead (Graziano et al., 1990). Blood leads are elevated among U.S.
 5    immigrants, especially those who migrated from countries where lead is still used as a gasoline
 6    additive (Centers for Disease Control and Prevention, 2000); indeed, blood leads are inversely
 7    associated with the number of years since migration (Centers for Disease Control and
 8    Prevention, 2000; Klitzman et al.,  2002; Rothenberg et al., 1999).
 9          In conclusion, the epidemiologic evidence indicates that lead freely crosses the placenta,
10    resulting in continued fetal exposure throughout  pregnancy.  Indeed, the evidence is strong that
11    exposure increases during the later half of pregnancy.  Exposure to the fetus is more pronounced
12    in high-risk populations, especially those who migrated from countries still using lead as a
13    gasoline additive.
14
15    6.6.3    Effects of Lead on Reproductive Function
16    6.6.3.1   Effects  on Male Reproductive Function
17          Male reproductive function is measured using the reproductive history  of the male (i.e.,
18    number of pregnancies fathered), time to pregnancy and direct measures of semen quality
19    (usually sperm count,  motility and morphology).  Most studies relating lead exposure to male
20    reproductive function are based on data collected in the occupational setting linked to population
21    birth registries and on studies directly collecting  questionnaire exposure and outcome data.
22
23    6.6.3.1.1   Sperm Count, Motility and Morphology
24          Recent publications which purport a decline in sperm concentration, motility, and
25    morphology seek the explanation in the rising use of man-made chemical endocrine disrupters
26    (Auger et al., 1995; Fisch et al., 1997; Farrow, 1994; Gyllenborg et al., 1999;  Kavlock et al.,
27    1996; Keiding et al., 1994; Kieding and Skakkebaek, 1996; Lerchl, 1995; Olsen et al., 1995;
28    Sherins, 1995). Several studies from the 1970s and early 1980s suggest aberrations in both
29    sperm count and morphology in men exposed to  relatively high levels of lead. In the earliest
30    study,  Lancranjan  et al. (1975) found decreased sperm counts and an increased prevalence of
31    morphologically abnormal sperm among workers heavily exposed to lead (mean blood lead

      December 2005                            6-211       DRAFT-DO NOT QUOTE OR CITE

-------
 1    74.5 |ig/dL) as well as those moderately exposed (mean blood lead 52.8 |ig/dL). These findings
 2    have been corroborated by results of studies in the U.S. (Cullen et al., 1984) and Italy (Assennato
 3    et al., 1986) which describe similar effects in workers with blood leads above 60 |ig/dL.
 4          More recently, corroborating data was described in a comprehensive review by Apostoli
 5    et al. (1998). In studies of men with blood leads above 40 |ig/dL, decreases in sperm count and
 6    concentration, motility and morphologic aberrations were found. Chowdhury et al. (1986) found
 7    a significant decrease in sperm count and motility and  an increase in the number of sperm with
 8    abnormal morphology in 10 men with occupational lead exposure; the average blood lead in the
 9    exposed group was 42.5 |ig/dL compared to 14.8 |ig/dL in the unexposed.  Similar results were
10    found in a group of 30 lead-exposed factory workers compared to controls (Lerda, 1992). In a
11    large study of male lead smelter workers, Alexander et al. (1996a) found a decreasing trend of
12    sperm concentrations with increasing lead exposure. In this cohort,  152 workers provided blood
13    specimens and 119 also provided semen samples.  Geometric mean  sperm  concentrations were
14    79.1, 56.5, 62.7, and 44.4 million cells/mL for blood leads of <15, 15-24, 25-39, and >40 |ig/dL,
15    respectively. Long-term body lead burden was estimated from current blood lead concentrations
16    and historical blood lead monitoring data.  Using this measure of long-term lead body burden, a
17    similar trend was found for sperm concentration, total  sperm count,  and total motile sperm count.
18    No associations were found for sperm morphology or serum concentrations of reproductive
19    hormones.  A study of traffic police in Peru, where leaded gasoline is still in use, found decreases
20    in sperm morphology, concentration, motility and viability among men with blood lead
21    >40 |ig/dL compared to men with blood lead <40 |ig/dL.
22          Using data from an international study of 503 men employed in the  lead industry,
23    Bonde et al. (2002) considered the lowest adverse effect level associated with perturbed semen
24    parameters.  Median sperm concentration was reduced by 49% in men with blood lead
25    >50 |ig/dL; regression analysis indicated a threshold value of 44 |ig/dL. These investigators
26    conclude that adverse effects on sperm quality were  unlikely at blood leads <45 |ig/dL.
27          In a population of couples undergoing either artificial insemination  or in vitro fertilization,
28    Benoff et al. (2003a,b) found higher concentrations of lead in seminal fluid in the male partner
29    among couples who did not conceive, compared to those who did conceive. While not directly
30    measuring the adverse effects of lead on sperm per se,  these data suggest a possible mechanism
31    for the transfer of lead from paternal exposure to the fetal environment.  Hernandez-Ochoa et al.

      December 2005                           6-212        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (2005) also provide evidence that lead concentrations in seminal fluid may be a better indicator
 2    of exposure than blood lead. Mean blood lead in this sample was lower than in most other
 3    studies, 9.3 |ig/dL.  Decreases in sperm concentration, motility, morphology, and viability were
 4    correlated with seminal fluid lead or lead in spermatozoa, but not with blood lead.
 5          Overall, the available evidence suggests a small association between exposure to lead,
 6    usually in the workplace, and perturbed semen quality. It appears that sperm count and
 7    morphology (% normal forms) may be decreased at exposures >45 |ig/dL. Future research
 8    should focus on studies of men exposed to lower levels of lead, as exposures in the very high
 9    range are associated primarily with occupational exposure.  These studies should also account for
10    variables known to be associated with semen quality and which may also be associated with
11    exposure, e.g., social class, other environmental exposures such as heat and vibration, and
12    lifestyle variables such as cigarette smoking and alcohol use.
13
14    6.6.3.1.2  Time to Pregnancy
15          Time to pregnancy represents a sensitive measure of fecundity.  Time to pregnancy is
16    important because it measures the end effect of perturbed reproductive function. While it is
17    important and necessary to understand the associations between prenatal exposures and
18    endocrine abnormalities and semen characteristics, they  represent possible antecedents to the
19    occurrence of pregnancy. Previous reports demonstrate  good validity and reliability for reports
20    of time to pregnancy in both males and females and when time to recall has been both long and
21    short (Weinberg et al., 1993, 1994).
22          One advantage to the use of this parameter, as compared to just an infertility measure, is
23    that it does not require categorization of men into fertile and infertile groups. Among couples
24    that succeed in establishing pregnancy, there is considerable variability in the time between
25    discontinuation  of contraception and conception (Weinberg et al., 1994). With the possible
26    exception of cigarette smoking and age, very little is known regarding this intercouple
27    variability. Delays in time  to pregnancy may be indicative of a range of reproductive
28    abnormalities of both partners, including impaired gametogenesis, hormonal  disruptions, and
29    very early unrecognized pregnancy loss. Time to pregnancy has the menstrual cycle as its
30    natural unit and is thus measured in integer units of menstrual cycles.
      December 2005                           6-213       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Usually, time to the most recent pregnancy is taken as the outcome (Baird et al., 1986).
 2    The measure of exposure in these studies usually is the fecundity density ratio, which is similar
 3    to an incidence density ratio. Fecundity density ratios can be interpreted as the risk of pregnancy
 4    among the exposed during an interval, compared to the risk of pregnancy among the unexposed
 5    during the same interval.  In such studies, the intervals of interest are menstrual cycles.
 6    Fecundity density ratios less than one indicate reduced fecundity (i.e., longer time to pregnancy)
 7    among the exposed compared to the unexposed, while those greater than one indicate enhanced
 8    fecundity (i.e., shorter time to pregnancy) in the exposed.  Usually fecundity density ratios are
 9    calculated using discrete time Cox proportional hazards regression models.
10          Several recent studies evaluate time to pregnancy when the male partner is occupationally
11    exposed to lead. The Asclepios Project, a large European collaborative cross-sectional study,
12    evaluated time to pregnancy in 1,108 men of whom 638 were exposed to lead (Joffe et al., 2003).
13    The reference group consisted of lead workers for whom exposure did not coincide with time of
14    pregnancy.  The investigators only included pregnancies which resulted in live births.  Fecundity
15    density ratios were 1.12 (95% CI:  0.84, 1.49), 0.96 (95% CI: 0.77, 1.19), 0.88 (95% CI:  0.70,
16    1.10) and 0.93 (95% CI: 0.76,  1.15) for blood leads <20, 20-29, 30-39, and  >40  |ig/dL,
17    respectively. These results indicate that no association was found between blood lead and
18    delayed time to pregnancy. Similar results were found when duration of exposure or cumulative
19    exposure was used as the exposure metric.
20          A separate report was published in the Italian group of men included  in the Asclepios
21    project (Apostoli et al., 2000).  Blood lead at the time closest to conception was used as the
22    measure of exposure.  Lead-exposed men (n = 251) who had experienced at least one completed
23    pregnancy were compared to nonexposed men (n = 45). Contrary to what was expected, time to
24    pregnancy was  significantly shorter among couples in which the male partner was exposed to
25    lead compared to those in which the male partner was not exposed. In secondary analyses, time
26    to pregnancy was longer among men with the highest blood lead (i.e., >40 |ig/dL). Limiting the
27    analysis solely to exposed men, time to pregnancy was longer among men with higher blood
28    leads.
29          Among 502 couples identified by Sallmen (2000) from the Finnish Institute of
30    Occupational Health in which the male partner was exposed to lead, time to pregnancy was
31    reduced among those with blood leads >10  |ig/dL compared to those with blood leads

      December 2005                          6-214        DRAFT-DO NOT QUOTE OR CITE

-------
 1    < 10 |ig/dL.  However, when blood lead was stratified, no concentration-response relationship
 2    was found. Fecundity density ratios were 0.92 (95% CI:  0.73, 1.16), 0.89 (95% CI: 0.66, 1.20),
 3    0.58 (95% CI: 0.33, 0.96) and 0.83 (95% CI:  0.50, 1.32) for exposures of 10-20, 21-30, 31-40,
 4    and >40 |ig/dL, respectively.  In this study, blood leads close to the time of conception were
 5    available on 62% of men, while in 38% it was estimated using blood leads obtained at other
 6    points or based on job descriptions.
 7          Among 280 pregnancies in 133 couples in which the male partner was employed in a
 8    battery plant, 127 were conceived during exposure while the remainder conceived prior to
 9    exposure (Shiau et al., 2004).  Time to pregnancy increased with increasing blood lead,
10    especially when  blood leads were >30 |ig/dL.  Fecundity density ratios were 0.50 (95% CI:
11    0.34, 0.74) and 0.38 (95% CI: 0.26, 0.56) for blood leads 30-39 and >39 |ig/dL, respectively.
12    In 41 couples, one pregnancy occurred prior to exposure and one during exposure - time to
13    pregnancy during exposure was significantly longer. Of note, this is the only study to estimate
14    decreases in time to pregnancy when blood lead was below 40 |ig/dL; time to pregnancy
15    increased by 0.15 months for each 1 |ig/dL increase in blood lead between 10 and 40 |ig/dL.
16
17    6.6.3.1.3   Reproductive History
18          Population-based birth registries in the  Scandanavian countries provide data on medically
19    diagnosed pregnancies.  These registries provide a basis for linking occupational data on lead
20    exposure obtained by place and duration of employment or by direct measures of blood lead
21    relative to the timing of marriage or conception. Using a roster of men employed in three battery
22    plants in Denmark, Bonde and Kolstad (1997) matched all births to the 1,349 employees when
23    they were age 20-49 years. A control group of 9,656 men who were not employed in a lead
24    industry was chosen. No associations were found between employment or, among those
25    employed in the  lead industry, duration of employment in the lead industry and  birth rate.
26          A similar study in Finland (Sallmen, 2000) examined the association between conception
27    and blood lead among men monitored for occupational exposure at the Finnish Institute of
28    Occupational Health (n = 2,111). Men were categorized as probably exposed and possibly
29    exposed based on their measured blood lead in relation to the time of marriage.  A nonexposed
30    group of 681 men with blood lead < 10  |ig/dL was similarly evaluated. Among  men in the
31    probable exposure group, the risk of failing to achieve a pregnancy increased with increased

      December 2005                           6-215       DRAFT-DO NOT QUOTE OR CITE

-------
 1    blood lead in a monotonic concentration-response fashion. Compared to the nonexposed, the
 2    risk ranged from 1.3 to 1.9 for blood leads 10-20 |ig/dL and >50 |ig/dL, respectively.
 3          Lin et al. (1996) linked records from the Heavy Metal Registry in New York State to birth
 4    certificates from the New York State Office of Vital Statistics for the period 1981 to 1992.
 5    Exposure was defined as having at least one blood lead measurement above 25 |ig/dL and
 6    identified 4,256 men.  A reference group of 5,148 men was frequency matched for age and
 7    residence. The exposed group had fewer births than expected, and was especially pronounced
 8    among men employed in the lead industry for over 5 years.
 9          Among 365 men occupationally exposed to metals, Gennart et al. (1992) identified
10    74 exposed continuously for more than 1 year  and with at least one blood lead measurement
11    >20 |ig/dL. Compared to a reference group with no occupational exposure, the probability of at
12    least one live birth was significantly reduced.  Fertility decreased with increasing duration of
13    exposure but no concentration-response relationship with blood lead was found (possibly due to
14    the small  sample size of exposed men).
15          A study of men exposed to lead in a French battery plant (Coste et al., 1991) reported no
16    effect on fertility. However, this study did not adequately control for potentially confounding
17    variables, particularly those related to the women. Further, nonexposed workers were defined as
18    those with no blood leads recorded which likely resulted in exposure misclassification.
19          One potential mechanism to explain the associations between lead exposure and male
20    reproductive outcomes may be through an effect of lead on circulating pituitary and testicular
21    hormones. Several studies have evaluated this hypothesis in groups of workers (Braunstein
22    et al., 1978; Cullen et al., 1984; Erfurth et al., 2001; Ng et al., 1991; Rodamilans et al., 1988).
23    In general these studies find perturbations in concentrations of follicle stimulating hormone,
24    luteinizing hormone, and testosterone.  Although many of these studies were limited by small
25    sample sizes, lack of control groups, and admixtures of exposure, taken together, they provide
26    evidence for this possible mechanism.
27
28    6.6.3.2   Genotoxicity and Chromosomal Aberrations
29          The potential genotoxicity and ability to induce chromosomal aberrations speak to the
30    mechanisms by which lead is a potential reproductive toxin. Two possible mechanisms by
      December 2005                           6-216       DRAFT-DO NOT QUOTE OR CITE

-------
 1    which lead may affect reproduction are through affinity with proteins and ability to mimic the
 2    actions of calcium (Silbergeld et al., 2000).
 3          Data from occupational studies regarding the effects of lead on chromosomes are
 4    contradictory; however, the bulk of evidence suggests that there may indeed be a genotoxic
 5    effect.  Early studies in occupational groups find associations between lead exposure and
 6    increased frequency of sister chromatid exchanges (Grandjean et al., 1983; Huang et al., 1988;
 7    Leal-Garza et al., 1986; Maki-Paakkanen et al., 1981).  Similar results were found in a group of
 8    environmentally-exposed children with blood leads ranging from 30 to 63  |ig/dL (Dalpra et al.,
 9    1983).  Increased frequencies of chromosomal aberrations, particularly chromatid aberrations,
10    were found in battery plant workers and were correlated with increased blood lead (Huang et al.,
11    1988).  A more marked increase was found when blood leads were above 50 |ig/dL. Other
12    occupational studies find similar associations (Al-Hakkak et  al., 1986; Forni et al., 1976, 1980;
13    Nordenson et al., 1978;  Schwanitz et al., 1970). Other studies find no evidence of chromosomal
14    aberrations when blood leads ranged from 38 to 120 |ig/dL (Bauchinger et al., 1977; Maki-
15    Paakkanen et al., 1981; O'Riordan and Evans, 1974; Schmid et al., 1972; Schwanitz et al., 1975).
16    More recently, two studies in battery plant workers (mean blood lead 40.1  |ig/dL) and controls
17    (mean blood lead 9.8 |ig/dL) found an increase in high-frequency cells and sister chromatid
18    exchanges among the workers, indicating the cytogenetic toxicity of lead (Duydu et al., 2001,
19    2005).  An increase in sister chromatid exchanges, although not statistically significant, was also
20    found in individuals exposed to lead and/or alcohol and tobacco (Rajah and Ahuja, 1995, 1996).
21    In the Lithuanian populations exposed to either environmental or occupational lead, a higher
22    incidence of sister chromotid exchanges and chromosomal aberrations was found (Lazutka et al.,
23    1999), although these populations were also exposed to other potentially genotoxic substances.
24    Recent data also indicates that lead may inhibit DNA repair responses among lead-exposed
25    workers (Karakaya et al., 2005).
26          Occupational  exposure to lead, particularly when blood leads were high (i.e., over
27    40 |ig/dL) was associated with increased mitotic activity in peripheral lymphocytes and with an
28    increased rate of abnormal mitosis (Forni et al., 1976; Minozzo et al., 2004; Sarto et al., 1978;
29    Schwanitz etal., 1970).
30
      December 2005                           6-217       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.6.3.2.1   Issues Concerning Studies of Male Fecundity Related to Lead Exposure
 2          In examining studies of fecundity and fertility, several issues relating to interpretation and
 3    bias must be addressed.  Infertility usually is defined as 12 months of continuous unprotected
 4    intercourse without pregnancy. Fecundity represents both a characteristic of the individuals and
 5    a characteristic of a couple, meaning that both partners must be biologically able to procreate.
 6    Thus, one possible explanation for observations of reduced fecundity related to occupational lead
 7    exposure in the male partner is the exposure he "takes home" via transport of dust on clothing
 8    and shoes, ultimately resulting in an effect related to the female partner.  Other possible
 9    interpretations need to account for measurement error, especially related to the outcomes of
10    reproductive history and time  to pregnancy, bias in the selection of subjects for study, and the
11    control for potentially confounding variables.
12          Both reproductive history and time to pregnancy are subject to errors of recall and rely on
13    the veracity of the subject.  Several studies have evaluated recall  and veracity of the male partner
14    using the female partner as the "gold standard." In general, these find good reliability between
15    the male and female (Weinberg et al., 1993, 1994).  Nevertheless, it is possible, at least for
16    studies using men as the sole informant, that the number of pregnancies a man has fathered is
17    underreported. If reporting is  nondifferential with regard to lead  exposure, then associations will
18    generally be biased towards the null value; however, since characteristics such as social
19    circumstances, ethnicity, and age may affect both exposure and reporting, it is difficult to
20    evaluate the role of bias.
21          It was not clear from many of the  studies that men with  medical conditions which affect
22    fecundity/fertility  were excluded. Further, several prescription and over-the-counter medications
23    also affect fecundity as does a history of surgery in the genital  area (e.g., varicocele). To the
24    extent that these conditions are related to the absence of employment in lead-industries, then the
25    results may be subject to a type of "healthy worker" effect. Because it is unclear whether many
26    of these  studies asked about these conditions, this cannot be ruled out as a possible source
27    of bias.
28          In retrospective studies it is often useful to use the outcome of the most recent pregnancy
29    in the primary analysis. The reason for this is to reduce any possible recall bias. This type of
30    bias may also be an issue in studies which use occupational registry data, i.e., men may have
31    fathered an additional pregnancy after employment in the industry ceased.

      December 2005                            6-218       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Variables considered potential confounders in studies of fertility and fecundity include
 2    sociodemographic characteristics (e.g., age, ethnicity, education, occupation); prenatal and recent
 3    lifestyle variables such as cigarette smoking, alcohol use, and medication use; exposures through
 4    occupation and hobby, and recent medication use.  Also important in these studies is control for
 5    factors which may affect the partner's fertility, e.g., cigarette smoking. Many of the studies
 6    reviewed did not carefully measure or adjust for confounding variables.
 7          The issues presented above potentially limit the interpretation of results from studies
 8    examining the association of lead exposure with male fecundity and fertility. Nevertheless, most
 9    studies find small associations between lead exposure at high levels (i.e., >45 |ig/dL) and slightly
10    reduced male fecundity or fertility.
11
12    6.6.3.3   Effects on Female Reproductive Function
13          Few data directly address the effects of lead exposure on fecundity in the female.
14    A recent retrospective study of time to pregnancy among wives of lead workers provides limited
15    support that lead exposure is associated with increased time to pregnancy.  Fecundity density
16    ratios were 0.92 (95% CI: 0.72, 0.16), 0.89 (95% CI: 0.66, 1.20), 0.58 (95% CI:  0.33, 0.96),
17    and 0.83 (95% CI: 0.50, 1.32) for blood leads in the male partners of 10-20, 21-30, 31-38 and
18    >39 |ig/dL compared to <10 |ig/dL, respectively. Note however, that exposure here is measured
19    in the male partners and not the females.
20          Time to pregnancy was evaluated in 121 women biologically monitored for lead exposure
21    at the Finnish Institute of Occupational Health between 1973 and 1983 (Sallmen et al.,  1995).
22    Fecundity did not differ with level of exposure (defined as <10 |ig/dL, 10-19 |ig/dL and
23    >20 jig/dL), but among women with blood leads between 29 and 50 |ig/dL, there was a
24    suggestion of reduced fecundity (longer time to pregnancy).  However, only a small number of
25    subjects (n = 8) were exposed in this range.
26          In the limited number of studies, there is little evidence regarding the associations
27    between lead exposure and fertility in the female to draw any conclusions at this time.
28
      December 2005                           6-219       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.6.4   Spontaneous Abortion
 2    6.6.4.1    Spontaneous Abortion and Maternal Exposure to Lead
 3         Historical observations suggest increased rates of spontaneous abortion among lead-
 4    exposed women, particularly those employed in cottage industries (Rom, 1976). Two early
 5    studies in  a smelter town in Sweden (Nordstrom et al., 1978a,  1979) suggest elevated rates of
 6    spontaneous abortion among female employees at the smelter and among female residents living
 7    in close proximity to the smelter.  Neither of these studies used biological markers of lead
 8    exposure.  Moreover, the Swedish smelter study included other exposures such as arsenic, zinc,
 9    and cadmium; thus the conclusions for these analyses should be tempered.
10         In contrast, a prospective study in and around a smelter town in Port Pirie, Australia
11    (McMichael et al., 1986) did not find an association between blood lead concentration and
12    spontaneous abortion. However, it was likely that complete ascertainment of spontaneous
13    abortions was not obtained (Rowland and Wilcox, 1987) since most women were recruited for
14    this study  after the first trimester of pregnancy. A retrospective cohort study in two towns in the
15    former Yugoslavia (Murphy et al., 1990) showed no associations between lead exposure and
16    spontaneous abortion in the first reported pregnancy.  One of these towns was a smelter town
17    with relatively high lead exposure (at recruitment during mid-pregnancy, the mean blood lead
18    concentration was 17.1  |ig/dL, while in the control town the mean blood lead was 5.1 jig/dL).
19    A similar  study in Poland (Laudanski et al., 1991) evaluated the association between lead-
20    exposed and nonexposed areas for their reproductive histories.  Among women in the exposed
21    areas, 11% reported having at least one prior spontaneous abortion, compared to 19.5% of
22    women in the unexposed areas.
23         Two studies in Finland (Lindbohm et al., 1991; Taskinen, 1988) used hospital registry
24    data to ascertain women with either spontaneous abortions or livebirths. Either maternal job
25    histories (Taskinen, 1988) or both maternal and paternal job histories were obtained from a
26    registry of occupational blood lead measurements. Neither study found evidence of an
27    association between maternal exposure and spontaneous abortion. In the Lindbohm et al. (1991)
28    study, maternal exposure was extrapolated from the occupation of the father.
29         In Bulgaria,  pregnant women residing in or near lead smelting areas or petrochemical
30    plants were prospectively followed for pregnancy outcomes (Tabacova and Balabaeva, 1993).
31    The investigators compared blood leads in those women with spontaneous abortions and those

      December 2005                           6-220       DRAFT-DO NOT QUOTE OR CITE

-------
 1    without. Blood lead concentrations in cases were significantly higher than in controls (mean
 2    blood lead 7.1 |ig/dL versus 5.2 |ig/dL, respectively).  However, this study did not fully describe
 3    the selection of women nor the definition for cases.
 4          Women employed by the U.S. Forest Service and exposed to lead-based paint (to mark
 5    trees for clearing) were studied using self-reported questionnaires (Driscoll,  1998). Adjustment
 6    was made for potential confounders and generalized estimating equations were used to adjust for
 7    multiple pregnancies per woman.  Significant associations were found for three types of paint
 8    containing lead pigment (odds ratios of 4.3 [95% CI: 2.0, 9.3], 2.0  [95% CI: 1.2, 3.3] and
 9    1.8 [95% CI:  1.2, 2.6]). While these findings are intriguing, the response rate was only 59%
10    (with no evaluation of selection bias) and the paint also contained solvents thought to be
11    associated with spontaneous abortions.
12          Borja-Aburto et al.  (1999) examined the association between blood lead concentrations
13    and spontaneous abortions in  a nested case-control study using incidence density methods and
14    matching for age, calendar time of study entry,  public versus private clinic, and gestational age at
15    study entry.  They ascertained 668 women  during the first trimester of pregnancy in Mexico
16    City. After contacting women biweekly to update pregnancy status, they found 35 cases (6.4%)
17    of spontaneous abortion among women not lost to follow up. An odds ratio of 1.8 (95% CI:  1.1,
18    3.1) per 5 |ig/dL increase in blood lead was observed after adjustment for spermicide use, active
19    and passive smoking, use of alcohol and coffee, maternal age, education, income,  physical
20    activity, hair dye use, use of video display terminals, and medical conditions. Mean blood lead
21    in cases (12.0 |ig/dL, range 3.1-29 |ig/dL) was slightly higher than in controls (10.1 |ig/dL, range
22    1.3-26 |ig/dL). Further, after  categorizing blood lead into 5 |ig/dL intervals, a concentration-
23    response relationship was  evident.
24          More recently, a small  study of 57 female workers in a battery plant in China and
25    62 controls found that 6 spontaneous abortions  occurred in the  exposed group, compared to none
26    in the controls (Tang and Zu,  2003).  A long-term follow-up of survivors of acute plumbism
27    (Hu, 1991) found increased risk of spontaneous abortions or stillbirths (odds ratio of 1.6
28    [95% CI: 0.6,4.0]). Although the study was based on small numbers, the data suggest a
29    persistent association between childhood exposure and outcomes later in life.
30          A review of eight studies (Borja-Aburto  et al., 1999; Driscoll, 1998; Laudanski et al.,
31    1991; Lindbohm et  al., 1991;  McMichael et al., 1986; Murphy  et al., 1990; Tabacova and

      December 2005                           6-221       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Balabaeva, 1993; Taskinen, 1988) evaluating maternal exposure to lead (blood lead >30 |ig/dL)
 2    and spontaneous abortion concluded that there was little evidence that lead exposure at these
 3    relatively high levels was associated with an increased risk in spontaneous abortions (Hertz-
 4    Picciotto, 2000).  However, Hertz-Picciotto also concluded that methodological difficulties in
 5    most of these studies (i.e., small sample sizes, inadequate ascertainment of outcome, and possible
 6    residual confounding) limited the confidence in these data. Further, she noted that exposure in
 7    many of these studies was either measured in an ecologic fashion or biological measures were
 8    available, but they were not ascertained during a biologically meaningful period.
 9          Collectively, there is little evidence to support an association between lead exposure in the
10    female and spontaneous abortion. The only well-designed study which finds an association is
11    that of Borja-Aburto et al. (1999); however, these results need to be confirmed in other
12    populations.  Studies of spontaneous abortion need be done carefully to avoid possible bias due
13    to recall, use of pregnancies other than the first, and confounding. Retrospective studies, for
14    example,  should take full pregnancy histories, including probing for spontaneous abortions
15    versus induced abortions versus stillbirths. In some cultures, for example, induced abortions are
16    frowned upon and women may report spontaneous abortions instead.  Additionally, some women
17    may confuse a stillbirth with spontaneous abortion, especially if she is unable to adequately date
18    her pregnancy using date of last menstrual period.  Although the use of the most recent
19    pregnancy may curtail problems of recall, other concerns dictate that the first pregnancy be used
20    in studies of spontaneous abortion because the risk of subsequent spontaneous abortion depends
21    on the history of spontaneous abortion.  Finally, while few variables are known confounders of
22    this relationship, the following should be controlled:  maternal age, education and other
23    socioeconomic indicators, cigarette smoking, and alcohol use.  Several studies of spontaneous
24    abortion did not properly adjust for these potentially  confounding variables.
25          One final concern regards the type of spontaneous abortion. Very early spontaneous
26    abortions, i.e., before a clinical pregnancy is diagnosed, may be missed; assuming, however, that
27    both exposed and unexposed women have the same rates of early spontaneous abortions, this
28    would bias the association towards the null. Indeed,  this may be true, as many very early
29    spontaneous abortions may be chromosomally abnormal and probably not attributable to lead
30    exposure.
31

      December 2005                            6-222       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.6.4.2   Spontaneous Abortion and Paternal Exposure to Lead
 2          Three studies evaluated paternal exposure to lead and spontaneous abortion.  Lindbohm
 3    et al. (1991), using national databases to identify pregnancy outcomes among 99,186 births in
 4    Finland, found no association between paternal employment in jobs with lead exposure and
 5    spontaneous abortion (odds ratio of 0.9 [95% CI: 0.9,1.0]). In a follow up case-control study
 6    (Lindbohm et al., 1991b), they ascertained paternal exposure status during the period of
 7    spermatogenesis in 213 cases of spontaneous abortion and 500 controls.  Exposure was
 8    ascertained using blood lead concentrations measured during spermatogenesis for 6% of men;
 9    for the remaining 94%, exposure was estimated using a regression model where the independent
10    variables were blood leads measured either prior to or after the period of spermatogenesis.
11    Blood lead (either measured or estimated) was not associated with spontaneous abortion.
12    When analysis was restricted to men with measured blood lead, blood lead concentrations
13    >30 |ig/dL were associated with an increased odds of spontaneous abortion (odds ratio of
14    3.8 [95% CI:  1.2, 2.0]); however, this result was only based on 12 cases and 6 controls.
15          The third study (Alexander et al., 1996b) found no association between men employed in
16    a lead smelter and spontaneous abortion.  For men with "moderate" exposure jobs the estimated
17    odds ratio was 0.8 (95% CI:  0.5, 1.5) and for those with "high" exposure jobs, the estimated
18    odds ratio was 1.4 (95% CI:  0.7, 2.5). Further when  blood lead 1 year prior to the pregnancy
19    was used as the exposure measure, no increased odds of spontaneous abortion was found.  These
20    results, however, are based on a low participation rate in eligible workers (37%) and should be
21    interpreted with caution. Overall, the available studies provide little evidence for an association
22    between lead exposure in the male and spontaneous abortions.
23
24    6.6.5  Fetal Growth
25          The results of epidemiologic studies regarding the association between lead exposure and
26    birth weight are inconsistent. Cross-sectional studies (Clark, 1977; Gershanik et al., 1974;
27    Moore et al., 1982; Rajegowda et al.,  1972) did not find significant correlations between blood
28    lead and birth weight,  nor did a study using placental  lead as the exposure variable (Wibberley
29    et al., 1977). A case-control study (Bogden et al., 1978) comparing 25 low birth weight babies
30    (1,500-2,500 grams) to 25  controls (>2,500 grams) matched on maternal age, race and social
31    class found a small, nonsignificant difference in maternal and cord blood leads. Mean maternal

      December 2005                          6-223        DRAFT-DO NOT QUOTE OR CITE

-------
 1    blood lead concentrations were 16.2 + 4.5 |ig/dL and 15.3 + 5.2 |ig/dL and mean cord blood
 2    leads were 13.8 + 4.4 |ig/dL and 13.1 + 4.3  |ig/dL in cases and controls, respectively. A further
 3    study (Huel et al., 1981) found no differences in maternal and fetal hair lead concentrations
 4    between infants born small-for-gestational-age compared to those of normal birth weight.
 5          In 1984, Needleman et al. (1984) reported on a cross-sectional study of 5,183 births of at
 6    least 20 weeks gestation in Boston, MA. No associations were found between the proportion of
 7    births under 2,500 grams and cord blood lead. Exposure levels in this study were relatively low
 8    for the time; cord blood leads ranged from <1 to 35 |ig/dL. A reanalysis of these data found no
 9    relationship between cord blood lead and birth weight when birth weight was considered as a
10    continuous variable (Bellinger, et al., 1991). However, when birth weight was categorized as
11    low birth weight (<2,500 grams), small for gestational age (<10th percentile for gestational age),
12    or intrauterine growth retarded (>2 standard deviations below the mean for gestational age),
13    relative risks of 1.6 (95% CI: 1.0, 2.6), 1.2  (95% CI:  0.8, 1.6) and 1.9 (95% CI:  1.0, 3.4),
14    respectively, were found for each 10 |ig/dL  increase in cord blood lead levels. Increased relative
15    risks also were found for cord blood lead levels > 15 |ig/dL, compared to cord blood lead
16    <15 |ig/dL; however, only 83 of the 5,183 women  had exposures in the high range, resulting in
17    imprecise estimates. These data suggest that lead-related modest reductions in birth weight are
18    perhaps plausible when birth weight is expressed as a function of gestational age.
19          The prospective study of lead exposure in and around Port Pirie, Australia (McMichael
20    et al., 1986) followed 749 pregnancies of at least 20 weeks duration. Mean maternal blood leads
21    at mid-pregnancy were 10.1 |ig/dL and 7.0 |ig/dL for women residing in Port Pirie and the
22    surrounding communities, respectively.  After excluding 9 sets of twins and 10 cases for which
23    the maternal last menstrual period could not be ascertained, no relationship was found between
24    either cord blood lead or maternal blood lead measured at mid-pregnancy or at delivery and birth
25    weight in a multivariate regression model controlling for known determinants of birth weight.
26          A prospective study in two towns in Kosovo, Yugoslavia evaluated relationships  between
27    birth weight (adjusted for gestational age using last menstrual period) and (a) maternal blood
28    lead at mid-pregnancy and delivery and (b)  cord blood lead (Factor-Litvak et al., 1991).  The
29    towns were vastly different in exposure patterns, as one was the site of a lead smelter, refinery
30    and battery plant (n = 401, mean mid-pregnancy blood lead 19.0 |ig/dL) and one was relatively
31    unexposed (n = 506, mean mid-pregnancy blood lead 5.6 |ig/dL). No associations were  found

      December 2005                           6-224       DRAFT-DO NOT QUOTE OR CITE

-------
 1    between any of the biomarkers of lead and birth weight in either crude analyses or analyses
 2    adjusted for potentially confounding variables.
 3          While the aforementioned studies generally found no association between environmental
 4    lead exposure and birth weight, three other studies have shown large reductions in birth weight
 5    related to lead exposure.  These studies, however, have questionable study designs. Nordstrom
 6    et al. (1978b, 1979) in a series of ecologic analyses known as the Swedish Smelter Study, found
 7    significant reductions in birth weight between the offspring of women either working at or living
 8    in close proximity to the smelter. The 125 gram deficit in birth weight among the offspring of
 9    women living closest to the smelter was confined to those with parity three or more, an
10    observation which does not appear to be biologically plausible. Moreover, the ecological nature
11    of the study did not allow for individual measurements of blood lead or for control of potentially
12    confounding variables. Hence, while suggestive, these data do not provide strong evidence for a
13    causal association between lead exposure and birth weight.
14          In a cross-sectional study of 100 "normal" singleton births, a negative correlation was
15    found between placental lead concentration and birth weight (Ward et al., 1987).  Mean placental
16    lead concentration in 21 infants weighing less than 3,000 grams was 2.35 + 0.9  |ig/g compared to
17    1.12 + 0.4 |ig/g in 10 infants weighting more than 4,000 grams.  This study has several
18    limitations. First, no statistical adjustment was made for multiple comparisons (many exposures
19    were studied).  Second, potentially confounding variables were not controlled. Third,  only 31 of
20    the 100 infants, representing the extremes of the birth weight distribution, were studied.  Hence,
21    this study also does not provide strong evidence for an association.
22          In Cincinnati, OH, the association between lead exposure and birth weight was examined
23    in offspring of a cohort of young (mean maternal age = 22.7 years), inner city women,
24    85% African American, 86% on public assistance, with a mean IQ of 75 (Dietrich et al.,  1987a).
25    The mean gestational period of the neonates, as determined by physical examination, was
26    39.5 weeks.  A decrement in birth weight of 172 grams was associated with an increase in blood
27    lead from 10 to 30 |ig/dL. Lead exposure in this group was relatively low with a mean blood
28    lead of 8.0 + 3.7 |ig/dL. In a sample of women from this cohort,  the interaction between blood
29    lead and maternal age was significantly associated with birth weight; the effect varied from a
30    decrease of 64 grams for 18 year old mothers to 660 grams for 30 year old mothers, as blood lead
31    rose from 10 to 30 |ig/dL (Bornschein et al., 1989). Although the Cincinnati study is highly

      December 2005                           6-225       DRAFT-DO NOT QUOTE OR CITE

-------
 1    suggestive of an effect (especially an effect which varies by maternal age) three factors should be
 2    considered in the interpretation of their findings.  First, length of gestation was estimated by
 3    examining the neurological and physical maturity of the neonate (Ballard et al., 1979); other
 4    investigators find assessment of gestational age using this scale overestimates gestational age in
 5    preterm infants (Constantine et al., 1987; Kramer et al., 1988; Shukla et al.,  1987; Spinnato et al.,
 6    1984).  Second, it is possible that the association between lead and birth weight differs by
 7    maternal characteristics such as race, ethnicity, and SES; however, no study has provided a
 8    population sufficiently heterogeneous to examine this possible source of difference. Finally, it is
 9    possible that confounding by unmeasured maternal lifestyle characteristics may account for the
10    reported association.
11          A hospital-based study of cord blood lead and pregnancy outcomes in Quebec, Canada,
12    between June 1993 and January 1995 found a slight increase in cord blood lead levels among
13    infants with birth weight <2,500 grams (Rhainds et al., 1999).  For those infants with birth
14    weight <2,500 grams, the geometric mean blood lead was 1.8 |ig/dL (95% CI:  1.6, 2.9)
15    compared to 1.6 |ig/dL (95% CI:  1.5, 1.7), 1.6 |ig/dL (95% CI: 1.5, 1.7), and 1.5 |ig/dL
16    (95% CI:  1.5, 1.6) among those with birth weights 2,500-2,990, 3,000-3,499, and >3,500 grams,
17    respectively. Although suggestive, the study did not control for potentially confounding
18    variables.  The investigators also measured cord blood  levels of mercury and organochlorine
19    compounds, and observed that mean levels of these toxicants were higher as well in infants who
20    weighed <2,500 g.
21          More recently Irgens et al. (1998) using data from the Norwegian birth registry found that
22    women occupationally exposed to lead (none/low compared to moderate/high) were more likely
23    to deliver a low birth weight infant (odds ratio of 1.3 [95%  CI:  1.1, 1.6]). No association was
24    found for paternal occupational lead exposure. Parental occupational exposure to lead was not
25    associated with low birth weight in the Baltimore-Washington Infant Study  database (Min et al.,
26    1996), although subgroup analysis suggested that high  paternal exposure may be associated with
27    small-for-gestational-age infants (odds ratio of 2.9 [95% CI: 0.9, 9.2]). Similar findings were
28    reported by Lin et al. (1998) who compared offspring of lead-exposed workers with those of bus
29    drivers. No associations were reported between lead exposure and low birth weight except
30    among the group of men with blood lead levels >25 |ig/dL for over 5 years (relative risk of 3.4
31    [95% CI:  1.4,8.4]).

      December 2005                            6-226        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Using bone lead as the metric of exposure, Gonzalez-Cossio et al. (1997) found
 2    associations with tibia bone lead (but not with patella bone lead or umbilical cord blood lead)
 3    and reduced birth weight. Bone lead was measured one month after delivery.  Infants with tibia
 4    bone lead in the highest quartile (>15.15 jig lead / g bone mineral) were, on average, 156 g
 5    lighter than those in the lowest quartile (<4.50 jig lead / g bone mineral).  Further analyses of
 6    these data (Hernandez-Avila et  al., 2002) found an association between infants in the highest
 7    quintile of tibia bone lead and shorter birth length (odds ratio of 1.8 [95% CI:  1.1, 3.2]).
 8          Two studies have considered the relationship between lead exposure and head
 9    circumference (Hernandez-Avila et al., 2002).  Among 233 women in Mexico City, high
10    maternal patella bone lead was associated with increased risk of a low head circumference score
11    at delivery (1.02 per jig lead / g bone mineral [95% CI:  1.01, 1.04]).  Similar findings were
12    reported by Rothenberg et al. (1999) who found a reduction in six-month head circumference of
13    1.9cm(95%CI: 0.9, 3.0) as maternal blood lead rose from 1 to 35 |ig/dL.  This study, however
14    was plagued by multiple comparisons as head circumference was measured nine times and
15    prenatal blood lead six times - only one  statistically significant result was found.
16          Potential confounders need to be adjusted for to properly assess the relationship between
17    lead exposure and fetal growth. Factors consistently associated with fetal growth include gender,
18    ethnic origin, maternal body build (i.e., pre-pregnancy weight, height), parity, SES, gestational
19    weight gain and nutritional intake during pregnancy, maternal illness, and cigarette smoking
20    (Kramer, 1987). Factors with less established associations include alcohol consumption (Kline
21    et al.,  1987; Kramer, 1987) and street drug use (Kline et al., 1987; Kramer, 1987; Zuckerman
22    et al.,  1989).  To the extent that these factors are associated with blood lead as well as with fetal
23    growth, they must be accounted for in the analysis.
24          Studies to date are inconsistent regarding the association between lead exposure and birth
25    weight.  Several large prospective studies find no association (Factor-Litvak et al., 1991;
26    McMichael et al.,  1990), while at least one (Bornschein et al., 1989) did find an association in
27    specific subgroups of women.  However, there is limited evidence (Bellinger et al., 1991) for an
28    association between lead exposure and low birth weight (i.e., <2,500 g), small for gestational age
29    (i.e., <10th percentile for gestational age), and intrauterine growth retardation (i.e., >2 standard
30    deviations below the mean for gestational age).  These prospective studies were all well-
31    conducted, adequately measured exposure and outcome, and controlled for potential confounding

      December 2005                           6-227       DRAFT-DO NOT QUOTE  OR CITE

-------
 1    variables. They did, however, take place in very different populations, suggesting that the
 2    association between lead and fetal growth may depend on the population being studied.  The
 3    Yugoslavia study (Factor-Litvak et al., 1991) took place in two towns in Kosovo, Yugoslavia,
 4    which were divergent on exposure and somewhat comparable on other variables. The Port Pirie
 5    study took place in a middle class area of Australia (McMichael et al., 1986).  The Boston study
 6    (Bellinger et al., 1991) took place in a range of social strata in Boston; the exposure in the
 7    highest social group was attributable to renovation of older housing stock. Finally, in the
 8    Cincinnati study (Bornschein et al., 1989), the study sample was comprised of lower social class
 9    African Americans; the mean IQ of the mothers was 75. It is possible that in this latter study,
10    there was some unmeasured variable which accounts for the observed interaction.  Thus, the
11    evidence suggests at most a small effect of lead exposure on birth weight and possibly a small
12    association between lead exposure and several dichotomized measures of fetal growth.
13
14    6.6.6   Preterm  Delivery
15          Early evidence regarding an association between environmental lead exposure and
16    preterm delivery was inconsistent.  In 1976, Fahim et al. found  a preterm delivery rate of 13% in
17    254 pregnant women living near a lead mining community in Missouri, compared to 3% in
18    249 women living in a control location. These investigators also found higher concentrations of
19    lead in amniotic membrane, but not higher placental or cord lead in preterm compared to term
20    deliveries, regardless of the women's residential locale. This observation prompted other studies
21    of lead and preterm delivery.
22          Of the cross-sectional studies, the three which show no association employed cord blood
23    lead as the exposure measure and restricted gestational age (Angell and Lavery, 1982; Bellinger
24    et al., 1991; Needleman et al., 1984; Rajegowda et al., 1972). In contrast, three other studies
25    used different exposure markers (placental lead,  maternal and cord blood lead, and maternal and
26    fetal hair lead) and found statistically significant associations (Huel et al., 1981; Moore et al.,
27    1982; Ward et al., 1987). Other studies evaluated pregnancy outcomes in relation to maternal
28    delivery blood lead  (McMichael et al.,  1986; Rahman and Hakeem, 2003).
29          Of the prospective studies, the Cincinnati study (Bornschein et al., 1989) found no
30    association between both maternal blood lead at mid-pregnancy or maternal blood lead during
31    the neonatal period  (10 days post delivery) and preterm delivery. However, gestational age was

      December 2005                           6-228       DRAFT-DO NOT QUOTE OR CITE

-------
 1    estimated by examining the neurological and physical maturity of the neonates (which tends to
 2    overestimate gestational age) and not actual dates. In Port Pirie, Australia (McMichael et al.,
 3    1986), a concentration-response relationship between maternal delivery blood lead and preterm
 4    delivery was reported. Odds ratios ranged from 2.1 to 4.4 in women with blood leads of
 5    7.7-10.6 |ig/dL and >13.5 |ig/dL, respectively, compared to those with blood lead <7.7 |ig/dL.
 6    Savitz et al. (1990) used data from the National Natality Survey and found an odds ratio of
 7    2.3 (95% CI:  0.7, 7.0) between maternal occupational exposure to lead and preterm delivery;
 8    however, the estimated odds ratio was based on only 7 cases. In the Yugoslavia study (Factor-
 9    Litvak et al., 1991) no associations were found between cord blood lead or blood lead measured
10    at mid pregnancy or delivery and either preterm delivery (defined as delivery <37 completed
11    weeks) or gestational  age. A registry study in Norway (Irgens et al., 1998) which linked births
12    between 1970 and 1993 to census-based occupation records found a slightly increased odds of
13    preterm delivery among moderate/high lead-exposed women, compared to those with no or low
14    exposure (odds ratio of 1.13 [95% CI: 0.98, 1.29]). Paternal exposure was not found to increase
15    the risk of preterm birth.
16          An ecologic study in Canada (Philion et al., 1997) examined 30 years of birth records,
17    corresponding to 9,329 births in a smelter city and a control city. Outcome variables were
18    intrauterine growth retardation defined as small for gestational age. The odds ratio for
19    intrauterine growth retardation in the smelter city compared to the control city was 0.83.
20    Further analysis, stratifying time into 5-year intervals also revealed no associations.
21          A case control  study in Mexico City (Torres-Sanchez et al., 1999) evaluated 161 preterm
22    births and 459 full term births. Cord blood lead was significantly higher in the preterm group
23    (9.8 + 2.0 |ig/dL) compared to the full term group (8.4 + 2.2 |ig/dL) only among primiparous
24    women.
25          Using data from the Baltimore-Washington Infant Study database, Min et al. (1996)
26    found a small association between paternal occupational exposure in the high range and preterm
27    delivery with appropriate weight for gestational age (odds ratio of 2.1 [95% CI: 0.7, 6.5]) and
28    preterm delivery with small for gestational age (odds ratio of 2.4 [95% CI: 1.9, 3.1]).  Similar
29    findings were reported by Lin et al. (1998). Comparing the offspring of lead exposed workers
30    with those of bus drivers, they found an elevated relative risk for preterm delivery (3.0 [95% CI:
31    1.6, 6.8]) only among men with blood leads >25 |ig/dL for over 5 years.

      December 2005                           6-229       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In contrast to fetal growth, few factors are consistently related to preterm delivery; thus in
 2    both developed and developing countries the majority of preterm deliveries remain unexplained
 3    (Kramer 1987; van den Berg and Oechsli, 1984).  Factors which are inconsistently associated
 4    with preterm delivery include maternal age, SES, pre-pregnant weight, prior history of preterm
 5    delivery or spontaneous abortion, and cigarette smoking (Kline et al.,  1987; Kramer, 1987).
 6    Thus, these factors must be evaluated as potentially confounding factors in studies of lead
 7    exposure and preterm delivery.
 8          For preterm delivery, or reduced length of gestation, the evidence for an association with
 9    lead exposure is contradictory. Several of the prospective studies find no evidence of an
10    association (Bornschein et al., 1989; Factor-Litvak et al.,  1991) while one finds a concentration-
11    response relationship (McMichael et al., 1986). Further, two well-done registry studies (Irgens
12    et al., 1998; Savitz et al., 1990) find some evidence of an  association,  albeit the number of
13    exposed cases was small. It seems unlikely that the association between lead exposure and
14    preterm delivery is large, but, more research is clearly necessary.
15
16    6.6.7   Congenital Abnormalities
17          Needleman et al.  (1984) found an association between cord blood lead and minor
18    congenital anomalies among 4,354 infants born in a single hospital in Boston, MA. All data
19    were obtained from hospital records, not from direct examination of the infants. The most
20    common anomalies were hemangiomas, lymphangiomas, minor skin problems (tags and
21    papillae), and undescended testicles. Blood lead levels were not found to be associated with
22    individual anomalies.
23          More  recently, a number of studies have considered parental lead related to occupational
24    exposure and risk of congenital anomalies in the offspring. In Finland, Sallmen et al. (1992)
25    evaluated the associations between congenital  malformations and paternal exposure during the
26    time of spermatogenesis. The overall estimated unadjusted odds ratio for men with blood lead
27    levels >20 |ig/dL was 2.4 (95% CI: 0.9, 6.5).  Due to small sample sizes, the investigators could
28    only adjust for one potentially confounding factor at a time; this resulted in odds ratios ranging
29    from 1.9 to 3.2.  Of note is the lack of consistency of malformations among the five men with the
30    highest blood lead. The malformation observed included congenital heart disease, oral cleft, club
31    foot, polydactyly, and anomalies of the adrenal gland.  The breadth of these anomalies suggests


      December 2005                            6-230       DRAFT-DO NOT QUOTE OR CITE

-------
 1    either that lead affects physical development throughout gestation or that this association
 2    represents a chance finding. Among 2,021 pregnancies, Alexander et al. (1996b) found slightly
 3    elevated odds ratios for congenital defects among men in the lead smelting industry with
 4    moderate exposure (odds ratio of 1.9 [95% CI: 0.6, 6.3]) and high exposure (odds ratio of
 5    2.7 [95% CI:  0.7, 9.6]).  These estimates are based on 30 birth defects and 12 stillbirths.
 6    No analyses were presented which considered individual birth defects. In Norway, neither
 7    maternal (odds ratio of 1.25 [95% CI: 0.8, 1.9]) nor paternal (odds ratio of 0.94 [95% CI:
 8    0.8, 1.1]) occupational lead exposure was associated with serious birth defects (Irgens et al.,
 9    1998).  Similar results were reported by Kristensen et al. (1993) between paternal lead exposure
10    and birth defects, with the exception of a fourfold increase in the risk of cleft lip among male
11    offspring.
12          The risk of parental lead exposure and neural tube defects was evaluated in a case-control
13    study of 88,449 births (363 neural tube defects) over a 25-year period in Fylde, England (Bound
14    et al., 1997). Women living in areas in which the water lead concentration was >10 |ig/L were
15    more likely to deliver a child with a neural tube defect. The association was consistent for
16    anencephaly (n = 169) and spina bifida/cranium bifidum (n = 195), even after adjusting for social
17    class. These authors posit that the association could be a direct effect of lead on neural tube
18    closure or an indirect effect, the latter meaning a reduction in uptake of zinc (due to lead
19    exposure) leading to a reduction in folate uptake. Irgens et al.  (1998) partially confirmed these
20    effects on neural tube defects in mothers occupationally-exposed to lead (relative risk of 2.87
21    [95% CI: 1.05, 6.38]), but not for paternal lead exposure.
22          The association between total anomalous pulmonary venous return and parental lead
23    exposure during pregnancy (self reported, obtained from industrial hygiene measures, or from a
24    job exposure matrix) was examined in the Baltimore-Washington Infant Study (Jackson et al.,
25    2004).  In this case-control study, maternal periconceptional (i.e., 3 months prior to conception
26    through the first trimester) exposure to lead resulted in an estimated odds ratio of 1.57 (95% CI:
27    0.64, 3.47). For lead-exposed men, the estimated odds ratio was 1.83 (95% CI:  1.00, 3.42).
28    Findings from this study support a possible association between paternal lead exposure and total
29    anomalous pulmonary venous return.
30          Taken together, the evidence suggests few associations between periconceptional or
31    prenatal exposure to lead and congenital anomalies.  There is a suggestion of small associations

      December 2005                            6-231        DRAFT-DO NOT QUOTE OR CITE

-------
 1    with high levels of exposure, but many of those studies relied on occupational histories rather
 2    than on actual measures of blood lead levels.
 3
 4    6.6.8   Summary of the Epidemiologic Evidence for the Rep reductive and
 5            Developmental Effects of Lead
 6          There is little doubt that maternal to fetal transmission of lead results from placental
 7    transport. This transport occurs throughout pregnancy and may increase in the later stages.
 8    Further, there may be populations with increased fetal susceptibility, including populations with
 9    high rates of smoking and alcohol use, those using ethnic remedies and cosmetics, and those who
10    use lead glazed pottery. Low levels of calcium intake may also increase fetal exposure.
11          The available evidence suggests small associations between exposure to lead and male
12    reproductive outcomes. These include perturbed semen quality and increased time to pregnancy.
13    These associations appear at blood lead levels greater the 45 |ig/dL, as most studies only
14    considered exposure in the occupational setting.  More research is needed regarding possible
15    male reproductive effects at exposure levels in the lower (and currently more relevant) range.
16    There are no adequate data to evaluate associations between lead exposure and female fertility.
17          With one exception, there is no evidence to suggest an association between either
18    maternal or paternal lead exposure and increased risk of spontaneous abortions.  One study in
19    Mexico where the mean maternal blood leads were in the moderate range (i.e. 10-12 |ig/dL)
20    suggests an association.
21          To date, the evidence suggests at most a small association between lead exposure and
22    birth weight and possibly small associations between lead exposure and several dichotomized
23    measures of fetal  growth. The reviewed studies occurred in very different populations, and the
24    small associations may reflect some unmeasured or unknown confounding variable.  It is
25    unlikely that further epidemiologic research will fully resolve this question. However, several
26    factors, such as maternal SES, maternal education, smoking prevention and reduced use of
27    alcohol, related to lead exposure are associated with increases in birth weight (and decreases in
28    blood lead) and are  candidates for intervention.
29          Similarly, the evidence suggests at most a small association between lead exposure and
30    preterm delivery or reduced length of gestation. The available data also suggest limited
31    associations between either periconceptional or prenatal lead exposure and congenital anomalies.

      December 2005                          6-232        DRAFT-DO NOT QUOTE OR CITE

-------
 1    There is a suggestion of small associations with high levels of exposure, but many of those
 2    studies relied on occupational histories rather than on actual measures of blood lead.
 3         Overall, since the 1986 Lead AQCD, a substantial body of work has evaluated the
 4    associations between lead exposure and reproductive outcomes. It is now clear that lead clearly
 5    crosses the placenta during all trimesters and maternal exposure results in fetal exposure.
 6    For many other outcomes, the observed associations are relatively small, especially at the levels
 7    of exposure that are currently of interest. Nevertheless, there may be populations that are highly
 8    susceptible to lead-related reproductive effects, especially if they have additional risk factors for
 9    these outcomes.
10
11
12    6.7    GENOTOXIC AND CARCINOGENIC EFFECTS OF LEAD
13    6.7.1   Summary of Key Findings from the 1986 Lead AQCD
14         The 1986 EPA Lead AQCD reviewed five epidemiologic studies of occupationally
15    exposed workers (Cooper and Gaffey, 1975; Davies, 1984; Selevan et al.,  1985; Sheffet et al.,
16    1982; McMichael and Johnson, 1982).  These workers were exposed to inorganic lead
17    compounds such as lead oxides and lead sulfides. The EPA noted that Cooper and Gaffey
18    reported a significant increase in lung and gastrointestinal cancer among battery and smelter
19    workers in the U.S. (standardized mortality ratios of 1.50 and 1.48 respectively among smelter
20    workers, and 1.32 and 1.23 among battery workers). The EPA further noted that much of this
21    exposure was by inhalation and ingestion of lead oxides, which are relatively insoluble, adding
22    some plausibility to the occurrence of cancer at these two sites.  Sheffet et al. (1982) provided
23    some corroborating  evidence for gastrointestinal cancer by finding a nonsignificant excess of
24    stomach cancer among U.S. lead chromate pigment workers.  However, Davies (1984) did not
25    find any cancer excess among lead chromate pigment workers in the U.K.  The EPA noted that
26    Selevan (1984) found a significant excess of kidney cancer among U.S. lead smelter workers
27    based on 6 cases.  This finding was judged striking because it mimicked the findings of kidney
28    cancer in animals. The EPA judged that the McMichael and Johnson (1982) study of lead
29    poisoned workers was not particularly informative because the non-poisoned workers may have
30    had substantial lead exposure and no details were given on how lead poisoning was determined.
     December 2005                           6-233        DRAFT-DO NOT QUOTE OR CITE

-------
 1   In summary the EPA felt the evidence was insufficient, stating that "little can now be reliably
 2   concluded from available epidemiologic studies."
 3         The studies by Cooper and Gaffey (1975) and Selevan (1984), which are both important
 4   because they are large occupational cohorts with documented high exposure, have been updated
 5   and are further reviewed below. A cohort study of U.K. battery workers (Malcolm and Barnett,
 6   1982) is also reviewed below.
 7         EPA in 1986 also presented data on human cytogenetic studies, reproducing data from
 8   an earlier 1980 International Agency for Research on Cancer (IARC) monograph for metals and
 9   metallic compounds (IARC, 1980).  For lead, 10 chromosomal aberration studies were judged to
10   be "positive" and 6 such studies were judged to be "negative." On the whole the EPA
11   considered that "under certain conditions lead compounds are capable of inducing chromosomal
12   aberrations in vivo and in tissue cultures." The EPA also reviewed more limited data from two
13   human studies of sister chromatid exchange (Dalpra et al., 1983; Grandjean et al., 1983), one of
14   which was positive and  one negative.
15
16   6.7.2   Summary of Key Findings by the  International Agency for Research
17           on Cancer and the National Toxicology Program
18         IARC reviewed inorganic and organic lead compounds in its monograph number 87 in
19   February of 2004 (IARC, 2005), and IARC concluded that inorganic lead compounds were
20   probable human carcinogens (Group IIA). This classification is one step down from a
21   classification as a "definite" human carcinogen (Group I). The IARC classification of inorganic
22   lead compounds as probable human carcinogens was based on limited evidence in humans and
23   sufficient evidence in animals.  Also, IARC noted that there was insufficient information
24   regarding organic lead compounds (e.g., tetraethyl  lead) to make any judgment.
25         Regarding the human studies, IARC based its evaluation largely on six occupational
26   cohort studies of highly-exposed workers, which were felt to be particularly informative (battery
27   workers in U.S. and U.K.,  smelter workers in Italy, Sweden, and the U.S.).  The IARC
28   assessment focused on four cancer sites, lung, stomach, kidney, and brain.  IARC noted that lung
29   showed a significant elevation in one study (Lundstrom et al., 1997) and nonsignificant
30   elevations in a number of others. However, the significant elevation of lung cancer in
31   Lundstrom et al. appeared to be inextricably associated with arsenic in addition to lead exposure

     December 2005                          6-234       DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Englyst et al., 2001). IARC concluded that the strongest epidemiologic evidence for lead
 2    carcinogenicity was for stomach cancer, noting that four cohort studies showed a consistent
 3    30-50% excess of stomach cancer vs. external referent populations.  IARC noted that
 4    confounding by ethnicity, diet, Helicobacter pylori infections, or SES may have played a role in
 5    the stomach cancer excesses.  Finally, IARC noted that while one cohort study showed a 2-fold
 6    excess of renal  cancer (Steenland et al., 1992), the other studies showed no excess. Similarly,
 7    there were no consistent excesses of brain cancer, although one study did find a significant
 8    positive dose-response between glioma and blood leads, based on small numbers (Anttila et al.,
 9    1996).
10          The National Toxicology Program (NTP) in 2003 evaluated the carcinogenicity of lead
11    and lead compounds.  A summary of its evaluation can be found in NTPs Report on Carcinogens
12    (www.ntp.niehs.nih.gov/ntp/roc/eleventh/profiles/sl011ead.pdf), and the detailed evaluation is
13    also available (NTP, 2004). NTP, like IARC, concluded that "lead and lead compounds are
14    reasonably anticipated to be human carcinogens based on limited evidence from studies in
15    humans and sufficient evidence from studies in experimental animals." The NTP considered that
16    "the strongest epidemiologic evidence was for lung and stomach cancer, which are consistently
17    but weakly associated with occupational and industries entailing lead exposure and with indices
18    of individual lead exposure, including job history and biological monitoring of occupationally
19    exposed and general populations. However, most studies of lead exposure and cancer reviewed
20    had limitations, including poor exposure assessment and failure to control for confounders (other
21    factors that could increase the risk of cancer, including lifestyle factors and concurrent
22    occupational exposure to other carcinogens), and did not demonstrate relationships between the
23    amount of exposure (concentration or duration, for example) and the magnitude of cancer risk."
24    NTP, like IARC, also relied heavily on occupational cohort studies in its evaluation of the
25    epidemiologic evidence. NTP (2003) noted that "the mechanisms by which lead causes cancer
26    are not understood. Lead compounds do not appear to cause genetic damage directly, but may
27    do so through several indirect mechanisms, including inhibition of DNA synthesis and repair,
28    oxidative damage, and interaction with DNA-binding proteins and tumor-suppressor proteins."
29          Both the IARC and NTP evaluations of human evidence relied primarily on occupational
30    studies of highly exposed workers, in which limited evidence of stomach and to some extent lung
31    carcinogenicity was found.  There are seven such studies with relatively large populations

      December 2005                          6-23 5        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Anttila et al., 1995; Carta et al., 2005; Fanning, 1988; Gerhardsson et al., 1995a; Lundstrom
 2    et al., 1997; Steenland et al., 1992; Wong and Harris, 2000). A further study (Ades and
 3    Kazantzis, 1988) also addresses lead exposure in a large occupational cohort, although it is
 4    compromised by the strong  correlation between arsenic and lead exposure in the cohort.
 5    It should be noted that the blood lead levels among these workers were generally three to five
 6    times higher than the blood  lead levels in the two studies of the general U.S. population (Jemal
 7    et al., 2002; Lustberg and Silbergeld, 2002; both based on NHANES II) with environmental
 8    exposures. For example, mean blood levels in two studies of U.S. lead smelter workers averaged
 9    56 |ig/dL in Steenland et al. (1990) in 1976 and 80 |ig/dL in Cooper et al. (1985) during the
10    period 1947-1972.  In contrast, blood levels in the U.S. population enrolled in NHANES II in
11    late  1976-1980 averaged 14 |ig/dL. General population blood lead levels have decreased
12    markedly since the 1970s in many industrial countries with the banning of leaded gasoline. For
13    example, in the U.S. in the early 1990s general population levels averaged 3 |ig/dL according to
14    NHANES III (www.atsdr.cdc.gov/toxprofilesA see lead toxicological profile, page 409).
15    Regarding the occupational  studies, while exposure is well documented, exposure-response data
16    are generally not available, making impossible any quantitative inference about likely cancer
17    effects in low exposure groups based on these studies. The high exposure occupational cohorts
18    are the most informative for deciding whether lead is likely to cause cancer, simply because high
19    doses are more likely to show detectable effects than low doses, if effects exist.  If lead does
20    cause cancer, and assuming there is no threshold below which exposure does not cause cancer
21    (which is generally true for  human carcinogens), current low level exposures to the general
22    public may result in some level of cancers related to lead exposure due to the potential exposure
23    of a large number of people.
24
25    6.7.3    Meta-Analyses  of Lead and  Cancer
26         There have been two published meta-analyses of the carcinogenicity of lead and lead
27    compounds.  The major findings of these studies are summarized in Table 6-7.1. Steenland and
28    Boffeta (2000) relied on eight occupational cohort studies  of highly-exposed workers (seven
29    cohort studies, one nested case-control), all of which had documentation of exposure levels.
30    Meta-analyses were conducted for lung, stomach, kidney,  and brain cancer.  The combined lung
31    cancer relative risk relative risk was  1.30 (95% CI:  1.15,1.46), based on 675 lung cancer deaths.


      December 2005                          6-236        DRAFT-DO NOT QUOTE OR CITE

-------
             Table 6-7.1. Results of Meta-Analyses Addressing the Association Between
                                   Lead Exposure and Cancer
Risk Estimate (95% CI) for indicated outcome
[Number of studies utilized in estimate]
Meta-Analysis
Fu and Boffetta
(1995)
Fu and Boffetta
(1995)
Steenland and
Boffetta (2000)
Lung Cancer
1.24(1.16, 1.33)
[n=15]
1.42(1.05, 1.92)
[battery /smelter only]
1.30(1.15, 1.46)
[n = 8 - cohort only]
Stomach Cancer
1.33(1.18, 1.49)
[n=10]
1.50(1.23, 1.83)
[battery /smelter only]
1.34(1.14, 1.57)
[n = 8 - cohort only]
Renal Cancer
1.19(0.96,1.48)
[n=5]
1.26(0.70,2.26)
[battery /smelter only]
1.01 (0.72, 1.42)
[n = 7 - cohort only]
 1    However, the authors noted that the lung cancer findings were not consistent across studies, and
 2    were influenced highly by one study (Lundstrom et al., 1997) in which confounding by arsenic
 3    was likely. Exclusion of this study dropped the combined lung cancer relative risk to 1.14
 4    (95% CI: 1.04, 1.73). The strongest positive evidence was for stomach cancer (relative risk 1.34
 5    [95% CI: 1.14, 1.57], 181 observed deaths).  There was little positive evidence for renal cancer
 6    (relative risk 1.01 [95% CI: 0.72, 1.42], 40 deaths), or brain cancer (relative risk 1.06 [95% CI:
 7    0.81, 1.40]). All meta-analyses used fixed effects models, given that no evidence of
 8    heterogeneity was found across studies (there was significant heterogeneity for lung cancer
 9    when the Lundstrom et al. study was included, but not when it was excluded).
10          Fu and Boffetta (1995) conducted an earlier meta-analysis  in which they reviewed
11    16 cohort and 7 case-control studies. Different numbers of studies were used for meta-analyses
12    of different outcomes, dependent on whether that outcome was reported separately, among other
13    factors.  These authors focused their analysis on the occupational  studies. Twelve of these
14    studies were used in a meta-analysis of lung cancer, resulting in a combined relative risk of 1.29
15    (95% CI: 1.10, 1.50) (random effects model). There was significant heterogeneity  of lung
16    cancer results across studies. Meta-analyses using fixed effects (no significant heterogeneity
17    between studies) resulted in relative risks of 1.33 (95% CI: 1.18,  1.49) for stomach cancer
18    (10 studies), of 1.19 (95% CI: 0.96, 1.48) for kidney cancer (5 studies), and 1.41 (95% CI: 1.16,
19    1.71) for bladder cancer (5 studies). No meta-analysis was conducted for brain cancer. Separate
20    analyses for stomach, lung, and kidney cancer were also conducted for those studies with the

      December 2005                            6-237        DRAFT-DO NOT QUOTE OR CITE

-------
 1    highest occupational exposure to lead (3 to 5 studies of battery and smelter workers), which
 2    resulted in slightly higher relative risks. The authors concluded that "the findings from the
 3    workers with heavy exposure to lead provided some evidence to support the hypothesis of an
 4    association between stomach and lung cancer and exposure to lead.  The main limitation of the
 5    present analysis is that the excess risks do not take account of potential confounders, because
 6    little information was available for other occupational exposures, smoking, and dietary habits.
 7    The excess risk of stomach cancer may also be explained, at least in part, by nonoccupational
 8    factors. For bladder and kidney cancers, the excess risks are only suggestive of a true effect
 9    because of possible publication bias.
10
11    6.7.4  Genotoxicity of Lead
12          The NTP reviewed in some detail the genotoxicity studies over the period 1970-2002.
13    These studies are  cross-sectional studies, mostly of occupationally exposed workers compared to
14    a control population. Usually blood lead levels are available to document exposure. Outcomes
15    consisted  of chromosomal aberrations (CA), sister chromatid exchange (SCE), micronuclei
16    formation (MN), and studies of DNA damage (often via the comet assay) and/or measures of the
17    mitotic activity. Of these outcomes, only CAs have been shown to have a positive relationship to
18    subsequent cancer (Hagmar et al., 2004, Rossner et al., 2005).  SCEs are generally considered a
19    marker of exposure to environmental agents which have some effect on DNA, but are not
20    thought to necessarily predict cancer risk.  MN and DNA damage are thought to indicate
21    genotoxicity with unknown effect on cancer risk. These outcomes are somewhat informative
22    regarding the possible human carcinogenicity of lead but are clearly secondary to direct
23    information on cancer risk from epidemiologic studies.
24          The most recent studies of the genotoxicity of lead are summarized in Annex Table
25    AX6-7.1.  Of eleven studies of chromosomal aberrations (CA),  six were judged to show a
26    positive relationship between CA and lead, four were judged negative, and one was neither
27    clearly positive or negative. In general, these studies were done in the 1970s and 1980s; only
28    one dates  from the 1990s.  There were nine studies of sister chromatid exchange.  Of these, four
29    were judged positive, three negative, and two could not be judged clearly one way or the other.
30    It is notable that the positive studies were generally the most recent. There were four MN
31    studies, all of which were judged positive. Finally, there were nine studies of DNA damage

      December 2005                           6-23 8       DRAFT-DO NOT QUOTE OR CITE

-------
 1    and/or mitotic activity.  These varied in the specific outcome, although many used a comet assay
 2    to measured oxidative damage to DNA. Eight of these nine studies were judged positive in the
 3    sense that increased DNA damage or mitotic activity was related to lead exposure, while one was
 4    judged negative.
 5          Since the NTP review, there have been three additional cytogenetic studies which are
 6    informative regarding lead (Palus et al., 2003, Minozzo et al., 2004, and Fracasso et al., 2002),
 7    as well as one mutation study (Van Larebeke et al., 2004).  All four of these studies (two of DNA
 8    damage one of MN, and one of a specific mutation frequency) were positive in significantly
 9    linking lead exposure to the outcome. The results  of these studies as well as those reviewed by
10    the NTP are summarized in Table 6-7.2.
11
12
         Table 6-7.2.  Results of Epidemiologic  Studies on the Genotoxicity of Lead Exposure3
Studied Outcome
Chromosomal Aberrations
(CA)
Sister Chromatid Exchange
(SCE)
Micronucleus Formation
(MN)
DNA Damage/Mitosis
Gene Mutation

Positive
6
4
5
10
1
Results
Mixed
1
2
0
0
0

Negative
4
o
J
0
1
0
        Results summarize the overall findings of epidemiologic studies addressing the potential genotoxic effects of
        lead exposure. Some studies addressed multiple aspects of genotoxicity; for these studies, their results for each
        of the listed categories of genotoxic outcomes are presented separately.
13          While the overall the evidence from cytogenetic studies is mixed, more recent studies
14   which were focused on DNA damage or mitotic activity have tended to be largely positive.
15   However, it is not known whether these outcomes predict subsequent cancer risk.
16
     December 2005                            6-239        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.7.5   Review of Specific Studies on the Carcinogenicity of Lead Since the
 2            1986 Lead AQCD
 3    6.7.5.1    Introduction
 4          The epidemiologic studies of lead exposure and cancer are listed in Table 6-7.3. The most
 5    relevant studies focus on exposure through occupational sources, wherein the most intense
 6    exposure to lead can be expected to occur. This exposure predominantly involves inorganic lead
 7    species.  Relevant studies are discussed below, beginning with the most key studies of the
 8    general population will then be presented followed by a brief summary of other relevant studies
 9    examining the occupational studies.
10
11    6.7.5.2    Key Studies of Occupational Populations in the U.S.
12          There are seven key occupational studies based on highly exposed worker populations;
13    these are all cohort studies with adequate numbers to address lung and/or stomach cancer.
14    There are two cohorts based in the U.S. and five based outside it.  Studies reviewed in this
15    section are summarized in Annex Table AX6-7.2.
16          Steenland et al. (1992) followed up 1,990 male U.S. lead smelter workers, employed from
17    1940 to 1965, through 1988. Standardized mortality ratios indicated an excess of lung, stomach,
18    kidney, and bladder cancer, but these excesses did not reach statistical significance. Focusing on
19    a subgroup of workers classified as highly lead exposed based on air-monitoring records yielded
20    a statistically significant excess for kidney cancer (standardized mortality ratio of 2.39 [95% CI:
21    1.03, 4.71]), although it did not appear to increase with increasing duration of exposure.
22    Estimates for the other cancers (standardized mortality ratio of 1.11  [95% CI:  0.82, 1.47] for
23    lung; 1.28 [95% CI:  0.61, 2.34] for stomach; 1.33 [95% CI: 0.48, 2.90] for bladder) showed
24    little change with restriction to the high-exposure group.  While neither arsenic nor cadmium
25    exposure could be controlled for, 1975 NIOSH monitoring data indicated less intense  exposure
26    to airborne cadmium or arsenic than to lead.  Lead averaged 3.1 mg/m3 and arsenic  14 |ig/m3,
27    compared to current OSHA standards of 0.05 mg/m3 for lead and 10 |ig/m3 for arsenic. It is
28    notable that a  1996 review of studies (Steenland et al., 1996) on arsenic-exposed workers
29    concluded that significantly elevated rates of lung cancer were concentrated in studies where
30    average exposures greatly exceeded OSHA standards (e.g., hundreds of |ig/m3).  No data on
31    workers' smoking status were available.

      December 2005                            6-240        DRAFT-DO NOT QUOTE OR CITE

-------
         Table 6-7.3.  Epidemiologic Studies of Lead Exposure and Cancer in Specific
                      Populations, by Geographic Region and Study Design3
                                                       Epidemiologic Study Design
 Specific Study Population
           Cohort
  Nested Case-control
Case-control
 United States

 Battery and lead production
 workers
 Copper workers (Utah)

 Lead and zinc pigment plant
 workers

 Lead smelter workers (Idaho)
 Sample of deaths due to cancer
 vs. noncancer deaths (Illinois)

 Brain cancer

 Central nervous system cancer

 Stomach cancer

 NHANES II cohort mortality
 follow-up, general U.S.
 population

 Canada

 Population-based cases

 Specific cancers versus all
 cancers

 Europe

 Glass workers (Finland)

 Registry-derived liver cancer
 cases vs. stomach cancer or
 myocardial infarctions (Finland)

 Workers via Cancer Registry
 (Finland)

 Renal-cell cancer vs. population
 controls (Germany)

 Laryngeal cancer among
 persons with no history of lead
 exposure (Greece)
Cooper and Gaffey (1975), Cooper  Cooper et al. (1989),
et al. (1985), Wong and Harris      Wong and Harris (2000)
(2000)                          (same publication as
                               cohort study)

Rencheretal. (1977)

Sheffetetal. (1982)
Selevan et al. (1985), Steenland et
al. (1992)
Jemal et al. (2002), Lustberg and
Silbergeld (2002)
 Sankilaetal. (1990)
 Anttilaetal. (1995)
                               Kauppinen et al. (1992)
Anttilaetal. (1996)
                                                       Mallinetal. (1989)


                                                       Coccoetal. (1998a)

                                                       Coccoetal. (1998b)

                                                       Coccoetal. (1999)
                                                       Rischetal. (1988)

                                                       Siemiatycki et al. (1991)
                                                       Pesch et al. (2000)
                                                       Kandiloris et al. (1997)
December 2005
                    6-241
    DRAFT-DO NOT QUOTE OR CITE

-------
    Table 6-7.3 (cont'd). Epidemiologic Studies of Lead Exposure and Cancer in Specific
                      Populations, by Geographic Region and Study Design3
                                                        Epidemiologic Study Design
    Specific Study Population
           Cohort
  Nested Case-control
Case-control
 Europe (cont'd)

 Glass workers (Italy)

 Lead and zinc miners:  females
 only (Sardinia)

 Lead and zinc miners:  male only
 (Sardinia)

 Lead and zinc smelter workers
 (Sardinia)

 Lead and zinc smelter workers
 (Sardinia, but different from
 Coccoetal. 1996)

 Glass workers (Sweden)
 Copper and lead smelter workers
 (Sweden)

 Copper and lead smelter workers
 (Sweden) (Lundstrom:  full
 cohort; Englyst:  sub-cohort)

 Lead-acid battery workers (U.K.)
 Chromate (including lead-
 chromate) workers (U.K.)

 Zinc, cadmium, and lead smelter
 workers (U.K.)


 Asia
 Gliomas vs. noncancer patients
 (China)
 Meningiomas vs. noncancer
 patients (China)

 Gall bladder cancer vs. gallstone
 patients (India)
 Prostate cancer cases versus
 benign prostate hyperplasia cases
 and normal controls (India)
Cordiolietal. (1987)

Coccoetal. (1994b)
Cocco et al. (1994a), Carta et al.
(1994); Cartaetal. (2003;

Coccoetal. (1996)
Coccoetal. (1997)
Wingren and Englander (1990)    Wingren and Axelson
                              (1985, 1987, 1993)

Gerhardsson et al. (1995a)
Gerhardsson et al. (1986),
Lundstrom et al. (1997), Englyst
etal. (2001)

Dingwall-Fordyce and Lane
(1963), Malcolm and Bamett
(1982)

Davies(1984a, 1984b)
                        Fanning (1988)
Ades and Kazantzis (1988)
Ades and Kazantzis
(1988) (same publication
as cohort study)
                                                      Huetal. (1998)

                                                      Huetal. (1999)


                                                      Shuklaetal. (1998)

                                                      Siddiqui et al. (2002)
 a Within regions, study populations are listed in chronological order based on the earliest published study on that specific
   worker population.  Publications considered to be key studies are italicized.
December 2005
                   6-242
    DRAFT-DO NOT QUOTE OR CITE

-------
 1          Wong and Harris (2002) extended follow-up on the battery and smelter worker cohort
 2    previously reported on by Cooper et al., 1985 through 1995, an additional 15 years. With the
 3    additional follow-up, standardized mortality ratios for lung, tracheal, or bronchial cancer
 4    decreased to 1.14 (95% CI: 0.99, 1.30) for battery workers but showed little change for smelter
 5    workers at 1.22 (95% CI:  1.00,1.47).  A significantly elevated standardized mortality ratio for
 6    stomach cancer (1.53 [95% CI:  1.12, 2.05]) persisted among battery workers, with a lesser
 7    elevation among smelter workers (1.33 [95% CI:  0.75,2.20]).  Among other cancers, only
 8    thyroid cancer among all workers combined showed a significantly elevated standardized
 9    mortality ratio (3.08 [95% CI:  1.33,6.07]). Cancer mortality did not increase with earlier year
10    of hire for lung, stomach, or thyroid cancer. Lung and stomach cancer mortality peaked among
11    workers with 10 to 19 years of factory employment and declined with longer employment
12    duration. Thyroid cancer mortality occurred exclusively among workers with 20 or more years
13    of exposure. As with earlier analyses based on this cohort, concomitant exposures to other
14    compounds could not be controlled for, but as these were likely to be most intense among lead
15    production workers, whose standardized mortality ratios were similar to or lower than those for
16    battery workers, any bias resulting from such exposure probably was toward the null.  No data
17    were available to  assess the possible role of smoking, diet, or other potential nonoccupational
18    risk factors in the results.
19          A nested case-control analysis was also conducted to further explore stomach cancer
20    mortality within workers employed at the Philadelphia lead battery plant in the cohort (Wong
21    and Harris, 2000). Among 30 workers who died of stomach cancer and  120 age-matched
22    controls, job title histories were used to estimate duration of employment and cumulative
23    exposure based on job-specific intensities of exposure. Duration of employment and estimated
24    degree of lead exposure showed no elevation among workers who died of stomach cancer, nor
25    did mortality increase across increasing tertiles of lead exposure.  Little information appeared to
26    be available on potential confounders.  The authors suggested that in light of historically higher
27    stomach cancer rates in Ireland and Italy and the observation of a higher proportion of Irish and
28    Italian immigrants among lung cancer cases in the case-control study, differences in ethnicity
29    may have contributed to the elevated standardized mortality ratios seen in the cohort as a whole.
30    The recent IARC  Working Group (IARC, 2005) concluded that, based on the ethnic composition
      December 2005                           6-243       DRAFT-DO NOT QUOTE OR CITE

-------
 1    of the control population (23% Irish or Italian), confounding by race could account for only part
 2    of the observed association, however.
 3          The extended follow-up and nested case-control analyses on the original Cooper et al.
 4    (1985) cohort thus continued to provide evidence for some increase in lung and stomach cancer
 5    among these lead workers, but no consistent evidence of increasing cancer risk with increasing
 6    exposure within the lead worker cohort itself, especially for stomach cancer.
 7          Fanning (1998) studied deaths due to specific cancer types among U.K. battery and other
 8    factory workers.  High to moderate lead exposure resulted in odds ratios for lung and digestive
 9    cancer of 0.93 and 1.13, respectively, with the latter elevation due mainly to stomach cancer
10    (odds ratio of 1.34).  No odds ratios reached nominal statistical significance, and no associations
11    were noted for other cancer types. The excess of digestive cancer deaths was restricted to the
12    1926 to 1965 period, during which lead exposures would have been most intense. Odd ratios for
13    other cancers did not vary by period. Because each cancer case group was compared with a
14    control group consisting of subjects who died from  all other causes, including other cancers,
15    odds ratios would have been biased  downward if some of these other deaths also were lead-
16    related. However, most deaths were due to nonmalignant respiratory or circulatory diseases
17    other than hypertension, mitigating the potential impact of such a bias.
18          Anttila et al. (1995) linked 20,700 Finnish workers whose blood lead was monitored
19    during 1973 to  1983 by the Finnish Institute of Occupational Health to the Finnish Cancer
20    Registry. Exposure was subdivided according to highest peak blood level measured: low (0 to
21    0.9 |imol/L [0 to 18.6 jig/dL]), moderate (1.0 to  1.9 |imol/L [20.7 to 39.4 jig/dL]), and high
22    (2.0 to 7.8 |imol/L [41.4 to  161.6 |ig/dL]). The total cohort showed no elevation in total or site-
23    specific cancer mortality, based on an  standardized  mortality ratio analysis.  Among male
24    workers with moderate exposure, incidence of total respiratory cancer and lung cancer both were
25    elevated (standardized incidence ratio  of 1.4 [95% CI:  1.0, 1.9 for both]). Risks of total
26    digestive, stomach, bladder, and nervous-system cancer also were modestly elevated.  However,
27    risks did not increase in the high-exposure group. Risks of mortality for all cancer for both men
28    and women (relative risk 1.4 of [95% CI:  1.1, 1.8]) and lung or tracheal cancer (relative risk of
29    2.0 [95% CI: 1.2, 3.2]) were even stronger when a person-year analysis was applied to compare
30    workers with moderate lead exposure to those with  low exposure. Again, risks did not increase
31    in the highest exposure group. This exposure group was smaller than the others, however, which

      December 2005                           6-244       DRAFT-DO NOT QUOTE OR CITE

-------
 1    limited the power of analyses specific for high exposure workers. Thus, for example, the
 2    numbers of lung or tracheal cancer deaths among men in the low-, moderate-, and high-exposure
 3    groups were 25, 34, and 11, respectively, for the person-year-based analyses.
 4          It should be noted that for cancer, cumulative exposure, particularly during the earlier part
 5    of the follow-up period, might be more relevant than peak exposure, although the two were
 6    reported to be highly correlated (Anttila et al., 1995). Case-referent substudies of lung cancer
 7    used different exposure criteria (Anttila et al.,  1995). Odds ratios increased most consistently
 8    with increasing cumulative exposure to lead. Among histologic subtypes, significantly elevated
 9    risk for squamous-cell cancer of the lung (odds ratio of 4.1 [95% CI:  1.1, 15]) for the highest
10    blood lead group persisted after adjustment for smoking, although with additional adjustment for
11    engine exhaust and solvent exposure, the risk declined (odds ratio of 3.4 [95% CI: 0.9, 13]).
12    Results for female workers are not considered, as too few cancers (3 total) occurred to permit
13    meaningful conclusions.  Although the  follow-up period was relatively short, the lung cancer
14    association was analyzed in much greater detail than in most studies, and smoking was adjusted
15    for, while the association between lead  exposure and lung cancer weakened with control for
16    engine exhaust and solvent exposure, the odds ratio remained well above 1. The highest odds
17    ratio of all was observed for estimated risk of lung cancer among workers with peak blood lead
18    levels of at least 0.8 jimol/L [> 16.6 |ig/dL] who were exposed to engine exhaust (odds ratio of
19    14.9 [95% CI:  1.3, 178]; 11 cases). If engine exhaust was acting as an effect modifier, directly
20    controlling for it might not have been appropriate. The exhaust could have served as a source of
21    organic lead, as well.
22          Gerhardsson et al. (1995a) followed up 664 male Swedish secondary lead smelter
23    workers, tracing their cancer morbidity from 1969 to 1989. Compared to the population of the
24    surrounding county, the workers' standardized incidence ratio for all cancers was 1.27 (95% CI:
25    0.91, 1.74), based on 40 tumors. Standardized incidence ratios for cancers at all specific sites
26    except the brain were elevated, notably those for the respiratory system (1.32 [95%  CI: 0.49,
27    2.88]), stomach (1.88 [95% CI: 0.39, 5.50]), and colon (1.46 [95% CI:  0.30, 4.28]). Because of
28    the small numbers of tumors (only 6, 3, and 3, respectively, even for the aforementioned sites),
29    the reliability of estimates for most sites is limited.  Restricting analyses to workers  in the highest
30    quartile of lead exposure based on routine blood lead monitoring data yielded a higher
31    standardized incidence ratio for total gastrointestinal cancer (2.43 [95% CI: 1.11,4.62];

      December 2005                           6-245        DRAFT-DO NOT QUOTE OR CITE

-------
 1    9 tumors), but not respiratory cancer. Availability of blood lead measurements is an advantage
 2    of this study, along with a lead-exposed worker population unlikely to have much exposure to
 3    arsenic, chromium, or cadmium. However, the cases were too few for detailed exposure-
 4    response analyses by cancer type. Lack of data on smoking further restricts interpretation of the
 5    results.
 6          Lundstrom et al. (1997) followed 3,979 Swedish smelter workers from 1928 to 1987.
 7    Workers were further subdivided into those with high cumulative blood lead  scores (mean times
 8    years exposed >10 |imol/L), and those exposed to "lead only" (excluding those from departments
 9    thought to have significant exposures to other potential carcinogens, such as arsenic, or little
10    exposure to lead). The lung cancer standardized mortality ratio was 2.8 (95% CI:  2.0, 3.8)  for
11    the total cohort, 2.8 (95% CI:  1.8, 4.5) for the high-exposure subgroup, and reportedly similar
12    for the subgroup exposed to lead only. With adjustment for a 15-year latency period, lung cancer
13    standardized incidence ratios likewise differed little between the total cohort  and high-exposure
14    subgroup; however, among workers with exposure to lead only, the standardized incidence ratio
15    rose from 3.1 (95% CI:  1.7, 5.2; 14 cases) for all workers to 5.1 (95% CI:  2.0, 10.5; 7 cases) for
16    those with the highest exposure. With a 15-year latency period, elevated standardized incidence
17    ratios also were observed for cancer of the brain and nervous system (1.6 [95% CI:  0.4, 4.2])
18    and renal pelvis, ureter, or bladder (1.8 [95% CI:  0.8, 3.4]) among the high-exposure subgroup.
19    Non-respiratory cancers were too infrequent (5 total) in the high-exposure lead-only subgroup
20    for meaningful analysis.  This study's size, extensive follow-up, and ability to integrate blood-
21    based and job-based exposure indices give it unusual power. The  apparent increase in cancer
22    risk with higher cumulative lead exposure that appeared when workers thought to be potentially
23    exposed to other metals, such as arsenic and nickel, were excluded also appeared to strengthen
24    the evidence for a specific link between lead  and respiratory cancer. A subsequent study by
25    Englyst et al. (2001), however, cast doubt on the efficacy of the "lead only" grouping.
26          Englyst et al. (2001) conducted additional analyses on one element of the Lundstrom et al.
27    (1997) cohort.  A total of 1,093 workers from the smelter's lead department was followed up
28    through 1997.  Significantly elevated lung cancer standardized incidence ratios were observed
29    in all subcohorts, including the subcohort who had never worked in arsenic-exposed areas
30    (3.6 [95% CI:  1.2, 8.3]; 5 cases). This subcohort is the same as the "lead-only" subgroup
31    evaluated by Lundstrom et al. (1997).  A review of detailed job histories obtained for all workers

      December 2005                           6-246       DRAFT-DO NOT QUOTE OR CITE

-------
 1    with lung cancer, however, indicated that 13 of the 15 had "considerable" exposure to arsenic as
 2    well as lead, including all but 1 in the "lead only" subcohort.
 3          Carta et al. (2003) followed up the mortality of 918 Sardinian lead smelter workers from
 4    1972 through 2001. Smelter workers as a whole displayed an overall cancer mortality no higher
 5    than expected based on regional rates (standardized mortality ratio of 1.01). Cancer-specific
 6    standardized mortality ratios were, however, nonsignificantly elevated for cancers of the lung
 7    (1.21) and stomach (1.22) as well as for lymphoma and leukemia (1.82). Use of blood and
 8    ambient lead monitoring data available by department and task to categorize estimated exposure
 9    yielded a statistically significant upward trend with increasing lead exposure for lung cancer; no
10    significant trend was seen for the other cancers, although in  light of the small number of gastric
11    cancer and lymphoma/leukemia deaths (4 and 6, respectively) interpretation of dose-response is
12    problematic for these outcomes.
13
14    6.7.5.3   Key Studies of the General Population
15          There are two key general population cohort studies in which lead exposure is assessed
16    via blood lead levels (see Annex Table AX6-7.3 for additional details). Jemal et al. (2002)
17    conducted the first biomarker-based general population cohort study of lead exposure and
18    cancer. The study employed the subsample of 3,592 white U.S. participants in NHANES II
19    (1976 to 1980) who had undergone blood lead level determinations at time of entry. Deaths
20    among this population were enumerated through 1992 by linkage to the National Death Index
21    (NDI) and Social Security Administration Death Master File. Median blood lead levels in this
22    population were 12 |ig/dL. Adjusted for age, smoking,  drinking, region, year, and gender, risk of
23    mortality from any cancer rose across quartiles of blood lead level, but this trend was not
24    statistically significant. The trend across quartiles was not consistent in gender-specific analyses,
25    although relative risks were elevated for the highest quartile of blood lead level in both men and
26    women (relative risk 2.0 for men and  1.6 for women).  The relative risk for lung cancer based on
27    comparison of subjects with blood lead levels above or  below the median was 1.5 in the
28    combined population, with higher risk observed among women (2.5 [95% CI:  0.7, 8.4]) than
29    men. The highest relative risks were observed for cancer of the esophagus (3.7 [95% CI: 0.2,
30    89]), pancreas (3.6 [95% CI: 0.6, 19.8]), and stomach (2.4 [95% CI: 0.3, 19.1]); no elevations
31    were noted for cancers of other sites.

      December 2005                            6-247      DRAFT-DO NOT QUOTE OR CITE

-------
 1          The lack of statistically significant results reflects the small number of deaths during
 2    follow-up, which limited the study's power; of the nine major sites examined, the number of
 3    deaths ranged between 5 and 16 for all sites except the lung. Detailed exposure-response
 4    analyses were restricted to all cancers combined, although potential effects could have been
 5    strongly target-organ specific.  In addition, the use of quartile cut points based on the distribution
 6    of lead concentrations estimated for the total U.S. population resulted in relatively  small numbers
 7    in the referent group (lowest exposure quartile) for males and in the high-exposure quartile for
 8    females. Use of a biomarker provided an objective measure of lead exposure. Nevertheless,
 9    reliance  on a single blood lead measurement produces less reliable  estimates than would be
10    obtained through multiple measurements and precludes addressing  temporal changes in lead
11    exposure over the follow-up period. Lack of control for exposure to occupational carcinogens
12    other than lead and potential residual confounding by duration and  intensity of tobacco smoking
13    also could have biased the results, especially for men. Lustberg and Silbergeld (2002) carried
14    out another biomarker-based general population study based on the same NHANES II mortality
15    cohort used by Jemal et al. (2002).  This study did not exclude nonwhites, however (thus gaining
16    524 subjects) and employed more extensive adjustment for potential confounding factors than
17    the Jemal et al. (2002) analyses (i.e., education, body mass index, and exercise were included in
18    the regression models,  although alcohol intake was not).  In addition, persons with blood lead
19    levels of 30 |ig/dL or higher were excluded in order to restrict comparisons to levels below the
20    OSHA standard for lead exposure.  Persons with levels below 10 |ig/dL served as the  referent
21    group. Survival analyses adjusted for potential confounders found a relative risk for cancer
22    mortality of 1.5 (95% CI: 0.9, 2.5) for those with blood lead levels of 10 to 19  |ig/dL, compared
23    with those with levels below 10 |ig/dL, rising  to 1.7 (95% CI: 1.0,  2.8) for those with levels of
24    20 to 29 |ig/dL. Separate analyses of lung-cancer and non-lung-cancer deaths yielded estimates
25    of increased risk for moderate- or high-exposure groups, compared with the referent population,
26    both for lung cancer and non-lung cancer.  However, none of the estimates reached the P < 0.05
27    level of statistical significance, and the results for non-lung cancers showed no evidence of an
28    exposure-response relationship.
29          As with Jemal et al. (2002), the use of a biomarker for exposure and the  prospective
30    design of the study are strengths. Its attempts to control for potential confounders were more
31    extensive, and its choice of cut points for the referent category yielded more males in the referent

      December 2005                           6-248       DRAFT-DO NOT QUOTE OR CITE

-------
 1    group, although that group still included less than 20% of the study population. However, it is
 2    notable that blood lead levels rose significantly with smoking level.  The models included terms
 3    for former smoking, current light smoking, and current heavy smoking (>1 pack per day).
 4    Nevertheless, some degree of residual confounding due to smoking might have remained, which
 5    could have contributed to the estimated risk of lung cancer for the highest exposure category
 6    (relative risk of 2.2 [95% CI: 0.8, 6.1]).  Such residual confounding would have had less effect
 7    on the results for non-lung cancer. As noted regarding the other NHANES-based study,
 8    however, mortality due to cancers of other sites was too uncommon to allow for reliable site-
 9    specific comparisons. In the Lustberg and Silbergeld analysis, all cause and cardiovascular
10    mortality increased monotonically with blood lead level, which might indicate residual
11    confounding from SES or smoking affecting both heart disease and cancer.
12
13    6.7.5.4   Other Lead Studies
14           There are a variety of other epidemiologic studies of lead exposure, which are less
15    important than the key studies above but which offer some information.  Studies reviewed in this
16    section are summarized in Annex Table AX6-7.4. Rencher et al. (1977) compared Utah copper
17    smelter workers' mortality with that of miners for the same company. Workers in lead-exposed
18    operations had a higher proportional mortality due to respiratory cancer in general and lung
19    cancer specifically than did other workers, with or without control for smoking status. Among
20    lead-exposed workers, those  who developed lung cancer had significantly higher estimated lead
21    exposure than the rest. Workers with lung cancer also had significantly higher estimated
22    exposure to arsenic and sulfur dioxide, however,  and these exposures were not adjusted for.
23          Ades and Kazantzis (1988) conducted a cohort study  of lung cancer mortality among
24    4,393 U.K. zinc, lead, and cadmium smelter workers. Smelter workers had a lung cancer
25    standardized mortality ratio of 1.25 (95% CI: 1.07, 1.44) compared with national rates, based on
26    182 lung cancer deaths Potential effects of lead could not be adjusted for arsenic exposure or
27    other exposures due to inadequate numbers.  Cancer-specific standardized mortality ratios were
28    calculated for production and maintenance workers from an  Italian lead and zinc smelter
29    followed from 1950 to 1992 by Cocco et al. (1997). Deaths from lung,, stomach,  and all cancer
30    were not elevated over regional rates. Cocco et al. (1996) followed 1,060 Sardinian lead and
31    zinc smelter workers whose glucose-6-phosphate dehydrogenase (G6PD) phenotype had been

      December 2005                          6-249       DRAFT-DO NOT QUOTE OR CITE

-------
 1    measured from 1973 through 1991. Despite the thought that G6PD-deficient workers might be
 2    more vulnerable to the depletion of red blood cell glutathione associated with lead toxicity,
 3    mortality from cancer and from all causes was slightly lower among G6PD-deficient workers
 4    than among G6PD-normal workers. Follow-up was subsequently extended through 2001 by
 5    Carta et al. (2003).
 6          Three European studies followed up cohorts of glass workers. Cordioli et al. (1987)
 7    studied 468 Italian glass workers. Workers producing low-quality glass containers were
 8    classified as being exposed to lead.  A small elevation in mortality from all cancer (standardized
 9    mortality ratio of 1.3 [95% CI:  0.8, 1.8]) among glass workers was driven by significant
10    excesses in lung cancer (2.1 [95% CI:  1.1, 3.6]) and laryngeal cancer (4.5 [95% CI:  1.2, 11.4]).
11    The small number of deaths among exposed workers (28 total, 13 lung, and 4 laryngeal cancer)
12    limited the study's statistical power.  Sankila et al. (1990) compared the incidence of cancer in
13    1,803 male and 1,946 female Finnish glass workers with that of the national population.
14    Glassblowers were considered to be a lead-exposed subgroup. Modest elevations in lung cancer
15    risk were observed among glass workers for both men (standardized incidence ratio of 1.3 [95%
16    CI: 1.0,  1.7])  and women (1.1 [95% CI:  0.5,2.3]). However, the increased risk of lung cancer
17    was not specific to glassblowers.  In the final study, Wingren and Englander (1990) compared
18    mortality in Swedish glass workers from work areas with airborne lead levels ranging from
19    <0.001 up to 0.110 mg of lead/m3, noting a significant elevation for pharyngeal cancer
20    (standardized mortality ratio of 9.9 [95% CI: 1.2, 36.1]) and nonsignificant elevations for lung
21    and colon cancer compared to national  rates.
22          Wingren and Axelson (1985, 1987, 1993) conducted a case-control analysis comparing
23    stomach, colon, and lung cancer mortality among Swedish glass workers with that of the
24    surrounding regional populations.  A small early study of three parishes (Wingren and Axelson,
25    1985) was expanded to include 11 parishes, thus encompassing most of the Swedish glass-work
26    industry (Wingren and Axelson 1987).  Mortality from cancer of the lung (odds ratio of 1.7 [90%
27    CI: 1.1, 2.5]), stomach (1.5 [90% CI:   1.1, 2.0]), and colon (1.6 [90% CI: 1.0, 2.5]) all were
28    elevated among glass workers as a whole (Wingren and Axelson, 1987).  Among specific classes
29    of glass workers, glassblowers had the  highest odds ratios (2.3, 2.6, and 3.1 for lung, stomach,
30    and colon cancer, respectively).  When the data were analyzed according to level of estimated
31    metal exposure, no consistent dose-response trend with lead was found for lung cancer, and the

      December 2005                           6-250       DRAFT-DO NOT QUOTE OR CITE

-------
 1    association with stomach cancer was weaker for lead than for arsenic, copper, and other metals.
 2    In general in this study it was difficult to separate the independent cancer effects of different
 3    metals.
 4          Sardinian lead and zinc miners were studied in a set of three papers published in 1994.
 5    Carta et al. (1994) studied a small group of workers and Cocco et al. (1994a,b) expanded
 6    coverage to follow 1,741 male and 526 female workers from two mines.  Number of cancer
 7    deaths were small. This study was limited because exposure characterization focused only on
 8    silica and radon daughters; no lead exposure specific analyses were performed.
 9          Davies (1984b) followed up 57 pigment factory workers who had been diagnosed with
10    nonfatal lead poisoning, finding a small excess of lung cancer deaths (relative risk of 1.45), but
11    with only 4 deaths in the lead-poisoned group this  result did not reach statistical significance.
12          Mallin et al. (1989) used death certificates for Illinois males to compare deaths from seven
13    specific cancers with a control group of 3,198 randomly selected deaths from other causes.
14    Based on occupations from death certificates, the odds ratio for cancer of the brain (3.0,
15    p < 0.05) was significantly elevated in white male  glass workers (as well as physicians and
16    communications workers).  No significant association was observed for other cancer sites,
17    including lung and stomach.  This isolated association is not consistent with the results for
18    Swedish glass workers summarized above. The National Cancer Institute, NIOSH, and the
19    National Center for Health Statistics have assembled a database that integrates industry,
20    occupation, and cause of death information from death certificates in 24 states.  This resource
21    provides a very large sample size for case-control analyses of occupational exposures, but results
22    are limited by a lack of detailed work history and no control over confounders.  Cocco et al.
23    (1998a) matched all 27,060 brain cancer deaths occurring among persons aged 35 or older during
24    1984 to 1992 with four gender-, race-, age-, and region-matched deaths from nonmalignant
25    causes. A job-exposure matrix was used to assign subjects to low, medium,  or high probability
26    and intensity of exposure. Risk of brain cancer mortality increased consistently with rising
27    intensity of lead exposure among African American men but not among the other three race-
28    gender groups. Cooco et al. (1998b) broadened the study  to CNS cancer deaths, and computed
29    odds ratios for specific industries and occupations  rather than particular substances. Statistically
30    significant associations were found with some industries and some race/sex groups, but little
31    inference can be made about lead carcinogenicity from these data.

      December 2005                           6-251       DRAFT-DO NOT QUOTE OR CITE

-------
 1          In the third 24-state death-certificate study, 41,957 stomach cancer deaths were matched
 2    with 83,914 deaths due to nonmalignant causes (Cocco et al., 1999). A job-exposure matrix was
 3    used to assign subjects to low, medium, or high probability and intensity of exposure to lead and
 4    11 other chemicals. Elevated odds ratios occurred among white women (1.53 [95% CI:  1.10,
 5    2.12]), African-American men (1.15 [95% CI:  1.01, 1.32]), and African-American women
 6    (1.76 [95% CI: 0.74, 4.16]) with high probability of lead exposure. Odds ratio in the moderate-
 7    probability group were elevated only for African-American women (1.37 [95% CI:  0.58, 3.21]),
 8    and not elevated for any exposure group among white males. Risk showed no consistent
 9    increase with intensity of exposure in any group. The absence of any association with lead
10    exposure among the largest race-gender group, white males, is notable, as is the general absence
11    of association with intensity of exposure.  More consistent elevations of odds ratios for stomach-
12    cancer mortality were observed for inorganic dust and nitrosamines than for lead.
13          Anttila et al. (1996) presented a nested case-control analysis of 26 Finnish male workers
14    with central nervous system (CNS) cancer and 200 controls, using the same Finnish occupational
15    cohort as in Anttila et al. (1995). For CNS cancer incidence, odds ratios rose with increasing
16    peak lifetime blood lead level; however, the trend was not  statistically significant.  Odds ratios
17    for glioma mortality rose consistently and significantly with increasing peak and mean blood
18    lead level,  as well as duration of and estimated cumulative lead exposure. A strength of this
19    study is the availability of blood lead measurements. Limitations include the small number of
20    cases (10 gliomas among workers with complete exposure information),  short follow-up time
21    (maximum of 15 years), potential  selection bias due to low response rates (60% for cases, 56%
22    for controls), and possible coexposures  such as solvents or other metals.
23          Risch et al. (1988) compared  826 Canadian men with histologically confirmed bladder
24    cancer with 792 Canadian population controls.  Reported occupational  exposure to lead yielded a
25    significantly elevated  smoking-adjusted odds ratio (2.0 [95% CI: 1.2, 3.5]) and a significant
26    trend with  duration of exposure. Of 17  other exposures examined, only one (tar and asphalt) was
27    significantly associated with bladder cancer.  These analyses relied  on self-reported exposure,
28    with the potential for inaccurate recall.  Siemiatycki et al. (1991) conducted a case-control study
29    in Canada  using 3,730 cases of various histologically confirmed cancers. Occupational exposure
30    to 293 substances, including lead,  was estimated from interview data. Elevated odds ratios were
31    noted for cancer of the lung (1.1 [90% CI: 0.9, 1.4]), stomach (1.2 [90% CI:  1.0, 1.6]), bladder

      December  2005                           6-252       DRAFT-DO NOT QUOTE OR CITE

-------
 1    (1.3 [90% CI: 1.0, 1.6]), and kidney (1.2 [90% CI: 1.0, 1.6]). Strengths of this study are
 2    adjustment for smoking and other potential risk factors and reliance on interview-obtained
 3    exposure data, further evaluated by experts. Limitations include potential confounding by the
 4    other 292 occupational exposures and low quantitative detail regarding lead exposure.
 5          Kauppinen et al. (1992) conducted a nested case-control study in Finland, matching 344
 6    primary liver cancer deaths by age and gender to 476 stomach cancer deaths and 385 myocardial
 7    infarct deaths. No association was found between lead and liver cancer, which was not an a
 8    priori  site of interest. Use of a control group with stomach cancer, which some other studies
 9    have linked to lead exposure, may have biased results toward a negative association.
10          In a Chinese hospital-based case-control study, Hu et al. (1998) compared 218 patients
11    with histologically confirmed primary gliomas with 436 patients with non-neurological,
12    nonmalignant disease, matched by age, gender, and residence. An odds ratio could not be
13    calculated for occupational exposure to lead because no glioma patients reported such exposure.
14          In a parallel study, Hu et al. (1999) compared 183 patients with histologically confirmed
15    primary meningiomas with patients with non-neurological, nonmalignant disease, matched by
16    age, gender, and residence. Reported occupational exposure to lead was associated with risk of
17    meningioma in both men (odds ratio of 7.20 [95% CI: 1.00, 51.72]) and women (5.69 [95% CI:
18    1.39, 23.39]).  Some elevation of odds ratios occurred in most of the 14 occupational exposures
19    examined, including exposure to cadmium.  Malczyk et al. (1999) measured urinary lead
20    concentrations in 24 Polish bladder cancer cases.  Ten out of the 24 cases had urinary lead levels
21    above 90 |ig/L, thus exceeding the upper limit of the range estimated as normal for a healthy
22    person (10-90 |ig/L). Results are limited by the lack of any measurements done on persons
23    without bladder cancer from the same area.
24          Pesch et al. (2000) compared occupational exposure to potential carcinogens among
25    935 Germans newly diagnosed with renal-cell cancer and 4,298 controls selected from regional
26    population registries and matched by age, gender, and area of residence. Lifetime job histories
27    and information on smoking habits  and other potential risk factors were collected by interview.
28    Cumulative exposure to lead,  as well as cadmium, solder fumes, welding fumes, and metals in
29    general, was estimated based on previously published job exposure matrices and grouped into
30    four ascending categories; separate  estimates of lead exposure were calculated based on British-
31    and German-developed matrices. After adjustment for age and smoking, odds ratios for renal

      December 2005                            6-253      DRAFT-DO NOT QUOTE OR CITE

-------
 1    cancer were elevated in men (1.5 [95% CI:  1.0, 2.3]) and women (2.6 [95% CI: 1.2, 5.5]) with
 2    the highest lead exposure, compared with the low-exposure groups based on the British matrix.
 3    When exposure was based on the German matrix, the odds ratio was less elevated among men
 4    (1.3 [95% CI:  0.9, 2.0]); no results for women were reported. Strengths of the study are its size
 5    and population base.  The primary limitation is uncertainty regarding the specificity of the results
 6    for lead. Significant associations also were noted for exposure to cadmium, solder fumes, and
 7    organic solvents among men, for example, but no analyses attempting to account for other
 8    exposures were reported.  It is thus unclear how much of the observed risk associated with lead
 9    exposure may be secondary to exposure to cadmium or other agents.
10          Siddiqui et al.  (2002) compared blood lead levels in Indian men with prostate cancer or
11    benign prostatic hyperplasia to levels seen in normal controls of similar  SES.  Lead levels were
12    significantly higher in both prostate cancer and benign prostatic hyperplasia cases than in
13    controls, while zinc levels were lower.
14         Kandiloris et al. (1997) found similar blood lead levels but lower aminolevulinic acid
15    dehydratase (ALAD) activity in 26 laryngeal carcinoma cases compared to 53 controls from the
16    same hospital. Shukla et al. (1998) found significantly higher mean bile lead in 38 newly
17    diagnosed, histologically confirmed gall bladder cancer cases compared to 58 patients with
18    gallstones diagnosed  at the same Indian hospital surgical unit (58.38 ± 1.76 mg/L versus
19    3.99 ± 0.43 mg/L). Cancer cases also showed elevated cadmium and chromium levels.
20
21    6.7.6  Confounding of Occupational Lead Studies Due to  Other Occupational
22           Exposures: Arsenic, Cadmium
23         A number of studies of lead workers come from smelters, where exposures to other metals
24    are common. Of particular concern are other lung carcinogens, especially arsenic (workers
25    exposed to high levels of arsenic historically have had a lung cancer relative risk of 3-4, see
26    Steenland et al. 1996), but also cadmium. Glass workers are also of limited use for inference
27    about lead effects, as they are also typically exposed to arsenic, cadmium, chromium, and nickel,
28    all of which are lung carcinogens (e.g., see Wingren and Axelson, 1993).
29         In some smelters, measurements have been taken which indicate  clearly that exposures to
30    these other carcinogens was minimal and the main suspect is lead (e.g., Steenland et al., 1992).
31    In others, however, one is unable to disentangle the effects of arsenic and lead (Ades and

      December 2005                           6-254       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Kazantis, 1988, Lundstrom et al., 1997).  As a result, these studies cannot yield strong evidence
 2    regarding the possible relation between lung cancer and lead specifically. The study by
 3    Lundstrom et al., 1997 is particularly important in this regard, because it had a high relative risk
 4    of 2.8 (95% CI:  2.0, 3.8), and had an important effect in raising the overall result when included
 5    in meta-analyses (e.g., Steenland and Boffetta [2000], where exclusion of the Lundstrom et al.
 6    study lowered the estimated combined lung cancer relative risk from 1.30 to  1.14).  A subsequent
 7    publication by Englyst et al. (2001) indicated that the smelter workers studied by Lundstrom
 8    et al. (1997) were likely to have had significant exposure to arsenic, and the authors concluded
 9    that it was impossible to separate the effects of lead and arsenic.
10
11    6.7.7   Confounding of Lead Studies: Smoking and Other Factors
12          The most informative studies of lead carcinogenicity are those comparing highly exposed
13    workers to general populations.  In these  comparisons one must consider typical differences
14    between worker populations and the general populations, in particular differences due to
15    smoking and diet.  Smoking can be a major confounder for lung cancer, while diet or SES can
16    be a confounder, albeit weaker, for stomach cancer.
17          Regarding smoking, it has been shown both theoretically and empirically that
18    confounding due to smoking differences between workers and the general population will
19    typically account for an observed relative risk of approximately 1.1 to 1.2, with a possible
20    maximum of about 1.4 (Axelson and Steenland,  1988; Siemiatycki et al., 1988). Furthermore,
21    most occupational cohort studies are retrospective and have little information on smoking,
22    making it impossible to control directly for potential confounding by this strong risk factor.
23    As noted above, the lung cancer relative risk in the meta-analysis of Steenland and Boffetta
24    (2000), after excluding the Lundstrom  et  al.  study, was 1.14 (95% CI: 1.04, 1.73), based on
25    seven occupational cohort studies, six of which used a non-worker external referent population,
26    and none of which controlled for smoking as a confounder. This relatively small excess relative
27    risk could plausibly be due to confounding by smoking. Unfortunately the occupational cohort
28    studies were usually not followed by nested-case control studies of lung cancer which could have
29    controlled for smoking, and furthermore they usually did not involve internal exposure-response
30    analyses, wherein confounding by smoking is usually minimal. An exception was the lung
31    cancer case-control study conducted by Anttila et al. (1995) within a large cohort of Finnish

      December 2005                            6-255        DRAFT-DO NOT QUOTE OR CITE

-------
 1   workers with known blood lead levels.  In this case-control study smoking-adjusted lung cancer
 2   odds ratios were increased among workers with higher estimated cumulative blood lead or higher
 3   peak blood lead exposure compared to workers with the lowest exposure, and the authors noted
 4   that smoking actually appeared to be a "weak negative confounder" for the high peak blood lead
 5   group. Also, in one large population-based case-control study with extensive information on
 6   other cancer risk factors, there remained an elevated odds ratio for lung cancer with substantial
 7   lead exposure after controlling for smoking (Siemiatycki et al., 1991). Hence there is some
 8   evidence that confounding by smoking does not explain the modest excess lung cancer risk seen
 9   in many studies.
10          Diet high in salt or smoked meats, Helicobacter pylori infection, and SES are possible
11   confounders for stomach cancer. Those of highest SES compared to those of lower SES have
12   been shown to have a relative risk of about 3 (Tomatis, 1990). None of the occupational cohort
13   studies, in which again stomach cancer in workers was compared to the general population,
14   controlled for these potential confounders.  However, these potential confounding factors are
15   much less powerful risk factors in respect to stomach cancer than smoking is with respect to lung
16   cancer, and hence are unlikely to account for relative risks higher than perhaps  1.1 or at most 1.2.
17   Given that the occupational cohort studies had a combined relative risk of 1.34  (95% CI: 1.14,
18   1.57) in the meta-analysis of Steenland et al. (2002) and 1.33 (95% CI: 1.18, 1.49) in that of Fu
19   and Boffetta (1995), it seems unlikely that confounding by these factors can fully account for the
20   excess stomach cancer risk observed in the occupational studies.
21
22   6.7.8  Summary of Epidemiologic Evidence for the Genotoxic and
23           Carcinogenic Effects of Lead
24          The availability of studies of cancer in lead-exposed populations was relatively limited at
25   the time of the 1986 Lead AQCD. The number and range of studies has notably expanded since
26   that time, including extended follow-ups of major extant cohorts, new cohort and case-control
27   studies, and analyses addressing not only cancer but genotoxicty. These new human data greatly
28   expand our knowledge of possible lead carcinogenicity. Animals studies are primarily based on
29   dermal exposure to lead acetate.  While the animal studies clearly show a carcinogenic effect,
30   they are of less relevance here because human exposures are usually to inhaled  lead oxides.
     December 2005                           6-256       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Studies of genotoxicity consistently link lead exposed populations with DNA damage and
 2    micronuclei formation, although less consistently with the more established indicator of cancer
 3    risk, chromosomal aberrations.  Epidemiologic studies, particularly those of the high exposed
 4    occupational cohorts, are the most informative for determining whether lead causes cancer,
 5    because in general we assume that any cancer effect will be strongest and most easily detected
 6    when exposure is highest. There are only two general population cohort studies at ambient
 7    levels, and these are of the same population (NHANES II in the late 1970s). These general
 8    population studies at lower exposure levels show internal dose-response trends but suffer at
 9    times from small numbers for site-specific analyses or lack of site-specific analyses altogether
10    and, also, from possible residual confounding by SES and smoking.
11          The strongest evidence in the key occupational studies linking lead exposure to actual
12    human cancers is that for cancers of the lung and those of the stomach.  Of seven large
13    occupational cohort studies available (Ades and Kazantzis, 1988; Anttila et al., 1995; Carta et al.,
14    2005; Gerhardsson et al., 1995; Lundstrom et al., 1997;  Steenland et al., 1992; Wong and Harris,
15    2000), for example, all showed results consistent with an increase in lung cancer risk among
16    lead-exposed workers, and in four of these studies the association was statistically significant.
17    Further, where workers  could be categorized as to their level of lead exposure, the greatest
18    magnitude of association for lung cancer was usually seen for the highest exposure category.
19    However, the modest elevation of lung cancer risk seen in most relevant studies is in the range of
20    possible confounding due to smoking or other occupational exposures, particularly arsenic,
21    which precludes the evidence from these studies being seen as conclusive.  In particular, the one
22    occupational study with the highest lung cancer risk (Lundstrom et al.) has been  subsequently
23    shown to be highly confounded by arsenic, and without this  study, the combined evidence for a
24    lung cancer elevation across studies is considerably reduced (e.g., the estimated relative risk falls
25    from 1.30 to  1.14). A moderate elevation of stomach cancer is also found in most studies of
26    occupationally exposed populations with applicable data on this outcome.  As with lung cancer,
27    it is possible  that other risk factors such as intake of smoked meats or H. pylori infection could
28    have contributed to the observed associations, but the  observed elevation (meta-analysis of 1.33
29    or 1.34) coupled with the known effect of diet makes it unlikely that the elevation in stomach
30    cancer is entirely due to confounding by diet. Data for other sites such as kidney, brain, and
      December 2005                           6-257        DRAFT-DO NOT QUOTE OR CITE

-------
 1   bladder show some indications of an excess, but the results across studies are not consistent and
 2   are based on small numbers.
 3          The epidemiologic data reviewed above from key high lead exposure occupational studies
 4   suggest a relationship between lead exposure and cancers of the lung and the stomach.  These are
 5   supported by two meta-analyses. This is limited by potential confounders such as other
 6   occupational exposures (arsenic, cadmium), smoking, and dietary habits. General population
 7   cohort studies in which low lead exposure was assessed via blood levels and adjusted for
 8   confounders showed trends for a relationship, but were limited by relatively small numbers for
 9   site-specific analysis. A cancer assessment on lead has not been conducted using the U.S. EPA
10   Guidelines for Cancinogen Risk Assessment (U.S. Environmental Protection Agency, 2005).
11   However, the most recent IARC (2005) review concluded that inorganic lead compounds were
12   probable human carcinogens (Group IIA), based on limited evidence in humans and sufficient
13   evidence in animals. This classification is one step down from a classification as "definite"
14   human carcinogen (Group I).
15
16
17   6.8   EFFECTS OF LEAD ON THE  IMMUNE SYSTEM
18   6.8.1   Summary of Key Findings of the Effects of Lead on the Immune
19            System from  the 1986 Lead  AQCD
20          The 1986 Lead AQCD concluded that studies conducted in laboratory animal models
21   provided evidence for immunosuppressive effects of lead; however, evidence for such effects in
22   humans was lacking.  Since then, the epidemiological  study of immunological effects of lead has
23   progressed considerably.  The currently available epidemiologic and clinical observations are
24   consistent with the greater body of evidence derived from studies in experimental animals
25   indicating that lead can suppress cellular and humor immunity and decrease host resistance to
26   infection agents and tumor cells (see Section 5.9). Findings from the epidemiologic studies
27   suggest that lead exposure (as reflected in blood lead concentration) may be associated  with
28   effects on cellular and humoral immunity.  These effects include changes in serum
29   immunoglobulin levels (e.g., elevated serum IgE); perturbation of peripheral lymphocyte
30   phenotype profiles, including decreases in peripheral blood T-cell abundance and changes in
     December 2005                          6-258       DRAFT-DO NOT QUOTE OR CITE

-------
 1    T-cell:B-cell abundance ratios; suppression of lymphocyte activation; and suppression of
 2    neutrophil chemotaxis and phagocytosis.
 3          Available studies of associations between lead exposure and immunological outcomes are
 4    summarized in Annex Tables AX6-8.1 and AX6-8.2. In general, while the studies provide
 5    support for associations between lead exposure and immunological outcomes, the studies have
 6    numerous limitations that complicate the assessment of the strength of the associations and
 7    causation. Furthermore, the health consequences of outcomes that have been associated with
 8    lead exposure are uncertain. All studies have been cross-sectional in design and most included
 9    relatively small cohorts. The studies implemented varying degrees of quantitative analysis of
10    potential covariables and confounders.  In most studies, a detailed analysis of covariables and
11    confounding was lacking, and many of the reports offered no analysis of covariables or
12    confounding.  Covariables that were considered (but not consistently) in multivariate analyses or
13    controlled by stratification included age, sex, race, smoking habits, alcohol consumption, and
14    illness and/or medications that might affect the immune system. Studies that offer the strongest
15    designs are discussed in greater detail below.
16
17    6.8.2  Host Resistance
18          Associations between lead exposure and host resistance have not been rigorously
19    examined in humans.  Two analyses of illness surveys in children (Rabinowitz et al., 1990) and
20    lead workers (Ewers et al., 1982) have been reported, which suggest a possible association
21    between increasing blood lead concentrations (>10 |ig/dL) and illness incidence or prevalence.
22    Both studies relied on personal surveys for assessment of illness and neither study considered
23    covariates or confounders in the analyses.
24
25    6.8.3  Humoral Immunity
26          Studies of biomarkers of humoral immunity in children have consistently found
27    significant associations between increasing blood lead concentration and serum immunoglobulin
28    levels, with increasing serum IgE in association with increasing blood lead concentration
29    (Table 6-8.1; Karmaus et al., 2005; Lutz et al., 1999; Sun et al., 2003). These effects were
30    evident at blood lead concentrations <10 |ig/dL.  Increasing serum IgE levels also have been
31    observed with increasing blood  lead concentration (blood lead >30 |ig/dL) in association with

      December 2005                            6-259       DRAFT-DO NOT QUOTE OR CITE

-------
December 2005





ON
to
ON
O
O
H
6
o
0
H
O
o
H
W
O
O
H
W
Table 6-8.1.
Study
Children
Annesi-Maesano et al. (2003)
Karmaus et al. (2005)
Lutzetal. (1999)
Sarasua et al. (2000)
Sun et al. (2003)
Adults
Heo et al. (2004)
Pinkerton et al. (1998)
Sarasua et al. (2000)
-, decrease; +, increase; o, no effect;
Summary of Results of Selected Studies of Associations Between Lead Exposure
and Serum Immunoglobulin Levels
Blood Lead Oig/dL)
Subjects na Mean (SD) Range J§A ^ ^
neonates 374 67 (48)b NR NR +c NR
children, 7-10 yr 331 3 l-5e o + o
children, 9 mo-6 yr 270 NR 1^5 NR + NR
children, 6-35 mo 372 7 ~2-16d + NR +
children, 3-6 yr 73 NR -3-40 NR +

batter manufacture workers 606 —22 (— 10)e NR o + o
smelter workers 229 39f <2-55 o NR -
general population 433 4.3 ~l-10d o NR o
NR, not reported, Ig, serum immunoglobulin level

IgM
NR
0
NR
+
-

0
0
0

a total number of subjects (including reference group)
b infants cord blood (maternal blood lead mean was 96 ug/dL (SD 58)
0 in association with increasing neonatal hair lead
d 5-95th percentile range
e mean of age-group means and SDs
f median













-------
 1    occupational exposures to lead (Heo et al., 2004). Outcomes for other immunoglobulin indices
 2    in adults have been less consistent (Pinkerton et al., 1998; Sarasua et al., 2000).
 3          Possible associations between lead exposure and biomarkers of humoral immunity in
 4    children have been examined in several cross-sectional studies (Annesi-Maesano et al., 2003;
 5    Karmaus et al., 2005; Lutz et al., 1999; Reigart and Graher, 1976; Sarasua et al., 2000; Sun et al.,
 6    2003; Wagnerova et al.,  1986).  Four studies warrant particular attention because they examined
 7    a relatively low range of blood lead concentrations and applied multivariate analyses to the data
 8    in attempts to control for possible covariables (Karmaus et al., 2005; Lutz et al., 1999; Sarasua
 9    et al., 2000; Sun et al., 2003). Three studies found significant associations between increasing
10    blood lead concentration and serum IgE levels (Karmaus et al., 2005; Lutz et al., 1999; Sun
11    et al., 2003).  The reported percent increase in serum IgE levels measured in these studies ranged
12    from approximately 50 to 400%  The Lutz et al. (1999)  study measured serum IgE and IgG
13    (against Rubella) in 270 children (age range 9 months to 2 years; blood lead range 1-45 jig/dL).
14    The observed blood lead-age-IgE relationship is shown  in Figure 6-8.1.  The highest IgE levels
15    (mean 211 lU/mL, SD 441, n = 17) were observed in children who had blood lead concentrations
16    in the range 15-19 |ig/dL; by comparison, mean IgE levels were blood lead concentrations in the
17    range of 15-19 |ig/dL; by comparison, mean IgE levels  were 52 ITJ/mL (SD 166) for subjects
18    who had blood lead concentrations <10 |ig/dL (n = 174).  The Karmaus et al. (2005) study
19    measured serum IgA, IgE, IgG, and IgM  levels in 331 children (age range 7-10 years).  Blood
20    lead concentrations were lower in this study than in the Lutz et al. (1999) study (1-5 |ig/dL).
21    A multivariate linear regression analysis revealed a significant association between blood lead
22    (p < 0.05) and serum IgE, however, the change in serum IgE level was not monotonic with
23    increasing blood lead concentration (Figure 6-8.2). The highest IgE levels (adjusted mean
24    59 IU/L) were observed in the children who had blood lead concentrations ranging from 2.8-3.4
25    |ig/dL (n = 86) and >3.4 |ig/dL (n = 82).  Sun et al. (2003) measured serum  IgE, IgG, and IgM
26    levels in children, ages 3-6 years (blood lead concentration range 2.6-44 |ig/dL, n = 73).
27    A nonparametric comparison of immunoglobulin levels between low (<10 |ig/dL) and high
28    (> 10 |ig/dL) blood lead strata revealed significantly higher IgE levels (Figure 6-8.3) and
29    significantly lower IgG and IgM levels in the high blood lead stratum.
30          The study by Annesi-Maesano et al. (2003) provides further suggestive evidence for
31    an association between lead exposure and increasing IgE levels.  The study  included 374

      December 2005                          6-261       DRAFT-DO NOT QUOTE OR CITE

-------
                    E \ It'll!!
                    Ap (bt
     Figure 6-8.1.   Relationship between blood lead concentration (PbB), age, and serum IgE
                    level in children.  Spearmen partial correlation between blood lead and
                    serum IgE is 0.22 p = 0.0004, n = 221).
     Source: Lutzetal. (1999).
 1   mother-infant pairs who had relatively high mean blood lead levels (maternal mean 96 |ig/dL,
 2   SD 58; infant cord 67 |ig/dL, SD 48).  Serum IgE level was significantly associated with
 3   increasing infant hair lead (p < 0.001), but not with cord blood lead or placental lead level.  The
 4   association between IgE and hair lead levels was evident in a subset of mother-infant pairs, in
 5   which mothers were classified as nonallergenic, and was unrelated to maternal smoking (i.e.,
 6   urinary cotinine).
 7         The ATSDR Multisite Lead and Cadmium Exposure Study (ATSDR, 1995) is one of the
 8   largest studies to assess humoral immune status in association with lead exposures; however, it
 9   did not include an assessment of IgE. The study included a cross-sectional analysis of serum
10   IgA, IgG, and IgM levels in 1,561 subjects (age range 6 months to 75 years) who resided in areas
     December 2005
6-262
DRAFT-DO NOT QUOTE OR CITE

-------
                  70

                  60

                  50-
                j*
                3
                S. 40 -
                LU
                _ra
                E 30-
                3
                ^
                o>
                W 20-

                  10-
                         <2.2(n=82)      2.2-2.8 (n=81)     2,8-3.4 (n=86)
                                          Blood Lead Category (ug/dL)
                            >3,4 (n=82)
     Figure 6-8.2.   Relationship between blood lead concentration and serum IgE level in
                    children. Mean serum IgE levels (standard deviations not reported) are
                    adjusted for age, number of infections in the previous 12 months, exposure
                    to passive smoke in the previous 12 months, and serum lipids (sum of
                    cholesterol and triglycerides). Means of serum IgE levels in blood lead
                    categories were significantly different (F-test p = 0.03).
     Source: Karmaus et al. (2005).
 1   impacted by lead mining and/or smelting operations and in 480 demographically-matched
 2   controls (Sarasua et al., 2000). A multivariate linear regression analysis of immunoglobulin
 3   levels and blood lead concentration (exposed and control groups combined) revealed
 4   associations between increasing blood lead and increasing serum IgA, IgG, and IgM levels
 5   in subjects 6-35 months of age (blood lead 5th-95th percentile range 1.7-16 |ig/dL,
 6   Figure 6-8.4).
 7          Possible associations between lead exposure and biomarkers of humoral immunity also
 8   have been examined in several cross-sectional studies of lead workers (Alomran and Shleamoon,
 9   1988; Anetor and Adeniyi, 1998; Ayatollahi, 2002; Coscia et al., 1987; Ewers et al., 1982; Heo
10   et al., 2004; Kimber et al., 1986; Pinkerton et al., 1998; Undeger et al., 1996). Outcomes from
     December 2005
6-263
DRAFT-DO NOT QUOTE OR CITE

-------
                    KID
                    400-
                    230 -
                      D-
                    -'•DO
                               .6
1C
  1,4
i.i
    Figure 6-8.3.  Relationship between blood lead concentration (lead) and serum IgE level in
                  lead children. Mean serum IgE levels in female children whose blood lead
                  concentrations were in the range 10-40 ug/dL (20.4 IU/L; n = 16) were
                  significantly higher than for children whose blood lead concentrations
                  <10 ug/dL (13.1 IU/L; n = 17).
    Source: Sun et al. (2003).
1   these studies, with respect to humoral immune parameters, measured as serum and/or salivary
2   immunoglobulin levels, are mixed.  Some studies finding positive associations with blood lead
3   (Heo et al., 2004), negative associations (Anetor and Adeniyi, 1998; Ewers et al., 1982;
4   Pinkerton et al., 1998), or no (or mixed) effects (Alomran and Shleamoon, 1988; Kimber et al.,
5   1986; Queiroz et al., 1994b; Sarasua et al., 2000; Undeger et al., 1996).
6          Based on study design considerations (e.g., cohort criteria, size, treatments of covariates),
7   three studies warrant particular attention (Heo et al., 2004; Pinkerton et al., 1998; Sarasua et al.,
8   2000). Of these, only Heo et al. (2004) assessed serum IgE levels consistent with outcomes
9   reported in children, increasing blood lead concentration was significantly associated with
    December 2005
6-264
DRAFT-DO NOT QUOTE OR CITE

-------
                 1.3
               & 1.2
               E
               =15
     Figure 6-8.4.   Relationship between blood lead concentration and serum immunoglobulin
                    (Ig) levels in children.  Shown are relative changes in serum Ig levels,
                    adjusted for age, sex, and exposure location.  P-values reflect comparison
                    to <5 ug/dL blood lead category mean (<5 mg/dL, n = 165; 5-9.9 ug/dL,
                    n = 136; 10-14.9 ug/dL, n = 47; ;>15 ug/dL, n = 24).
     Source: Sarasua et al. (2000).
 1   increasing serum IgE levels (Figure 6-8.5).  The study measured serum IgE, IL-4 and LFNy in
 2   606 battery manufacture workers.  Serum IgE levels were significantly higher in the blood lead
 3   stratum (>30 |ig/dL) compared to lower strata (<10 or 10-29 |ig/dL) for the age strata 30-39
 4   years, >40 years, and for all ages combined.
 5         Although the Pinkerton et al. (1998) study did not assess IgE outcomes, it offers the
 6   strongest study design of the three for assessment of other immunoglobulin classes.  Although it
 7   is a relatively small  cross-sectional study, it considered immune illnesses and immune
 8   suppressant drugs in the construction of the cohorts and examined a relatively large number of
 9   potential covariates  in the data analysis. Serum immunoglobulin levels were  measured in male
10   smelter (n = 145) workers and hardware workers (n = 84). Excluded (by blind evaluation) from
11   the study cohorts were individuals who had "serious" illnesses of the immune system, who were
12
     December 2005
    6-265
DRAFT-DO NOT QUOTE OR CITE

-------
                    2000
                O)
               HI

                E
                3
                (5
                                                   10-29
                       >=30
                                      Blood Lead Category (fjg/dL)
     Figure 6-8.5.   Relationship between blood lead concentration and serum IgE level in
                    lead workers. Mean serum IgE levels in high blood lead category were
                    significantly higher for all ages (shown), and within age categories
                    >40 years and 30-39 years, but not within age category <30 years.
     Source: Heo et al. (2004).
 1   taking immune suppressant drugs, or who had chemical exposures (other than to lead) that might
 2   affect immune function. Median blood lead concentrations were 39 |ig/dL (range 15-55) in the
 3   lead workers and <2 |ig/dL (range <2-12) in the reference group. Covariate-adjusted (logistic
 4   regression) geometric mean serum IgA, IgG, and IgM, and salivary IgA levels in the lead
 5   workers were not significantly different from the reference group; however, the adjusted
 6   regression coefficient for serum IgG and time-integrated (but not current) blood lead
 7   concentration was negative and significant.
 8         The Sarasua et al. (2000) study, described above for its assessment of children, also
 9   included a cross-sectional analysis of serum IgA, IgG, and IgM levels in adults (age 16-75 years,
10   n = 433; blood lead 5th-9th percentile range 1-10 |ig/dL) and found no significant associations
11   between blood lead and serum immunoglobulin levels (serum IgE outcomes were not assessed).
12         Also germane to the evidence for effects of lead on humoral  immunity in humans are the
13   results of a clinical study in which serum immunoglobulin levels were repeatedly measured in a
14   lead smelter worker who underwent CaEDTA therapy three times per week for a period of
     December 2005
6-266
DRAFT-DO NOT QUOTE OR CITE

-------
 1    10 weeks (Sata et al., 1998). Serum IgA, IgG, and IgM were significantly higher when assessed
 2    24 h after each CaEDTA treatment compared to assessments made prior to treatment.
 3    Furthermore, serum IgG levels were significantly negatively correlated with blood lead
 4    concentration during the treatment period. Before-treatment and after-treatment blood lead
 5    concentration means were 45.1 (SD 16.0) and 31.0 (SD 9.8), respectively.
 6
 7    6.8.4   Cell-mediated Immunity
 8          Studies of biomarkers of cellular immunity in children have found significant associations
 9    between increasing blood lead concentration and decreases in T-cell abundance, with
10    corresponding increases in B-cell abundance (Karmaus et al., 2005; Sarasua et al., 2000; Zhao
11    et al., 2004). These effects have been observed in children whose blood lead concentrations
12    were below 10 |ig/dL (Karmaus et al., 2005; Sarasua et al., 2000), although not all studies (e.g.,
13    Lutz et al., 1999) have found such associations at higher blood lead concentrations (e.g., 10-45
14    jig/dL).  Studies of occupational lead exposures have also found associations between increasing
15    blood lead concentration and changes in T-cell abundance (Fischbein et al., 1993; Pinkerton
16    et al., 1998; Sata et al.,  1997). Effects were observed in association with blood lead
17    concentrations below 25 |ig/dL (Fischbein et al., 1993) and in populations whose blood lead
18    concentrations ranged from approximately 7 to 55 |ig/dL (Pinkerton et al., 1998; Sata et al.,
19    1997). Outcomes from these studies are qualitatively summarized in Table 6-8.2 and are
20    discussed in greater detail below.
21          Several cross-sectional studies have examined possible associations between lead
22    exposure and biomarkers of cellular immunity in children (Karmaus et al., 2005; Lutz et al.,
23    1999; Sarasua et al., 2000; Zhao et  al., 2004). Three studies (Karmaus et al., 2005;  Sarasua
24    et al., 2000; Zhao et al., 2004) found significant associations between increasing lead exposure
25    and decreases in T-cell abundance (Table 6-8.2). The largest study (Sarasua et al., 2000)
26    examined abundance of total lymphocytes, T-cells (CD3+), B-cells (CD20+), NK cells, and CD4+
27    and CD8+ T-cell phenotypes in infants, children, and adolescents. Associations between
28    increasing blood lead concentration and increasing B-cell abundance (% and number), and
29    decreasing T-cell abundance (%) were found for children 6-35 months of age (n = 312), after
30    adjustment for age, sex, and study site (of four mining/smelting sites). Comparison of adjusted
      December 2005                           6-267       DRAFT-DO NOT QUOTE OR CITE

-------
o

O

cr
o
o
ON
to
ON
oo
                          Table 6-8.2. Summary of Results of Selected Studies of Associations Between Lead Exposure
                                                       and Serum Lymphocyte Abundances
Blood Lead (jig/dL)
Study
Children
Karmaus et al. (2005)
Lutzetal. (1999)
Sarasua et al. (2000)
Zhao et al. (2004)
Adults
Fischbein et al. 1993)
Pinkerton et al. (1998)
Sarasua et al. (2000)
Sataetal. (1997)
Subjects

children, 7-10 yr
children, 9 mo-6 yr
children, 6-35 mo
children, 3-6 yr

firearms instructors
smelter workers
general population
lead stearate workers
na

331
270
372
73

87
229
433
99
Mean (SD)

3
NR
7
NR

3 1(4?
39k
4.3
19
pTib
Range

1-51
1^5 o
-2-161
-3-40 o

NR
<2-55 o
-1-101 o
7-50 o
TC rri d rri C
H AC AHC

o - NR
NR NR NR
o o NR
+

o NR
000
000
o + NR
T f
AM

0
NR
NR
NR

NR
+
NR
-
NKg

0
NR
0
NR

0
0
0
NR
Bh

-
0
+
0

+
+
0
0
H
6
o

o
H
O
o
H
W
O
O
r-H
H
W
           -, decrease; +, increase; o, no effect; NR, not reported; SD, standard deviation.

         a total number of subjects (including reference group)
         b T-cells (CD3+)
         c T-helper cells (CD4+)
         d Cytotoxic T-cells (CD8+)
         e CD4+CD8+
         f T-memory cells (CD45RO+, CD45RA+)
         g Natural killer cells (e.g., CD16+, CD56+)
         h B-cells (e.g., CD19+, CD20+)
         1 5-95* percentile range
         J high exposure group
         k median

-------
 1    means for outcomes across blood lead strata revealed that the differences were significant for the
 2    > 15 |ig/dL stratum only, compared to the <5 |ig/dL stratum. The Karmaus et al. (2005) study
 3    examined children in the age range 7-10 years (n = 331) who had blood lead concentrations
 4    <5 |ig/dL.  In addition to age and sex, regression models relating outcomes to blood lead
 5    concentration included exposure to environmental tobacco smoke and infections in the previous
 6    year as covariates. Similar to the Sarasua et al. (2000) study, Karmaus et al. (2005) found
 7    significant associations between blood lead concentration and decreased T-cell abundance
 8    (CD3+, CD3+CD8+) and increased B-cell (CD19+) abundance (for the blood lead quartile
 9    2.2-2.8 |ig/dL; Figure 6-8.6). Zhao et al. (2004) examined lymphocyte phenotype abundance in
10    children in the age range 3-6 years (n = 73) and found significantly lower % abundance of T-cell
11    phenotypes CD3+CD4+, CD4+CD8+ and significantly higher abundance of D3+CD8+ cells in
12    children whose blood lead concentrations were > 10 |ig/dL compared to <10 |ig/dL. Lutz et al.
13    (1999) found no significant associations between blood lead concentration and age-adjusted
14    T-cell (CD3+) or B-cell (CD19+) abundance or abundance of various other lymphocyte
15    phenotypes (i.e., CD2+, CD25+, CD28+, CD71+) in children whose  blood lead concentrations
16    were 10-14, 15-19, or 20-45 |ig/dL compared to <10 |ig/dL.
17          A larger set of studies have evaluated potential associations  between lead exposure and
18    biomarkers of cellular immunity in adults (Basaran and Undeger, 2000; Cohen et al., 1989;
19    Coscia et al., 1987; Fischbein et al., 1993; Kuo et al., 2001; Mishra et al., 2003; Pinkerton et al.,
20    1998; Sarasua et al., 2000; Sata et al., 1998,  1997; Yucesoy et al., 1997b; Undeger et al.,  1996).
21    Four studies warrant particular attention because they implemented relatively stronger study
22    designs (i.e., cohort criteria, size, treatment of covariates): Fischbein et al., 1993; Pinkerton
23    et al., 1998; Sarasua et al., 2000; and Sata et al., 1998).  With one exception (Sarasua et al.,
24    2000), all were studies of relatively small occupational cohorts. The Sarasua et al. (2000) study
25    included a cross-sectional analysis of abundance of total lymphocytes, B-cells, NK cells,  and
26    CD4+ and CD8+ T-cell phenotypes in individuals (n = 433), age 16-75 years. Associations were
27    not found between blood lead concentration and either B-cell or T-cell abundance, after
28    adjustment for age, sex, and study site (of four mining/smelting sites). The study did detect
29    significant associations among these variables in infants and children (see above discussion of
30    cellular immunity outcomes in  children). However, all three occupational studies found
      December 2005                          6-269        DRAFT-DO NOT QUOTE OR CITE

-------
              1.12

              1.08-
            (B
            •Q 1.04-)
            3
            = 1.00-
            0)
            O
            .£ 0.96 H
            O)
            c
            5 O-92 ~
            o
            3.4 (n=82)
    Figure 6-8.6.   Relationship between blood lead concentration and T- and B-cell
                   abundances in children.  Shown are relative changes in covariate-adjusted
                   absolute cell numbers (cells/uL) compared to the lowest blood lead group;
                   adjusted for age, number of infections in the previous 12 months, exposure
                   to passive smoke in the previous 12 months, and serum lipids (sum of
                   cholesterol and triglycerides). Abundances for T-cells, cytotoxic T-cells,
                   and B-cells in the 2.2-2.8 ug/dL group were significantly different (p <, 0.05)
                   from the <2.2 ug/dL group. Receptor phenotypes assayed were: T-cells,
                   CD3+; T-helper cells, CD3+CD4+; cytotoxic T-cells, CD3+CD8+; memory
                   T-helper cells, CD4+CD45RO+; natural killer cells, CD16+CD56+; B-cells,
                   CD3+CD5+CD19+.

    Source: Karmaus et al. (2005).
1    significant associations between increasing blood lead concentration and changes in abundanceof

2    circulating T-cells with either no effect or an increasing B-cell abundance (Fischbein et al., 1993;

3    Pinkerton et al., 1998; Sata et al., 1997). The strengths of the Pinkerton et al. (1998) study have

4    been described previously with respect to outcome measures for humoral immunity. The study

5    included male smelter workers (n = 145, mean blood lead 39 |ig/dL; range 15-55) and hardware

6    workers (n = 84, mean <2 |ig/dL, range <2-12). Covariate-adjusted significant outcomes were

7    an increase in B-cell (CD19+) abundance (% and number) and increases in CD4+CD45RA+ cell
    December 2005
                     6-270
            DRAFT-DO NOT QUOTE OR CITE

-------
 1    abundance (%, number) in association with increasing blood lead concentration. Covariate-
 2    adjusted mean levels of monocytes (%), and T-cells (% D4+CD8+, CD8+CD56+) were lower in
 3    lead workers compared to the reference group.
 4          The Fischbein et al. (1993) study examined a small group of firearms instructors (n = 51)
 5    and age-matched reference subjects (n = 36). Fifteen of the instructors had blood lead
 6    concentration >25 |ig/dL (mean 31.4, SD 4.3), the mean of the remaining 21 subjects was
 7    4.6 |ig/dL (SD 4.6). Mean blood lead concentration of the reference group was reported as
 8    <10 |ig/dL.  Increasing blood lead concentration was significantly associated with decreasing
 9    covariate-adjusted T-cell (CD4+) abundance (Figure 6-8.7). Covariate-adjusted T-cell (CD3+ %
10    and number, CD4+ % and number, CD4+CD8+ number) abundance was significantly lower
11    and B-cell (CD20+ cells % and number) abundance was higher in the instructors than in the
12    reference group.
13          The Sata et al. (1998) study included male lead stearate manufacture workers (n = 71)
14    and a nonexposed reference group (n = 28). Mean blood lead concentration was 19 |ig/dL (range
15    7-50) in the lead workers (blood lead concentration for the reference group was not reported).
16    Categorical covariate-adjusted lead exposure classification (exposed, not exposed) was
17    significantly associated with lower T-cell (CD3+CD45RO+) number. Lead workers, relative to
18    the reference group, had significantly lower covariate-adjusted mean CD3+CD45RO+ number
19    and higher CD8+cells (%).
20          The above observations of decreasing T-cell abundance in association with lead exposure,
21    as assessed from blood lead concentrations, is supported by results of several smaller cross-
22    sectional studies, including Basaran and Undeger (2000), Coscia et al. (1987), and Undeger et al.
23    (1996), as well as a clinical study in which T-cell and NK cell abundance was found to increase
24    after CaEDTA chelation therapy of a lead smelter worker (Sata et al., 1997). Lower serum levels
25    of the cytokines that function in the regulation of cellular immune responses, including IL-1P
26    and IFN-y, in lead workers compared to nonexposed subjects have also been observed (Yiicesoy
27    etal., 1997a).
28
29
     December 2005                           6-271       DRAFT-DO NOT QUOTE OR CITE

-------
              1.8


              1.6
              0.4
- B-cells
• Cytotoxic T-cells
•T-helper cells
-Natural killer cells
-T-cells
                                                                            ***
                     Reference (<10, n=36)    Low Exposure (<25, n=36)  High Exposure (>=25, n=15)
                                    Exposure Category (Blood Lead, ug/dL)

    Figure 6-8.7.   Relationship between lead exposure and T- and B-cell abundances in
                   firearms instructors.  Shown are relative changes in absolute cell numbers
                   compared to the reference group. Comparisons of exposed relative to the
                   reference group are shown as:  * for p < 0.05; ** for p < 0.01; and *** for
                   p < 0.002. Receptor phenotypes assayed were: T-cells, CD3+; T-helper cells,
                   CD4+; cytotoxic T-cells, CD8+; natural killer cells, CD16+; B-cells, CD20+.
                   The CD4+/CD8+ ratio (not shown) was  significantly lower in both the low
                   exposure (1.38 [SD 0.5], p < 0.002) and higher exposure group (0.95
                   [SD 0.5], p < 0.002), compared to the reference group (1.95 [SD 0.66]).

    Source: Fischbein et al. (1993).
1   6.8.5   Lymphocyte Function
2         Limited evidence from occupational studies suggests that lead may suppress lymphocyte
3   activation in humans.  All available studies are of adults.  Several studies have examined
4   associations between lead exposure in adults and lymphocyte activation, assessed as a
5   proliferative response to mitogens and/or antigens (Alomran and Shleamoon, 1988; Cohen et al.,
6   1989; Fischbein et al., 1993; Kimber et al., 1986; Mishra et al., 2003; Pinkerton et al., 1998;
7   Queiroz etal., 1994b). Results of these have been mixed.  Three studies found no significant
8   associations between blood lead concentrations in lead workers and lymphocyte proliferative
    December 2005
                     6-272
             DRAFT-DO NOT QUOTE OR CITE

-------
 1    response to activating agents (; Kimber et al., 1986; Pinkerton et al., 1998; Queiroz et al.,
 2    1994b). Four studies found decreasing proliferative response with increasing blood lead
 3    concentration (Alomran and Shleamoon, 1988; Cohen et al., 1989; Fischbein et al., 1993; Mishra
 4    et al., 2003).  The Alomran and Shleamoon (1988), Cohen et al. (1989), Mishra et al. (2003), and
 5    Queiroz et al. (1994b) studies, which found significant lead associations, included subjects who
 6    had relatively high blood lead levels (>60 |ig/dL) compared to the Kimber et al. (1986) and
 7    Pinkerton et al. (1998) studies.  The inclusion of subjects with higher lead concentrations may
 8    have contributed to the differences in outcomes.
 9          As noted in the previous section, the Fischbein et al. (1993) and Pinkerton et al. (1998)
10    studies are particularly noteworthy because of the strengths of the cohort selection and the data
11    analyses which attempted to account for potential confounders.  Also, these are the only reported
12    studies that examined antigen-specific lymphocyte activation in humans. Mean blood lead
13    concentrations in the two studies were similar 31 |ig/dL (SD 4) in the Fischbein et al. (1993)
14    study and 39 |ig/dL (range 15-55) in the Pinkerton et al. (1998) study.  Both studies found no
15    significant associations between lead exposures (i.e., blood lead concentration) and antigen-
16    specific lymphocyte proliferation, assessed in the Pinkerton et al. (1998) study with tetanus
17    toxoid as the antigen and in the Fischbein et al. (1993) study with staphylococcus aureus as the
18    antigen.  However, the Fischbein et al. (1993) study also measured mitogen-induced lymphocyte
19    proliferation  (induced with PHA or PWM) and found a significantly lower proliferative response
20    to the mitogens in association with lead exposure.  This study also found a significant association
21    between increasing blood lead concentration and decreasing proliferative response in mixed
22    lymphocyte cultures (i.e., proliferative response of lymphocytes from exposed subjects when
23    incubated with inactivated lymphocytes from a reference subject).
24          Inorganic lead has been shown by in vitro studies to perturb several aspects of lymphocyte
25    function when introduced into primary isolates of human blood monocytes.  Activated
26    lymphocytes show altered lysosomal enzyme secretion and altered expression and secretion of
27    cytokines (Bairati et al., 1997; Guo et al.,  1996a; Hemdan et al., 2005).  Lymphocytes activated
28    with Salmonella enteritidis or to monoclonal antibodies of CD3, CD28 and CD40, and exposed
29    to inorganic lead had suppressed expression of T-helper cell type TH-!  cytokines, interferon
30    (LFN-y), interleukin (IL-1P), and tumor necrosis factor (TNF-a), whereas activation by CD
31    antibodies increased secretion of Tn-2  cytokines, IL-5, IL-6, and IL-10 (Hemdan et al. 2005).

      December 2005                           6-273       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Inorganic lead also activates transcription factor NK-icp in CD4+ cells (Pyatt et al., 1996), an
 2    important regulator of T-cell activation, and increases expression of MHC class II surface
 3    antigens (HLA-DR), an important surface antigen in the CD4+ response to exogenous antigens
 4    (Guo et al., 1996b). Lead increases antibody production in cultured human B-cells (McCabe and
 5    Lawrence, 1991). These observations suggest that they may perturb cellular immune function
 6    through a variety of mechanisms.
 7
 8    6.8.6   Phagocyte (Macrophage and Neutrophil) Function
 9          Studies of lymphocyte and phagocyte (i.e., macrophage, neutrophil) function have found
10    associations between blood lead concentrations and suppressed activation of macrophages in
11    children whose blood lead concentrations ranged from 4 to 50 |ig/dL  (Pineda-Zavaleta et al.,
12    2004). In addition, studies have observed suppressed PMNL chemotaxis in association with
13    occupational exposures that resulted in blood lead concentrations of 12-90 |ig/dL (Bergeret
14    et al., 1990; Queiroz et al., 1994a,  1993).
15          Pineda-Zavaleta et al. (2004) examined mitogen (PHA)- and cytokine (INFy)-induced
16    activation of blood monocytes collected from 65  children  (age range 6-11 years) who resided
17    near an active lead smelter. Mean blood lead  concentrations of subjects at three schools were:
18    7.0 |ig/dL (range 3-25  |ig/dL; 8,100 meters from smelter complex), 21 |ig/dL (range
19    11-49 |ig/dL; 1,750 meters from smelter), and 30 (range 10-48 |ig/dL; 650 meters from  smelter).
20    Endpoints measured were nitric oxide and superoxide anion production, a response generally
21    attributed to activated macrophages.  Increasing blood  lead concentration was significantly
22    associated with decreasing PHA-induced nitric oxide production and  increasing INFy-induced
23    superoxide anion production.  The mitogen, PHA, activates macrophages indirectly through
24    activation of lymphocytes, whereas INFy, a cytokine released from CD44 (Tnl) cells, directly
25    activates macrophages. Thus, one interpretation of this outcome is that lead suppressed  T-cell
26    mediated macrophage activation and stimulated cytokine-induced macrophage activation.
27          Possible associations between occupational lead exposure and PMNL chemotaxis and
28    phagocytic activity have been explored in several small cross-sectional studies. Consistent
29    findings are significantly reduced chemotactic response and phagocytic activity (i.e., respiratory
30    burst, luminal uptake) in lead workers compared to reference groups.  The largest study is that of
31    Queiroz et al. (1994a, 1993) which evaluated PMNL function in several (possibly overlapping)

      December 2005                          6-274        DRAFT-DO NOT QUOTE OR CITE

-------
 1    cohorts of lead battery manufacture workers (n = 60).  Blood lead concentrations in the study
 2    groups ranged from 12 to 90 |ig/dL. PMNL chemotaxis and lytic activity were significantly
 3    lower in the lead workers compared to the reference group.  Bergeret et al. (1990) assessed
 4    PMNL chemotaxis and phagocytosis in a group of battery smelting workers (n = 34) and in a
 5    group of reference subjects (n = 34) matched to the lead worker group by age, sex, ethnic origin,
 6    smoking and alcohol consumption habits, and intake of antibiotics and NS AIDs. Mean blood
 7    lead concentrations were 71 |ig/dL (SD 18) in the lead workers and 9 |ig/dL (SD 4) in the
 8    reference group. Significantly lower PMNL chemotactic response to FMLP and phagocytic
 9    response in opsonized zymosan were significantly lower in the lead workers than in the reference
10    group. Lead introduced into primary cultures of human PMNLs suppressed chemotaxis and
11    phagocytosis (Governa et al., 1987).
12
13    6.8.7   Summary of the Epidemiologic Evidence for the Effects of Lead
14           on the Immune System
15          Several studies have examined possible associations between lead exposures and
16    biomarkers of immune function. Findings from the epidemiologic studies suggest that lead
17    exposure (as reflected in blood lead concentration) may be associated with effects on cellular and
18    humoral immunity.  These effects include changes in serum immunoglobulin levels; perturbation
19    of peripheral lymphocyte phenotype profiles, including decreases in peripheral blood T-cell
20    abundance and changes in T-cell :B-cell abundance ratios; suppression of lymphocyte activation;
21    and suppression of neutrophil chemotaxis and phagocytosis.
22          Studies of biomarkers of humoral immunity in children have consistently found
23    significant associations between increasing blood lead concentration and serum Ig levels with
24    increasing serum IgE in association with increasing blood lead concentration (Karmaus et al.,
25    2005; Lutz et al., 1999; Sun et al., 2003).  These effects were evident at blood lead
26    concentrations below 10 |ig/dL. Findings of studies of adults have been mixed with significant
27    associations between blood lead (>30 |ig/dL) and serum immunoglobulin levels  (Heo et al.,
28    2004; Pinkerton et al., 1998) and no association in a study group in which blood lead
29    concentrations were <10 |ig/dL (Sarasua et al., 2000).
30          Studies of biomarkers of cellular immunity in children have found significant associations
31    between increasing blood lead concentration and decreases in T-cell abundance, with

      December 2005                           6-275       DRAFT-DO NOT QUOTE OR CITE

-------
 1    corresponding increases in B-cell abundance (Karmaus et al., 2005; Sarasua et al., 2000; Zhao
 2    et al., 2004).  These effects have been observed in children whose blood lead concentrations
 3    were below 10 |ig/dL (Karmaus et al., 2005; Sarasua et al., 2000), although not all studies have
 4    found such associations at higher blood lead concentrations (e.g., 10-45 |ig/dL; Lutz et al.,
 5    1999). Studies of occupational lead exposures have also found associations between increasing
 6    blood lead concentration and decreasing T-cell abundance (Pinkerton et al., 1998; Sata et al.,
 7    1997; Fischbein et al., 1993).  Effects were observed in association with blood lead
 8    concentrations below 25 |ig/dL (Fischbein et al., 1993) and in populations whose blood lead
 9    concentrations ranged from approximately 7 to 55 |ig/dL (Pinkerton et al., 1998;  Sata et al.,
10    1997).
11          Studies of lymphocyte and phagocyte (i.e., macrophage, neutrophil) function have found
12    associations between blood lead concentrations and suppressed activation of macrophages in
13    children whose blood lead concentrations ranged from 4 to 50 |ig/dL (Pineda-Zavaleta et al.,
14    2004); suppressed PMNL chemotaxis in association with occupational exposures that resulted in
15    blood lead concentrations of 12 to 90 |ig/dL (Bergeret et al.,  1990; Queiroz et al., 1994a, 1993),
16    and suppressed mitogen-induced  activation of peripheral lymphocytes in adults in association
17    with occupational exposures that  resulted in blood lead concentrations that ranged from 15 to
18    55 |ig/dL (Fischbein et al., 1993).
19
20
21    6.9   EFFECTS OF LEAD ON OTHER ORGAN SYSTEMS
22    6.9.1   Biochemical Effects of Lead
23    6.9.1.1    Summary of Key Findings of the Biochemical Effects of Lead from the
24             1986 Lead AQCD
25          The 1986 Lead AQCD provided an extensive discussion of the effects of lead on heme
26    biosynthesis and on quantitative relationships between exposure and effects in humans. Lead
27    interferes with heme synthesis by inhibiting the enzymes 5-aminolevulinic acid dehydratase
28    (ALAD) and ferrochelatase.  As a consequence, heme biosynthesis decreases, relieving the rate-
29    limiting enzyme of the heme synthesis pathway, 5-aminolevulinic synthetase (ALAS), from
30    negative feedback inhibition by heme (Figure 6-9.1).  The outcomes of decreased activity of
31    ALAD and ferrochelatase, and increased activity of ALAS are increased urinary excretion of

      December 2005                            6-276       DRAFT-DO NOT QUOTE OR CITE

-------
                                       MITOCHONDRION
                       GLYCINE
                    SUCCINYL^CoA
                                   MITOCHONDR1AL MEMBRANE
                                                     FERRO-
                                                   CHELATASE
                             ALA SYNTHETASE
                                (INCREASE)
                             .  Pb (DIRECTLY OR
                              BY DEREPRESSION)
                                                   IRON -i- PROTOPORPHYRIN
                 AMINOLEVULINIC ACID
                        (ALA)
                 ALA
              DEHYDRASE
              (DECREASE)
                    PORPHOBILINOGEN
       CORPROPORPHYRIN
           (INCREASE)
     Figure 6-9.1.   Effects of lead on heme biosynthesis.
     Source: Derived from EPA (1986).


 1   coproporphyrin (CP) and 5-aminolevulinic acid (ALA), increased level of ALA in blood plasma,
 2   and increased erythrocyte protoporphyrin (EP) levels.
 3          Associations between lead exposure and blood ALAD activity and EP levels, and urinary
 4   ALA and CP excretion have been studied extensively in adults and children, and quantitative
 5   relationships between exposure and effect are well understood. Much of this information was
 6   available prior to completion of the 1986 Lead AQCD and is summarized in that criteria
 7   document (e.g., Alessio et al., 1976; Hernberg et al.,  1970; Lilis et al., 1978; Piomelli et al.,
 8   1982; Roels et al.,  1979; Selander and Cramer, 1970; Valentine et al., 1982).  Numerous studies
 9   published since the 1986 AQCD  provide additional support for the  lead concentration-response
10   relationships in humans described in the 1986 AQCD. The most pertinent studies are
11   summarized in Annex Tables AX6-9.1 and AX6-9.2. The studies that provide the strongest basis
12   for empirically-derived expressions relating blood lead concentration, blood ALAD activity,
13   urinary ALA, and EP are listed in Table 6-9.1 and are discussed below.
14
     December 2005
6-277
DRAFT-DO NOT QUOTE OR CITE

-------
December 2005




Oi
to
oo

O
[>
H
6
O
0
H
O
O
H
W
O
O
H
W
Table 6-9.1. Blood Lead-Response Relationships for Heme Synthesis Biomarkers in Adults and
Study n
ALAD Activity Decrease
Roels and Lauwerys (1987) 143
Alessio et al. (1976, 1977) 169
Hernberg et al. (1970) 158
Morita et al. (1997) 58
Urinary ALA Increase
Roels and Lauwerys (1987) 37
Alessio et al. (1976, 1977) 3 16
Gennart etal. (1992) 183
Oishietal. (1996) 418
Selander and Cramer ( 1 970) 150
Roels and Lauwerys ( 1 987) 39
Roels and Lauwerys (1987) 36




Age

10-13 yr
Adult (m)
Adult (m, f)
Adult (m)

10-13 yr
Adult (m)
Adult (m, f)
Adult (m, f)
Adult (m, f)
Adult (m)
Adult (f)




Children
Blood Lead Change
Oig/dL)
Blood Lead Predicted to Halve or
(jig/dL) Regression Equation (r) Double Effect Biomarker

5-41 log[ALAD]= 1.864-0.015[blood lead] (r = 0.87)
15-150 log[ALAD]=3. 73-0.031 [blood lead] (r= 0.87)
5-95 log[ALAD]=2.274-0.018[blood lead] (r = 0.90)
2-82 log[ALAD]=1.8535-0.00971[blood lead] (r = 0.76)

20-41 log[ALAU]=0.94+0. 11 [blood lead] (r = 0.54)
10-150 log[ALAU]=1.25+0.014[blood lead] (r = 0.62)
4-75 log[ALAU]=0.37+0.008[blood lead] (r = 0.64)
10-99 log[ALAU]= -0.387+0.022[blood lead] (r = 0.71)
6-90 log[ALAU]= -1.0985+0. 157[blood lead] (r= 0.74)
10-60 log[ALAU]=0.37+0.006[blood lead] (r = 0.41)
7-53 log[ALAU]=0.15+0.015[bloodlead] (r= 0.72)





20.1
22.4
16.1
20.1

20.9
49.5
37.6
13.7
19.2
50.2
20.1





-------
December 2005



Oi
to
VO


O
H
6
O
0
H
O
O
H
W
O
O
H
W
Table 6-9.1 (cont'd). Blood Lead-Response Relationships for Heme Synthesis Biomarkers in Adults and Children
Study
EP Increase
Piomellietal. (1982)
Roels and Lauwerys (1987)
Soldin et al. (2003)
Alessioetal. (1976, 1977)
Alessioetal. (1976, 1977)
Gennart etal. (1992)
Roels and Lauwerys (1987)
Roels and Lauwerys (1987)
Wildtetal. (1987)
Wildtetal. (1987)
Blood Lead
n Age (jig/dL)

2,002 2-12 2-98
51 10-13 15-41
4,908 0-17 <1-103
95 Adult (m) 10-90
93 Adult (f) 10-70
183 Adult (m) 4-75
39 Adult (m) 10-60
36 Adult (f) 7-53
851 Adult (m) 10-80
139 Adult (f) 10-80
ALA, 5-aminolevulinic acid; ALAD, 6-aminolevulinic acid dehydratase;
PbB, blood lead concentration






Regression Equation (r)

log[EP]=1.099+0.016[blood lead] (r = 0.509)
log[EP]=1.321+0.025[blood lead] (r = 0.73)
EP = -0.0015[PbB]3+0.1854[blood lead]2-
2.7554[PbB]+30.911 (r = 0.999)
log[EP]=0.94+0.0117[blood lead]
log[EP]=1.60+0.0143[blood lead]
log[EP]=0.06+0.019[blood lead] (r = 0.87)
log[EP]=1.41+0.014[bloodlead] (r= 0.74)
log[EP]=1.23+0.027[blood lead] (r = 0.81)
log[EP]=1.21+0.0148[blood lead] (r= 0.72)
log[EP]=1.48+0.0113[blood lead] (r= 0.56)
ALAU, urinary 5-aminolevulinic acid; EP, erythrocyte



Blood Lead Change
Oig/dL)
Predicted to Halve or
Double Effect Biomarker

18.8
12.0
20.6
25.7
21.1
15.8
21.1
11.1
20.3
20.6
protoporphyrin;




-------
 1          Since completion of the 1986 Lead AQCD, a literature has developed on the effects of
 2    lead on serum and blood lipids, including cholesterol levels and indications of oxidative stress, in
 3    the form of lipid peroxides, depletion of erythrocyte reduced glutathione (GSH), and production
 4    of reactive oxygen species (ROS). These studies also are summarized in Annex Tables AX6-9.1
 5    and AX6-9.2,  and key findings are discussed below.
 6
 7    6.9.1.2   Heme Biosynthesis
 8    6.9.1.2.1   ALAD Inhibition
 9          Numerous studies published since completion of the 1986 AQCD have explored
10    associations between lead exposure and inhibition of ALAD activity, as assessed from
11    measurements of blood ALAD activity (Gurer-Orhan et al., 2004; Kim et al.,  2002; Lee et al.,
12    2000; Makino et al., 1997; Roels and Lauwerys, 1987; Schuhmacher et al., 1997), or urinary
13    ALA excretion (Gennart et al., 1992; Oishi et al.,  1996; Schuhmacher et al., 1997; Wildt et al.,
14    1987; Soldin et al., 2003).  Quantitative estimates derived from the larger, more recent studies
15    are presented in Table 6-9.1. Blood lead concentration is inversely correlated with the log of
16    blood ALAD activity and log of urinary ALA and quantitative estimates of the change in blood
17    ALAD activity per unit change in blood lead concentration are consistent across studies.
18    Halving of blood ALAD activity occurs with an increase in blood lead concentration of
19    approximately 20 |ig/dL in both children (Roels and Lauwerys, 1987) and adults (Morita et al.,
20    1997). These  estimates are consistent with earlier studies of adults (e.g., Hernberg et al., 1970)
21    and children (e.g., Alessio et al.,  1976, 1977), discussed in the 1986 AQCD. Greater variability
22    is apparent in  estimates of the change in urinary ALA per unit change in blood lead
23    concentration  (Table 6-9.1).  This may be related, in part, to gender-heterogeneity in the
24    relationship. Roels and Lauwerys (1987) estimated that urinary ALA doubles in association with
25    a 20 |ig/dL increase in blood lead concentration in females and 50 |ig/dL in males. In a much
26    larger study (Oishi et al., 1996), an analysis that combined data from males (n = 253) and
27    females (n = 165) found that a doubling of urinary ALA occurred in association with a 14 |ig/dL
28    increase blood lead concentration. Urinary ALA excretion increases as a linear function of
29    plasma ALA concentration (Oishi et al., 1996); thus,  the gender heterogeneity for the blood lead-
30    urinary ALA relationship may derive from a gender difference in the effect of lead on plasma
31    ALA concentration or from differences in renal plasma clearance of ALA.

      December 2005                          6-280       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.9.1.2.2     ALAD Polymorphism
 2          ALAD is a polymorphic enzyme with two alleles (ALAD1 and ALAD2) and three
 3    genotypes: ALAD1,1, ALAD1,2, and ALAD2,2 (Battistuzzi etal., 1981).  The corresponding
 4    phenotypes appear to have nearly identical catalytic properties (Battistuzzi et al., 1981).  The
 5    predominant genotype is ALAD1,1 which has a prevalence of approximately 90% (Astrin et al.,
 6    1987; Battistuzzi et al., 1981; Hsieh et al., 2000; Shen et al., 2001). A higher percentage of
 7    erythrocyte lead was bound to ALAD in carriers of the ALAD2 allele (84%) compared to
 8    carriers of the ALAD1 allele (81%); however, no differences were evident in the distribution of
 9    lead between erythrocytes and plasma (Bergdahl et al., 1997), and there is no evidence that the
10    ALAD genotype confers different sensitivity to inhibition of heme biosynthesis (Hsieh et al.,
11    2000; Perez-Brava et al., 2004; Schwartz et al., 1997;  Suzen et al., 2003).
12
13    6.9.1.2.3     Ferrochelatase Inhibition
14          Lead inhibition of ferrochelatase results in an accumulation of protoporphyrin IX in
15    erythrocytes  (EP, also referred to as zinc protoporphyrin, or ZPP, or iron protoporphyrin, FEP,
16    depending on the method used to make the measurement).  Numerous studies have examined
17    relationships between blood lead concentration and EP levels in adults and children.
18    Quantitative  estimates based on the most pertinent studies are presented in Table 6-9.1. Results
19    across these studies are similar. In both children and adults (males and females), a doubling of
20    EP levels occurs in association with an increase in  blood lead concentration of approximately
21    20 |ig/dL (Piomelli et al., 1982; Soldin et al., 2003; Wildt et al., 1987).  A pronounced gender
22    difference in the relationship between EP and blood lead concentration was observed by Roels
23    and Lauwerys (1987) which was  not observed in the much larger study of Wildt et al. (1987).
24          Inhibition of ferrochelatase also gives rise to an increase in urinary coproporphyrin, with a
25    similar relationship to blood lead concentration; a doubling of urinary EP occurs in association
26    with an increase in urinary coproporphyrin of approximately 20 |ig/dL (Alessio et al., 1976).
27
28    6.9.1.3   Effects on Blood Lipids
29          Associations between occupational exposure to lead and changes in blood lipid
30    composition  have been observed.  These include increased levels of lipid peroxides in blood
31    and/or serum (Ito et al.,  1985; Jiun and Hsien,  1994; Sugawara et al., 1991) and increased serum

      December 2005                           6-281       DRAFT-DO NOT  QUOTE OR  CITE

-------
 1    levels of total and HDL cholesterol (Kristal-Boneh et al., 1999). Increased levels of glucose-6-
 2    phosphate dehydrogenase (G6PD) in erythrocytes have also been observed in lead workers
 3    (Cocco et al., 1995; Gurer-Orhan et al., 2004).
 4          Kristal-Boneh et al. (1999) measured serum total, HDL, and LDL cholesterol, and
 5    triglycerides in a group of male battery manufacture workers. Covariate-adjusted serum total-
 6    cholesterol and HDL cholesterol levels were 6% and 12% higher, respectively, in lead workers
 7    (n = 56, mean blood lead 42 |ig/dL, SD 15) compared to reference group (mean blood lead:
 8    2.7 |ig/dL).  Increasing blood lead concentration was significantly associated with increasing
 9    covariate-adjusted total cholesterol and HDL cholesterol. A similar outcome was found in a
10    larger study (Ito et al.,  1985) of male steel workers (n = 712, blood lead range 5-62 |ig/dL).
11    When stratified by age, total and HDL cholesterol levels in serum were 3.6% and 7.5% higher,
12    respectively, in lead workers in the age range 40 to 49 years, compared to corresponding strata of
13    the office workers (n = 155). Although a smaller study, the Kristal-Boneh et al. (1999) study
14    considered a larger set of potential  covariables (e.g., dietary fat, cholesterol, and calcium intakes,
15    sport activities, alcohol consumption, cigarette smoking).
16          Oxidative changes in blood lipids (e.g., increased levels of lipid peroxides and
17    malondialdehyde levels)  as well as decreased levels of erythrocyte superoxide dismutase (SOD),
18    catalase, G6PD, and GSH peroxidase,  indicative of increased oxidative stress, have been
19    observed in lead workers, in comparison to reference groups (Ito et al., 1985; Jiun and Hsien,
20    1994; Solliway et al.,  1996;  Sugawara et al., 1991).  However, none of these studies have
21    developed concentration-response relationships that take into account potential confounders.
22    The largest study is that of (Ito et al., 1985), described above. When stratified by age, serum
23    HDL cholesterol and serum  lipoperoxide levels were 16% higher in the lead workers in the age
24    range 40 to 49 years, compared to corresponding strata of the reference group. Serum
25    lipoperoxide levels also appeared to increase as blood lead increased above 30 |ig/dL, while
26    erythrocyte  SOD appeared to decrease with increasing blood lead concentration (a statistical
27    evaluation was not reported).
28          Evidence for increased oxidative stress (increased reactive oxygen species) in
29    lymphocytes of lead workers has also been reported (Fracasso et al., 2002).  Peripheral
30    lymphocytes collected from battery manufacture workers (n = 37, mean blood lead:  40 |ig/dL)
31    exhibited increased DNA strand breaks, higher production of ROS and lower GSH levels

      December 2005                            6-282       DRAFT-DO NOT QUOTE OR CITE

-------
 1    compared to a reference group of office workers (n = 29, mean blood lead 4 |ig/dL). The
 2    covariate-adjusted odds ratios (exposed versus not exposed) were 1.069 (95% CI:  1.020, 1.120)
 3    for increased DNA strand breaks and 0.634 (95% CI:  0.488, 0.824) for lower GSH levels.
 4
 5    6.9.2     Effects of Lead on the Hematopoietic System
 6    6.9.2.1    Summary of Key Findings of the Effects of Lead on the Hematopoietic System
 7             from the 1986 Lead AQCD
 8         The 1986 Lead AQCD concluded that lead decreases heme production and shortens
 9    erythrocyte survival; both effects contributing to lead-induced anemia in children and adults,
10    which becomes evident in children at blood lead concentrations >40  |ig/dL and, in adults,
11    >50 |ig/dL.  The 1986 Lead AQCD also concluded that effects of lead on blood hemoglobin
12    level extend below 50 |ig/dL, with effects detected in lead workers at blood lead concentrations
13    <25 |ig/dL (Baker et al., 1979; Grandjean,  1979). More recent epidemiologic studies,
14    summarized below, provide additional information on concentration-response relationships for
15    hematopoietic effects of lead. The studies  support the conclusion that clinical anemia can occur
16    in children in association with blood lead concentrations >40 |ig/dL (Schwartz et al., 1990).  The
17    newer studies suggest that perturbation of erythropoiesis, indicated by changes in serum
18    erythropoietin, occurs in association with blood lead concentrations <40 |ig/dL and in the
19    absence of detectable changes in blood hemoglobin levels or hematocrit.  Details regarding the
20    design of these studies and outcomes are presented in Annex Tables AX6-9.3 and AX6-9.4.
21    Outcomes of the most pertinent studies are discussed below.
22
23    6.9.2.2    Blood Hemoglobin Levels
24         Several studies reported since the completion of the 1986 Lead AQCD have explored
25    associations between lead exposure and blood hemoglobin levels in children and adults.
26    Consistent findings have been a lack of discernable depression of blood hemoglobin levels in
27    study populations whose mean blood lead concentrations were <40 |ig/dL (Table 6-9.2).  Of note
28    is the findings relating patella bone lead to both blood hemoglobin levels and hematocrit.
29          The Kosovo prospective study of pregnancy outcomes is one  of the largest epidemiologic
30    evaluations of associations between lead exposure and blood hemoglobin levels in infants and
31    children (Graziano et al., 2004; Factor-Litvak et al., 1999, 1998). The study included pregnant

      December 2005                           6-283       DRAFT-DO NOT QUOTE OR CITE

-------
Decemt
w
^
to
o
o





ON
i
oo

o
i
H
I
0
o
0
H
O
o
H
W
O

O
HH
H
W
Table 6-9.2. Summary of Results of Selected Studies of Associations Between Lead Exposure and
Blood Hemoglobin Levels


Study Subjects
Children
Graziano et al. (2004) ages: 4.5-12 yr
Liebeltetal. (1999) ages: 1-6 yr
Adults
Graziano et al. (1990) pregnant women
Hu et al. (1994) male carpenters
Makino et al. (1997) male VCS workers
Solliway et al. (1996) male battery workers
Gennart et al. (1992) battery workers


Horiguchi et al. (1991) male lead refinery workers
Poulos et al. (1986) male lead workers


Blood Lead Oig/dL)

™a Mean (SD) Range

311 6-9,31-39b 3-70
86 18C 2-84

1,502 5, 17d 2-43
119 8 2-25
1,573 13 1-39
100 10 23-63
183 51 (8) 40-70


40 54 (16) NR
160 18-27 (5)e NR


-, decrease; +, increase; Hgb, hemoblogin; NR, not reported; PCV, packed cell volume SD, standard deviation; VCS,

a total number of subjects (including reference group)
b range of means of low and higher exposure groups
0 median
d mean of low- and high-exposure groups
e range of group means (standard deviation estimated for up ranj






>e based on reported standard error).


Blood
Hemoglobin Comment

o + erythropoietin
o - erythropoietin

o - erythropoietin
o - in association with patella
bone lead
+ (+) 1 g/dL per 10 ug/dL blood
lead
o RBC count
- hematocnt


- hematocnt
- hematocnt


vinyl chloride stabilizer








-------
 1    women (n = 1502) and their children (n = 311) who resided in one of two regions of Kosovo,
 2    Yugoslavia; one was heavily impacted by lead industries (high-lead area), the other had
 3    relatively little lead contamination (low-lead area).  Mean blood lead concentrations of children
 4    (measured at birth and at intervals to 12 years  of age) ranged from 30 to 40 |ig/dL in the high-
 5    lead area and 6 to 9 |ig/dL in the low-lead area. Mean blood hemoglobin levels in the low-lead
 6    and high-lead children, measured at 4.5, 6.5, 9.5, and 12 years of age, were not significantly
 7    different. These findings are consistent with those from a smaller cross-sectional study (n = 89;
 8    blood lead range 2 to 84 |ig/dL, 84% <35 |ig/dL) that also found no association between blood
 9    lead concentration and blood hemoglobin levels (Liebelt et al., 1999). Results from these two
10    studies suggest that, in the absence of iron deficiency, lead exposures that result in blood lead
11    concentrations <40  |ig/dL do not produce detectable changes in blood hemoglobin levels
12    in children.
13          Associations between lead exposure and blood hemoglobin levels in adults have been
14    examined in numerous epidemiological studies (Froom et al., 1999; Gennart et al.,  1992;
15    Horiguchi et al., 1991; Hu et al., 1994; Makino et al., 1997; Poulos et al., 1986; Romeo et al.,
16    1996; Solliway et al.,  1996). The Graziano et  al. (1990) and Makino et al. (1997) studies warrant
17    particular attention because of the design (longitudinal), relatively large size (>1000 subjects),
18    and relatively low blood lead levels of the subjects (<40 |ig/dL). Both studies support the
19    general conclusion that blood hemoglobin levels are not depressed in association with blood lead
20    concentrations <40  g/dL. In the Kosovo prospective study, no discernable effect of lead on
21    maternal blood hemoglobin levels was evident from a comparison of the high-lead exposure
22    group (mean blood lead 17 |ig/dL, range 7-43  |ig/dL) with the low lead exposure group (mean
23    blood lead 5.1 |ig/dL, range 2-11 jig/dL). Makino et al. (1997) found a positive  association
24    between increasing blood lead concentration and increasing blood hemoglobin levels in a
25    longitudinal survey of adult males (n = 1,573)  who worked in pigment or vinyl chloride
26    stabilizer manufacture (mean blood lead 13 |ig/dL, range 1-39 |ig/dL).  A simple linear
27    regression model predicted a 1 g/dL increase in blood hemoglobin per 10 |ig/dL increase in
28    blood lead concentration (typical level 10-20 g/dL).
29          Two other cross-sectional studies are also notable, because of design considerations
30    and/or blood lead concentration ranges of the subjects.  Solliway et al. (1996) observed no
31    differences in mean blood hemoglobin levels in a comparison of adult male battery manufacture

      December 2005                           6-285       DRAFT-DO NOT QUOTE OR CITE

-------
 1    workers (n = 34, mean blood lead 41 |ig/dL, range 23-63 |ig/dL) and a matched reference group
 2    (n = 56, mean blood lead 7 |ig/dL, range 1-13 |ig/dL). Hu et al. (1994) conducted a cross-
 3    sectional assessment of adult male carpentry workers (n = 119) whose blood lead concentrations
 4    were <25 |ig/dL.  Blood hemoglobin was not significantly associated with blood lead
 5    concentration.  Of note, however, was the finding that increasing patella bone lead was
 6    significantly associated with decreasing blood hemoglobin levels.  Covariate-adjusted blood
 7    hemoglobin levels were predicted to decrease by 1.1 g/dL per 37 jig/g increase (mean of first and
 8    fourth quartiles) in patella bone lead.
 9          Studies of lead workers whose blood lead levels were higher than in the studies noted
10    above have, in general, found lower blood hemoglobin levels in association with increasing
11    blood lead concentrations; these include Gennart et al. (1992) with a blood lead range of 40-70
12    ng/dL, Horiguchi et al. (1991) with a mean blood lead level of 54 |ig/dL (SD 16), and Poulos
13    et al. (1986) with mean blood lead range of 21-27 |ig/dL. In the latter study (Poulos et al.,
14    1986), blood hemoglobin levels decreased by 0.6-0.9 g/dL per 10 |ig/dL increase in blood lead
15    (simple linear regression) in adult males. Analyses or adjustments for potential covariables were
16    not reported for these studies.
17
18    6.9.2.3   Erythrocyte Volume and Number
19          Schwartz et al. (1990) conducted a concentration-response analysis of data collected at the
20    Bunker Hill smelter site in Idaho in 1974, shortly after the failure of the smelter bag house
21    resulted in extensive contamination of the surrounding area with uncontrolled smelter emissions.
22    This analysis is unique in that it collected hematocrit measurements in children (n = 579, age
23    range 1-5 years) who had relatively high blood lead levels (range 11-164  |ig/dL, approximately
24    40% exceeded 40 |ig/dL). A logistic model relating blood lead concentration and age to
25    hematocrit predicted a 10% decrease in hematocrit (from 39.5 to 35.5%) in association with
26    blood lead concentrations of 85, 115, and 145 |ig/dL at ages 1, 3, and 5 years, respectively
27    (Figure 6-9.2). A 10% probability of anemia (hematocrit <35%) was predicted in association
28    with a blood lead concentration of approximately 20 |ig/dL at age 1 year, 50 |ig/dL at age
29    3 years, and 75 |ig/dL at age 5 years (Figure 6-9.2).
30
      December 2005                           6-286       DRAFT-DO NOT QUOTE OR CITE

-------
           40
           38 -
         u
         s
         ra
         I 34
           32
           30
                               Age 3 yr
                                               Age 5 yr
                     20       40       60       80
                                   Blood Lead (ug/dL)
              100
           120
140
                    00		_	
                       0        50        IOO      ISO       200


Figure 6-9.2.   Relationship between blood lead and hematocrit in children. The top panel
              shows central tendency predictions based on a logistic regression model
              relating hematocrit and blood lead concentration, adjusted for age. The
              regression coefficients relating hematocrit and blood lead were (P = 0.0133
              [SE 0.0041], p = 0.0005). The bottom panel shows corresponding
              concentration-response (hematocrit <35%) relationships.

Source:  Schwartz etal. (1990)
December 2005
6-287
DRAFT-DO NOT QUOTE OR CITE

-------
 1          Numerous studies of associations between lead exposure and erythrocyte volume (e.g.,
 2    hematocrit) or number have been reported in adults (Gennart et al., 1992; Horiguchi et al., 1991;
 3    Hsiao et al., 2001; Hu et al., 1994; Makino et al., 1997; Osterode et al., 1999; Poulos et al., 1986;
 4    Solliway et al., 1996). The Hu et al. (1994) and Makino et al. (1997) studies examined groups of
 5    workers that had blood lead concentrations that were relatively low, compared to other studies,
 6    and found either no association or weak association between blood lead concentration and
 7    hematocrit and/or erythrocyte number.  The Hu et al. (1994) cross-sectional study of carpentry
 8    workers (n = 119, blood lead concentration range 2-25 |ig/dL) found no association between
 9    blood lead concentration and hematocrit; however, increasing patella bone lead was associated
10    with a significant decrease in hematocrit.  Covariate-adjusted blood hematocrit was predicted to
11    decrease by 0.03% (95% CI: 0.01, 0.05) per 37 |ig/g increase (mean of first and fourth quartiles)
12    in patella bone lead.  The Makino et al. (1997) longitudinal  study of pigment and vinyl  chloride
13    stabilizer manufacture workers (n = 1,573; blood lead concentration range 1-39 |ig/dL) found a
14    positive association between blood lead concentration and hematocrit, and erythrocyte count.
15    A simple linear regression model predicted an increase in hematocrit of 0.6 (typically 43) and an
16    increase in erythrocyte count of 0.07 x  106/mm3 (typically 4-7 x 106/mm3) per 10 |ig/dL increase
17    in blood lead concentration.
18          Studies that included subjects who had higher blood  lead concentrations (i.e., >40 |ig/dL)
19    have, in general, found negative associations between blood lead concentration and hematocrit
20    Gennart et al., 1992; Horiguchi et al., 1991; Poulos et al., 1986; Solliway et al., 1996), with two
21    exceptions, Hsiao et al. (2001) and Osterode et al. (1999). Hsiao et al. (2001) conducted an
22    11-year retrospective longitudinal analysis of blood lead concentration, hematocrit, and
23    erythrocyte count in a group of battery manufacture workers (n = 30; mean blood lead
24    concentration 30-60 |ig/dL). A repeated measures regression analysis (generalized estimation
25    equation) yielded a significant association between increasing blood lead concentration and
26    increasing hematocrit and erythrocyte count. Osterode et al. (1999) measured erythrocyte
27    number and packed cell volume in a group of lead workers (n = 20) and an age-matched
28    reference group  (n = 20). Mean blood lead concentration was 45.5 |ig/dL (range 16-91 |ig/dL)
29    in the lead workers and 4.1  |ig/dL (range 3-14 |ig/dL) in the reference group. Mean erythrocyte
30    number and packed cell volume in the lead workers and reference group were not different.
31

      December 2005                            6-288        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.9.2.4   Erythropoiesis
 2          Several studies have found associations between lead exposure and serum erythropoietin
 3    levels in children (Graziano et al., 2004; Liebelt et al.,  1999) and adults (Graziano et al., 2001;
 4    Osterode et al., 1999; Romeo et al. 1996).  A qualitative summary of outcomes from these
 5    studies are provided in (Table 6-9.3).
 6          Two studies have examined possible association between lead exposure and serum
 7    erythropoietin levels in children. In the Kosovo prospective study (Factor-Litvak et al., 1999,
 8    1998; Graziano et al., 2004) a significant association was evident between increasing blood lead
 9    concentration (3-70 ng/dL) and increasing serum erythropoietin levels after adjustment for age
10    and blood hemoglobin levels  (Figure 6-9.3). The association weakened with age; it was
11    significant at ages 4.5 and 6.5 years, but not at ages 9.5 or 12 years.  A multivariate linear
12    regression model predicted a  36% increase in serum erythropoietin per 10 |ig/dL increase
13    (3-13  ng/dL, hemoglobin 13 g/dL) in blood lead at age 4.5 years, and an 18% increase per
14    10 ng/dL at age 6.5 years.  These outcomes suggest that erythropoiesis is stimulated in children
15    in association with increasing blood lead concentrations below 40 |ig/dL and in the absence of
16    depressed blood hemoglobin levels.
17          A smaller cross-sectional  study examined serum erythropoietin levels in a group of
18    children (n = 89),  1 to 6 years of age (Liebelt et al., 1999). The blood lead concentration range in
19    the study group (2-84 |ig/dL) was similar to that in the Graziano et al. (2004) study and,
20    consistent with this study, Liebelt et al. (1999) found no association between blood lead
21    concentration and serum hemoglobin levels. However, in contrast to the Graziano et al. (2004)
22    study, blood hemoglobin-adjusted serum erythropoietin levels decreased in association with an
23    increase in blood lead concentration (0.3 mlU/mL decrease per 10 |ig/dL increase blood lead).
24    The Liebelt et al. (1999) study did not include age as a covariate in the regression model, which
25    was shown in the Kosovo prospective study to be a significant covariable in blood lead-serum
26    erythropoietin relationship (Graziano et al., 2004); this may have contributed to the different
27    outcome in the two studies. Liebelt et al. (1999) studied a convenience sample from a
28    lead/primary care  clinic (rather than a prospectively selected cohort) that specifically excluded
29    children who had symptoms of severe iron deficiency,  or were taking iron supplements or other
30    bone marrow suppressing drugs. Iron status of the children in the Graziano et al. (2004) study
31    was not reported.  However, serum ferritin levels in the mothers, at mid-pregnancy, was not

      December 2005                           6-289       DRAFT-DO NOT QUOTE OR CITE

-------
            Table 6-9.3.  Summary of Results of Selected Studies of Associations Between Lead Exposure and Serum Erythropoietin
jcember 2005





Study
Children
Graziano
Liebelt et
Adults
Graziano



et al. (2004)
al. (1999)

etal. (1990)
Osterode etal. (1999)
Oi
i
O
Romeo et

al. (1996)


Subjects

ages: 4.5-12yr
ages: 1-6 yr

pregnant women
male lead workers
male lead workers


Blood Lead (jig/dL)
na Mean (SD) Range

311 6-9, 3 1-3 9b 3-70
86 18C 2-84

48 NR 2-40
40 45 16-91
141 30, 65b'd 30-92



Serum
Erythropoietin Comment

+ adjusted for age, blood Hgb
adjusted for blood Hgb

stratified by blood Hgb
adjusted for blood PCV
no association with blood Hgb

H

6
o


o
H

O

o
H
W

O


O
HH
H
W
          -, decrease; +, increase; Hgb, hemoblogin; NR, not reported; PCV, packed cell volume SD, standard deviation.



          "total number of subjects (including reference group)

          b range of means of low and higher exposure groups

          0 median

          d reference group mean was 10 ug/dL (range 3-20)

-------
               10
            E
            3
            E
            &
            "o
            a.
            o
6-
            UJ
            S   4^
            4>
                    Age 12yr
                      \
                                   Age 4.5 yr
                          10       20       30       40        50
                                         Blood Lead (ug/dL)
                                                        60
                        70
    Figure 6-9.3.   Relationship between blood lead and serum erythropoietin in children.
                   Shown are central tendency predictions based a generalized estimating
                   equation (for repeated measures) relating serum erythropoietin and
                   cumulative lifetime average blood lead concentration, adjusted for age and
                   blood hemoglobin levels (13 g/dL). The regression coefficients relating
                   erythropoietin and blood lead were significant for ages 4.5 (P = 0.21 [95%
                   CI: 0.13, 0.30], p < 0.0001) and 6.5 years (P = 0.12 [95% CI:  0.03, 0.20],
                   p < 0.001).
    Source: Graziano etal. (2004).
1   indicative of iron deficiency (Graziano et al., 1990). Although the direction of the outcome
2   measure was different in the two studies, both studies (Graziano et al., 2004; Liebelt et al., 1999)
3   found evidence for an effect of lead exposure on serum erythropoietin levels in the absence of
4   significant lead-associated changes in blood hemoglobin levels.
5          Three studies have found associations between lead exposure and changes in
6   erythropoiesis biomarkers in adults.  As part of the Kosovo prospective study, serum
7   erythropoietin was measured at mid-pregnancy and at term in a subset of women enrolled in the
8   study (Graziano et al., 1991).  The high- and low-lead cohorts were constructed from the six
    December 2005
                          6-291
DRAFT-DO NOT QUOTE OR CITE

-------
 1    highest and lowest mid-pregnancy blood lead concentrations, within each of four blood
 2    hemoglobin strata, ranging from 9.0 to 12.9 g/dL.  Mean blood lead concentrations in the strata
 3    ranged from 17 to 39 |ig/dL in the high-lead group and 2.4 to 3.6 |ig/dL in the low lead group.
 4    Serum erythropoietin levels significantly decreased in association with increasing blood lead
 5    concentration, independently of an effect of blood hemoglobin (Figure 6-9.4). Romeo  et al.
 6    (1996) also found an association between increasing blood lead concentration and  decreasing
 7    serum erythropoietin, in the absence of discernable changes in blood hemoglobin levels, in a
 8    comparison of groups male lead workers (n = 28, blood lead range 30-92 |ig/dL) and a similar-
 9    aged reference group (n = 113, mean blood lead 10 |ig/dL, range 3-20).  Osterode  et al. (1999)
10    examined several measures of erythropoiesis in a group of lead workers (n = 20, mean  age
11    46 years) and in an age-matched reference group (n = 20). Mean blood lead concentration was
12    45.5 |ig/dL (range 16-91) in the lead workers and 4.1 |ig/dL (range 3-14) in the reference group.
13    Mean  blood hemoglobin levels in the lead worker and reference groups were not different. Lead
14    workers with had blood lead concentrations >60 |ig/dL had significantly lower circulating
15    erythrocyte progenitor cells than the reference group. Also, erythrocyte progenitor cell number
16    was significantly negatively correlated with blood lead concentration and urine lead
17    concentration.  Serum erythropoietin levels increased exponentially with decreasing packed
18    blood  cell volume in the reference group,  but not in the lead workers (i.e., serum erythropoietin
19    level was not significantly correlated with packed cell volume in the lead workers). Thus, unlike
20    the reference group (blood lead concentration < 14 jig/dL), lead workers appeared to have a
21    suppressed erythropoietin response to  declining blood cell volume.
22          Collectively, the results of the above studies suggest that lead exposure depresses serum
23    erythropoietin levels, in the absence of significant depression in blood hemoglobin levels.  Lead-
24    induced nephrotoxicity may contribute to a suppression of erythropoietin levels in  lead-exposed
25    individuals. Although this cannot be entirely ruled out in these studies, both the Romeo et al.
26    (1996) and Osterode et al. (1999) studies excluded people who had a history of hematological or
27    kidney disease.  Nevertheless, renal nephrotoxicity, including proximal tubular nephropathy,
28    could  have been a confounder in these studies which included subjects whose blood lead
29    concentrations were >40 |ig/dL.
30
      December 2005                        6-292          DRAFT-DO NOT QUOTE OR CITE

-------
        I
        o
OU "

70-

00-

50-

40 -


20-



10 -


n-



















(4,4)
J
!

o\x
X \ *
J$!
^
"v\x
§|
^
AX:-

X'\X
\N\S
"\\s
O-'VO
xx"1
•Xv




(


















_
f
I
^
""V*
y$
^
*'}f

^\j
K /^
Y y\
[K
\ S*i
^•^


















^. .- 	 _.







(3,9)
T
^J,^
§$
v\x

\ ^ x^
x^x
\sx
^\\x
' XNv^
s\V




























28 ft)
	 L
X,.

x^
^y^'
K \
..' -.
\s
g'-x



















-"Low" i

E2 -"Hiqh" Leod i
( ) — Mean BPb (ftg/dl)



(3,0)


\XX\] (33.21
§vN 	 L
\ ^v v^ « j£
\\t\ 'jcjc C^ 0*) {3& 2|
^XN ' X' --x^ I --
^ *s^\ '^^f' rO^'^sj x""* ^V '
S \\ vVl" %- ( K -,\ 1 \J- 1
X\ %^ 5*-s [j\J"ei,''M .^x /^J
^V1! K^y^ tv^il—' ^vl
9.0-9.9        10.0-109       11.0-11.9
            Hemoglobin concentration (gfdl)
                                                                    12.0-12.9
    Figure 6-9.4.   Association between blood lead concentration and serum erythropoietin in
                   pregnant women. Shown are combined data for mid-pregnancy and
                   delivery.  Each bar represents the mean (±SD) of 12 subjects. ANOVA of
                   the data at mid-pregnancy and at delivery showed blood lead effects
                   (p = 0.049, p = 0.055, respectively) and blood hemoglobin effects (p = 0.0001.
                   p = 0.009, respectively), with no significant interaction between the two
                   variables.

    Source: Graziano etal. (2001).
1   6.9.2.5    Other Effects on Erythrocyte Metabolism and Physiology

2   6.9.2.5.1    Erythrocyte Nucleotide Metabolism

3         Lead inhibits erythrocyte pyrymidine-5 'nucleotidase (P5N) and adenine dinucleotide

4   synthetase (NADS). Associations between increasing blood lead concentration and decreasing

5   blood P5N and NADS activity have been observed in studies of lead workers (Kim et al. 2002;

6   Mohammed-Brahim et al., 1985; Morita et al., 1997). Mean blood lead concentrations in these

7   study groups were >35 |ig/dL and ranged up to 80 |ig/dL.
    December 2005
                    6-293
DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.9.2.5.2     Erythrocyte Deformability
 2          Horiguchi et al. (1991) compared the deformability of erythrocytes collected from adult
 3    male secondary lead refinery workers (n = 17, age range 24-58 years) with a reference group of
 4    male subjects (n = 13, age range 22-44 years).  Erythrocyte deformability was assessed as
 5    microfilterability of erythrocytes under a negative (-10 cm H2O) pressure head. Erythrocytes
 6    from the lead workers showed significantly lower deformability compared to the reference
 7    group. The mean blood lead concentration in the lead workers was 53.5 |ig/dL (SD 16.1).
 8
 9    6.9.2.5.3     Erythrocyte Membrane Transport
10          Hajem et al. (1990) measured erythrocyte membrane activities of Na+-K+-ATAase, Na+-
11    K+-co-transport, Na+-Li+-antiport, and passive Na+ and K+ permeability in erythrocytes collected
12    from adult males (n = 122, geometric mean blood lead: 16 |ig/dL, range 8.0-33.0) and hair lead
13    was 5.3 |ig/g (95% CI:  4.44, 6.23, range 0.9-60). Na+-K+-co-transport activity was negatively
14    correlated with blood lead concentration but not with hair lead (geometric mean 5.3 jig/g, range
15    0.9-60), and Na+-K+-ATPase activity was negatively correlated with hair lead, but not with
16    blood lead.
17
18    6.9.3     Effects of Lead on the Endocrine System
19    6.9.3.1   Summary of Key Findings of the Effects of Lead on the Endocrine System from
20             the 1986 Lead AQCD
21          The 1986 Lead AQCD concluded that various endocrine processes may be affected by
22    lead at relatively high exposure levels.  These included effects on thyroid hormone levels (e.g.,
23    Refowitz, 1984; Robins et al., 1983), effects on male sex hormone levels  (e.g., Braunstein et al.,
24    1978), and impairment  of the production of 1,25-dihydroxy vitamin D (1,25-OH-D) (e.g., Rosen
25    et al., 1980).  Effects on these endocrine systems  were concluded to be apparent only at blood
26    lead concentrations exceeding 30-40 |ig/dL.  The 1986 Lead AQCD concluded that studies from
27    which the effects of lead on reproductive hormones in females could be assessed were lacking.
28          More recent epidemiologic studies have examined possible associations between lead
29    exposure (as reflected by blood and/or bone lead levels) and various biomarkers of endocrine
30    function, including the thyroid, male reproductive, and calcitropic endocrine systems. These
31    studies have examined endocrine outcomes at lower blood lead ranges and in the absence of
     December 2005                        6-294         DRAFT-DO NOT QUOTE OR CITE

-------
 1    overt clinical lead toxicity, and have more rigorously attempted to control for confounding
 2    factors. Evidence for lead effects on these systems, in association with blood lead concentrations
 3    below 30-40 |ig/dL, remains absent.  The strongest study designs have yielded no associations,
 4    or weak associations, between lead exposure and thyroid hormone status (Erfurth et al., 2001;
 5    Schumacher et al., 1998; Tuppurainen et al., 1988; Zheng et al., 2001).  Similarly, studies of the
 6    male reproductive system that attempted to control for confounding effects of age, have yielded
 7    mixed outcomes (Alexander et al., 1998, 1996; Erfurth et al., 2001; Gustafson et al., 1989;
 8    McGregor and Mason, 1990; Ng et al., 1991).  Results of a more  recent epidemiologic study of
 9    the calcitropic endocrine system in children suggest that associations between serum vitamin D
10    status and blood lead may not be present in calcium-replete children who have average lifetime
11    blood lead concentrations below 25 |ig/dL (Koo et al., 1991).  In  adults, exposures to lead that
12    result in blood lead concentrations >40-60 |ig/dL may increase, rather than decrease, circulating
13    levels of 1,25-OH-D and PTH (Kristal-Boneh  et al., 1999; Mason et al., 1990), possibly as a
14    compensatory response to increased urinary calcium losses, secondary to impaired kidney
15    function.  Details regarding the design of these studies and outcomes are presented in Annex
16    Tables AX6-9.5 and AX6-9.6. Outcomes of the most pertinent studies are summarized below.
17
18    6.9.3.2   Thyroid Endocrine Function
19          Several studies have examined possible associations between lead exposure and thyroid
20    hormone status.  Most of these have been studies of occupational exposures.  The results of these
21    studies have been mixed; some studies have found significant associations with lead exposure
22    (e.g., blood lead concentration), but most studies have found none or relatively weak
23    associations. In studies that have controlled for the effects of age, outcomes also have been
24    mixed, with the strongest study designs finding none or weak associations between lead
25    biomarkers and thyroid hormone status (Erfurth  et al., 2001; Schumacher et al., 1998;
26    Tuppurainen et al.,  1988; Zheng et al., 2001).  The strength of the association and, possibly, the
27    direction of the effect (i.e., increase or decrease in hormone levels) may change with exposure
28    duration or level (Robins et al., 1983; Tuppurainen et al., 1988).  The overall picture that
29    emerges is that those studies that have included subjects having blood lead concentrations
30    exceeding 100 |ig/dL have found depression of serum T3 and/or T4 levels, without a detectable
31    increase in serum TSH. However, studies in which the blood lead distribution was dominated by

      December 2005                        6-295          DRAFT-DO NOT  QUOTE OR CITE

-------
 1    levels well below 100 |ig/dL, have found either no effects or subclinical increases in serum T3,
 2    T4, with no change in TSH levels. Outcomes from the most pertinent studies are summarized
 3    qualitatively in Table 6-9.4 and are described in greater detail below.
 4          Siegel et al. (1989) measured serum total thyroxine (TT4) and free thyroxine (FT4) in
 5    children ages 11 months to 7 years (n = 68) who were outpatients at a clinical care facility.
 6    Mean blood lead concentration in the study group was 25 |ig/dL (range 2-77).  In a simple
 7    (univariate) linear regression analysis, hormone levels were not significantly associated with
 8    blood lead concentration.
 9          Zheng et al. (2001) measured concentrations of TT4 and transthyretin (TTR) in serum  and
10    cerebral spinal  fluid (CSF) of adult hospital patients (n = 82) admitted for evaluation of CSF
11    clinical chemistry (e.g., for head wounds, tumors, neurological symptoms). Mean blood lead
12    concentration was 14.9 |ig/dL (SD 8.3). Age-adjusted serum TT4 and TTR, and CSF TT4 were
13    not significantly associated with blood lead concentration; however, increasing CSF lead
14    concentration was associated with decreasing CSF TTR levels (r = -0.30, p = 0.023).
15          Possible associations between lead exposure and thyroid hormone status have been
16    examined in several studies of lead workers (Dursun and Tutus, 1999; Erfurth et al., 2001;
17    Gennart et al., 1992;  Gustafson et al., 1989; Horiguchi et al., 1987; Lopez et al., 2000; Refowitz,
18    1984; Robins et al., 1983; Schumacher et al.,  1998; Singh et al., 2000; Tuppurainen et al.,  1988).
19    Of these, six warrant particular attention because the design and/or analysis attempted to control
20    for effects of age (Erfurth et al., 2001; Dursun and Tutus, 1999; Gustafson et al.,  1989;
21    Schumacher et al., 1998;  Tuppurainen et al., 1988; Robins et al., 1983).  Outcomes of these
22    studies are summarized in Table 6-9.4. The largest studies were Erfurth et al. (2001),
23    Schumacher et al. (1998), and Tuppurainen et al. (1988).
24          Erfurth et al. (2001) was a cross-sectional study of secondary smelter workers (n = 62)
25    and a reference group of metal (not lead) workers (n = 26).  Excluded from the study were
26    individuals with ongoing thyroid disease or who were taking thyroid hormone supplements or
27    other drugs that would interfere with thyroid hormone levels (e.g., beta-blockers). Median blood
28    lead concentration in the lead workers was 31 |ig/dL (range 8-93  jig/dL). Age-adjusted basal
29    serum levels of FT3,  FT4, and TSH were not associated with blood, urine, or finger bone lead
30    levels.  Thyroid releasing hormone (TRH)-induced TSH secretion (area under serum TSH
31    concentration-time curve) was measured in an age-matched subset of the study group (9 lead

      December 2005                       6-296         DRAFT-DO NOT QUOTE OR CITE

-------
December 2005




Oi
i
VO


O
H
6
o
0
H
O
o
H
W
O
O
H
W
Table 6-9.4. Summary of Results of Selected Studies of Associations Between Lead Exposure
Thyroid Hormone Levels
Study
Children
Siegel etal. (1989)
Adults
Dursun and Tutus (1999)
Erfurth etal. (2001)
Gustafson etal. (1989)

Robins etal. (1983)
Schumacher etal. (1998)
Tuppurainen et al. (1988)
Zheng etal. (2001)
- , decrease; +, increase; o
Blood Lead (jig/dL)
SubJects "a Mean(SD) Range T3

children, 1 1 mo-7 yrs 68 25 2-77 NR

metal powder manufacture workers 57 17.1(9.0) 1-36 +
secondary smelter workers 88 31.1° 4-93 o
secondary smelter workers 42 39.4(2.1) NR o

brass foundry workers 47 NR 16-127 NR
primary smelter workers 151 24.1 15>40% o
battery manufacture workers 176 55.9(23.8) 5-134
general population 82 14.9(8.3) NR NR
, no effect; NR, not reported; T3, triiodothyronine; T4, thyroxine; TSH, thyroid stimulating hormone
and
T4 TSH

o NR

+ o/ob
0 0
+ 0

NR
0 0
0
o NR

a Total number of subjects (including reference group)
b basal/thyroid releasing hormone-stimulated
0 median













-------
 1    workers and 11 reference subjects) and was not significantly different in the two groups.  The
 2    Schumacher et al. (1998) study measured serum FT4, TT4, and TSH levels in a group of male
 3    workers (n = 151) at the Trail British Columbia smelter complex. Excluded from the study were
 4    individuals who had ongoing clinical thyroid disease. Mean blood lead concentration in the
 5    study group was 24 |ig/dL (15% >40 jig/dL). Covariate-adjusted (age, alcohol consumption)
 6    hormone levels were not significantly associated with current blood lead concentration or
 7    10-year average blood lead concentrations. Prevalence of abnormal hormone values was also
 8    unrelated to blood lead concentration.
 9          Tuppurainen et al. (1988) measured serum total triiodothyronine (TT3), FT4, TT4,
10    and TSH levels in a group of male battery manufacture workers (n = 176). Mean blood lead
11    concentration was 56 |ig/dL (range 14-134 |ig/dL).  Although, hormone levels were not
12    significantly associated with blood lead concentrations, increasing exposure (i.e., employment)
13    duration was significantly associated with decreasing FT4 (r2 = 0.071, p = 0.001) and TT4 (r2 =
14    0.059, p = 0.021) levels.  The r2 was not improved by including age or blood lead as covariables.
15    Strength of the association was greater when the analysis was restricted to workers who had an
16    exposure duration >7.6 years (FT4:  r2 = 0.33, p < 0.002; TT4: r2 = 0.21, p < 0.001). Consistent
17    with the results of the Tuppurainen et al. (1988) study, Robins et al. (1983) found a significant
18    association between increasing blood lead concentration and decreasing FT4 (r2 = 0.085,
19    p = 0.048) in a group of brass foundry workers (n = 47). The blood lead range in the subjects
20    was 16-127 |ig/dL. When stratified by race (black, white) the association was significant in the
21    black stratum (r2 = 0.21, p =  0.03), but not in the white stratum (r2 = 0.05, p = 0.27).  The
22    strength of association was not changed by including age in the regression model.  Both the
23    Robins et al. (1983) and Tuppurainen et al. (1988) included subjects with blood lead
24    concentrations >100 |ig/dL.
25          Blood lead concentrations were lower in the Dursun and Tutus (1999) and Gustafson et al.
26    (1989) studies than in the above studies, and both studies found significant associations between
27    lead exposure and increasing serum TT4 levels.  Dursun and Tutus (1999) measured serum FT3,
28    TT3, FT4, TT4, and TSH in a group of metal powder manufacture workers (n = 27) and a
29    reference group (n = 30). Mean blood lead concentration in the workers was 17 |ig/dL (range 9-
30    36 jig/dL). A linear regression model that included age, blood lead concentration, and exposure
31    duration, indicated a significant association between increasing exposure duration and increasing

      December 2005                        6-298           DRAFT-DO NOT QUOTE OR CITE

-------
 1    serum TT4 levels (r2 = 0.3, p = 0.03).  The Gustafson et al. (1989) study examined a group of
 2    male secondary smelter workers (n = 21) and reference subjects, individually matched to the lead
 3    workers by age, sex, and work shift. Mean blood lead concentration in the workers was
 4    39 |ig/dL (SD 2).  Serum TT4 levels were significantly higher (p < 0.02) in the lead workers
 5    compared to the reference group. The difference strengthened when the analysis was restricted
 6    to the age range <40 years (p = 0.01).
 7
 8    6.9.3.3   Reproductive Endocrine Function
 9    6.9.3.3.1     Male Reproductive Endocrine Function
10          Low testosterone (TES) levels, blunted sex hormone secretion in response to
11    gonadotropin releasing hormone (GNRH), and defects in spermatogenesis have been observed in
12    humans exhibiting clinical neurological symptoms of lead poisoning (Braunstein et al., 1978;
13    Cullen et al.,  1984). However, the effects of lower exposure levels on reproductive endocrine
14    status are less clear. Possible associations between lead exposure and changes in male
15    reproductive hormone levels have been examined in studies of lead workers. Of these, five
16    studies attempted to control for effects of age, an important determinant of testosterone levels
17    (Alexander et al.,  1998; Erfurth et al., 2001; Gustafson et al., 1989; McGregor and Mason, 1990;
18    Ng et al., 1991).  The outcomes from these studies are qualitatively summarized in Table 6-9.5.
19    Blood lead ranges in the latter studies were similar (4-90 |ig/dL), yet outcomes were mixed, with
20    no change (Erfurth et al., 2001; Gustafson et al., 1989; McGregor and Mason, 1990) or
21    subclinical decrease (Alexander et al., 1998, 1996; Ng et al., 1991) in serum testosterone (TES)
22    in association with lead exposure. Mixed effects were observed for the effect of lead exposure
23    on serum follicle stimulating hormone (FSH) and luteinizing hormone (LH), increases
24    (McGregor and Mason, 1990; Ng et al., 1991), decreases (Gustafson et al., 1989), and with no
25    change (Alexander et al., 1998, 1996; Erfurth et al., 2001) in hormone levels observed. The
26    inconsistency in the direction of effects on TES and the two androgen regulating pituitary
27    hormones, FSH and LH, is particularly noteworthy.  In the absence of an abnormality in the
28    hypothalamic-pituitary regulation of testosterone levels, an association between declining serum
29    TES (or free TES) and increasing FSH and LH levels would be expected.  Erfurth et al. (2001)
30    observed a suppressed FSH response to GNRH in a group of lead workers compared to an
     December 2005                        6-299          DRAFT-DO NOT QUOTE OR CITE

-------
0 Table 6-9.5.
o

^
to
o Study
Alexander et al. (1998, 1996)
Erfurthetal. (2001)
Gustafson et al. (1989)
McGregor and Mason (1990)
Ngetal. (1991)
Summary of Results of Selected Studies of Associations Between Lead Exposure and
Male Sex Hormone Levels in Adults
Blood Lead (jig/dL)
Subjects
primary smelter workers
secondary smelter workers
secondary smelter workers
lead workers
battery manufacture workers
"a
152
88
42
176
171
Mean (SD)
NR
31. 1°
39.4(2.1)
NR
35(13)
Range
5-58
4-93
NR
17-77
10-72
FSH LH PRL
o o NR
o/-^ o/od o/od
- - 0
+ + NR
+ + 0
TES
_b
od
0
0
-
H

6
o


o
H

O

o
H
W

O


O
HH
H
W
o

°         -, decrease; +, increase; o, no effect; NR, not reported, FSH, follicle stimulating hormone, LH, luteinizing hormone; PRL, prolactin; TES, testosterone


           a total number of subjects (including reference group)

           b in association with increasing semen lead levels, not with blood lead

           0 median

O         d basal/gonadotropin releasing hormone-stimulated
            effect was evident in comparison between groups, but not in multivariate regression that adjusted for age

-------
 1    age-matched reference group; however, the magnitude of the response was not significantly
 2    associated with lead exposure indices in a multivariate regression analysis that accounted for age.
 3          Alexander et al. (1998, 1996) examined serum FSH, LH, and TES in males (n = 152) who
 4    worked at the Trail British Columbia smelter complex. Covariate-adjusted hormone levels and
 5    prevalence of clinically abnormal values were unrelated (p > 0.05) to blood lead concentration
 6    (range 5-58 |ig/dL); however, increasing semen lead concentration (range 0.3-17 |ig/dL) was
 7    significantly associated with decreasing semen testosterone levels (p = 0.004).  Erfurth et al.
 8    (2001) measured serum TES, sex hormone binging globulin (SHBG), and GNRH-stimulated
 9    changes in serum FS, LH, and PRL in male secondary smelter workers (n = 62) and in
10    a reference group (n = 26). Mean blood lead in the lead workers was 31 |ig/dL (range
11    8-93 jig/dL). Age-adjusted basal hormone levels were unrelated to blood, plasma, or urine lead
12    concentrations. In an age-matched subset of the cohorts (n = 9 lead workers, n = 11 reference),
13    median GNRH-stimulated serum FSH was significantly lower in lead workers than in the
14    reference group; however, GNRH-stimulated LH, FSH, and PRL were not significantly
15    associated with any of the lead measures in a multivariate regression analysis.  Gustafson et al.
16    (1989) measured serum FSH, LH, and TES (total and free) in a group of male secondary smelter
17    workers (n = 21) and in a group of reference subjects individually matched to the lead workers
18    by age, sex, and work shift. Mean blood lead concentrations were 39 |ig/dL (SD 2) in the lead
19    workers and 5.0 |ig/dL (SD 0.2) in the reference group. Serum FSH levels were significantly
20    lower (p = 0.009) in lead workers compared to reference group. When the analysis was
21    restricted to the age range <40 years, lead workers had significantly lower FSH and LH
22    compared to the reference group. McGregor and Mason (1990) measured serum FSH, LH, TES,
23    and SHBG in a group of male lead workers (n = 90) and in a reference group (n = 86).  Blood
24    lead range in the lead workers was 17-77 |ig/dL; blood lead concentrations in the reference
25    subjects were <12 |ig/dL. Prevalences of abnormal hormone levels in the lead workers and
26    reference group were not different; however, age-adjusted serum FSH was significantly higher in
27    lead workers compared to reference group  and increasing FSH levels were significantly
28    associated with increasing blood lead concentrations.  Increasing  serum LH was significantly
29    associated with increasing exposure duration but not with blood lead concentration or age.
30    Serum TES or SHBG levels were unrelated to blood lead concentration or exposure duration.
31    Ng et al. (1991) measured serum FSH, LH, PRL, and TES  in a group of male battery

      December 2005                       6-301           DRAFT-DO NOT QUOTE OR CITE

-------
 1    manufacture workers (n = 122) and a reference group (n = 49). Mean blood lead concentrations
 2    were 35 |ig/dL (range 10-77 |ig/dL) in the lead workers and 8 |ig/dL (range 3-15 |ig/dL) in the
 3    reference group. When cohorts were stratified by age, serum FSH and LH levels were
 4    significantly higher in lead workers <40 years of age compared to corresponding age stratum of
 5    the reference group; serum TES was significantly lower in lead workers >40 years of age,
 6    compared to the same age stratum in the reference group.  Covariate-adjusted (age, tobacco
 7    smoking) serum TES levels were significantly lower in lead workers in the 10-year exposure
 8    duration stratum, compared to the reference group.  Covariate-adjusted serum FSH and LH were
 9    significantly higher in lead workers in the < 10-year exposure duration stratum, compared to the
10    reference group.
11
12    6.9.3.3.2     Female Reproductive Endocrine Function
13          Although delays in sexual  maturation in humans have been associated with blood lead
14    concentrations (Selevan et al., 2003; Wu et al., 2003), and lead has been shown to alter levels of
15    female sex hormones and the menstrual cycle in nonhuman primates (Foster, 1992; Franks et al.,
16    1989; Laughlin et al.,  1987), epidemiologic studies of interactions between lead exposure and
17    reproductive endocrinology in females have not been reported. Lead introduced into cultures of
18    human ovarian granulosa cells suppresses progesterone production (Paksy et al., 2001) and
19    suppresses expression of aromatase and estrogen receptor P (Taupeau et al., 2003).
20
21    6.9.3.4   Pituitary and Adrenal Endocrine Function
22          Several studies of possible associations between lead exposure and levels of pituitary
23    hormones that regulate production and secretion of thyroid hormones (see Section 6.9.3.2)  and
24    reproductive hormones (see Section 6.9.3.3) have been reported.  In  addition to the above
25    studies, Gustafson et al. (1989) found that serum  cortisol levels were lower in a group of male
26    secondary smelter workers (n = 21) compared to  a reference group individually matched to the
27    lead workers by age, sex, and work shift. Mean blood lead concentration were 39 |ig/dL (SD 2)
28    in the workers and 5.0 |ig/dL (SD 0.2) in the reference group. Campbell et al. (1985) measured
29    various biomarkers of status of the renin-angiotensin-aldosterone system in male welders (n = 5)
30    and reference subjects (n = 8).  Mean blood lead concentration was 35 |ig/dL (range 8-62 |ig/dL).
31    Significant positive correlations were observed between blood lead concentration and plasma

      December 2005                        6-302          DRAFT-DO NOT QUOTE OR CITE

-------
 1    aldosterone (r = 0.53, p < 0.002), which may have been, at least in part, secondary to a lead
 2    effect on plasma renin activity (r = -0.76, p < 0.001) and angiotensin I levels (r = 0.68,
 3    p < 0.002).  Saenger et al. (1984) found lower urinary levels of 6-p-OH-cortisol, but not cortisol,
 4    in children who had elevated urinary lead in an EDTA provocation test (>500 jig/24 h),
 5    compared to children who did not have elevated urinary lead levels, or whose blood lead
 6    concentrations were <30 |ig/dL.  The change in urinary excretion of 6-p-OH-cortisol in the
 7    absence of a change in cortisol levels may reflect an effect of lead on liver cytochrome P450
 8    activity, rather than an effect on the  adrenal gland (see Section 6.9.4).
 9
10    6.9.3.5   Calcitropic Endocrine Function
11          Children exposed to relatively high level of lead >30 |ig/dL may exhibit depressed levels
12    of circulating  1,25-OH-D (Mahaffey et al., 1982; Rosen et al., 1980). These effects were not
13    detected in a study of calcium-replete children with average lifetime blood lead levels below 25
14    |ig/dL (Koo et al., 1991). In adults,  lead exposures that result in blood lead concentrations
15    >40-60 |ig/dL may increase, rather than decrease, circulating levels of 1,25-OH-D and PTH.
16    These studies  also are summarized in Annex Tables AX6-9.5 and AX6-9.6.  Outcomes from the
17    more pertinent studies are qualitatively summarized in Table 6-9.6 and are discussed in greater
18    detail below.
19          Epidemiologic studies of possible associations between lead exposure and vitamin D
20    status in children have yielded mixed results. Mahaffey et al. (1982) and Rosen  et al. (1980)
21    observed lower 1,25-OH-D in association with increasing blood lead concentration. Koo et al.
22    (1991) found no association between 1,25-OH-D and blood lead concentration.  The Koo et al.
23    (1991) study was a longitudinal analysis of a subset of a prospective study of pregnancy
24    outcomes.  Serum calcium magnesium, phosphorus, PTH, CAL, 25-OH-D, 1,25-OH-D, and
25    bone mineral content were measured in children (n = 105) at ages 21, 27, and 33 months. Mean
26    lifetime average blood lead concentrations (based on quarterly assessments) was 9.7 |ig/dL
27    (range 4.8-23.6 |ig/dL). The range of highest values observed was 6-63 |ig/dL.  A structural
28    equation model was developed that initially considered age, sex, race, sampling season, and
29    dietary intake  of calcium, phosphorus, and vitamin D as covariables; the final model  retained
30    age, sex, race, and sampling season. Decreasing blood lead (In-transformed) was significantly
31    associated with covariate-adjusted decreasing serum phosphorus. No other covariate-adjusted

      December 2005                        6-303          DRAFT-DO NOT QUOTE OR CITE

-------
o
o
3
1 L <
V_)
^
to
o
o





ON
i
O


O
!*
i-rj
H
6
o
2;
0
H
O
o
H
W
O
O
H
W
Table 6-9.6. Summary of Results of Selected Studies of Associations Between Lead Exposure
and Calcitropic Hormones
Blood Lead (jig/dL)

Study Subjects na Mean (SD) Range PTH CAL !'25D
Children
Koo etal. (1991) ages: 21, 27, 33 mo 105 9.7 5-24 o o o
Mahaffey etal. (1982) ages: 1-16 yr 177 NR 12-120 o o -
Rosen etal. (1980) ages: 1-5 yr 45 18, 47, 74b 10-120 + o
Adults
Chalkley etal. (1998) smelter workers' 19 47 21-76 NR NR +c

Kristal-Boneh etal. (1998) battery manufacture workers 140 43 1-77 + NR +
Mason etal. (1990) lead workers 138 NR 15-95 o NR +

-, decrease; +, increase; o, no effect; NR, not reported, PTH, parathyroid hormone; CAL, calcitonin; 1,25D, 1,25-dihydroxyvitaminD; 25D, 25-
hydroxyvitamin D

a total number of subjects (including reference group)
b group means: low, moderate, high
0 cadmium, lead, zinc smelter workers, effect on 1,24D in association with high blood cadmium and lead and high urinary cadmium














25D

0
0
-

0

NR
NR


















-------
 1    outcomes were significantly associated with blood lead. The distribution of dietary calcium
 2    intakes was 4% for <600 mg/day, 55% for 600-1200 mg/day, and 41% for >1200 mg/day.
 3    Intakes of phosphorous were similar, suggesting that the subjects were nutritionally replete with
 4    respect to these two nutrients.
 5          The different outcomes in Koo et al. (1991) compared to the Mahaffey et al. (1982) and
 6    Rosen et al. (1980) studies may reflect, in part, the lower blood lead range in the subjects in Koo
 7    et al. (1991) (range of lifetime average 5-24 |ig/dL, range of observed highest values 6-63
 8    |ig/dL) compared to the Mahaffey et al. (1982) and Rosen et al. (1980) studies (10-120 (ig/dL).
 9    Subjects in the Koo et al. (1991) study also had higher calcium intakes (4% with <600 mg/day,
10    43% with >1200 mg/day) than in the Rosen et al. (1980) study (mean 580 mg/day [SE 15] in
11    high blood lead group).  Calcium intake (and/or related nutritional factors) may also have been
12    an uncontrolled confounder in the Rosen et al. (1980) study, as higher blood lead concentration
13    appeared to be associated with lower calcium  intakes (Sorrell et al.,  1977). Mahaffey et al.
14    (1982) did not report calcium intakes. Thus, the effect of lead exposure on vitamin D status may
15    be more pronounced at higher blood lead concentrations (i.e., >60 |ig/dL) and in combination
16    with lower intakes of calcium (or other nutritional limitations).
17          Studies of lead workers have found evidence for higher serum levels of 1,25-OH-D and
18    PTH in association with increasing blood lead concentration (Chalkley et al., 1998; Kristal-
19    Boneh et al.,  1998; Mason et al., 1990). The Chalkey et al. (1998) study was a small study
20    (n = 19) of subjects exposed to both cadmium and lead, and effects of lead and cadmium on
21    1,25-OH-D could not be isolated. The Kristal-Boneh et al. (1998) and Mason et al. (1990)
22    studies included larger samples of subjects whose exposure was primarily, but not exclusively,
23    to lead. Attempts were made to control for effects of age and, in the Kristal-Boneh et al. (1998)
24    study,  other potential covariables. Kristal-Boneh et al. (1998) measured  serum calcium,
25    magnesium, phosphorus, PTH, 25-OH-D, and 1,25-OH-D in a group of male battery
26    manufacture workers (n = 56) and a reference group (n = 90). Mean blood lead concentrations
27    were 43 |ig/dL (SD  14, range 1-77 |ig/dL) in the lead worker group and 4.5 |ig/dL (SD 2.6, range
28    1.4-19 (ig/dL) in the reference group.  Serum 1,25-OH-D and PTH, but not 25-OH-D, were
29    significantly higher in lead workers compared to the reference group. Increasing blood lead
30    concentration (In-transformed) was significantly associated with covariate-adjusted increasing
31    serum  PTH and 1,25-OH-D levels. No effects on serum calcium were apparent. Occupational

      December 2005                            6-305        DRAFT-DO  NOT QUOTE OR CITE

-------
 1    lead exposure was also significantly associated with increasing PTH and 1,25-OH-D level.
 2    Covariates retained in the multivariate model were age, alcohol consumption, smoking; calcium
 3    intake, magnesium intake, and calorie intake.  Mason et al. (1990) measured serum calcium,
 4    phosphate, PTH, and 1,25-OH-D in male lead workers (n = 63) and in a reference group (n = 75)
 5    and found significantly higher prevalence of elevated 1,25-OH-D (defined as >2 SD higher than
 6    reference mean) in lead workers (13%) compared to the reference group (1.3%). Serum levels of
 7    1,25-OH-D were also significantly higher in lead workers compared to the reference group.
 8    After stratification of the lead workers into exposure categories (high exposure: blood lead >40
 9    |ig/dL and bone lead >40 |ig/g; low exposure: blood lead ^40 |ig/dL and bone lead <40 |ig/g),
10    serum 1,25-OH-D levels were significantly higher in the high lead group.  Serum calcium levels
11    were not different in the two groups.  Increasing blood lead was significantly associated with
12    increasing 1,25-OH-D levels (r2 = 0.206; with age and bone lead included, r2 = 0.218).  After
13    excluding 12 subjects whose blood lead concentrations >60 |ig/dL, the regression coefficient was
14    no longer significant (r2 = 0.162, p = 0.26).
15
16    6.9.4    Effects of Lead on the Hepatic System
17    6.9.4.1    Summary of Key Findings of the Effects of Lead on the Hepatic System
18             from the  1986 Lead AQCD
19         The 1986 Lead AQCD noted that effects of lead on liver function in humans had not been
20    extensively studied.  Possible association between lead exposures (blood lead concentrations
21    >70 |ig/dL) and nonspecific liver injury (i.e., increases in liver enzymes in serum) were noted
22    based on studies of workers (e.g., Cooper et al., 1973; Hammond et al., 1980). Also noted was
23    evidence for possible association of suppression of hepatic cytochrome P450 activity with high
24    blood lead concentrations (>70 |ig/dL) (Meredith et al., 1977).
25         Few studies of hepatic effects of lead on humans have been reported since the 1986 Lead
26    AQCD.  Studies of hepatic enzyme levels in serum suggest that liver injury may be present in
27    lead workers; however, associations specifically with lead exposures are not evident (Al-Neamy
28    et al., 2001; Hsiao et al., 2001).  Studies of urinary metabolites of cytochrome P450 phenotypes
29    CYP2A6 and CYP3 A4 suggest possible associations between lead exposure and suppression of
30    hepatic enzyme activity. The effect on CYP2A6 activity was observed in children with high lead
31    burdens (i.e., blood lead concentration >40 |ig/dL, EDTA-provoked  urinary lead >500 |ig/dL).

      December 2005                          6-306       DRAFT-DO NOT QUOTE OR CITE

-------
 1    The effect on CYP3 A4 was observed in association with blood lead ranges of approximately
 2    30-112 |ig/dL (based on reported serum lead concentrations). These studies are summarized in
 3    Annex Table AX6-9.7 and the most pertinent findings are discussed below.
 4
 5    6.9.4.2   Non-specific Hepatic Injury
 6          Possible association between occupational lead exposure and liver injury has been
 7    assessed from measurements of serum enzymes (Al-Neamy et al., 2001; Hsiao et al., 2001).
 8    Al-Neamy et al. (2001) found significantly higher serum activity of alkaline phosphatase (AP)
 9    and lactate dehydrogenase (LDH), both within clinically normal ranges, in a group (n = 100) of
10    male lead workers (e.g., gas pump attendants, garage workers, printing workers, construction
11    workers), compared to an age-matched reference group (n = 100).  Serum levels of alanine
12    aminotransferase (ALT), aspartate aminotransferase (AST), and y-glutamyl transferase (y-GT)
13    were not different in the two groups. The mean lead concentrations were 78 |ig/dL (SD 43)
14    in the lead workers and 20 |ig/dL (SD 12) in the reference group.  Hsiao et al. (2001) found no
15    association between blood lead concentration and ALT  activity, in a longitudinal study of a
16    group  of battery manufactory workers (n = 30). Mean blood lead concentrations ranged from
17    60 |ig/dL (approximate range 25-100 |ig/dL) at the start of the study (1989) and 30 |ig/dL
18    (approximate range 10-60 |ig/dL) in the final year of the study (1999).
19
20    6.9.4.3   Hepatic Cytochrome P-450 Function
21          Urinary excretion of 6-p-hydroxy cortisol (6-p-OH-cortisol) derives primarily from
22    oxidation of cortisol through the hepatic cytochrome P450 phenotype CYP3 A4. A lower urinary
23    6-p-OH-corti sol: cortisol ratio is indicative of possible suppression of hepatic CYP3A4 activity.
24    Saenger et al. (1984) found significantly lower (-45% lower) urinary excretion of 6-P-OH-
25    cortisol and lower urinary 6-p-OH-cortisol:cortisol ratio in 2-9 year-old children (n = 26) who
26    qualified for chelation (EDTA-provoked urinary lead >500 |ig/24 h) than in children who did not
27    qualify, and significantly lower than in an age-matched reference group. Urinary 6-P-OH-
28    cortisol:cortisol ratio was significantly  correlated with blood lead (r =  -0.514, p < 0.001), urinary
29    lead, and EDTA-provoked urinary lead (r = -0.593, p < 0.001). Mean blood lead concentrations
30    were 46 |ig/dL (range 33-60 |ig/dL), prior to chelation, and 42 |ig/dL (range 32-60 |ig/dL) in the
31    children who did not qualify for chelation.

      December 2005                           6-307        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Satarug et al. (2004) measured urinary excretion of 7-hydroxy-coumarin (7-OH-
 2    coumarin) following a single oral dose of coumarin to assess effects of cadmium and lead
 3    exposure on cytochrome P450 phenotype CYP2A6.  The rationale for this approach is that
 4    7-hydroxylation of coumarin occurs solely through the CYP2A6 pathway. Coumarin-induced
 5    urinary 7-OH-coumarin was measured in a group (n = 118) selected from the general population
 6    in Bangkok, Thailand.  All subjects were nonsmokers. The study found a significant association
 7    between increasing urinary lead and decreasing covariate-adjusted urinary 7-OH-coumarin in
 8    males, but not in females.  Covariates retained included age and zinc excretion. A significant
 9    association, in opposite direction, was found between urinary cadmium and urinary 7-OH-
10    coumarin.  Mean urinary lead levels (blood lead concentrations were not reported) were 1.3 jig/g
11    creatinine (range 0.1-1.2 |ig/dL) in males, and 2.4 |ig/g creatinine (range 0.6-6.8 |ig/dL) in
12    females. Mean serum lead concentrations were 4 |ig/L (range 1-28 |ig/dL) in males and 3 |ig/dL
13    (range 1-12 |ig/dL) in females. The range 1-28 |ig/L serum would correspond to a blood lead
14    concentration range of approximately 30-112 |ig/dL (U.S. Environmental Protection Agency,
15    2003).  These results are consistent with observations of depressed excretion of metabolites of
16    the CYP2A6 substrate, phenazone, in association with overt clinical lead toxicity in lead workers
17    (Fischbein et al., 1977; Meredith et al., 1977).
18
19    6.9.5  Effects of Lead on the Gastrointestinal System
20    6.9.5.1    Summary of Key Findings of the Effects of Lead on the Gastrointestinal
21             System from the 1986 Lead AQCD
22          The 1986 Lead AQCD described gastrointestinal  colic (abdominal pain, constipation,
23    intestinal paralysis) as a consistent early symptom of lead poisoning in humans and noted that
24    such symptoms may be present in association with blood lead concentrations in the range of
25    30-80 |ig/dL. The 1986 Lead AQCD concluded that information was insufficient to establish
26    clear concentration (i.e., blood concentration)-response relationships in the general population in
27    association with environmental exposure. Subsequent to the 1986 AQCD several studies of
28    prevalence of symptoms of gastrointestinal colic in lead workers have been reported that provide
29    evidence for symptoms in association with blood lead concentrations >50-80 |ig/dL (Awad el
30    Karim et al., 1986; Holness and Nethercott, 1988; Lee et al., 2000; Matte et al., 1989).
     December 2005                           6-308       DRAFT-DO NOT QUOTE OR CITE

-------
 1    Summaries of these studies are presented in Annex Table AX6-9.8.  Similar types of studies of
 2    children have not been reported.
 3
 4    6.9.5.2   Gastrointestinal Colic
 5          Lee et al. (2000) collected data on symptoms (self-reported questionnaire) in male lead
 6    workers (n = 95) who worked in secondary smelters, PVC-stabilizer manufacture facilities, or
 7    battery manufacture facilities. A logistic regression model was applied to the prevalence data for
 8    gastrointestinal symptoms (loss of appetite, constipation or diarrhea, abdominal pain). The
 9    covariate-adjusted odds ratio for symptoms, in association with blood lead concentration
10    (>versus versus <260.5 |ig/4 h, the group
12    median) was also not significant (1.1, [95% CI: 0.4, 2.5]).  However, the odds ratio for
13    neuromuscular symptoms in association with DMSA-provoked urinary lead was significant
14    (7.8, [95% CI:  2.8, 24.5]), suggesting that neuromuscular symptoms may occur in association
15    with exposures that are insufficient to result in detectable gastrointestinal symptoms. Covariates
16    retained in the final regression models were age, tobacco smoking, and alcohol consumption.
17          Three other studies have attempted to quantify associations between lead exposure and
18    gastrointestinal symptoms in lead workers (Awad el Karim et al., 1986; Holness and Nethercott,
19    1988; Matte et al., 1989). Holness and Nethercott (1988) found a significantly (p < 0.05) higher
20    prevalence of symptoms in a group of demolition workers (n =  119) in association with a blood
21    lead concentration range 50-70 |ig/dL (n = 87), 37% for abdominal cramps and 42% for
22    constipation, or >70 |ig/dL (n = 19) 77% for abdominal cramps and 62% for constipation
23    compared to a group of workers in which the blood lead concentration range was <50 |ig/dL
24    (n = 13), prevalences of 8% and 6%. Awad el Karim et al. (1986) found higher prevalence of
25    gastrointestinal symptoms, for abdominal colic and constipation, respectively, in male battery
26    manufacture workers, 41.3% for abdominal colic and 41.4% for constipation, compared to a
27    reference group of workers, n = 40 prevalences of 7.5% and 10% for abdominal colic and
28    constipation, respectively. The blood lead ranges were 55-81 |ig/dL in the lead workers and
29    7-33 |ig/dL in the reference group. Matte et al. (1989) did not find a significant difference in
30    prevalence of gastrointestinal symptoms (decreased appetite, nausea, abdominal pain) among a
31    group of battery manufacture and repair workers (n = 63) when stratified by blood lead

      December 2005                           6-309       DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentration (60 |ig/dL, >60 |ig/dL). The prevalence ratio (high/low blood lead strata) for
 2    abdominal pain was 1.5 (95% CI:  0.5, 4.6).
 3          In a small study of environmentally-exposed adults, Bercovitz and Laufer (1991) found
 4    that the lead level in the dentine of patients with gastrointestinal ulcers (n = 11), even long after
 5    recovery, were significantly higher (mean lead 75.02 |ig/g [SE 8.15]) than that in healthy
 6    subjects (mean lead 25.62 jig/g [SE 10.15]).  Ten of the 11 peptic ulcer patients had a higher lead
 7    level than the healthy subjects. In these 10 patients, increased severity of the ulcer and longevity
 8    of suffering was associated with increased tooth lead levels. The authors suggested that
 9    increased absorption of lead was associated with damage to the epithelial mucosal cells of the
10    gastrointestinal tract.
11
12    6.9.6   Effects of Lead on the Respiratory System
13    6.9.6.1   Summary of Key Findings of the Effects of Lead on the Respiratory System
14             from the 1986 Lead AQCD
15          The 1986 Lead AQCD did not discuss effects of lead on the respiratory tract on humans.
16    Only one study since the 1986 document has examined the association between lead and
17    respiratory health outcomes.
18
19    6.9.6.2   Pulmonary Function
20          Bagci  et al. (2004) conducted pulmonary function tests on a group of male battery
21    manufacture workers (n = 22), automobile exhaust repair workers (n = 40), and a group of
22    hospital workers (n = 24). Mean blood lead  concentrations were 37 |ig/dL (SD 8) in the battery
23    manufacture group, 27 |ig/dL (SD 9) in the exhaust repair group, and 15  |ig/dL (SD 3) in the
24    hospital workers. Lead workers and the reference group had similar tobacco smoking
25    prevalences (51-56%). Battery manufacture workers had significantly lower forced expiratory
26    volume in one second (FEVi), FEVi:vital capacity (VC) ratio, FEVi/forced vital capacity (FVC)
27    ratio, forced expiration flow (FEF), and maximum voluntary ventilation (MVV) compared to
28    the hospital workers. Blood lead concentration was significantly negatively correlated with
29    FEVi/FVC (r = -0.31, p = 0.006) and FEF (r = -0.30, p = 0.009) after adjusting for age,
30    cigarette smoking, and exposure duration. Results from this study are further summarized in
31    Annex Table AX6-9.9.

      December 2005                           6-310        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.9.7  Effects of Lead on Bone and Teeth
 2    6.9.7.1    Summary of Key Findings of the Effects of Lead on Bone and Teeth from
 3             the 1986 Lead AQCD
 4         The 1986 Lead AQCD did not discuss the effects of lead on bone and teeth. Since
 5    completion of the 1986 AQCD, an additional development in lead epidemiology has been studies
 6    that have explored possible associations between lead exposure and risk of dental caries
 7    (Campbell et al., 2000; Dye et al., 2002; Gemmel et al., 2002; Moss et al., 1999). In addition,
 8    a limited number of studies also examined the toxic effect of lead on bone. These studies are
 9    summarized in Annex Table AX6-9.10.
10
11    6.9.7.2    Bone Toxicity
12         The number of papers dealing with direct toxicity of lead on bone is limited. Most papers
13    are reviews (Hu et al., 1991;  Puzas, 2000; Puzas et al.,  1992; Rabinowitz,  1991; Silbergeld,
14    1991; Silbergeld et al., 1993; Vig and Hu,  2000) or based on cellular studies (e.g., Pounds
15    et al., 1991) or animals.
16         Various authors have suggested that lead is a potential risk factor for osteoporosis because
17    of the pivotal role of the skeleton in lead toxicokinetics (Goyer et al.,  1994). Bone cells
18    accumulate lead  actively and earlier ideas suggested that lead was incorporated into the mineral
19    matrix of the bone (Wittmers et al., 1988). However, in an in vivo iliac bone biopsy using laser
20    microbeam mass analysis on a lead-intoxicated adult female following chelation therapy, Flood
21    et al.  (1988) found the extracellular lead was concentrated in the superficial 3 to 6 jim of the
22    osteoid zone of bony trabeculae. As lead was absent from the deeper parts of the mineralized
23    matrix,  the authors suggested that lead binds more strongly to the organic  matrix than to bone
24    mineral.
25         There is increasing evidence from cell culture experiments, animal  studies, and from
26    measurements in humans that lead may exert detrimental effects on bone mineral metabolism.
27    In humans this evidence comes from several studies. Following on from the earlier observations
28    of Rosen et al. (1980)  that 1,25 (OHh vitamin D levels are reduced in lead poisoned children,
29    Markowitz et al. (1988) found that osteocalcin levels were inversely related to lead body burden
30    in moderately lead poisoned children. During chelation treatment for lead, the osteocalcin levels
31    were  shown to increase.

      December 2005                           6-311       DRAFT-DO NOT QUOTE OR CITE

-------
 1          An inverse relationship between blood lead and stature and chest circumference has been
 2    observed in children from the NHANES II study (Schwartz et al., 1986). There are several
 3    explanations for the inverse correlation between blood lead and growth in children. First, blood
 4    lead level may be a composite factor for genetic, ethnic, nutritional, environmental, and
 5    sociocultural factors.  Second, nutritional deficits that retard growth also enhance lead
 6    absorption.  Finally, there may be a direct effect of low level lead on growth in children. This
 7    condition was explained by Dowd et al. (1994) as resulting from the inhibition by Pb2+ of
 8    binding of osteocalcin to hydroxyapatite. Effects similar to those described by Schwartz et al.
 9    (1986) were reported by Angle and Kuntzelman (1989), Lauwers et al. (1986), and Shukla et al.
10    (1989).
11          Puzas et al. (1992) suggested lead could upset the very  sensitive interactive metabolic
12    activity of osteoblasts and chondrocytes and thereby affect bone growth. In a later review, Puzas
13    (2000) enlarged upon his earlier paper and described in more detail the potential mechanism of
14    lead on growth plate cartilage  metabolism and effects of lead on osteoclasts and osteoblasts,
15    especially associated with osteoporosis.
16          Observational  studies by Spencer et al. (1992, 1994) suggested a link between
17    occupational exposure to lead  and Paget's disease in both males and females but the authors
18    declined to advocate a causal effect.  Later Spencer et al. (1995) found that 92% of a group of
19    48 patients with Paget's disease were exposed to lead either from occupational or environmental
20    sources.  Adachi  et al. (1998) explored a possible association between lead and bone disease
21    from XRF analyses of cortical and trabecular bone lead content in 117 patients who attended a
22    metabolic bone disease  clinic (n = 92) or were undergoing dialysis for renal failure (n = 25).
23    In patients suffering from Paget's disease, cortical bone lead content was higher than it was in
24    controls,  patients with osteoporosis, and patients on dialysis.  Trabecular bone lead content was
25    lowest in patients with Paget's disease or osteitis fibrosa. However, the  authors could not
26    distinguish between two alternatives, the first being that increased bone  turnover due to Paget's
27    disease releases lead from trabecular bone that is then available for deposition into cortical bone,
28    or secondly, that  an increased  lead content in cortical bone may cause increased turnover with
29    release of lead from trabecular bone.
30          In another facet of the Normative Aging Study, Shadick et al. (2000) investigated a
31    possible association between long-term lead accumulation and hyperuricemia and gouty arthritis

      December 2005                            6-312        DRAFT-DO NOT QUOTE OR CITE

-------
 1    in 777 male subjects.  They found a positive association between patella bone lead and uric acid
 2    levels (p = 0.022) but no association between bone or blood lead and gout in this
 3    environmentally-exposed group.
 4
 5    6.9.7.3    Dental Health
 6          Caries is considered an infectious disease arising from a multifactorial process involving
 7    particular flora, dietary exposures, and a susceptible host (Schafer and Adair, 2000). Increased
 8    caries risk has been detected in association with increasing blood lead concentrations in
 9    populations whose mean blood lead concentrations are approximately 2-3 |ig/dL (Dye et al.,
10    2002; Gemmel et al., 2002; Moss et al., 1999).
11          Several studies have examined relationships between lead exposure and the occurrence of
12    dental caries in children and  adults. The two largest studies were analyses of data collected in
13    the NHANES III; both found significant associations between increasing caries prevalence and
14    increasing blood lead concentrations in children and adolescent (Moss et al., 1999) and the adult
15    (Dye et al., 2002) populations, whose geometric mean blood lead concentration was -2.5 |ig/dL.
16    In the Moss et al. (1999) study, the odds ratios for caries in association with a 5 |ig/dL increase
17    in blood lead concentration (i.e., from <2 |ig/dL) was 1.8 (95% CI: 1.3,2.5). Outcomes of two
18    smaller studies were mixed, with one study finding no significant association between blood lead
19    and caries prevalence (Campbell et al., 2000) and one study finding significant associations
20    (Gemmel et al., 2002); the latter, in children whose mean blood lead concentration was 2.9
21    Hg/dL (maximum 13 |ig/dL).
22          The Moss et al. (1999) NHANES III analysis included the results of coronal caries
23    examinations on 24,901 subjects, stratified by age: 2-5 years (n = 3,547), 6-11 years (n = 2,894),
24    and > 12 years (n =  18,460).  Specific outcomes assessed varied by age group: for children 2-11
25    years who had at least one deciduous tooth, the number of deciduous teeth displaying decayed or
26    filled surfaces (DPS); for subjects >6 years and who had at least one permanent tooth,  the
27    number of permanent teeth displaying decayed or filled surfaces; and for subjects > 12 years, the
28    sum of decayed, missing, and filled surfaces on permanent teeth (DMFS). In a multivariate
29    linear regression model, increasing blood lead concentration (log-transformed) was significantly
30    associated with covariate-adjusted increases in dfs in the 2-5 year age group (P = 1.78 [SE 0.59],
31    p = 0.004) and  in the 6-11 year age group (P =  1.42 [SE 0.51], p  = 0.007). Log-transformed

      December 2005                           6-313       DRAFT-DO NOT QUOTE OR CITE

-------
 1    blood lead also was associated with increases in DPS in the 6-11 years age group (P = 0.48
 2    [SE 0.22], p = 0.03) and in the > 12 years age group (P = 2.50 [SE 0.69], p<0.001), and increases
 3    in DMFS in the > 12 years age group (P = 5.48 [SE 1.44], p = 0.01).  The odds ratios (compared
 4    to 1 st tertile, < 1.66 |ig/dL) for the binomial outcome, 0 or > 1 DMFS, were 1.36 (95% CI: 1.01,
 5    2.83) for the blood lead concentration range 1.66-3.52 |ig/dL, and 1.66 (95% CI: 1.12, 2.48) for
 6    the range >3.52 |ig/dL. Corresponding population risks attributable to blood lead concentration
 7    were 9.6% and 13.5% in the blood lead strata, respectively. An increase in blood lead of
 8    5 |ig/dL was associated with an odds ratio of 1.8 (95% CI:  1.3, 2.5).  Covariates included in
 9    the models were age, gender, race/ethnicity, poverty income ratio, exposure to cigarette smoke,
10    geographic region, educational level of head of household, carbohydrate and calcium intakes,
11    and frequency of dental visits.
12          Gemmel et al. (2002) conducted a cross-sectional study of associations between blood
13    lead concentration and dental caries in children, 6-10 years of age (n = 543), who resided either
14    in an urban (n = 290) or rural (n = 253) setting. Mean blood lead concentrations were 2.9 |ig/dL
15    (SD 2.0, maximum 13 |ig/dL) in the urban group and 1.7  |ig/dL (SD 1.0, maximum 7 |ig/dL) in
16    the rural group. Increasing blood lead concentration (In-transformed) was significantly
17    associated with covariate-adjusted number of caries (dfs + DPS) (In-transformed) in the urban
18    group (P = 0.22 [SE 0.08], p = 0.005), but not in the rural group (P = -0.15 [SE 0.09], p = 0.09).
19    When dfs counts were stratified by permanent or deciduous teeth, the blood lead association in
20    the urban group was significant for deciduous teeth (P = 0.28 [SE 0.09], p = 0.002), but not for
21    permanent teeth (P = 0.02 [SE 0.07], p = 0.8). Covariates retained in the linear regression model
22    were age, sex, ethnicity, family income, education of female guardian, maternal smoking,
23    frequency of tooth brushing, firmness of toothbrush bristles, and frequency of chewing gum.
24          Campbell et al. (2000) was a retrospective cohort study in which dfs were assessed in
25    children 7-12 years of age (n = 248) from Rochester, NY.  Mean blood lead concentration,
26    measured at ages 18 and 37 months of age, was 10.7 |ig/dL (range 18.0-36.8 |ig/dL). The
27    covariate-adjusted odds ratios for caries associated with a blood lead concentration >10 |ig/dL
28    compared to < 10 |ig/dL were 0.95 |ig/dL (95% CI: 0.43, 2.09) for permanent teeth and
29    1.77 |ig/dL (95% CI: 0.97, 3.24) for deciduous teeth. Covariates retained in the logistic model
30    were age, grade in school, number of tooth surfaces at risk. Other covariates examined in the
31    models, all of which had no significant effect on the outcome, were gender,  race/ethnicity,  SES,

      December 2005                           6-314        DRAFT-DO NOT  QUOTE OR CITE

-------
 1    parental education, residence in community supplied with fluoridated drinking water, and
 2    various dental hygiene variables. This study did not demonstrate that lead exposure >10 |ig/dL
 3    as a toddler was a strong predictor of caries among school-age children, but the authors noted
 4    that this might be due to limited statistical power.
 5          Dye et al. (2002) analyzed data collected in NHANES III on indices of periodontal bone
 6    loss. The analysis was confined to subjects 20-69 years of age (n = 10,033).  The geometric
 7    mean blood lead concentration of the study group was 2.5 |ig/dL (SE 0.08), with 2.4% of the
 8    group having blood lead levels >10 |ig/dL.  Increasing log-transformed blood lead was
 9    significantly associated with increasing prevalence of covariate-adjusted dental furcation
10    (P = 0.13  [SE 0.05], p = 0.005). Dental furcation is indicative of severe periodontal disease.
11    Covariates retained in the linear regression model were age, sex, race/ethnicity, education,
12    smoking,  and age of home. Smoking status was a significant interaction term when included in
13    the model (P = 0.10 [SE 0.05], p=0.034).  When stratified by smoking status, the association
14    between dental furcation and blood lead concentration was significant for current smokers
15    (P = 0.21  [SE 0.07], p = 0.004) and former smokers (P = 0.17 [SE 0.07], p=0.015), but not for
16    nonsmokers (P = -0.02 [SE 0.07], p = 0.747).
17          Some studies examined the relationship between tooth lead concentrations and dental
18    caries.  In their compilation of metal concentrations in 1,200 deciduous teeth from a Norwegian
19    population,  Tvinnereim et al. (2000) found that carious teeth had higher lead concentrations than
20    noncarious teeth. Gil et al. (1994) measured lead concentrations from  220 whole deciduous and
21    permanent teeth from Coruna,  Spain. The geometric mean lead level was 10.36 jig/g of tooth.
22    There was a significant increase in teeth lead levels with advancing age. Permanent teeth
23    showed higher mean lead values (13.09 jig/g [SEM 1.07]) than deciduous teeth (3.96 jig/g
24    [SEM 1.07]). The authors reported a possible relationship between increased lead content and
25    periodontal  pathology but did not observe any relationship between caries and lead
26    concentrations.
27
28
      December 2005                           6-315       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.9.8   Effects of Lead on Ocular Health
 2    6.9.8.1   Summary of Key Findings of the Effects of Lead on Ocular Health from the
 3             1986 Lead AQCD
 4          The 1986 Lead AQCD did not address effects of lead on ocular health in humans.
 5    Various disturbances of the visual system have been observed in association with overt clinical
 6    lead poisoning, including retinal stippling and edema, cataracts, ocular muscle paralysis, and
 7    impaired vision (see Otto and Fox, 1993 for review). Two longitudinal studies completed since
 8    1986 provide evidence for possible associations between lead exposure and visual evoked retinal
 9    responses in children of mothers whose blood lead concentrations in mid-pregnancy were in the
10    range of 10-32 |ig/dL (Rothenberg et al., 2002), and evidence for a possible association between
11    lead exposure and risk of cataracts in middle-aged males whose tibia bone lead levels were in
12    the range 31-126 |ig/g (Schaumberg et al., 2004). These studies are summarized in Annex
13    Table AX6-9.11.
14
15    6.9.8.2   Ocular Effects
16          In the Mexico City prospective lead study, Rothenberg et al. (2002) measured
17    flash-evoked electroretinograms (ERG) in a subset of the study group (n = 45) at ages 7-10
18    years. As part of the prospective study, blood lead concentrations had been measured during
19    pregnancy and in the children, at birth and every 6 months, thereafter.  Increasing maternal blood
20    lead, measured at 12 weeks of gestation, was  significantly associated with increasing ERG a-
21    wave and b-wave amplitude, with significant  increases in a-wave in the second maternal blood
22    lead tertile (range 6.0-10.0 |ig/dL), and a-wave and b-wave in the third maternal blood lead
23    tertile (range 10.5-32.5 |ig/dL), compared to the first blood lead tertile (range 2.0-5.5 |ig/dL).
24    No other blood lead measurements were significantly associated with any ERG outcomes.
25          As part of the longitudinal Normative Aging Study, Schaumberg et al. (2004) analyzed
26    prevalence of cataracts in adult males (n = 642), mean age 69 years (range 60-93).  Subjects
27    were stratified by blood lead, patella bone lead, or tibia bone lead quintiles for a logistic
28    regression analysis of the odds ratios for cataracts (first quintile as reference). Covariate
29    adjusted odds ratio for cataracts in the fifth tibia bone lead quintile was significant (3.19 [95%
30    CI:  1.48, .90]).  Odds ratios for cataracts were not significantly associated with patella bone lead
31    (1.88 [95% CI: 0.88, 4.02]) or blood lead (0.89 [95% CI: 0.46, 1.72]). The first and fifth

      December 2005                            6-316       DRAFT-DO NOT QUOTE OR CITE

-------
 1    quintile lead levels were 0-11 jig/g and 31-126 jig/g for tibia bone; 1-16 jig/g and 43-165 jig/g
 2    for patella bone; and 1.0-3.0 jig/g and 8-35 jig/g for blood. Covariates retained in the
 3    regression model were age, smoking, history of diabetes; and daily intake of vitamin C, vitamin
 4    E, and carotenoids.
 5          Cavalleri et al. (1982) measured visual fields of male workers in a polyvinyl pipe
 6    manufacturing facility (n = 35) who were exposed to lead stearate. Workers in a reference group
 7    (n = 350) were individually matched for age, smoking, and alcohol consumption.  Visual
 8    sensitivity was significantly lower in lead workers compared to the reference group; however,
 9    visual sensitivity index was not significantly associated with blood or urine lead. Prevalence of
10    scotoma in the mesopic field was 28.5% in the lead workers and 0% in the reference group.
11    Mean blood lead levels were 46 |ig/dL (range 21-82 |ig/dL) in the lead workers and 30 |ig/dL
12    (range 21-42 |ig/dL) in the reference group.
13
14    6.9.9    Summary of the Epidemiologic Evidence for the  Effects of Lead
15             on Other Organ Systems
16    Biochemical Effects of Lead
17          Evidence for disruption of heme synthesis derives from numerous studies in which lead
18    exposure has been associated with decreased activities of enzymes in the heme synthesis
19    pathway (i.e., ALAS, ferrochelatase) and increased levels of substrates for heme synthesis (i.e.,
20    ALA, coproporphyrin, erythrocyte protoporphrin) in both children and adults. Quantitative
21    relationships between blood lead concentration and the above biomarkers of impaired heme
22    synthesis are highly consistent across studies (e.g., Alessio et al., 1977, 1976; Gennart et al.,
23    1992; Hernberg et al., 1970; Morita et al., 1997; Oishi et al., 1996; Piomelli  et al., 1982; Roels
24    and Lauwerys, 1987; Selander and Cramer, 1970; Sol din et al., 2003; Wildt  et al., 1987).
25    Increases in blood lead concentration of approximately 20-30 |ig/dL are sufficient to halve
26    erythrocyte ALAD activity and sufficiently inhibit ferrochelatase to double erythrocyte
27    protoporphyrin levels.
28          Associations between occupational exposure to lead and changes in blood lipid
29    composition have been observed. These include increased levels of lipid peroxides in blood
30    and/or serum (Jiun and Hsien, 1994; Sugawara et al., 1991; Ito et al., 1985) and increased serum
31    levels of total and HDL cholesterol (Kristal-Boneh et al., 1999). Effects on  serum cholesterol

      December 2005                           6-317       DRAFT-DO NOT QUOTE OR CITE

-------
 1    levels were evident in association with a mean blood lead concentration of 42 |ig/dL (Kristal-
 2    Boneh et al., 1999) or a range of 5-62 |ig/dL (approximated mean 14 |ig/dL) (Ito et al., 1985).
 3    Oxidative changes in blood lipids (e.g.,  increased levels of lipid peroxides and malondialdehyde
 4    levels) as well as decreased levels of erythrocyte superoxide dismutase, catalase, G6PD, and
 5    GSH peroxidase; and increased lymphocyte reactive oxygen species and depleted GSH levels,
 6    indicative of increased oxidative stress,  have been observed in lead workers in association with
 7    blood lead concentrations >30 |ig/dL (Fracasso et al., 2002; Ito et al., 1985; Jiun and Hsien,
 8    1994; Solliway et al., 1996; Sugawara et al.,  1991).
 9
10    Disruption of Hemoglobin Synthesis and Declines in Erythrocyte Numbers
11          Exposures that result in blood lead concentrations below 40  |ig/dL appear to be tolerated
12    without a decline in blood hemoglobin levels or hematocrit. However, perturbation of
13    erythropoiesis, indicated by changes in  serum erythropoietin and progenitor cells, occurs in
14    association with blood  lead concentrations below 40 |ig/dL and in the absence of detectable
15    changes in blood hemoglobin levels or hematocrit in children  (Graziano et al., 2004; Liebelt
16    et al., 1999) and adults (Graziano et al., 1990; Osterode et al.,  1999; Romeo et al., 1996). Risk of
17    clinical anemia in children becomes appreciable at much higher blood lead concentrations; a
18    10% decrease in hematocrit has been estimated to occur in association with blood lead
19    concentrations >85 |ig/dL; a 10% probability of anemia (hematocrit <35%) was estimated to be
20    associated with a blood lead concentration of approximately 20 |ig/dL at age 1 year, 50 |ig/dL at
21    age 3 years, and 75 |ig/dL at age 5 years.  (Schwartz et al., 1990). In adults, with blood lead
22    levels below 25 |ig/dL, increasing patella bone lead, but not blood lead, was associated with a
23    significant decrease in hematocrit.
24
25    Effects on the Endocrine System
26          Several studies have examined possible associations between lead exposures in children
27    and adults and various biomarkers of endocrine function, including the thyroid, male
28    reproductive, and calcitropic endocrine  systems. The strongest study designs have yielded no
29    associations, or weak associations, between lead exposure and thyroid hormone status (Erfurth
30    et al., 2001; Schumacher et al., 1998; Tuppurainen et al., 1988; Zheng et al., 2001). Studies of
31    occupational exposures which included subjects having blood lead  concentrations exceeding

      December 2005                           6-318       DRAFT-DO NOT QUOTE OR CITE

-------
 1    100 |ig/dL have found depression of serum T3 and/or T4 levels, without a detectable increase in
 2    serum TSH; however, studies in which the blood lead distribution was dominated by levels well
 3    below 100 |ig/dL, have found either no effects or subclinical increases in serum T3, T4, with no
 4    change in TSH levels.
 5          Studies of the male reproductive system that attempted to control for confounding effects
 6    of age have yielded mixed outcomes (Alexander et al., 1998, 1996; Erfurth et al., 2001;
 7    Gustafson et al., 1989; McGregor and Mason, 1990; Ng et al., 1991). Blood lead ranges in these
 8    studies were similar (4-90 g/dL), yet outcomes were mixed, with no change (Erfurth et al., 2001;
 9    Gustafson et al., 1989; McGregor and Mason, 1990), or subclinical decrease (Alexander et al.,
10    1998,  1996; Ng et al., 1991) in serum  testosterone (TES) in association with lead exposure.
11    There are also mixed effects on serum follicle stimulating hormone (FSH) and luteinizing
12    hormone (LH) with increases (McGregor and Mason, 1990; Ng et al., 1991), decreases
13    (Gustafson et al., 1989), and with no change (Alexander et al., 1998, 1996; Erfurth et al., 2001)
14    in hormone levels observed.  The inconsistency  in the direction of effects on TES and the two
15    androgen-regulating pituitary hormones, FSH and LH, is particularly noteworthy, in the absence
16    of evidence for effects of lead exposure on GNRH-induced FSH (Erfurth et al. 2001).
17          Children exposed to relatively a high level of lead >30 |ig/dL may exhibit depressed
18    levels of circulating 1,25-OH-D (Mahaffey et al., 1982; Rosen et al., 1980). However,
19    associations between serum vitamin D status and blood lead may not be present in calcium-
20    replete children who have average lifetime blood lead concentrations below 25 |ig/dL (Koo
21    et al.,  1991). In adults, exposures to lead that result in blood lead concentrations >40-60 |ig/dL
22    may increase, rather than decrease, circulating levels of 1,25-OH-D and PTH (Kristal-Boneh
23    et al.,  1999;  Mason et al., 1990).
24
25    Effects on the Hepatic System
26          Few studies of hepatic effects of lead on humans have been reported since the 1986 Lead
27    AQCD.  Studies of hepatic enzyme levels in serum suggest that liver injury may be present in
28    lead workers; however, associations specifically with lead exposures are not evident (Al-Neamy
29    et al., 2001;  Hsiao et al., 2001).  Studies of urinary metabolites of cytochrome P450 phenotypes
30    CYP2A6 and CYP3A4 suggest possible associations between lead exposure and suppression of
31    hepatic enzyme activity.  The effect on CYP2A6 activity was observed in children with high lead

      December 2005                          6-319       DRAFT-DO NOT QUOTE OR CITE

-------
 1    burdens (i.e., blood lead concentration >40 |ig/dL, EDTA-provoked urinary lead >500 |ig/dL).
 2    The effect on CYP3 A4 was observed in association with blood lead ranges of approximately
 3    30-112 |ig/dL (based on reported serum lead concentrations).
 4
 5    Effects on the Gastrointestinal System
 6          Several studies of prevalence of symptoms of gastrointestinal colic in lead workers
 7    provide evidence for symptoms in association with blood lead concentrations >50-80 |ig/dL
 8    (Awad el Karim et al., 1986; Holness and Nethercott, 1988; Lee et al., 2000; Matte et al., 1989).
 9    Similar types of studies of children have not been reported.
10
11    Effect on Bone and Teeth
12          There is limited, but suggestive evidence of an association between lead exposure and
13    bone toxicity. However, in most studies, it is difficult to assess the direct contribution of lead on
14    bone diseases or reduced growth.  Several studies that have explored possible associations
15    between lead exposure and risk of dental caries (Campbell et al., 2000; Dye et al., 2002; Gemmel
16    et al., 2002; Moss et al., 1999). Increased caries risk has been detected in association with
17    increasing blood lead concentrations in populations whose mean blood lead concentrations are
18    approximately 2-3 |ig/dL (Dye et al., 2002; Gemmel et al., 2002; Moss et al., 1999).
19
20    Ocular Health
21          Various disturbances of the visual system have been observed in association with overt
22    clinical lead poisoning, including retinal stippling and edema, cataracts, ocular muscle paralysis,
23    and impaired vision (Otto and Fox, 1993).  Two longitudinal studies completed  since the 1986
24    Lead AQCD provide evidence for possible associations (a) between lead exposure and visual
25    evoked retinal responses in children of mothers whose blood lead concentrations in mid-
26    pregnancy was 10.5-32.5 |ig/dL (Rothenberg et al., 2002) and (b) between lead  exposure and
27    risk of cataracts in middle-aged males whose tibia bone lead levels were 31-126 jig/g
28    (Schaumberg et al., 2004).
29
30
      December 2005                           6-320       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.10  INTERPRETIVE ASSESSMENT OF THE EVIDENCE IN
 2          EPIDEMIOLOGIC STUDIES OF LEAD HEALTH EFFECTS
 3    6.10.1   Introduction
 4          A remarkable expansion has occurred since the 1990 Lead Supplement in the extent of the
 5    database available for drawing inferences about the various expressions of lead toxicity.
 6    Moreover, the nature of the evidence available has changed as well.  Many of the studies
 7    conducted prior to 1990 focused on the issue of whether an observed observation was likely to be
 8    real or the result of chance, selection bias, residual confounding, or some other methodological
 9    error.  The validity of any association still needs to be assured. The studies since 1990 mainly
10    focus on characteristics of the pertinent concentration-response relationships, including the
11    functional forms of the relationships, the slopes of the relationships, the natural histories of
12    adverse effects, and the effect modifying influences of various co-exposures and host
13    characteristics.
14
15    6.10.2   Exposure and Outcome Assessment in Lead Epidemiologic Studies
16    6.10.2.1   Assessment of Lead Exposure and Body Burdens  Using Biomarkers
17          For any health endpoint of interest, the most useful biomarker of exposure is one that
18    provides information about the lead dose at the critical target organ and, moreover, reflects the
19    exposure averaging time that is appropriate to the underlying pathogenetic processes (e.g.,
20    cumulative over lifetime, cumulative over a circumscribed age range, concurrent, etc.). In recent
21    studies of lead and health, the exposure biomarkers most frequently used are blood lead and bone
22    lead.  For outcomes other than those relating to hematopoeisis and bone health, these biomarkers
23    provide information about lead dose that is some distance  from the target organ.  For example,
24    given that the central nervous system is considered the critical target organ for childhood lead
25    toxicity,  it would be most helpful to be able to measure, in vivo, the concentration of lead at the
26    cellular site(s) of action in the brain. Because such measurements are not currently feasible,
27    however, investigators must rely on measurements of lead in the more readily accessible but
28    peripheral tissues. The relationship between brain lead and lead in each of these surrogate
29    tissues is still poorly understood, although the pharmacokinetics clearly differs among these
30    compartments.  In both rodents and nonhuman primates, brain lead level falls much more slowly
31    than blood lead level following chelation with succimer and, in the rodent, in nonchelated

      December 2005                          6-321        DRAFT-DO NOT QUOTE OR CITE

-------
 1    animals after cessation of exposure. These observations suggest that using blood lead as an
 2    index of lead in the brain will result in exposure misclassification, although the magnitude of this
 3    bias in any specific setting will be difficult to characterize.  The most likely direction, however,
 4    would be underestimation of the amount of lead in the brain, at least under scenarios involving
 5    chronic exposure.
 6          As an exposure biomarker, blood lead level has other limitations.  Only about 5% of an
 7    individual's total body lead burden resides in blood.  Furthermore, blood consists of several sub-
 8    compartments. More than 90% of lead in whole blood is bound to red cell proteins such as
 9    hemoglobin, with the balance in plasma. From a toxicological perspective, this unbound fraction
10    is likely to be the most important sub-compartment of blood lead because of the ease with which
11    it diffuses into soft tissues.  The concentration of lead in plasma is much lower than in whole
12    blood, however. For example, in a group of pregnant women with blood lead levels below
13    10 |ig/dL, plasma lead levels were less than 0.3% of the whole blood lead level.  The greater
14    relative abundance of lead in whole blood makes its measurement much easier (and more
15    affordable) than the measurement of lead in plasma.  The use  of whole blood lead as a surrogate
16    for plasma lead could be justified if the ratio of whole blood lead to plasma lead were well
17    characterized, but this is not so. At least some studies suggest that it varies several-fold among
18    individuals with the same blood lead level. Moreover, the ability of red cells to bind lead is
19    limited, so the ratio of blood lead to plasma lead would be expected to be nonlinear. Thus,
20    interpreting whole blood lead level as a proxy for plasma lead level, which, itself, is a proxy for
21    brain lead level, will result in some exposure misclassification.
22          Another limitation in the use of blood lead as the exposure biomarker is that its residence
23    time in blood  is closely linked to red cell lifetime, with a half-time on the order of 30 days.
24    Thus, a high blood lead level does not necessarily indicate a high body lead burden. Similarly,
25    individuals who have the same blood lead level will not necessarily have similar body burdens or
26    exposure histories.  The rate at which blood lead level changes with time/age depends on
27    exposure history due to re-equilibration of lead stored in the various body pools.  In nonchelated
28    children, the time for blood lead to decline to a value less than 10 |ig/dL was linearly related to
29    baseline blood lead level. A single blood lead measurement might therefore provide limited
30    information about  an individual's lead exposure history, a  difficulty frequently cited with respect
31    to the interpretation of cross-sectional studies of pediatric lead toxicity, in which children's blood

      December 2005                           6-322        DRAFT-DO NOT QUOTE OR CITE

-------
 1    lead level is often measured only once, and sometimes only well after the period when levels
 2    typically peak (18-30 months).  If it is exposures to lead in the early postnatal years that are most
 3    detrimental to children's development, categorizing a child's exposure status based on the blood
 4    lead level that is contemporaneous with the measurement of neurodevelopment at school-age
 5    could result in exposure misclassification.  Unless intra-individual stability of serial blood lead
 6    levels is very high within a study cohort, misclassification would probably be non-differential,
 7    more likely resulting in an underestimate rather than an overestimate of the effect of lead on
 8    child neurodevelopment (Jurek et al., 2005).  This concern must be qualified, however, by recent
 9    data from some longitudinal studies indicating that concurrent blood lead level, even at ages well
10    beyond 18 to 30 months, is sometimes the strongest predictor of late outcomes (Dietrich et al.,
11    1993a,b; Canfield et al., 2003a; long et al., 1996; Wasserman et al., 2000b).  Age-related
12    changes in vulnerability, and the reasons why it might differ across studies, remain uncertain.
13    It might be that among children with chronically elevated exposure, but not in children with
14    relatively low lifetime exposure, blood lead level measured at school-age is a reasonably good
15    marker of cumulative exposure. That  concurrent blood lead level is, under some circumstances,
16    a stronger predictor of school-age outcomes than is blood lead level in the early postnatal years
17    does not necessarily imply greater vulnerability of the brain to ongoing than to past exposure.
18          The development of X-ray-fluorescence (XRF) methods  for measuring lead in
19    mineralized tissues offers another approach for characterization and reconstruction of exposure
20    history.  Such tissues are long-term lead storage sites, with a half-life measured in decades and
21    contain approximately 90% of the total body  lead burden in adults and 70% in children. Thus,
22    bone lead is an index with a long exposure averaging time.  XRF methods have proven useful in
23    studying individuals with occupational lead exposure, those living in highly polluted
24    environments, and those for whom community lead exposures are or, in the past, were relatively
25    high (e.g., Korrick et al., 1999; Schwartz et al., 2000a,b,c,d).  In a relatively highly exposed
26    cohort of pregnant women in Mexico City, higher bone lead levels  at one month postpartum
27    were associated with reduced birth weight, less infant weight gain,  smaller head circumference
28    and birth length, and slower infant development (Gomaa et al., 2002; Gonzalez-Cossio et al.,
29    1997; Hernandez-Avila et al., 2002; Sanin et al., 2001).  Among children living near a large lead
30    smelter in Yugoslavia, IQ at age 10-12 years  was more strongly associated, inversely, with tibia
31    lead level than with blood lead level (Wasserman et al., 2003).

      December 2005                            6-323       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Current XRF methods for measuring bone lead levels have limitations, however.
 2    Temporal features of exposure history cannot readily be discerned.  Some progress has been
 3    made toward this goal by examining the spatial distribution of lead in teeth in relation to the
 4    relative abundance of stable lead isotopes, but the specialized technologies needed to carry out
 5    these analyses are unlikely ever to be widely available, and the unpredictability of tooth
 6    exfoliation makes this tissue difficult to collect unless the study design involves contact with
 7    (and the cooperation of) participants at the appropriate ages. Current XRF methods might not be
 8    sufficiently sensitive for studies of the health effects of low-dose community exposures.  The
 9    bone lead levels of a large percentage of subjects might be below the detection limit,  e.g., 80% in
10    a case-control study of bone lead levels and juvenile delinquency in which the minimum
11    detection limit was 21.5 |ig/g bone mineral (Needleman et al., 2002). Even among individuals
12    known to have histories of substantial lead exposures, such as adolescents and young adults who
13    grew up near the Bunker Hill smelter in Idaho (McNeill et al., 2000), bone lead levels tend to be
14    low. Lead appears to be deposited at sites of most active calcification. In children, this is
15    trabecular bone, in which the rate of fractional resorption in early childhood is high.  Depending
16    on the amount of the child's ongoing exposure, lead deposited in bone might not remain there for
17    decades, making bone lead level an imprecise index of lifetime lead exposure. This concern also
18    exists in the use of tooth lead to represent cumulative lead  exposure in children.  Rabinowitz
19    et al. (1993) observed that a child's tooth lead level  was more strongly related to blood lead level
20    around the time  of tooth exfoliation than to an integrated index of blood lead level prior to
21    exfoliation. Finally, it is difficult to compare the performance of different laboratories using
22    XRF methods to measure bone lead because of the absence of standard reference materials.
23    Nevertheless, efforts continue to modify the instrumentation or measurement protocols to reduce
24    the detection limit.
25          A major research need is the development and validation of biomarkers of critical dose
26    that, compared to blood lead or bone lead, are fewer toxicokinetic steps removed from the sites
27    of lead's actions in the brain.  One promising front in the effort to deduce the contents of the
28    "black box" separating external dose and clinical disease is the measurement of processes and
29    products that potentially mediate the association between them.  For example, magnetic
30    resonance spectroscopy (MRS) has been used in small case series to measure the ratio of
31    N-acetylaspartate (NAA) to creatine, which are a marker of neuronal and axonal damage and

      December 2005                           6-324        DRAFT-DO  NOT QUOTE OR CITE

-------
 1    thus, an early biological effect rather than a biomarker of exposure.  In children, higher lead
 2    exposures are associated with lower NAA to creatine ratios in the frontal gray matter and, to a
 3    lesser extent, in frontal white matter (Trope et al., 1998, 2001).  Similarly, an adult who had
 4    higher bone and blood lead levels than did his monozygotic twin had both greater
 5    neuropsychological deficits and lower NAA to creatine ratios in the hippocampus, frontal lobe,
 6    and midbrain (Weisskopf et al., 2004).  While much remains uncertain about the interpretation of
 7    MRS, the use of this and other biochemical imaging methods, in combination with more
 8    conventional structural and functional imaging methods, might bring us closer to understanding
 9    the mechanisms of lead neurotoxicity. With the number of toxicokinetic steps separating lead
10    levels at the critical target organs from the  usual exposure biomarkers, the progress made in
11    characterizing the concentration-response relationships is remarkable.
12
13    6.10.2.2   Assessment of Health Outcomes
14          Outcome measurement and outcome classification have generally received less attention
15    from investigators than have exposure measurement and misclassification. The specific
16    problems are, to some extent, endpoint domain-specific. With regard to neurodevelopmental
17    toxicities, critical issues are whether the assessment instruments used are psychometrically  sound
18    and appropriate for the study cohort, the data generated will support adequate tests of the study
19    hypotheses, and whether the instruments have been administered and scored consistently and
20    correctly. With regard to the cardiovascular toxi cities of increased blood pressure/prevalence of
21    hypertension, the critical issue is whether the blood pressure value recorded for a participant is
22    an accurate estimate.  Multiple measurements of blood pressure are frequently made in a study
23    but investigators usually have not taken advantage of the collected information to quantify the
24    amount of error in the measurements. This information  can be used to improve the reliability of
25    the measurements, which would be expected to improve the precision of the associations
26    estimated. Similarly, aggregating scores to estimate latent variables representing, for instance,
27    "language skills" or "visual-spatial skills" is an approach that might take advantage of the
28    overlapping information provided by the multiple tests included in neurobehavioral test batteries,
29    producing more reliable endpoint variables. This approach, however, has not been widely
30    applied in lead studies.
31

      December 2005                           6-325       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.10.3    Concentration-Response Relationship of Lead Health Effects
 2          Recent studies have not altered the consensus that the developing nervous system is the
 3    organ system that is most sensitive to lead toxicity in children and adults. Neurobehavioral
 4    deficits appear to occur at lower levels of exposure than do other adverse health effects, although
 5    adverse effects in other organ systems have been observed in some susceptible populations at
 6    similarly low levels (e.g., adverse renal outcomes in individuals with hypertension or chronic
 7    renal insufficiency).  Effects have been reported at blood lead levels as low as 1 to 2 |ig/dL in the
 8    case of neurobehavioral toxicity. Accumulating data appear to validate well the statement made
 9    in the 1996 AQCD and Addendum, and 1990 Supplement that adverse effects occur at blood
10    lead levels of 10 to 15  |ig/dL or "possibly lower."  In a recent study of 6 to  16 year old children
11    in the NHANES III survey, concentration-related deficits in reading and arithmetic scores were
12    found even when analyses were restricted to children with concurrent blood lead levels below
13    5 |ig/dL (Lanphear et al., 2000).
14          Canfield et al. (2003a) applied semi-parametric models with penalized splines to their
15    data, essentially allowing the data to reveal the functional form that best described them. These
16    analyses showed that the IQ decline per |ig/dL increase in blood lead was greater below
17    10 |ig/dL than it was above 10 |ig/dL.  The estimated slope of the IQ decline per |ig/dL was
18    greatest among children for whom the maximum blood lead level measured over the course of
19    the study never exceeded 10 |ig/dL.  A similarly steeper slope at lower than at higher blood lead
20    levels was found in a re-analysis of the Boston prospective study (Bellinger and Needleman,
21    2003).
22          Identifying the functional  form that best fits a particular set of data and that presumably
23    serves as the best description of the pertinent underlying concentration-response relationship is
24    clearly important.  The linear model (Figure 6-10.1) is, as the name implies, linear over the entire
25    range of the exposure data. For certain tests, the assumption is made that the residuals (observed
26    - predicted response) are normally distributed with constant variance, but violations of this
27    assumption (heteroscedasticity) have no real effect on the estimation and minimal effect  on the
28    tests of significance (see Annex Section AX6.10).  If heteroscedasticity is present but all other
29    conditions are met, regression still yields unbiased estimators, but the standard errors can be
30    larger than when remedial efforts such as using weighted regression are employed. The use of
31    regression requires no  assumption concerning the distribution of the independent variable

      December 2005                            6-326       DRAFT-DO NOT QUOTE OR CITE

-------
                  105
                  100
               V)
               c
               o
               Q.
               
               a:
                   90
                   85
                             5        10      15       20       25      30
                                              Exposure
                           — Linear Model         ---Log-linear Model
                                   35
     Figure 6-10.1.   Comparison of a linear and log-linear model to describe the relationship
                     between exposure and response.
 1   (lead exposure marker).  However, when the form of the heteroscedasticity is an increase in
 2   variance with level of blood lead and when the data are lognormally distributed or otherwise
 3   skewed, there are possibly a large number of influential data points at high blood lead where the
 4   data is least reliable.  In this case, a log transformation of blood leads may result in more precise
 5   estimation of the slope parameter.  The presence of heteroscedasticity and other departures from
 6   assumptions forming the basis for regression analysis can be detected by using diagnostic tests or
 7   graphics. These are rarely used in epidemiologic studies of lead health effects.
 8          The log-linear model (see Figure 6-10.1) is written as:

 9                           Response = a + P Ln(lead exposure marker),

10   where  Ln is the natural logarithmic function. The log-linear model is concave upwards
11   (assuming that the estimated coefficient is negative).  It approaches a linear function for very
12   high exposure values, but approaches infinity at very low exposure values. In other words, it is
13   assumed that the adverse effect of lead is greater at lower than at higher blood lead levels.
     December 2005
6-327
DRAFT-DO NOT QUOTE OR CITE

-------
 1    Blood lead levels have been shown repeatedly to follow a lognormal distribution (Azar et al.,
 2    1975; Billick et al., 1979; Hasselblad and Nelson, 1975; Yankel et al., 1977; Hasselblad et al.,
 3    1980; U.S. Environmental Protection Agency, 1986), but this fact is not an argument for
 4    choosing the log-linear model.  The choice of either log-linear or linear may be based on the
 5    Akaike's Information Criteria (Akaike, 1973), J-test (Davidson and MacKinnon, 1981), or other
 6    statistical tests if the choice is to be based on the best fitting model. Rothenberg and Rothenberg
 7    (2005) compared the linear lead model with the log-linear lead model for the pooled data from
 8    Lanphear et al. (2005) using the J-test.  The J-test showed that the log lead specification was still
 9    significant (p = 0.009) in a model that also included the linear lead specification, indicating that
10    the log lead specification described the data significantly better than did the linear lead
11    specification. Other models have been used, such as nonparametric models, spline functions,
12    and polynomial models, but the vast majority of the analyses have used either a linear model or a
13    log-linear model.
14          Nonlinear concentration-response relationships are not uncommon in toxicology, although
15    many of these are claimed to be examples of hormesis, with the lowest doses of a toxicant being
16    associated with a beneficial effect rather than a greater adverse effect. A biological mechanism
17    for a steeper slope at lower than at higher blood lead levels has not been identified.  Perhaps the
18    predominant mechanism at very low blood lead levels is rapidly saturated, and that a different,
19    less rapidly saturated process becomes predominant at blood lead levels greater than 10 |ig/dL.
20    This ad hoc explanation is more descriptive than explanatory, however, and the specific
21    processes that would produce this result have not yet been identified.  Nevertheless, relationships
22    of this apparent form have been observed in several data sets, indicating the need to determine
23    whether such a relationship is real or a statistical artifact.
24          An important caveat regarding efforts to  specify the functional form of the concentration-
25    response relationship is that the accuracy that can be achieved is constrained by the extent to
26    which the biomarker of lead concentration does, in fact, reflect the concentration at the critical
27    target organ, the brain. The greater the misclassification, the more uncertain will be the
28    biological relevance of the best statistical description of the concentration-response relationship.
29
      December 2005                            6-328       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.10.4    Interindividual Variability in Susceptibility to Lead Toxicity
 2          Although increased lead exposure has been linked to adverse health effects in many
 3    different organ systems, scatterplots reveal tremendous variability of observed points about the
 4    best fit lines representing the concentration-response relationships. In other words, individuals
 5    for whom the lead biomarker measured has the same value can have markedly different values
 6    on the health indicator measured. Even for neurobehavioral deficits in children, the correlation
 7    between biomarker level and test score rarely exceeds -0.2, indicating that the explained
 8    variance in the test score generally does not exceed 5%. A major challenge is therefore to
 9    decompose this variability, to distinguish components of it that reflect error from components
10    that reflect biological processes that determine an individual's response to lead.
11          Deviation of the observed points from the fitted point can have many sources. Exposure
12    misclassification is one source. The lead biomarker measured might not adequately capture the
13    lead dose delivered to the target organ and at the time that is most appropriate biologically.
14    In general, the error would be expected to be non-differential, i.e., it would not introduce a
15    systematic bias in estimation of the concentration-response relationship. On average, such
16    misclassification would be expected to result both in an attenuation of the  slope of the
17    concentration-response relationship and an increase in the scatter of the observations. As focus
18    shifts to the risks associated with lower and lower levels of lead exposure, the importance of
19    errors introduced by poor dosimetry will assume greater importance insofar as the effects at such
20    levels will presumably be more subtle and increasingly difficult to detect amid the noise
21    contributed by exposure misclassification. Outcome misclassification is another source of error
22    that is likely to contribute to apparent interindividual variability in response.  This results if the
23    indicator of the critical health effect that is measured is fallible, i.e., an imperfect measure of the
24    target function.  Such misclassification would generally be expected to be non-differential,
25    introducing random noise rather than a systematic bias.
26          Another likely source of scatter  in observed points is true interindividual variability in
27    response to a given lead dose. That is,  the magnitude of individual response to lead might
28    depend on other characteristics of that individual. Three major categories of such effect
29    modifying factors that might influence  susceptibility to lead toxicity are genetic polymorphisms,
30    nutritional status, and social environmental factors.  Adequate data are not available to  provide a
31    quantitative estimate of the amount of interindividual variability in susceptibility to lead.

      December 2005                           6-329       DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.10.4.1  Influence of Genetic Polymorphisms on Risk
 2          Genetic polymorphisms that are presumed to influence lead toxicokinetics and/or
 3    toxicodynamics have been identified, mostly in studies of adults who were occupationally
 4    exposed to lead. Compared to workers with the wild type allele of amino levulinic acid
 5    dehydratase, workers with the variant allele had a higher mean blood lead level, greater lead-
 6    associated renal dysfunction,  and an increased risk of amyotrophic lateral sclerosis (Kamel et al.,
 7    2003).  Lead workers with the ATP1 A2(3') polymorphism appear to be at increased risk of lead-
 8    associated effects on blood pressure (Glenn et al., 2001). The slope of the association between
 9    floor dust lead and blood lead is steeper among children with the less common variant of the
10    vitamin D receptor (Fox 1 or B) than among children with the wild-type allele (Haynes et al.,
11    2003).  In adults, these same alleles are associated with higher blood lead levels and increased
12    blood pressure (Schwartz et al., 2000c; Lee et al., 2001). Greater lead-associated reductions in
13    renal function have been observed in adults with a variant allele of nitric acid synthetase,
14    although cardiovascular outcomes, such as blood pressure and hypertension do not appear to
15    depend on eNOS (endogenous nitric oxide synthase) allele (Weaver et al., 2003). Adults with
16    variants of the hemochromatosis gene (C282Y and/or H63D) have higher patella lead levels
17    (Wright et al., 2004). Only one polymorphism has been shown to modify lead neurotoxicity.
18    Lead workers with the apolipoprotein E4 allele showed greater lead-associated decreases in
19    neurobehavioral function than did workers with the El, E2, or E3 alleles (Stewart et al., 2002).
20    This work is in its early stages, and while it promises to shed light on bases of susceptibility to
21    lead toxicity, firm conclusions cannot yet be drawn.
22
23    6.10.4.2  Influence of Nutritional Status on Risk
24          Only limited epidemiologic data are available on the role of nutritional status in
25    modifying an  individual's risk of lead toxicity. Adjusting for severity of environmental lead
26    contamination, iron-deficient children appear to have higher blood lead levels than iron-replete
27    children (Bradman et al., 2001). One interpretation of these data is that children experiencing the
28    same external lead dose can experience different internal doses. In another study of iron status, a
29    decline in blood lead level was associated with improved cognitive performance in iron-
30    sufficient but not in iron-deficient children (Ruff et al., 1996).  Among the possible explanations
31    for this finding is that iron deficiency contributes to pharmacodynamic variability, increasing the

      December 2005                           6-330       DRAFT-DO NOT QUOTE OR CITE

-------
 1    toxicity of a given lead dose. Some evidence suggests that the intellectual deficit associated with
 2    an elevated blood lead level is greater among undernourished children than well-nourished
 3    children (Gardner et al., 1998).
 4          Several studies have suggested that dietary calcium may have a protective role by
 5    decreasing absorption of lead in the gastrointestinal tract and decreasing the mobilization of lead
 6    from bone stores to blood, especially during periods of high metabolic activity of the bone such
 7    as pregnancy and lactation.  Lower calcium intake during pregnancy, especially the second half,
 8    appears to increase the mobilization of lead from bone compartments (Hernandez-Avila et al.,
 9    1996). However, in other studies, calcium supplementation had no effect on bone lead levels
10    pregnant and lactating women (Rothenberg et al., 2000; Tellez-Rojo et al., 2002).
11
12    6.10.4.3   Influence of Health Status on Risk
13          The influence of an individual's health status on susceptibility to lead toxicity has been
14    demonstrated most clearly for renal outcomes. Individuals with diabetes, hypertension, and
15    chronic renal insufficiency are at increased risk of lead-associated declines in renal function and
16    adverse effects have been demonstrated at blood  lead levels below 5 |ig/dL (Lin et al., 2001,
17    2003; Muntner et al., 2003; Tsaih et al., 2004). As discussed in an earlier section, children with
18    nutritional deficiencies also appear to be more vulnerable to lead-associated neurobehavioral
19    deficits.
20
21    6.10.4.4   Influence of Co-Exposures on Risk
22          Epidemiologic studies do not provide an adequate basis for determining whether cigarette
23    smoking and/or alcohol affect the nature or severity of the health effects associated with lead
24    exposure.  Both factors have often been included in models of both child and adult health
25    outcomes in order to adjust for potential confounding. In addition, both have been evaluated as
26    pertinent pathways of adult exposure. However,  their possible roles as effect modifiers have not
27    been well studied.
28          Although most individuals are not exposed to lead in isolation but to lead in combination
29    with other toxicants  including cadmium, arsenic, mercury, and polychlorinated biphenyls,
30    epidemiologic studies generally have focused solely on lead. Other toxicant exposures have
31    sometimes been measured but are usually treated as potential confounders in the statistical

      December 2005                            6-331        DRAFT-DO NOT QUOTE OR CITE

-------
 1    analyses, with their status as potential modifiers of lead toxicity left unexplored (Bellinger,
 2    2000).  As a result, epidemiologic studies do not provide an adequate basis for determining
 3    whether co-exposure to other toxicants affects the nature or severity of the health effects
 4    associated with lead exposure.
 5
 6    6.10.4.5   Influence of Timing of Exposure on Risk
 7    6.10.4.5.1    Children
 8          Studies do not provide a definitive answer to the question of whether lead-associated
 9    neurodevelopmental deficits are the result of exposure during a circumscribed critical period or
10    of cumulative exposure.  Although support can be cited for the conclusion that it is exposure
11    within the  first few postnatal years that is most important in determining long-term outcomes
12    (Bellinger et al., 1992), other studies suggest that concurrent blood lead level is as predictive,
13    and perhaps more predictive,  of long-term outcomes than are early blood lead levels (Canfield
14    et al., 2003a; Dietrich et al., 1993a,b; long et al., 1996; Wasserman et al., 2000b). Because of
15    the complex kinetics of lead, an accumulative toxicant, it is extremely difficult to draw strong
16    conclusions from these observational studies about windows of heightened vulnerability in
17    children. The high degree of intraindividual "tracking" of blood lead levels over time, especially
18    among children in environments providing substantial, chronic exposure opportunities (e.g.,
19    residence near a smelter or in older urban dwellings in poor repair), poses formidable obstacles
20    to identifying the time interval during which exposure to lead caused the health effects measured
21    in a study.  It could be that damage occurred during a circumscribed period when the critical
22    substrate was undergoing rapid development, but that the high correlation between serial blood
23    lead levels impeded identification of the special significance of exposure at that time. Under
24    such circumstances, an index of cumulative blood lead level or concurrent blood lead level,
25    which might be a good marker of overall body burden under conditions of relatively steady-state
26    exposure, might bear the strongest association with the adverse effect.
27
28    6.10.4.5.2    Aging Population
29          Increases in blood lead for postmenopausal women have been attributed to release of lead
30    from the skeleton associated with increased bone remodeling during menopause in both
31    occupationally- and environmentally-exposed women (Garrido-Latorre et al., 2003; Popovic
32    et al., 2005). In middle-aged to elderly males from the Normative Aging Study, patella lead

      December 2005                            6-332        DRAFT-DO NOT QUOTE OR CITE

-------
 1    accounted for the dominant portion of variance in blood lead (Hu et al., 1996). These findings
 2    provide evidence that the skeleton may serve as a potential endogenous source of lead in the
 3    aging population.
 4          Considerable evidence also suggests that indicators of cumulative or long-term lead
 5    exposure are associated with adverse effects in several organ systems, including the central
 6    nervous, renal, and cardiovascular systems. Among occupationally-exposed men, higher tibia
 7    lead levels have been associated with increased cognitive decline over repeated assessments
 8    (Schwartz et al., 2005). With regard to the renal system, increased lead exposure may accelerate
 9    the effects of normal aging, producing a steeper age-related decline in function. Weaver et al.
10    (2003)  observed that higher lead exposure and dose were associated with worse renal function in
11    older workers, but with lower blood urea nitrogen and serum creatinine in young workers.
12
13    6.10.4.5.3   Pregnancy
14          Potential mobilization of lead from the skeleton also occurs during pregnancy and
15    lactation due to increased bone remodeling (Hertz-Picciotto et al., 2000; Manton, 1985;
16    Silbergeld, 1991). In women who have been exposed to lead in childhood and have accumulated
17    large stores in their bones, there may be significant mobilization of lead from bone to blood
18    during late pregnancy and lactation. The greatest probability of lead toxicity for the mothers will
19    be in postpartum while they are lactating; the infants will be particularly vulnerable during the
20    prenatal period, especially in the last weeks of pregnancy (Manton et al., 2003).
21          A variety of adverse reproductive outcomes have been associated with higher paternal or
22    maternal lead exposures, including reduced fertility, spontaneous abortion, gestational
23    hypertension, congenital malformations, fetal  growth deficits, and neurobehavioral deficits in
24    offspring. The levels of exposure at which different adverse outcomes occur vary.  Increased
25    risks of spontaneous abortion,  neurobehavioral deficits in offspring and, in some  studies,
26    gestational hypertension, have been reported at pregnancy blood lead levels below  10 |ig/dL
27    (Bellinger, 2005).
28
29
      December 2005                           6-333        DRAFT-DO NOT QUOTE OR CITE

-------
 1    6.10.5    Reversibility of Lead Health Effects
 2    6.10.5.1   Natural History of Effects
 3          The absence of a clear operational definition of "reversibility" is a major impediment to
 4    drawing inferences about the natural history of any adverse effect associated with an
 5    accumulative neurotoxicant such as lead. Rather than indicating irreversibility, a performance
 6    deficit that remains detectable after external exposure has ended could reflect ongoing toxicity
 7    due to lead remaining at the critical target organ or lead deposited at the organ post-exposure as
 8    the result of redistribution of lead among body pools. As noted earlier, brain lead levels can
 9    remain elevated long after blood lead levels fall.  A rigorous test of reversibility would require
10    that every lead atom has been cleared from  the body.  This being unattainable, investigators must
11    exploit opportunities that permit only weaker tests of hypotheses about reversibility. These
12    include assessing the persistence of deficits previously associated with lead biomarkers and
13    evaluating performance changes associated with natural experiments, i.e., events  such as
14    chelation or a change in external exposure that would be expected to perturb the equilibrium of
15    lead among different body pools.
16          The likelihood of reversibility, as defined above, appears to be related, at least for the
17    adverse effects observed in certain organ systems, to both the age-at-exposure and the age-at-
18    assessment.  In occupationally-exposed adults, the central and peripheral nervous system
19    correlates of higher lead burdens appear to attenuate if exposure is reduced.
20          The prospective studies of childhood lead exposure, involving serial measurements of
21    lead biomarkers and health outcomes, provide the best opportunities available to assess the
22    natural history  of adversities associated with low-level lead exposures.  In some prospective
23    studies, associations observed in infancy between biomarkers of prenatal exposure and
24    neurodevelopment attenuated by the time children reached preschool age.  It can be difficult to
25    determine, however, whether this  reflects actual disappearance of the effect or an increased
26    difficulty in detecting it due to the emergence of associations between neurodevelopment and
27    lead biomarkers measured postnatally.  It is notable, however, that in some prospective studies of
28    children, associations between biomarkers of prenatal lead exposure and various outcomes in
29    middle adolescence have been reported, suggesting that the persistence of the associations might
30    be endpoint-specific. For example, among  children in Kosovo, Yugoslavia, IQ scores at the age
31    of 8 years were inversely associated with a  composite index of prenatal lead exposure (average

      December 2005                           6-334       DRAFT-DO NOT QUOTE OR CITE

-------
 1    of mothers' blood lead levels at midpregnancy and at delivery) (Wasserman et al., 2000). This
 2    association was independent of changes in postnatal blood lead levels.  Among 15 to 17 year old
 3    inner-city children in Cincinnati, OH, maternal blood lead levels in the 1st trimester (ranging
 4    from 1 to -30 |ig/dL) were inversely related to attention and visuoconstruction (Ris et al., 2004)
 5    and positively related to the frequency of self-reported delinquent behaviors (Dietrich et al.,
 6    2001).
 7          The results of the prospective studies are more consistent in showing that higher postnatal
 8    lead biomarkers are associated with neurocognitive deficits that persist, in some studies,  into
 9    early adulthood when the concurrent lead exposures are generally much lower. Ongoing external
10    exposure does not appear to be necessary to maintain the deficits, although, as noted previously,
11    it is not possible to exclude entirely a role for ongoing endogenous exposures of the target organs
12    resulting from the redistribution, over time, of lead stores among different compartments. These
13    data are  consistent with those from experimental nonhuman primate studies, in which the
14    temporal characteristics of exposure are manipulated as opposed to merely observed as in the
15    human studies.
16          In most epidemiologic studies, the potential for true longitudinal analysis of the data has
17    not been fully exploited, with the data evaluated in what is effectively a series of cross-sectional
18    analyses.
19
20    6.10.5.2   Medical Interventions
21          Data from the Treatment of Lead Poisoned Children (TLC) study, a randomized
22    controlled trial of the late outcomes of children treated for lead poisoning, support the hypothesis
23    that the deficits associated with exposures of such magnitude are persistent and, possibly,
24    permanent (Dietrich et al., 2004; Rogan et al., 2001). At 36-months post-treatment and at age 7
25    years, no significant differences in cognition or behavior were noted between the succimer and
26    placebo  groups. Current blood lead levels were significantly associated with cognitive
27    performance at baseline, 36-months post-treatment, and at 7 years of age, and the regression
28    coefficients were similar in magnitude to those estimated in observational studies (i.e., ~3 point
29    IQ decline per 10 |ig/dL increase in blood lead), providing a linkage between the results  of the
30    observational studies and those of this experimental  study. However, within-child analyses
      December 2005                           6-335        DRAFT-DO NOT QUOTE OR CITE

-------
 1    indicated that changes in developmental test scores over time were not consistently associated
 2    with changes over time in blood lead level.
 3
 4    6.10.6    Confounding of Lead Health Effects
 5    6.10.6.1   Adjustment for Confounding in Epidemiologic Studies of Lead
 6          The possibility that the adverse health effects associated with increased lead exposure in
 7    epidemiologic studies are, in fact, due to risk factors with which increased lead exposure is
 8    associated remains the most important impediment to drawing causal inferences.  Various
 9    approaches have been taken to reduce the uncertainty this creates. Some investigators have
10    specified the sampling frame  or the eligibility criteria so as to increase the homogeneity of the
11    study participants on factors known to be strong risk factors for the outcome of interest, thereby
12    reducing the correlation between them and lead, and their potential to confound any association
13    observed between increased lead exposure and poor outcome. Reducing confounding by means
14    of design decisions has the disadvantage that an investigator cannot determine whether the
15    impact of lead on the outcome varies depending on the factor whose range of potential values has
16    been restricted.  More frequently, however, investigators have relied on statistical procedures,
17    applied post data collection, to identify  and control for potential confounding.  Unlike sample
18    restriction, this approach preserves the opportunity to explore possible modification of the lead
19    effect by cofactors.
20          Adjustment for confounding has been performed primarily using multiple regression
21    analyses and data stratification. For multiple regression modeling, stepwise regression has been
22    frequently used for covariate  selection.  Stepwise regression  has many faults and is often less
23    acceptable then the use of a few well-chosen covariates. However, the stepwise regression
24    methodology may be considered less bias as it selects from a class of variables that represent a
25    wide scientific viewpoint rather than the narrower one of the investigator. One problem with
26    stepwise regression pointed out by Bellinger (2004) is that the usual adjustment strategy assumes
27    that all the variance in the response shared by the exposure and the confounder belongs to  the
28    confounder.  In some settings, this is likely to be excessively conservative, because confounders
29    can, to some extent, also be proxies for  exposure.  This is further discussed in the next section.
30          Splitting the data set into smaller data sets (partitioning or stratification) and analyzing
31    those data sets separately was used in some of the studies examining the relationship between

      December 2005                           6-336       DRAFT-DO NOT QUOTE OR CITE

-------
 1    blood pressure and lead.  This practice also has some advantages and disadvantages.
 2    An advanced statistical method could be used to determine how the partitioning should be done
 3    (Young and Hawkins, 1998), which could reveal relationships that would not be possible to
 4    detect using the usual regression techniques. A disadvantage of partitioning a small data set is
 5    that the smaller sample size may lack the power to detect otherwise detectable associations and
 6    to yield reliable estimates.
 7          The segmented line model consists of joined straight line segments where the joined
 8    points are chosen to best fit the data (Quandt, 1958).  The log-linear and the quadratic models
 9    have  shown in several cases to better fit the biomarker-response relationship than the linear
10    model.  However, these models are not considered practicable for extrapolation outside the range
11    of the biomarker variable. The segmented line model is suggested as a more reasonable model
12    for extrapolation into the low-concentration sparse-data region.
13
14    6.10.6.2   Confounding Adjustment on Lead Health Effect Estimates
15          The ability of the investigator to determine how much of the apparent association between
16    a lead biomarker and an outcome reflects residual confounding by a cofactor depends on the
17    characteristics of the joint distribution of lead and the cofactor. Co-factors for lead health effects
18    include maternal IQ, maternal smoking, alcohol use, birth weight, and many others depending on
19    the health outcome of interest.  Some of these cofactors are truly independent predictors and can
20    be adjusted for using multiple regression analyses. Under some circumstances, however, lead
21    and the cofactor may be so highly related that one cannot be confident that their associations
22    with the outcome have been disentangled by the statistical methods applied. Moreover, the true
23    causal relationships among lead, the cofactors, and the outcome might not be sufficiently well
24    understood that the outcome variance shared by lead  and the cofactors can be characterized
25    appropriately in the analyses.
26          In studies of lead and neurodevelopment, the magnitude of the lead coefficient, reflecting
27    the decline in test score per unit increase in the lead biomarker, is substantially reduced, often by
28    half or more, by adjusting for markers of the social environment.  During the 1980s, adjustment
29    for parental IQ and quality of the home environment  (e.g., HOME scores) became almost
30    mandatory if the findings of a study of lead and children's cognitive outcomes were to be
31    considered credible.  While both factors surely strongly influence child outcomes in ways that

      December 2005                           6-337       DRAFT-DO NOT QUOTE OR CITE

-------
 1    are independent of lead, a case can also be made that lead might contribute to the associations.
 2    A parent's IQ presumably reflects the parent's early lead exposure and, assuming that the
 3    physical environments in which a parent and child grow up are not completely unrelated to one
 4    another, provide similar lead exposure opportunities.  Adjusting for parent IQ in evaluating the
 5    association between a child's lead exposure and his or her IQ, therefore, will result in an
 6    underestimate of the contribution of the child's lead exposure to his or her IQ. Similarly, if early
 7    lead exposure alters child behavior, the transact!onal model of child development would generate
 8    the prediction that the changes will elicit different behaviors from parents, altering the
 9    characteristics of the child rearing environment. For instance, increased lead exposure might
10    result in an infant being more irritable, less soothable, and the parent less nurturing.  In so far as
11    measurement of the quality of the rearing environment in studies occurs after the children have
12    experienced some lead exposure, the hypothesis that lead is responsible for shaping some aspects
13    of that environment cannot be entirely dismissed, and control for HOME scores might be
14    excessively conservative.  For example, in the pooled analysis by Lanphear et al. (2005) that
15    included seven prospective studies, the crude coefficient for concurrent lead and childhood IQ
16    score was -4.66 (95%  CI: -5.72, -3.60), but the coefficient adjusted for study site, HOME
17    score, birth weight, maternal IQ, and maternal education was -2.70 (95% CI: -3.74, -1.66).
18          Other aspects of model building in assessing the association of lead with health outcomes
19    also warrant comment. In many studies of lead and cognitive outcomes in children, investigators
20    have adjusted for factors such as birth weight or length of gestation that might,  themselves,
21    reflect adverse effects of lead, i.e., mediating factors that lie between lead and condition on the
22    causal pathway. The coefficient estimated for lead in a model that contained such factors would
23    be smaller in magnitude than it would be if terms for such mediating factors had not been
24    included.
25          Recognizing imperfections in the ability to measure such factors well, a  concern is
26    expressed that the lead coefficient could be reduced further, perhaps all the way to the null,
27    if better, more comprehensive methods of measurement were applied.  On the other hand, the
28    methods used to adjust for such factors may be excessively conservative insofar as they attribute
29    to a factor all of the outcome variance that it shares with lead, despite the likelihood that the true
30    relationships among lead, social factors, and outcome are unlikely to be as simple as this model
31    assumes. Some factors might, in part, be markers of lead exposure opportunities. For example,

      December 2005                           6-338       DRAFT-DO NOT QUOTE OR CITE

-------
 1    both lead biomarker levels and lower cognitive function in children are associated with lower
 2    social class standing. Social class is a complex construct that conveys information about a
 3    multitude of factors that might influence children's health, including the amount of lead in
 4    environmental media.  Thus, some of the association between lower social class and poorer
 5    health might reflect the effect  of higher lead exposure.  If so, routine adjustment of health
 6    outcome for social  class in assessing the association between increased lead exposure and poorer
 7    health in children will fail to distinguish these lead-related and non-lead-related components of
 8    the association between social class and health, and, in fact, will assume that all of it is non-lead-
 9    associated.  It is nearly impossible to actually determine if the problem of overadjustment exists
10    in a particular data  set.  There  are several statistical methods which attempt to address this
11    problem.  These include using partial F tests, ridge regression, path analysis, and structural
12    equations. None of these methods are completely satisfactory.
13
14    6.10.7    Inferences of Causality
15          Even with more sophisticated and nuanced models, however, any conclusions about the
16    causal forces generating the results of any observational epidemiologic study are necessarily
17    uncertain. In the absence of random assignment to exposure group, residual confounding will
18    always be a possible explanation of an observed association. As in other areas of epidemiology,
19    a weight-of-evidence approach remains the best option  available as a basis for drawing of causal
20    inferences.  If the association between a lead biomarker and a health outcome of interest is
21    observed in settings that vary widely in terms of the characteristics of the social environment
22    including sociodemographic and cultural characteristics, characteristics of the study participants,
23    including nutritional status, genetic factors, and lifestyle factors, the likelihood that the
24    association is attributable, in its entirety, to residual confounding is reduced. For instance, the
25    pooled analyses of  data contributed by many of the international prospective studies provide a
26    compelling demonstration that the association between  blood lead level and child IQ  is
27    remarkably robust across disparate socio-cultural settings (Lanphear et al., 2005). Even such
28    consistency in the effect estimate across diverse settings is only indirect and weak evidence of
29    causality, however.  In general, epidemiologic  studies rarely provide data that enhance our
30    understanding of the "black box" between biomarkers of lead burden and indicators of health
31    status. Epidemiologic data identify associations between exposure biomarkers and health

      December 2005                           6-339       DRAFT-DO NOT QUOTE OR CITE

-------
 1    indicators, but are not highly informative regarding possible mechanisms of lead toxicity that
 2    underlie the associations.  A critical stage in applying the overall weight-of-evidence approach is
 3    the examination of the epidemiologic data in the context of data from experimental animal
 4    behavioral and mechanistic studies. Although such data have their own limitations, they are not
 5    subject to many of the most important potential biases that can becloud the interpretation of the
 6    epidemiologic data.
 7
 8    6.10.8    Effects on the Individual Versus Effects on the Population
 9          The critical distinction between population and individual risk, an issue pertinent to many
10    questions in chronic disease epidemiology, has frequently been blurred in discussions of the
11    public health implications of lead-associated decrements in health. With respect to
12    neurodevelopment, while it may be true that a two- or three-point decline in IQ may not be
13    consequential for an individual, the same level of decline observed in a population mean is of
14    great importance.  Similarly, although an increase of a few mm Hg in blood pressure may
15    generally not be of concern for an individual's well-being, a very modest increase in the
16    population mean is associated with substantial increases in the percentages of individuals with
17    values that are sufficiently extreme that they  exceed the criteria used to diagnose illness (Rose
18    and Day, 1990). In other words, the mean value conveys substantial information  about the
19    percentage of individuals with clinically relevant, extreme values of the indicator. Moreover,
20    interventions that shift the population mean by an amount that is without clinical consequence
21    for an individual have been shown to produce substantial changes in the  percentage of
22    individuals with indicator values that are clinically significant (Bellinger, 2004).  The following
23    subsections will discuss quantitatively lead-related effects of a population level change in IQ and
24    blood pressure.
25
26    6.10.8.1   Effects of Lead on Intelligence
27          The outcome most often examined to investigate neurotoxic effects of lead is IQ.
28    Although the definition of "intelligence" is quite  abstract,  IQ remains a useful outcome measure
29    as it is correlated with important measures of life success,  such as academic achievement,
30    earnings, and  social status (Bellinger, 2003; Weiss, 2000). Several studies reported quantitative
31    relationships between full scale IQ and current blood lead levels for children aged 5 to 11 years

      December 2005                           6-340        DRAFT-DO NOT QUOTE OR CITE

-------
 1
 2
 3
 4
old, and these are summarized in Table 6-10.1. The estimated relationships as reported by the
authors are used.
                Table 6-10.1. Summary of Studies with Quantitative Relationships for
                                         IQ and Blood Lead
      Reference
                     Study Location
 Estimated Slope
(IQ points/fig/dL) -
Blood Lead 10th to
  90th Percentile
 Estimated Slope
(IQ points/fig/dL) -
Blood Lead Under
    10
Bellinger etal. (1992)
Canfield et al. (2003a)
Dietrich etal. (1993a)
Ernhart etal. (1989)
Wasserman et al. (1997)
Baghurst etal. (1992)
Silva etal. (1988)
Lanphear et al. (2005)
Boston, Massachusetts
Rochester, New York
Cincinnati, Ohio
Cleveland, Ohio
Kosovo, Yugoslavia
Port Pirie, South Australia
Dunedin, New Zealand
International Pooled Analysis
116
182
221
160
231
324
579
1,333
-0.5
-0.7
-0.3
-0.1
-0.2
-0.2
-0.3
-0.5
NA
-0.8
-0.3
NA
NA
-0.4
-0.3
-0.2
 5          The curves over a range of blood lead levels from the 10th percentile to the 90th
 6   percentile are shown in Figure 6-10.2.  The curves are restricted to that range because log-linear
 7   curves become very steep at the lower end of the blood lead levels, and this may be an artifact of
 8   the model chosen. The percentiles are estimated using various methods and are only
 9   approximate values. Studies which estimated a linear relationship are shown as reported, and
10   similarly for the log-linear relationships.  Note that these are not forest plots of slopes or hazard
11   ratios - they are the actual estimated relationships.
12          The analysis by Lanphear et al. included the studies of Baghurst et al. (1992), Bellinger
13   et al. (1992), Canfield et al. (2003a), Dietrich et al.  (1993a), Ernhart et al. (1989) and Wasserman
14   et al. (1997). The pooled analysis also included the Mexico City study of Schnaas et al. (2000).
15   The results from Schnaas et al. are not included in Table 6-10.1 or Figure 6-10.2 because the
16
     December 2005
                                          6-341
 DRAFT-DO NOT QUOTE OR CITE

-------
             120
             115
             110
          _2 100 J
           S  95 J
          0)
          =  so J
          £  85
              80
              75
              70
              et al. (1992)
                    Silvaetal. (1988)
                                                                   Baghurstetal. (1992)
                             et al. (ZOOS)
Dietrich etal. (1993)
Ernhartetal. (1989)

                   (1997)
                                     10        15       20        25
                                                 30
                                    35
     Figure 6-10.2. Concentration-response relationships of IQ to blood lead for the individual
                    studies and the pooled analysis by Lanphear et al. (2005).
 1   authors did not provide regression coefficients in their paper, thus concentration-response
 2   relationship were not estimable. The study by Silva et al. (1988) is not included in the pooled
 3   analysis of Lanphear et al., but is included in this section as its results are comparable
 4   and informative.
 5           Several conclusions can be drawn from these graphs. First, note that the overall IQ levels
 6   are quite different. This results from different populations and from different applications of the
 7   IQ tests. Second, all studies showed a decreasing IQ score as the blood lead level increased.
 8   It is the slope of the studies that is relevant, not the actual IQ scores. Third, for studies with
 9   lower blood lead levels, the slopes appear to be steeper.  This is the reason that many authors
10   choose to use the log-linear model. However, for those studies where the blood leads were
11   generally high, the log-linear and linear models are almost identical. Thus it is not surprising
12   that some authors chose a linear model instead of a log-linear model. The curves in Figure
13   6-10.2 do not show evidence of a no-effect threshold because the slopes increase as the blood
14   lead levels become smaller.  The observed mean adjusted IQ levels (for blood lead <5, 5 to 10,
     December 2005
                   6-342
         DRAFT-DO NOT QUOTE OR CITE

-------
 1
 2
 3
 4
10 to 15, 15 to 20, and >20 |ig/dL) reported by Lanphear et al. (2005) also show no evidence of a
threshold, as seen in Figure 6-10.3.
                        105
                        100
                     g
                     "re
                     o
                     *  S5
                     3
                     c
                     re
                     o
                         90
                         85
                                      10    15    20    25    30
                                                              35
     Figure 6-10.3.   Mean blood lead levels adjusted for HOME Score, maternal education,
                     maternal IQ, and birth weight from the pooled analysis of seven studies by
                     Lanphear et al. (2005). Mean adjusted IQ levels at blood lead levels of <5,
                     5 to 10,10 to 15,15 to 20, and >20 ug/dL are shown.
 5         Weiss (1990) predicted, on purely statistical grounds, that a downward shift of five points
 6   in mean IQ, if the amount of dispersion in the distribution remained the same, should be
 7   accompanied by a doubling of the numbers of individuals with scores two or more standard
 8   deviations below the mean and a reduction by half of the number of individuals with scores two
 9   or more standard deviations above the mean.  With respect to lead, the general accuracy of this
10   prediction has been empirically demonstrated in two different datasets by Needleman et al.
11   (1982) and Bellinger (2004).  The example below provides further evidence of the change in
12   percentages of individuals with IQ <70 or <50 points after restricting the analysis to those with
13   blood lead levels less than 10 |ig/dL.
     December 2005
                                        6-343
DRAFT-DO NOT QUOTE OR CITE

-------
 1          The average slope was estimated for those studies with a significant portion of the
 2    subjects with blood lead levels less than 10 |ig/dL.  These average slopes are given in Table
 3    6-10.1. In addition, the results of Lanphear et al. (2005) were considered. The average slope
 4    at blood lead levels less than 10 |ig/dL from that pooled analysis was -0.5 IQ points per |ig/dL.
 5    Based on the individual studies and the pooled analysis it appears that the average slope
 6    is between -0.3 and -0.5 points per |ig/dL, with the exception of the large negative slope of
 7    -0.8 points per 10 |ig/dL from the study by Canfield et al. (2003a).  The value of -0.4  points per
 8    |ig/dL will be used in calculations of the implications of the slope at blood lead levels less than
 9    10 |ig/dL.
10           A nonexposed population was assumed to have a standard mean IQ of 100 and  standard
11    deviation of 15 at a blood lead exposure of 0 |ig/dL. The fraction of the population that would
12    have an IQ <70 or <50 as a function of blood lead level was then calculated.  The results are
13    shown in Figure 6-10.4. Note that the fraction with an IQ level below 70, a level often requiring
14    community support to live (World Health Organization, 1992) increases  from a little over
15    2 percent for no lead exposure to about 4 percent with a blood lead level of 10 |ig/dL.
16    In addition, the fraction with an IQ level below 50, a level often requiring continuous support to
17    live (World Health Organization, 1992) increases from a little over 4 per 100,000 for no lead
18    exposure to about 11 per 100,000 with a blood lead level of 10 |ig/dL.
19           A shift in the mean value of a health indicator has substantial importance for both
20    extremes of the distribution. In the case of lead, a downward shift in the mean IQ value is
21    associated not only with a substantial increase in the percentage of individuals achieving very
22    low scores, it  is associated as well with a substantial decrease in the percentage achieving very
23    high scores. Based on the study by Bellinger et  al. (1987) examining intelligence test scores of
24    lead-exposed  children, Weiss (1988) discussed the shift of the population distribution of IQ from
25    a mean of 100 and a standard deviation of 15 to  a mean of 95, a 5% reduction. When the mean
26    IQ level is 100, 2.3% of the individuals in a given population would score above 130. However,
27    with the population distribution shift and the resulting mean decline in IQ, only 0.99%  of the
28    individuals would score above 130.  Weiss states that the implication of such as loss transcends
29    the current circumscribed definitions of risk.
30
      December 2005                           6-344        DRAFT-DO NOT QUOTE OR CITE

-------
                          0.08
                      V
                      a
                      c
                      o
                      '&
                      o
                      (Q
                          0.04
0.02
                          0.00
                                           345678
                                                 10
                    o
                    in
                        0.0016
                    "g
                     c
                    _o
                     o
                     2
                    U_
                        0.0000
                             0123456789   10
    Figure 6-10.4.  Effect of blood lead on fraction of population with IQ level <70 or
                    <50 points.
1    6.10.8.2  Cardiovascular Effects of Lead
2          In studies investigating the cardiovascular effects of lead, blood pressure has been
3    examined most frequently.  Results from the Framingham Heart Study show that higher levels of
4    blood pressure, even within the nonhypertensive range, impose increased rates of cardiovascular
5    disease (Kannel, 2000a,b).  A continuous graded increase in cardiovascular risk is observed as
6    blood pressure increases, with no evidence of a threshold value. Most events arise not in the
7    most severe cases, but mainly in those with high normal blood pressure (i.e., mild hypertension).
8    This view is further supported by the Seventh Report of the Joint National Committee on
    December 2005
                    6-345
DRAFT-DO NOT QUOTE OR CITE

-------
 1    Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (Chobanian et al.,
 2    2003).  Kannel (2000b) states that reducing even moderate elevation in blood pressure is likely to
 3    be beneficial.
 4          Kannel (2000a) states that systolic blood pressure exerts a strong, influence on
 5    cardiovascular events, as it is the prime causal function of hypertension and its adverse
 6    cardiovascular sequelae. Cardiovascular events include coronary disease, stroke, peripheral
 7    artery disease, and cardiac failure. Risk ratios are larger for cardiac failure and stroke, but
 8    coronary disease (i.e., myocardial infarction, angina pectonis, sudden death) is the most common
 9    and most lethal sequela of hypertension (Kannel, 1996). Kannel (2000a) notes that the
10    Framingham Heart Study has recognized that elevated blood pressure tends to occur alongside
11    other major risk factors of cardiovascular disease such as glucose intolerance, dyslipidemia,
12    abdominal obesity, and left ventricular hypertrophy, among others.  If a cluster of multiple risk
13    factors is present, the hazard is formidable for coronary disease and stroke.
14          No critical level of blood pressure is evident. The risk appears to be simply proportional
15    from the lowest to the highest level recorded. In the Multiple Risk Factor Intervention Trial
16    (MRFIT), Neaton et al. (1995) confirmed a continuing and graded influence of systolic blood
17    pressure on cardiovascular disease mortality extending down into the range of <140 mm Hg.
18    The Prospective Studies Collaboration (2002) meta-analysis of 61 prospective studies relates
19    blood pressure to vascular mortality without indication of a threshold down to 115/75 mm Hg.
20    The absence of a demonstrable safe or critical level of blood pressure suggests using the range of
21    blood pressure rather than discrete categories such as hypertension.
22          Many studies have suggested a relationship between blood lead and systolic blood
23    pressure. In particular, the meta-analysis of Nawrot et al. (2002) indicated that a doubling of the
24    blood lead corresponded to a 1 mm Hg increase in systolic blood pressure. Although this
25    magnitude of increase is not clinically meaningful for an individual,  a population shift of
26    1  mm Hg is important.
27          The Framingham Heart Study results (Kannel, 2000a) were used to estimate a typical
28    population distribution of systolic blood pressure values (Figure 6-10.5).  The distribution of
29    systolic blood pressure values was approximated well by a lognormal distribution for both
30    women and men (p > 0.4). The relationship between systolic blood pressure and the risk of
      December 2005                           6-346       DRAFT-DO NOT QUOTE OR CITE

-------
                     (0
                     c
                     o

                     t
                     o
                     Q.
                     O
                     O
                     c
                     &
                     3
                     D"
                          75                    150                  225

                            Systolic Blood Pressure (mm Hg) in Women
                     (Q
                      o
                     c
                     o
                     t
                     o
                     Q.
                     O
                     IX
                     u

                     CD
                     3
                     CT

                     £
                     UL
                         75                    150                    225

                             Systolic Blood Pressure (mm Hg) in Men
    Figure 6-10.5.  Distribution of systolic blood pressure in women and men aged 35 to

                   64 years from the Framingham Heart Study (Kannel, 2000a).
1   cardiovascular events was also given by Kannel (2000a). The relationships are shown in


2   Figure 6-10.6.


3         To estimate population risk, it was assumed that the effect of blood lead on blood pressure


4   was to shift the entire distribution by the amount given by Nawrot et al. (2002).  For each shift in
    December 2005
6-347
DRAFT-DO NOT QUOTE OR CITE

-------
                    75

               LJJ c
               jo w 50
                 «
                 O
                  -25
               o S.
                     0
                      80           110          140           170          200
                              Systolic Blood Pressure (mm Hg) in Women
                    75
               HI c
               u.
                o o
               I -'25
               o 2L
                      80           110          140           170
                               Systolic Blood Pressure (mm Hg) in Men
                               200
    Figure 6-10.6.  Relationship of cardiovascular events (coronary disease, stroke,
                   peripheral artery disease, cardiac failure) to systolic blood pressure in
                   women and men aged 35 to 64 years from the Framingham Heart Study
                   (Kannel, 2000a).
1   the distribution, the entire distribution was integrated out over the risk given in Figure 6-10.6.
2   The result estimated was expected number of cardiovascular events per 1,000 person years, and
3   this was plotted for blood lead levels ranging from 5 to 15 |ig/dL for both women and men. The
4   results are shown in Figure 6-10.7. Although the effects are modest, they translate into a large
5   number of events for a moderate population size. For example, a decrease in blood lead from
6   10 to 5 ng/dL results in an annual decrease of 27 events per 100,000 women and 39 events per
7   100,000 men.
    December 2005
6-348
DRAFT-DO NOT QUOTE OR CITE

-------
i=
OJ
LU
_co
=s
CO
o
co
12
co
O
in
E
"5
c





efl
!•«
CO
^
C
o
en
<5
n
O
^
OJ
a.




27.5
25.0-
ff^f^j i»
22.5-
20.0-

17.5-

15.0-
12.5-
10.0-
7.5-

5.0-
2.5-
O/*
.0
















                                          8        10        12
                                           Blood Lead (pg/dL)
                          14
                        16
                                           Women
           --- Men
     Figure 6-10.7.   Effect of blood lead on expected annual risk of cardiovascular events per
                     1,000 person years.
 1   6.10.9   Summary of Key Findings and Conclusions Derived from Lead
 2             Epidemiology Studies
 3          The remarkable progress that has been made since the mid-1980s in understanding the
 4   effects of lead on health can be gauged by noting the changes that have occurred over time in
 5   the questions investigators have addressed. In the 1980s, the question of interest was often,
 6   "Does low-level lead exposure affect  health?" The questions asked in recent studies have more
 7   often focused on details of the associations, including the shapes of concentration-response
 8   relationships, especially at levels well within the range of general population exposures,
 9   biological and socio-environmental factors that either increase or decrease an individual's risk,
10   the prognoses associated with lead-associated effects, the efficacy of interventions to reduce
11   adverse effects, and so on. In fact, "low-level," a term long-used to describe exposures that are
12   not sufficiently high to produce clinical signs and symptoms, is increasingly being recognized as
     December 2005
6-349
DRAFT-DO NOT QUOTE OR CITE

-------
 1    a descriptor that has little biological meaning and is interpretable only in a specific historical
 2    context. What was considered "low" in the 1980s is an order of magnitude higher than the
 3    current mean level in the U.S. population, and the current mean remains perhaps as much as two
 4    orders of magnitude above "natural" background levels in humans. The current CDC screening
 5    guideline for children of 10 |ig/dL is not a "bright line" separating toxicity from safety, but
 6    merely a risk management tool. There is no level of lead exposure that can be clearly identified,
 7    with confidence, as "safe." Recent studies of lead neurotoxicity in infants have observed adverse
 8    effects at blood lead levels of only 1 or 2 |ig/dL and adverse renal outcomes have been reported
 9    at blood lead levels below 5 |ig/dL. Public health interventions have resulted in declines, over
10    the last 25 years, of more than 90% in the mean blood lead level within all age and gender
11    subgroups of the U.S. population, substantially decreasing the numbers of individuals at risk of
12    lead toxi cities.
13          The following are a listing of key health outcomes discussed earlier in the epidemiology
14    chapter:
15      •  Neurotoxic effects of lead in children. The effects of lead on neurobehavior in children
16         have been observed with remarkable consistency across numerous studies of various
17         designs, populations, and developmental assessment protocols. The negative impact of
18         lead on neurocognitive ability and other neurobehavioral outcomes persist in most recent
19         studies even after adjustment for numerous confounding factors including social class,
20         quality of caregiving, and parental intelligence.  An international pooled analysis of seven
21         prospective cohort studies offers evidence that exposure to lead has an effect on the
22         intellectual attainment of preschool and school age children even at blood lead levels
23         below 10 jig/dL.
24                 Epidemiologic studies have demonstrated that lead also may be associated with
25         increased risk for antisocial and delinquent behavior,  which may be a consequence of
26         attention problems and academic underachievement among children who have suffered
27         higher exposures to lead during their formative years.  Direct measures of brain damage
28         using Magnetic Resonance Imaging (MRI) and Magnetic Resonance Spectroscopy (MRS)
29         also are suggesting evidence of harm due to lead exposure.  Pharmacological or nutritional
30         intervention strategies generally have not shown to eliminate or reduce lead-associated
31         neurodevelopmental morbidities.

      December 2005                            6-350       DRAFT-DO NOT QUOTE OR CITE

-------
 1      •  Neurotoxic effects of lead in adults. Environmental lead exposure has not been found to
 2         be associated with impaired cognitive performance in the elderly if competing risk factors
 3         are considered. In adults, the effect of lead on the nervous system may not be detected
 4         through neurobehavioral testing due to cognitive reserve, the ability to compensate for
 5         brain impairment.
 6                 Numerous studies of occupational lead exposure observed associations of blood
 7         lead with peripheral sensory nerve impairment, visuomotor and memory impairment, and
 8         postural sway abnormalities. Past occupational exposure to lead also was associated with
 9         increased risk of developing Amyotrophic Lateral Sclerosis (ALS), motor neuron disease,
10         and essential  tremor. The odds of developing ALS and essential tremor were significantly
11         increased in individuals with the ALAD2 allele.  These neurobehavioral impairments in
12         occupationally-exposed individuals were typically associated with higher blood lead levels
13         (approximately 30-40 jig/dL); however, essential tremor was found to be associated with
14         much lower blood lead levels (mean 3 jig/dL).
15
16      •  Renal effects of lead.  In the general population, both cumulative and circulating lead was
17         found to be associated with longitudinal decline in renal functions. In the large NHANES
18         III study, renal dysfunction was observed in hypertensives at a mean blood lead of only
19         4.2 |ig/dL.  These results provide strong evidence that the kidney is a target organ for
20         adverse effects from lead in adults at current U.S. environmental exposure levels.  The
21         renal impact in children environmental lead exposure is difficult to assess since the most
22         studies have measured early biological effect markers and their prognostic value is
23         uncertain.
24                 Studies involving the longitudinal assessment of renal function decline in
25         susceptible patient populations observed that low levels of blood lead (<5 |ig/dL) and
26         chelatable lead levels were associated with decline in glomerular filtration rate over a
27         4 year follow-up period in patients with chronic renal insufficiency. Renal function in
28         these patients was found to stabilize and, in some cases,  improve after therapeutic
29         chelation.
30
      December 2005                           6-3 51       DRAFT-DO NOT QUOTE OR CITE

-------
 1      •  Cardiovascular effects of lead. Epidemiologic studies support the relationship between
 2         increased lead exposure and increased adverse cardiovascular outcome, including
 3         increased blood pressure and increased incidence of hypertension.  A recent meta-analysis
 4         reported that a doubling of blood lead level was associated with a 1.0 mm Hg increase in
 5         systolic blood pressure and a 0.6 mm Hg increase in diastolic pressure. Studies also have
 6         found that cumulative past lead exposure (e.g., bone lead) may  be as important, if not
 7         more, than present exposure in assessing cardiovascular effects. The evidence for an
 8         association of lead with cardiovascular morbidity and mortality is limited but supportive.
 9
10      •  Reproductive and developmental effects of lead. The epidemiologic evidence suggests
11         small associations between exposure to lead and male reproductive outcomes, including
12         perturbed semen quality and increased time to pregnancy.  These associations appear at
13         blood lead levels greater the 45 |ig/dL, as most studies only considered exposure in the
14         occupational setting. There are no adequate data to evaluate associations between lead
15         exposure and female fertility.  For many other outcomes, the observed associations are
16         fairly small, especially at the levels of exposure that are currently of interest. However,
17         there may be populations that are highly susceptible to lead-related reproductive effects,
18         especially if they have additional risk factors for these outcomes.
19
20      •  Genotoxic and carcinogenic effects of lead. Studies of genotoxicity consistently find
21         associations of lead exposure with DNA damage and micronuclei formation; however, the
22         associations with the more established indicator of cancer risk,  chromosomal aberrations,
23         are inconsistent. Epidemiologic studies of highly-exposed occupational populations
24         suggest a relationship  between lead and cancers of the lung and the stomach; however the
25         evidence is limited by the presence  of various potential confounders, including
26         coexposures (e.g., arsenic, cadmium), smoking, and dietary habits. The 2004 IARC
27         review concluded that lead was a probable carcinogen based on limited evidence in
28         humans and sufficient evidence in animals.
29
30      •  Effects of lead on the immune system.  Several studies have examined possible
31         associations between lead exposures and  biomarkers of immune function. Findings from
32         recent epidemiologic studies  suggest that lead exposure may be associated with effects on

      December 2005                           6-3 52       DRAFT-DO NOT QUOTE OR CITE

-------
 1         cellular and humoral immunity.  These effects include changes in serum immunoglobulin
 2         levels; perturbation of peripheral lymphocyte phenotype profiles, including decreases in
 3         peripheral blood T-cell  abundance and changes in T-cell to B-cell abundance ratios;
 4         suppression of lymphocyte activation; and suppression of neutrophil chemotaxis and
 5         phagocytosis.  Studies of biomarkers of humoral immunity in children have consistently
 6         found significant associations between increasing blood lead concentrations and serum
 7         IgE levels at blood lead levels below 10 |ig/dL.
 8
 9      •  Effects of lead on the hematopoietic system. Lead exposure has been associated with
10         disruption of heme synthesis in both children and adults. Increases in blood lead
11         concentration of approximately 20-30 |ig/dL are sufficient to halve erythrocyte ALAD
12         activity and sufficiently inhibit ferrochelatase to double erythrocyte protoporphyrin levels.
13         Perturbation of erythropoiesis, indicated by changes in serum erythropoietin and
14         progenitor cells,  occurs in the absence of detectable changes in blood hemoglobin levels or
15         hematocrit in children and adults at blood lead levels below 40 |ig/dL.  Risk of clinical
16         anemia in children becomes appreciable at much higher blood lead concentrations.
17
18      •  Effects of lead on the hepatic and gastrointestinal system.  Studies of hepatic enzyme
19         levels in serum suggest that liver injury may be present in lead workers; however,
20         associations specifically with lead exposures are not evident. Studies of urinary
21         metabolites of cytochrome P450 phenotypes CYP2A6 and  CYP3A4 suggest possible
22         associations between lead exposure and suppression of hepatic enzyme activity in adults
23         and children. Several studies observed an association between occupational lead exposure
24         and prevalence of symptoms of gastrointestinal colic. These hepatic and gastrointestinal
25         effects are largely observed only at blood lead concentrations (>40 |ig/dL).
26
27      •  Effects of lead on the endocrine system. Most studies have yielded no associations, or
28         weak associations, of lead exposure with thyroid hormone status and male reproductive
29         endocrine status  in highly-exposed occupational populations.  Children exposed to
30         relatively high levels of lead (blood lead >30 |ig/dL) exhibit depressed  levels of circulating
31         1,25-dihydroxy vitamin D (1,25-OH-D).  However, associations between serum vitamin D
      December 2005                           6-353       DRAFT-DO NOT QUOTE OR CITE

-------
 1         status and blood lead were not evident in a study of calcium-replete children who had
 2         average lifetime blood lead concentrations below 25 |ig/dL.
 3
 4      •  Effects of lead on bone and teeth. The epidemiologic evidence is limited, but suggestive
 5         of an association between lead exposure and bone toxicity. Studies have found an
 6         association between occupational exposure to lead and Paget's disease. However, it is
 7         difficult to assess whether increased lead results from bone diseases or the bone disease is
 8         a result of increase lead exposure.  Increased risk of dental caries has been associated with
 9         lead exposure in children and adults. Lead effects on caries were observed in populations
10         whose mean blood lead levels were less than 10 |ig/dL.
11
12      •  Effects of lead on ocular health.  Recent longitudinal studies provide evidence for
13         possible associations between lead exposure and adverse ocular health outcomes in low- to
14         moderately-exposed populations. In children whose mothers had blood lead levels of
15         10.5-32.5 |ig/dL in mid-pregnancy, an association was observed between lead exposure
16         and visual evoked retinal responses. Middle-aged males whose tibia bone lead levels were
17         31-126 |ig/g had increased risk of cataracts.
18
     December 2005                            6-3 54       DRAFT-DO NOT QUOTE OR CITE

-------
  1     6.11  REFERENCES

  2     Abadin, H. G.; Wheeler, J. S. (1993) Guidance for risk assessment of exposure to lead: a site-specific, multi-media
  3            approach. In: Andrews, J. S.; Frumkin, H.; Johnson, B. L.; Mehlman, M. A.; Xintaras, C.; Bucsela, J. A.,
  4            eds. Hazardous waste and public health: International Congress on the health effects of hazardous waste.
  5            Princeton, NJ: Princeton  Scientific Publishing Company, Inc.; pp. 477-485.
  6     Abbate, C.; Buceti, R.; Munao, F.  (1995) Neurotixicity induced by lead levels: an electrophysiological study.
  7            Int. Arch. Occup. Environ. Health 66: 389-392.
  8     Aberg, G.; Fosse, G.; Stray, H. (1998) Man, nutrition and mobility: a comparison of teeth and bone from the
  9            Medieval era and the present from Pb and Sr isotopes. Sci. Total Environ. 224: 109-119.
10     Abudhaise, B. A.; Alzoubi, M. A.; Rabi, A. Z.; Alwash, R. M.  (1996) Lead exposure in indoor firing ranges:
11            environmental impact and health risk to the range users. Int. J. Occup. Med. Environ. Health 9: 323-329.
12     Adachi, J. D.; Arlen, D.; Webber,  C. E.; Chettle, D. R.; Beaumont, L. F.; Gordon, C. L. (1998) Is there any
13            association between the presence of bone disease and cumulative exposure to lead? Calcif. Tissue Int.
14            63:429-432.
15     Ades, A. E.; Kazantzis,  G. (1988)  Lung cancer in a non-ferrous smelter: the role of cadmium. Br. J. Ind. Med.
16            45:435-442.
17     Agency for Toxic Substances and Disease Registry. (1993) Toxicological profile for cadmium. Atlanta,  GA:
18            U.S. Department of Health & Human Services, Public Health Service; report no. ATSDR/TP-92/06.
19            Available from: NTIS, Springfield, VA; PB93-182418.
20     Agency for Toxic Substances and Disease Registry. (1995) Multisite lead and cadmium exposure study with
21            biological markers incorporated. Final report. Atlanta, GA: U.S. Department of Health and Human
22            Services.
23     Aguilera de Benzo, Z.; Fraile, R.;  Carrion, N.; Loreto, D. (1989) Determination of lead in whole blood by
24            electrothermal atomisation atomic absorption spectrometry using tube and platform atomisers and dilution
25            with triton X-100. J. Anal. At. Spectrom. 4: 397-400.
26     Ahlgren, L.; Liden, K.; Mattsson,  S.; Tejning, S. (1976) X-ray  fluorescence analysis of lead in human skeleton in
27            vivo. Scand. J. Work Environ. Health 2: 82-86.
28     Akaike, H. (1973) Information theory and an  extension of the maximum likelihood principle. In: Petrov, B. N.;
29            Csaki, F., eds. 2nd International symposium on information theory; September 1971; Tsahkadsor, Armenia,
30            USSR. Budapest, Hungary: Akademiai Kiado; pp. 267-281.
31     Al-Ashban, R. M.; Aslam, M.; Shah, A. H. (2004) Kohl  (surma): a toxic traditional eye cosmetic study in Saudi
32            Arabia. Public Health. 118: 292-298.
33     Al-Hakkak, Z. S.; Hamamy, H. A.; Murad, A. M.; Hussain, A. F. (1986) Chromosome aberrations in workers at a
34            storage battery plant in Iraq. Mutat. Res. 171: 53-60.
35     Al-Neamy, F. R.; Almehdi, A. M.; Alwash, R.; Pasha, M. A. H.; Ibrahim, A.; Bener, A. (2001) Occupational lead
36            exposure and amino acid profiles and liver function tests in industrial workers. Int. J. Environ. Health Res.
37            11: 181-188.
38     Al-Saleh, I. A. (1995) Lead exposure in Saudi Arabia and its relationship to smoking. BioMetals 8: 243-245.
39     Al-Saleh, I.; Khalil, M. A.; Taylor, A. (1995)  Lead, erythrocyte protoporphyrin, and hematological parameters in
40            normal maternal  and umbilical cord blood from subjects of the Riyadh region, Saudi Arabia. Arch.
41            Environ. Health 50: 66-73.
42     Alessio, L.; Bertazzi, P. A.; Monelli, O.; Toffoletto, F. (1976) Free erythrocyte protoporphyrin as an indicator of the
43            biological effect of lead in adult males. III. Behavior of free erythrocyte protoporphyrin in workers with
44            past lead exposure. Int. Arch. Occup. Environ. Health 38: 77-86.
45     Alessio, L.; Castoldi, M. R.; Buratti, M.; Maroni, M.; Bertazzi, P. A. (1977) Behaviour of some indicators of
46            biological effect in female lead workers. Int. Arch. Occup. Environ. Health 40: 283-292.
47     Alexander, B. H.; Checkoway, H.; VanNetten, C.; Muller, C. H.; Ewers, T. G.; Kaufman, J. D.; Mueller, B. A.;
48            Vaughan, T. L.; Faustman, E. M. (1996a) Semen quality of men employed at a lead smelter. Occup.
49            Environ. Med. 53: 411-416.
50     Alexander, B. H.; Checkoway, H.; Van Netten, C.; Kaufman, J. D.; Vaughan, T. L.; Mueller, B. A.; Faustman, E. M.
51            (1996b) Paternal Occupational Lead Exposure and Pregnancy Outcome. Int. J. Occup. Environ. Health 2:
52            280-285.
53     Alexander, B. H.; Checkoway, H.; Faustman, E. M.; Van Netten C.; Muller,  C. H.; Ewers, T. G. (1998) Contrasting
54            associations of blood and semen lead concentrations with semen quality among lead smelter workers.
55            Am. J. Ind. Med. 34: 464-469.


       December 2005                               6-355         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Alfven, T.; Jarup, L.; Blinder, C.-G. (2002) Cadmium and lead in blood in relation to low bone mineral density and
  2             tubular proteinuria. Environ. Health Perspect. 110: 699-702.
  3     Alomran, A. H.; Shleamoon, M. N. (1988) The influence of chronic lead exposure on lymphocyte proliferative
  4             response and immunoglobulin levels in storage battery workers. J. Biol. Sci. Res. 19: 575-585.
  5     Altmann, L.; Weinsberg, F.; Sveinsson, K.; Lilienthal, H.; Wiegand, H.; Winneke, G. (1993) Impairment of long-
  6             term potentiation and learning following chronic lead exposure. Toxicol. Lett. 66: 105-112.
  7     Altmann, L.; Sveinsson, K.; Kramer, U.; Winneke, G.; Wiegand, H. (1997) Assessment of neurophysiologic and
  8             neurobehavioral effects of environmental pollutants in 5- and 6-year-old children. Environ. Res.
  9             73: 125-131.
10     Altmann, L.; Sveinsson, K.; Kramer, U.; Weishoff-Houben, M.; Turfeld, M.; Winneke, G.; Wiegand, H. (1998)
11             Visual functions in 6-year-old children in relation to lead and mercury levels. Neurotoxicol. Teratol.
12             20:9-17.
13     American Educational Research Association, American Psychological Association, National Council on
14             Measurement in Education (1999). Standards for Educational and Psychological Testing. Washington, DC:
15             American Psychological Association.
16     Anderson, L. A., Jr. (1995) A review of blood lead results from the third National Health and Nutrition Examination
17             Survey (NHANES III). Am. Ind. Hyg. Assoc. J. 56: 7-8.
18     Anderson etal. (1991)
19     Anetor, J. L; Adeniyi, F. A. A. (1998) Decreased immune status in Nigerian workers occupationally exposed to lead.
20             Afr. J. Med. Med. Sci. 28: 169-172.
21     Angell, N. F.; Lavery, J. P. (1982) The relationship of blood levels to obstetric outcome. Am. J. Obstet. Gynecol.
22             142:40-46.
23     Angle, C. R.; Kuntselman, D. R. (1989) Increased erythrocyte protoporphyrins and blood lead~a pilot study of
24             childhood growth patterns. J. Toxicol. Environ. Health. 26:  149-156.
25     Annesi-Maesano, L; Pollitt, R.; King, G.; Bousquet, J.; Hellier, G.; Sahuquillo, J.; Huel, G. (2003) Inutero exposure
26             to lead and cord blood total IgE. Is there a connection? Allergy 58: 589-594.
27     Anttila, A.; Heikkila, P.; Pukkala, E.; Nykyri, E.; Kauppinen, T.; Hernberg, S; Hemminki, K. (1995) Excess lung
28             cancer among workers exposed to lead. Scand. J. Work Environ. Health. 21: 460-469.
29     Anttila, A.; Heikkila, P.; Nykyri, E.; Kauppinen, T.; Pukkala, E.; Hernberg, S.; Hemminki, K. (1996) Risk of
30             nervous system cancer among workers exposed to lead. J. Occup. Environ. Med. 38: 131-136.
31     Apostoli, P.; Maranelli, G.; Dei Cas, L.; Micciolo, R. (1990) Blood lead and blood pressure:  a cross sectional study
32             in a general population group.  Cardiologia 3 5: 597-603.
33     Apostoli, P.; Maranelli, G.; Micciolo, R. (1992) Is hypertension a confounding factor in the assessment of blood lead
34             reference values? Sci. Total Environ. 120: 127-134.
35     Apostoli, P.; Kiss, P.; Porru, S.; Bonde, J. P.; Vanhoorne, M.; the ASCLEPIOS study group. (1998) Male
36             reproductive toxicity of lead in animals and humans. Occup. Environ. Med. 55: 364-374.
37     Apostoli, P.; Corulli, A.; Metra, M.; Dei Cas, L. (2004) Piombo e cardiopatie [Lead and cardiopathy]. Med.
38             Lav. 95: 124-132.
39     Araki, S.; Aono, H.; Yokoyama, K.; Murata, K. (1986) Filterable plasma concentration, glomerular filtration,
40             tubular balance, and renal clearance of heavy metals and organic substances in metal workers. Arch.
41             Environ. Health 41: 216-221.
42     Araki, S.; Sata, F.; Murata, K. (1990) Adjustment for urinary flow rate: an improved approach to biological
43             monitoring. Int. Arch. Occup. Environ. Health 62: 471 477.
44     Armon, C.; Kurland, L. T.; Daube, J. R.; Obrien, P. C. (1991) Epidemiologic correlates of sporadic amyotrophic
45             lateral sclerosis. Neurology 41: 1077-1084.
46     Arnvig, E.; Grandjean, P.; Beckmann, J. (1980) Neurotoxic effects of heavy lead exposure determined with
47             psychological tests. Toxicol. Lett. 5: 399-404.
48     Aro, A. C.; Todd, A. C.; Amarasiriwardena, C.; Hu, H. (1994) Improvements in the calibration of 109Cd K x-ray
49             fluorescence systems for measuring bone lead in vivo. Phys. Med. Biol. 39: 2263-2271.
50     Aro, A.; Amarasiriwardena, C.; Lee, M. L.; Kim, R.; Hu, H. (2000) Validation of K x-ray fluorescence bone lead
51             measurements by inductively coupled plasma mass spectrometry in cadaver legs. Med. Phys. 27: 119-123.
52     Arriada-Mendioca, N.; Rios-Castaneda, C.; Otero-Siliceo, E.; Corona-Vazquez, T. (2000) Amyotrophic lateral
53             sclerosis in a secluded region in Mexico possibly related to lead toxicity. Arch. Neurocien. 5: 2-5.
54     Aschengrau, A.; Zierler, S.; Cohen, A. (1993) Quality of community drinking water and the occurrence of late
55             adverse pregnancy outcomes. Arch. Environ. Health 48: 105-113.
       December 2005                                6-356         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Assennato, G.; Paci, C.; Baser, M. E.; Molinini, R.; Candela, R. G.; Altamura, B. M; Giorgino, R. (1986) Sperm
  2             count suppression without endocrine dysfunction in lead-exposed men. Arch. Environ. Health 41: 387-390.
  3     Assennato, G.; Baser, M.; Molinini, R.; Candela, R. G.; Altamura, B. M.; Giorgino, R.; Abbaticchio, G.; Paci, C.
  4             (1987) Sperm count suppression without endocrine dysfunction in lead-exposed men. Arch. Environ.
  5             Health 42: 124-127.
  6     Audesirk, G. (1985) Effects of lead exposure on the physiology of neurons. Prog. Neurobiol. 24: 199-231.
  7     Audesirk, G.; Shugarts, D.; Nelson,  G.; Przekwas, J. (1989) Organic and inorganic lead inhibit neurite growth in
  8             vertebrate and invertebrate neurons in culture. In Vitro Cell Dev. Biol. 25: 1121-1128.
  9     Auger, J; Kunstmann,  J. M.; Czyglik, F.; Jouannet, P. (1995) Decline in semen quality among fertile men in Paris
10             during the past 20 years. N. Engl. J.  Med. 332: 281-285.
11     Awad El Karim, M. A.; Hamed, A. S.; Elhaimi, Y. A.; Osman, Y. (1986) Effects of exposure to lead among lead-
12             acid battery factory workers in Sudan. Arch. Environ. Health 41: 261 -265.
13     Axelson, O.; Steenland, K. (1988) Indirect methods of assessing the effects of tobacco use in occupational studies.
14             Am. J. Ind. Med. 13: 105-118.
15     Ayatollahi, M. (2002)  Study of the impact of blood lead level on humoral immunity in humans. Toxicol. Ind. Health
16             18:39-44.
17     Azar, A.; Snee, R. D.;  Habibi, K.  (1975) An epidemiologic approach to community air lead exposure using personal
18             samplers. In:  Griffin, T. B.; Knelson, J. H., eds. Lead. Stuttgart, Federal Republic of Germany: Georg
19             Thieme Publishers; pp. 254-290. (Coulston, F.; Korte, F., eds. Environmental quality and safety:
20             supplement v. 2).
21     Azcona-Cruz, M. I.; Rothenberg,  S.  J.; Schnaas-Arrieta, L.; Romero-Placeres, M.; Perroni-Hernandez, E. (2000)
22             [Levels of plasmatic lead in children 8-10 years of age and its relation to changes in visual-motor system
23             and balance]. Salud Publica Mex. 42: 279-287.
24     Baer, R. D.; Garcia de Alba, J.; Mares Leal, R.; Plascencia Campos, A. R.; Goslin, N. (1998) Mexican use of lead in
25             the treatment of empacho: community, clinic, and longitudinal patterns. Soc. Sci. Med. 47: 1263-1266.
26     Bagci, C.; Bozkurt, A. I.; Cakmak, E. A.; Can,  S.; Cengiz, B. (2004) Blood lead levels of the battery and exhaust
27             workers and their pulmonary function tests. Int. J. Clin. Pract. 58: 568-572.
28     Baghurst, P. A. (1995) Getting the lead out...  Neurotoxicol. Teratol.  17: 213-214.
29     Baghurst, P. A.; McMichael, A. J.; Wigg, N.  R.; Vimpani, G. V.; Robertson, E. F.; Roberts, R. J.; Tong, S.-L.
30             (1992) Environmental exposure to lead and children's intelligence at the age of seven years: the Port Pirie
31             cohort study.  N. Engl. J.  Med.  327: 1279-1284.
32     Baghurst, P. A.; McMichael, A. J.; Tong, S.;  Wigg, N. R.; Vimpani, G. V; Robertson, E. F. (1995) Exposure to
33             environmental lead and visual-motor integration at age 7 years: the Port Pirie cohort study. Epidemiology
34             6: 104-109.
35     Baghurst, P. A.; Tong, S.; Sawyer, M. G.; Burns, J.; McMichael, A. J. (1999) Sociodemographic and behavioural
36             determinants  of blood lead concentrations in children aged 11-13 years. The Port Pirie Cohort Study. Med.
37             J. Aust. 170: 63-67.
38     Bairati, C.; Goi, G.; Bollini, D.; Roggi,  C.; Luca, M.; Apostoli, P.; Lombardo, A. (1997) Effects of lead and
39             manganese on the release of lysosomal enzymes in vitro and in vivo. Clin. Chim. Acta 261: 91-101.
40     Baird, D. D.; Wilcox, A. J.; Weinberg, C. R.  (1986) Use of time to pregnancy to study environmental exposures.
41             Am. J. Epidemiol. 124: 470-480.
42     Baker, E. L., Jr.; Landrigan, P. J.; Barbour, A. G.; Cox, D. H.; Folland, D. S.; Ligo, R. N.; Throckmorton, J. (1979)
43             Occupational lead poisoning in the United States: clinical and biochemical findings related to blood lead
44             levels. Br. J. Ind. Med. 36:  314-322.
45     Balbus, J. M.; Stewart, W.; Bolla, K. I.; Schwartz, B. S. (1997) Simple visual reaction time in organolead
46             manufacturing workers: comparison of different methods of modeling lead exposure and reaction time.
47             Am. J. Ind. Med. 32: 544-549.
48     Balbus, J. Stewart, W.; Bolla, K. I.;  Schwartz, B. S. (1998) Simple visual reaction time in organolead manufacturing
49             workers: influence of the interstimulus interval. Arch. Environ. Health 53: 264-270.
50     Balbus-Kornfeld, J. M.; Stewart, W.; Bolla, K.  I.; Schwartz, B. S. (1995) Cumulative exposure to inorganic lead and
51             neurobehavioural test performance in adults: an epidemiological review. Occup. Environ. Med. 52: 2-12.
52     Ball, G. V.;  Sorensen,  L. B. (1969) Pathogenesis of hyperuricemia in saturnine  gout. N. Engl. J. Med.
53             280: 1199-1202.
54     Ballard,  J. L.; Novak, K. K.; Driver, M. (1979) A simplified score for assessment of fetal maturation of newly born
55             infants. J. Pediatr. 95: 769-774.
       December 2005                                6-357         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Baloh, R. W.; Spivey, G. H.; Brown, C. P.; Morgan, D.; Campion, D. S.; Browdy, B. L.; Valentine, J. L. (1979)
  2             Subclinical effects of chronic increased lead absorption - a prospective study. II. Results of baseline
  3             neurologic testing. J. Occup. Med. 21: 490-496.
  4     Banks, E. C.; Ferretti, L. E.; Shucard, D. W. (1997) Effects of low level lead exposure on cognitive function in
  5             children: a review of behavioral, neuropsychological and biological evidence. Neurotoxicology
  6             18:237-281.
  7     Barltrop, D. (1968) Lead poisoning in childhood. Postgrad. Med. J. 44: 537-542.
  8     Barry, P. S. I. (1975) A comparison of concentrations of lead in human tissues. Br. J. Ind. Med. 32: 119-139.
  9     Basaran, N.; Undeger, U. (2000) Effects of lead on immune parameters in occupationally exposed workers. Am. J.
10             Ind. Med. 38: 349-354.
11     Bates, M.; Malcolm, M.; Wyatt, R.; Garrett, N.; Galloway, Y.; Speir, T.; Read, D. (1995) Lead in children from
12             older housing areas in the Wellington region. N. Z. Med. J. 108: 400-404.
13     Battistuzzi, G.; Petrucci, R.; Silvagni, L.; Urbani, F. R.; Caiola, S. (1981) "Delta"-aminolevulinate dehydrase: a new
14             genetic polymorphism in man. Ann. Hum. Genet. 45: 233-229.
15     Batuman, V.  (1993) Lead nephropathy, gout, and hypertension. Am. J. Med. Sci.  305: 241-247.
16     Batuman, V.; Maesaka, J. K.; Haddad, B.; Tepper,  E.; Landry, E.; Wedeen, R. P. (1981) The role of lead in gout
17             nephropathy. N. Engl. J. Med. 304: 520-523.
18     Batuman, V.; Landy, E.; Maesaka, J. K.; Wedeen, R. P. (1983) Contribution of lead to hypertension with renal
19             impairment. N. Engl. J. Med. 309: 17-21.
20     Bauchinger, M.; Dresp, J.;  Schmid, E.; Englert, N.; Krause, Chr. (1977) Chromosome analyses of children after
21             ecological lead exposure.  Mutat. Res. 56:  75-80.
22     Begerow, J.; Freier, I.; Turfeld, M.; Kramer, U.; Dunemann, L. (1994) Internal lead and cadmium exposure in
23             6-year-old children from western and eastern Germany. Int. Arch. Occup. Environ. Health 66: 243-248.
24     Behringer, D.; Craswell, P.; Mohl,  C.; Stoeppler, M.; Ritz, E. (1986) Urinary lead excretion in uremic patients.
25             Nephron 42:  323-329.
26     Bekkelman, L; Pfister, E. (2001) Neirotoksicheskie effekty mnogoletnei ekspozitsii svintsom [Neurotoxicity due to
27             long-standing exposure to lead]. Med. Tr.  Prom. Ekol. (5): 22-26.
28     Bellinger, D. (1995) Neuropsychologic function in children exposed to environmental lead (comment).
29             Epidemiology 6: 101-103.
30     Bellinger, D. C. (1995) Interpreting the literature on lead and child development: the neglected role of the
31             "experimental system." Neurotoxicol. Teratol. 17: 201-212.
32     Bellinger, D. C. (2000) Effect modification in epidemiologic studies of low-level neurotoxicant exposures and
33             health outcomes. Neurotoxicol. Teratol. 22: 133-140.
34     Bellinger, D. (2002) Perspectives on  incorporating human neurobehavioral end points in risk assessments. Risk
35             Anal. 22: 487-498.
36     Bellinger, D. C. (2003) Perspectives on incorporating human neurobehavioral end points in risk assessments. Risk
37             Anal. 23:  163-174.
38     Bellinger, D. C. (2004) Confounded by confounding. Epidemiology 15: 383-384.
3 9     Bellinger, D. C. (2005) Teratogen update: lead and pregnancy. Birth Defects Res. Part A 73: 409-420.
40     Bellinger, D.; Dietrich, K. N. (1994) Low-level lead exposure and cognitive function in children. Pediatr. Ann.
41             23:600-605.
42     Bellinger, D. C.; Needleman, H. L. (2003) Intellectual impairment and blood lead levels [letter]. N. Engl. J. Med.
43             349:500.
44     Bellinger, D.; Rappaport, L. (2002) Developmental assessment and interventions. In: Managing elevated blood lead
45             levels among young children: recommendations from the Advisory Committee on Childhood Lead
46             Poisoning Prevention. Atlanta, GA: Centers for Disease Control; pp. 79-95.
47     Bellinger, D.; Needleman, H. L.; Bromfield, R.; Mintz, M. (1984) A followup study of the academic attainment and
48             classroom behavior of children with elevated dentine lead levels. Biol. Trace Elem. Res. 6: 207-223.
49     Bellinger, D.; Leviton, A.; Waternaux, C.; Allred, E. (1985) Methodological issues in modelling the relationship
50             between low-level lead exposure and infant development: examples from the Boston lead study. In:
51             Bornschein, R. L.; Rabinowitz, M. B., eds. The second international conference on prospective studies of
52             lead; April 1984; Cincinnati, OH. Environ. Res. 38: 119-129.
53     Bellinger, D.; Leviton, A.; Needleman, H. L.; Waternaux, C.; Rabinowitz, M.  (1986) Low-level lead exposure and
54             infant development in the first year. Neurobehav.  Toxicol. Teratol. 8: 151-161.
55     Bellinger, D.; Leviton, A.; Rabinowitz, M.; Needleman, H.; Waternaux, C. (1986) Correlates of low-level lead
5 6             exposure in urban children at 2 years of age. Pediatrics 77: 826-83 3.
       December 2005                                6-358        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Bellinger, D.; Leviton, A.; Waternaux, C.; Needleman, H.; Rabinowitz, M. (1987) Longitudinal analyses of prenatal
  2             and postnatal lead exposure and early cognitive development. N. Engl. J. Med. 316: 1037-1043.
  3     Bellinger, D.; Leviton, A.; Waternaux, C.; Needleman, H.; Rabinowitz, M. (1988) Low-level lead exposure, social
  4             class, and infant development. Neurotoxicol. Teratol. 10: 497-503.
  5     Bellinger, D.; Leviton, A.; Waternaux, C. (1989) Lead, IQ and social class. Int. J. Epidemiol. 18: 180-185.
  6     Bellinger, D.; Leviton, A.; Waternaux, C.; Needleman, H.; Rabinowitz, M. (1989) Low-level lead exposure, social
  7             class, and infant development. Neurotoxicol. Teratol. 10: 497-503.
  8     Bellinger, D.; Leviton, A.; Sloman, J. (1990) Antecedents and correlates of improved cognitive performance in
  9             children exposed in utero to low levels of lead. Environ. Health Perspect. 89: 5-11.
10     Bellinger, D.; Leviton, A.; Rabinowitz, M.; Allred, E.; Needleman, H.; Schoenbaum, S. (1991) Weight gain and
11             maturity in fetuses exposed to low levels of lead. Environ. Res. 54:  151-158.
12     Bellinger, D.; Sloman, J.; Leviton, A.; Rabinowitz, M.; Needleman, H. L.; Waternaux, C. (1991) Low-level lead
13             exposure and children's cognitive function in the preschool years. Pediatrics 87: 219-227.
14     Bellinger, D. C.; Stiles, K. M.; Needleman, H. L. (1992) Low-level lead exposure, intelligence and academic
15             achievement: a long-term follow-up study. Pediatrics 90: 855-861.
16     Bellinger, D.; Hu, H.; Titlebaum, L.; Needleman, H.  L. (1994) Attentional correlates of dentin and bone lead levels
17             in adolescents. Arch. Environ. Health 49: 98-105.
18     Bellinger, D.; Leviton, A.; Allred, E.; Rabinowitz, M. (1994) Pre- and postnatal lead exposure  and behavior
19             problems in school-aged children. Environ.  Res. 66: 12-30.
20     Bellinger et al. (2003)
21     Bellinger, D. C.; Hu, H.; Kalaniti, K.; Thomas, N.; Rajan, P.; Sambandam, S.; Ramaswamy, P.; Balakrishnan, K.
22             (2005) A pilot study of blood lead levels and neurobehavioral function in children living in Chennai, India.
23             Int. J. Occup. Environ. Health 11: 138-143.
24     Benetou-Marantidou, A.; Nakou, S.; Micheloyannis,  J. (1988) Neurobehavioral estimation of children with life-long
25             increased lead exposure. Arch. Environ. Health 43: 392-395.
26     Benoff, S.; Hurley, I. R.; Millan,  C.; Napolitano, B.;  Centola, G. M. (2003a)  Seminal lead concentrations negatively
27             affect outcomes of artificial insemination. Fertil. Steril. 80: 517-525.
28     Benoff, S.; Centola, G. M. (2003b) Increased seminal plasma lead levels adversely affect the fertility potential of
29             sperm in IVF. Hum. Reprod. 18: 374-383.
30     Bercovitz, K.; Laufer, D.  (1991) Age and gender influence on lead accumulation in root dentine of human
31             permanent teeth. Arch. Oral Biol. 36: 671-673.
32     Bercovitz, K.; Laufer, D.  (1993) Carious teeth as indicators to lead exposure. Bull. Environ. Contam. Toxicol.
33             50:724-729.
34     Bergdahl, I. A.; Skerfving, S. (1997) Partition of circulating lead between plasma and red cells does  not seem to be
35             different for internal and external sources of lead [letter]. Am. J. Ind. Med. 32: 317-318.
36     Bergdahl, I. A.; Gerhardsson, L.;  Schutz, A.; Desnick, R. J.; Wetmur, J. G.; Skerfving, S. (1997) Delta-
37             aminolevulinic acid dehydratase polymorphism: influence on lead levels and kidney function in humans.
38             Arch. Environ. Health 52: 91-96.
39     Bergdahl, I. A.; Schutz, A.; Gerhardsson, L.; Jensen, A.; Skerfving, S. (1997) Lead concentrations in human plasma,
40             urine and whole blood. Scand. J. Work Environ. Health 23:  359-363.
41     Bergdahl, I. A.; Sheveleva, M.; Schutz, A.; Artamonova,  V. G.;  Skerfving, S. (1998) Plasma and blood lead in
42             humans: capacity-limited binding to "delta"-aminolevulinic acid dehydratase and other lead-binding
43             components. Toxicol. Sci. 46: 247-253.
44     Bergdahl, I. A.; Vahter, M.; Counter, S. A.; Schutz, A.; Buchanan, L. H.; Ortega, F.; Laurell, G.; Skerfving, S.
45             (1999) Lead in plasma and whole blood from lead-exposed children. Environ. Res. 80: 25-33.
46     Berger, O. G.; Gregg, D. J.; Succop, P. A. (1990) Using unstimulated urinary lead excretion to assess the need for
47             chelation in the treatment of lead poisoning. J. Pediatr.  116: 46-51.
48     Bergeret, A.; Pouget, E.; Tedone, R.; Meygert, T.; Cadot, R.; Descotes, J. (1990) Neutrophil functions in lead-
49             exposed workers. Hum. Exp. Toxicol. 9: 231-233.
50     Berkowitz, G. S.; Wolff, M. S.; Lapinski, R. H.; Todd, A. C. (2004) Prospective study of blood and tibia lead in
51             women undergoing surgical menopause. Environ. Health Perspect.  112: 1673-1678.
52     Bernard. (1999)
53     Bernard, A. (2004) Renal dysfunction induced by cadmium: biomarkers of critical effects. Biometals 17: 519-523.
54     Bernard, A. M.; Vyskocil, A.; Roels, H.; Kriz, J.; Kodl, M.; Lauwerys, R. (1995) Renal effects in children living in
55             the vicinity of a lead smelter. Environ. Res.  68: 91-95.
       December 2005                                6-359         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Bernard, A.; Thielemans, N.; Roels, H.; Lauwery s, R. (1995) Association between NAG-B and cadmium in urine
  2             with no evidence of a threshold. Occup. Environ. Med. 52: 177-180.
  3     Bhattacharya, A.; Shukla, R.; Bornschein, R.; Dietrich, K.; Kopke, J. E. (1988) Postural disequilibrium
  4             quantification in children with chronic lead exposure: a pilot study. Neurotoxicology 9: 327-340.
  5     Bhattacharya, A.; Shukla, R.; Bornschein, R. L.; Dietrich, K. N.; Keith, R. (1990) Lead effects on postural balance
  6             of children. In: Conference on advances in lead research: implications for environmental health; January
  7             1989; Research Triangle Park, NC. Environ. Health Perspect. 89: 35-42.
  8     Bhattacharya, A.; Shukla, R.; Dietrich, K.; Bornschein, R.; Berger, O. (1995) Effect of early lead exposure on
  9             children's postural balance. Dev. Med. Child Neurol. 37: 861-878.
10     Bhattacharya, A.; Shukla, R.; Dietrich, K. N.; Miller, J.; Bagchee, A.; Bornschein, R. L.; Cox, C.; Mitchell, T.
11             (1993) Functional implications of postural disequilibrium due to lead exposure. Neurotoxicology
12             14: 179-189.
13     Biagini, G.; Caudarella, R.; Vangelista, A. (1977) Renal morphological and functional modification in chronic lead
14             poisoning. In: Brown, S. S., ed. Clinical chemistry and chemical toxicology of metals. New York, NY:
15             Elsevier/North-Holland Biomedical Press; pp. 123-126.
16     Billick, I. H.; Curran, A. S.; Shier, D. R. (1979) Analysis of pediatric blood lead levels in New York City for
17             1970-1976. Environ. Health Perspect. 31: 183-190.
18     Binder, S.; Matte, T. (1993) Childhood lead poisoning: the impact of prevention. JAMA J. Am. Med. Assoc.
19             269: 1679-1681.
20     Bleecker, M.; Bolla-Wilson, K.; Kawas, C.; Agnew, J. (1988) Age-specific norms for the mini-mental state exam.
21             Neurology 38: 1565-1568.
22     Bleecker, M. L.; Lindgren, K. N.; Ford, D. P. (1997a) Differential contribution of current and cumulative indices of
23             lead dose to neuropsychological performance by age. Neurology 48: 639-645.
24     Bleecker, M. L.; Lindgren, K. N.; Tiburzi, M. J.; Ford, D. P. (1997b) Curvilinear relationship between blood lead
25             level and reaction time. Differential association with blood lead fractions derived from exogenous and
26             endogenous sources. J. Occup. Environ. Med. 39: 426-431.
27     Bleecker, M. L.; Lindgren, K. N.; Ford, D. P.; Tiburzi, M. J. (2002) The interaction of education and cumulative
28             lead exposure on the mini-mental state examination. J. Occup. Environ. Med. 44: 574-578.
29     Bleecker, M. L.; Ford, D. P.; Lindgren, K. N.; Hoese, V. M.; Walsh, K. S.; Vaughan, C. G. (2005a) Differential
30             effects of lead exposure on components of verbal memory. Occup. Environ. Med. 62:  181-187.
31     Bleecker, M. L.; Ford, D. P.; Baughan, C. G.; Lindgren, K.  N.; Tiburzi, M. J.; Walsh, K. S. (2005b) Effect of lead
32             exposure and ergonomic stressors on peripheral  nerve function. Environ. Health Perspect.  113: 1730-1734.
33     Bogden, J. D.; Thind, I. S.; Louria, D. B.; Caterini, H. (1978) Maternal and cord blood metal concentrations and low
34             birth weight~a case-control study. Am. J. Clin. Nutr. 31: 1181 -1187.
3 5     Bogden, J. D.; Oleske, J. M.; Louria, D. B. (1997) Lead poisoning-one approach to a problem that won't go away.
36             Environ. Health Perspect. 105: 1284-1287.
37     Boivin, M. J.; Giordani, B. (1995) A risk evaluation of the neuropsychological effects of childhood lead toxicity.
38             Dev. Neuropsychol. 11:  157-180.
39     Bonde, J. P. E.; Kolstad, H. (1997) Fertility of Danish battery workers exposed to lead. Int. J. Epidemiol.
40             26: 1281-1288.
41     Bonde, J. P.; Joffe, M.; Apostoli, P.; Dale, A. Kiss, P.; Spano, M.; Caruso, F.; Giwercman, A.; Bisanti, L.; Porru, S.;
42             Vanhoorne, M.; Comhaire, F.; Zschiesche, W. (2002) Sperm count and chromatin structure in men exposed
43             to inorganic lead: lowest adverse effect levels. Occup. Environ. Med. 59: 243-242.
44     Borja-Aburto, V. H.; Hertz-Picciotto, I.; Lopez, M.  R.; Farias, P.; Rios, C.; Blanco, J. (1999) Blood lead levels
45             measured prospectively and risk of spontaneous abortion. Am. J. Epidemiol. 150: 590-597.
46     Bornschein, R. L.; Rabinowitz, M. B. (1985) The second international conference on prospective studies of lead -
47             foreword. Environ. Res.  38: 1-2.
48     Bornschein, R. L.; Grote, J.; Mitchell, T., Succop. P. A.; Dietrich, K. N.; Krafft, K. M.; Hammond, P. B. (1989)
49             Effects of prenatal lead exposure on infant size at birth. In: Smith, M. A.; Grant, L. D.; Sors, A. I., eds.
50             Lead exposure and child development: an international assessment  [workshop organized by the
51             Commission of the European Communities and the U.S. Environmental Protection Agency]; September
52             1986; Edinburgh, United Kingdom. Dordrecht, The Netherlands: Kluwer Academic Publishers BV;
53             pp. 307-319.
54     Bornschein et al. (1990)
       December 2005                                6-360         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Bos, A. J. I; Van der Stop, C. C. A. H.; Valkovic, V.; Vis, R. D.; Verheul, H. (1985) Incorporation routes of
  2            elements into human hair: implications for hair analysis used for monitoring. Sci.Total Environ.
  3            42: 157-169.
  4     Boscolo, P.; De Gioacchino, M; Sabbioni, E.; Di Giacomo, F.; Reale, M; Volpe, A. R.; Di Sciascio, M. B.;
  5            Conti, P.; Giuliano, G. (2000) Lymphocyte subpopulations, cytokines and trace elements in asymptomatic
  6            atopic women exposed to an urban environment. Life Sci. 67: 1119-1126.
  7     Bost, L.; Primatesta, P.; Dong, W.; Poulter, N. (1999) Blood lead and blood pressure: evidence from the Health
  8            Survey for England 1995. J. Hum. Hypertens. 13: 123-128.
  9     Bound, J.  P.; Harvey, P. W.; Francis, B. J.; Awwad, F.; Gatrell, A.  C. (1997) Involvement of deprivation and
10            environmental lead in neural tube defects: a matched case-control study. Arch. Dis. Child. 76:  107-112.
11     Bourgoin, B. P.; Evans, D. R.; Cornell, J. R.; Lingard, S. M.; Quattrone, A. J. (1993) Lead content in 70 brands of
12            dietary calcium supplements. Am. J. Public Health 83: 1155-1160.
13     Bradbury, M. W. (1992) An approach to study of transport of trace metals at the blood-brain barrier. Prog. Brain
14            Res. 91: 133-138.
15     Bradman, A.; Eskenazi, B.; Sutton, P.; Athanasoulis,  M.; Goldman, L. R.  (2001) Iron deficiency associated with
16            higher blood lead in children living in contaminated environments. Environ. Health Perspect.
17            109: 1079-1084.
18     Braun, C. M. J.; Daigneault, S. (1991) Sparing of cognitive executive functions and impairment of motor functions
19            after industrial exposure to lead: a field study with control group. Neuropsychology, 5: 179-193.
20     Braunstein, G. D.; Dahlgren, J.; Loriaux, D. L. (1978) Hypogonadism in chronically lead-poisoned men. Infertility
21            1:33-51.
22     Bressler, J. P.; Goldstein, G. W. (1991) Mechanisms  of lead neurotoxicity. Biochem. Pharmacol. 41: 479-484.
23     Brito, J. A. A.; McNeill, F. E.; Stronach, I.; Webber,  C. E.; Wells, S.; Richard, N.; Chettle, D. R. (2001)
24            Longitudinal changes in bone lead concentration: implications for modelling of human bone lead
25            metabolism. J. Environ. Mo nit. 3: 343-351.
26     Brito, J. A. A.; McNeill, F. E.; Webber, C. E.; Wells, S.; Richard, N.; Carvalho, M. L.; Chettle, D. R. (2002)
27            Evaluation of a novel structural model to describe the endogenous release of lead from bone. J. Environ.
28            Monit. 4: 194-201.
29     Brockhaus, A.; Collet, W.; Dolgner, R.; Engelke, R.;  Ewers, U.; Freier, I.; Jermann, E.; Kramer, U.; Manojlovic, N.;
30            Turfeld, M.; Winneke, G. (1988) Exposure to lead and cadmium of children living in different areas of
31            north-west Germany: results of biological monitoring studies 1982-1986. Int. Arch. Occup. Environ. Health
32            60:211-222.
33     Brody, D. J.; Pirkle, J. L.; Kramer, R. A.; Flegal, K. M.; Matte, T. D.; Gunter, E. W.; Paschal, D. C. (1994) Blood
34            lead levels in the US population: phase 1 of the third National Health and Nutrition Examination Survey
35            (NHANES III, 1988 to 1991). JAMA J. Am. Med. Assoc. 272: 277-283.
36     Brown, A.; Tompsett, S. L. (1945) Poisoning due to mobilization of lead from the skeleton by leukaemic
37            hyperplasia of bone marrow. Br. Med. J. 2: 764-765.
38     Brown, M. J.; Hu, H.; Gonzales-Cossio, T.; Peterson, K. E.; Sanin, L.-H.; de Luz Kageyama, M.; Palazuelos, E.;
39            Aro, A.; Schnaas, L.; Hernandez-Avila, M. (2000) Determinants of bone and blood lead concentrations in
40            the early postpartum period. Occup. Environ. Med. 57: 535-541.
41     Buchet, J. P.; Lauwerys, R.; Roels, H.; Bernard, A.; Bruaux, P.; Claeys, F.; Ducoffre, G.; De Plaen, P.; Staesen, J.;
42            Amery, A.; Linjen, P.; Thijs, L.; Rondia, D.; Sartor, F.; Saint Remy, A.; Nick,  L. (1990) Renal effects of
43            cadmium body burden of the general population. Lancet 336: 699-702.
44     Budd, P.;  Montgomery, J.; Evans, J.; Trickett, M. (2004) Human lead exposure in England from approximately 5500
45            BP to the 16th century AD. Sci. Total Environ. 318: 45-58.
46     Burgstahler, A. W. (2003) Influence of fluoride and lead on children's IQ: U.S. tolerance standards in question.
47            Fluoride 36: 79-81.
48     Burns, J. M.; Baghurst, P. A.; Sawyer, M. G.; McMichael, A. J.; Tong, S.-L. (1999) Lifetime low-level exposure to
49            environmental lead and children's emotional and behavioral development at ages 11-13 years. The Port
50            Pirie cohort study. Am. J. Epidemiol. 149: 740-749.
51     Cake, K. M.; Bowins, R. J.; Vaillancourt, C.;  Gordon, C. L.; McNutt, N. H., Laporte, R.; Webber, C. E.; Chettle,
52            D. R. (1996) Partition of circulating lead between serum and red cells is different for internal and external
53            sources of lead. Am. J. Ind. Med. 29: 440-445.
54     Calderon, J.; Navarro, M. E.; Jimenez-Capdeville, M. E.;  Santos-Diaz, M. A.; Golden, A.; Rodriguez-Leyva, I.;
55            Borja-Aburto, V.; Diaz-Barriga, F. (2001) Exposure to arsenic and lead and neuropsychological
56            development in Mexican children. Environ. Res. 85: 69-76.
       December 2005                                6-361         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Calderon-Salinas, J. V.; Hernandez-Luna, C.; Valdez-Anaya, B.; Maldonado-Vega, M.; Lopez-Miranda, A. (1996)
  2             Evolution of lead toxicity in a population of children. Hum. Exp. Toxicol. 15: 376-382.
  3     Campara, P.; D'Andrea, F.; Micciolo, R.; Savonitto, C.; Tansella, M.; Zimmermann-Tansella, C. (1984)
  4             Psychological performance of workers with blood-lead concentration below the current threshold limit
  5             value. Int. Arch. Occup. Environ. Health 53: 233-246.
  6     Campbell, J. R.; Toribara, T. Y. (2001) Hair-root lead to screen for lead toxicity. J. Trace Elem. Exp. Med.
  7             14:69-72.
  8     Campbell, B.C.; Meredith, P. A.; Scott, J. J. C. (1985) Lead exposure and changes in the renin-angiotensin-
  9             aldosterone system in man. Toxicol. Lett. 25: 25-32.
10     Campbell, J. R.; Moss, M. E.; Raubertas, R. F. (2000) The association between caries and childhood lead exposure.
11             Environ. Health Perspect. 108: 1099-1102.
12     Campbell, T. F.; Needleman, H. L.; Riess, J. A.; Tobin, M. J. (2000) Bone lead levels and language processing
13             performance. Dev. Neuropsychol. 18: 171-186.
14     Canfield, R. L.; Henderson, C. R., Jr.; Cory-Slechta, D. A.; Cox, C.; Jusko, T. A.; Lanphear, B. P. (2003a)
15             Intellectual impairment in children with blood lead  concentrations below 10 micrograms per deciliter.
16             N. Engl. J. Med. 348: 1517-1526.
17     Canfield, R. L.; Kreher, D. A.; Cornwell, C.; Henderson, C. R., Jr. (2003b) Low-level lead exposure, executive
18             functioning, and learning in early childhood. Child Neuropsychol. 9: 35-53.
19     Canfield, R. L.; Gendle, M. H.; Cory-Slechta, D.  A. (2004) Impaired neuropsychological functioning in lead-
20             exposed children. Dev. Neuropsychol. 26: 513-540.
21     Cantarow, A.; Trumper, M. (1944) Lead poisoning. Baltimore, MD: Williams & Wilkins Co.
22     Carbone, R.; Laforgia, N.; Crollo, E.; Mautone, A.; lolascon, A. (1998) Maternal and neonatal lead exposure in
23             southern Italy. Biol. Neonate 73: 362-366.
24     Cardenas, A.; Roels, H.; Bernard, A. M.; Barbon, R.; Buchet, J. P.; Lauwerys, R. R.;  Rosello, J.; Hotter, G.;
25             Mutti, A.; Franchini, I.; Pels, L. M.; Stolte, H.; De Broe, M. E.; Nuyts, G. D.; Taylor, S. A.; Price, R. G.
26             (1993) Markers of early renal changes induced by industrial pollutants.  I. Application to workers exposed
27             to mercury vapor. Br. J. Ind. Med. 50: 17-27.
28     Cardenas, A.; Roels, H.; Bernard, A. M.; Barbon, R.; Buchet, J. P.; Lauwerys, R. R.;  Rosello, J.; Ramis, I.; Mutti,
29             A.; Franchini, I.; Pels, L. M.; Stolte, H.;  De Broe, M.  E.; Nuyts, G. D.; Taylor, S. A.; Price, R. G. (1993)
3 0             Markers of early renal changes induced by industrial pollutants. II. Application to workers exposed to lead.
31             Br. J. Ind. Med.  50:  28-36.
32     Cardozo dos Santos, A.; Colacciopo, S.; Bo, C. M. R. dal; Santos, N. A. G. dos. (1994) Occupational exposure to
33             lead, kidney function tests, and blood pressure. Am. J. Ind. Med. 26: 635-643.
34     Carsia, R. V.; Forman, D.; Hock, C.  E.; Nagele, R. G.; Mcllroy, P. J. (1995) Lead alters growth and reduces
3 5             angiotensin II receptor density of rat aortic smooth muscle cells. Proc. Soc. Exp. Biol. Med. 210: 180-190.
36     Carta, P.; Cocco, P.; Picchiri, G. (1994) Lung cancer mortality and airways obstruction among metal miners exposed
37             to silica and low levels of radon daughters. Am. J. Ind. Med. 25: 489-506.
38     Carta, P.; Am, G.; Cadeddu,  C.; Nieddu, V.; Polizzi, M.; Nurchis, P.; Flore, C.; Salis, S.; Sanna, R. F. (2003)
3 9             Mortalita per cancro polmonare in lavoratori di una fonderia di piombo della Sardegna [Mortality from
40             lung cancer among workers of a Sardinian lead smelter [Follow-up:  1972-2001]].  G. Ital. Med. Lav. Ergon.
41             25(suppl. 3): 17-18.
42     Carta, P.; Am, G.; Nurchis, P.; Cadeddu, C.; Polizzi, M.; Nieddu, V.; Sali, G.; Gaviano, L.; Flore, C.; Sanna, R. F.
43             (2005) Studio di mortalita per cause specifiche in lavoratori di una fonderia di piombo e zinco della
44             Sardegna. G. Ital. Med. Lav. Ergon. 27(suppl. 1): 43-45.
45     Casey, C. E.; Robinson, M. F. (1978) Copper, manganese, zinc, nickel, cadmium and lead in human foetal tissue.
46             Br. J. Nutr. 39: 639-646.
47     Cavalleri, A.; Trimarchi, F.; Gelmi, C.; Baruffini, A.; Minoia, C.; Biscaldi, G.; Gallo, G. (1982) Effects of lead on
48             the visual system of occupationally exposed subjects. Scand. J. Work Environ. Health 8(suppl. 1):  148-151.
49     Cecil, K. M.; Weihong, Y.; Holland, S.; Wessel,  S.; Dietrich, K.; Ris, D.; Lanphear, B. (2005) The influence of
50             childhood lead exposure on language function in young adults: an fMRI study. Presented at: International
51             Society for Magnetic Resonance Imaging: 12th scientific meeting and exhibition;  May; Miami, FL; A1443.
52     Centers for Disease Control and Prevention.  (1991) Preventing lead poisoning in young children: a statement by the
53             Centers for Disease Control. Atlanta, GA: U.S. Department of Health and Human Services; October.
54     Centers for Disease Control and Prevention.  (1993) Lead poisoning associated with use of traditional ethnic
55             remedies. Morb. Mortal. Wkly. Rep. 42: 521-524.
       December 2005                                6-362         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Centers for Disease Control and Prevention. (1995) Blood lead levels among children ~ Rhode Island, 1993-1995.
  2            Morb. Mortal. Wkly. Rpt. MMWR 44: 788-791.
  3     Centers for Disease Control and Prevention. (1995) Blood lead levels among children in a managed-care
  4            organization - California, October 1992-March 1993. Morb. Mortal. Wkly. Rep. 44: 627-635.
  5     Centers for Disease Control and Prevention. (1997) Adult blood lead epidemiology and surveillance - United States,
  6            third quarter,  1996. Morb. Mortal. Wkly Rpt. MMWR 46: 105-107.
  7     Centers for Disease Control and Prevention. (1997) Blood lead levels - United States, 1991-1994. Morb. Mortal.
  8            Wkly. Rep. MMWR 46:  141-146.
  9     Centers for Disease Control and Prevention. (2000) Blood lead levels in young children - United States and selected
10            states, 1996-1999. Morb. Mortal. Wkly. Rep. 49: 1133-1137.
11     Centers for Disease Control and Prevention. (2005) Third national report on human exposure to environmental
12            chemicals. Atlanta, GA: U.S. Department of Health and Human Services, National Center for
13            Environmental Health. NCEH Pub. No. 05-0570.
14     Chalkley, S. R.; Richmond, I; Barltrop, D. (1998) Measurement of vitamin D3 metabolites in smelter workers
15            exposed to lead and cadmium. Occup. Environ. Med. 55: 446-452.
16     Chamberlain, A. C.; Heard, M. I; Little, P.; Newton, D.; Wells, A. C.; Wiffin, R. D. (1978) Investigations into lead
17            from motor vehicles. Harwell, United Kingdom: United Kingdom Atomic Energy Authority; report no.
18            AERE-R9198.
19     Chancellor, A. M.; Slattery, J. M.; Fraser, H.; Warlow, C. P. (1993) Risk factors  for motor neuron disease: a case-
20            control study  based on patients from the Scottish Motor Neuron Disease Register. J. Neurol. Neurosurg.
21            Psychiatry 56: 1200-1206.
22     Chaube, S.; Swinyard, C. A.; Nishimura, H. (1972) A quantitative study of human embryonic and fetal lead with
23            considerations of maternal fetal lead gradients and the effect of lead on  human reproduction. Teratology
24            5:253.
25     Chen, A.; Dietrich, K.  N.; Ware, J. H.; Radcliffe, J.; Rogan,  W. J. (2005) IQ and blood lead from 2 to 7 years of age:
26            are the effects in older children the residual of high blood lead concentrations in 2-year-olds? Environ.
27            HealthPerspect. 113: 597-601.
28     Cheng, Y.; Willett, W. C.; Schwartz, J.; Sparrow, D.; Weiss, S.; Hu, H.  (1998a) Relation of nutrition to bone lead
29            and blood lead levels in middle-aged to elderly men. The Normative Aging Study. Am. J. Epidemiol.
30            147: 1162-1174.
31     Cheng, Y.; Schwartz, J.; Vokonas, P. S.;  Weiss, S. T.; Aro, A.; Hu, H. (1998b) Electrocardiographic conduction
32            disturbances in association with low-level lead exposure (the Normative Aging Study). Am. J.  Cardiol.
33            82: 594-599.
34     Cheng, Y.; Schwartz, J.;  Sparrow, D.; Aro, A.; Weiss, S. T.; Hu, H. (2001) Bone lead and blood lead levels in
3 5            relation to baseline blood pressure and the prospective development of hypertension: the Normative Aging
36            Study. Am. J. Epidemiol. 153: 164-171.
37     Chettle, D. R.; Fleming, D. E. B.; McNeill, F.  E.; Webber, C. E. (1997) Serum (plasma) lead, blood lead, and bone
38            lead [letter]. Am. J. Ind. Med. 32: 319-320.
39     Cheung etal. (1998)
40     Chia, S. E.; Chua, L. H.; Ng,  T. P.; Foo, S. C.; Jeyaratnam, J. (1994) Postural stability of workers exposed to lead.
41            Occup. Environ. Med. 51:768-771.
42     Chia, K. S.; Mutti, A.;  Alinovi, R.; Jeyaratnam, J.; Tan, C.; Ong, C. N.;  Lee, E. (1994) Urinary excretion of tubular
43            brush-border  antigens among lead exposed workers. Ann. Acad. Med. Singapore 23: 655-659.
44     Chia, K. S.; Mutti, A.;  Tan, C.; Ong, H. Y.; Jeyaratnam, J.; Ong, C. N.;  Lee, E. (1994) Urinary N-acetyl-"beta"-D-
45            glucosaminidase activity in workers exposed to inorganic lead. Occup. Environ. Med. 51: 125-129.
46     Chia, K. S.; Jeyaratnam, J.; Lee, J.; Tan,  C.; Ong, H. Y.; Ong, C. N.; Lee, E. (1995) Lead-induced nephropathy:
47            relationship between various biological exposure indices and early markers of nephrotoxicity. Am. J. Ind.
48            Med. 27: 883-895.
49     Chia, K. S.; Jeyaratnam,  J.; Tan, C.; Ong, H. Y.; Ong, C. N.; Lee, E. (1995) Glomerular function of lead-exposed
50            workers. Toxicol. Lett. 77: 319-328.
51     Chia, S.; Chia, K.; Chia, H.; Ong,  C.; Jeyaratnam, J. (1996a) Three-year follow-up of serial nerve conduction among
52            lead-exposed  workers. Scand. J. Work Environ. Health 22: 374-380.
53     Chia, S.; Chia, H.; Ong, C.; Jeyaratnam, J. (1996b) Cumulative blood lead levels and nerve conduction parameters.
54            Occup. Med.  46: 59-64.
55     Chia, S.; Chia, H.; Ong, C.; Jeyaratnam, J. (1996c) Cumulative concentrations of blood lead and postural stability.
56            Occup. Environ. Med. 53:264-268.
       December 2005                                6-363        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Chia, S.-E.; Chia, H.-P.; Ong, C.-N.; Jeyaratnam, J. (1997) Cumulative blood lead levels and neurobehavioral test
  2             performance. Neurotoxicology 18: 793-803.
  3     Chiodo, L. M; Jacobson, S. W.; Jacobson, J. L. (2004) Neurodevelopmental effects of postnatal lead exposure at
  4             very low levels. Neurotoxicol. Teratol. 26: 359-371.
  5     Chisolm, J. J., Jr.; Mellits, E. D.; Barrett, M. B. (1976) Interrelationships among blood lead concentration,
  6             quantitative daily ALA-U and urinary lead output following calcium EDTA. In: Nordberg, G. F., ed.
  7             Proceedings of third meeting of the subcommittee on the toxicology of metals under the Permanent
  8             Commission and International Association on Occupational Health; November 1974; Toyko, Japan.
  9             Amsterdam, The Netherlands: Elsevier Publishing Co.; pp. 416-433.
10     Chobanian, A. V.; Bakris, G. L.; Black, H. R.;  Cushman, W. C; Green, L. A.; Izzo, J. L., Jr.; Jones, D. W.;
11             Materson, B. J.; Oparil, S.; Wright, J.  T., Jr.; Roccella, E. J.; Joint National Committee on Prevention,
12             Detection, Evaluation, and Treatment of High Blood Pressure Institute; National High Blood Pressure
13             Education Program Coordinating Committee. (2003) Seventh report of the Joint National Committee on
14             Prevention,  Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 42:  1206-1252.
15     Chow, B. J. W.; Hassan, A. H.; Chan, K. L.; Tang, A. S. L. (2003) Prevalence and  significance of lead-related
16             thrombi in patients with implantable cardioverter defibrillators. Am. J. Cardiol. 91: 88-90.
17     Chowdhury, A. R.; Rao, R. V.; Gautam, A.  K.  (1986) Histochemical changes in the testes of lead induced
18             experimental rats. Folia Histochem. Cytobiol. 24: 233-237.
19     Christoffersson, J. O.; Schutz, A.; Ahlgren,  L.; Haeger-Aronsen, B.; Mattsson, S.; Skerfving, S. (1984) Lead in
20             finger-bone analysed in vivo in active and retired lead workers. Am. J. Ind. Med. 6: 447-457.
21     Christoffersson, J. O.; Ahlgren, L.; Schwartz, A.; Skerfving, S.; Mattsson, S. (1986) Decrease of skeletal lead levels
22             in man after end of occupational exposure. Arch. Environ. Health41: 312-318.
23     Christoffersson et al. (1994)
24     Chu, N.-F.; Liou, S.-H.;  Wu, T.-N.; Chang,  P.-Y. (1999) Reappraisal of the relation between blood lead
25             concentration and blood pressure among the general population in Taiwan. Occup. Environ. Med.
26             56:30-33.
27     Chuang, H.-Y.; Schwartz, J.; Tsai, S.-Y.; Lee, M.-L. T.; Wang, J.-D.; Hu, H. (2000) Vibration perception thresholds
28             in workers with long term exposure to lead. Occup. Environ. Med. 57: 588-594.
29     Chuang, H. Y.; Schwartz, J.; Gonzales-Cossio, T.; Lugo, M. C.; Palazuelos, E.; Aro, A.; Hu, H.; Hernandez-Avila,
30             M. (2001) Interrelations of lead levels in bone, venous blood, and umbilical cord blood with exogenous
31             lead exposure through maternal plasma lead in peripartum women. Environ. Health Perspect. 109:  527-532.
32     Chuang, H.-Y.; Yu, K.-T.; Ho, C.-K.; Wu, M.-T.; Lin, G.-T.; Wu, T.-N. (2004) Investigations of vitamin D  receptor
33             polymorphism affecting workers' susceptibility to lead. J. Occup. Health 46: 316-322.
34     Churchill, D.; Perry,  I. J.; Beevers, D. G. (1997) Ambulatory blood pressure in pregnancy and fetal growth.  Lancet
35             349:7-10.
36     Cicuttini, F. M.; Woodburn, C. M.; Golec, R.; Forbes, A.; Sim, M. (1998) Low lead levels in amniotic fluid and cord
37             blood in a public hospital population.  Aust. N. Z. J. Public Health 22: 628-629.
38     Clark, A. R. L. (1977) Placental transfer of lead and its effects on the newborn.  Postgrad. Med. J. 53: 674-678.
39     Clark, C. S.; Bornschein, R. L.; Succop, P. A.;  Que Hee, S. S.; Hammond, P. B.; Peace, B.  (1985) Condition and
40             type of housing as an indicator of poential environmental lead exposure and pediatric blood lead levels.
41             Environ. Res. 38: 46-53.
42     Cleymaet, R.; Bottenberg, P.; Relief, D. H.; Slop, D.; Michotte, Y.; Coomans, D. (1991)  In vivo use of a dual acid
43             etch biopsy  for the evaluation of lead profiles in human surface enamel. Caries Res. 25: 256-263.
44     Cocco,  P. L.; Carta, P.; Belli, S.; Picchiri, G. F.; Flore, M. V. (1994a) Mortality of Sardinian lead and zinc miners:
45             1960-88. Occup. Environ. Med. 51: 674-682.
46     Cocco,  P. L.; Carta, P.; Flore, V.; Picchiri, G. F.; Zucca, C. (1994b) Lung cancer mortality among female  mine
47             workers exposed to silica. J. Occup. Med. 36: 894-898.
48     Cocco,  P.; Salis, S.; Anni, M.; Cocco, M. E.; Flore, C.; Ibba, A. (1995) Effects of short-term occupational exposure
49             to lead on erythrocyte glucose-6-phosphate dehydrogenase activity and serum cholesterol. J. Appl. Toxicol.
50             15:375-378.
51     Cocco,  P.; Carta, P.;  Flore, C.; Congia, P.; Manca, M. B.; Saba, G.; Salis, S. (1996) Mortality of lead smelter
52             workers with the glucose-6-phosphate dehydrogenase-deficient phenotype. Cancer Epidemiol. Biomarkers
53             Prev. 5: 223-225.
54     Cocco,  P.; Hua, F.; Boffetta, P.; Carta, P.; Flore, C.; Flore, V.; Onnis, A.; Picchiri, G. F.; Colin, D. (1997) Mortality
55             of Italian lead smelter workers. Scand. J. Work Environ. Health 23: 15-23.
       December 2005                                6-364         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Cocco, P.; Dosemeci, M; Heineman, E. F. (1998a) Brain cancer and occupational exposure to lead. J. Occup.
  2            Environ. Med. 40: 937-942.
  3     Cocco, P.; Dosemeci, M.; Heineman, E. F. (1998b) Occupational risk factors for cancer of the central nervous
  4            system: A case-control study of death certificates from 24 U.S. States. Am. J. Ind. Med. 33: 247-255.
  5     Cocco, P.; Ward, M. H.; Dosemeci, M. (1999) Risk of stomach cancer associated with 12 workplace hazards:
  6            Analysis of death certificates from 24 states of the United States with the aid of job exposure matrices.
  7            Occup. Environ. Med. 56: 781-787.
  8     Cockcroft, D. W.; Gault, M. H. (1976) Prediction of creatinine clearance from serum creatinine. Nephron6: 31-41.
  9     Cohen, N.; Modai, D.; Golik, A.; Weissgarten, J.; Peller, S.; Katz, A.; Averbukh, Z.; Shaked, U. (1989) Increased
10            concanavalin A-induced suppressor cell activity in humans with occupational lead exposure. Environ. Res.
11            48:1-6.
12     Colleoni, N.; D'Amico, G. (1986) Chronic lead accumulation as a possible cause of renal failure in gouty patients.
13            Nephron 44: 32-35.
14     Colleoni, N.; Arrigo, G.; Gandini, E.; Corigliano, C.; D'Amico, G. (1993) Blood lead in hemodialysis patients.
15            Am. J. Nephrol.  13:  198-202.
16     Constantine, N. A.; Kraemer, H. C.; Kendall-Tackett, K. A.; Bennett, F. C.; Tyson, J. E.; Gross, R. T. (1987) Use of
17            physical and neurologic observations in assessment of gestational age in low birth weight infants. J.
18            Pediatr. (St. Louis, MO, U.S.) 110: 921-928.
19     Cooney, G. H. (1995) Lead research: where do we go from here? Neurotoxicol.  Teratol. 17: 215-218.
20     Cooney, G. H.; Bell, A.; McBride, W.; Carter, C. (1989a) Neurobehavioural consequences of prenatal low level
21            exposures to lead. Neurotoxicol. Teratol.  11: 95-104.
22     Cooney, G. H.; Bell, A.; McBride, W.; Carter, C. (1989b) Low-level exposures to lead: the Sydney lead study. Dev.
23            Med. Child Neurol. 31: 640-649.
24     Cooney, G.; Bell, A.; Stavrou, C. (1991) Low level exposures to lead and neurobehavioural development: the
25            Sydney study at seven years. In: Farmer, J. G., ed. International conference: heavy metals in the
26            environment, v. 1; September; Edinburgh, United Kingdom. Edinburgh, United Kingdom: CEP
27            Consultants, Ltd.; pp. 16-19.
28     Cooper, W.  C. (1988) Deaths from chronic renal disease in U. S. battery and lead production workers. In: Victery,
29            W., ed. Symposium on lead-blood pressure relationships; April 1987; Chapel Hill, NC. Environ. Health
30            Perspect. 78: 61-63.
31     Cooper, W.  C.; Gaffey, W. R. (1975) Mortality of lead workers. In: Cole, J. F., ed. Proceedings of the 1974
32            conference on standards of occupational lead exposure; February 1974; Washington, DC. J. Occup. Med.
33            17: 100-107.
34     Cooper, W.  C.; Tabershaw, I. R.; Nelson, K. W.  (1973) Laboratory studies of workers in lead smelting and refining.
35            In:  Earth, D.; Berlin, A.; Engel, R.; Recht, P.; Smeets, J., eds. Environmental health aspects of lead:
36            proceedings [of an] international symposium; October 1972; Amsterdam, The Netherlands. Luxembourg:
3 7            Commission of the European Communities; pp. 517-530; report no. EUR 5004 d-e-f.
38     Cooper, W.  C.; Wong, O.; Kheifets, L. (1985) Mortality among employees of lead battery plants and lead-producing
39            plants, 1947-1980. Scand. J. Work Environ. Health 11: 331-345.
40     Cooper, W.  C.; Wong, O.; Trent, L. (1989) Case-control study of gastric cancer  deaths in a lead battery plant. Report
41            to the International Lead Zinc Research Organization.
42     Coratelli, P.; Giannattasio, M.; Lomonte, C.; Marzolla, R.; Rana, F.; L'Abbate, N. (1988) Enzymuria to detect
43            tubular injury in workers exposed to lead: a 12-month follow-up. In:  Bianchi, C.; Bocci, V.; Carone, F. A.;
44            Rabkin, R., eds. Kidney and proteins in health and disease: fifth international symposium in health and
45            disease; July 1987; Montecatini Terme, Italy. Basel, Switzerland: S. Karger; pp. 207-211.
46     Cordioli, G.; Cuoghi, L.; Solari, P. L.; Berrino, F.; Crosignani, P.; Riboli, E. (1987) Mortalita per tumore inuna
47            coorte di lavoratori della industria del vetro  [Tumor mortality in a cohort of glass industry workers].
48            Epidemiol. Prev. (Italy) 9(30): 16-18.
49     Cory-Slechta, D. A. (1995) Bridging human and experimental animal studies of lead neurotoxicity: moving beyond
50            IQ. Neurotoxicol. Teratol. 17: 219-221.
51     Cory-Slechta, D. A. (1996) Legacy of lead exposure: consequences for the central nervous system. Otolaryngol.
52            Head Neck Surg. 114: 224-226.
53     Coscia, G. C.; Discalzi, G.; Ponzetti, C. (1987) Immunological aspects of occupational lead exposure. Med. Lav.
54            78:360-364.
55     Coscia, J. M.; Ris, M. D.; Succop, P. A.; Dietrich, K. N. (2003) Cognitive development of lead exposed children
56            from ages 6 to 15 years: an application  of growth curve analysis. Child Neuropsychol. 9: 10-21.
       December 2005                                6-365         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Coste, I; Mandereau, L.; Pessione, F.; Bregu, M.; Faye, C.; Hemon, D.; Spira, A. (1991) Lead-exposed workmen
  2             and fertility: a cohort study on 354 subjects. Eur. J. Epidemiol. 7: 154-158.
  3     Counter, S. A.; Vahter, M.; Laurell, G.; Buchanan, L. H.; Ortega, F.; Skerfving, S. (1997) High lead exposure and
  4             auditory sensory-neural function in Andean children. Environ. Health Perspect. 105: 522-526.
  5     Counter, S. A.; Buchanan, L. H.; Rosas, H. D.; Ortega, F.  (1998) Neurocognitive effects of chronic lead intoxication
  6             in Andean children. J. Neurol. Sci. 160: 47-53.
  7     Cramer, K.; Goyer, R. A.; Jagenburg, R.; Wilson, M. H. (1974) Renal ultrastructure, renal function, and parameters
  8             of lead toxicity  in workers with different periods of lead exposure. Br. J.  Ind. Med. 31: 113-127.
  9     Craswell, P. W.; Price, J.; Boyle, P. D.; Behringer,  D.; Stoeppler, M.; Ritz, E. (1987) Patterns of lead excretion in
10             patients with gout and chronic renal failure - a comparative German and Australian study. Sci. Total
11             Environ. 66: 17-28.
12     Cristofolini, A.; Del, D. M.; Manfrini, G.; Vitalone, V.; Ramponi, C.; De Santa, A.; Miori, R. (1995)  Blood lead
13             levels in the population of Trento. Ig. Mod. 104: 93-105.
14     Cristy, M. (1981) Active bone marrow distribution as a function of age in humans. Phys. Med. Biol. 26: 389-400.
15     Cullen, M. R.; Kayne, R. D.; Robins, J. M. (1984) Endocrine and reproductive dysfunction in men associated with
16             occupational inorganic lead intoxication. Arch. Environ. Health 39: 431-440.
17     Cunningham, R. D., Jr. (1993) Declining blood lead levels and cognitive change in children [letter]. JAMA J.
18             Am. Med. Assoc. 270: 828.
19     Dalpra, L.; Tibiletti, M. G.; Nocera, G.; Giulotto, P.; Auriti, L.; Carnelli, V.; Simoni, G. (1983) SCE analysis in
20             children exposed to lead emission from a smelting plant. Mutat. Res. 120: 249-256.
21     David, O. J.; Clark, J.; Voeller, K. (1972) Lead and hyperactivity. Lancet (7783):  900-903.
22     David, O. J.; Hoffman, S. P.; Sverd, J.; Clark, J.; Voeller, K. (1976) Lead and hyperactivity. Behavorial response to
23             chelation: a pilot study. Am. J. Psychiatry 133: 1155-1158.
24     David, O. J.; Clark, J.; Hoffman, S. (1979) Childhood lead poisoning: a re-evaluation. Arch. Environ. Health
25             34: 106-111.
26     Davidson, R.; MacKinnon, J. G. (1981) Several tests for model specification in the presence of alternative
27             hypotheses. Econometrica 49: 781-793.
28     Davies, J. M. (1984a) Lung  cancer mortality among workers making lead chromate and zinc chromate pigments at
29             three English factories. Br. J. Ind. Med. 41: 158-169.
30     Davies, J. M. (1984b) Long  term mortality study of chromate pigment workers who suffered lead poisoning. Br. J.
31             Ind. Med. 41: 170-178.
32     Davis, J. M.; Svendsgaard, D. J. (1987) Lead and child development. Nature (London) 329: 297-300.
33     Davis, J. M.; Svendsgaard, D. J. (1990) Nerve conduction velocity and lead: a critical review and meta-analysis. In:
34             Johnson, B. L.;  Anger, W. K.; Durao, A.; Xintaras, C., eds. Advances in  neurobehavioral toxicology:
3 5             applications in environmental and occupational health: [selected papers presented at the third international
36             symposium on neurobehavioral and occupational health]; December 1988; Washington, DC. Chelsea, Ml:
37             Lewis Publishers, Inc.; pp.  353-376.
38     Davis, J. M. (1990) Risk assessment of the developmental neurotoxicity of lead. Neurotoxicology 11: 285-292.
39     Davis, D. W.;  Chang, F.; Burns, B.; Robinson, J.; Dossett, D. (2004) Lead exposure and attention regulation in
40             children living in poverty. Dev. Med. Child Neurol. 46: 825-831.
41     De Burbure, C.; Buchet,  J. P.; Bernard, A.; Leroyer, A.; Nisse, C.; Haguenoer, J.-M.; Bergamaschi, E.; Mutti, A.
42             (2003) Biomarkers of renal effects in children and adults with low environmental exposure to heavy metals.
43             J. Toxicol. Environ. Health Part A 66: 783-798.
44     DeCastro, F. J.; Medley, J. (1997) Lead in bone and hypertension [letter]. Matern. Child Health J. 1: 199-200.
45     De Kort, W. L. A. M.; Verschoor, M. A.; Wibowo, A. A. E.; van Hemmen, J. J. (1987) Occupational exposure to
46             lead and blood pressure: a study in 105 workers. Am. J. Ind. Med. 11: 145-156.
47     Delves, H. T.; Campbell, M. J. (1988) Measurements of total lead concentrations and of lead isotope  ratios in whole
48             blood by use of inductively coupled plasma source mass spectrometry. J. Anal. At. Spectrom. 3: 343-348.
49     Den Hond, E.; Nawrot, T.; Staessen, J. A. (2002) The relationship between blood  pressure and blood lead in
50             NHANES III. J. Hum. Hypertens. 16: 563-568.
51     Denno, D. (1990) Biology and violence. From birth to adulthood. New York, NY: Cambridge University Press.
52     Despres, C.; Beuter, A.;  Richer, F.; Poitras, K.; Veilleux, A.; Ayotte, P.; Dewailly, E.; Saint-Amour, D.; Muckle, G.
53             (2005) Neuromotor functions in Inuit preschool children exposed to Pb, PCBs, and Hg. Neurotoxicol.
54             Teratol. 27: 245-257.
55     DiPietro, E. S.; Philips, D. L.; Paschla, D. C.; Neese, J. W. (1989) Determination of trace elements in human hair.
56             Biol.  Trace Elem. Res. 22: 83-100.
       December 2005                                6-366        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Diamond, G. L. (1988) Biological monitoring of urine for exposure to toxic metals. In: Clarkson, T. W.;
  2             Nordberg, G.; Sager, P., eds. Scientific basis and practical applications of biological monitoring of toxic
  3             metals. New York, NY: Plenum Press; pp. 515-529.
  4     Diamond, G. L. (1992) Review of default value for lead plasma-to-urine transfer coefficient (TPLUR) in the U.S.
  5             EPA uptake/biokinetic  model. Syracuse, NY: Syracuse Research Corporation. Prepared for Environmental
  6             Criteria and Assessment Office, U.S. Environmental Protection Agency.
  7     Dietrich, K. N. (1993) [Untitled letter concerning reporting practices of the international prospective studies of lead
  8             and child development]. Arch. Environ. Health 48: 125.
  9     Dietrich, K. N. (1995) A higher  level of analysis: Bellinger's,  interpreting the literature on lead and child
10             development. Neurotoxicol. Teratol. 17: 223-225.
11     Dietrich, K. N.; Krafft, K. M; Bier, M; Succop, P. A.; Berger, O.; Bornschein, R. L. (1986) Early effects of fetal
12             lead exposure: neurobehavioral findings at 6 months. Int. J. Biosoc. Res. 8: 151-168.
13     Dietrich, K. N.; Krafft, K. M.; Shukla, R.; Bornschein, R. L.;  Succop, P. A. (1987a) The neurobehavioral effects of
14             early lead exposure. In: Schroeder, S. R., ed. Toxic substances and mental retardation: neurobehavioral
15             toxicology and teratology. Washington, DC: American Association on Mental Deficiency;  pp. 71-95.
16             (Begab, M. J., ed. Monographs of the American Association on Mental Deficiency: no. 8).
17     Dietrich, K. N.; Krafft, K. M.; Bornschein, R. L.; Hammond, P. B.; Berger, O.; Succop, P. A.; Bier,  M. (1987b)
18             Low-level fetal lead exposure effect on neurobehavioral development in early infancy. Pediatrics
19             80:721-730.
20     Dietrich, K. N.; Succop, P. A.; Bornschein, R. L.; Krafft, K. M.; Berger, O.; Hammond, P. B.; Buncher, C. R. (1990)
21             Lead exposure and neurobehavioral development in later infancy. In: Conference on advances in lead
22             research: implications for environmental health; January  1989; Research Triangle Park, NC. Environ.
23             HealthPerspect. 89: 13-19.
24     Dietrich, K. N.; Succop, P. A.; Berger, O. G.; Hammond, P. B.; Bornschein, R. L. (1991) Lead exposure and the
25             cognitive development of urban preschool children: the Cincinnati lead study cohort at age 4 years.
26             Neurotoxicol. Teratol.  13: 203 -211.
27     Dietrich, K. N.; Succop, P. A.; Berger, O. G.; Keith, R. W.  (1992) Lead exposure and the central auditory processing
28             abilities and cognitive development of urban children: the Cincinnati lead study cohort at age 5 years.
29             Neurotoxicol. Teratol.  14: 51-56.
30     Dietrich, K. N.; Berger, O. G.; Succop, P. A.; Hammond, P. B.; Bornschein, R. L. (1993a) The developmental
31             consequences of low to moderate prenatal and postnatal lead exposure: intellectual attainment in the
32             Cincinnati Lead Study  Cohort following school entry. Neurotoxicol. Teratol. 15: 37-44.
33     Dietrich, K. N.; Berger, O. G.; Succop, P. A. (1993b) Lead exposure and the motor developmental status of urban
34             six-year-old children in the Cincinnati prospective study. Pediatrics 91: 301-307.
35     Dietrich, K. N.; Ris, M. D.; Succop, P. A.; Berger, O. G.; Bornschein, R. L. (2001) Early exposure to lead and
36            juvenile delinquency. Neurotoxicol. Teratol. 23: 511-518.
37     Dietrich, K. N.; Ware, J. H.; Salganik, M.; Radcliffe, J.; Rogan, W. J.; Rhoads, G. G.; Fay, M. E.; Davoli, C. T.;
38             Denckla, M. B.; Bornschein, R. L.; Schwarz, D.; Dockery, D. W.; Adubato,  S.; Jones, R. L.; forthe
3 9             Treatment of Lead-Exposed Children Clinical Trial Group. (2004) Effect of chelation therapy on the
40             neuropsychological and behavioral development of lead-exposed children after school entry. Pediatrics
41             114: 19-26.
42     Dietrich, K. N.; Eskenazi, B.; Schantz, S.; Yolton, K.; Rauh, V. A.; Johnson, C. B.; Alkon, A.; Canfield, R. L.;
43             Pessah, I. N.; Berman, R. F. (2005) Principles and practices of neurodevelopmental assessment in children:
44             lessons learned from the Centers for Children's Environmental Health and Disease Prevention Research.
45             Environ. Health Perspect. 113:  1437-1446.
46     Dingwall-Fordyce, I.; Lane, R. E. (1963) A follow-up study of lead workers. Br. J. Ind. Med. 20: 313-315.
47     Diouf, A.; Garcon, G.; Thiaw, C.; Diop, Y.; Fall, M.; Ndiaye B.; Siby, T.; Hannothiaux, M. H.; Zerimech, F.;
48             Ba, D.; Haguenoer, J. M.; Shirali, P. (2003) Environmental lead exposure and its relationship to traffic
49             density among Senegalese children: a pilot study. Hum. Exp. Toxicol. 22: 559-564.
50     Discalzi, G. L.; Capellaro, F.; Bottalo, L.; Fabbro, D.; Mocellini, A. (1992) Auditory  brainstem evoked potentials
51             (BAEPS) in lead-exposed workers. Neurotoxicology 13: 207-209.
52     Discalzi, G.; Fabbro, D.; Meliga, F.; Mocellini, A.; Capellaro, F. (1993) Effects of occupational exposure to mercury
53             and lead on brainstem auditory evoked potentials.  Int. J. Psychophysiol. 14:  21-25.
54     Dolenc, P.; Staessen, J. A.; Lauwerys, R. R.; Amery, A., on behalf of the Cadmibel Study Group. (1993) Short
55             report: low-level lead exposure does not increase the blood pressure in the general population. J. Hypertens.
56             11:589-593.
       December 2005                                6-367         DRAFT-DO NOT QUOTE OR CITE

-------
  1    Dowd, T. L.; Rosen, J. F.; Gundberg, C. M; Gupta, R. K. (1997) The displacement of calcium from osteocalcin at
  2             submicromolar concentrations of free lead. Biochim. Biophys. Acta. 1226: 131-137.
  3    Draisma, G.; Boer, R.; Otto, S. J.; Van der Cruijsen, I. W.; Damhuis, R. A. M.; Schroder, F. H.; de Koning, H. J.
  4             (2003) Lead times and overdetection due to prostate-specific antigen screening: estimates from the
  5             European randomized study of screening for prostate cancer. J. Natl. Cancer Inst. 95: 868-878.
  6    Drasch, G. A.; Bohm, J.; Baur, C. (1987) Lead in human bones. Investigations on an occupationally non-exposed
  7             population in southern Bavaria (F. R. G.). I. Adults. Sci. Total Environ. 64: 303-315.
  8    Drasch, G.; Wanghofer, E.; Roider, G. (1997) Are blood, urine, hair, and muscle valid biomonitors for the internal
  9             burden of men with the heavy metals mercury, lead and cadmium? Trace Elem. Electrolytes
10             14(3): 116-123.
11    Driscoll, R. J. (1998) Epidemiologic study of adverse reproductive outcomes among women in the U.S. Forest
12             Service. In: Driscoll, R. J.; Reh, B. D.; Esswein, E. J.; Mattorano, D. A. Health hazard evaluation report no.
13             93-1035-2686, section 2. Washington, DC: U.S. Department of Agriculture, Forest Service. Available
14             from: NTIS, Springfield, VA; PB99-152241.
15    Droz, P. O. (1989) Biological monitoring I: Sources of variability in human response to chemical exposure.
16             Appl. Ind. Hyg. 4: F20-F24.
17    Dudek, B.; Merecz, D. (1997) Impairment of psychological functions in children environmentally exposed to lead.
18             Int. J. Occup. Med. Environ. Health 10: 37-46.
19    Dundar, M. S.; Pala, M. F. (2003) Monitoring of lead,  zinc, cadmium, nickel, chromium and copper in street dust
20             samples in Adapazari, Turkey, after earthquake. Trace Elem. Electrol. 20: 104-107.
21    Durbin, P. W. (1992) Distribution of transuranic elements in bone. Neurotoxicology 13: 821-824.
22    Dursun, N.; Tutus, A. (1999) Chronic occupational lead exposure and thyroid function. J. Trace Elem. Exp. Med.
23             12:45-49.
24    Duydu, Y.; Suzen, H. S.; Aydin, A.; Cander, O.; Uysal, H.; Isimer, A.; Vural, N. (2001) Correlation between lead
25             exposure indicators and sister chromatid exchange (SCE) frequencies in lymphocytes from inorganic lead
26             exposed workers. Arch. Environ. Contam. Toxicol. 41: 241-246.
27    Duydu, Y.; Dur, A.; Suzen, H. S. (2005) Evaluation of increased proportion of cells with unusually high sister
28             chromatid exchange counts as a cytogenetic biomarker for lead exposure. Biol. Trace Elem. Res.
29             104: 121 129.
30    Dye, B. A.; Hirsch, R.; Brody, D. J. (2002) The relationship between blood lead levels and periodontal bone loss in
31             the United States, 1988-1994. Environ. Health Perspect. 110:  997-1002.
32    EL-Safty, I. A.; Afifi, A. M.; Shouman, A. E.; EL-Sady, A. K. R. (2004)  Effects of smoking and lead exposure on
33             proximal tubular integrity among Egyptian industrial workers. Arch. Med. Res. 35: 59-65.
34    Elwood, P. C.; Davey-Smith, G.; Oldham, P.  D.; Toothill, C. (1988a) Two Welsh surveys of blood lead and blood
35             pressure. In: Victery, W., ed. Symposium on lead-blood pressure relationships; April 1987; Chapel Hill,
36             NC. Environ. Health Perspect. 78: 119-121.
37    Elwood, P. C.; Yarnell, J. W. G.; Oldham, P.  D.; Catford, J. C.; Nutbeam, D.; Davey-Smith, G.; Toothill, C. (1988b)
3 8             Blood pressure and blood lead in surveys in Wales. Am. J. Epidemiol.  127: 942-945.
39    ESA Biosciences, Inc. (1998) Lead care(R) childhood blood lead testing. Chelmsford, MA: ESA Biosciences, Inc.
40    Egeland, G. M.; Burkhart, G. A.; Schnorr, T.  M.; Hornung, R. W.; Fajen, J. M.; Lee, S. T. (1992) Effects of
41             exposure to carbon disulphide on low density lipoprotein cholesterol concentration and diastolic blood
42             pressure. Br. J. Ind. Med. 49: 287-293.
43    Ehrlich, R.; Robins, T.; Jordaan, E.; Miller, S.; Mbuli,  S.; Selby, P.; Wynchank, S.; Cantrell, A.; De Broe, M.;
44             D'Haese, P.; Todd,  A.; Landrigan, P. (1998) Lead absorption and renal dysfunction in a South African
45             battery factory. Occup. Environ. Med. 55: 453-460.
46    Eller, P. M., ed. (1984) NIOSH manual of analytical methods. 3rd ed. Cincinnati, OH: U.S. Department of Health
47             and Human Services, National Institute for Occupational Safety and Health; publication no. DHHS
48             (NIOSH) 84-100.
49    Elmarsafawy, S. F.; Tsaih, S.-W.; Korrick, S.; Dickey, J. H.; Sparrow, D.; Aro, A.; Hu, H. (2002) Occupational
50             determinants of bone and blood lead levels in middle aged and elderly men from the general community:
51             the Normative Aging Study. Am. J. Ind. Med. 42: 38-49.
52    Emmerson, B. T. (1965) The renal excretion of urate in chronic lead nephropathy. Australas.  Ann. Med.
53             14:295-303.
54    Emmerson, B. T.; Ravenscroft, P. J.  (1975) Abnormal renal urate homeostasis in systemic disorders. Nephron
55             14:62-80.
      December 2005                                6-368         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Emory, E.; Pattillo, R.; Archibold, E.; Bayorh, M.; Sung, F. (1999) Neurobehavioral effects of low-level lead
  2             exposure in human neonates. Am. J. Obstet. Gynecol. 181: S2-S11.
  3     Endo, G.; Horiguchi, S.; Kiyota, I. (1990) Urinary N-acetyl-"beta"-D-glucosaminidase activity in lead-exposed
  4             workers. J. Appl. Toxicol. 10: 235-238.
  5     Endo, G.; Konishi, Y.; Kiyota, A.; Horiguchi, S. (1993) Urinary "alpha"! microglobulin in lead workers. Bull.
  6             Environ. Contam. Toxicol. 50: 744-749.
  7     Englyst, V.; Lundstrom, N. G.; Gerhardsson, L.; Rylander, L.; Nordberg, G. (2001) Lung cancer risks among lead
  8             smelter workers also exposed to arsenic. Sci. Total Environ. 273: 77-82.
  9     Erfurth, E. M.; Gerhardsson, L.; Nilsson, A.; Rylander, L.; Schutz, A.; Skerfving, S.; Borjesson, J. (2001) Effects of
10             lead on the endocrine system in lead smelter workers. Arch. Environ. Health 56: 449-455.
11     Ericson, J. E.; Smith, D. R.; Flegal, A. R. (1991) Skeletal concentrations of lead, cadmium, zinc, and silver in
12             ancient North American Pecos indians. Environ. Health Perspect.  93: 217-223.
13     Erkkila, J.; Armstrong, R.; Riihimaki, V.; Chettle, D. R.; Paakkari, A.; Scott, M.; Somervaille, L.; Stark, J.; Kock,
14             B.; Aitio, A. (1992) In vivo measurements of lead in bone at four anatomical sites: long term occcupational
15             and consequent edogenous exposure. Br. J. Ind. Med. 49: 631-644.
16     Ernhart, C. B. (1993) Declining blood lead levels and cognitive change in children [letter]. JAMA J. Am. Med.
17             Assoc. 270: 827-828.
18     Ernhart, C. B. (1994)  [Untitled letter concerning errors in "Effect of low-level body burdens of lead on the mental
19             development of children: limitations of meta-analysis in a review of longitudinal data"]. Arch. Environ.
20             Health 49: 77-78.
21     Ernhart, C. B. (1995) Inconsistencies in the lead-effects literature exist and cannot be explained by "effect
22             modification." Neurotoxicol. Teratol. 17: 227-233.
23     Ernhart, C. B. (1995) Environmental lead and children's intelligence: Cleveland study hypothesis was not confirmed
24             [letter]. Br. Med. J. 310: 397.
25     Ernhart, C. B.; Greene, T. (1990) Low-level lead exposure in the prenatal and early preschool periods: language
26             development. Arch. Environ. Health 45: 342-354.
27     Ernhart, C. B.; Landa, B.; Wolf, A. W. (1985) Subclinical lead level and developmental deficit; reanalyses of data.
28             J. Learning Disabilities 18: 475-479.
29     Ernhart, C. B.; Wolf, A. W.; Kennard, M. J.; Erhard, P.; Filipovich, H. F.; Sokol, R. J. (1986) Intrauterine exposure
30             to low levels of lead: the status of the neonate. Arch. Environ. Health 41: 287-291.
31     Ernhart, C. B.; Morrow-Tlucak, M.; Marler, M.  R.; Wolf, A. W. (1987) Low level lead exposure in the prenatal and
32             early preschool periods: early preschool development. Neurotoxicol.  Teratol. 9: 259-270.
33     Ernhart, C. B.; Morrow-Tlucak, M.; Wolf, A. W. (1988) Low level lead exposure and intelligence in the preschool
34             years. Sci. Total Environ. 71: 453-459.
35     Ernhart, C. B.; Morrow-Tlucak, M.; Wolf, A. W.; Super, D.; Drotar, D. (1989) Low level lead exposure in the
36             prenatal and early preschool periods: intelligence prior to school entry. Neurotoxicol. Teratol. 11: 161-170.
37     Espy, K. A. (1997) The Shape School: assessing executive function in preschool children. Dev. Neuropsychol.
38             13:495-499.
39     Esteban, E.; Rubin, C. H.; Jones, R. L.; Noonan, G. (1999) Hair and blood  substrates for screening children for lead
40             poisoning. Arch. Environ. Health 54: 436-440.
41     Ettinger, A. S.; Tellez-Rojo, M. M.; Amarasiriwardena, C.; Schwartz, J.; Hu, H.; Hernandez-Avila, M. (2003)
42             Influence of maternal bone lead burden and calcium intake on lead in breast milk. Am. J. Epidemiol.
43             157(suppl. 11):S105.
44     Ettinger, A. s.; Tellez-Rojo, M. M.; Amarasiriwardena, C.; Gonzalez-Cossio, T.; Peterson, K. E.; Aro, A.; Hu, H.;
45             Hernandez-Avila, M. (2004) Levels of lead in breast milk and their relation to maternal blood and bone
46             lead levels at one month postpartum. Environ. Health Perspect. 112: 926-931.
47     Ewers, U.; Stiller-Winkler, R.; Idel, H. (1982) Serum immunoglobulin, complement C3, and salivary IgA levels in
48             lead workers. Environ. Res. 29: 351-357.
49     Factor-Litvak, P.; Graziano, J. H.; Kline, J. K.; Popovac, D.; Mehmeti, A.;  Ahmedi, G.; Shrout, P.; Murphy, M. J.;
50             Gashi, E.; Haxhiu, R.; Rajovic, L.; Nenezic, D. U.; Stein, Z. A. (1991) A prospective study of birthweight
51             and length of gestation in a population  surrounding a lead smelter in Kosovo, Yugoslavia. Int. J. Epidemiol.
52             20:722-728.
53     Factor-Litvak, P.; Stein, Z.; Graziano, J. (1993)  Increased risk of proteinuria among a cohort of lead-exposed
54             pregnant women. Environ. Health Perspect. 101: 418-421.
55
       December 2005                                 6-369        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Factor-Litvak, P.; Graziano, J.; Kline, J. K. (1996) Letter to editor [re: Al-Saleh, I.; Khalil, M. A.; Taylor, A. (1995)
  2             Lead, erythrocyte protoporphyrin, and hematological parameters in normal maternal and umbilical cord
  3             blood from subjects of the Riyadh region, Saudi Arabia. Arch. Environ. Health 50: 66-73]. Arch. Environ.
  4             Health 51: 468-469.
  5     Factor-Litvak, P.; Kline, J. K.; Popovac, D.; Hadzialjevic, S.; Lekic, V.; Preteni-Rexhepi, E.; Capuni-Paracka, S.;
  6             Slavkovich, V.; Graziano, J. (1996) Blood lead and blood pressure in young children. Epidemiology 7:
  7             633-637.
  8     Factor-Litvak, P.; Slavkovich, V.; Liu, X.; Popovac, D.; Preteni, E.; Capuni-Paracka, S.; Hadzialjevic, S.; Lekic, V.;
  9             Lolacono, N.; Kline, J.;  Graziano, J. (1998) Hyperproduction of erythropoietin in nonanemic lead-exposed
10             children. Environ. Health Perspect. 106: 361-364.
11     Factor-Litvak, P.; Wasserman, G.; Kline, J. K.; Graziano, J. (1999) The Yugoslavia prospective study of
12             environmental lead exposure. Environ. Health Perspect. 107: 9-15.
13     Falcon, M.; Vinas, P.; Osuna, E.; Luna, A. (2002) Environmental exposures to lead and cadmium measured in
14             human placenta. Arch. Environ. Health 57: 598-602.
15     Falcon, M.; Vinas, P.; Luna, A. (2003) Placenta! lead and outcome of pregnancy. Toxicology 185: 59-66.
16     Fanning, D. (1988) A mortality study of lead workers, 1926-1985. Arch. Environ. Health 43: 247-251.
17     Farant, J.-P.; Wigfield, D. C. (1987) Interaction of divalent metal ions with normal and lead-inhibited human
18             erythrocytic porphobilinogen synthase in vitro. Toxicol. Appl. Pharmacol. 89: 9-18.
19     Farias, P.; Borja-Aburto, V. H.; Rios, C.; Hertz-Picciotto, I.; Rojas-Lopez, M.; Chavez-Ayala, R. (1996) Blood lead
20             levels in pregnant women of high and low socioeconomic status in Mexico City. Environ. Health Perspect.
21             104: 1070-1074.
22     Farias, P.; Hu, H.; Rubenstein, E.; Meneses-Gonzalez, F.; Fishbein, E.; Palazuelos, E.; Aro, A.; Hernandez-Avila,
23             M. (1998) Determinants of bone and blood lead levels among teenagers living in urban areas with high lead
24             exposure. Environ. Health Perspect.  106: 733-737.
25     Farrow, S. (1994) Falling sperm quality: fact of fiction? Br. Med. J. 309: 1-2.
26     Pels, L. M.; Herbert, C.; Pergande, M.; Jung, K.; Hotter, G.; Rosello, J.; Gelpi, E.; Mutti, A.; De Broe, M.;  Stolte, H.
27             (1994) Nephron target sites in chronic exposure to lead. Nephrol.  Dial. Transplant. 9: 1740-1746.
28     Pels, L. M.; Wunsch, M.; Baranowski, J.; Norska-Borowka, I.; Price, R. G.; Taylor, S. A.; Patel, S.; De Broe, M.;
29             Elsevier, M. M.; Lauwerys, R.; Roels, H.; Bernard, A.; Mutti, A.; Gelpi, E.; Rosello, J.;  Stolte, H.  (1998)
30             Adverse effects of chronic low level lead exposure on kidney function~a risk group study in children.
31             Nephrol. Dial. Transplant 13: 2248-2256.
32     Fergusson, D. M.; Horwood, L. J. (1993) The effects of lead levels on the growth of word recognition in middle
33             childhood. Int. J. Epidemiol. 22: 891-897.
34     Fergusson, D. M.; Fergusson, J. E.; Horwood, L. J.; Kinzett, N. G. (1988a) A longitudinal study of dentine  lead
3 5             levels, intelligence, school performance and behaviour. Part II. Dentine lead and cognitive ability.  J. Child
36             Psychol. Psychiatry Allied Discip. 29: 793-809.
37     Fergusson, D. M.; Fergusson, J. E.; Horwood, L. J.; Kinzett, N. G. (1988b) A longitudinal study of dentine  lead
3 8             levels, intelligence, school performance and behaviour. Part III. Dentine lead levels and attention/activity.
39             J. Child Psychol. Psychiatry Allied Discip. 29: 811-824.
40     Fergusson, D. M.; Fergusson, J. E.; Horwood, L. J.; Kinzett, N. G. (1988) A longitudinal study of dentine lead
41             levels, intelligence, school performance and behaviour. Part I. Dentine lead levels and exposure to
42             environmental risk factors. J. Child Psychol. Psychiatry Allied Discip. 29: 781-792.
43     Fergusson, J. E.; Kinzett, N. G.; Fergusson, D. M.; Horwood, L. J. (1989) A longitudinal study of dentin lead levels
44             and intelligence  school performance and behavior the measurement of dentin lead. Sci. Total Environ.
45             80:229-242.
46     Fergusson, D. M.; Horwood, L. J.; Lynskey, M. T. (1993) Early dentine lead levels and subsequent cognitive and
47             behavioural development. J. Child Psychol. Psych. Allied Disciplines 34: 215-227.
48     Fergusson, D. M.; Horwood, L. J.; Lynskey, M. T. (1997) Early dentine lead levels and educational outcomes at
49             18 years. J. Child Psychol. Psychiatry 38: 471-478.
50     Fewtrell, L. J.; Pruss-Ustun, A.; Landrigan, P.; Ayuso-Mateos, J. L. (2004) Estimating the global burden of disease
51             of mild mental retardation and cardiovascular diseases from environmental lead exposure. Environ. Res.
52             94: 120-133.
53     Fisch, H.; Andrews, H.; Hendricks, J.; Goluboff, E. T.; Olson, J. H.; Olsson, C. A. (1997) The relationship of sperm
54             counts to birth rates: a population based study. J. Urol. 157:  840-843.
       December 2005                                 6-370         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Fischbein, A.; Alvares, A. P.; Anderson, K. E.; Sassa, S.; Kappas, A. (1977) Lead intoxication among demolition
  2            workers: the effect of lead on the hepatic cytochrome P-450 systems in humans. J. Toxicol. Environ. Health
  3            3:431-437.
  4     Fischbein, A.; Tsang, P.; Luo, J.-C. J.; Roboz, J. P.; Jiang, J. D.; Bekesi, J. G. (1993) Phenotypic aberrations of the
  5            CD3+ and CD4+ cells and functional impairments of lymphocytes at low-level occupational exposure to
  6            lead. Clin. Immunol. Immunopathol. 66: 163-168.
  7     Flegal, A. R.; Smith, D. R. (1992) Lead levels in preindustrial humans [letter]. N. Engl. J. Med. 326: 1293-1294.
  8     Flegal, A. R.; Smith, D. R. (1995) Measurements of environmental lead contamination and human exposure. Rev.
  9            Environ. Contam. Toxicol. 143: 1-45.
10     Fleming, D. E. B.; Boulay, D.; Richard, N. S.; Robin, J.-P.; Gordon, C. L.; Webber, C. E.;  Chettle, D. R. (1997)
11            Accumulated body burden and endogenous release of lead in employees of a lead smelter. Environ. Health
12            Perspect.  105: 224-233.
13     Fleming, D. E. B.; Chettle, D. R.; Wetmur, J. G.; Desnick, R. J.; Robin, J.-P.; Boulay, D.; Richard, N.  S.; Gordon,
14            C. L.; Webber, C. E. (1998) Effect of the delta-aminolevulinate dehydratase polymorphism on the
15            accumulation of lead in bone and blood in lead smelter workers. Environ. Res. 77: 49-61.
16     Flood, P. R.; Schmidt, P. F.; Wesenberg, G. R.; Gadeholt, H. (1988) The distribution of lead in human hemopoietic
17            tissue and spongy bone after lead poisoning and Ca-EDTA chelation therapy: observations made by atomic
18            absorption spectroscopy, laser microbeam mass analysis and electron microbeam X-ray analysis. Arch.
19            Toxicol. 62: 295-300.
20     Forbes, G. B.; Braining, G. B. (1976) Urinary creatinine excretion and lean body mass. Am. J. Clin. Nutr.
21            29: 1359-1366.
22     Forni, A.; Cambiaghi, G.; Secchi, G. C. (1976) Initial occupational exposure to lead: chromosome  and biochemical
23            findings. Arch. Environ. Health 31: 73-78.
24     Forni, A.; Sciame, A.; Bertazzi, P.  A.; Alessio, L. (1980) Chromosome and biochemical studies in women
25            occupationally exposed to lead. Arch. Environ. Health 35: 139-146.
26     Fosse, G.; Wesenberg, G. B. R.; Tvinnereim, H. M, Eide, R.; Kristoffersen, O.; Nag, O. H.; Wierzbicka, M;
27            Banoczy, J.; De Oliveira,  A. A.; Srisopak, C.; Zamudio, A. (1995) Lead in deciduous teeth from larger
28            cities of some countries. Int. J. Environ. Stud. 47: 203-210.
29     Foster, W. G. (1992) Reproductive toxicity of chronic lead exposure in the female  cynomolgus monkey. Reprod.
30            Toxicol. 6: 123-131.
31     Fracasso, M. E.;  Perbellini, L.; Solda, S.; Talamini, G.; Franceschetti, P. (2002) Lead induced DNA strand breaks in
32            lymphocytes of exposed workers: role of reactive oxygen species and protein kinase C. Mutat. Res.
33            515: 159-169.
34     Frank, R. M.; Sargentini-Maier, M. L.; Leroy, M. J. F.; Turlot, J. C. (1988) Age-related lead increase in human
3 5            permanent teeth demonstrated by energy dispersive X-ray fluorescence. J. Trace Elem. Electrolytes Health
36            Dis. 2: 175-179.
37     Franklin, C. A.; Inskip, M. J.; Baccanale, C. L.; Edwards, C. M.; Manton, W. I.; Edwards, E.;  O'Flaherty, E. J.
3 8            (1997) Use of sequentially administered stable lead isotopes to investigate changes in blood lead during
39            pregnancy in a nonhuman primate (Macaca fascicularis). Fundam. Appl. Toxicol. 39: 109-119.
40     Franks, P. A.; Laughlin, N. K.; Dierschke, D. J.; Bowman, R. E.; Meller, P. A. (1989) Effects  of lead on luteal
41            function in rhesus monkeys. Biol. Reprod. 41: 1055-1062.
42     Froom, P.; Kristal-Boneh, E.; Benbassat, J.; Ashkanazi, R.; Ribak, J. (1998) Predictive value of determinations of
43            zinc protoporphyrin for increased blood lead concentrations. Clin. Chem.  44(6): 1283-1288.
44     Froom, P.; Kristal-Boneh, E.; Benbassat, J., Ashkanazi, R.; Ribak, J. (1999) Lead exposure in battery-factory
45            workers is not associated with anemia. J. Occup. Environ. Med. 41:  120-123.
46     Fu, H.; Boffetta,  P. (1995) Cancer and occupational exposure to inorganic lead compounds: a  meta-analysis of
47            published data. Occup. Environ. Med. 52: 73-81.
48     Fulton, M.; Raab, G.; Thomson, G.; Laxen, D.; Hunter, R.; Hepburn, W. (1987) Influence of blood lead on the
49            ability and attainment of children in Edinburgh. Lancet (8544): 1221-1226.
50     Furman, A.; Laleli, M. (2001) Maternal and umbilical cord blood lead levels: an Istanbul study. Arch.  Environ.
51            Health 56: 26-28.
52     Galeas, T.; Tselepatiotis, E.; Katsanos, D.; Lappas, C. (1996) Chronic renal failure, caused by lead poisoning,
53            among traditional-spirit drinkers in Greece. Nephrol. Dial. Transplant. 11: A127.
54     Garcon, G.; Leleu, B.; Zerimech, F.; Marez, T.; Haguenoer, J.-M.; Furon, D.; Shirali, P. (2004) Biologic markers of
5 5            oxidative stress and nephrotoxicity as studied in biomonitoring and adverse effects of occupational
56            exposure to lead and cadmium. J. Occup.  Environ. Med. 46: 1180-1186.
       December 2005                                6-371        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Gardner, J. M; Walker, S. P.; Chang, S. M; Vutchkov, M; Lalor, G. C. (1998) Undernutrition and elevated blood
  2            lead levels: effects on psychomotor development among Jamaican children. Public Health Nutr.
  3            1: 177-179.
  4     Garman, S.; Anderson, H. A.; Moen, T. (2000) Occupational and adult lead exposure in Wisconsin. Wis. Med. J.
  5            99:25-29.
  6     Garrido Latorre, F.; Hernandez-Avila, M.; Orozco, J. T.; Medina, C. A. A.; Aro, A.; Palazuelos, E.; Hu, H. (2003)
  7            Relationship of blood and bone lead to menopause and bone mineral density among middle-age women in
  8            Mexico City. Environ. HealthPerspect. Ill: 631-636.
  9     Gartside, P. S. (1988) The relationship of blood lead levels and blood pressure in NHANES II: additional
10            calculations. In: Victery, W., ed. Symposium on lead-blood pressure relationships; April 1987; Chapel Hill,
11            NC. Environ. Health Perspect. 78: 31-34.
12     Gemmel, A.; Tavares, M.; Alperin, S.; Soncini, J.; Daniel, D.; Dunn, J.; Crawford, S.; Braveman, N.; Clarkson,
13            T. W.; McKinlay, S.; Bellinger, D. C. (2002) Blood lead level and dental caries in school-age children.
14            Environ. Health Perspect. 110: A625-A630.
15     Gennart, J. P.; Bernard, A.; Lauwerys, R. (1992) Assessment of thyroid, testes, kidney and autonomic nervous
16            system function in lead-exposed workers. Int. Arch. Occup. Environ. Health 64: 49-57.
17     Gerhard, I.; Waibel, S.; Daniel, V.; Runnebaum, B.  (1998) Impact of heavy metals on hormonal and immunological
18            factors in women with repeated miscarriages. Hum. Reprod. Update 4: 301-309.
19     Gerhardsson, L.; Brune, D.;Nordberg, G. F.; Wester, P. O. (1986) Distribution of cadmium, lead and zinc in lung,
20            liver and kidney in long-term exposed smelter workers. Sci. Total Environ. 50: 65-85.
21     Gerhardsson, L.; Chettle, D. R.; Englyst, V.; Nordberg, G. F.; Nyhlin, H.; Scott, M. C.; Todd, A. C.; Vesterberg, O.
22            (1992) Kidney effects in long term exposed lead smelter workers. Br. J. Ind. Med. 49: 186-192.
23     Gerhardsson, L.; Attewell, R.; Chettle, D. R.; Englyst, V.; Lundstrom, N.-G.; Nordberg, G. F.; Nyhlin, H.; Scott,
24            M. C.; Todd, A. C. (1993) In vivo measurements of lead in bone in long-term exposed lead smelter
25            workers. Arch. Environ. Health 48: 147-156.
26     Gerhardsson, L.; Hagmar, L.; Rylander, L.; Skerfving, S.  (1995a) Mortality and cancer incidence among secondary
27            lead smelter workers. Occup. Environ. Med. 52:  667-672.
28     Gerhardsson, L.; Englyst, V.; Lundstrom, N.G.; Nordberg, G.; Sandberg, S.;  Steinvall, F. (1995b) Lead in tissues of
29            deceased lead smelter worker. J. Trace Elem. Med. Biol. 9: 136-143.
30     Gerr, F.; Letz, R.; Stokes, L.; Chettle, D.; McNeill, F.; Kaye, W. (2002) Association between bone lead
31            concentration and blood pressure among young adults. Am. J. Ind. Med. 42: 98-106.
32     Gershanik, J. J.; Brooks, G. G.; Little, J. A. (1974) Blood lead values in pregnant women and their offspring. Am. J.
33            Obstet. Gynecol. 119: 508-511.
34     Gidlow, D. A. (2004) Lead toxicity. Occup. Med. (London) 54: 76-81.
35     Gil, F.; Perez, M. L.; Facio, A.; Villanueva, E.; Tojo, R.; Gil, A. (1994) Dental lead levels in the Galacian
36            population, Spain. Sci. Total Environ. 156: 145-150.
37     Glenn, B.  S.; Stewart, W.  F.;  Schwartz, B. S.;  Bressler, J.  (2001) Relation of alleles of the sodium-potassium
3 8            adenosine triphosphatase alpha2 gene with blood pressure and lead exposure. Am. J.  Epidemiol.
39            153:537545.
40     Glenn, B.  S.; Stewart, W.  F.;  Links, J. M.; Todd, A. C.; Schwartz, B. S. (2003) The longitudinal association of lead
41            with blood pressure. Epidemiology 14: 30-36.
42     Glickman, L.; Valciukas, J. A.; Lilis, R.; Weisman,  I. (1984) Occupational lead exposure: effects on saccadic eye
43            movements. Int. Arch. Occup. Environ. Health 54: 115-125.
44     Gomaa, A.; Hu, H.; Bellinger, D.; Schwartz, J.; Tsaih, S.-W.;  Gonzalez-Cossio, T.; Schnaas, L.; Peterson, K.;
45            Aro, A.; Hernandez-Avila, M. (2002) Maternal bone lead as an independent risk factor for fetal
46            neurotoxicity: a prospective study. Pediatrics 110: 110-118.
47     Gompertz, D.; Chettle, D. R.; Fletcher, J. G.; Mason, H.; Perkins, J.; Scott, M. C.; Smith, N. J.; Topping, M. D.;
48            Blindt, M. (1983) Renal dysfunction  in cadmium smelters: relation to in vivo liver and kidney cadmium
49            concentrations. Lancet (8335): 1185-1187.
50     Gonick, H. C.; Behari, J. R. (2002) Is lead exposure the principal cause of essential hypertension? Med.  Hypotheses
51            59:239-246.
52     Gonick, H. C.; Cohen, A.  H.; Ren, Q.; Saldanha, L.  F.; Khalil-Manesh, F.; Anzalone, J.; Sun, Y. Y. (1996) Effect of
53            2,3-dimercaptosuccinic acid on nephrosclerosis in the Dahl rat. I.  Role of reactive oxygen species. Kidney
54            Int. 50: 1572-1581.
55     Gonzalez-Cossio, T.; Peterson, K. E.; Sanin, L.-H.; Fishbein, E.; Palazuelos,  E.; Aro, A.; Hernandez-Avila, M.;
56            Hu, H. (1997) Decrease in birth weight in relation to maternal bone-lead burden. Pediatrics 100: 856-862.
       December 2005                                6-372         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Gordon, C. L.; Chettle, D. R.; Webber, C. E. (1993) An improved instrument for the in vivo detection of lead in
  2             bone. Br. J. Ind. Med. 50: 637-641.
  3     Governa, M; Valentino, M; Visona, I. (1987) In vitro impairment of human granulocyte functions by lead. Arch.
  4             Toxicol. 59: 421-425.
  5     Govoni, S.; Battaini, F.; Fernicola, C.; Castelletti, L.; Trabucchi, M. (1987) Plasma prolactin concentrations in lead
  6             exposed workers. J. Environ. Pathol. Toxicol. Oncol. 7: 13-15.
  7     Goyer, R. A. (1990) Lead toxicity: from overt to subclinical to subtle health effects. Environ. Health Perspect.
  8             86: 177-181.
  9     Goyer, R. A.; Epstein, S.; Bhattacharyya, M.; Korach, K. S.; Pounds, J. (1994) Environmental risk factors for
10             osteoporosis. Environ. Health Perspect. 102:  390-394.
11     Grandjean, P. (1979) Occupational lead exposure in Denmark: screening with the haematofluorometer. Br. J. Ind.
12             Med. 36: 52-58.
13     Grandjean, P.; Arnvig, E.; Beckmann, J. (1978) Psychological dysfunctions in lead-exposed workers: relation to
14             biological parameters of exposure. Scand. J. Work Environ. Health 4: 295-303.
15     Grandjean, P.; Wulf, H. C.; Niebuhr, E. (1983) Sister chromatid exchange in response to variations in occupational
16             lead exposure. Environ. Res. 32: 199-204.
17     Grandjean, P.; Hollnagel, H.; Hedegaard, L.; Christensen, J. M.; Larsen, S. (1989) Blood lead-blood pressure
18             relations: alcohol intake and hemoglobin as confounders. Am. J. Epidemiol. 129: 732-739.
19     Graves, A. B.; VanDuijn, C. M.; Chandra, V.; Fratiglioni, L.; Heyman, A.; Jorm, A. F.; Kokmen, E.; Kondo, K.;
20             Mortimer, J. A.; Rocca, W. A.; Shalat, S. L.; Soininen, H.; Hofman, A. (1991) Occupational exposures to
21             solvents and lead as risk factors for Alzheimer's disease: a collaborative re-analysis of case-control studies.
22             Int. J. Epidemiol. 20(suppl. 2): S58-S61.
23     Graziano, J. H. (1994) Validity of lead exposure markers in diagnosis and surveillance. Clin. Chem. 40: 1387-1390.
24     Graziano, J. H.; Popovac, D.; Factor-Litvak, P.; Shrout, P.; Kline, J.; Murphy, M. J.; Zhao, Y.-H.; Mehmeti, A.;
25             Ahmedi, X.; Rajovic, B.; Zvicer, Z.; Nenezic, D. U.; Lolacono, N. J.;  Stein, Z. (1990) Determinants of
26             elevated blood lead during pregnancy in a population surrounding a lead smelter in Kosovo, Yugoslavia.
27             In: Conference on advances in lead research: implications for environmental health; January 1989;
28             Research Triangle Park, NC. Environ. Health Perspect. 89: 95-100.
29     Graziano, J.; Slavkovich, V.; Liu X., Factor-Litvak, P.; Todd, A. (2004) A prospective study of prenatal and
30             childhood lead exposure and erythropoietin production. J. Occup. Environ. Med. 46: 924-929.
31     Green, S.; Bradley, D. A.; Palethorpe, J. E.; Mearman, D.; Chettle, D. R.; Lewis, A. D.; Mountford, P. J.; Morgan,
32             W. D. (1993) An enhanced sensitivity K-shell x-ray fluorescence technique for tibial lead determination.
33             Phys. Med. Biol. 38: 389-396.
34     Griffin, T. B.; Coulston, F.;  Wills, H.; Russell, J. C.; Knelson, J. H. (1975) Clinical studies on men continuously
35             exposed to airborne paniculate lead. In: Griffin, T. B.; Knelson, J. H., eds. Lead. Stuttgart, Federal
36             Republic of Germany: Georg Thieme Publishers; pp. 221-240. (Coulston, F.; Korte, F., eds. Environmental
37             quality and safety:  supplement v. 2).
38     Groth-Marnat, G. (2003). Handbook of Psychological Assessment. 4th ed. Hoboken, NJ, John Wiley & Sons.
39     Guerra-Tamayo, J. L.; Hernandez-Cadena, L.; Tellez-Rojo, M. M.; Mercado-Garcia, A. del S.; Solano-Gonzalez,
40             M.; Hernandez-Avila, M.; Hu, H. L. (2003) Exposicion al plomo y su relacion con el tiempo requerido para
41             embarazo [Lead exposure and time to pregnancy]. Salud Publica  Mex. 45(suppl. 2): S189-S195.
42     Guidetti, D.; Bondavalli, M.; Sabadini, R.; Marcello, N.; Vinceti, M.; Cavalletti, S.; Marbini, A.; Gemignani, F.;
43             Colombo, A.; Ferriari, A.; Vivoli, G.; Solime, F. (1996) Epidemiological survey of amyotrophic lateral
44             sclerosis in the province of Reggio Emilia, Italy: influence of environmental exposure to lead.
45             Neuroepidemiology 15: 301-312.
46     Gulson, B. L. (1996) Tooth analyses of sources and intensity of lead exposure in children. Environ. Health Perspect.
47             104:306-312.
48     Gulson, B.; Wilson, D. (1994) History of lead exposure in children revealed from isotopic analyses of teeth. Arch.
49             Environ. Health 49: 279-283.
50     Gulson, B. L.; Mahaffey, K. R.; Mizon, K. J.; Korsch, M. J.; Cameron, M. A.;  Vimpani, G. (1995) Contribution of
51             tissue lead to blood lead in adult female subjects based on stable lead  isotope methods. J. Lab. Clin. Med.
52             125:703-712.
53     Gulson, B. L.; Jameson, C. W.; Mahaffey, K. R.; Mizon, K. J.; Korsch, M. J.; Vimpani, G. (1997) Pregnancy
54             increases mobilization of lead from maternal skeleton. J. Lab. Clin. Med. 130: 51-62.
       December 2005                                6-373         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Gulson, B. L.; Mahaffey, K. R.; Jameson, C. W.; Mizon, K. I; Korsch, M. I; Cameron, M. A.; Eisman, J. A.
  2             (1998a) Mobilization of lead from the skeleton during the postnatal period is larger than during pregnancy.
  3             J. Lab. Clin. Med. 131:  324-329.
  4     Gulson, B. L.; Jameson, C. W.; Mahaffey, K. R.; Mizon, K. J.; Patison, N.; Law, A. J.; Korsch, M. J.; Salter, M. A.
  5             (1998b) Relationships of lead in breast milk to lead in blood, urine, and diet of the infant and mother.
  6             Environ. Health Perspect. 106: 667-674.
  7     Gulson, B. L.; Gray, B.; Mahaffey, K. R.; Jameson, C. W.; Mizon, K. J.; Patison, N.; Korsch, M. J. (1999a)
  8             Comparison of the rates of exchange of lead in the blood of newly born infants and their mothers with lead
  9             from their current environment. J. Lab. Clin. Med. 133: 171-178.
10     Gulson, B. L.; Mahaffey, K. R.; Jameson, C. W.; Patison, N.; Law, A. J.; Mizon, K. J.; Korsch, M. J.; Pederson, D.
11             (1999b) Impact of diet on lead in blood and urine in female adults and relevance to mobilization of lead
12             from bone stores. Environ. Health Perspect. 107: 257-263.
13     Gulson, B. L.; Mizon, K. J.; Palmer, J. M.; Korsch, M. J.; Donnelly, J. B. (2000) Urinary excretion of lead during
14             pregnancy and postpartum. Sci. Total Environ. 262: 49-55.
15     Gulson, B. L.; Mizon, K. J.; Palmer, J. M.; Korsch, M. J.; Taylor, A. J. (2001) Contribution of lead from calcium
16             supplements to blood lead. Environ. Health Perspect. 109: 283-288.
17     Gulson, B.; Mizon, K.;  Smith, H.; Eisman, J.; Palmer, J.; Korsch, M.; Donnelly, J.; Waite, K. (2002) Skeletal lead
18             release during bone resorption: effect of bisphosphonate treatment in a pilot study. Environ. Health
19             Perspect. 110:  1017-1023.
20     Gulson, B. L.; Mizon, K. J.; Korsch, M. J.; Palmer, J. M.; Donnelly, J. B. (2003) Mobilization of lead from human
21             bone tissue during pregnancy and lactation~a summary of long-term research. Sci. Total Environ.
22             303:79-104.
23     Gulson, B. L.; Mizon, K. J.; Palmer, J. M.; Korsch, M. J.; Taylor, A. J.; Mahaffey, K. R. (2004) Blood lead changes
24             during pregnancy  and postpartum with calcium supplementation. Environ. Health Perspect.  112: 1499-
25             1507.
26     Gunnarsson, L. G.; Bodin, L.; Soderfeldt, B.; Axelson, O. (1992) A case-control study of motor neurone disease: its
27             relation to heritability, and occupational exposures, particularly to solvents. Br. J. Ind. Med. 49: 791-798.
28     Guo, T. L.; Mudzinski,  S. P.; Lawrence, D. A. (1996a) The heavy metal lead modulates the expression of both
29             TNF-"alpha" and TNF-"alpha" receptors in lipopolysaccharide-activated human peripheral blood
30             mononuclear cells. J. Leukoc. Biol. 59:  932-939.
31     Guo, T. L.; Mudzinski,  S. P.; Lawrence, D. A. (1996b) Regulation of HLA-DR and invariant chain expression by
32             human peripheral blood mononuclear cells with lead, interferon-"gamma", or interleukin-4.  Cell. Immunol.
33             171: 1-9.
34     Gurer-Orhan, H.; Sabir, H.D.; Ozgunes, H. (2004) Correlation between clinical indicators of lead poisoning and
35             oxidative stress parameters in controls and lead-exposed workers. Toxicology  195: 147-154.
36     Gustafson, A.; Hedner,  P.;  Schutz, A.; Skerfving, S. (1989) Occupational lead exposure and pituitary function. Int.
3 7             Arch. Occup. Environ. Health 61: 277-281.
38     Gustavsson, P.; Plato, N.; Hallqvist, J.; Hogstedt, C.; Lewne, M.; Reuterwall, C.; Scheele, P. (2001) Apopulation-
39             based case-referent study of myocardial infarction and occupational exposure to motor exhaust, other
40             combustion products, organic solvents, lead, and dynamite. Stockholm heart epidemiology program
41             (SHEEP) study group. Epidemiology 12: 222-228.
42     Gwiazda, R.; Campbell, C.; Smith, D. (2005) A noninvasive isotopic approach to estimate the bone lead
43             contribution to blood in children: implications for assessing the efficacy of lead abatement. Environ. Health
44             Perspect. 113:  104-110.
45     Gyllenborg, J.; Skakkebaek, N. E.; Nielsen, N. C.; Keiding, N.; Giwercman, A. (1999) Secular and seasonal changes
46             in semen quality among young Danish men: a statistical analysis of semen samples from 1927 donor
47             candidates during 1977-1995. Int. J. Androl. 22: 28-36.
48     Hafeez, A.; Malik, Q. U. (1996)  Blood lead levels in preschool children in Rawalpindi. JPMA J. Pak. Med. Assoc.
49             46:272-274.
50     Hagmar, L.; Stromberg, U.; Bonassi, S.; Hansteen, I.-L.; Knudsen, L. E.; Lindholm, C.; Norppa, H. (2004) Impact of
51             types of lymphocyte chromosomal aberrations on human cancer risk: results from Nordic and Italian
52             cohorts. Cancer Res. 64: 2258-2263.
53     Hajem, S.; Moreau, T.; Hannaert, P.; Lellouch, J.; Huel, G.; Hellier, G.; Orssaud, G.; Claude, J. R.; Juguet, B.;
54             Festy, B.; Garay, R. P. (1990) Influence of environmental lead on membrane ion transport in a French
55             urban male population.  Environ. Res. 53: 105-118.
       December 2005                                6-374        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Hammond, P. B.; Lerner, S. I.; Gartside, P. S.; Hanenson, I. B.; Roda, S. B.; Foulkes, E. C.; Johnson, D. R.; Pesce,
  2             A. J. (1980) The relationship of biological indices of lead exposure to the health status of workers in a
  3             secondary lead smelter. J. Occup. Med. 22: 475-484.
  4     Haenninen, H.; Hernberg, S.; Mantere, P.; Vesanto, R.; Jalkanen, M. (1978) Psychological performance of subjects
  5             with low exposure to lead. J. Occup. Med. 20: 683-689.
  6     Han, S.; Pfizenmaier, D. H.; Garcia, E.; Eguez, M. L.; Ling, M.; Kemp, F. W.; Bogden, J. D. (2000) Effects of lead
  7             exposure before pregnancy and dietary calcium during pregnancy on fetal development and lead
  8             accumulation. Environ. Health Perspect. 108: 527-531.
  9     Hanninen, H.; Aitio,  A.; Kovala, T.; Luukkonen, R.; Matikainen, E.; Mannelin, T.; Erkkila, J.; Riihimaki, V. (1998)
10             Occupational exposure to lead and neuropsychological dysfunction. Occup. Environ. Med. 55: 202-209.
11     Haraguchi, T.; Ishizu, H.; Takehisa, Y.; Kawai, K.; Yokota, O.; Terada, S.; Tsuchiya, K.; Ikeda, K.; Morita, K.;
12             Horike, T.; Kira, S.; Kuroda, S. (2001) Lead content of brain tissue in diffuse neurofibrillary tangles with
13             calcification (DNTC): the possibility of lead neurotoxicity. Neuroreport 12: 3887-3890.
14     Haraguchi, T.; Ishizu, H.; Takehisa, Y.; Kawai, K.; Yokota, O.; Terada, S.; Tsuchiya, K.; Ikeda, K.; Morita, K.;
15             Horike, T.; Kira, S.; Kuroda, S. (2002) Lead content of brain tissue in diffuse neurofibrillary tangles with
16             calcification (DNTC): the possibility of lead neurotoxicity [erratum to Neuroreport 12: 3887-3890].
17             Neuroreport 13(1): inside back  cover.
18     Harville, E. W.; Hertz-Picciotto, I.; Schramm, M.; Watt-Morse, M.; Chantala, K.; Osterloh, J.; Parsons, P. J.; Rogan,
19             W. (2005) Factors influencing the difference between maternal and cord blood lead. Occup. Environ. Med.
20             62:263-290.
21     Hasselblad, V.; Nelson, W. (1975) Additional analysis of the seven city lead study. In:  Griffin, T. B.; Knelson, J. H.,
22             eds. Lead. Stuttgart, Federal Republic  of Germany: Georg Thieme Publishers; pp. 147-151. (Coulston, F.;
23             Korte, F., eds. Environmental quality and safety: supplement v. 2).
24     Hasselblad, V.; Stead, A. G.; Galke, W. (1980) Analysis of coarsely grouped data from the lognormal distribution.
25             JASA J. Am. Stat. Assoc. 75: 771-778.
26     Hatzakis, A.; Salaminios, F.; Kokevi,  A.; Katsouyanni, K.; Maravelias, K.; Kalandidi, A.; Koutselinis,  A.; Stefanis,
27             K.; Trichopoulos, D. (1985) Blood lead and classroom behaviour of children in two communities with
28             different degree of lead exposure: evidence of a dose-related effect? In: Lekkas, T. D., ed. International
29             conference:  heavy metals in the environment, v. 1; September; Athens, Greece. Edinburgh, United
30             Kingdom: CEP Consultants,  Ltd.; p. 47.
31     Hatzakis, A.; Kokkevi, A.; Maravelias, C.; Katsouyanni, K.; Salaminios, F.; Kalandidi, A.; Koutselinis, A.; Stefanis,
32             C.; Trichopoulos, D. (1989) Psychometric intelligence deficits in lead-exposed children. In: Smith, M. A.;
33             Grant, L. D.; Sors, A. I., eds. Lead exposure and child development: an international assessment [workshop
34             organized by the Commission of the European Communities and the U.S. Environmental Protection
35             Agency]; September 1986; Edinburgh, United Kingdom. Dordrecht, The Netherlands: Kluwer Academic
36             Publishers BV; pp. 211-223.
37     Haynes, E. N.; Kalkwarf, H. J.; Hornung, R.; Wenstrup, R.; Dietrich, K.; Lanphear, B. P. (2003) Vitamin D receptor
3 8             Fokl polymorphism and blood lead concentration in children. Environ. Health Perspect. Ill:  1665-1669.
39     He, F. S.; Zhang, S. L.; Li, G.; Zhang, S. C.;  Huang, J. X.; Wu, Y. Q. (1988) An electroneurographic assessment of
40             subclinical lead neurotoxicity. Int. Arch. Occup. Environ. Health 61: 141-146.
41     Hellstrom, L.; Blinder,  C.-G.; Dahlberg,  B.; Lundberg, M.; Jarup, L.; Persson, B.; Axelson, O. (2001) Cadmium
42             exposure and end-stage renal disease. Am. J. Kidney Dis. 38:  1001-1008.
43     Hemdan, N. Y. A.; Emmrich, F.; Adham, K.; Wichmann, G.; Lehmann, I.; El-Massry, A.; Ghoneim, H.; Lehmann,
44             J.; Sack, U.  (2005) Dose-dependent modulation of the in vitro cytokine production of human immune
45             competent cells by lead salts. Toxicol. Sci. 86: 75-83.
46     Henderson, D. A. (1954) A follow-up of cases of plumbism in children. Australas. Ann. Med. 33: 219-224.
47     Henderson, D. A. (1955) Chronic nephritis in Queensland. Australas. Ann. Med.  4: 163-177.
48     Hense, H. W.; Filipiak, B.; Keil, U. (1993) The association of blood lead and blood pressure in population surveys.
49             Epidemiology 4: 173-179.
50     Hense, H. W.; Filipiak, B.; Keil, U. (1994) Alcohol consumption as a modifier of the relation between  blood lead
51             and blood pressure. Epidemiology 5: 120-123.
52     Heo, Y.; Lee, B.-K.;  Ahn, K.-D.; Lawrence, D. A. (2004) Serum  IgE elevation correlates with blood lead levels in
53             battery manufacturing workers. Hum. Exp. Toxicol. 23:  209-213.
54     Hernandez-Avila, M.; Gonzalez-Cossio, T.; Palazuelos, E.; Romieu, L; Aro, A.; Fishbein, E.; Peterson, K. E.;
55             Hu, H. (1996) Dietary and environmental determinants of blood and bone lead levels in lactating
56             postpartum women living in Mexico City. Environ. Health Perspect.  104: 1076-1082.
       December 2005                                6-375         DRAFT-DO NOT QUOTE OR CITE

-------
  1    Hernandez-Avila, M; Sanin, L. H.; Romieu, I.; Palazuelos, E.; Tapia-Conyer, R.; Olaiz, G.; Rojas, R.; Navarrete, J.
  2            (1997) Higher milk intake during pregnancy is associated with lower maternal and umbilical cord lead
  3            levels in postpartum women. Environ. Res. 74: 116-121.
  4    Hernandez-Avila, M.; Smith, D.; Meneses, F.; Sanin, L. H.; Hu, H. (1998) The influence of bone and blood lead on
  5            plasma lead levels in environmentally exposed adults. Environ. Health Perspect. 106: 473-477.
  6    Hernandez-Avila, M.; Villalpano, C. G.; Palazuelos, E.; Villapando, M. E. G. (2000) Determinants of blood lead
  7            levels across the menopausal transition. Arch. Environ. Health 53: 355-360.
  8    Hernandez-Avila, M.; Peterson, K. E.; Gonzalez-Cossio, T.; Sanin, L. H.; Aro, A.; Schnaas, L.; Hu, H. (2002) Effect
  9            of maternal bone lead on length and head circumference of newborns and 1-month-old infants. Arch.
10            Environ. Health 57: 482-488.
11    Hernandez-Avila, M.; Gonzalez-Cossio, T.; Hernandez-Avila, J.  E.; Romieu, L; Peterson, K. E.; Aro, A.;
12            Palazuelos, E.; Hu, H. (2003) Dietary calcium supplements to lower blood lead levels in lactating women: a
13            randomized placebo-controlled trial. Epidemiology 14: 206-212.
14    Hernandez-Ochoa, I; Garcia-Vargas, G; Lopez-Carrillo, L; Rubio-Andrade, M; Moran-Martinez, J; Cebrian, M. E.;
15            Quintanilla-Vega, B. (2005) Low lead environmental exposure alters semen quality and sperm chromatin
16            condensation in northern Mexico.  Reprod. Toxicol. 20: 221-228.
17    Hernberg, S.; Nikkanen, J.; Mellin, G.; Lilius, H. (1970) "delta"-aminolevulinic acid dehydrase as a measure of lead
18            exposure. Arch. Environ. Health 21: 140-145.
19    Hertz-Picciotto, L; Croft, J. (1993) Review of the relation between blood lead and blood pressure. Epidemiol. Rev.
20            15:352-373.
21    Hertz-Picciotto, L; Schramm, M.; Watt-Morse, M.; Chantala, K.; Anderson, J.; Osterloh, J. (2000) Patterns and
22            determinants of blood lead during pregnancy. Am. J. Epidemiol. 152: 829-837.
23    Hill, A. B. (1965) The environment and disease: association or causation? Proc. R. Soc. Med. 58: 295-300.
24    Hirata, M.; Kosaka, H.; Yoshida, T. (2004) A study on the effect of lead on event-related potentials among lead-
25            exposed workers. Ind. Health 42: 431-434.
26    Hisanaga, A.; Eguchi, Y.; Hirata, M.; Ishinishi, N. (1988) Lead levels in ancient and contemporary Japanese bones.
27            Biol. Trace Elem. Res. 16: 77-85.
28    Hogstedt, C.; Hane, M.; Agrell, A; Bodin, L. (1983) Neuropsychological test results and symptoms among workers
29            with well-defined long-term exposure to lead. Br. J. Ind. Med. 40:  99-105.
30    Holness, D. L.; Nethercott, J. R. (1988) Acute lead intoxication in a group of demolition workers. Appl. Ind. Hyg.
31            3:338-341.
32    Holstein, Y.; Pratt, H.; Goldsher, M.; Rosen, G.;  Shenhav, R.; Linn, S.; Mor, A.; Barkai, A. (1986) Auditory
33            brainstem evoked potentials in asymptomatic lead-exposed subjects. J. Laryngol. Otol. 100: 1031-1036.
34    Hoppin, J. A.; Aro, A.; Hu, H.; Ryan, P. B. (2000) Measurement variability associated with KXRF bone lead
3 5            measurement in young adults. Environ. Health Perspect. 108: 239-242.
36    Horiguchi, S.; Endo,  G.; Kiyota, I. (1987) Measurement of total triiodothyronine (T3), total thyroxine (T4) and
37            thyroid-stimulating hormone (TSH) levels in lead-exposed workers. Osaka City Med J. 33: 51-56.
38    Horiguchi, S.; Matsumura, S.; Fukumoto, K.; Karai, I.; Endo, G.; Teramoto, K.; Shinagawa, K.; Kiyota, I.;
39            Wakitani, F.; Takise, S.; Kawaraya, T. (1991) Erythrocyte deformability in workers exposed to lead. Osaka
40            City Med. J. 37: 149-155.
41    Hotz, P.; Buchet, J. P.; Bernard, A.; Lison,  D.; Lauwerys, R. (1999) Renal effects of low-level environmental
42            cadmium exposure: 5-year follow-up of a subcohort from the Cadmibel study. Lancet 354: 1508-1513.
43    Houston, D. K.; Johnson, M. A. (1999) Lead as a risk factor for hypertension in women. Nutr. Rev. 57: 277-279.
44    Hsiao, C. Y.; Wu, H. D.; Lai, J. S.; Kuo, H. W. (2001) A longitudinal study of the effects of long-term exposure to
45            lead among  lead battery factory workers in Taiwan (1989-1999). Sci. Total Environ. 279: 151-158.
46    Hsieh, L. L.; Liou, S. H.; Chen, Y. H.; Tsai, L. C.; Yang, T.; Wu, T. N. (2000) Association between aminolevulinate
47            dehydrogenase genotype and blood lead levels in Taiwan. J. Occup. Environ. Med. 42(2): 151-155.
48    Hu, H. (1991) A 50-year follow-up of childhood plumbism: hypertension, renal function, and hemoglobin levels
49            among survivors. Am. J.  Dis. Child. 145: 681-687.
50    Hu, H. (1998) Bone lead as a new biologic marker of lead dose: recent findings and implications for public health.
51            Environ. Health Perspect. 106(suppl. 4): 961-967.
52    Hu, H.; Hernandez-Avila, M. (2002) Lead, bones, women, and pregnancy~the poison within? Am. J. Epidemiol.
53            156: 1088-1091.
54    Hu, H.; Milder, F. L.; Burger, D. E. (1990) X-ray fluorescence measurements of lead burden in subjects with low-
55            level community lead exposure. Arch. Environ. Health 45: 335-341.
      December 2005                                6-376         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Hu, H.; Milder, F. L.; Burger, D. E. (1991) The use of K X-ray fluorescence for measuring lead burden in
  2            epidemiological studies: high and low lead burdens and measurement uncertainty. Environ. Health
  3            Perspect. 94:  107-110.
  4     Hu, H.; Watanabe, H.; Payton, M; Korrick, S.; Rotnitzky, A. (1994) The relationship between bone lead and
  5            hemoglobin. JAMA J. Am. Med. Assoc. 272: 1512-1517.
  6     Hu, H.; Aro, A.; Rotnitzky, A. (1995) Bone lead measured by X-ray fluorescence: epidemiologic methods. Environ.
  7            Health Perspect. 103(suppl.  1): 105-110.
  8     Hu, H.; Aro, A.; Payton, M.; Korrick, S.; Sparrow, D.; Weiss, S. T.; Rotnitzky, A. (1996) The relationship of bone
  9            and blood lead to hypertension. The Normative Aging Study. JAMA J. Am. Med. Assoc. 275: 1171-1176.
10     Hu, H.; Rabinowitz, M.; Smith, D. (1998) Bone lead as a biological marker in epidemiologic studies of chronic
11            toxicity: conceptual paradigms. Environ. Health Perspect. 106:  1-8.
12     Hu, J.; La Vecchia, C; Negri, E.; Chatenoud, L.; Bosetti, C;  Jia, X.; Liu, R.; Huang, G.; Bi, D.; Wang, C. (1999)
13            Diet and brain cancer in adults: a case-control study  in northeast China. Int. J. Cancer 81: 20-23.
14     Hu, H.; Wu, M.-T.; Cheng, Y.;  Sparrow, D.; Weiss, S.; Kelsey, K. (2001) The "delta"-aminolevulinic acid
15            dehydratase (ALAD) polymorphism and bone and blood lead levels in community-exposed men: the
16            Normative Aging Study. Environ. Health Perspect. 109: 827-832.
17     Huang, J.; He, F.; Wu, Y.; Zhang, S.  (1988) Observations on  renal function in workers exposed to lead. Sci. Total
18            Environ. 71:535-537.
19     Huel, G.; Boudene, C.; Ibrahim, M. A. (1981) Cadmium and  lead content of maternal and newborn hair: relationship
20            to parity, birth weight, and hypertension. Arch. Environ, health 36: 221-227.
21     Huel, G.; Campagna, D.; Moreau, T.; Tubert-Bitter, P. (1995) Environmental lead and children's intelligence: hair
22            lead studies were excluded [letter]. Br. Med. J. 310:  397-398.
23     Hunter, J.; Urbanowicz, M. A.; Yule, W.; Lansdown, R. (1985) Automated testing of reaction time and its
24            association with lead in children. Int. Arch. Occup. Environ. Health 57: 27-34.
25     Hwang, K.-Y.; Lee, B.-K.; Bressler, J. P.; Bolla, K. I.; Stewart, W. F.; Schwartz, B.  S.  (2002) Proteinkinase C
26            activity and the relations between blood lead and neurobehavioral function in lead workers. Environ.
27            Health Perspect. 110: 133-138.
28     Inskip, M. J.; Franklin, C. A.; Baccanale, C. L.; Manton, W. I.; O'Flaherty, E. J.; Edwards, C. M. H.; Blenkinsop,
29            J. B.; Edwards, E. B. (1996) Measurement of the flux of lead from bone to blood in a nonhuman primate
30            (Macaca fascicularis) by sequential administration of stable lead isotopes. Fundam. Appl. Toxicol.
31            33:235-245.
32     Irgens, A.; Kruger, K.; Skorve,  A. H.; Irgens, L. M. (1998) Reproductive outcome in offspring of parents
33            occupationally exposed to lead in Norway. Am. J. Ind. Med. 34: 431-437.
34     International Agency for Research on Cancer. (1980) Some metals and metallic compounds. World Health
3 5            Organization, International Agency for Research on Cancer, Lyon, France; IARC Monogr. Eval. Carcinog.
36            Risks Hum. Vol. 23.
37     International Agency for Research on Cancer. (2005) Inorganic and organic lead compounds. World Health
3 8            Organization, International Agency for Research on Cancer, Lyon, France; IARC Monogr. Eval. Carcinog.
39            Risks Hum. Vol. 87: in preparation.
40     International Commission on Radiological Protection. (1973) Alkaline earth metabolism in adult man. ICRP
41            Publication20. HealthPhys. 24: 125-221.
42     International Commission on Radiological Protection. (1981) Report of the task group on reference man. Oxford,
43            United Kingdom: Elsevier Science Publishers; ICRP Publication 23.
44     International Commission on Radiological Protection. (1996) Basic anatomical & physiological data for use in
45            radiological protection: the skeleton. Oxford, United Kingdom: Elsevier Science Publishers; ICRP
46            publication 70; Annals of the ICRP, v. 25, no. 2.
47     Ishida, M.; Ishizaki, M.; Yamada, Y. (1996) Decreases in postural change of finger blood flow in ceramic painters
48            chronically exposed to low level lead. Am. J. Ind. Med. 29: 547-553.
49     Ito, Y.; Niiya, Y.; Kurita, H.; Shima,  S.; Sarai, S. (1985) Serum lipid peroxide level  and blood superoxide dismutase
50            activity in workers with occupational exposure to lead. Int. Arch. Occup. Environ. Health 56: 119-127.
51     Jackson, L. W.; Correa-Villasenor, A.; Lee, P. S. J.; Dominici, F.; Stewart,  P. A.; Breysse, P. N. (2004) Parental lead
52            exposure and total anomalous pulmonary venous return. Birth Defects Res. Part A 70:  185-193.
53     Jarup, L.; Hellstrom, L.; Alfven, T.; Carlsson, M. D.; Grubb,  A.; Persson, B.; Pettersson, C.; Spang, G.; Schutz, A.;
54            Blinder, C.-G. (2000) Low level exposure to cadmium and early kidney damage: the OSCAR study. Occup.
55            Environ. Med. 57: 668-672.
       December 2005                                6-377        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Jemal, A.; Graubard, B. I.; Devesa, S. S.; Flegal, K. M. (2002) The association of blood lead level and cancer
  2             mortality among whites in the United States. Environ. Health Perspect. 110: 325-329.
  3     Jiun, Y. S.; Hsien, L. T. (1994) Lipid peroxidation in workers exposed to lead. Arch. Environ. Health 49: 256-259.
  4     Johnson, R. I; Kang, D.-H.; Feig, D.; Kivlighn, S.; Kanellis, I; Watanabe, S.; Tuttle, K. R.; Rodriguez-Iturbe, B.;
  5             Herrera-Acosta, J.; Mazzali, M.  (2003) Is there a pathogenetic role for uric acid in hypertension and
  6             cardiovascular and renal disease? Hypertension 41: 1183-1190.
  7     Jones, R. R. (1995) Environmental lead and children's intelligence: obvious hypothesis is ignored [letter]. Br. Med.
  8             J. 310:397.
  9     Jones, S. J.; Williams, A. J.; Kudlac, H.; Hainsworth, I. R.; Morgan, W. D. (1990) The measurement of bone lead
10             content in patients with end stage failure. In: Yasumura, S.; Harrison, J. E., eds. In vivo body composition
11             studies: recent advances. New York, NY: Plenum Press; pp. 259-262. (Basic life sciences: v.  55)
12     Jordan, C.; Lee, P.; Shapiro, E. (2000) Measuring developmental outcomes of lead exposure in an urban
13             neighborhood: the challenges of community-based research. J. Exposure Anal. Environ. Epidemiol.
14             10:-732-742.
15     Juarez-Perez, C. A.; Aguilar-Madrid, G.;  Smith, D. R.; Lacasana-Navarro, M.; Tellez-Rojo, M. M.; Piacitteli, G.;
16             Hu, H.; Hernandez-Avila, M. (2004) Predictors of plasma lead among lithographic print shop workers in
17             Mexico City. Am. J. Ind. Med. 46: 245-252.
18     Juberg, D. R.; Kleiman, C. F.; Kwon, S. C. (1997) Position paper of the American Council on Science and Health:
19             lead and human health. Ecotoxicol. Environ. Saf. 38: 162-180.
20     Jung, K.-Y.; Lee, S.-J.; Kim, J.-Y.; Hong, Y.-S.; Kim, S.-R.; Kim, D.-I.; Song, J.-B. (1998) Renal dysfunction
21             indicators in lead exposed workers. J. Occup. Health 40: 103-109.
22     Jurek, A. M.; Greenland, S.; Maldonado,  G.; Church, T. R. (2005) Proper interpretation of non-differential
23             misclassification effects: expectations vs observations. Int. J. Epidemiol. 34: 680-687.
24     Kahn, C. A.; Kelly, P. C.; Walker, W. O., Jr. (1995) Lead screening in children with attention deficit hyperactivity
25             disorder and developmental delay. Clin. Pediatr. 34: 498-501.
26     Kamel, F.; Umbach, D.; Munsat, T.; Shefner,  J.; Hu, H.; Sandier, D. (2002) Lead exposure and amyotrophic later
27             sclerosis. Epidemiology 13: 311-319.
28     Kamel, F.; Umbach, D. M.; Lehman, T. A.; Park, L. P.; Munsat, T. L.; Shefner, J. M.; Sandier, D. P.; Hu, H.;
29             Taylor, J. A. (2003) Amyotrophic lateral sclerosis, lead, and genetic susceptibility: polymorphisms in the
3 0             "delta"-aminolevulinic acid dehydratase and vitamin D receptor genes. Environ. Health Perspect.  Ill:
31             1335-1339.
32     Kandiloris, D. C.; Goletsos, G. A.; Nikolopoulos, T. P.;  Ferekidis, E. A.; Tsomis, A. S.; Adamopoulos, G. K. (1997)
3 3             Effect of subclinical lead intoxication on laryngeal cancer. Br. J.  Clin. Practice 51: 69-70.
34     Kannel, W. B. (1996) Blood pressure as a cardiovascular risk factor: prevention and treatment. JAMA J. Am. Med.
35             Assoc. 275: 1571-1576.
36     Kannel, W. B. (2000a) Elevated systolic blood presure as a cardiovascular risk factor.  Am. J. Cardiol.  85: 251-255.
37     Kannel, W. B. (2000b) Risk stratification in hypertension: new insights from the Framingham Study. Am. J.
38             Hypertens. 13: 3S-10S.
39     Kannel, W. B.; Wilson, P.  W. F.; Nam, B.-H.; D'Agostino, R. B.; Li, J. (2004) A likely explanation for the J-curve
40             of blood pressure cardiovascular risk. Am. J. Cardiol. 94: 380-384.
41     Karakaya, A. E.; Ozcagli, E.; Ertas, N.; Sardas, S. (2005) Assessment of abnormal DNA repair responses and
42             genotoxic effects  in lead exposed workers. Am. J. Ind. Med. 47: 358-363.
43     Karimi, P. G.; Moodley, J.; Jinabhai, C. C.; Nriagu, J. (1999) Maternal and fetal blood lead levels. S. Afr. Med. J.
44             89:676-679.
45     Karmaus, W.; Brooks, K. R.; Nebe, T.; Witten, J.; Obi-Osius, N.; Kruse, H. (2005) Immune functionbiomarkers in
46             children exposed to lead and organochlorine compounds: a cross-sectional study. Environ. Health Glob.
47             Access Sci. 4: 1-10.
48     Kasperczyk, S.; Dziwisz, M.; Kasperczyk, A.; Birkner, E. (2002) Wplyw olowiu na wystepowanie nadcisnienia
49             tetniczego [Influence of lead exposure on arterial hypertension]. Wiad. Lek. 55(suppl. 1): 230-234.
50     Kaufman, A. S.; Kaufman, N. L. (1983) Kaufman assessment battery for children. Circle Pines, MN: American
51             Guidance Service.
52     Kauppinen, T.; Riala, R.; Seitsamo, J.; Hernberg, S. (1992) Primary liver cancer and occupational exposure. Scand.
53             J. Work Environ.  Health. 18: 18-25.
54
       December 2005                                6-378         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Kavlock, R. I; Daston, G. P.; DeRosa, C.; Fenner-Crisp. P.; Gray, L. E.; Kaattari, S.; Lucier, G.; Luster, M.; Mac,
  2            M. I; Maczka, C.; Miller, R.; Moore, I; Rolland, R.; Scott, G.; Sheehan, D. M.; Sinks, T.; Tilson, H. A.
  3            (1996) Research needs for the risk assessment of health and environmental effects of endocrine disrupters:
  4            a report of the U.S. EPA-sponsored workshop. Environ. Health Perspect. Suppl. 104(4): 715-740.
  5     Kehoe, R. A. (1987) Studies of lead administration and elimination in adult volunteers under natural and
  6            experimentally induced conditions over extended periods of time. Food Chem. Toxicol. 25: 425-493.
  7     Keiding, N.; Skakkebaek, N. E. (1996) Sperm decline-real or artifact? Fertil. Steril. 65: 450-453.
  8     Keiding, N.; Giwercman, A.; Carlsen, E.; Skakkebaek, N. E. (1994) Comment on "Farrow, S. (1994) Falling sperm
  9            quality: fact or fiction? Br. Med. J. 309: 1-2." Br. Med. J. 309: 131.
10     Keinonen, M. (1992) The isotopic composition of lead in man and the environment in Finland 1966-1987: isotope
11            ratios of lead as indicators of pollutant source. Sci. Total Environ. 113: 251-268.
12     Kelada, S. N.; Shelton, E.; Kaufmann, R. B.; Khoury, M. J. (2001) "Delta"-aminolevulinic acid dehydratase
13            genotype and lead toxicity: a HuGE review. Am. J. Epidemiol. 154: 1-13.
14     Kemp, F. W.; Penn-Erskine, C.; Coba, V.; Wenger, P.; Palmer-Keenan, D.; Lundt, M. R.; Davidow, A.; Louria,
15            D. B.; Bogden, J. D. (2002) Recent increases in dietary calcium intake in young urban children at high risk
16            of lead poisoning. FASEB J. 16: A253.
17     Kemper, A. R.; Cohn,  L. M.; Fant, K. E.; Dombkowski, K. J.;  Hudson, S. R. (2005) Follow-up testing among
18            children with elevated screening blood lead levels. JAMA J. Am. Med. Assoc. 293: 2232-2237.
19     Kessler, M.; Durand, P. Y.; Huu, T. C.; Royer-Morot, M. J.; Chanliau, J.; Netter, P.; Due, M. (1999) Mobilization of
20            lead from bone in end-stage renal failure patients with secondary hyperparathyroidism. Nephrol. Dial.
21            Transplant. 14:2731-2733.
22     Khalil-Manesh, F.; Gonick, H. C. Cohen, A. H.; Alinovi, R.; Bergamaschi, E.; Mutti, A.; Rosen, V. J. (1992a)
23            Experimental model of lead nephropathy. I. Continuous high-dose lead administration. Kidney Int.
24            41: 1192-1203.
25     Khalil-Manesh, F.; Gonick, H. C.; Cohen, A.; Bergamaschi, E.; Mutti, A. (1992b) Experimental model of lead
26            nephropathy.  II. Effect of removal from lead exposure and chelation treatment with dimercaptosuccinic
27            acid (DMSA). Environ. Res 58: 35-54.
28     Khalil-Manesh, F.; Gonick, H. C.; Cohen, A. H. (1993) Experimental model of lead nephropathy. III. Continuous
29            low-level lead administration. Arch. Environ. Health  48: 271-278.
30     Kim, R.; Aro, A.; Rotnitzky, A.; Amarasiriwardena, C.; Hu, H. (1995) K x-ray fluorescence measurements of bone
31            lead concentration: the analysis of low-level data. Phys. Med. Biol. 40: 1475-1485.
32     Kim, R.; Hu, H.; Rotnitzky, A.; Bellinger, D.; Needleman, H. (1995) A longitudinal study of chronic lead exposure
33            and physical growth in Boston children. Environ. Health Perspect. 103:  952-957.
34     Kim, R.; Rotnitsky, A.; Sparrow, D.; Weiss, S. T.; Wager, C.;  Hu, H. (1996) A longitudinal study of low-level lead
35            exposure and impairment  of renal function. The Normative Aging Study. JAMA J. Am. Med. Assoc.
36            275: 1177-1181.
37     Kim, Y.; Lee, H.; Lee, C. R.; Park, D. U.; Yang, J. S.; Park, I.  J.; Lee, K. Y.; Lee, M.; King, T. K.; Sohn, N. S.; Cho,
38            Y. S.; Lee, N.; Chung, H.  K. (2002) Evaluation of lead exposure in workers at secondary lead smelters in
3 9            South Korea: with focus on activity of erythrocyte pyrimidine 5'-nucleotidase (P5N). Sci. Total Environ.
40            286: 181-189.
41     Kimber, I.; Stonard, M. D.; Gidlow, D. A.; Niewola, Z. (1986) Influence of chronic low-level exposure to lead on
42            plasma immunoglobulin concentration and cellular immune function in man. Int. Arch. Occup. Environ.
43            Health 57: 117-125.
44     Kline, J.; Stein, Z.; Hutzler, M. (1987) Cigarettes, alcohol and marijuana: varying associations with birthweight. Int.
45            J. Epidemiol. 16:44-51.
46     Klitzman, S.; Sharma,  A.; Nicaj, L.; Vitkevich, R.; Leighton, J. (2002) Lead poisoning among pregnant women in
47            New York City: risk factors and screening practices. Bull. N. Y.  Acad. Med. 79: 225-237.
48     Koller, K.; Brown, T.; Spurgeon, A.; Levy, L. (2004) Recent developments in low-level lead exposure and
49            intellectual impairment in children. Environ. Health Perspect. 112: 987-994.
50     Konishi, Y.; Endo, G.; Kiyota, A.; Horiguchi, S. (1994) Fractional clearances of low molecular weight proteins in
51            lead workers. Ind. Health  32: 119-127.
52     Koo, W. W. K.; Succop, P. A.; Bornschein, R. L.; Krugwispe, S. K.; Steinchen, J. J.; Tsang, R.  C.; Berger, O.G.
53            (1991) Serum vitamin D metabolites and bone mineralization in young children with chronic low to
54            moderate lead exposure. Pediatrics 87: 680-687.
       December 2005                                6-379        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Kordas, K.; Lopez, P.; Rosado, J. L.; Vargas, G. G.; Rico, J. A.; Ronquillo, D.; Cebrian, M. E.; Stoltzfus, R. J.
  2             (2004) Blood lead, anemia, and short stature are independently associated with cognitive performance in
  3             Mexican school children. J. Nutr. 134: 363-371.
  4     Kordas, K.; Stoltzfus, R. J.; Lopez, P.; Rico, J. A.; Rosado, J. L. (2005) Iron and zinc supplementation does not
  5             improve parent orteacher ratings of behavior in first grade Mexican children exposed to lead. J. Pediatr.
  6             147:632-639.
  7     Korrick, S.; Hunter, D.; Rotnitzky, A.; Hu, H.; Nurses' Health Study Research Group. (1996) Lead and hypertension
  8             in a sample of the Nurses' Health Study. Am. J. Epidemiol.  143(11 suppl.): S44.
  9     Korrick, S. A.; Hunter, D. J.; Rotnitzky, A.; Hu, H.; Speizer, F. E. (1999) Lead and hypertension in a sample of
10             middle-aged women. Am. J. Public Health 89: 330-335.
11     Korrick, S. A.; Schwartz, J.; Tsaih, S.-W.; Hunter, D. J.; Aro, A.;Rosner, B.; Speizer, F. E.; Hu, H. (2002) Correlates
12             of bone and blood lead levels among middle-aged and elderly women. Am. J. Epidemiol.  156: 335-343.
13     Kosnett, M. J.; Becker, C. E.; Osterloh, J. D.; Kelly, T. J.; Pasta, D. J. (1994) Factors influencing bone lead
14             concentration in a suburban community assessed by noninvasive K x-ray. JAMA J. Am. Med. Assoc.
15             271: 197-203.
16     Koster, J.; Erhardt, A.; Stoeppler, M.; Mohl, C.; Ritz, E. (1989) Mobilizable lead in patients with chronic renal
17             failure. Eur. J. Clin. Invest. 19: 228-233.
18     Kovala, T.; Matikainen, E.; Mannelin, T.; Erkkila, J.; Riihimaki, V.; Hanninen, H.; Aitio, A. (1997) Effects of low
19             level exposure to lead on neurophysiological functions among lead battery workers.  Occup. Environ. Med.
20             54:487-493.
21     Kramer,  M. S. (1987) Intrauterine growth and gestational duration determinants. Pediatrics 80: 502-511.
22     Kramer,  M. S.; McLean, F. H.; Boyd, M. E.; Usher, R. H. (1988) The validity of gestational age estimation by
23             menstrual dating in term, preterm, and postterm gestations.  JAMA J. Am. Med. Assoc. 260: 3306-3308.
24     Kristal-Boneh, E.; Froom, P.; Yerushalmi, N.; Harari, G.; Ribak, J. (1998) Calcitropic hormones and occupational
25             lead exposure. Am. J.  Epidemiol. 147: 458-463.
26     Kristal-Boneh, E.; Coller, D.; Froom, P.; Harari, G.; Ribak, J. (1999) The association between occupational lead
27             exposure and serum cholesterol and lipoprotein levels. Am. J. Public Health 89: 1083-1087.
28     Kristensen, P.; Irgens, L. M.; Daltveit, A. K.; Andersen, A. (1993) Perinatal outcome among children of men
29             exposed to lead and organic solvents in the printing industry. Am. J. Epidemiol. 137: 134-144.
30     Kromhout, D.; Wibowo, A. A.  E.; Herber, R. F. M.; Dalderup, L. M.; Heerdink, H.;  de Lezenne Coulander, C.;
31             Zielhuis, R. L. (1985) Trace metals and coronary heart disease risk indicators in 152 elderly men (the
32             Zutphen study). Am. J. Epidemiol.  122: 378-385.
33     Kumar, B. D.; Krishnaswamy,  K. (1995) Detection of occupational lead nephropathy using early renal markers.
34             J. Toxicol. Clin. Toxicol. 33: 331-335.
35     Kumar, A.; Dey, P. K.; Singla,  P. N.; Ambasht, R. S.; Upadhyay, S. K. (1998) Blood lead levels in children with
36             neurological disorders. J. Trop. Pediatr. 44: 320-322.
37     Kuo, H.-W.; Hsiao, T.-Y.;  Lai, J.-S. (2001) Immunological effects of long-term lead exposure among Taiwanese
38             workers. Arch. Toxicol.  75: 569-573.
39     Kurtin, D.; Therrell, B. L., Jr.; Patterson, P. (1997) Demographic risk factors associated with  elevated lead levels in
40             Texas children covered by Medicaid. Environ. Health Perspect. 105: 66-68.
41     Lacey, R. F.; Moore, M. R.; Richards, W. N. (1985) Lead in water, infant diet and blood: the  Glasgow Duplicate
42             Diet Study. Sci. Total Environ. 41: 235-257.
43     Lagerkvist, B. J.; Ekesrydh, S.; Englyst, V.; Nordberg, G.  F.; Soderberg, H.-A.; Wiklund, D.-E. (1996) Increased
44             blood lead and decreased calcium levels during pregnancy:  a prospective study of Swedish women living
45             near a smelter. Am. J.  Public Health 86: 1247-1252.
46     Lancranjan, I.; Popescu, H. I.; Gavanescu, O.; Klepsch, L; Serbanescu, M. (1975) Reproductive ability of workmen
47             occupationally exposed to lead. Arch. Environ. Health 30: 396-401.
48     Landrigan, P. J. (2000) Pediatric lead poisoning: is there a threshold? Public Health Rep. 115:530-531.
49     Lanphear, B. P. (2005) Childhood lead poisoning prevention: too little, too late. JAMA J. Am. Med. Assoc.
50             293:2274-2276.
51     Lanphear, B. P.; Winter, N. L.; Apetz, L.; Eberly, S.; Weitzman, M.  (1996) A randomized trial of the effect of dust
52             control on children's blood lead levels. Pediatrics 98: 35-40.
53     Lanphear, B. P.; Burgoon,  D. A.;  Rust, S. W.; Eberly, S.; Galke, W.  (1998) Environmental exposures to  lead and
54             urban children's blood lead levels. Environ. Res. 76: 120-130.
       December 2005                                6-3 80         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Lanphear, B. P.; Howard, C.; Eberly, S.; Auinger, P.; Kolassa, I; Weitzman, M; Schaffer, S. J.; Alexander, K.
  2             (1999) Primary prevention of childhood lead exposure: a randomized trial of dust control. Pediatrics
  3             103:772-777.
  4     Lanphear, B. P.; Dietrich, K.; Auinger, P.; Cox, C. (2000) Cognitive deficits associated with blood lead
  5             concentrations < 10 "mu"g/dL in U.S. children and adolescents. Public Health Rep. 115: 521-529.
  6     Lanphear, B. P.; Dietrich, K. N.; Berger, O. (2003) Prevention of lead toxicity in US children. Ambul. Pediatr.
  7             3:27-36.
  8     Lanphear, B. P.; Hornung, R.; Khoury, J.; Yolton, K.; Baghurst, P.; Bellinger, D. C.; Canfield, R. L.; Dietrich,
  9             K. N.; Bornschein, R.; Greene, T.; Rothenberg, S. J.; Needleman, H. L.; Schnaas, L.; Wasserman, G.;
10             Graziano, J.; Roberts, R. (2005) Low-level environmental lead exposure and children's intellectual
11             function: an international pooled analysis. Environ. Health Perspect. 113: 894-899.
12     Last, J. M. (2001) A dictionary of epidemiology. New York, NY: Oxford University Press.
13     Laudanski, T.; Sipowicz, M.; Modzelewski, P.; Bolinski, J.; Szamatowicz, J.; Razniewska, G.; Akerlund, M. (1991)
14             Influence of high lead and cadmium soil content on human reproductive outcome. Int.  J. Gynecol. Obstet.
15             36:309-315.
16     Laughlin, N. K. (1995) A new approach for the study of the neurotoxicity of lead. Neurotoxicol. Teratol.
17             17:235-236.
18     Laughlin, N. K.; Bowman, R. E.; Franks, P. A.; Dierschke, D. J. (1987) Altered menstural cycles in rhesus monkeys
19             induced by lead. Fundam. Appl. Toxicol. 9: 722-729.
20     Lauwers, M. C.; Hauspie, R. C.; Susanne, C.; Verheyden, J. (1986) Comparison of biometric data of children with
21             high and low levels of lead in the blood. Am. J. Phys. Anthropol. 69: 107-116.
22     Laxen, D. P. H.; Raab, G. M.; Fulton, M. (1987) Children's blood lead and exposure to lead in household dust and
23             water - a basis for an environmental standard for lead in dust. Sci. Total Environ. 66: 235-244.
24     Lazutka, J. R.; Lekevicius, R.; Dedonyte, V.; Maciuleviciute-Gervers, L.; Mierauskiene, J.; Rudaitiene, S.; Slapsyte,
25             G. (1999) Chromosomal aberrations and sister-chromatid exchanges in Lithuanian populations: effects of
26             occupational and environmental exposures. Mutat. Res. 445: 225-239.
27     Leal-Garza, C.; Moates,  D. O. R.; Cerda-Flores, R. M.; et al. (1986) Frequency of sister-chromatid exchanges (SCE)
28             in lead exposed workers. Arch. Invest. Med. 17: 267-276.
29     Lee, B.-K.; Ahn, K.-D.;  Lee, S.-S.; Lee, G.-S.; Kim, Y.-B.; Schwartz, B. S. (2000) A comparison of different lead
30             biomarkers in their associations with lead-related symptoms. Int. Arch. Occup. Environ. Health
31             73:298-304.
32     Lee, B.-K.; Lee, G.-S.; Stewart, W. F.; Ahn, K.-D.; Simon, D.; Kelsey, K. T.; Todd, A. C.; Schwartz, B. S. (2001)
3 3             Associations of blood pressure and hypertension with lead dose measures and polymorphisms in the
34             vitamin D receptor and "delta"-aminolevulinic acid dehydratase genes. Environ. Health Perspect.
35             109:383-389.
36     Leggett, R. W. (1993) An age-specific kinetic model of lead metabolism in humans. Environ. Health Perspect.
37             101:598-616.
3 8     Lerchl, A. (1995) Evidence for decreasing quality of sperm. Presentation of data on sperm concentration was
3 9             flawed. Br. Med. J. 311: 569-570.
40     Lerda, D. (1992) Study of sperm characteristics in persons occupationally exposed to  lead. Am. J. Ind. Med. 22:
41             567-571.
42     Levey, A. S.; Bosc, J. P.; Lewis, J. B.; Greene, T.; Rogers, N.; Roth, D. (1999) A more accurate method to estimate
43             glomerular filtration rate from serum creatinine: a new prediction equation. Ann. Intern. Med.
44             130:461-470.
45     Levey, A. S.; Coresh, J.; Balk, E.; Kausz, A. T.; Levin, A.; Steffes, M. W.; Hogg, R. J.; Perrone, R. D.; Lau, J.;
46             Eknoyan, G. (2003) National kidney foundation practice guidelines for chronic kidney disease: evaluation,
47             classification, and stratification. Ann. Intern. Med. 139:  137-147.
48     Leviton, A.; Bellinger, D.; Allred, E. N.; Rabinowitz, M.; Needleman, H.; Schoenbaum,  S. (1993) Pre- and postnatal
49             low-level lead exposure and children's dysfunction in school. Environ.  Res. 60:  30-43.
50     Lezak, M. (1995) Neuropsychological assessment. New York, NY: Oxford University Press.
51     Lidsky, T. I.; Schneider, J. S. (2003) Lead neurotoxicity in children: basic mechanisms and clinical correlates.
52             Brain 126: 5-19.
53     Liebelt, E. L.; Schonfeld, D. J.; Gallagher, P. (1999) Elevated blood lead levels  in children are associated with lower
54             erythropoietin concentrations. J. Pediatr. 134:  107-109.
       December 2005                                6-3 81         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Lilienthal, H.; Winneke, G.; Ewert, T. (1990) Effects of lead on neurophysiological and performance measures:
  2             animal and human data. In: Conference on advances in lead research: implications for environmental
  3             health; January 1989; Research Triangle Park, NC. Environ. Health Perspect. 89: 21-25.
  4     Lilis, R.; Gavrilescu, N.; Nestorescu, B.; Dumitriu, C.; Roventa, A. (1968) Nephropathy in chronic lead poisoning.
  5             Br. J. Ind. Med. 25:  196-202.
  6     Lilis, R.; Eisinger, J.; Blumberg, W.; Fischbein, A.; Selikoff, I. J. (1978) Hemoglobin, serum iron, and zinc
  7             protoporphyrin in lead-exposed workers. Environ. Health Perspect. 25: 97-102.
  8     Lim, J.; Kang, E.; Choi, B.; Seo, J.; Hong, Y.; Chang, I.; Park, J. (2000) The distribution of blood lead concentration
  9             by age and sex in growing students. Chungang Uidaechi 25: 87-91.
10     Lim, Y. C.; Chia, K. S.; Ong, H. Y.;Ng, V.; Chew, Y. L. (2001) Renal dysfunction in workers exposed to inorganic
11             lead. Ann. Acad. Med. Singapore 30: 112-117.
12     Lin, J. L.; Huang, P. T. (1994) Body lead stores and urate excretion in men with chronic renal disease. J. Rheumatol.
13             21:705-709.
14     Lin, J.-L.; Lim, P.-S. (1992) Elevated lead burden in Chinese patients without occupational lead exposure. Miner.
15             Electrolyte Metab .18:1-5.
16     Lin, J.-L.; Lim, P.-S. (1994) Does lead play a role in the development of renal insufficiency in some patients with
17             essential hypertension? J. Hum. Hypertens. 8: 495-500.
18     Lin, J.-L.; Yeh, K.-H.; Tseng, H.-C;  Chen, W- Y.; Lai, H.-H.; Lin, Y.-C; Green Cross Health Service Association
19             Study Group. (1993) Urinary N-acetyl-glucosaminidase excretion and environmental lead exposure. Am. J.
20             Nephrol. 13:442-447.
21     Lin, S.; Hwang, S. A.; Marshall, E. G.; Stone,  R.; Chen, J. (1996) Fertility rates among lead workers and
22             professional bus drivers: a comparative study. Ann. Epidemiol. 6: 201-208.
23     Lin, S.; Hwang, S.-A.; Marshall, E. G.; Marion, D. (1998) Does paternal occupational lead exposure increase the
24             risks of low birth weight or prematurity? Am. J. Epidemiol.  148: 173-181.
25     Lin, J.-L.; Ho, H.-H.; Yu, C.-C. (1999) Chelation therapy for patients with elevated body lead burden and
26             progressive renal insufficiency. A randomized, controlled trial. Ann. Intern. Med. 130: 7-13.
27     Lin, J.-L.; Tan, D.-T.; Hsu, K.-H.; Yu, C.-C. (2001) Environmental lead exposure and progressive renal
28             insufficiency. Arch. Intern. Med. 161: 264-271.
29     Lin, J.-L.; Yu, C.-C.; Lin-Tan, D.-T.; Ho, H.-H. (2001) Lead chelation therapy and urate excretion in patients with
30             chronic renal diseases and gout. Kidney Int. 60: 266-271.
31     Lin, J.-L.; Tan, D.-T.; Ho, H.-H.; Yu, C.-C. (2002) Environmental lead exposure and urate excretion in the general
32             population. Am. J. Med. 113: 563-568.
33     Lin, J.-L.; Lin-Tan, D.-T.; Hsu, K.-H.; Yu, C.-C. (2003) Environmental lead exposure and progression of chronic
34             renal diseases in patients without diabetes. N. Engl. J. Med.  348: 277-286.
35     Lin, C.; Kim, R.; Tsaih, S.-W.; Sparrow, D.; Hu, H. (2004) Determinants of bone and blood lead levels among
36             minorities living in the Boston area. Environ. Health Perspect. 112: 1147-1151.
37     Lindbohm, M.-L.; Hemminki, K.; Bonhomme, M. G.; Anttila, A.; Rantala, K.; Heikkila, P.; Rosenberg, M. J. (1991)
38             Effects of paternal occupational exposure on spontaneous abortions. Am. J. Public Health 81: 1029-1033.
39     Lindeman, R. D.; Tobin, J.; Shock, N. W. (1985) Longitudinal studies on the rate of decline in renal function with
40             age. J. Am. Geriatr.  Soc. 33: 278-285.
41     Lindgren, K.; Masten, V.; Ford, D.; Bleecker,  M. (1996) Relation of  cumulative exposure to inorganic lead and
42             neuropsychological  test performance. Occup. Environ. Med. 53: 472-477.
43     Lindgren, K. N.; Masten, V. L.; Tiburzi, M. J.; Ford, D. P.; Bleecker, M. L. (1999) The factor structure of the profile
44             of mood states (POMS) and its relationship to occupational  lead exposure. J. Occup. Environ. Med.
45             41:3-10.
46     Liu, X.; Dietrich, K. N.; Radcliffe, J.; Ragan, N. B.; Rhoads, G. G.; Rogan, W. J. (2002) Do children with falling
47             blood lead levels have improved cognition? Pediatrics 110:  787-791.
48     Lochen, M. L.; Rasmussen, K.; Macfarlane, P. W.; Arnesen, E. (1996) Can single lead computerized
49             electrocardiography predict myocardial infarction in young  and middle-aged men? The Tromso  Study. Eur.
50             Heart J. 17(abstr. suppl.): 432.
51     Lockett, C. J.; Arbuckle, D. (1987) Lead, ferritin, zinc, and hypertension. Bull. Environ. Contam. Toxicol.
52             38:975-980.
53     Loeber, R. (1991) Initiation, escalation and desistance in juvenile offending and their correlates. J. Criminal Law
54             Criminol. 82: 36-82.
55     Loghman-Adham, M. (1998) Aminoaciduria and glycosuria following severe childhood lead poisoning. Pediatr.
56             Nephrol. 12: 218-221.


       December 2005                                6-3 82         DRAFT-DO NOT QUOTE  OR CITE

-------
  1     Lopez, C. M.; Pineiro, A. E.; Nunez, N.; Avagnina, A. M.; Villaamil, E. C.; Roses, O. E. (2000) Thyroid hormone
  2             changes in males exposed to lead in the Buenos Aires area (Argentina). Pharmacol. Res. Commun.
  3             42(6): 599-602.
  4     Louis, E. D.; Applegate, L.; Graziano, J. H.; Parides, M.; Slavkovich, V.; Bhat, H. K. (2005) Interaction between
  5             blood lead concentration and delta-amino-levulinic acid dehydratase gene polymorphisms increases the
  6             odds of essential tremor. Mov. Disord. 20: 1170-1177.
  7     Lucchini, R.; Albini, E.; Cortesi, I.; Placidi, D.; Bergamaschi, E.; Traversa, F.; Alessio, L. (2000) Assessment of
  8             neurobehavioral performance as a function of current and cumulative occupational lead exposure.
  9             Neurotoxicology 21: 805-811.
10     Lundstrom, N. G.; Nordberg, G.; Englyst, V.; Gerhardsson, L.; Hagmar, L.; Jin, T.; Rylander, L.; Wall, S. (1997)
11             Cumulative lead exposure in relation to mortality and lung cancer morbidity in a cohort of primary smelter
12             workers. Scand. J. Work Environ. Health 23: 24-30.
13     Lustberg, M. (2003) I know lead exposure is dangerous for children, but what about adults? Health News
14             9(March): 12.
15     Lustberg, M.; Silbergeld, E. (2002) Blood lead levels and mortality. Arch. Intern. Med. 162: 2443-2449.
16     Lustberg, M. E.; Schwartz, B. S.; Lee, B. K.; Todd, A. C.; Silbergeld, E. K. (2004) The g(894)-t(894)polymorphism
17             in the gene for endothelial nitric oxide synthase and blood pressure in lead-exposed workers from Korea.
18             J. Occup. Environ. Med. 46: 584-590.
19     Lutz, P. M.; Wilson, T. J.; Ireland, A. L.; Gorman, J. S.; Gale, N. L.; Johnson, J. C.; Hewett, J. E. (1999) Elevated
20             immunoglobulin E (IgE) levels in children with exposure to environmental lead. Toxicology 134: 63-78.
21     Lyngbye, T. (1997) Methodological problems in assessing health-related, neuropsychological effects of lead
22             absorption in a very low-level exposed area. Cent. Eur.  J. Public Health 5: 70-74.
23     Lyngbye, T.; Hansen, O. N.;  Grandjean, P. (1991) Lead concentration in deciduous teeth from Danish school
24             children. Dan. Med. Bull. 38: 89-93.
25     Mahaffey, K. R. (1992) Exposure to lead in childhood. N. Engl.  J. Med. 327: 1308-1309.
26     Mahaffey, K. R.; Annest, J. L.; Roberts,  J.; Murphy, R. S. (1982) National estimates of blood  lead levels: United
27             States, 1976-1980. Association with selected demographic and socioeconomic factors. N. Engl. J. Med.
28             307:573-579.
29     Maheswaran, R.; Gill, J. S.; Beevers, D.  G. (1993) Blood pressure and industrial lead exposure. Am. J. Epidemiol.
30             137:645-653.
31     Maizlish, N. A.; Parra, G.; Feo, O. (1995) Neurobehavioural evaluation of Venezuelan workers exposed to inorganic
32             lead. Occup. Environ. Med. 52: 408-414.
33     Makino, S.; Shimizu, Y.; Takata, T. (1997) A study on the relationship between blood lead levels and anemia
34             indicators in workers exposed to low levels of lead.  Ind. Health 35: 537-541.
35     Maki-Paakkanen, J.; Sorsa, M.; Vainio, H. (1981) Chromosome  aberrations and sister chromatid exchanges in lead-
36             exposed workers. Hereditas 94: 269-275.
37     Malcolm, D.; Barnett, H. A. (1982) A mortality study of lead workers 1925-76. Br. J. Ind. Med. 39: 404-410.
38     Malczyk, E.; Darewicz, B.; Pawlak, D.; Darewicz, J.; Buczko, W. (1999) Investigations of urinary lead
3 9             concentration in patients with urinary bladder carcinoma.  Int. Urol. Nephrol. 31: 661 -663.
40     Mallin et al. (1989) [240,table 7.3]
41     Mamdani, M.; Sykora, K.; Li, P.; Normand, S.-L. T.; Streiner, D. L.; Austin, P. C.; Rochon, P. A.; Anderson, G. M.
42             (2005) Reader's guide to critical appraisal of cohort studies: 2. Assessing potential for confounding. Br.
43             Med. J. BMJ 330: 960-962.
44     Manea-Krichten, M.; Patterson, C.; Miller, G.; Settle, D.; Erel, Y. (1991) Comparative increases of lead and barium
45             with age in human tooth enamel, rib and ulna. Sci. Total Environ. 107:  179-203.
46     Mantere, P.; Hanninen, H.; Hernberg, S. (1982) Subclinical neurotoxic lead effects: two-year follow-up studies with
47             psychological test methods. Neurobehav. Toxicol. Teratol. 4: 725-727.
48     Manton, W. I. (1985) Total contribution of airborne lead to blood lead. Br. J. Ind. Med. 42: 168-172.
49     Manton, W. I.; Cook, J. D. (1984) High accuracy (stable isotope dilution) measurements of lead in serum and
50             cerebrospinal fluid. Br. J. Ind. Med. 41: 313-319.
51     Manton, W. I.; Malloy, C. R. (1983) Distribution of lead in body fluids after ingestion of soft  solder. Br. J. Ind. Med.
52             40:51-57.
53     Manton, W. I.; Angle, C. R.; Stanek, K. L.; Reese, Y. R.; Kuehnemann, T. J. (2000) Acquisition and retention of
54             lead by young children. Environ. Res.  82: 60-80.
55     Manton, W. I.; Rothenberg, S. J.; Manalo, M. (2001) The lead content of blood serum. Environ. Res. 86: 263-273.
       December 2005                                6-3 83         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Manton, W. I.; Angle, C. R.; Stanek, K. L.; Kuntzelman, D.; Reese, Y. R.; Kuehnemann, T. J. (2003) Release of
  2            lead from bone in pregnancy and lactation. Environ. Res. 92: 139-151.
  3     Markowitz, M. E.; Rosen, J. F. (1981) Zinc (Zn) and copper (Cu) metabolism in CaNa2 EDTA-treated children with
  4            plumbism. Pediatr. Res. 15: 635.
  5     Markowitz, M. E.; Shen, X.-M. (2001) Assessment of bone lead during pregnancy: a pilot study. Environ. Res.
  6            85:83-89.
  7     Markowitz, M. E.; Weinberger, H. L. (1990) Immobilization-related lead toxicity in previously lead-poisoned
  8            children. Pediatrics 86: 455-457.
  9     Markowitz, M. E.; Gundberg, C. M.; Rosen, J. F. (1988) Sequential osteocalcin campling as a biochemical marker
10            of the success of treatment of moderately lead poisoned children [abstract]. Pediatr. Res. 23: 393A.
11     Markowitz, M. E.; Shen, X. M.; Balbi, K.; Rosen, J. F. (1996) Growing up in an era of increased lead (Pb) exposure:
12            is there a risk to the next generation? Pediatr. Res. 39(4 Part 2):  109A.
13     Marti, J. B.; Cabo, M.; Marhuenda, D.; Cardona, A. (1999) Accumulation of lead and chromium in the teeth of the
14            population in the southeast of the province Alicante, Spain. Trace Elem. Electrol. 16: 137-141.
15     Martin, C. J.; Werntz, C. L., Ill; Ducatman, A. M. (2004) The interpretation of zinc protoporphyrin changes in lead
16            intoxication: a case report and review of the literature.  Occup. Med. (London) 54: 587-591.
17     Mason, H. J.; Somervaille, L. J.; Wright, A. L.; Chettle, D. R.; Scott, M. C. (1990) Effect of occupational lead
18            exposure on serum 1,25-dihydroxyvitamin D levels. Hum. Exp. Toxicol. 9: 29-34.
19     Mason, P. J.; Manson, J. E.; Sesso, H. D.; Albert, C. M.; Chown, M. J.; Cook,  N. R.; Greenland, P.; Ridker, P. M.;
20            Glynn, R. J. (2004) Blood pressure and risk of secondary cardiovascular events in women: the Women's
21            Cardiovascular Study (WACS). Circulation 109: 1623-1629.
22     Matte, T.; Binder, S. (1993) Costs and benefits of lead screening [reply]. JAMA J. Am. Med. Assoc.
23            270:2054-2055.
24     Matte, T. D.; Figueroa, J. P.; Burr, G.; Flesch, J. P.; Keenlyside, R. A.; Baker,  E. L. (1989) Lead exposure among
25            lead-acid battery workers in Jamaica. Am. J. Ind. Med. 16: 167-177.
26     McBride, W. G.; Carter, C. J.; Bratel, J. R.; Cooney, G.; Bell, A. (1989) The Sydney study of health effects of lead
27            in urban children. In: Smith, M. A.; Grant, L. D.; Sors, A. I., eds. Lead exposure and child development:  an
28            international assessment  [workshop organized by the Commission of the European Communities and the
29            U.S. Environmental Protection Agency]; September 1986; Edinburgh, United Kingdom. Dordrecht,
30            The Netherlands: Kluwer Academic Publishers BV; pp. 255-259.
31     McCabe, M. J.; Lawrence, D. A. (1991) Lead, a major environmental pollutant, is immunomodulatory by its
32            differential effects on CD4+ T cell subsets. Toxicol. Appl. Pharmacol. Ill: 13-23.
33     McDonald, J. (1981) The lead contamination problem with emphasis on the lead content of wine. Am. J. Enol. Vitic.
34            32:219-222.
35     McDonald, J. A.; Potter, N. U. (1996) Lead's legacy? Early and late mortality of 454 lead-poisoned children. Arch.
36            Environ. Health. 51: 116-121.
37     McGregor, A. J.; Mason, H. J. (1990) Chronic occupational lead exposure and testicular endocrine function. Hum.
38            Exp. Toxicol. 9: 371-376.
39     McMichael, A. J. (1993) [Untitled letter concerning a critique paper of five main cohort studies of environmental
40            lead and child mental development]. Arch. Environ. Health 48: 125-126.
41     McMichael, A. J. (1993) (Response to Thacker et al. article on lead and child mental development [letter]). Arch.
42            Environ. Health 48: 125-126.
43     McMichael, A. J. (1995) Environmental lead and intellectual development: strengths and limitations of
44            epidemiological research. Neurotoxicol. Teratol. 17: 237-240.
45     McMichael, A. J. (1997) Lead exposure and child intelligence: interpreting or  misinterpreting, the direction of
46            causality? J. Paediatr. Child Health 33: 7-8.
47     McMichael, A. J.; Johnson, H. M. (1982) Long-term mortality profile of heavily-exposed lead smelter workers.
48            J.  Occup. Med. 24: 375-378.
49     McMichael, A. J.; Vimpani, G. V.; Robertson, E. F.; Baghurst, P. A.; Clark, P. D. (1986) The Port Pirie cohort
50            study: maternal blood lead and pregnancy outcome. J. Epidemiol. Commun. Health 40:  18-25.
51     McMichael, A. J.; Baghurst, P. A.; Wigg, N. R.; Vimpani, G. V.; Robertson, E. F.; Roberts, R. J. (1988) Port Pirie
52            cohort study: environmental exposure to lead and children's abilities at the age of fouryears. N. Engl. J.
53            Med. 319:468-475.
54     McMichael, A. J.; Baghurst, P. A.; Vimpani, G. V; Robertson, E. F.; Wigg, N. R.; Tong, S.-L. (1992)
5 5            Sociodemographic factors modifying the effect of environmental lead on neuropsychological development
56            in early childhood. Neurotoxicol. Teratol. 14: 321-327.
       December 2005                                6-3 84        DRAFT-DO NOT QUOTE OR CITE

-------
  1     McMichael, A. I; Baghurst, P. A.; Vimpani, G. V.; Wigg, N. R.; Robertson, E. F.; Tong, S. (1994) Tooth lead
  2             levels and IQ in school-age children: the Port Pirie cohort study. Am. J. Epidemiol.  140: 489-499.
  3     McNeill, F. E.; Stokes, L.; Brito, J. A.; Chettle, D. R.; Kaye, W. E. (2000) 109Cd K x-ray fluorescence
  4             measurements of tibial lead content in young adults exposed to lead in early childhood. Occup. Environ.
  5             Med. 57: 465-471.
  6     Menditto, A.; Morisi, G.; Spagnolo, A.; Menotti, A.; NFR Study Group. (1994) Association of blood lead to blood
  7             pressure in men aged 55 to 75 years: effect of selected social and biochemical confounders. NFR study
  8             group. Environ. Health Perspect. 102(suppl. 9): 107-111.
  9     Meng, X.-M; Zhu, D.-M; Ruan, D.-Y.; She, J.-Q.; Luo, L. (2005) Effects of chronic lead exposure on H MRS of
10             hippocampus and frontal lobes in children. Neurology 64: 1644-1647.
11     Meredith, P. A.; Campbell, B.C.; Moore, M. R.; Goldberg, A. (1977) The effects of industrial lead poisoning on
12             cytochrome P450 mediated phenazone (antipyrine) hydroxylation. Eur. J. Clin. Pharmacol. 12: 235-239.
13     Merlo, J.; Asplund, K.; Lynch, J.; Rastam, L.; Dobson, A.; World Health Organization MONICA Project. (2004)
14             Population effects on individual systolic blood pressure: a multilevel analysis of the World Health
15             Organization MONICA Project. Am. J. Epidemiol. 159: 1168-1179.
16     Michaels, D.; Zoloth, S. R.; Stern, F. B. (1991) Does low-level lead exposure increase risk of death? A mortality
17             study of newspaper printers. Int. J. Epidemiol. 20: 978-983.
18     Milanov, L; Kolev, P.  (2001) Clinical  and electromyographic examinations of patients with tremor after chronic
19             occupational lead exposure. Occup. Med. (London) 51: 157-162.
20     Millstone, E.; Russell, J. (1995) Environmental lead and children's intelligence: Britain must replace its lead pipes to
21             meet WHO standards for drinking water [letter]. Br. Med. J. 310: 1408-1409.
22     Milne, R.; Gamble, G.; Whitlock, G.; Jackson, R. (2003) Framingham Heart Study risk equation predicts first
23             cardiovascular event rates in New Zealanders at the population level. N. Z. Med. J. 116(1185): U662.
24     Min, Y.-L; Correa-Villasenor, A.; Stewart, P. A. (1996) Parental occupational lead exposure and low birth weight.
25             Am. J. Ind. Med. 30: 569-578.
26     Minozzo, R.; Deimling, L. I.; Gigante, L. P.; Santos-Mello, R. (2004) Micronuclei in peripheral blood lymphocytes
27             of workers exposed to lead. Mutat. Res. 565: 53-60.
28     Miranda-Cams, E.; Mateos, F. A.; Sanz, A.  G.; Herrero, E.; Ramos, T.; Puig, J. G. (1997) Purine metabolism in
29             patients with gout: the role of lead. Nephron 75:  327-335.
30     Mirsky, A. F. (1987) Behavioral and psychophysiological makers of disordered attention. Environ. Health Perspect.
31             74:191-199.
32     Mishra, K. P.; Singh, V.  K.; Rani, R.; Yadav, V. S.; Chandran, V.; Srivastava, S. P.; Seth, P.  K.  (2003) Effect of
33             lead exposure on the immune response of some occupationally exposed individuals. Toxicology
34             188:251-259.
35     Moel, D. I.; Sachs, H. K. (1992) Renal function 17 to 23 years after chelation therapy for childhood plumbism.
36             Kidney Int. 42:  1226-1231.
37     Mohammed-Brahim, B.; Buchet, J. P.; Lauwerys, R. (1985) Erythrocyte pyrimidine 5'-nucleotidase activity in
38             workers exposed to lead, mercury or cadmium. Int. Arch.  Occup. Environ. Health 55: 247-252.
39     Moline, J.; Carrillo, L. L.; Sanchez, L. T.; Godbold, J.; Todd, A. (2000) Lactation and lead body burden turnover: a
40             pilot study  in Mexico. J. Occup. Environ. Med. 42: 1070-1075.
41     M011er, L.; Kristensen, T. S. (1992) Blood lead as a cardiovascular risk factor. Am. J. Epidemiol. 136:  1091-1100.
42     Montgomery, L. E.; Carter-Pokras, O. (1993) Health status by social class and/or minority status: implications for
43             environmental equity research. Toxicol. Ind. Health 9: 729-773.
44     Moore, M. R.; Goldberg, A.;  Bushnell, I. W. R.; Day, R.; Fyfe, W. M. (1982) A prospective study of the
45             neurological effects  of lead in children. Neurobehav. Toxicol.  Teratol. 4: 739-743.
46     Moore, M. R.; Mclntosh, M.  J.; Bushnell, I. W. R. (1986) The neurotoxicology of lead. Neurotoxicology 7: 541-556.
47     Morgan, J. M. (1975) Chelation therapy in lead nephropathy. South. Med. J. 68: 1001-1006.
48     Morgan, W. D.;  Ryde, S. J.; Jones, S. J.; Wyatt, R. M.; Hainsworth, I. R.; Cobbold, S. S.; Evans, C. J.; Braithwaite,
49             R. A. (1990) In vivo measurements of cadmium and lead in occupationally-exposed workers and an urban
50             population. Biol. Trace Elem. Res. 26-27: 407-414.
51     Morita, Y.; Sakai, T.; Araki, S.; Araki, T.; Masuyama, Y. (1997) Nicotinamide adenine dinucleotide synthetase
52             activity in erythrocytes as a tool for the biological monitoring of lead exposure. Int.  Arch. Occup. Environ.
53             Health 70:  195-198.
54     Morris, C.; McCarron, D. A.; Bennett, W. M. (1990) Low-level lead exposure, blood pressure, and calcium
55             metabolism. Am. J. Kidney Dis. 15: 568-574.
       December 2005                                6-385         DRAFT-DO NOT QUOTE OR CITE

-------
  1    Morrow-Tlucak, M; Ernhart, C. B. (1987) The relationship of low level lead exposure and language development in
  2            the pre-school years. In: Lindberg, S. E.; Hutchinson, T. C., eds. International conference: heavy metals in
  3            the environment, v. 1; September; New Orleans, LA. Edinburgh, United Kingdom: CEP Consultants, Ltd.;
  4            pp. 57-59.
  5    Mortada, W. I.; Sobh, M. A.; El-Defrawy, M. M.; Farahat, S. E. (2001) Study of lead exposure from automobile
  6            exhaust as a risk for nephrotoxicity among traffic policemen. Am. J. Nephrol. 21: 274-279.
  7    Mortada, W. I.; Sobh, M. A.; El-Defrawy, M. M. (2004) The exposure to cadmium, lead and mercury from smoking
  8            and its impact on renal integrity. Med. Sci. Monit. 10: CR112-CR116.
  9    Moss, M. E.; Lanphear, B. P.; Auinger, P. (1999) Association of dental caries and blood lead levels. JAMA J. Am.
10            Med.  Assoc. 281: 2294-2298.
11    Moura, M.; Valente, J. G. (2002) Blood lead levels during pregnancy in women living in Rio de Janeiro, Brazil. Sci.
12            Total Environ. 299: 123-129.
13    Muldoon, S. B.; Cauley, J. A.; Kuller, L. H. ; Morrow, L.; Needleman, H. L. ; Scott, J.; Hooper, F. J. (1996) Effects
14            of blood lead levels on cognitive function of older women. Neuroepidemiology 15: 62-72.
15    Muldoon, S. B.; Cauley, J. A.; Garzarella, L.; Salamone, L.; Bradshaw, P. (2000) Blood lead levels and blood
16            pressure in perimenopausal women.  Am. J. Epidemiol. 151(11 suppl.): S79.
17    Muntner, P.; He, J.; Hamm, L.; Loria, C.; Whelton, P. K. (2002) Renal insufficiency and subsequent death resulting
18            from cardiovascular disease in the United States. J.  Am. Soc. Nephrol.  13: 745-753.
19    Muntner, P.; He, J.; Vupputuri, S.; Coresh, J.; Batuman, V. (2003) Blood lead and chronic kidney disease in the
20            general United States population: results from NHANES III. Kidney Int. 63: 1044-1050.
21    Murphy, M. J.; Graziano, J. H.; Popovac, D.; Kline, J. K.; Mehmeti, A.; Factor-Litvak, P.; Ahmedi, G.; Shrout, P.;
22            Rajovic, B.; Nenezic, D.  U.; Stein, Z. A. (1990) Past pregnancy outcomes among women living in the
23            vicinity of a lead smelter in Kosovo, Yugoslavia. Am. J. Public Health 80: 33-35.
24    Mushak, P. (1991) Gastro-intestinal absorption of lead in children and adults: overview of biological and
25            biophysico-chemical aspects. Chem. Speciation Bioavailability 3(3/4): 87-104.
26    Mushak, P. (1993) New directions in the toxicokinetics of human lead exposure. Presented at: Ninth international
27            neurotoxicology conference; October 1991; Little Rock, AR. Neurotoxicology 14(2-3): 29-42.
28    Nash, D.; Silbergeld, E.; Magder,  L.; Stolley, P. (1998) Menopause, hormone replacement therapy (HRT), and blood
29            lead levels among adult women from NHANES III, 1988-1994. Am. J. Epidemiol. 147(suppl. 11): S93.
30    Nash, D.; Magder, L.; Lustberg, M.; Sherwin, R. W.; Rubin, R.  J.; Kaufmann, R. B.; Silbergeld, E. K. (2003) Blood
31            lead, blood pressure, and hypertension in perimenopausal and postmenopausal women. JAMA J. Am. Med.
32            Assoc. 289: 1523-1532.
33    Nash, D.; Magder, L. S.; Sherwin, R.; Rubin, R. J.; Silbergeld, E. K. (2004) Bone density-related predictors of blood
34            lead level among peri- and postmenopausal women in the United States: the Third National Health and
35            Nutrition Examination Survey, 1988-1994. Am. J. Epidemiol. 160: 901-911.
36    National Institute for Occupational Safety and Health. (1977a) Manual of analytical methods. 2nd ed. Cincinnati,
37            OH: U.S. Department of Health, Education, and Welfare, Public Health Service, Centers for Disease
3 8            Control, National Institute for Occupational Safety  and Health. DHEW (NIOSH) publication no. 77/157-A.
39            Method No. P&CAM 102. V. 1.
40    National Institute for Occupational Safety and Health. (1977b) Manual of analytical methods. 2nd ed. Cincinnati,
41            OH: U.S. Department of Health, Education, and Welfare, Public Health Service, Centers for Disease
42            Control, National Institute for Occupational Safety  and Health. DHEW (NIOSH) publication no. 77/157-A.
43            Method No. P&CW 195. V. 1.
44    National Institute for Occupational Safety and Health. (1977c) Manual of analytical methods. 2nd ed. Cincinnati,
45            OH: U.S. Department of Health, Education, and Welfare, Public Health Service, Centers for Disease
46            Control, National Institute for Occupational Safety  and Health, 200-1 to 200- ?. Method No. P&CAM 200.
47            Vol. 1.
48    National Institute for Occupational Safety and Health. (1977d) Manual of analytical methods. 2nd ed. Cincinnati,
49            OH: U.S. Department of Health, Education, and Welfare, Public Health Service, Centers for Disease
50            Control. National Institute for Occupational Safety  and Health. 214-1 to 214-8. Method No. P&CAM 214.
51            Vol.1.
52    National Institute for Occupational Safety and Health. (1977e) Manual of analytical methods. 2nd ed. Cincinnati,
53            OH: U.S. Department of Health, Education and Welfare. Public Health Service, Centers for Disease
54            Control, National Institute for Occupational Safety  and Health. Method No. P&CAM 262. Vol.  1.
      December 2005                                6-3 86         DRAFT-DO NOT QUOTE OR CITE

-------
  1     National Institute for Occupational Safety and Health. (1977f) Manual of analytical methods. 2nd ed. Cincinnati,
  2             OH: U.S. Department of Health, Education, and Welfare, Public Health Service, Centers for Disease
  3             Control, National Institute for Occupational Safety and Health. Method No. P&CAM 208. Vol. 1.
  4     National Institute for Occupational Safety and Health. (1984) Manual of analytical methods. 3rd ed. Cincinnati, OH:
  5             U.S. Department of Health and Human Services, Centers for Disease Control, National Institute for
  6             Occupational Safety and Health. Method No. 7300, 8003, and 8310. Vol. 1.
  7     National Institute for Occupational Safety and Health. (1994) Manual of analytical methods. 4rd ed. Cincinnati, OH:
  8             U.S. Department of Health and Human Services, Centers for Disease Control, National Institute for
  9             Occupational Safety and Health; DHHS (NIOSH) publication 94-113; method no. 7105.
10     National Institutes of Health. (1994) Consensus development panel on optimal calcium uptake. JAMA J. Am. Med.
11             Assoc. 272: 1942-1948.
12     National Toxicology Program. (2004) Lead (CAS no. 7439-92-1) and lead compounds. In: Report on carcinogens,
13             eleventh edition. Research Triangle Park, NC: U.S. Department of Health and Human Services. Available:
14             http://ntp.niehs.nih.gov/ntp/roc/eleventh/profiles/sl011ead.pdf [28 November, 2005].
15     Navarro, J. A.; Granadillo, V. A.; Salgado, O.; Rodriguez-Iturbe, B.; Garcia, R.; Delling, G.; Romero, R. A. (1992)
16             Bone metal content in patients with chronic renal failure.  Clin.  Chim. Acta211: 133-142.
17     Navas-Acien, A.; Selvin, E.; Sharrett, A. R.; Calderon-Aranda, E.; Silbergeld, E.; Guallar, E. (2004) Lead,
18             cadmium, smoking, and increased risk of peripheral arterial disease. Circulation 109: 3196-3201.
19     Nawrot, T. S.; Thijs, L.; DenHond, E. M.; Roels, H. A.; Staessen, J. A.  (2002) An epidemiological re-appraisal of
20             the association between blood pressure and blood lead: a  meta-analysis. J. Hum. Hypertens. 16: 123-131.
21     Neaton, J. D.; Kuller, L.; Stamler, J.; Wentworth, D. N. (1995) Impact of systolic and diastolic blood pressure on
22             cardiovascular mortality. In: Laragh, J.  H.; Brenner, B. M., eds. Hypertension: pathophysiology, diagnosis,
23             and management. 2nd ed. New York, NY: Raven Press Ltd.; pp. 127-144.
24     Needleman, H. L. (1983) Lead at low dose and the behavior of children. Neurotoxicology 4: 121-133.
25     Needleman, H. L. (1983) Low level lead exposure and neuropsychological performance. In: Rutter, M.; Russell
26             Jones, R., eds. Lead versus health. New York, NY: John Wiley & Sons, Ltd.; pp. 229-248.
27     Needleman, H. L. (1995) Environmental lead and children's intelligence: studies included in the meta-analysis are
28             not representative [letter]. Br. Med. J. 310: 1408.
29     Needleman, H. L. (1995) Making models of real world events: the use and abuse of inference. Neurotoxicol. Teratol.
30             17:241-242.
31     Needleman, H. L. (2004) Low level lead exposure and the development of children. Southeast Asian J. Trop. Med.
32             Public Health 35: 252-254.
3 3     Needleman, H. L.; Bellinger, D. (1988) Recent developments. Environ. Res. 46: 190-191.
34     Needleman, H. L.; Gatsonis, C. A. (1990) Low-level lead exposure and the IQ of children: a meta-analysis of
35             modern studies. JAMA J. Am. Med. Assoc. 263: 673-678.
36     Needleman, H. L.; Gunnoe, C.; Leviton, A.; Reed, R.; Peresie, H.; Maher, C.; Barrett, P. (1979) Deficits in
3 7             psychologic and classroom performance of children with elevated dentine lead levels. N. Engl. J.  Med.
38             300:689-695.
39     Needleman, H. L.; Leviton, A.; Bellinger, D. (1982) Lead-associated intellectual deficit. N. Engl. J. Med. 306: 367.
40     Needleman, H. L.; Rabinowitz, M.; Leviton, A.; Linn, S.; Schoenbaum, S. (1984) The relationship between prenatal
41             exposure to lead and congenital anomalies. JAMA J. Am. Med. Assoc.  251: 2956-2959.
42     Needleman, H. L.; Schell, A.; Bellinger, D.; Leviton, A.; Allred, E. N. (1990) The long-term effects of exposure to
43             low doses of lead in childhood; an 11-year follow-up report. N. Engl. J. Med. 322: 83-88.
44     Needleman, H. L.; Riess, J. A.; Tobin, M. J.; Biesecker, G. E.;  Greenhouse, J. B. (1996) Bone lead levels and
45             delinquent behavior. JAMA J. Am. Med. Assoc. 275:  363-369.
46     Needleman, H. L.; McFarland, C.; Ness, R. B.; Fienberg, S. E.; Tobin, M. J. (2002) Bone lead levels in adjudcated
47             delinquents. A case control study. Neurotoxicol. Teratol.  24: 711-717.
48     Neisser, U.; Boodoo, G.; Bouchard, T. J.; Boykin, A. W.; Brody, N.; Ceci, S. J.; Halpern, D. F.; Loehlin, J. C.;
49             Perloff, R.; Sternberg, R. J.; Urbina, S.  (1996) Intelligence: knowns and unknowns. Am. Psychol.
50             51:77-101.
51     Nenov, V. D.; Taal, M. W.; Sakharova, O. V.; Brenner, B. M. (2000) Multi-hit nature of chronic renal disease. Curr.
52             Opin. Nephrol. Hypertens. 9: 85-97.
53     Neri, L. C.; Hewitt,  D.; Orser, B. (1988) Blood lead and blood pressure: analysis of cross-sectional and longitudinal
54             data from Canada. In: Victery, W., ed. Symposium on lead-blood pressure relationships; April 1987;
55             Chapel Hill, NC. Environ. Health Perspect. 78: 123-126.
       December 2005                                6-3 87         DRAFT-DO NOT QUOTE OR CITE

-------
  1    Nevin, R. (2000) How lead exposure relates to temporal changes in IQ, violent crime, and unwed pregnancy.
  2            Environ. Res. 83: 1-22.
  3    Nezhdanova, M. V. (1996) Frequency and structure of renal pathology in Saransk in dependence of degree of lead
  4            pollution. Pediatriya (Moscow) (2): 72-73.
  5    Ng, T. P.; Goh, H. H.; Ng, Y. L.; Ong, H. Y.; Ong, C. N.; Chia, K. S.; Chia, S. E.; Jeyaratnam, J. (1991) Male
  6            endocrine functions in workers with moderate exposure to lead. Br. J. Ind. Med. 48: 485-491.
  7    Nilsson, U.; Attewell, R.; Christoffersson, J.-O.; Schutz, A.; Ahlgren, L.; Skerfving, S.; Mattsson, S. (1991) Kinetics
  8            of lead in bone and blood after end of occupational exposure. Pharmacol. Toxicol. (Copenhagen)
  9            68:477-484.
10    Niu, Q.; He, S. C.; Li, H. Y.; Wang, J. Y.; Dai, F. Y.; Chen, Y. L. (2000) A comprehensive neurobehavioral and
11            neurophysiological study for low level lead-exposed workers. G. Ital. Med. Lav. Ergon. 22: 299-304.
12    Noda, H.; Sugiyama, S.; Yamaguchi, M.; Tatsumi, S.; Sano, Y.; Konishi, S.; Furutani, A.; Yoshimura, M. (1993)
13            Studies on secular changes in the concentration of lead accumulated in organs and rib of Japanese. Jpn. J.
14            Leg. Med. 47: 147-152.
15    Nolte, J. (1993). The human brain: an introduction to its functional anatomy. Mosby Year Book Publishers:
16            St. Louis, MO.
17    Nomiyama, K.; Nomiyama, H.; Liu, S. J.; Tao, Y. X.; Nomiyama, T.; Omae, K. (2002) Lead induced increase of
18            blood pressure in female lead workers. Occup. Environ. Med. 59: 734-738.
19    Noonan, C. W.; Sarasua, S. M.; Campagna, D.; Kathman, S. J.; Lybarger, J. A.; Mueller, P. W. (2002) Effects of
20            exposure to low levels of environmental cadmium on renal biomarkers. Environ. Health Perspect.
21            110:151-155.
22    Nordberg, M.; Winblad, B.; Fratiglioni, L.; Basun, H. (2000) Lead concentrations in elderly urban people related to
23            blood pressure and mental performance: results from a population-based study. Am. J. Ind. Med.
24            38:290-294.
25    Nordenson, I.; Beckman, G.; Beckman,  L.; Nordstrom, S. (1978) Occupational and environmental risks in and
26            around a smelter in northern Sweden. IV. Chromosomal aberrations in workers exposed to lead. Hereditas
27            88:263-267.
28    Nordstrom, S.; Beckman, L.; Nordenson, I. (1978a) Occupational and environmental risks in and around a smelter in
29            northern Sweden: I. variations  in birth weight. Hereditas (Lund, Swed.) 88: 43-46.
30    Nordstrom, S.; Beckman, L.; Nordenson, I. (1978b) Occupational and environmental risks in and around a smelter
31            in northern Sweden: III. frequencies of spontaneous abortion. Hereditas  (Lund, Swed.) 88: 51-54.
32    Nordstrom, S.; Beckman, L.; Nordenson, I. (1979) Occupational and environmental risks in and around a smelter in
33            northern Sweden. V. Spontaneous abortion among female employees and decreased birth weight in their
3 4            offspring. Hereditas 90: 291 -296.
35    Nowack, R.; Wiecek, A.; Ritz, E. (1992) Lead and hypertension. In: Berlyne,  G. M., ed. The kidney today. Selected
36            topics in renal science. Basel, Switzerland: Karger; pp. 25-34. (Contributions to nephrology: v. 100).
37    Nuyts, G. D.; D'Haese, P. C.; Elseviers, M. M.; De Broe, M. E. (1989) Renal dysfunction and lead exposure [letter].
38            Am. J. Nephrol. 9: 85-86.
39    Nuyts, G. D.; Van Vlem, E.; Thys, J.; De Leersnijder, D.; D'Haese, P. C.; Elseviers, M. M.; De Broe, M.  E. (1995)
40            New occupational risk factors for chronic renal failure. Lancet  346: 7-11.
41    Nystrom-Rosander, C.; Lindh, U.; Friman, G.; Lindqvist, O.; Thelin, S.; Ilback, N. G. (2004) Trace element changes
42            in sclerotic heart valves from patients are expressed in their blood. Biometals 17: 121-128.
43    O'Dowd, P. (2002) Controversies regarding low blood lead level harm. Med. Health 85: 345-348.
44    O'Flaherty, E. J. (1991) Physiologically based models for bone-seeking elements. III. Human skeletal and bone
45            growths. Toxicol. Appl. Pharmacol.  Ill: 332-341.
46    O'Flaherty, E. J. (1993) Physiologically based models for bone-seeking elements. IV. Kinetics of lead disposition in
47            humans. Toxicol. Appl. Pharmacol.  118:  16-29.
48    O'Flaherty, E. J. (1995) Physiologically based models for bone-seeking elements: V. Lead absorption and
49            disposition in childhood. Toxicol. Appl. Pharmacol.  131: 297-308.
50    O'Flaherty, E. J.; Hammond, P. B.; Lerner, S. I. (1982) Dependence of apparent blood lead half-life on the length of
51            previous lead exposure in humans. Fundam. Appl. Toxicol. 2: 49-54.
52    O'Flaherty, E. J.; Inskip, M. J.; Franklin, C. A.; Durbin, P. W.; Manton, W. I.; Baccanale, C. L. (1998) Evaluation
53            and modification of a physiologically based model of lead kinetics using data from a sequential isotope
54            study in cynomolgus monkeys. Toxicol. Appl. Pharmacol. 149: 1-16.
      December 2005                                6-388         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Oishi, H.; Nomiyama, H.; Nomiyama, K.; Tomokuni, K. (1996) Comparison between males and females with
  2             respect to the porphyrin metabolic disorders found in workers occupationally exposed to lead. Int. Arch.
  3             Occup. Environ. Health 68: 298-304.
  4     Oliveira, S.; Aro, A.; Sparrow, D.; Hu, H. (2002) Season modifies the relationship between bone and blood lead
  5             levels: the Normative Aging Study. Arch. Environ. Health 57: 466-472.
  6     Oliver, T. (1911) Lead poisoning and the race. Br. Med. J. 1(2628): 1096-1098.
  7     Olsen, G. W.; Bodner, K. M; Ramlow, J. M; Ross, C. E.; Lipshultz, L. I. (1995) Have sperm counts been reduced
  8             50 percent in 50 years? A statistical model revisited. Fertil. Steril. 63: 887-893.
  9     Olsson, I.-M; Bensryd, L; Lundh, T.; Ottosson, H.; Skerfving, S.; Oskarsson, A. (2002) Cadmium in blood and
10             urine-impact of sex, age, dietary intake, iron status, and former smoking-association of renal effects.
11             Environ. Health Perspect. 110: 1185-1190.
12     Omae, K.; Sakurai, H.; Higashi, T.;  Muto, T.; Ichikawa, M.; Sasaki, N. (1990) No adverse effects of lead on renal
13             function in lead-exposed workers. Ind. Health 28: 77-83.
14     Onalaja, A. O.; Claudio, L. (2000) Genetic susceptibility to lead poisoning. Environ. Health Perspect. Suppl.
15             108(1): 23-28.
16     Oneglia, C.; Apostoli, P.; Rusconi, C. (1998) Vasospastic angina in a patient with chronic lead intoxication: a
17             possible cause-effect relationship? Cardiovasc. Drugs Ther. 12: 71-73.
18     Ong, C. N.; Endo, G.; Chia, K.  S.; Phoon, W. O.; Ong, H. Y. (1987) Evaluation of renal function in workers with
19             low blood lead levels.  In: Foa, V.; Emmett, E. A.; Maroni, M.; Colombi, A., eds. Occupational and
20             environmental chemical hazards: cellular and biochemical indices for monitoring toxicity. New York, NY:
21             Halstead Press; pp. 327-333.
22     Opler, M. G. A.; Brown, A. S.;  Graziano, J.; Desai, M.; Zheng, W.; Schaefer, C.; Factor-Litvak, P.; Susser, E. S.
23             (2004) Prenatal lead exposure, "delta"-aminolevulinic acid, and schizophrenia. Environ. Health Perspect.
24             112:548-552.
25     Orban B. (1953) Oral Histology and Embryology. Mosby Year Book Publishers: St. Louis, MO.
26     O'Riordan, M. L.; Evans, H. J. (1974) Absence of significant chromosome damage in males occupationally exposed
27             to lead. Nature (London) 247: 50-53.
28     Orssaud, G.; Claude, J. R.; Moreau, T.; Lellouch, J.; Juguet, B.; Festy, B. (1985) Blood lead concentration and blood
29             pressure. Br. Med. J. 290: 244.
30     Osman, K.; Pawlas, K.; Schutz, A.;  Gazdzik, M.; Sokal, J. A.; Vahter, M. (1999) Lead exposure and hearing effects
31             in children in Katowice, Poland. Environ. Res. 80: 1-8.
32     Osterberg, K.; Borjesson, J.; Gerhardsson, L.; Schutz, A.; Skerfving, S. (1997) A neurobehavioural study of long-
33             term occupational inorganic lead exposure. Sci. Total Environ. 201: 39-51.
34     Osterloh, J. D.; Kelly, T. J. (1999) Study of the effect of lactational bone loss on blood lead concentrations in
35             humans. Environ. Health Perspect. 107: 187-194.
36     Osterloh, J. D.; Selby, J. V.; Bernard, B. P.; Becker, C. E.; Menke, D. J.; Tepper, E.; Ordonez, J. D.; Behrens, B.
37             (1989) Body burdens of lead in hypertensive nephropathy. Arch. Environ. Health 44: 304-310.
3 8     Osterode, W.; Barnas, D.; Geissler,  K. (1999) Dose dependent reduction of erythroid progenitor cells and
39             inappropriate erythropoietin response in exposure to lead: new aspects of anaemia induced by lead. Occup.
40             Environ. Med. 56: 106-109.
41     Otto, D. A.; Fox, D. A. (1993) Auditory and visual dysfunction following lead exposure. Presented at: Ninth
42             international neurotoxicology conference; October 1991; Little Rock, AR. Neurotoxicology 14(2-
43             3): 191-207.
44     Otto, D.; Robinson, G.; Baumann, S.; Schroeder, S.; Mushak, P.; Kleinbaum, D.; Boone, L. (1985) Five-year
45             follow-up study of children with low-to-moderate lead absorption: electrophysiological evaluation.
46             Environ. Res. 38:  168-186.
47     Paksy, K.; Gati, I.; Naray, M.; Rajczy, K. (2001) Lead accumulation in human ovarian follicular fluid, and in vitro
48             effect of lead on progesterone production by cultured human ovarian granulosa cells. J. Toxicol. Environ.
49             Health A 62: 359-366.
50     Palus, J.; Rydzynski, K.; Dziubaltowska, E.; Wyszynska, K.; Natarajan, A. T.; Nilsson, R. (2003) Genotoxic effects
51             of occupational exposure to lead and cadmium. Mutat. Res. 540: 19-28.
52     Parkinson, D. K.; Hodgson, M.  J.; Bromet, E. J.; Dew, M. A.; Connell, M. M. (1987) Occupational lead exposure
53             and blood pressure. Br. J. Ind. Med. 44: 744-748.
54     Paschal, D. C.; Burt, V.; Caudill, S. P.; Gunter, E. W.; Pirkle, J. L.; Sampson, E. J.; Miller, D. T.; Jackson, R. J.
55             (2000) Exposure of the U.S. population aged 6 years and older to cadmium: 1988-1994. Arch. Environ.
56             Contam.  Toxicol.  38: 377-383.
       December 2005                                6-3 89         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Paterson, L. I; Raab, G. M; Hunter, R.; Laxen, D. P. H.; Fulton, M.; Fell, G. S.; Halls, D. I; Sutcliffe, P. (1988)
  2             Factors influencing lead concentrations in shed deciduous teeth. Sci. Total Environ. 74: 219-233.
  3     Payton, M.; Hu, H.; Sparrow, D.; Weiss, S. T. (1994) Low-level lead exposure and renal function in the normative
  4             aging study. Am. J. Epidemiol. 140: 821-829.
  5     Payton, M.; Riggs, K. M.; Spiro, A., Ill; Weiss, S. T.; Hu, H. (1998) Relations of bone and blood lead to cognitive
  6             function: the VA Normative Aging Study. Neurotoxicol. Teratol. 20: 19-27.
  7     Peplow, D.; Edmonds, R. (2004) Health risks associated with contamination of groundwater by abandoned mines
  8             near Twisp in Okanogan County, Washington, USA. Environ. Geochem.  Health 26: 69-79.
  9     Perez-Bravo, F.; Ruz, M; Moran-Jimenez, M. J.; Olivares, M.; Rebolledo, A.; Codoceo, J.; Sepulveda, J.; Jenkin, A.;
10             Santos, J. L.; Fontanellas, A.  (2004) Association between aminolevulinate dehydrase genotypes and blood
11             lead levels in children from a lead-contaminated area in Antofagasta, Chile. Arch. Environ. Contam.
12             Toxicol. 47(2): 276-280.
13     Pergande, M.; Jung, K.; Precht, S.; Pels, L. M.; Herbert, C.; Stolte, H. (1994) Changed excretion of urinary proteins
14             and enzymes by chronic exposure to lead. Nephrol. Dial. Transplant. 9: 613-618.
15     Pesch, B.; Haerting, J.; Ranft, U.; Klimpel, A.;  Oelschlagel, B.; Schill, W.; MURC Study Group. (2000)
16             Occupational risk factors for renal cell carcinoma: agent-specific results from a case-control study in
17             Germany. Int. J. Epidemiol. 29: 1014-1024.
18     Peterson, K. E.; Salganik, M.; Campbell, C.; Rhoads, G. G.; Rubin, J.; Berger, O.;  Ware, J. H.; Rogan, W. (2004)
19             Effect of succimer on growth of preschool children with moderate blood lead levels. Environ. Health
20             Perspect.  112:233-237.
21     Pilion, J. J.; Schmitt, N.; Rowe, J.; Gelpke, P. M. (1997) Effect of lead on fetal growth in a Canadian smelter city,
22             1961-1990. Arch. Environ. Health 52: 472-475.
23     Pineda-Zavaleta, A. P.; Gracia-Vargas, G.; Borja-Aburto, V. H.; Acosta-Saavedea, L. C.; Vera Aguilar, E.; Gomez-
24             Munoz, A.; Cebrian, M. E. Calderon-Aranda, E. S. (2004) Nitric oxide  and superoxide anion production in
25             monocytes from children exposed to arsenic and lead in region Lagunera, Mexico. Toxicol. Appl.
26             Pharmacol. 198: 283-290.
27     Pinkerton, L. E.; Biagini, R. E.; Ward, E. M.; Hull, R. D.; Deddens, J. A.; Boeniger, M. F.; Schnorr, T. M.;
28             MacKenzie, B. A.; Luster, M. I. (1998) Immunologic findings among lead-exposed workers. Am. J. Ind.
29             Med. 33:400-408.
30     Pinto de Almeida, A. R.; Carvalho, F. M.; Spinola, A. G.; Rocha, H. (1987) Renal dysfunction in Brazilian lead
31             workers. Am. J. Nephrol. 7: 455-458.
32     Piomelli, S.; Seaman, C.; Zullow, D.; Curran, A.; Davidow, B. (1982) Threshold for lead damage to heme synthesis
33             in urban children. Proc. Natl. Acad. Sci. U. S. A. 79: 3335-3339.
34     Pirkle, J. L.; Brody, D. J.; Gunter, E. W.; Kramer, R. A.; Paschal, D. C.; Flegal, K. M.; Matte, T. D. (1994) The
3 5             decline in blood lead levels in the United States: the National Health and Nutrition Examination Surveys
36             (NHANES). JAMA J. Am. Med. Assoc. 272: 284-291.
37     Pirkle, J. L.; Kaufmann, R. B.; Brody, D. J.; Hickman, T.; Gunter, E. W.; Paschal,  D. C. (1998) Exposure of the U.S.
38             population to lead, 1991-1994. Environ. Health Perspect. 106: 745-750.
39     Poblano, A.; Rothenberg, S. J.; Schnaas, L.; Elias, Y.; Cruz, M. L.  (2001) Spatial distribution of EEC theta activity
40             as a function of lifetime lead exposure in 9-year-old children. Neurotoxicology 22: 439-446.
41     Pocock, S. J.; Shaper, A. G.; Walker, M.; Wale, C. J.; Clayton, B.; Delves, T.; Lacey, R. F.; Packham, R. F.; Powell,
42             P. (1983)  The effects of tap water lead, water hardness, alcohol, and cigarettes on blood lead
43             concentrations. J. Epidemiol.  Community Health 37: 1-7.
44     Pocock, S. J.; Shaper, A. G.; Ashby, D.; Delves, T.; Whitehead, T. P. (1984) Blood lead concentration, blood
45             pressure, and renal function. Br. Med. J. 289: 872-874.
46     Pocock, S. J.; Ashby, D.; Smith, M. A. (1987) Lead exposure and children's intellectual performance.  Int. J.
47             Epidemiol. 16: 57-67.
48     Pocock, S. J.; Smith, M.; Baghurst, P.  (1994) Environmental lead and children's  intelligence: a systematic review of
49             the epidemiological evidence. Br. Med. J. 309:  1189-1197.
50     Pocock, S. J.; Smith, M.; Baghurst, P.  (1995) Environmental lead and children's  intelligence [author's  reply]. Br.
51             Med. J. 310: 1409.
52     Poleckinger, B.; Ulm, M. R.; Golaszewski, T.; Meisinger, V.; Suzin, J.; Grudzinska, M.; Zdziennicki,  A.; Dadak, D.
53             (1996) Lead, mercury, and cadmium exposure of neonates in Poland compared to Austria and other
54             European countries. Trace Elem. Electrol. 13: 22-25.
55     Pollock, C. A.; Ibels, L. S. (1988) Lead intoxication in Sydney Harbour bridge workers. Aust. N. Z. J. Med.
56             18:46-52.
       December 2005                                6-390         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Popovac, D.; Graziano, I; Seaman, C.; Kaul, B.; Colakovic, B.; Popovac, R.; Osmani, I.; Haxhiu, M.; Begraca, M.;
  2             Bozovic, Z.; Mikic, M. (1982) Elevated blood lead in a population near a lead smelter in Kosovo,
  3             Yugoslavia. Arch. Environ. Health 37: 19-23.
  4     Popovic, M.; McNeill, F. E.; Chettle, D. R.; Webber, C. E.; Lee, C. V.; Kaye, W. E. (2005) Impact of occupational
  5             exposure on lead levels in women. Environ. Health Perspect. 113: 478-484.
  6     Port, S.; Garfinkel, A.; Boyle, N. (2000) There is a non-linear relationship between mortality and blood pressure.
  7             Eur. Heart! 21: 1635-1638.
  8     Poulos, L.; Qammaz, S.; Athanaselis, S.; Maravelias, C.; Koutselinis, A. (1986) Statistically significant
  9             hematopoietic effects of low blood lead levels. Arch. Environ. Health 41: 384-386.
10     Pounds, J. G.; Long, G. I; Rosen, J. F. (1991) Cellular and molecular toxicity of lead in bone. Environ. Health
11             Perspect. 91: 17-32.
12     Powell, J. J.; Greenfield, S. M.; Thompson, R. P. H.; Cargnello, J. A.; Kendall, M. D.; Landsberg, J. P.; Watt, F.;
13             Delves, H. T.; House, I. (1995) Assessment of toxic metal exposure following the Camelford water
14             pollution incident: evidence of acute mobilization of lead into drinking water. Analyst (Cambridge, U. K.)
15             120:793-798.
16     Price, J.; Grudzinski, A. W.;  Craswell, P. W.; Thomas, B. J. (1992) Bone lead measurements in patients with chronic
17             renal disease studied over time. Arch. Environ. Health 47: 330-335.
18     Price, R. G.; Patel, S.; Olivers, I.; Milligan, P.; Taylor, S. A. (1999) Early markers of nephrotoxicity: detection of
19             children at risk from environmental pollution. Ren. Fail. 21: 303-308.
20     Pringle, E.; Phillips, C.; Thijs, L.; Davidson, C.; Staessen, J. A.; de Leeuw, P. W.; Jaaskivi, M.; Nachev, C.; Parati,
21             G.; O'Brien, E. T.; Tuomilehto, J.; Webster, J.; Bulpitt, C. J.; Fagard, R. H.; Syst-Eur Investigators. (2003)
22             Systolic blood pressure variability as a risk factor for stroke and cardiovascular mortality in the elderly
23             hypertensive population. J. Hypertens. 21: 2251-2257.
24     Proctor, S. P.; Rotnitzky, A.; Sparrow, D.; Weiss, S. T.; Hu, H. (1996) The relationship of blood lead and dietary
25             calcium to blood pressure in the normative aging study. Int. J. Epidemiol. 25: 528-536.
26     Prospective Studies Collaboration. (2002) Age-specific relevance  of usual blood pressure to vascular mortality: a
27             meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 360: 1903-1913.
28     Prpic-Majic, D.; Bobic, J.; Simic, D.; House, D. E.; Otto, D. A.; Jurasovic, J.; Pizent, A.  (2000) Parental education
29             as a confounder in the assessment of low level lead effect on psychological functions in children. Cent. Eur.
30             J. Public Health 8(suppl.): 69.
31     Prpic-Majic, D.; Bobicc, J.; Simicc, D.; House, D. E.; Otto, D. A.; Jurasovicc, J.; Pizent, A. (2000) Lead absorption
32             and psychological function in Zabreb (Croatia) school children. Neurotoxicol. Teratol. 22: 347-356.
33     Pueschel, S. M.; Kopito, L.; Schwachman, H. (1972) Children with an increased lead burden. A  screening and
34             follow-up study. JAMA J. Am. Med. Assoc. 222: 462-466.
35     Puzas, J. E.  (2000) Osteotoxicology: the role of lead in bone disease. Curr. Opin. Orthop. 11: 360-365.
36     Puzas, J. E.; Sickel, M. J.; Felter, M. E. (1992) Osteoblasts and chondrocytes are important target cells for the toxic
37             effects of lead. Neurotoxicology 13: 783-788.
3 8     Pyatt, D. W.; Zheng, J.-H.; Stillman, W. S.; Irons, R. D. (1996) Inorganic lead activates NF-kB in primary human
39             CD4+ T lymphocytes. Biochem. Biophys. Res. Commun. 227: 380-385.
40     Quandt, R. E. (1958) The estimation of the parameters of a linear regression system obeying two separate regimes.
41             J. Am. Stat. Assoc. 53:873-880.
42     Que Hee, S.S.; Boyle, J.R. (1988) Simultaneous multielemental analysis of some environmental  and biological
43             samples by inductively coupled plasma atomic emission spectrometry. Anal. Chem. 60: 1033-1042.
44     Que Hee, S. S.; MacDonald,  T. J.;  Bornschein, R. L. (1985) Blood lead by furnace-Zeeman atomic absorption
45             spectrophotometry.  Microchem. J. 32: 55-63.
46     Queiroz, M. L. S.; Almeida, M.; Gallao, M. I.; Hoehr, N. F. (1993) Defective neutrophil  function in workers
47             occupationally exposed to lead. Pharmacol. Toxicol. 72: 73-77.
48     Queiroz, M. L.; Costa, F. F.;  Bincoletto, C.; Perlingeiro, R. C. R.;  Dantas, D. C. M.; Cardoso, M. P.; Almeida, M.
49             (1994a) Engulfment and killing capabilities of neutrophils and phagocytic splenic function in persons
50             occupationally exposed to lead. Int. J. Immunopharmacol. 16: 239-244.
51     Queiroz, M. L. S.; Perlingeiro, R. C. R.; Bincoletto, C.; Almeida, M.; Cardoso, M. P.; Dantas, D. C. M. (1994b)
52             Immunoglobulin levels and cellular immune function in lead exposed workers. Immunopharmacol.
53             Immunotoxicol.  16: 115-128.
54     Raab, G. M.; Thomson, G. O. B.; Boyd, L.; Fulton, M.; Laxen, D. P. H. (1990) Blood lead levels, reaction time,
55             inspection time and ability in Edinburgh children. Br. J. Dev. Psychol. 8: 101-118.
56     Rabinowitz, M. B.  (1991) Toxicokinetics of bone lead. Environ. Health Perspect. 91: 33-37.
       December 2005                                 6-391         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Rabinowitz, M. (1993) Declining blood lead levels and cognitive change in children [letter]. JAMA J. Am. Med.
  2            Assoc. 270: 827-829.
  3     Rabinowitz, M. B. (1995) Environmental lead and children's intelligence: Taiwan results are not included [letter].
  4            Br. Med. 1310:397.
  5     Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1973) Lead metabolism in the normal human: stable isotope
  6            studies. Science (Washington, DC) 182: 725-727.
  7     Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1976) Kinetic analysis of lead metabolism in healthy humans.
  8            J. Clin. Invest. 58: 260-270.
  9     Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1977) Magnitude of lead intake from respiration by normal
10            man. J. Lab. Clin. Med. 90: 238-248.
11     Rabinowitz, M.; Needleman, H.; Burley, M.; Finch, H.; Rees, J. (1984) Lead in umbilical blood, indoor air, tap
12            water, and gasoline in Boston. Arch. Environ. Health 39: 299-301.
13     Rabinowitz, M.; Bellinger, D.; Leviton, A.; et al. (1987) Pregnancy hypertension, blood pressure during labor, and
14            blood lead levels. Hypertension 10: 447-451.
15     Rabinowitz, M. B.; Leviton, A.; Bellinger, D. C. (1989) Blood lead-tooth lead relationship among Boston Children.
16            Bull. Environ. Contam. Toxicol. 43: 485-492.
17     Rabinowitz, M. B.; Allred, E. N.; Bellinger, D. C.; Leviton, A.; Needleman, H. L. (1990) Lead and childhood
18            propensity to infectious and allergic disorders: is there an association? Bull. Environ. Contam. Toxicol.
19            44:657-660.
20     Rabinowitz, M. B.; Bellinger, D.; Leviton, A.; Wang, J.-D. (1991) Lead levels among various deciduous tooth types.
21            Bull. Environ. Contam. Toxicol. 47: 602-608.
22     Rabinowitz, M. B.; Wang, J.-D.; Soong, W. T. (1992) Children's classroom behavior and lead in Taiwan. Bull.
23            Environ. Contam. Toxicol. 48: 282-288.
24     Rabinowitz, M. B.; Leviton, A.; Bellinger, D. (1993) Relationships between serial blood lead levels and exfoliated
25            tooth dentin lead levels: models of tooth lead kinetics. Calcif. Tissue  Int. 53: 338-341.
26     Rahman, A.; Hakeem, A.  (2003) Blood lead levels during pregnancy and pregnancy outcome in Karachi women.
27            J.Pakistan Med.  Assoc. 53:529-533.
28     Rahman, A.; Maqbool, E.; Zuberi, H. S. (2002) Lead-associated deficits in stature, mental ability and behaviour in
29            children in Karachi. Ann. Trop. Paediatr. 22: 301-311.
30     Rajah, T.; Ahuja, Y. R. (1995) In vivo genotoxic effects of smoking and occupational lead exposure in printing press
31            workers. Toxicol. Lett. 76: 71-75.
32     Rajah, T. T.; Ahuja, Y. R. (1996) In vivo genotoxicity of alcohol consumption and lead exposure in printing press
33            workers. Alcohol 13: 65-68.
34     Rajegowda, B. K.; Glass, L.; Evans, H. E. (1972) Lead concentrations in the newborn infant. J. Pediatr. 80: 116-117.
35     Ratzon, N.; Froom, P.; Leikin, E.; Kristal-Boneh, E.; Ribak, J. (2000) Effect of exposure to lead on postural control
36            in workers. Occup. Environ. Med. 57: 201-203.
37     Refowitz, R. M. (1984) Thyroid function and lead: no clear relationship. J. Occup. Med. 26: 579-583.
38     Reigart, J. R.; Graber, C. D. (1976) Evaluation of the humoral immune response of children with low level lead
39            exposure. Bull. Environ. Contam. Toxicol. 16: 112-117.
40     Reimer, W.; Tittelbach, U. (1989) Verhalten von Herzfrequenz, Blutdruck und systolischen Zeitintervallen in Ruhe
41            und wahrend Einhandarbeit bei Bleiexponierten und Kontrollpersonen [Heart rate, blood pressure and
42            systolic time  interval in rest and during single-hand exertion in persons exposed to lead and in control
43            subjects]. Z. Gesamte Hyg. Ihre Grenzgeb. 35: 491-492.
44     Rencher, A. C.; Carter, M. W.; McKee, D. W. (1977) A retrospective epidemiological study of mortality at a large
45            western copper smelter. J. Occup. Med. 19: 754-758.
46     Research Triangle Institute. (1999) Health effects. In: Toxicological profile for lead. Atlanta, GA: U.S. Department
47            of Health and Human Services, Agency for Toxic Substances and Disease Registry.
48     Rhainds, M.; Levallois, P. (1997) Effects of maternal cigarette smoking and alcohol consumption on blood lead
49            levels of newborns. Am. J. Epidemiol. 145: 250-257.
50     Rhainds, M.; Levallois, P.; Dewailly, E.; Ayotte, P.  (1999) Lead, mercury, and organochlorine compound levels in
51            cord blood in Quebec, Canada. Arch. Environ. Health 54: 40-47.
52     Rhodes, D.; Spiro, A.; Aro, A.; Hu,  H. (2003) Relationship of bone and blood lead levels to psychiatric symptoms:
53            The Normative aging Study. J. Occup. Environ. Med. 45: 1144-1151.
54     Rico, J.; Kordas, K.; et al. (2005) The efficacy of iron and/or zinc supplementation on cognitive performance of
55            lead-exposed mexican school children: a randomized, placebo-controlled trial. Pediatrics: in press.
       December 2005                                6-392        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Ris, M. D. (2003) Causal inference in lead research: introduction to the special section on the neurobehavioral
  2             effects of environmental lead.  Child Neuropsychol. 9: 1-9.
  3     Ris, M. D.; Dietrich, K. N.; Succop, P.  A.; Berger, O. G.; Bornschein, R. L. (2004) Early exposure to lead and
  4             neuropsychological outcome in adolescence. J. Int. Neuropsychol. Soc. 10: 261-270.
  5     Risch, H. A.; Burch, J. D.; Miller, A. B.; Hill, G. B.; Steele, R.; Howe, G. R. (1988) Occupational factors and the
  6             incidence of cancer of the bladder in Canada. Br. J. Ind. Med. 45: 361-367.
  7     Robertson, W. O. (1996) Elevated environmental lead levels in a day care setting.  Arch. Pediatr. Adolesc. Med.
  8             150:556.
  9     Robins, J. M.; Cullen, M. R.; Connors,  B. B.; Kayne, R. D.  (1983) Depressed thyroid indexes associated with
10             occupational exposure to inorganic lead. Arch. Intern. Med. 143: 220-224.
11     Rodamilans, M.; Osaba, M. J. M.; To-Figueras, J.; Rivera Filial, F.; Marques, J. M.; Perez, P.; Corbella, J. (1988)
12             Lead toxicity on endocrine testicular function in an occupationally exposed population. Hum. Toxicol.
13             7: 125-128.
14     Roderer, G.; Doenges, K. H. (1983) Influence of trimethyl lead and inorganic lead on the in vitro assembly of
15             microtubules from mammalian brain. Neurotoxicology 4: 171-180.
16     Roelofs-Iverson, R.  A.; Mulder, D. W.; Elveback, L. R.; Kurland, L. T.; Molgaard, C. A. (1984) Amyotrophic
17             lateral sclerosis and heavy metals: a pilot case study. Neurology 34: 393-395.
18     Roels, H.; Lauwerys, R. (1987) Evaluation of dose-effect and dose-response relationships for lead exposure in
19             different Belgian population groups (fetus, child, adult men and women). Trace Elem. Med. 4: 80-87.
20     Roels, H. A.; Balis-Jacques, M. N.; Buchet, J.-P.; Lauwerys, R. R. (1979) The influence of sex and of chelation
21             therapy on erythrocyte protoporphyrin and urinary "delta"-aminolevulinic acid in lead-exposed workers.
22             J. Occup. Med. 21:527-539.
23     Roels, H. A.; Lauwerys, R. R.; Buchet, J. P.; Bernard, A. M.; Vos, A.; Oversteyns, M.  (1989) Health significance of
24             cadmium induced renal dysfunction: a five year follow up. Br. J. Ind. Med. 46: 755-764.
25     Roels, H.; Lauwerys, R.; Konings, J.; Buchet, J.-P.; Bernard, A.; Green, S.; Bradley, D.; Morgan, W.; Chettle, D.
26             (1994) Renal function and hyperfiltration capacity in lead smelter workers with high bone lead. Occup.
27             Environ. Med. 51:505-512.
28     Roels, H.; Konings, J.; Green, S.; Bradley, D.; Chettle, D.; Lauwerys, R. (1995) Time-integrated blood lead
29             concentration is a valid surrogate for estimating the cumulative lead dose assessed by tibial lead
30             measurement. Environ. Res. 69: 75-82.
31     Roels, H. A.; Van Assche, F. J.; Oversteyns, M.; De Groof,  M.; Lauwerys, R. R.; Lison, D. (1997) Reversibility of
32             microproteinuria in cadmium workers with incipient tubular dysfunction  after reduction of exposure. Am.
33             J. Ind. Med. 31:645-652.
34     Rogan, W. J.; Ware, J. H. (2003) Exposure to lead in children ~ how low is low enough? N.  Engl. J. Med. 348:
35             1515-1516.
36     Rogan, W. J.; Treatment of Lead-Exposed Clinical Trial Group. (1998) The treatment of lead-exposed children
3 7             (TLC) trial: design and recruitment for a study of the effect of oral chelation on growth and develpment in
3 8             toddlers. Paediatr. Perinat. Epidemiol. 12:313-333.
39     Rogan, W. J.; Dietrich, K.  N.; Ware, J.  H.; et al. (2001) The effect of chelation therapy with  succimer on
40             neuropsychological development in children exposed to lead. New Engl.  J. Med. 344: 1421-1426.
41     Rom, W. N. (1976)  Effects of lead on the female and reproduction: a review. Mt. Sinai J. Med. 43: 542-552.
42     Romeo, R.; Aprea, C.; Boccalon, P.; Orsi. D; Porcelli, B.; Sartorelli, P. (1996) Serum erthropoietin and blood lead
43             concentrations. Int. Arch. Occup. Environ. Health 69: 73-75.
44     Romieu, I.; Carreon, T.; Lopez, L.; Palazuelos, E.; Rios, C.; Manuel, Y.; Hernandez-Avila, M. (1995)
45             Environmental urban lead exposure and blood lead levels in children of Mexico City. Environ. Health
46             Perspect. 103: 1036-1040.
47     Rose, G.; Day, S. (1990) The population mean predicts the number of deviant individuals. Br. Med. J.
48             301: 1031-1034.
49     Rosen, J. F.; Mushak, P. (2001) Primary prevention of childhood lead poisoning ~ the  only solution [comment].
50             N. Engl. J.  Med. 344: 1470-1471.
51     Rosen, J. F.; Pounds, J. G.  (1998) "Severe chronic lead insult that maintains body burdens of lead related to those in
52             the skeleton": observations by Dr. Clair Patterson conclusively demonstrated.  Environ. Res. 78: 140-151.
53     Rosen, J. F.; Chesney, R. W.; Hamstra, A.; DeLuca, H. F.; Mahaffey, K. R. (1980) Reduction in 1,25-
54             dihydroxyvitamin D in children with increased lead absorption. N. Engl. J. Med. 302: 1128-1131.
       December 2005                                6-393         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Rosen, J. F.; Markowitz, M. E.; Bijur, P. E.; Jenks, S. T.; Wielopolski, L.; Kalef-Ezra, J. A.; Slatkin, D. N. (1989)
  2             L-line x-ray fluorescence of cortical bone lead compared with the CaNa2EDTA test in lead-toxic children:
  3             public health implications. Proc. Natl. Acad. Sci. U. S. A. 86: 685-689,
  4     Rosenthal, R. (1984) Meta-analytic procedures for social research. Beverly Hills, CA: Sage Publications.
  5     Roses, O. E.; Alvarez, S.; Conti, M. L; Nobile, R. A.; Villaamil, E. C. (1989) Correlation between lead and prolactin
  6             in males exposed and unexposed to lead in Buenos Aires (Argentina) area. Bull. Environ. Contam. Toxicol.
  7             42:438-442.
  8     Rossner, P.; Boffetta, P.; Ceppi, M.; Bonassi, S.; Smerhovsky, Z.; Landa, K.; Juzova, D.; Sram, R. J. (2005)
  9             Chromosomal aberrations in lymphocytes of healthy subjects and risk of cancer. Environ. Health Perspect.
10             113:517-520.
11     Rothenberg, S. J.; Rothenberg, J.  C. (2005) Testing the dose-response specification in epidemiology: public health
12             and policy consequences for lead. Environ. Health Perspect.: in press.
13     Rothenberg, S. J.; Schnaas, L.; Cansino-Ortiz, S.; Perroni-Hernandez, E.; de la Torre, P.; Neri-Mendez, C.;
14             Ortega, P.; Hidalgo-Loperena, H.; Svendsgaard, D. (1989) Neurobehavioral deficits after low level lead
15             exposure in neonates: the Mexico City pilot study. Neurotoxicol. Teratol. 11: 85-93.
16     Rothenberg, S. J.; Karchmer, S.; Schnaas, L.; Perroni, E.; Zea, F.; Alba, J. F. (1994) Changes in serial blood lead
17             levels during pregnancy. Environ. Health Perspect. 102: 876-880.
18     Rothenberg, S. J.; Cansino, S.; Sepkoski, C.; Torres, L. M.; Medina, S.; Schnaas, L.; Poblano, A.; Karchmer, S.
19             (1995) Prenatal and perinatal lead exposures alter acoustic cry parameters of neonate. Neurotoxicol.
20             Teratol. 17:  151-160.
21     Rothenberg, S. J.; Karchmer, S.; Schnaas, L.; Perroni, E.; Zea, F.; Salinas, V.; Alba, J. F. (1996) Maternal influences
22             on cord blood lead levels. J. Exposure Anal. Environ. Epidemiol. 6: 211-227.
23     Rothenberg, S. J.; Manalo, M.; Jiang, J.; Cuellar, R.; Reyes, S.; Sanchez, M.; Diaz, M.; Khan, F.; Aguilar, A.;
24             Reynoso, B.; Juaregui, M.; Acosta, S.; Johnson, C. (1999) Blood lead level and blood pressure during
25             pregnancy in south central Los Angeles. Arch. Environ. Health 54: 382-389.
26     Rothenberg, S. J.; Khan, F.; Manalo, M.; Jian, J.; Cuellar, R.; Reyes, S.; Acosta, S.; Jauregui, M.; Diaz, M.;
27             Sanchez, M.; Todd, A. C.; Johnson, C. (2000) Maternal bone lead contribution to blood lead during and
28             after pregnancy. Environ. Res. 82: 81-90.
29     Rothenberg, S. J.; Kondrashov, V.; Manalo, M.; Manton, W. L; Khan, F.; Todd, A. C.; Johnson, C. (2001) Seasonal
30             variation in bone lead contribution to blood lead during pregnancy. Environ. Res. 85: 191-194.
31     Rothenberg, S. J.; Kondrashov, V.; Manalo, M.; Jiang, J.; Cuellar, R.; Garcia, M.; Reynoso, B.; Reyes, S.; Diaz, M.;
32             Todd, A. C. (2002) Increases in hypertension and blood pressure during pregnancy with increased bone
33             lead levels. Am. J. Epidemiol. 156: 1079-1087.
34     Rothenberg, S. J.; Schnaas, L.; Salgado-Valladares, M.; Casanueva, E.; Geller, A. M.; Hudnell, H. K.; Fox, D. A.
3 5             (2002) Increased ERG a- and b-wave amplitudes in 7- to 10-year-old children resulting from prenatal lead
36             exposure. Invest. Ophthalmol. Vis. Sci. 43: 2036-2044.
37     Rowe, J. W.; Andres, R.; Tobin, J. D.; Norris, A. H.; Shock, N. W. (1976) Age adjusted standards for creatinine
38             clearance. Ann. Intern. Med. 84: 567-569.
39     Rowland, A.; Wilcox, A. (1987) Maternal blood lead  [letter]. J. Epidemiol. Community Health 41:  184.
40     Rucoba, R. J.; Cajolet, L.; Loy, G.; Binns, H. J. (1998) Prevalence of elevated blood lead levels in inner-city
41             pregnant women. Pediatr. Res. 43: 118A.
42     Ruff, H. A. (1999) Population-based data and the development of individual children: the case of low to moderate
43             lead levels and intelligence. J. Dev. Behav. Pediatr. 20: 42-49.
44     Ruff, H. A.; Bijur P. E.; Markowitz, M.; Ma, Y.-C.; Rosen, J. F. (1993) Declining blood lead levels and cognitive
45             changes in moderately lead-poisoned children. JAMA J. Am. Med.  Assoc. 269: 1641-1646.
46     Ruff, H. A.; Markowitz, M. E.; Bijur, P. E.; Rosen, J. F.; Ma, Y.-C. (1993) Declining blood lead levels and cognitive
47             change in children [author reply]. JAMA J. Am. Med. Assoc. 270: 828-829.
48     Ruff, H. A.; Markowitz, M. E.; Bijur, P. E.; Rosen, J. F. (1996) Relationships among blood lead levels, iron
49             deficiency, and cognitive development in two-year-old children. Environ. Health Perspect. 104: 180-185.
50     Ryu, J. E.; Ziegler, E. E.; Nelson, S. E.; Fomon, S. J. (1983) Dietary intake of lead and blood lead concentration in
51             early infancy. Am. J. Dis. Child. 137:  886-891.
52     Saenger, P.; Markowitz, M. E; Rosen, J. F. (1984) Depressed excretion of 6Beta-hydroxycortisol in lead-toxic
53             children. J. Clin. Endocrinol. Metab. 58: 363-367.
54     Salkever, D. S. (1995) Updated estimates of earnings benefits from reduced exposure of children to environmental
55             lead. Environ. Res. 70: 1-6.
       December 2005                                6-394         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Sallmen, M; Lindbohm, M.-L.; Anttila, A.; Taskinen, H.; Hemminki, K. (1992) Paternal occupational lead exposure
  2             and congenital malformations. J. Epidemiol. Community Health 45: 519-522.
  3     Sallmen, M.; Anttila, A.; Lindbohm, M.-L.; Kyyronen, P.; Taskinen, H.; Hemminki, K. (1995) Time to pregnancy
  4             among women occupationally exposed to lead. J. Occup. Environ. Med. 37: 931-934.
  5     Sallmen, M.; Lindbohm, M. L.; Anttila, A.; Taskinen, H.; Hemminki, K. (2000) Time to pregnancy among the
  6             wives of men occupationally exposed to lead. Epidemiology 11: 141-147.
  7     Salmon, P. L.; Bondarenko, O. A.; Henshaw, D. L. (1999) DOSE210, a semi-empirical model for prediction of
  8             organ distribution and radiation doses from long-term exposure to 210Pb and 210Po. Radiation Protection
  9             Dosimetry 82: 175-192.
10     Sanchez, S. E.; Larrabure, G.; Zhang, C.; Williams, M. A. (2001) Red blood cell selenium, zinc and lead levels in
11             relation to preeclampsia risk among Peruvian women. Am. J. Epidemiol. 153(11 suppl.): S157.
12     Sanchez-Fructuoso, A. I.; Torralbo, A.; Arroyo, M.; Luque, M.; Ruilope, L. M.; Santos, J. L.; Cruceyra, A.;
13             Barrientos, A. (1996) Occult lead intoxication as a cause of hypertension and renal failure. Nephrol. Dial.
14             Transplant. 11:  1775-1780.
15     Sanin, L. H.; Gonzalez-Cossio, T.; Romieu, I.; Peterson, K. E.; Ruiz, S.; Palazuelos, E.; Hernandez-Avila, M.;
16             Hu, H. (2001) Effect of maternal lead burden on infant weight and weight gain at one month of age among
17             breastfed infants. Pediatrics 107: 1016-1023.
18     Sankila, R.; Karjalainen, S.; Pukkala, E.; Oksanen, H.; Hakulinen, T.; Teppo, L.; Hakama, M. (1990) Cancer risk
19             among glass factory workers: an excess of lung cancer? Br. J. Ind. Med. 47: 815-818.
20     Sarasua, S. M.; Vogt, R. F.; Henderson, L. O.; Jones, P. A.; Lybarger, J. A. (2000) Serum immunoglobulins and
21             lymphocyte subset distributions in children and adults living in communities assessed for lead and
22             cadmium exposure. J. Toxicol. Environ. Health A. 60(1): 1-15.
23     Sarasua, S. M.; Mueller, P.; Kathman, S.; Campagna, D.; Uddin, M. S.; White, M. C. (2003) Confirming the utility
24             of four kidney biomarker tests in a longitudinal follow-up study. Renal Failure 25: 797-817.
25     Sargent, J. D.; Dalton, M. A.; O'Connor, G. T.; Olmstead, E. M.; Klein, R. Z. (1999) Randomized trial of calcium
26             glycerophosphate-supplemented infant formula to prevent lead absorption. Am. J. Clin. Nutr.
27             69: 1224-1230.
28     Sarto, F; Stella, M; Acqua, A. (1978) Cytogenetic study of a group of workers with increased lead absorption
29             indices. Med. Lav. 69: 172-180.
30     Sata, F.; Araki, S.;  Sakai, T.; Nakata, A.; Yamashita, K.; Morita, Y.; Tanigawa, T.; Miki, A. (1997) Immunological
31             effects of CaEDTA injection: observations in two lead workers. Am. J. Ind. Med. 32: 674-680.
32     Sata, F.; Araki, S.; Tanigawa, T.; Morita, Y.; Sakurai, S.; Nakata, A.; Katsuno, N. (1998) Changes in T cell
33             subpopulations in lead workers. Environ. Res. 76: 61-64.
34     Satarug, S.; Nishijo, M.; Ujjin, P.; Vanavanitkun, Y.; Baker, J. R.; Moore, M. R. (2004) Evidence for concurrent
3 5             effects of exposure to environmental cadmium and lead on hepatic CYP2A6 phenotype and renal function
36             biomarkers in nonsmokers. Environ. HealthPerspect. 112: 1512-1518.
37     Satarug, S.; Ujjin, P.; Vanavanitkun, Y.; Nishijo, M.; Baker, J. R.; Moore, M. R. (2004) Effects of cigarette smoking
3 8             and exposure to cadmium and lead on phenotypic variability of hepatic CYP2A6 and renal function
39             biomarkers in men. Toxicology 204: 161-173.
40     Savitz, D. A.; Whelan, E. A.; Rowland, A. S.; Kleckner, R. C. (1990) Maternal employment and reproductive risk
41             factors.  Am. J. Epidemiol. 132: 933-945.
42     Schafer, J. H.; Glass, T. A.; Bressler, J.; Todd, A. C.; Schwartz, B. S. (2005) Blood lead in a predictor of
43             homocysteine levels in a population-based study of older adults. Environ. Health Perspect. 113:31-35.
44     Scharer, K.; Veits,  G.; Brockhaus, A.; Ewers, U. (1991) High lead content of deciduous teeth in chronic renal
45             failure. Pediatr. Nephrol. 5: 704-707.
46     Schaumberg, D. A.; Mendes, F.; Balaram, M.; Dana, M. R.; Sparrow, D.; Hu, H. (2004) Accumulated lead exposure
47             and risk of age-related cataract in men. JAMA J. Am. Med. Assoc. 292: 2750-2754.
48     Schell, L. M.; Stark, A. D.; Gomez, M. I.; Grattan, W. A. (1997) Blood lead level, by year and season, among poor
49             pregnant women. Arch. Environ. Health 52: 286-291.
50     Schell, L. M.; Czerwinski,  S.; Stark, A. D.; Parsons, P. J.; Gomez, M.; Samelson, R. (2000) Variation in blood lead
51             and hematocrit levels during pregnancy in a socioeconomically disadvantaged population. Arch. Environ.
52             Health 55: 134-140.
53     Schildcrout, J. S.; Heagerty, P. J. (2005) Regressions analysis of longitudinal binary data with time-dependent
54             environmental covariates: bias and efficiency. Biostatistics: doi: 10.1093/biostatistics/kxi033.
55     Schindler, A. M.; Haecker, T.; Gould, J.; Turner, E.; Torchia, M.; Kaye, R.; Cockerill, M.; Spachman, S. (1993)
56             Declining blood lead levels and cognitive change in children [letter]. JAMA J. Am. Med. Assoc. 270: 828.
       December 2005                                6-395        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Schmid, E.; Bauchinger, M.; Pietruck, S.; Hall, G. (1972) Die cytogenetische Wirkung von Blei in menschlichen
  2            peripheren Lymphocyten in vitro und in vivo [The cytogeneticeffect of lead in human peripehral
  3            lymphocytes in vitro and in vivo]. Mutat. Res. 16: 401-406.
  4     Schnaas, L.; Rothenberg, S. J.; Perroni, E.; Martinez, S.; Hernandez, C.; Hernandez, R. M. (2000) Temporal pattern
  5            in the effect of postnatal blood lead level on intellectual development of young children. Neurotoxicol.
  6            Teratol. 22:  805-810.
  7     Schoen, E. J. (1993) Childhood lead poisoning: definitions and priorities. Pediatrics 91: 504-505.
  8     Schroeder, S. R.; Hawk, B. (1987) Psycho-social factors, lead exposure, and IQ. In: Schroeder, S. R., ed. Toxic
  9            substances and mental retardation: neurobehavioral toxicology and teratology. Washington, DC: American
10            Association on Mental Deficiency; pp. 97-137. (Begab, M. J., ed. Monographs of the American Association
11            on Mental Deficiency: no.  8).
12     Schroeder, H. A.; Tipton, I. H. (1968) The human body burden of lead. Arch. Environ. Health 17: 965-978.
13     Schuhmacher, M.; Hernandez, M.; Domingo, J. L.; Fernandez-Ballart, J. D.; Llobet, J. M.; Corbella, J. (1996)
14            A longitudinal study of lead mobilization during pregnancy: concentrations in maternal and umbilical cord
15            blood. Trace Elem. Electrol. 13: 177-181.
16     Schuhmacher, M.; Patemain, J. L.; Domingo, J. L.; Corbella, J. (1997) An assessment of some biomonitors
17            indicative of occupational exposure to lead. Trace Elem. Electrolytes 14(3): 145-149.
18     Schumacher, C.; Brodkin, C. A.; Alexander, B.; Cullen, M.; Rainey, P. M.; vanNetten, C.; Faustman, E.;
19            Checkoway, H. (1998) Thyroid function in lead smelter workers: absence of subacute or cumulative effects
20            with moderate lead burdens. Int. Arch. Occup. Environ. Health 71: 453-458.
21     Schutz, A.; Skerfving, S.; Christoffersson, J. O.; Ahlgren, L.; Mattson, S. (1987) Lead in vertebral bone biopsies
22            from active and retired lead workers. Arch. Environ. Health 42: 340-346.
23     Schutz, A.; Skerfving, S.; Christoffersson, J. O.; Tell, I. (1987) Chelatable lead versus lead in human trabecular and
24            compact bone. Sci. Total Environ. 61: 201-209.
25     Schutz, A.; Skerfving, S.; Ranstam,  J.; Christoffersson, J.-O. (1987) Kinetics of lead in blood after the end of
26            occupational exposure.  Scand. J. Work Environ. Health 13: 221 231.
27     Schwanitz, G.; Lehnert, G.; Gebhart, E. (1970) Chromosomenschadenbei beruflicher Bleibelastung [Chromosome
28            damage after occupational  exposure to lead]. Dtsch. Med. Wochenschr. 95: 1636-1641.
29     Schwanitz, G.; Gebhart, E.; Rott, H.-D.; Schaller, K.-H.; Essing, H.-G.; Lauer, O.; Prestele, H.  (1975)
3 0            Chromosomenuntersuchungen bei Personen mit beruflicher Bleiexposition [Chromosome investigations in
31            subjects with occupational  lead exposure]. Dtsch. Med. Wochenschr. 100: 1007-1011.
32     Schwartz, J. (1985) Evidence for a blood lead-blood pressure relationship  [memorandum to the Clean Air Science
33            Advisory Committee]. Washington, DC: U.S. Environmental Protection Agency, Office of Policy Analysis.
34            Available for inspection at: U.S. Environmental Protection Agency, Central Docket Section, Washington,
35            DC; docket no. ECAO-CD-81-2 IIA.F.60.
36     Schwartz, J. (1991) Lead, blood pressure, and cardiovascular disease in men and women. Environ. Health Perspect.
37            91:71-75.
3 8     Schwartz, J. (1994) Low-level lead exposure and children's IQ: a meta-analysis and search for a threshold. Environ.
39            Res. 65: 42-55.
40     Schwartz, J. (1995) Lead, blood pressure, and cardiovascular disease in men. Arch. Environ. Health 50: 31-37.
41     Schwartz, J.; Otto, D. (1987) Blood lead, hearing thresholds, and neurobehavioral development in children and
42            youth. Arch. Environ. Health 42: 153-160.
43     Schwartz, J.; Otto, D. (1991) Lead and minor hearing impairment. Arch. Environ. Health 46: 300-305.
44     Schwartz, B. S.; Stewart, W. F. (2000) Different associations of blood lead, meso 2,3-dimercaptosuccinic acid
45            (DMSA)-chelatable lead, and tibial lead levels with blood pressure in 543 former organolead
46            manufacturing workers.  Arch.  Environ. Health 55: 85-92.
47     Schwartz, J.; Angle, C.; Pitcher, H.  (1986) Relationship between childhood blood lead and stature. Pediatrics
48            77:281-288.
49     Schwartz, J.; Landrigan, P. J.; Baker, E. L., Jr.; Orenstein,  W. A.; von Lindern, I. H. (1990) Lead-induced anemia:
50            dose-response relationships and evidence for a threshold. Am. J. Public. Health 80: 165-168.
51     Schwartz, B. S.; Bolla, K. I.; Stewart, W.; Ford, D. P.; Agnew, J.; Frumkin, H. (1993) Decrements in
52            neurobehavioral performance associated with mixed exposure to organic and inorganic lead. Am. J.
53            Epidemiol. 137: 1006-1021.
54     Schwartz, B. S.; Lee, B.-K.; Stewart, W.; Ahn, K.-D.; Kelsey, K.; Bresssler, J. (1997) Associations of subtypes of
55            hemoglobin with delta-aminolevulinic acid dehydratase genotype and dimercaptosuccinic acid-chelatable
56            lead levels. Arch. Environ. Health 52: 97-103.
       December 2005                                6-396         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Schwartz, B. S.; Lee, B.-K.; Stewart, W.; Sithisarankul, P.; Strickland, P. T.; Ahn, K.-D.; Kelsey, K. (1997)
  2             13 Aminolevulinic acid dehydratase genotype modifies four hour urinary lead excretion after oral
  3             administration of dimercaptosuccinic acid. Occup. Environ. Med. 54: 241-246.
  4     Schwartz, B. S.; Lee, B.-K.; Lee, G.-S.; Stewart, W. F.;  Simon, D.; Kelsey, K.; Todd, A. C. (2000a) Associations of
  5             blood lead, dimercaptosuccinic acid-chelatable lead, and tibia lead with polymorphisms in the vitamin D
  6             receptor and "delta"-aminolevulinic acid dehydratase genes. Environ. Health Perspect. 108: 949-954.
  7     Schwartz, B. S.; Stewart, W. F.; Todd, A. C.; Simon, D.; Links, J. M. (2000b) Different associations of blood lead,
  8             meso 2,3-dimercaptosuccinic acid (DMSA)-chelatable lead, and tibial lead levels with blood pressure in
  9             543 former organolead manufacturing workers. Arch. Environ. Health. 55: 85-92.
10     Schwartz, B. S.; Stewart, W. F.; Kelsey, K. T.; Simon, D.; Park, S.; Links, J. M.; Todd, A. C. (2000c) Associations
11             of tibial lead levels with BsmI polymorphisms  in the vitamin D receptor in former organolead
12             manufacturing workers. Environ. Health Perspect. 108: 199-203.
13     Schwartz, B. S.; Stewart, W. F.; Bolla, K. I.; Simon, P. D.; Bandeen-Roche, K.; Gordon, P. B.; Links, J. M.; Todd,
14             A. C. (2000d) Past adult lead exposure is associated with longitudinal decline in cognitive function.
15             Neurology 55: 1144-1150.
16     Schwartz, B. S.; Lee, B. K.; Lee, G. S.; Stewart, W. F.; Lee, S. S.; Hwang, K. Y.; Ahn, K.-D.; Kim, Y.-B.; Bolla,
17             K. I.; Simon, D.; Parsons, P.  J.; Todd, A.  C. (2001a) Associations of blood lead, dimercaptosuccinic acid-
18             chelatable lead, and tibia lead with neurobehavioral test scores in South Korean lead workers. Am. J.
19             Epidemiol. 153:453-464.
20     Schwartz, B. S.; Stewart, W. F.; Bolla, K. I.; Simon, P. D.; Bandeen-Roche, K.; Gordon, P. B.; Links, J. M.; Todd,
21             A. C. (200 Ib) Past adult lead exposure is associated with longitudinal decline in cognitive function.
22             [erratum to Neurology 55: 1144-1150]. Neurology 56: 283.
23     Schwartz, B. S.; Lee, B.-K.; Bandeen-Roche, K.; Stewart, W.; Bolla, K. I.; Links, J.; Weaver, V.; Todd, A. (2005)
24             Occupational lead exposure and longitudinal decline in neurobehavioral test scores. Epidemiology
25             16: 106-113.
26     Schwela, D. (2000) Air pollution and  health in urban areas. Rev. Environ. Health 15: 13-42.
27     Sciarillo, W. G.; Alexander, G.; Farrell, K. P. (1992) Lead exposure and child behavior. Am. J. Public Health
28             82: 1356-1360.
29     Selander, S.; Cramer, K. (1970) Interrelationships between lead in blood, lead in urine, and ALA in urine during
30             lead work. Br. J. Ind. Med. 27: 28-39.
31     Selbst, S. M.; Sokas, R. K.; Henretig,  F. M.; Weller, S. C.; Tershakovec, A. M. (1993) The effect of blood lead on
32             blood pressure in children. J. Environ. Pathol. Toxicol. Oncol. 12: 213-218.
33     Selevan, S.  G.; Landrigan, P. J.; Stern, F. B.; Jones, J. H. (1985) Mortality of lead smelter workers. Am. J.
34             Epidemiol. 122: 673-683.
35     Selevan, S.  G.; Rice, D.  C.; Hogan, K. A.; Euling, S. Y.; Pfahles-Hutchens, A.; Bethel, J. (2003) Blood lead
36             concentration and delayed puberty in girls. N. Engl. J. Med. 348: 1527-1536.
37     Selvester, R. H. S.; Ahmed, J.; Tolan, G. D. (1996) Asymptomatic coronary artery disease detection: update 1996:
38             a screening protocol using 16-lead high-resolution ECG, ultrafast CT, exercise testing, and radionuclear
39             imaging. J. Electrocardiol. 29(suppl.) 135-144.
40     Seshadri, S.; Wolf, P. A.; Beiser, A.; Vasan, R. S.; Wilson, P. W. F.; Kase, C.  S.; Kelly-Hays, M.; Kannel, W. B.;
41             D'Agostino, R.  B. (2001) Elevated midlife blood pressure increases stroke risk in elderly persons: the
42             Framingham Study. Arch. Intern. Med. 161: 2343-2350.
43     Shadick, N. A.; Kim, R.; Weiss, S.; Liang, M. H.;  Sparrow, D.; Hu, H. (2000) Effect of low level lead exposure on
44             hyperuricemia and gout among middle aged and elderly men: the normative aging study. J. Rheumatol.
45             27: 1708-1712.
46     Shannon, M.; Graef, J. W.  (1996) Lead intoxication in children with pervasive developmental disorders. J. Toxicol.
47             Clin. Toxicol. 34: 177-181.
48     Shannon, M.; Woolf, A.; Binns, H. (2001) Chelation therapy in children exposed to lead [comment]. N. Engl. J.
49             Med. 345: 1212-1213.
50     Shapiro, I. M.; Dobkin, B.; Tuncay, O. C.; Needleman,  H. L. (1973) Lead levels in dentin and circumpulpal dentin
51             of deciduous teeth of normal and lead poisoned children. Clin. Chimica. Acta. 46: 119-23.
52     Shapiro, I. M.; Burke, A.; Mitchell, G.; Bloch, P. (1978) X-ray fluorescence analysis of lead in teeth of urban
53             children in situ: correlation between the tooth lead level and the concentration of blood lead and  free
54             erythroporphyrins. Environ. Res.  17: 46-52.
55     Sharma, K.; Reutergardh, L. B. (2000) Exposure of preschoolers to lead in the Makati area of Metro Manila, the
56             Philippines. Environ. Res. A 83: 322-332.
       December 2005                                6-397         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Sharp, D. S.; Benowitz, N. L.; Osterloh, J. D.; Becker, C. E.; Smith, A. H.; Syme, S. L. (1990) Influence of race,
  2             tobacco use, and caffeine use on the relation between blood pressure and blood lead concentration. Am. J.
  3             Epidemiol. 131:845-854.
  4     Sheffet, A.; Thind, L; Miller, A. M; Louria, D. B. (1982) Cancer mortality in a pigment plant utilizing lead and zinc
  5             chromates. Arch. Environ. Health 37: 44-52.
  6     Shen, X.-M; Yan, C.-H.; Quo, D.; Wu, S.-M; Li, R.-Q.; Huang, H.; Ao, L.-M; Zhou, J.-D.; Hong, Z.-Y.; Xu, J.-D.;
  7             Jin, X.-M.; Tang, J.-M. (1998) Low-level prenatal lead exposure and neurobehavioral development of
  8             children in the first year of life: a prospective study in Shanghai. Environ. Res. 79: 1-8.
  9     Shen, X.-M.; Wu, S.-H.; Yan, C.-H.; Zhao, W.; Ao, L.-M.; Zhang, Y.-W.; He, J.-M.; Ying, J.-M.; Li, R.-Q.; Wu,
10             S.-M.; Guo, D. (2000) Delta-aminolevulinate dehydratase polymorphism and blood lead levels in Chinese
11             children. Environ. Res. 85: 185-190.
12     Sherins, R. J. (1995) Are semen quality and male fertility changing? N. Engl. J. Med.  332: 327-328.
13     Sherlock, J. C.; Quinn, M. J. (1986) Relationship between blood and lead concentrations and dietary lead intake in
14             infants: the Glasgow Duplicate Diet Study  1979-1980. Food Addit. Contam.  3:  167-176.
15     Sherlock, J.; Smart, G.; Forbes, G. I.; Moore, M. R.; Patterson, W. J.; Richards, W. N.; Wilson, T. S. (1982)
16             Assessment of lead intakes and dose-response for a population in Ayr exposed to a plumbosolvent water
17             supply. Hum. Toxicol. 1: 115-122.
18     Sherlock, J. C.; Ashby, D.; Delves, H. T.; Forbes, G. I.; Moore, M. R.; Patterson, W. J.; Pocock, S. J.; Quinn, M. J.;
19             Richards, W. N.; Wilson, T. S. (1984) Reduction in exposure to lead from drinking water and its effect on
20             blood lead concentrations. Hum. Toxicol. 3: 383-392.
21     Shiau, C.-Y.; Wang, J.-D.; Chen, P.-C. (2004) Decreased fecundity among male lead workers. Occup. Environ.
22             Med. 61:915-923.
23     Shukla, H.; Atakent, Y. S.; Ferrara, A.; Topsis, J.; Antoine, C. (1987) Postnatal overestimation of gestational age in
24             preterm infants. Am. J. Dis. Child. 141: 1106-1107.
25     Shukla, R.; Bornschein, R. L.; Dietrich, K. N.; Buncher, C. R.; Berger, O. G.; Hammond, P. B.; Succop, P. A.
26             (1989) Fetal and infant lead exposure: effects on growth in stature. Pediatrics 84: 604-612.
27     Shukla, V. K.; Prakash, A.; Tripathi, B. D.; Reddy, D. C.; Singh, S. (1998) Biliary heavy metal concentrations in
28             carcinoma of the gall bladder: case-control study. Br. Med. J. 317: 1288-1289.
29     Siddiqui, M. K.; Srivastava, S.; Mehrotra, P. K. (2002) Environmental exposure to lead as a risk for prostate cancer.
30             Biomed. Environ. Sci. 15: 298-305.
31     Siegel, M.; Forsyth, B.; Siegel, L.; Cullen, M. R. (1989) The effect of lead on thyroid  function in children. Environ.
32             Res. 49: 190-196.
33     Siemiatycki, J.; Germ, M.; Stewart, P.; Nadon, L.; Dewar, R.; Richardson, L. (1988) Associations between several
34             sites of cancer and ten types of exhaust and combustion products: results from a case-referent study in
35             Montreal. Scand. J. Work Environ. Health  14: 79-90.
36     Siemiatycki, J.; Germ, M.; Dewar, R.; Nadon, L.; Lakhani, R.; Begin, D.; Richardson, L. (1991) Associations
37             between occupational circumstances and cancer. In: Siemiatycki, J., ed. Risk factors for cancer in the
38             workplace. Boca Raton, FL: CRC Press; pp. 141-145.
39     Silbergeld, E. K. (1991) Lead in bone: implications  for toxicology during pregnancy and lactation. Environ. Health
40             Perspect. 91:63-70.
41     Silbergeld, E. K.; Schwartz, J.; Mahaffey, K.  (1988) Lead and osteoporosis: mobilization of lead from bone in
42             postmenopausal women. Environ. Res. 47: 79-94.
43     Silbergeld, E. K.; Sauk, J.; Somerman, M.; Todd, A.; McNeill, F.; Fowler, B.; Fontaine, A.; vanBuren, J. (1993)
44             Lead in bone: storage site, exposure source, and target organ. Presented at: Ninth international
45             neurotoxicology conference; October 1991; Little Rock, AR. Neurotoxicology 14(2-3): 225-236.
46     Silbergeld, E. K.; Waalkes, M.; Rice, J. M. (2000) Lead as a carcinogen: experimental evidence and mechanisms of
47             action. Am. J. Ind. Med.  38: 316-323.
48     Silva, P.  A.; Hughes, P.; Williams, S.; Faed, J. M. (1988) Blood lead, intelligence, reading attainment, and
49             behaviour in eleven year old children in Dunedin, New Zealand. J. Child Psychol. Psychiatr.  Allied Discipl.
50             29:43-52.
51     Singh, B.; Chandran, V.; Bandhu, H. K.; Mittal, B. R.; Bhattacharya, A.; Jindal, S. K.; Varma, S. (2000) Impact of
52             lead exposure on pituitary-thyroid axis in humans. BioMetals 13: 187-192.
53     Sirivarasai, J.; Kaojarern, S.; Wananukul, W.; Deechakwan, W.; Srisomerarn, P. (2004) Non-occupational lead and
54             cadmium exposure and blood pressure in Thai men. Asia Pac. J. Public Health 16: 133-137.
       December 2005                                6-398         DRAFT-DO NOT QUOTE OR CITE

-------
  1     SkerfVing, S. (1988) Biological monitoring of exposure to inorganic lead. In: Clarkson, T. W.; Friberg, L.;
  2            Nordberg, G. F.; Sager, R. P., eds. Biological monitoring of toxic metals. New York, NY: Plenum Press;
  3            pp. 169-197.
  4     SkerfVing, S.; Ahlgren, L.; Christoffersson, J.-O.; Haeger-Aronsen, B.; Mattsson, S.; Schutz, A. (1983) Metabolism
  5            of inorganic lead in occupationally exposed humans. Arh. Hig. Rada Toksikol. 34: 341-350.
  6     SkerfVing, S.; Nilsson, U.; Schutz, A.; Gerhardsson, L. (1993) Biological monitoring of inorganic lead. Scand. J.
  7            Work Environ. Health 19(suppl. 1): 59-64.
  8     Slobozhanina, E. L; Kozlova, N. M; Lukyanenko, L. M; Oleksiuk, O. B.; Gabbianelli, R.; Fedeli, D.; Caulini, G.
  9            C.; Falcioni, G. (2005) Lead-induced changes in human erythrocytes and lymphocytes. J. Appl. Toxicol.
10            25: 109-114.
11     Smith, F. L., 2nd; Rathmell, T. K.; Marcil, G. E. (1938) The early diagnosis of acute and latent plumbism. Am. J.
12            Clin. Pathol. 8: 471-508.
13     Smith, C.  M.; Wang, X.; Hu, H.; Kelsey, K. T. (1995) A polymorphism in the "delta"-aminolevulinic acid
14            dehydratase gene may modify the pharmacokinetics and toxicity of lead. Environ. Health Perspect.
15            103:248-253.
16     Smith, D.  R.; Osterloh, J. D.; Flegal, A. R. (1996) Use of endogenous, stable lead isotopes to determine release of
17            lead from the skeleton. Environ. Health Perspect. 104: 60-66.
18     Smith, D.; Hernandez-Avila, M.; Tellez-Rojo, M.M.; Mercado, A.; Hu, H. (2002) The relationship between lead in
19            plasma and whole blood in women. Environ. Health Perspect. 110: 263-268.
20     Sokas, R.  K.; Simmens, S.; Sophar, K.; Welch, L. S.; Liziewski, T. (1997) Lead levels in Maryland construction
21            workers. Am. J. Ind. Med. 31: 188-194.
22     Soldin, O. P.; Pezzullo, J. C.; Hanak, B.; Miller, M.; Soldin, S. J. (2003)  Changing trends in the epidemiology of
23            pediatric lead exposure: interrelationship of blood lead and ZPP concentrations and a comparison to the US
24            population. Ther. Drug Monit. 25: 415 -420.
25     Solliway,  B. M.; Schaffer, A.; Pratt, H.; Yannai, S. (1996) Effects of exposure to lead on selected biochemical and
26            haematological variables. Pharmacol. Toxicol. 78:  18-22.
27     Somervaille, L. J.; Chettle, D. R.; Scott, M. C.; Tennant, D. R.; McKiernan, M.  J.; Skilbeck, A.; Trethowan, W. N.
28            (1988) In vivo tibia lead measurements as an index of cumulative exposure in occupationally exposed
29            subjects. Br. J. Ind. Med. 45:  174-181.
30     Somervaille, L. J.; Nilsson, U.; Chettle, D. R.; Tell, L; Scott, M. C.; Schutz, A.; Mattsson, S.; SkerfVing, S. (1989)
31            In vivo measurements of bone lead~a comparison of two x-ray  fluorescence techniques used at three
32            different bone sites. Phys. Med. Biol. 34: 1833-1845.
33     Sonmez, F.; Donmez, O.; Sonmez, H. M.; Keskinoglu, A.; Kabasakal, C.; Mir, S. (2002) Lead exposure  and urinary
34            N-acetyl "beta" D glucosaminidase activity in adolescent workers in auto repair workshops. J. Adolesc.
35            Health 30: 213-216.
36     Sorel, J. E.; Heiss, G.; Tyroler, H. A.; Davis, W. B.; Wing, S. B.; Ragland, D. R. (1991) Black-white differences in
37            blood pressure among participants in NHANES II: the contribution of blood lead. Epidemiology
38            2:348-352.
39     Sorrell, M.; Rosen, J. F.; Roginsky, M. (1977) Interactions of lead, calcium, vitamin D, and nutrition in lead-
40            burdened children. Arch. Environ. Health  32: 160-164.
41     Sowers, M.; Jannausch, M.; Scholl, T.; Li, W.; Kemp, F. W.; Bogden, J.  D.  (2002) Blood lead concentrations and
42            pregnancy outcomes. Arch. Environ. Health 57: 489-495.
43     Spencer, H.; O'Sullivan, V.; Sontag, S. J. (1992) Does lead play a role in Paget's disease of bone? A hypothesis. J.
44            Lab. Clin. Med. 120: 798-800.
45     Spencer, H.; O'Sullivan, V.; Sontag, S. J. (1994) Occupational exposure to lead: preliminary observations in Paget's
46            disease of bone in women and in family members of affected patients.  J. Trace Elem. Exp. Med. 7: 53-58.
47     Spencer, H.; O'Sullivan, V.; Sontag, S. J. (1995) Exposure to lead, a potentially hazardous toxin - Paget's disease of
48            bone. J. Trace Elem. Exp. Med. 8: 163-171.
49     Spinnato,  J. A.; Sibai, B. M.; Shaver, D.  C.; Anderson, G. D. (1984) Inaccuracy of Dubowitz gestational age  in low
50            birth weight infants. Obstet. Gynecol. 63:  491-495.
51     Spivey, G. H.; Baloh, R. W.; Brown, C. P.; Browdy, B. L.; Campion, D.  S.; Valentine, J. L.; Morgan, D. E.; Culver,
52            B. D. (1980) Subclinical effects of chronic increased lead absorption~a prospective study. III. Neurologic
53            findings at follow-up examination. J. Occup. Med. 22: 607-612.
54     Spreen, O.; Risser, A. T.; Edgell, D. (1995) Developmental neuropsychology. New York, NY: Oxford University
55            Press.
       December 2005                                6-399        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Srivastava, S.; Mehrotra, P. K.; Srivastava, S. P.; Tandon, I.; Siddiqui, M. K. J. (2001) Blood lead and zinc in
  2             pregnant women and their offspring in intrauterine growth retardation cases. J. Anal. Toxicol. 25: 461-465.
  3     Staessen, J. (1995) Low-level lead exposure, renal function and blood pressure. Verh.-K. Acad. Geneeskd. Belg.
  4             57: 527-574.
  5     Staessen, J.; Yeoman, W. B.; Fletcher, A. E.; Markowe, H. L.; Marmot, M. G.; Rose, G.; Semmence, A.; Shipley,
  6             M. J.; Bulpitt, C. J. (1990) Blood lead concentration, renal function, and blood pressure in London civil
  7             servants. Br. J. Ind. Med. 47: 442-447.
  8     Staessen, J.; Amery, A.; Bernard, A.; Bruaux, P.; Buchet, J. P.; Bulpitt, C. J.; Claeys, F.; De Plaen, P.; Ducoffre, G.;
  9             Fagard, R.; Lauwerys, R. R.; Lijnen, P.; Nick, L.; Saint Remy, A.; Roels, H.; Rondia, D.; Sartor, F.;
10             Thijs, L. (1991) Blood pressure, the prevalence of cardiovascular diseases, and exposure to cadmium: a
11             population study.  Am. J. Epidemiol. 134: 257-267'.
12     Staessen, J. A.; Lauwerys,  R. R.; Buchet, J.-P.; Bulpitt, C. J.; Rondia, D.; VanRenterghem, Y.; Amery, A. (1992)
13             Impairment of renal function with increasing blood lead concentrations in the general population. N. Engl.
14             J. Med. 327: 151-156.
15     Staessen, J. A.; Dolenc, P.; Amery, A.; Buchet, J. P.; Claeys, F.; Fayard, R.; Lauwerys, R.; Lijnen, P.; Roels, H.;
16             Rondia, D.;  Sartor, F.; Thijs, L.; Vyncke, G. (1993) Environmental lead exposure does not increase blood
17             pressure in the population at large: evidence from the Cadmibel study. J. Hypertens. ll(suppl.2): S35-S41.
18     Staessen, J. A.; Bulpitt, C.  J.; Fagard, R.; Lauwerys, R. R.; Roels, H.; Thijs, L.; Amery, A. (1994) Hypertension
19             caused by low-level lead exposure: myth or fact? J. Cardiovasc. Risk 1: 87-97.
20     Staessen, J. A.; Roels, H.; Lauwerys, R.  R.; Amery, A. (1995) Low-level lead exposure and blood pressure. J. Hum.
21             Hypertens. 9: 303-328.
22     Staessen, J. A.; Buchet, J.-P.; Ginucchio, G.; Lauwerys, R. R.; Lijnen, P.; Roels, H.; Fagard, R. (1996) Public health
23             implications of environmental exposure to cadmium and lead: an overview of epidemiological studies in
24             Belgium. J.  Cardiovasc. Risk 3: 26-41.
25     Staessen, J. A.; Roels, H.; Fagard, R. (1996) Lead exposure and conventional and ambulatory blood pressure: a
26             prospective  population study. JAMA J. Am. Med. Assoc. 275: 1563-1570.
27     Staessen, J. A.; Nawrot, T.; Den Hond, E.; Thijs, L.; Fagard, R.; Hoppenbrouwers, K.; Koppen, G.; Nelen, V.;
28             Schoeters, G.; Vanderschueren, D.; VanHecke, E.; Verschaeve, L.; Vlietinck, R.; Roels, H. A. (2001)
29             Renal function, cytogenetic measurements, and sexual developments in adolescents in relation to
30             environmental pollutants: a feasibility study of biomarkers. Lancet 357: 1660-1669.
31     Steele, G.; Kattouf, V. (2000) Blood lead levels and vision. Optometry 71: 217-220.
32     Steenhout, A.; Pourtois, M. (1981) Lead accumulation in teeth as a function of age with different exposures. Br. J.
33             Ind. Med. 38: 297-303.
34     Steenland, K.; Boffetta, P.  (2000) Lead and cancer in humans: where are we now? Am. J.  Ind. Med. 38: 295-299.
35     Steenland, K.; Thun, M. J.; Ferguson, C. W.; Port, F. K. (1990) Occupational and other exposures associated with
36             male end-stage renal disease: a  case/control study. Am. J. Public Health. 80: 153-157.
37     Steenland, K.; Selevan, S.; Landrigan, P. (1992) The mortality of lead smelter workers: an update. Am. J. Public
38             Health 82: 1641-1644.
39     Steenland, K.; Loomis, D.; Shy, C.; Simonsen, N. (1996) Review of occupational lung carcinogens.  Am. J. Ind.
40             Med. 29: 474-490.
41     Steenland, K.; Mannetje, A.; Boffetta, P.; Stayner, L.; Attfield, M.; Chen, J.; Dosemeci, M.; DeKlerk, N.;
42             Hnizdo, E.;  Koskela, R.; Checkoway, H. (2002) Pooled exposure-response analyses  and risk assessment for
43             lung cancer in 10  cohorts of silica-exposed workers: an IARC multi-centric study (vol 12, pg 773, 2001).
44             Cancer Causes Control 13: 777.
45     Stefanak, M. A.; Bourguet, C. C.; Benzies-Styka, T. (1996) Use of the Centers for Disease Control and Prevention
46             childhood lead poisoning risk questionnaire to predict blood lead elevations in pregnant women. Obstet.
47             Gynecol. 87: 209-212.
48     Stevens, L. A.; Levey, A. S. (2005a) Measurement of kidney function. Med. Clin. N. Am. 89: 457-473.
49     Stevens, L. A.; Levey, A. S. (2005b) Chronic kidney disease in the elderly ~ how to assess risk. N. Engl. J. Med.
50             352:2122-2124.
51     Stewart, W. F.; Schwartz, B. S.; Simon,  D.; Bolla, K. I.; Todd, A. C.; Links, J.  (1999) Neurobehavioral function and
52             tibial and chelatable lead levels in 543 former organolead workers. Neurology 52: 1610-1617.
53     Stewart, W. F.; Schwartz, B. S.; Simon,  D.; Kelsey, K.;  Todd, A. C.  (2002) ApoE genotype, past adult lead
54             exposure, and neurobehavioral  function. Environ. Health Perspect. 110: 501-505.
55     Stiefel, Th.; Schulze, K.; Zorn, H.; Toelg, G. (1980) Toxicokinetic and toxicodynamic studies of beryllium.
56             Arch. Toxicol. 45: 81-92.
       December 2005                                6-400         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Stiles, K. M; Bellinger, D. C. (1993) Neuropsychological correlates of low-level lead exposure in school-age
  2            children: a prospective study. Neurotoxicol. Teratol. 15: 27-35.
  3     Stojek, E.; Skoczynska, A. (2003) Oddzialywanie olowiu na srodblonek naczyniowy [The effect of lead on vascular
  4            endothelium]. Med. Pr. 54: 87-93.
  5     Stokes, L.; Letz, R.; Gerr, F.; Kolczak, M.; McNeill, F. E.; Chettle, D. R.; Kaye, W. E. (1998) Neurotoxicity in
  6            young adults 20 years after childhood exposure to lead: the Bunker Hill experience. Occup. Environ. Med.
  7            55:507-516.
  8     Stollery, B. T. (1996) Reaction time changes in workers exposed to lead. Neurotoxicol. Teratol. 18: 477-483.
  9     Stollery, B. T.; Banks, H. A.; Broadbent, D. E.; Lee, W. R. (1989) Cognitive functioning in lead workers. Br. J. Ind.
10            Med. 46: 698-707.
11     Stollery, B. T.; Broadbent, D. E.; Banks, H. A.; Lee, W. R. (1991) Short term prospective study of cognitive
12            functioning in lead workers. Br. J. Ind. Med. 48: 739-749.
13     Stone, B. M.; Reynolds, C. R. (2003) Can the National Health and Nutrition Examination Survey III (NHANES III)
14            data help resolve the controversy over low blood lead levels and neuropsychological development in
15            children? Arch. Clin. Neuropsychol. 18: 219-244.
16     Stretesky, P. B.; Lynch, M. J. (2001) The relationship between lead exposure and homicide. Arch. Pediatr. Adolesc.
17            Med. 155: 579-582.
18     Succop, P.; Bornschein, R.; Brown, K.; Tseng, C. -Y. (1998) An empirical comparison of lead exposure pathway
19            models. Environ. Health Perspect. Suppl. 106(6): 1577-1583.
20     Sugawara, E.; Nakamura, K.; Miyake, T.; Fukumura, A.; Seki, Y. (1991) Lipid peroxidation and concentration of
21            glutathione in erythrocytes from workers exposed to lead. Br. J. Ind. Med. 48: 239-242.
22     Sun, C. A.; Chang, Y. C.; Liou, S. H.; Yang, G. Y.; Wu, T. N.; Ko, Y. C.; Lee, C. C.; Ho, S. T.; Lai, J. M.; Wu,
23            Y. Q.; Chiang, H. C.; Ko, K. N.; Chang, P.  Y. (1997) Test reliability of blood lead levels in a multicenter
24            epidemiological study: the Taiwan experience. Am. J. Epidemiol. 145(11 suppl.): S48.
25     Sun, L.; Hu, J.; Zhao, Z.; Li, L.; Cheng, H. (2003) Influence of exposure to environmental lead on serum
26            immunoglobulin in preschool children. Environ. Res. 92: 124-128.
27     Susser, M. (1991) What is a cause and how do we know one? A grammar for pragmatic epidemiology. Am. J.
28            Epidemiol. 133: 635-648.
29     Suzen, H. S.; Duydu, Y.; Aydin, A.; Isimer, A; Vural,  N. (2003) Influence of the delta-aminolevulinic acid
3 0            dehydratase (ALAD) polymorphism on biomarkers of lead exposure in Turkish storage battery
31            manufacturing workers. Am. J. Ind. Med. 43: 165-171.
32     Symanski, E.; Hertz-Picciotto, I. (1995) Blood lead levels in relation to menopause, smoking, and pregnancy
33            history. Am. J. Epidemiol. 141: 1047-1058.
34     Syracuse Research Corporation. (2003) Evaluation of the ICRP lead biokinetics model: empirical comparisons with
3 5            observations of plasma-blood lead concentration relationships in humans [draft final]. Washington, DC:
36            U.S. Environmental Protection Agency, Office of Emergency and Remedial Response; contract no. GS-
37            10F-0137K; SRC no. FA332.
38     Tabacova, S.;Balabaeva, L. (1993) Environmental pollutants in relation to complications of pregnancy. Environ.
39            Health Perspect. 101(suppl. 2): 27-31.
40     Tang, N.; Zhu, Z. Q.  (2003) Adverse reproductive effects in female workers of lead battery plants. Int. J. Occup.
41            Med. Environ. Health 16: 359-361.
42     Tang, H.-W.; Liang, Y.-X.; Hu, X.-H.; Yang, H.-G.  (1995) Alterations of monoamine metabolites and
43            neurobehavioral function in lead-exposed workers. Biomed. Environ. Sci.  8: 23-29.
44     Taskinen, H.  (1988) Spontaneous abortions among women occupationally exposed to lead. In: Hogstedt, C.;
45            Reuterwall,  C., eds. Progress in occupational  epidemiology. New York, NY: Elsevier Science Publishers;
46            pp. 197-200.
47     Tassler, P.; Schwartz, B. S.; Coresh, J.; Stewart, W.; Todd, A. (2001) Associations  of tibia lead, DMSA-Chelatable
48            lead, and blood lead with measures of peripheral nervous system function  in former organolead
49            manufacturing workers. Am. J. Ind. Med. 39: 254-261.
50     Taupeau, C.;  Poupon, J.; Treton, D.; Brosse, A.; Richard, Y.; Machelon, V. (2003)  Lead reduces messenger RNA
51            and protein levels of cytochrome P450 aromatase and estrogen receptor "Beta" in human ovarian granulosa
52            cells. Biol. Reprod. 68: 1982-1988.
53     Telisman, S.; Cvitkovic, P.; Jurasovic, J.; Pizent, A.; Favella, M.; Rocic, B. (2000)  Semen quality  and reproductive
54            endocrine function in relation to biomarkers of lead, cadmium, zinc, and copper in men. Environ. Health
55            Perspect. 108: 45-53.
      December 2005                                6-401         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Talisman, S.; Pizent, A.; Jurasovic, I; Cvitkovic, P. (2004) Lead effect on blood pressure in moderately lead-
  2             exposed male workers. Am. J. Ind. Med. 45: 446-454.
  3     Tellez-Rojo, M. M; Hernandez-Avila, M; Gonzalez-Cossio, T.; Romieu, I.; Aro, A.; Palazuelos, E.; Schwartz, J.;
  4             Hu, H. (2002) Impact of reastfeeding on the mobilization of lead from bone. Am. J. Epidemiol.
  5             155:420-428.
  6     Tellez-Rojo, M. M.; Hernandez-Avila, M.; Lamadrid-Figueroa, H.; Smith, D.; Hernandez-Cadena, L.; Mercado, A.;
  7             Aro, A.; Schwartz, J.; Hu, H. (2004) Impact of bone lead and bone resorption on plasma and whole blood
  8             lead levels during pregnancy. Am. J. Epidemiol. 160: 668-678.
  9     Tepper, A.; Mueller,  C.; Singal, M.; Sagar, K. (2001) Blood pressure, left ventricular mass, and lead exposure in
10             battery manufacturing workers. Am. J. Ind. Med. 40: 63-72.
11     TerraGraphics Environmental Engineering, Inc. (2001) Final human health risk assessment for the Coeur d'Alene
12             Basin extending from Harrison to Mullan on the Coeur d'Alene River and tributaries remedial
13             investigation/feasibility study. Washington, DC: U.S. Environmental Protection Agency, prepared for the
14             Idaho Department of Health and Welfare, Idaho Department of Environmental Quality.
15     Teruya, K.; Sakurai, H.; Omae, K.; Higashi, T.; Muto, T.; Kaneko, Y. (1991) Effect of lead on cardiac
16             parasympathetic function. Int. Arch. Occup. Environ. Health 62: 549-553.
17     Thacker, S. B.; Hoffman, D. A.; Smith, I; Steinberg, K.; Zack, M. (1992) Effect of low-level body burdens of lead
18             on the mental development of children: limitations of meta-analysis in a review  of longitudinal data. Arch.
19             Environ. Health 47: 336-346.
20     Thacker, S. B.; Hoffman, D. A.; Smith, J.; Steinberg, K.; Zack, M. (1993) [Untitled author response to letters
21             concerning "Effect of low-level body burdens of lead on the mental development of children:  limitations of
22             meta-analysis in a review of longitudinal data." Arch. Environ. Health 48: 126-127.
23     Theppeang, K.; Schwartz, B. S.; Lee, B.-K.; Lustberg, M. E.; Silbergeld, E. K.; Kelsey, K. T.; Parsons, P. J.; Todd,
24             A. C. (2004) Associations of patella lead with polymorphisms in the vitamin D receptor, "delta"-
25             aminolevulinic acid dehydratase and endothelial nitric oxide synthase genes. J. Occup. Environ. Med.
26             46: 528-537.
27     Thompson, S. G.; Pocock,  S. J. (1992) Can meta-analysis be trusted? Lancet 338:1127-1130.
28     Thomson, G. O. B.; Raab, G. M.; Hepburn, W. S.; Hunter, R.; Fulton, M.; Laxen, D. P. H. (1989) Blood-lead levels
29             and children's behaviour - results from the Edinburgh lead study. J.  Child Psychol. Psychiatr. 30: 515-528.
30     Todd, A. C.; McNeill, F. E.; Palethorpe, J. E.; Peach, D. E.; Chettle, D. R.; Tobin, M. J.;  Strosko, S. J.; Rosen, J. C.
31             (1992) In vivo X-ray fluorescence of lead in bone using K X-ray excitation with 109Cd sources: radiation
32             dosimetry studies. Environ. Res. 57: 117-132.
33     Todd, A. C.; Carroll, S.; Godbold,  J. H.; Moshier, E. L.; Khan, F. A. (2000) Variability in XRF-measured tibia lead
34             levels. Phys. Med. Biol. 45: 3737-3748.
35     Todd, A. C.; Buchanan, R.; Carroll, S.; Moshier, E. L.; Popovac, D.; Slavkovich, V.; Graziano, J. H. (2001) Tibia
36             lead levels and methodological uncertainty in 12-year-old children.  Environ. Res. 86: 60-65.
37     Todd, A. C.; Spencer, C.; Geraghty, C.; Khan, F. A.; Moshier, E. L.; Tang, S.; Parsons, P. J. (2002) L-shell x-ray
38             fluorescence measurements of lead in bone: accuracy and precision. Phys. Med. Biol. 47: 1399-1419.
39     Tollestrup, K.; Daling, J. R.; Allard, J. (1995) Mortality in a cohort of orchard workers exposed to lead arsenate
40             pesticide spray. Arch. Environ. Health 50: 221-229.
41     Tomatis, L. (1990) Cancer: causes, occurrence, and control. Lyon, France: International Agency for Research on
42             Cancer. (IARC scientific publications: v.  100).
43     Tong, S. (1998) Lead exposure and cognitive development: persistence and a dynamic pattern. J. Paediatr. Child
44             Health 34: 114-118.
45     Tong, I.  S.; Lu, Y. (2001) Identification of confounders in the assessment of the relationship between lead exposure
46             and child development. Ann. Epidemiol. 11: 38-45.
47     Tong, S.; Baghurst, P.; McMichael, A.; Sawyer, M.; Mudge, J. (1996) Lifetime exposure to environmental lead and
48             children's intelligence at 11-13 years: the Port Pirie cohort study. Br. Med. J. 312: 1569-1575.
49     Tong, S.; Baghurst, P. A.; Sawyer, M. G.; Burns, J.; McMichael, A. J. (1998) Declining blood lead levels and
50             changes in cognitive function during childhood: the Port Pirie cohort study. JAMA J. Am. Med. Assoc.
51             280:1915-1919.
52     Tong, S.; McMichael, A. J.; Baghurst, P.  A. (2000) Interactions between environmental lead exposure and
53             sociodemographic factors on cognitive development. Arch. Environ. Health 55:  330-335.
54     Torres-Sanchez, L. E.; Berkowitz,  G.; Lopez-Carrillo, L.; Torres-Arreola, L.; Rios, C.; Lopez-Cervantes, M. (1999)
55             Intrauterine lead exposure and preterm birth. Environ. Res. 81: 297-301.
       December 2005                                6-402         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Trope, I.; Lopez-Villegas, D.; Lenkinski, R. E. (1998) Magnetic resonance imaging and spectroscopy of regional
  2            brain structure in a 10-year-old boy with elevated blood lead levels. Pediatrics 101(6): E7.
  3     Trope, I.; Lopez-Villegas, D.; Cecil, K. M; Lenkinski, R. E. (2001) Exposure to lead appears to selectively alter
  4            metabolism of cortical gray matter. Pediatrics 107: 1437-1443.
  5     Tsaih, S. W.; Schwartz, I; Ting Lee, M. L.; Amarasiriwardena, C.; Aro, A.; Sparrow, D.; Hu, H. (1999) The
  6            independent contribution of bone and erythrocyte lead to urinary lead among middle-aged and elderly men:
  7            the normative aging study. Environ. Health Perspect. 107:391-396.
  8     Tsaih, S.-W.; Korrick, S.; Schwartz, I; Lee, M.-L. T.; Amarasiriwardena, C.; Aro, A.; Sparrow, D.; Hu, H. (2001)
  9            Influence of bone resorption on the mobilization of lead from bone among middle-aged and elderly men:
10            the Normative Aging Study. Environ. Health Perspect. 109: 995-999.
11     Tsaih, S.-W.; Korrick, S.; Schwartz, I; Amarasiriwardena, C.; Aro, A.; Sparrow, D; Hu, H. (2004) Lead, diabetes,
12            hypertension, and renal function: the normative aging study. Environ. Health Perspect. 112: 1178-1182.
13     Tuppurainen, M.; Wagar, G.; Kurppa, K.; Sakari, W.; Wambugu, A.; Froseth, B.; Alho, I; Nykyri, E. (1988)
14            Thyroid function as assessed by routine laboratory tests of workers with long-term lead exposure. Scand. J.
15            Work Environ. Health 14: 175-180.
16     Tuthill, R. W. (1996) Hair lead levels related to children's classroom attention-deficit behavior. Arch. Environ.
17            Health 51: 214-220.
18     Tvinnereim, H. M.; Eide, R.; Riise, T.; Wesenberg, G. R.; Fosse, G.; Steinnes, E. (1997) Lead in primary teeth from
19            Norway: changes in lead levels from the 1970s to the 1990s. Sci. Total Environ. 207: 165-177.
20     Tvinnereim, H. M.; Eide, R.; Riise, T. (2000) Heavy metals in human primary teeth: some factors influencing the
21            metal concentrations. Sci. Total Environ. 255: 21-27.
22     U.S. Environmental Protection Agency. (1986a) Air quality criteria for lead. Research Triangle Park, NC: Office of
23            Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
24            EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.
25     U.S. Environmental Protection Agency. (1986b) Lead effects on cardiovascular function, early development, and
26            stature: an addendum to U.S. EPA Air Quality Criteria for Lead (1986). In: Air quality criteria for lead, v.
27            1. Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
28            and Assessment Office; pp. A1-A67; EPA report no. EPA-600/8-83/028aF. Available from: NTIS,
29            Springfield, VA; PB87-142378.
30     U.S. Environmental Protection Agency. (1990) Air quality criteria for lead: supplement to the 1986 addendum.
31            Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
32            Assessment Office; report no. EPA/600/8-89/049F. Available from: NTIS, Springfield, VA; PB91-138420.
33     U.S. Environmental Protection Agency. (2001) Final human health risk assessment for the Coeur d'Alene Basin
34            extending from  Harrison to Mullan on the  Coeur d'Alene River and Tributaries remedial
35            investigation/feasibility study. Washington, DC: U.S. Environmental Protection Agency, Idaho Department
36            of Environmental Quality.
37     U.S. Environmental Protection Agency. (2003) Evaluation of the ICRP Lead Biokinetics Model: Empirical
3 8            Comparisons with Observations of Plasma-Blood Lead Concentration Relationships in Humans. Prepared
39            for U. S. Environmental Protection Agency, Office of Emergency and Remedial Response, Washington,
40            DC. Syracuse Research Corporation under Contract No. GS-10F-0137K, FEDSIM Order No. DABT63-01-
41            F-0133-00.
42     U.S. Environmental Protection Agency. (2004) Air quality criteria for paniculate matter. Research Triangle Park,
43            NC: National Center for Environmental Assessment; report no. EPA/600/P-99/002aF-bF. 2v. Available:
44            http://cfpub.epa.gov/ncea/ [9 November, 2004].
45     U.S. Environmental Protection Agency. (2005) Guidelines for carcinogen risk assessment. Washington, DC: Risk
46            Assessment Forum; report no. EPA/630/P-03/001F. Available: http://cfpub.epa.gov/ncea/index.cfm
47            [30 November, 2005].
48     U.S. Renal Data System. (2004) Outcomes: hospitalization & mortality. In: Annual data report. Minneapolis, MN:
49            USRDS Coordinating Center; pp. 118-138. Available:
50            http://www.usrds.org/2004/pdf/06_hosp_morte_04.pdf [21 November, 2005].
51     Undeger, U.; Basaran, N.; Canpinar, H.; Kansu, E. (1996) Immune alterations in lead-exposed workers. Toxicology
52            109: 167-172.
53     Vacca, C.  V.; Hines, J. D.; Hall, P. W., III. (1986) The proteinuria of industrial lead intoxication. Environ.  Res.
54            41:440-446.
55     Valciukas, J. A.; Lilis, R.; Eisinger, J.; Blumberg, W. E.; Fischbein, A.;  Selikoff, I. J. (1978) Behavioral indicators
56            of leadneurotoxicity: results of a clinical field survey. Int. Arch. Occup. Environ. Health 41: 217-236.
      December 2005                               6-403         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Valentine, J. L.; Baloh, R. W.; Browdy, B. L.; Gonick, H. C.; Brown, C. P.; Spivey, G. H.; Culver, B. D. (1982)
  2             Subclinical effects of chronic increased lead absorption~a prospective study. J. Occup. Med. 24: 120-125.
  3     Valentino, M; Governa, M; Marchiseppe, L; Visona, I. (1991) Effects of lead on polymorphonuclear leukocyte
  4             (PMN) functions in occupationally exposed workers. Arch. Toxicol. 65: 685-688.
  5     VanDe Vyver, F. L.; D'Haese, P. C.; Visser, W. J.; Elseviers, M. M.; Knippenberg, L. J.; Lamberts, L. V.; Wedeen,
  6             R. P.; De Broe, M. E. (1988) Bone lead in dialysis patients. Kidney Int. 33: 601-607.
  7     Van Den Berg, B; Oechsli, F. (1984) Prematurity. In: Bracken, M., ed. Perinatal Epidemiology. New York, NY:
  8             Oxford University Press; pp. 69-85.
  9     Van Den Hoogen, P.; Seidell, J. Nagelkerke, N.; Menotti, A.; Kromhout, D. (2001) Relation between blood pressure
10             and mortality: is there a threshold? Eur. Heart J. 22: 2132-2133.
11     Van Larebeke, N.; Koppen, G.; Nelen, V.; Schoeters, G.; Van Loon H, Albering H, Riga L, Vlietinck, R.; Kleinjans,
12             J.; Flemish Environment and Health Study Group. (2004) Differences in HPRT mutant frequency among
13             middle-aged Flemish women in association with area of residence and blood lead levels. Biomarkers
14             9:71-84.
15     Vaziri, N. D.; Sica, D. A. (2004) Lead-induced hypertension: role of oxidative stress. Curr. Hypertens. Rep.
16             6:314-320.
17     Verberk, M. M.; Willems, T. E. P.; Verplanke, A. J. W.; De Wolff, F. A. (1996) Environmental lead and renal
18             effects in children. Arch. Environ. Health 51: 83-87.
19     Verschoor, M.; Wibowo, A.; Herber, R.; vanHemmen, J.; Zielhuis, R. (1987) Influence of occupational low-level
20             lead exposure on renal parameters. Am. J. Ind. Med. 12: 341-351.
21     Vig, E.; Hu, H. (2000) Lead toxicity in older adults. J. Am. Geriatr. Soc. 48: 1501-1506.
22     Vijayalakshhmi, P.; Serrano, J.; Sparrow, D.; Hu, H. (1999) Relationship of lead in drinking water to bone lead
23             levels twenty years later in Boston men: The Normative Aging Study. J. Occup. Environ. Med. 41: 349-
24             355.
25     Vimpani, G. V; Wigg, N. R.; Robertson, E. F.; McMichael, A. J.; Baghurst, P. A.; Roberts, R. J. (1985) The Port
26             Pirie cohort study: blood lead concentration and childhood developmental assessment. In: Goldwater, L. J.;
27             Wysocki, L. M.; Volpe, R. A., eds. Edited proceedings: Lead environmental health - the current issues;
28             May; Durham, NC. Durham, NC: Duke University; pp. 139-146.
29     Vinceti, M.; Guidetti, D.; Bergomi, M.; Caselgrandi, E.; Vivoli, R.; Olmi, M.; Rinaldi, L.; Rovesti, S.; Solime, F.
30             (1997) Lead, cadmium, and selenium in the blood of patients with sporadic amyotrophic lateral sclerosis.
31             Ital. J. Neurol. Sci. 18: 87-92.
32     Vupputuri, S.; He, J.; Muntner, P.; Bazzano, L. A.; Whelton, P. K.; Batuman, V. (2003) Blood lead level is
33             associated with elevated blood pressure in blacks. Hypertension 41: 463-468.
34     Wagnerova, M.; Wagner, V.; Madlo, Z.; Zavazal, Y.; Wokounova, D.; Kriz, J.; Mohyla, O. (1986) Seasonal
3 5             variations in the level of immunoglobulins and serum proteins of children differing by exposure to air-
36             borne lead. J. Hyg. Epidemiol. Microbiol. Immunol. 30(2): 127-138.
3 7     Walkowiak, J.; Altmann, L.; Kramer, U.; Sveinsson, K.; Turfeld, M.; Weishoff-Houben, M.; Winneke,  G. (1998)
3 8             Cognitive and sensorimotor functions in 6-year-old children in relation to lead and mercury levels:
39             adjustment for intelligence and contrast sensitivity in computerized testing. Neurotoxicol. Teratol.
40             20:511-521.
41     Wang, S. T.; Pizzolato, S.; Demshar, H. P.; Smith, L. F. (1997) Decline in blood lead in Ontario children correlated
42             to decreasing consumption of leaded gasoline, 1983-1992 [letter]. Clin. Chem. 43: 1251-1252.
43     Wang, C.-L.; Chuang, H.-Y.; Ho, C.-K.; Yang, C.-Y.; Tsai, J.-L.; Wu, T.-S.; Wu, T.-N. (2002) Relationship between
44             blood lead concentrations and learning achievement among primary school children in Taiwan. Environ.
45             Res. 89:  12-18.
46     Wang, V.-S.; Lee, M.-T.; Chiou, J.-Y.; Guu, C.-F.; Wu, C.-C; Wu, T.-N.; Lai, J.-S. (2002) Relationship between
47             blood lead levels and renal function in lead battery workers. Int. Arch. Occup. Environ. Health
48             75: 569-575.
49     Ward, N. L; Watson, R.; Bryce-Smith, D. (1987) Placenta! element levels in relation to fetal development for
50             obstetrically 'normal' births: a study of 37 elements. Evidence for effects of cadmium, lead and zinc on fetal
51             growth, and for smoking as a source of cadmium. Int. J. Biosoc. Res. 9: 63-81.
52     Wasserman, G. A. (1995) Effects of early lead exposure: time to integrate and broaden our efforts. Neurotoxicol.
53             Teratol.  17: 243-244.
54     Wasserman, G. A.; Factor-Litvak, P.  (2001) Methodology, inference and causation:  environmental lead exposure
55             and childhood intelligence. Arch. Clin. Neurospychol. 16: 343-352.
       December 2005                                6-404        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Wasserman, G.; Graziano, J. H.; Factor-Litvak, R.; Popovac, D.; Morina, N.; Musabegovic, A.; Vrenezi, N.;
  2             Capuni-Paracka, S.; Lekic, V.; Preteni-Redjepi, E.; Hadzialjevic, S.; Slavkovich, V.; Kline, J.; Shrout, P.;
  3             Stein, Z. (1992) Independent effects of lead exposure and iron deficiency anemia on developmental
  4             outcome at age 2 years. J. Pediatr. 121: 695-703.
  5     Wasserman, G. A.; Graziano, J. H.; Factor-Litvak, P.; Popovac, D.; Morina, N.; Musabegovic, A.; Vrenezi, N.;
  6             Capuni-Paracka, S.; Lekic, V.; Preteni-Redjepi, E.; Hadzialjevic, S.; Slavkovich, V.; Kline, J.; Shrout, P.;
  7             Stein, Z. (1994) Consequences of lead exposure and iron supplementation on childhood development at age
  8             4 years. Neurotoxicol. Teratol. 16: 233-240.
  9     Wasserman, G. A.; Liu, X.; Lolacono, N. J.; Factor-Litvak, P.; Kline, J. K.; Popovac, D.; Morina, N.; Musabegovic,
10             A.; Vrenezi, N.; Capuni-Paracka, S.; Lekic, V.; Preteni-Redjepi, E.; Hadzialjevic, S.; Slavkovich, V.;
11             Graziano, J. H. (1997) Lead exposure and intelligence in 7-year-old children: the Yugoslavia prospective
12             study. Environ. Health Perspect. 105: 956-962.
13     Wasserman, G. A.; Graziano, J.; Factor-Litvak, P. (1998) Lead effects research [response]. Am. J. Pub. Health
14             88: 1879-1880.
15     Wasserman, G. A.; Staghezza-Jaramillo, B.; Shrout, P.; Popovac, D.; Graziano, J. (1998) The effect of lead exposure
16             on behavior problems in preschool children. Am. J. Pub. Health 88 (3): 481-486.
17     Wasserman, G. A.; Musabegovic, A.; Liu, X.; Kline, J.; Factor-Litvak, P.; Graziano, J. H. (2000a) Lead exposure
18             and motor functioning in 4 1/2-year-old children: the Yugoslavia prospective study. J. Pediatr.
19             137:555-561.
20     Wasserman, G. A.; Liu, X.; Popovac, D.; Factor-Litvak, P.; Kline, J.; Waternaux, C.; Lolacono, N.; Graziano, J. H.
21             (2000b) The Yugoslavia prospective lead industry study: contributions of prenatal and postnatal lead
22             exposure to early  intelligence. Neurotoxicol. Teratol. 22: 811-818.
23     Wasserman, G. A.; Liu, X.; Pine, D.  S.; Graziano, J. H. (2001) Contribution of maternal smoking during pregnancy
24             and lead exposure to early child behavior problems. Neurotoxicol. Teratol. 23: 13-21.
25     Wasserman, G. A.; Factor-Litvak, P.; Liu, X.; Todd, A. C.; Kline, J. K.; Slavkovich, V.; Popovac, D.; Graziano, J.
26             H. (2003) The relationship between blood lead, bone lead and child intelligence. Child Neuropsychol.
27             9:22-34.
28     Watt, G. C. M.; Britton, A.; Gilmore, W. H.; Moore, M. R.; Murray, G. D.; Robertson, S. J.; Womersley, J. (1996)
29             Is lead in tap water still a public health problem?  An observational study in Glasgow. Br. Med. J.
30             313:979-981.
31     Weaver, V M.; Lee, B.-K.; Ahn, K.-D.; Lee, G.-S.; Todd, A. C.; Stewart, W. F.; Wen, J.;  Simon, D. J.; Parsons,
32             P. J.; Schwartz, B. S. (2003a) Associations of lead biomarkers with renal function in Korean lead workers.
33             Occup. Environ. Med. 60: 551-562.
34     Weaver, V M.; Schwartz,  B. S.; Ahn, K.-D.; Stewart, W. F.; Kelsey, K. T.; Todd, A. C.; Wen, J.; Simon, D. J.;
35             Lustberg, M. E.; Parsons, P. J.; Silbergeld, E. K.; Lee, B.-K. (2003b) Associations of renal function with
36             polymorphisms in the "delta"-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide
37             synthase genes in Korean lead workers. Environ.  Health Perspect. Ill: 1613-1619.
38     Weaver, V M.; Jarr, B. G.; Schwartz, B. S.; Todd, A. C.; Ahn, K.-D.; Lee, S.-S.; Wen, J.; Parsons, P. J.; Lee, B.-K.
3 9             (2005a) Associations among lead dose biomarkers, uric acid, and renal function in Korean lead workers.
40             Environ. Health Perspect. 113: 36-42.
41     Weaver, V M.; Lee, B.-K.; Todd, A. C.; Jaar, B. G.; Ahn, K.-D.; Wen, J.; Shi, W.; Parsons, P. J.; Schwartz, B. S.
42             (2005b) Associations of patella lead and other lead biomarkers with renal function in lead workers. J.
43             Occup. Environ. Med. 47: 235-243.
44     Weaver, V M.; Schwartz,  B. S.; Jaar, B. G.; Ahn, K.-D.; Todd, A. C.; Lee, S.-S.; Kelsey, K. T.; Silbergeld, E. K.;
45             Lustberg, M. E.; Parsons, P. J.; Wen, J.; Lee-B.-K. (2005c) Associations of uric acid with polymorphisms
46             in the "delta"-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide synthase genes in
47             Korean lead workers. Environ. Health Perspect. 113: 1509-1515.
48     Webber, C. E.; Chettle, D. R.; Bowins, R. J.; Beaumont, L. F.; Gordon, C. L.; Song, X.; Blake, J. M.; McNutt, R.  H.
49             (1995) Hormone replacement therapy may reduce the return of endogenous lead from bone to the
50             circulation. Environ. Health Perspect. 103: 1150-1153.
51     Wedeen, R. P. (1992) Removing lead from bone: clinical implications of bone lead stores. Neurotoxicology
52             13:843-852.
53     Wedeen, R. P.; Maesaka, J. K.; Weiner, B.; Lipat, G. A.; Lyons, M. M.; Vitale, L. F.; Joselow, M. M. (1975)
54             Occupational lead nephropathy. Am. J. Med. 59:  630-641.
55     Wedeen, R. P.; Mallik, D. K.; Batuman, V. (1979) Detection and treatment of occupational lead nephropathy. Arch.
56             Intern. Med. 139: 53-57.
       December 2005                                6-405         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Wedeen, R. P.; Batuman, V.; Landy, E. (1983) The safety of the EDTA lead-mobilization test. Environ. Res.
  2             30:58-62.
  3     Weinberg, C. R.; Baird, D. D.; Rowland, A. S. (1993) Pitfalls inherent in retrospective time-to-event studies:
  4             the example of time to pregnancy. Stat. Med. 12: 867-879.
  5     Weinberg, C. R.; Baird, D. D.; Wilcox, A. J. (1994) Sources of bias in studies of time to pregnancy. Stat. Med.
  6             13:671-681.
  7     Weiss, B. (1988) Neurobehavioral toxicity as a basis for risk assessment. Trends Pharmacol. Sci. 9: 59-62.
  8     Weiss, B. (1990) Risk assessment: the insidious nature of neurotoxicity and the aging brain. Neurotoxicology
  9             11:305-314.
10     Weiss, B. (2000) Vulnerability of children and the developing brain to neurotoxic hazards. Environ. Health Perspect.
11             Suppl. 108(3): 375-381.
12     Weiss, S. T.; Munoz, A.; Stein, A.; Sparrow, D.; Speizer, F. E. (1986) The relationship of blood lead to blood
13             pressure in a longitudinal study of working men. Am. J. Epidemiol.  123: 800-808.
14     Weisskopf, M. G.; Wright, R. O.; Schwartz, J.; Spiro, A., Ill; Sparrow, D.; Aro, A.; Hu, H. (2004) Cumulative lead
15             exposure and prospective change in cognition among elderly men. The VA Normative Aging Study. Am. J.
16             Epidemiol. 160: 1184-1193.
17     West, W. L.; Knight, E. M.; Edwards, C. H.; Manning, M.; Spurlock, B.; James, H.; Johnson, A. A.; Oyemade,
18             U. J.; Cole, O. J.; Westney, O. E.; Laryea, H.; Jones, S.; Westney, L. S. (1994) Maternal low level lead and
19             pregnancy outcomes. J. Nutr. 124(suppl.): 981S-986S.
20     Wetmur, J. G.; Lehnert, G.; Desnick, R. J. (1991) The "delta"-aminolevulinate dehydratase polymorphism: higher
21             blood lead levels in lead workers and environmentally exposed children with the 1-2 and 2-2 isozymes.
22             Environ. Res. 56:  109-119.
23     Weyermann, M.; Brenner, H. (1998) Factors affecting bone demineralization and blood lead levels of
24             postmenopausal women~a population-based study from Germany. Environ. Res.  76:  19-25.
25     Wibberley, D. G.; Khera, A. K.; Edwards, J. H.; Rushton, D. I. (1977) Lead levels in human placentae from normal
26             and malformed births. J. Med. Genet. 14: 339-345.
27     Wigg, N. R. (2001) Low-level lead exposure and children. J. Paediatr. Child Health 37: 423-425.
28     Wigg, N. R.; Vimpani, G. V.; McMichael, A. J.; Baghurst, P. A.; Robertson, E. F.; Roberts, R. J. (1988) Port Pirie
29             cohort study: childhood blood lead and neuropsychological development at age two years. J. Epidemiol.
30             Community Health 42: 213-219.
31     Wildt, K.; Berlin, M.; Isberg, P. E. (1987) Monitoring of zinc protoporphyrin levels in blood following occupational
32             lead exposure. Am. J. Ind. Med. 12: 385-398.
33     Wilhelm, M.; Lombeck, I.; Hafner, D.; Ohnesorge, F. K. (1989) Hair lead levels in young children from the F.R.G.
34             J. Trace Elem. Electrolytes Health Dis. 3: 165-170.
35     Wilhelm, M.; Pesch, A.;  Rostek, U.; Begerow, J.; Schmitz, N.; Idel, H.; Ranft, U. (2002) Concentrations of lead in
3 6             blood, hair and saliva of German children living in three different areas of traffic density. Sci. Total
37             Environ. 297: 109-118.
38     Wingren, G.; Axelson, O. (1985) Mortality pattern in a glass producing area in SE Sweden. Br. J. Ind. Med.
39             42:411-414.
40     Wingren, G.; Axelson, O. (1987) Mortality in the Swedish glassworks industry. Scand. J. Work Environ. Health
41             13:412-416.
42     Wingren, G.; Axelson, O. (1993) Epidemiologic studies of occupational cancer as related to complex mixtures of
43             trace elements in the art glass industry. Scand. J. Work Environ. Health 19(suppl.  1): 95-100.
44     Wingren, G. Englander, V. (1990) Mortality and cancer morbidity in a cohort of Swedish glassworkers. Int. Arch.
45             Occup. Environ. Health 62: 253-257.
46     Winkelstein, W.; Balfour, J. L. (1996) [Untitled letter concerning bone lead levels and antisocial and delinquent
47             behavior]. JAMA J. Am. Med. Assoc. 275: 1727-1728.
48     Winneke, G. (1992) Cross species extrapolation in neurotoxicology: neurophysiological and neurobehavioral
49             aspects. Neurotoxicology 13: 15-25.
50     Winneke, G. (1995) Lead and child development: uncertainties, possibilities, and explanations. Neurotoxicol.
51             Teratol. 17: 245-247.
52     Winneke, G. (1995) Endpoints of developmental neurotoxicity in environmentally exposed children. Toxicol.  Lett.
53             77: 127-136.
54     Winneke, G.; Kraemer, U. (1984) Neuropsychological effects of lead in children: interactions with social
55             background variables. Neuropsychobiology 11: 195-202.
       December 2005                                6-406        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Winneke, G.; Kramer, U. (1997) Neurobehavioral aspects of lead neurotoxicity in children. Cent. Eur. J. Public
  2             Health 5: 65-69.
  3     Winneke, G.; Collet, W.; Lilienthal, H. (1988) The effects of lead in laboratory animals and environmentally-
  4             exposed children. Toxicology 49: 291-298.
  5     Winneke, G.; Brockhaus, A.; Collet, W.; Kramer, U. (1989) Modulation of lead-induced performance deficit in
  6             children by varying signal rate in a serial choice reaction task. Neurotoxicol. Teratol. 11:  587-592.
  7     Winneke, G.; Brockhaus, A.; Ewers, U.; Kramer, U.; Neuf,  M. (1990) Results from the European multicenter study
  8             on lead neurotoxicity in children: implications for risk assessment. Neurotoxicol. Teratol. 12: 553-559.
  9     Winneke, G.; Altmann, L.; Kramer, U.; Turfeld, M.; Behler, R.; Gutsmuths, F. J.; Mangold, M. (1994)
10             Neurobehavioral and neurophysiological observations in six year old children with low lead levels in East
11             and West Germany. Neurotoxicology 15: 705-713.
12     Winterberg, B.; Korte, R.; Bertram, H. P. (1991) Response: bone lead is elevated in renal failure [letter]. Nephron
13             58:496-497.
14     Wittmers, L. E.; Aufderheide, A. C.; Wallgren, J.; Rapp, G.; Alich, A. (1988) Lead in bone. IV. Distribution of lead
15             in the human skeleton. Arch. Environ. Health 43:381-391.
16     Wolf, A. W.; Ernhart, C. B.; White, C. S. (1985) Intrauterine lead exposure and early development. In: Lekkas,
17             T. D., ed. International conference: heavy metals in the environment, v. 2; September; Athens, Greece.
18             Edinburgh, United Kingdom: CEP Consultants, Ltd.; pp. 153-155.
19     Wolf, C.; Wallnofer, A.; Waldhoer, T.; Vutuc, C.; Meisinger, V.; Rudiger, H. W. (1995) Effect of lead on blood
20             pressure in occupationally nonexposed men. Am. J. Ind. Med. 27: 897-903.
21     Wong, O.; Harris, F. (2000) Cancer mortality study of employees at lead battery plants and lead smelters,
22             1947-1955. Am. J. Ind. Med. 38: 255-270.
23     World Health Organization. (1977) Lead. Geneva, Switzerland: World Health Organization. (Environmental health
24             criteria: v.3). Available: http://www.inchem.org/documents/ehc/ehc/ehc003.htm [11 March, 2005].
25     World Health Organization. (1992) International statistical classification of diseases  and related health problems:
26             tenth revision (ICD-10). World Health Organization, Geneva, pp. 369-370.
27     World Health Organization. (1995) Inorganic lead. Geneva, Switzerland: World Health Organization, International
28             Programme on Chemical  Safety. (Environmental health criteria 165).
29     Wright, R. O.; Hu, H.; Silverman, E. K.; Tsaih, S. W.; Schwartz, J.; Bellinger,  D.; Palazuelos, E.; Weiss, S. T.;
30             Hernandez-Avila, M. (2003)  Apolipoprotein E genotype predicts 24-month bayley scales infant
31             development score. Pediatr. Res. 54: 819-825.
32     Wright, R. O.; Silverman, E. K.; Schwartz, J.; Tsaih, S.-W.; Senter, J.;  Sparrow, D.; Weiss, S. T.; Aro, A.; Hu, H.
3 3             (2004) Association between hemochromatosis genotype and lead exposure among elderly men: the
34             Normative Aging Study. Environ. Health Perspect. 112: 746-750.
35     Wu, M. T.; Kelsey,  K.; Schwartz, J.; Sparrow, D.; Weiss, S.; Hu, H. (2003) A "delta"-aminolevulinic acid
3 6             dehydratase (ALAD) polymorphism may modify the relationship of low-level lead exposure to uricemia
3 7             and renal function: the normative aging study. Environ. Health Perspect. 111:335-341.
38     Wu, T.-N.; Shen, C.-Y.; Ko, K.-N.; Guu, C.-F.; Gau, H.-J.; Lai, J.-S.; Chen, C.-J.; Chang, P.-Y. (1996) Occupational
39             lead exposure and blood pressure. Int. J. Epidemiol. 25: 791-796.
40     Wu, T.; Buck, G. M.; Mendola, P.  (2003) Blood lead levels and sexual maturation in U.S.  girls: the Third National
41             Health and Nutrition Examination Survey, 1988-1994. Environ. Health Perspect.  Ill: 737-741.
42     Yagoda, B.; Miller,  S. A. (1987) Lead in calcium supplements  [letter and answer]. JAMA J. Am. Med Assoc.
43             257: 1810.
44     Yankel, A. J.; Von Lindern, I. H.; Walter, S. D. (1977) The Silver Valley lead study: the relationship between
45             childhood blood lead levels and environmental exposure. J. Air Pollut. Control Assoc. 27: 763-767.
46     Yassin, A. S.; Martonik, J. F.; Davidson, J. L. (2004) Blood lead levels in U.S. workers, 1988-1994.  J. Occup.
47             Environ. Med. 46: 720-728.
48     Ye, X. B.; Wu, C. E.; Fu, H.; Yang, S.-L.; Lu, Y.-W.; Ni, W.-M. (2003) Associations of blood lead levels, kidney
49             function, and blood pressure  with "delta"-aminolevulinic acid dehydratase and vitamin D receptor gene
50             polymorphisms. Toxicol.  Mech. Methods 13: 139-146.
51     Yokoyama, K.; Araki, S.; Murata,  K.; Morita, Y.; Katsuno, N.; Tanigawa, T.; Mori, N.; Yokota, J.; Ito, A.;
52             Sakata, E. (1997) Subclinical vestibulo-cerebellar,  anterior cerebellar  lobe and spinocerebellar effects in
53             lead workers in relation to concurrent and past exposure. Neurotoxicology 18: 371-380.
54     Yokoyama, K.; Araki, S.; Aono, H.; Murata, K. (1998) Calcium disodium ethylenediaminetetraacetate-chelated lead
55             as a predictor for subclinical  lead neurotoxicity: follow-up study on gun-metal foundry workers. Int. Arch.
56             Occup. Environ. Health 71: 459-464.
       December 2005                                6-407         DRAFT-DO NOT QUOTE OR CITE

-------
  1    Yokoyama, K.; Araki, S.; Yamashita, K.; Murata, K.; Nomiyama, K.; Nomiyama, H.; Tao, Y.-X.; Liu, S.-J. (2002)
  2            Subclinical cerebellar anterior lobe, vestibulocerebellar and spinocerebellar afferent effects in young
  3            female lead workers in China: computerized posturography with sway frequency analysis and brainstem
  4            auditory evoked potentials. Ind. Health 40: 245-253.
  5    Younes, B.; Al-Meshari, A. A.; Al-Hakeem, A.; Al-Saleh, S.; Al-Zamel, F.; Al-Shammari, F.; Alwarthan, A. (1995)
  6            Lead concentration in breast milk of nursing mothers living in Riyadh. Ann. SaudiMed. 15:249-251.
  7    Young, S. S.; Hawkins, D. M. (1988) Using recursive partitioning to analyze a large SAR data set. Struct.-Act.
  8            Relat. Quant. Struct.-Act. Relat. 8: 183-193.
  9    Young, B. A.; Boyko, E. I; Ross, H. I; Fihn, S.; Bryson, C. L. (2004) Association of urine cadmium with
10            hypertension, microalbuminuria and reduced renal function: results from the NHANES III study [abstract].
11            J. Am. Soc. Nephrol. 15: 146A.
12    Yu, C.-C.; Lin, J.-L.; Lin-Tan, D.-T. (2004) Environmental exposure to lead and progression of chronic renal
13            diseases: a four-year prospective longitudinal study. J. Am. Soc. Nephrol. 15: 1016-1022.
14    Yule, W.; Lansdown, R.; Millar, I. B.; Urbanowicz, M.-A. (1981) The relationship between blood lead
15            concentrations, intelligence and attainment in a school population: a pilot study. Dev. Med. Child Neurol.
16            23: 567-576.
17    Yule, W.; Urbanowicz, M.-A.; Lansdown, R.; Millar, I. B. (1984) Teachers' ratings of children's behaviour in
18            relation to blood lead levels. Br. J. Dev. Psychol. 2: 295-305.
19    Yucesoy, B.; Turhan, A.; Ure, M.; Imir, T.; Karakaya, A. (1997a) Effects of occupational lead and cadmium
20            exposure on some immunoregulatory cytokine levels in man.  Toxicology 123: 143-147.
21    Yucesoy, B.; Turhan, A.; Ure, M.; Imir, T.; Karakaya, A. (1997b) Simultaneous effects of lead and cadmium onNK
22            cell activity and some phenotypic parameters. Immunopharmacol. Immunotoxicol. 19: 339-348.
23    Zentner, L. E. A.; Rondo, P. H. C. (2004)  Lead contamination among pregnant Brazilian women living near a lead
24            smelter. Int. J. Gynecol. Obstet. 87: 147-148.
25    Zhang, Z.-W.; Shimbo, S.; Ochi, N.; Eguchi, M.; Watanabe, T.; Moon, C.-S.; Ikeda, M. (1997) Determination of
26            lead and cadmium in food and blood inductively coupled plasma mass spectrometry: a comparison with
27            graphite furnace atomic adsorption spectrometry. Sci. Total Environ. 205: 179-187.
28    Zhao, Z. Y.; Li, R.; Sun, L.; Li, Z. Y.; Yang, R. L. (2004) Effect of lead exposure on the immune function of
29            lymphocytes and erythrocytes. in preschool children. J. Zhejiang Univ. Sci. 5(8): 1001-1004.
30    Zheng, W.; Lu, Y. M.; Lu, G. Y.; Zhao, Q.; Cheung, 0.; Blaner, W. S. (2001) Transthyretin, thyroxine, and retinol-
31            binding protein in human cerebrospinal fluid: effect of lead exposure. Toxicol. Sci. 61(1):  107-114.
32    Zimmermann-Tansella, C.; Campara, P.; Andrea, F. D.; Savontto, C.; Tansella, m. (1983) Psychological and
33            physical complaints of subjects with low exposure to lead. Hum. Toxicol. 2: 615-623.
34    Zuckerman, B.; Amaro, H.; Cabral, H. (1989) Validity of serf-reporting of marijuana and cocaine use among
35            pregnant adolescents. J. Pediatr. ( St. Louis) 115: 812-815.
36
      December 2005                                6-408         DRAFT-DO NOT QUOTE OR CITE

-------
 i               8. ENVIRONMENTAL EFFECTS OF LEAD
 2
 3
 4    8.1    TERRESTRIAL ECOSYSTEMS
 5    8.1.1   Introduction
 6          Surface soils across the United States are enriched in lead (Pb) relative to levels expected
 7    from natural (geogenic) inputs (Erel and Patterson, 1994; Francek, 1992; Friedland et al., 1984;
 8    Marsh and Siccama, 1997; Murray et al., 2004; Yanai et al., 2004). While some of this
 9    contaminant Pb is attributed to paint, salvage yards, and the use of Pb arsenate as a pesticide in
10    localized areas (Francek, 1997), Pb contamination of surface soils is essentially ubiquitous
11    because of atmospheric pollution associated with waste incineration, the combustion of fossil
12    fuels, and metal smelting and production (Newhook et al., 2003; Polissar et al., 2001). However,
13    lead inputs to terrestrial ecosystems in the United States have declined dramatically in the past
14    30 years.  The primary reason for this decline has been the almost complete elimination of alkyl-
15    lead additives in gasoline in North America. Also, emissions from smelters have declined as
16    older plants have been shut down or fitted with improved emissions controls.
17          Lead released from forest floor soils in the past has been largely immobilized in mineral
18    soils (Miller and Friedland, 1994; Johnson et al., 1995b, 2004; Johnson and Petras, 1998;
19    Watmough et al., 2004). The amount of Pb that has leached into the mineral soil appears to be
20    on the order of 20 to 50% of the total anthropogenic Pb deposition.
21          Most terrestrial ecosystems in North America remain sinks for lead,  despite reductions in
22    atmospheric Pb deposition of more than 95%.  Although inputs of Pb to ecosystems are currently
23    low, Pb export from watersheds via groundwater and streams is substantially lower. Reported
24    concentrations of Pb in waters draining natural terrestrial ecosystems have always been low
25    (Bacon and Bain, 1995; Johnson et al., 1995b; Wang et al., 1995; Vinogradoff et al., 2005),
26    generally less than 1 ng L"1, even at moderately polluted sites (Laskowski et al., 1995).
27    Therefore, even at current input levels, watersheds are accumulating industrial Pb.
28          The current chapter summarizes the most relevant information from  the 1986 Air Quality
29    Criteria Document (AQCD) and reviews new information that has become  available on the
30    potential effects of atmospheric lead inputs on the terrestrial ecosystem. It has been organized to
31    address: methodologies used in terrestrial ecosystem research (Section 8.1.2); the distribution of

      December 2005                        8-1           DRAFT-DO NOT QUOTE OR CITE

-------
 1    atmospherically delivered lead in terrestrial ecosystems (Section 8.1.3); lead uptake and
 2    mechanisms of action (Section 8.1.4); toxic effects of lead on terrestrial organisms (Section
 3    8.1.5); and, lead effects on natural terrestrial ecosystems (Section 8.1.6).  The major conclusions
 4    and recommendations from each of these subject areas are summarized here.
 5
 6    8.1.1.1  Methodologies in Terrestrial Ecosystem Research
 7          Several methodologies used in terrestrial ecosystems research are described in Sections
 8    8.1.2 and 8.1.3. One of the key factors necessary for understanding ecological risks is related to
 9    bioavailability. The National Research Council (NRC) 2002 review on bioavailability defined
10    the "bioavailability processes" in terms of three key processes.  One of these processes,
11    contaminant interactions between phases, is more commonly referred to as "speciation." For a
12    given metal or metalloid, the term speciation describes the chemical's ability to interact with its
13    biological or chemical surroundings by characterizing its physicochemical properties that are
14    relevant to bioavailability.
15          Methods to address bioavailability (speciation), and methods used to reduce Pb
16    bioavailability, are summarized in this section.
17
18    Analytical Tools and Models
19          A wide variety of analytical tools have been used to characterize a metal's speciation as it
20    is found in various media:
21
22       •  XRD - X-ray diffraction;
23       •  EPMA - electron probe microanalysis;
24       •  PIXE and //PIXE - particle induced X-ray emission;
25       •  XPS - X-ray photoelectron spectroscopy;
26       •  XAS - X-ray absorption spectroscopy;
27       •  SIMS - secondary ion mass spectrometry;
28       •  sequential extractions; and,
29       •  single chemical extractions.
30
31          EMPA techniques provide the greatest information on metal speciation.  Other techniques,
32    such as EXAFS (extended X-ray absorption fine structure) and EXANES (extended X-ray
      December 2005                         8-2           DRAFT-DO NOT QUOTE OR CITE

-------
 1    absorption near edge spectroscopy), show great promise and will be important in solving key
 2    mechanistic questions.
 3          The tools that have been used most often to evaluate speciation for metal particles in
 4    solution include the following computer-based models: MINTEQL, REDEQL2, ECOSAT,
 5    MINTEQA2, HYDRAQL, PHREEQE, and WATEQ4F.
 6
 7    Metal Speciation for Plants
 8          When considering the bioavailability of a metal to plants from  soils and sediments, it is
 9    generally assumed that both the kinetic rate of supply and the speciation of the metal to either the
10    root or shoot are highly important. In soils and sediments, generally only a small volume of
11    water is in contact with the chemical form, and although the proportion of a metal's
12    concentration in this pore water to the bulk soil/sediment concentration is small, it is this phase
13    that is directly available to plants. Therefore, pore water chemistry (i.e., metal concentration as
14    simple inorganic species, organic complexes, or colloid complexes) is most important.
15          Tools currently used for metal speciation for plants include (1) in situ measurements using
16    selective electrodes (Gundersen et al., 1992; Archer et al., 1989; Wehrli et al., 1994); (2) in situ
17    collection techniques using diffusive equilibrium thin films (DET) and diffusive gradient thin
18    films (DGT) followed by laboratory analyses (Davison et al., 1991, 1994; Davison and Zhang,
19    1994; Zhang et al., 1995); and (3) equilibrium models (SOILCHEM) (Sposito and Coves, 1988).
20
21    Influence of Soil Amendments on Bioavailability
22          The removal of contaminated soil to mitigate exposure of terrestrial ecosystem
23    components to Pb can often present both economic and logistic problems. Because of this,
24    recent studies have focused on in situ methodologies to lower soil-Pb  RBA (Brown et al.,
25    2003a,b). To date, the most common methods studied include the addition of soil amendments
26    in an effort to either lower the solubility of the Pb form or to provide sorption sites for fixation of
27    pore-water Pb.  These amendments typically fall within the categories of phosphate, biosolid,
28    and Al/Fe/Mn-oxide amendments.
29          Phosphate amendments have been studied extensively and, in some cases, offer the most
30    promising results (Brown et al., 1999; Ryan et al., 2001; Cotter-Howells and Caporn, 1996;
31    Hettiarachchi et al., 2001, 2003; Yang et al., 2001; Ma et al., 1995). A number of potentially

      December 2005                        8-3          DRAFT-DO NOT QUOTE OR CITE

-------
 1    significant problems associated with phosphate amendments have been recognized, including
 2    both phyto- and earthworm toxicity (Ownby et al., 2005; Cao et al., 2002; and Rusek and
 3    Marshall, 2000). Both of these toxicities are primarily associated with very high applications of
 4    phosphorous and/or decreased soil pH.  Indications of phytotoxicity are often balanced by studies
 5    such as Zhu et al. (2004) that illustrate a 50 to 70% reduction in shoot-root uptake of Pb in
 6    phosphate-amended soils. Additionally, the added phosphate poses the potential risk of
 7    eutrophication of nearby waterways from soil  runoff.
 8          Biosolids have been used historically in the restoration of coal mines (Haering et al.,
 9    2000;  Sopper, 1993).  More recently, workers have demonstrated the  feasibility of their use in
10    the restoration of mine tailings (Brown  et al., 2000), and urban soils (Brown et al., 2003a; Farfel
11    et al., 2005). As with phosphate amendments, problems with biosolid application have also been
12    documented.  Studies have shown that metal transport is significantly accelerated in soils
13    amended with biosolids (Al-Wabel et al., 2002; McBride et al., 1999, 1997; Lamy et al., 1993;
14    Richards et al., 1998, 2000).
15
16    8.1.1.2  Distribution of Atmospherically Delivered Lead in Terrestrial Ecosystems
17          Advances in technology since  the 1986 AQCD have allowed for a more quantitative
18    determination of the mobility, distribution, uptake, and fluxes of atmospherically-delivered Pb in
19    terrestrial ecosystems.
20
21    Lead Speciation in Solid Phases
22          Selective chemical extractions have been employed extensively for quantifying amounts
23    of a particular metal phase (e.g., PbS, Pb-humate, Pb-Fe, Mn  oxide) present in soil rather than
24    total metal concentration. Selective extractions can be a relatively rapid, simple, and inexpensive
25    means for determining metal phases in soils, and the generated data can be linked to potential
26    mobility and bioavailability of the metal (Tessier and  Campbell,  1987). However, some
27    problems persist with the selective extraction technique.  First, extractions are rarely specific to  a
28    single phase.  For example, while peroxide (H2O2) is often used to remove metals bound in
29    organic matter in soils, some researchers have demonstrated that this  reagent destroys clay
30    minerals and sulfides (Ryan et al., 2002). Peroxide solutions may also be inefficient at removing
31    metals bound to humic acids, and in fact could potentially result in the precipitation of metal-

      December 2005                         8-4          DRAFT-DO NOT QUOTE OR CITE

-------
 1    humate substances. In addition to non-selectivity of reagents, significant metal redistribution has
 2    been documented during sequential chemical extractions (Ho and Evans, 2000), and many
 3    reagents may not extract targeted phases completely.  Therefore, while chemical extractions do
 4    provide some useful information on metal phases in soil, the results should be treated as
 5    "operationally defined," e.g., "H2O2 liberated-Pb" rather than "Organic Pb."
 6          Synchrotron radiation (X-rays) allows researchers to probe the electron configuration of
 7    metals in untreated soil samples. Since different elements have different electron binding
 8    energies, X-rays can be focused in an energy window specific to a metal of interest.  The precise
 9    energy required to dislodge a core electron from a metal will be a function of the oxidation state
10    and covalency of the metal.  Since the electron configuration  of a lead atom will be directly
11    governed by its speciation (e.g., Pb bound to organics, Pb adsorbed to oxide surfaces, PbS, etc.),
12    X-ray absorption experiments are a powerful in-situ technique for determining speciation that
13    does not suffer from some of the problems of chemical extractions (Bargar et al., 1997a; Bargar
14    et al., 1997b; Bargar et al., 1998).
15          Selective chemical extractions and synchrotron-based X-ray studies have shown that
16    industrial Pb can be strongly sequestered by organic matter and secondary minerals such as clays
17    and oxides of Al, Fe, and Mn (Miller and McFee, 1983; Jersak et al., 1997; Johnson and Petras,
18    1998; Kaste et al., 2005). More recent X-ray studies have demonstrated the importance of
19    biomineralization of Pb in soils by bacteria and nematodes (Jackson et al., 2005; Templeton
20    et al., 2003a,b; Xia et al., 1997).
21
22    Lead Solid-Solution Partitioning
23          The concentration of Pb  species dissolved in soil solution is probably controlled by some
24    combination of a) Pb mineral solubility equilibria, b) adsorption reactions of dissolved Pb phases
25    on inorganic surfaces (e.g., oxides  of Al, Fe, Si, Mn, etc., clay minerals), and c)  adsorption
26    reactions of dissolved Pb phases on soil organic matter.  Dissolved Pb phases in soil solution can
27    be some combination of Pb2+ and its  hydrolysis species, Pb bound to dissolved organic matter,
28    and Pb complexes with inorganic ligands such as Cl~ and SO42  . Alkaline soils  typically have
29    solutions supersaturated with respect to PbCO3; Pb3(CO3)2(OH)2, Pb(OH)2, Pb3(PO4)2,
30    Pb5(PO4)3(OH), and Pb4O(PO4)2 (Badawy et al., 2002).  Pb phosphate minerals  in particular, are
31    very insoluble, and thermodynamic data predict that these phases will control dissolved Pb in

      December 2005                         8-5           DRAFT-DO NOT QUOTE OR CITE

-------
 1    soil solution under a variety of conditions (Nriagu, 1974; Ruby et al., 1994).  However, certain
 2    chelating agents, such as dissolved organic matter can prevent the precipitation of Pb minerals
 3    (Lang and Kaupenjohann, 2003).
 4          Soil solution dissolved organic matter content and pH typically have a very strong
 5    positive and negative correlation, respectively, with the concentration of dissolved Pb species
 6    (Badawy et al., 2002; Sauve et al., 1998, 2000a,b, 2003; Tipping et al., 2003; Weng et al., 2002).
 7    In the case of adsorption phenomena, the partitioning of Pb2+ to the solid phase is also controlled
 8    by total metal loading: high Pb loadings will result in a lower fraction partitioned to the solid
 9    phase.  Sauve et al. (1998;  1997) demonstrated that only a fraction of the total Pb in solution was
10    actually Pb2+ in soils treated with leaf compost.  The fraction of Pb2+ to total dissolved Pb ranged
11    from <1 to 60%,  depending on pH and the availability of Pb-binding ligands.  In acidic soils,
12    Al species can compete for sites on natural organic matter and inhibit Pb binding to surfaces
13    (Gustafsson  et al., 2003).
14
15    Tracing the Fate of Atmospherically-delivered Lead
16          Radiogenic Pb isotopes offer a powerful tool for separating anthropogenic Pb from natural
17    Pb derived from mineral weathering (Erel and Patterson, 1994; Erel et al., 1997). This is
18    particularly useful for studying the mineral soil, where geogenic Pb often dominates. The ore
19    bodies from  which anthropogenic Pb are typically derived are usually enriched in 207Pb relative
20    to 206Pb and  208Pb when compared with Pb found in granitic rocks.  Uranium-238 series 210Pb
21    also provides a tool for tracing atmospherically-delivered Pb in soils.  Fallout 210Pb is deposited
22    onto forests via wet and dry deposition, similar to anthropogenic Pb deposition in forests, and is
23    thus useful as a tracer for non-native Pb in soils.  210Pb is convenient to use for calculating the
24    residence time of Pb in soil layers because it's atmospheric and  soil fluxes can be assumed to be
25    in steady-state at undisturbed sites (Dorr, 1995; Dorr and Munnich, 1989; Kaste et al., 2003).
26          Researchers assessing the fate of atmospheric Pb in soils have also relied on repeated
27    sampling of soils and vegetation for total Pb. This technique works best when anthropogenic Pb
28    accounts for the vast majority of total Pb in a particular reservoir. Johnson et al. (1995b), Yanai
29    et al. (2004), and Friedland et al. (1992) used O horizon (forest floor) time series data to evaluate
30    the movement of gasoline-derived Pb in the soil profile. Surface soils sampled relatively
31    recently demonstrate that the upper soil horizons (O + A horizons) are retaining most of the

      December 2005                          8-6           DRAFT-DO NOT QUOTE OR CITE

-------
 1    anthropogenic Pb burden introduced to the systems during the 20th century (Evans et al., 2005).
 2    Miller and Friedland (1994) and Wang and Benoit (1997) suggested that the movement of
 3    organic particulates dominated Pb transport in the soil profile.
 4
 5    8.1.1.3  Species Response/Mode of Action
 6          The current document expands upon and updates knowledge since 1986 related to the
 7    uptake, detoxification, physiological effects, and modifying factors of lead toxicity to terrestrial
 8    organisms.  Terrestrial organisms discussed in this chapter include soil organisms, plants, birds,
 9    and mammals.
10
11    Uptake into Plants and Invertebrates
12          Recent work supports previous  results and conclusions that surface deposition of lead
13    onto above-ground vegetation from airborne sources may be significant (Dalenberg and Van
14    Driel, 1990; Jones and Johnston, 1991; Angelova et al., 2004). In addition, most lead is taken up
15    by plants via the symplastic route (through cell membranes) (Sieghardt, 1990) and remains in the
16    roots, with little translocation to shoots, leaves, or other plant parts.  Different species of plants
17    and invertebrates accumulate different amounts of lead (Pizl and Josens, 1995; Terhivuo et al.,
18    1994; Wierzbicka, 1999).
19          Recent work supports previous  conclusions that the form of metal tested,  and its
20    speciation in soil, influence uptake and toxicity to plants and invertebrates. The  oxide form is
21    less toxic than the chloride or acetate forms, which are less toxic than the nitrate form of lead
22    (Khan and Frankland, 1983; Lock and Janssen, 2002; Bongers et al., 2004). However, these
23    results must be interpreted with caution, as the counterion (e.g., the nitrate ion) may be
24    contributing to the observed toxicity (Bongers et al., 2004).
25
26    Detoxification in Plants and Invertebrates
27          Lead may be deposited in root cell walls as a detoxification mechanism, and this may be
28    influenced by calcium (Antosiewicz, 2005). Yang et al. (2000) suggested that the oxalate
29    content in root  and root exudates reduced the bioavailability of lead in soil, and that this was an
30    important tolerance mechanism. Other hypotheses put forward recently include  the presence of
      December 2005                         8-7          DRAFT-DO NOT QUOTE OR CITE

-------
 1    sulfur ligands (Sharma et al., 2004) and the sequestration of lead in old leaves (Szarek-
 2    Lukaszewska et al., 2004) as detoxification mechanisms.
 3         Lead detoxification has not been studied extensively in invertebrates. Glutathione
 4    detoxification enzymes were measured in two species of spider (Wilczek et al., 2004). Lead may
 5    be stored in waste nodules in earthworms (Hopkin, 1989) or as pyromorphite in the nematode
 6    (Jackson et al., 2005).
 7
 8    Physiological Effects
 9         The effects on heme synthesis (as measured by 5-aminolaevulinic acid dehydratase
10    [ALAD] activity and protoporphyrin concentration, primarily) had been well-documented in the
11    1986 criteria document and continue to be studied (Schlick et al., 1983;  Scheuhammer, 1989;
12    Redig et al., 1991; Henny et al., 1991; Beyer et al., 2000; Hoffman et al., 2000a, b).  However,
13    Henny et al. (1991) caution that changes in ALAD and other enzyme parameters are not always
14    related to adverse effects, but simply indicate exposure. Other effects on plasma enzymes, which
15    may damage other organs, have been reported (Brar et al., 1997a, b). Lead also may cause lipid
16    peroxidation (Mateo and Hoffman, 2001) which may be alleviated by Vitamin E, although lead
17    poisoning may still result (Mateo et al., 2003b).  Changes in fatty acid production have been
18    reported, which may influence immune response and bone formation (Mateo et  al., 2003a).
19
20    Response Modification
21         Genetics, biological factors, physical/environmental factors, nutritional factors and other
22    pollutants can modify terrestrial organism response to lead. Fisher 344 rats were found to be
23    more sensitive to lead than Sprague-Dawley rats (Dearth et al., 2004). Younger animals  are
24    more sensitive than older animals (Eisler, 1988;  Scheuhammer,  1991), and females generally are
25    more sensitive than males (Scheuhammer, 1987; Tejedor and Gonzalez, 1992; Snoeijs et al.,
26    2005). Monogastric animals are more sensitive than ruminants (Humphreys, 1991).
27    Insectivorous mammals may be more exposed to lead than herbivores (Beyer et al., 1985;
28    Sample et al., 1998), and higher tropic-level consumers may be less exposed than lower trophic-
29    level organisms (Henny et al., 1991). Nutritionally-deficient diets (including low calcium) cause
30    increased uptake of lead (Snoeijs et al., 2005) and greater toxicity (Douglas-Stroebel et al., 2005)
31    in birds.

      December 2005                         8-8           DRAFT-DO NOT QUOTE OR CITE

-------
 1          Mycorrhizal fungi may ameliorate lead toxicity until a threshold is surpassed (Malcova
 2    and Gryndler, 2003), which may explain why some studies show increased uptake into plants
 3    (Lin et al., 2004) while others show no difference or less uptake (Dixon, 1988).  Lower soil pH
 4    generally increases uptake of lead into plants and soil invertebrates. However, calcium content,
 5    organic matter content, and cation exchange capacity of soils also had a significant influence on
 6    uptake of lead into plants and invertebrates (Beyer et al., 1987; Morgan and Morgan, 1988).
 7          Interactions of lead with other metals are inconsistent, depending on the endpoint
 8    measured, the tissue analyzed, the animal species, and the metal combination (Phillips  et al.,
 9    2003;  An et al., 2004; Garcia and Corredor, 2004; He et al., 2004; Perottoni et al., 2005).
10
11    8.1.1.4  Exposure/Response of Terrestrial Species
12          The current document expands upon and updates knowledge related to the effects of lead
13    on terrestrial primary  producers, consumers and decomposers.
14
15    Primary Producers
16          Effects of lead on terrestrial plants include decreased photosynthetic and transpiration
17    rates, and decreased growth and yield. The phytotoxicity of lead is considered to be relatively
18    low, and there are few reports of phytotoxicity from lead exposure under field conditions.
19    Phytotoxicity data recently were reviewed for the development of the ecological soil screening
20    levels (Eco-SSL) (U.S. Environmental Protection Agency, 2005b). Many of the toxicity data
21    presented in EPA (2005b) are lower than those discussed in the 1986 AQC document, although
22    both documents acknowledge that toxicity is observed over a wide range of concentrations of
23    lead in soil (tens to thousands of mg/kg soil). This may be due to many factors,  such as the soil
24    conditions (e.g., pH, organic matter) and differences in bioavailability of the lead in spiked soils,
25    perhaps due to lack of equilibration of the lead solution with the soil after spiking.  Most
26    phytotoxicity data continue to be developed for agricultural plant species (i.e., vegetable and
27    grain crops). Few data are available for trees or native herbaceous plants, although two of the
28    five ecotoxicological  endpoints used to develop the Eco-SSL were for trees  and two were  for
29    clover.
30
      December 2005                         8-9          DRAFT-DO NOT QUOTE OR CITE

-------
 1    Consumers
 2          Effects of lead on avian and mammalian consumers include decreased survival,
 3    reproduction, and growth, as well as effects on development and behavior.  There remain few
 4    field effects data for consumers, except from sites with multiple contaminants, for which it is
 5    difficult to attribute toxicity specifically to lead.  Avian and mammalian toxicity data recently
 6    were reviewed for the development of Eco-SSLs (U.S. Environmental Protection Agency,
 7    2005b). Many of the toxicity data presented in EPA (2005b) are lower than those discussed in
 8    the 1986 AQC document,  although EPA (2005b) recognizes that toxicity is observed over a wide
 9    range of doses (<1 to >1,000 mg Pb/kg  bw-day). Most toxicity data for birds are derived from
10    chicken and quail studies,  and most data for mammals are derived from laboratory rat and mouse
11    studies. Data derived for other species would contribute to the understanding of lead toxicity,
12    particularly for wildlife species with different gut physiologies.  In addition, data derived using
13    environmentally-realistic exposures, such as from lead-contaminated  soil and food may be
14    recommended. Finally, data derived from inhalation exposures, which evaluate endpoints such
15    as survival, growth, and reproduction, would contribute to understanding the implications of
16    airborne releases of lead.
17
18    Decomposers
19          Effects of lead on soil invertebrates include decreased survival, growth and reproduction.
20    Effects on microorganisms include changes in nitrogen mineralization, and changes in enzyme
21    activities. Recent data on  lead toxicity to soil invertebrates and microorganisms are consistent
22    with those reported in EPA (1986a), with toxicity generally observed at concentrations of
23    hundreds to thousands of mg/kg soil.  Studies on microbial processes may be influenced
24    significantly by  soil parameters and the significance of the test results is not clear.
25
26    Ecological Soil Screening Levels (Eco-SSLs)
27          Eco-SSLs are concentrations of contaminants in soils that are protective of ecological
28    receptors (U.S. Environmental Protection Agency, 2005a). They were developed following
29    rigorous scientific protocols, and were subjected to two rounds of peer review.  The Eco-SSLs
30    for terrestrial plants, birds, mammals, and soil invertebrates are 120 mg/kg, 11 mg/kg, 56 mg/kg
31    and 1700 mg/kg, respectively. See Section 8.1.5 for additional information.

      December 2005                         8-10          DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.1.5  Effects of Lead on Natural Terrestrial Ecosystems
 2          At present, industrial point sources such as smelter sites represent the greatest Pb-related
 3    threat to the maintenance of sustainable, healthy, diverse, and high-functioning terrestrial
 4    ecosystems in the United States. However, assessing the risks specifically associated with Pb is
 5    impossible because these sites also experience elevated concentrations of other metals and
 6    impacts related to 862 emissions. Terrestrial ecosystems may respond to stress in a variety of
 7    ways, including reductions in the vigor and/or growth of vegetation, reductions in biodiversity,
 8    and effects on microbial processes.
 9
10    Influence of Dissolved Organic Matter (DOM)
11          Since the movement and fate of Pb in terrestrial ecosystems is strongly related to the
12    organic matter cycle, stresses that could lead to disruption or alteration of the soil organic matter
13    pool are of particular concern in the assessment of effects of ecosystem stress on Pb cycling. By
14    binding soluble Pb, soil organic matter acts as a barrier to the release of lead to drainage waters
15    (Wang et al., 1995; Kaste et al., 2003; Watmough and Hutchinson, 2004). The release of soluble
16    Pb does not typically result in elevated surface water Pb concentrations because: (1) organic
17    matter has a relatively long residence time in most temperate soils (Gosz et al., 1976;
18    Schlesinger,  1997), so the fraction of the organic matter pool that is dissolved at any time is
19    small; (2) DOM-Pb complexes solubilized in upper soil horizons may be precipitated or
20    adsorbed lower in the soil profile; and, (3) the DOM to which Pb is bound may be utilized by
21    microbes, allowing the associated Pb to bind anew to soil organic matter. These factors
22    moderate the release of Pb  to surface waters in temperate terrestrial ecosystems.  As a result,
23    concentrations of Pb in soil solutions and drainage waters tend to be low (Driscoll et al., 1988;
24    Bacon and Bain,  1995; Johnson et al., 1995b; Wang et al., 1995). However, stresses or
25    disturbances that result in increased release of DOM from soils could potentially result in the
26    unanticipated release of Pb to groundwater and/or surface waters.
27
28    Influence ofpH
29          Like most metals, the solubility of Pb is increased at lower pH (Stumm  and Morgan,
30    1995), suggesting that acidification should result in enhanced mobility of Pb in ecosystems.
31    However, reductions in pH may also cause a decrease in the solubility of DOM, due to the

      December 2005                         8-11           DRAFT-DO NOT QUOTE OR CITE

-------
 1    protonation of carboxylic functional groups (Tipping and Woof, 1990). Because of the
 2    importance of complexation with organic matter to Pb mobility in soils, lower DOM
 3    concentrations resulting from acidification may offset the increased solubility of the metal.
 4    The increased mobility was only observed in very acid soils, those with pH <4.5 (Blake and
 5    Goulding, 2002).  Acidification also may enhance Pb export to drainage water in very sandy
 6    soils, with limited ability to retain organic matter (Swanson and Johnson, 1980; Turner et al.,
 7    1985).
 8
 9    Influence of Forest Harvesting
10          Observations from clear-cut sites in the United States and Europe indicate that forest
11    harvesting causes little or no mobilization of Pb from forest soils (Fuller et al., 1988; Johnson
12    et al., 1995a). The principal risk associated with forest harvesting is the loss of Pb in particulate
13    form to drainage waters through erosion.
14
15    Influence of Climate Change
16          The potential linkages between climate-related stress and Pb cycling are very poorly
17    understood. Effects related to alterations in organic matter cycling may influence Pb migration.
18    For example,  an increase in temperature leading to increased rates of organic matter
19    decomposition could lead to temporary increases in DOM concentrations and smaller steady -
20    state pools of soil organic matter. Either of these factors could result in increased concentrations
21    of Pb in waters draining terrestrial ecosystems.
22
23    Influence on Microbial Processes
24          Recent research has documented significant inhibitory effects of Pb and other metals on
25    the activities of several enzymes believed to be crucial to nitrogen mineralization in soils (Senwo
26    and Tabatabai, 1999; Acosta-Martinez and Tabatabai, 2000; Ekenler and Tabatabai, 2002). This
27    suggests that the inhibitory effect of Pb and other metals is broad-based, and not specific to any
28    particular metabolic pathway. In reducing environments, the rate of denitrification is also
29    depressed by trace metals. Fu and Tabatabai (1989) found that 2.5 jimol g"1 of Pb  (ca. 500 mg
30    kg"1) was sufficient to  cause 0, 27, and 52% decreases in nitrogen reductase activity in three
31    different soils.

      December 2005                         8-12          DRAFT-DO NOT QUOTE OR CITE

-------
 1   Effects Observed Around Industrial Point Sources
 2          Assessing the effects of lead exposure on natural ecosystems is complicated by the fact
 3   that lead exposure cannot be decoupled from other factors that could also impact the ecosystem
 4   under consideration. Principal among these factors are other trace metals and acidic deposition.
 5   Emissions of Pb from smelting and other industrial activities are accompanied by other trace
 6   metals (e.g., Zn, Cu, Cd) and sulfur dioxide (862) that may cause toxic effects independently or
 7   in concert with Pb.
 8          Natural terrestrial ecosystems near smelters, mines, and other industrial activities have
 9   exhibited a variety of effects related to ecosystem structure and function. These effects include
10   decreases in species diversity, changes in floral and faunal community composition, and
11   decreasing vigor of terrestrial vegetation.  All of these effects were observed in  ecosystems
12   surrounding the Anaconda copper smelter, in southwestern Montana, which operated between
13   1884 and 1980 (Galbraith et al., 1995; Kapustka et al.,  1995). Similar observations were made in
14   the area surrounding Palmerton, Pennsylvania, where two zinc smelters operated between 1898
15   and 1980 ( Jordan, 1975; Sopper, 1989; Storm et al.,  1994). Subsequent to the effects on
16   vegetation, wind and erosion may remove litter and humus, leaving bare mineral soil, a nearly
17   sterile environment in which very little energy transfer takes place (Little and Martin,  1972;
18   Galbraith et al., 1995). Metal pollution around a Pb-Zn smelter near Bristol, England has not
19   resulted in the loss of oak woodlands within 3 km of the smelter, despite significant
20   accumulation of Pb, Cd, Cu, and Zn in soils and vegetation (Martin and Bullock, 1994).
21   However, the high metal concentrations have favored the growth of metal-tolerant  species in the
22   woodland.
23          The effects of Pb on terrestrial ecosystems near smelters and other industrial sites
24   decrease downwind from the source.  Several studies using the soil Pb burden as an indicator
25   have shown that much of the contamination occurs within a radius of 20-50 km  around the
26   emission source (e.g., Miller and McFee, 1983; Martin and Bullock, 1994;  Galbraith et al., 1995;
27   Spurgeon and Hopkin, 1996a).
28
      December 2005                        8-13          DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.2   Methodologies Used in Terrestrial Ecosystems Research
 2    8.1.2.1  Introduction
 3          The distribution of Pb throughout the ecosystem, via aerial deposition, has been discussed
 4    throughout this document. Its further impacts on soil, sediment, and water provide numerous
 5    pathways that may promote unacceptable risk to all levels of biota.  Stable isotopes of Pb have
 6    been found useful in identifying sources and apportionment to various sources. One of the key
 7    factors required to assess this risk is an understanding, and perhaps quantification, of
 8    bioavailability. Therefore, the bioavailability of Pb is a key issue to the development of ambient
 9    air quality criteria (AAQC).  However, the discussion of all methods used in characterizing
10    bioavailability is beyond the scope of this  chapter.  The following topics are discussed in this
11    chapter.
12          •   Lead Isotopes and Apportionment
13          •   Methodologies to determine Pb speciation
14          •   Lead and the Biotic Ligand Model (BLM)
15          •   In situ methods to reduce Pb bioavailability
16
17    8.1.2.2  Lead Isotopes and Apportionment
18          Determination of the extent of Pb contamination from an individual source(s) and its
19    impact are of primary importance in risk assessment. The identification of exposure pathway(s)
20    is fundamental to the risk analysis and critical in the planning of remediation scenarios.
21          Although societies have been consuming Pb for nearly 9,000 years, production of Pb in
22    the United States peaked in 1910 and 1972, at approximately 750 and 620 kt/year, respectively
23    (Rabinowitz, 2005). The diversity of potential Pb sources (paint pigments, gasoline additives,
24    solders, ceramics, batteries) and associated production facilities (mining, milling, smelting-
25    refining) make fingerprinting of sources difficult.  Therefore, dealing with multiple sources
26    (point and non-point), a reliable and specific fingerprinting technique is required.  It has been
27    well established (Sturges and Barrie, 1987; Rabinowitz, 1995) that the stable isotope
28    composition of Pb is ideally suited for this task. Lead isotopic ratio differences often allow
29    multiple sources to be distinguished, with  an apportionment of the bulk Pb concentration made to
30    those sources.
      December 2005                        8-14          DRAFT-DO NOT QUOTE OR CITE

-------
 1          Lead has four stable isotopes: 204Pb, 206Pb, 207Pb, and 208Pb in natural abundances of 1.4,
 2    24.1, 22.1, and 52.4%, respectively.  The radiogenic 206Pb, 207Pb, and 208Pb are produced by
 3    radioactive decay of 238U, 235U, and 232Th, respectively. Thus, the isotopic composition of Pb
 4    varies based on the U:Pb and Th:Pb ratios of the original ore's source and age (Faure, 1977).
 5    Because of the small fractional mass differences of the Pb isotopes, ordinary chemical and
 6    pyrometallurgical reactions will not alter their original composition. Therefore, anthropogenic
 7    sources reflect the isotopic composition of the ores from which the Pb originated.
 8          To acquire the Pb isotopes, a sample, generally in aqueous form, is analyzed on an
 9    ICP/MS (quadrapole, magnetic sector, or time-of-flight).  Studies reviewing the most common
10    analytical and sample preparation procedures include Ghazi and Millette (2004), Townsend et al.
11    (1998), and Encinar et al. (2001a,b). The correction factor for mass discrimination biases are
12    generally made by analyzing the National Institute for Standards and Technology (NIST), SRM
13    981 and/or spiked 203T1 and 205T1 isotopes (Ketterer et al., 1991; Begley and Sharp, 1997).  The
14    overall success of Pb isotope fingerprinting is generally dependent on analysis precision, which
15    in turn depends on the type of mass analyzer used (Table 8-1.2.1).
                 Table 8-1.2.1. Relative Standard Deviation (RSD) for Lead Isotope
                               Ratios on Selected Mass Spectrometers
RSD
204pb.206pb
207pb.206pb
208pb.206pb
Quadrapole
0.0031
0.0032
0.0026
Double-Focusing
0.0032
0.0027
0.0024
Single-Focusing
Magnetic Sector
0.00053
0.00053
0.00053
High-Resolution
Magnetic Sector
ICP/MS
0.0011
0.00048
0.00046
16          An extensive database comprising primarily North American Pb sources can be assembled
17    from Doe and Rohrbough (1977), Doe and Stacey (1974), Doe et al. (1968), Heyl et al. (1974),
18    Leach et al. (1998), Stacey et al. (1968), Zartman (1974), Cannon and Pierce (1963), Graney
19    et al. (1996), Unruh et al. (2000), James and Henry (1993), Rabinowitz (2005), and
20    Small  (1973).
      December 2005                        8-15          DRAFT-DO NOT QUOTE OR CITE

-------
 1          The use of Pb isotopes to quantitatively apportion source contributions follows the simple
 2    mixing rule when only two sources are possible (Faure, 1977). Once multiple sources need to be
 3    considered, a unique solution can no longer be calculated (Fry and Sherr, 1984).  Phillips and
 4    Gregg (2003) have designed a model to give feasible source contributions when multiple sources
 5    are likely.
 6          Many studies have demonstrated the usefulness of this technique. Media of all types have
 7    been studied:  water (Flegal et al.,  1989a,b; Erel et al.,  1991; Monna et al., 1995), ice (Planchon
 8    et al., 2002), dust (Adgate et al., 1998; Sturges et al., 1993), and soil/sediments (Hamelin et al.,
 9    1990; Farmer et al., 1996; Bindler et al., 1999; Haack et al., 2004; Rabinowitz and Wetherill,
10    1972; Rabinowitz, 2005; Ketterer et al., 2001).
11
12    8.1.2.3  Speciation in Assessing Lead Unavailability in the Terrestrial Environment
13          The National Research Council (NRC) 2002 review on bioavailability defined the
14    "bioavailability processes" in terms of three key processes. One of these processes,  contaminant
15    interactions between phases, is more commonly referred to as "speciation."  For a given metal or
16    metalloid, the term speciation describes the chemical's ability to interact with its biological or
17    chemical surroundings by characterizing its physicochemical properties that are relevant to
18    bioavailability.
19          A wide variety  of analytical (XRD, BMP A, PIXIE, XPS, XAS,  SIMS) and chemical
20    species modeling (SOILCHEM, MINTEQL, REDEQL2, ECOSAT, MINTEQA2, HYDRAQL,
21    PHREEQE, WATEQ4F) tools have been used to  characterize a metal's speciation as it is found
22    in various media. Currently, for risk assessment purposes (not considering phytotoxicity), where
23    large sites with numerous media, pathways, and metals must often be characterized in a
24    reasonable time frame, EMPA techniques provide the greatest information on metal  speciation.
25    Other techniques such as EXAFS and EXANES show great promise and will be important in
26    solving key mechanistic questions. In the case of phytotoxicity, the speciation of metals by
27    direct measurement or chemical models of pore water chemistry is most valuable. Further work
28    needs to  be done in  developing analytical tools for the speciation of the methyl-forming metals
29    (Hg, As, Sb, Se, and Sn) in soils and sediments.
30
     December 2005                         8-16          DRAFT-DO NOT QUOTE OR CITE

-------
 1    Concept
 2          As stated above, for a given metal or metalloid (hereafter also referred to as metal), the
 3    term speciation refers to its chemical form or species, including its physicochemical
 4    characteristics that are relevant to bioavailability.  As a result of the direct impact these factors
 5    often have on a metal's bioavailability, the term "bioaccessibility" has been adopted to define
 6    those factors.
 7
 8    Speciation Role
 9          The accumulation of metals in the lithosphere is of great concern.  Unlike organic
10    compounds, they do not degrade and, thus, have a greater tendency to bioaccumulate. It is now
11    well known that knowledge of the bulk, toxic characteristic leaching procedure (TCLP), or
12    synthetic leaching procedure (SLP) concentrations for any metal is not a controlling factor in
13    understanding a metal's environmental behavior or in developing remedies for its safe
14    management. Although these  tests are essential to site characterization and management, they
15    offer no insight into risk assessment. Rather, it is the metal's bioavailability (the proportion of a
16    toxin that passes a physiological membrane [the plasma membrane in plants or the gut wall in
17    animals] and reaches a target receptor [in cytosol or blood]), which plays a significant role in the
18    risk assessment of contaminated media.
19          The National Research  Council (NRC) review (NRC, 2002) on bioavailability defined
20    bioavailability processes in terms of three key processes:
21
22          •  contaminant interactions between phases (association-dissociation/bound-released),
23          •  transport of contaminants to organism, and
24          •  passage across a physiological membrane.
25
26          This first process, contaminant interactions between phases, is more commonly referred to
27    as speciation. As described above, the speciation of a toxic metal in the environment is a critical
28    component of any ecosystem health risk assessment. The concept of speciation  describes a
29    chemical's ability to interact with its biological or chemical surroundings by characterizing its
30    physicochemical properties. Four important toxicologic and toxicokinetic determinants relating
31    speciation to bioavailability are the (1) chemical form or species, (2) particle size of the metal
32    form, (3) lability of the chemical form, and (4) source.

      December 2005                        8-17          DRAFT-DO NOT QUOTE OR CITE

-------
 1   Chemical Form of Species
 2         The solid phase in a medium controls the activity of a metal in solution, whether the
 3   solution is surface, ground, or pore water or GI fluids, and plays a profound role in metal
 4   bioavailability.  This is perhaps best illustrated by in vivo and in vitro results for many of the
 5   common Pb-bearing minerals (Drexler, 1997) (Figure 8-1.2.1). The metal species found in
 6   media are often diverse, and data suggest that their bioavailability may be significantly
 7   influenced by site-specific variations within these identified metal species (Davis et al., 1993;
 8   Ruby et al., 1992; Drexler and Mushak, 1995).
               LOW  (1-20)
               PbS
               Slag
               Pb-FeOOH
               Native Pb

MEDIUM  (20-60)

    Pb-MnOOH
    PbSO4
    PbP04
                 U     X220  mv»
          15.6KII     X120  :88"ii
        HIGH (60-100)
         PbCO3
         PbO/PbAsO
         Paint
         PbCL
       Figure 8-1.2.1.  Relationship of bioaccessibility versus speciation.


 9   Particle Size of Metal Species
10         Particle size of a metal form is an important factor in the mobilization of the metal,
11   primarily because as size decreases, the surface area of the particle increases, thereby increasing
12   solubility. Thus, although solubility is not the only control for bioavailability, an increase in
13   bioavailability has been directly attributed to a decrease in particle size:  Barltrop and Meek
14   (1979) observed that "the smaller the lead particle, the higher blood lead level." Similar
     December 2005
     8-18
DRAFT-DO NOT QUOTE OR CITE

-------
1    observations were made by Healy et al. (1992) using galena (PbS) and an in vitro dissolution
2    technique. Drexler (1997) presented in vitro results on numerous Pb-bearing phases ranging in
3    particle size from 35 to 250 jim.  While all phases studied showed increased bioavailability with
4    decreasing particle size, more significantly, not all forms showed the same degree or magnitude
5    of change (Figure 8-1.2.2).  Finally, such laboratory data have been supported by extensive
6    epidemiologic evidence, enforcing the importance of particle size (Bornschein et al., 1987;
7    Brunekreef et al., 1983; Angle et al., 1984).
                               00
                                                Anglesite
                                            20        40
                                                Minutes
-e-250|jm
                             60
                                                             38|jm
          b.
             150r
                              Slag
                c.
                        20        40
                           Minutes
         60
                                                      100r
                                                    CO
              PbO
    Figure 8-1.2.2.   Variation of bioavailability with particle size.
    December 2005
         8-19
DRAFT-DO NOT QUOTE OR CITE

-------
 1   Particle Lability
 2          The impact on bioavailability of a metal particle's lability (its associations within the
 3   medium matrix) is not well documented, but it follows the premise put forth by many of the
 4   developing treatment technologies regarding its being bound or isolated from its environment.
 5   Data from several EPA Superfund sites and the Region VIII swine study (U.S. Environmental
 6   Protection Agency, 2004b) suggest that matrix associations, such as liberated versus enclosed,
 7   can play an important part in bioavailability. As illustrated in Figure 8-1.2.3, two different
 8   media with similar total Pb concentrations and Pb forms (slag, Pb-oxide, and Pb-arsenate)
 9   exhibit significantly different bioavailabilities. In the Murray, UT sample (bioaccumulation
10   factor [BAF] = 53%), a greater fraction of the more bioavailable Pb-oxides are liberated and not
11   enclosed in the less-soluble glass-like slag as observed in the Leadville, CO sample
12   (BAF = 17%).  Other evidence is more empirical, as illustrated in Figure 8-1.2.4, where a large
13   particle of native  Pb is shown to have developed a weathering ring of highly bioavailable Pb-
14   chloride and Pb-oxide.  Such observations can be useful in understanding the mechanistic
15   phenomena controlling bioavailability. In addition, they will aid in developing and validating
16   models to predict metal-environment interactions.
17
                 Murry
                 BAF 53%
                 11500mg/kg   Pb
                 20% liberated
                                                                    Leadville AV
                                                                    BAF 17%
                                                                    10600 mg/kg Pb
                                                                    5% liberated
     Figure 8-1.2.3.  Illustration of particle lability and bioavailability.
     December 2005                        8-20          DRAFT-DO NOT QUOTE OR CITE

-------
                                    m
                   BEI Baseline
     Figure 8-1.2.4.  Scanning electron micrograph of a large native Pb particle.
 1   Source
 2          Although the source of a metal is not directly related to bioavailability, it plays an
 3   important role in risk assessment with the evaluation of metal (1) pathways, (2) background,
 4   and (3) apportionment. It is important to understand a metal's pathway before any remedial
 5   action can be taken; otherwise, recontamination of the primary pathway and reexposure can
 6   occur. Knowledge of background is required by statute, as an action level cannot be established
 7   below natural background levels. Finally, cost recovery can be an important factor in a remedial
 8   action, as it is the Agency's responsibility to identify and, if possible, seek the assistance of
 9   responsible parties.

10   Plants
11          When considering the bioavailability of a metal to plants from soils and sediments, it is
12   generally assumed that both the kinetic rate of supply and the speciation of the metal to either the
13   root or shoot are highly important. In soils and sediments, generally only a small volume of
     December 2005
8-21
DRAFT-DO NOT QUOTE OR CITE

-------
 1   water is in contact with the chemical form, and although the proportion of a metal's
 2   concentration in this pore water to the bulk soil/sediment concentration is small, it is this phase
 3   that is directly available to plants. Therefore, pore water chemistry (i.e., metal concentration as
 4   simple inorganic species, organic complexes, or colloid complexes) is most important.
 5         Tools currently used for metal speciation for plants include (1) in situ measurements using
 6   selective electrodes (Gundersen et al., 1992;  Archer et al., 1989; Wehrli et al., 1994); (2) in situ
 7   collection techniques using diffusive equilibrium thin films (DET) and diffusive gradient thin
 8   films (DOT) followed by laboratory analyses (Davison et al., 1991, 1994; Davison and Zhang,
 9   1994; Zhang et al., 1995); and (3) equilibrium models ( SOILCHEM) (Sposito and Coves, 1988).
10
11   8.1.2.4  Tools for Bulk Lead Quantification and Speciation
12   Bulk Quantification
13         The major analytical methods most commonly used for bulk analyses outlined in the  1986
14   Pb ACQD included:
15
16         •   Atomic Absorption Spectrometry  (AAS)
17         •   Emission Spectrometry (Inductively coupled plasma/atomic emission spectrometry)
18         •   X-ray Fluorescence (XRF)
19         •   Isotope Dilution Mass Spectrometry (ID/MS)
20         •   Colorimetric
21         •   Electrochemical (anodic stripping voltametry and differential pulse polarography).
22
23         The choice of analytical method today for bulk quantification is generally ICP/AES or
24   ICP/MS (U.S. Environmental Protection Agency, 2001).  Since 1986, numerous standard
25   reference materials (SRM) have been developed for Pb (Table 8-1.2.2), and several significant
26   technological improvements have been developed.
27         Modern spectrometry systems have replaced photomultiplier tubes with a charge-coupled
28   device (CCD).  The CCD is a camera that can detect the entire light spectrum (>70,000 lines)
29   from 160 to 785 nm. This allows for the  simultaneous measurement of all elements, as well as
30   any interfering lines (a productivity increase), and increases precision.  The detection limit for Pb
31   in clean samples can now be as low as 40 ppb.
32

     December 2005                        8-22          DRAFT-DO NOT QUOTE OR CITE

-------
             Table 8-1.2.2. National Institute of Standards and Technology Lead SRMs
NIST SRM
2710
2711
2709
2587
2586
2783
1648
1649a
2584
2583
1515
1575
Medium
Soil
Soil
Soil
Soil (paint)
Soil (paint)
Filter (PM2 5)
Urban participate
Urban dust
Indoor dust
Indoor dust
Apple leaves
Pine needles
Mean Pb
mg/kg
5532
1162
18.9
3242
432
317
6550
12,400
9761
85.9
0.47
0.167
 1         Modern ICP/AES systems offer a choice of either axial viewed plasma (horizontal),
 2   which provides greater sensitivity (DL= 0.8 |ig/L Pb), or radial (vertical) viewed plasma, which
 3   performs best with high total dissolved samples (DL = 5.0 |ig/L Pb).
 4         The development of reaction or collision cells have expanded the capabilities of ICP/MS
 5   and lowered detection limits for many elements that were difficult to analyze because of
 6   interferences such as Se, As, Ti, Zn, Ca, Fe, and Cr.  The cells provide efficient interference
 7   (isobaric, polyatomic, and argide) removal independent of the analyte and sample matrix by
 8   using various reaction gases (H2, He, NFL?), eliminating the need for interference correction
 9   equations.
10
11   Speciation Tools
12         A wide variety of analytical and chemical techniques have been used to characterize a
13   metal's speciation (as defined above) in various media (Hunt et al., 1992; Manceau et al.,  1996,

     December 2005                        8-23          DRAFT-DO NOT QUOTE OR CITE

-------
 1   2000a; Welter et al., 1999; Szulczewski et al., 1997; Isaure et. al., 2002; Lumsdon and Evans,
 2   1995; Gupta and Chen, 1975; Ma and Uren, 1995; Charlatchka et al., 1997). Perhaps the most
 3   important factor that one must keep in mind in selecting a technique is that, when dealing with
 4   metal-contaminated media, one is most often looking for the proverbial "needle in a haystack."
 5   Therefore, the speciation technique must not only provide the information outlined above, but it
 6   must also determine that information from a medium that contains very little of the metal.
 7   As illustrated in Figure 8-1.2.5, for a Pb-contaminated soil, less than 1% (modally) of a single
 8   species can be responsible for a bulk metals concentration above an action level. This factor is
 9   even more significant for other metals (i.e., As, Cd, or Hg) were action levels are often below
10   100  mg/kg.

                                              ¥S
                              SLAG
                     re
                     .c
                     0.
                     Q_
                     5?
                     o
                                                     7500
                                         Bulk Pb mg/kg
                   10000
     Figure 8-1.2.5.  Bulk lead versus single species modality.

11          Of the techniques tested (physicochemical, extractive, and theoretical), the tools that have
12   been used most often to evaluate speciation for particle-bound metal include X-ray absorption
     December 2005
8-24
DRAFT-DO NOT QUOTE OR CITE

-------
 1    spectroscopy (XAS), X-ray diffraction (XRD), particle induced X-ray emission (PIXE and
 2    jiPIXE), electron probe microanalysis (EPMA), secondary ion mass spectrometry (SIMS),
 3    X-ray photoelectron spectroscopy (XPS), sequential extractions, and single chemical extractions.
 4    The tools that have been used most often to evaluate speciation for metal particles in solution
 5    include the following computer-based models: MINTEQL, REDEQL2, ECOSAT, MINTEQA2,
 6    HYDRAQL, PHREEQE, andWATEQ4F.  These tools are briefly described below.
 7
 8    Particle-Bound Metal
 9    Direct Approaches
10          Over the past decade, numerous advances in materials science have led to the
11    development of a wide range in analytical tools for the determination of metal concentration,
12    bonding, and valance of individual particles on a scale that can be considered useful for the
13    speciation of environmentally important materials (i.e., soils, wastes, sediments, and dust). This
14    review will provide the reader with a brief description of these techniques, including their
15    benefits, limitations (cost, availability, sample preparation, resolution), and usability as well as
16    references to current applications. Although most of these tools are scientifically sound and
17    offer important information on the mechanistic understanding of metal occurrence and behavior,
18    only a few currently provide useful information on metal bioavailability at a "site" level.
19    However, one may still find other techniques essential to a detailed characterization of a selected
20    material to describe the chemical or kinetic factors controlling a metal's release, transport,
21    and/or exposure.
22          X-Ray absorption Spectroscopy (XAS). X-ray absorption spectroscopy (XAS) is a
23    powerful technique using the tunable, monochromatic (white light) X-rays produced by a
24    synchrotron (2-4 GeV) to record oscillations in atomic absorption within a few 100 eV of an
25    element's absorption edge. Spectra provide both information on chemical state and atomic
26    structure. Measurements are theoretically available for all elements and are not surface-sensitive
27    nor sample-sensitive (i.e.,  gases, liquids, solids, and amorphous materials are testable).
28          High-energy spectra within 30 eV of the edge are termed XANES (X-ray absorption near
29    edge structure spectroscopy (Fendorf et al., 1994; Maginn, 1998) are particularly suited for
30    determination and quantification (10 to 100 ppm) of metal in a particular oxidation state
      December 2005                        8-25          DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Szulczewski et al., 1997; Shaffer et al., 2001; Dupont et al., 2002). The lower-energy spectra
 2    persist some 100 eV above the edge.  These oscillations are termed EXAFS (extended X-ray
 3    absorption fine structure) and are more commonly used for speciation analyses (Welter et al.,
 4    1999; Manceau et al., 1996, 2000a; Isaure et al., 2002).
 5          The main limitations to XAS techniques are (1) the lack of spatial resolution; (2) XAS
 6    techniques provide only a weighted average signal of structural configurations, providing
 7    information on the predominant form of the metal, while minor species, which may be more
 8    bioavailable, can be overlooked; (3) access to synchrotrons is limited and the beam time required
 9    to conduct a site investigation would be prohibitive; (4) a large spectral library must be
10    developed; (5) generally, poor fits to solution models are achieved when the compound list is
11    large; and (6) high atomic number elements have masking problems based on compound density.
12         X-Ray Diffraction (XRD). In X-ray diffraction, a monochromatic Fe, Mo, Cr, Co, W, or
13    Cu X-ray beam rotates about a finely powdered sample and is reflected off the interplanar
14    spacings of all crystalline compounds in the sample, fulfilling Bragg's law (nX = 2dsin9).  The
15    identification of a species from this pattern is based upon the position of the lines (in terms of
16    0 or 20) and their intensities as recorded by an X-ray detector. The diffraction angle (20) is
17    determined by the spacing between a particular set of atomic planes.  Identification of the species
18    is empirical, and current available databases contain more than 53,000 compounds.
19          If a sample contains multiple compounds, interpretation becomes more difficult and
20    computer-matching programs are essential.  In some instances, by measuring the intensity of the
21    diffraction lines and comparing them to standards, it is possible to quantitatively analyze
22    crystalline mixtures; however, if the species is a hydrated form or has a preferred orientation, this
23    method is only semi quantitative at best. Since this technique represents a bulk analysis, no
24    particle size or lability information can be extracted from the patterns.
25         Particle Induced X-Ray Emission (PIXE and (J.PIXE). Particle induced X-ray emission
26    (PIXE) uses a beam, ~4 jim in diameter, of heavy charged particles (generally He) to irradiate
27    the sample. The resulting characteristic X-rays are emitted and detected in a similar manner as
28    XRF, using Si-Li detectors. Particles generated from a small accelerator or cyclotron, with a
29    potential of 2 to 4 MeV, are commonly used. Detection limits on the order of 1 mg/kg are
30    achieved on thin-film samples.  Disadvantages to its use for speciation include (1) only a small
31    volume of material can be analyzed (1 to 2 mg/cm2);  (2) no particle size information is provided;

      December 2005                        8-26          DRAFT-DO NOT QUOTE OR CITE

-------
 1    (3) peak overlaps associated with Si-Li detectors limit identification of species; (4) limited
 2    availability; and (5) high cost. For a further review of PIXE analysis and applications, see
 3    Maenhaut(1987).
 4          Electron Probe Microanalysis (EPMA). Electron probe microanalysis uses a finely
 5    focused (1 |im) electron beam (generated by an electron gun operating at a 2 to 30 Kv
 6    accelerating voltage and pico/nanoamp currents) to produce a combination of characteristic
 7    X-rays for elemental quantification along with secondary electrons and/or backscatter electrons
 8    for visual inspection of a sample. Elements from beryllium to uranium can be nondestructively
 9    analyzed at the 50 ppm level with limited sample preparation. X-ray spectra can be rapidly
10    acquired using either wavelength dispersive spectrometers (WDS) or energy dispersive
11    spectrometers (EDS).
12          With WDS, a set of diffracting crystals, of known d-spacing, revolve in tandem with a
13    gas-filled proportional counter inside the spectrometer housing so that Bragg's law is satisfied
14    and a particular wavelength can be focused. Photon energy pulses reflecting off the crystal are
15    collected for an individual elemental line by the counter as  a first approximation to
16    concentration.  For quantitative analysis, these intensities are compared to those of known
17    standards and must be corrected for background, dead time, and elemental  interactions (ZAF)
18    (Goldstein et al., 1992). ZAF correction is in reference to the three components of matrix
19    effects:  atomic number (Z), absorption (A), and fluorescence (F).
20          With EDS,  a single Si-Li  crystal detector is used in conjunction with a multichannel
21    analog-digital converter (ADC) to sort electrical pulses (with heights approximately proportional
22    to the quantum energy of the photon that generated them), producing a spectrum of energy
23    (wavelength) versus  counts. The net area under a particular peak (elemental line) is proportional
24    to its concentration in the sample. For quantitative  analyses, corrections similar to WDS analysis
25    must be performed. Although EDS detectors are more efficient than WDS, detection limits are
26    significantly greater (-1000 ppm), because of elevated backgrounds and peak overlaps.
27    For speciation analysis, the EDS system must NEVER be used as the primary detector, as
28    numerous errors in species identification are often made. These are generally the result of
29    higher-order X-ray line overlaps.
30          This technique has been routinely used for site characterizations (Linton et al., 1980;
31    Hunt et al., 1992; Camp, Dresser, and McKee (COM), 1994; U.S. Environmental Protection

      December 2005                        8-27          DRAFT-DO NOT QUOTE OR CITE

-------
 1    Agency, 2002).  Currently this technique offers the most complete data package on metal
 2    speciation than any of the other tools.  The method is relatively fast and inexpensive, available,
 3    and provides all of the required information for bioavailability assessments (i.e., particle size,
 4    species, lability, and sourcing). A number of limitations still need to be addressed including:
 5    (1) its inability to quickly isolate a statistically significant population of particles in soils with
 6    low bulk metal concentrations (<50 mg/kg), meaning that for some metals with low
 7    concentrations of concern (i.e., Cd, Mo, Sb, Se), this method may be less useful; (2) the more
 8    volatile metals (i.e., Hg, Tl) are often volatilized under the electron beam or lost during sample
 9    preparation.
10          Secondary Ion Mass Spectrometry (SIMS). Secondary ion mass spectrometry (also known
11    as ion microprobes or ion probes) is a well-known primarily surface technique that uses a 0.5 to
12    20 KV O, Ar, Ga, In, or Cs ion beam  in bombarding (sputtering) the surface of a sample while
13    emitting secondary ions that are detected by either quadrapole, time-of-flight (TOF), or magnetic
14    sector mass spectrometers.  Sensitivity is very high, in the ppb range for elements hydrogen to
15    uranium, providing quantitative results on elemental or isotopic  metals and organic compounds.
16    With the advent of liquid metal (In and Ga) ion beams, beam sizes of less than 1 jim are possible,
17    although 20 jim is more commonly used.
18          The major disadvantage of SIMS to species identification is that each element or isotope
19    must be tuned and analyzed sequentially.  This makes the identification of a metal form highly
20    time-consuming and, thus, the characterization of a multiphase medium impractical.
21         X-Ray Photoelectron Spectroscopy (XPS). X-ray photoelectron spectroscopy (XPS) or
22    ESCA (electron spectroscopy for chemical analysis, as it was previously known) is a classical
23    surface, 10 to 50 A in depth, analytical technique for determinating qualitative elemental
24    concentrations of elements greater than He in atomic number and provides limited structural and
25    oxidation state information.  In XPS, the high-energy (15 Kv) electrons are typically produced
26    from a dual-anode (Al-Mg) X-ray tube. The excitation or photoionization of atoms within the
27    near surface of the specimen emit a spectrum of photoelectrons.  The measured binding energy is
28    characteristic of the individual atom to which it was bound. Monochromatic sources are often
29    employed to improve energy resolution, allowing one to infer oxidation states of elements or
30    structure of compounds (organic  and inorganic) by means of small chemical shifts in binding
31    energies (Hercules, 1970). The major disadvantages of XPS for environmental speciation studies

      December 2005                        8-28          DRAFT-DO NOT QUOTE OR CITE

-------
 1    is its poor sensitivity, especially in complex matrices and its large, 100-200 jim, spatial
 2    resolution. Direct speciation techniques discussed above are summarized in Table 8-1.2.3.
 3
 4    Indirect Approaches
 5          A more indirect approach to speciation than the methods previously described include the
 6    functional or operational extraction techniques that have been used extensively over the years
 7    (Tessier et al, 1979; Tessier and Campbell, 1988; Gupta and Chen, 1975). These methods use
 8    either a single or sequential extraction procedure to release species associated with a particular
 9    metal within the media. Single chemical extractions are generally used to determine the
10    bioavailable amount of metal in a functional class: water-soluble, exchangeable, organically
11    bound, Fe-Mn bound, or insoluble.
12          In a similar approach, sequential extractions treat a sample with a succession  of reagents
13    intended to specifically dissolve different, less available phases. Many of these techniques have
14    been proposed, most of which are a variation on the classical method of Tessier et al. (1979),
15    in which metal associated with exchangeable, carbonate-bound, Fe-Mn bound, organically
16    bound, and residual species can be determined.  Beckett (1989), Kheboian and Bauer (1987),
17    and Foerstner (1987) provide excellent reviews on the use and abuse of extractions.  These
18    techniques can be useful in a study of metal uptake in plants, where transfer takes place
19    predominately via a solution phase.  However, one must keep in mind that they are not
20    "selective" in metal species, give no particle size information and, above all, these teachable
21    fractions have never been correlated to bioavailability.
22          Solution Speciation Using Computer-Based Models.  Computer-based models are either
23    based upon equilibrium constants or upon Gibb's free energy values in determining metal
24    speciation from solution chemistry conditions (concentration, pH, Eh, organic complexes,
25    adsorption/desorption sites, and temperature). Both approaches are subject to mass balance and
26    equilibrium conditions. These models have undergone a great deal of development in recent
27    years, as reliable thermodynamic data has become available and can provide some predictive
28    estimates of metal behavior. A good review of these models and their applications is provided
29    by Lumsdon and Evans (1995).
      December 2005                        8-29          DRAFT-DO NOT QUOTE OR CITE

-------
o



cr
to
o
o
                                            Table 8-1.2.3.  Characteristics for Direct Speciation Techniques
oo

OJ
o

Tools
XRD
EMPA
SIMS
XPS
XAS
PIXIE
'2
sS
-J
I
a.
No
Yes
No
No
No
No
es Particle
a. a
C/3 C/3
No
Yes
Yes
No
No
No
es Valance
I!
!/5 !/5
No
Yes+
No
Yes
Yes
No
=
•3
i
1
a.
No
No
No
Yes
Yes
No
=
_o
S ®
.S« ft

No#
Yes
Yes*
Yes*
Yes*
Yes
OJ
o
8I
II
No
Yes?
Yes**
Yes**
Yes**
Yes**
if
II
No
B-U
Li-U
H-U
He-U
B-U
o3
O ^"
J u
No
No***
Yes
No
No
No
il 1
11 1
3-4vol% Bulk
50 ppm 0.5-1 n
1 ppb 10 n
wt.% 100 n
ppb 2 n
10 ppm 4 n
2
-2
>• 0
< U
1 $
2 $$
4 $$$
2 $$
5 $$$$
$$$$
H

6
o


o
H

O


o
H
W

O


O
HH
H
W
          *Technique requires each element be tuned and standardized, requiring unreasonable time limits.

          ** Techniques designed and tested only on simple systems. Multiple species require lengthy analytical times and data reduction.

          *** Limited when combined with ICP/MS/LA.

          # Identifies crystalline compounds and stoichiometric compositions only.

          7 Technique has limitations based on particle counting statistics.

          + Valance determined by charge balance of complete analyses.

-------
 1          Speciation can be controlled by simple reactions; however, in many cases, particularly in
 2    contaminated media, their state of equilibrium and reversibility are unknown.  In addition, these
 3    models suffer from other limitations such as a lack of reliable thermodynamic data on relevant
 4    species, inadequacies in models to correct for high ionic strength, reaction kinetics are poorly
 5    known, and complex reactions with co-precipitation/adsorption are not modeled.
 6          The first limitation is perhaps the most significant for contaminated media.  For example,
 7    none of the models would predict the common, anthropogenic, Pb phases, i.e., paint, solder,
 8    and slag.
 9
10    8.1.2.5  Biotic Ligand Model
11          The Biotic ligand model (BLM) is an equilibrium-based model that has been incorporated
12    into regulatory agencies guidelines (including the EPA) to predict effects of metals on aquatic
13    and terrestrial biota and to aid in the understanding of their interactions with biological surfaces.
14    Most recent directions in research are directed toward extending the model to predict metal
15    toxicity in soils.
16          As initially presented by Paquin et al. (1999), the BLM evolved from both the gill surface
17    interaction model (GSIM) of Pagenkopf (1983) and the free ion activity model (FIAM) of Morel
18    (1983). The model can be used to define  site-specific ambient water quality criteria (AWQC) by
19    providing the rational as to how metal toxi city to an aquatic organism is controlled by variations
20    in water chemistry.
21          By integrating the interaction of a metal in solution with its predicted speciation and
22    subsequent interaction with either a receptor site (e.g., root, gill, whole body) of an organism
23    (biotic ligand) a lethal concentration (LCso) estimate is made, replacing expensive, time
24    consuming bioassay testing. The biotic ligand is assumed to be independent and homogeneously
25    distributed and is essentially described using an affinity constant (Ks [M-l]) that have been
26    generated from laboratory studies.  A current version (v 2.12) of the BLM can be downloaded
27    from:  http://www.hydroqual.com/blm.
28          Currently, a limited metal/organism set ([Cu, Ag, Cd, and Zn] and [flathead minnow,
29    rainbow trout, Daphnia magna, Daphniapulex, and Ceidaphia dubia], respectively) are
30    provided. However, users are able to input site-specific metal/organism datasets if available.
31    The literature contains numerous studies on additional metals (i.e., Co, Ni, Pb, U, Sr, and Ba)

      December 2005                            8-31        DRAFT-DO NOT QUOTE OR CITE

-------
 1    and aquatic organisms, references to which can be found in Slaveykova and Wilkinson (2005)
 2    and Niyogi and Wood (2004). Site-specific water chemistry is entered as temperature, pH, metal
 3    (Cu, Ag, Cd, and Zn), dissolved organic carbon (DOM), humic acid (HA), cations (Ca, Mg, Na,
 4    and K), anions (Cl and 804), and alkalinity for speciation calculations.
 5          Currently, there is no acute BLM for Pb; however the work of MacDonald et al. (2002) on
 6    gill-Pb in rainbow trout and that of Slaveykova and Wilkinson (2002) on algae suggest that Ca2+,
 7    DOM, and perhaps Na+ competitively inhibit Pb2+ uptake and thus exhibit a much lower (<100x)
 8    affinity for the biotic ligand.  Further toxicity testing must be conducted before an acute BLM for
 9    Pb is established. Presently, affinity constants for Pb are limited to a few organisms
10    (Table 8-1.2.4).
                             Table 8-1.2.4.  Affinity Constants for Lead
Organism
Phytoplankton
Bacteria

Fish
Cladoceran
Species
Chlorella kesslerii
Bacillus subtilis
Bacillus lichiformis
Rainbow trout
Hyalella azteca
log Ks [M-l]
5.5
3.4,5.1
4.4, 5.7
6.0
5.8,6.9
Reference
Slaveykova and Wilkinson (2002)
Daughney and Fein (1998)
Daughney and Fein (1998)
MacDonald et al. (2002)
Borgmann et al. (1993, 2004)
MacLean et al. (1996)
11
12
13
14
15
16
17
18
19
20
8.1.2.6  Soil Amendments
      The removal of contaminated soil to mitigate exposure of terrestrial ecosystem
components to Pb can often present both economic and logistic problems.  Because of this,
recent studies have focused on in situ methodologies to lower soil-Pb RBA (Brown et al.,
2003a,b). To date, the most common methods studied include the addition of soil amendments
in an effort to either lower the solubility of the Pb form or to provide sorbtion sites for fixation of
pore-water Pb. These amendments typically fall within the categories of phosphate, biosolid,
and Al/Fe/Mn-oxide amendments.
     December 2005
                                         8-32
DRAFT-DO NOT QUOTE OR CITE

-------
 1   Phosphate Amendments
 2          Phosphate amendments have been studied extensively and, in some cases, offer the most
 3   promising results (Brown et al., 1999; Ryan et al., 2001; Cotter-Howells and Caporn, 1996;
 4   Hettiarachchi et al., 2001, 2003; Yang et al., 2001; Ma et al., 1995). Research in this area stems
 5   from early work by Nriagu (1973) and Cotter-Howells and Caporn (1996), who pointed out the
 6   very low solubilities for many Pb-phosphates (Ksp -27 to -66), particularly chloropyromorphite
 7   [Pbs(PO4)3Cl].  The quest to transform soluble Pb mineralogical forms into chloropyromorthite
 8   continues to be the primary focus of most studies. Sources of phosphorous have included
 9   phosphoric acid (H3PO4), triple-super phosphate (TSP), phosphate rock, and/or hydroxyapatite
10   (HA).  Various studies have combined one or more of these  phosphorous sources with or without
11   lime, iron, and/or manganese in an attempt to enhance amendment qualities. Most amendments
12   are formulated to contain between 0.5 and 1.0% phosphorous by weight. They are then either
13   applied wet or dry and then mixed or left unmixed with the contaminated soil. Success of
14   phosphate amendments has been variable, and the degree of success appears to depend  on
15   available phosphorous and the dissolution rate of the original Pb species.
16          A number of potentially significant problems associated with phosphate amendments have
17   been recognized, including both phyto- and earthworm toxicity (Ownby et al., 2005; Cao et al.,
18   2002; and Rusek and Marshall, 2000). Both of these toxicities are primarily associated with very
19   high applications of phosphorous and/or decreased soil pH.  Indications of phytotoxicity are
20   often balanced by studies such as Zhu et al. (2004) that illustrate a 50 to 70% reduction in shoot-
21   root uptake of Pb in phosphate-amended soils. Additionally, the added phosphate poses the
22   potential risk of eutrophication of nearby waterways from soil runoff.
23
24   Biosolid Amendments
25          Historically, biosolids have been used in the restoration of coal mines (Haering et al.,
26   2000; Sopper, 1993).  More recently, workers have demonstrated the feasibility of their use in
27   the restoration of mine tailings (Brown et al., 2003a),  and urban soils (Brown et al., 2003b;
28   Farfel et al., 2005).  Mine tailings are inherently difficult to remediate in that they pose  numerous
29   obstacles to plant growth. They are most often (1) acidic; (2) high in metal content, thus prone to
30   phytotoxicity; (3) very low in organic content; and (4) deficient in macro- and micronutrients.
      December 2005                            8-33        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Stabilization (i.e., the establishment of a vegetative cover) of these environments is essential to
 2    the control of metal exposure or migration from soil/dust and groundwater pathways.
 3          At Bunker Hill, ID, Brown et al. (2003b) demonstrated that a mixture of high nitrogen
 4    biosolids and wood pulp or ash, when surface applied at a rate of approximately 50 and 220
 5    tons/ha, respectively, increased soil pH from 6.8 to approximately 8.0, increased plant biomass
 6    from 0.01 mg/ha to more than 3.4 tons/ha, and resulted in a healthy plant cover within 2 years.
 7    Metal mobility was more difficult to evaluate.  Plant concentrations of Zn and Cd were generally
 8    normal for the first 2 years of the study; however, Pb concentrations in vegetation dramatically
 9    increased two to three times in the first year. Additionally, macronutrients (Ca, K, and Mg)
10    decreased in plant  tissue.
11          Urban soils, whether contaminated from smelting, paint, auto emissions, or industrial
12    activity,  are often contaminated with Pb (Agency for Toxic Substances and Disease Registry
13    [ATSDR], 1988) and can be a significant pathway to elevated child blood Pb levels (Angle et al.,
14    1974). Typically,  contaminated residential soils are replaced under Superfund rules. However,
15    urban soils are less likely to be remediated unless  a particular facility is identified as the
16    contaminate source.  Application of biosolids to such soils may be a cost-effective means for
17    individuals or communities to lower Pb RBAs.
18          A field study by Farfel et al. (2005) using the commercial biosolid ORGO found that,
19    over a 1-year period, Pb in the dripline soils of one residence had reduced RBAs by
20    approximately 60%. However, soils throughout the remainder of the yard showed either no
21    reduction in RBA or a slight increase.  A more complex study was conducted by Brown et al.
22    (2003a) on an urban dripline soil in the lab. The study used an assortment of locally derived
23    biosolids (raw, ashed, high-Fe  compost, and compost) with and without lime. All amendments
24    were incubated with approximately 10% biosolids for a little more than 30 days. In vitro and in
25    vivo data both indicated a 3 to 54% reduction in Pb RBA, with the high-Fe compost providing
26    the greatest reduction.
27          As with phosphate amendments, problems with biosolid application have also been
28    documented. Studies have shown that metal transport is significantly accelerated in soils
29    amended with biosolids (Al-Wabel et al., 2002; McBride et al., 1999, 1997; Lamy et al., 1993;
30    Richards et al., 1998, 2000).  Some of these studies indicate that metal concentrations in soil
31    solutions up to 80 cm below the amended surface  increased by 3- to 20-fold in concentration up

      December 2005                            8-34        DRAFT-DO NOT QUOTE OR CITE

-------
 1    to 15 years after biosolid application.  The increase in metal transport is likely the result of
 2    elevated dissolved organic carbon (DOC) in the amended soil. Anodic stripping voltammetry
 3    has indicated that very low (2 to 18%) of the soluble metals are present as ionic or inorganic
 4    complexes (McBride, 1999; Al-Wabel et al., 2002).
 5
 6    8.1.2.7  Future Needs
 7          Since the 1986 Lead ACQD, considerable data has been generated on the bioavailability
 8    process. The understanding of bioavailability is central to improving risk assessments and
 9    designing efficient, cost-effective remediations. Four key areas for future research can be
10    identified.
11          •  A set of bioavailability and speciation standards should be developed for traceability
12             and quality assurance to aid researchers in developing new or refining existing tools.
13          •  An effort should be made to develop in vitro bioassays for nonhuman biota in order to
14             provide site-specific, rapid, cost-effective estimates of bioavailability/toxicity for all
15             levels of the ecosystem evaluated in a risk assessment.
16          •  Research should continue on the development of in situ amendments to lower Pb
17             bioavailability, with a strong emphasis on long-term field validation studies.
18          •  Finally, toxi city testing for expanding organism/metal affinity constants for the BLM
19             should  be continued.
20
21    8.1.3   Distribution of Atmospherically Delivered Lead in
22            Terrestrial Ecosystems
23    8.1.3.1  Introduction
24          The 1986 Lead Air Quality Criteria Document (Pb AQCD) (U.S. Environmental
25    Protection Agency, 1986a) contains only a few minor sections that detail the speciation,
26    distribution, and behavior of atmospherically delivered Pb in terrestrial ecosystems. The
27    document concluded that the majority of Pb in the atmosphere at that time was from gasoline
28    consumption:  of the 34,881 tons of Pb emitted to the atmosphere in 1984, 89% was from
29    gasoline use and minor amounts were from waste oil combustion, iron and steel production, and
30    smelting. Lead in the atmosphere today, however, does not come from gasoline consumption;
31    instead it results largely  from waste incineration, metal smelting, and metal production (Polissar
32    et al., 2001; Newhook et al., 2003).  The emission source can determine the  species of Pb that are
      December 2005                            8-35        DRAFT-DO NOT QUOTE OR CITE

-------
 1    delivered to terrestrial ecosystems.  For example, Pb species emitted from automobile exhaust is
 2    dominated by particulate Pb halides and double salts with ammonium halides (e.g., PbBrCl,
 3    PbBrCl2NH4Cl), while Pb emitted from smelters is dominated by Pb-sulfur species (Habibi,
 4    1973). The halides from automobile exhaust break down rapidly in the atmosphere, possibly via
 5    reactions with atmospheric acids (Biggins and Harrison,  1979). Lead phases in the atmosphere,
 6    and presumably the compounds delivered to the surface of the earth (i.e., to vegetation and soils),
 7    are suspected to be in the form of PbSO/t, PbS, and PbO (Olson and Skogerboe, 1975; Clevenger
 8    et al., 1991; Utsunomiya et al., 2004).
 9          There are conflicting reports of how atmospherically derived Pb specifically behaves in
10    surface soils. This disagreement may represent the natural variability of the biogeochemical
11    behavior of Pb in different terrestrial systems, or it may be a function of the different analytical
12    methods employed. The importance of humic and fulvic acids (Zimdahl and Skogerboe, 1977;
13    Gamble  et al., 1983) and hydrous Mn- and Fe-oxides (Miller and McFee, 1983) for scavenging
14    Pb in soils are discussed in some detail in the 1986 Pb AQCD. Nriagu (1974) used
15    thermodynamics to argue that Pb-orthophosphates (e.g., pyromorphite) represented the most
16    stable Pb phase in many soils and sediments. He further suggested that, because of the extremely
17    low solubility of Pb-phosphate minerals, Pb deposition could potentially reduce phosphorous
18    availability.  Olson and Skogerboe (1975) reported that solid-phase PbSO4 dominated gasoline-
19    derived Pb speciationin surface soils from Colorado, Missouri, and Chicago, while
20    Santillan-Medrano and Jurinak (1975) suggested that Pb(OH)2, Pb(PO4)2, and  PbCO3 could
21    regulate  Pb  speciation in soils.  However, insoluble organic material can bind strongly to Pb and
22    prevent many inorganic phases from ever forming in soils (Zimdahl and Skogerboe, 1977).
23          The vertical distribution and mobility of atmospheric Pb in soils was poorly documented
24    prior to 1986. Chapter 6 of the 1986 AQCD cited a few references suggesting that the
25    atmospheric Pb is retained in the upper 5 cm of soil (Reaves and Berrow, 1984). Techniques
26    using radiogenic Pb isotopes had been developed to discern between gasoline-derived Pb and
27    natural, geogenic (native) Pb, but these techniques were mostly applied to only sediments
28    (Shirahata et al., 1980) prior to the 1986 Pb AQCD. Without using these techniques, accurate
29    determinations of the depth-distribution and potential migration velocities for atmospherically
30    delivered Pb in soils were largely unavailable.
      December 2005                           8-36        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Some technological advances, combined with the expansion of existing technologies after
 2    1986, resulted in the publication of a large body of literature detailing the speciation,
 3    distribution, and geochemical behavior of gasoline-derived Pb in the terrestrial environment.
 4    Most notably, the development of selective chemical extraction (SCE) procedures as a rapid and
 5    inexpensive means for partitioning Pb into different soil and sediment phases (e.g., Pb-oxides,
 6    Pb-humate, etc.) has been exploited by a number of researchers (Tessier et al., 1979; Johnson
 7    and Petras, 1998; Ho and Evans, 2000; Scheckel et al., 2003). Also, since 1986, several workers
 8    have exploited synchrotron-based XAS in order to probe the electron coordination environment
 9    of Pb in soils, organic matter, organisms, and sediments (Manceau et al., 1996; Xia et al., 1997;
10    Trivedi et al., 2003). X-ray absorption studies can be used for the in situ determination of the
11    valence state of Pb and can be used to quantify Pb speciation in a variety of untreated samples.
12    Biosensors, which are a relatively new technology coupling biological material such as an
13    enzyme with a transducer,  offer a new, simple, and inexpensive means for quantifying available
14    Pb in ecosystems (Verma and Singh, 2005). Advances in voltammetric, diffusive gradients in
15    thin films (DOT), and TCP techniques have also increased the abilities of researchers to quantify
16    Pb phases in solutions  (Berbel et al., 2001;  Scally et al., 2003).  In addition to the development of
17    techniques for describing and quantifying Pb species in the soils and solutions, researchers have
18    used radiogenic Pb isotopes (206Pb, 207Pb, 208Pb) to quantify the distribution, speciation, and
19    transport of anthropogenic Pb in soil profiles and in vegetation (Bindler et al., 1999; Erel et al.,
20    2001; Kaste et al., 2003; Klaminder et al., 2005).
21          Over the past several decades, workers have also developed time-series data for Pb in
22    precipitation, vegetation, organic horizons,  mineral soils, and surface waters. Since
23    atmospherically delivered Pb often comprises a significant fraction of the "labile" Pb (i.e., Pb not
24    associated with primary minerals), these data have been useful for developing transport and
25    residence time models of Pb in different terrestrial reservoirs (Friedland et al.,  1992; Miller and
26    Friedland, 1994; Johnson et al., 1995b; Wang and Benoit, 1997). Overall, a significant amount
27    of research has been published on the distribution, speciation, and behavior of anthropogenic Pb
28    in the terrestrial environment since 1986. However, certain specific details on the behavior of Pb
29    in the terrestrial environment and its potential effects on soil microorganisms remain elusive.
30
      December 2005                            8-37        DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.3.2  Speciation of Atmospherically-Delivered Lead in Terrestrial Ecosystems
 2    8.1.3.2.1   Lead in the Solid Phases
 3          Selective chemical extractions have been employed extensively over the past 20 years for
 4    quantifying amounts of a particular metal phase (e.g., PbS, Pb-humate, Pb-Fe/Mn-oxide) present
 5    in soil or sediment rather than total metal concentration.  Sometimes selective chemical
 6    extractions are applied sequentially to a particular sample. For example, the exchangeable metal
 7    fraction is removed from the soil using a weak acid or salt solution (e.g., BaCb), followed
 8    sequentially by an extraction targeting organic matter (e.g., F^C^ or NaOCl), an extraction
 9    targeting secondary iron oxides (e.g., NFI^OH-HCl), and finally, a reagent cocktail targeting
10    primaiy minerals (e.g., HNOs-HCl-HF). Tessier et al. (1979) developed this technique.  More
11    recently, this technique has been modified and developed specifically for different metals and
12    different types of materials (Keon et al., 2001).  Alternatively, batch-style selective chemical
13    extractions have been used on soils and sediments to avoid the problems associated with
14    nonselective reagents (Johnson  and Petras, 1998).  Selective extractions can be a relatively rapid,
15    simple, and inexpensive means  for determining metal phases in soils and sediments, and the
16    generated data can be linked to  potential mobility and bioavailability of the metal (Tessier and
17    Campbell,  1987). However, some problems persist with the selective extraction technique.
18    First, extractions are rarely specific to a single phase.  For example, while FI^C^ is often used to
19    remove metals bound to organic matter in soils,  others have demonstrated that this reagent
20    destroys clay minerals and sulfides (Ryan et al., 2002). Peroxide solutions may also be
21    inefficient in removing metals bound to humic acids, and in fact could potentially result in the
22    precipitation of metal-humate substances. In addition to the nonselectivity of reagents,
23    significant metal redistribution has been documented to occur during sequential chemical
24    extractions (Ho and Evans, 2000), and many reagents may not completely extract targeted
25    phases. While chemical extractions provide some useful information on metal phases in soil or
26    sediment, the results should be treated as "operationally defined," e.g., "H2O2-liberated Pb"
27    rather than "organic Pb."
28          Lead forms strong coordination complexes with oxygen on mineral  surfaces and organic
29    matter functional groups (Abd-Elfattah and Wada, 1981), because of its high electronegativity
30    and hydrolysis constant. Therefore, Pb is generally not readily exchangeable, i.e., the amount of
31    Pb removed from soils by dilute acid or salts is usually less that 10% (Karamanos et al., 1976;

      December 2005                             8-38         DRAFT-DO NOT QUOTE OR CITE

-------
 1    Sposito et al., 1982; Miller and McFee, 1983; Johnson and Petras, 1998; Bacon and Hewitt,
 2    2005). Lead is typically adsorbed to organic and inorganic soil particles strongly via inner-
 3    sphere adsorption (Xia et al., 1997; Bargar et al., 1997a,b,  1998). Kaste et al. (2005) found that a
 4    single extract of 0.02 M HC1 removed 15% or less Pb in organic horizons from a montane forest
 5    in New Hampshire. The fact that relatively concentrated acids, reducing agents, oxidizing
 6    agents, or chelating agents are required to liberate the majority of Pb from soils is used as one
 7    line of evidence that Pb migration and uptake by plants in soils is expected to be low.
 8         Lead that is "organically bound" in soils is typically quantified by extractions that
 9    dissolve/disperse or destroy organic matter.  The former approach often employs an alkaline
10    solution (NaOH), which deprotonates organic matter functional groups, or a phosphate solution,
11    which chelates structural cations.  Extractions used to destroy organic matter often rely on H2O2
12    or NaOCl.  Both organic and mineral horizons typically have significant Pb in this soil phase.
13    Miller and McFee (1983) used Na4P2O? to extract organically bound Pb from the upper 2.5 cm of
14    soils sampled from northwestern Indiana. They found that  organically bound Pb accounted for
15    between 25  and 50% of the total Pb present in the sampled topsoils.  Jersak et al.  (1997), Johnson
16    and Petras (1998), and Kaste et al. (2005) selectively extracted Pb from spodosols from the
17    northeastern United States. Using acidified H2O2, Jersak et al. (1997) found that very little
18    (<10 %) of the Pb in mineral soils (E, B, C) sampled from New York and Vermont was organic.
19    Johnson and Petras (1998) used K4P2Oy to quantify organically bound Pb in the Oa horizon and
20    in mineral soils from the Hubbard Brook Experimental Forest in New Hampshire. They reported
21    that 60% of the total Pb in the Oa horizon was organic and that between 8 and 17% of the total
22    Pb in the mineral soil was organic. However, in the E, Bh, and Bsl horizons, organically bound
23    Pb dominated the total "labile" (non-mineral lattice) Pb. Kaste et al. (2005) used selective
24    chemical extractions on organic horizons from montane forests in Vermont and New Hampshire.
25    They found that repeated extractions with Na4P2O? removed between 60 and 100% of the Pb
26    from their samples.  Caution should be used when interpreting the results of pyrophosphate
27    extractions.  Although they are often used to quantify organically-bound metals, this reagent can
28    both disperse and dissolve Fe phases (Jeanroy and Guillet,  1981;  Shuman, 1982). Acidified
29    H2O2 has also been reported to destroy and release elements associated with secondary soil
30    minerals (Papp et al., 1991; Ryan et al., 2002).
      December 2005                           8-39        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Aside from organic forms, Pb is often found to be associated with secondary oxide
 2    minerals in soils. Pb can be partitioned with secondary oxides by a variety of mechanisms,
 3    including (1) simple ion exchange, (2) inner-sphere or outer-sphere adsorption, and (3) co-
 4    precipitation and/or occlusion (Bargar et al., 1997a,b, 1998, 1999). As discussed above, very
 5    little Pb is removed from soil via dilute acid or salt solutions,  so adsorption and co-precipitation
 6    are likely the dominant Pb interaction with secondary mineral phases.  Reagents used to quantify
 7    this phase are often solutions of EDTA, oxalate, or hydroxylamine hydrochloride (HH). Miller
 8    and McFee (1983) used an EDTA solution followed by an HH solution to quantify Pb occluded
 9    by Fe and Mn minerals, respectively, in their surface-soil samples from Indiana. They reported
10    that approximately 30% of the total soil Pb was occluded in Fe minerals, and 5 to 15% was
11    occluded in Mn phases.  In soils  from the northeastern United States, Jersak et al. (1997) used
12    various strengths of HH solutions and  concluded that negligible Pb was associated with Mn-
13    oxides and that 1 to 30% of the Pb was associated with Fe phases in the mineral soils in their
14    study. Johnson and Petras (1998) reported that no Pb was removed from the Oa horizon at the
15    HBEF by oxalate, but that 5 to 15% of the total Pb in mineral soils was removed by this
16    extraction, presumably because it was  bound to amorphous oxide minerals.  Kaste et al. (2005),
17    however, reported that HH removed 30 to 40% of the Pb from organic horizons in their study.
18    They concluded that Fe phases were important in scavenging  Pb, even in soil horizons
19    dominated by organic matter.
20          Synchrotron radiation  (X-rays)  allows researchers to probe the electron configuration of
21    metals in untreated soil and sediment samples.  This type of analysis has been extremely valuable
22    for determining the coordination environment of Pb in a variety of soils in sediments. Since
23    different elements have different electron binding energies (Eb), X-rays can be focused in an
24    energy window specific to a metal of interest. In experiments involving XAS, X-ray energy is
25    increased until a rapid increase in the amount of absorption occurs; this absorption edge
26    represents Eb. The precise energy required to dislodge a core electron from a metal (i.e., Eb) will
27    be a function of the oxidation state and covalency of the metal. X-ray absorption studies that
28    focus on the location of the absorption edge are referred to as XANES (X-ray absorption near
29    edge structure). In the energy region immediately after the absorption edge, X-ray absorption
30    increases and decreases with  a periodicity that represents the wave functions of the ejected
31    electrons and the constructive and destructive interference with the wave functions of the nearby

      December 2005                            8-40       DRAFT-DO NOT QUOTE OR CITE

-------
 1    atoms.  X-ray absorption studies used to investigate the periodicity of the absorption after Eb are
 2    referred to as EXAFS (extended X-ray absorption fine structure). Since the electron
 3    configuration of a Pb atom will be directly governed by its speciation (e.g., Pb bound to organics,
 4    Pb adsorbed to oxide surfaces, PbS, etc.) X-ray absorption studies provide a powerful in situ
 5    technique for determining speciation without some of the problems associated with chemical
 6    extractions (Bargar et al., 1997a,b, 1998).
 7          Manceau et al. (1996) used EXAFS to study soil contaminated by gasoline-derived Pb in
 8    France and found that the Pb was divalent and complexed to salicylate and catechol-type
 9    functional  groups of humic substances. He concluded that the alkyl-tetravalent Pb compounds
10    that were added to gasoline were relatively unstable and will not dominate the speciation of Pb
11    fallout from the combustion of leaded gasoline. The binding mechanism of Pb to organics is
12    primarily inner-sphere adsorption (Xia et al., 1997). DeVolder et al.  (2003) used EXAFS to
13    demonstrate that Pb phases were shifting to the relatively insoluble PbS when contaminated
14    wetland soils were treated with sulfate. More recent XAS studies have demonstrated the
15    importance of biomineralization of Pb in soils by bacteria and nematodes (Xia et al., 1997;
16    Templeton et al., 2003a,b; Jackson et al., 2005).  Templeton et al. (2003a,b) demonstrated that
17    biogenic precipitation of pyromorphite was the dominant source of Pb uptake by Burkholderia
18    cepacia biofilms below pH 4.5.  Above pH 4.5, adsorption complexes began to form in addition
19    to Pb mineral precipitation.
20          In addition to XAS studies of Pb in environmental samples, numerous experimental-based
21    XAS studies have documented in detail the coordination environment of Pb adsorbed to Fe-
22    oxides, Mn-oxides, Al-oxides, and clay minerals (Manceau et al., 1996, 2000a,b, 2002; Bargar
23    et al., 1997a,b, 1998, 1999; Strawn and Sparks, 1999; Trivedi et al., 2003). Bargar et al. (1997a)
24    showed that Pb can adsorb to FeOe octahedra on three different types of sites: on corners, edges,
25    or faces. Ostergren et al. (2000a,b) showed that the presence of dissolved carbonate and sulfate
26    increased Pb adsorbtion on goethite. The relative fraction of corner-sharing complexes can be
27    greatly increased by the presence of these ligands, as bridging complexes between the metal and
28    the corners are formed (Ostergren et al., 2000a,b).
29          Recently, Jackson et al. (2005) used microfocused synchrotron-based X-ray fluorescence
30    (//SXRF) to detail the distribution of Pb and Cu in the nematode Caenorhabditis elegans. They
31    found that, while Cu was evenly distributed throughout the bodies of exposed Caenorhabditis

      December 2005                            8-41        DRAFT-DO NOT QUOTE OR CITE

-------
 1    elegans, Pb was concentrated in the anterior pharynx region.  Microfocused X-ray diffraction
 2    indicated that the highly concentrated Pb region in the pharynx was actually comprised of the
 3    crystalline Pb mineral, pyromorphite. The authors concluded that C. elegans precipitated
 4    pyromorphite in the pharynx as a defense mechanism to prevent spreading the toxic metal to the
 5    rest of the organism's body. They further suggested that, because of the high turnover rate of
 6    nematodes, biomineralization could play an important role in the speciation of Pb in certain soils.
 7
 8    8.1.3.2.2   Lead Solid-Solution Partitioning
 9          The concentration of Pb species dissolved in soil solution is probably controlled by some
10    combination of (a) Pb-mineral solubility equilibria, (b) adsorption reactions of dissolved Pb
11    phases on inorganic surfaces (e.g., crystalline or amorphous oxides of Al, Fe, Si, Mn, etc.; clay
12    minerals), and (c) adsorption reactions of dissolved Pb phases on soil organic matter.  Dissolved
13    Pb phases in soil solution can be some combination of Pb2+ and its hydrolysis species, Pb bound
14    to dissolved organic matter, and Pb complexes with inorganic ligands such as Cl~ and SO42  .
15    Alkaline soils typically have solutions supersaturated with respect to PbCOs, PbsCCOsHOFTh,
16    Pb(OH)2, Pb3(PO4)2, Pb5(PO4)3(OH), and Pb4O(PO4)2 (Badawy et al., 2002). Pb-phosphate
17    minerals in particular are very insoluble, and thermodynamic data predict that these phases will
18    control dissolved Pb in soil solution under a variety of conditions (Nriagu, 1974; Ruby et al.,
19    1994). However, certain chelating agents, such as dissolved organic matter, can prevent the
20    precipitation of Pb minerals (Lang and Kaupenjohann, 2003).
21          Soil solution dissolved organic matter content and pH  typically have very strong positive
22    and negative correlations, respectively, with the concentration of dissolved Pb species (Sauve
23    et al.,  1998, 2000b, 2003; Weng et al., 2002; Badawy et al., 2002; Tipping et al., 2003).  In the
24    case of adsorption phenomena, the partitioning of Pb2+ to the solid phase is also controlled by
25    total metal loading, i.e., high Pb loadings will result in a lower fraction being partitioned to the
26    solid phase.  Sauve et al. (1997, 1998) demonstrated that only a fraction of the total Pb in
27    solution was actually Pb2+ in soils treated with leaf compost.  The fraction of Pb2+ to total
28    dissolved Pb ranged from <1 to 60%, depending on pH and the availability of Pb-binding
29    ligands. Nolan et al. (2003) used Donnan dialysis to show that 2.9 to 48.8% of the dissolved Pb
30    was Pb2+ in pore waters of agricultural and contaminated soils from Australia and the United
      December 2005                            8-42        DRAFT-DO NOT QUOTE OR CITE

-------
 1    States.  In acidic soils, Al species can compete for sites on natural organic matter and inhibit Pb
 2    binding to surfaces (Gustafsson et al., 2003).
 3          Differential pulse anodic stripping voltammetry (DPASV) is a technique that is useful for
 4    identifying relatively low concentrations of Pb2+ and has found many applications in adsorption
 5    and partitioning experiments. This technique has been particularly useful for quantifying the Kd,
 6    or partitioning ratio of Pb in the solid-to-liquid phase (Kd = [total solid-phase metal in mg kg"1] /
 7    [dissolved metal in mg I/1]). While the exact Kd value is a function of pH, organic matter
 8    content, substrate type, total metal burden,  and concentrations of competing ligands, such studies
 9    typically show that Pb has very strong solid-phase partitioning.  Partitioning ratios determined by
10    DPASV generally range from 103 to 106 in soils in the typical pH range (Sauve et al., 2000a).
11    Aualiitia and Pickering (1987) used thin film ASV to compare the relative affinity of Pb for
12    different inorganic particulates. They reported that Mn(IV) oxides completely adsorbed the Pb,
13    regardless of pH in the range of 3 to 9, and had the highest affinity for Pb in their study. The
14    adsorption of Pb to pedogenic Fe-oxides, Al-hydroxides,  clay minerals, and Fe ores was reported
15    to be pH-dependent. Sauve et al. (1998) used DPASV to study the effects of organic matter and
16    pH on Pb adsorption in an orchard soil. They demonstrated that Pb complexation to dissolved
17    organic matter (DOM) increased Pb solubility, and that 30 to 50% of the dissolved Pb was bound
18    to DOM at pH 3 to 4, while >80% of the dissolved Pb was bound to DOM at neutral pH.  They
19    concluded that in most soils, Pb in solution would not be  found as Pb2+ but as bound to DOM.
20    Sauve et al.  (2000b) compared the relative  affinity of Pb2+ for synthetic ferrihydrite, leaf
21    compost, and secondary oxide minerals collected from soils. They reported that the inorganic
22    mineral phases were more efficient at lowering the amount of Pb2+ that was available in solution
23    than the leaf compost.  Glover et al. (2002) used DPSAV in studying the effects of time and
24    organic acids on Pb adsorption to goethite.  They found that Pb adsorption to geothite was very
25    rapid, and remained unchanged after a period of about 4 h. Lead desorption was found to be
26    much slower, however, and adsorption was not reversible on a time scale of 8 h. The presence of
27    salicylate appeared to increase the amount of Pb that desorbed from goethite more so than
28    oxalate.
29
      December 2005                            8-43        DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.3.3  Tracing the Fate of Atmospherically Delivered Lead in Terrestrial Ecosystems
 2          Radiogenic Pb isotopes offer a powerful tool for separating anthropogenic Pb from natural
 3    Pb derived from mineral weathering (Erel and Patterson,  1994; Erel et al., 1997).  This is
 4    particularly useful for studying mineral soil, where geogenic Pb often dominates.  The three
 5    radiogenic stable Pb isotopes (206Pb, 207Pb, and 208Pb) have a heterogeneous distribution in the
 6    earth's crust primarily because of the differences in the half-lives of their respective parents
 7    (238U, Ti/2 = 4.7 x  109 year; 235U, T1/2 = 0.7 x 109 year; 232Th, T1/2 = 14 x 109 year).  The result is
 8    that the ore bodies from which anthropogenic Pb are typically derived are usually enriched in
 9    207Pb relative to 206Pb and 208Pb when compared with Pb found in granitic rocks. Graney et al.
10    (1995) analyzed a  dated core from Lake Erie, and found that the 206Pb:207Pb value in sediment
11    deposited in the late 1700s was 1.224, but in 20th-century sediment, the ratio ranged from 1.223
12    to 1.197.  This shift in the Pb isotopic composition represents the introduction of a significant
13    amount of anthropogenic Pb into the environment. Bindler et al. (1999) and Emmanuel  and Erel
14    (2002) analyzed the isotopic composition of Pb in  soil profiles in Sweden and the Czech
15    Republic, respectively, and determined that mineral soils immediately below the organic horizon
16    had a mixture of both anthropogenic and geogenic Pb.
17          Erel  and Patterson (1994) used radiogenic Pb  isotopes to trace the movement of industrial
18    Pb from topsoils to groundwaters to streams in a remote mountainous region of Yosemite
19    National Park in California.  They calculated that total 20th-century industrial Pb input to their
20    study site was approximately 0.4 g Pb nT2.  Lead concentrations in organic material were highest
21    in the upper soil horizons, and decreased with depth. During snowmelt, Pb in the snowpack was
22    mixed with the anthropogenic  and geogenic Pb already in the topsoil, and spring melts contained
23    a mixture of anthropogenic and geogenic particulate  Pb.  During base flows, however, 80% of
24    the Pb export from groundwater and streams was from natural granite weathering (Erel and
25    Patterson, 1994).
26          Uranium-238 series 210Pb also provides a tool for tracing atmospherically delivered Pb in
27    soils.  After 222Rn (Ti/2 = 3.8 days) is produced from the decay of 226Ra (Ti/2 = 1600 years), some
28    fraction of the 222Rn escapes from rocks and soils to the atmosphere. It then decays relatively
29    rapidly to 210Pb (Ti/2 = 22.3 years), which has a tropospheric residence time of a few weeks
30    (Koch et al., 1996). Fallout 210Pb is deposited onto forests via wet and dry deposition, similar to
31    anthropogenic Pb deposition in forests, and is thus useful as a tracer for non-native Pb in soils.
      December 2005                            8-44        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Lead-210 is convenient to use for calculating the residence time of Pb in soil layers, because its
 2    atmospheric and soil fluxes can be assumed to be in steady state at undisturbed sites (Dorr and
 3    Munnich, 1989; Dorr, 1995; Kaste et al., 2003).  Atmospheric 210Pb (210Pbex hereafter,210Pb in
 4    "excess" of that supported by 222Rn in the soil) must be calculated by subtracting the amount of
 5    210Pb formed in soils by the in situ decay of 222Rn from the total 210Pb (Moore and Poet, 1976;
 6    Nozakietal., 1978).
 7          Benninger et al. (1975) measured fallout 210Pb in soils and streamwater at Hubbard Brook
 8    and at an undisturbed forest in Pennsylvania. They estimated atmospheric 210Pb export in
 9    streamwaters to be <0.02% of the standing 210Pb crop in the organic horizons. They used a
10    simple steady-state model to calculate the residence time of Pb in the organic horizons to be
11    5,000 years.  This overestimate of the Pb residence time in the organic horizons was likely a
12    result of the low resolution of their sampling.  Since they only sampled the upper 6 cm of soil
13    and the drainage waters, they did not accurately evaluate the distribution of 210Pb in the soil
14    column in between. Dorr and Munnich (1989, 1991) used 210Pb profiles in soils of southern
15    Germany to evaluate the behavior of atmospherically delivered Pb. They calculated the vertical
16    velocity of Pb by dividing the relaxation depth (i.e., the depth at which 210Pb activity decreases to
17    1/e, or approximately 37% of its surface value) by the 210Pb mean  life of 32 years.  They reported
18    downward transit velocities of atmospherically deposited Pb at 0.89 ± 0.33 mm year"1. The
19    downward transport of atmospheric Pb was not affected by pH or soil type. However, since Pb
20    velocities in the soil profile where identical to carbon velocities calculated with  14C, they
21    concluded that Pb movement in forest soils is probably controlled  by carbon transport. Kaste
22    et al.  (2003) used 210Pb to model the response time of atmospherically delivered Pb in the O
23    horizon at Camel's Hump Mountain in Vermont. They concluded that the forest floor response
24    time was between 60 and 150 years, depending on vegetation zone and elevation.  Using
25    206Pb:207Pb, they also demonstrated that some gasoline-derived Pb migrated out of the O horizon
26    and into the mineral soil in the deciduous vegetation zone on the mountain, while all of the
27    atmospheric Pb was retained in the upper 20 cm of the soil profile.
28          Researchers assessing the fate of atmospheric Pb in soils have also relied on repeated
29    sampling of soils and vegetation for total Pb. This technique works best when anthropogenic Pb
30    accounts for the vast majority of total Pb in a particular reservoir.  Johnson et al. (1995b), Yanai
31    et al.  (2004), and Friedland et al. (1992) used O horizon (forest floor) time-series data to evaluate

      December 2005                           8-45        DRAFT-DO NOT QUOTE OR CITE

-------
 1    the movement of gasoline-derived Pb in the soil profile.  These studies have concluded that the
 2    distribution of Pb in the upper soil horizons has changed over the past few decades. Yanai et al.
 3    (2004) documented a decline in Pb from the Oie horizon between the late 1970s to the early
 4    1990s in remote forest soils in New Hampshire. Johnson et al. (1995b) and Friedland et al.
 5    (1992) demonstrated that some fraction of Pb had moved from the O horizon to the mineral soil
 6    during the 1980s at Hubbard Brook and at selected remote sites in the northeastern United States,
 7    respectively.  Evans et al. (2005) demonstrated that Pb concentrations in the litter layer (fresh
 8    litter + Oi horizon) sampled in a transect from Vermont to Quebec decreased significantly
 9    between 1979 and 1996, reflecting a decrease in Pb deposition to forests and upper soil horizons
10    during that time period.  Miller et al. (1993) and Wang and Benoit (1997) used forest floor time-
11    series data to model the response time (e folding time, the time it takes a reservoir to decrease to
12    the l/e, (ca. 37%) of its original amount ) of Pb in the forest floor. Miller et al. (1993) calculated
13    O horizon response times of 17 years for the northern hardwood forest and 77 years in the
14    spruce-fir zone on Camel's Hump Mountain in Vermont. Wang and Benoit (1997) determined
15    that the O horizon would reach steady state with respect to Pb (1.3 jig g"1 Pb) by 2100. Both
16    suggested that the movement of organic particulates dominated Pb transport in the soil profile.
17
18    8.1.3.4   Inputs/Outputs of Atmospherically Delivered Lead in Terrestrial Ecosystems
19          The concentration of Pb in rainfall in the northeastern United States is on the order of
20    500 pg g"1 (Wang  et al., 1995). Assuming a precipitation rate of 1 m year"1, then loadings to
21    terrestrial ecosystems via wet deposition are currently on the order of 0.5 mg m"2 year"1.  Since
22    dry deposition may account for anywhere between 10 and 40% of total Pb deposition (Galloway
23    et al., 1982), total loadings to ecosystems are approximately 1 mg m"2 year"1.  This is a relatively
24    small annual flux of Pb if compared to the reservoir of approximately  0.5 to 4 g m"2 of gasoline-
25    derived Pb that is already in surface soils over much of the United States (Friedland et al., 1992;
26    Miller and Friedland, 1994; Erel and Patterson, 1994; Marsh and Siccama, 1997; Yanai et al.,
27    2004; Johnson et al., 2004; Evans et al., 2005). While vegetation can play an  important role in
28    sequestering Pb from rain and dry deposition (Russell et al., 1981), direct uptake of Pb from soils
29    by plants appears to be low (Klaminder et al., 2005).  High elevation areas, particularly those
30    near the base level of clouds often have higher burdens of atmospheric contaminants (Siccama,
31    1974).  A Pb deposition model by Miller and Friedland (1994) predicted 2.2 and 3.5 g Pb m"2
      December 2005                           8-46        DRAFT-DO NOT QUOTE OR CITE

-------
 1    deposition for the 20th century in the deciduous zone (600 m) and the coniferous zone (1000 m),
 2    respectively. More recently, Kaste et al. (2003) used radiogenic isotope measurements on the
 3    same mountain to confirm higher loadings at higher elevation.  They measured 1.3 and 3.4 g
 4    gasoline-derived Pb m"2 in the deciduous zone and coniferous zones, respectively. Higher
 5    atmospheric Pb loadings to higher elevations are attributed to (1) the higher leaf area of
 6    coniferous species,  which are generally more prevalent at high elevation; (2) higher rainfall at
 7    higher elevation; and (3) increased cloudwater impaction at high elevation (Miller et al.,  1993).
 8          Although inputs of Pb to ecosystems are currently low, Pb export from watersheds via
 9    groundwater and streams is substantially lower.  Therefore, even at current input levels,
10    watersheds are accumulating industrial Pb. Seeps and streams at the HBEF have Pb
11    concentrations on the order of 10 to 30 pg Pb g'1 (Wang et al., 1995). At a remote valley in the
12    Sierra Nevada, Pb concentrations in streamwaters were on the order of 15 pg Pb g'1 (Erel and
13    Patterson,  1994).  Losses of Pb from soil horizons are assumed to be via particulates (Dorr and
14    Munnich, 1989; Wang and Benoit, 1996, 1997). Tyler (1981) noted that Pb losses from an a
15    horizon in Sweden were influenced by season; with highest Pb fluxes being observed during
16    warm, wet months.  He suggested that DOC production and Pb movement were tightly linked.
17          Surface soils across the United States are enriched in Pb relative to levels expected from
18    solely natural geogenic inputs (Friedland et al., 1984; Francek, 1992; Erel and Patterson, 1994;
19    Marsh and Siccama, 1997; Yanai et al., 2004; Murray et al., 2004).  While some of this
20    contaminant Pb is attributed to paint,  salvage yards, and the use of Pb-arsenate as a pesticide in
21    localized areas (Francek, 1997), Pb contamination of surface soils is essentially ubiquitous
22    because of atmospheric pollution associated with the metal production industry and the
23    combustion of fossil fuels. Surface soils in Michigan, for example, typically range from 8 to
24    several hundred ppm Pb (Francek,  1992; Murray et al., 2004).  Soils collected and analyzed
25    beneath 50 cm in Michigan, however, range only from 4 to 60 ppm Pb (Murray et al., 2004).
26    In remote surface soils from the Sierra Nevada Mountains, litter and upper  soil horizons are 20 to
27    40 ppm Pb, and approximately 75% of this Pb has been attributed to atmospheric deposition
28    during the 20th century (Erel and Patterson, 1994).  Repeated sampling of the forest floor (O
29    horizon) in the northeastern United States demonstrates that the organic layer has retained much
30    of the Pb load deposited on ecosytems during the 20th century.  Total Pb deposition during the
31    20th century has been estimated at 1 to 3 g Pb m"2, depending on elevation and proximity to

      December 2005                            8-47        DRAFT-DO NOT QUOTE OR CITE

-------
 1    urban areas (Miller and Friedland, 1994; Johnson et al., 1995b). Forest floors sampled during
 2    the 1980s and 1990s, and in early 2000 had between 0.7 and 2 g Pb nT2 (Friedland et al., 1992;
 3    Miller and Friedland, 1994; Johnson et al., 1995b; Kaste et al., 2003; Yanai et al., 2004; Evans
 4    et al., 2005).  The pool of Pb in above- and below-ground biomass at the FffiEF is on the order of
 5    0.13 g Pb nT2 (Johnson et al., 1995b).
 6          The amount of Pb that has leached into mineral soil appears to be on the order of 20 to
 7    50% of the total anthropogenic Pb deposition. Kaste et al. (2003) and Miller and Friedland
 8    (1994) demonstrated that Pb loss from the forest floor at Camel's Hump Mountain in Vermont
 9    depended on elevation. While the mineral soil in the deciduous forest had between 0.4 and 0.5 g
10    Pb nT2 (out of 1 to 2 g Pb nT2 in the total soil profile), at higher elevations the thicker coniferous
11    forest floor retained more than 90% of the total Pb deposition (Kaste et al., 2003).  Johnson et al.
12    (1995b) determined that the forest floor at FfflEF in the mid-1980s had  about 0.75 g Pb nT2.
13    Compared to their estimated 20th-century atmospheric Pb deposition of 0.9 g Pb nT2, the forest
14    floor has retained 83% of the atmospheric Pb loadings (Johnson et al., 1995b).  Johnson et al.
15    (2004) noted that gasoline-derived Pb was a significant component of the labile Pb at the HBEF.
16    They calculated that Pb fluxes to the HBEF by atmospheric pollution were essentially equivalent
17    to the Pb released by mineral weathering over the past 12,000 years. Marsh and Siccama (1997)
18    used the relatively homogenous  mineral soils underneath formerly plowed land in New
19    Hampshire, Connecticut,  and Rhode Island to assess the depth-distribution of atmospheric Pb.
20    They reported that 65% of the atmospheric Pb deposited during the 20th century is in the mineral
21    soil and 35% is in the forest floor. At their remote study site in the Sierra Nevada Mountains,
22    Erel and Patterson (1994) reported that most of the anthropogenic Pb was associated with the
23    humus fraction of the litter layer and soils  sampled in the upper few  cm.
24          Atmospherically delivered Pb is probably present in ecosystems  in a variety of different
25    biogeochemical phases.  A combination of Pb adsorbtion processes and the precipitation of Pb
26    minerals will typically keep dissolved Pb species low in soil solution, surface waters, and
27    streams (Sauve et al., 2000b; Jackson et al., 2005).  While experimental and theoretical evidence
28    suggest that the precipitation of inorganic Pb phases and the adsorption of Pb on inorganic
29    phases can control the biogeochemistry of contaminant Pb (Nriagu,  1974; Ruby et al., 1994;
30    Jackson et al., 2005), the influence of organic matter on the biogeochemistry of Pb in terrestrial
31    ecosystems cannot be ignored in many systems. Organic matter can bind to Pb, preventing Pb

      December 2005                           8-48        DRAFT-DO NOT QUOTE OR CITE

-------
 1    migration and the precipitation of inorganic phases (Manceau et al., 1996; Xia et al., 1997; Lang
 2    and Kaupenjohann, 2003). As the abundance of organic matter declines in soil, Pb adsorption to
 3    inorganic soil minerals and the direct precipitation of Pb phases may dominate the
 4    biogeochemistry of Pb in terrestrial ecosystems (Ostergren et al., 2000a,b; Sauve et al., 2000b).
 5
 6    Conclusions
 7          Advances in technology since the 1986 Pb  AQCD have allowed for a quantitative
 8    determination of the mobility, distribution, uptake, and fluxes of atmospherically delivered Pb in
 9    ecosystems.  Among other things, these studies have shown that industrial Pb  represents a
10    significant fraction of total labile Pb in watersheds. Selective chemical extractions and
11    synchrotron-based X-ray studies have shown that  industrial Pb can be strongly sequestered by
12    organic matter and by secondary minerals such as clays and oxides of Al, Fe, and Mn.  Some of
13    these studies have provided compelling evidence that the biomineralization of Pb phosphates by
14    soil organisms can play an important role in the biogeochemistry of Pb.  Surface soils sampled
15    relatively recently demonstrate that the upper soil  horizons (O + A horizons) are retaining most
16    of the industrial Pb burden introduced to the systems during the 20th century.  The migration and
17    biological uptake of Pb in ecosystems is relatively low. The different biogeochemical behaviors
18    of Pb reported by various studies may be a result of the many different analytical techniques
19    employed, or they may be a result of natural variability in the behavior of Pb in different
20    systems.
21
22    8.1.4   Species Response/Mode of Action
23    8.1.4.1   Introduction
24          The 1986 Pb AQCD, Volume II (U.S. Environmental Protection Agency, 1986a)
25    reviewed the literature on the uptake of Pb  into plants, soil organisms, birds, and mammals.
26    This chapter expands upon the major conclusions  from the EPA (U.S. Environmental Protection
27    Agency, 1986a) related to those organisms. It summarizes the recent (since 1986) critical
28    research conducted on Pb uptake into terrestrial organisms (Section 8.1.4.2), mechanisms of
29    resistance to Pb toxicity (Section 8.1.4.3), the physiological effects of Pb (Section 8.1.4.4),
30    and, the factors that modify organism response to  Pb (Section 8.1.4.5). A summary is presented
      December 2005                           8-49        DRAFT-DO NOT QUOTE OR CITE

-------
 1    in Section 8.1.4.6. All concentrations are expressed as mg Pb/kg dw (dry weight) unless
 2    otherwise indicated.
 3          Areas of research that are not addressed include those that used irrelevant exposure
 4    conditions relative to airborne emissions of Pb (e.g., Pb shot, Pb paint, injection studies, studies
 5    conducted on mine tailings or using hyperaccumulator plants for phytoremediation, and studies
 6    conducted with hydroponic solutions) except when these studies provided critical information for
 7    understanding physiologic effects.
 8
 9    8.1.4.2  Lead Uptake
10          Since the 1986 Pb AQCD, there have been several studies that evaluated the uptake of Pb
11    into plants and invertebrates. The mechanisms associated with Pb uptake and translocation are
12    described in this section.  The methods used by the EPA (U.S. Environmental Protection
13    Agency, 2005b) to estimate Pb uptake into plants, earthworms, and small mammals as part of
14    Ecological Soil Screening Level (Eco-SSL) development are also presented.
15          The accumulation of Pb into the various tissues of consumers (birds and mammals) is
16    discussed only when it was described relative to either environmental concentrations or
17    organismal effects. Numerous  other monitoring studies measuring only the Pb  concentrations in
18    various tissues of birds and mammals were not included in this chapter; their data cannot be used
19    to develop an air standard without information on environmental concentrations or organismal
20    effects.
21
22    Lead Uptake into Plants
23          Plants take up Pb via their foliage and through their root systems (U.S. Environmental
24    Protection Agency, 1986a; Pahlsson, 1989).  Surface deposition of Pb onto plants may represent
25    a significant contribution to the total Pb in and on the plant, as has been observed for plants near
26    smelters and along roadsides (U.S. Environmental Protection Agency, 1986a).  The importance
27    of atmospheric deposition on above-ground plant Pb uptake is well-documented (Dalenberg and
28    Van Driel, 1990; Jones and Johnston, 1991; Angel ova et al., 2004).  Data examined from
29    experimental grassland plots in southeast England demonstrated that atmospheric Pb is a greater
30    contributor than soil-derived Pb in crop plants and grasses (Jones and Johnston, 1991).  A study
31    by Dalenberg and Van Driel (1990) showed that 75 to 95% of the Pb found in field-grown test

      December 2005                            8-50       DRAFT-DO NOT QUOTE OR CITE

-------
 1    plants (i.e., the leafy material of grass, spinach, and carrot; wheat grain; and straw) was from
 2    atmospheric deposition. Angel ova et al. (2004) found that tobacco grown in an industrial area
 3    accumulated significant amounts of Pb from the atmosphere, although uptake from soil was also
 4    observed. The concentration of Pb in tobacco seeds was linearly related to the concentration of
 5    Pb in the exchangeable and carbonate-bound fractions of soil, as measured using sequential
 6    extraction (Angelova et al., 2004).  Lead in soil is more significant when considering uptake into
 7    root vegetables (e.g., carrot, potato), since, as was noted in the 1986 Pb AQCD (U.S.
 8    Environmental Protection Agency, 1986a), most Pb remains in the roots of plants.
 9          There are two possible mechanisms (symplastic or apoplastic) by which Pb may enter the
10    root of a plant. The symplastic route is through the cell membranes of root hairs; this is the
11    mechanism of uptake for water and nutrients.  The apoplastic route is an extracellular route
12    between epidermal cells into the intercellular spaces of the root cortex.  Previously, Pb was
13    thought to enter the plant via the symplastic route, probably by transport mechanisms similar to
14    those involved in the uptake of calcium or other divalent cations (i.e., transpirational mass flow,
15    diffusion, or active transport).  However, it also had been speculated that Pb may enter the plant
16    via the apoplastic route  (U.S. Environmental Protection Agency, 1986a). Sieghardt (1990)
17    determined that the mechanism of Pb uptake was via the symplastic route only and that the
18    apoplastic pathway of transport was stopped in the primary roots by the endodermis. He studied
19    the uptake of Pb into two plants, Minuartia verna (moss sandwort) and Silene vulgaris (bladder
20    campion) that colonize metal-contaminated sites.  In the roots of both plants, Pb was found
21    mainly in the root cortex. Active ion uptake was required to transport the Pb into the stele and
22    then into the shoots of the plant (Sieghardt, 1990).
23          Although some plants translocate more Pb to the shoots than others, most Pb remains in
24    the roots of plants. Two mechanisms have been proposed to account for this relative lack of
25    translocation to the shoots:  (1) Pb may be deposited within root cell wall material, or (2) Pb may
26    be sequestered within root cell organelles  (U.S. Environmental Protection Agency, 1986a).
27    Pahlsson (1989) noted that plants can accumulate large quantities of Pb  from the soil but that
28    translocation to shoots and leaves is limited by the binding of Pb ions at root surfaces and cell
29    walls. In a study by Wierzbicka (1999), 21 different plant species were exposed to Pb2+ in the
30    form of Pb-chloride. The plant species included cucumber (Cucumis sativus), soy bean (Soja
31    hispida), bean (Phaseolus vulgaris), rapeseed (Brassica napus),  rye (Secale cereale), barley

      December 2005                             8-51        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Hordeum vulgare\ wheat (Triticum vulgare\ radish (Raphanus sativus), pea (Pisum sativum),
 2    maize (Zea mays), onion (Allium cepa\ lupine (Lupinus luteus), bladder campion (Silene
 3    vulgaris), Buckler mustard (Biscutella laevigata), and rough hawkbit (Leontodon hispidus).
 4    Although, the amount of Pb taken up by the plant varied with species, over 90% of absorbed Pb
 5    was retained in the roots.  Only a small amount of Pb was translocated (~2 to 4%) to the shoots
 6    of the plants. Lead in roots was present in the deeper layers of root tissues (in particular, the root
 7    cortex) and not only on the root surface. There was no correlation between Pb tolerance
 8    (measured as root mass increase expressed as a percentage of controls) and either root or shoot
 9    tissue concentrations (Wierzbicka, 1999).  The study by Wierzbicka (1999) was the first to report
10    that plants developing from bulbs, in this case the onion, were more tolerant to Pb than plants
11    developing from seeds. This tolerance was assumed to be related to the large amounts of Pb that
12    were transported from the roots and stored in the bulb of the plant (Wierzbicka, 1999).
13          Uptake of Pb from soil into plants was modeled as part of Eco-SSL development (U.S.
14    Environmental Protection Agency, 2005b). The relationship derived between Pb in the soil and
15    Pb in a plant was taken from Bechtel Jacobs Company (BJC) (1998) and is as follows:
16
18                          Ln(Cp) = 0.561 * Ln(Csoil) - 1.328                           (8-1)
20
21    where Cp is the concentration of Pb in the plant (dry weight) and Csoil is the concentration of Pb
22    in the soil. This equation recognizes that the ratio of Pb concentration in plant to Pb
23    concentration in soil is not constant.
24
25    Invertebrates
26          There was no clear evidence suggesting a differential uptake of Pb into different species
27    of earthworm (Lumbricus terrestris, Aporrectodea rosea, and A. caliginosd) collected around a
28    smelter site near Avonmouth, England (Spurgeon and Hopkin, 1996a).  This is in contrast to Pizl
29    and Josens (1995)  and Terhivuo et al. (1994) who found Aporrectodea spp. accumulated more
30    Pb than Lumbricus. The authors suggested that these differences could be due to different
31    feeding behaviors, as Lumbricus feeds on organic material and Apporectodea species are
32    geophagus, ingesting large amounts of soil during feeding.  The differences between species also
33    may be related to differing efficiencies in excretory mechanisms (Pizl and Josens, 1995).

      December 2005                           8-52        DRAFT-DO NOT QUOTE OR CITE

-------
 1    However, the interpretation of species difference is complicated by a number of potentially
 2    confounding variables, such as soil characteristics (e.g., calcium or other nutrient levels)
 3    (Pizl and Josens, 1995).
 4          The bioaccumulation of Pb from contaminated soil was tested using the earthworm
 5    Eisemafetida, and the amount of Pb accumulated did not change significantly until the
 6    concentration within soil reached 5000 mg/kg (Davies et al., 2003).  This coincided with the
 7    lowest soil concentrations at which earthworm mortality was observed. The ratio of the
 8    concentration of Pb in worms to the concentration in soil decreased from 0.03 at 100 mg/kg to
 9    0.001 at 3000 mg/kg, but then increased quickly to 0.02 at 5000 mg/kg. The authors concluded
10    that earthworms exhibit regulated uptake of Pb at levels of low contamination (<3000 mg/kg)
11    until a critical concentration is reached, at which point this mechanism breaks down, resulting in
12    unregulated accumulation and mortality.  This study was conducted using test methods where
13    soil was not allowed to equilibrate following the addition of Pb and prior to the addition of the
14    test organisms.  This may have resulted in an increased bioavailability and overestimated Pb
15    toxicity relative to actual environmental conditions (Davies et al., 2003).
16          Lock and Janssen (2002) and Bongers et al. (2004) found that Pb-nitrate was more toxic
17    than Pb-chloride to survival and reproduction of the springtail Folsomia Candida. However,
18    percolation (removal of the chloride or nitrate counted on) caused a significant decrease in Pb-
19    nitrate toxicity  such that there was no difference in toxicity once the counted on was removed
20    (Bongers et al., 2004). No change in toxicity was observed for Pb-chloride once the chloride
21    was removed from the soil. Bongers et al. (2004) suggested that the nitrate ion was more toxic
22    than the chloride ion to springtails.
23          Uptake of Pb from soil into earthworms was also modeled as part of Eco-SSL
24    development (U.S. Environmental Protection Agency, 2005b).  The relationship derived between
25    Pb in the soil and Pb in an earthworm was taken from Sample et al. (1999) and is as follows:
26
28                        Ln(Cworm) =  0.807*Ln(Csoil)-0.218                        (8-2)
30
31    where Cworm is the concentration of Pb in the earthworm (dry weight) and Csoil is the
32    concentration of Pb in the soil. This equation recognizes that the ratio of Pb concentration in
33    worm to Pb concentration in soil is not constant.
      December 2005                           8-53        DRAFT-DO NOT QUOTE OR CITE

-------
 1    Wildlife
 2          Research has been conducted to determine what Pb concentrations in various organs
 3    would be indicative of various levels of effects. For example, Franson (1996) compiled data to
 4    determine what residue levels were consistent with three levels of effects in Falconiformes (e.g.,
 5    falcons, hawks, eagles, kestrels, ospreys), Columbiformes (e.g., doves, pigeons), and Galliformes
 6    (e.g., turkey, pheasant, partridge, quail, chickens).  The three levels of effect were (1) subclinical,
 7    which are physiological effects only, such as the inhibition of 5-aminolevulinic acid dehydratase
 8    (ALAD; see Section 8.1.4.4); (2) toxic, a threshold level marking the initiation of clinical signs,
 9    such as anemia, lesions in tissues, weight loss,  muscular incoordination, green diarrhea, and
10    anorexia; and (3) compatible with death, an approximate threshold value associated with death in
11    field, captive, and/or experimental cases of Pb  poisoning. The tissue Pb levels associated with
12    these levels of effects are presented in Table 8-1.4.1.
13
14
Table 8-1.4.1. Tissue Lead
Order
Falconiformes
Subclinical
Toxic
Compatible with death
Columbiformes
Subclinical
Toxic
Compatible with death
Galliformes
Subclinical
Toxic
Compatible with death


Levels in Birds Causing Effects (taken from
Blood Liver
(ug/dL) (ppm wet wt.)

0.2-1.5 2-4
>1 >3
>5 >5

0.2-2.5 2-6
>2 >6
>10 >20

0.2-3 2-6
>5 >6
>10 >15


Franson, 1996)
Kidney
(ppm wet wt.)

2-5
>3
>5

2-20
>15
>40

2-20
>15
>50
15          Tissue residue levels below the subclinical levels in Table 8-1.4.1 should be considered
16   "background" (Franson, 1996). Levels in the subclinical range are indicative of potential injury
17   from which the bird would probably recover if Pb exposure was terminated. Toxic residues
     December 2005
8-54
DRAFT-DO NOT QUOTE OR CITE

-------
 1    could lead to death. Residues above the compatible-with-death threshold are consistent with
 2    Pb-poisoning mortality (Franson, 1996).  Additional information on residue levels for
 3    Passeriformes (e.g., sparrows, starlings, robins, cowbirds), Charadriiformes (e.g., gulls, terns),
 4    Gruiformes (e.g., cranes), Ciconiformes (e.g., egrets), Gaviformes (e.g., loons), and Strigiformes
 5    (e.g., owls) is available (Franson, 1996).  Scheuhammer (1989) found blood Pb concentrations of
 6    between 0.18 and 0.65 |ig/mL in mallards corresponded to conditions associated with greater
 7    than normal exposure to Pb but that should not be considered Pb  poisoning.
 8          Lead concentrations in various tissues of mammals also have been correlated with toxicity
 9    (Ma, 1996). The tissues commonly analysed for Pb are blood, liver, and kidney. Typical
10    baseline levels of blood Pb are approximately 4 to 8  |ig/dL for small mammals, and 2 to  6 |ig/dL
11    for mature cattle. Typical  baseline levels of Pb in liver are 1 to 2 mg/kg dw for small mammals.
12    Typical baseline levels of Pb in kidney are 0.2 to 1.5 mg/kg dw for mice and voles, but shrews
13    typically have higher baseline levels of 3 to 19 mg/kg dw. Ma (1996) concluded that Pb levels
14    less than 5 mg/kg dw in liver and 10 mg/kg dw in kidney were not associated with toxicity, but
15    that levels greater than 5 mg/kg dw in liver and greater than 15 mg/kg dw in kidney could be
16    taken as a chemical biomarker of toxic exposure to Pb in mammals. Humphreys (1991)  noted
17    that the concentrations of Pb in liver and kidney can be elevated in  animals with normal blood Pb
18    concentrations (and without exhibiting clinical signs of Pb toxicity), because Pb persists  in these
19    organs longer than in blood.
20          Uptake of Pb from soil into small mammals was also modeled as part of Eco-SSL
21    development (U.S. Environmental Protection Agency, 2005b). The relationship derived between
22    Pb in the soil and Pb in the whole-body of a small mammal was taken from Sample et al. (1998)
23    and is  as follows:
25
27                      Ln(Cmammal) = 0.4422 * Ln(Csoil) + 0.0761                       (8'3)
29
30    where Cmammal is the concentration  of Pb in small mammals (dry weight) and Csoil is the
31    concentration of Pb in the  soil. This equation recognizes that the ratio of Pb concentration in
32    small mammals to Pb concentration in soil is not constant.
33
     December 2005                            8-55        DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.4.3  Resistance Mechanisms
 2          Many mechanisms related to heavy metal tolerance in plants and invertebrates have been
 3    described, including avoidance (i.e., root redistribution, food rejection), exclusion (i.e., selective
 4    uptake and translocation), immobilization at the plant cell wall, and excretion (i.e., foliar
 5    leakage, moulting) (Tyler et al., 1989; Patra et al., 2004). The following section reviews the
 6    recent literature on the resistance mechanisms of plants and invertebrates through mitigation of
 7    Pb (1) toxicity (8.1.4.3.1) or (2) exposure (8.1.4.3.2).
 8
 9    8.1.4.3.1   Detoxification Mechanisms
10          Lead sequestration in cell walls may be the most important detoxification mechanism in
11    plants. Calcium may play a role in this detoxification by regulating internal Pb concentrations
12    through the formation of Pb-containing precipitates in the cell wall (Antosiewicz, 2005). Yang
13    et al. (2000) screened 229 varieties of rice (Oryza sativa) for tolerance or sensitivity to Pb and
14    found that the oxalate content in the root and root exudates was increased in Pb-tolerant varieties.
15    The authors suggested that the oxalate reduced Pb bioavailability, and that this was an important
16    tolerance mechanism (Yang et al., 2000). Sharma et al. (2004) found Pb-sulfur and Pb-sulfate in
17    the leaves, and Pb-sulfur in the roots of Sesbania drummondii (Rattlebox Drummond), a Pb
18    hyperaccumulator plant grown in Pb-nitrate solution. They hypothesized that these sulfur
19    ligands were indicative of glutathione and phytochelatins, which play a role in heavy metal
20    homeostasis and detoxification (Sharma et al., 2004).
21          Sea pinks (Armeria maritima) grown on a metal-contaminated site (calamine spoils more
22    than 100 years old) accumulated 6x the concentrations of Pb in brown (dead and withering)
23    leaves than green leaves (Szarek-Lukaszewska et al., 2004). The concentration of Pb in brown
24    leaves was similar to that in roots.  This greater accumulation of Pb into older leaves was not
25    observed in plants grown hydroponically in the laboratory.  The authors hypothesized that this
26    sequestering of Pb into the oldest leaves was a detoxification mechanism (Szarek-Lukaszewska
27    et al., 2004).
28          Terrestrial invertebrates also mitigate Pb toxicity. Wilczek et al. (2004) studied two
29    species of spider, the web-building Agelena labyrinthica and the active hunter wolf spider
30    Pardosa lugubris. The activity of metal detoxifying enzymes  (via the glutathione metabolism
31    pathways) was greater in A. labyrinthica and in females of both species (Wilczek et al., 2004).

      December 2005                            8-56         DRAFT-DO NOT QUOTE OR CITE

-------
 1          Marinussen et al. (1997) found that earthworms can excrete 60% of accumulated Pb very
 2    quickly once exposure to Pb-contaminated soils has ended. However, the remainder of the body
 3    burden is not excreted, possibly due to the storage of Pb in waste nodules that are too large to be
 4    excreted (Hopkin,  1989). Gintenreiter et al. (1993) found that Lepidoptera larvae (in this case,
 5    the gypsy moth Lymantria dispar) eliminated Pb, to some extent, in the meconium (the fluid
 6    excreted shortly after emergence from the chrysalis).
 7          Lead, in the form of pyromorphite (Pbs^O^Cl), was localized in the anterior pharynx
 8    region of the nematode Ceanorhabditis elegans (Jackson et al., 2005). The authors hypothesized
 9    that the nematode may detoxify Pb via its precipitation into pyromorphite, which is relatively
10    insoluble (Jackson et al., 2005).
11
12    8.1.4.3.2   Avoidance Response
13          Studies with soil invertebrates hypothesize that these organisms may avoid soil with high
14    Pb concentrations. For example, Bengtsson et al. (1986) suggested that the lower Pb
15    concentrations in earthworm tissues may be a result of lowered feeding activity of worms at
16    higher Pb concentrations in soil.
17
18    8.1.4.4  Physiological Effects of Lead
19          Several studies have measured decreased blood ALAD activity in birds and mammals
20    exposed to Pb (U.S. Environmental Protection Agency, 1986a). Recent studies on the
21    physiological effects of Pb to consumers have focused on heme synthesis (as measured by
22    ALAD activity and protoporphyrin concentration), lipid peroxidation, and production of fatty
23    acids.  Effects on growth are covered in Section 8.1.5.
24          Biochemically, Pb adversely affects hemoglobin synthesis in birds and mammals.  Early
25    indicators of Pb exposure in birds and mammals include decreased blood ALAD concentrations
26    and increased protoporphyrin IX activity.  The effects of Pb on blood parameters and the use of
27    these parameters as sensitive biomarkers of exposure has been well documented (Eisler, 1988;
28    U.S. Environmental Protection Agency, 2005a). However, the linkage between these
29    biochemical indicators and ecologically-relevant effects is less well understood.  Low-level
30    inhibition of ALAD is not generally considered a toxic response, because this enzyme is thought
      December 2005                            8-57        DRAFT-DO NOT QUOTE OR CITE

-------
 1    to be present in excess concentrations; rather, it may simply indicate that the organism has
 2    recently been exposed to Pb (Henny et al., 1991).
 3          Schlick et al. (1983) studied ALAD inhibition in mouse bone marrow and erythrocytes.
 4    They estimated that an absorbed dose of between 50 and 100 jig Pb-acetate/kg body weight per
 5    day would result in long-term inhibition of ALAD.
 6          Beyer et al. (2000) related blood Pb to sublethal effects in waterfowl along the Coeur
 7    d'Alene River near a mining site in Idaho. The sublethal effects measured included, among
 8    others, red blood cell ALAD activity and protoporphyrin levels in the blood. As found in other
 9    studies, ALAD activity was the most sensitive indicator of Pb exposure, decreasing to 3% of the
10    reference value at a blood Pb concentration of 0.68 mg/kg ww (wet weight). Protoporphyrin
11    concentrations showed a 4.2-fold increase at this same concentration.
12          Henny et al. (1991) studied osprey along the Coeur d'Alene River.  There were no
13    observations of death, behavioral abnormalities, or reduced productivity related to Pb exposure,
14    although inhibition of blood ALAD and increased protoporphyrin concentrations were measured
15    in ospreys. Henny et al. (1991) hypothesized that no impacts to osprey were observed, even
16    though swan mortality was documented in the area because swans feed at a lower trophic level
17    (i.e., Pb does not biomagnify, and thus is found at higher concentrations in lower trophic level
18    organisms).
19          Hoffman et al. (2000a) also studied the effects of Coeur d'Alene sediment on waterfowl,
20    focusing on mallard ducklings for 6 weeks after hatching. The study revealed that a 90%
21    reduction in ALAD activity and a greater than 3-fold increase in protoporphyrin concentration
22    occurred when blood Pb reached  a concentration of 1.41 mg/kg ww as a result of the ducklings
23    being fed a diet composed of 12% sediment (3449 mg/kg Pb). Those ducklings fed a diet
24    composed of 24%  sediment were found to have a mean blood Pb concentration of 2.56 mg/kg
25    ww and a greater than 6-fold increase in protoporphyrin concentration.  Hoffman et al. (2000b)
26    also studied Canada Geese (Branta canadensis) goslings in a similar fashion. The results
27    revealed that, while blood Pb concentrations in goslings were approximately half (0.68 mg/kg
28    ww) of those found in ducklings under the same conditions (12% diet of 3449 mg/kg sediment
29    Pb), goslings showed an increased sensitivity to Pb exposure. Goslings experienced a 90%
30    reduction in ALAD activity and a 4-fold increase in protoporphyrin concentration, similar to
31    conditions found in the ducklings, although blood Pb concentrations were half those found in the

      December 2005                            8-58        DRAFT-DO NOT QUOTE OR CITE

-------
 1    ducklings. More serious effects were seen in the goslings when blood Pb reached 2.52 mg/kg,
 2    including decreased growth and mortality.
 3          Redig et al. (1991) reported a hawk LOAEL of 0.82 mg/kg-day for effects on heme
 4    biosynthetic pathways. Lead dosages as high as 1.64 to 6.55 mg/kg-day caused neither mortality
 5    nor clinical signs of toxicity. A dose of 6.55 mg/kg-day resulted in blood Pb levels of
 6    1.58 |ig/ml. There were minimal changes in immune function (Redig et al., 1991).
 7          Repeated oral administration of Pb resulted in biochemical alterations in broiler chickens
 8    (Brar et al., 1997a,b). At a dose of 200 mg/kg-day Pb-acetate, there were significant increases in
 9    plasma levels of uric acid and creatinine and significant declines in the levels of total proteins,
10    albumin, glucose, and cholesterol. Brar et al. (1997a) suggested that increased uric acid and
11    creatinine levels could be due to an accelerated rate of protein catabolism and/or kidney damage.
12    They also suggested that the decline in plasma proteins and albumin levels may be caused by
13    diarrhea and liver dysfunction due to the Pb exposure. Brar et al.  (1997b) also found that
14    significant changes in plasma enzymes may be causing damage to other organs.
15          Lead can cause an increase in tissue lipid peroxides and changes in glutathione
16    concentrations, which may be related to peroxidative damage of cell membranes (Mateo and
17    Hoffman, 2001).  There are species-specific differences in resistance to oxidative stress (lipid
18    peroxidation), which may explain why Canada geese are more sensitive to Pb poisoning than
19    mallards (Mateo and Hoffman, 2001).  Lead also caused an increase in the production of the fatty
20    acid arachidonic acid, which has been associated with changes in bone formation and immune
21    response (Mateo et al., 2003a). The effects observed by Mateo et al.  (2003a,b) were associated
22    with very high concentrations of Pb in the diet (1840 mg Pb/kg diet), much higher than would be
23    found generally in the environment, and high enough that birds decreased their food intake.
24          Lead also induces lipid peroxidation in plants. Rice plants exposed to a highly toxic level
25    of Pb (1000 jiM in nutrient solution) showed elevated levels of lipid peroxides, increased activity
26    of superoxide dismutase, guaiacol peroxidase, ascorbate peroxidase, and glutatione reductase
27    (Verma and Dubey, 2003). The elevated levels of these enzymes suggest the plants may have an
28    antioxidative defense mechanism against oxidative injury caused by Pb (Verma and Dubey,
29    2003).
30
      December 2005                           8-59        DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.4.5  Factors that Modify Organism Response
 2          Research has demonstrated that Pb may affect survival, reproduction, growth,
 3    metabolism, and development in a wide range of species. These effects may be modified by
 4    chemical, biological, and physical factors.  The factors that modify responses of organisms to Pb
 5    are described in the following sections.
 6
 7    8.1.4.5.1   Genetics
 8          Uptake and toxicity of Pb to plants are influenced strongly by the type of plant.  Liu et al.
 9    (2003) found that Pb uptake and translocation by rice plants differed by cultivar (a cultivated
10    variety of plant produced by selective breeding) but was not related to genotype.  Twenty
11    cultivars were tested from three genotypes. The differences in Pb concentrations among
12    cultivars were smallest when  comparing concentrations in the grains at the ripening stage. This
13    study also found that toxicity  varied by cultivar; at 800 mg Pb/kg soil, some cultivars were
14    greatly inhibited, some were significantly improved, and others showed no change.
15          Dearth et al. (2004) compared the response of Fisher 344 (F344) rats and Sprague-Dawley
16    (SD) rats to exposure via gavage to 12 mg Pb/mL as Pb-acetate. Blood Pb levels in the F344
17    dams were higher than those of the SD dams. Lead delayed the timing of puberty and
18    suppressed hormone levels in F344 offspring.  These effects were not observed in the offspring
19    of SD rats, even when the dose was doubled.  The authors conclude that F344 rats are more
20    sensitive to Pb (Dearth et al., 2004).
21
22    8.1.4.5.2   Biological Factors
23          Several biological factors may influence Pb uptake and organisms response, including
24    organism age, sex, species, feeding guild, and, for plants, the presence of mycorrhizal fungi.
25    Monogastric animals are more sensitive to Pb than ruminants (Humphreys,  1991).
26          Younger organisms may be more susceptible to Pb toxicity (Eisler, 1988; Humphreys,
27    1991). Nestlings are more sensitive to the effects of Pb than  older birds, and young altricial birds
28    (species unable to  self-regulate body heat at birth, such as songbirds), are considered more
29    sensitive than precocial birds  (species that have a high degree of independence at birth,  such as
30    quail, ducks, and poultry) (Scheuhammer, 1991).
      December 2005                           8-60        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Sex can also have an effect on the accumulation of Pb by wildlife (Eisler, 1988).  Female
 2    birds accumulate more Pb than males (Scheuhammer, 1987; Tejedor and Gonzalez, 1992).
 3    These and other authors have related this to the increased requirement for calcium in laying
 4    females.
 5          Different types of invertebrates accumulate different amounts of Pb from the environment
 6    (U.S. Environmental Protection Agency, 1986a).  There may be species-and sex-specific
 7    differences in accumulation of Pb into invertebrates, specifically arthropods.  This has been
 8    shown by Wilczek et al. (2004) who studied two species of spider, the web-building
 9    A. labyrinthica and the active hunter wolf spider P.  lugubris.  The body burdens of Pb in the
10    wolf spider were higher than in the web-building spider, and this may be due to the more
11    effective use of glutathione metabolism pathways in A. labyrinthica. Body burdens of females
12    were lower than those of males in both species. This was also observed in spiders by Rabitsch
13    (1995b). Females are thought to be able to detoxify and excrete excess metals more effectively
14    than males (Wilczek et al., 2004). Lead accumulation has been measured in numerous species of
15    arthropods with different feeding strategies. Differences were observed between species
16    (Janssen and Hogervorst, 1993; Rabitsch, 1995a) and depending upon sex (Rabitsch, 1995a),
17    developmental stage (Gintenreiter et al., 1993; Rabitsch, 1995a), and season (Rabitsch, 1995a).
18          Uptake of Pb may be enhanced by symbiotic associations between plant roots and
19    mycorrhizal fungi.  Similar to the mechanism associated with increased uptake of nutrients,
20    mycorrhizal fungi also may cause an increase in the uptake of Pb by increasing the surface area
21    of the roots, the ability of the root to absorb particular ions, and the transfer of ions through the
22    soil (U.S. Environmental Protection Agency, 1986a).  There have been contradictory results
23    published in the literature regarding the influence of mycorrhizal organisms on the uptake and
24    toxicity of Pb to plants (see review in Pahlsson, 1989). Lin et al. (2004) found that the
25    bioavailability of Pb increased in the rhizosphere of rice plants, although the availability varied
26    with Pb concentration in soil. Bioavailability was measured as the soluble plus exchangeable Pb
27    fraction from sequential extraction  analysis. The authors hypothesized that the enhanced
28    solubility of Pb may be due to a reduced pH in the rhizosphere or, more likely, the greater
29    availability of organic ligands, which further stimulates microbial growth (Lin et al., 2004).
30    Increased bioavailability of Pb in soil may increase  the uptake of Pb into plants, although
31    this was not assessed by Lin et al. (2004). However, Dixon (1988) found that red oak

      December 2005                           8-61        DRAFT-DO NOT QUOTE OR CITE

-------
 1    (Quercus rubra) seedlings with abundant ectomycorrhizae had lower Pb concentrations in their
 2    roots than those seedlings without this fungus, although only at the 100 mg Pb/kg sandy loam
 3    soil concentration (no differences were found at lower Pb concentrations). Lead in soil also was
 4    found to be toxic to the ectomycorrhizal fungi after 16 weeks of exposure to 50 mg Pb/kg or
 5    more (Dixon, 1988). Malcova and Gryndler (2003) showed that maize root exudates from
 6    mycorrhizal fungi can ameliorate heavy metal toxicity until a threshold metal concentration was
 7    surpassed.  This may explain the conflicting results in the past regarding the uptake and toxicity
 8    of Pb to plants with mycorrhizal fungi.
 9          The type of food eaten is a major determinant of Pb body burdens in small mammals, with
10    insectivorous animals accumulating more Pb than herbivores or granivores (U.S. Environmental
11    Protection Agency,  1986a). In fact, the main issue identified by the EPA (U.S. Environmental
12    Protection Agency,  1986a) related to invertebrate uptake of Pb was not toxicity to the
13    invertebrates, but accumulation of Pb to levels that may be toxic to their consumers.  Several
14    authors suggest that shrews are a good indicator of metal contamination, because they tend to
15    accumulate higher levels of metals than herbivorous small mammals (see data summary in
16    Sample et al. (1998)). Shrews accumulate higher levels of metals in contaminated habitats,
17    because their diet mainly consists of detritivores  (i.e., earthworms) and other soil invertebrates in
18    direct contact with the soil (Beyer et al., 1985).
19
20    8.1.4.5.3   Physical/Environmental Factors
21    Plants
22          The uptake and distribution of Pb into higher plants from the soil is affected by various
23    chemical and physical factors including the chemical form of Pb, the presence of other metal
24    ions,  soil type,  soil pH, cation exchange capacity (CEC), the amount of Fe/Mn-oxide films
25    present, organic matter content, temperature, light, and nutrient availability. A small fraction of
26    Pb in soil may be released to the soil moisture, which is then available to be taken up by plants
27    (U.S. Environmental Protection Agency, 1986a).
28          The form of Pb has an influence on its toxicity to plants. For example, Pb-oxide is less
29    toxic than more bioavailable forms such as Pb-chloride or Pb-acetate.  In a study by Khan and
30    Frankland (1983), radish plants were exposed to Pb-oxide and Pb-chloride in a loamy sand at pH
31    5.4, in a 42-day study. In a tested concentration range of 0 to 5000 mg/kg, root growth was

      December 2005                            8-62        DRAFT-DO NOT QUOTE OR CITE

-------
 1    inhibited by 24% at 500 mg/kg for Pb-chloride and an ECso of 2400 mg/kg was calculated from a
 2    dose-response curve. Plant growth ceased at 5000 mg/kg and shoots exhibited an ECso of
 3    2800 mg/kg. For Pb-oxide exposure (concentration range of 0 to 10,000 mg/kg), reported results
 4    indicate an ECso of 12,000 mg/kg for shoot growth and an ECso of 10,000 mg/kg for root growth.
 5    There was no effect on root growth at 500 mg/kg and a 26% reduction at 1000 mg/kg Pb oxide.
 6          Soil pH is the most influential soil property with respect to uptake and accumulation of Pb
 7    into plant species.  This is most likely due to increased bioavailability of Pb created by low soil
 8    pH. At low soil pH conditions, markedly elevated Pb toxicity was reported for red spruce
 9    (P. rubens) (Seller and Paganelli, 1987).  At a soil pH of 4.5, ryegrass (Lolium hybridum)
10    and oats (Avena sativd) had significantly higher Pb concentrations after 3 months of growth
11    compared to plants grown at pH 6.4 (Allinson and Dzialo, 1981).
12
13    Invertebrates
14          The uptake  of Pb into invertebrates depends on the physical environment and parameters
15    such as pH, calcium concentration, organic matter content, and CEC.  Greater accumulation is
16    found generally when the soil pH or organic content is lower (U.S. Environmental Protection
17    Agency,  1986a).
18          Soil pH has a significant influence on uptake of Pb into invertebrates.  Peramaki et al.
19    (1992) studied the  influence of soil pH on uptake into the earthworm Aporrectodea caliginosa.
20    Lead accumulation was lowest at the highest pH values, but there was no statistical difference
21    due to variability in the data.  Variability in the response also was found by Bengtsson et al.
22    (1986), who reared earthworms (Dendrobaena rubida) in acidified soils at pH 4.5, 5.5, or 6.5.
23    Lead uptake into worms was pH-dependent, although the highest concentrations were not always
24    found at the lowest pH. There was no clear relationship between Pb concentration in cocoons
25    and soil pH, and Pb concentrations were higher in the hatchlings than in the cocoons.  As has
26    been reported  in many  other studies (Neuhauser et al., 1995), concentration factors (ratio of Pb in
27    worm to Pb in soil) were lower at higher Pb concentrations in soil.  The authors attribute some of
28    this to a lowered feeding activity in worms at higher Pb  concentrations (Bengtsson et al., 1986).
29         Beyer et al.  (1987) and Morgan and Morgan (1988) recognized that other factors beyond
30    soil pH could influence the uptake of Pb into earthworms, which may be the cause of the
31    inconsistencies reported by several authors. Both studies evaluated worm uptake of Pb relative

      December 2005                           8-63        DRAFT-DO NOT QUOTE OR CITE

-------
 1    to pH, soil calcium concentration, and organic matter content.  Morgan and Morgan (1988) also
 2    considered CEC, and Beyer et al. (1987) considered concentrations of phosphorus, potassium, or
 3    magnesium in soil. Both studies found that calcium concentrations in soil were correlated with
 4    soil pH. Morgan and Morgan (1988) also found that CEC was correlated with percentage
 5    organic matter.  Soil pH (coupled with CEC) and soil calcium were found to play significant
 6    roles in the uptake of Pb into  worms (Beyer et al., 1987; Morgan and Morgan, 1988).  Beyer
 7    et al. (1987) noted that concentrations of phosphorus in soil had no effect.
 8
 9    8.1.4.5.4  Nutritional Factors
10          Diet is a significant modifier of Pb absorption and of toxic effects in many species of
11    birds and mammals (Eisler, 1988).  Dietary deficiencies in calcium, zinc, iron, vitamin E, copper,
12    thiamin, phosphorus, magnesium, fat, protein, minerals, and ascorbic acid increased Pb
13    absorption and its toxic effects (Eisler,  1988).
14          Mateo et al. (2003b) studied intraspecies sensitivity to Pb-induced oxidative stress, by
15    varying the vitamin E content of mallard diets.  Vitamin E can protect against peroxidative
16    damage and was found to decrease the lipid peroxidation in nerves of birds; however, it did not
17    alleviate any sign of the Pb poisoning.  The authors hypothesize that inhibition of antioxidant
18    enzymes and interaction with sulfhydryl groups of proteins may have a greater influence on Pb
19    toxicity than lipid peroxidation (Mateo et al., 2003b). The effects observed by Mateo et al.
20    (2003b) were associated with very high concentrations of Pb in diet (1840 mg Pb/kg diet), much
21    higher than would be found generally in the environment, and high enough that the birds
22    decreased their food intake.
23          Mallard ducklings were exposed to Pb-contaminated sediment and either a low nutrition
24    or optimal nutrition diet (Douglas-Stroebel et al., 2005).  Lead exposure combined with a
25    nutritionally-inferior diet caused more changes in behavior (as measured by time bathing,
26    resting, and feeding) than Pb  exposure or low-nutrition diet alone. These effects may be due to
27    the low-nutrition diet being deficient in levels of protein, amino acids, calcium, zinc, and other
28    nutrients.
29          Zebra finches (Taeniopygia guttata) were exposed to Pb-acetate via drinking water at
30    20 mg/L for 38 days, along with either a low- or high-calcium diet (Snoeijs et al., 2005).  Lead
31    uptake  into tissues was enhanced by a low-calcium diet.  Lead did not affect body weight,

      December 2005                           8-64        DRAFT-DO NOT QUOTE OR CITE

-------
 1    hematocrit, or adrenal stress response. Lead suppressed the humoral immune response only in
 2    females on a low-calcium diet, suggesting increased susceptibility of females to Pb (Snoeijs
 3    et al., 2005).
 4
 5    8.1.4.5.5   Interactions with Other Pollutants
 6          Lead can interact with other pollutants to exert toxicity in an antagonistic (less than
 7    additive), independent, additive, or synergistic (more than additive) manner. Concurrent
 8    exposure to Pb and additional pollutant(s) can affect the ability of plants to uptake Pb or the
 9    other pollutant.  However, the uptake and toxic response of plants, exposed to Pb combined with
10    other metals, is inconsistent (Pahlsson, 1989).  Therefore, no generalizations can be made about
11    the relative toxicity of metal mixtures. For example, An et al. (2004) conducted acute, 5-day
12    bioassays on cucumber exposed to Pb, Pb + copper, Pb + cadmium, or Pb + copper + cadmium
13    in a sandy loam soil of pH 4.3. Shoot and root growth were measured. Depending on the tissue
14    and metal combination, additivity, synergism, or antagonism was observed in the responses to
15    these metals. In fact, the response in roots was  not consistent with the response in shoots for the
16    binary mixtures. However, the combined effects were greater in the roots than the shoots, which
17    may be explained by the tendency for Pb and other heavy metals to be retained in the roots of
18    plants.  In addition, the pattern of metal bioaccumulation into plant tissue did not always
19    correlate with the toxic response. However, antagonism was observed in the response of roots
20    and shoots exposed to all three metals, and this  was reflected in the decreased accumulation of
21    metals into plant tissues.  The authors hypothesized that this may be  due to the formation of less
22    bioavailable metal complexes (An et al., 2004).
23          He et al. (2004) found that selenium and zinc both inhibited the uptake of Pb into Chinese
24    cabbage (Brassica rapci) and lettuce (Lactuca sativa). Zinc applied at 100 mg/kg or selenium
25    applied at 1 mg/kg decreased the uptake of Pb (present in soil at 10 mg/kg as Pb-nitrate) into
26    lettuce by 15% and 20%, respectively, and into Chinese cabbage by 23 and 20%, respectively.
27          Selenium compounds were evaluated to determine whether they could change the
28    inhibition of ALAD in liver, kidney, or brain of mice exposed to Pb-acetate (Perottoni et al.,
29    2005).  Selenium did not affect the inhibition of ALAD in the kidney or liver, but it did reverse
30    the ALAD inhibition in mouse brain.
      December 2005                           8-65        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Co-occurrence of cadmium with Pb resulted in reduced blood Pb concentrations in rats
 2    (Garcia and Corredor, 2004).  The authors hypothesized that cadmium may block or antagonize
 3    the intestinal absorption of Pb, or the metallothionein induced by cadmium may sequester Pb.
 4    However, this was not observed in pigs, where blood Pb concentrations were greater when
 5    cadmium was also administered (Phillips et al., 2003).  The effect on growth rate also was
 6    additive when both metals were given to young pigs (Phillips et al., 2003).
 7
 8    8.1.4.6  Summary
 9          The current document expands upon and updates knowledge related to the uptake,
10    detoxification, physiological effects, and modifying factors of Pb toxicity to terrestrial
11    organisms.
12
13    Surface Deposition onto Plants
14          Recent work (Dalenberg and Van Driel, 1990; Jones and Johnston, 1991; Angelova et al.,
15    2004) has supported previous results and conclusions that surface deposition of Pb onto above-
16    ground vegetation from airborne sources may be significant (U.S. Environmental Protection
17    Agency, 1986a).  Similarly, it has been well documented previously that Pb in soil also is taken
18    up by plants, although most remains in the roots, there is little translocation to shoots, leaves, or
19    other plant parts (U.S. Environmental Protection Agency, 1986a). More recent work continues
20    to support this finding (Sieghardt, 1990), and one study found increased tolerance in species with
21    bulbs, possibly due to the storage of Pb in the bulb (Wierzbicka, 1999).
22
23    Uptake Mechanism into Plants
24          Lead was thought previously to be taken up by plants via the  symplastic route (through
25    cell membranes), although it was unknown whether some Pb also may be taken up via the
26    apoplastic route (between cells) (U.S.  Environmental Protection Agency, 1986a).  Recent work
27    has shown that the apoplastic route of transport is stopped in the primary roots by the endodermis
28    (Sieghardt, 1990), supporting the previous conclusion that the symplastic route is the most
29    significant route of tranport into plant cells.
30
      December 2005                           8-66        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Species Differences in Uptake into Earthworms
 2          Different species of earthworm accumulated different amounts of Pb, and this was not
 3   related to feeding strategy (U.S. Environmental Protection Agency, 1986a). This is supported by
 4   recent work, which has shown Aporrectodea accumulated more than Lumbricus (Terhivuo et al.,
 5    1994; Pizl and Josens, 1995),  although this is not consistently observed (Spurgeon and Hopkin,
 6    1996a).
 7
 8   Speciation and Form of Lead
 9          Recent work supports previous conclusions that the form of metal tested, and its
10   speciation in soil, influence uptake and toxicity to plants and invertebrates (U.S. Environmental
11   Protection Agency, 1986a). The oxide form is less toxic that the chloride or acetate forms,
12   which are less toxic that the nitrate form of Pb  (Khan and Frankland, 1983; Lock  and Janssen,
13   2002; Bongers et al., 2004). However, these results must be interpreted with caution, as the
14   counterion (e.g., the nitrate  ion) may be contributing to the observed toxicity (Bongers et al.,
15   2004).
16
17   Detoxification in Plants
18          Lead may be deposited in root cell walls as a detoxification mechanism (U.S.
19   Environmental Protection Agency, 1986a), and this may be influenced by calcium concentrations
20   (Antosiewicz, 2005).  Yang et al. (2000) suggested that the oxalate content in root and root
21   exudates reduced the bioavailability of Pb in soil, and that this was an important tolerance
22   mechanism.  Other hypotheses put forward recently include the presence of sulfur ligands
23   (Sharma et al., 2004) and the sequestration of Pb in old leaves (Szarek-Lukaszewska et al., 2004)
24   as detoxification mechanisms.
25
26   Detoxification in Invertebrates
27          Lead detoxification has not been studied extensively in invertebrates.  Glutathione
28   detoxification enzymes were measured in two species of spider (Wilczek et al., 2004). Lead may
29   be stored in waste nodules in earthworms (Hopkin, 1989) or as pyromorphite in the nematode
30   (Jackson et al., 2005).
31

     December 2005                            8-67        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Physiological Effects
 2          The effects on heme synthesis (as measured by ALAD activity and protoporphyrin
 3   concentration, primarily) have been well-documented (U.S. Environmental Protection Agency,
 4   1986a) and continue to be studied (Schlick et al., 1983; Scheuhammer, 1989; Henny et al., 1991;
 5   Redig et al., 1991; Beyer et al., 2000; Hoffman et al., 2000a,b).  However, Henny et al. (1991)
 6   caution that changes in ALAD and other enzyme parameters are not always related to adverse
 7   effects, but simply indicate exposure.  Other effects on plasma enzymes, which may damage
 8   other organs, have been reported (Brar et al., 1997a,b). Lead also may cause lipid peroxidation
 9   (Mateo and Hoffman, 2001), which may be alleviated by vitamin E, although Pb poisoning may
10   still result (Mateo et al., 2003b). Changes in fatty acid production have been reported, which
11   may influence immune response and bone formation (Mateo et al., 2003a).
12
13   Response Modification
14          Genetics, biological factors, physical/environmental factors, nutritional  factors, and other
15   pollutants can modify terrestrial organism response to Pb. Fisher 344 rats were found to be more
16   sensitive to Pb than  Sprague-Dawley rats (Dearth et al., 2004). Younger animals are more
17   sensitive than older  animals (Eisler, 1988; Scheuhammer, 1991), and females generally are more
18   sensitive than males (Scheuhammer, 1987; Tejedor and Gonzalez, 1992; Snoeijs et al., 2005).
19   Monogastric animals are more sensitive than ruminants (Humphreys,  1991). Insectivorous
20   mammals may be more exposed to Pb than herbivores (Beyer et al., 1985; Sample et al.,  1998),
21   and higher tropic-level consumers may be less exposed than lower trophic-level organisms
22   (Henny etal., 1991). Nutritionally-deficient diets (including low calcium) cause increased
23   uptake of Pb (Snoeijs et al., 2005) and greater toxicity (Douglas-Stroebel et al., 2005) in  birds.
24          Mycorrhizal  fungi may ameliorate Pb toxicity until a threshold is surpassed (Malcova and
25   Gryndler, 2003), which may explain why some studies show increased uptake  into plants (Lin
26   et al., 2004) while others show no difference or less uptake (Dixon, 1988).  Lower soil pH
27   generally increases uptake of Pb into plants and soil invertebrates. However, calcium content,
28   organic matter content, and cation exchange capacity of soils also have had a significant
29   influence on uptake of Pb into plants and invertebrates (Beyer et al., 1987; Morgan and Morgan,
30   1988).
     December 2005                            8-68       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Interactions of Pb with other metals are inconsistent, depending on the endpoint
 2    measured, the tissue analyzed, the animal species, and the metal combination (Phillips et al.,
 3    2003; An et al., 2004; He et al., 2004; Garcia and Corredor, 2004; Perottoni et al., 2005).
 4
 5    8.1.5   Exposure-Response of Terrestrial Species
 6    8.1.5.1  Introduction
 7          Section 8.1.4 summarized the most important factors related to uptake of Pb by terrestrial
 8    organisms, the physiological effects of Pb, and the factors that modify terrestrial organism
 9    responses to Pb. Section 8.1.5 outlines and highlights the critical recent advancements in the
10    understanding of the toxicity of Pb to terrestrial organisms.  This section begins with a summary
11    of the conclusions from the 1986 Pb AQCD and then summarizes the critical research conducted
12    on effects of Pb on primary producers, consumers, and decomposers. All  concentrations are
13    expressed as mg Pb/kg soil dw, unless otherwise indicated.
14          The summary of recent critical advancements in understanding toxicity relies heavily on
15    the work completed by a multi-stakeholder group, consisting of federal, state, consulting,
16    industry, and academic participants, led by the EPA to develop Ecological Soil Screening Levels
17    (Eco-SSLs). Eco-SSLs describe the concentrations of contaminants in soils that are protective of
18    ecological receptors (U.S. Environmental Protection Agency, 2005b).  They were developed to
19    identify contaminants requiring further evaluation in an ecological risk assessment and were not
20    designed to be used as cleanup levels.  Eco-SSLs were derived for terrestrial plants, soil
21    invertebrates,  birds, and mammals. Detailed procedures using an extensive list of acceptability
22    and exclusion criteria (U.S. Environmental Protection Agency, 2005b) were used in screening
23    the toxicity studies to ensure that only those that met minimum quality standards were used to
24    develop the Eco-SSLs.  In addition, two peer reviews were completed during the Eco-SSL
25    development process. The first was a consultation with the EPA Science Advisory Board (SAB)
26    in April 1999, and the second was a peer review workshop in July 2000, which was open to
27    the public.
28          Areas of research that were not addressed are effects from irrelevant exposure conditions
29    relative to airborne  emissions of Pb (e.g.,  Pb shot, Pb paint, injection studies, studies conducted
30    on mine tailings, and studies conducted with hydroponic solutions); mixture toxicity  (addressed
31    in Section 8.1.4); issues related to indirect effects (e.g., effects on predator/prey interactions,

      December 2005                            8-69        DRAFT-DO NOT QUOTE OR CITE

-------
 1    habitat alteration, etc.); and human health-related research (e.g., hypertension), which is
 2    addressed in other sections of this document.
 3          The toxicity data presented herein should be reviewed with a note of caution regarding
 4    their relevance to field conditions.  Laboratory studies, particularly those using Pb-spiked soil,
 5    generally do not allow the soil to equilibrate following the addition of Pb and prior to the
 6    addition of test organisms.  This may result in increased bioavailability and overestimation of Pb
 7    toxicity relative to actual environmental conditions (Davies et al., 2003).
 8
 9    8.1.5.2  Summary of Conclusions from the 1986 Lead Criteria Document
10          The previous Pb AQCD, Volume II (U.S. Environmental Protection Agency, 1986a)
11    reviewed the literature on the toxicity of Pb to plants, soil organisms, birds, and mammals. The
12    main conclusions from this document are provided below.
13
14    Primary Producers
15          Commonly reported effects of Pb on vascular plants include the inhibition of
16    photosynthesis, respiration, and/or cell elongation, all of which reduce plant growth.  However, it
17    was noted that studies of other effects on plant processes such as maintenance, flowering, and
18    hormone development had not been conducted; therefore, no conclusion could be reached
19    concerning effects of Pb on these processes.
20          The EPA (U.S. Environmental Protection Agency, 1986a) concluded that most plants
21    experience reduced growth when Pb concentrations in soil moisture (the film of moisture
22    surrounding soil particles in the root zone of soil) exceed 2 to 10 mg/kg.  It also was concluded
23    that most plants would experience reduced growth (inhibition of photosynthesis, respiration, or
24    cell elongation) in soils of > 10,000 mg/kg when soil composition and pH are such that
25    bioavailability of Pb in the soil is low (see  Section 8.1.4 for details on factors affecting
26    bioavailability of Pb in soil). Acid soils or soils with low organic matter tend to increase Pb
27    bioavailability and would inhibit plants at much lower Pb concentrations (e.g., as low as
28    <100 mg/kg).
29          Many effect levels have been reported at Pb concentrations much lower than
30    10,000 mg/kg soil. For  example, effects on rye grass (Lolium rigidum) exposed to Pb in soil
31    included inhibition of germinating root elongation (at <2.5 mg/kg), absence of root growth

      December 2005                           8-70       DRAFT-DO NOT QUOTE OR CITE

-------
 1    (at 5 mg/kg), or 55% inhibition of seed germination (at 20 to 40 mg/kg). Stunted growth in
 2    radish (Raphanus sativus) was observed at 1000 mg/kg soil, with complete growth inhibition at
 3    5000 mg/kg, when Pb was added as Pb-chloride; effects were less severe when the Pb was added
 4    as Pb-oxide.
 5
 6    Consumers
 1         The EPA (U.S. Environmental Protection Agency, 1986a) concluded that food is the
 8    largest contributor of Pb to animals, with inhalation rarely accounting for more than 10 to 15%
 9    of daily intake of Pb and drinking water exposures being quite low.  It also was concluded that a
10    regular dose of 2 to 8 mg/kg-day causes death in most animals. Grazing animals may consume
11    more than 1 mg/kg-day in habitats near smelters and roadsides, but no toxic effects were
12    documented in these animals.
13
14    Decomposers
15         Lack of decomposition has been observed as a particular problem around smelter sites.
16    Lead concentrations between 10,000 and 40,000 mg/kg soil can eliminate populations of
17    decomposer bacteria and fungi (U.S. Environmental Protection Agency, 1986a). Lead may
18    affect decomposition processes by direct toxicity to specific groups of decomposers, by
19    deactivating enzymes excreted by decomposers to break down organic matter, or by binding with
20    the organic matter and rendering it resistant to the action of decomposers.
21         Microorganisms are more sensitive than plants to Pb in soil. Delayed decomposition may
22    occur at between 750 and 7500 mg/kg soil  (depending on soil type and other conditions).
23    Nitrification is inhibited by 14% at 1000 mg/kg soil.
24
25    U.S. Environmental Protection Agency Staff Review of 1986 Criteria Document
26         The EPA reviewed the 1986 Pb AQCD and presented an overall summary of conclusions
27    and recommendations (U.S. Environmental Protection Agency, 1990).  The major conclusion
28    was that available laboratory and field data indicated that high concentrations of Pb can affect
29    certain plants and alter the composition of soil microbial communities.  It was noted that few
30    field studies were available in which Pb exposures and associated effects in wildlife were
31    reported.

      December 2005                           8-71        DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.5.3  Recent Studies on the Effects of Lead on Primary Producers
 2          Several studies published since 1986 have reports of terrestrial plant exposure to Pb in
 3    soil, many of which were reviewed during the development of the Eco-SSLs (U.S.
 4    Environmental Protection Agency, 2005a). The relevant information from the Eco-SSL
 5    document (U.S. Environmental Protection Agency, 2005a) is summarized below. A literature
 6    search and review also was conducted to identify critical papers published since 2002, which is
 7    when the literature search was completed for Eco-SSL development, and no new papers were
 8    identified as critical to the understanding of Pb toxicity to terrestrial primary producers.
 9          Effects observed  in studies conducted since the 1986 Pb AQCD are similar to those
10    reported previously and  include decreased photosynthetic and transpiration rates and decreased
11    growth and yield (U.S. Environmental Protection Agency, 2005a). The phytotoxicity of Pb is
12    considered relatively low, due to the limited availability and uptake of Pb from soil and soil
13    solution and minimal translocation of Pb from roots to shoots (Pahlsson, 1989).  Although many
14    laboratory toxicity studies have reported effects on plants, there are few reports of phytotoxicity
15    from Pb exposure under field conditions. For example, Leita et al. (1989) and Sieghardt (1990)
16    reported high concentrations of Pb and other metals in soil and vegetation collected around
17    mining areas in Europe,  with no toxicity symptoms observed in plants or fruit.
18          The literature search completed for the terrestrial plant Eco-SSL development identified
19    439 papers for detailed review, of which 28 met the minimum criteria (U.S. Environmental
20    Protection Agency, 2005b). Thirty ecotoxicological endpoints  were gleaned from these 28
21    papers and were further  evaluated; most of those evaluated growth (biomass), which was
22    considered the most sensitive and ecologically-relevant endpoint (U.S. Environmental Protection
23    Agency, 2005a).  Five of the endpoints, representing four species tested under three different
24    combinations of pH and organic matter content, were used to develop the Eco-SSL of 120 mg/kg
25    (115 mg/kg rounded to two significant digits) (Table 8-1.5.1).
26          The 25 ecotoxicological endpoints that were not used to develop the Eco-SSL for plants
27    are presented in Table 8-1.5.2. The first six endpoints were considered eligible for Eco-SSL
28    derivation but were not used; the remainder did not meet all of the requirements to be considered
29    for inclusion in the Eco-SSL derivation process.
30
31
      December 2005                           8-72        DRAFT-DO NOT QUOTE OR CITE

-------
                  Table 8-1.5.1. Plant Toxicity Data Used to Develop the Eco-SSL
Plant Species
Loblolly pine (Pinus taeda)
Red maple (Acer rubrum)
Berseem clover
(Trifolium alexandrium)
Berseem clover
Rye grass (Lolium rigidum)

Soil pH
4
4
6.3
6.7
5.6

% Organic
Matter
2.5
2.5
0.94
3.11
0.1

Toxicity
Parameter
MATC* (growth)
MATC (growth)
MATC (growth)
MATC (growth)
MATC (growth)
Geometric Mean
Pb in Soil
(mg/kg dw)
144
144
316
141
22
115
       *MATC = Maximum Acceptable Threshold Concentration, or the geometric mean of the NOAEC
       (no-observed-adverse-effect concentration) and LOAEC (lowest-observed-adverse-effect concentration).
       Source: U.S. Environmental Protection Agency (2005a).
 1   8.1.5.4  Recent Studies on the Effects of Lead on Consumers
 2         Since the 1986 Pb AQCD, there have been several studies in which birds and mammals
 3   were exposed to Pb via ingestion  (primarily through dietary Pb). Many of these were reviewed
 4   during development of the Eco-SSLs (U.S. Environmental Protection Agency, 2005a).  The
 5   relevant information from the Eco-SSL document (U.S. Environmental Protection Agency,
 6   2005a) is described below. A literature search and review was conducted to identify critical
 7   papers published since 2002. These recent critical papers are described briefly below.
 8   No studies were found that used inhalation exposures to evaluate endpoints such as survival,
 9   growth, and reproduction in birds or mammals.  All studies described below exposed organisms
10   via ingestion (drinking water or diet) or gavage.
11         The Eco-SSLs for avian and mammalian consumers are presented as Pb concentrations in
12   soil. These concentrations were calculated by assuming exposure to Pb via incidental soil
13   ingestion and ingestion of Pb-contaminated food, and using a no-observed-adverse-effect level
14   (NOAEL) as the toxicity reference value (TRV) (U.S. Environmental Protection Agency,
15   2005b). A simplified version of the equation used to calculate the Eco-SSL is:
16
     December 2005
8-73
DRAFT-DO NOT QUOTE OR CITE

-------
            Table 8-1.5.2. Plant Toxicity Data Not Used to Develop the Eco-SSL
Plant Species Soil pH
Studies eligible for Eco-SSL derivation,
Berseem clover
(Trifolium alexandrium)
Tomato
(Lycopericon esculentum)
Tomato
Fenugreek
(Trigonella foenum-graecum)
Spinach (Spinacea oleracea)
Com (Zea mays)
Sow thistle (Sonchus oleraceus)
but not used
6.7

7.73
8.20
8.3
6.7
6.5
7.23
% Organic
Matter

3.11

1.70
0.86
0.5
3.0
2.1
1.6
Toxicity Parameter

MATC

MATC
MATC
MATC
MATC
MATC
MATC
Pb in Soil
(mg/kg dw)

141

71
71
283
424
158
2,263
Studies not eligible for Eco-SSL derivation
Loblolly pine (Finns taeda)
Red oak (Quercus rubra)
Spinach
Alfalfa (Medicago sativa)
Alfalfa
Alfalfa
Radish (Raphanus sativus)
Radish
Radish
Onion (Allium cepa)
Radish
Carrot (Daucus carota)
Peas (Pisum sativum)
Barley (Hordeum vulgare)
Alfalfa
Tomato
Spinach
Radish
Radish
5.5
6
6.7
6.4
6.9
6.9
6.9
6.9
6.9
8.3
5.1
7.0
7.0
6.0
6.9
7.45
6.7
6.2
7.1
3.4
1.5
0.0
1.0
1.7
1.7
1.0
1.0
1.0
0.5
8.0
0.6
0.6
2.5
4.8
2.06
8.0
8.0
8.0
NOAEC
LOAEC
NOAEC
NOAEC
NOAEC
NOAEC
LOAEC
LOAEC
LOAEC
LOAEC
NOAEC
NOAEC
NOAEC
NOAEC
NOAEC
MATC
NOAEC
NOAEC
NOAEC
480
100
600
250
250
250
500
100
100
50
600
85
85
1,000
250
35
600
600
600
 *MATC = Maximum Acceptable Threshold Concentration, or the geometric mean of the NOAEC
  (no-observed-adverse-effect concentration) and LOAEC (lowest-observed-adverse-effect concentration).

 Source: U.S. Environmental Protection Agency (2005a).
December 2005
8-74
DRAFT-DO NOT QUOTE OR CITE

-------
 2                      HQ = rrCgnji x IRsoiT) + (C^ x IRg^l / BW                       (8-4)
 4                                        TRY
 5
 6    where:
 7
 8    HQ   = hazard quotient (1 mg Pb/kg bw/day)
 9    CSoii   = concentration of Pb in soil (mg Pb/kg soil)
10    IRsoii  = incidental soil ingestion rate (kg soil/day)
11    Cfood  = concentration of Pb in food (mg Pb/kg food)
12    IRfood  = food ingesti on rate (kg food/day)
13    BW   = body weight (kg)
14    TRV  = toxicity reference value (mg Pb/kg bw/day)
15
16         Food ingesti on was estimated by modeling the uptake of Pb from soil into each diet
17    component (e.g., vegetation, invertebrates, etc.).  Bioavailability of Pb in soil and food was
18    assumed to be 100%.  The Eco-SSL is equivalent to the concentration of Pb in soil that results in
19    an HQ = 1.  The two factors that may have the most significant influence on the resulting Eco-
20    SSL are the assumption of 100% bioavailability of Pb in soil and diet and the selection of the
21    TRV. The toxicity data that were reviewed to develop the TRV  are presented in the following
22    subsections.
23         Representative avian and mammalian wildlife  species were selected for modeling Pb
24    exposures to wildlife with different diets and calculating the Eco-SSL.  The avian species
25    selected were dove (herbivore), woodcock (insectivore), and hawk (carnivore).  The mammalian
26    species selected were vole (herbivore), shrew (insectivore), and weasel  (carnivore). The lowest
27    of the three back-calculated soil concentrations, which resulted in an HQ = 1, was selected as the
28    Eco-SSL. For Pb,  the lowest values were for the insectivorous species  of bird and mammal.
29
30    Avian Consumers
31         Effects on birds observed in studies conducted since the 1986 Pb AQCD  are similar to
32    those reported previously: mortality, changes in juvenile growth rate and weight gain, effects on
33    various reproductive measures, and changes in behavior (U.S. Environmental Protection Agency,
34    2005a). Reproductive effects following Pb exposure  included declines  in clutch size, number of
35    young hatched, and number of young fledged as well as decreased fertility or eggshell thickness.
      December 2005                           8-75        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Few significant reproductive effects have been reported in birds at Pb concentrations below
 2   100 mg/kg in the diet (Scheuhammer, 1987).
 3          The literature search completed for Eco-SSL development identified 2,429 papers for
 4   detailed review for either avian or mammalian species, of which 54 met the minimum criteria for
 5   further consideration for avian Eco-SSL development (U.S. Environmental Protection Agency,
 6   2005a). The 106 toxicological data points for birds that were further evaluated included
 7   biochemical, behavioral, physiological, pathological, reproductive, growth, and survival effects.
 8   Growth and reproduction data were used to derive the Eco-SSL (Table  8-1.5.3; Figure 8-1.5.1),
 9   as these were determined to be the most ecologically-relevant endpoints.  The geometric mean of
10   the NOAELs was calculated as 10.9 mg/kg-day, which was higher than the lowest bounded
11   LOAEL (the term "bounded" means that both a NOAEL and LOAEL were obtained from the
12   same study). Therefore, the highest bounded NOAEL that was lower than the lowest bounded
13   LOAEL for survival, growth, or reproduction (1.63 mg Pb/kg bw-day) was used as the TRV.
14   The TRV was used to back-calculate the Eco-SSL of 11 mg/kg soil for  avian species.
15          Many of the toxicity data presented in the Eco-SSL document (U.S. Environmental
16   Protection Agency, 2005a) are lower than those discussed in the 1986 Pb AQCD. The TRV and
17   resulting Eco-SSL were derived using many conservative assumptions.  For example, the EPA
18   (U.S. Environmental Protection Agency, 2005a) recognizes that toxicity is observed over a wide
19   range of doses (<1 to >100 mg Pb/kg bw/day), even when considering only reproductive effects
20   in the same species. In addition, the TRV of 1.63 mg/kg-day is lower than most of the reported
21   doses that have been associated with measured effects. This is true not only for survival, growth,
22   and reproductive effects but also biochemical, behavioral, physiological, and pathological
23   effects, which generally are observed at lower concentrations than effects on growth or
24   reproduction. In addition, the Eco-SSL was back-calculated using conservative modeling
25   assumptions. Therefore, the Eco-SSL of 11 mg/kg may be considered a conservative value.
26          Very little research has been done to expand the knowledge of the toxicity of Pb to birds
27   since the Eco-SSL work was done. However, several studies have been conducted on waterfowl.
28   Toxicity data for waterfowl (in particular, mallards) were included in the soil Eco-SSL
29   development process (Table 8-1.5.3), although mallards may be more exposed to contaminants
30   in sediment than soil. Effects on waterfowl may vary depending on the form of Pb,
31   characteristics of the sediment, the foraging strategy of the species (which may vary during

     December 2005                           8-76        DRAFT-DO NOT QUOTE OR CITE

-------
O
O
3
cr
^
to
O
O








oo
i
^j





^
H
I
O
O
n
s.**
H
O
O
H
W
O
O
H
W
Table 8-1.5.3. Avian Toxicity Data Used to Develop the Eco-SSL

Avian
Species
Reproduction
Japanese
quail
Chicken
Chicken
Mallard
American
kestrel
Japanese
quail

Japanese
quail
Japanese
quail
Japanese
quail
Japanese
quail

Chicken
Ringed
turtle dove
Japanese
quail

Japanese
quail




No. of
Doses

4

3
4
2
3

5


5

3

5

4


5
2

2


2





Route of
Exposure

FD

FD
FD
FD
FD

FD


FD

FD

FD

FD


FD
DR

FD


FD





Exposure
Duration

5

4
30
76
6

5


5

32

12

12


10
11

1


27





Du ratio Age Lifestag Se Effect
n Units Age Units e x Type

w 6 w LB F REP

w NR NR LB F REP
d 22 w LB F EGG
d NR NR SM F EGG
mo 1-6 yr AD F REP

w 6 d JV M REP


w Id JV M REP

d NR NR AD F REP

w 0 d LB B REP

w NR NR LB F REP


w NR NR LB F REP
w NR NR AD M REP

w 14 w JV F REP


d NR NR AD F REP




NOAEL
Effect Response (nig/kg
Measure Site bw/day)

PROG WO 0.194

PROG WO 1.63
ALWT EG 2.69
ESTH EG 5.63
RSUC WO 12.0

TEWT TE 12.6


TEWT TE 67.4

PROG WO 125

EGPN EG

PROG WO


PROG WO
TEWT TE

TPRD WO


PROG WO




LOAEL
(mg/kg
bw/day)

1.94

3.26
4.04



126


135



0.110

0.194


3.26
11.8

93.1


377





-------
o
CD
3

^
o
o









oo
oo



M
\^
£j
H
I
0
o
0
H

O
o
H
W
o

O
H
W
Table 8-1.5.3 (cont'd). Avian Toxicity Data Used to Develop the Eco-SSL

Avian
Species
Growth

Japanese
quail
Japanese
quail
Japanese
quail
Japanese
quail
Chicken
Chicken
Japanese
quail
Japanese
quail
Japanese
quail
Japanese
quail

Chicken
Duck

American
kestrel
Chicken

Japanese
quail





No. of
Doses


3

3

2

3

4
4
5

5

5

5


2
3

4

2

5






Route of
Exposure


FD

FD

FD

FD

FD
FD
FD

FD

FD

FD


FD
GV

GV

FD

FD






Exposure
Duration


5

2

2

4

4
4
12

12

2

1


21
3

10

20

14






Du ratio Age
n Units Age Units


w 1 d

w 1 d

w 1 d

w 0 d

w 4 w
w 4 w
w 0 d

w 1 w

w 6 d

w 1 d


d 1 d
mo 24 w

d 1 d

d 1 d

d 1 d






Lifestag Se Effect
e x Type


JV F GRO

JV B GRO

JV NR GRO

JV F GRO

JV NR GRO
JV NR GRO
JV F GRO

JV F GRO

JV NR GRO

JV NR GRO


JV B GRO
MA F GRO

JV NR GRO

JV B GRO

JV B GRO






Effect
Measure


BDWT

BDWT

BDWT

BDWT

BDWT
BDWT
BDWT

BDWT

BDWT

BDWT


BDWT
BDWT

BDWT

BDWT

BDWT






Response
Site


wo

wo

wo

wo

wo
wo
wo

wo

wo

wo


wo
wo

wo

wo

wo





NOAEL
(mg/kg
bw/day)


1.56

2.77

4.64

5.93

6.14
7.10
11.1

11.2

12.6

13.5


14.2
20.0

25.0

28.4

34.5





LOAEL
(mg/kg
bw/day)


15.6





59.3

61.4
71.0
111

112

126

67.4





125










-------
Table 8-1.5.3 (cont'd). Avian Toxicity Data Used to Develop the Eco-SSL
V J
CD
3
cr
to
o
o






oo
i
VO


O
H
6
o
0
H
O
o
H
W
O
O
H
W
Avian
Species
American
kestrel
Chicken
Mallard
Chicken
Chicken
Japanese
quail
Chicken
Chicken
Chicken
Chicken
Chicken
Chicken
Chicken
Chicken
Chicken

AD = adult
thinning; F
or numbers
Source: U.


No. of Route of
Doses Exposure
4 FD
5 FD
4 FD
5 FD
2 FD
3 FD
2 FD
3 FD
4 FD
2 FD
2 OR
2 FD
2 FD
2 FD
2 FD

; ALWT = albumin weight; B
= female; FD = food; GRO =
; REP = reproduction; RSUC
Exposure Duratio Age
Duration n Units Age Units
60
2
8
20
3
32
19
2
14
20
4
7
2
7
14

d
w
d
d
w
d
d
w
d
d
w
d
w
d
d

1-2 yr
1 d
9 d
1 d
1 d
NR NR
1 d
1 d
8 d
1 d
NR NR
1 d
1 d
1 d
8 d

Lifestag
e
AD
JV
JV
JV
JV
AD
JV
JV
JV
JV
JV
JV
JV
IM
JV

Se
X
B
M
NR
M
M
F
M
M
M
M
B
M
M
NR
M

Effect
Type
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO

NOAEL LOAEL
Effect Response (nig/kg (nig/kg
Measure Site bw/day) bw/day)
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT

WO
WO
WO
WO
WO
WO
WO
WO
WO
WO
WO
WO
WO
wo
wo

54.3
61.3 123
66.9
38.2
53.1
64.3
76.3
124
152
163
200
262
270
273
282

= both; BDWT = body weight changes; d = days; DR = drinking water; EG = egg; EGG = effects on eggs; EGPN = egg production; ESTH = eggshell
growth; GV = gavage; JV = juvenile; LB = laying bird; MA = mature; M = male; mo = months; NR = not reported; OR = other oral; PROG = progeny counts
= reproductive success; SM = sexually mature; TE = testes; TEWT = testes weight; TPRD = total production; w = weeks; WO = whole organism; yr = years.
S. Environmental Protection Agency (2005a)





















-------
•innn
*>*
(0 tnn
^ 100
.Q
0)
-^ -in^
^ 10
0)
Q)
(A
Out
o
04
.1

oooo
Qo
AAAO oo o°
oo°A o A ..• Qo°
•••• °
?--.AA.,«??A 	
•** Legend:
A* A
A • Growth NOAEL
0 Growth LOAEL
A Reproduction NOAEL
A Reproduction LOAEL
A A - Eco-SSL (mg/kg dw)
A
Reproduction and Growth






     Figure 8-1.5.1.  Avian toxicity data considered in development of the Eco-SSL.
     Source: U.S. Environmental Protection Agency (2005a).
 1   reproduction), and the nutritional status of the animal.  Sediment is recognized as an important
 2   route of exposure for waterfowl, particularly those species that dabble (i.e., forage on
 3   invertebrates in the sediment) (Beyer et al., 2000; Douglas-Stroebel et al., 2005).  Douglas-
 4   Stroebel et al. (2005) found that mallard ducklings exposed to Pb-contaminated sediment and a
 5   low nutrition diet exhibited more changes in behavior (as measured by time bathing, resting, and
 6   feeding) than Pb exposure or low nutrition exposure alone.  These effects may be due to the low
 7   nutrition diet being deficient in levels of protein, amino acids, calcium, zinc, and other nutrients.
 8          Beyer et al. (2000) related blood Pb to sublethal effects in waterfowl along the Coeur
 9   d'Alene River near a mining site in Idaho. The authors suggested that 0.20 mg/kg ww blood Pb
10   represents the no-effect level.  This no-effect blood concentration corresponds to a sediment Pb
11   concentration of 24 mg/kg. A sediment concentration of 530 mg/kg, associated with a blood Pb
12   concentration of 0.68 mg/kg ww, is suggested to be the lowest-effect concentration. These
13   results are consistent with those of Scheuhammer (1989) who found blood Pb concentrations of
14   0.18 |ig/mL to 0.65 |ig/mL in mallards corresponded to conditions associated with greater than
     December 2005
8-80
DRAFT-DO NOT QUOTE OR CITE

-------
 1   normal exposure to Pb, but that that should not be considered Pb poisoning.  The study by Beyer
 2   et al. (2000) related blood Pb to waterfowl mortality and concluded that some swan mortality
 3   may occur at blood Pb levels of 1.9 mg/kg ww, corresponding to a sediment Pb concentration of
 4   1800 mg/kg. Using the mean blood level of 3.6 mg/kg ww from all moribund swans in the
 5   study, it was predicted that half of the swans consuming sediment at the 90th percentile rate
 6   would die with chronic exposure to sediment concentrations of 3600 mg/kg.
 7
 8   Mammalian Consumers
 9         Effects on mammals observed in studies conducted since the 1986 AQCD are similar to
10   those reported previously:  mortality, effects on reproduction, developmental effects, and
11   changes in growth (U.S. Environmental Protection Agency, 2005a).  Very little research has
12   been done to expand the knowledge of the toxicity of Pb to mammalian wildlife, since the
13   Eco-SSL work was done.  Most studies conducted on mammals use laboratory animals to study
14   potential adverse effects of concern for humans, and such studies are summarized in other
15   sections of this document.
16         Of the 2,429 papers identified in the literature search for Eco-SSL development, 219 met
17   the minimum criteria for further consideration for mammalian Eco-SSL development (U.S.
18   Environmental Protection Agency, 2005a). The 343 ecotoxicological endpoints for mammals
19   that were further evaluated included biochemical, behavioral, physiological, pathological,
20   reproductive, growth, and survival effects.  Growth and reproduction data were used to derive
21   the Eco-SSL (Table 8-1.5.4, Figure 8-1.5.2), as these were determined to be the most
22   ecologically-relevant endpoints. The geometric mean of the NOAELs was calculated as
23   40.7 mg/kg-day, which was higher than the lowest bounded LOAEL for survival, growth or
24   reproduction. Therefore, the highest bounded NOAEL that was lower than the lowest bounded
25   LOAEL for survival, growth, or reproduction (4.7 mg Pb/kg bw-day) was used  as the TRV. The
26   TRV was used to back-calculate the Eco-SSL of 56 mg/kg soil.
27         A review of the data presented in the Eco-SSL document (U.S. Environmental  Protection
28   Agency, 2005a) reveals that effects on survival generally are observed at Pb doses much greater
29   than those reported in the  1986 Pb AQCD, where it was concluded that most animals would die
30   when consuming a regular dose of 2 to 8 mg Pb/kg bw-day (U.S. Environmental Protection
31   Agency,  1986a). However, the data presented in the Eco-SSL document (U.S. Environmental

     December 2005                        8-81            DRAFT-DO NOT QUOTE OR CITE

-------
December 2005







oo
1
oo
to



o
§
H
6
O
0
H
O
o
H
W
O
O
HH
H
W
Table 8-1.5.4. Mammalian Toxicity Data Used
Mammalian
Species
Reproduction
Rat
Rat
Rat
Rat
Sheep
Rat
Guinea pig
Rat
Rat
Rat
Cotton rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat

No.
of
Doses

5
6
3
4
3
6
3
5
4
5
3
4
3
2
4
4
3
4
5
4

Route of
Exposure

DR
DR
DR
DR
FD
DR
DR
FD
DR
DR
DR
GV
DR
FD
DR
DR
DR
GV
DR
DR

Exposure
Duration

62
21
35
62
27
21
40
92
23.8
23.8
7
9
100
35
60
56
31
41
1
30

Duration
Units

d
d
d
d
w
d
d
w
d
d
w
w
d
d
d
d
d
d
w
d

Age

21
NR
NR
21
NR
NR
NR
21
21
21
NR
10
21
70
NR
70
NR
NR
94
NR

Age
Units

d
NR
NR
d
NR
NR
NR
d
d
d
NR
w
d
d
NR
d
NR
NR
d
NR

Lifestage

GE
GE
AD
GE
GE
GE
GE
JV
LC
GE
AD
JV
GE
LC
SM
LC
LC
GE
JV
SM

to Develop the Eco-SSL
Sex

F
F
M
B
F
F
F
M
F
F
M
M
F
F
M
F
F
F
M
M

Effect
Type

REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP
REP

Effect
Measure

PRWT
PRWT
RSUC
PRWT
RSUC
PRWT
PRWT
TEWT
Other
Other
RHIS
SPCV
PRWT
PRWT
TEWT
PROG
PRWT
PRWT
SPCL
Other

Response
Site

wo
wo
wo
wo
wo
wo
wo
TE
WO
WO
RT
TE
WO
WO
TE
WO
WO
WO
SM
SV

NOAEL
(mg/kg
bw/day)

0.71
1.00
2.60
3.00
4.50
5.00
5.50
7.50
8.90
9.10
12.4
18.0
25.4
27.5
31.6
32.5
33.3
41.0
47.3
56.0

LOAEL
(mg/kg
bw/day)

7.00
5.00
26.0
6.0
—
10.0
—
74.9

45.0
170
180
	
63.2
—
Ill
54.6
82.0
285


-------
December ^
i^-^
o
o







oo
i
oo
OJ


o
f?
H
I
o
o
0
H

O
o
H
W
O
O
H
W

Mammalian
Species
Hamster

Hamster
Rat
Rat
Rat
Rat
Rat
Mouse
Mouse


Mouse
Rat
Rat
Rat
Rat
Rat


Rat
Mouse

Rat

Rat
Mouse
Mouse
Rat



No.
of
Doses
2

2
4
5
3
4
2
3
7


7
4
5
2
2
4


4
4

3

2
2
4
2


Table
Route of
Exposure
DR

DR
DR
GV
DR
DR
DR
FD
FD


FD
DR
DR
GV
DR
FD


DR
GV

FD

DR
DR
GV
DR


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
51

14
37
12
68
77
21
8
30


30
21
10
102
9
4


13
60

339

9
6
52
120


Duration
Units
d

d
d
d
d
d
d
w
d


d
d
w
d
mo
d


w
d

d

mo
mo
d
d


Age
15

11
NR
NR
25
25
NR
2
NR


NR
NR
NR
30
NR
NR


NR
NR

26-27

21
21
2
1


Age
Units
w

w
NR
NR
d
d
NR
mo
NR


NR
NR
NR
d
NR
NR


NR
NR

d

d
d
mo
d


Lifestage
GE

GE
GE
GE
GE
GE
LC
GE
LC


LC
GE
AD
GE
SM
LC


JV
AD

JV

JV
JV
GE
GE


Sex
F

F
F
F
F
F
F
M
F


F
F
M
F
M
F


M
F

B

F
F
F
M


Effect
Type
REP

REP
REP
REP
REP
REP
REP
REP
REP


REP
REP
REP
REP
REP
REP


REP
REP

REP

REP
REP
REP
REP


Effect
Measure
PROG

PROG
PRWT
RSEM
PRWT
PRWT
PRWT
SPCV
PRWT


PRWT
DEYO
TEWT
PRWT
RHIS
PRWT


PERT
RPRD

PRWT

DEYO
DEYO
PROG
SPCL


Response
Site
wo

wo
wo
EM
WO
wo
wo
TE
WO


WO
WO
MT
WO
TE
WO


WO
0V

WO

WO
WO
EM
TE


NOAEL
(mg/kg
bw/day)
64.8

64.9
90.1
100
115
116
120
144
202


202
276
294
441
600
601


639
—

	

—
—
—



LOAEL
(mg/kg
bw/day)
	

—
270
150
—
—
—
1,440
506


506
552
587
—
—
1,500


—
2.00

2.49

2.94
3.62
5.50
6.76



-------
December ^
i^-^
o
o







oo
i
oo
-^


o
f?
H
I
o
o
0
H

O
o
H
W
O
O
H
W

Mammalian
Species
Mouse

Mouse
Rat
Rat
Rat
Rat
Mouse
Mouse
Mouse


Mouse
Mouse
Rat
Rat
Rat
Rat


Rat
Rat

Rat

Rat
Rat
Rat
Rat



No.
of
Doses
2

2
2
2
2
3
2
2
2


2
4
2
5
2
3


2
2

4

2
4
3
2


Table
Route of
Exposure
DR

GV
FD
GV
DR
DR
DR
DR
DR


DR
FD
DR
DR
DR
DR


DR
DR

FD

FD
FD
FD
DR


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
5

2
102
3
18
90
23
62
18


12
18
4
6
22
30


13
21

3

1
3
25
21


Duration
Units
w

w
d
mo
d
d
d
d
w


w
d
w
w
d
d


w
d

w

w
w
d
d


Age
NR

NR
NR
8
NR
NR
NR
NR
6-8


9
NR
99
4
NR
52


NR
80

NR

19
NR
NR
NR


Age
Units
NR

NR
NR
w
NR
NR
NR
NR
w


w
NR
d
mo
NR
d


NR
d

NR

w
NR
NR
NR


Lifestage
AD

JV
GE
SM
GE
AD
GE
GE
LC


SM
GE
JV
GE
GE
JV


GE
JV

LC

LC
LC
LC
LC


Sex
M

M
F
M
F
M
F
F
F


M
F
M
F
F
M


F
F

F

F
F
F
F


Effect
Type
REP

REP
REP
REP
REP
REP
REP
REP
REP


REP
REP
REP
REP
REP
REP


REP
REP

REP

REP
REP
REP
REP


Effect
Measure
TEDG

SPCL
PROG
TEDG
PRWT
SPCL
PRWT
PRWT
PRWT


PRFM
PRWT
SPCL
RHIS
PRWT
GREP


PRWT
PRWT

PRWT

PRWT
PRWT
PRWT
PRWT


Response
Site
TE

SM
WO
TE
WO
SM
WO
wo
wo


wo
wo
SM
WO
wo
PG


WO
WO

wo

wo
wo
wo
wo


NOAEL LOAEL
(mg/kg (mg/kg
bw/day) bw/day)
— 16.6

— 46.4
— 49.6
— 50.0
— 55.5
— 61.2
— 78.6
— 99.8
— 137


— 139
— 154
— 171
— 175
— 178
— 198


— 200
— 218

— 221

— 222
— 230
— 258
— 330



-------
December ^
l^-^
O
O







oo
i
oo
^


o
f?
H
I
o
o
0
H

O
o
H
W
O
O
H
W

Mammalian
Species
Rat

Rat
Rat
Rat
Rat
Mouse
Mouse
Rat
Mouse


Rat
Mouse
Rat
Rat
Rat
Mouse


Rat
Rat

Mouse

Rat
Rat
Rat
Mouse



No.
of
Doses
2

2
2
2
2
2
2
2
2


2
2
2
2
2
2


2
2

2

2
2
2
4


Table
Route of
Exposure
DR

DR
DR
DR
DR
DR
DR
DR
DR


DR
DR
DR
DR
DR
FD


DR
DR

DR

FD
FD
FD
DR


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
30

17
24
12
30
44
14
50
45


22
48
9
9
3
7


126
20

4

2
7
21
18


Duration
Units
d

d
d
d
d
d
d
d
d


d
d
mo
mo
w
d


d
w

d

w
d
d
w


Age
52

NR
NR
NR
27
NR
NR
24
50-100


NR
NR
3
NR
14
NR


1
10

NR

NR
NR
NR
11


Age
Units
d

NR
NR
NR
d
NR
NR
d
d


NR
NR
mo
NR
w
NR


d
w

NR

NR
NR
NR
w


Lifestage
JV

GE
LC
GE
JV
GE
LC
JV
GE


GE
GE
SM
SM
LC
GE


GE
GE

LC

LC
LC
LC
JV


Sex
M

F
F
F
M
F
F
F
F


F
F
M
M
F
F


F
F

F

F
F
F
F


Effect
Type
REP

REP
REP
REP
REP
REP
REP
REP
REP


REP
REP
REP
REP
REP
REP


REP
REP

REP

REP
REP
REP
REP


Effect
Measure
SPCL

PRWT
PRWT
PRWT
SPCL
PRWT
PRWT
RBEH
ODVP


PRWT
PRWT
SPCL
TEDG
PRWT
RSUC


PROG
PRWT

PRWT

PRWT
PRWT
PRWT
TEWT


Response
Site
SM

WO
wo
wo
SM
WO
wo
wo
wo


wo
wo
TE
TE
WO
EM


WO
WO

WO

wo
wo
wo
wo


NOAEL LOAEL
(mg/kg (mg/kg
bw/day) bw/day)
— 354

— 360
— 360
— 362
— 364
— 381
— 381
— 381
— 404


— 420
— 437
— 579
— 600
— 635
— 646


— 651
— 750

— 762

— 828
— 833
— 991
— 1,370



-------
December ^
i^-^
o
o







oo
i
oo
ON


O
f?
H
I
o
o
0
H
O
o
H
W
O
O
H
W

Mammalian
Species
Rat

Mouse
Rat
Rat
Rat
Rat
Mouse
Rat
Growth


Horse
Rat
Rat
Rat
Cattle
Rat


Rat
Rat

Dog
Rat
Rat
Cattle
Rat



No.
of
Doses
2

2
2
2
2
M
2
2



2
2
6
5
4
3


2
4

3
3
4
3
2


Table
Route of
Exposure
FD

DR
FD
FD
FD
FD
DR
FD



FD
FD
DR
DR
OR
DR


DR
DR

FD
DR
DR
OR
OR


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
30

14
16
7
25
27
14
17



15
21
21
7
7
14


332
7

7
30
23
84
6


Duration
Units
d

w
d
d
d
d
w
d



w
d
d
d
w
d


d
w

mo
d
d
d
w


Age
NR

NR
NR
NR
NR
NR
21
NR



20-21
0
NR
50
1
21


28
21

NR
22-24
22
NR
NR


Age
Units
NR

NR
NR
NR
NR
NR
d
NR



w
d
NR
d
w
d


d
d

NR
d
d
NR
NR


Lifestage
LC

GE
LC
LC
LC
LC
JV
LC



JV
JV
GE
AD
JV
JV


JV
GE

JV
JV
JV
JV
AD


Sex
F

B
F
F
F
C
B
F



M
F
F
F
M
F


B
F

NR
M
F
M
M


Effect
Type
REP

REP
REP
REP
REP
REP
REP
REP



GRO
GRO
GRO
GRO
GRO
GRO


GRO
GRO

GRO
GRO
GRO
GRO
GRO


Effect
Measure
PRWT

PROG
PROG
PRWT
PRWT
PROG
PROG
PRWT



BDWT
BDWT
BDWT
BDWT
BDWT
BDWT


BDWT
BDWT

BDWT
BDWT
BDWT
BDWT
BDWT


Response
Site
wo

wo
wo
wo
wo
wo
wo
wo



wo
wo
wo
wo
wo
wo


wo
wo

wo
wo
wo
wo
wo


NOAEL
(mg/kg
bw/day)
	

—
—
—
—
—
—
—



0.15
0.5
1.00
1.27
1.99
2.40


2.98
4.70

4.71
5.64
5.80
7.79
9.10


LOAEL
(mg/kg
bw/day)
1,770

1,990
2,570
2,570
2,570
2,840
3,630
6,170



—
—
5.00
13.0
—
_


	
8.90

—
28.2
29.0
—




-------
December ^
i^-^
o
o







oo
i
oo
^


o
f?
H
I
o
o
0
H

O
o
H
W
O
O
H
W

Mammalian
Species
Rat

Rat
Rabbit
Rat
Rat
Rat
Rat
Rat
Mouse


Rat
Rat
Rat
Rat
Rat
Sheep


Rat
Cattle

Rat

Rat
Rat
Rat
Rat



No.
of
Doses
2

3
3
2
2
2
2
2
3


4
3
4
7
4
5


2
4

2

5
4
2
6


Table
Route of
Exposure
GV

DR
GV
DR
DR
FD
OR
DR
DR


GV
FD
GV
DR
DR
FD


DR
FD

GV

DR
GV
FD
DR


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
8

6
10
140
6
10
6
7
14


9
339
29
10
56
84


10
7

28

4
12
4
10


Duration
Units
w

mo
d
d
w
w
w
w
d


w
d
d
w
d
d


w
w

d

w
d
w
w


Age
NR

NR
1
21
NR
NR
NR
NR
0


10
26-27
NR
NR
70
NR


NR
16

2

94
2
NR
NR


Age
Units
NR

NR
d
d
NR
NR
NR
NR
d


w
d
NR
NR
d
NR


NR
w

d

d
d
NR
NR


Lifestage
JV

AD
JV
JV
JV
JV
AD
JV
JV


JV
JV
SM
JV
LC
JV


JV
JV

JV

JV
JV
JV
JV


Sex
F

M
F
M
M
M
M
M
NR


M
B
F
M
F
M


M
M

B

M
B
M
M


Effect
Type
GRO

GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO


GRO
GRO
GRO
GRO
GRO
GRO


GRO
GRO

GRO

GRO
GRO
GRO
GRO


Effect
Measure
BDWT

BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT


BDWT
BDWT
BDWT
BDWT
BDWT
BDWT


BDWT
BDWT

BDWT

BDWT
BDWT
BDWT
BDWT


Response
Site
wo

wo
wo
wo
wo
wo
wo
wo
wo


wo
wo
wo
wo
wo
wo


wo
wo

wo

wo
wo
wo
wo


NOAEL
(mg/kg
bw/day)
10.0

10.6
10.7
10.7
15.1
15.4
15.5
16.1
16.3


18.0
18.3
18.9
24.3
32.5
32.7


38.5
43.0

50.0

71.5
75.0
100
120


LOAEL
(mg/kg
bw/day)
	

532
50.4
—
—
—
—
—
163


180
—
—
—
—
_


	
—

	

178
225
—
383



-------
December ^
i^-^
o
o







oo
i
oo
oo


o
f?
H
I
o
o
0
H

O
o
H
W
O
O
H
W

Mammalian
Species
Mouse

Mouse
Mouse
Rat
Rat
Rat
Mouse
Rat
Rat


Rat
Rat
Rat
Mouse
Rat
Rat


Rat
Rat

Rat

Mouse
Mouse
Rat
Rat



No.
of
Doses
3

2
2
3
2
4
3
4
2


2
4
4
5
2
4


5
2

2

4
7
2
2


Table
Route of
Exposure
FD

DR
DR
DR
DR
GV
DR
GV
GV


DR
FD
DR
DR
DR
GV


GV
FD

GV

GV
FD
DR
DR


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
4

18
12
30
4
18
6
18
91


21
1
30
10
30
14


14
14

102

12
30
126
20


Duration
Units
w

w
w
d
w
d
w
d
d


d
w
d
w
d
d


d
mo

d

d
d
d
w


Age
3

6-8
NR
52
99
3
7
2
NR


80
NR
NR
NR
52
14


20
0

30

6
NR
1
10


Age
Units
mo

w
NR
d
d
d
w
d
NR


d
NR
NR
NR
d
d


d
d

d

d
NR
d
w


Lifestage
JV

LC
GE
JV
JV
JV
SM
JV
JV


JV
LC
JV
JV
JV
JV


JV
JV

LC

JV
LC
GE
GE


Sex
B

F
M
M
B
M
M
B
M


F
F
M
M
M
NR


NR
NR

F

M
F
F
F


Effect
Type
GRO

GRO
GRO
GRO
GRO
GRO
GRO
GRO
GRO


GRO
GRO
GRO
GRO
GRO
GRO


GRO
GRO

GRO

GRO
GRO
GRO
GRO


Effect
Measure
BDWT

BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT
BDWT


BDWT
BDWT
BDWT
BDWT
BDWT
BDWT


BDWT
BDWT

BDWT

BDWT
BDWT
BDWT
BDWT


Response
Site
wo

wo
wo
wo
wo
wo
wo
wo
wo


wo
wo
wo
wo
wo
wo


wo
wo

wo

wo
wo
wo
wo


NOAEL
(mg/kg
bw/day)
136

137
139
169
171
180
187
200
200


218
230
285
362
364
400


400
431

441

534
632
651
750


LOAEL
(mg/kg
bw/day)
1360

—
—
508
—
—
373
—
_


—
460
—
—
—
800


800
—

	

—
1264
—




-------
December ^
i^-^
o
o







oo
i
oo
VO


o
f?
H
I
o
o
0
H

O
o
H
W
O
O
H
W

Mammalian
Species
Mouse

Rat
Rat
Cattle
Rat
Rat
Rat
Rat
Rat


Rat
Mouse
Dog
Shrew
Rat
Pig


Rat
Rat

Rat

Rat
Rat
Rat
Rat



No.
of
Doses
7

4
2
2
3
4
2
2
2


2
2
2
4
3
2


2
2

2

2
2
2
4


Table
Route of
Exposure
FD

FD
DR
FD
DR
DR
DR
DR
DR


DR
GV
OR
FD
GV
FD


GV
FD

GV

DR
DR
GV
GV


8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop the Eco-SSL
Exposure
Duration
28

18
9
283
92
7
5
26
14


10
3
5
31
58
13


29
5

6

30
50
28
14


Duration
Units
d

d
d
d
d
d
d
d
d


d
w
w
d
d
w


d
w

d

d
d
d
d


Age
NR

NR
21
7
25
25
26
22
26


26
NR
<1
NR
2
4


2
NR

1

27
24
2
18


Age
Units
NR

NR
d
mo
d
d
d
d
d


d
NR
yr
NR
d
w


d
NR

d

d
d
d
d


Lifestage
LC

LC
JV
JV
GE
GE
JV
JV
JV


JV
JV
JV
JV
JV
JV


JV
MA

JV

JV
JV
JV
JV


Sex
F

F
M
M
F
F
F
F
F


F
M
NR
B
B
NR


F
NR

B

M
M
M
NR


Effect
Type
GRO

GRO
GRO
GRO
MPH
GRO
GRO
GRO
MPH


GRO
GRO
GRO
GRO
GRO
GRO


GRO
GRO

GRO

GRO
GRO
GRO
GRO


Effect
Measure
BDWT

BDWT
BDWT
BDWT
GMPH
BDWT
BDWT
BDWT
Other


BDWT
BDWT
BDWT
BDWT
BDWT
BDWT


BDWT
BDWT

BDWT

BDWT
BDWT
BDWT
BDWT


Response
Site
wo

wo
wo
wo
TB
WO
WO
WO
TA


WO
WO
WO
WO
wo
wo


wo
wo

wo

wo
wo
wo
wo


NOAEL LOAEL
(mg/kg (mg/kg
bw/day) bw/day)
1260 2530

1500
— 3.30
— 15.0
— 28.7
— 29.0
— 29.0
— 29.5
— 29.9


— 30.4
— 46.4
— 50.0
— 61.5
— 100
— 173


— 200
— 272

— 328

— 354
— 371
— 400
— 400



-------
December 200f
•^s i






oo
OD
o
DRAFT-DO I
^
0
H
O
O
H
W
O
O
H
W
Table 8-1.5.4 (cont'd). Mammalian Toxicity Data Used to Develop
No.
Mammalian of Route of Exposure
Species Doses Exposure Duration
Mouse
Rat
Rat
Mouse
Rat
Rat
Rat
Rat
Rat
Rat
Rat
2 DR 45
4 FD 1
2 DR 6
4 DR 10
2 FD 21
2 GV 18
2 FD 2
4 GV 14
2 FD 2
3 GV 14
2 FD 14
the Eco-SSL
NOAEL LOAEL
Duration Age Effect Effect Response (mg/kg (mg/kg
Units Age Units Lifestage Sex Type Measure Site bw/day) bw/day)
d 50-100
w NR
w 14
w 11
d NR
d 2
w 0
d 24
w 60-80
d 16
d 60
d GE F GRO
NR LC F GRO
w LC F GRO
w JV F GRO
NR LC F GRO
d JV B GRO
d JV NR GRO
d JV NR GRO
d JV M GRO
d JV NR GRO
d JV M GRO
BDWT WO — 404
BDWT WO — 442
BDWT WO — 638
BDWT WO — 748
BDWT WO — 991
BDWT WO — 1000
BDWT WO — 1430
BDWT WO — 1600
BDWT WO — 2390
BDWT WO — 2400
BDWT WO — 2650
AD = adult; B = both; BDWT = body weight changes; d = days; DEYO = death of young; DR = drinking water; F = female; FD = food; PERT = fertility; GMPH = general morphology;
GRO = growth; GV = gavage; JV = juvenile; LC = lactation; M = male; MA = mature; mo = months; MPH = morphology; NR = not reported; ODVP = offspring development; OR = other oral;
PG = prostate gland; PROG = progeny counts or numbers; PRWT = progeny weight; RBPH = reproductive behavior; REP = reproduction; RHIS = reproductive organ histology; RSEM = resorbed
embryos; RSUC = reproductive success (general); RT = reproductive tissue; SM = sexually mature; SPCL = sperm cell counts; SPCV = sperm cell viability; TA = tail; TB = tibia; TE = testes;
TEDG = testes degeneration; TEWT = testes weight; w = weeks; WO = whole organism; yr = years.
Source: U.S. Environmental Protection Agency (2005a)































-------
                 10000
                                                            • Growth NOAEL
                                                            o Growth LOAEL
                                                            A Reproduction NOAEL
                                                            A Reproduction LOAEL
                                                            - Eco-SSL (mg/kg dw)
                               Reproduction and Growth

     Figure 8-1.5.2.  Mammalian toxicity data considered in development of the Eco-SSL.
      Source:  U.S. Environmental Protection Agency (2005a).
 1   Protection Agency, 2005a) generally do not support this.  While five studies reported decreased
 2   survival at these levels, 34 other studies reported no mortality or a LOAEL for mortality at
 3   significantly higher doses (U.S. Environmental Protection Agency, 2005a). The five studies that
 4   supported this low toxic level were conducted on three species (mouse, rat, and cow) and used
 5   either gavage or drinking water as the exposure method. The 34 other studies included data on
 6   these three species as well as five other species (rabbit, dog, pig, hamster, and shrew) and
 7   included gavage and drinking water as well as food ingestion exposure methods.  The NOAELs
 8   for survival ranged from 3.5 to 3200 mg/kg-day (U.S. Environmental Protection Agency, 2005a).
 9   Therefore, the review of data in the Eco-SSL document suggests effects on survival generally
10   would occur at doses greater than those reported to be toxic in the 1986 Pb AQCD (U.S.
11   Environmental Protection Agency, 1986a).
12
     December 2005
8-91
DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.5.5  Recent Studies on the Effects of Lead on Decomposers
 2          Recent studies on effects of Pb to two groups of decomposers are summarized in this
 3    subsection.  Effects on terrestrial invertebrates, such as earthworms and springtails, are described
 4    first, followed by effects on microorganisms.
 5
 6    Effects on Invertebrates
 1          Since the 1986 Pb AQCD, there have been several studies in which terrestrial
 8    invertebrates were exposed to Pb in soil. Many of these were reviewed during the development
 9    of the Eco-SSLs (U.S. Environmental Protection Agency, 2005a). The relevant information
10    from the Eco-SSL document is described below.
11          A literature search and review was conducted to identify critical papers published since
12    2002.  Effects on earthworms and other invertebrates observed in studies conducted since the
13    1986 Pb AQCD are similar to those reported previously:  mortality and decreased growth and
14    reproduction (Lock and Janssen, 2002; Davies et al., 2002; Rao et al., 2003; Bongers et al., 2004;
15    Nursita et al., 2005; U.S. Environmental Protection Agency, 2005a).
16          The literature search completed for terrestrial invertebrate Eco-SSL development
17    identified 179 papers for detailed review, of which  13 met the minimum criteria for further
18    consideration (U.S. Environmental Protection Agency, 2005a).  Most of the 18 ecotoxicological
19    endpoints that were further evaluated measured reproduction or survival as the ecologically
20    relevant endpoint. Four of these, representing one species under three different pH test
21    conditions were used to develop the Eco-SSL of 1700 mg/kg soil (Table 8-1.5.5).
22          In a study designed to test the toxicity of Pb  to the earthworm Eiseniafetida, Davies
23    et al. (2002) found that the 28-day LC50 for Pb in soils  contaminated with Pb(NO3)2 was
24    4379 + 356 mg/kg. Twenty-eight day ECso values for  weight change and cocoon production
25    were 1408 + 198  and 971 + 633 mg/kg, respectively.  Significant mortalities were noted at
26    concentrations of 2000 mg/kg. These data are consistent with those reported in the Eco-SSL
27    document (U.S. Environmental Protection Agency, 2005a) for the same species of earthworm.
28          Nursita et al. (2005) found no mortality and no adverse effects on reproduction (i.e.,
29    number of juveniles) of the collembolan Proisotoma minuta exposed for 42 days to 300, 750,
30    1500, or 3000 mg Pb/kg as Pb-nitrate in an acidic (pH  = 4.88) sandy loam soil.  It was noted that
31    the soils were allowed to equilibrate for 4 weeks after adding the Pb-nitrate before the organisms

      December 2005                         8-92           DRAFT-DO NOT QUOTE OR CITE

-------
               Table 8-1.5.5.  Invertebrate Toxicity Data Used to Develop the Eco-SSL
Invertebrate Species
Collembola
(Folsomia Candida)
Collembola
Collembola
Collembola

Soil pH
6.0
4.5
5.0
6.0

% Organic
Matter
10
10
10
10

Toxicity Parameter
MATC1 (reproduction)
MATC (reproduction)
MATC (reproduction)
MATC (reproduction)
Geometric Mean
Pb in Soil
(mg/kg dw)
3162
3162
894
894
1682
      *  MATC = Maximum Acceptable Threshold Concentration, or the geometric mean of the NOEC (no-observed-
         effect concentration) and LOEC (lowest-observed-effect concentration).
      Source: U.S. Environmental Protection Agency (2005a).
 1   were added. The observation of no effect at 3000 mg/kg is consistent with that of Sandifer and
 2   Hopkin (1996). Sandifer and Hopkin (1996) determined a NOEC and LOEC for collembolan
 3   reproduction of 2000 and 5000 mg/kg, respectively. (A MATC of 3162 mg/kg was used to
 4   devel op the Eco- S SL).
 5         The remaining 14 toxicity endpoints that were not used to develop the Eco-SSL for
 6   invertebrates are presented in Table 8-1.5.6. None of these endpoints were considered eligible
 7   for Eco-SSL derivation.
 8         Lock and Janssen (2002) exposed the potworm Enchytraeus albidus to Pb, as Pb-nitrate.
 9   The 21-day LCso was 4530 mg/kg, and the 42-day ECso for juvenile reproduction was
10   320 mg/kg. The Fl generation was then grown to maturity in the same concentration soil and
11   subsequently used in a reproduction test. The ECso for the Fl generation (394 mg/kg) was
12   similar to that of the P  generation. The authors concluded that the two-generation assay did not
13   increase the sensitivity of the test (Lock and Janssen, 2002). None of the 18 toxicity endpoints
14   evaluated in detail during development of the Eco-SSLs used this species.  The LCso reported for
15   the potworm was higher than reported for nematodes and  similar to that reported for the
16   earthworm. The ECso for reproduction was lower than reported for the earthworm or collembola.
17
     December 2005                        8-93           DRAFT-DO NOT QUOTE OR CITE

-------
             Table 8-1.5.6. Invertebrate Toxicity Data Not Used to Develop the Eco-SSL
Invertebrate Species
Nematode
Nematode
Nematode
Nematode
Nematode
Nematode
Earthworm
Earthworm
Earthworm
Earthworm
Nematode
Nematode
Nematode
Nematode
Soil pH
4
4
4
4
6.2
5.1
6.3
6.1
6.0
6.5
4
4
6.1
6.2
% Organic
Matter
1.14
1.14
4.2
4.2
1.7
3.0
10.0
10.0
10.0
10.0
10
10
3.4
2.2
Toxicity Pb in Soil
Parameter (mg/kg dw)
LC50 285
LC50 297
LC50 847
LC50 1341
LC50 1554
LC50 891
EC50 1940
EC50 1629
LC50 3716
ILL 1.16
LC50 1434
NOAEC 2235
LC50 13.9
LC50 11.6
      *NOAEC (no-observed-adverse-effect concentration); LC50 (concentration lethal to 50% of test population);
       EC50 (effect concentration for 50% of test population); ILL (incipient lethal level).
      Source:  U.S. Environmental Protection Agency (2005a).
1          Recent work by Bongers et al. (2004) cautioned against attributing all toxicity observed in
2    a spiked-soil toxicity test to Pb. They found that the counterion may also contribute to
3    thetoxicity of Pb in the springtail Folsomia Candida. This may have implications on the
4    interpretation of the Eco-SSL data, because the toxicity of the counterion (nitrate) was not taken
5    into account during Eco-SSL development.  Percolation (removal of the counterion) had no
6    statistically significant effect on Pb-chloride toxicity (LCso = 2900 mg/kg for both non-
7    percolated and percolated soil; ECso for reproduction = 1900 mg/kg or 2400 mg/kg for non-
8    percolated or percolated soil, respectively).  However, percolation did have a significant effect
     December 2005
8-94
DRAFT-DO NOT QUOTE OR CITE

-------
 1    on Pb-nitrate toxicity (LCso = 980 mg/kg or 2200 mg/kg for non-percolated and percolated soil,
 2    respectively; ECso for reproduction = 580 mg/kg or 1700 mg/kg for non-percolated or percolated
 3    soil, respectively). Lead nitrate was more toxic than Pb-chloride for survival and reproduction.
 4    However, the toxicity of Pb, from chloride or nitrate, was not significantly different after the
 5    counted on was percolated out of the test soil.  It is noted that the soil was left for 3 weeks to
 6    equilibrate before testing.  Lock and Janssen (2002) also found that Pb-nitrate was more toxic
 7    than Pb-chloride, and they used Pb-nitrate in their experiments because 1000 mg/kg Pb-chloride
 8    did not produce any mortality in their range-finding tests. This difference in chloride and nitrate
 9    toxicity has not been found for earthworms (Neuhauser et al., 1985; Bongers et al., 2004).
10          Rao et al. (2003) exposed the earthworm Eisenia fetida to Pb-oxide in an artificial soil
11    with a pH of 6 at the LCso concentration of 11 mg/kg. Exposure for 14 days resulted in a number
12    of effects including body fragmentation, protrusions, rupture of the cuticle, etc. Many of these
13    effects may trigger defensive mechanisms. For example, fragmentation of the affected posterior
14    region was followed by regeneration and a new ectoderm layer was formed to cover affected
15    areas, both of which processes may serve to prevent soil bacteria from further affecting the
16    earthworm (Rao et al.,  2003 ).
17
18    Effects on Microorganisms and Microbial Processes
19          Microorganisms and microbial processes were not included in the Eco-SSL development
20    process (see Attachment 1-2 of OSWER Direction 92857-55 dated November 2003). Many
21    reasons were given, including that it is unlikely that site conditions would only pose
22    unacceptable risk to microbes and not be reflected as unacceptable risks to higher organisms; that
23    the significance of laboratory-derived effects data to the ecosystem is uncertain; and that the
24    spatial (across millimeter distances) and temporal (within minutes to hours) variation makes
25    understanding ecological consequences challenging. Microbial endpoints often vary
26    dramatically based on moisture, temperature, oxygen, and many non-contaminant factors.
27    Therefore, the recommendation arising from the Eco-SSL development process was that risks to
28    microbes or microbial processes not be addressed through the chemical screening process but
29    that they should be addressed within a site-specific risk assessment (U.S. Environmental
30    Protection Agency, 2005a).
      December 2005                        8-95           DRAFT-DO NOT QUOTE OR CITE

-------
 1          Few studies on the effects of Pb to microbial processes have been published since 1986.
 2    As the direct toxicity to fungi and bacterial populations are difficult to determine and interpret,
 3    indicators for soil communities are often measured as proxies for toxicity (e.g., urease activity in
 4    soil). Recent studies of this nature  (Doelman and Haanstra, 1986; Wilke, 1989; Haanstra and
 5    Doelman, 1991) are summarized in this subsection. The Pb concentrations in these recent
 6    studies (1000 to 5000 mg/kg) are consistent with those reported in the 1986 Pb AQCD as
 7    associated with effects on microbial processes (750 to 7500 mg/kg).
 8          The effects of Pb-chloride on the processes of nitrification and nitrogen mineralization
 9    were studied in a 28-day experiment by Wilke (1989). The authors reported that nitrification
10    was increased by 12 and 16% at levels of 1000 and 4000 mg/kg, respectively, and that nitrogen
11    mineralization was reduced by 32 and 44% at concentrations of 1000 and 4000 mg/kg,
12    respectively.
13          The effects of Pb on arylsulfatase (Haanstra and Doelman, 1991) and urease activity
14    (Doelman and Haanstra, 1986) in soil were investigated. LCsos for decreases in arylsulfatase
15    activity were reported at Pb concentrations of 3004 and 4538 mg/kg in a silty loam soil, at pH 6
16    and 8, respectively. The LCso for a decrease in urease activity was 5060 mg Pb/kg in a sandy
17    loam soil.
18
19    8.1.5.6  Summary
20          The current document expands upon and updates knowledge related to the effects of Pb
21    on terrestrial primary producers, consumers, and decomposers.
22
23    Primary Producers
24          The effects of Pb on terrestrial plants include decreased photosynthetic and transpiration
25    rates in addition to decreased growth and yield. The phytotoxicity of Pb is considered to be
26    relatively low, and there are few reports of phytotoxicity from Pb exposure under field
27    conditions.  Recently, phytotoxicity data were reviewed for the development of the Eco-SSL
28    (U.S. Environmental Protection Agency, 2005 a). Many of the toxicity data presented in the Eco-
29    SSL document (U.S. Environmental Protection Agency, 2005a) are lower than those discussed in
30    the 1986 Pb AQCD, although both  documents acknowledged that toxicity is observed over a
31    wide range of concentrations of Pb in soil (tens to thousands of mg/kg soil).  This may be due to

      December 2005                         8-96           DRAFT-DO NOT QUOTE OR CITE

-------
 1    many factors, such as the soil conditions (e.g., pH, organic matter) and differences in
 2    bioavailability of the Pb in spiked soils perhaps due to lack of equilibration of the Pb solution
 3    with the soil after spiking. Most phytotoxicity data continue to be developed for agricultural
 4    plant species (i.e., vegetable and grain crops). Few data are available for trees or native
 5    herbaceous plants, although two of the five toxicity endpoints used to develop the Eco-SSL were
 6    for trees and two were for clover.
 7
 8    Consumers
 9          Effects of Pb on avian and mammalian consumers include decreased survival,
10    reproduction, and growth as well as effects on development and behavior. There remain few
11    field effects data for consumers, except from sites with multiple contaminants, for which it is
12    difficult to attribute toxicity specifically to Pb. Avian and mammalian toxicity data recently
13    were reviewed for the development of Eco-SSLs (U.S. Environmental Protection Agency,
14    2005a).  Many of the toxicity data presented in the Eco-SSL document (U.S. Environmental
15    Protection Agency, 2005a) are lower than those discussed in the 1986 Pb AQCD, although the
16    EPA (U.S. Environmental Protection Agency, 2005a) recognizes that toxicity is observed over a
17    wide range of doses (<1 to >1000 mg Pb/kg bw-day). Most toxicity data for birds have been
18    derived from chicken and  quail studies, and most data for mammals have been derived from
19    laboratory rat and mouse studies. Data derived for other species would contribute to the
20    understanding of Pb toxicity, particularly for wildlife species with different gut physiologies.  In
21    addition, data derived using environmentally-realistic exposures, such as from Pb-contaminated
22    soil and food, may be recommended. Finally, data derived from inhalation exposures, which
23    evaluate endpoints such as survival,  growth, and reproduction, would contribute to understanding
24    the implications of airborne releases of Pb.
25
26    Decomposers
27          Effects of Pb on soil invertebrates include decreased survival, growth, and  reproduction.
28    Effects on microorganisms include changes in nitrogen mineralization and enzyme activities.
29    Recent data on Pb toxicity to soil invertebrates and microorganisms are consistent with those
30    reported in the 1986 Pb AQCD (U.S. Environmental Protection Agency,  1986a), with toxicity
31    generally observed at concentrations of hundreds to thousands of mg/kg soil. Studies on

      December 2005                         8-97           DRAFT-DO NOT QUOTE OR CITE

-------
 1    microbial processes may be influenced significantly by soil parameters, and the significance of
 2    the test results is not clear.
 3
 4    Ecological Soil Screening Levels (Eco-SSLs)
 5          Eco-SSLs are concentrations of contaminants in soils that are protective of ecological
 6    receptors (U.S. Environmental Protection Agency, 2005b).  They were developed following
 7    rigorous scientific protocols and were subjected to two rounds of peer review.  The Eco-SSLs for
 8    terrestrial plants, birds, mammals, and soil invertebrates are 120, 11, 56, and 1700 mg Pb/kg soil,
 9    respectively.
10
11    8.1.6   Effects of Lead on Natural Terrestrial Ecosystems
12    8.1.6.1  Introduction
13          The concept that organisms are part of larger systems that include both biotic and abiotic
14    components of the environment dates back to the naturalists of the Victorian era. However, the
15    breakthrough in what we now consider the ecosystem approach to ecology occurred in the 1950s
16    and 1960s when E.P. and H.T. Odum pioneered the quantitative analysis of ecosystems
17    (Odum, 1971).  This approach encouraged the calculation of energy flows into, out of, and
18    within explicitly defined ecosystems.  The rapid development of computer technology aided in
19    the growth of ecosystem ecology by allowing the development and use of increasingly complex
20    models for estimating  fluxes that could not be directly measured.
21          It was not long  before the quantitative analysis of ecosystems was extended to examine
22    the flows of nutrients and other chemical compounds. In temperate terrestrial systems, the
23    watershed was identified as a convenient and informative experimental unit (Bormann and
24    Likens, 1967). A major conceptual breakthrough in the watershed approach was that drainage
25    water chemistry could be used as an indicator of the "health" of the ecosystem.  In a system
26    limited by nitrogen, for example, elevated concentrations of NOs  in drainage waters indicate
27    that the ecosystem is no longer making optimal use of available nutrients.
28          The ecosystem  approach can also be used effectively in the study of trace element
29    biogeochemistry.  Input-output budgets can be used to determine whether an ecosystem is a net
30    source or sink of a trace element. Changes to the input-output balance over time can be used to
31    assess the effects of natural or experimental changes in deposition, land use, climate, or other

      December 2005                         8-98            DRAFT-DO NOT QUOTE OR CITE

-------
 1    factors.  In addition, examination of fluxes within the ecosystem (in plant uptake, soil solutions,
 2    etc.) can be used to understand the processes that are most influential in determining the fate and
 3    transport of the trace element.
 4          Many published ecosystem studies include data for 1 to 3 years, the typical duration of
 5    research grant funding or doctoral dissertation research.  While these studies enrich our
 6    understanding of terrestrial ecosystems, the most valuable studies are those that are maintained
 7    over many years. Natural variations in climate, pests, animal migrations, and other factors can
 8    make inferences from short-term studies misleading (Likens, 1989). To nurture long-term
 9    research, the National Science Foundation supports a network of Long-Term Ecological
10    Research (LTER) sites that represent various biomes.
11          This section describes terrestrial ecosystem research on Pb, focusing on work done since
12    the  1986 Pb AQCD and highlighting key long-term studies. Unfortunately, there are few studies
13    that feature long-term data on trace metal behavior. Therefore, this examination of the effects of
14    Pb on terrestrial ecosystems combines insights from long- and short-term investigations as well
15    as including observations from whole-ecosystem and single-component studies.
16
17    8.1.6.2  Effects  of Terrestrial Ecosystem Stresses  on Lead Cycling
18          Terrestrial ecosystems may respond to stressors in a variety of ways, including reductions
19    in the vigor and/or growth of vegetation, reductions in biodiversity, and effects on microbial
20    processes. Each  of these effects may lead to the "leakage" of nutrients, especially nitrogen, in
21    drainage waters.  The reduced vigor or growth of vegetation results in a lower uptake of nitrogen
22    and other nutrients from soils. Reduced biodiversity accompanied by lower total net primary
23    productivity for the ecosystem would also result in a lower nutrient uptake. Effects of stress in
24    microbial populations are less obvious.  If the stress reduces microbial activity rates, then
25    nutrients bound in soil organic matter (e.g., organic nitrogen compounds) will likely be
26    mineralized at a lower rate and retained in the system. On the other hand, disturbances such as
27    clear-cutting,  ice-storm damage, and soil freezing can result in substantial nutrient losses from
28    soils (Bormann et al., 1968; Likens et al.,  1969; Mitchell et al., 1996; Groffman et al., 2001;
29    Houlton et al., 2003).
30          Since the movement and fate of Pb in terrestrial ecosystems is strongly related to the
31    organic matter cycle (Section 8.1.3), stressors that could lead to disruption or alteration of the

      December 2005                         8-99           DRAFT-DO NOT QUOTE OR CITE

-------
 1    soil organic matter pool are of particular concern in assessing effects of ecosystem stress on Pb
 2    cycling. By binding soluble Pb, soil organic matter acts as a barrier to the release of Pb to
 3    drainage waters (Wang et al., 1995; Kaste et al., 2003; Watmough and Hutchinson, 2004).  As a
 4    result, concentrations of Pb in soil solutions and drainage waters tend to be low (Driscoll et al.,
 5    1988; Wang et al., 1995; Bacon and Bain, 1995; Johnson et al., 1995b). Through decomposition
 6    and leaching, soluble organic matter is released to solution, and with it, some Pb is also
 7    mobilized. Wang and Benoit (1996) found that essentially all of the Pb in soil solutions in  a
 8    hardwood forest in New Hampshire was bound to dissolved organic matter (DOM). This release
 9    of soluble Pb does not typically result in elevated surface water Pb concentrations, because
10    (1) organic matter has a relatively long residence time in most temperate soils (Gosz et al.,  1976;
11    Schlesinger, 1997), so only a small fraction of the organic matter pool is dissolved at any time;
12    (2) DOM-Pb complexes solubilized in upper soil horizons may be precipitated or adsorbed lower
13    in the soil profile; (3) the DOM to which Pb is bound may be utilized by microbes, allowing the
14    associated Pb to bind anew to soil organic matter.  Together, these factors tend to moderate the
15    release of Pb to surface waters in temperate terrestrial ecosystems.  However, stressors or
16    disturbances that result in increased release of DOM from soils could result in the unanticipated
17    release of Pb to groundwater and/or surface waters.
18
19    8.1.6.2.1   Acidification
20          The effect of acidification on ecosystem cycling of Pb is difficult to predict.  Like most
21    metals, the solubility of Pb is increased at lower pH (Stumm and Morgan, 1995), suggesting that
22    enhanced mobility of Pb should be found in ecosystems under acidification stress. However,
23    reductions in pH may also decrease the solubility of DOM, via protonation of carboxylic
24    functional groups (Tipping and Woof, 1990).  Because of the importance  of complexation with
25    organic matter to Pb mobility in soils, lower DOM concentrations resulting from acidification
26    may offset the  increased solubility of the metal.
27          In a study of grassland and forest soils at the Rothamsted Experiment Station in England,
28    long-term (i.e., >100 years) soil acidification significantly increased the mobility of Pb in the soil
29    (Blake  and Goulding, 2002).  However, the increased mobility was only observed in very acid
30    soils, those with pH of <4.5.  The fraction of exchangeable Pb  (extracted with 0.1 M CaCb)
31    increased from about 3% to 15% of the total Pb in the most acidified soils.  Similarly, the

      December 2005                        8-100           DRAFT-DO NOT QUOTE OR CITE

-------
 1    fraction of organically bound Pb increased from about 2% of total Pb in neutral soils to 12% of
 2    total Pb in the most acidified soils.  Similarly, Nouri and Reddy (1995) observed higher levels of
 3    DTPA-extractable Pb in soils in a loblolly pine forest treated with simulated acid rain, but only
 4    in the most acidic treatment, with simulated rain with a pH of 3.5.
 5          Although acidification may increase the mobility of Pb in soils, it is not clear that this Pb
 6    is actually moving through or out of the soil profile. In an examination of running waters in
 7    Sweden, Johansson et al. (1995) found no relationship between acidification and Pb
 8    concentrations and concluded that Pb concentrations were governed by the DOM concentration,
 9    which masked any association with acidification.  In an in situ lysimeter study, Bergkvist (1986)
10    measured lower concentrations of Pb in soil solutions draining experimentally acidified plots
11    than in unacidified plots.  In a laboratory study using large soil columns, Merino and Garcia-
12    Rodeja (1997) observed no effect of experimental acidification on the release of Pb to soil
13    solution.  Thus, while acidification may increase the potential mobility of Pb in soils, as
14    indicated by increases in labile  soil fractions such as exchangeable and DTPA-extractable Pb, the
15    actual movement of Pb in the soil is limited by DOM solubilization and transport.  It is worth
16    noting that in all of these studies,  significant effects of acidification were observed for other trace
17    metals (Bergkvist, 1986; Johansson et al., 1995; Merino and Garcia-Rodeja, 1997).
18          Acidification may enhance Pb export to drainage water in very sandy soils,  soils with
19    limited ability to retain organic matter.  Studies in the McDonald's Branch watershed in the
20    New Jersey pine barrens, where soil texture is  similar to beach sands, suggested little Pb
21    retention in the mineral soil (Swanson and Johnson, 1980; Turner et al., 1985).  If acidification
22    results in the mobilization of Pb and organic matter into these mineral soils, then increased
23    streamwater Pb concentrations would likely follow.
24
25    8.1.6.2.2   Forest Harvesting
26          Forest harvesting represents a severe disruption of the organic  matter cycle in forest
27    ecosystems.  Litter inputs are severely reduced for several years after cutting (e.g., Hughes  and
28    Fahey,  1994).  The removal of the forest canopy results in reduced interception of precipitation,
29    and, therefore, increased water flux to the soil  surface.  Also, until a new canopy closes, the soil
30    surface is exposed to increased solar radiation  and higher temperatures. Together,  the higher
31    moisture and temperature in surface soils tend  to increase the rate of organic matter

      December 2005                         8-101          DRAFT-DO NOT QUOTE OR CITE

-------
 1    decomposition.  Several studies have estimated decreases of up to 40% in the organic matter
 2    content of forest floor soils after clear-cutting (Covington, 1981; Federer, 1984; Johnson et al.,
 3    1995a).  This loss of organic matter from the forest floor could result in the mobilization of
 4    organically complexed Pb. However, observations from clear-cut sites in the United States and
 5    Europe indicate that forest harvesting causes little or no mobilization of Pb from forest soils.
 6          At the Hubbard Brook Experimental Forest in New Hampshire, whole-tree harvesting, the
 7    most intensive form of clear-cutting, resulted in very small increases in Pb concentrations in soil
 8    solutions draining the Oa soil horizon despite substantial reductions in the organic matter mass of
 9    that horizon (Fuller et al.,  1988; Johnson et al., 1995a).  These increases were associated with
10    similarly small increases in dissolved organic carbon (DOC) concentrations in the Oa horizon
11    soil water.  Output of Pb from the watershed stream was unaffected by clear-cutting. Similarly,
12    Berthelsen and Steinnes (1995) observed small decreases in the Pb content of the Oa horizon
13    ("H" in the European system of soil classification) in clear-cut sites in Norway,  compared to
14    uncut reference  sites.  This mobilization of Pb from the Oa horizon was accompanied by an
15    increase in the Pb content of the upper mineral soil horizons. The Pb  decline in the Oa horizon
16    was accompanied by  a decrease in the organic matter content, leading the authors  to attribute the
17    Pb dynamics to  leaching with DOM.  In a study conducted in Wales, Durand et  al. (1994)
18    observed lower  Pb outputs from a stream draining a clear-cut watershed than from where the
19    stream drained the upper reaches of the watershed, which were uncut. The DOC and H+ outputs
20    were also lower in the clear-cut area.  These patterns persisted in all 5 years of the study.
21          Forest harvesting is a severe form of ecosystem disturbance,  and, thus, it is somewhat
22    surprising that studies of clear-cutting have shown little or no effect on Pb mobility or loss from
23    forest ecosystems. Perhaps the strong complexation behavior of Pb with natural organic matter
24    results in the retention of Pb in forest soils.  Even in cases where Pb is mobilized in forest floor
25    soils (Fuller et al., 1988; Berthelsen and Steinnes, 1995), there is no evidence of loss of Pb from
26    the ecosystem, indicating that mineral soils are efficient in capturing and retaining any Pb that is
27    mobilized in the forest floor. Therefore, the principal risk associated with forest harvesting is the
28    loss of Pb in particulate form to drainage waters through erosion.
29
30
      December 2005                        8-102           DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.6.2.3  Land Use and Industry
 2          Changes in land use also represent potentially significant changes in the cycling of
 3    organic matter in terrestrial ecosystems. Conversion of pasture and croplands to woodlands
 4    changes the nature and quantity of organic matter inputs to the soil. In temperate climates, forest
 5    ecosystems tend to accumulate organic matter in an O horizon on the forest floor, whereas
 6    organic matter in grasslands and agricultural fields is concentrated in an A horizon at the soil
 7    surface.  Andersen et al. (2002) compared the trace metal concentrations in arable fields in
 8    Denmark to nearby sites that had been converted to forest land. After 34 years of afforestation,
 9    the soils showed no significant difference in Pb concentration or fractionation, despite significant
10    acidification of the soils. Afforestation had no effect on the soil carbon concentration,
11    suggesting that land use change may have little effect on Pb cycling unless soil carbon pools are
12    affected.
13          Similarly,  the introduction of industrial activity may have consequences for organic
14    matter cycling, and subsequently, Pb mobilization. In a rare long-term study of polluted soils,
15    Egli et al. (1999) studied the changes in trace metal concentrations in forest soils at a site in
16    western  Switzerland between 1969 and 1993. The site is 3 to 6 km downwind from an aluminum
17    industrial plant that operated between the 1950s and 1991. In the 24-year period of study, the
18    site experienced significant declines in organic carbon in surface (0 to 5 cm depth) and
19    subsurface (30 to 35 cm) soils.  In the 30 to 35 cm layer, the organic carbon concentration
20    declined by more than 75%. Extractable Pb (using an ammonium acetate and EDTA mixture)
21    declined by 35%  in the same layer.  The authors suggested that the Pb lost from the soil had been
22    organically bound. While this study indicates that loss of soil carbon can induce the mobilization
23    and loss of Pb from terrestrial ecosystems, it is also worth noting that the decline in soil Pb was
24    considerably smaller than the decline in organic carbon. This suggests that Pb mobilized during
25    organic matter decomposition can resorb to remaining organic matter or perhaps to alternate
26    binding  sites (e.g., Fe and Mn oxides).
27          The effects of industries that emit Pb to the atmosphere are discussed in Sections 8.1.6.3
28    and 8.1.6.4 below.
29
      December 2005                        8-103           DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.1.6.2.4   Climate Change
 2          Atmospheric Pb is not likely to contribute significantly to global climate change. Lead
 3    compounds have relatively short residence times in the atmosphere, making it unlikely that they
 4    will reach the stratosphere. Also, Pb compounds are not known to absorb infrared radiation and,
 5    therefore, are unlikely to contribute to stratospheric ozone depletion or global warming.
 6          Climate change does, however, represent a disturbance to terrestrial ecosystems.
 7    Unfortunately, the potential linkages between climate-related stress and Pb cycling are poorly
 8    understood.  As in the previous examples, effects related to alterations in organic matter cycling
 9    may influence Pb migration.  For example, an increase in temperature leading to increased rates
10    of organic matter decomposition could lead to temporary increases in DOM concentrations and
11    smaller steady-state pools of soil organic matter.  Either of these factors  could result in increased
12    concentrations of Pb in waters draining terrestrial ecosystems.
13          Climate change may also affect the fluctuations of temperature and/or precipitation in
14    terrestrial ecosystems. For example, there is some evidence for recent increases in the frequency
15    of soil freezing events in the northeastern United States (Mitchell et al., 1996). Soil freezing
16    occurs when soils have little to no snow cover to insulate them from cold temperatures and
17    results in an increased release of nitrate and DOC from the O horizons of forest soils (Mitchell
18    et al., 1996; Fitzhugh et al., 2001). Increased DOC losses from O horizons subjected to freezing
19    may also increase Pb mobilization.
20          Increased fluctuations in precipitation may induce more frequent flooding, with
21    potentially significant consequences for Pb contamination of floodplain ecosystems. Soils
22    collected from the floodplain of the Elbe River, in Germany, contained elevated concentrations
23    of Pb and other trace metals (Kruger and Grongroft, 2004).  Tissues of plants from floodplain
24    sites did not, however, contain higher Pb concentrations than control sites. More frequent or
25    more severe flooding would likely result in increased inputs of Pb and other metals to floodplain
26    soils.
27
28    8.1.6.3  Effects of Lead Exposure on Natural Ecosystem Structure and Function
29          The effects of Pb exposure on natural ecosystems are confounded by the fact that Pb
30    exposure cannot be decoupled from other factors that may also  affect the ecosystem under
31    consideration.  Principal among these factors are other trace metals and acidic deposition.

      December 2005                        8-104           DRAFT-DO NOT QUOTE OR CITE

-------
 1    Emissions of Pb from smelting and other industrial activities are accompanied by other trace
 2    metals (e.g., Zn, Cu, Cd) and sulfur dioxide (862) that may cause toxic effects independently or
 3    in concert with Pb. Reductions in the use of alkyl-Pb additives in gasoline have resulted in
 4    significant decreases in Pb deposition to natural ecosystems in the northeastern United States
 5    (Johnson et al., 1995b). However, the period in which Pb deposition has declined (ca. 1975 to
 6    the present) has also seen significant reductions in the acidity (i.e., increased pH) of precipitation
 7    in the region (Likens et al., 1996; Driscoll et al., 1998). Therefore, changes in ecosystem Pb
 8    fluxes may be the result of reduced Pb inputs and/or reduced acidity.
 9          Experimental manipulation  studies do not suffer from these confounding effects, because
10    Pb can be added in specific amounts, with or without other compounds.  Unfortunately,
11    ecosystem-level manipulations involving Pb additions have not been undertaken. Therefore, we
12    must use observations from field studies of Pb behavior in sites exposed to various forms of Pb
13    pollution to assess the effects of Pb on terrestrial ecosystems. This section includes a discussion
14    of effects of Pb in the structure and function of terrestrial ecosystems. Effects on energy flows
15    (food chain effects) and biogeochemical cycling are discussed in Section 8.1.6.4.
16
17    8.1.6.3.1   Sites Affected by Nearby Point Sources of Lead
18          Natural terrestrial ecosystems near smelters, mines, and other industrial plants have
19    exhibited a variety of effects related to ecosystem structure and function. These effects include
20    decreases in species diversity, changes in floral and faunal community composition, and
21    decreasing vigor of terrestrial vegetation.
22          All of these effects were observed in ecosystems surrounding the Anaconda smelter in
23    southwestern Montana, which operated between 1884 and 1980 (Galbraith et al., 1995).  Soils in
24    affected areas around the Ananconda smelter were enriched in Pb, arsenic, copper, cadmium, and
25    zinc; had very low pH; and were determined to be phytotoxic to native vegetation (Kapustka et
26    al., 1995). The elevated soil arsenic and metal concentrations occurred despite significantly
27    lower organic matter concentrations in affected soils relative to reference sites (Galbraith et al.,
28    1995). Line-transect measurements indicated that affected sites had an average of 6.9 species per
29    10-m of transect, compared to 20.3 species per 10-m in the reference areas. More than 60% of
30    the reference sites supported coniferous (58%) or deciduous (3%) forest communities, whereas
31    less than 1% of the affected sites retained functioning forest stands. Abundant dead timber and

      December 2005                        8-105           DRAFT-DO NOT QUOTE OR CITE

-------
 1    stumps confirmed that the affected sites were once as forested as the reference sites.  Affected
 2    grassland sites were also less diverse and had higher abundances of invasive species than
 3    reference grasslands.  More than 50% of the affected sites were classified as bare ground.  The
 4    occurrence of bare ground was significantly correlated with the phytotoxicity scores derived by
 5    Kapustka et al. (1995), indicating a link between phytotoxicity and the loss of vegetation in the
 6    affected area.
 7          Because of the plant community changes near the Anaconda smelter, the vertical diversity
 8    of habitats in the affected ecosystems decreased, with only shrubs and soil remaining as viable
 9    habitats.  Galbraith et al. (1995) also used the Bureau of Land Management's habitat evaluation
10    procedure (HEP) to estimate habitat suitability indices (HSI) for two indicator species, marten
11    (Maries americand) and elk (Cervus elaphus).  The HSI value ranges from 0 (poor habitat) to 1
12    (ideal habitat). In sites affected by the Anaconda smelter, HSI values for marten averaged 0.0,
13    compared to 0.5 to 0.8 for the reference sites. For elk, affected  sites had an average HSI of 0.10,
14    compared to 0.31 at reference sites.
15          Similar observations were made in the area surrounding Palmerton, Pennsylvania, where
16    two zinc smelters operated between 1898 and 1980.  Soils in the area were enriched in Cd, Zn,
17    Pb, and Cu, with concentrations decreasing with distance from the smelter sites (Beyer et al.,
18    1985; Storm et al., 1994).  Smelting was determined to be the principal source of Pb in soils  in
19    residential and undeveloped areas around Palmerton (Ketterer et al., 2001), which lies on the
20    north side of a gap in Blue Mountain, a ridge running roughly east-west in east-central
21    Pennsylvania. Much of the north-facing side of Blue Mountain within 3 km of the town is bare
22    ground or sparsely vegetated, whereas the surrounding natural landscape is predominantly oak
23    forest (Sopper, 1989; Storm et al., 1994). Biodiversity in affected areas is considerably lower
24    than at reference sites, a pattern attributed to emissions from the smelters (Beyer et al., 1985;
25    Sopper,  1989). The history is complicated, however, by the land use history of the area.
26    Logging and fire in the early 20th century may also have played a role in the changes in the
27    terrestrial ecosystems (Jordan, 1975). Extensive logging occurred after the smelters began
28    operation, suggesting that some of the logging may have been salvage logging in affected areas.
29    Regardless, the smelter emissions appear to have inhibited the regrowth of ecosystems compared
30    to those in nearby unaffected areas. As in Anaconda, MT, the changes in the structure and
31    function of the Palmerton ecosystem changed its suitability as a habitat for fauna that would

      December 2005                        8-106           DRAFT-DO NOT QUOTE OR CITE

-------
 1    normally inhabit the area.  Storm et al. (1994) did not find amphibians or common invertebrates
 2    in two study sites nearest to the smelters. In the larger study area, they documented elevated
 3    concentrations of Pb, Cd, Cu, and Zn in tissues of species ranging in size from red-backed
 4    salamanders (Pletheron cenereus) to white-tailed deer (Odocoilius virginianus).
 5          Metal pollution around a Pb-Zn smelter near Bristol, England has not resulted in the loss
 6    of oak woodlands within 3 km of the smelter, despite significant accumulation of Pb, Cd, Cu,
 7    and Zn in soils and vegetation (Martin and Bullock, 1994). However, the high metal
 8    concentrations have favored the growth of metal-tolerant species in the woodland.
 9          The effects of Pb on terrestrial ecosystems near smelters and other industrial  sites
10    decrease downwind from the Pb source. Several studies using the soil Pb burden as an indicator
11    have shown that much of the contamination occurs within a radius of 20 to 50 km around the
12    emission source (Miller and McFee, 1983; Martin and Bullock, 1994; Galbraith et al., 1995;
13    Spurgeon and Hopkin,  1996a; see also Section 8.1.3.). For example, the concentration of Pb in
14    forest litter declined downwind from a Pb-Zn smelter near Bristol, UK, from 2330 to 3050 ppm
15    in a stand 2.9 km from the smelter to 45 to 110 ppm in a stand 23 km from the smelter (Martin
16    and Bullock,  1994). Thus, while sites near point sources of Pb may experience profound effects
17    on ecosystem structure and function, the extent of those effects is limited spatially.  Most
18    terrestrial ecosystems are far enough from point sources that long-range Pb transport is the
19    primary mechanism for Pb inputs.
20
21    8.1.6.3.2   Sites Affected by Long-Range Lead Transport
22          Because the effects of anthropogenic Pb emissions tend to be restricted in geographic
23    extent, most natural terrestrial ecosystems in the U.S. sites have Pb burdens derived primarily
24    from long-range atmospheric transport.  Pollutant Pb represents a large fraction of the Pb in
25    many of these ecosystems. In particular, many of these sites have accumulated large amounts of
26    Pb in soils. For example, at the Hubbard Brook Experimental Forest in New Hampshire, the
27    amount of Pb in the forest floor was estimated to have increased from about 1.35 kg ha'1 in 1926
28    (before the introduction of alkyl-Pb additives in gasoline) to 10.5 kg ha"1 in 1977 (Johnson et al.,
29    1995b). They also estimated the atmospheric Pb deposition from  1926 to 1987 to be 8.7 kg ha'1,
30    an amount that could account for nearly all of the increase in Pb in the forest floor during the
31    period. The input of precipitation Pb to the Hubbard Brook ecosystem in the six decades

      December 2005                       8-107          DRAFT-DO NOT QUOTE OR CITE

-------
 1    spanning 1926 to 1987 was more than half of the total Pb estimated to have been released by
 2    mineral weathering in the entire 12,000- to 14,000-year post-glacial period (14.1 kg ha'1:
 3    (Johnson et al., 2004)).  Other studies employing Pb budgets (Miller and Friedland, 1994;
 4    Watmough et al., 2004), and Pb isotopes (Bacon et al., 1995, 1996; Watmough et al., 1998;
 5    Bindler et al., 1999; Hansmann and Koppel, 2000; Kaste et al., 2003), have also shown that
 6    pollutant Pb, primarily from gasoline combustion, represents a quantitatively significant fraction
 7    of labile Pb in temperate soils, especially in the upper, organic-rich horizons.
 8          Despite years of elevated atmospheric Pb inputs and elevated concentrations in soils, there
 9    is little evidence that sites affected primarily by long-range Pb transport have experienced
10    significant effects on ecosystem structure or function. Low concentrations of Pb in soil
11    solutions, the result of strong complexation of Pb by soil organic matter, may explain why few
12    ecological effects have been observed.  At Hubbard Brook, for example, the concentration of Pb
13    in soil solutions draining the Oa horizon is <0.1 |iM and is even lower in solutions draining
14    mineral-soil horizons (Driscoll et al., 1988; Wang et al., 1995).  Friedland and Johnson (1985)
15    measured similar concentrations in soil solutions collected from deciduous and spruce-fir stands
16    on Camel's Hump Mountain in Vermont.
17          In ecosystems where Pb concentrations in soil solutions are low, toxicity levels for
18    vegetation are not likely to be reached regardless of the soil Pb concentration. Furthermore,
19    mycorrhizal infection of tree roots appears to reduce the translocation of Pb from roots to shoots
20    (Marschner et al., 1996; Jentschke et al., 1998).  In a study of mycorrhizal and non-mycorrhizal
21    Norway spruce (Picea abies (L.) Karst), mycorrhizal infection of roots was not affected by Pb
22    dose.  Some, but not all, species of mycorrhizae showed reductions in the amount of
23    extrametrical mycelium with Pb exposure but only at solution concentrations of 5 //M, a level at
24    least 50 times greater than typical concentrations in forest soils. In a related study, the growth
25    rate of mycorrhizal  fungi was unaffected at solution Pb concentrations of 1 and 10 fjM, but
26    decreased at 500 |iM (Marschner et al., 1999).
27          Low soil solution Pb concentrations and the influence of mycorrhizal symbionts also
28    result in low uptake of Pb by terrestrial vegetation. The net flux of Pb into vegetation in the
29    northern hardwood  forest at Hubbard Brook in the 1980s was estimated as  only 1  g ha'1 year"1
30    (Johnson et al., 1995b), representing 3% of the precipitation input. Klaminder et al. (2005) also
31    measured a Pb uptake of 1 g ha"1 year"1 in a spruce-pine forest in northern  Sweden. Despite

      December 2005                       8-108           DRAFT-DO NOT QUOTE OR CITE

-------
 1    plant uptake fluxes being very low, they are sensitive to differences and changes in Pb
 2    deposition.  Berthelsen et al. (1995) observed decreases in the Pb content of stem, twig, leaf, and
 3    needle tissues of a variety of tree species in Norway between 1982 and 1992, when atmospheric
 4    Pb deposition declined by approximately 70%. They also observed significantly lower Pb
 5    concentrations in tree tissues collected in northern Norway versus southern Norway, where
 6    atmospheric Pb deposition is greater.
 7          Even at subtoxic concentrations, Pb and other metals may influence species diversity in
 8    terrestrial ecosystems. However, little work has been done on the effect of low-level metal
 9    concentrations on species diversity.  In one study, plant species diversity was positively
10    correlated to the concentration of available Pb in natural and artificial urban meadows in Britain
11    (McCrea et al., 2004). The authors hypothesized that Pb may inhibit phosphorous uptake by
12    dominant species, allowing less abundant (but more Pb-tolerant) ones to succeed.
13
14    8.1.6.4  Effects of Lead on Energy Flows and Biogeochemical Cycling
15          In terrestrial ecosystems, energy flow  is closely  linked to the carbon cycle.  The principal
16    input of energy to terrestrial ecosystems is through photosynthesis, in which CC>2 is converted to
17    biomass carbon.  Because of this link between photosynthesis and energy flow, any effect that Pb
18    has on the structure and function of terrestrial ecosystems  (as discussed in Section 8.1.6.3.)
19    influences the flow of energy into the ecosystem. This section focuses on how Pb  influences
20    energy transfer within terrestrial ecosystems, which begin with the decomposition  of litter and
21    other detrital material by soil bacteria and fungi, and cascade through the various components of
22    the detrital food web. Because the mobility of Pb in soils is closely tied to organic matter
23    cycling, decomposition processes are central  to the biogeochemical cycle of Pb.  This section
24    concludes with a discussion of how biogeochemical cycling of Pb has  changed in response to the
25    changing Pb inputs to terrestrial ecosystems.
26
27    8.1.6.4.1    Effects of Lead on Detrital Energy Flows
28          Lead can have a significant effect on energy flows in terrestrial  ecosystems. At some sites
29    severely affected by  metal pollution, death of vegetation can occur, dramatically reducing the
30    input of carbon to the ecosystem (Jordan, 1975; Galbraith  et al., 1995). Subsequently, wind and
      December 2005                        8-109           DRAFT-DO NOT QUOTE OR CITE

-------
 1    erosion may remove litter and humus, leaving bare mineral soil, a nearly sterile environment in
 2    which very little energy transfer can take place (Little and Martin, 1972; Galbraith et al., 1995).
 3          At Pb-affected sites that can retain a functioning forest stand, the rate of decomposition of
 4    litter may be reduced, resulting in greater accumulation of litter on the forest floor than in
 5    unpolluted stands. Numerous investigators have documented significant declines in litter
 6    decomposition rates (Cotrufo et al., 1995; Johnson and Hale, 2004) and/or the rate of carbon
 7    respiration (Laskowski et al., 1994; Cotrufo et al., 1995; Saviozzi et al., 1997; Niklinska et al.,
 8    1998; Palmborg et al., 1998; Aka and Darici, 2004) in acid- and metal-contaminated soils or soils
 9    treated with Pb. The resulting accumulation of organic matter on the soil surface can be
10    dramatic. For example, an oak woodland 3 km from a smelter in Bristol, England had a litter
11    layer mass 10 times greater than the mass in a similar stand 23 km from the smelter (Martin and
12    Bullock, 1994).
13          Lower decomposition rates in polluted ecosystems  are the result of the inhibition of soil
14    bacteria and fungi and its effects on microbial community structure (Baath, 1989). Kuperman
15    and Carreiro (1997) observed 60% lower substrate-induced respiration in heavily polluted
16    grassland soils near the U.S. Army's  Aberdeen Proving Ground in Maryland.  This decline in
17    carbon respiration was associated with 81% lower bacterial biomass and 93% lower fungal
18    biomass.  Similar declines in the activities of carbon-, nitrogen-, and phosphorus-acquiring
19    enzymes were also observed. Such dramatic effects have  only been observed in highly
20    contaminated ecosystems. In a less contaminated grassland site near a Pb factory in Germany,
21    Chander et al. (2001) observed a lower ratio of microbial biomass carbon to soil organic carbon
22    in polluted soils.  The ratio of basal respiration to microbial biomass (the "metabolic quotient,"
23    qCC>2) declined with increasing metal concentration, though this observation depended on the
24    procedure for measuring microbial biomass (substrate-induced respiration versus fumigation-
25    extraction). The combined effect of lower microbial biomass per unit soil carbon and similar or
26    lower qCC>2 on polluted sites indicates that the ability of soil microorganisms to process carbon
27    inputs is compromised by metal pollution.
28          The type of ecosystem also plays  a role in determining the effects of Pb and other metals
29    on the microbial processing of litter.  Forest soils in temperate zones accumulate organic matter
30    at the soil surface to a greater degree than in grasslands. This organic-rich O horizon can support
31    a large microbial biomass; but it is also an effective trap for Pb inputs, because of the association

      December 2005                        8-110           DRAFT-DO NOT QUOTE OR CITE

-------
 1    between Pb and soil organic matter. At highly contaminated forest sites, microbial biomass and
 2    enzyme activities may be depressed (Fritze et al., 1989; Baath et al., 1991), causing slower
 3    decomposition of the litter.
 4          In addition to effects on decomposition and carbon transformations, Pb and other trace
 5    metals can also influence key nitrogen cycling processes. Studies in the 1970s demonstrated that
 6    Pb and other metals inhibit the mineralization of nitrogen from soil organic matter and
 7    nitrification (Liang and Tabatabai, 1977, 1978),  resulting in lower nitrogen availability to plants.
 8    More recent research has documented significant inhibitory effects of Pb and other metals on the
 9    activities of several enzymes believed to be crucial to nitrogen mineralization in soils (Senwo
10    and Tabatabai, 1999; Acosta-Martinez and Tabatabai, 2000; Ekenler and Tabatabai, 2002). This
11    suggests that the inhibitory effect of Pb and other metals is  broad-based, and not specific to any
12    particular metabolic pathway.  In reducing environments, the rate of denitrification is also
13    depressed by trace metals.  Fu and Tabatabai (1989) found  that 2.5 jimol g"1 of Pb (ca.
14    500 mg/kg"1) was sufficient to cause 0, 27, and 52% decreases in nitrogen reductase activity in
15    three different soils.
16          Metal pollution can also affect soil invertebrate populations. Martin and Bullock (1994)
17    observed lower abundances of a variety of woodlice, millipedes, spiders, insects, and earthworms
18    in an oak woodland site 3 km from a Pb-Zn smelter in Bristol, England, compared to a reference
19    site 23 km from the smelter. The differences were most dramatic when expressed per unit mass
20    of litter. Several species that were abundant in the reference site were not found in the
21    contaminated woodland. For example, the abundance of the woodlice Trichoniscus pusillus
22    was 151 individuals per m2 in the reference woodland, but none were found in the contaminated
23    soils. This was also true of 2 of the 3 millipede species, and 4 of the 5 earthworm species
24    studied. At six sites within 1 km from the smelters, no earthworms were present at all (Spurgeon
25    and Hopkin, 1996a). Contamination at this site has apparently reduced both the population and
26    biodiversity of the soil invertebrate community.
27          The effect of metal pollution on soil invertebrates may be a threshold-type response. In a
28    study conducted in  woodlands near two zinc smelters in Noyelles-Godault, in northern France,
29    soils at the most polluted site were devoid of mites and millipedes, while the remaining sites had
30    diversity measures  similar to control sites (Grelle et al., 2000).
      December 2005                        8-111           DRAFT-DO NOT QUOTE OR CITE

-------
 1          While Pb pollution affects the population and diversity of soil fauna, there is little
 2    evidence of significant bioaccumulation of Pb in the soil food web (see also Section 8.1.4.).
 3    In the Bristol, England study, Pb concentrations in earthworms were lower than soil Pb
 4    concentrations and much lower than litter Pb concentrations (Martin and Bullock, 1994). Litter-
 5    dwelling mites had Pb concentrations that were 10% of the average litter concentration.  The
 6    predator centipedes Lithobius forficatus and L. variegatus had mean Pb concentrations of
 7    18.6 and 44.0 mg kg"1, respectively, two orders of magnitude lower than the Pb concentration of
 8    litter (2193 mg kg"1) and lower than the concentrations of their known prey species.  In a study
 9    conducted in a Norway spruce forest affected primarily by automobile exhaust from a nearby
10    highway, earthworms had Pb concentrations similar to the soil (Roth, 1993).  Almost all of the
11    litter decomposers, however, had Pb concentrations that were less than 20% of the litter. All but
12    3 of the zoophagous arthropods had Pb concentrations that were less than 40% of their prey; the
13    remaining 3 had Pb concentrations similar to their prey. Because of the absence of significant
14    bioaccumulation in the soil food web, predator species will be affected by Pb pollution primarily
15    through  effects on the abundance of their prey (Spurgeon and Hopkin, 1996b).
16          Taken as  a whole, ecosystem-level studies of the soil food web indicate that Pb can affect
17    energy flows in terrestrial ecosystems through two principal mechanisms. In the most severely
18    polluted sites, the death of primary producers directly decreases the flow of energy into the
19    ecosystems. More commonly, the accumulation of toxic levels of Pb or other metals in litter and
20    soil decreases the rate of litter decomposition through decreases in microbial biomass and/or
21    respiration.  These reductions can subsequently affect higher trophic levels that depend on these
22    organisms.  It is important to note that sites that have exhibited significant disruption to energy
23    flows and the terrestrial food web are sites that have experienced severe metal contamination and
24    adverse effects from 862 from smelters or other metals-related activities.
25
26    Lead Dynamics in Terrestrial Ecosystems
27          Lead inputs to terrestrial ecosystems in the United States have declined dramatically in the
28    past 30 years, primarily because of the almost complete elimination of alkyl-Pb additives in
29    gasoline in North America.  Also, Pb emissions from smelters have declined as older plants have
30    been shut down or fitted with improved emissions controls. Unfortunately, there are few long-
31    term data sets of precipitation inputs to terrestrial ecosystems. At the Hubbard Brook

      December 2005                         8-112           DRAFT-DO NOT QUOTE OR CITE

-------
 1    Experimental Forest, in New Hampshire, Pb input in bulk deposition declined by more than 97%
 2    between 1976 and 1989 (Johnson et al., 1995b).  Studies of freshwater sediments also indicate a
 3    dramatic decline in Pb inputs since the mid-1970s (Graney et al., 1995; Johnson et al., 1995b;
 4    Farmer et al., 1997; Brannvall et al., 2001a,b).
 5          Reported concentrations of Pb in waters draining natural terrestrial ecosystems have
 6    always been low (Wang et al., 1995; Bacon and Bain, 1995; Johnson et al., 1995b; Vinogradoff
 7    et al., 2005), generally less than 1 ng I/1, even at moderately polluted sites (Laskowski et al.,
 8    1995).  Consequently, most terrestrial ecosystems in North America and Europe remain sinks for
 9    Pb despite reductions in atmospheric Pb deposition of more than 95%.  At Hubbard Brook, for
10    example, the input of Pb in bulk precipitation declined from 325 g ha l year"1 between 1975 and
11    1977 compared to 29 g ha"1 year"1 between 1985 and 1987 (Johnson et al., 1995b). During the
12    same period, the output of Pb in stream water declined from 6 g ha'1 year"1 to 4 g ha l year"1.
13    Thus, despite the  decline in Pb input, 85% of the incoming Pb was still  retained in the terrestrial
14    ecosystem in the later time period.  Similar observations have been made in Europe, where the
15    use of leaded gasoline has also declined in the last few decades.  At the Glensaugh Research
16    Station in Scotland, the input of Pb to the forest ecosystem was estimated as 42.6 g ha"1 year"1
17    between 2001  and 2003, about six times the stream export of 7.2 g ha l year"1 (Vinogradoff et
18    al., 2005).  Similarly, Huang and Matzner (2004) reported a throughfall flux of 16.5 g ha"1 year"1
19    at the forested Lehstenbach catchment in Bavaria, about six times the efflux in runoff of 2.82 g
20    ha"1 year"1.
21          Lead pollution has resulted in the accumulation of large Pb burdens in terrestrial
22    ecosystems (see Sections 8.1.3. and 8.1.6.3.2). Despite reductions in emissions, this
23    accumulation of Pb continues, though at markedly lower rates.  The large pool of Pb bound in
24    soils may potentially be a threat to aquatic ecosystems, depending on its rate of release from the
25    soil. Early estimates of the residence time of Pb in the forest floor ranged from 220 to 5,000
26    years (Benninger et al., 1975; Friedland and Johnson, 1985; Turner et al., 1985).  However, more
27    recent literature suggests that Pb is transported more rapidly within soil profiles than previously
28    believed.  The pool of Pb in forest floor soils of the northeastern United States declined
29    significantly in the late 20th century. Friedland et al. (1992) reported a 12% decline in the
30    amount of Pb in forest floor soils at 30 sites in the region between 1980 and 1990, a much greater
31    decline than would be expected for a pool with a residence time of 220 to 5,000 years.

      December 2005                       8-113           DRAFT-DO NOT QUOTE OR CITE

-------
 1    At Hubbard Brook, the pool of Pb in the forest floor declined by 29% between 1977 and 1987,
 2    an even more rapid rate of loss than reported by Friedland et al. (1992). More recently, Evans
 3    et al. (2005) reported significant declines in the Pb content of forest floor soils in the
 4    northeastern United States and eastern Canada between 1979 and 1996. The magnitude of the
 5    decrease in Pb content was greatest at their sites in southern Vermont, and smallest at sites on the
 6    Gaspe Peninsula in Quebec, reflecting the historic gradient in Pb deposition in the region.
 7          Since drainage water Pb concentrations remain low, the Pb released from forest floor soils
 8    in the past has been largely immobilized in mineral soils (Miller and Friedland, 1994; Johnson
 9    et al., 1995b; Johnson and Petras, 1998; Watmough and Hutchinson, 2004; Johnson et al., 2004).
10    This is supported by evidence from Pb-isotope analyses. Gasoline-derived Pb has a 206Pb:207Pb
11    ratio that can be easily discriminated from Pb in the rocks from which soils are derived. Using
12    isotopic mixing models with gasoline-Pb and Pb in soil parent materials as end members,
13    a number of researchers have documented the accumulation of pollutant Pb in mineral soils
14    (Bindler et al., 1999; Kaste et al., 2003; Watmough and Hutchinson, 2004; Bacon and Hewitt,
15    2005; Steinnes and Friedland, 2005).  In a hardwood stand on Camel's Hump Mountain in
16    Vermont, as much as 65% of the pollutant Pb deposited to the stand had moved into mineral
17    horizons by 2001 (Kaste et al., 2003). In a spruce-fir stand, containing a thicker organic forest
18    floor layer, penetration of pollutant Pb into the mineral soil was much lower.
19          This recent research has resulted in a reevaluation of the turnover time of Pb in forest
20    floor soils. The Camel's Hump data suggest that Pb resides in the forest floor of deciduous
21    stands for about 60 years  and about 150 years in coniferous stands (Kaste et al., 2003).  These
22    values are somewhat greater than those published previously  by Miller and Friedland (1994),
23    who used a Pb budget approach. Extremely rapid turnover of Pb was observed in some
24    hardwood forest floor soils in south-central Ontario (Watmough et al., 2004). Their estimated
25    turnover times of 1.8 to 3.1 years are much lower than any other published values, which they
26    attribute to the mull-type  forest floor at their sites. Mull-type forest floors are normally underlain
27    by organic-rich A horizons, capable of immobilizing Pb released from the forest floor.  Indeed,
28    at the same site in Ontario, Watmough and Hutchinson (2004) found that 90% of the pollutant Pb
29    could be found in this A horizon.
30          The time period over which the accumulated Pb in soils may be released to drainage
31    waters remains unclear. If Pb moves as a pulse through the soil, there may be a point in the

      December 2005                       8-114           DRAFT-DO NOT QUOTE OR CITE

-------
 1    future at which problematic Pb concentrations occur. However, several authors have argued
 2    against this hypothesis (Wang and Benoit, 1997; Kaste et al., 2003; Watmough et al., 2004),
 3    contending that the strong linkage between Pb and DOM will result in a temporally dispersed
 4    release of Pb in the form of Pb-DOM complexes. Thus, the greatest threat is likely to be in the
 5    most highly contaminated areas surrounding point sources of Pb, where the amount of Pb
 6    accumulated in the soil is high, and the death of vegetation has resulted in reduced soil organic
 7    matter levels.
 8
 9    8.1.6.5  Summary
10          Atmospheric Pb pollution has resulted in the accumulation of Pb in terrestrial ecosystems
11    throughout the world.  In the United States, pollutant Pb represents a significant fraction of the
12    total Pb burden in soils, even in sites remote from smelters and other industrial plants.  However,
13    few significant effects of Pb pollution have been observed at sites that are not near point sources
14    of Pb. Evidence from precipitation collection and sediment analyses indicates that atmospheric
15    deposition of Pb has declined dramatically (>95%) at sites unaffected by point  sources of Pb, and
16    there is little evidence that Pb accumulated in soils at these sites represents a threat to
17    groundwaters or surface water supplies.
18          The highest environmental risk for Pb in terrestrial ecosystems exists at  sites within about
19    50 km of smelters and other Pb-emitting industrial sites. Assessing the risks specifically
20    associated with Pb is difficult, because these sites also experience elevated concentrations of
21    other metals and because of effects related to SC>2 emissions.  The concentrations of Pb in soils,
22    vegetation, and fauna at these sites can be two to three orders of magnitude higher than in
23    reference areas (see Sections 8.1.3. and 8.1.4.). In the most extreme cases, near smelter sites, the
24    death  of vegetation causes a near-complete collapse of the detrital food web, creating a terrestrial
25    ecosystem in which energy and nutrient flows are minimal. More commonly, stress in soil
26    microorganisms and detritivores can cause reductions in the rate of decomposition of detrital
27    organic matter. Although there is little evidence of significant bioaccumulation of Pb in natural
28    terrestrial ecosystems, reductions in microbial and detritivorous populations can affect the
29    success of their predators. Thus, at present, industrial point sources represent the greatest Pb-
30    related threat to the maintenance of sustainable, healthy, diverse, and high-functioning terrestrial
31    ecosystems in the United States.

      December 2005                        8-115          DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2   AQUATIC ECOSYSTEMS
 2    8.2.1   Introduction
 3          The overall intent of Section 8.2 is to provide sufficient information to support
 4    development of an air quality criterion for lead that is protective of aquatic ecosystems.
 5    To achieve this objective, the logical starting points are to (1) gain a general understanding of the
 6    current distribution and concentrations of lead in the aquatic environment and (2) identify the
 7    threshold levels for lead effects on aquatic populations, communities, and ecosystems.  For this
 8    latter objective, compatible with the EPA's continuing goal of developing environmental criteria
 9    using a holistic approach, development of air quality criteria should be integrated with EPA's
10    current ambient water quality criteria (AWQC)1 for lead and sediment quality benchmarks.
11    Ambient water quality criteria for lead and other chemicals represent surface water
12    concentrations that are intended to be protective of aquatic communities, including recreationally
13    and commercially important species. The EPA derives AWQC to provide guidance to States and
14    Tribes that are authorized to establish water quality standards under the Clean Water Act
15    (CWA).  Similarly, EPA has recommended sediment quality benchmarks for lead and other
16    divalent metals, although not truly criteria, that represent concentrations in sediment that are
17    derived to be protective of benthic (sediment) organisms.  As summarized further below and in
18    subsequent sections, the EPA has increasingly focused on developing AWQC and sediment
19    quality benchmarks for lead and other metals that account for the bioavailability of the metal to
20    aquatic life. These criteria  and benchmark concentrations in water and sediment represent
21    appropriate starting points to ensure that air quality criteria for lead are adequately protective of
22    aquatic life.
23          Since publication of the 1986 air quality criteria document for lead, knowledge has
24    expanded on the fate and effects of lead in aquatic ecosystems and on the distribution and
25    concentrations of lead in surface waters throughout the United States.  In addition, chemical,
26    physical, and biological properties of lead are discussed.  The following provides a general
27    overview of the key information from Section 8.2.
28
            :Lead AWQC are currently in the process of being updated.

      December 2005                        8-116           DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.1.1  Methodologies in Aquatic Ecosystem Research
 2    Ambient Water Quality Criteria and Bioavailability
 3          The primary form of lead in freshwater and marine environments is divalent lead (Pb2+).
 4    In surface waters, the bioavailability of lead to aquatic biota is driven by a variety of factors,
 5    including calcium, dissolved organic carbon (DOC), pH, alkalinity, and total suspended solids
 6    (TSS). Accounting for the influence of calcium and magnesium ions on lead bioavailability, the
 7    current AWQC for lead are normalized to the hardness of the receiving water (Table 8-2.1.1).
 8    More recently, the biotic ligand model (BLM), which considers the binding of free metal ion to
 9    the site of toxic action and competition between metal species and other ions, has been
10    developed to predict the toxicity of several metals under a variety of water quality conditions.
11    The BLM has been incorporated into the draft AWQC for copper and is currently being
12    researched for lead.
          Table 8-2.1.1.  Summary of Lead Ambient Water Quality Criteria for Freshwater
                              Organisms at Different Hardness Levels
        Hardness                         Acute Criterion                 Chronic Criterion
        (mg/L as CaCO3)                      (jig/L)
        50                                   34                            1.3
        100                                  82                            3.2
        200                                 200                            7.7
13    Sediment Quality Benchmarks and Bioavailability
14          As in surface waters, there are a number of factors in sediment that can influence lead
15    bioavailability to benthic (sediment) organisms. Although sediment quality criteria have not
16    been formally adopted, the EPA has published an equilibrium partitioning procedure for
17    developing sediment criteria for metals (U.S. Environmental Protection Agency 2005c).
18    Equilibrium partitioning (EqP) theory predicts that metals partition in sediment between acid
19    volatile sulfide, pore water, benthic organisms, and other sediment phases, such as organic
20    carbon. When the sum of the molar concentrations of simultaneously extracted metal (ZSEM)

      December 2005                         8-117          DRAFT-DO NOT QUOTE OR CITE

-------
 1

 2

 3

 4

 5

 6

 7
10
11
minus the molar concentration of AVS is less than zero, it can accurately be predicted that

sediments are not toxic because of these metals.  Further, if ZSEM-AVS is normalized to the

fraction of organic carbon (i.e., (ZSEM-AVS)//bc), mortality can be more reliably predicted by

accounting for both the site-specific organic carbon and AVS concentrations (Table 8-2.1.2).

An alternative approach for developing sediment quality guidelines is to use empirical

correlations between metal concentrations in bulk sediment to associated biological effects,

based on sediment toxicity tests (Table 8-2.1.2).  These guidelines are based on total metal

concentrations in sediment and do not account for the bioavailability of metals between

sediments.
                     Table 8-2.1.2. Summary of Sediment Quality Benchmarks and
                                           Guidelines for Lead

       Benchmark/
       Guideline Type    Source                       Effect Level        Value
       Equilibrium
       partitioning
                  U.S. Environmental
                  Protection Agency
                  (2005c)
Low risk of adverse
 biological effects
                                                   May have adverse
                                                   biological effects

                                                   Adverse biological
                                                    effects expected
(SEM-AVS)//OC < 130 umol/goc
                                                                  130 umol/goc < (SEM-AVS)//oC
                                                                  < 3,000 umol/goc
                                                                  (SEM-AVS)//OC> 3,000 umol/g
                                                                                            ,oc
Bulk sediment MacDonald et al. (2000)

Ingersolletal. (1996)

Long etal. (1995)

TEC
PEC
ERL
ERM
ERL
ERM
35.8 ug/gdry wt.
128 ug/g dry wt.
55 ug/g dry wt.
99 ug/g dry wt.
46.7 ug/g dry wt.
218 ug/gdry wt.
       AVS = Acid volatile sulfide; ERL = Effects range - low (sediment concentration below which adverse effects are
       rarely observed or predicted among sensitive species, Long et al. [1995]); ERM = Effects range - median
       (sediment concentration above which effects are frequently or always observed or predicted among most species,
       Long et al. [1995]); oc = Organic carbon (foc = fraction organic carbon, g0c = grams organic carbon); PEC =
       Probably effect concentration (sediment concentration above which harmful effects are likely to be observed,
       MacDonald et al. [2000]); SEM = Simultaneously extracted metal; TEC = Threshold effect concentration
       (sediment concentration below which harmful effects are unlikely to be observed, MacDonald et al. [2000])
      December 2005
                                         8-118
              DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.1.2  Distribution of Lead in Aquatic Ecosystems
 2    Speciation of Lead in Aquatic Ecosystems
 3         The speciation of lead in the aquatic environment is controlled by many factors, such as,
 4    pH, salinity, sorption, and biotransformation processes. Lead is typically present in acidic
 5    aquatic environments as PbSC>4, PbCU, ionic lead, cationic forms of lead hydroxide, and ordinary
 6    hydroxide Pb(OH)2. In alkaline, waters common species of lead include anionic forms of lead
 7    carbonate Pb(CO3) and hydroxide Pb(OH)2. In freshwaters, lead typically forms strong
 8    complexes with inorganic OH" and CO32 and weak complexes with Cl (Bodek et al., 1988;
 9    Long & Angino, 1977). The primary form of lead in freshwaters at low pH (<6.5) is
10    predominantly Pb2+ and less abundant inorganic forms include Pb(HCO)3, Pb(SC>4)22 , PbCl,
11    PbCO3, and Pb2(OH)2CO3. At higher pH (>7.5) lead forms hydroxide complexes (PbOH+,
12    Pb(OH)2, Pb(OH)3 , Pb(OH)42 ). Lead speciation in  seawater is a function of chloride
13    concentration and the primary species are PbCl3 > PbCO3 > PbCl2 > PbCl+ > and Pb (OH)+
14    (Fernando, 1995).
15         Lead sorption to suspended or bed sediments or suspended organic matter typically
16    increases with increasing pH, increasing amounts of iron or manganese; and with the polarity of
17    particulate matter (e.g., clays). Adsorption decreases with water hardness (Syracuse Research
18    Corporation [SRC], 1999). At higher pH, lead precipitates as Pb(OH)+ and PbHCO3+ into bed
19    sediments (Weber, 1993). Conversely, at low pH, lead is negatively sorbed (repelled from the
20    adsorbent surface) (U.S. Environmental Protection Agency, 1979; Gao et al., 2003). In addition,
21    lead may be remobilized from sediment due to a decrease in metal concentration in the solution
22    phase, complexation with chelating agents (e.g., EDTA), and changing redox conditions (Gao
23    et al., 2003). Changes in water chemistry (e.g., reduced pH or ionic composition) can cause
24    sediment Pb to become re-mobilized and potentially bioavailable to aquatic organisms (Weber,
25    1993). Methylation may result in lead's remobilization and reintroduction into the aqueous
26    environment compartment and its subsequent release into the atmosphere (SRC, 1999).
27    However, methylation is not  a significant environmental pathway controlling lead fate in the
28    aquatic environment.
29
     December 2005                        8-119           DRAFT-DO NOT QUOTE OR CITE

-------
 1   Lead Concentrations in United States Surface Waters
 2          Nationwide lead data in surface waters, from 1991 onward, were compiled using the
 3   United States Geological Survey's (USGS) National Water-Quality Assessment (NAWQA)
 4   database.  Data were compiled from locations categorized as "ambient" or "natural."  Ambient
 5   refers to data collected from all sampling locations, while natural referred to data collected from
 6   sampling locations categorized as forest, rangeland, or reference. Summary statistics for surface
 7   water, sediment (bulk, <63  jim), and fish tissue (whole body and liver) are summarized in
 8   Table 8-2.1.3. Overall atmospheric sources of lead are generally decreasing as regulations have
 9   removed lead from gasoline and other products (Eisenreich et al., 1986); however, elevated lead
10 concentrations remain near sites with ongoing sources, such as near mining wastes or wastewater
1 1 effluents.
12
13
Table 8-2.1.3.


















Summary of Lead Concentrations in United States Surface
Sediment, and Fish Tissue
Surface Water -
Dissolved
Statistic Ambien
t
n 3,445
%ND 86
Min 0.04
Mean 0.66
95th %ile 1.10
Max 29.78
Natural

430
88
0.04
0.52
0.50
8.40
Sediment
Bulk, <63
Ambien
t
1,466
0.48
0.50
120
200
12,000
uM
wt.)
Natural

258
1.2
0.50
109
162
12,000



Water,
Fish Tissue (jig/g dry wt.)
Whole
Organism
Ambient Natural

332
39
0.08
1.03
1.06
22.6

93
51
0.08
0.95
1.26
22.6
Liver
Ambient

559
71
0.01
0.36
3.24
12.7

Natural

83
89
0.01
0.28
2.50
3.37
      %ND = Percentage not detected
14          In addition to directly measuring lead concentrations in various aquatic compartments, it
15   is useful to study the vertical distribution of lead.  Sediment profiling and core dating is a method
16   used to determine the extent of accumulation of atmospheric lead and provides information on
17   potential anthropogenic sources.  Sediment concentration profiles are typically coupled with lead
18   isotopic analysis. The isotope fingerprinting method utilizes measurements of the abundance of
     December 2005
8-120
DRAFT-DO NOT QUOTE OR CITE

-------
 1    common lead isotopes (204Pb, 206Pb, 207Pb, 208Pb) to distinguish between natural lead over
 2    geologic time and potential anthropogenic sources.  Studies of sediment profiles have suggested
 3    that observed increases in lead concentrations in the upper sediment layer are concomitant with
 4    increases in anthropogenic inputs (Bloom and Crecelius, 1987; Case et al., 1989; Ritson et al.,
 5    1999; Chillrud  et al., 2003).  Isotopic ratios have been used to link increases in sediment
 6    concentrations  with specific anthropogenic sources and to estimate historic records of lead fluxes
 7    to surface waters and sediments (Flegal et al., 1987, 1989; Blais, 1996; Bindler et al., 1999).
 8
 9    8.2.1.3  Species Response/Mode of Action
10    Lead Uptake
11          Lead can bioaccumulate in the tissues of aquatic organisms through ingestion of food and
12    water, and adsorption from water, and can subsequently lead to adverse effects if exposed to
13    sufficiently high concentrations (Vink, 2002). The accumulation of lead is influenced by pH and
14    decreasing pH  favors bioavailability and bioaccumulation.  In general, aquatic organisms have
15    three bioaccumulation strategies for lead: (1) significant accumulation with a low rate of loss;
16    (2) excretion of lead roughly in balance with the availability of the metal in the environment; and
17    (3) weak net accumulation due to a very low metal uptake rate and no significant excretion
18    (Rainbow 1996). Bioconcentration factors (BCFs) have been reported in the scientific literature
19    for various organisms and range from 840 - 20,000 (aquatic plants), 499 - 3,670 (aquatic
20    invertebrates),  and 42 - 45 (fish). Organisms that bioaccumulate lead with little excretion must
21    partition the metal such that it has limited bioavailability, otherwise toxicity will occur if a
22    sufficiently high concentration is reached. As previously mentioned, the biotic ligand model
23    (BLM) has been developed to explore the speciation of metals and to understand how metals
24    accumulate and cause effects in aquatic organisms.
25
26    Resistance Mechanisms
27          Aquatic organisms have various methods to resist the toxic effects of metals such as lead.
28    Resistance processes include detoxification and avoidance responses.  Mechanisms of resistance
29    and detoxification vary among aquatic biota.  These processes can include translocation,
30    excretion, chelation, adsorption,  and vacuolar storage and deposition.  For example, protists and
31    plants produce  intracellular polypeptides that form complexes with lead (Zenk, 1996; Morelli

      December 2005                        8-121           DRAFT-DO NOT QUOTE OR CITE

-------
 1    and Scarano, 2001).  Some macrophytes and wetland plants have developed translocation
 2    strategies for tolerance and detoxification (Knowlton et al., 1983; Deng et al., 2004).  Various
 3    aquatic invertebrates may sequester lead in the exoskeleton (Boisson et al., 2002; Knowlton
 4    et al., 1983) or have developed specialized excretion processes (Vogt and Quinitio, 1994).
 5    Fish scales and mucous may chelate lead in the water column and potentially reduce lead uptake
 6    (Coello and Khan, 1996).
 7          Avoidance responses  are actions performed to evade a perceived threat. Some aquatic
 8    organisms have been shown  to be quite adept at avoiding lead in aquatic systems, while others
 9    seem incapable of detecting its presence. Snails have been shown to be sensitive to lead, and
10    avoid it at high concentrations (Lefcort et al., 2004).  Conversely, anuran (frog and toad) species
11    lack an avoidance response up to 1000 jig Pb/L (Steele et al., 1991). Fish avoidance of chemical
12    toxicants has been well established, and is a dominant sublethal response in polluted waters
13    (Svecevicius, 2001).  However, studies examining avoidance behaviour of lead in fish are
14    lacking. In addition to the presence of toxic metals, light and pH, can also alter preference-
15    avoidance responses.
16
17    Physiological Effects of Lead
18          Physiological effects of lead on aquatic biota can occur at the biochemical, cellular and
19    tissue levels of organization. Lead has been shown to affect brain receptors in fish (Rademacher
20    et al. 2005) and serum  enzyme activity (e.g., EROD and ALAD) in fish and amphibians (Kutlu
21    and Susuz, 2004; Blasco and Puppo, 1999; Gill et al., 1991; Vogiatzis and Loumbourdis, 1999).
22    Studies examining the  effects of lead on fish blood chemistry have indicated alterations from
23    acute and chronic exposures  ranging from 100 to 10,000 |ig/L (Gill et al., 1991; Allen, 1993;
24    Gopal et al., 1997). Lead exposure has also been shown to negatively affect the growth of
25    aquatic invertebrates (Arai et al., 2002).
26
27    Factors that Modify Organism Response to Lead
28          There are several factors that may influence organism response to lead exposure.  These
29    may include the size or age of an organism, genetics, environmental factors (e.g., pH, salinity),
30    nutrition, and the presence of other contaminants. Lead  accumulation in living organisms is
31    controlled, in part, by metabolic rates (Farkas et al., 2003) and by the physiological conditions of

      December 2005                        8-122           DRAFT-DO NOT QUOTE OR CITE

-------
 1    an organism. Relationships between age, size and lead body burden in aquatic invertebrates and
 2    fish are variable and depend on many environmental variables (e.g., exposure) (Farkas et al.,
 3    2003). For example, examination of lead exposure (up to 100 |ig/L) in aquatic invertebrates
 4    showed little relationship between body size and lead accumulation (MacLean et al., 1996; Canli
 5    and Furness, 1993) while lead accumulation and fish size was found to be positively correlated
 6    (Douben, 1989; Kock et al.,  1996).
 7          The genetics of an organism and/or population may alter the response to lead exposure
 8    through one of two processes: (1) a contaminant may influence selection, by selecting for certain
 9    phenotypes that enable populations to better cope with the chemical, or (2) a contaminant can be
10    genotoxic, meaning it can produce alterations in nucleic acids at sublethal exposure
11    concentrations, resulting in changes in hereditary characteristics or DNA inactivation (Shugart,
12    1995). Genetic selection has been observed in aquatic organisms due to lead tolerance. Because
13    tolerant individuals have a selective advantage over vulnerable individuals in polluted
14    environments, the frequency of tolerance genes will increase in exposed populations over time
15    (Beaty et al., 1998).  Several studies have shown that heavy metals can alter population gene
16    pools resulting in decreased genetic diversity (Duan et al., 2000; Kim et al., 2003). Laboratory
17    studies have shown that exposure to lead at 10 mg Pb2+/mL of blood leads to chromosomal
18    aberrations in some aquatic organisms (Cestari et al., 2004).  Low level (50 ug/L) lead exposure
19    in water over four weeks resulted in DNA strand breakage in the freshwater mussel Anodonta
20    grandis (Black et al., 1996).  More recently, Cestari et al. (2004) observed similar results
21    (increase in the frequency of chromosomal aberrations and DNA damage in kidney cell cultures)
22    in fish (Hoplias malabaricus) that were fed lead contaminated food over 18, 41 and 64 days.
23          Environmental factors can alter the availability, uptake and toxicity of lead to aquatic
24    organisms.  Van Hattum et al. (1996) studied the influence of abiotic variables, including
25    dissolved organic carbon (DOC) on lead concentrations in freshwater isopods and found that as
26    DOC concentrations increased, BCFs decreased in P. meridianus and A aquaticus, indicating
27    that DOC acts to inhibit the availability of lead to these isopods. Schwartz et al. (2004) collected
28    natural organic matter (NOM) from several aquatic sites across Canada and investigated the
29    effects of NOM on lead toxicity in rainbow trout (Oncorhynchus mykiss). The results showed
30    that NOM in test water almost always increased LT50 (time to reach 50% mortality), and
31    optically dark NOM tended to decrease lead toxicity more than did optically light NOM in

      December 2005                        8-123           DRAFT-DO NOT QUOTE OR CITE

-------
 1    rainbow trout.  Studies generally agree that as pH increases the toxicity of Pb decreases
 2    (MacDonald et al., 2002; Home and Dunson, 1995a,b,c). As pH decreases, lead becomes more
 3    soluble and more readily bioavailable to aquatic organisms (Weber, 1993). Acute and chronic
 4    toxicity of lead increases with decreasing water hardness, as lead becomes more soluble and
 5    bioavailable to aquatic organisms (Home and Dunson, 1995c; Borgmann et al., 2005).  There is
 6    some evidence that water hardness and pH work together to increase or decrease the toxicity of
 7    lead. High Ca2+ concentrations have been shown to protect against the toxic effects of lead
 8    (Sayer et al., 1989; Rogers and Wood, 2004; MacDonald et al., 2002; Hassler et al., 2004).
 9    Ca2+ affects the permeability and integrity of cell membranes and intracellular contents (Sayer
10    et al., 1989). As Ca2+ concentrations decrease, the passive flux of ions (e.g., lead) and water
11    increases.  Finally, increasing salinity was found to decrease lead toxicity (Verslycke et al.,
12    2003).  The reduction in toxicity was attributed to increased complexation of Pb2+ with Cl  ions.
13         Nutrients (e.g., nitrate, carbonate) have been shown to affect lead toxicity in some aquatic
14    organisms.  Jampani (1988) looked at the impact of various nutrients (i.e., sodium acetate,  citric
15    acid, sodium carbonate, nitrogen, and phosphates) on reducing growth inhibition in blue-green
16    algae (Synechococcus aeruginosus) exposed to 200 mg/L of lead. Results indicated that
17    additional nitrogen, phosphates, and some carbon sources, including sodium acetate, citric acid
18    and sodium carbonate, all protected the algae from lead toxicity. One hypothesis was that
19    nutrients were able to reverse toxic effects.  The second hypothesis was that nutrients directly
20    interacted with lead, in some way sequestering the metal so as to inhibit its metabolic interaction
21    with the organism (Rao and Reddy, 1985; Jampani, 1988).  Rai and Raizada (1989) investigated
22    the effects of lead on nitrate and ammonium uptake and results indicated that lead exposure can
23    affect the uptake of some nutrients in N. muscorum. Thus, nutrients seem to be capable of
24    reducing toxicity, though the mechanisms have not been well established.
25
26    Interactions with Other Pollutants
27         Predicting the response of organisms to mixtures of chemicals is a daunting task
28    (Norwood et al., 2003). Antagonism, synergism, and additivity are the primary responses that
29    occur following exposure to multiple contaminants. When two or more metals compete for the
30    same binding sites or interfere with transport through cell walls or membranes, the interaction is
31    termed less than strictly additive or antagonistic.  Antagonistic interactions can reduce metal

      December 2005                       8-124           DRAFT-DO NOT QUOTE OR CITE

-------
 1    bioavailability when metals are present in combination, and may lead to reduced potential for
 2    toxicity (Hassler et al., 2004).  There are a number of elements (Ca2+, Cd2+, Mg2+, Na+ and Cl")
 3    that act in an antagonistic fashion with Pb (Niyogi and Wood, 2004; Rogers and Wood, 2003,
 4    2004; Ahern and Morris, 1998; Li et al., 2004). For example, Pb is a well-known antagonist to
 5    Ca2+ (Hassler et al., 2004; Niyogi and Wood, 2004).  Calcium is an essential element, required
 6    for a number of physiological processes in most organisms.
 7          Synergism occurs when the interaction of two or more metals causes an effect that is
 8    greater than the effect observed from the individual metals themselves (Hagopian-Schlekat et al.,
 9    2001).  Synergism is likely the result of increased bioavailability of one or more of the metal ions
10    due to the presence of other metals (Hassler et al., 2004). Hassler  et al. (2004) reported that in
11    the presence of copper (Cu2+) there was a significantly higher rate of internalization of Pb in the
12    green algae Chlorella kesserii. It was suggested that Cu2+ may have affected organism
13    physiology through the disruption of cell membrane integrity. This would allow increased cation
14    (i.e., Pb2+) permeability and therefore substantially increased internalization of Pb. Synergistic
15    interactions have also been observed with lead and other metals (Cd, Cu, Ni, and Zn) (Hagopian-
16    Schlekat et al.,  2001).
17          The combined effects of two or more metals may result in additivity when the observed
18    effects are greater than that observed with individual metals but equivalent to a  summation of the
19    effects from multiple metals.  Norwood et al. (2003) reported that  in a review and re-
20    interpretation of published data on the interactions of metals in binary mixtures (n = 15  studies),
21    antagonistic (6) and additive interactions (6) were the most common for lead. The two most
22    commonly reported lead-element interactions are between lead and calcium and lead and zinc.
23    Both calcium and zinc  are essential elements in organisms and the interaction of Pb with these
24    ions can lead to adverse effects both by increased Pb uptake and by a decrease in Ca and Zn
25    required for normal metabolic functions.
26
27    8.2.1.4 Exposure/Response of Aquatic Species
28    Effects of Lead on Primary Producers
29          In the previous air quality criteria document (U.S. Environmental Protection Agency,
30    1986a), several authors reported that some algal species  (e.g., Scenedesmus sp.) were found to
31    exhibit physiological changes when exposed to high lead or organolead concentrations in situ.

      December 2005                        8-125          DRAFT-DO NOT QUOTE OR CITE

-------
 1    The observed changes included increasing numbers of vacuoles, deformations in cell organelles,
 2    and increased autolytic activity.  Increased vacuolization was assumed to be a tolerance
 3    mechanism by which lead was immobilized within cell vacuoles.
 4          Several studies have been conducted since the 1986 air quality criteria document on the
 5    toxicity of lead to primary producers (Rai and Raizada, 1989; Jampani, 1988; Adam and Abdel-
 6    Basset, 1990; Gaur et al., 1994; Gupta and Chandra, 1994). Effects to algal growth (Chlorella
 1    vulgaris, Closterium acerosum, Pediastrum simplex, Scenedesmus quadricaudd), ranging from
 8    minimal to complete inhibition, have been reported at lead concentrations between 100 and
 9    200,000 ng/L (Bilgrami and Kumar, 1997; Jampani, 1988). The toxicity of lead to aquatic plant
10    growth has been studied using Spirodelapolyrhiza, Azollapinnata, and Lemna gibba (Gaur
11    et al., 1994; Gupta and Chandra, 1994; Miranda and Ilangovan, 1996). Test durations ranged
12    from 4 to 25 days and test concentrations ranged between 49.7 and 500,000 |ig/L (Gaur et al.,
13    1994; Miranda and Ilangovan, 1996).  Research on aquatic plants has been focussed on the
14    effects of lead on aquatic plant growth, chlorophyll and protein content.
15          Algae and aquatic plants have a wide range in sensitivity to the effects of lead in water.
16    Both groups of primary producers experience EC50 values for growth inhibition between
17    approximately 1,000 and >100,000  |ig/L (Bilgrami and Kumar, 1997; Jampani,  1988; Gaur et al.,
18    1994). The most sensitive primary producers reported in the literature for effects to growth were
19    Closterium acersoum and Azollapinnata (Bilgrami and Kumar, 1997; Gaur et al., 1994).  The
20    least sensitive primary producers reported in the literature for effects to growth were
21    Synechococcus aeruginosus and Lemna gibba (Jampani, 1988; Miranda and Ilangovan, 1996).
22    Exposure to lead in combination with other metals is generally less toxic to growth than
23    exposure to lead alone.  Studies have shown that lead adversely affects the metabolic processes
24    of nitrate uptake, nitrogen fixation, ammonium uptake, and carbon fixation (Rai and Raizada,
25    1989). Lead in combination with nickel or chromium produced synergistic effects for nitrate
26    uptake, nitrogenase activities, ammonium uptake, and carbon fixation (Rai and Raizada,  1989).
27
28    Effects of Lead on Consumers
29          The 1986 AQCD (U.S. Environmental Protection Agency, 1986a) reported that
30    hematological and neurological responses are the most commonly reported effects to aquatic
31    vertebrates. These effects include red blood cell destruction and inhibition of the enzyme

      December 2005                        8-126           DRAFT-DO NOT QUOTE OR CITE

-------
 1    ALAD, required for hemoglobin synthesis.  The lowest reported exposure concentration causing
 2    either hematological or neurological effects was 8 |ig/L (U.S. Environmental Protection Agency,
 3    1986a).
 4          Recent literature on the toxicity of lead to fish and aquatic invertebrates has been
 5    summarized by Eisler (2000).  Exposure of invertebrates to Pb can lead to adverse effects on
 6    reproduction, growth, survival, and metabolism (Eisler, 2000). Water-borne lead is highly toxic
 7    to aquatic organisms, with toxicity varying depending on the species and life stage tested,
 8    duration of exposure, the form of lead tested, and water quality characteristics. Among the
 9    species tested, aquatic invertebrates, such as amphipods and water fleas, were the most sensitive
10    to the effects of lead with adverse effects being reported at concentrations ranging from 0.45 to
11    8000 |ig/L. Freshwater fish demonstrated adverse effects at concentrations ranging from 10 to
12    >5400 |ig/L, generally depending upon water quality parameters (e.g., pH, hardness, salinity).
13    Amphibians tend to be relatively tolerant of lead, however, may exhibit decreased enzyme
14    activity (e.g., ALAD reduction) and changes in behavior (e.g., hypoxia response behavior). Lead
15    tends to be more toxic in longer-term exposures, with chronic toxicity thresholds for
16    reproduction in water fleas ranging as low as 30 jig/L (e.g., Kraak et al., 1994).
17
18    8.2.1.5  Effects of Lead on Natural Aquatic Ecosystems
19          The effects of lead on natural aquatic ecosystems were examined following the conceptual
20    framework developed by the EPA Science Advisory Board (Young and Sanzone, 2002).  The
21    essential attributes used to describe ecological condition include landscape condition, biotic
22    condition,  chemical and physical characteristics, ecological processes, hydrology and
23    geomorphology and natural disturbance regimes. The majority of the published literature
24    pertaining  to lead and natural aquatic ecosystems focuses on the biotic condition and identifies
25    effects on energy flow or nutrient cycling, community structure, and predator-prey interactions.
26          Recent studies have attributed the presence of lead to reduced primary productivity,
27    respiration, and alterations of community structure.  Specifically, lead (6-80 mg/L) was found to
28    reduce primary productivity and increase respiration in an algal community (Jayaraj et al., 1992).
29    Laboratory microcosm studies have indicated reduced species abundance and diversity in
30    protozoan  communities exposed to 0.02 - 1 mg/L (Fernandez-Leborans and Novillo, 1992, 1994;
31    Fernandez-Leborans and Antonio-Garcia, 1988). Numerous field studies have associated the

      December 2005                        8-127           DRAFT-DO NOT QUOTE OR CITE

-------
 1    presence or bioaccumulation of lead with reductions of species abundance, richness, or diversity,
 2    particularly in benthic macroinvertebrate communities (Deacon et al., 2001; Mize and Deacon,
 3    2002; Mucha et al., 2003; Poulton et al., 1995; Rhea et al., 2004; Maret et al., 2003). However,
 4    in natural aquatic ecosystems, lead is often found coexisting with other metals and other
 5    stressors. Thus, understanding the effects of lead in natural systems is challenging given that
 6    observed effects may be due to cumulative toxicity from multiple stressors.
 7          Exposure to lead in laboratory studies and simulated ecosystems may  alter species
 8    competitive behaviors, predator-prey interactions, and contaminant avoidance behaviors.
 9    Alteration of these interactions may have negative effects on species abundance and community
10    structure. For example, reduced avoidance behaviors have been observed at  lead concentrations
11    ranging from 0.3 - 1.0 mg/L (Weber, 1996; Steele et al., 1991; Weis and Weis, 1998).  The
12    feeding behaviors  of competitive species in some aquatic organisms are also  influenced by the
13    presence of lead (Lefcort et al., 2000).
14          The effects of lead have primarily been studied in instances of point source pollution
15    rather than area-wide atmospheric deposition. Thus, the effects of atmospheric lead on aquatic
16    ecological condition remains to be defined.  There is a paucity of data in the general literature
17    that explores the effects of lead in conjunction with all or several of the various components of
18    ecological condition as defined by the EPA (Young and Sanzone, 2002). However, numerous
19    studies are available associating the presence of lead with effects on biotic conditions.
20
21    8.2.2   Methodologies in Aquatic Ecosystem Research
22    8.2.2.1  Introduction
23          As discussed  in previous sections, aerial deposition is one source of Pb deposition to
24    aquatic systems. Consequently, to develop air quality criteria for Pb, consideration must be
25    given to not only the environmental fate of Pb, but also to the environmental  effects of Pb in the
26    aquatic environment through consideration of laboratory toxicity studies and field evaluations.
27    Perhaps the most straightforward approach for evaluating the effects of Pb is to consider extant
28    criteria for Pb in aquatic ecosystems, i.e., water and sediment quality criteria. A key issue in
29    developing Pb water and sediment criteria that are broadly applicable to a range of water bodies
30    is properly accounting for Pb bioavailability and the range in species sensitivities. This section
31    summarizes how these criteria are derived, the types of toxicity studies considered,  and key

      December 2005                        8-128          DRAFT-DO NOT QUOTE OR CITE

-------
 1    factors that influence the bioavailability of Pb in surface water and sediment to aquatic life.
 2    Because Pb in the aquatic environment is often associated with other metals (e.g., cadmium,
 3    copper, zinc), the importance of considering the toxicity of metal mixtures is also discussed.
 4    Finally, some issues related to background Pb concentrations are briefly addressed. It is beyond
 5    the scope of this section to review all methodologies in aquatic system research, but good
 6    reviews can be found in summary books,  such as Rand et al. (1995).
 7
 8    8.2.2.2  Analytical Methods
 9          Common analytical methods for measuring Pb in the aquatic environment are summarized
10    in Table 8-2.2.1. For relevance to the ambient water quality criteria (AWQC) and sediment
11    quality criteria for Pb discussed below, minimum detection limits should be in the low parts per
12    billion (ppb) range for surface water and the low parts per million (ppm) range for sediment.
13
14
              Table 8-2.2.1.  Common Analytical Methods for Measuring Lead in Water,
                                         Sediment, and Tissue
       Analysis Type                                                 Analytical Method
       Direct-Aspiration (Flame) Atomic Absorption                  EPA SW-846 Method 7420a,
       Spectroscopy (AAS)                                            EPA Method 239. lb,
                                                                   Standard Method 31 llc
       Graphite Furnace Atomic Absorption Spectroscopy             EPA SW-846 Method 7421a,
       (GFAAS)                                                     EPA Method 239.2b,
                                                                   Standard Method 3113C
       Inductively Coupled Plasma                                EPA SW-846 Method 601 OBa,
       (ICP)                                                         EPA Method 200.7b,
                                                                   Standard Method 3120C
       Inductively Coupled Plasma-Mass Spectrometry                EPA SW-846 Method 6020a,
       (ICP-MS)                                                     EPA Method 200.8b
       a U.S. Environmental Protection Agency (1986c) Test Methods for Evaluating Solid Waste, Physical/Chemical
        Methods (SW-846). Third Edition, September  1986; Final Updates I (7/92), IIA (8/93), II (9/94), IIB (1/95),
        III (12/96), IIIA (4/98), IIIB (11/04).
       b U.S. Environmental Protection Agency (1991) Methods for the Determination of Metals in Environmental
        Samples. EPA/600/4-91-010. June 1991 (Supplement I, EPA/600/R-94-111, May 1994).
       0 American Public Health Association (1995) Standard Methods for the Examination of Water and Wastewater,
        19th Edition. American Public Health Association, American Water Works Association, Water Pollution
        Control Federation.
      December 2005                         8-129           DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.2.3   Ambient Water Quality Criteria:  Development
 2          The EPA's procedures for deriving AWQC are described in Stephan et al. (1985) and are
 3    summarized here. With few exceptions, AWQC are derived based on data from aquatic toxicity
 4    studies conducted in the laboratory. In general, both acute (short term) and chronic (long term)
 5    AWQC are developed.  Depending on the species, the toxicity studies considered for developing
 6    acute criteria range in length from  48 to 96 hours. Acceptable endpoints for acute AWQC
 7    development are mortality and/or immobilization, expressed as the median lethal concentration
 8    (LCso) or median effect concentration (ECso). For each species, the geometric mean of the
 9    acceptable LCso/ECso data is calculated to determine the species mean acute value (SMAV).
10    For each genera, the geometric mean of the relevant SMAVs is then calculated to determine the
11    genus mean acute value (GMAV).  The GMAVs are then ranked from high to low, and the final
12    acute value (FAV; the 5th percentile of the GMAVs, based on the four GMAVs surrounding the
13    5th percentile) is determined.  Because the FAV is based on LCso/ECso values (which represent
14    unacceptably high levels of effect), the FAV is divided by two to estimate a low-effect level.
15    This value is then termed the acute criterion, or criterion maximum concentration (CMC). Based
16    on the most recent AWQC document for Pb (U.S. Environmental Protection Agency, 1985),
17    Table 8-2.2.2 shows the freshwater SMAVs and GMAVs for Pb, and the resulting freshwater
18    CMC.  Note that the freshwater AWQC are normalized for the hardness of the site water, as
19    discussed further below in Section 8.2.2.4.
20          To develop chronic AWQC, acceptable chronic toxicity studies should encompass the full
21    life cycle of the test organism, although for fish, early life stage or partial life cycle toxicity
22    studies are considered acceptable.  Acceptable endpoints include reproduction, growth and
23    development, and survival, with the effect levels expressed as the chronic value, which is the
24    geometric mean of the no-observed-effect concentration (NOEC)2 and the lowest-observed-
25    effect concentration (LOEC)3. Although a chronic criterion could be calculated as the 5th
26    percentile of genus mean chronic values (GMCVs), sufficient chronic toxicity data are generally
27    lacking, as is the case for Pb. Consequently, an acute-chronic ratio (ACR) is typically applied to
             2 The NOEC is the highest concentration tested that did not result in statistically significant effects relative
      to the control.
             3 The LOEC is the lowest concentration tested that resulted in statistically significant effects relative to
      the control.
      December 2005                        8-130           DRAFT-DO NOT QUOTE OR CITE

-------
             Table 8-2.2.2. Development of Current Acute Freshwater Criteria for Lead
                          (U.S. Environmental Protection Agency, 1985)1
Rank
10
9
8
7
6
5
4
3
2
1

Species
Midge (Tanytarsus dissimilis)
Goldfish (Carassius auratus)
Guppy (Poecilia reticulates}
Bbluegill (Lepomis macrochirus)
Fathead minnow (Pimephales promelas)
Brook trout (Salvelinus fontinalis)
Rainbow trout (Oncorhynchus mykiss)
Snail (Aplexa hypnomm)
Cladoceran (Daphnia magna)
Amphipod (Gammams pseudolimnaeus)

GMAV
(Mg/L)
235,900
101,100
66,140
52,310
25,440
4,820
2,448
1,040
447.8
142.6
FAV
CMC
SMAV
(Hg/L)
235,900
101,100
66,140
52,310
25,440
4,820
2,448
1,040
447.8
142.6
= 67.54 jig/L
= 33.77 ug/L
       All values are normalized to a hardness of 50 mg/L (see Section 8.2.2.4).
 1   the FAV to derive the chronic criterion.  As the name applies, the ACR is the ratio of the acute
 2   LCso to the chronic value, based on studies with the same species and in the same dilution water.
 3   For Pb, the final ACR is 51.29, which results in a final chronic value (FCV) of 1.317 |ig/L (at a
 4   hardness of 50 mg/L).
 5          Subsequent sections summarize some of the toxicity studies that meet the AWQC
 6   development guidelines, with an emphasis on key studies published since the last Pb AWQC
 7   were derived in 1984.
 8
 9   8.2.2.4 Ambient Water Quality Criteria: Bioavailability Issues
10          In surface waters, the environmental  fate of metal contaminants is mitigated through
11   adsorption, complexation, chelation, and other processes that affect bioavailability. The toxicity
     December 2005
8-131
DRAFT-DO NOT QUOTE OR CITE

-------
 1    of divalent cations tends to be highest in soft waters with low concentrations of dissolved organic
 2    matter and suspended particles. In an acidic environment (pH <4), the ionic form of most metals
 3    generally predominates and is considered to be the more toxic form. As the pH increases,
 4    carbonate, oxide, hydroxide, and sulfide complexes of the metals tend to predominate, and tend
 5    to be less toxic (Florence, 1977; Miller and Mackay, 1980). The portion of dissolved metal
 6    available for uptake or bioaccumulation is influenced by modifying factors that "sequester" the
 7    metal in an environmental matrix, thereby reducing the bioavailability of the metal at the sites of
 8    action.  Metals can become complexed (bound) to a ligand that can make metals either more
 9    toxic (via transport mechanisms) or less toxic (by changing the metal's biological activity).
10    Metals that complex tightly to ligands generally are not readily bioavailable and, thus, are less
11    toxic to aquatic biota than their free-metal ion counterparts (Carlson et al., 1986; McCarthy,
12    1989).  There are many kinds of ligands, organic and inorganic, as well as natural and man-
13    made. Ligands found in natural surface waters and municipal and industrial effluent discharges
14    include glycine, ammonia, oxalate, humic or fulvic acids, hydroxide, carbonate, bicarbonate,
15    chloride, and hydrogen sulfide (Stumm and Morgan, 1970; Martin,  1986; Pagenkopf, 1986).
16         Recognizing the importance of calcium and magnesium ions (hardness) in modifying Pb
17    toxicity, the current freshwater AWQC for Pb are normalized based on the hardness of the site
18    water. The acute freshwater criteria, for example, are 34, 82, and 200 jig/L at hardness levels of
19    50, 100, and 200 mg/L (as CaCOs). Although it has been known for some time that other water
20    quality parameters such as pH, dissolved organic carbon (DOC), and alkalinity affect the
21    bioavailability  of metals to aquatic biota, it was the relatively recent development of the biotic
22    ligand model (BLM) that allowed AWQC to consider all of these factors.  Paquin et al. (2002)
23    provided a thorough review of the factors influencing metal bioavailability and how research
24    over the last few decades has culminated in the development of the BLM.
25         By understanding the binding affinities of various natural ligands in surface waters and
26    how the freshwater fish gill interacts with free cations in the water,  one can predict how metals
27    exert their toxic effects (Schwartz et al., 2004). Early precursors to the BLM were the free-ion
28    activity model  (FIAM) and the gill surface interaction model (GSIM).  The FIAM is a conceptual
29    model that accounts for the binding of free metal ion and other metal complexes to the site of
30    toxic action in  an organism; it also considers competition between metal species and other
31    cations  (Paquin et al., 2002). The GSIM is fundamentally similar to the FIAM in that it accounts

      December 2005                       8-132           DRAFT-DO NOT QUOTE OR CITE

-------
 1    for competition between metal ions and hardness cations at the physiological active gill sites, but
 2    whereas the FIAM is largely conceptual, the GSIM was used in interpreting toxicity test results
 3    for individual metals and metal mixtures (Pagenkopf, 1983). The BLM was adapted from the
 4    GSIM and uses the biotic ligand, rather than the fish gill as the site of toxic action (Di Toro et al.,
 5    2001; Paquin et al., 2002).  This approach, therefore, considers that the external fish gill surface
 6    contains receptor sites for metal binding (Schwartz et al., 2004) and that acute toxicity is
 7    associated with the binding of metals to defined sites (biotic ligands)  on or within the organism
 8    (Paquin et al., 2002). The model is predicated on the theory that mortality (or other toxic effects)
 9    occurs when the concentration of metal bound to biotic ligand exceeds a threshold concentration
10    (Di Toro et al., 2001; Paquin et al., 2002).  Direct uptake via the gills is thought to be the
11    pathway for Pb uptake in freshwater fish (Merlini and Pozzi, 1977; Hodson et al., 1978).  Free
12    metal cations "out compete" other cations and bind to the limited number of active receptor sites
13    on the gill surface, possibly suffocating and/or disrupting ionoregulatory mechanisms in the fish,
14    leading to death (Di  Toro et al., 2001; Paquin et al., 2002).  Because the BLM uses the biotic
15    ligand (not the fish gill) as the site of action, the model can be applied to other aquatic
16    organisms, such as crustaceans, where the site of action is directly exposed to the aqueous
17    environment (Di Toro et al., 2001). Dietary metals have also been  shown to contribute to uptake
18    by aquatic biota and, in  some cases, increased toxicity.  For example, Besser et al. (2005)
19    observed that chronic (42-day) Pb toxicity to the amphipod Hyalella azteca was greater from a
20    combined aqueous and dietary exposure than from a water-only exposure. The feasibility of
21    incorporating dietary metals into BLMs is under investigation.
22          To date, the EPA has incorporated the BLM into draft freshwater criteria for copper, but
23    the BLM is likely to be  also included in the revised Pb criteria.
24
25    8.2.2.5  Sediment Quality Criteria: Development and Unavailability Issues
26          As with metals in surface waters, the environmental  fate  of metal contaminants in
27    sediments is  moderated  through various binding processes that reduce the concentration of free,
28    bioavailable  metal.  Sediments function as a sink for Pb, as with most metals.  Lead compounds
29    such as Pb-carbonates, Pb-sulfates, and Pb-sulfides predominate in sediments (Prosi, 1989).
30    Total Pb has a higher retention time and a higher percentage is retained in sediments compared to
31    copper and zinc (Prosi,  1989). Lead is primarily accumulated in sediments as insoluble Pb

      December 2005                        8-133           DRAFT-DO NOT QUOTE OR CITE

-------
 1    complexes adsorbed to suspended particulate matter. Naturally occurring Pb is bound in
 2    sediments and has a low geochemical mobility (Prosi, 1989).  Organic-sulfide and moderately
 3    reducible fractions are less mobile, whereas cation-exchangeable fractions and easily-reducible
 4    fractions are more mobile and more readily bioavailable to biota (Prosi, 1989). Most Pb
 5    transported in surface waters is in a particulate form, originating from the erosion of sediments in
 6    rivers or produced in the water column (Prosi, 1989).
 7          Sediment quality criteria have yet to be adopted by the EPA, but an equilibrium
 8    partitioning procedure has recently been published (U.S. Environmental Protection Agency,
 9    2005c).  The EPA has selected an equilibrium partitioning approach because it explicitly
10    accounts for the bioavailability of metals. This approach is based on mixtures of cadmium,
11    copper, Pb, nickel, silver, and zinc. Equilibrium partitioning (EqP) theory predicts that metals
12    partition in sediment between acid-volatile sulfide, pore water, benthic organisms, and other
13    sediment phases such as organic carbon. When the sum of the molar concentrations of
14    simultaneously extracted metal (ZSEM) minus the molar concentration of AVS is less than zero,
15    it can accurately be predicted that sediments are not toxic because of these metals.  Note that this
16    approach can be used to predict the lack of toxicity, but not the presence of toxicity. It is
17    important to emphasize that metals must be evaluated as a mixture using this approach.
18    If ndividual metals, or just two or three  metals, are measured in sediment, ZSEM would be
19    misleadingly small and it may inaccurately appear that ZSEM - AVS is less than 1.0.
20          If ZSEM - AVS is normalized to the organic carbon fraction (i.e., (ZSEM - AVS)//bc),
21    mortality can be more reliably predicted by accounting for both the site-specific organic carbon
22    and AVS concentrations.  When evaluating a metal mixture containing cadmium, copper, Pb,
23    nickel, silver, and zinc, the following predictions can be made (U.S. Environmental Protection
24    Agency, 2005c):
25       •  A sediment with (SEM - AVS)//bc < 130 jimol/goc should pose low risk of adverse
26          biological effects  due to these metals.
27       •  A sediment with 130 jimol/goc < (SEM - AVS)//0c< 3000  jimol/goc may have adverse
28          biological effects  due to these metals.
29       •  In a sediment with (SEM - AVS)//bc > 3000 jimol/goc, adverse biological effects may
30          be expected.
      December 2005                        8-134          DRAFT-DO NOT QUOTE OR CITE

-------
 1          A third approach is to measure pore water concentrations of cadmium, copper, Pb, nickel,
 2   and zinc and then divide the concentrations by their respective FCVs. If the sum of these
 3   quotients is <1.0, these metals are not expected to be toxic to benthic organisms.
 4          Many alternative approaches for developing sediment quality guidelines are based on
 5   empirical correlations between metal concentrations in sediment to associated biological effects,
 6   based on sediment toxicity tests (Long et al., 1995; Ingersoll et al., 1996; MacDonald et al.,
 7   2000). However, these guidelines are based on total metal concentrations in sediment and do not
 8   account for the bioavailability of metals between sediments.  Sediment quality guidelines
 9   proposed for Pb from these other sources are shown in Table 8-2.2.3.

                 Table 8-2.2.3.  Recommended Sediment Quality Guidelines for Lead
Source
MacDonald et al. (2000)
Ingersoll etal. (1996)
Long etal. (1995)
Water Type
Freshwater
Freshwater
Saltwater
Guideline Type
TEC
PEC
ERL
ERM
ERL
ERM
Cone, (mg/kg dw)
35.8
128
55
99
46.7
218
      TEC = Threshold effect concentration; PEC = Probable effect concentration; ERL = Effects range - low;
      ERM = Effects range - median
10   8.2.2.6  Metal Mixtures
11         As discussed above, the EPA's current approach for developing sediment criteria for Pb
12   and other metals is to consider the molar sum of the metal concentrations (ZSEM). Although a
13   similar approach has not been applied to AWQC, metal mixtures have been shown to be more
14   toxic than individual metals (Spehar and Fiandt, 1986; Enserink et al., 1991).  Spehar and Fiandt
15   (1986) evaluated the acute and chronic toxicity of a metal mixture (arsenic, cadmium, chromium,
16   copper, mercury, and Pb) to fathead minnows (Pimephalespromelets) and a daphnid
17   (Ceriodaphnia dubid).  In acute tests, the joint toxicity of these metals was observed to be more
18   than additive for fathead minnows and nearly strictly additive for daphnids. In chronic tests, the
19   joint toxicity of the metals was less than additive for fathead minnows and nearly strictly

     December 2005                       8-135          DRAFT-DO NOT QUOTE OR CITE

-------
 1    additive for daphnids.  One approach for considering the additive toxicity of Pb with other metals
 2    is to use the concept of toxic units (TUs). Toxic units for each component of a metal mixture are
 3    derived by dividing metal concentrations by their respective acute or chronic criterion. The TUs
 4    for all the metals in the mixture are then summed.  A ETU > 1.0 suggests the metal mixture is
 5    toxic (note that this is the same approach as discussed above for developing metal sediment
 6    criteria based on pore water concentrations).  According to Norwood et al. (2003), the TU
 7    approach is presently the most appropriate model for predicting effects of metal mixtures based
 8    on the currently available toxicity data. However, it should also be emphasized that the TU
 9    approach is most appropriate at a screening level, because the true toxicity of the  mixture is
10    dependent on the relative amounts of each metal. The TU approach is also recommended with
11    mixtures containing less than six metals.
12          For accessing Pb effects on aquatic ecosystems, it is not truly feasible to account for metal
13    mixtures, because these will obviously vary highly from site to site.  However, the toxicity of
14    metal mixtures in surface water should be considered on a site-specific basis.
15
16    8.2.2.7  Background Lead
17          Because Pb is naturally occurring, it is found in all environmental compartments
18    including surface water, sediment, and aquatic biota. Background Pb concentrations are spatially
19    variable depending on geological features and local characteristics that influence  Pb speciation
20    and mobility. In the European Union risk assessments for metals, an "added risk" approach has
21    been considered that assumes only the amount of metal added above background  is  relevant in a
22    toxicological evaluation. However, this approach ignores the possible contribution  of
23    background metal levels to toxic effects, and background metal levels  are regionally variable,
24    precluding the approach from being easily transferable between sites.  In terms of deriving
25    environmental criteria for Pb, background levels should be considered on a site-specific basis if
26    there is sufficient information that Pb concentrations are naturally elevated. As discussed
27    previously, the use of radiogenic Pb isotopes is useful for source apportionment.
28
      December 2005                        8-136           DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.3  Distribution of Lead in Aquatic Ecosystems
 2    8.2.3.1  Introduction
 3         Atmospheric Pb is delivered to aquatic ecosystems primarily through deposition (wet or
 4    dry) or through erosional transport of soil particles (Baier and Healy, 1977; Dolske and
 5    Sievering, 1979). A number of physical and chemical factors govern the fate and behavior of Pb
 6    in aquatic systems.  The EPA summarized some of these controlling factors in the 1986 Pb
 7    AQCD (U.S. Environmental Protection Agency, 1986a). For example, the predominant form of
 8    Pb in the environment is in the divalent (Pb2+) form and complexation with inorganic and
 9    organic ligands is dependent on pH (Lovering, 1976; Rickard and Nriagu, 1978). A significant
10    portion of Pb in the aquatic environment exists in the undissolved form (i.e., bound to suspended
11    particulate matter).  The ratio of Pb in suspended solids to Pb in filtrate varies from 4:1 in rural
12    streams to 27:1 in urban streams (Getz et al., 1977). In still waters, Pb is removed through
13    sedimentation at a rate determined by temperature, pH, oxidation-reduction (redox) potential,
14    organic content, grain size, and chemical form of Pb in the water and biological activities (Jenne
15    and Luoma, 1977).  Since the publication of the 1986 Pb AQCD, knowledge of the properties of
16    Pb in aquatic ecosystems has expanded.  This section will provide further detail on the chemical
17    species and the environmental factors affecting speciation of Pb in the aquatic environment. In
18    addition, quantitative distributions of Pb in water, sediment, and biological tissues will be
19    presented for aquatic ecosystems throughout the United States. Finally, recent studies discussing
20    the tracing of Pb in aquatic systems will be summarized.
21
22    8.2.3.2  Speciation of Lead in Aquatic Ecosystems
23         The speciation of Pb in the aquatic environment is controlled by many factors. The
24    primary form of Pb in aquatic environments is divalent (Pb2+), while Pb4+ exists only under
25    extreme oxidizing conditions (Rickard and Nriagu,  1978). Labile forms of Pb (e.g., Pb2+,
26    PbOH+, PbCOs) are a significant portion of the Pb inputs to aquatic systems from atmospheric
27    washout. Lead is typically present in acidic aquatic environments as PbSC>4, PbCU, ionic Pb,
28    cationic forms of Pb-hydroxide, and ordinary Pb-hydroxide (Pb(OH)2).  In alkaline waters,
29    common species of Pb include anionic forms of Pb-carbonate (Pb(CC>3)) and Pb(OH)2.
30    Speciation models have been developed based on the chemical equilibrium model developed by
31    Tipping (1994) to assist in examining metal speciation.  The EPA MINTEQA2 computer model

      December 2005                       8-137           DRAFT-DO NOT QUOTE OR CITE

-------
 1    (http://www.epa.gov/ceampubl/mmedia/minteq/) is one such equilibrium speciation model that
 2    can be used to calculate the equilibrium composition of dilute aqueous solutions in the laboratory
 3    or in natural aqueous systems.  The model is useful for calculating the equilibrium mass
 4    distribution among dissolved species, adsorbed species, and multiple solid phases under a variety
 5    of conditions, including a gas phase with constant partial pressures.  In addition to chemical
 6    equilibrium models, the speciation of metals is important from a toxicological perspective.
 7    The BLM was developed to study the toxicity of metal ions in aquatic biota and was previously
 8    described in Section 8.2.2.4.  Further detail on speciation models is not provided herein, rather a
 9    general overview of major speciation principles  are characterized in the following sections.
10
11    Acidity (pH)
12    Freshwater
13          Most of the Pb in aquatic environments is in the inorganic form (Sadiq, 1992). The
14    speciation of inorganic Pb in freshwater aquatic  ecosystems is dependent upon pH and the
15    available complexing ligands.  Solubility varies  according to pH, temperature, and water
16    hardness (Weber, 1993).  Lead rapidly loses solubility above pH 6.5 (Rickard and Nriagu, 1978)
17    and as water hardness increases.  In freshwaters, Pb typically forms strong complexes with
18    inorganic OFTand CO32  and weak complexes with Cl  (Long and Angino, 1977; Bodek et al.,
19    1988).  The primary form of Pb at low pH (<6.5) is predominantly Pb2+ and less abundant
20    inorganic forms include Pb(HCO)3, Pb(SO4)22~, PbCl, PbCO3, and Pb2(OH)2CO3 (Figure
21    8-2.3.1). At higher pH (>7.5), Pb forms hydroxide complexes (PbOH+, Pb(OH)2, Pb(OH)3 ,
22    Pb(OH)42 ).
23          Organic compounds in surface waters may originate from natural (e.g., humic or fulvic
24    acids) or anthropogenic sources (e.g., nitrilotriacetonitrile and ethylenediaminetetraaceitc acid
25    [EDTA]) (U.S. Environmental Protection Agency, 1986b). The presence of organic complexes
26    has been shown to increase the rate of solution of Pb bound as Pb-sulfide (Lovering, 1976).
27    Soluble organic Pb compounds are present at pH values near 7 and may remain bound at pH as
28    low as 3 (Lovering,  1976; Guy and Chakrabarti, 1976).  At higher pH (7.4 to  9), Pb-organic
29    complexes are partially decomposed. Water hardness and pH were found to be important in Pb-
30    humic acid interactions (O'Shea and Mancy, 1978). An increase in pH increased the
31    concentration of exchangeable Pb complexes, while an increase in hardness tended to decrease

      December 2005                        8-138            DRAFT-DO NOT QUOTE OR CITE

-------
                     100
    Figure 8-2.3.1.  Distribution of lead aqueous species as a function of pH based on
                    a concentration of 1 ug/L lead (U.S. Environmental Protection
                    Agency, 1999).
1
2
O
4
5
     the humic acid-Pb interactions.  Thus, the metals involved in water hardness apparently inhibit
     the exchangeable interactions between metals and humic acids.

     Marine Water
            In marine systems, an increase in salinity increases complexing with chloride and
 6   carbonate ions and reduces the amount of free Pb2+.  In seawaters and estuaries at low pH, Pb is
 7   primarily bound to chlorides (PbCl, PbCl2, PbCl37, PbCU2 ) and may also form inorganic
 8   Pb(HCO)3, Pb(SO4)22~, or PbCO3. Elevated pH in saltwater environments results in the
 9   formation of Pb hydroxides (PbOH+, Pb(OH)2, Pb(OH)3  , Pb(OH)42 ) (Figure 8-2.3.2).  A recent
10   examination of Pb species in seawater as a function of chloride concentration suggested that the
11   primary species were PbCl3~ > PbCO3 > PbCl2 > PbCl+ > and Pb(OH)+ (Fernando, 1995). Lead
12   in freshwater and seawater systems is highly complexed with carbonate ligands suggesting that
13   Pb is likely to be highly available for sorption to suspended materials (Long and Angino, 1977).
14   Current information suggests that inorganic Pb is the dominant form in seawater; however, it has
    December 2005
                                          8-139
DRAFT-DO NOT QUOTE OR CITE

-------
                    100
                 c
                 o
                 u
                 c
                 o
                o
     Figure 8-2.3.2.  Lead speciation versus chloride content (Fernando, 1995).
 1   been shown that organically bound Pb complexes make up a large portion of the total Pb
 2   (Capodaglio et al., 1990).
 3
 4   Sorption
 5          Sorption processes (i.e., partitioning of dissolved Pb to suspended particulate matter or
 6   sediments) appear to exert a dominant effect on the distribution of Pb in the environment
 7   (U.S. Environmental Protection Agency, 1979).  Sorption of Pb results in the enrichment of
 8   bed sediments, particularly in environments with elevated organic matter content from
 9   anthropogenicsources. Lead adsorption to aquatic sediments is correlated with pollution in sites
10   containing high levels of anthropogenic organic content, even under acidic conditions (Tada and
11   Suzuki, 1982; Brook and Moore,  1988; Davis and Galloway, 1993; Botelho et al., 1994; Davis
12   et al., 1996).  Particulate-bound forms are more often linked to urban runoff and mining effluents
13   (Eisler, 2000).
14          Solid Pb complexes form when Pb precipitates or adsorbs to suspended particulate matter
15   and sediments. Inorganic Pb adsorption to  suspended organic matter or sediments is dependent
16   on parameters such as, pH, salinity, water hardness, and the composition of the organic matter
17   (U.S. Environmental Protection Agency, 1979).  In addition to suspended organic matter, Pb can
     December 2005
8-140
DRAFT-DO NOT QUOTE OR CITE

-------
 1    adsorb to biofilms (i.e., bacteria) (Nelson et al., 1995; Wilson et al., 2001).  Adsorption typically
 2    increases with increasing pH, increasing amounts of iron or manganese; and with a higher degree
 3    of polarity of the particulate matter (e.g., clays). Adsorption decreases with water hardness
 4    (Syracuse Research Corporation., 1999).  At higher pH, Pb precipitates as Pb(OH)+and
 5    PbHCC>3+ into bed sediments (Weber, 1993). Conversely, at low pH, Pb is negatively sorbed
 6    (repelled from the adsorbent surface) (U.S. Environmental Protection Agency, 1979; Gao et al.,
 7    2003). In addition, Pb may be remobilized from sediment with a decrease in metal concentration
 8    in the solution phase, complexation with chelating agents (e.g., EDTA), and changing redox
 9    conditions (Gao et al., 2003). Changes in water chemistry (e.g., reduced pH or ionic
10    composition) can cause sediment Pb to become remobilized and potentially bioavailable to
11    aquatic organisms (Weber, 1993).
12
13    Biotransfor matron
14          Methylation may result in Pb remobilization and reintroduction into the aqueous
15    environment compartment and its subsequent release into the atmosphere (Syracuse Research
16    Corporation., 1999).  However, methylation is not a significant environmental pathway
17    controlling the fate of Pb in the aquatic environment. The microbial methylation of Pb in aquatic
18    systems has been demonstrated experimentally, but evidence for natural occurrence is limited
19    (Beijer and Jernelov, 1984; DeJonghe and Adams, 1986). Reisinger et al. (1981) examined the
20    methylation of Pb in the presence of numerous bacteria known to alkylate metals and did not find
21    evidence of Pb methylation under any test condition. Tetramethyl-Pb may be formed by the
22    methylation of Pb-nitrate or Pb-chloride in sediments (Bodek et al., 1988).  However,
23    tetramethyl-Pb is unstable and may degrade in aerobic environments after being released from
24    sediments (U.S. Environmental Protection Agency, 1986b). Methylated species of Pb may also
25    be formed by the decomposition of tetralkyl-Pb compounds (Radojevic and Harrison, 1987;
26    Rhue et al.,  1992).  Sadiq (1992) reviewed the methylation of Pb compounds and suggested that
27    chemical methylation of Pb is the dominant process and that biomethylation is of secondary
28    importance.
29
     December 2005                        8-141          DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.3.3  Spatial Distribution of Lead in Aquatic Ecosystems
 2          The 1986 Pb AQCD did not describe the distribution and concentration of Pb throughout
 3    aquatic ecosystems of the United States.  Consequently, an analysis of readily available data on
 4    Pb concentrations was conducted to determine the distribution of Pb in the aquatic environment.
 5    Data from the United States Geological Survey (USGS) National Water-Quality Assessment
 6    (NAWQA) program were queried and retrieved.  NAWQA contains data on Pb concentrations in
 7    surface water, bed sediment, and animal tissue for more than 50 river basins and aquifers
 8    throughout the country, and it has been used by the EPA for describing national environmental
 9    concentrations for use in developing AWQC.
10          NAWQA data are collected during long-term, cyclical investigations wherein study units
11    undergo intensive sampling for 3 to 4 years, followed by low-intensity monitoring and
12    assessment of trends every  10 years. The NAWQA program's  first cycle was initiated in 1991;
13    therefore, all available data are less than 15 years old. The second cycle began in 2001 and is
14    ongoing; data are currently  available through 30  September 2003.  The NAWQA program study
15    units were selected to represent a wide variety of environmental conditions and contaminant
16    sources; therefore, agricultural, urban, and natural areas were all included.  Attention was also
17    given to selecting sites covering a wide variety of hydrologic and ecological resources.
18          NAWQA sampling protocols are designed to promote data consistency within and among
19    study units while minimizing local-scale spatial variability.  Water-column sampling is
20    conducted via continuous monitoring, fixed-interval sampling,  extreme-flow sampling, as well as
21    seasonal, high-frequency sampling in order to characterize spatial, temporal, and seasonal
22    variability as a function of hydrologic conditions and contaminant sources.  Sediment and tissue
23    samples are collected during low-flow periods during the summer or fall to reduce seasonal
24    variability. Where possible, sediment grab samples are collected along a 100-m stream reach,
25    upstream of the location of the water-column sampling. Five to ten deposit!onal zones at various
26    depths, covering left bank, right bank, and center channel, are sampled to ensure a robust
27    representation of each site.  Fine-grained samples from the surficial 2 to 3 cm of bed sediment at
28    each depositional zone are sampled and composited.  Tissue samples are collected following a
29    National Target Taxa list and decision trees that  help guide selection from that list to
30    accommodate local variability.
      December 2005                       8-142           DRAFT-DO NOT QUOTE OR CITE

-------
 1          The NAWQA dataset was chosen over other readily available national databases (i.e. the
 2    USEPA-maintained database for the STOrage and RETrieval [STORET] of chemical, physical,
 3    and biological data), because the study design and methods used to assess the water quality of
 4    each study unit are rigorous and consistent, and, as such, these data may be presented with a high
 5    level of confidence. This is in stark contrast to the STORET database, which essentially serves
 6    as a depot for any organization wishing to share data they have generated.  This lack of a
 7    consistent methodology or QA/QC protocol has lead to the STORET data being highly qualified
 8    and offered with only a mild level of confidence. Furthermore, because there is no standard for
 9    site selection within STORET, the database may be biased toward contaminated sites.  Finally,
10    and, perhaps most importantly, the majority of the available Pb data in STORET predate the use
11    of clean techniques for Pb quantification.4
12
13    Data Acquisition and Analysis
14          The following data were downloaded for the entire United States (all states) from the
15    NAWQA website (http://water.usgs.gov/nawqa/index.html): site information, dissolved Pb
16    concentration in surface water (|ig/L), total Pb concentration (|ig/g) in bed sediment (<63|im)5,
17    and Pb concentration in animal tissue (|ig/g dw). Using the land use classification given for each
18    site, the data were divided into two groups: "natural" and "ambient" (Table 8-2.3.1).
19    All  samples were considered to fall within the ambient group (the combined contribution of
20    natural and anthropogenic sources), whereas the natural group comprised "forest," "rangeland,"
21    or "reference" samples only6. These groups follow those defined and recommended for use by
22    the EPA's Framework for Inorganic Metals Risk Assessment (U.S. Environmental Protection
23    Agency, 2004c). Finally, in addition to the natural/ambient classification, tissue samples were
24    further divided into "whole organism" and "liver" groups.
25
             4 The authors recognize the existence of several local and regional datasets that may be of quality equal to
      NAWQA; however, due to the national scope of this assessment, these datasets were not included in the following
      analyses. We were unable to identify any monitoring data of similar quality for the marine/estuarine environment.
             5 NAWQA sediment samples are sieved to <63um to promote the collection of fine-grained surficial
      sediments, which are natural accumulators of trace elements.
             6 The authors acknowledge that while Pb samples collected from sites classified under these three land use
      categories will most closely reflect natural background concentrations, atmospheric input of lead may be present.

      December 2005                        8-143           DRAFT-DO NOT QUOTE OR CITE

-------
          Table 8-2.3.1. NAWQA Land Use Categories and Natural/Ambient Classification
NAWQA Land Use Categories
Agricultural
Commercial/Industrial
Cropland
Forest
Mining
Mixed
NA
Orchard/Vineyard
Other/Mixed
Pasture
Rangeland
Reference
Residential
Urban
Classification
Ambient
Ambient
Ambient
Ambient/Natural
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient/Natural
Ambient/Natural
Ambient
Ambient
 1         All data were compiled in spreadsheets wherein non-detect values were converted to one-
 2   half of the detection limit and the total number of samples, percentage of non-detect
 3   values(percent censorship), minimum, maximum, median, standard deviation, and cumulative
 4   density functions were calculated for each endpoint for both the natural and ambient groups.
 5   Since all data were geo-referenced, a geographic information system (GIS; ArcGIS) was used to
 6   generate maps, conduct spatial queries and analyses, and calculate statistics.
 7
 8   Lead Distributions Generated from the NA WQA Database
 9   Natural versus Ambient Groups
10         There were four to eight times more ambient surface water (Table 8-2.3.2) and bulk
11   sediment (Table 8-2.3.3) samples in the compiled dataset than natural samples. This is most
12   likely a function of both the NAWQA program site selection process and the fact that sites
13   unaffected by human activities are extremely limited. The spatial distributions of natural and

     December 2005                       8-144           DRAFT-DO NOT QUOTE OR CITE

-------
          Table 8-2.3.2. Summary Statistics of Ambient and Natural Levels of
                          Dissolved Lead in Surface Water
Surface Water Dissolved Pb (ug/L)
Statistic
% Censorship
N
Minimum
Maximum
Mean
Standard Deviation
95th Percentile
96th Percentile
97th Percentile
98th Percentile
99th Percentile
Table 8-2.3.3. Summary
Natural
87.91
430
0.04
8.40
0.52
0.59
0.50
0.67
1.00
1.79
2.48
Statistics of Ambient and Natural
<63 jim Bulk Sediment
Ambient
85.66
3445
0.04
29.78
0.66
1.20
1.10
2.00
2.34
3.58
5.44
Levels of Total Lead in
Bulk Sediment <63 um Total Lead (ug/g)
Statistic
% Censorship
N
Minimum
Maximum
Mean
Standard Deviation
Median
95th percentile
Natural
1.16
258
0.50
12000
109.07
786.74
22.00
161.50
Ambient
0.48
1466
0.50
12000
120.11
672.41
28.00
200.00
December 2005                      8-145          DRAFT-DO NOT QUOTE OR CITE

-------
 1    ambient surface water/sediment sites were fairly comparable, with natural samples located in
 2    almost all of the same areas as ambient samples except in the Midwest (Ohio, Illinois, Iowa, and
 3    Michigan), where natural sites were not present (Figure 8-2.3.3). This exception may be because
 4    these areas are dominated by agricultural and urban areas.  The same spatial distributions were
 5    observed for the natural and ambient liver and whole organism tissue samples (Figure 8-2.3.4
 6    and Figure 8-2.3.5).
 7
 8    Surface Water
 9          The total number of surface water Pb samples was 3,445; however these data were highly
10    censored with 85.66% of the ambient samples (2951/3445) and 87.91% of the natural samples
11    (378/430) below the detection limit7 (Table 8-2.3.2).  Consequently, the majority of the
12    variability between these two datasets fell between the 95th and 100th (maximum) percentiles,
13    as was shown by the frequency distributions of the two groups deviating only at the upper and
14    lower tails with most of the overlapping data falling at 0.50 |ig/L (one-half of the most common
15    detection limit, 1.0 |ig/L; Figure 8-2.3.6). As expected, due to the definitions of the natural and
16    ambient groups, the 95th and 100th percentiles were consistently higher for the ambient samples
17    than the natural samples.  Similarly, the mean ambient Pb concentration (0.66 |ig/L) was higher
18    than the mean natural Pb concentration (0.52 |ig/L).8
19          Due to the preponderance of non-detectable (ND) measurements, assessing national trends
20    in surface water-dissolved Pb concentrations was not possible. However, areas with elevated Pb
21    concentrations were identified by classifying the data with detectable Pb concentrations  above
22    and below the 99th percentile9. Areas with high surface water Pb concentrations were observed
23    in Washington, Idaho, Utah, Colorado,  Arkansas, and Missouri (Figure 8-2.3.7).  The maximum
             7 The NAWQA dataset contains multiple detection limits for Pb in surface water. While the majority of
      data were analyzed with a detection limit of 1.0 ug/L (before 2000/2001), the most recent samples were analyzed
      with either a 0.5, 0.2, 0.16, or 0.08 ug/L detection limit (after 2000/2001), and some older samples (N = 20) were
      analyzed with a detection limit of 2.0 ug/L.
             8 The same pattern was observed upon calculating the mean Pb concentrations based on detect data only
      (ambient mean =1.66 ug/L, natural mean =0.87 ug/L); however, deletion of non-detect data has been shown to
      increase the relative error in the mean to a greater extent than inclusion of non-detects as  1A of the detection limit
      (Newman et al., 1989); therefore means were calculated using the latter method for this analysis.
             9 The 99th percentile (versus the 95th percentile) was chosen in this instance to represent extreme
      conditions given the small window of variability in the dataset.  The 95th percentile will be used in subsequent
      analyses of this type.
      December 2005                         8-146           DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
oo
fe
H

6
o


o
H

O


O
H
W

O


O
HH
H
W
                                                                   Legend
                                                                  •  Natural Surface Water/Sediment Sites

                                                                  o  Ambient Surface Water/Sediment Sites
      Figure 8-2.3.3.  Spatial distribution of natural and ambient surface water/sediment sites (Surface water: natural N = 430,

                    ambient N = 3445;  Sediment: natural N = 258, ambient N = 1466).

-------
o

o

3
CT*
to
o
o
oo
oo
fe
H

6
o


o
H

O


O
H
W

O
Legend


  •  Natural Liver Sites

  o  Ambient Liver Sites
O    Figure 8-2.3.4.  Spatial distribution of natural and ambient liver tissue sample sites (Natural N = 83, Ambient N = 559).
H
W

-------
o

O

to
O
O
oo
VO
fe
H

6
o


o
H

O


O
H
W

O


O
HH
H
W
                                                                    Legend


                                                                      •  Natural Whole Organism Sites

                                                                      O  Ambient Whole Organism Sites
      Figure 8-2.3.5. Spatial distribution of natural and ambient whole organism tissue sample sites (Natural N = 93,

                    Ambient N = 332).

-------
                      100
                  T  80
                  5?

                      60
                   3  40
                   D
                  O  20
                                      •Ambient    Natural
                        0.0           0.1           1.0          10.0
                                 Surface Water Dissolved Pb (M9/L)
                                100.0
     Figure 8-2.3.6.  Frequency distribution of ambient and natural levels of surface water
                     dissolved lead (ug/L).
 1   measured Pb concentration was located in Canyon Creek at Woodland Park, ID, a site classified
 2   as mining land use.
 3
 4   Sediment
 5         There were approximately one-half of the number of surface water data available for
 6   sediments (N = 1466). In contrast to the surface water data, however, very few sediment data
 7   were below the detection limit (7/1466 ambient ND, 3/258 natural ND; Table 8-2.3.3).
 8   As expected, the mean ambient Pb concentration was higher than the mean natural Pb
 9   concentration (120.11 and 109.07 |ig/g, respectively).  Similarly, the median ambient Pb
10   concentration was higher than the median natural Pb concentration (28.00 and 22.00 |ig/g,
11   respectively) and the ambient 95th percentile was higher than the natural 95th percentile
12   (200.00 and 161.50 |ig/g, respectively). While the natural and ambient surface water Pb
13   distributions differed only at the extremes, the natural  sediment Pb percentiles were consistently
14   lower than the ambient percentiles throughout the distributions (Figure 8-2.3.8). Unlike the
     December 2005
8-150
DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
oo
H

6
o


o
H

O

o
H
W

O


O
HH
H
W
                           OB"
                                               ^  o  o


                                               T3-0	©-
                                                                      Legend

                                                                         Surface Water Dissolved Pb (M9/L)

                                                                        o  Non-detect

                                                                        0  0.51 - 5.44 (<99th precentile)

                                                                            5.45 - 29.78 (>99th percentile)
       Figure 8-2.3.7.  Spatial distribution of dissolved lead in surface water (N = 3445).

-------
                    100
                                 1        10       100      1000     10000    100000
                               Bulk Sediment <63um Total Pb (M9/g)
     Figure 8-2.3.8.  Frequency distribution of ambient and natural levels of bulk sediment
                     <63 jim total Pb (ug/g).
 1   surface water dataset, because the sediment dataset was not heavily censored, assessing national

 2   trends in sediment Pb concentrations was possible. The data were mapped and categorized into

 3   the four quartiles of the frequency distribution (Figure 8-2.3.9).  The following observations

 4   were made:

 5          •  Sediment Pb concentrations generally increased from west to east (the majority of
 6             sites along East Coast had Pb concentrations in the fourth quartile of the sediment Pb
 7             concentration frequency distribution).

 8          •  Several "hot spots" of concentrated sites with elevated sediment Pb concentrations
 9             were apparent in various western states.

10          •  Sediment Pb concentrations were generally lowest in the midwestern states
11             (the majority of sites in North Dakota, Nebraska, Minnesota, and Iowa had Pb
12             concentrations in the first or second quartile of sediment Pb concentration
13             frequency distribution).

14         As was seen with surface water Pb concentrations, the highest measured sediment Pb

15   concentrations were found in Idaho, Utah, and Colorado.  Not surprisingly, of the top 10

16   sediment Pb concentrations recorded, 7 were measured at sites classified as mining land use.

17
     December 2005
8-152
DRAFT-DO NOT QUOTE OR CITE

-------
o

O

to
O
O
oo
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
                                                                            Legend

                                                                               Bulk Sediment <63|j Total Lead (|jg/g)


                                                                                  0 0.50- 18.00 (1st Quartile)


                                                                                  0 18.01 - 28.00 (2nd Quartile)


                                                                                  • 28.01 -49.00 (3rd Quartile)


                                                                                  • 49.01 -12000.00 (4th Quartile)
        Figure 8-2.3.9.  Spatial distribution of total lead in bulk sediment <63 um (N = 1466).

-------
 1    Tissue
 2          As was true for the surface water data, there were a high number of tissue samples below
 3    the detection limit (47/93 natural whole organism ND, 130/332 ambient whole organism ND,
 4    74/83 natural liver ND, 398/559 ambient liver ND; Table 8-2.3.4).  In general, more
 5    non-censored data were available for whole organism samples than liver samples, and for
 6    ambient sites than natural  sites. As expected, for whole organism samples, the 95th percentile Pb
 7    concentration measured at ambient sites was higher than that measured at natural sites (3.24 and
 8    2.50  |ig/g, respectively); however, Pb liver concentration 95th percentiles for ambient and
 9    natural samples were very similar, with the natural 95th percentile actually higher than the
10    ambient 95th percentile (1.26 and 1.06  |ig/g, respectively).  In addition, as expected, the median
11    and mean Pb liver concentrations of ambient samples (0.15 and 0.36 |ig/g, respectively) were
12    higher than the median and mean Pb liver concentrations of natural samples (0.11 and 0.28 |ig/g,
13    respectively). The same pattern was observed in the whole organism median and mean Pb
14    concentrations  (ambient: median = 0.59, mean = 1.03; natural: median = 0.35, mean =
15    0.95  |ig/g).  In  addition, the frequency distributions of the liver and whole organism Pb
16    concentrations  followed the same trends, with the natural percentiles consistently lower than the
17    ambient percentiles throughout the distributions (Figure 8-2.3.10 and Figure 8-2.3.11).
18          These whole organism results were compared with findings from the 1984 U.S. Fish and
19    Wildlife Service (USFWS) National Contaminant Biomonitoring Program (NCBP) (Schmitt and
20    Brumbaugh, 1990).  As part of this program, 321 composite samples of 3 to 5 whole, adult fish
21    of a single species were collected from 109 river and Great Lake stations throughout the country.
22    Samples were analyzed for Pb concentrations (|ig/g ww) and the geometric mean, maximum, and
23    85th  percentile were calculated. Upon comparing these summary statistics with the equivalent
24    NAWQA ambient group value (NCBP  stations were representative of both natural and
25    anthropogenically influenced conditions), a very strong agreement between the two analyses was
26    observed for each endpoint (Table 8-2.3.5).  For example, NCBP and NAWQA geometric mean
27    Pb concentrations were nearly identical10 (0.55 and 0.54 |ig/g dw, respectively) and the 85th
28    percentiles only differed by 0.5 |ig Pb/g dw (NCBP, 1.10 and NAWQA, 1.60).
            10 The authors acknowledge that a high degree of censorship is present in both datasets and no firm
      conclusions can be drawn by comparing these means. The objective of this exercise was limited to showing how
      the NAWQA data compare to other national datasets.

      December 2005                        8-154           DRAFT-DO NOT QUOTE OR CITE

-------
        Table 8-2.3.4. Summary Statistics of Ambient and Natural Levels of Lead
                        in Whole Organism and Liver Tissues

                      Tissue Pb (jig/g dry weight)
                      Whole Organism
                        Liver

% Censorship
N
Minimum
Maximum
Mean
Standard Deviation
Median
95th percentile
Natural
50.54
93
0.08
22.60
0.95
2.53
0.11
1.26
Ambient
39.16
332
0.08
22.60
1.03
1.74
0.15
1.06
Natural
89.16
83
0.01
3.37
0.28
0.54
0.35
2.50
Ambient
71.20
559
0.01
12.69
0.36
0.96
0.59
3.24
              0.01
0.1             1             10
 Liver Pb (M9/9 dry weight)
                     100
Figure 8-2.3.10.  Frequency distribution of ambient and natural levels of lead in liver
                tissue (ug/g dry weight).
December 2005
       8-155
DRAFT-DO NOT QUOTE OR CITE

-------
                 100
                       * Ambient l=l Natural
                   0.01           0.1            1             10           100

                          Whole Organism Pb (M9/9 dry weight)
    Figure 8-2.3.11.  Frequency distribution of ambient and natural levels of lead in whole
                    organism tissue (ug/g dry weight).
             Table 8-2.3.5. Comparison of NCBP and NAWQA Ambient Lead Levels
                                  in Whole Organism Tissues

Statistic
Geometric Mean
Maximum
85th Percentile
Whole Organism Lead Concentration (jig/gd
NCBP1
0.55
24.40
1.10
ry weight)
NAWQA
0.54
22.60
1.60
     1 To convert between wet and dry weight, wet weight values were multiplied by a factor of five.
1         As was the case with surface water data, the high amount of non-detectable measurements
2   did not allow for a national assessment of spatial trends in Pb tissue concentrations. Instead,
3   areas with high Pb tissue concentrations were identified by classifying the data above and below
    December 2005
8-156
DRAFT-DO NOT QUOTE OR CITE

-------
 1    the 95th percentile. Similar to surface water and sediments, tissue concentrations were found to
 2    be elevated in Washington, Idaho, Utah, Colorado, Arkansas, and Missouri; however, several of
 3    the highest measured Pb concentrations were also found in study units in the southwestern and
 4    southeastern states (Figure 8-2.3.12 and Figure 8-2.3.13). As expected, the majority of the
 5    samples with elevated Pb concentrations were taken from sites classified as urban,
 6    commercial/industrial, or mining.
 7
 8    8.2.3.4  Tracing the Fate and Transport of Lead in Aquatic Ecosystems
 9          The following section presents a generalized framework for the fate and transport of Pb in
10    aquatic systems (Figure 8-2.3.14).  The primary source of Pb in natural systems is atmospheric
11    deposition (Rickard and Nriagu,  1978; U.S.Environmental Protection Agency, 1986a).
12    Estimated median global atmospheric emission for anthropogenic and natural sources are
13    332 x 106 kg/year and  12 x 106 kg/year, respectively (summarized by Giusti et al., 1993).
14    Inorganic and metallic Pb  compounds are nonvolatile and will partition to airborne particulates
15    or water vapors (Syracuse Research Corporation., 1999). Dispersion and deposition of Pb is
16    dependent on the particle size (U.S.Environmental Protection Agency, 1986a;  Syracuse Research
17    Corporation., 1999). More soluble forms  of Pb will be removed from the atmosphere by
18    washout in rain.
19          In addition to atmospheric deposition, Pb may enter aquatic ecosystems through industrial
20    or municipal wastewater effluents,  storm water runoff, erosion, or direct point  source inputs
21    (e.g., Pb shot or accidental spills).  Once in the aquatic environment, Pb will partition between
22    the various compartments  of the  system (e.g., dissolved phase,  solid phase, biota). The
23    movement of Pb between dissolved and particulate forms is governed by factors such as pH,
24    sorption, and biotransformation (see Section 8.2.3.2). Lead bound to organic matter will settle to
25    the bottom sediment layer, be assimilated by aquatic organisms, or be resuspended in the water
26    column. The uptake, accumulation, and toxicity of Pb in aquatic organisms from water and
27    sediments are influenced by various environmental factors (e.g., pH, organic matter, temperature,
28    hardness, bioavailability).  These factors are further described in  Sections 8.2.4.5.3 and
29    8.2.4.5.4). The remainder of this section discusses some methods for describing the distribution
30    of atmospheric Pb in the aquatic  environment.
31

      December 2005                        8-157          DRAFT-DO NOT  QUOTE OR CITE

-------
o

O

to
O
O
oo
oo
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
                                                                                Legend

                                                                                   Liver Lead (|jg/g dry weight)

                                                                                   O  Non-detect


                                                                                   0  0.10 -1.06 (<95th precentile)


                                                                                  V  1.06-12.69 (>95th precentile)


                                                                 -  „ 	 -  '    Qrto    >V

                                                                   '3~W??®i.   '^$fe&i^
                                                                 r, !>3V-<1, ,• ,-i'Hl    •- -ftWHfiU
                                                                                 A
        Figure 8-2.3.12.  Spatial distribution of lead in liver tissues (N = 559).

-------
o

O

to
O
O
oo
VO
H

6
o


O
H

O


o
H
W

O


O
HH
H
W
                                                                          Legend

                                                                            Whole organism Lead (\iglg dry weight)

                                                                            O  Non-detect


                                                                            0 0.20 - 3.24 (<95th precentile)


                                                                               3.24 - 22.60 (>95th precentile)
        Figure 8-2.3.13.  Spatial distribution of lead in whole organism tissues (N = 332).

-------
                                    *
                                   Dissolved
                                    Organic
                                  — Pb
     Figure 8-2.3.14.      Lead cycle in an aquatic ecosystem.
 1   Sediment Core Dating and Source Tracing
 2          In addition to directly measuring Pb concentrations in various aquatic
 3   compartments (see Section 8.2.3.3), it is useful to study the vertical distribution of Pb.
 4   Sediment profiling and core dating is a method used to determine the extent of
 5   accumulation of atmospheric Pb and provide information on potential anthropogenic
 6   sources.  Sediment concentration profiles are typically coupled with Pb isotopic analysis.
 7   The isotope fingerprinting method utilizes measurements of the abundance of common
 8   Pb isotopes (i.e., 204Pb, 206Pb, 207Pb, 208Pb) to distinguish between natural Pb over
 9   geologic time and potential anthropogenic sources. Details of this method were
10   described in Section 8.1.2. The concentration of isotope 204Pb has remained constant
11   throughout time, while the other isotope species can be linked to various anthropogenic
12   Pb sources.  Typically, the ratios  or signatures of isotopes (e.g., 207Pb:206Pb) are compared
13   between environmental samples to indicate similarities or differences in the site being
14   investigated and the potential known sources.
15          Generally, Pb concentrations in sediment vary with depth.  For example, Chow
16   et al. (1973) examined sediment Pb profiles in southern California. Lead concentrations
17   were increased in the shallower sediment depths  and comparatively decreased at greater
     December 2005
8-160    DRAFT-DO NOT QUOTE OR CITE

-------
 1    depths. These changes in sediment vertical concentration were attributed to higher
 2    anthropogenic Pb fluxes from municipal sewage, storm runoff, and atmospheric
 3    deposition.  Similar experiments conducted throughout the United States have also
 4    suggested an increase in Pb concentrations in the upper sediment layer concomitant with
 5    increases in anthropogenic inputs (Bloom and Crecelius, 1987; Case et al., 1989; Ritson
 6    et al., 1999; Chillrud et al., 2003).
 7          Sediment Pb concentration profiles and isotope analysis have also been used to
 8    identify specific anthropogenic sources. For example, Flegal et al. (1987) used isotopic
 9    ratios to trace sources of Pb in mussels from Monterey Bay, CA to a specific slag deposit.
10    Several investigators have examined isotopic tracers to determine potential regional
11    sources of Pb in eastern North America and the Great Lakes (Flegal et al., 1989b; Graney
12    et al., 1995; Blais, 1996). Water samples from Lake Erie and Lake Ontario were
13    collected and analyzed.  Lead isotope ratios (206Pb:207Pb) from the lakes were compared
14    to known ratios for Pb aerosols derived from industrial sources in Canada and the United
15    States and found to correlate positively. This indicated that a majority of Pb  in the lakes
16    was derived from those industrial sources (Flegal et al., 1989b). Lead isotopes in
17    sediment cores from Quebec and Ontario, Canada were also used to distinguish between
18    the amount of Pb deposited from local Canadian sources (28.4 to 61.7%) and U.S.
19    sources (38.3 to 71.6%) (Blais, 1996). Examination of Pb isotopes in  sediment and
20    suspended sediment in the St. Lawrence River were used to identify potential
21    anthropogenic Pb sources from Canada (Gobeil et al., 1995, 2005).  Graney et al. (1995)
22    used Pb isotope measurements to describe the differing historic sources of Pb in Lake
23    Erie, Ontario and in Michigan. Temporal changes in Pb isotopic ratios were  found to
24    correspond to sources such as regional deforestation from 1860 through!890, coal
25    combustion and or smelting through 1930, and the influence of leaded gasoline
26    consumption from 1930 to 1980.
27          The historic record of atmospheric Pb pollution has been studied to understand the
28    natural background Pb concentration and the effects of Pb accumulation on ecosystems
29    (Bindler et al., 1999; Renberg et al., 2000, 2002; Brannvall et al., 2001a,b).  The most
30    extensive work in this area has been conducted at pristine locations in Sweden (Bindler et
31    al., 1999).  In this study, soil,  sediment, and tree rings were sampled for Pb

      December 2005                         8-161    DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentrations and isotopic analyses were conducted on the soil samples.  From this
 2    record, historic Pb concentrations and Pb accumulation rates were estimated. Present day
 3    concentrations in the forest soils ranged from 40 to 100 mg/kg, while a natural
 4    background concentration was estimated at <1 mg/kg. The authors were able to model
 5    Pb concentrations for the past 6,000 years and also to project Pb concentrations for the
 6    next 400 years, given an assumed atmospheric deposition rate of 1 mg Pb nT2/year.
 7    Models such as this are useful tools in determining the critical limits of metals in soils or
 8    sediments (Bindler et al., 1999; Renberg et al., 2002).
 9          Lead source association may also be assessed through retrospective measurements.
10    Squire et al. (2002) used a time-series approach to evaluate the change in Pb in
11    San Francisco Bay, CA from 1989 to 1999.  This approach involved the use of detailed
12    linear regression models and long-term monitoring data to determine changes in Pb
13    concentrations and to identify events  corresponding to those  changes. Sediment and
14    water samples were collected throughout the bay and combined with data on effluent
15    discharges, urban runoff, atmospheric deposition, and river discharges.  The authors
16    identified a 40% decline of Pb  in the  southern portion of the  bay but found no change in
17    the northern reach. The decline was attributed to a reduction in wastewater source
18    loadings over the previous decade.
19
20    8.2.3.5  Summary
21          Lead is widely distributed in aquatic ecosystems, predominantly originating from
22    atmospheric deposition or point source contribution. The fate and behavior of Pb in
23    aquatic systems is regulated by physical and chemical factors such as pH, salinity,
24    sediment sorption, transformation, and uptake by aquatic biota.  In the United States, Pb
25    concentrations in surface waters, sediments, and fish tissues range from 0.04 to 30 |ig/L,
26    0.5 to 12,000 mg/kg, and 0.08 to 23 mg/kg, respectively. Atmospheric sources are
27    generally decreasing, as the United States has removed Pb from gasoline and other
28    products.  However, elevated Pb concentrations remain at sites associated with mining
29    wastes or wastewater effluents. Since the 1986 Pb AQCD, much has been learned about
30    the processes affecting Pb fate and transport. Detailed analyses are currently available
31    (i.e., Pb isotope dating) to allow for constructing the history of Pb accumulation and
      December 2005                        8-162    DRAFT-DO NOT QUOTE OR CITE

-------
 1    identifying specific Pb contaminant sources. Continued source control along with
 2    examination of the physical and chemical properties will further allow for the reduction
 3    of Pb concentrations throughout the United States.
 4
 5    8.2.4   Species Response/Mode of Action
 6    8.2.4.1  Introduction
 7          Recent advancements in understanding the responses of aquatic biota to Pb
 8    exposure are highlighted in this section.  A summary of the conclusions on the review of
 9    aquatic responses to Pb from the appropriate sections of the 1986 Pb AQCD, Volume II
10    (U.S. Environmental Protection Agency, 1986a) and the subsequent conclusions and
11    recommendations contained in the EPA staff review of that document (U. S.
12    Environmental Protection Agency, 1990) are also provided. In addition, this section
13    summarizes research subsequent to the 1986 Pb AQCD on Pb uptake into aquatic biota,
14    effects of Pb speciation on uptake, resistance mechanisms to Pb toxicity, physiological
15    effects of Pb, factors that affect responses to Pb, and factors associated with global
16    climate change.  Areas of research that are not addressed here include literature related to
17    exposure to Pb shot or pellets and studies that examine human health-related endpoints
18    (e.g., hypertension), which are described in other sections of this document.
19
20    8.2.4.2  Lead Uptake
21          Lead is nutritionally nonessential and non-beneficial and is toxic to living
22    organisms in all  of its forms (Eisler, 2000). Lead can bioaccumulate in the tissues of
23    aquatic organisms through ingestion of food and water and adsorption from water
24    (Vazquez et al.,  1999; Vink, 2002) and subsequently lead to adverse effects (see Section
25    8.2.5).  Lead concentrations in the tissues of aquatic organisms are generally higher in
26    algae and benthic organisms and lower in higher trophic-level consumers (Eisler, 2000).
27    Metals are not metabolized; therefore, they are good integrative indicators of exposure in
28    aquatic biota (Luoma  and Rainbow, 2005). Metal uptake is complex, being influenced by
29    geochemistry, route of exposure (diet and adsorption), depuration,  and growth (Luoma
30    and Rainbow,  2005).  This section discusses the factors affecting uptake of Pb by aquatic
31    biota and the state of current research in this area.
      December 2005                       8-163    DRAFT-DO NOT QUOTE OR CITE

-------
 1          As described in Section 8.2.3, the solubility of Pb in water varies with pH,
 2   temperature, and ion concentration (water hardness) (Weber, 1993).  Lead becomes
 3   soluble and bioavailable under conditions of low pH, organic carbon content, suspended
 4   sediment concentrations, and ionic concentrations (i.e., low Cd, Ca, Fe, Mn, Zn) (Eisler,
 5   2000).  Lead rapidly loses solubility above pH 6.5 (Rickard and Nriagu, 1978) and
 6   precipitates out as Pb(OH)+ and PbHCC>3+ into bed sediments.  However, at reduced pH
 7   levels or ionic concentrations, sediment Pb can remobilize and potentially become
 8   bioavailable to aquatic organisms (Weber, 1993).
 9          The most bioavailable inorganic form of Pb is divalent Pb (Pb2+), which tends to
10   be more readily assimilated by organisms than complexed forms (Erten-Unal et al.,
11   1998).  On the other hand, the low solubility of Pb salts restricts movement across cell
12   membranes, resulting in less accumulation of Pb in fish in comparison to other metals
13   (e.g., Hg,  Cu) (Baatrup,  1991).
14          The accumulation of Pb in aquatic organisms is, therefore, influenced by water
15   pH, with lower pHs favoring bioavailability and accumulation. For example, fish
16   accumulated Pb at a greater rate in acidic lakes (pH = 4.9 to 5.4) than in more neutral
17   lakes (pH = 5.8 to 6.8) (Stripp et al.,  1990). Merlini and Pozzi (1977) found that
18   pumpkinseed sunfish exposed to Pb at pH 6.0 accumulated three-times as much Pb as
19   fish kept at pH 7.5.  However, Albers and Camardese (1993a,b) examined the effects of
20   pH on Pb  uptake in aquatic plants and invertebrates in acidic (pH -5.0) and nonacidic
21   (pH -6.5) constructed wetlands, ponds, and small lakes in Maine and Maryland.  Their
22   results suggested that low pH had little effect on the accumulation of metals by aquatic
23   plants and insects and on the concentration of metals in the waters of these aquatic
24   systems (Albers and Camardese, 1993a,b).
25          Three geochemical factors that influence metal bioaccumulation in aquatic
26   organisms include speciation,  particulate metal form, and metal form in the tissues of
27   prey items (Luoma and Rainbow, 2005).  Lead is typically present in acidic aquatic
28   environments as PbSC>4, PbCl4, ionic Pb, cationic forms of Pb-hydroxide, and ordinary
29   hydroxide Pb(OH>2. In alkaline waters, common species of Pb include anionic forms of
30   Pb-carbonate (Pb(CO3)) and Pb(OH)2.  Labile forms of Pb (e.g., Pb2+, PbOH+, PbCO3)
31   are a significant portion of the Pb inputs to aquatic systems from atmospheric washout.

     December 2005                        8-164   DRAFT-DO NOT QUOTE OR CITE

-------
 1    Particulate-bound forms are more often linked to urban runoff and mining effluents
 2    (Eisler, 2000).  Little research has been done to link the complex concepts of chemical
 3    speciation and bioavailability in natural systems (Vink, 2002). The relationship between
 4    the geochemistry of the underlying sediment and the impact of temporal changes (e.g.,
 5    seasonal temperatures) to metal speciation are particularly not well studied (Vink, 2002;
 6    Hassler et al., 2004).
 7          Generally speaking, aquatic organisms exhibit three Pb accumulation strategies:
 8    (1) accumulation of significant Pb concentrations with a low rate of loss, (2) excretion of
 9    Pb roughly in balance with availability of metal in the environment, and (3) weak net
10    accumulation due to very low metal uptake rate and no significant excretion (Rainbow,
11    1996).  Species that accumulate nonessential metals such as Pb and that have low rates of
12    loss must partition it internally in such a way that it is sparingly available metabolically.
13    Otherwise, it may cause adverse toxicological  effects (Rainbow, 1996). Aquatic
14    organisms that exhibit this type of physiological response have been recommended for
15    use both as environmental indicators of heavy  metal pollution (Borgmann et al., 1993;
16    Castro et al., 1996;  Carter and Porter, 1997) and, in the case of macrophytes, as
17    phytoremediators, because they accumulate heavy metals rapidly from surface water and
18    sediment (Gavrilenko and Zolotukhina, 1989;  Simoes Gon9alves et al., 1991; Carter and
19    Porter, 1997).
20          Uptake experiments with aquatic plants and invertebrates (e.g., macrophytes,
21    chironomids, crayfish) have shown steady increases in Pb uptake with increasing Pb
22    concentration in solution (Knowlton et al., 1983; Timmermans et al., 1992). In crayfish,
23    the process of molting can cause a reduction in body Pb concentrations, as Pb
24    incorporated into the crayfish shell is eliminated (Knowlton et al., 1983). Vazquez et al.
25    (1999) reported on the uptake of Pb from  solution to the extracellular and intracellular
26    compartments of 3 species of aquatic bryophytes. Relative to the 6 metals tested, Pb was
27    found to accumulate to the largest degree  in the extracellular compartments of all  3
28    bryophytes. The extracellular metals were defined as those that are incorporated into the
29    cell wall or are found  on the outer surface of the plasma membrane (i.e., adsorbed)
30    (Vazquez et al., 1999). Intracellular metals were defined as metals introduced into the
31    cell through a metabolically controlled process.

      December 2005                        8-165    DRAFT-DO NOT QUOTE OR CITE

-------
 1          Aral et al. (2002) examined the effect of growth on the uptake and elimination of
 2   trace metals in the abalone Haliotos. They reported that older abalones had generally
 3   lower whole body concentrations of heavy metals than did younger, rapidly growing
 4   individuals.  During the rapid growth of juveniles, the organism is less able to distinguish
 5   between essential (e.g., Zn), and nonessential metals (e.g., Pb).  Once they reach
 6   maturity, they develop the ability to differentiate these metals. Li et al. (2004) reported a
 7   similar response in zebra fish embryo-larvae. Li et al. (2004) suggested that mature
 8   physiological systems are not developed in the embryo-larvae to handle elevated
 9   concentrations of metals. Therefore, metals are transported into the body by facilitated
10   diffusion. Both the zebra fish and juvenile abalone demonstrate a rapid accumulation
11   strategy followed by a low rate of loss as described above. There are insufficient data
12   available to determine whether this phenomenon is true for other aquatic organisms.
13   Growth rates are generally thought to be an important consideration in the comparison of
14   Pb levels in individuals of the same species.  The larger the individual the more the metal
15   content is diluted by body tissue (Rainbow, 1996).
16          Once Pb is absorbed it may sequester into varying parts of the organism. Calcium
17   appears to have an important influence on Pb transfer.  For example, Pb uptake and
18   retention in the skin and skeleton of coho salmon was reduced when dietary Ca was
19   increased (Varanasi and Gmur, 1978).  Organic Pb compounds tend to accumulate in
20   lipids, and are taken up and accumulated in fish more readily than inorganic Pb
21   compounds (Pattee and Pain, 2003).
22          Given the complexities of metal uptake in natural systems, a model incorporating
23   some of the factors mentioned above is desirable.  The EPA Environmental Research
24   Laboratory developed a thermodynamic equilibrium model, MINTEQ that  predicts
25   aqueous  speciation, adsorption, precipitation, and/or dissolution of solids for a defined set
26   of environmental  conditions (MacDonald et al., 2002; Playle, 2004). Although not
27   specifically designed to model uptake, MINTEQ provides an indication of what forms of
28   the metal are likely to be encountered by aquatic organisms by estimating the formation
29   of metal  ions, complexation of metals, and the general bioavailability of metals from
30   environmental parameters. More recently, a mechanistic model centered on biodynamics
31   has been proposed by Luoma and Rainbow (2005) as a method of tying together

     December 2005                        8-166    DRAFT-DO NOT QUOTE OR CITE

-------
 1    geochemical influences, biological differences, and differences among metals to model
 2    metal bioaccumulation. The biodynamic model would be useful in determining the
 3    potential adverse effects on aquatic biota, which species are most useful as indicators of
 4    metal effects, and how ecosystems may change when contaminated by metals.
 5          Two prominent models examine trace metal bioavailability and its link to effects
 6    (Hassler et al., 2004).  These include the free ion activity model (FIAM) and the biotic
 7    ligand model (BLM), which is partially based upon FIAM.  Generally, FIAM explores
 8    the activity of free ions in solution.  More specific information on these models is
 9    provided in Section 8.2.2.  The FIAM has been used to examine cationic binding to
10    sensitive sites in algae and takes into account dissolved organic matter in complexation
11    reactions (Niyogi  and Wood, 2004). The BLM explores the activity of free ions at
12    biologically reactive sites (i.e., fish gill tissue). Both of these models  can increase our
13    awareness of the processes governing  the movement of Pb into aquatic biota. They
14    provide insight into the speciation of Pb under certain environmental conditions (e.g., pH,
15    DOC, hardness) and are important in helping understand how Pb  and  other metals move,
16    accumulate, and cause effects in aquatic organisms.  To date, there has been no BLM
17    model of Pb, although research has been conducted on a Pb-gill binding model for
18    rainbow trout (MacDonald et al., 2002; Niyogi and Wood, 2003, 2004).
19
20    Bioconcentration Factors (BCF)
21          BCFs for Pb are reported  for various aquatic plants in Table 8-2.4.1.  The green
22    alga Cladophora glomerata is reported as having the highest BCF (Keeney  et al., 1976).
23    Duckweed (Lemna minor} exhibited high BCF values ranging from 840 to 3560
24    depending on the method of measurement (Rahmani and Sternberg, 1999).  Duckweed
25    that was either previously exposed or not exposed to Pb was exposed to  a single dose of
26    Pb-nitrate at 5000 jig/L for 21  days. Duckweed that was previously exposed to Pb
27    removed 70 to 80% of the Pb from the water, while the previously unexposed duckweed
28    removed 85 to 90%. Both plant  groups were effective at removing Pb from the water at
29    sublethal levels.
30          BCFs for Pb are reported  for various invertebrates in Table 8-2.4.2. BCFs for
31    freshwater snails were 738 for a  28-day exposure (Spehar et al., 1978) and 1,700 for a
     December 2005                        8-167    DRAFT-DO NOT QUOTE OR CITE

-------
                  Table 8-2.4.1.  Bioconcentration Factors for Aquatic Plants
BCF
840 to 2700
(measured digestion)
1150 to 3560
(measured solution)
16,000 to 20,000
Species
Duckweed (Lemna minor)
Duckweed
Alga (Cladophora glomerata)
Test Conditions
21 days, Pb-nitrate
21 days, Pb-nitrate
not specified
Reference
Rahmani and
Steinberg (1999)
Rahmani and
Sternberg (1999)
Keeney et al. (1976)
               Table 8-2.4.2. Bioconcentration Factors for Aquatic Invertebrates
       BCF
Species
Test Conditions
Reference
       738
       1700
       499
       1120
       1930
       3670
Snail (Physa Integra)
Snail (Lymnaea palustris)
Caddis fly (Brachycentrus sp.)
Stonefly (Pteronarcys dorsata)
Scud (Hyalella azteca)
28 days, Pb-nitrate
120 days, Pb-nitrate
28 days, Pb-nitrate
28 days, Pb-nitrate
4 days, Pb-chloride
Midge larvae (Chironomus riparius)   28 days
Speharetal. (1978)
Borgmann et al. (1978)
Speharetal. (1978)
Speharetal. (1978)
MacLean et al. (1996)
Timmermans et al. (1992)
 1    120-day exposure (Borgmann et al., 1978).  Other reported values for invertebrates
 2    included a BCF of 1930 for the scud during a 4-day exposure (MacLean et al., 1996),
 3    and BCFs of 499 and 1120 for the caddis fly and stonefly, respectively, in 28-day
 4    exposures (Spehar et al., 1978).  In a 28-day exposure, midge larvae were reported with
 5    a BCF of 3670 (Timmermans et al., 1992).
 6          BCFs for freshwater fish were 42 and 45 for brook trout and bluegill, respectively
 7    (Holcombe et al.,  1976; Atchison et al., 1977).  Although no BCFs have been reported for
 8    amphibians, Pb-nitrate was reported to accumulate mainly in the ventral skin and in the
 9    kidneys of frogs (Vogiatzis and Loumbourdis, 1999).
10
     December 2005
                             8-168    DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.4.3  Resistance Mechanisms
 2    8.2.4.3.1   Detoxification Mechanisms
 3          Detoxification includes the biological processes by which the toxic qualities, or
 4    the probability and/or severity of harmful effects, of a poison or toxin are reduced by the
 5    organism.  In the case of heavy metals, this process frequently involves the sequestration
 6    of the metal, rendering it metabolically inactive. Recent research into heavy metal
 7    detoxification in aquatic biota has focused on several physiological and biochemical
 8    mechanisms for detoxifying Pb. This section examines these mechanisms and the ability
 9    of plants, protists, invertebrates, and fish to mitigate Pb toxicity.
10
11    Plants and Protists
12          Deng et al. (2004) studied the uptake and translocation of Pb in wetland plant
13    species surviving in contaminated sites. They found that all plants tended to sequester
14    significantly larger amounts of Pb in their roots than in their shoots. Deng et al. (2004)
15    calculated  a translocation factor (TF), the amount of Pb found in the shoots divided by
16    the amount of Pb found in the root system, and found that TFs ranged from 0.02 to 0.80.
17    Concentrations of Pb in shoots were maintained at low levels and varied within a narrow
18    range. Deng et al. (2004) observed that plants grown in Pb-contaminated sites usually
19    contained higher concentrations than the 27 mg/kg toxicity threshold established for
20    plants by Beckett and Davis (1977). Some of the wetland plants examined by Deng et al.
21    (2004) also accumulated high concentrations of metals in shoot tissues; however, these
22    metals were  assumed to be detoxified (metabolically unavailable), as no toxic response to
23    these elevated concentrations was observed.  Deng et al. (2004) suggested that this ability
24    is likely related to discrete internal metal detoxification tolerance mechanisms.
25          Phytochelatins are thiol-containing intracellular metal-binding polypeptides that
26    are produced by plants and protists in response to excessive uptake of heavy  metals
27    (Zenk, 1996). Phytochelatins are synthesized by the enzyme phytochelatin synthase that
28    is activated by the presence of metal ions and uses glutathione as a substrate. When
29    phytochelatins are synthesized in sufficient amounts to chelate the metal ion, the enzyme
30    is deactivated (Morelli and Scarano, 2001).
      December 2005                        8-169    DRAFT-DO NOT QUOTE OR CITE

-------
 1          Morelli and Scarano (2001) studied phytochelatin synthesis and stability in the
 2    marine diatom Phaeodactylum tricornutum in the presence of Pb. They found that when
 3    metal exposure was alleviated, significant cellular Pb-phytochelatin complex content was
 4    lost.  Their findings support a hypothesis of vacuolarization proposed for higher plants
 5    (Zenk,  1996), in which metal-phytochelatin complexes are actively transported from the
 6    cytosol to the vacuole, where they undergo rapid turnover. Zenk (1996) suggested that
 7    the complex dissociates, and the metal-free peptide is subsequently degraded. Morelli
 8    and Scarano (2001) proposed concomitant occurrence of phytochelatin synthesis and
 9    release during metal exposure, as a coincident detoxification mechanism in P.
10    tricornutum.
11
12    A quatic Inverte brates
13          Like plants and protists, aquatic animals detoxify Pb by preventing it from being
14    metabolically available, though their mechanisms for doing so vary.  Invertebrates use
15    lysosomal-vacuolar systems to sequester and process Pb within glandular cells
16    (Giamberini and Pihan,  1996). They also accumulate Pb as deposits on and within
17    skeletal tissue (Knowlton et al., 1983; Anderson et al., 1997; Boisson et al., 2002), and
18    some can efficiently excrete Pb (Vogt and Quinitio,  1994; Prasuna et al., 1996).
19          Boisson et al. (2002) used radiotracers to evaluate the transfer of Pb into the food
20    pathway of the starfish Asterias rubens as well as its distribution and retention in various
21    body compartments. Boisson et al. (2002) monitored Pb elimination after a single
22    feeding of Pb-contaminated molluscs and found that Pb was sequestered and retained in
23    the skeleton of the starfish, preventing it from being metabolically available in other
24    tissues. Elimination (as percent retention in the  skeleton) was found to follow an
25    exponential time course. Elimination was rapid at first, but slowed after 1 week, and
26    eventually stabilized, implying an infinite biological half-life for firmly bound Pb.
27    Results of radiotracer tracking suggest that Pb migrates within the body wall from the
28    organic matrix to the calcified skeleton.  From there, the metal is either absorbed directly
29    or adsorbed on newly-produced ossicles (small calcareous skeletal structures), where it is
30    efficiently retained as mineral deposition and is not metabolically active (Boisson  et al.,
31    2002).

      December 2005                         8-170    DRAFT-DO NOT QUOTE OR CITE

-------
 1          AbdAllah and Moustafa (2002) studied the Pb storage capability of organs in the
 2    marine snail Nerita saxtilis. Enlarged electron-dense vesicles and many granules were
 3    observed in digestive cells of these snails and are suggested to be the site of storage of
 4    detoxified metals.  N. saxtilis were found to be capable of concentrating Pb up to 50 times
 5    that of surrounding marine water without exhibiting signs of histopathologic changes.
 6    This ability has been attributed to chelation with various biochemical compounds, such as
 7    thionine (forming metallothionine) (Rainbow,  1996), or complexation with carbonate,
 8    forming lipofuchsin (AbdAllah and Moustafa, 2002).  Granules observed in lysosomal
 9    residual  bodies were presumed to be the result of Pb accumulation.  The presence of large
10    vacuoles and residual bodies were indicative of the fragmentation phase of digestion,
11    suggesting that Pb was also processed chemically in the digestive cells.
12          The podocyte cells of the pericardial gland of bivalves  are involved in the
13    ultrafiltration of the hemolymph (Giamberini and Pihan, 1996).  A microanalytical study
14    of the podocytes in Dreissena polymorpha exposed to Pb revealed lysosomal-vacuolar
15    storage/processing similar to that in the digestive cells of Nerita saxtilis.  The lysosome is
16    thought to be the target organelle for trace metal accumulation in various organs of
17    bivalves (Giamberini and Pihan,  1996). Epithelial secretion is the principal detoxification
18    mechanism of the tiger prawn Penaeus monodon. Vogt and Quinitio (1994) found that
19    Pb granules tended to accumulate in the epithelial cells of the  antennal gland (the organ
20    of excretion) of juveniles exposed for 5 and 10 days to waterborne Pb.  The metal is
21    deposited in vacuoles belonging to the lysosomal system. Continued deposition leads to
22    the formation of electron-dense granules.  Mature granules are released from the cells by
23    apocrine secretion into the lumen of the gland, and presumably excreted through the
24    nephridopore (i.e., the opening of the antennal gland). Apocrine secretion is
25    predominant, so that as granules form, they are kept at low levels. Excretion was also
26    found to be a primary and efficient detoxification mechanism  in the shrimp Chrissia halyi
27    (Prasunaetal., 1996).
28          Crayfish exposed to Pb have been shown to concentrate the metal in their
29    exoskeleton and exuvia through adsorption processes. More than 80% of Pb found in
30    exposed crayfish has been found in exoskeletons (Knowlton et al., 1983; Anderson et al.,
31    1997). Following exposure, clearance is most dramatic from the exoskeleton. The result

      December 2005                        8-171    DRAFT-DO NOT QUOTE OR CITE

-------
 1    of a 3-week Pb-clearance study with red swamp crayfish Procambarus clarkia, following
 2    a 7-week exposure to 150 jig/L Pb, showed an 87% clearance from the exoskeleton due,
 3    in part, to molting. Other organs or tissues that take up significant amounts of Pb include
 4    the gills, hepatopancreas, muscle, and hemolymph, in decreasing order. These parts
 5    cleared >50% of accumulated Pb over the 3-week clearance period, with the exception of
 6    the hepatopancreas. The hepatopancreas is the organ of metal storage and detoxification,
 7    although the molecular mechanisms of metal balance in crayfish have yet to be
 8    extensively investigated (Anderson et al.,  1997).
 9
10    Fish
11          Most fish use mucus as a first line of defense against heavy metals (Coello and
12    Khan, 1996).  In fish, some epithelia are covered with extracellular mucus secreted from
13    specialized cells.  Mucus contains glycoproteins, and composition varies among species.
14    Mucosal glycoproteins chelate Pb, and settle, removing the metal  from the water column.
15    Fish may secrete large amounts of mucus when they come into contact with potential
16    chemical and biochemical threats. Coello and Khan (1996) investigated the role of
17    externally added fish mucus and scales in  accumulating Pb from water, and the
18    relationship of these with the toxicity of Pb in fmgerlings of green sunfish, goldfish and
19    largemouth bass.  The authors compared trials in which fish scales from black sea bass
20    (Centropristis striatd) and flounder (Pseudopleuronectes americanus) and mucus from
21    largemouth bass were added to green sunfish, goldfish, and largemouth bass test systems
22    and to reference test systems.  On exposure to Pb, fish  immediately started secreting
23    mucus from epidermal cells in various parts of the body.  Metallic Pb stimulated
24    filamentous secretion, mostly  from the ventrolateral areas of the gills, while Pb-nitrate
25    stimulated diffuse molecular mucus secretion from all  over the body.  The addition of
26    largemouth bass mucus significantly increased the LT50 (the time  to kill 50%) for green
27    sunfish and goldfish exposed to 250 mg/L of Pb-nitrate.  In contrast, Tao et al. (2000)
28    found that mucus reduced the overall bioavailability of Pb to fish but that the reduction
29    was insignificant. Coello and Khan (1996) found that  scales were more significant in
30    reducing LTso than mucous. Fish scales can accumulate high concentrations of metals,
31    including Pb, through chelation with keratin.  Scales were shown to buffer the pH of
      December 2005                       8-172    DRAFT-DO NOT QUOTE OR CITE

-------
 1    Pb-nitrate in solution and remove Pb from water after which they settled out of the water
 2    column. Addition of scales to test water made all species (green sunfish, goldfish, and
 3    largemouth bass) more tolerant of Pb.
 4
 5    Summary of Detoxi fiction Processes
 6          Mechanisms of detoxification vary among aquatic biota and include processes
 7    such as translocation, excretion, chelation, adsorption, vacuolar storage, and deposition.
 8    Lead detoxification has not been studied extensively in aquatic organisms, but existing
 9    results indicate the following:
10    •  Protists and plants produce intracellular polypeptides that form complexes with Pb
11       (Zenk,  1996; Morelli and Scarano, 2001).
12    •  Macrophytes and wetland plants that thrive in Pb-contaminated regions have
13       developed translocation strategies for tolerance and detoxification (Knowlton et al.,
14       1983; Deng et al., 2004).
15    •  Some starfish (asteroids) sequester the metal via mineral deposition into the
16       exoskeleton (Boisson et al., 2002).
17    •  Species of mollusc employ lysosomal-vacuolar systems that store and chemically
18       process Pb in the cells of their digestive and pericardial glands (Giamberini and
19       Pihan,  1996; AbdAllah and Moustafa, 2002).
20    •  Decapods can efficiently excrete Pb (Vogt and Quinitio,  1994; Giamberini and Pihan,
21       1996) and sequester metal through adsorption to the exoskeleton (Knowlton et al.,
22       1983).
23    •  Fish scales and mucous chelate Pb in the water column, and potentially reduce
24       visceral exposure.
25
26    8.2.4.3.2   Avoidance Response
27          Avoidance is the evasion of a perceived threat. Recent research into heavy metal
28    avoidance  in aquatic organisms has looked at dose-response relationships as well as the
29    effects of coincident environmental factors.  Preference/avoidance response to Pb has  not
30    been extensively studied in aquatic organisms.  In particular, data for aquatic
31    invertebrates is lacking. Using recent literature, this  section examines preference-
      December 2005                        8-173    DRAFT-DO NOT QUOTE OR CITE

-------
 1    avoidance responses of invertebrates and fish to Pb and some other environmental
 2    gradients.
 3
 4    Aquatic Invertebrates
 5          Only one study was identified on avoidance response in aquatic invertebrates.
 6    Lefcort et al. (2004) studied the avoidance behavior of the aquatic pulmonate snail
 7    Physella columbiana from a pond that had been polluted with heavy metals for over
 8    120 years. In a Y-maze test, first generation P. columbiana from the contaminated site
 9    avoided Pb at 9283 |ig/L (p < 0.05) and moved toward Pb at 6255 |ig/L (p < 0.05). It is
10    thought that attraction to Pb at certain elevated concentrations is related to Pb neuron-
11    stimulating properties (Lefcort et al., 2004). These results are consistent with those from
12    similar studies. Control snails from reference  sites, and first and second generation snails
13    from contaminated sites were capable of detecting and avoiding heavy metals, although
14    the first generation was better than the second generation, and the second was better than
15    the controls at doing so. This suggests that detection and avoidance of Pb is both genetic
16    and environmentally based for P. columbiana.  Lefcort et al. (2004) observed heightened
17    sensitivity to, and avoidance of, heavy metals by the snails when metals where present in
18    combination.
19
20    Aquatic Vertebrates
21          Steele et al. (1989) studied the preference-avoidance response of bullfrog  (Rana
22    catesbeiana) to plumes of Pb-contaminated water following 144 h exposure to 0  to
23    1000 jig Pb/L. In this laboratory experiment, tadpoles were exposed to an influx of
24    1000 |ig/L at five different infusion rates (i.e., volumes per unit time into the test system).
25    Experiments were videotaped and location data from the tank were used to assess
26    response. No significant differences were seen in preference-avoidance responses to
27    Pb either nonexposed or previously exposed animals.  In a similar subsequent study,
28    Steele et al.  (1991) studied preference-avoidance response to Pb in American toad
29    (Bufo mericanus) using the same exposure range (0 to 1000 |ig/L).  B. americanus did not
30    significantly avoid Pb,  and behavioral stress responses were not observed. The results do
31    not indicate whether the tadpoles were capable of perceiving the contaminant.  Lack of

      December 2005                         8-174   DRAFT-DO NOT QUOTE OR CITE

-------
 1    avoidance may indicate insufficient perception or the lack of physiological stress (Steele
 2    etal., 1991).
 3          The olfactory system in fish is involved in their forming avoidance response to
 4    heavy metals (Brown et al., 1982;  Svecevicius, 1991).  It is generally thought that
 5    behavioral avoidance of contaminants may be a cause of reduced fish populations in
 6    some water bodies, because of disturbances in migration and distribution patterns
 7    (Svecevicius, 2001). Unfortunately, avoidance of Pb by fish has not been studied as
 8    extensively as for other heavy metals (Woodward et al., 1995).
 9          Woodward et al. (1995) studied metal mixture avoidance response in brown trout
10    (Salmo tmttd), as well as the added effects of acidification.  A 1-fold (Ix) mixture
11    contained 1.1 |ig/L Cd, 12 |ig/L  Cu,  55 |ig/L Zn,  and 3.2 |ig/L Pb (all metals were in the
12    form of chlorides).  Avoidance was quantified as time spent in test water, trip time to test
13    water, and number of trips. Brown trout avoided the Ix mixture as well as the 0.5x,  2x,
14    4x, and lOx mixtures, but not the O.lx  mixture. Reduced avoidance was observed at
15    higher concentrations (4x and lOx).  The authors proposed that the reduced avoidance
16    response was due to impaired perception due to injury.  These responses are typical of
17    other fish species to individual metals of similar concentrations (Woodward et al.,  1995).
18    This study does not conclusively indicate which of the metals in the mixture may be
19    causing the avoidance response.  However, given the neurotoxic effects of Pb, impaired
20    perception is a likely response of Pb-exposed fish.
21          When test water was reduced in pH from 8 to 7, 6 to 5, brown trout avoidance
22    increased, but with no significant difference between metal treatments and controls.
23    However, in the 1 x  metal mixture treatment, brown trout made fewer trips  into the test
24    water chamber at the lower pHs  (Woodward et al., 1995).  This response may be related
25    to an increased abundance of Pb cations at lower pH values in the test system.
26          Scherer and McNicol  (1998) investigated the preference-avoidance responses  of
27    lake whitefish (Coregonus clupeaformis) to overlapping gradients of light and Pb.
28    Whitefish were found to prefer shade in untreated water. Lead concentrations under
29    illumination ranged from 0 to 1000 |ig/L, and from 0 to 54,000 |ig/L in the shade.  Under
30    uniform illumination, Pb was avoided at concentrations above 10 |ig/L, but avoidance
31    behavior lacked a dose-dependent increase over concentrations ranging from 10 to

      December 2005                        8-175    DRAFT-DO NOT QUOTE OR CITE

-------
 1    1000 |ig/L Pb.  Avoidance in shaded areas was strongly suppressed, and whitefish only
 2    avoided Pb at concentrations at or above 32,000 |ig/L.
 3          In summary, of those aquatic organisms studied, some are quite adept at avoiding
 4    Pb in aquatic systems, while others seem incapable of detecting its presence. Snails have
 5    been shown to be sensitive to Pb and to avoid it at high concentrations.  Conversely,
 6    anuran (frog and toad) species lack an avoidance response to the metal. Fish avoidance
 7    of many chemical toxicants has been well established, and it is a dominant sublethal
 8    response in polluted waters (Svecevicius, 2001).  However, no studies have been located
 9    specifically  examining avoidance behavior for Pb in fish. Environmental gradients, such
10    as light and  pH, can alter preference-avoidance responses.
11
12    8.2.4.4 Physiological Effects of Lead
13          This section presents a review of the physiological effects and functional growth
14    responses associated with the exposure of aquatic biota to Pb. Physiological effects of Pb
15    on aquatic biota can occur at the biochemical, cellular, and tissue levels of organization
16    and  include  inhibition of heme formation, adverse effects to blood chemistry, and
17    decreases in enzyme levels. Functional growth responses resulting from Pb exposure
18    include changes in growth patterns, gill binding affinities, and absorption rates.
19
20    Biochemical Effects
21          Lead was observed to have a gender-selective effect on brain endocannabinoid
22    (eCB) (e.g., 2-arachidonylglycerol [2-AG] and TV-arachidonylethanolamine [AEA]) levels
23    in fathead minnow Pimephalespromelas (Rademacher et al., 2005).  Cannabinoids, such
24    as eCB, influence locomotor activity in organisms. Increased levels of cannabinoids have
25    been shown to stimulate locomotor activity and decreased levels slow locomotor activity
26    (Safiudo-Pefia et al., 2000). Male and female fathead minnows were exposed to 0 and
27    1000 |ig/L of Pb.  Female minnows in the control group contained significantly higher
28    levels of AEA and 2-AG compared to males. At a concentration of 1000 |ig/L, this
29    pattern reversed, with males showing significantly higher levels of AEA in the brain than
30    females (Rademacher et al., 2005). After 14-days exposure to the  1000 jig/L treatment,
      December 2005                       8-176    DRAFT-DO NOT QUOTE OR CITE

-------
 1    significantly higher levels of 2-AG were found in male fathead minnows, but no effect on
 2    2-AG levels in females was observed (Rademacher et al., 2005).
 3         Lead acetate slightly inhibited 7-ethoxyresorufm-o-deethylase (7-EROD) activity
 4    in Gammaruspulex exposed for up to 96 h to a single toxicant concentration (ECso)
 5    (Kutlu and Susuz, 2004).  The exact concentration used in the study was not reported.
 6    The EROD enzyme is required to catalyze the conjugation and detoxification of toxic
 7    molecules and has been proposed as a biomarker for contaminant exposure. The authors
 8    believe more detailed studies are required to confirm EROD as a biomarker for Pb
 9    exposure. The enzyme group alanine transferases (ALT) has been suggested as a
10    bioindicator/biomarker of Pb stress (Blasco and Puppo, 1999). A negative correlation
11    was observed between Pb accumulation and ALT concentrations in the gills and soft
12    body of Ruditapes philippinarum exposed to 350 to 700  |ig/L of Pb for 7 days (Blasco
13    and Puppo, 1999).
14          Studies have identified ALAD in fish and amphibians as a useful indicator of Pb
15    exposure (Gill et al., 1991; Nakagawa et al., 1995a,b). ALAD catalyzes the formation of
16    hemoglobin and early  steps in the synthesis of protoporphyrin (Gill et al., 1991;
17    Nakagawa et al., 1995b).  The absence of an inhibitory effect on this enzyme following
18    exposure to cadmium, copper, zinc, and mercury suggests that this enzyme reacts
19    specifically to Pb  (Johansson-Sjobeck and Larsson, 1979; Gill et al., 1991). A 0%
20    decrease in ALAD activity was reported in common carp (Cyprinus carpio) exposed to a
21    Pb concentration of 10 |ig/L for 20 days (Nakagawa et al., 1995b).  The recovery the
22    ALAD activity after exposure to Pb has also been examined in carp (Nakagawa et al.,
23    1995a). After 2-week exposure to 200 |ig/L, ALAD activity decreased to approximately
24    25% of value reported for controls (Nakagawa et al.,  1995a). Fish removed from the test
25    concentration after 2 weeks and placed in a Pb-free environment recovered slightly, but
26    ALAD activity was only 50% of the controls even after 4 weeks (Nakagawa et al.,
27    1995a). Vogiatzis and Loumbourdis (1999) exposed the frog (Rana ridibunda) to a Pb
28    concentration of 14,000 |ig/L over 30 days and a 90% decrease in ALAD activity was
29    observed in the frogs.
     December 2005                       8-177    DRAFT-DO NOT QUOTE OR CITE

-------
 1   Blood Chemistry
 2         Numerous studies have examined the effects of Pb exposure on blood chemistry in
 3   aquatic biota.  These studies have primarily used fish in acute and chronic exposures to
 4   Pb concentrations ranging from 100 to 10,000 |ig/L. Decreased erythrocyte, hemoglobin,
 5   and hemocrit levels were observed in rosy barb (Barbuspuntius) during an 8-week
 6   exposure to 126 |ig/L of Pb-nitrate (Gill et al., 1991).
 7         No difference was found in red blood cell counts and blood hemoglobin in yellow
 8   eels (Anguilla cmguilld) exposed to 0 and 300 jig/L of Pb for 30 days (Santos and Hall,
 9   1990).  The number of white blood cells, in the form of lymphocytes, increased in the
10   exposed eels.  The authors concluded this demonstrates the lasting action of Pb as a
11   toxicant on the immune system (Santos and Hall, 1990).  Significant decreases in red
12   blood cell counts and volume was reported in blue tilapia (Oreochromis aureus) exposed
13   to Pb-chloride at a concentration of 10,000 |ig/L for 1 week (Allen,  1993).
14         Blood components, such as plasma glucose, total plasma protein, and total plasma
15   cholesterol, were unaffected in yellow eels exposed to 300 jig/L of Pb for 30 days
16   (Santos and Hall, 1990). Effects on plasma chemistry were  observed in Oreochromis
17   mossambicus exposed to 0, 18,000, 24,000, and 33,000 |ig/L of Pb (Ruparelia et al.,
18   1989).  Significant decreases in plasma glucose (hypoglycemic levels) were reported at
19   concentrations of 24,000 and 33,000 //g/L after  14 and 21 days of exposure, and at
20   18,000  |ig/L after 21 days of exposure (Ruparelia et al., 1989). Plasma cholesterol levels
21   dropped significantly in comparison to controls after 14 days of exposure to 33,000 |ig/L
22   and in all test concentrations after 21 days of exposure (Ruparelia et al., 1989).
23   Similarly, concentrations of blood serum protein, albumin, and globulin were identified
24   as bioindicators of Pb stress in carp (Cyprinus carpid) exposed to Pb-nitrates at
25   concentrations of 800 and 8000 |ig/L (Gopal et al., 1997).
26
27   Tissues
28         In fish, the gills serve as an active site for ion uptake. Recent studies have
29   examined the competition between cations for binding sites  at the fish gill (e.g., Ca2+,
30   Mg2+, Na+, H+, Pb2+) (MacDonald et al., 2002; Rogers and Wood, 2003, 2004). Studies
31   suggest that Pb2+ is an antagonist of Ca2+ uptake (Rogers and Wood, 2003, 2004).

     December 2005                        8-178    DRAFT-DO NOT QUOTE OR CITE

-------
 1    MacDonald et al. (2002) proposed a gill-Pb binding model that assumes Pb2+ has a
 2    > 100 times greater affinity for binding sites at the fish gill than other cations. More
 3    toxicity studies are required to quantify critical Pb burdens that could be used as
 4    indicators of Pb toxicity (Niyogi and Wood, 2003).
 5
 6    Growth Responses
 1          A negative linear relationship was observed in the marine gastropod abalone
 8    (Haliotis) between shell length and muscle Pb concentrations (Arai et al., 2002).
 9    Abalones were collected from two sites along the Japanese coast. Haliotis discus hannai
10    were collected from along the coast at Onagawa; Haliotis discus were collected from
11    along the coast at Amatsu Kominato. The authors did not report significant differences
12    between the two sampling sites. From samples collected at Onagawa, Pb concentrations
13    of 0.03 and 0.01 jig/g were associated with abalone shell lengths of 7.7 cm (3 years old)
14    and 12.3 cm (6 years old), respectively.  From samples collected at Amatsu Kominato,
15    Pb concentrations of 0.09 and 0.01  jig/g were associated with abalone shell lengths of
16    3.9 cm (0 years old) and 15.3 cm (8 years old), respectively (Arai et al., 2002).  The
17    authors theorized that young abalones, experiencing rapid growth, do not discriminate
18    between the uptake of essential and nonessential metals.  However, as abalones grow
19    larger and their rate of growth decreases, they increasingly favor the uptake of essential
20    metals over nonessential metals.  This is demonstrated by the relatively consistent
21    concentrations of Cu, Mn, and Zn that were reported for the abalone samples (Arai et al.,
22    2002).
23
24    Other Physiological Effects
25          Increased levels of Pb  in water were found to increase fish production of mucus:
26    excess mucus coagulates were observed over the entire body of fishes. Buildup was
27    particularly high around the gills, and in the worst cases, interfered with respiration and
28    resulted in death by anoxia (Aronson, 1971; National Research Council of Canada.,
29    1973).
30
      December 2005                        8-179    DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.4.5  Factors That Modify Organism Response to Lead
 2          A great deal of research has been undertaken recently to better understand the
 3    factors that modify aquatic organism response to Pb. The driving force behind this
 4    research is the development of the BLM approach to AWQC development. A discussion
 5    of research on the many factors that can modify aquatic organism response to Pb is
 6    provided in this section.
 7
 8    8.2.4.5.1  Organism Age and Size Influence on Lead Uptake and Response
 9          It is generally accepted that Pb accumulation in living organisms is controlled, in
10    part, by metabolic rates (Farkas et al., 2003). Metabolic rates are, in-turn, controlled by
11    the physiological conditions of an organism, including such factors as size, age, point in
12    reproductive cycle, nutrition,  and overall health.  Of these physiological conditions,  size
13    and age  are the most commonly investigated in relation to heavy metal uptake.  This
14    section reviews recent research focusing on  relationships between body size, age, and Pb
15    accumulation in aquatic invertebrates and fish.
16
17    Invertebrates
18          MacLean et al. (1996)  investigated bioaccumulation kinetics and toxicity of Pb in
19    the amphipod Hyalella azteca. Their results indicated that body size did not greatly
20    influence Pb accumulation in  H. azteca exposed to 50 or 100 |ig/L of PbC^ for 4 days.
21    Canli and Furness (1993) found similar results in the Norway lobster Nephrops
22    norvegicus exposed to 100 |ig/L of Pb(NO3)2 for 30 days.  No significant sex- or size-
23    related differences were found in concentrations of Pb in the tissue. The highest tissue
24    burden was found in the carapaces (42%). Several studies have determined that Pb can
25    bind to the exoskeleton of invertebrates and sometimes dominate the total Pb
26    accumulated (Knowlton et al., 1983). This adsorption of Pb  to the outer surface of
27    invertebrates can result in strong negative relationships for whole-body Pb concentration
28    as a function of body mass (i.e., concentrations decrease rapidly with  increased body size
29    and then stabilize) (MacLean  et al., 1996).
30          Drava et al. (2004) investigated Pb concentrations in the muscle of red shrimp
31    Aristeus antennatus from the  northwest Mediterranean. Lead concentrations ranged from


      December 2005                        8-180     DRAFT-DO NOT QUOTE OR CITE

-------
 1    0.04 to 0.31 |ig/g dw. No significant relationships between size and Pb concentration in
 2    A. antennatus were found, and concentrations were not related to reproductive status.
 3          Arai et al. (2002) analyzed abalones (Haliotis) at various life stages from coastal
 4    regions of Japan. They investigated growth effects on the uptake and elimination of Pb.
 5    Results indicated a significant negative linear relationship between age, shell length and
 6    Pb concentrations in muscle tissue.  The relationship was consistent despite habitat
 7    variations in Pb concentrations between the study sites, suggesting that Pb concentrations
 8    changed with growth in the muscle tissue of test specimens and implying that abalone can
 9    mitigate Pb exposure as they age.
10
11    Fish
12          Douben (1989) investigated the effects of body size and age on Pb body burden in
13    the stone loach (Noemacheilus barbatulus L.). Fish were caught during two consecutive
14    springs from three Derbyshire rivers.  Results indicated that Pb burden increased slightly
15    with age. Similarly, Kock et al. (1996) found that concentrations of Pb in the liver and
16    kidneys of Arctic char (Salvelinus alpinus) taken from oligotrophic alpine lakes were
17    positively correlated with age. It has been suggested that fish are not able to eliminate Pb
18    completely, and that this  leads to a stepwise accumulation from year to year (Kock et al.,
19    1996).  In contrast, Farkas et al. (2003) found a negative relationship between Pb
20    concentrations and muscle and gill Pb concentrations in the freshwater fish Abramis
21    brama. Fish were taken from a low-contaminated site and contained between 0.44 and
22    3.24 |ig/g Pb dw. Negative correlations between metal concentration and fish size in
23    low-contaminated waters likely results from variations in feeding rates associated with
24    developmental stages.  This  hypothesis is consistent with the fact that in low-
25    contaminated waters, feeding is the main route of uptake and feeding rates decrease with
26    development in fish (Farkas et al., 2003).
27          In summary, relationships between  age, size, and Pb body burden in aquatic
28    invertebrates and fish are interspecifically  variable and depend on many  environment-
29    related variables (e.g., exposure) (Farkas et al., 2003).
30
      December 2005                        8-181    DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.4.5.2    Genetics
 2          There are few studies documenting the effects of Pb on organismal and population
 3    genetics, although rapid advances in biotechnology have prompted recent research in this
 4    area (Beaty et al., 1998). There are two principal effects that sublethal exposure to a
 5    contaminant can have on the genetics of an organism and/or population: (1) a
 6    contaminant may influence selection by selecting for certain phenotypes that enable
 7    populations to better cope with the chemical; or (2) a contaminant can be genotoxic,
 8    meaning it can produce alterations in nucleic acids at sublethal exposure concentrations,
 9    resulting in changes in hereditary characteristics or DNA inactivation (Shugart, 1995).
10    Laboratory studies have shown that exposure to Pb2+ at 10 mg/mL of blood produces
11    chromosomal aberrations (i.e., deviations in the normal structure or number of
12    chromosomes) in some organisms (Cestari et al., 2004). Effects of genotoxicity and
13    toxin-induced selection do not preclude one another, and may act together on exposed
14    populations. This section reviews Pb genotoxicity and the effects of Pb-induced selection
15    in aquatic populations.
16
17    Selection
18          Evidence for genetic selection in the natural environment has been observed in
19    some aquatic populations exposed to metals (Rand et al., 1995; Beaty et al., 1998; Duan
20    et al., 2000; Kim et al., 2003). Because tolerant individuals have a selective advantage
21    over vulnerable individuals in polluted environments, the frequency of tolerance genes
22    will increase in exposed populations over time (Beaty et al., 1998).  Several studies have
23    shown that heavy metals can alter population gene pools in aquatic invertebrates. These
24    changes have resulted in decreased genetic diversity and are thought to  be a potential
25    source of population instability (Duan et al., 2000; Kim et al., 2003).
26          Kim et al. (2003) investigated genetic differences and population structuring in the
27    gastropod Littorina brevicula from heavy-metal polluted and unpolluted environments.
28    Organisms from polluted sites contained a mean of 1.76 |ig/g Pb, while organisms from
29    unpolluted sites contained 0.33 |ig/g Pb. They found significant differences in
30    haplotypes between the test groups and allelic diversity was significantly lower among
31    L. brevicula from polluted regions. In contrast, Yap et al. (2004) performed a similar

      December 2005                        8-182    DRAFT-DO NOT QUOTE OR CITE

-------
 1    experiment with the green-lipped mussel Perna viridis; they found that mussels from
 2    contaminated sites containing between 4 and 10 jig/g Pb, as well as other heavy metals,
 3    exhibited a higher percentage of polymorphic loci and excess heterozygosity compared to
 4    those from uncontaminated sites. The higher level of genetic diversity was attributed to
 5    greater environmental heterogeneity (i.e., variation due to pollution gradients) in
 6    contaminated sites (Yap et al., 2004).
 7          Duan et al. (2000) investigated amphipod (Hyalella aztecd) selective mortality and
 8    genetic structure following acute exposure to Pb (5.47 mg/L Pb(NC>2)2) as well as
 9    exposure to other heavy metals.  They found that genetic differentiation consistently
10    increased among survivors from the original population, supporting the hypothesis that
11    heavy metals, including Pb, have the potential to alter the gene pools of aquatic
12    organisms.
13
14    Genotoxicity
15          Low-level (50 |ig/L) Pb exposure in water over 4 weeks resulted in DNA strand
16    breakage in the freshwater mussel Anodonta grandis (Black et al., 1996), although higher
17    concentrations  (up to 5000 |ig/L) did not result in significant breakage by the end of the
18    study period. These results suggest that a threshold effect for DNA damage and repair
19    exists, where DNA repair only occurs once a certain body exposure level has been
20    reached. More recently, Cestari et al. (2004) observed similar results in neotropical fish
21    (Hoplias malabaricus) that were fed Pb-contaminated food over 18,  41, and 64 days.
22    Lead body burdens in H. malabaricus were approximately 21 jig Pb 2+/g. Results
23    indicated that exposure to Pb significantly increased the frequency of chromosomal
24    aberrations and DNA damage in kidney cell cultures, although when assessed at the end
25    of the longer exposure  periods, aberrations were less common.
26
27    8.2.4.5.3  Environmental Biological Factors
28          Environmental factors that are biological in origin can alter the availability, uptake
29    and toxicity of Pb to aquatic organisms.  These factors can be grouped into living and
30    non-living constituents. For example, living organisms may sequester Pb from the water
31    column, reducing the availability and toxicity of the metal in the water column to other


      December 2005                        8-183    DRAFT-DO NOT QUOTE OR CITE

-------
 1    biota, thus reducing potential toxic effects in other organisms.  Non-living organic
 2    material (e.g., components of sloughed-off scales, mucus, carcasses, and other
 3    decomposing, humic material) can similarly combine with Pb from the water column,
 4    rendering it unavailable. This section will review the literature on biological
 5    environmental factors and their influence on the bioavailability, uptake, and toxicity
 6    ofPb.
 7          Van Hattum et al. (1996) studied the influence of abiotic variables, including
 8    DOC on Pb concentrations in freshwater isopods (Proasellus meridianus and Asellus
 9    aquaticus). They found that BCFs were significantly negatively correlated with DOC
10    concentrations.  Thus, as DOC concentrations increased, BCFs decreased in P.
11    meridianus and A. aquaticus, indicating that DOC acts to inhibit the availability of Pb
12    to these isopods.
13          Kruatrachue et al. (2002) investigated the combined effects of Pb and humic acid
14    on total chlorophyll content, growth rate, multiplication rate, and Pb uptake of common
15    duckweed. When humic acid was added to the Pb-nitrate test solutions (50, 100, and
16    200 mg Pb(NO3)2/ L), toxicity of Pb to L. minor was decreased.  The addition of humic
17    acid to the Pb-nitrate solution increased the pH. The authors suggested that there was a
18    proton dissociation from the carboxyl group in the humic acid that complexed with Pb,
19    resulting in a decrease in free Pb ions available to the plant.
20          Schwartz et al. (2004) collected natural organic  matter (NOM) from several
21    aquatic sites across Canada and investigated the effects of NOM on Pb toxicity in
22    rainbow trout (Oncorhynchus mykiss).  They also looked at toxicity effects as they related
23    to the optical properties of the various NOM samples.  The results showed that NOM in
24    test water almost always increased LTso and that optically dark NOM tended to decrease
25    Pb toxicity more than did optically light NOM in rainbow trout.
26          In summary, non-living constituents of biological origin in the environment have
27    been shown to reduce Pb availability and, therefore, toxicity in some aquatic organisms.
28    It is generally thought that this occurs through complexation, or chelation processes that
29    take place  in the water column.
30
      December 2005                        8-184    DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.4.5.4   Physical Environmental Factors
 2          This section reviews the literature on physical environmental factors and their
 3    influence on the bioavailability, uptake, and toxicity of Pb in aquatic organisms. These
 4    factors are discussed with regard to their influence individually and in combination.
 5          Studies generally agree that as pH increases, the toxicity of Pb decreases (Home
 6    and Dunson, 1995b; MacDonald et al., 2002). As pH decreases, Pb becomes more
 7    soluble and more readily bioavailable to aquatic organisms (Weber, 1993).  Significantly
 8    lower survival, decreased hatching success, slower development, and increased egg mass
 9    and larval mortality were observed in Jefferson salamanders (Ambystoma jeffersoniammi)
10    and wood frogs (Rana sylvaticd) exposed to Pb at a pH of 4.5 versus a pH of 5.5 (Home
11    and Dunson, 1995b). Contradictory results have been reported for invertebrates.  Over a
12    96-h exposure period, mortality increased with decreasing pH for the bivalve Pisidium
13    casertanum, while pH-independent mortality was reported for gastropods and Crustacea
14    under similar exposure conditions (Mackie, 1989). Cladocerans (C. dubid) and
15    amphipods (H. azteca) were also more sensitive to Pb toxicity at pH  6 to 6.5 than at
16    higher pH levels (Schubauer-Berigan et al., 1993). Lead was 100 times more toxic to the
17    amphipod Hyalella azteca at a pH range of 5.0 to 6.0 (Mackie,  1989) than at a pH range
18    of 7.0 to 8.5 (Schubauer-Berigan et al., 1993). Lead was also more toxic to fathead
19    minnows at lower pH levels (Schubauer-Berigan et al., 1993).
20          The influence of pH on Pb accumulation has also been observed in sediments.
21    Accumulation of Pb by the isopod Asellus communis was enhanced at low pH, after
22    a 20-day exposure to Pb-contaminated sediments (Lewis and Mclntosh, 1986).
23    In A. aquaticus, temperature increases were found to be more important than increased
24    pH in influencing Pb accumulation (Van Hattum et al., 1996). Increased water
25    temperature was also found to reduce Pb uptake fluxes in green microalga (Chlorella
26    kesslerif) (Hassler et al., 2004). Lead and zinc body concentrations in Asellus sp. were
27    found to vary markedly with seasonal temperature changes, with greater concentrations
28    present in spring and summer (Van Hattum et al., 1996).
29          Acute and chronic toxicity of Pb increases with decreasing water hardness, as Pb
30    becomes more soluble and bioavailable to aquatic organisms (Home and Dunson, 1995a;
31    Borgmann et al., 2005). There is some evidence that water hardness and pH work

      December 2005                        8-185    DRAFT-DO NOT QUOTE OR CITE

-------
 1   together to increase or decrease the toxicity of Pb. Jefferson salamanders exposed to Pb
 2   for 28 days at low pH and low water hardness experienced 50% mortality, while
 3   exposure to Pb at high pH and high water hardness resulted in 91.7% survival (Home and
 4   Dunson, 1995a). Exposure to Pb at high pH and low water hardness or low pH and high
 5   water hardness resulted in 75 and 41.7% survival, respectively (Home and Dunson,
 6   1995a). Similar results were reported for Jefferson salamanders during a 7-day exposure
 7   and wood frogs during 7- and 28-day exposures (Home and Dunson, 1995c).  In some
 8   cases, water hardness and pH in the absence of Pb have been shown to affect survival
 9   adversely.  Mean acute survival of wood frogs and Jefferson salamanders exposed to low
10   pH and low water hardness, in the absence of Pb, was 83.3 and 91.7%, respectively.
11   Mean chronic survival of wood frogs and Jefferson salamanders exposed to low pH and
12   low water hardness, in the absence of Pb, was 79.2 and 41.7%, respectively (Home and
13   Dunson, 1995c).
14         High Ca2+ concentrations have been shown to protect against the toxic effects of
15   Pb (Sayer et al., 1989; MacDonald et al., 2002; Hassler et al., 2004; Rogers and Wood,
16   2004).  Calcium affects the permeability and integrity of cell membranes and intracellular
17   contents (Sayer et al., 1989). As Ca2+ concentrations decrease, the passive flux of ions
18   (e.g., Pb) and water increases. At the lowest waterborne Ca2+ concentration (150
19   |imol/L), Pb accumulation in juvenile rainbow trout (Oncorhynchus mykiss) branchials
20   significantly increased as Pb concentration in water increased (Rogers and Wood, 2004).
21   At higher Ca2+ concentrations, Pb accumulation did not significantly increase with Pb
22   concentration in water.  This result demonstrates the protective effects of waterborne
23   Ca2+ and supports the suggestion that the Ca2+ component of water hardness determines
24   the toxicity of Pb to fish (Rogers and Wood, 2004). Rogers and Wood (2004) reported
25   that the uptake of Ca2+ and Pb2+ involves competitive inhibition of apical entry at
26   lanthanum-sensitive Ca2+ channels and interference with the function of the ATP-driven
27   baso-lateral Ca2+ pump.  High mortality was reported in brown trout (Salmo truttd) fry
28   exposed to Pb at a waterborne Ca2+ concentration of 20 jimol/L, while negligible
29   mortality was reported at the same Pb concentration but at a waterborne Ca2+
30   concentration of 200  jimol/L (Sayer et al., 1989). Adverse effects to mineral uptake and
31   skeletal development were observed in the latter test group (Sayer et al., 1989).

     December 2005                        8-186   DRAFT-DO NOT QUOTE OR CITE

-------
 1          The bioavailability of Pb and other metals that can be simultaneously extracted in
 2    sediments may be modified through the role of acid volatile sulfide (AVS) under anoxic
 3    conditions (Tessier and Campbell, 1987; Di Toro et al., 1992; Casas and Crecelius,
 4    1994).  The term AVS (iron  sulfide is an example) refers to the fraction of the sediment
 5    that consists of a reactive pool of solid-phase sulfide.  This phase is available to bind
 6    divalent metals that then become unavailable for uptake by aquatic biota.  The models
 7    proposed by Di Toro et al. (1992) and Casas and Crecelius (1994) predict that when the
 8    molar ratio of simultaneously extractable metals (SEM) to AVS in sediments is less than
 9    one, the metals will not be bioavailable due to complexation with available sulfide.
10          Salinity is an important modifying factor  to metal toxicity.  Verslycke et al. (2003)
11    exposed the estuarine mysid Neomysis integer to individual metals, including Pb, and
12    metal mixtures under changing salinity.  At a salinity of 5%, the reported LCso for Pb was
13    1140 |ig/L (95% CL = 840, 1440 |ig/L).  At an increased salinity of 25%o, the toxicity of
14    Pb was substantially reduced (LC50 = 4274 |ig/L [95% CL = 3540, 5710 |ig/L])
15    (Verslycke et al., 2003).  The reduction in toxicity  was attributed to increased
16    complexation of Pb2+ with Cl  ions.
17
18    8.2.4.5.5   Nutritional Factors
19          The relationship between nutrition and Pb toxicity has not been thoroughly
20    investigated in aquatic organisms. In fact, algae species are the only aquatic organisms to
21    have been  studied fairly frequently.  Although nutrients have been found to have an
22    impact on Pb toxicity, the mechanisms involved are poorly understood. It is unclear
23    whether the relationship between nutrients and toxicity comprises organismal nutrition
24    (the process by which a living  organism assimilates food and uses it for growth and for
25    replacement of tissues), or whether nutrients have interacted  directly with Pb, inhibiting
26    its metabolic interaction in the organism. This section reviews the little information that
27    has been gathered from studies documenting apparent Pb-nutrition associations in aquatic
28    organisms.
29          Jampani (1988) looked  at the impact of various nutrients (i.e., sodium acetate,
30    citric acid, sodium carbonate, nitrogen, and phosphates) on reducing growth inhibition in
31    blue-green algae (Synechococcus aeruginosus) exposed to 200 mg/L of Pb.  Exposure to
      December 2005                        8-187    DRAFT-DO NOT QUOTE OR CITE

-------
 1    this Pb treatment concentration caused 100% mortality in algae. Results indicated that
 2    additional nitrogen, phosphates, and some carbon sources, including sodium acetate,
 3    citric acid and sodium carbonate, all protected the algae from Pb toxicity.  Algae that had
 4    been starved prior to the experiment were found to be significantly more sensitive to Pb
 5    exposure. Glucose was the only nutrient tested that did not have a significant impact on
 6    Pb toxicity in S. aeruginosus. In a similar study by Rao and Reddy (1985) on
 7    Scenedesmus incrassatulus, nitrogen, phosphate and carbon sources (including glucose),
 8    all had protective effects, and reduced Pb toxicity at 300 and 400 mg Pb/L. Both studies
 9    proposed similar hypotheses regarding nutrient-Pb mechanisms that led to reduced
10    toxicity. One hypothesis was that the nutrients were able to reverse toxic effects.
11    The second hypothesis was that the nutrients interacted directly with Pb, in some way
12    sequestering the metal so as to inhibit its metabolic interaction with the organism
13    (Rao and Reddy, 1985; Jampani, 1988).
14          Rai and Raizada (1989) investigated the effects of Pb on nitrate and ammonium
15    uptake as well as carbon dioxide and nitrogen fixation in Nostoc muscorum over a 96-h
16    period. Test specimens were exposed to 10, 20, and 30 mg/L of Pb. At 20 mg Pb/L,
17    nitrate uptake was inhibited by 64% after 24 h and by 39% after 96 h.  Ammonium
18    uptake was inhibited, and similarly, inhibition decreased from 72% inhibition after 24 h
19    to 26% inhibition after 96 h of exposure. Carbon dioxide fixation  and nitrogenase
20    activity followed similar patterns, and results indicated that Pb exposure can affect the
21    uptake of some nutrients in N. muscorum.
22          Adam and Abdel-Basset (1990) studied the  effect of Pb on metabolic processes of
23    Scenedesmus obliquus. They found that nitrogenase activity was inhibited by Pb nitrate,
24    but enhanced by Pb-acetate. As photosynthetic products and respiratory substrates,
25    carbohydrate and lipid levels were altered by Pb. Above 30 mg/L  of Pb-nitrate, both
26    macronutrients were reduced. However, Pb-acetate was found to increase carbohydrate
27    levels. Results suggest that Pb can have an effect on macronutrients in S. obliquus and
28    that effects may vary depending on the chemical species.
29          Simoes Gon9alves et al. (1991) studied the impact of light, nutrients, air flux, and
30    Pb, in various combinations, on growth inhibition in the green algae Selenastrum
31    capricornutum.  Results indicated that at lower Pb  concentrations (<0.207 mg/L) and

      December 2005                       8-188   DRAFT-DO NOT QUOTE OR CITE

-------
 1    increased nutrient concentrations, algae release more exudates that form inert complexes
 2    with Pb anions in the water. This suggests that S. capricornutum can use exudates as a
 3    protection and that this protective mechanism depends on nutrient supply.  These results
 4    are consistent with those of Capelo et al. (1993), who investigated uptake of nitrogen and
 5    phosphorus in the algae Selenastrum capricornutum over time in the absence and
 6    presence of 0.207 mg Pb/L. They found that the presence of Pb had no significant
 7    influence on the assimilation of nitrogen and phosphorus. However, they did find that in
 8    the presence of Pb, algae released higher concentrations of organics with Pb-chelating
 9    groups.
10          Amiard et al.  (1994) investigated the impact on soft tissue Pb concentrations of
11    various feeding regimes on oysters (Crassostrea gigas) during their spat rearing.  They
12    fed test groups of C.  gigas  different amounts of Skeletonema costatum and additional
13    natural phytoplankton grown in test solutions. Results showed that size and food intake
14    both negatively correlated with metal concentrations in soft tissue.  The authors
15    hypothesized that this relationship was due in part to a diluting effect of the food.
16          In summary, nutrients affect Pb toxicity in those aquatic organisms that have been
17    studied.  Some nutrients seem to be  capable of reducing toxicity, though the mechanisms
18    have not been well established. Exposure to Pb has not been shown to reduce nutrient
19    uptake ability, though it has been demonstrated that Pb exposure may lead to increased
20    production and loss of organic material (e.g., mucus and other complex organic ligands)
21    (Capelo etal., 1993).
22
23    8.2.4.5.6   Interactions with Other Pollutants
24          Most of the scientific literature reviewed in this section considered how Pb and
25    other elements combine to  affect uptake and exert toxicity.  Research on the interactions
26    of Pb with complexing ligands and other physical and biological factors was more
27    thoroughly discussed in Sections 8.2.4.5.3 and 8.2.4.5.4.  Predicting the response of
28    organisms to mixtures of chemicals  is difficult (Norwood et al., 2003).  For example, at
29    low zinc concentrations, (2:1 Pb:Zn ratio) a synergistic effect was observed in the frog,
30    Bufo arenarum (Herkovits  and Perez-Coll, 1991).  At high concentrations of zinc, an
31    antagonistic effect was observed as Pb toxicity was reduced.  This demonstrates the
      December 2005                        8-189    DRAFT-DO NOT QUOTE OR CITE

-------
 1    complexity of metal mixture interactions as different metal concentrations, environmental
 2    conditions (e.g., temperature, pH), and other factors can cause marked changes in the
 3    effects observed (Norwood et al., 2003). In describing Pb interactions with other
 4    elements, interaction types are classified here as antagonistic, synergistic, and additive.
 5    Each of these will be discussed below with specific reference to known Pb-metal
 6    interactions and implications on Pb uptake and toxicity.
 7
 8    Antagonistic Interactions
 9          When two or more  metals compete for the same binding sites or interfere with
10    transport through cell walls or membranes, the interaction is termed less than strictly
11    additive or antagonistic. Antagonistic interactions can reduce metal bioavailability when
12    metals are present in combination, and may lead to reduced potential for toxicity (Hassler
13    et al., 2004). A number of elements act in an antagonistic fashion with Pb. For example,
14    Pb is a well-known antagonist to Ca2+ (Niyogi and Wood, 2004; Hassler et al., 2004),
15    which is an essential element, required for a number of physiological processes in most
16    organisms. Lead ions have an atomic structure similar to Ca2+ and can be transported
17    either actively or passively across cell membranes in place of Ca2+. An example of this
18    interaction was reported by Behra (1993a,b) where Pb was shown to activate calmodulin
19    reactions in rainbow trout  (O. mykiss) and sea mussel (Mytilus sp.) tissues in the absence
20    of calcium.  Calmodulin (CaM) is a major intracellular calcium receptor and regulates the
21    activities of numerous enzymes and cellular processes. Allen (1994) reported that Pb can
22    replace calcium in body structures (e.g., bones, shells); replace zinc in ALAD, which is
23    required for heme biosynthesis; and react with sulfhydryl groups, causing conformation
24    protein distortion and scission of nucleic acids (Herkovits and Perez-Coll, 1991).  Lead is
25    also a known antagonist to Mg2+, Na+, and Cl  regulation in fish (Ahern and Morris,
26    1998; Rogers and Wood, 2003, 2004; Niyogi and Wood, 2004).  Li et al. (2004) reported
27    on the interaction of Pb2+ with Cd2+ in the context of adsorption from solution by
28    Phanerochaete chrysosporium, a filamentous fungus. The authors found that cadmium
29    uptake decreased with increasing concentration of Pb ions with Pb outcompeting Cd2+ for
30    binding sites.
      December 2005                        8-190    DRAFT-DO NOT QUOTE OR CITE

-------
 1    Synergistic Interactions
 2          Synergism occurs when the interaction of two or more metals causes an effect that
 3    is greater than the effect observed from the individual metals themselves (Hagopian-
 4    Schlekat et al., 2001) or,  put another way, a greater than the strictly additive effect of the
 5    individual metals in a mixture (Playle, 2004). Synergism is likely the result of increased
 6    bioavailability of one or more of the metal ions due to the presence of other metals
 7    (Hassler et al., 2004). Hassler et al. (2004) reported that in the presence of copper (Cu2+),
 8    there was a significantly higher rate of internalization of Pb in the green algae Chlorella
 9    kesserii. It was suggested that Cu2+ may have affected organism physiology through the
10    disruption of cell membrane integrity. This would allow increased cation (i.e., Pb2+)
11    permeability and, therefore, substantially increase internalization  of Pb.  Hagopian-
12    Schlekat et al. (2001) examined the impact of individual metals and complex metal
13    mixtures containing Cd, Cu, Ni, Zn, and Pb to the estuarine copepod Amphiascus
14    tenuiremis. The copepods were exposed to metal-spiked sediment and pore water.  The
15    mixed metal sediment toxicity tests demonstrated greater than additive toxicity to
16    A. tenuiremis.  It was postulated that the Synergism observed was due to two or more
17    metals affecting the same biological function. Herkovits and Perez-Coll (1991) exposed
18    Bufo arenarum larvae to various Pb and zinc concentrations in solution.  At low zinc
19    concentrations, (2:1 Pb:Zn ratio), a synergistic toxic effect was observed in the frog
20    larvae relative to the effects observed from exposure to the individual metals and at
21    higher zinc concentrations. Enhanced Pb toxicity was attributed to the interference of Pb
22    with cellular activities due to binding with sulfhydryl polypeptides and nucleic acid
23    phosphates (Herkovits and Perez-Coll, 1991). Allen (1994) reported on the accumulation
24    of numerous metals and ions  into specific tissues of the tilapia Oreochromis aureus.
25    Tilapia exposed to low concentrations of Pb and mercury (both at 0.05 mg/L)  had
26    significantly higher concentrations of Pb in internal organs than those fish exposed to Pb
27    alone. Similarly, low concentrations of cadmium with low concentrations of Pb caused
28    increased uptake of Pb in certain organs (e.g., liver, brain, and caudal muscle).
29
      December 2005                        8-191    DRAFT-DO NOT QUOTE OR CITE

-------
 1   Additive Interactions
 2          The combined effects of two or more metals may result in additivity when the
 3   observed effects are greater than that observed with individual metals but equivalent to a
 4   summation of the effects from multiple metals.  Lead has been shown to complex with
 5   OF in aquatic systems. For example, Verslycke et al. (2003) exposed the estuarine mysid
 6   Neomysis integer to six different metals, including Pb, and a combined metal mixture
 7   under changing salinity conditions. At a salinity of 5%, the reported LCso for Pb was
 8   1140 |ig/L (840, 1440 |ig/L).  At an increased salinity of 25%o, the toxicity of Pb was
 9   substantially reduced (LC50 = 4274 |ig/L [3540, 5710 //g/L]) (Verslycke et al., 2003).
10   This reduction in toxicity was attributed to the increased concentration of Cl~ ion due to
11   increased salinity, in that it complexed with divalent Pb in the test system. Exposure of
12   N. integer to Pb in combination with the other five metals (Hg, Cd, Cu, Zn, Ni) resulted
13   in roughly strictly additive toxicity (Verslycke et al., 2003).
14
15   Summary of Interactions  With Other Pollutants
16          Norwood et al. (2003) reported that in a review and reinterpretation of published
17   data on the interactions of metals in binary mixtures (n  = 15 studies), antagonistic (6) and
18   additive interactions (6) were the most common for Pb.  The complexity of the
19   interactions and possible modifying factors makes determining the impact of even binary
20   metal mixtures to aquatic biota difficult (Norwood et al., 2003; Playle, 2004).  The two
21   most commonly reported Pb-element interactions are between Pb and calcium and
22   between Pb and zinc. Both calcium and zinc are essential elements in organisms and the
23   interaction of Pb with these ions can lead to adverse effects both by increased Pb uptake
24   and by a decrease in Ca and Zn required for normal metabolic functions.
25
26   8.2.4.6  Factors Associated with Global Climate Change
27          It is highly unlikely that Pb has any influence on generation of ground-level ozone,
28   depletion of stratospheric ozone, global warming, or other indicators of global climate
29   change.  Lead compounds have relatively short residence times in the atmosphere,
30   making it unlikely that they will reach the stratosphere, and they do not absorb infrared
31   radiation, making them unlikely to contribute to  stratospheric  ozone depletion or global

     December 2005                         8-192    DRAFT-DO NOT QUOTE OR CITE

-------
 1    warming. Also, these compounds are unlikely to have a significant interaction with
 2    ground-level nitrogen oxides or volatile organic compounds, thus precluding generation
 3    of ground-level ozone.
 4          Approached from another viewpoint, climate change can have a major impact on
 5    the fate/behavior of Pb in the environment and, therefore, can subsequently alter
 6    organism or ecosystem responses. For example, changes in temperature regime (Q10
 7    rule), changes in precipitation quantity and quality (e.g., acidic deposition) may influence
 8    fate, transport, uptake, and bioavailability of Pb (Syracuse Research Corporation., 1999).
 9
10    8.2.4.7  Summary
11          There have been a number of advancements in the understanding of Pb behavior in
12    the environment and its impact on aquatic organisms since 1986.  In particular, greater
13    knowledge of factors that influence Pb accumulation in aquatic organisms, mechanisms
14    of detoxification and avoidance of Pb, and greater understanding of the interactions of Pb
15    in aquatic systems. Recently, the development of the Biotic Ligand Model (BLM) and its
16    exploration of the activity of free metal ions at biologically reactive sites (i.e., fish gill
17    tissue) have been a large contributor to the understanding of metal speciation and
18    movement into and effects to aquatic biota. To date, there has been no BLM model of Pb
19    although research has been conducted on a Pb-gill binding model for rainbow trout.
20    Further research in support of BLM model development for Pb is recommended to
21    further our understanding of these issues.
22
23    8.2.5   Exposure/Response of Aquatic Species
24    8.2.5.1  Introduction
25          This section outlines and highlights the critical recent advancements in the
26    understanding of the toxicity of lead to aquatic biota. The section begins with a review of
27    the major findings and conclusions from the 1986 Pb AQCD. The following sections
28    summarize the research conducted since 1986 on effects of lead on primary producers,
29    consumers, and decomposers. Issues related to indirect effects (e.g., effects on
30    predator/prey interactions, habitat alteration,) are not to be addressed.
31
      December 2005                       8-193    DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.5.2  Summary of Conclusions From the Previous Criteria Document
 2          The 1986 Pb AQCD (U.S. Environmental Protection Agency, 1986a) reviewed
 3    data in the context of the sublethal effects of lead exposure. The document focused on
 4    describing the types and ranges of lead exposures in ecosystems likely to adversely
 5    impact domestic animals. As such, the criteria document did not provide a
 6    comprehensive analysis of the effects of lead to most aquatic primary producers,
 7    consumers, and decomposers. For the aquatic environment, general reviews of the
 8    effects of lead to algae, aquatic vertebrates, and invertebrates were undertaken.
 9    A summary of these reviews is provided below.
10
11    Algae
12          The 1986 Pb AQCD (U.S. Environmental Protection Agency, 1986a) reported that
13    some algal species (e.g., Scenedesmus sp.) were found to exhibit physiological changes
14    when exposed to high lead or organolead concentrations in situ.  The observed changes
15    included increased numbers of vacuoles, deformations in cell organelles, and increased
16    autolytic activity. Increased vacuolization was assumed to be a tolerance mechanism by
17    which lead was immobilized within cell vacuoles.
18
19    Aquatic Vertebrates
20          The 1986 Pb AQCD (U.S. Environmental Protection Agency, 1986a) reported that
21    hematological and neurological responses were the most commonly reported effects in
22    aquatic vertebrates. These effects include red blood cell destruction and inhibition of the
23    enzyme ALAD, required for hemoglobin synthesis. At high lead concentrations,
24    neurological responses included neuromuscular distortion, anorexia, muscle tremor, and
25    spinal curvature (e.g., lordosis).  The lowest reported exposure concentration causing
26    either hematological or neurological effects was 8 |ig/L (U.S. Environmental Protection
27    Agency, 1986a).
28
29    Aquatic Invertebrates
30          Numerous studies were cited on the effects of lead to aquatic invertebrates in the
31    1986 Pb AQCD (U.S. Environmental Protection Agency, 1986a). In general, lead
32    concentrations in aquatic invertebrates were found to be correlated closely with

      December 2005                       8-194    DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentrations in water rather than food.  Freshwater snails were found to accumulate
 2    lead in soft tissue, often in granular bodies of precipitated lead.  Mortality and
 3    reproductive effects were reported to begin at 19 |ig/L for the freshwater snail Lymnea
 4   palutris and 27 |ig/L for Daphnia sp.
 5         The review of the NAAQS for Lead (U.S. Environmental Protection Agency,
 6    1990) made only one recommendation on the sections of the 1986 Pb AQCD reporting on
 7    effects to aquatic biota. This was the need to consider the impact of water hardness on
 8    lead bioavailability and toxicity, to be consistent with the recommendations of the
 9    AWQC for the protection of aquatic life (U.S. Environmental Protection Agency, 1985).
10
11    8.2.5.3   Recent Studies on Effects of Lead on Primary Producers
12         Using literature published since the 1986 Pb AQCD, this section examines the
13    toxicity of lead (individually and in metal mixtures) to algal and aquatic plant growth, its
14    effects on metabolic processes (e.g., nutrient uptake), and its impact on primary
15    productivity in natural systems.
16
17    Toxicity of Lead to Algae
18         The toxicity of lead to algal growth has been investigated for a number of species
19    including Chlorella vulgaris, Closterium acerosum, Pediastrum simplex, Scenedesmus
20    quadricauda, Scenedesmu obliquus, Syneschoccus aeruginosus, and Nostoc muscorum
21    (Jampani, 1988; Rai and Raizada, 1989; Adam and Abdel-Basset, 1990; Fargasova, 1993;
22    Bilgrami and Kumar, 1997).  Study durations ranged from 7 to 20 days and lead nitrate
23    was the most commonly used form of lead. Effects to algal growth (Chlorella vulgaris,
24    Closterium acerosum, Pediastrum simplex, Scenedesmus quadricauda), ranging from
25    minimal to complete inhibition, have been reported at lead concentrations between 100
26    and 200,000 |ig/L (Jampani, 1988; Bilgrami and Kumar, 1997). Most studies report the
27    percent inhibition in test groups compared to controls rather than calculating the LOEC,
28    NOEC, or ECso values. Clinical signs of lead toxicity include the deformation and
29    disintegration of algae cells and a shortened exponential growth phase (Jampani, 1988;
30    Fargasova, 1993). Other effects of lead block the pathways that lead to pigment
     December 2005                        8-195    DRAFT-DO NOT QUOTE OR CITE

-------
 1    synthesis, thus affecting photosynthesis, the cell cycle and division, and ultimately result
 2    in cell death (Jampani, 1988).
 3          From the studies reviewed, Closterium acerosum is the most sensitive alga species
 4    tested (Bilgrami and Kumar, 1997). Exposure of this alga to 1000 and 10,000 |ig/L of
 5    lead nitrate for 6 days resulted in cell growth that was 52.6 and 17.4%, respectively, of
 6    controls (Bilgrami and Kumar, 1997). Chlorella vulgaris, Pediastrum simplex, and
 7    Scenedesmus quadricauda were also exposed to lead nitrate in this study. Compared to
 8    controls, cell growth at 1000 and 10,000 |ig lead nitrate/L was 65.3 and 48.7%, 64.5 and
 9    42.7%, and 77.6 and 63.2%, respectively (Bilgrami and Kumar, 1997).  Scenedesmus
10    quadricauda exhibited a similar magnitude of effects when exposed to lead (Pb2+) for
11    20 days at 0, 5500, 11,000, 16,500, 22,000, 27,500, and 33,000 |ig/L (Fargasova, 1993).
12    This study reported an EC50 for growth inhibition at 13,180 |ig/L (95% CI: 10,190,
13    14,620). Decreased cell number, but increased cell size, was observed in Selenastrum
14    capricornutum11 exposed to lead (Pb2+) at 207.2 |ig/L and a Q/V (flux of air [Q] divided
15    by volume of the culture [V]) of 4.7 x 10~3 sec"1 for 9 days (Simoes Gon9alves et al.,
16    1991).  The Q/V is a measure of culture growth where an increase in the Q/V ratio
17    indicates growth. The pigment concentration per cell decreased with  exposure to lead, so
18    while the algae cells were larger, they were less healthy (Simoes Gon9alves et al., 1991).
19    Growth rates were not reported, making comparison with other studies difficult.
20          High lead concentrations were required to elicit effects in Nostoc muscorum and
21    Scenedesmus aeruginosus (Jampani,  1988; Rai  andRaizada, 1989). Following 15 days
22    of exposure, test groups exposed to 10,000, 20,000, and 30,000 |ig/L lead experienced
23    growth rates that were 90.5, 76.9, and 66.7% of the controls (Rai and  Raizada, 1989).
24    Synechococcus aeruginosus experienced little inhibition of growth from exposure to lead
25    nitrate up to a concentration of 82,000 |ig/L (Jampani,  1988). At a test concentration of
26    100,000 |ig/L, complete inhibition of growth was observed, and at a concentration of
27    200,000 |ig/L, algae failed to establish a single  colony (Jampani, 1988).  Scenedesmus
28    obliquus are quite tolerant to the effects of lead nitrate and lead acetate on growth.
              The species name Selenastrum capricornutum has been changed to Pseudokirchneriella
      subcapitata. The former species name is used in this report.
      December 2005                        8-196    DRAFT-DO NOT QUOTE OR CITE

-------
 1    Algae exposed to lead nitrate or lead acetate up to 180,000 jig/L had higher cell numbers
 2    than controls (Adam and Abdel-Basset, 1990). Exposure to the highest concentration of
 3    300,000 |ig/L lead nitrate or lead acetate resulted in cell numbers that were 81 and 90%
 4    of the controls, respectively (Adam and Abdel-Basset, 1990).
 5         Lead in combination with other metals (e.g., Pb and Cd, Pb and Ni, etc.) is
 6    generally less toxic than exposure to lead alone (Rai and Raizada, 1989).  Nostoc
 1    muscorum exposed to chromium and lead in combination demonstrated better growth
 8    than when exposed to either of the metals alone (Rai and Raizada, 1989).  Antagonistic
 9    interaction was observed in the exposure of Nostoc muscorum to lead and nickel in
10    combination (Rai and Raizada, 1989).  When applied separately, these metals
11    demonstrated different levels of toxicity; however, in combination, they exerted similar
12    effects (Rai  and Raizada, 1989).  More information on toxic interactions of lead with
13    other metals is provided in Section 8.2.4.5.6.
14
15    Aquatic Plants
16         The toxicity of lead to aquatic plant growth has been studied using Spirodela
17    polyrhiza, Azolla pinnata, and Lemna gibba (Gaur et al., 1994; Gupta and Chandra, 1994;
18    Miranda and Ilangovan, 1996).  Test durations ranged from 4 to 25 days and test
19    concentrations ranged between 49.7 and 500,000 |ig/L (Gaur et al., 1994;  Miranda and
20    Ilangovan, 1996).  Research on aquatic plants has focused on the effects of lead on
21    aquatic plant growth and chlorophyll and protein content.
22         Of the species reviewed here, the effects of lead on aquatic plant growth are most
23    pronounced in Azollapinnata (Gaur et al., 1994).  An ECso of 1100 |ig/L was reported for
24    Azolla pinnata exposed to lead nitrate for 4 days.  Spirodela polyrhiza exposed to lead
25    nitrate under the same test conditions had a reported ECso for growth of 3730 |ig/L (Gaur
26    et al., 1994). Lemna gibba was shown to be the least sensitive plant species to lead:
27    significant growth inhibition was reported at concentrations of 200,000 |ig/L  or greater
28    after 25 days of exposure to concentrations of 30,000, 50,000, 100,000, 200,000,
29    300,000, or  500,000 |ig/L (Miranda and Ilangovan, 1996).  The maximum growth rate for
30    Lemna gibba was observed at 10 days of exposure. After this point, the growth rate
31    declined in controls and test concentrations (Miranda and Ilangovan, 1996). Clinical
      December 2005                       8-197    DRAFT-DO NOT QUOTE OR CITE

-------
 1    signs of lead toxicity include yellowing and disintegration of fronds, reduced frond size,
 2    and chlorosis (Gaur et al., 1994; Miranda and Ilangovan, 1996). Toxicity results suggest
 3    that effects to growth from exposure to lead occur in a dose-dependent manner (Gaur
 4    etal., 1994).
 5
 6    Effects of Lead on Metabolic Processes
 1          Algal and aquatic plant metabolic processes are variously affected by exposure to
 8    lead, both singularly and in combination with other metals. Lead adversely affects the
 9    metabolic processes of nitrate uptake, nitrogen fixation,  ammonium uptake, and carbon
10    fixation at concentrations of 20,000 |ig/L or greater (Rai and Raizada, 1989). Lead in
11    combination with nickel has an antagonistic effect on nitrogen fixation and ammonium
12    uptake, but a synergistic effect on nitrate uptake and  carbon fixation (Rai and Raizada,
13    1989).  Lead in combination with chromium has an antagonistic effect on nitrate uptake,
14    but it has a synergistic effect on nitrogen fixation, ammonium uptake, and carbon fixation
15    (Rai and Raizada,  1989).
16          Lead effects on nitrate uptake in Nostoc muscorum (jig NOs/jig Chi a) were
17    greatest after 24 h, when exposure to 20,000 jig/L reduced nitrate uptake by 64.3%
18    compared to controls. Nitrate uptake reported after 48, 72, and 96 h was reduced by 30.0,
19    37.5, and 38.9%, respectively, compared to controls (Rai and Raizada, 1989).  Lead in
20    combination with chromium, both at a test concentration of 20,000 |ig/L, demonstrated
21    antagonistic effects on nitrate uptake.  Compared to controls, nitrate uptake was reduced
22    by 52.4, 30, 25, and 22.2% at 24, 48, 72 and 96 h, respectively (Rai and Raizada,  1989).
23    The greatest effect on uptake occurred at 24 h when,  compared to controls, a 52.4%
24    reduction was reported in the test concentration. Lead and nickel in combination  at test
25    concentrations of 20,000 and 1000 |ig/L, respectively, resulted in a greater reduction of
26    nitrate uptake than lead alone at 48, 72, and 96 h (Rai and Raizada, 1989).
27          After 24, 48, and 72 h of exposure to lead at 20,000 |ig/L, nitrogenase activity
28    (nmol C2H4/ jig protein/hr) in Nostoc muscorum was reduced by 39.3, 61.8, and 14.1%,
29    respectively, compared to controls (Rai and Raizada, 1989).  A concentration of
30    207.2 |ig/L had little effect on nitrogen or phosphorus assimilation in Selenastrum
31    capricornutum over 7 days (Capelo et al., 1993). An antagonistic effect on nitrogenase
      December 2005                        8-198    DRAFT-DO NOT QUOTE OR CITE

-------
 1    activity was generally reported for Nostoc muscorum exposed to lead in combination with
 2    nickel at 20,000 and 1,000 |ig/L, respectively (Rai and Raizada,  1989). Compared to
 3    controls, nitrogenase activity was reduced by 42.9, 32.7, and 13.6% at 24, 48, and 72 h,
 4    respectively (Rai and Raizada, 1989). Lead and chromium, both administered at a
 5    concentration of 20,000 |ig/L, had a synergistic impact on nitrogenase activity in Nostoc
 6    muscorum. Nitrogenase activity in the test group was reduced by 60.7, 60, and 50%
 7    compared to the controls at 24, 48, and 72 h, respectively (Rai and Raizada, 1989).
 8          Lead-induced inhibition of ammonium uptake (jig NH4 uptake/jig Chi a) was
 9    greatest in Nostoc muscorum after 48 h of exposure to 20,000 |ig/L of lead.  Compared to
10    controls, the lead test concentration 20,000 |ig/L reduced ammonium uptake by 72, 82,
11    61, and 26 % at 24, 48, 72, and 96 h, respectively (Rai and Raizada, 1989).  Lead in
12    combination with nickel at concentrations of 20,000 and 1,000 |ig/L, respectively,
13    demonstrated an antagonistic effect on ammonium uptake. Compared to controls,
14    ammonium uptake in the test group was reduced by 44.9, 54.1, 23.3, and 4% at 24, 48,
15    72, and 96 h, respectively (Rai and Raizada, 1989). Lead in combination with chromium,
16    both at concentrations of 20,000 |ig/L, demonstrated a synergistic interaction with 24, 48,
17    72, and 96 h uptake rates reduced by 87.2, 88.5, 72.5, and 50 %,  respectively, compared
18    to controls (Rai and Raizada, 1989).
19          Nostoc muscorum exposed to 20,000 |ig/L of lead experienced the greatest
20    reduction in carbon fixation at 0.5  h of exposure: 62% compared to controls. Inhibition
21    of carbon fixation in the test group was less pronounced after 1 and 2 h of exposure:
22    29 and 13% of controls (Rai and Raizada, 1989). Lead in combination with nickel or
23    chromium had  synergistic effects to carbon fixation. Lead and nickel concentrations of
24    20,000 and 1000 |ig/L, respectively, resulted in 0.5, 1, and 2 h carbon fixation rates
25    reduced by 93,  92, and 91%, respectively, compared to controls (Rai and Raizada, 1989).
26    Lead with chromium at concentrations of 20,000 |ig/L resulted in 0.5, 1, and 2 h carbon
27    fixation rates reduced by 65, 58, and 50%, respectively, compared to controls.
28          Nutrients such as nitrogen, phosphate, sodium acetate, sodium carbonate, and
29    citric acid have been shown to protect against the toxic effects of lead to algae (Jampani,
30    1988). Nitrogen compounds (ammonium chloride, potassium nitrate, sodium nitrate,
31    sodium nitrite)  protected Synechococcus aeruginosus from a lethal lead nitrate dose of

      December 2005                       8-199    DRAFT-DO NOT  QUOTE OR CITE

-------
 1    200,000 |ig/L (Jampani, 1988). Two phosphates (K2HPO4 and Na2HPO4) were found to
 2    improve Synechococcus aeruginosus survival from 0 to 72% at 200,000 |ig/L of lead
 3    nitrate (Jampani, 1988).
 4          Compared to controls, protein content was reduced by 54.2 and 51.9% in aquatic
 5    plants Vallisneria spiralis and Hydrilla verticillata, respectively, exposed to lead for
 6    7 days at 20,720 |ig/L (Gupta and Chandra, 1994).  Decreased soluble protein content has
 7    been observed in Scenedesmus obliquus exposed to lead nitrate or lead acetate at
 8    concentrations greater than 30,000 |ig/L, and in Lemna gibba at concentrations greater
 9    than 200,000 |ig/L (Adam and Abdel-Basset,  1990; Miranda and Ilangovan, 1996).
10    Lemna gibba also showed increased loss of soluble starch at concentrations >200,000
11    Hg/L (Miranda and Ilangovan, 1996).  Under the conditions described previously (Gupta
12    and Chandra, 1994), EC50 values for chlorophyll content were 14,504 and 18,648 |ig/L
13    for Vallisneria spiralis and Hydrilla verticillata, respectively (Gupta and Chandra, 1994).
14    Effects to chlorophyll a content have been observed in Scenedesmus obliquus at lead
15    nitrate and lead acetate concentrations >30,000 |ig/L (Adam and Abdel-Basset, 1990).
16
17    Summary of Toxic Effects Observed in Single-Species Bioassays
18          Algae and aquatic plants have a wide range in sensitivity to the effects of lead in
19    water. Both groups of primary producers experience ECso values for growth inhibition
20    between approximately 1000 and >100,000 |ig/L (Jampani, 1988; Gaur et al., 1994;
21    Bilgrami and Kumar, 1997).  The most sensitive primary producers reported in the
22    literature for effects to growth were Closterium acersoum and Azollapinnata (Gaur et al.,
23    1994; Bilgrami and Kumar, 1997).  The least sensitive primary producers reported in the
24    literature for effects to growth were Synechococcus aeruginosus and Lemna gibba
25    (Jampani, 1988; Miranda and Ilangovan, 1996). Exposure to lead in combination with
26    other metals is generally less toxic to growth than exposure to lead alone. Studies have
27    shown that lead adversely affects the metabolic processes of nitrate uptake, nitrogen
28    fixation, ammonium uptake, and carbon fixation (Rai and Raizada, 1989). Lead in
29    combination with nickel or chromium produced synergistic effects for nitrate uptake,
30    nitrogenase activities, ammonium uptake, and carbon fixation (Rai and Raizada, 1989).
31

      December 2005                        8-200    DRAFT-DO NOT QUOTE OR CITE

-------
 1   Leads Effects on Primary Productivity
 2          Lead nitrate and lead acetate have been shown to have adverse effects on the
 3   primary productivity of aquatic plants in two water bodies in India (Jayaraj et al.,  1992).
 4   One of the two water bodies was a freshwater tank that receives wastewater and supports
 5   a rich population of hyacinths, and the other was a wastewater stabilization pond.  Water
 6   quality characteristics in the freshwater tank were pH = 7.5, dissolved oxygen = 6 mg/L,
 7   and water hardness (CaCOs) =100 mg/L. Water quality characteristics in the wastewater
 8   pond were pH = 8.1, dissolved oxygen = 6.2 mg/L, and water hardness (CaCOs) =160
 9   mg/L (Jayaraj et al.,  1992). Lead nitrate concentrations of 500,  5000, 10,000, 25,000,
10   and 50,000 |ig/L were combined with appropriate water samples in light and dark bottles
11   and suspended in each of the water bodies for 4 h.  The concentrations of lead acetate
12   (5000,  10,000, 25,000, 50,000, and 100,000 |ig/L) were applied in the same manner.  The
13   ECso values were determined based on the concentration required to inhibit gross
14   productivity (GP) and net productivity (NP) by 50%  (Jayaraj et  al., 1992). The results
15   demonstrated that lead nitrate was more toxic to primary production than lead acetate. In
16   the freshwater tank, lead nitrate ECso values for GP and NP were 25,100 and 6310 |ig/L,
17   respectively, compared to lead acetate EC50 values of 50,100 and 28,200 jig/L for GP and
18   NP, respectively (Jayaraj et al., 1992). In the stabilization pond, lead nitrate ECso values
19   for GP and NP were 31,600 and 28,200 |ig/L, respectively, compared to lead acetate ECso
20   values of 79,400 and 316 |ig/L for GP and NP, respectively (Jayaraj et al., 1992).  The
21   higher toxicity reported in the freshwater tank was attributed to  differences in species
22   composition and diversity.  The freshwater tank was  dominated by water hyacinths that
23   decreased the photic zone available for photosynthesis and consumed a  great deal of
24   available nutrients. The stabilization pond had a rich nutrient budget, resulting  in
25   improved alga growth and species diversity (Jayaraj et al., 1992).
26
27   8.2.5.4  Recent Studies on Effects of Lead on Consumers
28          This section focuses on the effects of lead to aquatic biota including invertebrates,
29   fish, and other biota with an aquatic life stage (e.g., amphibians). It is not intended to be
30   a comprehensive review of all research conducted. Rather, the intent is to illustrate the
31   effects Pb can have on freshwater and marine aquatic species. Eisler (2000) provides an


     December 2005                       8-201    DRAFT-DO NOT QUOTE OR CITE

-------
 1    overview of much of the recent available literature on the toxicity of lead to fish and
 2    aquatic invertebrates. An extensive literature search was conducted using numerous
 3    electronic bibliographic and database services (e.g., DIALOG, EPA ECOTOX) and
 4    limited temporally from 1986 to present.  This temporal limit was due to the availability
 5    of the EPA water quality criteria report for the protection of aquatic life, released in 1986
 6    (U.S. Environmental Protection Agency, 1986b). Based on the results of the literature
 7    search and recent reviews of the toxicity of lead (Eisler, 2000), numerous studies have
 8    been published on the toxicity of lead to aquatic consumers.  Hardness, pH, temperature,
 9    and other factors are important considerations when characterizing the acute and chronic
10    toxicity of lead (Besser et al., 2005) (Section 8.2.4.5). However, many of the studies
11    reviewed did not report critical information on control mortality, water quality
12    parameters, or statistical methods, making comparing effects between studies difficult.
13    Studies reporting only physiological responses to lead exposure (e.g., reduction of
14    ALAD) are not discussed here, as this topic was covered more completely in
15    Section 8.2.4.4.  This section provides a review of toxicity studies conducted with
16    invertebrates, fish, and other aquatic organisms.
17
18    Invertebrates
19          Exposure of invertebrates to Pb can lead to adverse effects on reproduction,
20    growth, survival, and metabolism (Eisler, 2000).  The following presents information on
21    the toxicity of lead to invertebrates in fresh and marine waters.
22
23    Freshwater Invertebrates
24          Acute and chronic lead toxicity data for freshwater invertebrates are summarized
25    in Table 8-2.5.1.  As described in  Section 8.2.4.5.4, water hardness is a critical factor
26    governing the solubility, bioavailability, and ultimately the toxicity of lead.  The acute
27    and chronic toxicity of lead increases with decreasing water hardness as lead becomes
28    more soluble and bioavailable to aquatic organisms.  For example, Borgmann et al.
29    (2005)  examined the toxicity of 63 metals, including Pb, to Hyalella azteca at two levels
30    of water hardness (soft water hardness, 18 mg CaCOs/L; hard water, 124 mg CaCOs/L).
31    Lead was 23 times more acutely toxic to H. azteca in soft water than hard water.  Besser
32    et al. (2005) found that acute toxicity to H. azteca was also modified by water hardness.

      December 2005                        8-202    DRAFT-DO NOT QUOTE  OR CITE

-------
December
to










2,700
>2,700

LC50:
>8,000
>8,000
>8,000



LC50:
<90
>5,400
>5,400


LC50:
27(20.1-36.4)


LC50:
60 (53.6-67.3)


LC50:
1090
(400-133200)

LC50:
5000

of Lead to Freshwater and Marine Invertebrates
Duration of
Exposure Water Chemistry


48 h pH:
6-6.5
7-7.5
8-8.5
Hardness: 280-300 mg/L CaCO3
96 h pH:
6-6.5
7-7.5
8-8.5
Hardness: 280-300 mg/L CaCO3


96 h pH:
6-6.5
7-7.5
8-8.5
Hardness: 280-300 mg/L CaCO3

8 days Hardness
130 mg/L
pH 7.8-8.6

8 days Hardness
130 mg/L
pH 7.8-8.6

96 h pH:
4.5


96 h pH
7.0

Test Type - Effect


static-survival




static-survival






static-survival





renewal, 1 -week-
old amphipods


renewal, 10- to 16-
week old
amphipods

acute - survival



acute - survival


Reference


Schubauer-Berigan
etal. (1993)



Schubauer-Berigan
etal. (1993)





Schubauer-Berigan
etal. (1993)




MacLean et al.
(1996)


MacLean et al.
(1996)


Gerhardt (1994)



Gerhardt (1994)



-------
o
o
3
CT*
to
O
<*s\








oo
1
to
o



o

m
H
6
O
0
h-j
o

o
H
W
O

o
1 — I
H
W
Table 8-2.5.1 (cont'd). Effects of Lead to Freshwater and Marine

Species
Amphipod
(Hyalella azteca)


Bivalve
(Pisidium compressum)


Bivalve
(Pisidium casertanum)




Gastropod
(Amnicola limosa)


Mussel
(Dreissena polymorpha)


Mussel
(Dreissena polymorpha)
Amphipod
(Hyalella azteca)


Amphipod
(Hyalella azteca)





Endpoint:
Chemical Cone. (jig/L)*
lead
nitrate


lead
nitrate


lead
nitrate




lead
nitrate


lead
nitrate


lead
nitrate
lead
nitrate


lead
nitrate




LC50:
10
21
18
LC50:
38,000
21,300
11,400
LC50:
23,600
23,500
56,000


LC50:
10,300
20,600
9,500
EC50:
370
91

LT50:
358
LC50:
4.8(3.3-7.1)


LC50:
113(101-126)




Duration of
Exposure Water Chemistry
96 h pH:
5.0
5.5
6.0
96 h pH:
3.5
4.0
4.5
96 h pH:
3.5
4.0
4.5


96 h pH:
3.5
4.0
4.5
48 h pH = 7.9; Hardness = 150 mg
10 wks CaCO3/L; Temp = 15 °C


72 days pH = 7.9; Hardness = 150 mg
CaCO3/L; Temp = 15 °C
7 days pH = 7.37 - 8.27
Hardness = 18 mg CaCO3/L
DOC = 0.28 mg/L

7 days pH = 8.2 1-8.46
Hardness = 124 mg CaCO3/L;
DOC =1.1 mg/L



Invertebrates

Test Type - Effect Reference
acute-survival Mackie (1989)



acute- survival Mackie (1989)



acute- survival Mackie (1989)





acute- survival Mackie (1989)



renewal - filtration Kraak et al. (1994)



renewal - filtration Kraak et al. (1994)

renewal - survival Borgmann et al.
(2005)


renewal - survival Borgmann et al.
(2005)





-------
o
CD
3
cr
to
o
o









oo
i
to
o
01



O
^
H
I
0
o
0
H
O
o
w
o
o
HH
H
W


Species
Mayfly
(Leptophlebia marginata)


Cladoceran
(D. magna)

Cladoceran
(D. magna)
Cladoceran
D. magna)
Amphipod
(Hyalella azteca)


Tubificid worm
(Tubifex tubifex)

Marine
Copepod
(Amphiascus tenuiremis)


Bivalve
(Mytilus galloprovincialis)


Table 8-2.

Chemical
lead
chloride


lead
nitrate

lead
chloride
lead
chloride
lead



lead
nitrate


lead



lead
nitrate


* - Brackets after effect concentration are






5.1 (cont'd). Effects of Lead
Endpoint:
Cone. (^ig/L)*
LC50:
1090 (95% C.I.
133.2)
>5000
LC50:
0.45

NOEC:
260
NOEC:
270
LOEC:
(Dissolved Pb)
192
(Total Pb) 466
EC50:
237(183-316)
142 (107-184)

LC50: sediment
2462 ug
metal/dry
sediment
EC50 : 221
(58.9-346.3)
LOEC : 50

95% confidence



Duration of
Exposure
96 h
=


48 h


12 to 21 d

10 d

96 h



24 h
48 h


96 h







intervals.



to Freshwater and Marine Invertebrates

Water Chemistry
pH = 4.5-6.5;
DOC^l.emgCr1;
Cond = 7.0 ,uS cm"1

pH = 8.3± 0.2
Hardness (CaCO3) = 150 mg/L
Temp= 20 °C
Not specified

Not specified

pH = 8.27
Hardness (CaCO3) = 275 mg/L
Temp = 21.1 °C

pH = 7.5-7.7
Hardness = 245 mg/L
Temp = 29.5-31 °C

pH = 7.7± 0.1
Dissolved O2 -6.3 ± 0.3 mg/L
Salinity - 32 ppt

artificial seawater








Test Type - Effect
renewal - survival



static -
embryogenesis

renewal -
reproduction
renewal - growth

flow through -
survival


static -
immobilization






embryogenesis








Reference
Gerhardt (1994)



Bodaretal. (1989)


Enserink et al.
(1991)
Enserink et al.
(1991)
Besser et al. (2005)



Khangarot(1991)



Hagopian-Schlekat
etal. (2001)


Beiras and
Albentosa (2003)







-------
 1    At a mean pH of 7.97 in soft water (hardness (CaCOs) = 71 mg/L) mortality was >50% for
 2    H. azteca at a dissolved Pb concentration of 151 |ig/L.  The LOEC for survival in hard water
 3    (hardness (CaCO3) = 275 mg/L) at pH 8.27 was 192 |ig/L as dissolved Pb and 466 |ig/L as total
 4    Pb. Both waterborne and dietary Pb were found to contribute to reduced survival ofH. azteca
 5    (Besser et al., 2005).
 6          Exposure duration may also play an important role in lead toxicity in some species. For
 7    example, Kraak et al. (1994) reported that filtration in the freshwater mussel Dreissena
 8    polymorpha was adversely affected at significantly lower Pb concentrations over 10 weeks of
 9    exposure than was the case after 48 h of exposure.
10          The influence of pH on lead toxicity in freshwater invertebrates varies between
11    invertebrate species. Over a 96-h exposure period, mortality increased with decreasing pH in the
12    bivalve Pisidium casertanum, while pH-independent mortality was reported for gastropod and
13    crustacean species under similar exposure conditions (Mackie, 1989). Cladocerans
14    (Ceriodaphnia dubia), amphipods (H. azteca), and mayflies (Leptophlebia marginata) were also
15    more sensitive to lead toxicity at lower  pH levels (Schubauer-Berigan et al., 1993; Gerhardt,
16    1994). Lead was  100 times more toxic  to the amphipod, Hyalella azteca, at a pH range of 5.0 to
17    6.0 (Mackie, 1989) than at a pH range of 7.0 to 8.5 (Schubauer-Berigan et al., 1993).
18          The physiology of an aquatic organism at certain life stages may be important when
19    determining the toxicity of metals to test organisms. For example, Bodar et al. (1989) exposed
20    early life stages of Daphnia magna to concentrations of Pb(NC>3)2. The test medium had a pH of
21    8.3 + 0.2, water hardness (CaCO3) of 150 mg/L, and temperature of 20 + 1  °C. Lead
22    concentrations of < 100 mg/L had no impact on Daphnia egg development.  The authors
23    suggested that this may due to the Daphnia egg structure, which consists of two layers: the inner
24    vitelline layer and outer chlorion layer.  The chlorion layer in other species  (e.g., rainbow trout)
25    is known to adsorb metals, thereby, preventing ionic injury  to the developing embryo.
26          Exposures to sediment-associated lead can be toxic to sediment-dwelling organisms.
27    In freshwater sediments, 48-h exposure of water fleas (Daphnia magna) to 7000 mg/kg dw
28    significantly reduced mobility, while exposure to 13,400 mg/kg dw for 24 h produced the same
29    effect (Dave, 1992a,b). Longer-term (i.e.,  14-day) exposure of midges (Chironomus tentans) to
30    sediments containing 31,900 mg/kg dw of lead resulted in 100% mortality.
31

      December 2005                           8-206        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Marine Invertebrates
 2          In estuarine environments, salinity is an important modifying factor to Pb toxicity.
 3   Verslycke et al. (2003) exposed the estuarine mysid Neomysis integer to individual metals,
 4   including Pb, and metal mixtures under changing salinity. Water temperature (20+1 °C) and
 5   salinity were reported, although no other water quality parameters were available (e.g., pH, water
 6   hardness).  At a salinity of 5%o, the reported LC50 for Pb was 1140 |ig/L (95% CI: 840,
 7   1440 |ig/L). At an increased salinity of 25%o, the toxicity of lead was substantially reduced
 8   (LC50 = 4274 |ig/L [3540, 5710 |ig/L]) (Verslycke et al., 2003).
 9          Sensitivity to Pb can also vary between genders in some aquatic organisms. For example,
10   Hagopian-Schlekat et al. (2001) examined the toxicity of lead chloride in sediment and sediment
11   pore water to female and male estuarine copepods Amphiascus tenuiremis. The reported LCso
12   for total lead was 2462 mg Pb/kg dw (95% CI: 2097, 2891 mg Pb/kg dw). Gender effects were
13   observed in that male copepods were more sensitive (p = 0.038) to Pb than females as
14   determined by generalized linear model analysis.
15          Beiras and Albentosa (2003) examined the inhibition of embryo development in
16   commercial bivalves Ruditapes decussatus and Mytilus galloprovincialis after exposure to
17   concentrations of Pb(NC>3)2 in seawater.  No water chemistry parameters other than temperature
18   were reported (test conducted at 20 °C).  An ECso range for R. decussatus was reported as 156 to
19   312 |ig/L, as insufficient data were available to calculate the actual ECso. The lowest observable
20   effect concentration (LOEC) was 156 |ig/L. ForM galloprovincialis, the ECso was 221 |ig/L
21   (95% CI: 58.9, 346.3) while the LOEC was reported as 50 |ig/L.
22
23   Fish
24          The general symptoms of lead toxicity in fish include production of excess mucus,
25   lordosis, anemia,  darkening of the dorsal tail region, degeneration of the caudal fin, destruction
26   of spinal neurons, ALAD inhibition, growth inhibition, renal pathology, reproductive effects,
27   growth inhibition, and mortality (Eisler, 2000).  Toxicity in fish has been closely  correlated with
28   duration of exposure and uptake (Eisler, 2000).  The following presents information on the
29   toxicity of lead to fish in fresh and marine waters. Table 8-2.5.2 summarizes the  effects of Pb on
30   freshwater and marine fish.
31

     December 2005                           8-207        DRAFT-DO NOT QUOTE OR CITE

-------
December 2005





oo
1
o
oo

DRAFT-DO >
-^H
0
H
0
0
H
W
O
O
HH
H
W
Table 8-2.5.2. Effects of Pb to Freshwater and Marine Fish
Species
Freshwater
Fathead minnow
(Pimephales promelas)



Rainbow trout - mature
males
(Oncorhynchus mykiss)
Fathead minnow
(Pimephales promelas)

Rainbow trout - Juvenile
(Oncorhynchus mykiss)
Common carp
(Cyprinus carpio)



Endpoint:
Chemical Cone. (ug/L)

lead LC50:
chloride 810
>5,400
>5,400


lead Reproductive effects:
nitrate 10
lead Reproductive Effects:
acetate 500

lead LC50: 1000
nitrate (800 - 1400)
not LC50:
reported 6.5 cm fish- 1030
3. 5 cm fish -300



Duration of
Exposure Water Chemistry

96 h pH:
6-6.5
7-7.5
8-8.5
Hardness: 280-300
mg/L CaCO3
12 days Hardness 128 mg/L
CaC03
29 days pH:
7.5-8.5;
Hardness 130 mg/L
CaCO3;
22-25 °C
(Pb 95% soluble)
96 h pH:
7.9-8.0
DOC = 3 mg/L
Hardness (CaCO3) =
140 mg/L
96 h pH:
7.1
Temperature- 15 C
Oxygen sat. 6.4 mg/L



Comments Reference

static, measured Schubauer-Berigan
etal. (1993)



Decreased Ruby et al. (1993)
spermatocyte
development
Fewer viable eggs Weber (1993)
produced, testicular
damage

Flow through - Survival Rogers and Wood
(2003)
static -renewal - Alam and Maughan
Survival (1995)




-------
 1   Freshwater Fish
 2          Many of the toxicity modifying factors described above (e.g., pH, DOC) for invertebrates
 3   are also important modifying factors for lead toxicity to fish species.  The effects of pH on lead
 4   bioavailability and subsequent toxicity have been well studied (Sayer et al., 1989; Spry and
 5   Wiener, 1991; Schubauer-Berigan et al., 1993; Stouthart et al., 1994; MacDonald et al., 2002;
 6   Rogers and Wood, 2003).  Schubauer-Berigan et al. (1993) exposed fathead minnow to lead
 7   chloride over 96 hours. The reported LCso ranged from 810 to >5400 |ig/L at pH 6 to 6.5 and pH
 8   7 to 8.5, respectively.
 9          Water hardness also has a strong influence on the effects of lead to fish.  Chronic
10   exposure of rainbow trout fry to lead in soft water resulted in spinal deformities at 71 to
11   146 |ig/L after 2 months of exposure (Sauter et al., 1976) or 13.2 to 27 |ig/L (Davies and
12   Everhart, 1973; Davies et al., 1976), after  19 months of exposure. When exposed to lead in hard
13   water, only 0 and 10% of the trout (Oncorhynchus mykiss) developed spinal deformities at
14   measured lead concentrations of 190 and 380 |ig/L, respectively. In soft water, 44 and 97% of
15   the trout developed spinal deformities at concentrations of 31 and 62 |ig/L, respectively (Davies
16   et al., 1976). The maximum acceptable toxicant concentration (MATC) for rainbow trout fry in
17   soft water was 4.1 to 7.6 |ig/L (Davies et al., 1976), while the MATC for brook trout was 58 to
18   119 |ig/L (Holcombe et al., 1976). Histological reproductive abnormalities were noted in mature
19   male rainbow trout at 10 |ig/L lead nitrate  (Ruby et al., 1993).
20          Schwartz et al. (2004) examined the influence of NOM on lead toxicity to rainbow trout
21   exposed for 96 h in a static system. The pH of the exposure system ranged between 6.5 and 7.0,
22   temperature was maintained between 9 and 11 °C, and lead was added as PbCb. NOM from a
23   number of U.S. rivers and lakes was then added to the test system, and the LT50 was reported.
24   NOM was  found to reduce the toxic effects of Pb to rainbow trout.
25          Fish size is an important variable in determining the adverse effects of lead.  Alam and
26   Maughan (1995) exposed two different sizes of common carp (Cyprinus carpio) to lead
27   concentrations to observed effects on carp mortality. Water chemistry parameters were reported
28   (pH = 7.1;  temperature = 20 °C). Smaller fish (3.5 cm) were found to be more sensitive to Pb
29   than were larger fish (6.5 cm).  The reported LCsoS were 0.44 mg/L and 1.03  mg/L, respectively.
30
      December 2005                           8-209        DRAFT-DO NOT QUOTE OR CITE

-------
 1   Marine Fish
 2          There were no studies available that examined the toxicity of lead to marine fish species
 3   for the time period examined (1986 to present). However, Eisler (2000) reviewed available
 4   research on lead toxicity to marine species and reported studies done prior to 1986. Acute
 5   toxicity values ranged from 50 |ig/L to 300,000 |ig/L in plaice (Pleuronectesplatessa) exposed
 6   to organic and inorganic forms of lead (Eisler, 2000). Organolead compounds (e.g., tetramethyl
 7   lead, tetraethyl lead, triethyl lead, diethyl lead) were generally more toxic to plaice than
 8   inorganic lead (Maddock and Taylor, 1980).
 9
10   Other A quatic Biota
11          A paucity of data exist on the effects of lead to growth, reproduction, and survival of
12   aquatic stages of frogs and turtles. Rice et al. (1999) exposed frog larvae (Rana catesbeiand) to
13   780 jig Pb/L and two oxygen concentrations (3.5 or 7.85 mg/L) for 7 days (Table 8-2.5.3).
14   Exposure conditions included water hardness of 233 to 244 mg CaCOs/L, pH  from 7.85 to 7.9,
15   and temperature at 23 °C. Frog larvae were found to display little to no activity in the low
16   oxygen and high Pb treatment.  Hypoxia-like behavior was exhibited in larvae exposed to both
17   low and high oxygen concentrations and high Pb.  Therefore, larvae of R. catesbeiana showed
18   sensitivity to Pb and responded with hypoxia-like behavior.  Additionally, the larvae in the lead
19   treatment were found to have lost body mass relative to controls and the other treatments. Rice
20   et al. (1999) suggested that the decrease in mass likely indicated the beginning of a period of
21   reduced growth rate. Larvae exposed for longer periods (>4 weeks) were smaller and
22   metamorphosed later compared to unexposed individuals.
23          Herkovits and Perez-Coll (1991) examined lead toxicity to amphibian larvae (Bufo
24   arenarum).  Larvae (n = 50) were exposed for up to 120 h at two Pb concentrations, 8 mg Pb2+/L
25   and 16 mg Pb2+/L. Relative to controls, the 8 mg Pb2+/L treatment group exhibited 40%
26   mortality and the 16 mg Pb2+/L group 60% mortality after 120 h (p < 0.05). The authors reported
27   behavioral effects, erratic swimming, and  loss of equilibrium during the tests,  symptoms that are
28   consistent with the action of lead on the central  and peripheral nervous systems (Rice et al.,
29   1999).
      December 2005                           8-210       DRAFT-DO NOT QUOTE OR CITE

-------
December 2005


Table 8-2.5.3. Nonlethal Effects in Amphibians
Species
Frogs
(Rana ridibunda)
Frogs
(Bufo arenamm)
Frogs
Endpoint:
Chemical Concentration
lead nitrate Biochemical effects:
14,000 ug/L
Mortality:
16 mg Pb2+/L
Hypoxia-like behavior:
Duration of
Exposure
30 days
5 days
7 days
Water Chemistry
not specified
not specified
O2 = 3. 5-7.85 mg/L
Comments
Hepatic ALAD decreased by
90%
Effects reported include erratic
swimming, loss of equilibrium
Larvae used
Reference
Vogiatzis and
Loumbourdis (1999)
Herkovits and Perez-Coll
(1991)
Puce etal. (1999)
oo

to
          (Rana catesbeiana)
                                780 ug/L
Turtle Hatchlings     lead acetate  NOEL: 100 ug/g

(Trachemys scripta)               (Survival and

                                behavior)
4 weeks
pH = 7.85-7.9

Temp = 23 °C

CaCO3 = 233-244

mg/L



N/A
Exposure via single injection    Burger et al. (1998)
O
O

2
o
H

O
c
o
H
W

O
&
O
HH
H
W

-------
 1          Behavior (i.e., righting, body turnover, seeking cover), growth, and survival of
 2    hatchling slider turtles (Trachemys scripta) exposed to lead acetate were investigated in
 3    one study (Burger et al., 1998). In the first part of the study, 6-month-old hatchlings
 4    received single lead acetate injections at 50 or 100 jig/g body weight (bw). In the second
 5    part of the study, 3-week-old turtles were injected once with doses of 250, 1000 or
 6    2500 |ig/g bw. There were no differences in survival, growth, or behavior for hatchlings
 7    in the first study, however, several effects were reported from the second part of the study
 8    at doses in the range of 250 to 2,500 jig/g bw. As the dose increased, so did the plastron
 9    length (i.e., ventral section of the shell), carapace length,  and weight. The highest dose
10    group had the lowest survival rate with an LD50 of 500 jig/g bw. Behavioral  effects
11    included slower times of righting behavior and seeking cover.  The authors suggested a
12    NOEL of 100 |ig/g bw for slider turtles for survival and behavior.
13
14    8.2.5.5  Recent Studies on Effects of Lead on Decomposers
15          In this section, decomposers are defined as being bacteria and other
16    microorganisms. Many invertebrates are also potentially considered decomposers, but
17    the effects of lead to invertebrates have been described in previous sections.  There were
18    no toxicity studies located on the effects of lead to aquatic decomposers in the time
19    period of interest.
20
21    8.2.5.6  Summary
22          Lead in all its forms is known to cause adverse effects in aquatic organisms
23    (Eisler, 2000). Effects to algal growth have been observed at lead concentrations ranging
24    from 100 to 200,000 |ig/L. Clinical signs of lead toxicity in plants include the
25    deformation  and disintegration of algae cells and  a shortened exponential growth phase.
26    Other effects of lead include a blocking of the pathways that lead to pigment  synthesis,
27    thus affecting photosynthesis,  cell cycle and  division, and ultimately resulting in death.
28    The toxicity  of lead to macrophyte growth has been studied using Spirodelapolyrhiza,
29    Azollapinnata, and Lemna gibba.  Test durations ranged  from 4 to 25 days and test
30    concentrations ranged between 49.7 and 500,000  |ig/L.
      December 2005                        8-212    DRAFT-DO NOT QUOTE OR CITE

-------
 1          Waterborne lead is highly toxic to aquatic organisms, with toxicity varying with
 2    the species and life stage tested, duration of exposure, form of lead tested, and water
 3    quality characteristics. Among the species tested, aquatic invertebrates, such as
 4    amphipods and water fleas, were the most sensitive to the effects of lead, with adverse
 5    effects being reported at concentrations ranging from 0.45 to 8000 |ig/L. Freshwater fish
 6    demonstrated adverse effects at concentrations ranging from 10 to >5400 |ig/L,
 7    depending generally upon water quality parameters. Amphibians tend to be relatively
 8    tolerant of lead; however, they may exhibit decreased enzyme activity (e.g., ALAD
 9    reduction) and changes in behavior (e.g., hypoxia response behavior). Lead tends to be
10    more toxic with longer-term exposures.
11
12    8.2.6   Effects of Lead on Natural Aquatic Ecosystems
13    8.2.6.1  Introduction
14          This section discusses the effects of lead on natural aquatic ecosystems. Such
15    effects include changes in species composition and richness, ecosystem function, and
16    energy flow due to lead stress. The format of this section generally follows a conceptual
17    framework for discussing the effects of a stressor such as lead on an ecosystem. This
18    conceptual framework was  developed by the EPA Science Advisory Board (Young and
19    Sanzone, 2002). The essential attributes used to describe ecological condition include:
20    landscape condition, biotic  condition, chemical and physical characteristics, ecological
21    processes, hydrology and geomorphology and natural disturbance regimes.  The majority
22    of the published literature pertaining to lead and  aquatic ecosystems focuses on the biotic
23    condition, one of several  essential attributes of an ecosystem as described in Young and
24    Sanzone (2002). For the biotic condition, the SAB framework identifies community
25    extent, community composition, trophic structure, community dynamics, and physical
26    structure as factors for assessing ecosystem health. Other factors for assessing the biotic
27    condition such as effects  of lead on species, populations, and organism conditions (e.g.,
28    physiological status) were discussed in Sections  8.2.4 and 8.2.5.
29          For natural aquatic ecosystems, the focus  of study in the general literature has
30    been on evaluating ecological stress where the sources of lead were from urban and
31    mining effluents (Poulton et al., 1995; Deacon et al., 2001; Mucha et al., 2003). The


      December 2005                       8-213    DRAFT-DO NOT QUOTE OR CITE

-------
 1    statistical methods used when evaluating the effects of lead on aquatic ecosystems are
 2    important, as more than one variable may be related to the observed effect.  Studied
 3    variables include water hardness, pH, temperature, and physical factors such as
 4    embeddedness, dominant substrate, and velocity.  In most cases single variable statistical
 5    techniques were used to evaluate the data. However, in other cases multivariate
 6    techniques were used. Therefore, where appropriate, some detail on the statistical
 7    methods used is presented.
 8          Although most of the available studies discussed in this section focus on the biotic
 9    condition, one case study examining multiple components of the EPA conceptual
10    framework is also included. The remainder of this section describes the effects of lead on
11    the biotic condition.
12
13    8.2.6.2  Case Study: Coeur d'Alene River Watershed
14          The Coeur d'Alene River watershed is an area of Idaho impacted by lead and other
15    metals from historic mining waste releases. Maret et al. (2003) examined several
16    ecological components to determine any negative associations with metals and the
17    watershed communities. The variables examined and associated ecological conditions
18    are presented in Table 8-2.6.1.  In addition to measurements of non-metal variables (e.g.,
19    dissolved oxygen levels, water temperature and pH, embeddedness), Cd, Pb, and Zn
20    levels were also compared in affected sites versus reference sites.
21          Some of the above non-metal variables are important to macroinvertebrate
22    communities.  For example, a stream with highly embedded substrate can have a lower
23    number of individuals within a species or a different species composition compared to a
24    stream with less embeddedness (Waters, 1995). Macroinvertebrates from the
25    Ephemeroptera (mayflies), Plecoptera (stoneflies), and Trichoptera (caddisflies) (EPT)
26    group inhabit the surface of cobble and the interstitial spaces between and underneath
27    cobble. When substrate is embedded, these interstitial spaces are filled, leaving less
28    habitat space for EPT taxa.  In  another example, water temperature is important; some
29    macroinvertebrates (e.g., stoneflies) are usually only found in cooler water (Harper and
30    Stewart, 1984).
31

      December 2005                       8-214   DRAFT-DO NOT QUOTE OR CITE

-------
              Table 8-2.6.1. Ecological Attributed Studies by Maret et al. (2003)
                                in the Coeur d'Alene Watershed
      Ecological Attribute
Subcategory
Measure
      Landscape condition
      Biotic Condition
Area! extent
landscape pattern

Organism condition
population
structure/dynamics
      Chemical/physical characteristics   Chemical/physical
                                    parameters
      Ecological processes

      Hydrology /geomorphology
Channel morphology and
distribution
      Natural disturbance regimes
Basin area (km)
Production mine density/km2

Caddisfly tissue concentrations
(mg/kg)
Number of EPT taxa
Density of EPT individuals
(no./m2)

Dissolved oxygen (mg/L)
Specific conductance (uS/cm)
Water temperature (E °C)
pH
Water hardness (mg/L)
Total NO3 (mg/L)
Total P (mg/L)
Dissolved NH3 (mg/L)
Sediment Cd, Pb, Zn (mg/kg)
Dissolved Cd, Pb, Zn in water
(mg/L)

None measured

Site elevation (m)
Stream gradient (%)
Stream discharge (m3/s)
Stream width (m)
Stream depth (m)
Open canopy (%)
Stream velocity (m/s)
Embeddedness (%)
Dominant substrate (mm)

None measured
1           Of the variables examined only metal concentrations, mine density, site elevation,

2    and water temperature were significantly different between reference and mine-affected

3    sites.  A Mann-Whitney t-test was used to evaluate statistical differences between

4    reference and test sites for physical and water quality parameters, while Spearman's rank

5    correlation matrices were used to compare all  possible response and explanatory

6    variables. Lead concentrations were significantly correlated with the number of mines in
     December 2005
         8-215    DRAFT-DO NOT QUOTE OR CITE

-------
 1    proximity to the watershed.  Lead concentrations in sediment and water were strongly
 2    correlated to lead levels in whole caddisflies, r2 = 0.90 and 0.63, respectively.
 3    Furthermore, mine density was significantly correlated to lead in tissue, r2 = 0.64.
 4    Although temperature was significantly different between reference and mine-affected
 5    sites, temperature conditions were concluded to be non-limiting to aquatic life. For
 6    example, reference and mine-affected sites had at least 15  and 13 obligate cold-water
 7    taxa, respectively.
 8          A significant negative correlation between lead in the water column (0.5 to
 9    30 ug/L dissolved) and total taxa richness, EPT taxa richness, and the number of metal-
10    sensitive mayfly species was observed.  Similar, significant negative correlations were
11    found between sediment lead levels (132 to 6252 ug/g) and the same macroinvertebrate
12    community metrics and caddisfly tissue levels. Negative correlations were also found
13    between Cd and Zn in the water and sediment and the macroinvertebrate community
14    metrics. In an analysis of cumulative toxicity, lead was judged to be the most significant
15    metal in sediment related to the cumulative toxicity measured.  This study provided
16    multiple lines of evidence (i.e., mine  density, metal concentrations, bioaccumulation in
17    caddisfly tissue and benthic invertebrate assemblage structure) of the negative impacts of
18    mining in the Coeur d'Alene River, suggesting that lead (and other metals) were primary
19    contributors to the effects observed in the Coeur d'Alene River watershed (Maret et al.,
20    2003).
21
22    8.2.6.3  Biotic Condition
23          In an evaluation of the biotic condition, the SAB framework described by Young
24    and Sanzone (2002) identifies  community extent, community composition, trophic
25    structure, community dynamics, and physical structure as essential ecological  attributes
26    for assessing ecosystem health. The following two sections describes the effects of lead
27    on community  composition, community dynamics, and trophic structure. To date, no
28    available studies were located  on the  effects of lead on physical structure (e.g., change in
29    tree canopy height, ecosystem  succession).
30
      December 2005                        8-216    DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.6.3.1   Ecosystems and Communities, Community Composition
 2          To measure community composition, an inventory of the species/taxa found in the
 3    ecological system must be conducted.  According the SAB framework, useful measures
 4    of composition include the total number of species or taxonomic units, their relative
 5    abundance, presence and abundance of native and non-native species, and information on
 6    the presence and abundance of focal or special interest species (Young and Sanzone,
 7    2002). Focal or special species of interest can be those that play a critical role in
 8    ecosystem processes such as flows of materials or energy within complex food-webs
 9    (Young and Sanzone, 2002). Community composition as assessed in lead studies has
10    included the following  measures.
11
12       •  Changes in energy flow or nutrient cycling:
13           o   Increased or decreased respiration or biomass
14           o   Increased or decreased turnover/cycling of nutrients
15
16       •  Changes to community structure:
17           o   Reduced species abundance (i.e., the total number of individuals of a species
18               within a given area or community)
19           o   Reduced species richness (i.e., the number of different species present in
20               a community)
21           o   Reduced species diversity (i.e., a measure of both species abundance and
22               species richness)
23          Investigators have evaluated the effects of lead on aquatic communities through
24    microcosm and mesocosm studies in natural aquatic systems. Field studies in the general
25    literature have focused on natural systems that were affected by metal stress from various
26    anthropogenic sources. In most of those natural systems,  the sources evaluated were
27    from direct mining waste inputs, rather than atmospheric deposition, of lead. Studies
28    published since the  1986 Pb AQCD that describe the effects of  lead on natural aquatic
29    ecosystems are presented below and summarized in Table 8-2.6.2. Studies included here
30    evaluated the effects of lead on watersheds, landscapes, aquatic ecosystems, aquatic
31    communities, biodiversity, lakes, rivers, streams, estuaries, wetlands, and species
32    interaction.
33

      December 2005                       8-217    DRAFT-DO NOT QUOTE OR CITE

-------
 1   Aquatic Microcosm Studies
 2          The examination of simulated aquatic ecosystems (i.e., microcosms) provides
 3   limited information on the effects of pollutants on natural systems. Microcosm studies
 4   typically focus on only a few aspects of the natural system and do not incorporate all of
 5   the ecological, chemical, or biological interactions. Nevertheless, a few microcosm
 6   studies have been conducted that indicate potential effects of lead on the community
 7   structure of aquatic ecosystems. Fernandez-Leborans and Antonio-Garcia (1988)
 8   evaluated the effect of lead on a natural community of freshwater protozoans in simulated
 9   aquatic ecosystems and found a reduction in the abundance and composition of protozoan
10   species with increasing lead concentrations (0.05 to 1.0 mg/L) compared to controls.
11   Studies with marine protozoan communities in laboratory microcosms indicated that
12   waterborne lead exposure reduced protozoan abundance, biomass, and diversity at
13   concentrations of 0.02 to 1.0 mg/L Pb.  (Fernandez-Leborans and Novillo, 1992, 1994).
14          Austen and McEvoy (1997) studied the effects of lead on an estuarine meiobenthic
15   community (mainly nematodes) in a microcosm setting using sediment samples collected
16   offshore from England. A multivariate analysis of similarities (ANOSIM) test with
17   square root-transformed data was used to evaluate differences between treatments and
18   controls.  Lead was found to significantly affect species abundance at 1343 mg/kg dw
19   relative to a control at 56 mg/kg dw, but no significant adverse effects were observed at
20   the highest dose tested, 1580 mg/kg dw.  The authors did not explain why the 1580
21   mg/kg dw dose was not significant while the 1343 mg/kg dw dose was.  None of the lead
22   exposures were significantly different than the controls based on separate univariate tests
23   of abundance, richness, and diversity.  There were no other confounding metals in the
24   lead tests, as the experiments were with a single metal dose. In one other mesocosm
25   study, the effects of a mixture of metals (Cu, Cd, Pb, Hg, and Zn) on a salt marsh
26   meiofaunal community were evaluated (Millward et al., 2001).  After 30 days exposure,
27   significant reductions in copepod, gastropod, and bivalve abundances were observed at
28   the highest lead exposure concentration,  177 mg/kg dw. Ostracods and nematodes were
29   not affected.  The authors believed that the response of the meiofauna taxa to metals was
30   in part due to the various feeding strategies in that deposit feeders were most affected.
31

     December 2005                        8-218    DRAFT-DO NOT QUOTE OR CITE

-------
 1   Natural Aquatic Ecosystem Studies
 2          Lead stress in aquatic ecosystems has also been evaluated in natural communities.
 3    Studies examining community-scale endpoints, however, are complex, and interpretation
 4    can be confounded by the variability found in natural systems and the presence of
 5    multiple stressors. Natural systems frequently contain multiple metals, making it difficult
 6    to attribute observed adverse effects to single metals.  For example, macroinvertebrate
 7    communities have been widely studied with respect to metals contamination and
 8    community composition and species richness (Winner et al., 1980; Chadwick et al., 1986;
 9    Clements, 1994). In these studies, multiple metals are evaluated and correlations
10    between observed community  level effects are ascertained.  The results often indicate a
11    correlation between the presence of one or more metals (or total metals) and the negative
12    effects observed. While, correlation may imply a relationship between two variables, it
13    does not imply causation of effects.  The following studies suggest an association
14    between lead concentration and an alteration of community structure and function (see
15    summary in Table 8-2.6.2):
16
17   Reduced Primary Productivity and Respiration
18          Jayaraj et al. (1992) examined the effects of lead on primary productivity and
19    respiration in an algal community of two water bodies. Concentrations of lead in water
20    (6 to 80 mg/L) were found to significantly reduce primary productivity and increase
21    respiration.  The authors suggested that increased respiration indicated a greater tolerance
22    to or adaptive mechanisms of the resident heterotrophs to cope with lead stress.
23
24   Alterations of Community Structure
25          Deacon et al. (2001) studied a macroinvertebrate community in mine-affected
26    waters of Colorado.  Initially, transplanted bryophytes were used to assess whether metals
27    could bioaccumulate at various mine-affected and unaffected sites (Deacon et  al., 2001;
28    Mize and Deacon, 2002).  Lead was bioaccumulated by the bryophytes, and median
29    tissue concentrations at mine-affected sites (34 to 299 ug/g dw) were higher than at
30    reference sites (2.5 to 14.7 ug/g dw).  Lead concentrations in surface water and sediment
31    ranged from <0.001 to 0.02 mg/L and 145 to 850 mg/kg dw (<63 urn fraction),
      December 2005                        8-219    DRAFT-DO NOT QUOTE OR CITE

-------
December 2005





oo
to
to
o


O
H
6
o
2
0
H
O
o
H
W
O
O
HH
H
W
Table 8-2.6.2. Essential Ecological
Category Species
Biotic Condition
Ecosystems and Protozoan community
Communities-
Community
Composition
Protozoan community


Protist community
Meiofauna community

Algal community
Algal community
Meiobenthic
community (primarily
nematodes)




Condition
Measures

Reduced species
abundance and
diversity
Reduced species
abundance


Reduced species
abundance and
diversity
Reduced abundance

Increased respiration
Decreased primary
productivity
Reduced species
abundance
No effect on
abundance




Attributes for
Exposure
Medium

Marine water

Freshwater
water


Marine water
Marine
sediment
Freshwater
Freshwater
Marine
sediment




Natural Aquatic
Location

Laboratory microcosm

Laboratory microcosm


Laboratory microcosm
Laboratory microcosm

Domestic water
stabilization pond
Sharana Basaveshwara
Tank, India
Laboratory microcosm




Ecosystems Affected by Lead
Other
Exposure Metals
Concentrations Present

0.02- 0.05 mg/L N

0.05-1 mg/L N


1 mg/L N
177mg/kgdw Y

25-80 mg/L ?
6-32 mg/L ?
1343mg/kgdw N
1580mg/kgdw





Reference

Femandez-
Leborans and
Novillo(1992)
Femandez-
Leborans and
Antonio-Garcia
(1988)
Femandez-
Leborans and
Novillo(1994)
Millward et al.
(2001)
Jayaraj etal. (1992)
Jayaraj etal. (1992)
Austen and
McEvoy(1997)





-------
O
o
3
^ '
CD
^
to
O
O
u\









oo
i
to
to
1—1



O

rr5
H
6
o

O
H
O
o
w
o

Q
1 — I
H
W
Table 8-2.6.2 (cont'd).


Category Species
Macroinvertebrate
community





Macroinvertebrate
community

Macroinvertebrate
community

Fish, crustacean and
macroinvertebrate
community

Chironomid community




Macroinvertebrate
community




Macroinvertebrate
community





Essential Ecological Attributes for Natural Aquatic Ecosystems Affected by

Condition
Measures
Lower total
abundance, decreased
taxa, and EPT
richness, larger
percentage of tolerant
species of benthic
macroinvertebrates.
Negatively correlated
with species richness
and diversity indices
Reduced species
abundance

Correlation with
changes in species
abundance and
distribution
Reduced chironomid
richness



Lead in tissues
negatively correlated
with taxa richness,
EPT richness,
chironomid richness,
and species density.
Lead in tissues
negatively correlated
with EPT richness
and abundance.




Exposure
Medium
Freshwater
and sediment





Estuary
sediment

Freshwater
sediment

Marine
Sediment


Whole
organism
residue


Whole
organism
residue



Biofilm
residues







Location
Mining sites in the
Upper Colorado Basin





Douro Estuary, Portugal


River 111 and tributaries,
France

Spencer Gulf, South
Australia


New Brunswick, Canada




Clark Fork River, MT





Boulder River, MT







Exposure
Concentrations
<0.001-0.02mg/L
145-850 mg/kg dw
(<63 uM fraction)




0.25-1 92 mg/kg dw


1-16 mg/kg dw


156-5270 mg/kg
dw


40. 3-1, 387 mg/kg
dw (periphyton)
1.6-131 mg/kg dw
(chironomid tissue)

32.2-67.1 mg/kg
dw




32-1 540 mg/kg dw






Other
Metals
Present
Y






Y


Y


Y



Y




Y





Y






Lead


Reference
Deacon et al.
(2001);Mizeand
Deacon, (2002)




Mucha et al. (2003)


Rosso etal. (1994)


Ward and Young
(1982);
Ward and
Hutchings(1996)
Swansburg et al.
(2002)



Poulton et al.
(1995)




Rhea et al. (2004)







-------
o

O

cr

^
to
O
O
(^









Table 8-2.6.2 (cont'd)


Category Species

Macroinvertebrate
Community


Fish Community



Ecosystems and Snails and tadpoles
Communities-
. Essential Ecological Attributes for Natural Aquatic Ecosystems Affected by Lead


Condition Measures

Lead in tissues and
sediment not correlated
to diversity and
richness
Lead in tissues and
sediment not correlated
to diversity and
richness
Lead affected predator-
prey interactions

Exposure
Medium

Sediment
and whole
organism
residue
Sediment
and whole
organism
residue
Sediment



Location

Aquashicola Creek
tributaries, Palmerton,
PA

Aquashicola Creek
tributaries, Palmerton,
PA

Outdoor mini-
ecosystems

Exposure
Concentrations

7.5-59.5 mg/kg dw
(sediment)
0.25-6.03 mg/kg dw
(macroinvertebrates)
7.5-59.5 mg/kg dw
(sediment)
0.1 -0.86 mg/kg dw
(fish)
Not cited

Other
Metals
Present Reference





Y Carline and Jobsis
(1993)





Y Carline and Jobsis
(1993)


Y Lefcort et al.




(1999)

oo
to
to
to
Community
Dynamics and
Trophic Structure
                               Snails and caddisflies
                                           No avoidance of
                                           predator by snail.
                                           Caddisfly did respond
                                           to predator
Water         Field microcosm for
               snail; in-stream
               disturbance for
               caddisfly
27.7-277.6 mg/kg
dw (snail tissue)
223-13,507 mg/kg
dw (caddisfly tissue)
Y
Lefcort et al. (2000)
£
£
H
I
0
O

H
O
O
H
W
O
O
HH
H
W
Fathead minnow Feeding behavior Water Laboratory microcosm 0.5-l.Omg/L
altered


American toad No avoidance of lead Water Laboratory microcosm 0.5-l.Omg/L

Mummichog Feeding behavior Water Laboratory 0.3-l.Omg/L
altered and predator
avoidance affected







N Weber (1996)



N Steeleetal. (1991)

N Weis and Weis
(1998)









-------
 1    respectively. The same sites were also evaluated for the effects of various metals on
 2    macroinvertebrate communities.  Values of total abundance, taxa richness, mayfly, and stonefly
 3    abundance were reduced at mining sites. Lead levels along with Cd, Cu, and Zn were correlated
 4    with reduced abundance and diversity indices.
 5          Macrobenthic communities studied in an estuary off Portugal were affected by lead at a
 6    range from 0.25 to 192 mg/kg dw (Mucha et al., 2003).  Species richness was decreased in areas
 7    with increased lead concentrations in the sediment.  Interpretation of lead effects was
 8    complicated by other non-metal stressors, namely sediment particle size and organic matter
 9    content. Furthermore, other metals were present (e.g., Al, Cu, Cr, Mn, Zn) and may have
10    affected the community (Mucha et al., 2003).
11          The effects of lead on oligochaetes in the 111 River and its tributaries in France were
12    evaluated by Rosso et al. (1994). Lead in sediment (5 to 16 ug/g dw at affected sites) was
13    positively correlated to the abundance of the oligochaete, Nais sp., and negatively correlated to
14    Tubificidae abundance. Lead was the only metal that was positively correlated to Nais species,
15    while other metals were negatively correlated to Tubificidae (Rosso et al., 1994).
16          The effects of metals and particle size on structuring epibenthic sea grass fauna (fish,
17    mollusks, crustaceans, and polychaetes) was evaluated near a lead smelter in South Australia
18    (Ward and Young, 1982; Ward and Hutchings, 1996). Effluent from the smelter was the primary
19    source of lead and other metal contamination.  Species richness and composition were evaluated
20    near the lead smelter along with metal concentrations in sediment. Lead levels in sediment (up
21    to 5270 mg/kg dw) correlated to the negative effects on species richness and composition, while
22    the other metals evaluated had similar correlations.  Therefore, lead alone could not be identified
23    as the sole metal causing stress.
24
25    Tissue Bioaccumulation Associated with Alterations of Community Structure
26          Several studies have examined the bioaccumulation of lead in aquatic systems with
27    indices of community structure and function. A focused study on changes in Chironomidae
28    community composition in relation to metal mines (New Brunswick, Canada) identified changes
29    in Chironomidae richness (Swansburg et al., 2002). Lead was not detected (detection limit not
30    given for any matrix) in the water column at any site.  However, lead levels in periphyton were
31    significantly higher at mining sites (40.3 to  1387 mg/kg  dw) compared to reference sites (not
      December 2005                           8-223       DRAFT-DO NOT QUOTE OR CITE

-------
 1    detected [ND], 33.3 mg/kg dw).  Furthermore, lead in chironomids was significantly higher at
 2    mine-affected sites (1.6 to 131 mg/kg dw) compared to reference sites (ND,10.2 mg/kg dw). The
 3    concentrations in biota indicate that lead is mobile and available to the aquatic community even
 4    though water concentrations were undetectable.  Chironomidae richness was reduced at the sites
 5    receiving mining effluent containing Pb, Cd, Cu, and Zn.
 6          In another study, macroinvertebrate lead tissue concentrations (32.2 to 67.1 mg/kg dw at
 7    affected sites) collected from the Clark Fork River, Montana correlated negatively with total
 8    richness, EPT richness, and density (Poulton et al., 1995). Mean lead levels were as high as
 9    67.1 mg/kg dw at sites most affected by lead. However, other metals, including Cd, Cu, and Zn,
10    also were negatively correlated with total richness and EPT richness. Therefore, attribution of
11    the observed effects to lead is difficult, as other metals may be contributing factors.
12          In Montana, the potential effects of metals on macroinvertebrate communities in the
13    Boulder River watershed were evaluated (Rhea et al., 2004).  Similar to the approach taken by
14    Poulton  et al. (1995), the effects on richness and abundance of EPT taxa were compared to metal
15    concentrations in tissue (i.e., biofilm and macroinvertebrates). Lead levels in biofilm (32 to
16    1540 mg/kg dw) were significantly correlated with habitat scores and macroinvertebrate indices
17    (e.g., EPT taxa).  However, macroinvertebrate tissue lead levels were not significantly correlated
18    with macroinvertebrate community level metrics. As with most natural systems with potential
19    mine impacts, other metals also correlated with community level effects. However, the authors
20    indicated that  lead concentrations in biofilm appeared to have the most significant impact on
21    macroinvertebrate metrics.
22          A detailed investigation of sediment, macroinvertebrates, and fish was conducted for
23    tributaries in the Aquashicola Creek watershed near a former zinc smelter in Palmerton, PA
24    (Carline and Jobsis, 1993). The smelter deposited large amounts of Cd, Cu, Pb, and Zn on the
25    surrounding landscape during its operation from 1898 to 1980. The goal of the study was to
26    evaluate if there was a trend in the metal levels in sediment, macroinvertebrate and fish tissue,
27    and community indices going away from the smelter. Sites were chosen, from 7.8 to 24.6 km
28    from the smelter. There were no clear associations between proximity to the smelter and lead
29    levels in sediment, macroinvertebrate tissue, and fish tissue.  Furthermore, there were no
30    associations between proximity to the smelter and macroinvertebrate and fish diversity and
      December 2005                           8-224       DRAFT-DO NOT QUOTE OR CITE

-------
 1    richness.  The authors suggested that the transport of metals in the watershed has decreased since
 2    the smelter ceased operation, and thereby no effects were observed.
 3
 4    8.2.6.3.2   Ecosystems and Communities, Community Dynamics, and Trophic Structure
 5          As described in the SAB framework, community dynamics include interspecies
 6    interactions such as competition, predation, and succession (Young and Sanzone, 2002).
 7    Measures of biotic interactions (e.g., levels of seed dispersal, prevalence of disease in
 8    populations of focal species) provide important information about community condition.  If the
 9    community dynamics are disrupted, then the trophic structure may also be disrupted. According
10    to the SAB framework, trophic structure refers to the distribution of species/taxa and functional
11    groups across trophic levels.  Measures of trophic structure include food web complexity and the
12    presence/absence of top predators or dominant herbivores.  Therefore, this section discusses how
13    aquatic species interactions can be affected by lead.  Examples of species interactions can
14    include:
15       •  Predator-prey interactions (e.g., reduced avoidance of predators)
16       •  Prey consumption rate (e.g., increase or decrease in feeding)
17       •   Species competition (e.g., interference with another species, increased aggressive
18          behavior)
19       •   Species tolerance/sensitivity (e.g., the emergence of a dominant species due to
20          contaminant tolerance or sensitivity)
21          Species interactions are relevant to a discussion about the effects of lead on natural
22    aquatic ecosystems, because effects on species interactions could potentially affect ecosystem
23    function and diversity. Some examples of lead induced changes in species interactions are
24    presented below (see summary in Table 8-2.6.2).
25
26    Predator-Prey Interactions
27          Lefcort et al. (1999) examined the competitive and predator avoidance behaviors of snails
28    and tadpoles in outdoor mini-ecosystems with sediment from a metals-contaminated Superfund
29    site (i.e., Pb, Zn, Cd). Previous investigations of aquatic invertebrates and vertebrates yielded
30    lead tissue concentrations of 9 to 3800 mg/kg dw and 0.3 to 55 mg/kg dw, respectively.  Several
31    species interactions were studied in the presence of metal-contaminated sediment:

      December 2005                           8-225        DRAFT-DO NOT QUOTE OR CITE

-------
 1          Snails and tadpoles have similar dietary behaviors. Thus, when placed in the same habitat
 2    they will compete for the same food items and negatively affect one another. However, when
 3    tadpoles exposed to a predator (i.e., through biweekly additions of 20 mL water from tanks
 4    housing sunfish—10 mL from sunfish-fed snails, 10 mL from sunfish-fed tadpoles) were placed
 5    with snails, the tadpoles reduced sediment ingestion, while snails increased ingestion. Thus,
 6    snails were exposed to greater quantities of metals in sediment.
 7          In an uncontaminated environment, snail recruitment (i.e., reproduction) was reduced in
 8    the presence of tadpoles. The addition of tadpoles increased the competition for food in the form
 9    of floating algae and the snails switched to feeding on algae that grew on the sediment. This
10    decrease was due to competition alone. The effects on snail recruitment were even higher when
11    tadpoles, the influence of a predator (i.e.,  sunfish extract), and metals in the sediment were all
12    present. However, the predator effect was indirect in that the tadpoles hid in the algae mats
13    forcing the snails to feed primarily on the benthic algae that grew on the sediment with high
14    metal levels. Furthermore, although not significant,  lead levels  in snails were higher when
15    tadpoles and sunfish extract were present than when only metals in the sediment were present.
16          Finally, snail predator avoidance was assessed.  Snails (control and lead-exposed) were
17    stimulated with a predator indicator (i.e., crushed snails and an extract of crushed snail). Control
18    snails changed behaviors in the presence of the predator indicator, while exposed snails did not
19    alter their behavior.  The authors suggested that metal exposure caused behavioral changes that
20    alter competitive interactions and the perception of predators by the snails. Thus, lead may
21    affect the predator avoidance response of snails.
22          In further study, Lefcort et al. (2000) examined the predator avoidance behaviors of snails
23    and caddisflies.  In separate experiments,  the avoidance behavior of the snail, Physella
24    columbiana, and four caddisfly genera (Agrypnia, Hydropsyche, Arctopsyche, Neothremma)
25    were evaluated.  The snails were collected from reference lakes and lakes downstream of the
26    Bunker Hill Superfund site. The snails from the affected lakes generally had higher cadmium,
27    lead, and zinc tissue levels implying previous exposure to these metals.  Snail predator  avoidance
28    behavior was tested by exposure to crushed snail  extract. Snails from the affected lakes did not
29    reduce  their activity when exposed to the  snail extract, implying a reduced predator avoidance.
30    The lack of response may make the snails at the affected lakes more prone to predation.
      December 2005                           8-226        DRAFT-DO NOT QUOTE OR CITE

-------
 1          The caddisflies were evaluated at 36 sites from six different streams.  As with the snails,
 2    the caddisflies from the affected streams had higher cadmium, lead and zinc tissue levels. The
 3    time for caddisfly larvae to respond (i.e., how long immobile) to disturbance (i.e., lifted from
 4    water for 3 seconds and moved to a new location) was evaluated.  There was no correlation
 5    between tissue metal level and any response variable (Lefcort et al., 2000). Therefore, the
 6    authors concluded that preexposure to metals did not reduce predator avoidance for caddisflies.
 7          Weber (1996) examined juvenile fathead minnows exposed to 0, 0.5, or 1.0 ppm lead in
 8    water during a 2-week preexposure and 2-week testing period (4 weeks total exposure).  Feeding
 9    behavior was evaluated by presenting two prey sizes (2-day-old and  7-day-old Daphnia magna).
10    Control fish began switching from larger, more difficult-to-capture 7-day-old daphnids to
11    smaller, easier-to-catch 2-day-old prey by day 3. Lead-exposed fish displayed significant
12    switching at day 3 (at 0.5 ppm) or day 10 (at 1.0 ppm). Thus, exposure to lead delayed the
13    altering of prey size choices to less energetically costly prey.
14          Lefcort et al. (1998) exposed spotted frogs (Rana luteiventris) to 0.05 to 50 ppm Pb in
15    water for 3 weeks. High levels of lead reduced the fright response of tadpoles; suggesting a
16    reduced avoi dance of predators.
17          Bullfrog larvae exposed to lead in water (0.78 mg/L) and high or low dissolved oxygen
18    were monitored for respiratory surfacing behavior (Rice et al., 1999). Larvae had a significantly
19    increased number of trips to the water surface regardless of oxygen content. Thus, the authors
20    suggest that lead may affect oxygen uptake such that larvae are under greater predation pressure
21    due to increased time spent at the surface.
22          Weis and Weis (1998) evaluated the effect of lead exposure on mummichog (Fundulus
23    heteroclitus) larvae prey capture rate, swimming behavior, and predator avoidance. Prey capture
24    rates were affected after 4 weeks exposure at 1.0 mg/L lead. The larvae were also more
25    vulnerable to predation by grass shrimp (Palaemonetespugio) at  1.0 mg/L lead.  Finally, the
26    swimming behavior of mummichog larvae was affected at 0.3 and 1.0 mg/L lead. Once  the
27    larvae were no longer exposed to lead, they recovered their ability to capture prey and avoid
28    predators.
29          Clearly,  exposure to lead does affect the predator-prey interactions and the ability of prey
30    to avoid predators. The effect of lead on these ecological functions may alter community
31    dynamics.

      December 2005                           8-227       DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.2.6.4  Summary
 2          The effects of lead have primarily been studied in instances of point source pollution
 3    rather than area-wide atmospheric deposition; thus, the effects of atmospheric lead on ecological
 4    condition remains to be defined.  The evaluation of point source lead within the EPA Ecological
 5    Condition Framework has been examined primarily in relation to biotic conditions.  The
 6    available literature focuses on studies describing the effects of lead in natural aquatic ecosystems
 7    with regard to community composition and species interactions. The effects of lead on the biotic
 8    condition of natural aquatic systems can be summarized as follows:  there is a paucity of data in
 9    the general literature that explores the effects of lead in conjunction with all or several of the
10    various components of ecological condition as defined by the EPA.  However, numerous studies
11    are available associating the presence of lead with effects on biotic conditions.
12          In simulated microcosms or natural systems, environmental exposure to lead in water and
13    sediment has been shown to affect energy flow and nutrient cycling and benthic community
14    structure. In field studies, lead contamination has been shown to significantly alter the aquatic
15    environment through bioaccumulation and alterations of community structure and function.
16    Exposure to lead in laboratory studies and simulated ecosystems may alter species competitive
17    behaviors, predator-prey interactions, and contaminant avoidance behaviors.  Alteration of these
18    interactions may have negative effects on species abundance and community structure. In
19    natural aquatic ecosystems, lead is often found coexisting with other metals and other stressors.
20    Thus, understanding the effects of lead in natural systems is challenging given that observed
21    effects may be due to cumulative toxicity from multiple stressors.
22
23
24    8.3     CRITICAL LOADS FOR LEAD IN TERRESTRIAL AND
25            AQUATIC ECOSYSTEMS
26    8.3.1    Introduction
27          This section defines critical loads, describes various concepts and methods that are related
28    to the estimation of critical loads, and provides a review of the relevant literature on critical
29    loads.
30
      December 2005                           8-228        DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.3.1.1  Definitions
 2          Critical loads are defined in a variety of ways depending on the chemicals and endpoints
 3    of concern (Paces, 1998; Skeffmgton, 1999; U.S. Environmental Protection Agency, 2004a).
 4    For the purposes of this section, critical loads are defined as threshold deposition rates of air
 5    pollutants that current knowledge indicates will not cause long-term adverse effects to ecosystem
 6    structure and function. A critical load is related to an ecosystem's sensitivity to anthropological
 7    inputs of a specific chemical.  If future inputs of a chemical exceed the critical load for an
 8    ecosystem, the chemical is expected to reach or persist at potentially toxic levels in the future.
 9    A critical load indicates a potential for future impacts only; a current exceedance of a critical
10    load does not specify whether the current deposition rate of a chemical presents a hazard to the
11    ecosystem.
12          In order to determine a critical load, the lowest concentration in the receiving medium that
13    poses a potential hazard to a defined ecosystem must first be determined. This concentration,
14    known in the critical loads literature as the critical limit (De Vries et al., 2004), is equal to the
15    effects-based criteria for the most sensitive endpoint in the ecosystem. The critical limit
16    indicates the current potential for adverse effects to an ecosystem.
17          In contrast to a critical load, a stand-still load is the highest deposition rate of a chemical
18    that will not result in future increases of its concentrations in the environmental media,
19    regardless of the potential for adverse effects at those concentrations.  Stand-still loads are also
20    called "acceptable loads" or critical loads calculated using a "stand-still" approach (De Vries
21    et al., 2004) and should not be confused with effects-based critical loads.
22
23    8.3.1.2  Historical Perspective
24          In the 1960s, scientists demonstrated that sulfur emissions on the European  continent
25    were contributing to the acidification of Scandinavian lakes.  During the 1970s, evidence
26    mounted that air pollutants could travel thousands of miles before deposition occurred, implying
27    that international cooperation was necessary to control acidification. To this  end, the European
28    Community (EC) and 34 governments signed the Convention on Long-range Transboundary of
29    Air Pollution (CLRTAP) in 1979 under the auspices of the United Nations  Economic
30    Commission for Europe (United Nations Economic Commission for Europe (UNECE), 2004).
      December 2005                            8-229       DRAFT-DO NOT QUOTE OR CITE

-------
 1          CLRTAP has since been extended to include eight protocols that regulate air pollutants
 2    such as sulfur, nitrogen oxides, heavy metals, persistent organic pollutants, volatile organic
 3    compounds, and ozone.  In 1988, CLRTAP adopted the critical-load concept, making it basic to
 4    the future development of international agreements concerning limitation of the emissions of air
 5    pollutants. In 1991, The Coordination Center for Effects (CCE) issued a Technical Report
 6    entitled "Mapping Critical Loads for Europe" which presented the first maps of critical loads that
 7    were produced as part of the work conducted under the UNECE. Each individual country
 8    created maps detailing critical loads and levels of acidity within its boundaries. The maps were
 9    then used by CCE to create a Europe-wide map of critical loads (Hettelingh et al., 1991) that is
10    used in combination with air emissions and deposition data to guide negotiations between
11    nations and reduce the gap between critical loads and deposition (Skeffington, 1999).  The first
12    international agreement on pollution control based on critical loads was the second Sulfur
13    Protocol, which was established in Oslo (United Nations Economic Commission for Europe
14    (UNECE), 1994) within CLRTAP.
15          Since 1991, CCE has issued biennial technical status reports on critical loads and critical
16    thresholds of acidification, eutrophication, sulfur, nitrogen, and nitrogen oxide (Coordination
17    Center for Effects (CCE), 2005).  Progress on data and methodologies is reviewed annually in
18    CCE Mapping workshops. Recent CCE reports focus on scientific and technical support for the
19    revision of protocols as well as time horizons for recovery from ecosystem damage.
20          Many of the signatory governments to CLRTAP have adopted the critical load concept for
21    determining national emission control polices. Canada has also committed to a critical load
22    approach for controlling acid deposition.  In 1998, federal, provincial, and territorial Energy and
23    Environment Ministers signed The Canada-wide Acid Rain Strategy for Post-2000.  According
24    to Environment Canada, the primary long-term goal of the Strategy is to achieve critical  loads
25    (or the threshold level) for acid deposition across Canada (Environment Canada, 2003).
26          The Ministry of Environment in the Netherlands took the initiative to develop analogous
27    methods for the calculation of critical loads for heavy metals, methods that would be valid in the
28    context of CLPTRP (De Vries et al., 2004). Beginning in the mid-1990s, these methods were
29    developed through a series of manuals, international workshops, and expert meetings (De Vries
30    et al., 2004). Participating nations completed a voluntary preliminary critical load mapping
31    exercise for Pb and cadmium in Europe in 2002 (Hettelingh et al., 2002).

      December 2005                            8-230        DRAFT-DO NOT QUOTE OR CITE

-------
 1          The EPA Pb AQCD, Volume II (U.S. Environmental Protection Agency, 1986a) largely
 2    predates the development of the concept of critical loads, and does not include this topic. The
 3    EPA 2004 Air Quality Criteria for Particulate Matter, Volume I (U.S. Environmental Protection
 4    Agency, 2004a) include a brief discussion of the key elements of the critical loads framework
 5    general to any air pollutant. To date, the critical loads framework has not been used for
 6    regulatory purposes in the United States for any chemical.
 7
 8    8.3.2    Application  of Critical Loads to Terrestrial and Aquatic Ecosystems
 9          A combinatorial application of critical limit and critical load allows one to assess current
10    risk while simultaneously estimating future risk from exposure to a chemical (De Vries et al.,
11    2004).  Figure 8-3.1 shows that four combinations of critical load and limit exceedance or
12    non-exceedance are possible for a given ecosystem (Figure 1 of De Vries et al. [2004]).
13    For example, if a current risk is indicated by an exceedance of the critical limit for Pb due to
14    historical Pb deposition, but current inputs of Pb to the ecosystem are below the critical load
15    (lower left corner), the critical load model predicts that Pb concentrations will fall below the
16    critical limit at some point in the future if Pb deposition is maintained at the present level.
17    If current soil concentrations are below the critical limit (upper right corner), inputs greater than
18    the critical load will not result in exceedance of the critical limit for some period of time, but
19    continued exceedance of a critical load will eventually lead to an exceedance of the critical limit.
20          The time until a critical limit is exceeded (critical time) can also be predicted using the
21    critical load model (Paces, 1998).  This requires knowledge of current concentrations, the critical
22    load, and predicted deposition rates. Critical times may be useful for setting priorities between
23    ecosystems with critical load exceedances or between different chemicals.
24
25    8.3.3    Calculation  of Critical Loads
26          This section summarizes the various methods used to calculate critical loads (De Vries
27    et al., 2001, 2002, 2004; Groenenberg et al., 2002), with an emphasis on the most recent
28    material.
29
      December 2005                           8-231       DRAFT-DO NOT QUOTE OR CITE

-------
                      No critical load
                        exceedance
                       Critical load
                       exceedance
                                   No critical limit exceedance
                                    No damage at present or foreseen:

                                     Concentration
                                     critical
                                    present
                   •CL
                    PL2
                   • SL
                    PL1
                                                      Time
                                   •^ Keep the Present Load
                                   (more stringent than Critical
                                   Load)
   Ftttarf dtitttiigc foresee tt:

Concentration
                                     critical
                                    present
                                                          PL4
                                                          PU
                                                         • CL

                                                      Time
                                   ^ Consider Cridcal Load
                                   (emissions must decrease, even if
                                   concentrations in the ecosystem
                                   are allowed to increase further at
                                   critical load)
                            Critical limit excecdance
                          Present damage but recovery in progress:

                             Concentrator!
                                                                 present

                                                                 critical
                 •CL
                 . PU
                 -TL
                 • PL1
                                  TT         Time

                          •^ Keep the Present Load
                          (more stringent than Critical
                          Load)
                          or
                          •^ Consider Target Load to reach
                          the critical limit in a defined time
                          period (more stringent than
                          Critical Load)
                                                                Present damage, no recovery foreseen:
                                                                   Concentration
                             present
                              critical
                                                                                        PL4
                   PL3
                  -CL

                                               Time

                          •^ Consider Critical Load
                          (decrease of concentrations in the
                          ecosystem down to critical limit
                          in the long term)
                          or
                          •^ Consider Target Load to reach
                          die critical limit in a defined time
                          period (more stringent than
                          Critical Load)
                      CL - Critical load; PL - present load (2 cases); SL - Stand-still load; TL - Target load; TT - Target time
     Figure 8-3.1.    The predicted development of metal concentrations in ecosystems for four
                       cases of exceedance or non-exceedance of critical limits and of critical loads
                       of heavy metals, respectively.

     Source: Taken from DeVries et al. (2004).
1    8.3.3.1   Critical Limits
2           To determine the critical limit, effects-based criteria for the major ecological endpoints
3    should be developed for the ecosystem of concern.  Criteria may be developed for any receptor
4    that is exposed to the chemical of concern deposited in the ecosystem.  In terrestrial ecosystems,
     December 2005
                 8-232
DRAFT-DO NOT QUOTE OR CITE

-------
 1    possible ecological endpoints include effects from direct contact of invertebrates or plants with
 2    soil and ingestion of plants by herbivores. Effects-based criteria for use in defining the critical
 3    limit should be derived from ecotoxicological data appropriate to the most sensitive endpoint
 4    (De Vries et al., 2004). Regardless of the selected endpoint, the critical limit should be defined
 5    as a concentration in the medium that receives the depositional load, typically soil in terrestrial
 6    ecosystems and surface water in aquatic ecosystems. To derive these values, uptake and/or food
 7    chain modeling may be necessary.
 8          Many critical load calculations rely on ecological effects criteria developed by
 9    government agencies in individual countries (Paces, 1998; De Vries et al.,  1998; Van Den Hout
10    et al., 1999; Skjelkvale et al., 2001). Criteria for Pb vary widely and can be the largest source of
11    uncertainty in a critical load calculation (Van Den Hout et al., 1999).  One reason for the wide
12    range in estimates of effects criteria is that Pb speciation is often not taken into account. This
13    can result in variation in estimates of concentration for total Pb that is associated with adverse
14    effects, since the fraction of Pb available to cause a toxic effect depends on chemical factors such
15    as the pH or organic matter content (Lofts et al., 2004). To develop effects-based criteria that are
16    applicable to media with a pH or organic matter content different from the test medium, it is
17    more appropriate to develop criteria based on the free concentration of Pb rather than the total
18    concentration of Pb.
19
20    8.3.3.2  Models
21          Critical loads for heavy metals are typically calculated using a steady state model that
22    ignores internal metal cycling and keeps the calculations as simple as possible (De Vries et al.,
23    2004).  The critical load is equal to the atmospheric  input flux, which equals the sum of the
24    output fluxes from the system minus the other input fluxes (e.g., weathering) when the
25    concentration of Pb is at the critical limit. The input flux of heavy metals via weathering is
26    sometimes neglected, because quantitative estimates are highly uncertain, and weathering  is
27    generally thought to be a relatively minor process (De Vries et al., 2004).
28          More complex methods may be used to calculate critical loads.  For example, dynamic
29    models can be used to model the  change of concentrations in soil or water over time (Paces,
30    1998).  These models are most valuable when the time  to steady state is very long compared to
31    the time of interest. Using these models, the critical load is the deposition rate that leads to
      December 2005                            8-233       DRAFT-DO NOT QUOTE OR CITE

-------
 1    concentrations equal to the critical limit as the model approaches steady state. Fate and transport
 2    models that include internal cycling can also be used in place of simple mass balance models
 3    (Doyle et al., 2003) that may improve the accuracy of the models.
 4
 5    Terrestrial Model
 6          If internal cycling and weathering of Pb is neglected and atmospheric deposition is the
 7    only important source of Pb to the system, the critical load in a terrestrial ecosystem is equal to
 8    the sum of the most important fluxes out of the system, leaching, and uptake by harvested plants:
 9
10                                 CL(Pb) = Pbu + Pble(cnt)                                  (8-5)
11
12    where:
13           CL(Pb)   =  critical load of Pb (mass per area-year)
14           Pbu      =  metal net uptake in harvestable parts of plants at the critical limit
15                       (mass per area-year)
16           Pbie(crit)   =  leaching flux of Pb (dissolved and paniculate) from the soil layer at
17                       the critical limit (mass per area-year)
18
19          When applying a mass balance model, it is important to define the boundaries of the
20    compartment such that all significant fluxes in and out of the compartment can be accounted for.
21          Uptake of Pb by harvested vegetation may be an important flux out of agricultural soil or
22    forested soil that is actively logged.  In ecosystems that are not harvested, the steady state model
23    assumes that uptake by plants is balanced by deposition of Pb from decaying vegetation.
24          The flux out of the system due to uptake in harvested plants is calculated as follows:
25
26                                 Pbu = fPb,U!z * Yha * [Pb]ha                                (8-6)
27
28    where:
29           fpb,u,z  =   fraction of net Pb uptake from soil within the considered layer
30                     (dimensionless)
31           Yha   =   annual yield of harvestable biomass (mass per area-year)
32           [Pb]ha  =   metal concentration of harvestable parts of plants (Pb per unit mass)
33
34          The net fraction of metal uptake from soil within the considered layer corrects for Pb
35    measured in harvested vegetation that is taken up via direct deposition onto the plant or from soil
36    outside of the considered soil layer.

      December 2005                            8-234        DRAFT-DO NOT QUOTE OR CITE

-------
 1          The yield of harvestable biomass should only include the parts of plants that are removed
 2    from the system. Tree leaves, stalks remaining after harvest of agricultural land, and other parts
 3    that remain in the considered terrestrial ecosystem should not be included in the yield.
 4          De Vries et al. (2004) recommends that data for metal content in harvestable biomass
 5    should be taken from unpolluted areas. This leads to more conservative critical loads than using
 6    the metal content at the critical load. If the selected endpoint for the critical limit is related to the
 7    concentration in harvested plants rather than a concentration in soil, that critical concentration
 8    should be used in place of actual metal content in harvestable biomass.
 9          The critical leaching flux from the topsoil can be calculated as follows:
10
11                                 McKcrit) = Qle * [Pb]tot,Sdw(crit)                             (8-7)
12
13    where:
14          Qie           =  flux of drainage water leaching from the considered soil layer
15                            (volume/year)
16          [Pb]tot,sdw(crit)  =  critical total concentration of Pb in soil drainage water
17                            (mass per volume)
18
19          The total concentration of Pb in soil drainage water is the sum of all species of dissolved
20    and particulate Pb that leach out of the system in drainage water. De Vries et al. (2004) suggests
21    that Pb that is sorbed to suspended parti culate matter should be neglected so that total Pb is equal
22    to dissolved Pb, as concentrations of suspended solids are difficult to estimate.  Dissolved Pb
23    may exist as free ions, organic complexes, or inorganic complexes.
24          The drainage water flux leaching from the topsoil can be calculated as follows:
25
26                                 Qle = P-E1-Es-fEt;Z*Et                              (8-8)
27
28    where:
29          P     =    Precipitation (volume per area-time)
30          E;    =    Interception evaporation (volume per area-time)
31          Es   =    Soil evaporation within the topsoil (volume per area-time)
32          fEt,z  =    Plant transpiration (volume per area-time)
33          Et    =    Fraction of water uptake within the topsoil by roots (unitless)
34
      December 2005                            8-23 5       DRAFT-DO NOT QUOTE OR CITE

-------
 1          De Vries et al. (2004) recommends default values for some of these parameters and
 2    provides an alternative calculation method for sites with detailed hydrologic data as part of the
 3    guidance document.
 4
 5    Aquatic Model
 6          If internal cycling and weathering of Pb is neglected and atmospheric deposition is the
 7    only important source of Pb to the system, the critical load in an aquatic ecosystem is equal to
 8    the sum of the most important fluxes out of the system, uptake by harvested plants in the
 9    catchment, sedimentation, and lateral outflow from the catchment:
10
11                          CL(Pb) = Pbu + Pbsed(crit) * A! / Ac + Pbioc,crit                     (8-9)
12
13    where:
14           Pbu      =   removal of Pb by harvesting of vegetation in the catchment
15                        (mass per area-time)
16           PbSed(crit)  =   removal of Pb by sedimentation at the critical load
17                        (mass per area-time)
18           Pbioc,crit  =   lateral Pb outflow from the catchment at the critical load
19                        (mass per area-time)
20           AI       =   lake area
21           Ac       =   catchment area
22
23          It is important to carefully define the boundaries of the aquatic  system, so that all inflows
24    and outflows may be fully accounted for. Current guidance recommends including the entire
25    watershed within the system, rather than confining the system to a single lake or stream (De
26    Vries et al., 2004). In stream water, removal  of Pb due to sedimentation does not need to be
27    considered, simplifying the equation to the following:
28
29                               CL(Pb) = Pbu  + Pbioc,cnt                                (8-10)
30
31          De Vries et al. (2004) recommends that critical  loads should be calculated for stream
32    waters only, due to a high level  of uncertainty in the rate of removal via sedimentation or other
33    removal mechanisms within a lake. Critical loads for streams are protective of nearby lakes,
34    because the critical loads calculated using this methodology will be lower for streams than for
35    lakes.
      December 2005                           8-236       DRAFT-DO NOT QUOTE OR CITE

-------
 1          Calculation of removal of Pb by harvesting of vegetation in the catchment is similar to
 2    that in terrestrial ecosystems, with fPb,u equal to 1, since the entire catchment is now included.
 3          The critical lateral Pb outflow from the catchment is the product of the lateral outflow flux
 4    of water and the total concentration of Pb in the outflow water at the critical limit. The outflow
 5    flux of water is calculated from the outflow divided by the catchment area.
 6
 7    8.3.4   Critical Loads in Terrestrial Ecosystems
 8          Critical loads of Pb have been calculated using  simple mass balance, dynamic, and
 9    probabilistic models for forested and agricultural land  in Europe and Canada in a handful of
10    preliminary studies.  The methods and model assumptions used to calculate critical loads vary
11    widely between these studies and little attempt has been made to validate the models that were
12    used, so it is not known how much various simplifying assumptions affect the results.
13          Paces (1998) used data from a small agricultural catchment in the Czech Republic that is
14    typical of agricultural land in that country to calculate  critical loads for Pb and other heavy
15    metals. The critical loads were calculated using a simple dynamic box model.  The fluxes into
16    the system included atmospheric deposition, agricultural inputs, and weathering of bedrock and
17    the fluxes out of the  system included biological uptake and runoff. The model assumed that
18    inputs of metals to the system are independent of their concentrations in soil but that outputs are
19    proportional to the concentration of biologically active metal.  The author defined biologically-
20    active metal as the concentration of metal in soil that can be extracted in a 2 M nitric acid
21    solution.  This method was used to set a Czech state norm designed to be protective for soil
22    systems that is used as the critical limit in this study. Using the model, Paces determined that the
23    critical limit was not presently exceeded, but that the critical load is exceeded.  However, the
24    critical time was almost 1,000 years. Therefore, the model predicts that Pb  will continue to
25    accumulate in Czech agricultural soil and will eventually pose a potential risk if current inputs
26    continue. The author identified the simplifying assumptions used to calculate fluxes out of the
27    system as the major source of uncertainty.
28          Van den Hout et al. (1999) calculated critical loads for Pb and other pollutants in the
29    organic and mineral  soil layers of forested ecosystems. Atmospheric deposition was assumed to
30    be the only inflow, and outflows from soil were assumed to occur due to biological uptake and
31    leaching. Net heavy metal uptake by the forest was set equal to the rate of water uptake by
      December 2005                           8-237       DRAFT-DO NOT QUOTE OR CITE

-------
 1    vegetation multiplied by the water concentration and a "preference factor" that indicates the
 2    preference of the vegetation for the metal relative to water. Water flux was estimated from
 3    precipitation, soil evaporation, and transpiration data. An equilibrium speciation model that
 4    takes inorganic and organic ligands into account was used to estimate dissolved concentrations
 5    of Pb in leachate. Results were strongly dependant on the critical limits that were chosen.  Using
 6    the most stringent levels, critical loads were exceeded over much of Europe. The time to steady
 7    state was estimated to be hundreds of years. Speciation of Pb was identified as an important
 8    source of uncertainty.
 9          Reinds et al. (2002) used the guidance prepared by De Vries et al. (2002) to calculate
10    critical loads in the mineral topsoil of forested and agricultural ecosystems across 80,000 areas of
11    the European continent. The median critical load for Pb in Europe was 25 g ha'1 year"1 using
12    this methodology. The drainage water flux leaching from the topsoil was the dominant term in
13    the model, so critical loads followed the spatial pattern of net runoff (excess precipitation)  across
14    Europe.
15          Probst et al. (2003) calculated critical loads for Pb for forested sites in France.
16    Weathering rates were determined using a model for representative French soil samples. The
17    biomass uptake of Pb was derived using National Forestry Inventory data for the average annual
18    biomass growth and data for the Pb content in biomass. An uptake factor scaled down to the
19    considered depth was applied. Leaching of Pb was calculated using runoff data and dissolved Pb
20    concentrations in soil solution.  Critical loads at the French site varied over a wide range (4.9 to
21    133 g ha- year l). Critical  loads were controlled mainly by net runoff.  Weathering rates were
22    small compared to leaching and biomass uptake rates.
23          Doyle et al. (2003) used a probabilistic assessment to calculate critical loads in terrestrial
24    and aquatic (see following  section) ecosystems on the Canadian Shield.  The terrestrial model
25    used an analytical solution to the convection/dispersion equation. The model only considered
26    soluble metal in the flux to soil and assumed that the insoluble fraction was not available.  Metals
27    were assumed to  be sorbed onto immobile soil solids according to an equilibrium  distribution
28    (Kd) relationship. The input parameters were selected to represent boreal forest and Canadian
29    Shield conditions. Best estimate inputs were used for deterministic evaluation and distributions
30    of values were used in a probabilistic assessment. The model inputs included net  water flux,
31    effective water velocity, moisture content of soil, pH, dispersion coefficient, and Kd. The  25th

      December 2005                           8-23 8        DRAFT-DO NOT  QUOTE OR CITE

-------
 1    percentile critical loads (47 mg/m3 per year for Pb) were compared to current deposition rates to
 2    evaluate risk.
 3          In spite of the variation in methods and model assumptions used to calculate critical loads
 4    for Pb in the studies discussed above, some general conclusions may be drawn.  The critical limit
 5    is the most important value for determining the value of the critical load. Wide variations in
 6    available effects levels, makes this parameter one of the most important sources of uncertainty
 7    when calculating critical loads in terrestrial ecosystems. Spatial variations in critical loads for Pb
 8    are largely controlled by net runoff.  Weathering and uptake by harvestable vegetation were less
 9    important.  The time to reach  steady state is several hundred years in the two studies that used
10    dynamic models to determine critical loads.
11
12    8.3.5   Critical Loads in Aquatic Ecosystems
13          Doyle et al. (2003) modeled critical loads in surface water bodies assuming complete
14    mixing with dilution water entering from the terrestrial catchment area.  Loss of metal was also
15    assumed to occur though downstream flushing and burial in sediment.  Transfer of metal to
16    sediment was modeled as a first-order process dependant on the dissolved concentration and pH.
17    The inputs to the model included the following:  water body area, terrestrial catchment area,
18    water body depth,  sediment accumulation rate, thickness of biologically active sediment, net
19    precipitation, and water pH. The fist-order rate constant for transfer to sediment were correlated
20    to pH. The model reaches steady state within a few years. Transfer of Pb from the terrestrial
21    catchment to the water body was neglected, because the time to steady state could be on the
22    order of 10,000 years if the model included this source of Pb.  However, the authors cited a
23    separate calculation that indicated that neglect of transfer of Pb from the catchment may lead to a
24    5-fold underestimation of Pb concentrations in the surface water.
25          These results indicate that Pb run-off from soil is more important than direct atmospheric
26    deposition to the surface water bodies considered in this study. Due to the long times required to
27    achieve steady state, the critical load methodology may not be appropriate for Pb in aquatic
28    systems.
29
      December 2005                           8-239       DRAFT-DO NOT QUOTE OR CITE

-------
 1    8.3.6   Limitations and Uncertainties
 2          The largest sources of uncertainty identified in studies of critical loads for Pb include the
 3    following:
 4           •   Steady-state assumption
 5           •   Derivation of the critical limit
 6           •   Lead speciation
 7           •   Soil runoff as an input to aquatic ecosystems
 8
 9          The critical load is calculated for steady state conditions, but the time for Pb to reach
10    steady state concentrations can be as long as several centuries. Thus, dynamic models are often
11    used to predict Pb concentrations over shorter time frames. Dynamic modeling requires
12    additional knowledge about current concentrations in the considered ecosystem. For regulatory
13    purposes, use of dynamic modeling requires that a target time be set in order to calculate a
14    critical load.
15          Criteria for the protection of soil and for the protection of aquatic organisms vary over a
16    wide range from country to country. Use of the critical loads method for international
17    negotiations will require implementation of a consistent calculation methodology that takes into
18    account the effect of Pb speciation on toxicity over a range of soil types and chemical conditions.
19          Speciation strongly influences the toxicity of Pb in soil and water and partitioning
20    between dissolved and solid phases determines the concentration of Pb  in soil drainage water,
21    but it has not been taken into account in most of the critical load calculations for Pb performed to
22    date. Recent guidance for heavy metals has begun to emphasize the importance of speciation to
23    critical load calculations and suggest methods to calculate speciation (De Vries et al.,  2004). To
24    this end, Lofts et al., (2004) developed critical limit functions for several metals, including Pb,
25    that take into account the effects of pH, organic matter, and the protective effects of cations on
26    speciation.
27          Runoff of Pb from soil may be the major source of Pb into aquatic systems.  However,
28    little attempt has been made to include this source into critical load calculations  for aquatic
29    systems due to the complexity of including this source in the critical load models.
30
      December 2005                           8-240        DRAFT-DO NOT QUOTE OR CITE

-------
 1   8.3.7   Conclusions
 2          Preliminary efforts to calculate critical loads for Pb in terrestrial and aquatic ecosystems
 3   have so far relied on a variety of calculation methods and model assumptions. Efforts to refine
 4   and standardize methods for the calculation of critical loads for heavy metals which are valid in
 5   the context of CLPTRP are ongoing. At this time, the methods and models commonly used for
 6   the calculation of critical loads have not been validated for Pb.  Many of the methods neglect the
 7   speciation of Pb when estimating critical limits, the uptake of Pb into plants, and the outflux of
 8   Pb in drainage water, limiting the utility of current models.
 9          Future efforts should focus on fully incorporating the role of Pb speciation into critical
10   load models, and validating the assumptions used by the models.
11
     December 2005                            8-241        DRAFT-DO NOT QUOTE OR CITE

-------
  1     8.4    REFERENCES

  2     Abd-Elfattah, A.; Wada, K. (1981) Adsorption of lead, copper, zinc, cobalt, and cadmium by soils that differ in
  3            cation-exchange material. J. Soil Sci. 32: 271-283.
  4     AbdAllah, A. T.; Moustafa, M. A. (2002) Accumulation of lead and cadmium in the marine prosobranch
  5            Nerita saxtilis, chemical analysis, light and electron microscopy. Environ. Pollut. 116: 185-191.
  6     Acosta-Martinez, V.; Tabatabai, M. A. (2000) Arylamidase activity of soils: effect of trace  elements and
  7            relationships to soil properties and activities of amidohydrolases. Soil Biol. Biochem. 33: 17-23.
  8     Adam, M. S.; Abdel-Basset, R. (1990) Effect of lead nitrate and lead acetate on the growth and some metabolic
  9            processes of Scenedesmus obliquus. Acta Hydrobiol. 32: 93-99.
10     Adgate, J. L.; Willis, R. D.; Buckley, T. J.; Chow, J. C; Watson, J. G.; Rhoads, G. G.; Lioy, P. J. (1998) Chemical
11            mass balance source apportionment of lead in house dust. Environ. Sci. Technol. 32: 108-114.
12     Agency for Toxic Substances and Disease Registry.  (1988) The nature and extent of lead poisoning in children in
13            the United States: a report to Congress. Atlanta, GA: U.S. Department of Health and Human Services,
14            Public Health Service. Available from: NTIS, Springfield, VA; PB89-100184.
15     Ahern, M. D.; Morris, S. (1998) Accumulation of lead and its effects on Na balance in the freshwater crayfish
16            Cher ax destructor. J. Exp.  Zool. 281: 270-279.
17     Aka, H.; Darici, C. (2004) Carbon and nitrogen mineralization of lead treated soils in the eastern Mediterranean
18            region, Turkey. Soil Sediment Contam. 13: 255-265.
19     Al-Wabel, M. A.; Heil, D. M.; Westfall, D. G.; Barbarick, K. A. (2002) Solution chemistry influence on metal
20            mobility in biosolids-amended soils. J. Environ. Qual. 31:1157-1165.
21     Alam, M.  K.; Maughan, O. E. (1995) Acute  toxicity of heavy metals to common carp (Cyprinus carpio). J. Environ.
22            Sci. Health A 30:  1807-1816.
23     Albers, P. H.; Camardese, M. B.  (1993a) Effects of acidification on metal accumulation by  aquatic plants and
24            invertebrates. 1. Constructed wetlands. Environ. Toxicol. Chem. 12: 959-967.
25     Albers, P. H.; Camardese, M. B.  (1993b) Effects of acidification on metal accumulation by aquatic plants and
26            invertebrates. 2. Wetlands, ponds and small lakes. Environ. Toxicol. Chem. 12: 969-976.
27     Allen, P. (1993) Effects of acute exposure to cadmium (II) chloride and lead (II) chloride on the haematological
28            profile of Oreochromis aureus (Steindachner). Comp. Biochem. Physiol. C: Pharmacol. Toxicol.
29            Endocrinol. 105C: 213-217.
30     Allen, P. (1994) Accumulation profiles of lead and the influence of cadmium and mercury in Oreochromis aureus
31            (Steindachner) during chronic exposure. Toxicol. Environ. Chem. 44: 101-112.
32     Allinson, D. W.; Dzialo, C. (1981)  The influence of lead, cadmium and nickel on the growth of ryegrass and oats.
33            Plant Soil 62: 81-89.
34     American Public Health Association. (1995) Standard methods for the examination of water and wastewater.
3 5            Method 625 IB. Disinfection by-products: haloacetic acids and trichlorophenol. 19th ed. Washington, DC:
36            American Public Health Association, pp. 6-67-6-76.
37     Amiard, J.-C.; Metayer, C.; Baud, J.-P.; Ribeyre, F. (1994) Influence de facteurs ecologiques et biologiques  sur la
3 8            bioaccumulation d'elements metalliques chez de jeunes huitres (Crassostrea gigas thunberg) au cours du
3 9            pregrossissement en nourricerie (Influence of some ecological and biological factors on metal
40            bioaccumulation in young oysters (Crassostrea gigas Thunberg) during their spat rearing]. Water Res.
41            28:219-231.
42     An, Y.-J.; Kim, Y.-M.; Kwon, T.-M.; Jeong, S.-W. (2004) Combined effect of copper, cadmium, and lead upon
43            Cucumis sativus growth and bioaccumulation. Sci. Total Environ. 326: 85-93.
44     Andersen, M. K.; Raulund-Rasmussen, K.; Hansen, H. C. B.; Strobel, B. W. (2002) Distribution and fractionation of
45            heavy metals in pairs of arable and afforested soils in Denmark. Eur. J. Soil Sci. 53: 491-502.
46     Anderson, M. B.; Preslan, J. E.; Jolibois, L.; Bellinger, J. E.; George, W. J. (1997) Bioaccumulation of lead nitrate
47            in red swamp crayfish (Procambrus clarkii). J. Hazard. Mat. 54: 15-29.
48     Angelova, V.; Ivanov, K.; Ivanova, R. (2004) Effect of chemical forms of lead, cadmium and zinc in polluted soils
49            on their uptake by tobacco. J. Plant Nutr. 27: 757-773.
50     Angle, C.  R.; Mclntire, M. S.; Colucci, A. V. (1974) Lead in air, dustfall, soil, housedust, milk and water:
51            correlation with blood lead of urban and suburban school children. In: Hemphill, D. D., ed. Trace substances
52            in environmental health - VIII: [proceedings of University of Missouri's 8th annual conference on trace
53            substances in environmental  health]; June; Columbia, MO. Columbia, MO: University of Missouri;
54            pp. 23-29.
      December 2005                                8-242        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Angle, C. L.; Marcus, A.; Cheng, I.-H.; Mclntire, M. S. (1984) Omaha childhood blood lead and environmental
  2            lead: a linear total exposure model. Environ. Res. 35: 160-170.
  3     Antosiewicz, D. M (2005) Study of calcium-dependent lead-tolerance on plants differing in their level of
  4            Ca-deficiency tolerance. Environ. Pollut. 134: 23-34.
  5     Arai, T.; Maeda, M.; Yamakawa, H.; Kamatani, A.; Miyazaki, N. (2002) Growth effect on the uptake and
  6            elimination of trace metals in the abalones Haliotis. Fish. Sci. 68: 1094-1098.
  7     Archer, D.; Emerson, S.; Reimers, C. (1989) Dissolution of calcite in deep-sea sediments: pH and oxygen
  8            microelectrode results. Geochim. Cosmochim. Acta 53: 2831-2845.
  9     Aronson, A. L. (1971) Biologic effects of lead in fish. J. Wash. Acad. Sci. 61: 124-128.
10     Atchison, G. J.; Murphy, B. R.; Bishop, W. E.; Mclntosh, A. W.; Mayes, R. A. (1977) Trace metal contamination of
11            bluegill (Lepomis macrochirus) from two Indiana lakes. Trans. Am. Fish Soc. 106: 637-640.
12     Aualiitia, T. U.; Pickering, W. F. (1987) The specific sorption of trace amounts of Cu, Pb, and Cd by inorganic
13            particulates. Water Air Soil Pollut. 35: 171-185.
14     Austen, M. C.; McEvoy, A. J. (1997) The use of offshore meiobenthic communities in laboratory microcosm
15            experiments: response to heavy metal contamination. J. Exp. Mar. Biol. Ecol.  21: 247-261.
16     Baath, E. (1989) Effects of heavy metals in soil on microbial processes and populations (a review). Water Air Soil
17            Pollut. 47: 335-379.
18     Baath, E.; Arnebrant, K.; Nordgren, A. (1991) Microbial biomass and ATP in smelter-polluted forest humus. Bull.
19            Environ. Contam. Toxicol. 47: 278-282.
20     Baatrup, E. (1991) Structural and functional effects of heavy metals on the nervous system, including sense organs,
21            offish. Comp. Biochem. Physiol. C 100: 253-257.
22     Bacon, J. R.; Bain, D. C. (1995) Characterization of environmental water samples using strontium and lead stable-
23            isotope compositions. Environ. Geochem. Health 17: 39-49.
24     Bacon, J. R.; Hewitt, I. J. (2005) Heavy metals deposited from the atmosphere on upland Scottish soils: chemical
25            and lead isotope studies of the association of metals with soil components. Geochim. Cosmochim. Acta 69:
26            19-33.
27     Bacon, J. R.; Berrow, M. L.; Shand, C. A. (1995) The use of isotopic composition in field studies of lead in upland
28            Scottish soils (U.K.). Chem. Geol. 124: 125-134.
29     Bacon, J. R.; Jones, K. C.; McGrath, S. P.; Johnston, A. E. (1996) Isotopic character of lead deposited from the
30            atmosphere at a grassland site in the United Kingdom since 1860. Environ. Sci. Technol. 30: 2511-2518.
31     Badawy, S. H.; Helal, M. I. D.; Chaudri, A. M.; Lawlor, K.; McGrath, S. P. (2002) Soil solid-phase controls lead
32            activity in soil solution. J. Environ. Qual. 31: 162-167.
33     Baier, R. W.; Healy, M. L. (1977) Partitioning and transport of lead in Lake Washington. J. Environ. Qual.
34            6:291-296.
35     Bargar, J. R.; Brown, G. E.; Parks, G. A. (1997a) Surface complexation of Pb(II) at oxide-water interfaces. II. XAFS
3 6            and bond-valence determination of mononuclear and polynuclear Pb(II)  sorption products and surface
37            functional groups on iron oxides. Geochim. Cosmochim. Acta 61: 2639-2652.
38     Bargar, J. R.; Brown, G. E.; Parks, G. A. (1997b) Surface complexation of Pb(II) at  oxide-water interfaces. I. XAFS
3 9            and bond-valence determination of mononuclear and polynuclear Pb(II)  sorption products on aluminum
40            oxides. Geochim. Cosmochim. Acta 61:  2617-2637.
41     Bargar, J. R.; Brown, G. E.; Parks, G. A. (1998)  Surface complexation of Pb(II) at oxide-water interfaces: III. XAFS
42            determination of Pb(II) and Pb(II)-chloro adsorption complexes on goethite and alumina. Geochim.
43            Cosmochim. Acta 62: 193-207.
44     Bargar, J. R.; Persson, P.; Brown, G. E. (1999) Outer-sphere adsorption of Pb(II) EDTA on goethite. Geochim.
45            Cosmochim. Acta 63: 2957-2969.
46     Barltrop, D.; Meek, F. (1979) Effect of particle size on lead absorption from the gut. Arch. Environ. Health
47            34:280-285.
48     Beaty, B. J.; Black, W. C.; Carlson, J. O.; Clements, W. H. DuTeau, N.; Harrahy, E.; Nucklos, J.; Olson, K. E.;
49            Rayms-Keller, A. (1998) Molecular and  genetic ecotoxicologic approaches to aquatic environmental
50            bioreporting. Environ. Health Perspect. 106(86): 1395-1407.
51     Bechtel Jacobs Company (BJC). (1998) Empirical models for the uptake of inorganic chemicals from soil by plants.
52            Oak Ridge, TN: U.S. Department of Energy; BJC/OR-133.
53     Beckett, P. H. T. (1989) The use of extractants in studies on trace metals in soil, sewage sludges, and sludge-treated
54            soils. Adv. Soil Sci. 9: 143-176.
55     Beckett, P.; Davis, R. (1977) Upper critical levels of toxic elements in plants. New Phytol. 79: 95-106.
       December 2005                                8-243        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Begley, I. S.; Sharp, B. L. (1997) Characterization and correction of instrumental bias in inductively coupled plasma
  2            quadrapole mass spectrometry for accurate measurement of lead isotope ratios. J. Anal. Atomic Spectrom.
  3            12:395-402.
  4     Behra, R. (1993a) In vitro effects of cadmium, zinc and lead on Calmondulin-dependent actions in Oncorhynchus
  5            mykiss, Mytilus Sp., and Chlamydomonas reinhardtii. Arch. Environ. Contam. Toxicol. 24: 21-27.
  6     Behra, R. (1993b) Interaction of cadmium, lead and zinc with calmodulin from rainbow trout, sea mussels, and a
  7            green alga. Sci. Total Environ. (1 suppl.): 647-653.
  8     Beijer, K.; Jernelov, A. (1984) Microbial methylation of lead. In: Grandjean, P., ed. Biological effects of organolead
  9            compounds. Boca Raton, FL: CRC Press, Inc.; pp. 13-20.
10     Beiras, R.; Albentosa, M. (2003) Inhibition of embryo development of the commercial bivalves Ruditapes
11            decussatus and Mytilus galloprovincialis by trace metals; implications for the implementation of seawater
12            quality criteria. Aquaculture 230: 205-213.
13     Bengstsson, G.; Gunnarsson, T.; Rundgren, S. (1986) Effects of metal pollution on the earthworm Dendrobaena
14            mbida (Sav.) in acidified soils. Water Air Soil Pollut. 28: 361-383.
15     Benninger, L. K.; Lewis, D. M.; Turekian, K. K. (1975) The use of Pb-210 as a heavy metal tracer in the river-
16            estuarine system. In: Church, T.  M., ed. Marine chemistry in the coastal environment: a special symposium
17            sponsored by the Middle Atlantic Region at the 169th meeting of the American Chemical Society; April;
18            Philadelphia, PA; pp. 202-210. (ACS Symposium Series 18).
19     Berbel, F.; Diaz-Cruz, J. M.; Arino, C.; Esteban, M.; Mas, F.; Garces, J. L.; Puy, J. (2001) Voltammetric analysis  of
20            heterogeneity in metal ion binding by humics. Environ. Sci. Technol. 35: 1097-1102.
21     Bergkvist, B. (1986) Leaching of metals  from a spruce forest soil as influenced by experimental acidification.
22            Water Air Soil Pollut. 31: 901-916.
23     Berthelsen, B. O.; Steinnes, E. (1995) Accumulation patterns of heavy-metals in soil profiles as affected by forest
24            clear-cutting. Geoderma 66: 1-14.
25     Berthelsen, B. O.; Steinnes, E.; Solberg,  W.; Jingsen, L. (1995) Heavy metal concentrations in plants in relation to
26            atmospheric heavy metal deposition. J. Environ. Qual. 24: 1018-1026.
27     Besser, J. M.; Brumbaugh, W. G.; Branson, E. L.; Ingersoll, C. G. (2005) Acute and chronic toxicity  of lead in water
28            and diet to the amphipod Hyalella azteca. Environ. Toxicol. Chem. 24: 1807-1815.
29     Beyer, W. N.; Pattee, O. H.; Sileo, L.; Hoffman, D. J.; Mulhern, B. M. (1985) Metal contamination in wildlife living
30            near two zinc smelters. Environ. Pollut. Ser. A 38: 63-86.
31     Beyer, W. N.; Hensler, G.; Moore,  J. (1987) Relation of pR and other soil variables to concentrations of Pb, Cu, Zn,
32            Cd, and Se in earthworms.  Pedobiologia 30: 167-172.
33     Beyer, W. N.; Audet, D. J.; Heinz,  G. H.; Hoffman, D. J.; Day, D. (2000) Relation of waterfowl poisoning to
34            sediment lead concentrations in the Coeur d'Alene River basin. Ecotoxicology 9: 207-218.
35     Biggins, P. D. E.; Harrison, R. M. (1979) Atmospheric chemistry of automotive lead. Environ. Sci. Technol.
36            13:558-565.
37     Bilgrami, K. S.; Kumar, S. (1997) Effects of copper, lead and zinc on phytoplankton growth. Biol. Plant.
38            (Biologica Plantaram) 39: 315-317.
39     Bindler, R.; Brannvall, M.-L.; Renberg, I. (1999) Natural lead concentrations in pristine boreal forest soils and past
40            pollution trends: a reference for critical load models. Environ. Sci. Technol. 33:  3362-3367.
41     Black, M. C.; Ferrell, J. R.; Horning, R. C.; Martin, L. K., Jr. (1996) DNA strand breakage in freshwater mussels
42            (Anodonta grandis) exposed to lead in the laboratory and field. Environ. Toxicol. Chem. 15:  802-808.
43     Blais, J. M. (1996) Using isotopic tracers in lake sediments to assess atmospheric transport of lead in Eastern
44            Canada. Water Air Soil Pollut. 92:329-342.
45     Blake, L.; Goulding, K. W. T. (2002) Effects of atmospheric  deposition,  soil pH and acidification on heavy metal
46            contents in soils and vegetation of semi-natural ecosystems at Rothamsted Experimental Station, UK.  Plant
47            Soil 240: 235-251.
48     Blasco, J.; Puppo, J. (1999) Effect of heavy metals (Cu, Cd and Pb) on aspartate and alanine aminotransferase in
49           Ruditapesphilippinarum (Mollusca: Bivalvia). Comp. Biochem. Physiol. Part C: Pharmacol. Toxicol.
50            Endocrinol. 122C: 253-263.
51     Bloom, N. S.; Crecelius, E. A. (1987) Distribution of silver, mercury, lead, copper, and cadmium in central Puget
52            Sound sediments. Mar. Chem. 21: 377-390.
53     Bodar, C. W. M.; Zee, V. D.; Voogt, P. A.; Wynne, H.; Zandee, D. I. (1989) Toxicity of heavy metals to early life
54            stages ofDaphnia magna. Ecotoxicol. Environ. Saf.  17: 333-338.
55     Bodek, I.; Lyman, W. J.; Reehl, W. F.; Rosenblatt, D. H., eds. (1988) Environmental inorganic chemistry properties,
56            processes, and estimation methods. Pergamon Press,  pp.7.8.1-7.8-9.
       December 2005                                8-244        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Boisson, F.; Cotret, O.; Fowler, S. W. (2002) Transfer and distribution of lead in the asteroid Asterias rubens
  2            following ingestion of contaminated food: a radiotracer study. Mar. Pollut. Bull. 44: 1003-1009.
  3     Bongers, M. V.; Rusch, B.; Van Gestel, C. A. M. (2004) The effect of couterion and percolation on the toxicity of
  4            lead for the springtail Folsomia Candida in soil. Environ. Toxicol. Chem. 23: 195-200.
  5     Borgmann, U.; Kramar, O.; Loveridge, C. (1978) Rates of mortality, growth, and biomass production ofLymnaea
  6            palustris during chronic exposure to lead. J. Fish. Res. Board Can. 35:  1109-1115.
  7     Borgmann, U.; Norwood, W. P.; Clarke, C. (1993) Accumulation, regulaton and toxicity of copper, zinc, lead and
  8            mercury in Hyalella azteca. Hydrobiologia 259: 79-89.
  9     Borgmann, U.; Norwood, W. P.; Dixon, D. G. (2004) Re-evaluation of metal bioaccumulation and chronic toxicity
10            in Hyalella azteca using saturation curves and the biotic ligand model.  Environ. Pollut. 131: 469-484.
11     Borgmann, U.; Couillard, Y.; Doyle, P.; Dixon, D. G. (2005) Toxicity of sixty-three metals and metalloids to
12            Hyalella azteca at two levels of water hardness. Environ. Toxicol. Chem. 24: 641-652.
13     Bormann, F. H.; Likens, G. E. (1967) Nutrient cycling. Science (Washington, DC) 155: 424-429.
14     Bormann, F. H.; Likens, G. E.; Fisher, D. W.; Pierce, R. S. (1968) Nutrient loss accelerated by clear-cutting of a
15            forest ecosystem. Science (Washington, DC) 159: 882-884.
16     Bornschein, R. L.;  Succop, P. A.; Krafft, K. M.; Clark,  C. S.; Peace, B.; Hammond, P. B.  (1987) Exterior surface
17            dust lead, interior house dust lead and childhood lead exposure in an urban environment. In: Hemphill,
18            D. D.  ed. Trace substances in environmental health-XX, proceedings of the University of Missouri's 20th
19            annual Conference, pp. 322-332; June 1986; Columbia, MO.
20     Botelho, C. M. S.; Boaventura, R. A. R.; Goncalves, M. L. S.  S.; Sigg, L. (1994) Interactions of lead(II) with natural
21            river water. Part II. Paniculate matter. Sci. Total Environ. 151: 101-112.
22     Brannvall, M.-L.; Bindler, R.; Emteryd, O.; Renberg, I. (2001a) Vertical distribution of atmospheric pollution lead
23            in Swedish boreal forest soils. Water Air Soil Pollut. Focus 1: 357-370.
24     Brannvall, M.-L.; Kurkkio, H.; Bindler, R.; Emteryd, O.; Renberg, I. (200 Ib) The role of pollution versus natural
25            geological sources for lead enrichment in recent lake sediments and surface forest soils. Environ. Geol.
26            40: 1057-1065.
27     Brar, R. S.; Sandhu, H. S.; Grewal, G. S. (1997a) Biochemical alterations induced by repeated oral toxicity of lead
28            in domestic fowl. Indian Vet. J. 74: 380-383.
29     Brar, R. S.; Sandhu, H. S.; Randhawa, S. S.; Grewal, G. S. (1997b) Effect of repeated oral toxicity of lead on
30            activities of some plasma enzymes in domestic fowls. Indian J. Anim. Sci. 67: 878-879.
31     Brook, E. J.; Moore, J. N. (1988) Particle-size and chemical control of As, Cd,  Cu, Fe, Mn, Ni, Pb, and Zn in bed
32            sediment from the Clark Fork River, Montana (U.S.A.). Sci. Total Environ. 76: 247-266.
33     Brown, S. B.;  Evans, R. E.; Thompson, B. E.; Hara, T.  J. (1982) Chemoreception and aquatic  pollutants. In: Hara,
34            T. J., ed. Chemoreception in fishes. New York, NY: Elsevier Scientific Publishing Co.; pp. 363-393.
3 5            [Developments in aquaculture and fisheries science, v. 8].
36     Brown, S. L.;  Chaney, R.; Berti, B. (1999) Field test of amendments to reduce the in situ availability of soil lead. In:
37            Wenzel, W. W.; Adriano, D. C.; Doner, H. E.; Keller, C.; Lepp, N. W.; Mench, M. W.; Naidu, R.;
38            Pierzynski, G. M., eds. Abstracts of the 5th international conference on biogeochemistry of trace elements;
39            July; Vienna, Austria. Vienna, Austria: International Society for Trace  Element Research.
40     Brown, S. L.; Henry,  C. L.; Compton, H.; Chaney, R. L.; DeVolder, P. S. (2000) Using municipal biosolids in
41            combination  with other residuals to restore a vegetative cover on heavy metal mining tailings. In: Daniels,
42            W. L.; Richardson, S. G., eds. Proceedings of the national meeting of the American Society of Surface
43            Mining and Reclamation; June; Tampa, FL.
44     Brown, S. L.;  Chaney, R.  L.; Hallfrisch, J. G.; Xue, Q. (2003a) Effects of biosolids processing on the bioavailability
45            of lead in urban soils. J. Environ. Qual. 32: 100-108.
46     Brown, S. L.; Henry,  C. L.; Chaney, R.; Compton, H.; DeVolder, P. S. (2003b) Using municipal biosolids in
47            combination  with other residuals to restore metal-contaminated mining areas. Plant Soil 249: 203-215.
48     Brunekreef, B.; Noy,  D.; Biersteker, K.; Boleij, J. (1983) Blood lead levels in Dutch city children and their
49            relationship to lead in the environment. J. Air Pollut. Control Assoc. 33: 872-876.
50     Burger, J.; Carruth-Hinchey, C.; Ondroff, J.; McMahon, M.; Gibbons, J. W.; Gochfeld, M. (1998) Effects of lead on
51            behavior, growth, and survival of hatchling slider turtles. J. Toxicol. Environ. Health Part A. 55: 495-502.
52     Camp, Dresser, and McKee (COM). (1994) Metal speciation report. California Gulch CERCLA site, Leadville,
53            Colorado. Denver, CO: U.S. Environmental Protection Agency, Region VIII.
54     Canli, M.; Furness, R. W. (1993) Toxicity of heavy metals dissolved in sea water and influences of sex and size on
5 5            metal accumulation and tissue distribution in the Norway lobster Nephrops norvegicus. Mar. Environ. Res.
56            36:217-236.
       December 2005                                8-245         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Cannon, R. S.; Pierce, A. P. (1963) Lead isotope variation with growth zoning in a single crystal. Science
  2            (Washington, DC) 142: 574-576.
  3     Cao, X.; Ma, Q. Y.; Chen, M; Singh, S. P.; Harris, W. G. (2002) Impacts of phosphate amendments on lead
  4            biogeochemistry at a contaminated site. Environ. Sci. Technol. 36: 5296-5304.
  5     Capelo, S.; Vilhena, M. F.; Goncalves, M. L. S.S.; Sampayo, M. A. (1993) Effect of lead on the uptake of nutrients
  6            by unicellular algae. Water Res. 27: 1563-1568.
  7     Capodaglio, G.; Coale, K. H.; Coale, K. W. (1990) Lead speciation in surface waters of the eastern North Pacific.
  8            Mar. Chem. 29:  221-233.
  9     Carline, R. F.; Jobsis, G. J. (1993) Assessment of aquatic animal communities in the vicinity of the Palmerton,
10            Pennsylvania, zinc smelters. Environ. Toxicol. Chem. 12:  1661-1670.
11     Carson, A. R. ; Nelson H.; Hammermeister, D. (1986) Development and validation of site-specific water quality
12            criteria for copper. Environ. Toxicol. Chem. 5:997-1012.
13     Carter, L. F.; Porter, S. D. (1997) Trace-element accumulation by Hygrohypnum ochraceum in the upper
14            Rio Grande basin, Colorado and New Mexico, USA. Environ. Toxicol. Chem. 16: 2521-2528.
15     Casas, A. M.; Crecelius, E. A. (1994) Relationship between acid volatile sulfide and the toxicity of zinc, lead and
16            copper in marine sediments. Environ. Toxicol. Chem. 13: 529-536.
17     Case, J. M.; Reif, C. B.; Timko, A. (1989) Lead in the bottom sediments of Lake Nuangola and fourteen other
18            bodies of water in Luzerne County, Pennsylvania. J. Pennsylvania Acad. Sci. 63: 67-72.
19     Castro, L.; Carmo, C.; Peres, I.; Pihan, J. C. (1996) The clam, Ruditapes decussatus L., as a pollutionbioindicator:
20            zinc and lead accumulation and depuration. Ecologie (Brunoy) 27: 263-268.
21     Cestari, M. M.; Lemos, P. M. M.; Ribeiro, C.; Costa, J. R.; Pelletier, E.; Ferraro, M.; Mantovani, M. S.; Fenocchio,
22            A. S. (2004) Genetic damage induced by trophic doses of lead in the neotropical fish Hoplias malabaricus
23            (Characiformes, Erythrinidae) as revealed by the comet assay and chromosomal aberrations. Genet. Mol.
24            Biol. 27: 270-274.
25     Chadwick, J. W.; Canton, S. P.; Dent, R. L. (1986) Recovery of benthic invertebrate communities in Silver Bow
26            Creek, Montana, following improved metal mine wastewater treatment. Water Air Soil Pollut. 28: 427-438.
27     Chander, K.; Dyckmans, J.; Hoeper, H.; Joergensen, R. G.; Raubuch, M. (2001) Long-term effects on soil microbial
28            properties of heavy metals from industrial exhaust deposition. J. Plant Nutr. Sci. 164: 657-663.
29     Charlatchka, R.;  Cambier, P.; Bourgeois, S. (1997) Mobilization of trace metals in contaminated soils under
30            anaerobic conditions. In: Prost, R., ed. Contaminated soils, proceedings of the 3rd international conference
31            on the biogeochemistry of trace elements; May,  1995; Paris, France.
32     Chillrud, S. N.; Hemming,  S.; Shuster, E. L.; Simpson, H. J.; Bopp, R. F.; Ross, J. M.; Pederson, D. C.; Chaky,
33            D. A.; Tolley, L.-R.; Estabrooks, F. (2003) Stable lead isotopes, contaminant metals and radionuclides in
34            upper Hudson River sediment cores:  implications for improved time stratigraphy and transport processes.
35            Chem. Geol. 199: 53-70.
36     Chow, T. J.; Bruland, K. W.; Bertine, K.; Soutar, A.; Koide, M.; Goldberg, E. D. (1973) Lead pollution: records in
37            Southern California coastal sediments. Science (Washington, DC, U.S.) 181: 551-552.
3 8     Clements, W. H. (1994) Benthic invertebrate community responses to heavy metals in the Upper Arkansas River
39            Basin, Colorado. J. N. Am. Benthol.  Soc. 13: 30-44.
40     Clevenger, T. E.;  Saiwan, C.; Koirtyohann, S. R. (1991) Lead speciation of particles on air filters collected in the
41            vicinity of a lead smelter. Environ. Sci. Technol.  25: 1128-1133.
42     Coello, W. F.; Khan, M. A. Q. (1996) Protection against heavy metal toxicity by mucus and scales in fish. Arch.
43            Environ. Contam. Toxicol. 30: 319-326.
44     Coordination Center for Effects (CCE). (2005) Methods and models. Available: http://www.rivm.nl/cce/methmod/
45            [22 June, 2005].
46     Cotrufo, M. F.; De Santo, A. V.; Alfani, A.; Bartoli, G.; De Cristofaro, A. (1995) Effects of urban heavy metal
47            pollution on organic matter decomposition in Quercus ilex L. woods. Environ. Pollut. 89: 81-87.
48     Cotter-Howells, J.; Caporn, S. (1996) Remediation of contaminated land by formation of heavy metal phosphates.
49            Appl. Geochem. 11: 335-342.
50     Covington, W. W. (1981) Changes in forest floor organic matter and nutrient content following clear cutting in
51            northern hardwoods. Ecology 62: 41-48.
52     Dalenberg, J. W.; Van Driel, W. (1990) Contribution of atmospheric deposition to heavy-metal concentrations in
53            field crops. Neth. J. Agric. Sci. 38: 369-379.
54     Daughney, C. J.; Fein, J. B. (1998) The effect of ionic strength on the adsorption of H+, Cd2+, Pb2+, and Cu2+ by
55            bacillus subtilis and bacillus licheniformis: a surface complexation model. J. Colloid Interf. Sci. 198: 53-77.
       December 2005                                8-246         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Dave, G. (1992a) Sediment toxicity and heavy metals in eleven lime reference lakes of Sweden. Water Air Soil
  2            Pollut. 63: 187-200.
  3     Dave, G. (1992b) Sediment toxicity in lakes along the river Kolbacksan, central Sweden Hydrobiologia
  4            236:419-433.
  5     Davies, P. H.; Everhart, W. H. (1973) Effects of chemical variations in aquatic environments: volume 3 lead toxicity
  6            to rainbow trout and testing application factor concept. Washington, DC: U.S. Environmental Protection
  7            Agency; report no. EPA-R3-73-011C. Available from: NTIS, Springfield, VA; PB-221345.
  8     Davies, P. H.; Goetti, J. P.; Sinley, J.  R.; Smith, N. F. (1976) Acute and chronic toxicity of lead to rainbow trout
  9            (Salmo gairdneri) in hard and soft water.  Water Res. 10: 199-206.
10     Davies, N. A.; Hodson, M. E.; Black, S. (2002) Changes in toxicity and bioavailability of lead in contaminated soils
11            to the earthworm Eisenia Fetida (Savigny 1826) after bone meal amendments to the soil. Environ. Toxicol.
12            Chem. 21:2685-2691.
13     Davies, N. A.; Hodson, M. E.; Black, S. (2003) The influence  of time on lead toxicity and bioaccumulation
14            determined by the OECD earthworm toxicity test. Environ. Pollut. 121:55-61.
15     Davis, A.; Galloway J. N. (1993) Distribution of Pb between sediments and pore water in Woods Lake, Adirondack
16            State Park, New York, U.S.A. Appl. Geochem. 8:51-65.
17     Davis, A.; Drexler, J. W.; Ruby, M. V.; Nicholson, A. (1993) Micromineralogy of mine wastes in relation to lead
18            bioavailability, Butte, Montana. Environ.  Sci. Technol. 27: 1415-1425.
19     Davis, A.; Sellstone, C.; Clough, S.; Barrick, R.; Yare, B. (1996) Bioaccumulation of arsenic, chromium and lead in
20            fish: constraints imposed by sediment geochemistry. Appl. Geochem. 11: 409-423.
21     Davison, W.; Zhang, H.  (1994) In situ speciation measurements of trace components in natural waters using thin-
22            film gels. Nature 367: 546-548.
23     Davison, W.; Grime, G.  W.; Morgan, J. A. W.; Clarke, K. (1991) Distribution of dissolved iron in sediment pore
24            waters at submillimetre resolution. Nature (London) 352: 323-325.
25     Davison, W.; Zhang, H.; Grime, G. W. (1994) Performance characteristics of gel probes  used for measuring pore
26            waters. Environ. Sci. Technol. 28: 1623-1632.
27     De Jonghe, W. R. A.; Adams, F. C. (1986) Biogeochemical cycling of organic lead compounds. In: Nriagu, J.  O.;
28            Davidson, C. I., eds. Toxic metals in the atmosphere. New York, NY: John Wiley & Sons; pp. 561-594.
29            (Advances in environmental science and technology: v. 17).
30     De Vries, W.; Bakker, D. J.; Groenenberg, J. E.; Reinds, G. J.; Bril, J.; Van  Jaarsveld, J.  A. (1998) Calculation and
31            mapping of critical loads for heavy metals and persistent organic pollutants for Dutch forest soils. J. Hazard.
32            Mat. 61:99-106.
33     De Vries, W.; Schiitze, G.; Romkens, P.; Hettelingh, J.-P. (2001) Guidance for the calculation of critical loads for
34            cadmium and lead in terrestrial and aquatic  ecosystems. In: Hettelingh, J.-P.; Slootweg, J.; Posch, M.;
3 5            Dutchak, S.; Ilyin, I., eds. Preliminary modelling and mapping of critical loads for cadmium and lead in
36            Europe;  RIVM report no. 259101011. Bilthoven, The  Netherlands: National Institute of Public Health and
37            the Environment; pp. 17-36.
38     De Vries, W.; Schtitze, G.; Lots, S.; Meili, M.; Romkens, P.; Terytze, K.; Scholz, K.; Farret, R.; Jakubowski, M.
39            (2002) Critical limits for cadmium, lead and mercury related to ecotoxicological  effects on soil organisms,
40            aquatic organisms, plants,  animals and humans: background document for the expert meeting on critical
41            limits for heavy metals and methods for their application. In: Proceedings of the expert meeting on critical
42            limits for heavy metals and methods for their application; December; Berlin. Geneva, Switzerland: United
43            Nations Economic Commission for Europe  (UN-ECE) Convention  on long range transboundary air
44            pollution. Available: http://www.oekodata.com/pub/mapping/workshops/ws_berlin/proceedings.pdf [19
45            October, 2005].
46     De Vries, W.; Schtitze, G.; Lofts, S.; Tipping, E.;  Meili, M.; Romkens, P. F. A.  M.; Groenenberg, J. E. (2004)
47            Calculation of critical loads for cadmium, lead and mercury: background document to a mapping manual on
48            critical loads of cadmium,  lead and mercury. Wageningen, The Netherlands:  Alterra report no. 1104.
49            Available: http://www.oekodata.com/pub/mapping/manual/reportl 104.pdf [22 June, 2005].
50     De Voider, P. S.;  Brown, S. L.; Hesterberg, D.; Pandya, K. (2003) Metal bioavailability and speciation in a wetland
51            tailings repository amended with biosolids compost, wood ash, and  sulfate. J. Environ. Qual. 32: 851-864.
52     Deacon, J. R.; Spahr, N. E.; Mize,  S. V.; Boulger, R. W. (2001) Using water, bryophytes, and macroinvertebrates to
53            assess trace element concentrations in the Upper Colorado River basin. Hydrobiologia 455: 29-39.
54     Dearth, R. K.; Hiney, J. K.; Srivastava, V.; Les Dees, W.; Bratton, G. R. (2004) Low level lead (Pb) exposure  during
5 5            gestation and lactation: assessment of effects on pubertal development in Fisher  344 and Sprague-Dawley
56            female rats. Life Sci. 74: 1139-1148.
       December 2005                                8-247        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Deng, H.; Ye, Z. H.; Wong, M. H. (2004) Accumulation of lead, zinc, copper and cadmium by 12 wetland species
  2            thriving in metal-contaminated sites in China. Environ. Pollut. 132: 29-40.
  3     Di Toro, D. M.; Mahoney, J. D.; Hansen, D. I; Scott, K. I; Carlson, A. R. (1992) Acid volatile sulfide predicts the
  4            acute toxicity of cadmium and nickel in sediments. Environ. Sci. Technol. 26: 96-101.
  5     Di Toro, D. M.; Allen, H. E.; Bergman, H. L.; Meyer, J. S.; Paquin, P. R.; Santore, R. C. (2001) Biotic ligand model
  6            of the acute toxicity of metals. 1. Technical basis. Environ. Toxicol. Chem. 20: 2383-2396.
  7     Dixon, R. K. (1988) Response of ectomycorrhizal Quercus rubra to soil cadmium, nickel and lead. Soil Biol.
  8            Biochem. 20: 555-559.
  9     Dixon, F. M.; Freer, J. R.; Abdi, A. N. (1995) Metal levels in garden vegetables raised onbiosolids amended soil.
10            Compost Sci. Util. 3:55-63.
11     Doe, B.; Rohrbough, R. (1977) Lead isotope data bank: 3,458 samples and analyses cited. U.S. Geological Survey
12            open-file report 79-661.
13     Doe, B.; Stacey, J. (1974) Application of lead isotopes to question of ore genesis and ore project evaluation. Econ.
14            Geol. Bull. Soc. Econ. Geol. 69: 757-776.
15     Doe, B.; Tilling, R.; Hege, C.; Klepper, M. (1968) Lead and  strontium isotope studies of the Boulder Batholith.
16            Econ. Geol. Bull. Soc. Econ. Geol. 63: 884-906.
17     Doelman, P.; Haanstra, L. (1986) Short- and long-term effects of heavy metals on urease activity in soils. Biol.
18            Fertil. Soils 2: 213-218.
19     Dolske, D. A.; Sievering,H. (1979) Trace element loading of southern Lake Michigan by dry deposition of
20            atmospheric aerosol. Water Air Soil Pollut. 12: 485-502.
21     Dorr, H. (1995) Application of 210Pb in Soils.  J. Paleolimnol.  13: 157-168.
22     Dorr, H.; Mtinnich, K. O. (1989) Downward movement of soil organic-matter and its influence on trace-element
23            transport (210Pb, 137Cs) in the soil. Radiocarbon 31: 655-663.
24     Dorr, H.; Mtinnich, K. O. (1991) Lead and cesium transport in European forest soils. Water Air Soil Pollut.
25            57/58: 809-818.
26     Douben, P. E. T. (1989) Lead and cadmium in stone loach (Noemacheilus barbatulus L.) from three rivers in
27            Derbyshire. Ecotoxicol. Environ. Saf. 18: 35-58.
28     Douglas-Stroebel, E. K.; Brewer, G. L.; Hoffman, D. J. (2005) Effects of lead-contaminated sediment and nutrition
29            on mallard duckling behavior and growth. J. Toxicol. Environ. Health Part A 68: 113-128.
30     Doyle, P. J.; Gutzman, D. W.; Sheppard, M. I.; Sheppard, S.  C.; Bird, G. A.; Hrebenyk, D. (2003) An ecological risk
31            assessment of air emissions of trace metals from copper and zinc production facilities. Hum. Ecol. Risk
32            Assess. 9:  607-636.
33     Drava, G.; Capelli, R.; Minganti,  V.; De Pellegrini, R.; Relini, L.; Ivaldi, M. (2004) Trace elements in the muscle of
34            red shrimp Aristeus antennatus (Risso, 1816) (Crustacea, Decapoda) fromLigurian sea (NW
35            Mediterranean): variations related to the reproductive cycle. Sci. Total Environ. 321: 87-92.
3 6     Drexler, J. W. (1997) Validation of an in vitro method: a tandem approach to estimating the bioavailability of lead
3 7            and arsenic in humans, quantifying the real toxicity of common soil contaminants, IBC conference on
38            bioavailability; December; Scottsdale, AZ.
39     Drexler, J. W.; Mushak, P. (1995) Health risks from extractive industry  wastes: characterization of heavy metal
40            contaminants and quantification of their bioavailability and bioaccessability. Presented at: The third
41            international conference on the biogeochemistry of trace elements; May; Paris, France.
42     Driscoll, C. T.;  Fuller, R. D.; Simone, D. M. (1988) Longitudinal variations in trace metal concentrations in a
43            northern forested ecosystem. J. Environ. Qual. 17: 101-107.
44     Driscoll, C. T.;  Likens,  G. E.; Church, M. R. (1998) Recovery of surface waters in the northeastern U.S. from
45            decreases in atmospheric deposition of sulfur. Water Air Soil Pollut. 105: 319-329.
46     Duan, Y.; Guttman, S.;  Oris, J.; Bailer, J. (2000) Genotype and toxicity  relationships among Hyalella azteca: I.
47            acute exposure to metals  or low pH. Environ. Toxicol. Chem. 19: 1414-1421.
48     Dupont, L.; Guillon, E.; Bouanda, J.; Dumonceau, J.; Aplincourt, M. (2002) EXAFS and XANES studies of
49            retention of copper and lead by a lignocellulosic biomaterial. Environ. Sci. Technol. 36: 5062-5066.
50     Durand, P.; Neal, C.; Jeffery, H. A.; Ryland, G. P.; Neal, M.  (1994) Major, minor and trace-element budgets in the
51            Plynlimon afforested catchments (Wales): general trends, and effects of felling and climate variations. J.
52            Hydrol. 157: 139-156.
53     Egli, M.; Fitze, P.; Oswald, M. (1999) Changes in heavy metal contents in an acidic forest soil affected by depletion
54            of soil organic matter within the time span 1969-93.  Environ. Pollut. 105: 367-379.
55     Eisenreich, S. J.; Metzer, N. A.; Urban, N. R.; Robbins, J. A. (1986) Response of atmospheric lead to decreased use
56            of lead in gasoline. Environ.  Sci. Technol. 20:  171-174.
       December 2005                                8-248         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Eisler, R. (1988) Lead hazards to fish, wildlife, and invertebrates: a synoptic review. Washington, DC: U.S.
  2            Department of the Interior, Fish and Wildlife Service; biological report 85(1.14); contaminant hazard
  3            reviews report no. 14.
  4     Eisler, R. (2000) Handbook of chemical risk assessment: health hazards to humans, plants, and animals. Volume 1:
  5            metals. New York, NY: Lewis Publishers; pp. 201-311.
  6     Ekenler, M; Tabatabai, M. (2002) Effects of trace metals on p-glucosaminidase activity in soils. Soil Biol.
  7            Biochem. 34:  1829-1832.
  8     Emmanuel, S.; Erel, Y. (2002) Implications from concentrations and isotopic data for Pb partitioning processes in
  9            soils. Geochim. Cosmochim. Acta 66: 2517-2527.
10     Encinar, J. R.; Garcia Alonso, J. I.; Sanz-Medel, A.; Main, S.; Turner, P. J. (2001a) A comparison between
11            quadrupole, double  focusing and multicollector ICP-MS: Part I. Evaluation of total combined uncertainty
12            for lead isotope ratio measurements. J. Anal. At. Spectrom. 16: 315-321.
13     Encinar, J. R.; Garcia Alonso, J. I.; Sanz-Medel, A.; Main, S.; Turner, P. J. (200Ib) A comparison between
14            quadrupole, double  focusing and multicollector ICP-MS: Part II. Evaluation of total combined uncertainty
15            in the determination of lead in biological matrices by isotope dilution. J. Anal. At. Spectrom. 16: 322-326.
16     Enserink, E. L.; Maas-Diepeveen, J. L.; VanLeeuwen, C. J. (1991) Combined effects of metals; an ecotoxicological
17            evaluation. Water Res.  25: 679-687.
18     Environment Canada.  (2003) What is acid rain? Available: http://www.on.ec.gc.ca/wildlife/acidrain/arl-e.html
19            [19 October, 2005].
20     Erel, Y.; Patterson, C.  C. (1994) Leakage of industrial lead into the hydrocycle. Geochim. Cosmochim. Acta
21            58: 3289-3296.
22     Erel, Y.; Morgan, J. J.; Patterson, C. C. (1991) Natural levels of lead and cadmium in a remote mountain stream.
23            Geochim. Cosmochim. Acta 55: 707-719.
24     Erel, Y.; Veron, A.; Halicz,  L. (1997) Tracing the transport of anthropogenic lead in the atmosphere  and in soils
25            using isotopic ratios. Geochim. Cosmochim. Acta 61: 4495-4505.
26     Erel, Y.; Emmanuel, S.; Teutsch, N.; Halicz, L.; Veron, A. (2001) Partitioning of anthropogenic and natural lead in
27            soils. Abst. Papers Am. Chem. Soc. 221:  U467.
28     Erten-Unal, M.; Wixson, B. G.; Gale, N.; Pitt, J. L. (1998) Evaluation of toxicity, bioavailability and speciation of
29            lead, zinc, and cadmium in mine/mill wastewaters. Chem. Spec. Bioavail. 10: 37-46.
30     Evans, G. C.; Norton,  S. A.; Fernandez, I. J.; Kahl, J. S.; Hanson, D. (2005) Changes in concentrations of major
31            elements and trace metals in northeastern U.S.-Canadian sub-alpine forest floors. Water Air  Soil Pollut.
32            163:245-267
33     Farfel, M. R.; Orlova,  A. O.; Chaney, R. L.; Lees, P. S. J.; Rohde, C.; Ashley, P. J. (2005) Biosolids  compost
34            amendment for reducing soil lead hazards: a pilot study of Orgro® amendment and grass seeding in urban
35            yards. Sci. Total Environ. 340: 81-95.
36     Fargasova, A. (1993) Effect of five toxic metals on the alga Scenedesmus quadricauda. Biologia (Bratislava)
37            48:301-304.
38     Farkas, A.; Salanki, J.; Specziar, A. (2003) Age- and size-specific patterns of heavy metals in the organs of
39            freshwater fishAbramis brama L. populating a low-contaminated site.  Water Res. 37: 959-964.
40     Farmer, J. G.; Eades, L. J.; MacKenzie, A. B.; Kirika, A.; Bailey-Watts, T. E. (1996) Stable lead isotope record of
41            lead pollution in Loch Lomond sediments since 1630 A.D. Environ. Sci. Technol. 30: 3080-3083.
42     Farmer, J. G.; MacKenzie, A. B.; Sugden, C. L.; Edgar, P. J.; Eades, L. J. (1997) A comparison of the historical lead
43            pollution records  in peat and freshwater lake sediments from central Scotland. Water Air Soil Pollut.
44            100:253-270.
45     Faure, G. (1977) Principles of isotope geology. New York, NY: John Wiley &  Sons.
46     Federer, C. A. (1984) Organic matter and nitrogen content of the forest floor in even-aged northern hardwoods.
47            Can.  J. For. Res. 14: 763-767.
48     Fendorf, S. E.; Sparks, D. L.; Lamble, G. M.; Kelley, M. J. (1994) Applications of x-ray absorption fine structure
49            spectroscopy to soils. Soil Sci. Soc. Am.  J. 58: 1583-1595.
50     Fernandez-Leborans, G.; Novillo, A. (1992) Hazard evaluation of lead effects using marine protozoan communities.
51            Aquat.  Sci. 54: 128-140.
52     Fernandez-Leborans, G.; Novillo, A. (1994) Effects of periodic addition of lead on a marine protistan community.
53            Aquat.  Sci. 56: 191-205.
54     Fernandez-Leborans, G.; Antonio-Garcia, M. T. (1988) Effects of lead and cadmium in a community of protozoans.
55            ActaProtozool. 27: 141-159.
       December 2005                                8-249         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Fernando, Q. (1995) Metal speciation in environmental and biological systems. Environ. Health Perspect. Suppl.
  2            103(1): 13-16.
  3     Fitzhugh, R. D.; Driscoll, C. T.; Groffman, P. M; Tierney, G. L.; Fahey, T. J.; Hardy, J. P. (2001) Effects of soil
  4            freezing disturbance on soil solution nitrogen, phosphorus, and carbon chemistry in a northern hardwood
  5            ecosystem. Biogeochemistry 56: 215-238.
  6     Flegal, A. R; Rosman, K. J. R.; Stephenson, M. D. (1987) Isotope systematics of contaminant leads in Monterey
  7            Bay. Environ. Sci. Technol. 21: 1075-1079.
  8     Flegal, A. R.; Duda, T. F.; Niemeyer, S. (1989a) High gradients of lead isotopic composition in north-east Pacific
  9            upwelling filaments. Nature 339: 458-460.
10     Flegal, A. R.; Nriagu, J. O.; Niemeyer,  S.; Coale, K. H. (1989b) Isotopic tracers of lead contamination in the
11            Great Lakes. Nature (London) 339: 455-458.
12     Florence, T. M. (1977) Trace metal species in fresh waters.  Water Res. 25: 681-687.
13     Foerstner, U. (1987) Sediment-associated contaminants—an overview of scientific basis for developing remidial
14            options. Hydrobiologia 149: 221-246.
15     Francek, M. A. (1992) Soil lead levels in a small town environment: a case study from Mt. Pleasant, Michigan.
16            Environ. Pollut. 76: 251-257.
17     Francek, M. A. (1997) Soil lead levels in orchards and roadsides of Mission Peninsula, Michigan. Water Air Soil
18            Pollut. 94: 373-384.
19     Franson, J. C. (1996) Interpretation of tissue lead residues in birds other than waterfowl. In: Beyer, W. N.; Heinz,
20            G. H.; Redmon-Norwood, A. W., eds. Environmental contaminants in wildlife. Interpreting tissue
21            concentrations. Boca Raton, FL: CRC Press, pp. 265-279. [SETAC special publication series].
22     Friedland, A. J.; Johnson, A. H. (1985) Lead distribution and fluxes in a high-elevation forest in northern Vermont.
23            J. Environ. Qual.  14: 332-336.
24     Friedland, A. J.; Johnson, A. H.; Siccama, T. G. (1984) Trace metal content of the forest floor in the Green
25            mountains of Vermont: spatial  and temporal patterns. Water Air Soil Pollut. 21: 161-170.
26     Friedland, A. J.; Craig, B. W.; Miller, E. K.; Herrick, G. T.; Siccama, T. G.; Johnson, A. H. (1992) Decreasing lead
27            levels in the forest floor of the northeastern USA. Ambio 21: 400-403.
28     Fritze, H.; Niini, S.; Mikkola, K.; Makinen, A. (1989) Soil microbial effects of Cu-Ni smelter in southwestern
29            Finland. Biol. Pert. Soils  8: 87-94.
30     Fry, B.; Sherr, E. B. (1984) C13 measurements as indicators of carbon flow in marine and freshwater ecosystems.
31            Contrib. Mar. Sci. 27: 13-47.
32     Fu, M. H.;  Tabatabai, M. A. (1989) Nitrate reductase in soils: effects of trace elements. Soil Biol. Biochem.
33            21:943-946.
34     Fuller, R. D.; Simone, D. M.; Driscoll,  C. T. (1988) Forest clearcutting effects on trace metal concentrations: spatial
35            patterns in soil solutions and streams. Water Air Soil Pollut. 40: 185-195.
36     Galbraith, H.; LeJeune, K.; Lipton, J. (1995) Metal and arsenic impacts to soils,  vegetation communities and wildlife
37            habitat in southwest Montana uplands contaminated by smelter emissions: I. Field evaluation. Environ.
38            Toxicol. Chem. 14: 1895-1903.
39     Galloway,  J. N.; Thornton, J. D.; Norton, S. A.; Volchok, H. L.; McLean, R. A.  N. (1982) Trace metals in
40            atmospheric deposition: a review and assessment. Atmos. Environ.  16: 1677-1700.
41     Gamble, D. S.; Schnitzer, M.; Kerndorff, H.; Langford, C. H. (1983) Multiple metal ion exchange equilibria with
42            humic-acid. Geochim.  Cosmochim. Acta47: 1311-1323.
43     Gao, Y.; Kan, A. T.; Tomson, M. B. (2003) Critical evaluation of desorption phenomena of heavy metals from
44            natural sediments. Environ. Sci. Technol. 37: 5566-5573.
45     Garcia, T. A.; Corredor, L. (2004) Biochemical changes in the kidneys after perinatal intoxication with lead and/or
46            cadmium and their antagonistic effects when coadministered. Ecotoxicol. Environ. Saf. 57: 184-189.
47     Gaur, J. P.; Noraho, N.; Chauhan, Y. S. (1994) Relationship between heavy metal accumulation and toxicity in
48            Spirodelapolyrhiza (L.) Schleid. wAAzollapinnata R. Br. Aquat. Bot.  49: 183-192.
49     Gavrilenko, Y. Y.; Zolotukhina, Y. Y. (1989) Accumulation and interaction of copper, zinc, manganese, cadmium,
50            nickel and lead ions adsorbed by aquatic macrophytes. Hydrobiol. J. 25: 54-61.
51     Gerhardt, A.  (1994) Short term toxicity of iron (Fe) and lead (Pb) to the  mayfly Leptophlebia  marginata (L.)
52            (Insecta) in relation to freshwater acidification. Hydrobiologia 284: 157-168.
53     Getz, L. L.; Haney, A. W.; Larimore, R. W.; McNurney, J. W.; Lelend, H. V; Price, P. W.; Rolfe, G. L.; Wortman,
54            R.  L.; Hudson, J. L.; Solomon, R. L.; Reinbold, K. A. (1977) Transport  and distribution in a watershed
55            ecosystem. In: Boggess, W. R., ed. Lead in the environment. Washington, DC: National Science
56            Foundation; pp. 105-133; NSF report no. NSF/RA-770214.
       December 2005                                8-250         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Ghazi, A. M; Millette, J. R. (2004) Environmental forensic application of lead isotope ratio determination: a case
  2            study using laser ablation sector ICP-MS. Environ. Forensics 5: 97-108.
  3     Giamberini, L.; Pihan, J.-C. (1996) The pericardial glands of the zebra mussel: ultrastructure and implication in lead
  4            detoxication process. Biol. Cell (Paris) 86: 59-65.
  5     Gill, T. S.; Tewari, H.; Pande, J. (1991) Effects of water-borne copper and lead on the peripheral bloodin the rosy
  6            barb, Barbus (Puntius) conchonius Hamilton. Bull. Environ. Contam. Toxicol. 46: 606-612.
  7     Gintenreiter, S.; Ortel, J.; Nopp, H. J. (1993) Bioaccumulation of cadmium, lead, copper, and zinc in successive
  8            developmental stages ofLymantria dispar L. (Lymantriidae, Lepid)—a life cycle study. Arch. Environ.
  9            Contam. Toxicol. 25:  55-61.
10     Giusti, L.; Yang, Y.L.; Hewitt, C.N.; Hamilton-Taylor, J.; Davidson, W. (1993) The solubility and partitioning of
11            atmospherically derived trace metals in artificial and natural waters; A review. Atmos. Environ. A27:
12            1567-1578
13     Glover, L. J., II; Eick, M. J.; Brady, P. V. (2002) Desorption kinetics of cadmium2+ and Iead2+ from goethite:
14            influence of time and organic acids. Soil Sci. Soc. Am. J. 66: 797-804.
15     Gobeil, C.; Johnson, W. K.; Macdonald, R. W.; Wong, C. S. (1995) Sources and burden of lead in St. Lawrence
16            estuary sediments: isotopic evidence. Environ. Sci. Technol. 29: 193-201.
17     Gobeil, C.; Bondeau, B.; Beaudin, L. (2005) Contribution of municipal effluents to metal fluxes in the St. Lawrence
18            River. Environ. Sci. Technol. 39: 456-464.
19     Goldstein, J. I.; Newbury, D. E.; Echlin, P.; Joy, D. C.; Roming, A. D., Jr.; Lyman, C. E.; Fiori, C.; Lifshin, E.
20            (1992) Scanning electron microscopy and x-ray microanalysis: a text for biologists, materials scientists, and
21            geologists. 2nd. Ed. New York, NY: Plenum Press.
22     Gopal, V.; Parvathy, S.; Balasubramanian, P. R. (1997) Effect of heavy metals on the blood protein biochemistry of
23            the fish Cyprinus carpio and its use as a bio-indicator of pollution stress. Environ. Monit. Assess.
24            48: 117-124.
25     Gosz, J. R.; Likens, G. E.; Bormann, F. H. (1976)  Organic matter and nutrient dynamics of the forest and forest floor
26            in the Hubbard Brook forest. Oecologica 22: 305-320.
27     Graney, J. R.; Halliday, A. N.; Keeler, G. J.; Nriagu, J. O.; Robbins, J. A.; Norton, S. A. (1995) Isotopic record of
28            lead pollution in lake  sediments from the northeastern United States. Geochim. Cosmochim.  Acta
29            59: 1715-1728.
30     Graney, J.; Keeler, G.; Norton, S.; Church, S.; Halliday, A. (1996) Historical record of mining at Leadville, CO
31            preserved in sediment from Emerald Lake in Rocky Mountain National Park. In: Geological  Society of
32            America abstracts with programs, v. 28, no.  7. Boulder, CO: Geological Society of America; pp. 156-157.
33     Grelle, C.; Fabre, M.-C.; Lepretre, A.; Descamps,  A. (2000) Myriapod and isopod communities in soils
34            contaminated by heavy metals in northern France. Eur. J. Soil Sci. 51:  425-433.
35     Groenenberg, B. J.; Romkens, P.; Tipping, E.; Pampura, T.; Vries, W. D.;  Schuetze, G.  (2002) Transfer functions for
3 6            the calculation of critical loads for lead, cadmium and mercury: background document for the expert
37            meeting on critical limits for heavy metals and methods for their application; December; Berlin, Germany.
3 8            [draft]. Geneva, Switzerland: United Nations Economic Commission for Europe (UNECE) convention on
3 9            long range transboundary air pollution.
40     Groffman, P. M.;  Driscoll, C.  T.; Fahey, T. J.; Hardy, J. P.; Fitzhugh, R. D.; Tierney, G. L. (2001) Effects of mild
41            winter freezing on soil nitrogen and carbon dynamics in a northern hardwood forest. Biogeochemistry
42            56:-191-213.
43     Gundersen, J. K.; Jorgensen, E. L.; Larsen, E.; Jannasch, H. W. (1992) Mats of giant sulfur bacteria on deep-sea
44            sediments due to fluctuating hydrothermal flow. Nature 360: 454-456.
45     Gupta, S.; Bakre,  P. P. (1996) Influence of lead in various organs ofLymnaea acuminata after supplementation with
46            calcium chloride. Geobios (Jodphur) 23: 251-258.
47     Gupta, M.; Chandra, P. (1994) Lead accumulation and toxicity in Vallisneria spiralis (L.) and Hydrilla verticillata
48            (l.f.) Royle. J. Environ. Sci. Health Part A 29: 503-516.
49     Gupta, S. K.; Chen, K. Y. (1975) Partitioning of trace elements in selective chemical fractions of nearshore
50            sediments. Environ. Lett. 10: 129-158.
51     Gustafsson, J. P.;  Pechova, P.; Berggren, D. (2003) Modeling metal binding to soils: the role of natural organic
52            matter. Environ. Sci. Technol. 37: 2767-2774.
53     Guy, R. D.; Chakrabarti, C. L. (1976) Studies of metal-organic interactions in model systems pertaining to natural
54            waters. Can. J. Chem. 54: 2600-2611.
55     Haack, U.; Kienholz, B.; Reimann, C.; Schneider, J.; Stumpfl, E. F. (2004) Isotopic composition of lead in moss and
56            soil of the European Arctic. Geochim. Cosmochim. Acta 68: 2613-2622.
       December 2005                                8-251         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Haanstra, L.; Doelman, P. (1991) An ecological dose-response model approach to short- and long-term effects of
  2            heavy metals on arylsulfatase activity. Biol. Fertil. Soils 11:  18-23.
  3     Habibi, K. (1973) Characterization of paniculate matter in vehicle exhaust. Environ. Sci. Technol. 7: 223-234.
  4     Haering, K.  C.; Daniels, W. L.; Feagly, S. E. (2000) Reclaiming mined lands withbiosolids, manures, and papermill
  5            sludges. In: Barnhisel, R., ed. Reclamation of drastically disturbed lands. Madison, WI: Soil Science Society
  6            of America; pp. 615-644.
  7     Hagopian-Schlekat, T.; Chandler, G. T.; Shaw, T. J. (2001) Acute toxicity of five sediment-associated metals,
  8            individually and in a mixture, to the estuarine meiobenthic harpacticoid copepodAmphiascus tenuiremis.
  9            Mar. Environ. Res. 51: 247-264.
10     Hamelin, B.; Grousset, F.;  Sholkovitz, E. R. (1990) Pb isotopes in surficial pelagic sediments from the North
11            Atlantic. Geochim. Cosmochim. Acta 54: 37-47.
12     Hansmann, W.; Koppel, V. (2000) Lead-isotopes as tracers of pollutants in soils. Chem. Geol. 171: 123-144.
13     Harper, P. P.; Stewart, K. W. (1984) Plecoptera. In: Merritt, R. W.; Cummins, K. W. eds. An introduction to the
14            aquatic insects of North America. Dubuque, IA: Kendall-Hunt Publishing Company.
15     Hassler, C. S.; Slaveykova, V. L; Wilkinson, K. J. (2004) Some fundamental (and often overlooked) considerations
16            underlying the free ion activity and biotic ligand models. Environ. Toxicol. Chem. 23: 283-291.
17     He, P. P.; Lv, X. Z.; Wang, G. Y. (2004) Effects of Se and Zn supplementation on the antagonism against Pb and Cd
18            in vegetables. Environ. Int.  30: 167-172.
19     Healy, M. A.; Harrison, P.  G.; Aslam, M.; Davis, S. S.; Wilson, C. G. (1992) Lead sulphide and traditional
20            preparations: routes for ingestion, and solubility and reactions in gastric fluid. J. Clin. Hosp. Pharm.
21            7: 169-173.
22     Henny, C. J.; Blus, L. J.; Hoffman, D. J.; Grove, R. A.; Hatfield, J. S. (1991) Lead accumulation and osprey
23            production near a mining site on the  Coeur d'Alene River, Idaho. Arch. Environ. Contam. Toxicol.
24            21:415-424.
25     Hercules, D. M. (1970) Electron spectroscopy.  Anal. Chem. 42: 20a-40a.
26     Herkovits, J.; Perez-Coll, C. S.  (1991) Antagonism and synergismbetween lead and zinc in amphibian larvae.
27            Environ. Pollut. 69: 217-221.
28     Hettelingh, J.-P.; Downing, R. J.; De Smet, P.A.M. (1991) Mapping critical loads for Europe: CCE technical report
29            no.  1. Bilthoven, The Netherlands: National Institute of Public Health and Environmental Protection;
30            Coordination Center for Effects; RIVM report no. 259101001.
31     Hettelingh, J.-P.; Slootweg, J.; Posch, M., eds. (2002) Preliminary modelling and mapping of critical loads for
32            cadmium and lead in Europe. Bilthoven, The Netherlands: National Institute of Public Health and the
33            Environment; RIVM report no. 259101011.
34     Hettiarachchi, G. M.; Pierzynski, G. M.; Oehne, F. W.; Sonmez, O.; Ryan, J. A. (2001) In situ stabilization of soil
35            lead using phosphorus. J. Environ. Qual. 30: 1214-1221.
36     Hettiarachchi, G. M.; Pierzynski, G. M.; Oehne, F. W.; Sonmez, 0.; Ryan, J. A. (2003) Treatment of contaminated
37            soil with phosphorus and manganese oxide reduces lead absorption by Sprague-Dawley rats. J. Environ.
38            Qual. 32: 1335-1345.
39     Heyl, A. V.; Landis, G. P.; Zartman, R. E. (1974) The isotopic evidence for the origins of Mississippi Valley-type
40            mineral deposits. Econ. Geol. Bull. Soc. Econ.  Geol. 69: 992-1006.
41     Ho, M. D.; Evans, G. J. (2000)  Sequential extraction of metal contaminated soils with radiochemical assessment of
42            readsorption effects. Environ. Sci. Technol. 34: 1030-1035.
43     Hodson, P. V.; Blunt, B. R.; Spry, D. J. (1978)  Chronic toxicity of water-borne and dietary lead to rainbow trout
44            (Salmo gairdnerf)  in Lake Ontario water. Water Res. 12: 869-878.
45     Hoffman, D. J.; Heinz, G. H.; Sileo, L.; Audet, D. J.; Campbell, J. K.; LeCaptain, L. J. (2000a) Developmental
46            toxicity of lead-contaminated sediment to mallard ducklings. Arch. Environ. Contam. Toxicol. 39: 221-232.
47     Hoffman, D. J.; Heinz, G. H.; Sileo, L.; Audet, D. J.; Campbell, J. K.; LeCaptain, L. J.; Obrecht, H. H., III. (2000b)
48            Developmental toxicity of lead-contaminated sediment in Canada geese (Branta Canadensis). J. Toxicol.
49            Environ. Health A 59: 235-252.
50     Holcombe, G. W.; Benoit,  D. A.; Leonard, E. N.; McKim, J. W.  (1976) Long-term effects of lead exposure on three
51            generations of brook trout (Salvelinusfontinalis). J. Fish Res. Board Can. 33: 1731-1734.
52     Hopkin, S. P. (1989) Ecophysiology of metals in terrestrial invertebrates. London, United Kingdom: Elsevier
53            Applied Science.
54     Home, M. T.; Dunson, W.  A. (1995a) Toxicity of metals and low pH to embryos and larvae of the Jefferson
55            salamander, Ambystomajeffersonianum. Arch. Environ. Contam. Toxicol. 29: 110-114.
       December 2005                                8-252         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Home, M. T.; Dunson, W. A. (1995b) The interactive effects of low pH, toxic metals, and DOC on a simulated
  2            temporary pond community. Environ. Pollut. 89: 155-161.
  3     Home, M. T.; Dunson, W. A. (1995c) Effects of low pH, metals, and water hardness on larval amphibians.
  4            Arch. Environ.  Contam. Toxicol. 29: 500-505.
  5     Houlton, B. Z.; Driscoll, C. T.; Fahey, T. I; Likens, G. E.;  Groffman, P. M.; Bernhardt, E. S.; Buso, D. C. (2003)
  6            Nitrogen dynamics in ice storm-damaged forest  ecosystems: implications for nitrogen limitation theory.
  7            Ecosystems 6: 431-443.
  8     Huang, J.-H.; Matzner, E. (2004) Biogeochemistry of trimethyllead and lead in a forested ecosystem in Germany.
  9            Biogeochemistry  71: 125-139.
10     Hughes, J. W.; Fahey, T. J. (1994) Litterfall dynamics and  ecosystem recovery during forest development. For. Ecol.
11            Manage. 63: 181-198.
12     Humphreys, D. J. (1991) Effects of exposure to excessive quantities of lead on animals. Br. Vet. J. 147: 18-30.
13     Hunt, A.; Johnson, D. L.;  Watt, J. M.; Thornton, I. (1992) Characterizing the sources of paniculate lead in house
14            dust by automated scanning electron microscopy. Environ. Sci. Technol. 26: 1513-1523.
15     Ingersoll, C. G.; Haverland, P. S.; Branson, E. L.; Canfield, T. J.; Dwyer, F. J.; Henke,  C. E.; Kemble, N. E.; Mount,
16            D.R.;Fox,R. G.  (1996) Calculation and evaluation of sediment effect concentrations for the amphipod
17            (Hyalella azteca) and the midge (Chironomus riparius). J.  Great Lakes Res. 22: 602-623.
18     Isaure, M.-P.; Laboudigue, A.; Manceau, A.; Sarret, G.; Tiffreau, C.; Trocellier, P.; Lamble, G.; Hazemann, J.-L.;
19            Chateigner, D. (2002) Quantitative Zn speciation in a contaminated dredged sediment by  u-PIXE, u-SXRF,
20            and EXAFS spectroscopy and principal component analysis. Geochim. Cosmochim.  Acta 66: 1549-1567.
21     Jackson, B. P.; Williams,  P. L.; Lanzirott, A.; Bertsch, P. M. (2005) Evidence forbiogenic pyromorphite formation
22            by the nematode caenorhabditis elegans. Environ. Sci. Technol. 39: 5620-5625.
23     James, E.; Henry, C. (1993) Lead isotopes of ore deposits in Trans-Pecos, Texas and northeastern Chihuahua,
24            Mexico:  basement, igneous, and sedimentary sources of metals. Econ. Geol. Bull. Soc. Econ. Geol.
25            88: 934-947.
26     Jampani, C. S. R. (1988) Lead toxicity to alga Synechococcus aeruginosus and its recovery by nutrients. J. Environ.
27            Biol. 9: 261-269.
28     Janssen, M. P. M.; Hogervorst, R. F. (1993) Metal accumulation in soil arthropods in relation to micro-nutrients.
29            Environ. Pollut. 79: 181-189.
30     Jayaraj, Y. M.; Mandakini, M.; Nimbargi, P. M. (1992) Effect of mercury and lead on primary productivity of two
31            water bodies. Environ. Ecol. 10: 653-658.
32     Jeanroy, E.; Guillet, B. (1981) The occurrence of suspended ferruginous particles in pyrophosphate extracts of some
33            soil horizons. Geoderma 26: 95-105.
34     Jenne, E. A.;  Luoma, S. N. (1977) Forms of trace elements in soils, sediments, and associated waters: an overview
35            of their determination and biological availability. In: Dracker, H.; Wildung, R.  E., eds. Biological
3 6            implication of metals in the environmnet. Proceedings of the fifteenth annual Hanford life sciences
37            symposium; September-October 1975; Richland, WA. Washington, DC: Energy Research and Developmnet
38            Administration; pp. 110-143. Available from: NTIS, Springfield, VA; CONF-750920.
39     Jentschke, G.; Marschner, P.; Vodnik, D.; Marth, C.; Bredemeier, M.; Rapp, C.; Fritz, E.; Gogala, N.; Godbold,
40            D. L. (1998) Lead uptake by Picea abies seedlings: effects of nitrogen source and mycorrhizas. J. Plant
41            Physiol.  153: 97-104.
42     Jersak, J.; Amundson, R.; Brimhall, G., Jr. (1997) Trace  metal geochemistry in spodosols of the northeastern United
43            States. J. Environ. Qual. 26: 511 -521.
44     Johansson, K.; Bringmark, E.; Lindevall, L.; Wilander, A. (1995) Effects of acidification on the concentrations of
45            heavy metals in running waters in Sweden. Water Air Soil Pollut. 85: 779-784.
46     Johansson-Sjobeck, M.-L.; Larsson, A. (1979) Effects of inorganic lead on delta-aminolevulinic dehydratase activity
47            and haematological variables in the rainbow trout,  Salmo gairdnerii. Arch. Environ. Contam. Toxicol.
48            8:419-431.
49     Johnson, D.; Hale, B. (2004) White birch (Betulapapyri/era Marshall) foliar litter decomposition in relation to trace
50            metal atmospheric inputs at metal-contaminated and uncontaminated sites near Sudbury, Ontario and
51            Rouyn-Noranda, Quebec, Canada. Environ. Pollut. 127: 65-72.
52     Johnson, C. E.; Petras, R.  J. (1998) Distribution of zinc and lead fractions within a forest spodosol. Soil Sci. Soc.
53            Am. J. 62: 782-789.
54     Johnson, C. E.; Driscoll, C. T.; Fahey, T. J.; Siccama, T. G.; Hughes, J. W. (1995a) Carbon dynamics following
5 5            clear-cutting of a northern hardwood forest. In: Kelly, J. M.; McFee, W. W., eds.  Carbon forms and function
56            in forest soils. Madison, WI: Soil Science Society of America; pp.  463-488.
       December 2005                                8-253        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Johnson, C. E.; Siccama, T. G.; Driscoll, C. T.; Likens, G. E.; Moeller, R. E. (1995b) Changes in lead
  2            biogeochemistry in response to decreasing atmospheric inputs. Ecol. Appl. 5: 813-822.
  3     Johnson, C. E.; Petras, R. J.; April, R. H.; Siccama, T. G. (2004) Post-glacial lead dynamics in a forest soil. Water
  4            Air Soil Pollut. 4: 579-590.
  5     Jones, K. C.; Johnston, A. E. (1991) Significance of atmospheric inputs of lead to grassland at one site in the United
  6            Kingdom since 1869. Environ.  Sci. Technol. 25:  1174-1178.
  7     Jordan, M. J. (1975) Effects of zinc smelter emissions and fire on a chestnut-oak woodland. Ecology 56: 78-91.
  8     Kapustka, L. A.; Lipton, J.; Galbraith, H.; Cacela, D.; Lejeune, K. (1995) Metal and arsenic impacts to soils,
  9            vegetation communities, and wildlife habitat in southwest Montana uplands contaminated by smelter
10            emissions: II. Laboratory phytotoxicity studies. Environ. Toxicol. Chem. 14: 1905-1912.
11     Karamanos, R. E.; Bettany, J. R.; Rennie, D. A. (1976) Extractability of added lead in soils using lead-210. Can. J.
12            Soil Sci. 56: 37-42.
13     Kaste, J.; Friedland, A.; Stirrup, S. (2003) Using stable and radioactive isotopes to trace atmospherically deposited
14            Pb in montane forest soils. Environ. Sci. Technol. 37: 3560-3567.
15     Kaste, J. M.; Friedland, A. J.; Miller, E. K. (2005) Potentially mobile lead fractions in montane organic-rich soil
16            horizons.  Water Air Soil Pollut. 167: 139-154.
17     Keeney, W. L.; Breck, W. G.; VanLoon, G. W.; Page, J. A. (1976) The determination of trace metals in Cladophora
18            glomerata - C. glomerata as a potential biological monitor. Water Res. 10: 981-984.
19     Keon, N. E.; Swartz, C. H.; Brabander, D. J.; Harvey, C.; Hemond, H. F. (2001) Validation of an arsenic sequential
20            extraction method for evaluation mobility in sediments. Environ. Sci. Technol. 35: 2778-2784.
21     Ketterer, M. E.; Peters, M. J.; Tisdale, P. J. (1991) Verification of a correction procedure for measurement of lead
22            isotope rations by inductively coupled plasma mass spectrometry. J. Anal. At. Spectrom. 6: 439-443.
23     Ketterer, M. E.; Lowry, J. H.; Simon, J.; Humphries, K.; Novotnak, M. P. (2001) Lead isotopic and chalcophile
24            element compositions in the environment near a zinc smelting-secondary zinc recovery facility, Palmerton,
25            Pennsylvania, USA. Appl. Geochem. 16: 207-229.
26     Khan, D. H.; Frankland, B. (1983) Effects of cadmium and lead on radish plants with particular reference to
27            movement of metals through soil profile and plant. Plant Soil 70: 335-345.
28     Khangarot, B.  S. (1991) Toxicity of metals to a freshwater tubificid worm, Tubifex tubifex (Muller). Bull. Environ.
29            Contam. Toxicol. 46: 906-912.
30     Kheboian, C.;  Bauer, C. F. (1987) Accuracy of selective extraction procedures for metal speciation in model aquatic
31            sediments. Anal. Chem. 59: 1417-1423.
32     Kim, S.-J.; Rodriguez-Lanetty, M.; Suh, J.-H.; Song, J.-I. (2003) Emergent effects of heavy metal pollution at a
33            population level: Littorina brevicula a case study. Mar. Pollut. Bull. 46: 74-80.
3 4     Klaminder, J.; Bindler, R.; Emteryd, O.; Renberg, I. (2005) Uptake and recycling of lead by boreal forest plants:
35            quantitative estimates from a site in northern Sweden. Geochim. Cosmochim. Acta 69: 2485-2496.
36     Knowlton, M.  F.;  Boyle, T. P.; Jones, J. R. (1983) Uptake of lead from aquatic sediment by submersed macrophytes
37            and crayfish. Arch. Environ.  Contam. Toxicol. 12: 535-541.
38     Koch, D. M.; Jacob, D. J.; Graustein, W. C. (1996) Vertical transport of tropospheric aerosols as indicated by 7Be
39            and 210Pb in a chemical tracer model. J. Geophys. Res. (Atmos.) 101(D13): 18651-18666.
40     Kock, G.; Triendl, M.; Hofer, R. (1996) Seasonal patterns of metal accumulation in Arctic char (SalveUnits alpinus)
41            from an oligotrophic Alpine lake related to temperature. Can.  J. Fish. Aquat. Sci. 53: 780-786.
42     Kraak, M. H. S.; Wink, Y. A.; Stuijfzand, S. C.; Buckert-de Jong, M. C.; de Groot, C. J.; Admiraal, W. (1994)
43            Chronic ecotoxicity of Zn and Pb to the zebra mussel Dreissenapolymorpha. Aquat. Toxicol. 30: 77-89.
44     Kruatrachue, M.; Jarupan, W.; Chitramvong, Y. P.; Pokethitiyook, P.; Upatham, E. S.; Parkpoomkamol, K. (2002)
45            Combined effects of lead and humic acid on growth and lead uptake of duckweed, Lemna  minor. Bull.
46            Environ. Contam. Toxicol. 69: 655-661.
47     Kruger, F.; Grongroft, A. (2004) The difficult assessment of heavy metal contamination of soils and plants in Elbe
48            River  floodplains. Acta Hydrochim. Hydrobiol. 31: 436-443.
49     Kuperman, R.  G.; Carreiro, M. M. (1997) Soil heavy metal concentrations, microbial biomass and enzyme activities
50            in a contaminated grassland ecosystem. Soil Biol. Biochem. 29:  179-190.
51     Kutlu, M.; Susuz, F. (2004) The effects of lead as an environmental pollutant on EROD enzyme in Gammaruspulex
52            (L.) (Crustacea: Amphipoda). Bull. Environ. Contam. Toxicol. 72: 750-755.
53     Lamy, I.; Bourgeosis, S.; Bermond, A. (1993)  Soil cadmium mobility as a consequence of sewage sludge disposal.
54            J. Environ. Qual. 22:731-737.
5 5     Lang, F.; Kaupenjohann, M. (2003) Effect of dissolved organic matter on the precipitation and mobility of the lead
56            compound chloropyromorphite in solution. Eur. J. Soil Sci. 54: 139-147.
       December 2005                                8-254         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Laskowski, R.; Maryanski, M.; Nikliriska, M. (1994) Effect of heavy-metals and mineral nutrients on forest litter
  2            respiration rate. Environ. Pollut. 84: 97-102.
  3     Laskowski, R.; Maryanski, M.; Niklinska, M. (1995) Changes in the chemical composition of water percolating
  4            through the soil provile in a moderately polluted catchment. Water Air Soil Pollut. 85: 1759-1764.
  5     Leach, D.; Hofstra, A.; Church, S.; Snee, L.; Vaughn, R.; Zartman, R. (1998) Evidence for proterozoic and late
  6            cretaceous - early tertiary ore-forming events in the Coeur D'Alene district, Idaho and Montana. Econ. Geol.
  7            Bull. Soc. Econ. Geol. 93:347-359.
  8     Lefcort, H.; Meguire, R. A.; Wilson, L. H.;  Ettinge, W. F. (1998) Heavy metals alter the survival, growth,
  9            metamorphosis, and antipredatory behavior of Columbia spotted frog (Rana luteiventris) tadpoles. Arch.
10            Environ. Contam. Toxicol. 35: 447-456.
11     Lefcort, H.; Thomson, S. M.; Cowles, E. E.; Harowicz, H. L.; Livaudais, B. M.; Roberts, W. E.; Ettinger, W. F.
12            (1999) Ramifications  of predator avoidance: predator and  heavy-metal-mediated competition between
13            tadpoles and snails. Ecol. Appl. 9: 1477-1489.
14     Lefcort, H.; Ammann, E.; Eiger, S. M. (2000) Antipredatory behavior as an index of heavy-metal pollution? A test
15            using snails and caddisflies. Arch. Environ. Contamin. Toxicol. 38:311-316.
16     Lefcort, H.; Abbott, D. P.; Cleary, D. A.; Howell, E.; Kellar, N. C.; Smith, M. M. (2004) Aquatic snails from mining
17            sites  have evolved to detect and avoid heavy metals. Arch. Environ.  Contam. Toxicol. 46: 478-484.
18     Leita, L.; De  Nobili, M.; Pardini, G.; Ferrari, F.; Sequi, P. (1989) Anomolous contents of heavy metals in soils and
19            vegetation of a mine area in S.W. Sardinia, Italy. Water Air Soil Pollut. 48: 423-433.
20     Lewis, T. E.; Mclntosh, A. W. (1986) Uptake of sediment-bound lead and zinc by the freshwater isopodAsellus
21            communis at three different pH levels. Arch. Environ. Contam. Toxicol. 15: 495-504.
22     Li, W.-H.; Chan, P. C. Y.; Chan, K. M. (2004) Metal uptake in zebrafish embryo-larvae exposed to metal-
23            contaminated sediments.  Mar. Environ. Res. 58: 829-832.
24     Liang, C.  N.; Tabatabai, M. A. (1977) Effects of trace elements on nitrogen mineralisation in soils. Environ. Pollut.
25            12: 141-147.
26     Liang, C.  N.; Tabatabai, M. A. (1978) Effects of trace elements on nitrification in soils. J. Environ. Qual.
27            7:291-293.
28     Likens, G. E., ed. (1989) Long-term studies in ecology: approaches and alternatives. Papers from the Second Gary
29            Conference; May, 1987; Millbrook, NY. New York, NY: Springer-Verlag, Inc.
30     Likens, G. E.; Bormann, F. H.; Johnson, N.  M. (1969) Nitrification: importance to nutrient losses from a cutover
31            forested ecosystem. Science (Washington, DC) 163: 1205-1206.
32     Likens, G. E.; Driscoll, C. T.;  Buso, D. C. (1996) Long-term effects of acid rain: response and recovery of a forest
33            ecosystem. Science (Washington, DC) 272: 244-246.
34     Lin, Q.; Chen, Y. X.; He, Y. F.; Tian,  G. M. (2004) Root-induced changes of lead availability in the rhizosphere of
35            Oryza sativa L. Agric. Ecosyst. Environ.  104: 605-613.
36     Linton, R. W.; Natusch, D. F.  S.;  Solomon,  R. L.; Evans, C. A., Jr. (1980) Physicochemical characterization of lead
37            in urban dusts. A microanalytical approach to lead tracing. Environ.  Sci. Technol. 14: 159-164.
38     Little, P.;  Martin, M. H. (1972) A survey of zinc, lead and cadmium in soil and natural vegetation around a smelting
39            complex. Environ. Pollut. 3: 241-254.
40     Liu, J.;  Li, K.; Xu, J.; Zhang, Z.; Ma, T.; Lu, X.; Yang, J.; Zhu, Q. (2003) Lead toxicity, uptake, and translocation in
41            different rice cultivars. Plant Sci. 165: 793-802.
42     Lock, K.; Janssen, C. R. (2002) Multi-generation toxicity of zinc, cadmium,  copper and lead to the potworm
43            Enchytraeus albidus. Environ. Pollut. 117: 89-92.
44     Lofts, S.;  Spurgeon, D. J.; Svendsen, C.; Tipping, E. (2004) Deriving soil critical limits for Cu, Zn, Cd, and Pb: a
45            method based on free  ion concentrations. Environ. Sci. Technol. 38(13): 3623-3631.
46     Long, D. T.; Angino, R. E. (1977) Chemical speciation of Cd, Cu, Pb, and Zn, in mixed freshwater, seawater, and
47            brine solutions. Geochim. Cosmochim. Acta.  41:1183-1191.
48     Long, E. R.; MacDonald, D. D.; Smith, S. L.; Calder, F. D. (1995) Incidence of adverse biological effects within
49            ranges of chemical concentrations in marine and estuarine sediments. Environ. Manage.  19: 81-97.
50     Levering, T.  G., ed. (1976) Lead in the environment. Washington, DC: U.S.  Department of the Interior, Geological
51            Survey; Geological Survey professional paper no. 957. Available from: GPO, Washington, DC; S/N
52            024-001-02911-1.
53     Lumsdon, D. G.; Evans, L. J. (1995) Predicting chemical speciation and computer simulation. In: Ure, A. M.;
54            Davidson, C. M., eds. Chemical speciation in the environment. London, United Kingdom: Blackie;
55            pp. 86-134.
       December 2005                                8-255         DRAFT-DO NOT QUOTE OR CITE

-------
  1    Luoma, S. N.; Rainbow, P. S. (2005) Why is metal bioaccumulation so variable? Biodynamics as a unifying
  2            concept. Environ. Sci. Technol. 39: 1921-1931.
  3    Ma, W.-C. (1996) Lead in mammals. In: Beyer, W. N.; Heinz, G. H.; Redmon-Norwood, A. W., eds. Environmental
  4            contaminants in wildlife: interpreting tissue concentrations. Boca Raton, FL: CRC Press. [SETAC special
  5            publications series].
  6    Ma, Y. B.; Uren, N. C. (1995) Application of new fractionation scheme for heavy metals in soils. Commun. Soil Sci.
  7            Plant Anal. 26: 3291-3303.
  8    Ma, Q. Y.; Logan, T. J.; Traina, S. J. (1995) Lead immobilization from aqueous solutions and contaminated soils
  9            using phosphate rocks. Environ. Sci. Technol. 29: 1118-1126.
10    MacDonald, D. D.; Ingersoll, C. G.; Berger, T. A. (2000) Development and evaluation of consensus-based sediment
11            quality guidelines for freshwater ecosystems. Arch. Environ. Contam. Toxicol. 39: 20-31.
12    MacDonald, A.; Silk, L.; Schwartz, M.; Playle, R. C. (2002) A lead-gill binding model to predict acute lead toxicity
13            to rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. Part C: Toxicol. Pharmacol.
14            133C: 227-242.
15    Mackie, G. L. (1989) Tolerances of five benthic invertebrates to hydrogen ions and metals  (Cd, Pb, Al). Arch.
16            Environ. Contam. Toxicol. 18: 215-223.
17    MacLean, R. S.; Borgmann, U.; Dixon, D. G. (1996) Bioaccumulation kinetics and toxicity of lead inffyalella
18            azteca (Crustacea, Amphipoda). Can. J. Fish. Aquat. Sci. 53: 2212-2220.
19    Maddock, E.G.; Taylor, D. (1980) The acute toxicity and bioaccumulation of some lead alkyl compounds in marine
20            animals. In: Branica, M.; Konrad, Z., eds. Lead in the marine environment. Oxford, United Kingdom:
21            Pergamon Press; pp. 233-261.
22    Maenhaut, W. (1987) Particle-induced x-ray emission spectrometry: an accurate technique in the analysis of
23            biological environmental and geological samples. Anal. Chim. Acta. 195: 125-140.
24    Maginn, S. J. (1998) Analytical applications of synchrotron radiation.  Analyst 123: 19-29.
25    Malcova, R.; Gryndler, M. (2003) Amelioration of Pb and Mn toxicity to arbuscular mycorrhizal fungus Glomus
26            intraradices\yy maize root exudates. Biol. Plant.  47: 297-299.
27    Manceau, A.; Boisset, M.; Sarret, G.; Hazemann, J.; Mench, M.; Cambier, P.; Prost, R. (1996) Direct determination
28            of lead speciation in contaminated soils by EXAFS spectroscopy. Environ. Sci. Technol. 30: 1540-1552.
29    Manceau, A.; Lanson, B.; Schlegel, M. L.; Harge, J. C.; Musso, M.; Eybert-Berard, L.; Hazemann, J.-L.; Chateigner,
30            D.; Lamble, G. M. (2000a) Quantitative Zn speciation in smelter-contaminated soils by EXAFS
31            spectroscopy. Am. J. Sci. 300: 289-343.
32    Manceau, A.; Schlegel, M. L.; Musso, M.; Sole, V. A.; Gauthier, C.; Petit, P. E.; Trolard, F. (2000b) Crystal
33            chemistry of trace elements in natural and synthetic goethite. Geochim. Cosmochim. Acta 64: 3643-3661.
34    Manceau, A.; Lanson, B.; Drits, V. A. (2002) Structure of heavy metal sorbedbirnessite. Part III: results from
3 5            powder and polarized extended x-ray absorption fine structure spectroscopy. Geochim. Cosmochim. Acta
36            66:2639-2663.
37    Maret, T. R.; Cain, D. J.; MacCoy, D. E.; Short, T. M. (2003) Response of benthic invertebrate assemblages to metal
3 8            exposure and bioaccumulation associated with hard-rock mining in northwestern streams, USA.  J. N. Am.
39            Benthol. Soc. 22: 598-620.
40    Marinussen, M. P. J. C.; Van der Zee, S. E. A. T. M.; de Haan, F. A. M.; Bouwman, L. M.; Hefting, M. M. (1997)
41            Heavy metal (copper, lead, and zinc) accumulation and excretion by the earthworm, Dendrobaena veneta. J.
42            Environ. Qual. 26: 278-284.
43    Marschner, P.; Godbold, D. L.; Jentschke, G. (1996) Dynamics of lead accumulation in mycorrhizal and non-
44            mycorrhizal Norway spruce (Picea abies (L.) Karst). Plant Soil 178: 239-245.
45    Marschner, P.; Klam, A.; Jentschke, G.; Godbold, D. L. (1999) Aluminium and lead tolerance in ectomycorrhizal
46            fungi. J. Plant Nutr. Soil Sci. 162: 281-286.
47    Marsh, A. S.; Siccama, T. G. (1997) Use of formerly plowed land in New England to monitorthe vertical
48            distribution of lead, zinc and copper in mineral soil. Water Air Soil Pollut. 95: 75-85.
49    Martin, R. B. (1986) Bioinorganic chemistry of metal ion toxicity. In:  Sigel, H., ed. Concepts on metal ion toxicity.
50            New York, NY: Marcel Dekker; pp. 21-66. (Metal ions in biological systems: v. 20).
51    Martin, M. H.; Bullock, R. J. (1994) The impact and fate  of heavy metals in an oak woodland ecosystem. In: Ross,
52            S. M., ed. Toxic metals in soil-plant systems. Chichester, England: John Wiley & Sons; pp.  327-365.
53    Mateo, R.; Hoffman, D. J. (2001) Differences in oxidative stress between young Canada geese and mallards exposed
54            to lead-contaminated sediment. J. Toxicol. Environ. Health Part A 64: 531-545.
      December 2005                               8-256         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Mateo, R.; Beyer, W. N.; Spann, J. W.; Hoffman, D. J. (2003a) Relation of fatty acid composition in lead-exposed
  2            mallards to fat mobilization, lipid peroxidation and alkaline phosphatase activity. Comp. Biochem. Physiol.
  3            C Pharmacol. Toxicol. 135: 451-458.
  4     Mateo, R.; Beyer, W. N.; Spann, J. W.; Hoffman, D. J.; Ramis, A. (2003b) Relationship between oxidative stress,
  5            pathology, and behavioral signs of lead poisoning in mallards. J. Toxicol. Environ. Health A 66: 1371-1389.
  6     McBride, M. B.; Richards, B. K.; Steenhuis, T.; Russo, J. J.;  Suave, S. (1997) Mobility and solubility of toxic metals
  7            and nutrients in soil fifteen years after sludge application. Soil Sci. 162: 487-500.
  8     McBride, M. B.; Richards, B. K.; Steenhuis, T. Spiers, G. (1999) Long-term leaching of trace elements in a heavily
  9            sludge-amended silty clay loam soil. Soil Sci. 164: 613-624.
10     McCarthy, J. F. (1989) Bioavailability and toxicity of metals and hydrophobic organic contaminants. In: Suffet, I.
11            H.; MacCarthy, P., eds. Aquatic humic substances: influence on fate and treatment of pollutants;
12            pp. 263-277'. Washington, DC: American Chemical Society. (Advances in chemistry series no. 219).
13     McCrea, A. R.; Trueman, I. C.; Fullen, M. A. (2004) Factors relating to soil fertility and species diversity in both
14            semi-natural and created meadows in the west midlands of England. Eur. J. Soil Sci. 55: 335-348.
15     Merino, A.; Garcia-Rodeja, E. (1997) Heavy metal and aluminium mobilization in soils from Galicia (NW spain) as
16            a consequence of experimental acidification. Appl. Geochem. 12: 225-228.
17     Merlini, M.; Pozzi, G. (1977) Lead and freshwater fishes: part 1—lead accumulation and water pH. Environ. Pollut.
18            12: 167-172.
19     Miller, E. K.;Friedland,  A. J. (1994) Lead migration in forest soils: response to changing atmospheric inputs.
20            Environ. Sci. Technol. 28: 662-669.
21     Miller, T. G.; Mackay, W.  C. (1980) The effects of hardness, alkalinity and pH of test water on the toxicity of
22            copper to rainbow  trout (Salmo gairdneri). Water Res.  14: 129-133.
23     Miller, W.  P.; McFee, W. W. (1983) Distribution of cadmium, zinc, copper, and lead in soils of industrial
24            northwestern Indiana. J. Environ. Qual. 12:  29-33.
25     Miller, E. K.; Friedland,  A. J.; Arons, E. A.; Mohnen, V. A.;  Battles, J. J.; Panek, J. A.; Kadlecek, J.; Johnson, A. H.
26            (1993) Atmospheric deposition to forests along an elevational gradient at Whiteface-Mountain, NY, USA.
27            Atmos. Environ. 27: 2121-2136.
28     Millward, R. N.; Carman, K. R.; Fleeger, J. W.; Gambrell, R. P.; Powell, R. T.; Rouse, M.-A. (2001) Linking
29            ecological impact to metal concentrations and speciation: a microcosm experiment using a salt marsh
30            meiofaunal community. Environ. Toxicol. Chem. 20: 2029-2037.
31     Miranda, M. G.; Ilangovan, K.  (1996) Uptake ofleadbyLemnagibbaL.: influence on specific growth rate and
32            basic biochemical  changes.  Bull. Environ. Contam. Toxicol. 56: 1000-1007.
33     Mitchell, M. J.; Driscoll, C. T.; Kahl, J. S.; Likens, G. E.; Murdoch, P. S.; Pardo, L. H.  (1996) Climatic  control of
34            nitrate loss from forested watersheds in the  northeast United States. Environ. Sci. Technol. 30: 2609-2612.
35     Mize, S. V.; Deacon, 1. R. (2002)  Relations of benthic macroinvertebrates to concentrations of trace elements in
36            water, streambed sediments, and transplanted bryophytes and stream habitat conditions in nonmining and
37            mining areas of the Upper Colorado River Basin, Colorado, 1995-98. Denver, CO: U.S. Geological Survey;
38            Water-Resources Investigations Report 02-4139. Available:
39            http://pubs.usgs.gov/wri/wri024139/pdfAVRI02-4139.pdf [24 October, 2005].
40     Monna, F.; Othman, D. B.; Luck, J.  M. (1995) Pb isotopes and Pb, Zn and Cd concentrations in the rivers feeding a
41            coastal pond (Thau, southern France): constraints on the origin(s) and flux(es) of metals. Sci. Total Environ.
42            166: 19-34.
43     Moore, H.  E.; Poet, S. E. (1976) 210Pb fluxes determined from 210Pb and 226Ra soil profiles. J. Geophys. Res.
44            (Oceans and Atmos.) 81:  1056-1058.
45     Morel, F. (1983) Principles of Aquatic Chemistry. New York, NY: John Wiley and Sons.
46     Morel, F. M. M.; Westall, J. E.; O'Melia, C. R.; Morgan, J. J. (1975) Fate of trace metals in Los Angeles County
47            wastewater discharge. Environ.  Sci. Technol. 9: 756-761.
48     Morelli, E.; Scarano, G. (2001) Synthesis and stability of phytochelatins induced by cadmium and lead in the marine
49            diatomPhaeodactylum tricornutum. Mar. Environ. Res. 52: 383-395.
50     Morgan, J. E.; Morgan, A.  J. (1988) Earthworms as  biological monitors of cadmium, copper, lead and zinc in
51            metalliferous soils. Environ. Pollut. 54: 123-138.
52     Mucha, A.  P.; Vasconcelos M.  T. S. D.; Bordalo A.  A. (2003) Macrobenthic community in the Douro estuary:
53            relations with trace metals and natural sediment characteristics. Environ. Pollut. 121:  169-180.
54     Murray, K. S.; Rogers, D. T.; Kaufman, M. M.  (2004) Heavy metals in an urban watershed in southeastern
55            Michigan. J.  Environ. Qual. 33: 163-172.
       December 2005                                8-257         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Nakagawa, H.; Nakagawa, K.; Sato, T. (1995a) Evaluation of erythrocyte 5-aminolevulinic acid dehydratase activity
  2            in the blood of carp Cyprinus carpio as an indicator in fish with water lead pollution. Fish. Sci. 61:  91 -95.
  3     Nakagawa, H.; Sato, T.; Kubo, H. (1995b) Evaluation of chronic toxicity of water lead for carp (Cyprinus carpio)
  4            using its blood 5-aminolevulinic acid dehydratase. Fish. Sci. 61: 956-959.
  5     National Research Council of Canada. (1973) Lead in the Canadian Environment. Ottawa, Canada: National
  6            Research Council of Canada; NRCC no. 13682; Environmental Secretariat publication BY73-7 (ES).
  7     National Research Council (NRC), Committee on Bioavailability of Contaminants in Soils and Sediments. (2002)
  8            Bioavailability of contaminants in soil and sediments: processes, tools and applications. Washington, DC:
  9            National Academies Press.
10     Nelson, Y. W.; Lo, W.; Lion, L. W.;  Shuler, M. L.; Ghiorse, W. C. (1995) Lead distribution in a simulated aquatic
11            environment: effects of bacterial biofilms and iron oxide. Wat. Res. 29(8): 1934-1944.
12     Neuhauser, E. F.; Loehr, R. C.; Milligan, D. L.; Malecki, M. R. (1985) Toxicity of metals to the earthworm Eisenia
13           foetida. Biol. Fertil. Soils 1:  149-152.
14     Neuhauser, E. F.; Cukic, Z. V.; Malecki, M. R.; Loehr, R. C.; Durkin, P. R. (1995) Bioconcentration andbiokinetics
15            of heavy metals in the earthworm. Environ. Pollut. 89: 293-301.
16     Newhook, R.; Hirtle, H.; Byrne, K.; Meek, M.E. (2003) Releases from copper smelters and refineries and zinc plants
17            in Canada: human health exposure and risk characterization. Sci. Total Environ. 301: 23-41.
18     Newman, M. C.; Dixon, P. M.; Looney, B. B.; Finder, J. E. I. (1989) Estimating mean and variance for
19            environmental samples with below detection limit observations. Water Resour. Bull. 25: 905-910.
20     Nikliriska, M.; Laskowski, R.; Maryariski, M. (1998) Effect of heavy metals and storage time on two types of forest
21            litter: basal respiration rate and exchangeable metals. Ecotoxicol. Environ. Saf. 41: 8-18.
22     Niyogi, S.; Wood, C. M. (2003) Effects of chronic waterborne and dietary metal exposures on gill metal-binding:
23            implications for the Biotic Ligand Model (BLM). Hum. Ecol. Risk Assess. 9: 813-846.
24     Niyogi, S.; Wood, C. M. (2004) Biotic ligand model, a flexible tool for developing site-specific water quality
25            guidelines for metals. Environ. Sci. Technol. 38: 6177-6192.
26     Nolan, A. L.; McLaughlin, M. J.; Mason, S. D. (2003) Chemical speciation of Zn, Cd, Cu, and Pb in pore waters of
27            agricultural and contaminated soils using Donnan dialysis. Environ. Sci. Technol.  37: 90-98.
28     Norwood, W. P.; Borgmann, U.; Dixon, D. G.; Wallace, A. (2003) Effects of metal mixtures on aquatic biota: a
29            review of observations and methods. Hum. Ecol. Risk Assess. 9: 795-811.
30     Nouri, P. A.; Reddy, G. B. (1995) Influence of acid-rain and ozone on soil heavy  metals under loblolly-pine trees:
31            a field-study. Plant Soil 171: 59-62.
32     Nozaki, Y.; DeMaster, D. J.; Lewis, D. M.; Turekian, K. K. (1978) Atmospheric 210Pb fluxes determined from soil
33            profiles. J. Geophys. Res. (Oceans Atmos.) 83: 4047-4051.
34     Nriagu, J. O. (1973) Lead orthophosphates—II. Stability of chloropyromorphite at 25 C. Geochim. Cosmochim.
35            Acta 38: 367-377.
36     Nriagu, J. O. (1974) Lead orthophosphates—IV. Formation and stability in environment. Geochim. Cosmochim.
37            Acta 38: 887-898.
38     Nursita, A. I.; Singh, B.; Lees, E. (2005) The effects of cadmium, copper, lead, and zinc on the growth and
39            reproduction ofPmisotoma minuta Tullberg (Collembola). Ecotoxicol. Environ. Saf. 60: 306-314.
40     O'Shea, T. A.; Mancy, K. H. (1978) The effect of pH and hardness metal ions on the competitive interaction
41            between trace metal ions and inorganic and organic complexing agents found in natural  waters. Water Res.
42            12:703-711.
43     Odum, E. P. (1971) Fundamentals of ecology. 3rd ed. Philadelphia, PA: W. B. Saunders Company; pp. 1-38,
44            106-136.
45     Olson, K. W.; Skogerboe, R. K. (1975) Identification of soil lead compounds from automotive sources. Environ. Sci.
46            Technol.  9: 227-230.
47     Ostergren, J. D.; Trainor, T. P.; Bargar, J. R.; Brown, G. E.; Parks, G. A. (2000a) Inorganic ligand effects on Pb(II)
48            sorption to goethite (a- FeOOH) -1. Carbonate. J. Colloid Interface Sci. 225: 466-482.
49     Ostergren, J. D.; Trainor, T. P.; Bargar, J. R.; Brown, G. E.; Parks, G. A. (2000b) Inorganic ligand effects on Pb(II)
50            sorption to goethite (a-FeOOH) - II.  Sulfate.  J. Colloid.  Interface Sci. 225: 483-493.
51     Ownby, D. R.; Galvan, K. A.; Lydy, M. J. (2005) Lead and zinc bioavailability to Eisniafetida after phosphorus
52            amendment to repository soils. Environ. Pollut. 136: 315-321.
53     Paces, T. (1998) Critical loads of trace metals in soils: a method of calculation. Water Air Soil Pollut. 105: 451-458.
54     Pagenkopf, G. K. (1983) Gill surface interaction model for trace-metal toxicity to fishes: role of complexation, pH,
55            and water hardness. Environ. Sci. Technol. 17: 342-347.
       December 2005                                8-258         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Pagenkopf, G. K. (1986) Metal ion speciation and toxicity in aquatic systems. In: Sigel, H., ed. Concepts on metal
  2            ion toxicity. New York, NY: Marcel Dekker; pp. 101-118. (Metal ions in biological systems: v. 20).
  3     Pahlsson, A.-M. B. (1989) Toxicity of heavy metals (Zn, Cu, Cd, Pb) to vascular plants. Water Air Soil Pollut.
  4            47:287-319.
  5     Palmborg, C.; Bringmark, L.; Bringmark, E.; Nordgren, A. (1998) Multivariate analysis of microbial activity and
  6            soil organic matter at a forest site subjected to low-level heavy metal contamination. Ambio 27: 53-57.
  7     Papp, C. S. E.; Filipek, L .H.; Smith, K. S. (1991) Selectivity and effectiveness of extractants used to release metals
  8            associated with organic-matter. Appl. Geochem. 6: 349-353.
  9     Paquin, P. R.; Di Toro, D. M.; Santore, R. C.; Trivedi, D.; Wu, B.  (1999) A biotic ligand model of the acute toxicity
10            of metals. III. Application to fish and Daphnia exposure to silver. In: Review of the biotic ligand model of
11            the acute toxicity of metals. Washington, DC: U.S. Environmental Protection Agency, Science Advisory
12            Board. EPA 822-E-99-001.
13     Paquin, P. R.; Gorsuch, J. W.; Apte, S.; Batley, G. E.; Bowles, K.  C.; Campbell, P.  G. C.; Delos, C. G.; Di Toro,
14            D. M.; Dwyer, R. L.; Galvez, F.; Gensemer, R. W.; Goss,  G. G.; Hogstrand, C.; Janssen, C. R.; McGeer,
15            J. C.; Naddy, R. B.; Playle, R. C.; Santore,  R. C.; Schneider, U.; Stubblefield, W. A.; Wood,  C. M.; Wu,
16            K. B. (2002) The biotic ligand model: a historical overview. Comp. Biochem. Physiol. C 133: 3-35.
17     Patra, M.; Bhowmik, N.; Bandopadhyay, B.; Sharma, A. (2004) Comparison of mercury, lead and arsenic with
18            respect to genotoxic effects on plant systems and the development of genetic tolerance. Environ. Exper. Bot.
19            52:  199-223.
20     Pattee, O. H.; Pain, D. J. (2003) Lead in the environment. In: Hoffman, D. J.; Rattner, B. A.; Burton,  G. A., Jr.;
21            Carins, K., Jr. Handbook of ecotoxicology. Boca Raton, Fl. Lewis Publishers, pp. 373-408.
22     Peramaki, P.; Itamies, J.; karttunen, V.; Lajunen, L. H. J.; Pulliainen, E. (1992) Influence of pH on the accumulation
23            of cadmium and lead in earthworms (Aporrectodea caliginosa) under controlled conditions. Ann. Zool.
24            Fenn. 29: 105-111.
25     Perottoni, J.; Meotti, F.; Folmer, V.; Pivetta, L.; Nogueira, C. W.; Zeni,  G.; Rocha,  J. B. (2005) Ebselen and
26            diphenyl diselenide do not change the inhibitory effect of lead acetate on delta-aminolevulinate dehidratase.
27            Environ. Toxicol. Pharmacol. 19: 239-248.
28     Petrosyan, V.; Orlova, A.; Dunlap, C. E.; Babayan, E.; Farfel, M.; von Braun, M. (2004) Lead in residential soil and
29            dust in a mining and smelting district in Northern Armenia: a pilot study. Environ. Res. 94: 297-308.
30     Phillips, D. L.; Gregg, J. W. (2003) Source partitioning using stable isotopes: coping with too many sources.
31            Oecologia 136: 261-269.
32     Phillips, C.; Gyori, Z.; Kovacs, B. (2003) The effect of adding cadmium and lead alone or in combination to the diet
33            of pigs on their growth, carcase composition and reproduction. J. Sci. Food Agric. 83: 1357-1365.
34     Pizl, V.; Josens,  G. (1995) Earthworm communities along a gradient of urbanization. Environ. Pollut. 90:  7-14.
35     Planchon, F. A. M.; VanDe Velde, K.; Rosman, K. J. R.; Wolff, E. W.; Ferrari, C.  P.; Boutron, C. F. (2002)
36            One hundred fifty-year record of lead isotopes in Antarctic snow from Coats Land. Geochim. Cosmochim.
37            Acta 67: 693-708.
3 8     Playle, R. C. (2004) Using multiple metal-gill binding models and the toxic unit concept to help reconcile multiple-
39            metal toxicity results. Aquat. Toxicol. 67: 359-370.
40     Polissar, A.  V.; Hopke, P. K.; Poirot, R. L. (2001) Atmospheric aerosol over Vermont: chemical composition and
41            sources. Environ. Sci. Technol. 35: 4604-4621.
42     Poulton, B. C.; Monda, D.  P.; Woodward, D. F.; Wildhaber, M. L.; Brumbaugh, W. G. (1995) Relations between
43            benthic community structure and metals concentrations in aquatic macroinvertebrates: Clark  Fork River,
44            Montana. J. Freshwater Ecol. 10: 277-293.
45     Prasuna, G.; Zeba, M.; Khan, M. A. (1996) Excretion of lead as a mechanism for survival on Chrissia halyi
46            (Ferguson, 1969). Bull. Environ. Contam. Toxicol. 57: 849-852.
47     Probst, A.; Moncoulon, D.; Godderis, Y.; Hernandez, L.; Party, J.-P. (2003) Critical loads for lead in France:
48            first results on forest soils. J. Phys. IV. 107: 1111-1114.
49     Prosi, F. (1989) Factors controlling biological availability and toxic effects of lead  in aquatic organisms. Sci. Total
50            Environ. 79:  157-169.
51     Rabinowitz, M. B. (1995) Stable isotopes of lead for source identification. (Selected proceedings of the 5th world
52            congress for the World Federation of Associations of Clinical Toxicology Centers and Poison Control
53            Centers, November 8-11, 1994, Taipei, Taiwan, R.O.C.). J. Toxicol. Clin. Toxicol. 33: 649-655.
54     Rabinowitz, M. B. (2005) Lead isotopes in soils near five historic  American lead smelters and refineries. Sci. Total
55            Environ. 346: 138-148.
       December 2005                                 8-259         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Rabinowitz, M. B.; Wetherill, G. W. (1972) Identifying sources of lead contamination by stable isotope techniques.
  2            Environ. Sci. Technol. 6: 705-709.
  3     Rabitsch, W. B. (1995a) Metal accumulation in arthropods near a lead/zinc smelter in Arnoldstein, Austria. Environ.
  4            Pollut. 90: 221-237.
  5     Rabitsch, W. B. (1995b) Metal accumulation in arthropods near a lead/zinc smelter in Arnoldstein, Austria. II.
  6            Formicidae. Environ. Pollut. 90: 249-257.
  7     Rademacher, D. I; Weber, D. N.; Hillard, C. J. (2005) Waterborne lead exposure affects brain endocannabinoid
  8            content in male but not female fathead minnows (Pimephalespromelas). Neurotoxicology 26: 9-15.
  9     Radojevic, M.; Harrison, R. M. (1987) Concentrations and pathways of organolead compounds in the environment:
10            a review. Sci. Total Environ. 59: 157-180.
11     Rahmani, G. N. H.; Sternberg, S. P. K. (1999) Bioremoval of lead from water using Lemna minor. Bioresour.
12            Technol. 70: 225-230.
13     Rai, L. C.; Raizada, M. (1989) Effect of bimetallic combinations of Ni, Cr and Pb on growth, uptake of nitrate and
14            ammonia, 14CO2 fixation, and nitrogenase activity of Nostoc muscorum. Ecotoxicol. Environ. Saf.
15            17:75-85.
16     Rainbow, P. S. (1996) Heavy metals in aquatic invertebrates. In: Beyer, W. N.; Heinz, G. H.; Redmon-Norwood,
17            A. W., eds. Environmental contaminants in wildlife: interpreting tissue concentrations. Boca Raton, FL:
18            CRC Press; pp. 405-425.
19     Rand, G. M.; Wells, P. G.; McCarty, L. S. (1995) Introduction to aquatic toxicology. In: Rand, G. M., ed.
20            Fundamentals of aquatic toxicity: effects, environmental fate, and risk assessment. 2nd ed. Washington, DC:
21            Taylor & Francis; pp. 3-67.
22     Rao, J.; Reddy, T. (1985)  Response of Scenedesmus incrassatulus to lead toxicity in presence of nutrients. J. Biol.
23            Res. 1:51-56.
24     Rao, J. V.; Kavitha, P.; Rao, A. P. (2003) Comparative toxicity of terra ethyl lead and lead oxide to earthworms,
25            Eisenia fetida (Savigny). Environ. Res. 92: 271-276.
26     Reaves, G. A.; Berrow, M. L. (1984) Total lead concentrations in Scottish soils. Geoderma 32: 1-8.
27     Redig, P. T.; Lawler, E. M.; Schwartz, S.; Dunnette, J. L.; Stephenson, B.; Duke, G. E. (1991) Effects of chronic
28            exposure to sublethal concentrations of lead acetate on heme synthesis and immune function in red-tailed
29            hawks. Arch. Environ. Contam. Toxicol. 21: 72-77'.
30     Reinds, G. J.; Vries, W.D.; Groenenberg, J. E. (2002) Annex 2: updated assessment of critical loads of lead and
31            cadmium for European forest soils. In: Hettelingh, J. P.;  Slootweg, J.; Posch, M.; Dutchak, S.; Ilyin, L, eds.
32            Preliminary modelling and mapping of critical loads for cadmium and lead in Europe; RIVM report no.
33            259101011. Bilthoven, The Netherlands: National Institute of Public Health and the Environment;
34            pp. 123-127.
35     Reisinger, K.; Stoeppler, M.; Nurnberg, H. W. (1981) Evidence forthe  absence of biological methylation of lead in
36            the environment. Nature (London) 291: 228-230.
37     Renberg, L; Brannvall, M.-L.; Bindler, R.; Emteryd, O. (2000) Atmospheric lead pollution history during four
38            millennia (2000 BC to 2000 AD) in Sweden. Ambio 29: 150-156.
39     Renberg, L; Brannvall, M. L.; Bindler, R.; Emteryd, O. (2002) Stable lead isotopes and lake sediments—a useful
40            combination for the study of atmospheric lead pollution history. Sci. Total Environ. 292:45-54.
41     Rhea, D. T.; Harper, D. D.; Brumbaugh, W. G.; Farag, A. M. (2004) Biomonitoring in the Boulder River Watershed,
42            Montana: metal concentrations in biofilm and macroinvertebrates, and relations with macroinvertebrate
43            assemblage. Reston, VA: U.S. Geological Survey; report no. USGS-CERC-91340.
44     Rhue, R. D.; Mansell, R. S.; Ou, L.-T.; Cox, R.; Tang, S. R.; Ouyang, Y. (1992) The fate  and behavior of lead alkyls
45            in the environment: a review. Crit. Rev. Environ. Control 22: 169-193.
46     Rice, T. M.; Blackstone, B. J.; Nixdorf, W. L.; Taylor, D. H. (1999) Exposure to lead induces hypoxia-like
47            responses in bullfrog larvae (Rana catesbeiana). Environ. Toxicol. Chem. 18: 2283-2288.
48     Richards, B. K.; Steenhuis, T. S.; Peverly, J. H.; McBride, M. B. (1998) Metal mobility at an old, heavily loaded
49            sludge application site. Environ. Pollut. 99: 365-377.
50     Richards, B. K.; Steenhuis, T. S.; Peverly, J. H.; McBride, M. B. (2000) Effect of sludge-processing mode, soil
51            texture, and pH on metal mobility in undisturbed soil columns under accelerated loading. Environ. Pollut.
52            109:327-346.
53     Rickard, D. T.; Nriagu, J.  O. (1978) Aqueous environmental chemistry  of lead.  In: Nriagu, J. O., ed. The
54            biogeochemistry of lead in the environment. Part A. Ecological cycles.  Amsterdam, The Netherlands:
55            Elsevier/North-Holland Biomedical Press; pp. 219-284. (Topics in environmental health: v. 1A).
       December 2005                                8-260        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Ritson, P. I.; Bouse, R. M; Flegal, A. R.; Luoma, S. N. (1999) Stable lead isotopic analyses of historic and
  2            contemporary lead contamination of San Francisco Bay estuary. Marine Chem. 64: 71-83.
  3     Rogers, J. T.; Wood, J. G. (2003) lonoregulatory disruption as the acute toxic mechanism for lead in the rainbow
  4            trout (Oncorhynchus mykiss). Aquat. Toxicol. 64: 215-234.
  5     Rogers, J. T.; Wood, C. M. (2004) Characterization of branchial lead-calcium interaction in the freshwater rainbow
  6            trout (Oncorhynchus mykiss). J. Exp. Biol. 207: 813-825.
  7     Rosso, A.; Lafont, M.; Exinger, A. (1994) Impact of heavy metals on benthic oligochaete communities in the river
  8            111 and its tributaries. Water Sci.  Technol. 29: 241-248.
  9     Roth, M. (1993) Investigations on lead in the soil invertebrates of a forest ecosystem. Pedobiologia 37: 270-279.
10     Ruby, M. V.; Davis, A.; Kempton, J. H.; Drexler, J. W.; Bergstrom, P. D.  (1992) Leadbioavailability: dissolution
11            kinetics under simulated gastric conditions. Environ. Sci. Technol. 26: 1242-1248
12     Ruby, S. M.;  Jaroslawski, P.; Hull, R. (1993) Lead and cyanide toxicity in sexually maturing rainbow trout,
13            Oncorhynchus mykiss during spermatogenesis. Aquat. Toxicol. 26: 225-238.
14     Ruby, M. V.; Davis, A.; Nicholson, A. (1994) In situ formation of lead phosphates in soils as a method to
15            immobilize lead. Environ. Sci. Technol. 28: 646-654.
16     Ruparelia, S.  G.; Verma, Y.; Mehta, N. S.; Salyed, S. R. (1989) Lead-induced biochemical changes in freshwater
17            fish Oreochrois mossambicus. Bull. Environ. Contam. Toxicol. 43:310-314.
18     Rusek, J.; Marshall, V. G. (2000) Impact of airborne pollutants on soil fauna. Annu. Rev. Ecol. System.
19            31:395-423.
20     Russell, I. J.;  Choquette, C. E.; Fang, S.-L.; Dundulis, W. P.; Pao, A. A.; Pszenny, A. A. P. (1981) Forest vegetation
21            as a sink for atmospheric particulates: quantitative studies in rain and dry deposition. J. Geophys. Res.
22            (Oceans & Atmos.) 86(C6): 5347-5363.
23     Ryan, J. A.; Zhang, P.; Hesterberg, D.; Chou, J.; Sayers, D. E. (2001) Formation of chloropyromorphite in
24            lead-contaminated soil amended with hydroxyapatite. Environ. Sci. Technol. 35: 3798-3803.
25     Ryan, P. C.; Wall, A. J.; Hillier, S.; Clark, L. (2002) Insights into sequential chemical extraction procedures from
26            quantitative XRD: a study of trace metal partitioning in sediments related to frog malformities.  Chem. Geol.
27            184:337-357.
28     Sadiq, M. (1992) Lead in marine environments. In: Toxic metal chemistry in marine environments, v. 1. New York,
29            NY: Marcel Dekker, Inc.; pp. 304-355.  (Environmental science and pollution control series: v. 1).
30     Sample, B. E.; Beauchamp, J. J.; Efroymson, R. A.; Suter, G. W., II; Ashwood, T. L. (1998) Development and
31            validation of bioaccumulation models for earthworms. Oak Ridge, TN: Oak Ridge National Laboratory;
32            ES/ER/TM-220.
33     Sample, B.; Beauchamp, J. J.; Efroymson, R.; Suter, G. W., II. (1999) Literature-derived bioaccumulation models
34            for earthworms: development and validation. Environ. Toxicol. Chem. 18: 2110-2120.
3 5     Sandifer, R. D.; Hopkin, S. P. (1996) Effects on pH on the toxicity of cadmium, copper, lead and zinc to Folsomia
36            Candida Willem, 1902 (Collembola) in  a standard laboratory test system. Chemosphere 33: 2475-2486.
37     Santillan-Medrano, J.; Jurinak, J. J. (1975) The chemistry of lead and cadmium in soil: solid phase formation.
38            Soil Sci. Soc. Am. Proc. 39: 851-856.
3 9     Santos, M. A.; Hall, A. (1990) Influence of inorganic lead on the biochemical blood composition of the eel, Anguilla
40            anguilla L. Ecotoxicol. Environ. Saf. 20: 7-9.
41     Sanudo-Pena, M. C.; Romero, J.; Seale, G. E.; Fernandez-Ruiz, J. J.; Walker, J. M. (2000) Activational role of
42            cannabinoids on movement. Eur. J. Pharmacol. 391: 269-274.
43     Sauter, S.; Buxton, K. S.; Macek, K. J.; Petrocelli, S. R. (1976) Effects of exposure to heavy metals on selected fresh
44            water fish: toxicity of copper, cadmium, chromium, and lead to eggs and fry of seven fish species. Duluth,
45            MN:  U.S. Environmental Protection Agency, Office of Research and Development; report no. EPA-600/3-
46            76-105. Available from: NTIS, Springfield, VA; PB-265612.
47     Sauve, S.; McBride, M.; Hendershot, W. (1997) Speciation of lead in contaminated soils. Environ. Pollut.
48            98: 149-155.
49     Sauve, S.; McBride, M.; Hendershot, W. (1998) Soil solution speciation of lead(II): effects of organic matter and
50            pH. Soil Sci. Soci. Am. J. 62: 618-621.
51     Sauve, S.; Hendershot, W.; Allen, H. E. (2000a) Solid-solution partitioning of metals in contaminated soils:
52            dependence on pH, total metal burden, and organic matter. Environ. Sci. Technol. 34: 1125-1131.
53     Sauve, S.; Martinez, C. E.; McBride, M.; Hendershot, W. (2000b) Adsorption of free lead (Pb2+)  by pedogenic
54            oxides, ferrihydrite, and leaf compost. Soil Sci. Soc. Am. J. 64: 595-599.
55     Sauve, S.; Manna, S.; Turmel, M. C.; Roy, A. G.; Courchesne, F. (2003) Solid—solution partitioning of Cd, Cu, Ni,
56            Pb, and Zn in the organic horizons of a forest soil. Environ. Sci. Technol. 37: 5191-5196.
       December 2005                                8-261         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Saviozzi, A.; Levi-Minzi, R.; Cardelli, R.; Riffaldi, R. (1997) The influence of heavy metals on carbon dioxide
  2            evolution from a typic xerochrept soil. Water Air Soil Pollut. 93: 409-417.
  3     Sayer, M. D. I; Reader, J. P.; Morris, R. (1989) The effect of calcium concentration on the toxicity of copper, lead
  4            and zinc to yolk-sac fry of brown trout, Salmo trutta L., in soft, acid water. J. Fish Biol. 35: 323-332.
  5     Scally, S.; Davison, W.; Zhang, H. (2003) In situ measurements of dissociation kinetics and labilities of metal
  6            complexes in solution using DGT. Environ. Sci. Technol. 37: 1379-1384.
  7     Scheckel, K.G.; Impellitteri, C.A.; Ryan,  J.A.; McEvoy, T. (2003) Assessment of sequential extraction procedure for
  8            perturbed lead-contaminated samples with and without phosphorus amendments. Environ. Sci. Technol.
  9            37:  1892-1898.
10     Scherer, E.; McNicol, R. E.  (1998) Preference-avoidance responses of lake whitefish (Coregonus dupeaformis) to
11            competing gradients of light and copper, lead, and zinc. Water Res. 32: 924-929.
12     Scheuhammer, A. M. (1987) The chronic toxicity of aluminum, cadmium, mercury and lead in birds: a review.
13            Environ. Pollut. 46: 263-295.
14     Scheuhammer, A. M. (1989) Monitoring wild bird populations for lead exposure. J. Wildl. Manage. 53: 759-765.
15     Scheuhammer, A. M. (1991) Effects of acidification on the availability of toxic metals  and calcium to wild birds and
16            mammals. Environ. Pollut. 71: 329-375.
17     Schlesinger, W. H. (1997) Biogeochemistry: an analysis of global change. 2nd ed. San Diego, CA: Academic Press.
18     Schlick, E.; Mengel, K.; Friedberg, K. D. (1983) The effect of low lead doses in vitro and in vivo on the d-ala-d
19            activity of erythrocytes, bone marrow cells, liver and brain of the mouse. Arch. Toxicol. 53: 193-205.
20     Schmitt, C. J.; Brumbaugh, W. G. (1990) National contaminant biomonitoring program: concentrations of arsenic,
21            cadmium, copper, lead, mercury, selenium, and zinc in U.S. freshwater fish, 1976-1984. Arch. Environ.
22            Contam. Toxicol. 19: 731-747.
23     Schubauer-Berigan, M. K.; Dierkes, J. R.; Monson, P. D.; Ankley, G. T. (1993) pH-dependent toxicity of Cd, Cu,
24            Ni, Pb and Zn to Ceriodaphnia dubia, Pimephales promelas, Hyalella azteca and Lumbriculus variegatus.
25            Environ. Toxicol. Chem. 12: 1261-1266.
26     Schwartz, M. L.; Curtis, P. J.; Playle, R. C. (2004) Influence of natural organic matter source on acute copper, lead,
27            and cadmium toxicity to rainbow trout (Oncorhynchus mykiss). Environ. Toxicol. Chem. 12: 2889-2899.
28     Seiler, J. R.; Paganelli, D. J. (1987) Photosynthesis and growth response of red spruce and loblolly pine to soil-
29            applied lead and simulated acid rain. For. Sci. 33: 668-675.
30     Senwo, Z. N.; Tabatabai, M. A. (1999) Aspartase activity in soils: effects of trace elements and relationships to other
31            amidohydrolases. Soil Biol. Biochem. 31: 213-219.
32     Shaffer, R. E.; Cross, J. O.; Rose-Pehrsson, S. L.; Elam,  W. T. (2001) Speciation of chromium in simulated soil
33            samples using x-ray absorption spectroscopy and multivariate calibration. Anal. Chim. Acta 442: 295-304.
34     Sharma, N.; Gardea-Torresday, J. L.; Parson, J.; Sahi, S. V. (2004) Chemical speciation and cellular deposition of
35            lead inSesbania drummondii. Environ. Toxicol. Chem.  23: 2068-2073.
36     Shirahata, H.; Elias, R. W.; Patterson, C. C.; Koide, M. (1980) Chronological variations in concentrations and
3 7            isotopic compositions of anthropogenic atmospheric lead in sediments of a remote subalpine pond.
38            Geochim. Cosmochim. Acta 44: 149-162.
39     Shugart, L. R. (1995) Environmental genotoxicology. In: Rand,  G. M., ed. Fundamentals of aquatic toxicology:
40            effects, environmental fate and risk assessment.  2nd ed. Washington, DC: Taylor and Francis; pp. 405-419.
41     Shuman, L. M. (1982) Separating soil iron-oxide and manganese-oxide fractions for micro-element analysis. Soil
42            Sci. Soc. Am. J. 46: 1099-1102.
43     Siccama, T. G. (1974) Vegetation, soil, and climate on Green Mountains of Vermont. Ecol. Monogr. 44: 325-349.
44     Sieghardt, H. (1990) Heavy-metal uptake and distribution in Silene vulgaris w&Minuartia verna growing on
45            mining-dump material containing lead and zinc. Plant Soil 123: 107-111.
46     Simoes Goncalves, M. L. S.; Vilhena, M. F. C.; Fernandes Sollis, J. M.; Castro Romero, J. M.; Sampayo, M. A.
47            (1991) Uptake of lead and its influence in the alga Selenastrum capricornutum Printz. Talanta
48            38:  1111-1118.
49     Skeffington, R. A. (1999) The use of critical loads in environmental policy making: a critical appraisal. Environ. Sci.
50            Technol. 33: 245A-252A.
51     Skjelkvale, B. L.; Andersen, T.; Field, E.; Mannio, J.; Wilander, A.; Johansson, K.; Jensen, J. P.; Moiseenko, T.
52            (2001) Heavy metal surveys in Nordic lakes; concentrations, geographic patterns and relation to critical
53            limits. Ambio 30: 2-10.
54     Slaveykova, V. L; Wilkinson, K. J. (2002) Physicochemical aspects of lead bioaccumulation by Chlorella vulgaris.
55            Environ. Sci. Technol.  36: 969-975.
       December 2005                                8-262         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Slaveykova, V. I.; Wilkinson, K. J. (2005) Predicting the bioavailability of metals and metal complexes: critical
  2            review of the biotic ligand model. Environ. Chem. 2: 9-24.
  3     Small, W. (1973) Isotopic compositions of selected ore leads from northwestern Washington. Can. J. Earth Sci.
  4            10:670-674.
  5     Snoeijs, T.; Dauwe, T.; Pinxten, R.; Darras, V. M.; Arckens, L.; Eens, M. (2005) The combined effect of lead
  6            exposure and high or low dietary calcium on health and immunocompetence in the zebra finch (Taeniopygia
  1            guttatd). Environ. Pollut. 134: 123-132.
  8     Sopper, W. E. (1989) Revegetation of a contaminated zinc smelter site. Landscape Urban Plann. 17: 241-250.
  9     Sopper, W. E. (1993) Municipal sludge use for land reclamation. Chelsea, MI: Lewis Publishers.
10     Spehar, R. L.; Anderson, R. L.; Fiandt, J. T. (1978) Toxicity and bioaccumulation of cadmium and lead in aquatic
11            invertebrates. Environ. Pollut. 15: 195-208.
12     Spehar, R. L.; Fiandt, J. T. (1986) Acute and chronic effects of water quality criteria-based metal mixtures on
13            aquatic species. Environ. Toxicol. Chem. 5: 917-931.
14     Sposito, G.; Coves, J. (1988) SOILCHEM: A computer program for the calculation of chemical speciation in soils.
15            Berkeley, CA: University of California, Kerney Foundation of Soil Science.
16     Sposito, G.; Lund, L. J.; Chang, A. C. (1982) Trace metal chemistry in arid-zone field soils amended with sewage
17            sludge. 1. Fractionation of Ni, Cu, Zn, Cd, and Pb in solid-phases. Soil Sci. Soc. Am. J. 46: 260-264.
18     Spry, D. J.; Wiener, J. G. (1991) Metal bioavailability and toxicity to fish in low-alkalinity lakes: a critical review.
19            Environ. Pollut. 71:243-304.
20     Spurgeon, D. J.; Hopkin, S. P. (1996a) The effects of metal contamination on earthworm populations around a
21            smelting works: quantifying species effects. Appl. Soil Ecol. 4: 147-160.
22     Spurgeon, D. J.; Hopkin, S. P. (1996b) Risk assessment of the threat of secondary poisoning by metals to predators
23            of earthworms in the vicinity of a primary smelting works. Sci. Total Environ. 187: 167-183.
24     Squire, S.; Scelfo, G. M.; Revenaugh, J.; Flegal, A. R. (2002) Decadal trends of silver and lead contamination in
25            San Francisco Bay surface waters. Environ. Sci. Technol. 36: 2379-2386.
26     Stacey, J.  S.; Zartman, R. E.; Komo, T. N. (1968) A lead isotope study of galena and selected feldspars from mining
27            districts in Utah. Econ. Geol. Bull. Soc. Econ. Geol. 63: 796-814.
28     Steele,  C. W.; Strickler-Shaw, S.; Taylor, D. H. (1989) Behavior of tadpoles of the bullfrog, Rana catesbeiana,
29            in response to sublethal lead exposure. Aquat. Toxicol. 14: 331-344.
30     Steele,  C. W.; Strickler-Shaw, S.; Taylor, D. H. (1991) Failure ofBufo americanus tadpoles to avoid lead-enriched
31            water. J. Herpetol. 25: 241-243.
32     Steinnes, E.; Friedland, A. J. (2005) Lead migration in podzolic soils from Scandanavia and the United States of
33            America. Can. J. Soil Sci. 85: 291-294.
34     Stephan, C. E.; Mount, D. I.; Hansen, D. J.; Gentile, J. H.; Chapman, G. A. (1985) Guidelines for deriving numerical
3 5            national water quality criteria for the protection of aquatic organisms and their uses. Washington, D.C.: U. S.
36            Environmental Protection Agency; report no. EPA/822-R85-100. Available from: NTIS, Springfield, VA;
37            PB85-227049.
38     Storm,  G. L.; Fosmire, G. J.;Bellis, E. D. (1994) Persistence of metals in soil and selected vertebrates in the vicinity
39            of the Palmerton zinc smelters. J. Environ. Qual. 23: 508-514.
40     Stouthart, A. J. H. X.; Spanings, F. A. T.; Lock, R. A. C.; Wendelaar Bonga, S. E. (1994) Effects of low water pH
41            on lead toxicity to early life stages of the common carp (Cyprinus carpio). Aquat. Toxicol. 30: 137-151.
42     Strawn, D. G.; Sparks, D. L. (1999) The use of XAFS to distinguish between inner- and outer-sphere lead adsorption
43            complexes on montmorillonite. J. Colloid Interface Sci. 216: 257-269.
44     Strickler-Shaw, S.; Taylor, D. H. (1990) Sublethal exposure to lead inhibits acquisition and retention of discriminate
45            avoidance learning in green frog (Rana clamitans) tadpoles. Environ. Toxicol. Chem. 9: 47-52.
46     Stripp,  R. A.; Heit, M.; Bogen, D. C.; Bidanset, J.; Trombetta, L. (1990) Trace element accumulation in the tissues
47            of fish from lakes with different pH values. Water Air Soil Pollut. 51: 75-87.
48     Stumm, W.; Morgan, J. J. (1970) Aquatic chemistry: an introduction emphasizing chemical equilibria in natural
49            waters. New York, NY: Wiley-Interscience.
50     Stumm, W.; Morgan, J. J. (1995) Aquatic chemistry: chemical equilibria and rates in natural waters. 3rd ed.
51            New York, NY: Wiley Interscience. [Schnoor, J. L.; Zehnder, A., eds. Environmental Science and
52            Technology series].
53     Sturges, W.  T.; Barrie, L. A. (1987) Lead 206/207 isotope ratios in the atmosphere of North America as tracers of
54            US and Canadian emissions. Nature (London) 329: 144-146.
55     Sturges, W.  T.; Hopper, J. F.; Barrie, L. A.; Schnell, R. C. (1993)  Stable lead isotope ratios in Alaskan Arctic
56            aerosols. Atmos. Environ. 27A: 2865-2871.
       December 2005                                 8-263         DRAFT-DO NOT QUOTE OR CITE

-------
  1     Svecevicius, G. (1991) The role of olfaction in avoidance reactions to pollutants by vimba Vimba vimba (L.).
  2            Ekologija4:3-8.
  3     Svecevicius, G. (2001) Avoidance response of rainbow trout Oncorhynchus mykiss to heavy metal model mixtures:
  4            a comparison with acute toxicity tests. Bull. Environ. Contam. Toxicol. 67: 680-687.
  5     Swansburg, E. O.; Fairchild, W. L.; Fryer, B. J.; Ciborowski, J. J. H. (2002) Mouthpart deformities and community
  6            composition of chironomidae (Diptera) larvae sownstream of metal mines in New Brunswick, Canada.
  7            Environ. Toxicol. Chem. 21: 2675-2684.
  8     Swanson, K. A.; Johnson, A. H. (1980) Trace metal budgets for a forested watershed in the New Jersey Pine
  9            Barrens. Water Resour. Res. 16: 373-376.
10     Syracuse Research Corporation (SRC). (1999) The environmental fate of lead and lead compounds. Washington,
11            DC: U.S. Environmental Protection Agency; contract no. SRC 68-D5-0012.
12     Szarek-Lukaszewska, G.;  Slysz, A.; Wierzbicka, M. (2004) Response ofArmeria maritima (Mill.) Willd. to Cd, Zn,
13            and Pb. Acta Biol. Cracoviensia Ser. Bot. 46: 19-24.
14     Szulczewski, M. D.; Helmke, P. A.; Bleam, W. F. (1997) Comparison of XANES analyses and extractions to
15            determine chromium speciation in contaminated soils. Environ. Sci. Technol. 31:  2954-2959.
16     Tada,F.; Suzuki, S. (1982) Adsorption and desorption of heavy metals in bottom mud of urbanrivers. Water Res.
17            16: 1489-1494.
18     Tao, S. Li,  H.; Liu, C.; Lam, K. C. (2000) Fish uptake of inorganic and mucus complexes  of lead. Ecotoxicol.
19            Environ. Saf. 46:  174-180.
20     Tejedor, M. C.; Gonzalez, M. (1992) Comparison between lead levels in blood and bone tissue of rock doves
21            (Columba livia) treated with lead acetate or exposed to the environment of Alcala-de-Henares. Bull.
22            Environ. Contam. Toxicol. 48: 835-842.
23     Templeton, A. S.; Spormann, A. M.; Brown,  G. E. (2003a) Speciation of Pb(II) sorbed by Burkholderia
24            cepac/a/goethite composites. Environ. Sci. Technol. 37: 2166-2172.
25     Templeton, A. S.; Trainor, T. P.; Spormann, A. M.; Newville, M.; Sutton, S. R.; Dohnalkova, A.; Gorby, Y.;
26            Brown, G. E. (2003b) Sorption versus biomineralization of Pb(II) within Burkholderia cepacia biofilms.
27            Environ. Sci. Technol. 37: 300-307.
28     Terhivuo, J.; Pankakoski, E.; Hyvarinen, H.; Koivisto, I.  (1994) Pb uptake by ecologically dissimilar earthworm
29            (Lumbricidae) species near a lead smelter in south Finland. Environ. Pollut. 85: 87-96.
30     Tessier, A.; Campbell, P. G. C.; Bisson, M. (1979) Sequential extraction procedure for the speciation of paniculate
31            trace-metals. Anal. Chem. 51:  844-851.
32     Tessier, A.; Campbell, P. G. C. (1987) Partitioning of trace-metals in sediments - relationships withbioavailability.
33            Hydrobiologia 149: 43-52.
34     Tessier, A.; Campbell, P. G. C. (1988) Partitioning of trace metals in sediments. In: Kramer, J. R.; Allen, H. E., eds.
35            Metal speciation:  theory, analysis and application. Chelsea, MI: Lewis Publishers, pp. 183-199.
36     Timmermans, K. R.; Peeters, W.; Tonkes, M. (1992) Cadmium, zinc, lead and copper in Chironomus riparius
37            (Meigen) larvae (Diptera, Chironomidae): uptake and effects. Hydrobiologia 241: 119-134.
3 8     Tipping, E. (1994) WHAMC—A chemical equilibrium model and computer code for waters, sediments, and soils
39            incorporating a discrete site/electrostatic model of ion-binding by humic substances. Comput. Geosci.
40            20:973-1023.
41     Tipping, E.; Woof, C. (1990) Humic substances in acid organic soils: modelling their release to the soil solution in
42            terms of humic charge. J. Soil Sci. 41: 573-586.
43     Tipping, E.; Rieuwerts, J.; Pan, G.; Ashmore, M. R.; Lofts, S.; Hill, M. T. R.; Farago, M. E.; Thornton, I. (2003)
44            The solid-solution partitioning of heavy metals (Cu, Zn, Cd, Pb) in upland soils of England and Wales.
45            Environ. Pollut. 125: 213-225.
46     Townsend, A.  T.; Yu, Z.; McGoldrick, P.; Hutton, J. A. (1998) Precise lead isotope ratios in Australian galena
47            samples by high resolution inductively coupled plasma mass spectrometry. J. Anal. At. Spectrom.
48            13:809-813.
49     Trivedi, P.; Dyer, J. A.; Sparks, D. L. (2003)  Lead sorption onto ferrihydrite.  1.  A macroscopic and spectroscopic
50            assessment. Environ. Sci. Technol. 37: 908-914.
51     Turner, R.  S.; Johnson, A. H.; Wang, D. (1985) Biogeochemistry of lead in McDonalds Branch Watershed,
52            New Jersey Pine Barrens. J. Environ. Qual. 14: 305-314.
53     Tyler, G. (1981) Leaching of metals from the A -horizon of a spruce forest soil. Water Air  Soil Pollut. 15: 353-369.
54     Tyler, G.; Balsberg Pahlsson, A.-M.; Bengtsson,  G.; Baath, E.;  Tranvik, L. (1989) Heavy-metal ecology of
55            terrestrial plants, microorganisms and invertebrates. Water Air Soil Pollut. 47: 189-215.
       December 2005                                8-264        DRAFT-DO NOT QUOTE OR CITE

-------
  1    U.S. Environmental Protection Agency. (1979) Water-related environmental fate of 129 priority pollutants. Volume
  2            I: Introduction and technical background, metals and inorganics, pesticides and PCBs. Washington, DC:
  3            Office of Water Planning and Standards; report no. EPA-440/4-79-029a. Available from: NTIS, Springfield,
  4            VA;PB80-204373.
  5    U.S. Environmental Protection Agency. (1985) Ambient water quality criteria for lead -1984. Washington, DC:
  6            Office of Water Regulations and Standards, Criteria and Standards Division; report no. EPA/440/5-84/027.
  7            Available: http://www.epa.gov/npdes/pubs/owm586.pdf [26 October, 2005].
  8    U.S. Environmental Protection Agency. (1986a) Air quality criteria for lead. Research Triangle Park, NC: Office of
  9            Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
10            EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.
11    U.S. Environmental Protection Agency. (1986b) Quality criteria for water 1986. Washington, DC: Office of Water
12            Regulations and Standards; EPA report no. EPA/440/5-86/001. Available from: NTIS, Springfield, VA;
13            PB87-226759.
14    U.S. Environmental Protection Agency. (1986c) Test methods for evaluating solid waste. Volumes 1A through 1C:
15            laboratory manual physical/chemical methods. Volume 11: field manual physical/chemical methods.
16            Washington, DC:  Office of Solid Waste and Emergency Response; report no. SW 846. Available from:
17            NTIS, Springfield, VA; PB88-239223.
18    U.S. Environmental Protection Agency. (1990) Review of the national ambient air quality standards for lead:
19            assessment of scientific and technical information:  OAQPS staff paper. Research Triangle Park, NC: Office
20            of Air Quality Planning and Standards; report no. EPA-450/2-89/022. Available from: NTIS, Springfield,
21            VA;PB91-206185.
22    U.S. Environmental Protection Agency. (1991) Methods for the determination of metals in environmental samples.
23            Washington, DC:  U.S. Environmental Protection Agency; EPA/600/4-91-010.
24    U.S. Environmental Protection Agency. (1999) Understanding variation in partition coefficient, Kd, values. Volume
25            II: review of geochemistry ad available Kd values for cadmium, cesium, chromium, lead, plutonium, radon,
26            strontium, thorium, tritium (3H), and uranium. Washington, DC: Office of Air and Radiation; report no.
27            EPA 402-R-99-004B. Available: http://www.epa.gov/radiation/docs/kdreport/vol2/402-r-99-004b.pdf
28            [26 October, 2005].
29    U.S. Environmental Protection Agency. (2001) Test methods for evaluating solid waste, physical/chemical methods
30            (SW-846 integrated manual with updates through final update III). Washington, DC: Office of Solid Waste
31            and Emergency Response; report no. SW 846. Available from: NTIS, Springfield, VA; PB97-156111.
32            Available: http://www.epa.gov/epaoswer/hazwaste/test/sw846.htm [1 November, 2005].
33    U.S. Environmental Protection Agency. (2002) Site technology capsule: demonstration of Rocky Mountain
34            remediation services soil amendment process. Cincinnati, OH: National Risk Management Research
35            Laboratory; EPA/540/R-02/501A. Available:
36            http://www.epa.gov/ORD/NRMRL/pubs/540r02501/540R02501 A.pdf [28 September, 2005].
37    U.S. Environmental Protection Agency. (2004a) Air quality criteria for paniculate matter. Research Triangle Park,
38            NC: National Center for Environmental Assessment; report no. EPA/600/P-99/002aF-bF. 2v. Available:
39            http://cfpub.epa.gov/ncea/ [9 November, 2004].
40    U.S. Environmental Protection Agency. (2004b) Estimation of relative bioavailability of lead in soil and soil-like
41            materials using in vivo and in vitro methods. Washington, DC: Office of Solid Waste and Emergency
42            Response; report no. OSWER 9285.7-77.
43    U.S. Environmental Protection Agency. (2004c) Framework for inorganic metals risk assessment [external review
44            draft]. Washington, DC: Risk Assessment Forum; report no. EPA/630/P-04/068B. Available:
45            http://cfpub2.epa.gov/ncea/raf/recordisplay.cfm?deid=88903 [26 October, 2005].
46    U.S. Environmental Protection Agency. (2005a) Ecological soil screening levels for lead. Interim final. Washington,
47            DC: Office of Solid Waste and Emergency Response, OSWER directive 9285.7-70.
48    U.S. Environmental Protection Agency. (2005b) Guidance for developing ecological soil screening levels (Eco-
49            SSLs). Washington, DC: Office of Solid Waste and Emergency Response, OSWER directive 9285.7-55,
5 0            November 2003 -revised February 2005. Available:
51            http://www.epa.gov/superfund/programs/risk/ecorisk/ecossl.pdf [29 September, 2005].
52    U.S. Environmental Protection Agency. (2005c) Procedures for the derivation of equilibrium partitioning sediment
53            benchmarks (ESBs) for the protection of benthic organisms: metal mixtures (cadmium, copper, lead, nickel,
54            silver and zinc). Washington, DC: Office of Research and Development; EPA-600-R-02-011.
      December 2005                                8-265         DRAFT-DO NOT QUOTE OR CITE

-------
  1     United Nations Economic Commission for Europe (UNECE). (1994) Protocol to the convention on long-range
  2            transboundary air pollution on further reduction of sulphur emissions (1994 Sulphur Protocol). Geneva,
  3            Switzerland: United Nations Economic Commission for Europe (UNECE). Available:
  4            http://www.unece.Org/env/lrtap/full%20text/1994.Sulphur.e.pdf [31 October, 2005].
  5     United Nations Economic Commission for Europe (UN-ECE). ( 2004) Convention on long-range transboundary air
  6            pollution. Available: http://www.unece.org/env/lrtap/lrtap_hl.htm [19 October, 2005].
  7     Unruh, D. M; Fey, D. L.; Church, S. E. (2000) Chemical data and lead isotopic compositions of geochemical
  8            baseline samples from streambed sediments and smelter slag, lead isotopic compositions in fluvial tailings,
  9            and dendrochronology results from the Boulder River watershed, Jefferson County, Montana. Denver, CO:
10            U.S. Department of the Interior, U.S. Geological Survey. USGS open file report 00-0038. Available:
11     Utsunomiya, S.; Jensen, K. A.; Keeler, G. J.; Ewing, R. C. (2004) Direct identification of trace metals in fine and
12            ultrafine particles in the Detroit urban atmosphere. Environ. Sci. Technol. 38: 2289-2297.
13     Van Den Hout, K. D.; Bakker, D. J.; Berdowski, J. J. M.; Van Jaarsveld, J. A.; Reinds, G. J.; Bril, J.; Breeuwsma,
14            A.; Groenenberg, J. E.; De Vries, W.;  Van Pagee, J. A.; Villars, M.; Sliggers,  C. J. (1999) The impact of
15            atmospheric deposition of non-acidifying substances on the quality of European forest soils and the North
16            Sea. Water Air Soil Pollut. 109: 357-396.
17     VanHattum, B.; Van Straalen, N. M.; Covers, H. A. J. (1996) Trace metals in populations of freshwater isopods:
18            influence of biotic and abiotic variables. Arch. Environ. Contam.  Toxicol. 31: 303-318.
19     Varanasi, U.; Gmur, D. J. (1978) Influence of water-borne and dietary calcium on uptake and retention of lead by
20            coho salmon (Oncorhynchus kisutch).  Toxicol. Appl. Pharmacol. 46: 65-75.
21     Vazquez, M. D.; Lopez, J.; Carballeira, A. (1999) Uptake of heavy metals to the extracellular and intracellular
22            compartments in three species of aquatic bryophyte. Ecotoxicol. Environ. Saf. 44:  12-24.
23     Verma, S.; Dubey, R. S. (2003) Lead toxicity induces lipid peroxidation and alters the activities of antioxidant
24            enzymes in growing rice plants. Plant  Sci. 164: 645-655.
25     Verma, N.; Singh, M. (2005) Biosensors for heavy metals. Biometals 18:  121-129.
26     Verslycke, T.; Vangheluwe, M.; Heijerick, D.; De Schamphelaere, K.; Van Sprang, P.; Janssen, C. R. (2003)
27            The toxicity of metal mixtures to the estuarine mysid Neomysis integer (Crustacea: Mysidacea) under
28            changing salinity. Aquat. Toxicol. 64:  307-315.
29     Vink, J. P. M.  (2002) Measurement of heavy metal speciation over redox gradients in natural water—sediment
30            interfaces and implications for uptake  by benthic organisms. Environ. Sci. Technol. 36: 5130-5138.
31     Vinogradoff, S. I.; Graham, M.  C.; Thornton, G. J. P.; Dunn, S. M.; Bacon, J. R.; Farmer, J. G. (2005) Investigation
32            of the concentration and isotopic composition of inputs and outputs of Pb in waters at an upland catchment
33            inNE Scotland. J. Environ. Monit. 7: 431-444.
34     Vogiatzis, A. K.; Loumbourdis, N. S. (1999) Exposure ofRana ridibunda to lead I. Study of lead accumulation in
35            various tissues and hepatic 5-aminolevulinic acid dehydratase activity. J. Appl. Toxicol. 19: 25-29.
36     Vogt, G.; Quinitio, E. T. (1994) Accumulation and excretion of metal granules in the prawn, Penaeus monodon,
37            exposed to water-borne copper, lead, iron and calcium. Aquat. Toxicol. 28: 223-241.
3 8     Wang, E. X.; Benoit, G. (1996) Mechanisms controlling the mobility of lead in the spodosols of a northern
39            hardwood forest ecosystem. Environ. Sci. Technol.  30: 2211-2219.
40     Wang, E. X.; Benoit, G. (1997) Fate and transport of contaminant lead in spodosols: a simple box model analysis.
41            Water Air Soil Pollut. 95: 381-397.
42     Wang, E. X.; Bormann, F. H.; Benoit, G. (1995) Evidence of complete retention of atmospheric lead in the soils of
43            northern hardwood forested ecosystems. Environ. Sci. Technol. 29: 735-739.
44     Ward, T. J.;  Hutchings, P. A. (1996) Effects of trace metals  on infaunal species composition in polluted intertidal
45            and subtidal marine sediments near a lead smelter, Spencer Gulf,  South Australia. Mar. Ecol. Prog.  Ser.
46            135: 123-135.
47     Ward, T. J.;  Young, P. C. (1982) Effects of sediment trace metals and particle size on the community structure of
48            epibenthic seagrass fauna near a lead smelter, South Australia. Mar. Ecol. Prog.  Ser. (Oldendorf)
49            9: 137-146.
50     Waters, T. F. (1995) Sediment in streams: sources, biological effects  and control. Bethesda, MD: American
51            Fisheries Society. (Monograph 7).
52     Watmough,  S. A.; Hutchinson, T. C. (2004) The quantification and distribution of pollution Pb at a woodland in
53            rural south central Ontario, Canada. Environ. Pollut. 128: 419-428.
54     Watmough,  S. A.; Hutchinson, T. C.; Sager, E. P. S. (1998) Changes  in tree ring chemistry in sugar maple
55            (Acer saccharum) along an urban-rural gradient in southern Ontario. Environ. Pollut. 101: 381-390.
       December 2005                                8-266        DRAFT-DO NOT QUOTE OR CITE

-------
  1     Watmough, S. A.; Hutchinson, T. C.; Dillon, P. J. (2004) Lead dynamics in the forest floor and mineral soil in
  2            south-central Ontario. Biogeochemistry 71: 43-68.
  3     Weber, D. N. (1993) Exposure to sublethal levels of waterborne lead alters reproductive behavior patterns in fathead
  4            minnows (Pimephalespromelas). Neurotoxicology 14: 347-358.
  5     Weber, D. N. (1996) Lead-induced metabolic imbalances and feeding alterations in juvenile fathead minnows
  6            (Pimephalespromelas). Environ. Toxicol. Water Qual. 11: 45-51.
  7     Wehrli, B.; Dinkel, C.; Muller, B. (1994) Measurement of benthic gradients in deep lakes with ion selective
  8            electrodes and video endoscopy. Mineral. Mag. 58A: 961-962.
  9     Weis, J. S.; Weis, P. (1998) Effects of exposure to lead on behavior of mummichog (Fundulus heteroditus L.)
10            larvae. J. Exp. Mar. Biol. Ecol. 222: 1-10.
11     Welter, E.; Calmano, W.; Mangold, S.;  Troger, L. (1999) Chemical speciation of heavy metals in soils by use of
12            XAFS spectroscopy and electron microscopical techniques. Fresenius J. Anal. Chem. 364: 238-244.
13     Weng, L.; Temminghoff, E. J. M.; Lofts, S.; Tipping, E.; Van Riemsdijk, W.  (2002) Complexation with dissolved
14            organic matter and solubility control of heavy metals in a sandy soil.  Environ. Sci. Technol. 36: 4804-4810.
15     Wierzbicka, M. (1999) Comparison of lead tolerance mAllium cepa with other plant species. Environ. Pollut.
16            104:41-52.
17     Wilczek, G.; Babczynska, A.; Augustyniak, M.; Migula, P. (2004) Relations between metals (Zn, Pb, Cd and Cu)
18            and glutathione-dependent detoxifying enzymes in spiders from a heavy metal pollution gradient. Environ.
19            Pollut. 132: 453-454.
20     Wilke, B.-M. (1989) Long-term effects of different inorganic pollutants on nitrogen transformations in a sandy
21            cambisol. Biol. Fertil. Soils 7: 254-258.
22     Wilson, A. R.; Lion, L. W.; Nelson, Y.  M.; Shuler, M. L.; Ghiorse, W. C. (2001) The effects of pH and surface
23            composition on Pb adsorption to natural freshwater biofilms. Environ. Sci. Technol. 35:  3182-3189.
24     Winner, R. W.; Boesel, B. W.; Farrell, M. P. (1980) Insect community structure as an index of heavy-metal
25            pollution in lotic ecosystems. Can. J. Fish. Aquatic Sci. 37: 647-655.
26     Woodward, D. F.; Hansen, J. A.; Bergman, H. L.; Little, E. E.; DeLonay, A. J. (1995) Brown trout avoidance of
27            metals in water characteristic of the Clark Fork River, Montana. Can. J. Fish. Aquat. Sci. 52: 2031-2037.
28     Xia, K.; Bleam, W.; Helmke, P. A. (1997) Studies of the nature of Cu2+ and Pb2+ binding sites in soil humic
29            substances using X-ray absorption spectroscopy. Geochim. Cosmochim. Acta61: 2211-2221.
30     Yanai, R. D.; Ray, D. G.; Siccama, T. G. (2004) Lead reduction and redistribution in the forest floor in New
31            Hampshire northern hardwoods. J. Environ. Qual. 33: 141-148.
32     Yang, Y.-Y.;  Jung, J.-Y.; Song, W.-Y.; Suh, H.-S.; Lee, Y. (2000) Identification of rice varieties with high tolerance
33            or sensitivity to lead and characterization of the mechanism of tolerance. Plant Physiol. 124:  1019-1026.
34     Yang, J.; Mosby, D. E.; Casteel,  S. W.; Blanchar, R. W. (2001) Lead immobilization using phosphoric acid in a
35            smelter-contaminated urban soil. Environ. Sci. Technol. 35: 3553-3559.
36     Yap, C. K.; Tan, S. G.; Ismail, A.; Omar, H. (2004) Allozyme polymorphism and heavy metal levels in the green-
3 7            lipped mussel Perna viridis (Linnaeus) collected from contaminated and uncontaminated sites in Malaysia.
38            Environ. Int. 30: 39-46.
39     Young, T. F.; Sanzone, S., eds. (2002) A framework for assessing and reporting on ecological condition: an SAB
40            report. Washington, DC: U.S. Environmental Protection Agency, Science Advisory Board; report no.
41            EPA-SAB-EPEC-02-009. Available: http://www.epa.gov/sab/pdf/epec02009.pdf [9 December, 2003].
42     Zartman, R. (1974) Lead isotopic provinces in the Cordillera of the western United States and their geologic
43            significance. Econ. Geol. Bull.  Soc. Econ. Geol. 69: 792-805.
44     Zenk, M. H. (1996) Heavy metal detoxification in higher plants—a review. Gene  179: 21-30.
45     Zhang, H.; Davidson, W.; Miller, S.; Tych, W. (1995) In situ high resolution measurements effluxes of Ni, Cu,
46            Fe and Mn and concentrations of Zn and Cd in pore waters by DOT.  Geochim. Cosmochim. Acta
47            59:4181-4192.
48     Zhu, Y. G.; Chen, S. B.; Yang, J.  C. (2004) Effects of soil amendments on lead uptake by two vegetable crops from
49            a lead-contaminated soil from Anhui, China.  Environ. Int. 30: 351-356.
50     Zimdahl, R. L.; Skogerboe, R. K. (1977) Behavior of lead in soil. Environ. Sci. Technol. 11: 1202-1207.
       December 2005                                8-267        DRAFT-DO NOT QUOTE OR CITE

-------
vvEPA
                           Please make all necessary changes in the below label,        PRESORTED STANDARD
                           detach copy or copy, and return to the address in the upper     POSTAGE & FEES PAID
                           left-hand corner.                                            EPA
United States                                                                  DCDMIT M^ r* Q^
Environmental Protection                                               m        PERMIT Na G'35
Agency                     If you do not wish to receive these reports CHECK HERE LJ;
                           detach copy or copy, and return to the address in the upper
                           left-hand corner.


 National Center for
  Environmental Assessment
 Research Triangle Park, NC 27711

 Official Business
 Penalty for Private Use
 $300

 EPA/600/R-05/144aA
 December 2005

-------