United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/9-81-001
January 1981
Research and Development
Proceedings of the
Research  Planning
Workshop on  Health
Effects of Oxidants

-------
                                                  EPA 600/9-81-001
                                                  January 1981
                   PROCEEDINGS OP THE
RESEARCH PLANNING WORKSHOP ON HEALTH EFFECTS OF OXIDANTS

                Raleigh, North Carolina
                  January 28-30, 1980
                      Edited by

                Northrop Services, Inc.
                Environmental Sciences
                    P.O. Box 12313
           Research Triangle Park, NC  27709
                   Project Officer

                       Jim Smith
     Research Advisory and Special Studies Office
              Health Effects Research Lab
           Research Triangle Park, NC  27711
           HEALTH EFFECTS RESEARCH LABORATORY
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S. ENVIRONMENTAL PROTECTION AGENCY
     RESEARCH TRIANGLE PARK, NORTH CAROLINA  27711

-------
                                  DISCLAIMER
     This report has been reviewed by the Health Effects Research Laboratory,




U.S. Environmental Protection Agency, and approved for publication.   Mention




of trade names or commercial products does not constitute endorsement or




recommendation for use.
                                      ii

-------
                                   ABSTRACT
     Presentations at the Research Planning Workshop on Health Effects of




Oxidants (Raleigh, North Carolina, January 28-30, 1980) are documented.  The




participants include scientists, administrators, and regulatory personnel from




the following agencies and institutions:  U.S. Environmental Protection




Agency, Brookhaven National Laboratory, Oak Ridge National Laboratory,




Lawrence Berkeley Laboratory, Inhalation Toxicology Research Institute, EPA




Science Advisory Board, California Air Resources Board, University of




California - Los Angeles, University of Southern California, University of




California - Davis, University of California - Santa Barbara, and University




of Rochester.









     The focus of the entire volume is the EPA-funded research that is planned




or in progress under Theme 1 ("Health Effects of Criteria and Non-Criteria




Pollutants from Fossil Fuel Combustion") of the Energy Interagency Health and




Ecological Effects Program.  The relevance of component projects to EPA




regulatory activities is a frequent topic of informal discussion.









     Chapter topics fall into five general categories:  regulatory activities




and concerns, general overviews of research programs, specific animal studies,




specific human studies, and biomathematical modeling.  Regulatory chapters






                                     iii

-------
review the legislative framework within which EPA operates and suggest lines




of research relevant to regulatory needs.  General program overviews are




presented for each of the agencies and institutions represented.









     Approximately half of the total 40 chapters describe animal




experimentation.  Most of the reported studies involve inhalation exposures to




ozone, nitrogen dioxide, or sulfur dioxide followed by examination of




biochemical, morphometric, or functional parameters.  In vitro methods receive




some attention.









     With respect to human studies, both epidemiologic and experimental




projects are reported.  The experimental protocols involve inhalation




exposures to ozone or nitrogen dioxide followed by measurements of pulmonary




function.  Chapters on biomathematical modeling discuss projections of human




absolute dosage and of the health effects of energy-related air pollution.









     The final chapter is a transcript of the Panel Discussion which concluded




the workshop.  The issues raised involve formal review of EPA-funded research




for scientific merit and relevance to regulatory activities.
                                      iv

-------
                                  CONTENTS
    ABSTRACT	    iii

    ABBREVIATIONS 	   xiii

1.  INTRODUCTION  	      1
    Robert E. Lee

2.  PROGRAM PLANNING FOR ENERGY HEALTH EFFECTS FROM CRITERIA
    AND NON-CRITERIA POLLUTANTS 	      3
    William Frietsch, III

         Energy Interagency Health and Ecological Effects Program ...      3
         Health Effects Program Area  	      7
         Concluding Remarks	     14
         Reference	-	     15
         Workshop Commentary  	     15

3.  ROLE OF THE SCIENCE ADVISORY BOARD IN RESEARCH PLANNING
    AND REVIEW	     16
    James L. Whittenberger

         Introduction	     16
         Committee Makeup and Orientation 	     17
         Steps in the Review Process	     18
         Recommendations	     19
         Concluding Remarks	     19
         Reference  ..... 	     20

4.  OVERVIEW OF REGULATORY AND COMPLIANCE REQUIREMENTS  	     21
    Michael H. Jones

         Introduction and Overview  	     21
         Regulatory Options Under the Clean Air Act	     24
         Development of National Ambient Air Quality Standards  ....     24
         Issues in Development of the Ozone Standard  	     27
         Current Research Needs	     29
         References	•	     30

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION 	    31
    Beverly E. Tilton

         Introduction 	    31
         The Framework	    32
         The Procedures	    34
         The Content	    40
         Current Research Needs 	    43
         References	    46
         Workshop Commentary  	    46

6.  TOXICOLOGICAL RESEARCH STRATEGIES IN RELATION
    TO EPA REGULATORY NEEDS	    47
    Edward L. Alpen

         Introduction 	    47
         Comments on EPA Recommendations	•	    48
         Reference  ..... 	    49
         Workshop Commentary  	    49

7.  CORRELATING EPIDEMIOLOGIC, CLINICAL, AND BIOLOGICAL
    RESEARCH ON OXIDANTS  	    53
    David L. Coffin

         Introduction	    53
         Previous Research  	  ..    55
         Promising Avenues for Future Research  	  ....    57
         Concluding Remarks 	    59

8.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
    AT BROOKHAVEN NATIONAL LABORATORY 	    60
    Robert T. Drew

         Life Sciences Research at Brookhaven National Laboratory ...    60
         Brookhaven Inhalation Toxicology Facility  	  ....    61

9.  INCORPORATING OXIDANTS IN ASSESSMENTS
    OF ENERGY-RELATED HEALTH EFFECTS  	    70
    Samuel C. Morris

         The Nature of a Health Effects Assessment	    70
         Special Challenges of Energy-Related Assessments 	    72
         Incorporating the Consideration of Oxidants  	    74
         Research Needs	•	    76
         Acknowledgment	    77
         References	•	    77
                                     vx

-------
10.   INTERACTIONS BETWEEN HYPERTENSION
     AND OXIDANT AIR POLLUTANTS	     78
     Robert T. Drew, Daniel L. Costa,
     Sonja Haber, and Junichi Iwai

          Introduction ..... 	     78
          Methods	     79
          Results and Discussion	     81
          Workshop Commentary  	     88

11.   THE EFFECTS OF OZONE ON RAT ERYTHROCYTES
     AFTER EXPOSURE IN VIVO	     91
     Karen M. Schaich and Donald C. Borg

          Introduction	     91
          Initial Studies  	     96
          Effect of Dietary Tocopherol 	    101
          Discussion	    106
          Summary and Overview of Current Research	    118
          References	    119
          Workshop Commentary	    121

12.  PATHOGENESIS OF CHRONIC LUNG DISEASE:
     THE ROLE OF TOXICOLOGICAL INTERACTION	    126
     Hanspeter Witschi

          Introduction	    126
          Background	    127
          Methods and Results	    128
          Discussion	    129
          Continuing Studies 	    131
          References	    132
          Workshop Commentary  	    133

13.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
     BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY
       PART 1.  RAT TRACHEAL TRANSPLANT SYSTEM	    135
     Ann C. Marchok

          Introduction and Background  	  .....    135
          Controlled Delivery of the Test Agent(s)	    137
          Increased Specification of End Points  	    139
          References	    143
          Workshop Commentary  	    144
                                     vii

-------
14.   OVERVIEW OF CURRENT AND PLANNED RESEARCH
     BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY
       PART 2.  TRACHEAL WASHING SYSTEM
       AND OXIDANT INHALATION EXPERIMENTS  	   145
     Walden E. Dalbey

          Introduction 	   145
          Irritant Gases as Cofactors
            in Nitrosamine-Induced Tumorigenesis	   14*>
          Tracheal Washing System  	   148
          Effects of Inhaled Nitrogen Dioxide
            on Pulmonic Lesions in Rats	•	

15.  EARLY STAGES OF RESPIRATORY TRACT CANCER:  A REVIEW 	   153
     Carol A. Heckman, C. C. Scott/
     A. C. Olson, and F. Snyder

          Introduction	   153
          The Ether Lipid Marker 	   156
          The Cell Shape Marker	   160
          Anomalies in Respiratory Tract Carcinogenesis  	   164
          References	   167
          Workshop Commentary  	   169

16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM INHALATION
     IN THE RAT	   171
     Nathanial W. Revis, T. Major, and Walden E. Dalbey

          Introduction	   171
          Background	   172
          Methods and Results	   174
          Recommendations for Further Research ... 	   178
          Workshop Commentary	   179

17.  EFFECTS OF NITROGEN DIOXIDE AND 3-METHYLFURAN INHALATION
     ON THE SMALL AIRWAYS IN THE MOUSE	   180
     Wanda M. Haschek

          Introduction 	   180
          Methods	   182
          Results	   183
          Discussion	   184
          References	   186
          Workshop Commentary  	   187
                                     viii

-------
18.   OVERVIEW OF RESEARCH AT LAWRENCE BERKELEY LABORATORY  	    189
     Edward L. Alpen

          Introduction	    189
          Cocarcinogenic Effects of Nitrogen Dioxide
            and Sulfur Dioxide in the" Mouse	    189
          Effects of Ozone on Serum Lipoprotein Concentrations
            in the Guinea Pig	*..    194

19.   OVERVIEW OF RESEARCH
     AT THE UNIVERSITY OF CALIFORNIA - DAVIS	    198
     Marvin Goldman

          Introduction 	    198
          Approach to Research 	    199
          Current Efforts  	    199

20.   PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY	    201
     Donald L. Dungworth

          Introduction 	    201
          Choice of Animal Models  	    201
          Exposure Regimens	    202
          Studies of Adaptation
            During Chronic Low-Level Exposure  	  .  	    203
          Other Studies	    205
          Comparison of Experimental and Epidemiologic
            Data for Ozone and Sulfur Oxides	    206
          References	    206
          Workshop Commentary  .. 	    207
                                        t
21.  BIOLOGICAL EFFECTS OF FLY ASH FROM COAL COMBUSTION	    209
     Otto Raabe

          Introduction	    209
          Sample Collection  	    210
          Sample Characterization  	 • 	    210
          Biological Test Systems  ..•.	    211
          Exposure Techniques  	    211
          Preliminary Results  	    213
          Current Studies  	    214
          Workshop Commentary  	    214
                                      ix

-------
22.  OVERVIEW OF RESEARCH
     AT THE INHALATION TOXICOLOGY RESEARCH INSTITUTE ..........   216
     Joe L. Mauderly

                                                                           9 1 6
          Introduction ........................ »•
          Operation and Funding  ....................
          Facilities and Staff .....................   217
          Research Capabilities  ....................   218
          Current Projects
23.  CELLULAR (IN VIVO) AND BIOCHEMICAL CHANGES
     FOLLOWING INHALATION OF ACID SULFATES:
     RAPID SCREENING TESTS TO DETERMINE
     THE PULMONARY RESPONSE TO COMBINED POLLUTANT EXPOSURES  ......   222
     Rogene F. .Henderson

          Problem  ...........................   222
          Approach ......................... ••   223
          Planned Research .......................   225
          Acknowledgments  ...... .................   226
          References ..........................   226
          Workshop Commentary  .. ..... . .............   227

24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES .............   229
     John A. Pickrell

          Problem  ...........................   229
          Approach ... ........................   231
          Planned Research ....... .... ............   235
          Acknowledgments  ....... ..... ...........   238
          References ..........................   239
          Workshop Commentary  .... ..... . ...........   239

25.  ACUTE-CHRONIC LOSS OF LUNG FUNCTION
     FOLLOWING INHALATION OF ACID SULFATES ...............   243
     Steven A. Silbaugh

          Problem  ...........................   243
          Approach ...........................   243
          Planned Research .....  ........ <.... .....   246
          Acknowledgments  ............. . ........  .   250
          Reference  ..........................   250
          Workshop Commentary  ......... . ...........   250

-------
26.  LUNG CLEARANCE MECHANISMS
     FOLLOWING INHALATION OF ACID SULFATES 	    256
     Ronald K. Wolff

          Problem	    256
          Approach	    257
          Planned Research ..... 	    263
          Acknowledgments	    264
          References	    264
          Workshop Commentary	    265

27.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
     BY THE INHALATION TOXICOLOGY BRANCH 	    266
     Donald E. Gardner

          Introduction	    266
          Staff, Facilities, and Budget  	    266
          Goals for Criteria Pollutant Research	    268
          Pollutants Studied 	    270
          Model Systems in Use	    270
          Model Systems Under Development  ... 	    272
          Studies to Be Replicated	    274
          Collaboration with Other  Investigators 	    275
          References	    276

28.  SOME SPECIFIC STUDIES PLANNED
     BY THE INHALATION TOXICOLOGY BRANCH 	    277
     Judith A. Graham

          Introduction	    277
          Chronic Nitrogen Dioxide  Exposure
            Using Mouse Infectivity Model  	    277
          Pentobarbital-Induced Sleeping Time
            in the Mouse	    279
          Sensitivity to Bronchoconstricting Agents  	    282
          Studies to Identify Susceptible Populations  	    284
          Workshop Commentary	•	    285
                                      xi

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY  ...........   289
     Fred J. Miller

          Introduction .........................   289
          Nasal Pharyngeal Removal ...................   290
          Gas Transport  ........................
          Modeling Process ...........  •  ...........   291
          Modeling Absolute Dosage:  Some Preliminary Results  .....   292
          Quantitative Biochemical Data on the Mucous Layer:
            A Major Research Need  ...................
          References ............  • .............   297
          Workshop Commentary  .....................   298

30.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
     BY THE HUMAN STUDIES DIVISION ...................   3°1
     Robert E. Lee

          Introduction ................ .  ........   301
          Epidemiologic Program  ..........  . .........   302
          Clinical Program .............  . .........   303
          Workshop Commentary  .....................   303

31.  HUMAN PULMONARY ADAPTATION TO OZONE ................   304
     Edward D. Haak, Jr.

          Introduction .........................   304
          Protocol ...........................   305
          Results and Discussion ....................   306
          Concluding Remarks ......................   312
          Workshop Commentary  .....................   313

32.  OZONE-INDUCED HYPERREACTIVITY AS MEASURED
     BY HISTAMINE CHALLENGE IN NORMAL HEALTHY SUBJECTS .........   314
     Milan J. Hazucha

          Introduction .................. .. .....   314
          Protocol .....................  . .....   315
          Results  ...........................   316
          Discussion ...... . .......  .  ...........   322
          References ..........................   324
          Workshop Commentary  .....................   325
                                     xii

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS
     TO LOW-LEVEL NITROGEN DIOXIDE 	    328
     Joel F. Ginsberg

          Introduction	    328
          Protocol	    329
          Preliminary Results  	    331
          Preliminary Conclusions  	  ...    338
          References	    340
          Workshop Commentary  	 ...............    340

34.  HEALTH EFFECTS OF AIR POLLUTANTS
     IN THE TEXAS GULF COAST AREA	    341
     Robert S. Chapman

          Introduction ..... 	    341
          Background	    342
          Planning Study 	    343
          Feasibility Assessment ...... 	    347
          Tentative Protocol 	    350
          References	    353
          Workshop Commentary	    353

35.  OVERVIEW OF RESEARCH AND REGULATORY ACTIVITIES
     OF THE CALIFORNIA AIR RESOURCES BOARD	    358
     John R. Holmes

          Introduction ......... 	    358
          Background	    358
          Research Program	    359
          California Air Pollution Standards 	    364
          Workshop Commentary	•	    367

36.  LUNG INJURY AND DEPLETION IN THE MOUSE
     FOLLOWING EXPOSURE TO NITROGEN DIOXIDE  	    368
     Russell P. Sherwin

          Introduction	'•	    368
          Background	•	    368
          Methods and Results	    370
          Discussion	    376
          References	    377
          Addendum	*	    377
                                     xiii

-------
37.  PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE  	   378
     Steven M. Horvath

          Introduction 	   378
          Pulmonary Effects of Ozone
            in Combination with Exercise	   378
          Pulmonary Habituation to Ozone
            in Combination with Exercise	   380
          Psychophysiological Studies  	   381
          References	•	   381
          Workshop Commentary	   382

38.  EPIDEMIOLOGIC STUDIES OF OXIDANT
     HEALTH EFFECTS IN THE LOS ANGELES AREA	   383
     Roger Detels

          Introduction 	   383
          Study Sites	   384
          Pulmonary Function Testing 	   385
          Preliminary Results	•	   387
          Discussion	   389
          Workshop Commentary  .... 	   390

39.  PROPOSED SCIENTIFIC PROGRAM
     OF THE COOPERATIVE STUDY OF OXIDANT
     HEALTH EFFECTS IN THE LOS ANGELES AREA	   395
     Stanley V- Dawson

          Introduction 	   395
          Epidemiologic Studies  	   395
          The Integrated Program	   398
          Improvement of Population Exposure Estimates . 	   400
          Human Laboratory Exposures	   400
          Animal Laboratory Studies  	   403
          Overall Aspects  	   405
          Workshop Commentary	   406

40.  PANEL DISCUSSION  	   411

     MAILING ADDRESSES:  AUTHORS AND DISCUSSANTS 	   428
                                     xiv

-------
                                ABBREVIATIONS
Ab's
AChE
APH
ARB
BHT
BP
CASAC
CEQ
COMT
DEE
DEN
DMBA
DNA
DOE
2 , 3-DPG
ECAO
FEV1
FIV
FRC
FVC
FWS
FY
GPx
GRase
6SH
G6PD
Hb
Hct
HRP
ITB
IVPL
LALN
LDH
MAA
MAO
MDH
MetHb
MMAD
antibodies
acetylcholinesterase
acetylphenyl hydrazine
Air Resources Board
butylated hydroxytoluene
ben zopyrene
Clean Air Scientific Advisory Committee
Council on Environmental Quality
catechol O methyltransferase
Department of Energy and Environment
diethylnitrosamine
dimethylbenzanthracene
deoxyribonucleic acid
Department of Energy
2,3-diphosphoglycerate
Environmental Criteria and Assessement Office
forced expiratory volume
forced inspiratory volume
functional residual capacity
forced vital capacity
Fish and Wildlife Service
fiscal year
glutathione peroxidase
glutathione reductase
reduced glutathione
glucose-6-phosphate dehydrogenase
hemoglobin
hematocrit
horseradish peroxidase
Inhalation Toxicology Branch
isolated ventilated perfused lung
lung-associated lymph nodes
median lethal concentration
lactate dehydrogenase
macroaggregated albumin
monoamine oxidase
malate dehydrogenase
methemoglobin
mass median aerometric diameter
                                      xv

-------
MMEF       —  maximal midexpiratory flow rate
MW        —  maximal voluntary ventilation
NAAQS      —  National Ambient Air Quality Standards
NAS        —  National Academy of Science
NASA       —  National Aeronautics and Space Administration
NBS        —  National Bureau of Standards
NIEHS      --  National Institute of Environmental Health Sciences
NIH        —  National Institutes of Health
NIOSH      —  National Institute of Occupational Safety and Health
NMU        —  N-methyl-N-nitrosourea
NOAA       —  National Oceanic and Atmospheric Administration
NOEL       —  "no observed effects" level
NPSH       —  nonprotein sulfhydryl
OAQPS      —  Office of Air Quality Planning and Standards
OMB        —  Office of Management and Budget
ORD        —  Office of Research and Development
ORNL       —  Oak Ridge National Laboratory
PCH        —  polycyclic hydrocarbons
pdf        —  probability density function
PFT        —  pulmonary function test
PK         —  pyruvate kinase
PMN        —  polymorphonuclear
RAW        —  airway resistance
RBC's      —  red blood cells
RTP        —  Research Triangle Park
SAB        —  Science Advisory Board
S.D.       —  standard deviation
SOD        —  superoxide dismutase
SRAW       --  specific airway response
TBA        —  thiobarbituric acid reactive products
TGV        —  thoracic gas volume
TIC        —  trypsin inhibitory capacity
TLC        —  thin layer chromatography
TVA        —  Tennessee Valley Authority
UCD        —  University of California - Davis
UCI        —  University of California - Irvine
UCLA       —  University of California - Los Angeles
USDA       —  U.S. Department of Agriculture
USGS       —  U.S. Geological Survey
UV         —  ultraviolet
VC         —  vital capacity
3MF        --  3-methylfuran
6PGD       —  6-phosphoglycerate dehydrogenase
                                     xvi

-------
                                1.  INTRODUCTION

                                Robert E. Lee

                  Health Effects Research Laboratory, MD-51
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
     On January 28-30,  1980, the Health Effects Research Laboratory (Office of

Research and Development, U.S. Environmental Protection Agency, Research

Triangle Park, North Carolina) sponsored a Research Planning Workshop on

Health Effects of Oxidants.  The meeting was held in Raleigh, North Carolina

and was attended by representatives of several governmental and private

agencies and institutions.



     The workshop focused on the research projects that are planned or in

progress under Theme 1  ("Health Effects of Criteria and Non-Criteria

Pollutants from Fossil  Fuel Combustion") of the Energy Interagency Health and

Ecological Effects Program.  Over the three days, attendees heard

approximately 30 presentations on these efforts.  Also presented were the

views of EPA personnel  who will use the results of such research in the

regulatory process.  The purpose of the present volume is to document each

presentation and the often lively discussion that followed.

-------
1.  INTRODUCTION	I»ee






     The subsequent chapters underscore the fact that research is a very human




activity.  In addition to the essential ingredients of money, facilities, and




instrumentation, it is still the individual researcher's creativity,




dedication, and often luck that may mean the difference between successful,




reproducible results and failure.  The individual researcher will always be




the key to successful research.  Accordingly, the basic workshop goals were to




encourage the exchange of knowledge, understanding,  and interpretation of




oxidants health effects research and, most importantly, to provide a forum for




individual creativity, ideas, and perspective on the additional research




needed to develop the data base that is essential for protecting the public




health.  To that end, the workshop was a notable success.

-------
                2.  PROGRAM PLANNING FOR ENERGY HEALTH EFFECTS
                  FROM CRITERIA AND NON-CRITERIA POLLUTANTS

                            William Frietsch, III

                        Energy Effects Division, RD682
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                            Washington, DC  20460
ENERGY INTERAGENCY HEALTH AND ECOLOGICAL EFFECTS PROGRAM



Background;  The First Five Years



     The Energy Interagency Health and Ecological Effects Program involves

coordination and funding of energy-related research by EPA and 10 other

federal agencies.  The program was created in 1973 after the Arab oil embargo

emphasized the impending acceleration of our country's energy development and

the need to address resultant environmental and health concerns.  Creation of

the program was driven by what is commonly known as the "King-Muir Report"

written by Don King of the State Department and Warren Muir of the Council on

Environmental Quality (CEQ).  With the assistance of Dr. Steven Gage and

personnel from about 20 federal agencies, King and Muir assembled an initial

five-year energy program that involved EPA and 10 other agencies.  The Office

of Management and Budget (OMB) and CEQ approved the program for a five-year

period to begin with fiscal year (FY) 1975 and end with FY 1979.

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	         Frietsch






     Table 2-1 indicates the basic program that remained in place for the




first five years.  The same agencies continue to participate.  An important




distinction is that research at the National Laboratories, unlike research at




the listed agencies, is not funded on a "pass through" basis.  ("Pass through"




refers to the procedure by which funds appropriated to EPA are "passed




through" to other agencies.)  In the case of the National Laboratories,




existing projects were transferred to EPA.









Recent Planning;  The Second Five Years









     About 18 months ago, OMB and CEQ approved the continuation of the program




for a second five years.  At that time, those of us involved in program




management decided to evaluate what program changes, if any, should be




implemented.  Two major criteria for program development in the second five




years were identified:  relevancy to agency programmatic needs and a client




relationship.  The first of these refers not only to EPA needs but also to the




needs of the other participating pass through agencies.  The second criterion




stems in part from frequent criticism that EPA research lacks a "client" or




"customer."  We in EPA have became very cognizant of this criticism and have




found that, in order to survive the budget cycles, we need to have a client




who wants the research, who works with us throughout development and conduct




of the research, and who receives the output of the research.









     With respect to the sources of research requirements, our first goal was




to respond to national energy needs and problems.  We negotiated with other

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING                            Frietsch
    TABLE 2-1.  ENERGY INTERAGENCY HEALTH AND ECOLOGICAL EFFECTS PROGRAM:
        PARTICIPANTS AND MAJOR PROGRAM AREAS FOR THE FIRST FIVE YEARS
                 	Program Area	

                                 Air Transport        Measurement
                  Ecological          and                 and           Health
   Agency*         Effects       Transformation     Instrumentation    Effects
EPA                   X

National
  Laboratories        X
DOE X
FWS X
NBS
NOAA X X
TVA X X
NIOSH
NASA
USGS
USDA X
NIEHS
X

X
X
X
X
X
X


X




X



X
 Definitions of agency abbreviations:   DOE, Department of Energy; FWS, Fish
  and Wildlife Service;  NBS,  National Bureau of  Standards; NOAA, National
  Oceanic  and Atmospheric Administration; TVA, Tennessee Valley Authority;
  NIOSH, National Institute of  Occupational Safety and Health; NASA, National
  Aeronautics and Space Administration;  USGS, U.S. Geological Survey; USDA,
  U.S.  Department of Agriculture; NIEHS, National Institute of Environmental
  Health Sciences.

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch




agencies to ensure sufficient emphasis in such areas as acid rain,  synthetic



fuels, air transport, visibility, and complex terrain.






     Of course, consideration of EPA programmatic needs was also extremely



important in the planning effort.  About three years ago, Dr. Gage  set up four



pilot Research Committees (Mobile Sources, Drinking Water, Pesticides, and



Gaseous and Inhalable Particulates).  In these Committees he brought together



staff members of EPA program offices and researchers working for the Office of



Research and Development (ORD).  The objective was to jointly plan  the ORD



program, to assure its responsiveness to program office needs.






     This Research Committee system has worked extremely well:  there are 12


Committees for FY 1980.  With very minor exceptions, the whole ORD  program is



strongly influenced by the Committees.  From the researchers' point of view,



the Committees provide the advantage of close coordination with the program



offices (the "clients").  Researchers have a forum in which to express their



own particular concerns (e.g., the need for longer-term research).   On the



other side of the fence, the advantage to the regulatory offices is clear:
                                                                         I


they have a strong voice in deciding what research will be performed.






     As a third source of research requirements, we considered the needs of



the other participating agencies.  The whole character of the interagency



program is based on negotiation of research projects that have high priority



in terms of mutual concern.  For example, DOE is extremely concerned about



synthetic fuels; so is EPA.  Our challenge in negotiating such programs is to

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING      	Frietsch





identify the particular projects which are most relevant to both agencies.




The system forces and fosters a good working relationship between EPA and the




other participating agencies.









HEALTH EFFECTS PROGRAM AREA









     Table 2-2 outlines the Health Effects program area for FY 1980. This




research effort is divided into a total of six "Themes"; funding totals ~$18




million.  Theme 1, Health Effects from Criteria and Non-Criteria Pollutants




from Fossil Fuel Combustion, has so far been the major focus of our FY 1981




planning exercise  (discussed below).









     Theme 3, Fossil Fuel Leachate Hazards, refers to the leaching of




materials from fossil fuel combustion into waters and soils.  The main EPA




clients for this work are the regulatory personnel concerned with drinking




water.









     Theme 4, Toxicological Studies of Specific Energy-Related Agents,




consists of research on cadmium performed at the Comparative Animals Research




Laboratory at Oak  Ridge National Laboratory.  Originally, this Theme involved




four or five compounds, but we subsequently discovered that most of the other




agents are covered by ORD base program activities.









     Theme 5, Relative Risk Assessments for Specific Energy Systems, seeks to




compare the known  risks of conventional combustion with those of the newer

-------
                                       TABLE 2-2.   HEALTH  EFFECTS  PROGRAM AREA:
                            THEMES, PARTICIPANTS,  AND FUNDING LEVELS FOR FISCAL YEAR 1980
                   Theme
                                                               Funding (thousands of dollars)8
                                             HEKL-RTP     HERL-CIN     DOE.,
                                                                                DOE     NIEHS     NIOSH
                                                               Total
                                                              by Theme
                                                                        M
                                                                        8
                                                                        *
00
1.  Health Effects of Criteria
    and Non-Criteria Pollutants
    from Fossil Fuel Combustion

2.  Development and Validation
    of Bioassay Screens and
    Predictor Test Protocols

3.  Fossil Fuel Leachate Hazards

4.  Toxicological Studies of
    Specific Energy-Related
    Agents (Cadmium)

5.  Relative Risk Assessments
    for Specific Energy Systems
    (Fluidized Bed Combustion
    Versus Conventional Combustion)

6.  Advanced Fossil
    Fuel Cycle Hazards
                                               2050
                                                             500
                                                             600
                        2368
                                                                       4065
                                                                                         1150
                                                                                         2150
                                                                        775
                                                400
                                                                                1217
                                                                                          500      1912
                                                     300
                                        5568




                                        7215




                                         600

                                         775




                                        2312





                                        1517
                                                                        W
                                                                        *1

                                                                        I
                                                                        CO
                                                                        V
                                                                        &
                                                                        8
                                                                        ?a
                                                                        >
                                                                        3
                                                                        a
     Total by Agency:
2450
                                                       1100
7208
1217
3800
2212
                                                                                                              17987
     aDefinitions of agency abbreviations:  HERL-RTP, Health Effects Research Laboratory  (EPA-ORD), Research Triangle
      Park, NC; HERL-CIN, Health Effects Research Laboratory (EPA-ORD),  Cincinnati, OH; DOEX, Department of Energy
      Projects transferred to EPA; DOE, Department of Energy; NIEHS, National Institute of Environmental Health
      Sciences; NIOSH, National Institute of Occupational Safety and Health.


-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch






fluidized bed combustion.  The National Institute of Occupational Safety and




Health has a major role in this Theme.  There are ~14 epidemiologic projects




focusing on workplace environments.









Planning for 1980;  Phase I









     One of the  first steps  in our planning  exercise for FY  1980 was to




discuss with the EPA program offices their energy-related research




requirements.  Walter Barber of the  Office of Air Quality Planning and




Standards  (OAQPS) indicated  a primary need for  additional health effects




studies to support review and revision of National Ambient Air Quality




Standards  for  both Criteria  and Non-Criteria Pollutants.  Barber also stressed




the need  for additional epidemiologic and clinical studies of human health.




His reason for this request  was  that he finds human data much more convincing




to Congress and to the  public interest groups.   These groups are not easily




convinced of  the relevance  of animal tojtficological data in establishing or




revising standards.  Barber was  careful to point out that he recognizes the




 importance of animal data,  and only advocates more of a balance between human




and animal studies.








      For the Energy Health Effects program,  at least, Barber's was a valid




 criticism:  at that time we had no epidemiology and a fairly low level  of




 clinical work.  We responded by dedicating almost half  of the 1980 Energy




 Health Effects program to epidemiology and clinical work.   We still  have a




 large animal toxicology program (as does the ORD base program).   The key

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch





question of what constitutes an "appropriate balance" between human and animal




work is a difficult one for which there may be no definitive answer.









     Following these discussions with OAQPS, the Energy Health Effects program




was structured into the six Themes shown in Table 2-2.  Theme 1 consists




partly of animal inhalation toxicology projects inherited from the National




Laboratories via an OMB mandate that $14 million worth of existing projects in




the health/ecological effects area be transferred to EPA.  The intent of the




transfer was to broaden EPA capabilities in this area by allowing us to draw




upon the tremendous talent and expertise residing in the National




Laboratories.  At the same time, ~$14 million worth of control technology work




was transferred from EPA to DOE.  OMB attached several ground rules to the




transfer:  Only projects involving conventional combustion would be




transferred to EPA; no human work would be affected.  Also, no project changes




would take place in FY 1979 and 1980.  The two-year moratorium on project




changes gave those of us in EPA a welcome opportunity to deal with our own




heavy workload and to become better acquainted with the National Laboratories




investigators.









     With respect to any changes for FY 1981, we prefer to keep project




resources within the particular National Laboratory in which they reside, and




we are absolutely committed to maintaining the total resources within the




National Laboratory system.  Thus, we might shift a particular project from




Brookhaven to Oak Ridge, for example, but we will not take those project




resources out of the National Laboratory system.







                                      10

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch





     Also included in Theme  1 are some clinical projects carried over from the




first five years, as well as some new epidemiologic work*









Planning for  1981;  Phase II









     About  six months ago we embarked upon  Phase  II of planning for the Energy




Health  Effects program.  As  it  would have been  impossible to perform an




in-depth evaluation  of  the entire $18-million program, our  strategy was to




focus on that part of the program which  responds  to OAQPS needs.  With the




assistance  of Richard Dowd we  assembled  a small Science Advisory Board (SAB)




subcommittee to  assist  us  in this  effort.  Dr.  James  L. Whittenberger was




named Chairman of the  subcommittee  (see  Chapter 3 of  this volume) .









      Our first public meeting was  held in Washington  on November 13-14, 1979.




This was an "information meeting":   no particular advice was requested or




 received.   Rather, the intent was to explain our  general planning philosophy




 and procedure.   The meeting agenda was split into three major  areas.  First,




 OAQPS staff described their research requirements in  as much detail  as




 possible.   Secondly, ORD staff presented the ORD base program.  In this, our




 goal was to promote coordination and minimum overlap between  the Energy Health




 Effects program and the base program (which has been in existence  for -15




 years).  The final presentation was on the Energy Health Effects program




 itself.  After outlining the program as  it stood at that time, we  asked the




 SAB subcommittee to work with us over the next few meetings to identify
                                        11

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	      Frietsch






changes needed to make the program more responsive to OAQPS needs  and to




ensure better coordination with the ORD base program.









     Subsequent to this meeting, we developed what is called the "straw man"




document (Miller et al. 1979).  The straw man document proposes changes in the




Energy Health Effects program to enhance its responsiveness to regulatory




needs and improve coordination with the ORD base program.









     A second public meeting was held in Washington on December 18-19,  1979.




The entire two days were devoted to discussions of the straw man document.




The SAB subcommittee concurred with most of the recommendations contained in




that document and added recommendations of its own.









     In January 1980, Dr. Charles Nauman, our health expert in the Energy




Effects Division, and this author made a quick "round robin" trip  to the




National Laboratories that would be affected by the proposed changes.  We




discussed the impact of the changes and received their feedback.  In several




cases we discovered misinterpretations, on our part, of their work; thus, the




discussions were of considerable benefit.









     In our visits to the National Laboratories, we repeatedly encountered




concern about the sometimes short-range nature of EPA programs. With regard




to Theme 1, it is significant that the Clean Air Act requires standard




revisions and revaluations on a five-year recurring basis.  Walter Barber of




OAQPS has stressed the importance that Theme 1 programs continue for five-year






                                      12

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch






periods, to permit acquisition of the best information to make regulatory




decisions.  And, indeed, we have endeavored to structure the Theme 1 projects




on five-year recurring bases.









     Aside from the needs of EPA (OAQPS), our Phase II planning exercise has




also considered the needs of the other agency that participates in Theme 1:




the National Institute for Environmental Health Sciences (NIEHS).  NIEHS has




expressed concern that the Theme 1 research program gives too much attention




to EPA regulatory needs.  Clearly, we must work closely with NIEHS to ensure a




program  consisting of projects that are of high priority to both agencies.









     Following the Research Planning Workshop on Health Effects of Oxidants




(January 28-30, 1980; documented by this volume), our next step will be to




meet again with some of the National Laboratories investigators.  A revised




straw man document will be published.  The SAB subcommittee will hold another




public meeting on March 11-12, 1980, at-which we will present the final or




semifinal set  of proposed changes.  We hope to complete the final 1981 program




decisions in April or May of  1980.









Implementation and Management









     Aside from the planning  exercise and the resultant program changes, an




equally  important question is how these changes will be implemented and the




entire Energy  Health Effects program managed.  With respect to management by




the Energy Effects Division, our very small staff represents a major







                                      13

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch





constraint:  it is not possible to run an $18-million program with one or two




individuals.  In addition, now that the Energy Health Effects program is more




closely coordinated with the ORD base program, it is increasingly apparent




that these efforts mesh and that most of the technical knowledge (for Theme 1,




at least) exists within ORD at Research Triangle Park, North Carolina (RTP).




Thus, our current plans are to assign project officers from the Health Effects




Research Laboratory at ORD-RTP where possible.









CONCLUDING REMARKS









     A similar planning effort is being mounted for Theme 2 (a ~$7.5 million




research effort).  Dr. Alan Moghissi from Dr. Gage's office will lead this




planning exercise.  Once again we have requested SAB's assistance.  To provide




continuity, we have asked Dr. Whittenberger to chair the Theme 2 SAB




subcommittee.  Of course, the subcommittee membership will be different (a




different expertise is needed).









     Major program changes are always difficult, and the Theme 1 planning




exercise could not have proceeded this far without tremendous cooperation from




all of the participants:  the ORD base program, SAB, and particularly the




National Laboratories.  We appreciate the feedback we have received, and we




will continue our efforts to acknowledge that feedback in proposals for




program revisions.  With everyone's help, the Theme 1 planning exercise will




be drawn to a successful conclusion.
                                      14

-------
2.  ENERGY HEALTH EFFECTS PROGRAM PLANNING	Frietsch


REFERENCE
Miller, F. J., J. A. Graham, M. Hazucha, C. Hayes, and W. Riggan.  1979.
     Preliminary Relevance Review of EPA-DOE Projects on Health Effects of
     Criteria and Non-Criteria Pollutants  from Fossil Fuel Combustion (Theme
     1).  Unpublished document, Health Effects Research Laboratory, Office of
     Research and Development, U.S. Environmental Protection Agency, Research
     Triangle Park, North Carolina, December 12.
WORKSHOP COMMENTARY
D. C. Borg;   As  someone at one of  the National Labs, I'd like to ask if there
is any  intention to involve us in  the "takeoff" as well as the "landing."
Will you involve us in the planning exercise or will you just tell us its
results?

W. Frietsch;   Yes,  we do plan to involve  you in the "takeoff."  In fact, we
plan to hold  a workshop in February 1980  at which all Theme 2 participants
will give verbal presentations. Also present will be representatives of our
base program  and (if assembled) the SAB subcommittee.

     As we go through this planning exercise, we're learning that it would be
very desirable to involve you more in the "up front" end of it.  That's
exactly our intent in having a workshop to kick off the whole planning
exercise.  We will also require (at the meeting) some written material, as we
did in  the first Theme.  But we definitely want people to give verbal
presentations and answer questions, etc.  about their work.
                                       15

-------
    3.  ROLE OF THE SCIENCE ADVISORY BOARD IN RESEARCH PLANNING AND  REVIEW

                            James L. Whittenberger

                           Department of Physiology
                           School of Public Health
                              Harvard University
                            665 Huntington Avenue
                              Boston, MA  02115
INTRODUCTION



     This report briefly describes the role of the Science Advisory Board

(SAB) in the ongoing revisions to Theme 1  of the Energy Health Effects program

(see Chapter 2 of this volume).  It is important to bear in mind that SAB's

role is advisory;  we are assisting EPA in redirecting Theme 1 to make it more

responsive to the needs of the Office of Air Quality Planning and Standards

and to achieve a better balance among the components of relevant animal

toxicological, clinical, and epidemiologic research.



     Having participated in the review of the criteria document leading to

revision of the National Ambient Air Quality Standard for photochemical

oxidants, and having participated in the Health Effects Research Review Group

which, at the request of Congress, examined health effects research in

relation to various EPA programmatic needs, this author feels very strongly

about the importance of providing EPA regulatory personnel with the very best


                                      16

-------
3.  ROLE OF THE SCIENCE ADVISORY BOARD	Whittenberger






possible data.  Those who work with criteria documents and who consider the




problem of translating scientific  information  (particularly health effects




information) into standards realize that, among the large number of studies




cited in a given criteria document, usually only a very small number are




really appropriate  for standard-setting.  It is thus extremely important that




those few studies be of high quality.  Making  such assessments of scientific




merit is perhaps the most difficult problem in planning research.









COMMITTEE MAKEUP AND ORIENTATION









     SAB participation in the Theme 1 planning exercise dates from September




12,  1979, when Mr.  Frietsch and I  met to discuss SAB's role.  The SAB




subcommittee assembled as a result of that meeting consists of Dr. Jack




Hackney, Dr. Ian Higgins, Dr. Daniel Menzel, Dr. Richard Riley, and myself as




Chairman.









     Most of the subcommittee members have a background in reviewing research




protocols as study  section members; thus, we prefer to know a lot about each




project, including  the methods to  be used, the techniques of data analysis and




interpretation, etc.  Accordingly, in the Theme 1 exercise, assessing




scientific merit has remained  (for. us, at least) an essential part of our job




even though the official subcommittee charge relates to research planning




rather than evaluation.  Because of this orientation, we have experienced a




number of problems  in reviewing both the written material given to us as well




as presentations at the meetings we have attended.  For example, we understand







                                      17

-------
3.  ROLE OF THE SCIENCE ADVISORY BOARD	Whittenberger





that there will be two new epidemiologic studies:   one in El Paso and one in




the Southern California Air Basin.  We have not been able to find out much




about either of these studies; therefore, we feel  we can express no judgment




whatsoever about them.









STEPS IN THE REVIEW PROCESS









     Between the November 13-14 and December 18-19 (1979) public meetings (see




Chapter 2 of this volume), EPA conducted its own review of the Theme 1




projects.  The EPA staff members involved in this  exercise were Dr. Fred




Miller, Dr. Judith Graham, Dr. Milan Hazucha, Dr.  Carl Hayes, and Dr. Wilson




Riggan.  These EPA scientists understood that they were to make a relevance




review rather than a scientific merit review.  Personally, I question whether




scientific merit can be separated from relevance.   All too often I have heard




that a particular study is subject to various significant criticisms and may




be of doubtful validity, but "cannot be ignored in standard-setting."  My view




is:  Pending replication, such a study not only can be but should be ignored




in standard-setting.









     At the December 18-19 meeting our subcommittee received the EPA staff




report (Miller et al. 1979), which included both general recommendations and




specific recommendations on individual projects.  In general, the subcommittee




agreed with these recommendations.  We could not determine, however, the




extent to which the various projects had undergone peer review on an




individual basis.  We knew that the National Laboratories employ laboratory







                                      18

-------
3.  ROLE OF THE SCIENCE ADVISORY BOARD	Whittenberger






reviews (program reviews); we could not ascertain the extent to which they




employ project reviews.









RECOMMENDATIONS









     Following the December meeting, the SAB subcommittee assembled our own




tentative recommendations which we consider to be consistent with the EPA




recommendations.  First, we recommend that rigorous external as well as




internal project review be undertaken for fiscal year 1981 and thereafter.  We




strongly endorse the  exchange between investigators of protocols for clinical




research.  We find certain projects to be of low priority (at least, as EPA




priorities have been  defined to us), and we recommend a major reprogramming of




effort.  Redirection  of Theme 1 projects can lead, we believe, to an increase




in  clinical  and epidemiologic work.  At present, our subcommittee lacks




sufficient information to determine what that balance should be.  One of the




principal goals of future meetings will be-to obtain such information.









CONCLUDING REMARKS









     With regard to comments by Frietsch (Chapter 2 of this volume) concerning




the needs of other participating  agencies, it is not clear why the needs of




any agency other than EPA should  be of primary concern.  In this exercise it




is  EPA requirements that are at issue.  EPA operates under laws that require




its research to be relevant to standard-setting and other forms of regulation;




if  that isn't the case (regardless of whether the studies are long-term or







                                      19

-------
3.  ROLE OF THE SCIENCE ADVISORY BOARD	Whittenberger


short-term), the research is simply not appropriate for EPA's purposes.  We in

the SAB subcommittee view the Research Planning Workshop on Health Effects of

Oxidants (January 28-30, 1980; documented by this volume)  as an opportunity to

become much better informed about the individual projects that comprise Theme

1 of the Energy Health Effects program.



REFERENCE
Miller, F. J., J. A. Graham, M. Hazucha, C.  Hayes,  and W.  Riggan.   1979.
     Preliminary Relevance Review of EPA-DOE Projects on Health Effects of
     Criteria and Non-Criteria Pollutants from Fossil Fuel Combustion (Theme
     1).  Unpublished document. Health Effects Research Laboratory,  Office of
     Research and Development, U.S.  Environmental Protection Agency,  Research
     Triangle Park, North Carolina,  December 12.
                                      20

-------
            4.  OVERVIEW OF REGULATORY AND COMPLIANCE REQUIREMENTS

                               Michael H. Jones

                 Strategies and Air Standards Division, MD-12
                 Office of Air Quality Planning and Standards
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION AND OVERVIEW



     This report attempts to provide some insight into the regulatory process

and into how EPA's Office of Air Quality Planning and Standards (OAQPS),  as a

customer for the output of the scientific community, uses information

generated by scientists in developing standards and regulations.   The key

point, perhaps, is that setting a standard is not strictly a scientific

enterprise.  Certainly at the heart of a standard's development is the

scientific information upon which a choice must be based.  It is important to

recognize, however, that there are other ingredients in setting National

Ambient Air Quality Standards (NAAQS) (or any other standards, for that

matter).



     Interpretation of the Clean Air Act and the legislative history is not an

activity of OAQPS alone.  We rely on EPA's Office of General Counsel and  their

attorneys to provide us with information on how far to go (if at all) in
                                      21

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	Jones





considering economics, for example, when selecting a standard.  Do we consider




the issue of attainability?  Although we're on record on most of these issues,




we nonetheless must rely on the guidance of the Office of General Counsel and




their attorneys to give us the proper direction.  As an example, probably the




key procedural issue during review of the photochemical oxidants standard was




the so-called "Shy Panel" that was convened early in the regulatory process.




Critics argued that procedural irregularities were associated with that




meeting.  This controversy flags the importance of working closely with our




attorneys to ensure that our actions are proper and in accordance with the




law.









     There is, of course, the obvious issue of interpretation of the health




evidence.  This is certainly the area where we spend most of our time and




where we need the most communication with the scientific community.









     Control effectiveness is another scientific element in development of a




standard.  In the case of ozone, for example, atmospheric chemistry bears




importantly on the effectiveness of control options.









     Determining the numbers of persons at risk is yet another scientific




enterprise.  This information is of interest to the Administrator in choosing




the level of a standard.









     An understanding of economic impacts is also important, even if that




information is not used in setting a standard level.  While it is our







                                      22

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS  	Jones





interpretation of the Clean Air Act that economics not be considered in




setting NAAQS, it is also our position that economic impacts be estimated and




made clear to the public so that the electorate can, if it sees fit, act




through elected officials to modify the way the law is structured.  In other




words, if the ozone standard is set at 0.12 ppm and provides a certain amount




of protection, it is important that the public realize the economic cost




associated with that protection.   If the electorate feels so inclined, it can




make choices for new directions.   Therefore, we strive to delineate economic




impacts as accurately and definitively as possible.









     The decisionmaker  (the Administrator of EPA) represents the




Administration.  His task involves a judgmental element and is not, as noted




earlier, a strictly scientific enterprise.  There is a margin of safety.  How




that margin of safety is characterized and what magnitude of risk the




decisionmaker will accept are important issues, and we must do our best to




clearly present the relevant information to him.  Concerns in this area




include marginal choices on what constitutes an adverse health effect.  In the




case of ozone, there is certainly  disagreement not only among policymakers but




also within the scientific community as to what constitutes an adverse effect.




It  is  unlikely that disagreements  of this sort will ever be totally resolved.




None the less, judgments must be made that reflect our best understanding of




science and of the health policy of our elected officials.
                                       23

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	     Jones





REGULATORY OPTIONS UNDER THE CLEAN AIR ACT









     NAAQS are not the only regulatory tool available to EPA for control of




air pollutants.  Section 112 of the Clean Air Act provides for National




Emission Standards for Hazardous Air Pollutants;  these exist for mercury,




beryllium, asbestos, and vinyl chloride.  Standards of Performance for New




Stationary Sources (Section 111) are technology-based standards that are, in




many cases, designed to support NAAQS.  Emission Standards for Moving Sources




(Section 202) are likewise tied in and support the attainment of NAAQS.









     Other options include the regulation of fuels and fuel additives,




inspection maintenance programs, and related control strategies.  Also, the




Emergency Powers clause (Section 303) permits the Administrator to take




immediate action in certain situations if there exists significant risk to




individuals.









     All of these regulatory options are available to the Administrator to




control air pollutants.  The Clean Air Act provides clear guidance as to how




he can and cannot use each option.









DEVELOPMENT OP NATIONAL AMBIENT AIR QUALITY STANDARDS









      In the 1970 Clean Air Act, EPA is tasked to list those ubiquitous




pollutants which, in the Administrator's judgment, have adverse effects  on




public health and welfare.  Primary standards protect public health; secondary







                                      24

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	jones






standards protect the public welfare against known or anticipated adverse




effects.  The second step is to issue criteria documents containing the latest




scientific knowledge on identifiable effects of pollutants on public health




and welfare.  After publication of each criteria document, a standard is




proposed.  Next is a period of public comment, followed by promulgation.




Finally, there is a requirement for periodic review and (where appropriate)




revision of each criteria document and standard.









     These provisions are changed and updated in the 1977 Clean Air Act




Amendments.  With regard to NAAQS, the 1977 Amendments set a timetable for




review of all the standards by the end of the 1980 calendar year.  The




Amendments further require (1) review on a five-year basis of all NAAQS, (2)




issuance of nitrogen dioxide criteria for a short-term standard, and (3)




establishment of a scientific committee to review criteria and standards.









     Determination of a primary standard level entails definition of the




adverse effect against which the standard is to protect, definition of the




population or group which is most sensitive to the particular effect, and




evaluation of all existing medical evidence set forth in the criteria




document.  In addition, outstanding uncertainties must be defined and




reasonable provision made for scientific and medical knowledge not yet




acquired.









     A number of judgmental choices need to be made in conjunction with the




scientific community.  EPA interprets the Clean Air Act and the  legislative







                                      25

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	 Jones





history to indicate that Congress has made a policy judgment in favor of




prevention of harm and in favor of exercising caution with regard to the




public health.









     A margin of safety is necessary to protect against hazards which research




has yet to define.  This is a legal as well as a scientific issue.  In my




view, a simple absence of evidence does not confirm that safety exists.  While




this is a very difficult issue, it is nonetheless something that the




Administrator must consider in selecting the final level of a standard.









     To protect the most sensitive groups in the population, the Administrator




must weigh risks to public health at pollutant levels below those convincingly




shown to impact health.  Once again, assessment of risk to public health is




based on scientific evidence but not based on fact alone.  In judging what




constitutes an adequate margin of safety, uncertainties in the evidence,




conflicting evidence, extrapolation, trends of known facts, and projections




from imperfect data must all be weighed.  We simply cannot ignore incomplete




or inconclusive evidence.









     Scientific information must be generated and then presented and compiled




into a criteria document.  Subsequently there is involvement of the scientific




community and review groups to ensure the adequacy of that document and that




we have the best possible basis upon which to set a standard.
                                      26

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	Jones






     Since the 1977 Amendments, a  staff paper  has been  employed to signal to




the public, prior to proposal, the way in which  EPA  interprets the medical




evidence in the criteria document.   The staff  paper  is  a public document which




is subject to public review.  In the carbon monoxide staff paper, we did our




best to present to the Clean Air Scientific Advisory Committee (CASAC) of the




Science Advisory Board (SAB) how we  plan to use  the  criteria document




information in the regulatory process.









     The next step is an internal  review process within the Agency.  Then the




Administrator makes a decision on  the standard.  Following the Administrator* s




decision, a standard is proposed.









     To indicate the length of time  involved,  our meeting with CASAC(SAB) to




discuss the carbon monoxide staff  paper was held on  June 11, 1979; hopefully




we'll  be able to propose a standard  in February  or March of 1980.  Development




of NAAQS is not a speedy process.  In the case of the original oxidants




standard, it was probably  a 30-month period from the time the scientific




information was compiled and presented in the  first  draft of the criteria




document until we proposed or promulgated the  standard.









ISSUES IN DEVELOPMENT OF THE OZONE STANDARD









     The 1977 Clean Air Act Amendments required  EPA  to  review the 1971




National Ambient Air Quality Standard for Photochemical Oxidants.  Today, the




revised standard (the "Ozone" standard) is probably  the most expensive






                                      27

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	Jones





environmental program in the country, with an estimated $5-10 billion annual




cost.  Most of that cost is associated with the emission standards on




automobiles.









     Of course, there was (and is) a wide divergence of medical opinion with




regard to the scientific evidence on ozone.  Therefore, the scientific




community's assistance was essential in providing our Administrator with the




kind of information that he needed in order to make a reasoned judgment.









     Besides medical questions, a number of other issues surfaced during




development of the ozone standard.  One such issue was severity of effects.  A




reinterpretation of evidence from the study upon which the original oxidants




standard was based (the Schoettlin/Landau study) led to designation of a new




estimated effect level in the revised criteria document.  There was a




continuing debate with the attorneys as to whether that action was appropriate




or inappropriate.  Review of the criteria document by CASAC(SAB) pointed up




differences of opinion as to the quality of the document.









     Another issue centered on measurement methods for singlet oxygen.  The




State of Texas brought to our attention what it considered to be an important




finding:  that all the clinical studies were simply invalid due to artifact




production by the ozone generators.  This issue had to be resolved prior to




promulgation of the standard.
                                      28

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	  Jones






     The Council of Economic Advisors, after reviewing  the  standard, suggested




that the level could be relaxed above what was proposed.  The Council felt




that the Federal Government should  consider cost across public health and




safety fields:  For example, are health dollars spent to protect people from




ozone as cost-effective as those that could be spent on highway guard rails?




We simply don't feel that this is an issue, because the Clean Air Act




emphasizes the need for protection  of public health from air pollutants.









     Other issues included:  economics, our modification of the nomenclature




from "photochemical oxidants" to "ozone," the contribution of natural ozone,




and adequate margins of safety.









CURRENT RESEARCH NEEDS









     During development of the ozone standard, we flagged several areas in




which a need for additional scientific evidence exists.  For example, to help




resolve the issue of what constitutes an adverse health effect, we need to




improve the characterization of the kind and magnitude  of various pulmonary




function detriments, biochemical changes, and other responses.









     We require a better understanding of the importance of the animal




studies, particularly with regard  to increased resistance to bacterial




infection.  Although  (to the author's knowledge) this effect has not been




demonstrated in clinical studies,  we remain concerned due to the seriousness
                                       29

-------
4.  REGULATORY AND COMPLIANCE REQUIREMENTS	Jones






of any such potential effect.  Any information on its relevance to human




health is certainly needed.










     Determination of the long-term chronic effects of photochemical oxidants




and of the potential for such damage as accelerated aging of body tissues,




chromosomal damage, mutagenesis, carcinogenesis, etc. is another area in which




good information is sparse.  Often we don't understand the significance or




importance of the effects that have been observed.










     Identification of particularly sensitive groups and investigation of




adverse effects on materials are additional scientific needs.









REFERENCES



              \





Clean Air Act.  42 USC 7401 et seq.




Clean Air Act Amendments of 1977.  Public Law 95-95, August 7, 1977.
                                      30

-------
                  5.  CRITERIA DOCUMENT REVIEW AND REVISION

                              Beverly E. Tilton

             Environmental Criteria and Assessment Office, MD-52
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     The review and revision of criteria documents constitute a broad topic,

but may be considered to consist of three aspects:  the framework,  the

procedures, and the content.  The framework is made up of the Clean Air Act

requirements, the organizational structure of the pertinent EPA organizations,

and the division of responsibilities among those organizations.  The

procedures followed are those stipulated by the Clean Air Act and those

established by EPA on the bases of precedent, practice, and practicality.   The

content of a criteria document is determined by the Clean Air Act,  by the

nature of the pollutant under consideration, and by the intended use of the

document.
                                      31

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	   Tiltqn


THE FRAMEWORK



     Section 108 of the Clean Air Act as amended in 1977 specifies that the

Administrator issue air quality criteria for each air pollutant
          (a)(1)(A) emissions of which, in his judgment, cause or
     contribute to air pollution which may reasonably be anticipated to
     endanger public health or welfare;

          (B) the presence of which in the ambient air results from
     numerous or diverse mobile or stationary sources;...
The Act further specifies that
          (2)  ...Air quality criteria for an air pollutant shall
     accurately reflect the latest scientific knowledge useful in
     indicating the kind and extent of all identifiable effects on public
     health or welfare which may be expected from the presence of such
     pollutant in the ambient air, in varying quantities.  The criteria
     for an air pollutant, to the extent practicable, shall include
     information on—

          (A) those variable factors (including atmospheric conditions)
     which of themselves or in combination with other factors may alter
     the effects on public health or welfare of such air pollutant;

          (B) the types of air pollutants which, when present in the
     atmosphere, may interact with such pollutant to produce an adverse
     effect on public health or welfare; and

          (C) any known or anticipated adverse effects on welfare.
     It should be noted at the outset that the Clean Air Act does not employ

the terminology "criteria document."  It specifies the "issuance" of criteria,

but does not specify the form or format in which such criteria are to be
                                      32

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	     Tilton







issued or disseminated.  Criteria documents as they presently exist are the




product of an evolving process that had  its genesis in legislation prior to




the Clean Air Act and that was first manifested  in the criteria documents




issued in 1970.  Early legislation did specify the publication of information




on the effects of and control of air pollutants, but subsequent legislation




did not refer specifically to publication of criteria.









     Not only does  it not specify the form in which criteria are to be issued,




the Clean Air Act also does not define "air quality criteria."  By dictionary




definition, criteria are "rules, tests,  or indices that can be used as a basis




for decisions or judgments."  It is clear from the portion of Section 108




cited above that criteria are to be descriptions of the effects of air




pollutants on man and his environment.   Combining the dictionary definition




with the definition implicit in Section  108, "air quality criteria" are




descriptions of air pollution effects presented  in a manner that renders them




suitable as the scientific basis for National Ambient Air Quality Standards




(NAAQS).  Thus, in  practice, air quality criteria are descriptions of




cause-and-effect relationships expressed, to the extent possible, as




dose-response functions.  Criteria documents, then, while they may serve a




number of worthwhile purposes, are prepared primarily as the vehicle for




issuance of the criteria that serve as the scientific basis for NAAQS.  Air




quality criteria, as indicated in Section 108, must reflect the latest




scientific information available that pertains to the effects of an air




pollutant, direct or indirect, on man and his environment.  Scientific




information needed  to formulate control  strategies—such as that useful in







                                      33

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	    Tilton






defining atmospheric transformations and source-receptor relationships—is to




be included only to the extent practicable.









     The legislative framework that determines the basic philosophy and




content of criteria documents has been discussed above.   A brief examination




follows of the organizational framework within which document review and




revision proceed.









     Responsibility for the preparation of air quality criteria documents




rests with the Environmental Criteria and Assessment Office (ECAO).  Located




at Research Triangle Park, North Carolina, ECAO is one of two such offices




within the Office of Health and Environmental Assessment; the latter is in




turn part of the Office of Research and Development (ORD) (Figure 5-1).




Responsibility for deriving and promulgating NAAQS, and for issuing control




techniques information, lies with the Office of Air Quality Planning and




Standards (OAQPS), which is part of the Office of Air, Noise and Radiation.




Because the Clean Air Act stipulates that standards be proposed and




information on control techniques be issued at the same time that criteria are




issued, close coordination between ECAO and OAQPS is required.  Most




day-to-day communication occurs directly between these offices.









THE PROCEDURES









     The responsibilities of ECAO versus those of OAQPS in the review and




revision of criteria and standards are shown in Figure 5-2.  Segments in which







                                      34

-------
                                                                                            OFFICE OF RESEARCH AND DEVELOPMENT
              OFFICE OF RESEARCH PROGRAM
                     MANAGEMENT
                   OPERATIONS OFFICE
             TECHNICAL INFORMATION OFFICE
u>
en
               CENTER FOR ENVIRONMENTAL
                 RESEARCH INFORMATION
                    CINCINNATI. OH
                                                                                               ASSISTANT ADMINISTRATOR FOR
                                                                                                RESEARCH AND DEVELOPMENT
                               OFFICE OF THE PRINCIPAL SCIENCE
                                        ADVISOR
                                                OFFICE OF MONITORING AND
                                                  TECHNICAL SUPPORT
                                               PROGRAM OPERATIONS STAFF
   NATIONAL WORKFORCE
   DEVELOPMENT STAFF
 QUALITY ASSURANCE AND
MONITORING SVSTEMS DIVISION
                                               TECHNICAL SUPPORT DIVISION
ENVIRONMENTAL MONITORING
   SVSTEMS LABORATORY
RESEARCH TRIANGLE PARK, NC
                                                ENVIRONMENTAL MONITORING
                                                AND SUPPORT LABORATORY
                                                     CINCINNATI. OH
                                                ENVIRONMENTAL MONITORING
                                                  SVSTEMS LABORATORY
                                                   LAS VEGAS. NEVADA
                                 OFFICE OF ENVIRONMENTAL
                                    ENGH. AND TECH.
                                                                                PROGRAM OPERATIONS STAFF
                                                                                 WASTE MANAGEMENT DIV.
                                                                                ENERGY PROCESSES DIVISION
                                 INDUSTRIAL AND EXTRACTIVE
                                    PROCESSES DIVISION
                                                                                 MO GRAM INTEGRATION
                                                                                   AND POLICY STAFF
INDUSTRIAL ENVIRONMENTAL
  RESEARCH LABORATORY
RESEARCH TRIANGLE PARK. NC
                                 INDUSTRIAL ENVIRONMENTAL
                                  RESEARCH LABORATORY
                                     CINCINNATI. OH
                                 MUNICIPAL ENVIRONMENTAL
                                  RESEARCH LABORATORY
                                      CINCINNATI, OH
                                                                               SENIOR ORD OFFICIAL.RESEARCH
                                                                                   TRIANGLE PARK. NC
                                                                                  SENIOR ORD OFFICIAL
                                                                                    CINCINNATI. OH
                                 OFFICE OF ENVIRONMENTAL
                               PROCESSES AND EFFECTS RESEARCH
                                                                                                                PROGRAM OPERATIONS STAFF
                                AGRICULTURE AND NON POINT
                               SOURCES MANAGEMENT DIVISION
                                                                MEDIA DUALITY MANAGEMENT
                                                                        DIVISION
                                                                                                                ENERGY EFFECTS DIVISION
                                                                ECOLOGICAL EFFECTS DIVISION
                                                                  ENVIRONMENTAL SCIENCES
                                                                   RESEARCH LABORATORY
                                                                RESEARCH TRIANGLE PARK. NC
                                 ENVIRONMENTAL RESEARCH
                                 LABORATORY, ATHENS. GA
                               ROBERT S. KEHR ENVIRONMENTAL
                               RESEARCH LABORATORY, ADA, OK
                                                                 ENVIRONMENTAL RESEARCH
                                                              LABORATORY, CORVALLIS, OREGON
                                                                 ENVIRONMENTAL RESEARCH
                                                                 LABORATORY, DULUTH, MN
ENVIRONMENTAL RESEARCH
LABORATORY, NARRAGANSETT,
Rl
                                                                                                                ENVIRONMENTAL RESEARCH
                                                                                                               LABORATORY. GULF BREEZE. Fl
                                                                                                                                                OFFICE OF HEALTH RESEARCH
                                                                                                                                                 PROGRAM OPERATIONS STAFF
                                                                                                                                                 HEALTH EFFECTS DIVISION
 HEALTH EFFECTS RESEARCH
      LABORATORY
RESEARCH TRIANGLE PARK. NC
                                                                                                 HEALTH EFFECTS RESEARCH
                                                                                                 LABORATORY CINCINNATI. OH
                                                                                                  NATIONAL CENTER FOR
                                                                                                  TOXICOLOGICAL RESEARCH
                                                                                                     JEFFERSON. AR
                                                                                                                                   OFFICE OF HEALTH AND
                                                                                                                                  ENVIRONMENTAL ASSESSMENT
                                                                                                                                  CARCINOGEN ASSESSMENT
                                                                                                                                         GROUP
                                                                                                                                                                                EXPOSURE ASSESSMENT GROUP
                                                                                                   REPRODUCTIVE EFFECTS
                                                                                                    ASSESSMENT GROUP
 ENVIRONMENTAL CRITERIA
  AND ASSESSMENT OFFICE
RESEARCH TRIANGLE PARK. NC
                                 ENVIRONMENTAL CRITERIA
                                 AND ASSESSMENT OFFICE
                                    CINCINNATI, OH
                                                                                                                                   o
                                                                                                                                   S
                                                                                                                                   a
                                                                                                                                   8
                                                                                                                                                                  CO
                                                                                                                                                                  H
                                                                                                                                                                  i
                       Figure  5-1.    Office  of   Research  and  Development,  U.S.   Environmental   Protection  Agency.
                                                                                                                                                                 §

-------
                                                                                                                                O
                                                                                                                                B
                    ORO AND
                   SCIENTIFIC
                   COMMUNITY
                                                                                                     OAQPS (OANR)
                                                                                         B
ECAO (ORD)
ECAO (ORD)
OAQPS (OANR)
OAQPS (OANR)
                                                           REGULATORY IMPACT
                                                                ANALYSIS
u>
SCIENTIFIC
RESEARCH

1 	
l
CRITERIA
DOCUMENT
1 i
i 	 '
t
ORD
RESEARCH
COMMITTEES





STAFF PAPER
INTERPRETING KE>
STUDIES IN
CRITERIA DOCUMEN
1 '
PUBLIC AND
SCIENTIFIC PEER
REVIEW

f
T


REGULATORY
DECISION
PACKAGE
i
\ r
PUBLIC AND
SCIENTIFIC PEER
REVIEW


AGENCY
REVIEW



ADMINISTRATOR'S
DECISION
1 '
PROPOSAL
                                                                                                    PUBLIC MEETINGS
                                                                                                    AND COMMENTS
                                                                                   AGENCY
                                                                                   REVIEW
                                                          REGULATORY DECISION
                                                           PACKAGE REFLECTING
                                                            PUBLIC COMMENTS
                         (ORD) • OFFICE OF RESEARCH AND DEVELOPMENT
                         (ECAO) - ENVIRONMENTAL CRITERIA AND ASSESSMENT OFFICE

                         (OANR) • OFFICE OF AIR, NOISE, AND RADIATION
                         (OAQPS) • OFFICE OF AIR QUALITY PLANNING AND STANDARDS
                                                             ADMINISTRATOR
                                                                 DECISION
                                                                                                    PROMULGATION
                         Figure 5-2.  National Ambient  Air Quality Standard-setting process.
                                                                                        O
                                                                                        a

-------
JL»  CRITERIA DOCUMENT REVIEW AND REVISION            	Tilton






ECAO has primary responsibility or participates  are  highlighted  in bold




outline.  ECAO has  responsibility  for preparing  the  criteria document, for




reviewing and incorporating (as needed)  comments resulting  from  the public




review process, and for helping OAQPS prepare  the staff paper that forms a




significant part of the regulatory information developed for NAAQS proposal




and  subsequent promulgation.   In addition,  ECAO  provides technical assistance




to OAQPS by helping brief the  Administrator on the scientific data base and




its  significance and by attending  public hearings on the standard to present




the  EPA viewpoint and  to answer questions regarding  criteria for the pollutant




under consideration.









     The  recent  inclusion of ECAO  staff as  members of  ORD research committees




is a significant development.   This involvement, along with the  principal role




played on these  committees by OAQPS staff,  provides  a  mechanism  for pointing




out  the research needs and problems identified during  criteria and standards




development.   This  important feedback mechanism  will help ORD plan its




research  programs and  will assist  those EPA offices  and personnel responsible




for  developing timetables and allocating funds.   In  time, the mechanism will




also help those  of  us  in ECAO and  OAQPS who are  in constant search of the data




base needed for  standard-setting.









     Figure 5-2  does not give an  adequate view of the  effort that must be




expended  to prepare a  criteria document.  A better view of  the complexity,




time, and demands of the review and revision process is shown in Table 5-1. As




indicated,  the preparation of a criteria document proceeds  in six phases.







                                       37

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	Tilton


             TABLE 5-1.  PHASES IN CRITERIA DOCUMENT PREPARATION
      Phase                      Activity                         Time
                                                                 (days)
        I             document planning and initiation           60-120

       II               preparation of working draft             60-200

      III           preparation of external review draft         60-120
                        (revisions to working draft)

       IV               public and CASAC(SAB)* review            60-90
                           of external review draft

        V               preparation of final document            60-100
                    (revisions to external review draft)

       VI                  closure by CASAC(SAB)                 60-90
                        on document and publication
                                                Total (months):  13-24
 Clean Air Scientific Advisory Committee (CASAC) of the Science Advisory
 Board (SAB).
     Phase I consists of (1) literature search and acquisition; (2)

publication of a Federal Register notice to solicit information on the

pollutant under consideration;  (3) selection of the ECAO project manager and

team members; (4) development of a work plan, document outline, and schedule;

(5) selection of a task force composed of EPA personnel; (6) recruitment of

outside consultants to write the document and additional consultants to review

it.
                                      38

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	        Tilton






     Phase II consists of  (1)  reading  and  analyzing  literature; (2) continuing




the acquisition of literature;  (3) writing;  (4) holding meetings between the




ECAO team and consultants;  (5)  typing, proofreading, and printing the working




draft; and (6) distributing the draft  for  internal review by EPA personnel and




consultants.









     Phase III begins with the convening of  a workshop, which is one of the




best and least time-consuming  mechanisms for ensuring that ECAO receives




adequate constructive criticism on  its documents early in their preparation.




Follow-up meetings with  consultants  are  held and the working draft is revised,




resulting in  an  external review draft, the availability of which is announced




to the public in a Federal Register notice.








      In Phase IV, the document is reviewed by the public, by the Clean Air




Scientific Advisory  Committee  (CASAC)  of the Science Advisory Board (SAB), by




EPA personnel (including the task force),  and by other Federal agencies.  The




review culminates in presentation and  discussion of the external review draft




at a  public meeting  of CASAC(SAB).








      Phase V  entails assessing comments  received  (1) by mail from the general




public and governmental  sectors and (2)  at the  public meeting from CASAC (SAB)




members and other attendees.   All comments are  logged in, are reviewed, and as




appropriate are  incorporated in revisions  to the  document.  All comments and




responses to  those comments are kept as  part of the  public  record for the




document.






                                       39

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	Tilton






     In the final phase, the revised draft is recirculated to CASAC(SAB) if




requested.  If major changes have been made to it, the document is also




recirculated to the public for a second review and is presented and discussed




at a second public meeting of CASAC(SAB).  When the document meets its




approval, CASAC(SAB) submits a letter of concurrence and a written report to




the Administrator of EPA.  After further editing for clarity, style, and




format, the document is published.









THE CONTENT









     The Clean Air Act stipulates that information on the effects of air




pollutants on the public health and welfare constitute the minimum and primary




content of any air quality criteria document.  It recommends, but does not




dictate, the inclusion of additional information, as noted earlier in the




quotation from Section 108.  As possible, or where necessary, information on




air quality management is included (e.g., emission inventories, modeling, and




atmospheric transport and transformation).  Such information is particularly




appropriate for secondary pollutants whose source-receptor relationships can




only be ascertained through understanding the transformation and transport of




the primary pollutants that serve as precursors.









     Criteria documents are prepared for substances that have been clearly




identified as pervasive pollutants posing threats to public health and the




environment.  When EPA begins to prepare a criteria document, then, it usually
                                      40

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	Tilton


encounters a sizeable and comprehensive body of  literature.  Typically, an air

quality criteria document contains the following:
     1.  Extensive chapters on animal toxicology, human clinical studies,
         and human epidemiologic  studies, both occupational and
         nonoccupational.  Such chapters include the pertinent
         information on human metabolism and pharmacokinetics, including
         uptake, distribution, metabolism or detoxification, and
         excretion.

     2.  One or more chapters on  the effects of the pollutant on plants,
         animals, acquatic and terrestrial  ecosystems, and natural and
         man-made materials.

     3.  One or more chapters that  treat fairly exhaustively the
         methodology available for  measuring the pollutant in ambient
         air, for measuring the pollutant in biological tissues and
         fluids (where pertinent),  and, in  a separate chapter or in the
         health effects chapters, for measuring observed effects.  Where
         possible, these  chapters include estimates of the precision,
         accuracy, and reliability  of the methods actually employed to
         obtain the ambient air,  exposure,  and effects data presented in
         the document.

     4.  Condensed chapters dealing, respectively, with chemical and
         physical properties; sources and emissions; geographic and
         temporal distribution of sources;  ambient air concentrations;
         the global cycle of the  pollutant  in all its forms; and
         atmospheric transport and  transformation of the pollutant.

     5.  Where possible,  a chapter  that assesses quantitatively the
         potential risk for the most sensitive segment of the population.

     6.  A summary and conclusions  chapter  focusing on the interpretation
         and significance of the  data cited in the document.  This
         chapter relates  ambient  air concentrations to observed effects,
         "no observed effects," and "least  observed effects" levels for
         sensitive or at-risk populations.
     As this  enumeration indicates,  air quality  criteria result from the

synthesis of  essentially two  lines of  inquiry  (Figure 5-3).
                                       41

-------
5.  CRITERIA DOCUMENT REVIEW AND  REVISION
                                 Tilton
           AMBIENT AIR MEASUREMENTS

           MEASURED LEVELS
           VALIDITY OF MEASUREMENTS
           SOURCE IDENTIFICATION
HEALTH/WELFARE EFFECTS OBSERVATIONS

 IDENTIFICATION OF EFFECTS
 ESTABLISHMENT OF LEVELS/DOSES
 IDENTIFICATION OF POPULATIONS AT RISK
                               AIR QUALITY CRITERIA

                   AMBIENT AIR LEVELS AT WHICH EFFECTS WILL OCCUR
                   SIGNIFICANCE FOR POPULATIONS OF CORRELATED
                      EFFECTS AND LEVELS
 Figure 5-3.  Lines of inquiry resulting in synthesis  of air quality criteria.



      Without health or welfare effects we would have  no criteria upon which to

 base regulations, but it  is equally true that the observation of effects is

 meaningless in the context of standard-setting if the effects cannot be

 correlated with the true  levels at which they occur.   Thus, it is important to

 be able to measure the criteria pollutant with specificity, accuracy, and

 precision in the ambient  air and in experimental settings.   Equally important,

 though not shown in Figure 5-3, is the ability to measure the effect itself

 with specificity, accuracy, and precision.  Qualitative information on

 pollutant effects is never disregarded in preparing a criteria document, but

 it serves mainly to help  establish the validity of available quantitative

 data.  To be of use in setting an ambient air standard, air quality criteria

 should be of a quantitative nature.  Air quality criteria,  then, are the

 synthesis of these two basic lines of inquiry and documentation.
                                        42

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION        	Til ton


     It should be noted that air quality criteria-must be objective statements

of the most recent available scientific evidence.  They are to be free from

bias, from speculation stated as fact, from  economic considerations relative

to the protection of public health, and from such language and presentation as

might cause them to be construed as standards rather than criteria.



     Criteria do not include a margin of safety but, wherever possible, they

include an assessment of any uncertainties associated (1) with the severity

and nature of the effects  caused by the pollutant in question and (2) with the

measurement and determination of the concentrations at which such effects

occur.



CURRENT RESEARCH NEEDS



     No discussion of the  review and revision of criteria and criteria

documents would be complete without brief mention of areas in which further

research and guidance are  needed from the scientific community, both within

and outside EPA.  The following list identifies recurring or continuing

problems that EPA encounters in preparing criteria documents:



      1.  What is a health  effect?  An adverse health effect?

     2.  Can we extrapolate from animal studies to human health effects?
         From clinical  studies, with their limited numbers of subjects,
         to populations?   From high-level, short-term exposures to
         low-level, long-term exposures?

     3.  Can we establish  "no observed effects" levels (NOEL'S)?  Are
         there thresholds? Are there threshold ranges?


                                      43

-------
    CRITERIA DOCUMENT REVIEW AND REVISION      	      Tilton


     4.  Have we overlooked important confounding or intervening
         variables in our epidemiologic studies,  such that the reported
         results are made suspect?

     5.  Have we overlooked important factors in  what may be a
         multifactorial etiology?

     6.  How good are our measurements of ambient air concentrations?  Of
         doses administered?  Of background levels?  Of responses
         elicited?

     7.  Does ambient air monitoring of the pollutant in question
         adequately reflect actual human exposure?  That is, can we
         adequately describe source-receptor relationships?

     8.  What kinds of groups should be excluded  from consideration as
         sensitive population(s)?
     The first question of this list may be unanswerable by scientific

inquiry, since part of the answer may be judgmental.   For example,  the Clean

Air Act itself allows the Administrator of EPA to make certain decisions

according to his "best judgment."  Furthermore, even health researchers may

differ about what constitutes a health effect and particularly about what

constitutes an adverse health effect.  Certain kinds of experiments, however,

can at least provide a partial answer to this question.  Long-term, low-level

exposure experiments plus specific experiments on adaptation may be of value.

Such low-level chronic exposures may also help resolve whether thresholds

exist and whether the NOEL1s we try to identify are real or only a product of

the insensitivity of present test systems.



     Some of the recurring problems listed here arise during derivation of the

standard rather than in preparation of the criteria document itself.  The last

question, for example, is usually treated in the staff paper that is part of


                                      44

-------
5 .  CRITERIA DOCUMENT REVIEW AND REVISION	Tilton


the regulatory package forwarded to the Administrator.  If smokers, for

example, were excluded from consideration as the sensitive population to be

protected by a carbon monoxide primary standard, such an exclusion would be a

policy decision rather than a scientific judgment.  Writers of a criteria

document, however, must try to identify all groups at risk and to provide

scientific documentation that will aid OAQPS and the Administrator in

determining whether a given group should be considered the sensitive group to

be protected, or whether a given group can in fact be protected by an ambient

air standard.



     Finally, broad areas in which ECAO believes research information is

presently inadequate include:
      1.   Human base-line  data  defining what is "normal" for certain
          pulmonary function  tests and biochemical indicators.

      2.   Human health  effects  data  resulting  from long-term, low-level
          exposures to  air pollutants, especially gas-phase pollutants.

      3.   Human studies of good design and  execution, especially
          epidemiologic studies.

      4.   Studies  at the biochemical/mechanistic level for gas-phase
          pollutants.

      5.   Correlation of classic pulmonary  function tests with underlying
          biochemical/physiological  changes.

      6.   Biochemistry, biophysics,  and pharmacokinetics of cardiovascular
          and cardiopulmonary systems in  general, and of the lung in
          particular.
                                       45

-------
5.  CRITERIA DOCUMENT REVIEW AND REVISION	Tilton


     In view of the basic problems and questions remaining, and in view of the

areas in which good scientific data are still lacking, those of us who daily

engage in the writing of air quality criteria and related documents look to

the scientific community for the continuation of high-quality research, both

in the specific areas mentioned above and in the general pathonomy of air

pollution related sickness and disease.



REFERENCES



Clean Air Act.  42 USC 7401 et seq.

Clean Air Act Amendments of 1977.  Public Law 95-95, August 7,  1977.



WORKSHOP COMMENTARY
K. M. Schaich;  You make recommendations which seem quite at variance with
some of the recommendations of [Miller et al. 1979; see Chapter 2 of this
volume for details of reference].  Where is the kind of mechanistic
information that you seek to fund into your program?

B. E. Tilton;  Let me point out that ECAO does not do any funding.  I hope
that question can be addressed by the appropriate committees, but it may be
that the needs I have indicated are not the consensus of others involved.
                                      46

-------
  6.   TOXICOLOGICAL RESEARCH STRATEGIES IN RELATION TO EPA REGULATORY NEEDS

                               Edward L. Alpen

                         Lawrence Berkeley Laboratory
                           University of California
                             Berkeley, CA  94720
INTRODUCTION



     Prior to the Research Planning Workshop on Health Effects of Oxidants,

prospective attendees were sent an EPA staff report known as a "straw man"

document.  This document (Miller et al. 1979) proposes certain revisions in

the EPA-funded research program designated as "Theme 1" ("Health Effects of

Criteria and Non-Criteria Pollutants from Fossil Fuel Combustion").   (For

further information on Theme 1 and the entire Energy Interagency Health and

Ecological Effects Program, see Chapter 2 of this volume.)  As an

"old-fashioned" toxicologist, I felt strong personal dismay in reviewing the

recommendations contained in the straw man document.  In my opinion, the

suggested changes diverge so sharply from toxicological dogma (be it good or

bad) that an examination of their merits is warranted.



     The EPA statements and/or recommendations that I wish to challenge are:
                                      47

-------
6.  TOXICOLOGICAL RESEARCH STRATEGIES	Alpen


     1.  Data from studies involving exposures of >1 ppm ozone (03) or >5
         ppm nitrogen dioxide (NO2) are not significant to EPA.   Sulfur
         dioxide particulates of >25 mg/m3 have no impact and have no
         significance to EPA.

     2.  All experiments should be conducted at ambient levels.

     3.  Negative results are as important to EPA as positive results;
         environmentally relevant exposure regimens should be used.

     4.  In vitro studies have essentially no value to EPA.
COMMENTS ON EPA RECOMMENDATIONS



      In my opinion, to perform studies at ambient levels and to expect other

than  negative data is to concede that control programs have failed.

Obviously, control programs are designed to incorporate margins of safety (see

Chapter 4 of this volume).



      Negative studies, in my view, have very little value except in completing

the bottom ends of dose-response curves that are already well developed.

Standard toxicological practice is to range-find first, and to then work down

to the bottom of the dose-response curve.  As a researcher, I do not pursue

negative studies because their results are trivial to users.



      To me, the fourth statement listed above (regarding _in vitro  studies) is

especially strange.  Clearly, EPA needs in vitro studies in order  to know what

to study in vivoI  Due to  limited resources of time and money, it  is simply

not possible to perform  every study  in the whole animal.  Results  from  the
                                       48

-------
6.  TOXICOLOGICAL RESEARCH STRATEGIES
                                                                         Alpen
relatively fast and inexpensive jji vitro  systems provide  important input in

designing fruitful jLn vivo experimentation.
REFERENCE
Miller, F. J., J. A. Graham, M.  Hazucha, C. Hayes, and W. Riggan.  1979.
     Preliminary Relevance Review of EPA-DOE Projects on Health Effects of
     Criteria and Non-Criteria Pollutants  from Fossil Fuel Combustion (Theme
     1).  Unpublished document,  Health Effects Research Laboratory, Office of
     Research and Development, U.S. Environmental Protection Agency, Research
     Triangle Park, North Carolina, December 12.
WORKSHOP COMMENTARY
J. A. Graham;   I don't  think we're as  far apart as some of your statements
imply.  It seems to me  that we all agree that concentration-response studies
are necessary;  it' s just  a question of the concentration range over which
those studies are  to be performed.  The straw man document [Miller et al.
1979] uses,  I believe,  the wording "near ambient levels."  We agree that it
would be  silly  to  begin a study at an  ambient concentration; rather, one
should start at a  slightly higher concentration and work down.  If somebody
else had  done some research at, say, 0.5 ppm, one might decide to start there.

     The  straw  man offers broad guidelines.  When we speak of starting at a
"high concentration" and  going lower,  we would consider 1 ppm to be a "high
concentration"  for O3.  That is definitely not an ambient concentration; 0.1
ppm would be more  likely  in an urban environment, although higher
concentrations  have been  reported.

     So,  in  my  opinion, we're not so far apart.  It's just that our
definitions  of  "high" may be different.  Our problem with high-concentration
studies also relates to the amount, of  work involved.  In other words, if 100%
of the work  effort for  a  given laboratory is at 20 ppm O3, that represents one
category.  If 5% of the effort is at 2 or 3 ppm 03, that's a different case
altogether.  We're talking about the majority of the research.

     It is indeed  true  that papers on, 10 ppm O3 are not even reviewed for
inclusion in the criteria document.  The reason is that, with the past 03
criteria  document, there  were so many  data available in the literature that it
became necessary to make  a cut somewhere.  The flavor we got from all those
                                       49

-------
6.  TOXICOLOGICAL RESEARCH STRATEGIES	Alpen


papers (and for nitrogen oxides toxicology, as well) is that higher
concentrations (e.g., 20-30 ppm) might even cause different types of effects.
For example, I believe several authors have reported O3 studies in which high
concentrations caused echinosis of the red cells while lower concentrations
caused spherocytosis [see Chapter 11 of this volume].

Comment;  Those "high concentrations" were 2 to 4 ppm, not 20 to 30.

J. A. Graham;  We get in trouble by using words like "high" and "low."  The
studies showed that 4 ppm caused a different type of effect.

E. L. Alpen;  But that's not what you said.  I've been peddling my business to
the government for nearly 30 years, and I've discovered that when somebody who
works for the government writes things down, you're not there to explain
them—

J. A. Graham;  That* s the purpose of this [Research Planning Workshop on
Health Effects of Oxidants].  That's why the front page of the straw man uses
the word "preliminary."

E. L. Alpen;  Toxicology has proven over and over that even very high level
studies are extremely useful in providing insights on mechanisms.  If one
finds high-level studies or dose-dependency curves that go through maxima (and
we know lots of these in toxicology), it is crucial to know what side of the
maximum one is on.

     I would point out one other fact:  Almost every major epidemiologic
study, whether published or in progress, is not at ambient levels.  Such
studies focus on high-risk populations, because that's where investigators
hope to find something.  For example, our group is conducting a hydrocarbon
toxicology study for the Department of Energy.  We're not out looking at
people on the street.  Rather, we're looking at three populations of high-risk
workers:  ramp workers at airports, industrial retail petroleum product
deliverers, and petrochemical refinery workers.  Most epidemiologic studies
are conducted on high-risk populations, because that's where one finds what to
look for in the low-risk groups.  The same generalization applies in
toxicology, too.

J. A. Graham;  We in EPA agree.

E. L. Alpen;  I don't think we agree.

J. A. Graham;  One starts at higher concentrations and works down.

E. L. Alpen;  Your position is:  If a researcher starts at a high
concentration and finds something, he works to lower concentrations, and other
investigators should not return to the higher level.  My position is:  They
should, because the first investigator might have missed a lot!
                                      50

-------
6.  TOXICOLOGICAL RESEARCH STRATEGIES
                                                                         Alpen
J. A. Graham;  It depends on the concentration  you1 re talking about.

E. L. Alpen;  I'll just read your own numbers to you:   "Any results above 1
ppm 03 are of no relevance or  significance  to EPA."

J. A. Graham;  The straw man states that  they don't go  into the criteria
document, not that they are "of no relevance or significance."

E. L. Alpen;  That doesn't mean they're neither relevant nor significant;
they're both, whether  they go  into the criteria document or not.

J. A. Graham;  Here again, we're in terminology.

E. L. Alpen;  Would anybody else like to  say something?

Comment;  As a toxicologist, I understand Dr. Alpen1 s position very well.  But
levels above 1 ppm for O3 and  above, say, 5 ppm for NO2 are so far out of the
potential for human exposure under air pollution circumstances that the data
from such studies are  unlikely to be important  to determining end points of
biological effects.  They may  be important  to understanding mechanisms, but
they1re not likely to  be important to understanding the potential hazard to a
population that is commonly exposed at factors  tenfold below this.

     With respect to the practical issue  of setting standards, I do therefore
believe that the limitations recommended  in the straw man are appropriate
limitations.  They are not limitations on toxicology.

J. A. Graham;  With respect to your statement on in vitro studies, we did not
say that they have "no value," period.  We  said that jLn vitro studies have
value under certain circumstances and that  those circumstances are quite
limited.  The reason is that j.n vitro data  simply cannot be used for
regulatory purposes; they can  be used to  substantiate an in vivo effect, and
one might choose an jln vitro system to help elucidate the mechanism of a
certain effect.  There are thousands of chemicals to be considered, so one
might want to use in vitro screening methods.   There are other cases where in
vitro studies are appropriate, but in vitro data in and of themselves cannot
be used for regulation.  So when a decision is  made to use in vitro exposure
experiments, they must be viewed from that  standpoint.  In vitro studies are
not of general value;  they are of value in  certain circumstances, and those
circumstances need to  be evaluated.

Question;  May I have  some clarification  of the statement, "cannot be used for
regulation?"  For those of us  not familiar  with EPA's procedures, is that an
internal policy or practice of EPA or is  that mandated by some law?  Where
does the "cannot" come from?

J. A. Graham;  Generally speaking, in a criteria document, when a decision is
made on what levels are harmful for man and what types of effects occur in
man, the emphasis is on the concentration-response data.  In other words, one


                                      51

-------
6.  TOXICOLOGICAL RESEARCH STRATEGIES	 Alpen


looks at the human data for 03, let's say, and sees that "X" effects in
pulmonary function are demonstrated at a given concentration.  An in vitro
study cannot relate concentration to the human exposure condition as can a
human clinical study.  When EPA views a human clinical study on 03, we see
what concentrations cause what effects and can then make decisions about
safety factors to be included.  When one is dealing with an in vitro study (no
matter what the pollutant), one cannot really use those data in the regulation
on a microgram per cubic meter basis.
                                      52

-------
           7.  CORRELATING EPIDEMIOLOGIC, CLINICAL, AND BIOLOGICAL
                             RESEARCH ON OXIDANTS

                               David L. Coffin

                  Health Effects Research Laboratory, MD-70
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     Developing pollution standards requires us to ask certain basic

questions.  Typically, the first question is:  "How much is too much?"  The

sequel to that is:  "How much are we willing to accept because of economic and

other factors?"  The first question must be answered by toxicologists, the

second question by politicians and philosophers.



     Another basic question that is certainly important to those who have the

unfortunate and rather unrewarding job of setting standards is:  "By which

discipline or disciplines do we attack the problem?  What can we use to

convince people that this agent is toxic at a certain level or not toxic at a

certain level?"  As a toxicologist, I have always been rather sensitive to

that question.  My work has generally been with mice; when significant results

are obtained, critics sometimes complain:  "Well, a mouse is only that long

and a human being is this tall.  How can they correlate?"


                                      53

-------
7.  CORRELATING LINES OF RESEARCH	   Coffin






     To accumulate data in support of standard-setting activities,  three




general disciplines are available:  epidemiology, clinical experimentation,




and biological experimentation.  Each carries certain advantages and




disadvantages.









     Epidemiologic studies are sometimes appropriate for community problems,




but epidemiology was originally developed and is more appropriate for




relatively simple problems of infectious disease (e.g., when one traces




typhoid to a contaminated well).  Environmental epidemiologic investigations




are plagued with two problems.  First, the dose is very low.  Secondly, there




are innumerable confounding variables.  In short, environmental epidemiology




carries great potential but is difficult to apply in practice.









     Clinical experimentation has the advantage of yielding data that are




directly applicable to human beings.  There may be difficulties in obtaining




dose responses, however, because subjects cannot be exposed to toxicants at




concentrations that are much above ambient levels.  Also, because of the small




number of subjects in any given study, there is a tendency to overlook




individual differences in susceptibility or resistance.  But such individual




differences may be very important, particularly for oxidant air pollutants.









     Biological experimentation offers several advantages:  higher doses,




primary detection of effect, better construction of the dose-response curve,




and exposure of relatively large numbers of individuals of defined genetic




pattern.  Under strictly controlled conditions, it is usually possible to






                                      54

-------
7.  CORRELATING LINES OF RESEARCH	Coffin






determine whether susceptibility or resistance  is. the result of genetic




structure or is acquired.  Of course, the disadvantages are also great.




Extrapolation of the obtained experimental data to human beings may be




difficult because of discrepancies in lung size, pulmonary frequency, etc.




Also, there may be certain biochemical differences between the experimental




animals and humans.









     It appears, therefore, that the ideal strategy would be to apply these




three disciplines in correlative studies that yield unifying answers.  The




problem is how to do this.  How do we detect primary human effect?  How do we




identify susceptible or resistant subgroups in  the human population?  How do




we develop a dose-response curve?  Obviously, dose-response information is




best obtained in animals or other biological systems permitting a sufficient




spread of dose.  If the investigator is limited to ambient or near-ambient




levels, the top dose is so small that there is very little room to work.









PREVIOUS RESEARCH









     In the case of oxidants, few correlative parameters have been




successfully pursued from animal biological through clinical experimental and




epidemiologic studies.   This  is an area deserving more attention.









     Eye irritation is  a primary problem of oxidant air pollution.  This




parameter cannot be studied in  animals; it is,  apparently, a peculiarly human




affair.  When studied experimentally using artificial smog, eye irritation







                                      55

-------
7.  CORRELATING LINES OF RESEARCH	Coffin






does not appear to correlate with ozone level or other obvious measurements of




oxidant pollution.  It is associated, however, with the oxidizing atmosphere.









     Pulmonary function is another parameter that has been examined in humans.




Pulmonary functions are easily obtained with experimental subjects, and




epidemiologic data can be acquired.  Also,  some information is available  on




individual differences.  For example, in Bates and Hackney's switch




experiments, subjects from Canada were found to be more susceptible to oxidant




exposure as measured by pulmonary function than subjects living in Los




Angeles.  Such findings probably reflect some sort of acquired resistance.









     A number of parameters measurable in animals may be applicable to human




beings.  Pulmonary pathology is one of the most sensitive and relatively




successful parameters in animals.  Oxidant exposure results in definite




disease of the small airway.  When animals are exposed to appreciable amounts,




the resultant lung changes, which include loss of recoil, appear to be rather




profound.  It appears that no one has attempted comparable measurements in




humans; perhaps larger doses would be required.  Positive results would




suggest that oxidants contribute to pulmonary emphysema in man; already there




may be some evidence of this.









     Another successful animal parameter is activation £f infection.  Several




studies have shown that animals react to both nitrogen dioxide and ozone at




relatively low levels and become more susceptible to bacterial infection of




the deep lung, probably as a result of oxidant action on the macrophages and







                                      56

-------
7 .  CORRELATING LINES OF RESEARCH	Coffin






other lung defense systems.  Thus, activation of infection would appear to be




a highly appropriate model with which to "crosswalk" to humans.  With humans




there are certain problems, however.  First, bacterial pneumonia has become an




uncommon disease in man.  When it does occur, the disease is frequently




secondary to viral infection.  Insofar as certain viral infections affect the




macrophage system as profoundly as (or more profoundly than) oxidant exposure,




human data might be difficult to interpret.  Secondly, epidemiologic data for




Los Angeles show no correlation between the oxidant season and the "pneumonia




season"  (if there is such a thing).  Perhaps some other area (e.g., Phoenix or




Houston) would  show a better correlation.  Finally, as mentioned earlier,




there are problems of species differences and human individual differences.









PROMISING AVENUES FOR FUTURE RESEARCH









     One goal for future epidemiologic studies might be to screen subjects to




find a  susceptible subgroup.  Such screening could be based, perhaps, on a




correlation between individuals who  are exceedingly susceptible to eye




irritation and  those who demonstrate lung-reactivity.  Experimental work in




Cincinnati showed that  some individuals are highly susceptible to eye




irritation while others are positively resistant to it.  There may be some




theoretical objections  to using this parameter as a screen:  the eye




irritation is probably  due to other  agents within photochemical smog (rather




than the oxidant itself).  But, as noted earlier, eye irritation does




correlate roughly with  the oxidizing atmosphere.
                                       57

-------
7.  CORRELATING LINES OF RESEARCH	Coffin






     Another potentially fruitful line of research would be to investigate




whether human "acquired resistance" to oxidants resembles the "tolerance" seen




in animals.  Over 20 years ago, Stokinger and others showed that low-dose




exposures to one gas will protect animals against exposures to large doses and




even against exposures to other gases (e.g., phosgene will protect against




ozone, or vice versa).  In our laboratory, we showed that such results are




associated with protection or inhibition of edema.  We postulated that this is




a laboratory artifact that probably does not occur in nature:  the




"edemagenic" exposures do not obtain in ambient air, and our cellular studies




did not reflect any tolerance phenomenon.  Still, clinical experiments have




shown that humans do acquire some sort of pulmonary functional resistance to




oxidants, and this remains a potentially productive line of inquiry.









     One of the obvious places to look for correlations would be in pulmonary




function studies.  Involved here are noninvasive tests that can be applied




across the board in epidemiology, clinical experimentation, and animal




experimentation.  We already know, of course, that the human response to




oxidants entails such physiological responses as constricted airway.  The




question is:  Are these responses significant—are they health-impairing—or




do they represent a physiological response that is easily accommodated?  The




other question is:  Does a 4-h acute exposure in animals result in a




measurable pulmonary physiological response?  In studies performed in the




early 1960's, Murphy demonstrated certain changes, including increased




frequency of respiration.  But he could not convincingly demonstrate any




changes in pulmonary resistance, and subsequent studies have been equally






                                      58

-------
7.  CORRELATING LINES OF RESEARCH	Coffin






unsuccessful.  Perhaps the problem  is one of primitive techniques:




anesthetizing the animals may wipe  out these minor changes.  To effectively




employ pulmonary resistance as a "crosswalk" between animals and humans, then,




will likely require us to improve our techniques  in animals.  There are




techniques under development that do not require  anesthetization of the




animal; this may enhance the response.









CONCLUDING REMARKS









     In conclusion,  I would  suggest that the oxidants problem be reassessed as




a  "new" problem.  Are oxidants a problem?   Are  the present standards adequate?




Rather than reapplying  techniques and parameters  that have been used in the




past, investigators  should devote more effort to  innovation.  We need to




develop parameters that provide  "crosswalks" from species to species.  Only in




this way  can we develop a coherent  body of  knowledge with which to approach




standard-setting.
                                       59

-------
                 8.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
                      AT BROOKHAVEN NATIONAL LABORATORY

                                Robert T. Drew

                              Medical Department
                        Brookhaven National Laboratory
                        Associated Universities, Inc.
                               Upton, NY  11973
LIFE SCIENCES RESEARCH AT BROOKHAVEN NATIONAL LABORATORY



     Brookhaven National Laboratory is located approximately 70  miles east of

New York City in the geographic center of Long Island.   While the major focus

at Brookhaven has been in high energy physics, there are a number of

departments which conduct life sciences research.   The  Safety and

Environmental Protection Division devotes a portion of  its efforts to studying

health effects of radiation.  The relatively new Department of Energy and

Environment (DEE) has several programs associated with  health effects

research.  One group is assessing health effects of population exposures at

the national level and of a variety of energy alternatives.  A second group in

DEE is developing the ability to assess mutagenic and carcinogenic effects of

chemicals using short-term in vitro systems.  Within the Biology Department,

the major focus, of course, is on biological research of all kinds.  One

effort that relates to oxidants is a program developed by the late Dr. Arnold

Sparrow to investigate the mutagenic effects of air pollutants on the higher


                                      60

-------
8.  BROOKHAVEN NATIONAL LABORATORY	Drew






Plant Tradescantia.  This  study  is  underway; trailers have been set up within




and adjacent to industrial complexes  in  the Northeast.









BROOKHAVEN INHALATION TOXICOLOGY FACILITY









     Inhalation toxicology is  a  fairly new effort within the Medical




Department.  Before  this author  joined Brookhaven in 1976, the only oxidants




research being conducted was that of  Drs.  Schaich and Borg (see Chapter 11 of




this volume) .  The majority of my time at  Brookhaven has been spent in




developing a laboratory for exposure  of  rodents to any compound, regardless of




the hazard associated with that  compound.  This facility, which is now




operational, is described  below.









Facility Layout








     The first two figures illustrate the  layout of the new Brookhaven




Inhalation Toxicology Facility.   The  inset in  Figure 8-1 shows the conceptual




layout  of the space. On the right-hand  side is nonregulated space; on the




left-hand-side, regulated  space. The chambers and glove boxes, indicated by




"C" in  the inset, are maintained within  the regulated space.  The air handling




system  for the nonregulated space is  a conventional system with common supply




and local exhaust ventilation.  The air  handling system for the regulated




space is completely  separate and independent from that of the nonregulated




space and has a common exhaust system providing several degrees of filtration




for exiting air.  A  third  separate  air handling system is provided for the






                                       61

-------
                                                                         00
                                V

     V
        	1
to
    i	i
 REGULATED
AIR SUPPLY
                   NON-REGULATED
                      AIR SUPPLY
                                V
c

R
r
NR
                                                                         H
                                                                         i
   Figure 8-1.  Outline of Brookhaven Inhalation Toxicology Facility.
            handling systems.
                                Inset:  Separation of air
o
H

-------
8.  BROOKHAVEN NATIONAL LABORATORY	Drew






inhalation chambers and glove boxes.   The air systems  for  the regulated space




and the chambers are  totally redundant.   In  operation,  the regulated space is




maintained at ~0.08 inH^O pressure less  than the nonregulated space.  Chambers




are operated at -0.5  to -1.0 inH2O with  respect to  the  regulated space.









     The actual layout of the building is shown in  Figure  8-2.  Within the




nonregulated space are the  electron microscopy suite, two  conventional




laboratories, two  offices,  and a glass washing area.  The  regulated space




consists of two large chamber rooms separated by a  bank of small laboratories;




total  separation between the two chamber rooms is accomplished by closing one




door.  We  intend to use Chamber Room  A for highly hazardous materials and




Chamber Room B for more conventional  compounds.  Two small animal rooms are




adjacent to Chamber Room B.  Access to the regulated space is through three




small  vestibules,  indicated in Figure.8-1, which are ventilated at ~1 air




change/min with clean, fresh air. Two large viewing windows allow observation




of operations behind  the barrier (Figure 8-3).  Thus, visitors can view the




exposure facilities without entering  the regulated  space.   Catwalks over the




inhalation chambers provide space for contaminant generation and for chamber




monitoring (Figure 8-4).








      Figure 8-5  is a  close-up of one  of  the  5-ft inhalation chambers.  The




chambers are  installed in  pits and arranged  so that a  rack of animals can be




wheeled in.   Each  rack houses 12 cage units, each capable  of holding 8 rats in




individual cages.  Thus, we can expose up to 96 rats in 1  rack, or a total of




192 rats in  1  inhalation chamber.  The cages are capable of housing either 1






                                       63

-------
8.  BROOKHAVEN NATIONAL LABORATORY
Drew
u
I C
-*
INUALATI0N -TOXICOLOGY FACILITY
SCALt- J4"« t'-Q*




, a
=r c=r
t
=]
    Figure 8-2.  Floor plan of Brookhaven Inhalation Toxicology Facility.
                                      64

-------
                                                                                    00
                                                                                    H
                                                                                    s
                                                                                    «M>

                                                                                    1


                                                                                    I
Figure 8-3.  Chamber Room B  (through  the  viewing window).

-------
8.  BROOKHAVEN NATIONAL LABORATORY
                                                                          Drew

  Figure 8-4.  Generation and monitoring equipment located above the chambers.
                                       66

-------
Figure 8-5.  Close-up of caging unit.
i

-------
8.  BROOKHAVEN NATIONAL LABORATORY	    Drew






rat, 2 hamsters, or 3 mice; thus, this design is capable of exposing up to 576




mice in 1 system.









     These facilities came on-line in December 1979 and are now in operation.




Chamber Room A will be used to expose animals to highly hazardous compounds,




such as polycyclic aromatic hydrocarbons and other known carcinogens.  For




these compounds, we intend to employ chambers that will house and expose




animals in a totally attached unit.  Design of these chambers^ is underway.









Instrumentation and Techniques








     Our  goal is developing this research effort was to build a team capable




of  measuring functional or physiological changes, anatomical changes, and




biochemical changes in small animals.  To that end, we recruited a small




animal respiratory physiologist. Dr. Daniel Costa, who trained under Dr. Mary




Amdur.  He has purchased and assembled equipment necessary for measuring the




physiological status of the rat pulmonary system in terms of spirometry and




mechanical function.  We are using plethysmography and gas dilution techniques




We  believe these methods represent the state of the art.









     We will employ a recently acquired scanning electron microscope to assess




the anatomical structure1 of lungs of animals exposed to a variety of airborne




materials.  Conventional pathologic evaluation will be included, and we intend




to  develop morphometric techniques at the level of the light microscope.   In




addition  to histopathologic evaluation, we will also measure a number  of  lung






                                      68

-------
8.  BROOKHAVEN NATIONAL LABORATORY	Drew






biochemical indices which  indicate  response  to  challenges of environmental




agents.









Current and Planned Studies









     Projects that are underway include:   a  study of the deposition and




translocation of sized glass fibers,  supported  in part by the Thermal




Insulation Manufacturers Association;  a comparison of functional change versus




anatomical change after  exposure of rodents  to  a series of known pulmonary




toxicants (ozone was  one of the first compounds chosen as a model pulmonary




toxicant); a  study of the  interaction between hypertension and air pollution




 (see Chapter  10 of this  volume); the use of  bleomycin to produce a model of




 fibrosis in  the rat;  and exposure of animals to coal dust in order to develop




a rodent model of coal workers' pneumoconiosis.








      Since epidemiologic evidence shows 'that people who are already sick are




 at a much greater risk during episodes of high air pollution, we intend to




 develop animal models of various aspects of chronic lung  disease.  Having




 developed these models,  we hope to then superimpose exposure to some of the




 common air pollutants (such as nitrogen dioxide and ozone)  in  an attempt to




 understand why already-stressed persons are at a greater  risk  during episodes




 of high air pollution.
                                       69

-------
  9.  INCORPORATING OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS

                               Samuel C. Morris

               Biomedical and Environmental Assessment Division
                        Brookhaven National Laboratory
                        Associated Universities, Inc.
                               Upton, NY  11973
THE NATURE OF A HEALTH EFFECTS ASSESSMENT



     A health effects assessment is a documented, quantitative description of

knowledge and uncertainty regarding the potential health impact of a program

or policy action.  Such assessments are useful in evaluating or formulating

research and development in both the energy and environmental areas.



     An assessment considers the potential health implications of an installed

future energy industry, for example, based on current knowledge of the

technology, its environmental residuals, and effects on workers and the

general population.  Computer modeling and simulation studies form the

framework of such an evaluation.  Figure 9-1 is the schematic outline of a

model for air pollution effects.  One or more scenarios are developed, plants

are sized and sited, and other activities (e.g., automobile traffic) are

distributed geographically.  Based on technological characterizations,

emission estimates are made.  Air transport, dispersion, and chemical modeling
                                      70

-------
9.  OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS
                                                              Morris
                      Scenario Development
                       Population Exposure
                            Estimates
                                                Population
                                                Projections
                                                   Dose-Response
                                                     Functions
                                          Health Effect Estimates
                                               Health Costs
 Figure 9-1.
Model  for assessing  the health effects of  energy-related air

pollution.
                                          71

-------
9.  OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS	Morris






are performed.  The results of these activities are combined with population




projections to estimate population exposure.  Such factors as population




time-activity patterns are often included.  In some cases, detailed lung and




metabolic models are used to translate exposure into dose.  Using




dose-response or health damage functions, the population exposure or dose




estimates are used to estimate health damage.  In some instances, health




damage is translated into monetary terms.









SPECIAL CHALLENGES OF ENERGY-RELATED ASSESSMENTS








     One major problem in assessing energy-related health effects is our




near-total lack of health damage functions for the kinds of low-level




exposures expected from most new energy developments.  We are dealing in an




area of great uncertainty.  The uncertainty is compounded when we project




emission characteristics of a new technology for which only pilot plant




measurements may exist.  In fact, the degree of uncertainty may itself be a




more important consideration than our best estimate of effect*









     To the extent possible, we must deal explicitly with this uncertainty.




One way is to express parameters as probability density functions (pdf's)




rather than as "best estimates."  As an example, Figure 9-2 displays the




probability that one particular coefficient will fall in a given range.  It




takes some judgment to come up with such density functions, but they are




important when one needs to combine several uncertain independent parameters.




It would be very unlikely that all would be at the high (or low) end of the






                                      72

-------
9.  OXIDANTS IN ASSESSMENTS  OF ENERGY-RELATED HEALTH EFFECTS
Morris
                                    PERCENT/HOUR


                  Figure 9-2.  A probability density function.
                 0.00
                    °
                         POPULATION EXPOSURE, 106 PERSON -wi/m*
 Figure 9-3.   Result of combination of probability density '«"<*»• *>

               characterize total population exposure (after Morgan et al.  1978),
                                        73

-------
9.  OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS	  Morris






range at once.  Figure 9-3 illustrates the result of combining several pdf's




to yield the likelihood of various population exposure levels.









     Another way to use assessment models is through sensitivity analysis,




where we test the effects of various hypotheses of environmental transport,




dose-response functions, etc. on the final estimate of damage.  This can be




particularly useful in finding the importance of one line of research over




another.









     Ideally, we would prefer a complete dose-response function for man in the




range of exposure.  Certainly this is what investigators should aim for.




Realistically, we settle for much less, and it becomes important to draw




information from many sources to form a hypothesis of the function.  High-dose




information helps define the shape of the curve and identify mechanisms.  In




vitro studies may also provide clues to mechanisms.  In addition to helping




guide the design of ambient level studies, these data lead to hypotheses that




can be directly applied in assessment models.









INCORPORATING THE CONSIDERATION OF OXIDANTS









     With regard to oxidants, what has been accomplished in energy-related




health assessments?  Very little.  There have been a number of assessments of




the health effects of various energy systems:
                                      74

-------
9.  OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS            Morris
     An Assessment of National Consequences of Increased Coal Utilization
       (U.S. Department of Energy 1979a)

     Regional Issue Identification and Assessment (U.S. Department of
       Energy 1979b)

     Energy in Transition, 1985-2010 (National Research Council 1980)

     The Environmental Impacts of Production and Use of Energy (Part I,
       Fossil Fuels) (United Nations Environment Programme 1979)
None of these includes quantitative assessments of the health effects of

oxidants.  Estimates have generally been based on sulfur and particulates;

this is a major shortcoming.  For example, in the first study listed above,

increasing emission controls were projected to lead to decreasing sulfur

dioxide and sulfate levels despite a considerable increase in coal

consumption.  Nitrogen oxide emissions and presumably ozone levels, however,

would increase.  These were not included for two reasons:  First, a credible

atmospheric model having a large regional basis is not available.  Thus, it is

impossible to make estimates of population exposure.  Second, no acceptable

health damage function is available.  The first problem is expected to be

resolved by 1982; the latter is of more interest here.



     There have been several difficulties in developing an oxidant health

damage function.  Epidemiologic studies are generally negative or

inconclusive.  Clinical and animal studies show strong effects, but these

effects generally involve subclinical physiological factors.  Changes in lung

compliance, pulmonary function, and blood chemistry are difficult to relate

directly to morbidity and mortality.  We have designated substantive disease
                                      75

-------
9.  OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS	Morris






or death as the only end points qualifying for consideration.  This stems, in




part, from the fact that the health effects documented in our assessments are




in some sense "weighed" (by others) against the cost of controls.  We do not




feel likely to convince anyone to enter into a multi billion dollar control




program on the basis of predictions of subclinical effects.  Even large




numbers of headaches and respiratory symptoms do not constitute a strong




counterweight•








RESEARCH NEEDS








     What can be done?  First, we need more information on the contributions




of oxidants to chronic disease.  As we all know, there are some suggestions of




chronic respiratory disease.  One way to attack this problem is through animal




models of chronic disease*  Another avenue is further epidemiology.  Still




another possibility is the indirect route of linking changes in pulmonary




function, changes in blood chemistry, biochemical effects, and minor




respiratory symptoms with chronic disease.









     Secondly, we can form hypotheses on the relation of oxidants to chronic




respiratory disease and test the implications of those hypotheses in




assessment models.  Sensitivity testing can provide some notion of the degree




to which changes in the hypotheses affect predictions of overall effect.  It




appears that we will have to build models that are much more specific about




sensitive subpopulations, physical activity, and (perhaps) change of exposure




level with time.






                                      76

-------
9.  OXIDANTS IN ASSESSMENTS OF ENERGY-RELATED HEALTH EFFECTS            Morris
ACKNOWLEDGMENT
     This research was supported by the Health and Environmental Risk Analysis
program, Office of Health and Environmental Research, U.S. Department of
Energy.
REFERENCES
Morgan, M. G., S. C. Morris, A. K. Meier, and D. L. Shank.  1978.  A
     probabilistic methodology for estimating air pollution health effects
     from coal-fired power plants.  Energy Systems and Policy, 2:278-309.

National Research Council (Committee on Nuclear and Alternative Energy
     Systems).   1980.  Energy in Transition, 1985-2010.  W. H. Freeman, San
     Francisco.

United Nations Environment Programme.  1979.  The Environmental Impacts of
     Production  and Use of Energy.  Part I.  Fossil Fuels.  United Nations
     Environment Programme, Nairobi, Kenya.

U.S. Department  of Energy (Office of Technology Impacts, Division of
     Technology  Assessments).  1979a.  An Assessment of National Consequences
     of Increased Coal Utilization.  TID-2945, Vol. 2.  U.S. Department of
     Energy, Washington, D.C.

U.S. Department  of Energy (Office of Technology Impacts, Regional Assessment
     Division).  1979b.  Regional Issue Identification and Assessment.  First
     Annual Report.  U.S. Department of Energy, Washington, D.C.
                                      77

-------
       10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANT AIR POLLUTANTS

                                Robert T. Drew
                               Daniel L. Costa
                                 Sonja Haber
                                 Junichi Iwai

                              Medical Department
                        Brookhaven National Laboratory
                        Associated Universities, Inc.
                               Upton, NY  11973
INTRODUCTION



     In the 1960's when the word "relevance" was not so important, one very

interesting project at Brookhaven National Laboratory was an investigation of

the causes of hypertension.  By developing an animal model of hypertension,

the late Dr. Lewis K. Dahl proved that both genetic and environmental factors

are involved in the development of hypertension.  Over the ensuing 15 years or

so at Brookhaven, we have refined this model, selectively breeding two lines

of rats from the same initial breeding stock.  One of the lines is susceptible

(S) to salt-induced hypertension, and the other is resistant (R) to

salt-induced hypertension (Figure 10-1).  We have used this model to

investigate interactions between commonly encountered air pollutants and the

development of hypertension.  Since the experiment is loaded in favor of

producing a positive effect, a negative finding (i.e., one in which no

increased effects occur subsequent to a significant challenge by the pollutant


                                      78

-------
10.   INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS
Drew et al.
in question) can be interpreted favorably.  This paper summarizes our studies

investigating the interaction of two air pollutants, sulfur dioxide (802) and

ozone (03), with the hypertensive model.
UJ
a:
5> 200
UJ °>
(Y ~^~
f\ P
| 160
Q C
0
0
m 120
1 1 1 1 1 1 1 1

m " • s-
_ • s-

D R -
OR-
_ * • *
-8 9 o °
0 o
1 1 1 1 1 1 1 1
1 1

HIGH -
LOW _

HIGH
LOW ~~
—
1 1
                         10        20       30       40       50
                                   WEEKS ON DIET

                            MALE RATS ON 4%  SALT
 Figure 10-1.   Blood pressure of S and R male rats on high- and low-salt diets
              as a function of time.
METHODS




     Both studies were designed as 2 x 2 x 2 matrix studies with 3 variables:

air vs. pollutant, high dietary salt vs. low dietary salt, and S vs. R line of

rats.  This design required 8  groups of 10 rats each.
                                    79

-------
10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS	Drew et al.






Sulfur Dioxide Study









     For studies with SC>2, the high-salt diet consisted of 4% salt; the




low-salt diet, 0.4% salt.  Forty male S rats and 40 male R rats were weaned at




21 d, arranged in groups of 10 animals each, and immediately established on




either the high- or low-salt diet.









     Exposures to S(>2 started the following week.  All exposures were done in




27- x 27-in stainless steel and glass chambers.  The animals were exposed to




SO2 at 50 ppm for 6 h/d, 5 d/week, for 31 weeks.  Control animals were exposed




to air for the same 6-h regimen.








     Blood pressures were measured under ether anesthesia using a tail cuff




method after the fourth week of exposure and on alternate weeks thereafter.




Animals were allowed food and water ad libitum (except during exposure) and




were housed under a 12-h on/off light cycle.  The measurements were made




immediately after exposure to SC>2 on Thursday and Friday.









Ozone Study









     Ozone exposures were carried out in similar chambers at a concentration




of 2 ppm for 6 h/d, 5 d/week, for 20 weeks.  In this study, however, female




rather than male rats were used.  The animals were allowed 1 week from the




time of weaning until establishment on the low- or high-salt diet.  The




high-salt diet was 8% salt (rather than 4% salt as in the S(>2 experiment).  In






                                      80

-------
10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS	Drew et al.






addition, because of the limited availability of sibling animals, each group




was divided into subgroups of 5 animals.  Thus, 20 S females and 20 R females




were weaned on 2 consecutive weeks prior to random assignment to 1 of the 8




experimental groups.  Exposures of the first sibling group (half of the




animals) began 1 week prior to exposures of the second group.  Subsequently,




all animals were exposed simultaneously.  Blood pressure was measured after




the fourth week of exposure and on alternate weeks thereafter.  Data were




accumulated on the basis of exposure duration.









RESULTS AND DISCUSSION









Sulfur Dioxide Study









     Figures 10-2 and 10-3 show blood pressure as a function of time for SC>2




animals on low and high dietary salt, respectively.  The figures clearly show




that the initial blood pressures in S animals were markedly higher than in R




animals.  It is also evident (Figure 10-2) that there were minimal differences




over the course of the experiment between S(>2- and air-exposed rats fed




low-salt diets.  The mean blood pressure of S animals maintained on high-salt




diets increased as expected (Figure 10-3).  In this case, S02-exposed animals




had blood pressures higher than those of air-exposed counterparts.  This




difference was statistically significant at certain time points while not




significant at others.  However, all differences disappeared after the last




exposure to
                                       81

-------
CO
to
-,200
      ,
UJ
ct:
z>
C/)
CO
UJ
       180
       160
       140
    Q
    O
    q 120
    GO
       100
        S

        R
LOW SALT
                                                   LAST S02
                                                  EXPOSURE
                                                 AIR
 SEM FOR ALL ANIMALS  -1-3
   i	I	i	I	i
                       10     14     18   22    26
                                 TIME (weeks)
                                                 30    34
  Figure 10-2. Blood pressure of S and R male rats exposed to air or S02 and maintained on low
           dietary salt.
3
H

CO
0)
                                                 W



                                                 co
                                                                  O
                                                                  x
                                                                  H
                                                                 CO
                                                                 (D
                                                                 ft

                                                                 Ql

-------
00
OJ
        220-
      E200-
      E
LU
o:
^
c/)
a.

o
o
o
«j
CD
         180
         160
         140
         120
         100
              R
                        1	1	T

                         HIGH  SALT
                                             —\	r
                                              LAST  S02

                                             EXPOSURE

                                                  *
                        S02
               SEM FOR R RATS 1-3
         46     10    14    18    22    26

                            TIME (weeks)
                                                      30    34
   Figure 10-3. Blood pressure of S and R male rats exposed to air or S0_ and maintained on high

            dietary salt.
                                                                        H



                                                                        CO
                                                                        CO
                                                                  H
                                                                  O
                                                                        O
                                                                        H
                                                                        (D
                                                                        ft

-------
10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS	Drew et al.






     In these studies, the percentages of dietary salt and the concentration




of SC>2 were chosen in order to severely challenge the animals.  Based on




previous studies, it was anticipated that the S animals given high dietary




salt would succumb to hypertension in 4 to 6 mo.  This did not happen.  Only 4




S rats on high salt, 2 from the exposed group and 2 from the control group,




died during the SC>2 study.  Furthermore, no differences in growth rates were




observed between any of the groups in question.  Although the exposed S rats




on high-salt diets always had slightly higher blood pressures, any apparent




relationship between SC>2 and hypertension is tenuous.  Thus, we conclude from




these studies that exposure to 50 ppm SC>2 does not alter the development of




salt-induced hypertension.








Ozone Study








     The results of our 03 exposures are displayed in Figures 10-4 and 10-5,




which plot blood pressure as a function of time for the animals on low and




high dietary salt-f—respectively.  It is apparent from these figures that the




Oj-exposed animals in both diet groups had lower blood pressures than their




air-exposed counterparts.  This difference is significant at the probability




level of 0.05.









     Of greater interest is the effect of 03 on mortality (Figure  10-6).  The




level of dietary salt was sufficient to cause a 50% mortality in the  S animals




exposed to air on high dietary salt; no other air-exposed animals  died.  The




03 challenge (2 ppm for 6 h/d, 5 d/week) was sufficient to cause 3 and 2






                                      84

-------
00
       160
     o»
     X
       140
UJ
a: 120
ID
en
CO
LJ 100
cr
a.

a   80

o
_J

m   60
                                                   S-AIR-LOW
                                   S-03-LOW
                                                  ^R-AIR-LOW
                                                     R-03-LOW _
                                  1
1
                2   4    6   8    10   12   14   16   18   20  22
                            WEEKS OF EXPOSURE
   Figure 10-4.  Blood pressure of S and R female rats exposed to air or O_ and maintained on low
            dietary salt.
                                §
                                TO
                                td
                                3

                                i
                                D

                                0>

-------
                                                                   H
    160
    140
 c/)
 UJIOO
 a:
 a.

 §  80
 o
     60
                         S-0-3-HIGH
                             *j
                                       1
                                               S-AIR-HIGH
R-AIR-HIGH _
R-0* -HIGH
i
i
                     68   10   12   14   16   18   20  22
                        WEEKS  OF  EXPOSURE
                   H

                   1

                   W
                   CO
                   H

                   §
                   8
                   CO
Figure 10-5. Blood pressure of S and R female rats exposed to air or 0  and maintained on high
         dietary salt.

-------
CO
                                                  XR-03-HIGH
                      S-0,-HIGH
                                                     18  20  22
                      WEEKS OF03  EXPOSURE


Figure 10-6.  Mortality of S and R female rats exposed to air or 0 and maintained on high or low
          dietary salt.
                                                                          CO

                                                                          u
                                                                          M
                                                                          CO
                                                                          H

                                                                          §
                                                                          H
                                                                          O
                                                                          0>


                                                                          PI

-------
10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS	Drew et al.


deaths in the R rats on high or low salt, respectively.   In contrast, all S

animals died within 16 weeks of exposure to 03,  regardless of the presence or

absence of salt in the diet.  These deaths appear to be  unrelated to

hypertension.



     What we may have observed is a susceptibility to 03 that is independent

of dietary salt and blood pressure.  We are excited about these results, since

this difference in 03 susceptibility with regard to S vs. R animals may be

very useful in understanding how the lung interacts with 03 from a

biochemical/physiological standpoint.  There is, however, the possibility of a

general difference in susceptibility to all stresses.  If the susceptibility

is unique to 03, this model will allow us to evaluate the biochemical changes

resulting from 03 exposure and perhaps to intervene with various treatment

regimes to ameliorate the response to 03.



WORKSHOP COMMENTARY



Question;  Was this systolic pressure?

R. T. Drew;  Yes, we measured systolic blood pressure using a tail cuff
method.

Question;  Was there any change in diastolic pressure?

R. T. Drew;  We cannot measure diastolic pressure with this technique.

Question;  Were these changes in the pulmonary epithelium as a result of the
high salt in the diet?

R. T. Drew;  Not to my knowledge, but we have not yet systematically evaluated
all of these studies.  The tissue has been prepared and is being evaluated at
the present time.  I'm not aware of anything in the past literature or  in any
                                      88

-------
10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS                Drew et al.
of Dahl1s previous work that suggests that there would be differences in the
respiratory epithelium.

J. D. Hackney;  Is it possible that these animals are susceptible to any kind
of stress?

R. T. Drew;  Yes, it is possible.  This is one of the reasons that I'm a
little reserved when I say that we may have a biochemical model; there is a
suggestion that S rats are generally susceptible to stress.  I don't know
exactly what this means.

Question;  Did (^-exposed animals lose more weight?

R. T. Drew;  In animals exposed to SO2, there were no differences.  The week
that the exposure started, the SO2-exposed group did not gain weight.  From
then on, all groups gained at comparable rates.  The S02~exposed group never
caught up and was always a little bit lighter.  With O3, there were some
differences at the start of the study.  This is unfortunate, but when you're
working with a limited number of animals you really can1t select for uniform
weight as with a larger number of animals.  However, in every case, O3-exposed
animals weighed less than their air-exposed counterparts.

Question;  Did you identify the factor causing hypertension?

R. T. Drew;  We believe it is a genetic factor.

Question;  Is it a kidney or a vascular problem?  What* s causing the animals
to be more susceptible to high salt?

K. M. Schaich;  There have been very few studies to characterize these rats in
terms of physiology and biochemistry.  There were some early studies in which
hormonal effects were sought, but none were observed that could cause the
increased blood pressure.  Presently, we are working on biochemical
characterization of these rats, but we don't yet know how they differ.  There
is some evidence that liver function may be different in the S vs. R rats, but
we don't have detailed information yet.

Question;  Is there any difference in the pathology of these animals?  Do they
have more extensive lesions?  Do the lesions appear earlier?  Obviously at 2
ppm the animals will develop lesions.

R. T. Drew;  Dr. Slatkin, a colleague, is working on this evaluation now.  I
can't answer the question.  The animals that died in the 03 study clearly died
a respiratory death.  It was not a hypertensive death; on gross pathology, the
lungs were deeply involved (as one would expect).  The micropathology is yet
to come.
                                       89

-------
10.  INTERACTIONS BETWEEN HYPERTENSION AND OXIDANTS                Drew et al.
     We've done respiratory function studies on the R animals in this group
(there are no S animals left).  Our results appear consistent with small
airway disease.

Comment;  With that concentration, you might be getting some effect on the
converting enzyme.  Have you looked at that?

R. T. Drew;  No, we haven't.  This is an exciting observation and we will
certainly pursue it.

Comment;  I don't know the data on ordinary species of rats.  Is it possible
that you have a very resistant strain as opposed to a sensitive one?  Was the
difference because one strain is extraordinarily resistant to this very high
level of 03?

R. T. Drew;  I don't believe so.  I don't believe my data on 03 toxicity are
in conflict with the literature on mortality alone.

Comment;  That's what I'm asking.  What is the experience with other strains?

R. T. Drew;  Well, I'm presently exposing Fisher 344 rats to 03 for the
National Toxicology Program.  We are exposing animals toO, 0.2, 0.8, and 2
ppm 03 and are beginning to see deaths in the 2 ppm group 8-9 weeks after
beginning the exposures.  So, the results are consistent.

Question;  Are you saying that your S rat is more susceptible to the same
doses of 03 than other strains of rats?

R. T. Drew:  Yes, I think that is correct.
                                      90

-------
     11.  THE EFFECTS OF OZONE ON RAT ERYTHROCYTES AFTER EXPOSURE IN VIVO

                               Karen M. Schaich
                                Donald C. Borg

                              Medical Department
                        Brookhaven National Laboratory
                        Associated Universities, Inc.
                               Upton, NY  11973
INTRODUCTION



     For some time, concern has been expressed that toxic effects of ozone

(03) may not be limited to the lung but may extend to other body tissues.

Most biochemical investigations of systemic toxicity have focused on

erythrocytes:  the cells which have first contact with the exposed lung and

which, along with plasma, presumably will be the carriers of cytotoxic

potential to other sites.  Observations from a large number of studies in

several species have been published (Table 11-1), but the issue of

extrapulmonary effects of 03 reflected in red blood cells is anything but

clear.  As can be seen in Table 11-1, a variety of end points have been

monitored on blood, including hematological/morphological, physiological, and

biochemical/metabolic.  There has been little effort, however, to integrate

different types of effects into an overall pattern of response.  In terms of

cellular biochemistry, most statistically significant changes have been noted

only when rats were made tocopherol (vitamin E) deficient before exposure or


                                     91

-------
TABLE 11-1.
Species
rat





rat
Sprague-Dawley
male
3 mo
± 45 ppm tocopherol


mouse
Caesarian-del ivered- 1
male
adult
vo
to mouse

rat
Spr agvie-Dawl ey
male
mouse


rat
Sprague-Dawley
male
5 mo
± 45 ppm tocopherol
rat
Spr ague-Da wl ey
male
2 mo



EFFECTS IN RED BLOOD CELLS FOLLOWING IN VIVO EXPOSURE TO OZONE
Exposure Conditions Observationsa Reference
1.5 ppm x 3 d no effect SOD, GPx, K+ flux Larkin et al. 1978
6 ppm x 4 h tretics. (all levels)
8 ppm x 4 h tHb, Hct, echinocytes II s III (6 & 8 ppm)
( echinocytes correlated with petechia in
lungs, indicative of vascular endothelial
damage)
0.8 ppm continuous x 7 d -Vit. E: (-KGSH, 4/GPx, tPK Chow and Kaneko 1979
+O3 : *GSH, tGPx, tPK
tLDH, no effect SOD,
catalase, TBA, G6PD
MetHb, reticulocytes
+(Vit. E + 03): no 03 effects

0.85 ppm x 4 h tHeinz bodies Menzel et al. 1975
(4/with continued exposure)



8 ppm x 4 h +AChE, (+JGSH, tHb Goldstein et al. 1968

5 ppm x 1.5 h tH2O2, ~50% 4- catalase Goldstein 1973


6.7 ppm x 1.5 h tH2O2, ~16% 4- catalase Goldstein 1973
1.7 ppm x 1 . 5 h no effects
1.0 ppm x 4 h no effects
0.8 ppm x 7 d +O3 : tGPx, 4-GSH (both groups) Chow et al. 1976
-Vit. E: 4-GPx, GSH
tG6PD & 2,3-DPG


0.5 ppm x 8 h/d x 7 d lung: tGSH, GPx, GRase, G6PD Chow et al. 1975
(20-25%)
RBC's: nonsignificant tGPx,
4- GSH

(continued)

q
i
01
0
"a
O
O
M


50
T*
1-3
M
3
3
S
g
K<
K!
CO











01
&
01
H*
O
V
I
0

-------
                                                         TABLE 11-1  (continued)
                 Species
                              Exposure Conditions
                                                                                  Observations*
                                                                                                                      Reference
rhesus monkey
  male
  2 yr
rat
  HLA-Greenacres
  male
                                     0.5 ppm x 8 h/d x 7 d
                                     1.5-12.5 ppm x 1-5 h
                                                           lung:  tGSH, GPx, GRase, G6PD
                                                                  (10-15%)
                                                           RBC's:  nonsignificant +GPx,
                                                                   4-GSH

                                                           tneutrophil:lymphocyte ratio
                                                                                                                Chow et al. 1975
                                                                                                                Bobb and Fairchild 1967
                                                                         3
                                                                         CO
                                                                         O

                                                                         O
rat
  C.R. Fischer
  male
                                     1-2 ppm x 2 or 7 d
                                                           no blood chemistry changes
                                                                                                                Cavender et al. 1977
guinea pig
  C.R. Hartley
  male
                                     1-2 ppm x 2 or 7 d
                                                           no blood chemistry changes
                                                                                                                Cavender et al. 1977
VO
w
          squirrel monkey
                            0.75 ppm x 4 h/d x 4  d
                                                           lung:  tlipid oxidation
                                                                  + tocopherol
                                                                  tG6PD, GRase, LDH
                                                                  no effect MDH, SOD
                                                           RBC's:  tRBC fragility (H2O2>
                                                                   +GSH, (AChE)
                                                                   no effect LDH, G6PD
                                                                   (paired analyses)
                                                                                                                Clark et al. 1978
                                                                                                                                              CQ
          rabbit
                                     0.2 ppm x 4 h
                                                           fosmotic fragility
                                                           spherocytosis
                                                                                                                Brinkman et al. 1964
 rabbit
  New Zealand White
  male and  female
  ~1  kg
          rabbit
                            0.4 ppm, 1 ppm
                            6 h/d x 5 d/wk x 10  mo
                            10 ppm x 1 h/wk x 6 wk
after ~105 d:
+serum albumin
to- S Y-globulin
no change total protein
^appetite last 3 mo
(effects greater in males)

production of  serum Ab's
that reacted with ozonized
egg albumin but not native
ovalbumin
                                                                                                                P'an and Jegier 1976
                                                                                                                Scheel et al.  1959
                                                                                                                                    «
                                                                                                                                    u
                                                                                                                                    I
                                                                  ( continued)
                                                                                                                                              d

-------
                 Species
                                                         TABLE 11-1  (continued)
                                       Exposure  Conditions
                                             Observations3
                                                                                                                     Reference
         Homo sapiens
           male

         Homo sapiens

         Homo sapiens
           male
         Homo sapiens
           male (20)
           female (2)
           asthmatic
0.15 ppm x 1 h ± exercise
0.30 ppm x 1 h ± exercise

0.2  ppm x 30-60 min

0.5  ppm x 165 min
0.25 ppm x 2 h ±  exercise
no effect NPSH,  G6PD
6PGD, GRase, Hb

spherocytosis

tosmotic fragility
+ AChE, GSH,  G6PD
+serum TEA,  tocopherol,  LDH

+H2O2 fragility
+GSH, AChE
tG6PD, LDH
no effect GRase, GPx,
23-OPG, Hct
DeLucia and Adams 1977


Brinkman et al. 1964

Buckley et al. 1975



Linn et al. 1978
         Definitions of abbreviations:   Ab's, antibodies; AChE, acetylcholinesterase;  2,3-DPG,  2,3-diphosphoglycerate; GPx,
          glutathione peroxidase;  GRase,  glutathione reductase; GSH, reduced glutathione;  G6PD,  glucose-6-phosphate dehydrogenase;
VO        Hb,  hemoglobin;  Hct,  hematocrit; LDH,  lactate dehydrogenase; MDH, malate dehydrogenase;  MetHb, methemoglobin; NPSH,
*        nonprotein sulfhydryl; PK, pyruvate kinase; RBC's, red blood cells; 6PGD, 6-phosphoglycerate dehydrogenase; SOD, superoxide
          disrautase; TBA,  thiobarbituric  acid reactive products.
                                                                                                         n

                                                                                                         W
                                                                                                                                              0)
                                                                                                                                              O
                                                                                                                                              n
                                                                                                                                              o

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg






when, in primate studies, paired comparisons  (each animal or human subject




serves as its own control) were used in analyzing the data.  These important




points will be discussed later.









     Our involvement in this kind of research began with the publication of




research (Zelac et al. 1971) showing strikingly high levels of chromosome




aberrations in circulating lymphocytes of Chinese hamsters exposed to




approximately industrial tolerance  levels of  03.  These findings raised




questions regarding the carcinogenic potential of 63 as well as the mechanisms




that could be operative in affecting lymphocyte chromatin.  Within the Medical




Department at Brookhaven, C.J. Shellabarger's program offers the




Sprague-Dawley rat mammary tumor model  (Huggins et al. 1959) calibrated for




tumor production by various radiations and chemical carcinogens (Shellabarger




1971, 1972; Shellabarger and Soo 1973; Shellabarger and Straub 1972).  Hence,




this model seemed suitable for a direct test  of possible extrapulmonary




carcinogenicity from 03.  Furthermore, mammary tissue seemed an appropriate




"target tissue" considering our expectations  that oxidizing lipids or their




products might play a major role in amplifying any systemic cytotoxicity




(Goldstein et al. 1969; Roehm et al. 1971).   One such oxidation product,




malonaldehyde, shows a weak mutagenic potential in the Ames screening test




(Mukai and Goldstein  1976).  In conjunction with the tumor studies, then, we




began to look for biochemical markers indicating either (1) passage of




oxidative products through the lung barrier or  (2) other systemic changes




attributable to 03  (Borg and Shellabarger  1978).
                                      95

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg







INITIAL STUDIES









     Using an O3 exposure system constructed from a modified controlled




atmosphere chamber, we first focused on finding evidence of lipid oxidation




and directly related oxidative damage in red blood cells and mammary tissue:




destruction of red cell membrane integrity, appearance of thiobarbituric acid




reactive products (TEA) or fluorescence characteristic of iminopropene




compounds, and changes in thin layer chromatography (TLC) patterns of




extracted lipids.  Subsequently, analyses for reduced glutathione (GSH),




enzymes, and hemolysis were added to the list of parameters monitored.  We




found great variability in the resistance of individual rats and different




batches of rats.  ("Resistance" is here defined as the absence of observable




changes in red cells; all exposed rats showed pulmonary lesions typical of 03




toxicology.)  Changes in the various parameters following acute exposures




(e.g.:  3-8 ppm, 2-6 h, 1-3 d) were seldom observed in old rats (~>80 d),




whereas rats exposed at ~40 d of age for only 1 d often showed dramatic




responses in iminopropene fluorescence (Figures 11-1 through 11-3), oxidation




products apparent on thin layer chromatographs (Figure 11-4), and depression




of acetylcholinesterase activity.  When such effects were manifested in red




cells, they were short-lived (no more than 2 d), thereafter declining to




approximately control level, then gradually increasing again when exposures




were extended.
                                     96

-------
11.   EFFECTS OF OZONE ON RAT ERYTHROCYTES
                                                           Schaich and Borg
         A. ERYTHROCYTES                    EMISSION
                               EXCITATION
                                                          EXPOSED
                                                          CONTROL
           300   320
360        400
FLUORESCENCE (nm)
440
480
         B. MAMMARY TISSUE
           300    320
360        400
FLUORESCENCE (nm)
440
480
Figure 11-1.  Typical iminopropene fluorescence patterns of erythrocyte (A)
             and mammary tissue (B) lipids from female Sprague-Dawley rats
             exposed to 2 ppm 03 for 5 h.
                                   97

-------
                                                                                                       M
00
                                             D VITAMIN E DEFICIENT

                                             O VITAMIN E ENRICHED

                                             A PURINA
                                                                                   30 CALENDAR DAY
                                                                                      EXPOSURE
                                                                                      (ppm 03 for 5 h/d)
    Figure  11-2.  Iminopropene fluorescence  in erythrocyte lipids from female Sprague-Dawley rats  exposed
                 chronically to O-.
                                                                                                       I
2
M
a

I
                                                                                                       M
                                                                                                       CO
                                                                                                       CO
                                                                                                       8-
o

01

a.

-------
10
                                              D  VITAMIN E DEFICIENT

                                              O  VITAMIN E ENRICHED

                                              A  PURINA
                                                                                                        M
                                                                                                        *a
                                                                                                        3

                                                                                                        9
                                                                                                        1
                                                                                                        w
                                                                                                        a
         0
         1—2-
8    10   12    14    16   18   20    22   24   26    28    30 CALENDAR DAY
 I-H                                                      EXPOSURE
   I	2	1    1—2	1            I	2	1    (ppm 03 for 5 h/d)
    Figure 11-3.  Iminopropene fluorescence in mammary tissue lipids from female Sprague-Dawley rats
                  exposed chronically to 0 .
                                                                          O
                                                                          H
                                                                          Kq

-------
 11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES
                                                Schaich and Borg
SOLVENT
FRONTS



A

o
< \
0
'-..
6
SOLVENT
FRONnr


^

t>
o
o

•i
'0
SOLVENT
rOrtftlT* 	 ,
FRONT
a

Vh

Cl
n
a

©
0

r >
O
0
y I
o i
'-'Jl>l
VIA L,l
" 1
• 1
[/
O
1

"o~
D
O
_ f» 1
A

J^'l
F 1
g,'U
S. ,1
(J 'l
i


S
o
0
0
e 1
1
c=>

O
1
A
l"3! O <=
1 O •=•

1 is A

Q ^>
0 Q

-L,l
© C c:
2 4 7
B
a o <=»
O 0

_

n o 0
c> **
0 0 ey
*~*
C: ,-, o
247
C
<£Z^
s S
o
m fh m

0 o n
o
° (i *
\j

247

0 I


•s,i

°

^=>
i <
CM
gIM,

~^~
1
(ffh, C
p '

1
1
«=> 1
1
w N 1
9


^^ 1

®S-'
1
0 1
0 1
^ 1

/.* M I
9 '
Figure 11-4,
               OZONE EXPOSURE DAY

Thin layer chromatograms of erythrocyte lipids  from  Sprague-
Dawley rats exposed chronically to 2  ppm 03  x 5 h/d.   Solvent is
hexane:ethyl etherracetic acid = 80:20:1.  Spot visualization
(see right margin if unlabeled):  I - iodine vapor;  S  -  short
UV; L - long UV; F - fluorescent; N - ninhydrin.   Cross-hatching
indicates UV light.  Diets:  Purina (A), vitamin  E enriched (B),
vitamin E deficient (C).
                                     100

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTBS	     Schaich and Borq






EFFECT OF DIETARY TOCOPHEROL










     The difficulties we encountered  in  inducing measurable oxidative effects




in red cells of rats fed the standard Purina Laboratory chow led us to




consider dietary and antioxidant  factors.  Lab chow contains a relatively high




level of tocopherol (for females  more than twice and for males ~8 times the




level necessary for maintaining reproductive capacity), so rats subsisting on




it should be well protected against oxidants.  Therefore, to find a more




sensitive population in which potential  systemic effects of 03 would be more




readily observed and defined, we  used a  synthetic "Draper" diet (Draper et al.




1964) to induce tocopherol deficiency in some rats before exposure.  Other




rats were maintained on the Draper diet  supplemented with a-tocopherol (75




mg/kg chow, ~10 times the tocopherol  level of standard chow).









     Although  the Draper diet is  the  classic diet most commonly employed in




tocopherol deficiency studies, we found  it to be less  than satisfactory as a




supporting diet.  Females transferred to the diet (with or without




supplementary  tocopherol) immediately after weaning never grew properly or




matured sexually, and hence could not be used for tumor studies (mammary tumor




production in  our model is hormone-dependent), and relatively mild acute




exposures to 03  (e.g., <2 ppm for 2 h) were occasionally lethal.  As a




compromise protocol, weanling rats were  fed for 3 weeks on standard laboratory




chow (by which time most had reached  estrus) before transfer to one of the




synthetic diets for 7 to 10 d (7  d was sufficient to attain a deficiency




state, as measured by hemolytic susceptibility) and subsequent exposure to O3.







                                      101

-------
11.  EFFECTS OF OZONE ON RAT SRYTHROCYTSS	Schaich and Borg







Even so, deficiencies (or toxicants) other than tocopherol existed in the




synthetic diets, and led to some anomalous response behavior, especially in




the tocopherol-supplemented rats.  A comparison of diet effects alone on




control values shows some biochemical parameters (e.g., GSH, fragility,




methemoglobin) to be consistent with antioxidant protection of extra




tocopherol, whereas patterns of enzyme activities cannot be attributed




entirely to tocopherol deficiency or supplementation (Table 11-2).









     In extending our studies from acute to short-term chronic exposures, we




included analyses of some red cell metabolic enzymes.  Dietary factors




represented a second set of variables.  Our goal was to gain a clearer




understanding of sequences of 03 effects (if any) and to determine whether the




apparent absence of lipid oxidation and glutathione destruction were real as




opposed to reflecting enhanced intracellular decomposition and regeneration,




respectively.  We exposed rats on each of the three diets (Purina, Draper ±




tocopherol) to 2.0 ppm 03 for 5 h/d for 1 and 10 d.  The animals were 50 d of




age at first exposure.









     As shown in Table 11-3, acute exposures under these conditions induced no




changes in the red cell parameters measured, nor were effects much more marked




after 10 d of exposure.  What is interesting about the pattern of effects in




the chronic exposures is that all statistically significant changes occurred




in rats on the Draper synthetic diet—both with and without tocopherol (Table




11-4):  decreased GSH (± tocopherol), increased TEA values and decreased




osmotic fragility (+ tocopherol), decreased glucose-6-phosphate dehydrogenase







                                     102

-------


Test
Parameter3
GSH (mg %)
TBAb
osmotic
fragility0
G6PD (units)
AChE (units)
o
w MetHb (%)
TABLE 11-2. DIET EFFECTS ON BIOCHEMICAL PARAMETERS
IN RED BLOOD CELLS OF UNEXPOSED SPRAGUE-DAWLEY RATS
Diet Probability ( 1-p) That
Vitamin E vitamin E Difference Between Diet
Purina Enriched Deficient Groups Is Not Due to Chance
(P) (E) (D) P-E P-D E-D
75.16 ± 0.86 88.13 ± 3.24 68.47 ± 2.32 1.00 1.00 1.00
6.71 ± 0.76 8.27 ± 0.98 8.97 ± 1.46 0.99 0.99 0.82

0.476 ± 0.003 0.461 ± 0.006 0.470 ± 0.007 0.96 0.81 0.85
13.27 ± 0.04 12.42 ± 0.62 10.09 ± 0.44 0.98 1.00 1.00
11.86 ± 0.48 11.56 ± 0.21 11.35 ± 0.77 0.88 0.90 0.72
4.81 ± 0.34 2.72 ± 0.35 3.60 ± 0.62 1.00 ~1.00 0.99
aDefinitions of abbreviations: TBA, thiobarbituric acid reactive products; MetHb, methemoglobin ; GSH,
glutathione; AChE, acetylcholinesterase ; G6PD, glucose-6-phosphate dehydrogenase . ^felues given are mean
± standard deviation.
bMoles x 10~6
thiobarbituric acid products/ml packed cells.
c% NaCl solution giving 50% hemolysis.








^
. EFFECTS OF OZONE
§
1
M
i
1
1


CO
g.
a
I
0
3
Qt
W
O
1
Q

-------
TABLE 11-3. BIOCHEMICAL CHANGES IN RAT RED BLOOD CELLS FOLLOWING ACUTE EXPOSURE TO OZONE ^

Diet —
Test Purina Vitamin E Enriched Vitamin E Deficient Jjj
Parameter" Control Exposed p° Control
TBAd 6.64 ± 0.53 7.14 ± 0.91 NS 6.23 ± 1.16
HetHb (%) 4.35 ± 0.60 4.49 ± 0.30 NS 3.88 ± 1.16
osmotic
fragilitye — -_ 0.526 ± 0.020
GSH (mg %) 79.3 ± 14.8 82.8 ± 13.0 NS 89.77 ± 6.22
AChE (units) 13.6 ± 1.5 14.2 ± 2.3 NS 14.1 ± 3.7
G6PD (units) 8.2 ± 0.9 8.2 ± 0.9 NS 8.6 ±1.5
aPurina diet: 3 ppm 03 x 4 h. Vitamin E Enriched and Deficient diets:

Exposed pc Control Exposed pc (•}
3
5.73 ± 0.82 NS 5.77 ± 0.48 5.85 ± 0.69 NS O
3.53 ± 0.53 NS 4.27 ± 0.67 4.57 ± 0.60 NS Q
§
0.503 ± 0.013 NS 0.533 ± 0.05 0.527 ± 0.03 NS W
85.86 i 3.44 0.06 115.3 ± 2.8 116.4 ± 7.6 NS 3
13.2 ± 2.5 NS 12.4 ± 1.6 12.5 ± 2.0 NS V*
9.0 ± 1.2 NS 11.5 ± 0.6 11.7 ± 1.1 NS RJ
2 ppm 03 x 4 h. S
Q
Definitions of abbreviations: TEA, thiobarbituric acid reactive products; MetHb, methemoglobin j GSH, reduced glutathione; AChe, O
acetylcholinesterasei G6PD, glucose-6-phosphate dehydrogenase. Values
°NS - not significant (p > 0.10 in one-tailed t-test) .
dMoles x 10~* thiobarbituric acid reactive products/ml packed cells.
e% NaCl solution giving 50% hemolysis.










given are mean ± standard deviation. 2
m
^
01


01
o
3-
h«.
n
3-
(h
a
Qi
a
o
a

-------
Ul
TABLE 11-4. BIOCHEMICAL CHANGES IN
RAT RED BLOOD CELLS FOLLOWING CHRONIC EXPOSURE TO OZONE3 ^
Diet M
Test Purina
Parameter^ Control Exposed p=
TBAd 6.71 ± 0.76 7.43 ± 1.21 NS
MetHb (%) 4.81 ± 0.34 4.28 ± 0.76 0.10
osmotic
fragility8 0.476 -± 0.03 0.474 ± 0.002 NS
GSH (mg t) 75.16 ± 0.86 75.35 ± 3.31 NS
AChE (units) 11.86 ± 0.48 12.10 ± 0.35 NS
G6PD (units) 13.27 ± 0.04 14.49 i 0.47 NS
Heinz
bodies (%) 0.10 ± 0.00 0.10 ± 0.00 NS
a2 ppm 03 x 4 h/d x 10 d.
3
Vitamin E Enriched Vitamin E Deficient HI
Control Exposed pc Control Exposed p° O
CO
8.27 ± 0.98 9.39 ± 1.49 0.06 8.97 ± 1.46 9.48 ± 2.10 NS O
"g
2.72 ± 0.35 2.77 ± 0.30 NS 3.60 ± 0.62 3.81 ± 0.47 NS Q
O
0.461 ± 0.006 0.449 ± 0.006 -(0.06) 0.470 ± 0.008 0.471 ± 0.000 NS M
88.13 ± 3.24 69.63 ± 5.27 0.005 68.47 ± 2.32 62.21 ± 4.13 0.005 §
11.56 ± 0.20 11.77 ± 0.99 NS 11.35 ± 0.77 11.11 ± 1.15 NS g
12.42 ± 0.62 12.24 ± 1.08 NS 10.09 ± 0.44 9.21 ± 0.63 0.01
3
0.50 i 0.28 0.60 ± 0.36 NS 1.40 ± 0.28 2.20 ± 0.00 0.03 J|
I
^Definitions of abbreviations: TBA, thiobarbituric acid reactive products; MetHb, methemoglobin ; GSH, reduced gluta thione ; AChE, acetylcholinesterase; 3
G6PD, glucose-6-phosphate dehydrogenase . Values given are , mean ± standard deviation. (Q
°NS = not significant (p > 0.10).
dMoles x 10"6 thiobarbituric acid reactive products/ml packed
e%NaCl solution giving 50% hemolysis.





cells.

a
u
j.
•f
u
3
Q*
to
3
-Q

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTSS	Schaich and Borg


(G6PD) and increased Heinz bodies (- tocopherol).  The seemingly anomalous

Decrease in fragility after 03 exposure in the supplemented rats may be

explained by observations of substantial hemolysis in blood drawn from vitamin

E supplemented rats; washed cells used for this test could be a residual, more

resistant population.



DISCUSSION



     From this inconsistent pattern of changes alone (not to mention much of

our early experience with acute and other short-term chronic exposures) it is

tempting to conclude that cytotoxic or oxidative potential from 03 is not

transferred from lung tissue to circulating red cells.  However, we are not

ready to dismiss the issue of systemic effects for three primary reasons:
     1.  TLC patterns of lipids extracted from red cells and mammary
         tissue indicate oxidative degradation of lipids;

     2.  Unpublished data from our lab and Dr. Drew's bring into
         question the utility of red cell studies in screening for
         oxidant systemic effects;

     3.  Rats, and perhaps all rodents, are unsuitable as models
         for human response to atmospheric oxidants.
The following paragraphs consider each of these points in more detail.
                                     106

-------
11.  EFFECTS OF OZONE ON BAT ERYTHROCYTES	Schaich and Borq





Evidence for Oxidative Degradation of Lipids









      After nearly every short-term  chronic exposure, some degree of oxidative




response was apparent, but the pattern of response varied between batches of




animals and with age, exposure level, etc.  The only effect that occurred




repeatedly and consistently with all diet groups, batches, and exposures




(including high-level acute exposures) was changes in TLC patterns of




extracted lipids.









     New TLC spots typical of those  expected  from a variety of oxidation




products appeared in the nonpolar fractions of lipids extracted from red cells




(Figure 11-5).  The classes of lipids that seem to have been involved are




cholesteryl esters, diglycerides and triglycerides, and free fatty acids (and




perhaps their esters), indicating glyceride hydrolysis and fatty acid




oxidation.  Tocopherol showed little effect.








     In contrast, in mammary tissue  the  phospholipid and polar lipid fractions




were most affected  (Figure  11-6), and the TLC changes were cumulative with




exposure time (Figure  11-7).  From comparisons with pure standards, it seems




that, in mammary tissue, the phospholipids phosphatidylethanolamine,




phosphatidylcholine (lecithin), phosphatidylserine, and phosphatidylinositol




are  the most likely targets.  The changes are consistent with fatty acid




oxidation, hydrolysis to lysolecithins and phosphatic acids, and formation of




iminopropene fluorescent products from reaction of fatty acid carbonyls with




amino groups.






                                      107

-------
11.   EFFECTS OF OZONE ON RAT ERYTHROCYTES
                                Schaich and Borg
   SOLVENT
    FRONT
      CONTROL   EXPOSED  CONTROL  EXPOSED  CONTROL   EXPOSED
          |	1         |	1           )	,
           PURINA
VITAMIN E ENRICHED    VITAMIN E DEFICIENT
Figure 11-5.  Thin layer chromatograms of erythrocyte lipids from female
             Sprague-Dawley rats exposed to 03  (2 ppm x 4 h x 10 d).
             Solvent is hexane:ethyl ether:acetic acid = 80:20:1.  Spot
             visualization:   I - iodine vapor?  S - short UV; L - long UV;
             N - ninhydrin.
                                   108

-------
11.   EFFECTS OF OZONE ON RAT ERYTHROCYTES
                                Schaich and Borg
  SOLVENT
   FRONT
       IS
           \
         \J
          0
            a i
              N
                                         A
                               0
                                        A   i
                                            i
                         0        0
                        ON
                                                       N.I
                                                                  ''L'F
ON
9   NN''
                                       N
      CONTROL  EXPOSED  CONTROL  EXPOSED    CONTROL   EXPOSED
          1	1         I	1          1	1
   DIET:    PURINA
VITAMIN E ENRICHED   VITAMIN E DEFICIENT
 Figure  11-6.  Thin layer chromatograms of mammary tissue lipids from female
              Sprague-Dawley rats exposed to 03 (2 ppm x 4 h x 10  d).  Solvent
              is chloroform:methanol:acetone:acetic acid:water =
              65:10:20:10:3.   Spot visualization:  I - iodine vapor; S - short
              UV; L - long UV;  F - fluorescent; N - ninhydrin.
                                    109

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTSS
                                                             Schaich and Borg
                   SOLVENT

                                     0    0
                                      0 N-'

                               o    o
                               o
                               1
                             I
                             N I
                              '
                                              oN,l  o
                   SOLVENT

                          0    0    0
                                1
Figure 11-7.
               OZONE EXPOSURE DAY
Thin layer chromatograms of mammary tissue  lipids from female
Sprague-Dawley rats exposed chronically to  2 ppm 03 x 5 h/d.
Solvent is chloroform:methanol:acetone:acetic acid:water =
65:10:20:10:3.  Spot visualization (see right margin if
unlabeled):  I - iodine vapor; S  - short UV; L - long UV;
F - fluorescent; N - ninhydrin.   Cross-hatching indicates UV
light.  Diets:  Purina (A), vitamin E  enriched (B), vitamin E
deficient (C).
                                    110

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg






     These changes show clearly that oxidative potential initiated in the lung




can be carried to distant sites.  Whether these specific changes are




functionally meaningful is another matter, since they cannot be correlated




with any of the other parameters measured.  Nevertheless, they do indicate




latent damage which, under our exposure  conditions, is either (1) not




sufficiently extensive to affect cell  functioning, or (2) rapidly repaired




(e.g., by the spleen's clearing of damaged red cells).









The Dynamic Nature of the Systemic Response to Oxidants








     The concept  of  latent damage brings us to some unreported observations in




our  laboratory  suggesting that systemic  responses to oxidants and related




secondary  stress  are dynamic.   From  these observations we can infer that




"spot-checking" a few parameters  for evidence of deleterious effects might not




yield an accurate picture of what is happening in vivo.








     A  few reports in the  literature have  considered red cell morphology to be




affected  by O3, noting spherocytosis and crenation  of red cells  (Larkin et al.




 1978) as well as  reticulocytosis and changes in  the proportional distributions




of blood  cell types (Bobb and Fairchild 1967).   We  have  noted but not yet




quantified similar changes.   In our studies,  spherocytosis  seemed most common




 for low-level exposures whereas formation of echinocytes (crenated or spiny




cells)  was quite  marked at high-level acute exposures  or after  chronic




 exposures.  Each of these types of cells denotes latent damage  that  alters




      response to the osmotic environment without,  for the most  part, affecting






                                       111

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	.	Schaich and Borg





cell function.  We have also seen a variable reticulocytotic response.  When




reticulocytosis seemed most marked (with attendant increases in hematocrit),




biochemical measures showed the greatest change and the animals themselves




showed the greatest respiratory distress.  Thus, we are left with several




questions:  Is the reticulocytosis a direct response to red cell damage jLn




vivo, or is it a secondary reaction to pulmonary pathology?  Since enzyme




activities in young erythrocytes are generally higher than in mature cells, is




it possible that the apparent lack of observable changes results from larger




numbers of reticulocytes counterbalancing decreased activities of defective




cells?  To what extent do these types of latent damage indicate the presence




of a rat response threshold below which "systemic" effects of 03 are




detectable only in hypersensitive populations, such as in tocopherol or other




nutrient deficiencies?  We are currently looking for answers to these




questions.








     Consider, also, that few studies have focused on tissues other than




blood, presumably on the assumption that any extrapulmonary effects will be




first noticeable in the bloodstream.  In some preliminary studies by Dr.




Drew's group (Costa and Drew 1980), rats exposed to sulfur dioxide for only a




few hours showed a reproducible destruction of sulfhydryl groups in the lung




and liver but not in blood.  While extrapolation would be premature, these




findings certainly raise questions about the utility of red cell studies in




screening for systemic effects of oxidants.
                                     112

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg






Nonsuitability of Rats as Models for Human Response to Oxidants









     This brings us to the last point:  the nonsuitability of rats (and




perhaps all rodents) as models for human response, either pulmonary or




systemic, to 03 and oxidant gases.  The difficulties we have encountered in




producing in rats a consistent pattern of measurable systemic damage from 03




similar to that observed in other species, coupled with known physical and




biochemical/metabolic characteristics of the rat, lead us to contend that the




rat  is in fact a poor surrogate for humans.  Three lines of reasoning support




this contention.








     First, there are significant differences in respiratory structure between




rodents and primates.  Rats are obligate nasal breathers; thus O3 uptake in




the  upper airway is greater than if breathing were by mouth (as in, for




example, humans who are exercising or who have nasal obstruction).  Unlike




primates, rats do not have well develope'd respiratory bronchioles (the




transition phase between the  terminal bronchiole and alveolar ducts) (Mellick




et al.  1977; Castleman et  al.  1977).  Both of these factors contribute to the




decreased distribution of  oxidant gases to respiratory exchange tissues in the




rodent.  Miller et al.  (1978)  developed mathematical models of transport and




removal of oxidant gases  (03  and nitrogen dioxide) in lungs of different




species.  These authors predicted the following relative respiratory




bronchiolar doses  (assuming tracheal doses of >0.05 ppm):  man,  100; rabbit,




80;  guinea pig, 40.   Relative comparisons of lung morphology and structure




would place rats on the  lower end of this scale.  The net  implication  is that,






                                      113

-------
 11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES  	Schaich  and Borg






 for  identical closes, much less 03 reaches peripheral lung cells and the




 gas/blood  barrier  in rats than in humans.  In addition, the GSH peroxidase




 system for detoxifying and preventing accumulation of oxidative products  in




 lung tissue normally operates at a much higher level in rats than in




 primates—e.g., four times that in monkeys (Chow et al. 1975)—and is  also




 more readily stimulated by oxidant challenge.  Thus, neither primary oxidant




 (03) nor secondary toxic products such as peroxides would be expected  to pass




 into blood at levels comparable to those in humans and other primates.








     "Antioxidative" defense mechanisms in red cells are also much more active




 in rats than in man.  Kurian and Iyer (1977), for example, have shown  that




 rats and other rodents are much less susceptible than man and other primates




 to stress  with the oxidant acetylphenyl hydrazine (APH) (Figure 11-8).  Heinz




 body production is also correspondingly decreased.  Higher levels of GSH, GSH




 peroxidase, and G6PD protect against even intense oxidative stress while




 efficiently regenerating active levels of intracellular reducing agents




 (Table 11-5).  Thus, in comparison to human red cells, rat red cells should be




 better able to cope with any 03 stress.









     Finally, there are problems with statistical treatment of data necessary




 to show "significant" differences between exposed and control values.  In




 studies citing significant 03 effects on red cell enzyme levels in monkeys




 (Clark et al. 1978) and humans (Buckley et al. 1975; Linn et al. 1978), the




actual  differences in activities were small.  (For example, Table 11-6




presents some results of the Linn et al. study.)  However, the data were






                                     114

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES
                                                              Schaich and Borg
                 100-
                                                  CAT

                                                  DOG
                                                  MAN

                                                  COW
                                                  MONKEY
                                                  GOAT
                                                  RAT
                                                  RABBIT
                                                  GUINEA PIG
 Figure 11-8.
                                30     40

                                TIME (min)
Species differences in sensitivity to oxidative  (APH) stress.
A 10% suspension of washed erythrocytes in  Ringer phosphate was
treated with APH (2 mg/ml) at 38°C;  at 15-min  intervals 0.5-ml
aliquots were withdrawn, diluted with 5 ml  ice-cold water, and
MetHb estimated.  Data from:   Kurian and Iyer  (1977).
    TABLE 11-5.  SPECIES DIFFERENCES IN LEVELS OF SOME PROTECTIVE ENZYMES
                FOLLOWING STRESS WITH ACETYL PHENYLHYDRAZINEa
               Parameter^
                         Rat
Man
Monkey
SOD (units)
GPx (yM GSH/g
Catalase (K x

Hb/min)
103)
52
122
8.6
30
12
36.5
28
5
38.7
aData from:  Kurian and Iyer (1977).

^Definitions of abbreviations:   GPx,  glutathione peroxidase; GSH, reduced
 glutathione; Hb,  hemoglobin;  SOD,  superoxide dismutase.
                                     115

-------
TABLE 11-6. EFFECTS OF OZONE IN ASTHMATICS*1
variable11
Hb (g/100 ml)

Hct (%)
HjO2 fragility

GSH

AChE


G6PD

LDH

GRase
GPx
2,3-DPG
a Adapted from:
bDefinitions of
Sharac
14.99 ± 0.96
1
44.4 ± 2.9
24.1 ± 7.0
I
30.8 ± 5.4
M
1
22.2 ± 2.5
I

1 	

5. 12 ± 1.06
t
| 	 	 	

104.0 ± 12.8
f
2.64 ± 0.79
10.87 ± 3.05
14.30 t 1.67
Odor-Shamc
14.97 ± 0.94
i

44.6 t 2.7
25.0 ± 6.9
t

28.7 ± 5.4
	 I

21.5 ± 2.4
	 I I 	



5.45 ± 1. 16
t


108.4 ± 16.1

2.66 ± 0.78
10.67 ± 2.73
14.52 ± 1.78
Linn et al. (1978) (Table 6). Exposure - 0.25 ppm O3
abbreviations : AChE,
GSH, reduced glutathione (red blood
red blood cell
fragility expressed a
acetylcholinesterase ; 2,3-DPG, 2
cell) ; G6PD, glucose-6-phosphate
s % hemolysis when incubated with
03 ExposureC F(2,42)d pd Change (%)
14.88 ± 0.99 5.2" 0.009 0.7
| |
I
44.2 ± 2.7 1.83 NS 1
26.8 ± 7.2 7.46 0.002 11
i !
	 	 	 i
29.3 ± 5.0 5.53 0.007 5
I
20.9 ± 2.5 24.94 <0.001 6
I t



5.67 ± 1.20 9.98 <0.001 11
i

111.4 ± 17.0 4.82 0.013 7
I
2.68 ± 0.74 0.89 NS
10.99 ± 2.54 0.57 NS 3
14.58 ± 2.53 0.50 NS 2
x 2 h. N = 22.
, 3-diphosphoglycerate ; GPx, glutathione peroxidase; GRase, glutathione reductase;
dehydrogenase; Hb, hemoglobin concentration; Hct, hematocrit; H202 fragility.
2% hydrogen peroxide solution; LDH, lactate dehydrogenase.
c values shown are mean * S.D. Pairwise significant differences are indicated by solid arrows for p < 0.05; by dashed arrows for p < 0.01.
dOverall variation is indicated by the F statistic (degrees of freedom
























in parentheses) and the p value. NS = not significant.








^
3
a
Q
CO
o

o
i

tt
3
d
3
3
in









CO
o
o
rr
91
s
Oi
td
0
n

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg






obtained as paired comparisons (comparisons of measures on the same subject




before and after exposure) rather than as comparisons between exposed and




nonexposed populations.  We took the post-exposure mean and standard deviation




data from some paired-comparison studies and applied several other standard




statistical tests of group differences.  As we had suspected, under these




conditions—conditions commonly used in small animal studies—we found no




statistical differences, thus illustrating the power of paired comparisons as




a  statistical tool.









     In small animals, the problems are twofold:  (1) paired comparisons are




not physically possible;  (2) differences in individual animal responses give




group variances that are too large to allow statistical detection of small but




probably real group differences.  If, then, the issue of 03 effects in red




cells is largely a "numbers game," the rules are predesigned for "no effects"




in any small animal study that doesn't involve very large numbers of animals




or show large differences in group mean -values.  With respect to the latter,




comparisons of exposed and control lung tissue (the first site of interaction




with oxidant gases) in rats and other species have shown 25% or greater




differences in, for example, enzyme or GSH levels.  But comparable effects in




red blood cells—especially the red blood cells of rodents—would not be




expected and in fact have not been seen, even in the presence of superimposed




stress such as tocopherol deficiency.









     This analysis does not imply that the use of rats and other rodents in




oxidants research should be discontinued.  However, the considerations






                                     117

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg






outlined here urge strong caution in the use of rats as surrogates for human




beings.  Especially in the case of extrapulmonary effects, it is extremely




doubtful whether "no effect" observations in rats can be credibly extrapolated




to predict "no" human responses.









SUMMARY AND OVERVIEW OF CURRENT RESEARCH









     In summary, acute and chronic exposures of rats to relatively high levels




of 03 have failed to provide evidence for statistically significant




extrapulmonary effects in red cells except for oxidation in component lipids.




Hematological observations and trends in GSH and enzyme changes, however,




suggest the possibility of a response threshold not reached in animals exposed




under our experimental conditions.  Observations of damage to lipid and GSH in




mammary and liver tissues, respectively, indicate the potential even in rats




for systemic effects at sites distant from the lung—effects that have been




largely unexplored.









     Finally, although laboratory rats are quite useful for research purposes,




they may not be appropriate or adequate models for directly predicting human




response to gaseous oxidative pollutants.  We must identify an animal having a




respiratory structure comparable to that of the human but which is less




expensive than primates and large enough to allow sequential tissue sampling




and paired before-and-after comparisons.
                                     118

-------
11.   EFFECTS OF OZONE ON RAT ERYTHROCYTES     	Schaich and Borg

     Efforts to respond to these research needs are currently underway within

the Medical Department at Brookhaven, where we are (1) exposing rats to 03 for

longer periods (up to 67 d) and correlating measures of lung function,

biochemistry, and pathology with red cell biochemistry and metabolism,

hematology, serum measures of hepatic function, and cytogenetic end points;

and  (2) exposing sheep to oxidant gases and investigating mechanisms of a wide

variety of effects.  In the latter system, paired before-and-after comparisons

as well as selective exposure of single lobes (using another nonexposed lobe

in the same animal as the control) are possible.  Also, the species is

sufficiently large to allow simultaneous studies of systemic functional or

biochemical, immunological, and pulmonary function and biochemistry studies in

a single animal.  Such coordinated studies should provide considerable insight

into the total and overall effects of oxidative pollutants, the mechanisms

responsible, and the relationship between effects on different body systems.



REFERENCES


Bobb, G. A., and E. J. Fairchild.   1967.  Toxicol. Appl. Pharmacol., 11:558.

Borg, D. C., and C. J. Shellabarger.   1978.  Internal Report 24400 (EPA
     Contract IAG-D5-0410), Brookhaven National Laboratory, Upton, New York.

Brinkman, R., H. B. Lamberts, and T. S. Veninga.   1964.  Lancet, 1:133.

Buckley, R.  D., J. D. Hackney, K. Clark, and C. Posin.   1975.  Arch. Environ.
     Health, 30:40.

Castleman, W. L., W. S. Tyler, and D. L. Dungworth.   1977.  Exp. Mol. Pathol.,
     26:384.

Cavender, F. L., W. H. Steinhagen, C. E. Ulrich, et al.   1977.  J. Toxicol.
     Environ. Health, 3:521.
                                      119

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES     	Schaich and Borg


Chow, C. K., and J. J. Kaneko.   1979.  Environ. Res.,  19:49.

Chow, C. K., M. G. Mustafa, C. E. Cross, and B. K. Tarkington.   1975.
     Environ. Physiol. Biochem., 5:142.

Chow, C. K., W. M. Aufderheide,  J. J. Kaneko, et al.   1976.   Fed.  Proc.,
     35:741.

Clark, K. W., C. I. Posin, and R. D. Buckley.   1978.   J. Toxicol.  Environ.
     Health, 4:471.

Costa, D.,  and R. Drew (Medical  Department, Brookhaven National Laboratory).
     1980.  Unpublished data.

De Lucia, A. J., and W. C. Adams.  1977.  J. Appl. Physiol.:   Resp. Environ.
     Exercise Physiol., 43:75.

Draper, H.  H., J. G. Bergan, M.  Chin, A. S. Csallany,  and A.  V. Boaro.   1964.
     J.-Nutrit., 84:395.

Goldstein,  B. D.  1973.  Arch. Environ. Health, 26:279.

Goldstein,  B. D., B. Pearson, C. Lodi, et al.   1968.   Arch. Environ. Health,
     16:648.

Goldstein,  B. D., C. Lodi, C. Collison, and O. J. Balchum.  1969.  Arch.
     Environ. Health, 18:631.

Huggins, C., G. Briziarelli, and H. Button.  1959.  J. Exp. Med.,  109:25.

Kurian, M., and G. Y. N. Iyer.   1977.  Can. J. Biochem., 55:597.

Larkin, E.  C., S. L. Kinzey, and K. Siler.  1978.  J.  Appl. Physiol.:   Resp.
     Environ. Exercise Physiol., 45:893.

Linn, W. S., R. D. Buckley, C. E. Spier, et al.  1978.  Amer. Rev. Resp.
     Dis.,  117:835.

Mellick, P. W., D. L. Dungworth, L. W. Schwartz, and W. S. Tyler.   1977.
     Lab. Invest., 36:82.

Menzel, D.  B., R. J. Slaughter,  A. M. Bryant, and H. O. Jauregui.   1975.
     Arch.  Environ. Health, 30:296.

Miller, F. J., D. B. Menzel, and D. L. Coffin.  1978.  Environ.  Res.,  17:84.

Mukai, F. H., and B. D. Goldstein.  1976.  Science, 191:868.

P'an, A. Y. S., and Z. Jegier.   1976.  J. Am. Assoc. Ind. Hyg.,  37:329.
                                      120

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg


Roehm, J. N., J. 6. Hadley, and D. B. Mengel.   1971.  Arch. Environ. Health,
     23:142.

Scheel, L. D., O. J. Dobrogorski, J. T. Mountain, et al.   1959.  J. Appl.
     Physiol., 14:67.

Shellabarger, C. J.  1971.  Proc. Soc. Exp. Biol. Med., 136:1103.

Shellabarger, C. J.  1972.  Cancer Res., 32:883.

Shellabarger, C. J., and V. A. Soo.  1973.  Cancer Res., 33:1567.

Shellabarger, C. J., and R. F. Straub.   1972.   J. Natl. Cancer Inst., 48:185.

Zelac, R. E., H. L. Cromrory, W. E. Bolch, Jr., B. G. Dunavant, and H. A.
     Bevis.  1971.  Environ. Res., 4:262, 325.
WORKSHOP COMMENTARY



Question;  Did you test  for hemolysis  in  these  studies?

K. M. Schaich;  Yes.

Question;  And you did not find  hemolysis?

K. M. Schaich;  We tried a number of hemolysis  stress tests and finally used
osmotic hemolysis—differential  response  to  salt—because it gave the most
consistent results.  However,  this  is  perhaps not the right measure for
hemolysis.  Osmotic hemolysis  reportedly  is  related to the sphericity of the
cells (the amount that they are  puffed up).  It results from pressure from
the inside.  Thus, in the absence of overt damage to the cell membrane or
gross alterations to intracellular  chemistry, observations of osmotic
stress effects are unlikely.

     Also, the rat red cells seem to be much more resistent to osmotic stress
than human erythrocytes.

Question;  What is the possibility  that you  have a red cell subpopulation
which is so sensitive that it  breaks down, and  is accordingly not tested?

K. M. Schaich;  That is  precisely why  I mentioned our findings with
tocopherol-supplemented  rats.  We have not yet  been able to do the
hematological studies in a controlled  quantitative way at the same time as the
biochemical studies. However,  in the supplemented rats we could never draw
blood without substantial hemolysis; thus, we suspect a supersensitive
                                      121

-------
 11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg


 population of cells in those rats.  The same kind of hemolysis did not occur
 in handling the other rats, but those animals' spleens may have cleared the
 damaged cells more efficiently and rapidly.

     I  can't give you a confirmed answer.  I suspect that we may in fact be
 looking at a more resistant, residual population, but until we finish the
 current study in which we are monitoring hematological and biochemical
 changes,  I cannot say anything more definite.

 Question;  In your interesting results on extrapulmonary effects, are you able
 to control for lung injury as a possible nonspecific kind of extrapulmonary
 effect?

 K. M. Schaich;  We cannot.  That is one of the problems of this type of animal
 study.   To my knowledge, in the literature there have been no reports directly
 comparing and correlating pulmonary and extrapulmonary effects of oxidant
 gases.   Recently, there have been some studies which monitored what was
 happening morphologically in the lung with what was happening biochemically
 in the  bloodstream, but I think the correlations were more limited than the
 implications of your question.  We have always been concerned that the changes
 we observe in red cells may be secondary responses to infection rather than
 primary effects of O3.  It is precisely this consideration that led us, in our
 current studies, to use specific pathogen free animals (thereby eliminating
 endemic pulmonary infections) and to simultaneously monitor hematological
 changes (if any), red cell and serum biochemistry, and lung function and
 pathology.  In this way we hope to be able to differentiate primary and
 secondary effects.

 Question;  What was the age of the rats when you started the studies?

 K. M. Schaich;  In the chronic studies, the rats were ~50 d of age when we
 started the exposures, and that bothers me.  Age at initial exposure is one
 reason  why we handled our deficient diets (the Draper diets) in a manner
 different from that usually reported in the literature.  Normally, rats are
 maintained on lab chow for a month or two to allow them to grow and establish
 approximately adult weight before transferal to the deficient diets.

     Considering our earlier results, however, we were afraid we would miss
 potential early effects if we followed such a procedure.  We would have
 preferred to maintain rats on the Draper diets from the weanling age (19-21 d)
 so that they would be deficient by 40 to 42 d—the age at which we saw the
 early responses in rats on lab chow.  Regardless of diet, animals did not show
 "first  day" or acute effects if they were older than ~45-48 d.

     There is a marked age dependence in the response of rats to various
 radiations and chemicals.  This age dependence is at least partly related to
hormone levels.  Shellabarger showed in his mammary tumor model that there is
a critical period between ~45 and ~60 d where hormonal changes proceed so
                                     122

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	      Schaich and Borg


rapidly that the overall response of the rats to systemic challenges is
altered.

     I cannot pinpoint how potential responses were affected by our regimen.

H. P. Witschi;  Would you expand on your comment that rats are not appropriate
experimental animals for pulmonary studies?

K. M. Schaich;  I don't think rats are adequate models for human response,
since blood studies in humans have consistently demonstrated some effects in
red cells and red cell enzymes while rat studies have not.  Compared to
humans, I don't expect as much 03 or oxidative potential to get through the
lung to the blood because of the rat lung structure and because of the
biochemical protective mechanisms which are much more active in rat lung.

H. P. Witschi;  Oxygen readily diffuses through membranes.  How do you know
that 03 isn't acting similarly, without reacting in the membrane?

K. M. Schaich;  I don't think that 03 itself gets into the blood; the more
likely effective toxic agent(s) is some breakdown product of 03 or some
secondary oxidation product.  There has been a long-standing dispute regarding
the disposition of 03 in the respiratory tree and what happens to 03 when it
traverses membranes (if it does).  Ozone is so reactive that when it contacts
anything (e.g., surfactant), it decomposes or interacts immediately with a
sensitive group.  Therefore, in the rat it is unlikely that 03 itself, having
to traverse the nasal passages as well as the upper respiratory tree, would
ever get into distal portions of the lung.  Oxidation products such as
superoxide anion may get through to the air/blood interface, but that species
is also very reactive.  I think it is more likely that secondary oxidation
products of lipids in the membranes—hydroperoxides or epoxides or product
aldehydes—provide the mobile and more stable oxidative potential that may be
picked up by the blood and carried on to initiate subsequent damage.

Comment;  Ozone is more reactive than oxygen, and oxygen doesn't diffuse very
well through the upper airway-  That region, therefore, is not an important
locus for gas exchange.  As others have shown in terms of the depth to which
one can see damage in the lung following cytotoxic 03 exposures, damage hardly
gets below the basal membrane, so in thick tissues not much gets through.
Ozone and gas diffusion in the lungs is quite different than irn cell
suspensions in vitro.

J. A. Graham;  Using a different systemic parameter and agents other than O3,
we've found that the female is much more sensitive than the male.  Were you
using males or females?

K. M. Schaich;  We were using females in the beginning; we are using males
now.  We started with Sprague-Dawley rats and now we're using Fischer rats.
For the first couple of years we used "scrap rats" from wherever we could get
them.
                                     123

-------
 11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg


     When we did have male rats, our impressions were that they were less
 sensitive.  But they were usually also older, so that confounds the issue.

 Comment;  Some of these parameters are reminiscent of aging red cells.  Have
 you determined the age distributions of the red cells you sampled?  Were they
 comparable?

 K. M. Schaich;  No, we had no hematological backup in the early studies.  You
 can see, however, there is a long list of things that we're beginning to
 follow up now.  In a study that is just beginning, we will look directly at a
 variety of hematological parameters in order to understand the red cell
 population being monitored.  If we don't have the same red cell populations
 and type distributions in exposed and normal rats, then obviously we should
 see some differences in the enzyme responses.  We are not presently equipped,
 however, to measure red cell kinetics and turnover, which would provide a more
 direct answer to your question.

 E. Hu;  Are you testing the effects of oxidants on the tumorous animal, too?

 K. M. Schaich;  Shellabarger did those studies and we worked in conjunction
 with him.  Those studies were essentially negative.  There were some
 borderline positive responses but nothing consistent.

 Question;  There are two forms of glutathione peroxidase.  Which form were you
 talking about?

 K. M. Schaich;  I don't know that we differentiate; we use standard
 glutathione peroxidase assays for the red blood cells as outlined by E.
 Beutler  [1975.  Red Cell Metabolism:  A Manual of Biochemical Methods.  Grune
 and Stratton, New York].

 Question;  How do you measure?  Do you use hydrogen peroxide?

 K. M. Schaich;  No, t-butyl hydroperoxide.

 Question;  Because there are different levels of the peroxidase and other
 materials that might decompose 03 in the blood, you're assuming that it is in
 fact the 03 that's causing the problem.  But maybe the decomposition products
 are effecting the damage.

 K. M. Schaich;  I don't think the O3 is causing damage directly in the red
 cells.  As I said before, I believe it to be more likely that ozonides or
 other oxidation products carry the damage potential from the lungs to other
 tissues.  That is why we initially looked only for oxidation effects,
 especially in lipids.

Question;  Since peroxidase activities are higher in rats than in humans or  in
primates, is it possible that you saw even more effect, because breakdown
products enter the bloodstream faster?
                                     124

-------
11.  EFFECTS OF OZONE ON RAT ERYTHROCYTES	Schaich and Borg


K. M. Schaich;  No.  Hydroperoxides are decomposed by peroxidases to products
with lower oxidation potential and generally more hydrophobia character.
These products, therefore, are less likely to be picked up by the bloodstream,
and are less damaging if they are.  More rapid peroxide decomposition,
furthermore, means that lower concentrations of the toxic hydroperoxides will
be available for transport by the bloodstream.  Hydrogen peroxide and lipid
hydroperoxides that do get into circulation provide quite active substrates
for the iron in the blood, forming an in vivo Fenton-like system which
produces, at a minimum, the more reactive and cytotoxic hydroxy and alkoxy
radicals.  Thus, in terms of potential damage it should be of advantage to
keep hydroperoxide levels as low as possible.

guestion;  These agents occur in the blood, and your measurements were in the
blood?

K. M. Schaich;  Yes.  Let me restate my reasons for contending that rats are
poor models for providing data which may be directly extrapolated to man—at
least in terms of pulmonary and systemic oxidant effects.  First:  Rat lung
tissue has a substantially greater capacity than does the human lung for
decomposing oxidation products formed during normal metabolism or after
exposure to gaseous or other oxidants.  Thus, lower concentrations of the most
potently damaging products are available to be picked up and circulated by
the blood.  Consequently, initial challenge to circulating red blood cells
is lower in rats than in man.  Second:  Levels of glutathione peroxidase in
red cells are much higher in rats than in man.  Since glutathione peroxidase
is located at the membrane and acts on organic peroxides as well as hydrogen
peroxide, effects in rat red cells will be counteracted at the membrane
level—at the site of initial oxidant reaction.  In contrast, in man higher
levels of catalase alone are ineffective in counteracting sustained oxidant
stress, especially at the membrane level.  Damage to the membrane allows
cytotoxic agents to be absorbed or otherwise taken up by the cell, thereby
affecting metabolic machinery and other components inside the cell.  Thus,
the site of major reactions and red cell damage are quite different for rats
as compared to man.  As a consequence, patterns of response to oxidant
challenge would not be expected to be comparable.
                                      125

-------
                  12.  PATHOGENESIS OP CHRONIC LUNG DISEASE:
                    THE ROLE OF TOXICOLOGICAL INTERACTION

                              Hanspeter Witschi

                               Biology Division
                        Oak Ridge National Laboratory
                              Post Office Box Y
                             Oak Ridge, TN  37830
INTRODUCTION



     It is known that simultaneous or successive exposure to two or possibly

several toxic agents will sometimes elicit a biological response that is

considerably more severe than the effects produced by one chemical alone

(National Research Council 1980).  To consider toxicological interactions

should be an important element in attempts to estimate and evaluate the health

effects of air pollution.  These attempts require an understanding of the

nature and mechanism of each interaction.



     Our group recently developed an experimental model to help describe and

analyze how an acute interaction between two toxic agents produces a chronic

form of lung damage.  The two compounds are the antioxidant butylated

hydroxytoluene (BHT) and oxygen (02).  We hope that it will be possible to

examine whether the principles governing the interaction between BHT and ©2
                                     126

-------
t2.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE	Witschi






are of general significance and might be applied .to better understand some




health effects of oxidants.









BACKGROUND









     In mice BHT produces extensive and uniform death of the Type I alveolar




cells.  Within 24 h after a single administration of BHT, necrotic Type I




cells disintegrate and slough off, leaving large areas of denuded basement




membrane.  If the animals are left undisturbed, the changes in the alveolar




zone are successfully repaired within the next 2 to 6 d.  Recovery is




accomplished, as in many other forms of acute toxic lung injury, by




proliferation of Type II alveolar cells which eventually transform into Type I




cells.  A normal air/blood barrier is restored ~1 week after BHT, and the




lungs of BHT-exposed animals look essentially normal again.  Therefore, a




single acute episode of toxic lung damage appears to be without any serious




long-term consequences (Hirai et al. 1977; Adamson et al. 1977).








     It is, however, possible to interfere with tissue recovery.  Once Type II




alveolar cells begin to divide following BHT-induced lung injury, they appear




to become quite susceptible to elevated concentrations of ©2 in the inspired




air and may be killed.  Oxygen concentrations as low as 50% have been found to




have an adverse effect on epithelial cell division in lung.  Interestingly,




dividing interstitial or dividing capillary endothelial cells appear resistant




to the cytotoxic action of O2.  This observation led us to predict that the




presence of O2 following lung injury would compromise reepithelialization of






                                     127

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE	 Witschi


the alveolar zone without, however, interfering with the proliferation of

fibroblasts.  A possible consequence would be the development of fibrosis
              t >
(Witschi and Cote 1977a).  The experiments summarized below confirmed this

hypothesis.



METHODS AND RESULTS



     Animals treated with BHT and exposed to O2 during the phase of epithelial

cell proliferation developed extensive and diffuse fibrosis within 1 to 2

weeks.  Quantitative determination of lung hydroxyproline showed that exposure

to 50% to 80% 02 alone for up to 6 d did not increase lung collagen.

Injection of BHT alone produced only slight fibrosis.  However, combined

treatment with BHT and 02 increased total lung collagen 2 to 3 times above

controls.  The combined effects of BHT and 02 were synergistic (Haschek and

Witschi 1979).  Histopathology showed the animals to have developed diffuse

interstitial fibrosis with many ultrastructural features common to the human

disease known as Hamman-Rich syndrome (Brody 1980).  Similar observations were

made when epithelial cell proliferation was inhibited by low and nonfibrogenic

doses of x-rays (<200 rads) (Haschek et al. 1980).  Animals exposed to 02 or

to x-rays once reepithelialization of the alveolar zone was complete showed no

development of fibrosis, however (Haschek and Witschi 1979; Haschek et al.

1980).



     In subsequent studies, we used lung hydroxyproline as a quantitative end

point to further study this interaction between a bloodborne lung toxic  agent


                                     128

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE	Witschi






and inhaled 02.  Several observations were made.  Abnormally high levels of




total lung collagen persisted for up to 6 mo  (possibly longer) after a single




episode of exposure to BHT followed by 02 or  x-rays  (Witschi et al. 1980).




Interestingly, lung morphology appeared to return to practically normal,




although lung collagen levels remained almost twice  as high as in controls.




The influence of persisting high collagen concentrations on lung function




remains to be established.  Increased hydroxyproline was also found if




exposure to 70% 02, a concentration which in  itself  does not produce fibrosis,




was limited to 48 h only.  The maximum fibrogenic response developed when




animals were exposed to 02 immediately after  BHT.  Delay of 02 exposure for 48




or even 96 h still produced fibrosis, but its development was much diminished.




Significant fibrosis also developed if animals given BHT were exposed for 2 d




to O2 concentrations of >70%, for 3 d to >60%, or for 6 d to >50%; only 40% O2




had no significant effect.  The most frequently used dose of BHT (400 mg/kg)




itself produced some accumulation of lung hydroxyproline.  If the dose of BHT




was lowered to 200 or 100 mg/kg, no fibrosis  developed.  Yet exposure to 70%




02 following a nonfibrogenic dose of BHT was  accompanied by the development of




fibrosis (Witschi et al. 1980b).









DISCUSSION








     From these experiments we  draw several conclusions.  A form of chronic




lung damage—interstitial fibrosis—may result from  an acute interaction of




two toxic agents in the lung.   Timing is of crucial  importance for the




occurrence of this interaction.  Fibrosis only develops if O2, or possibly any






                                     129

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE	   Witschi






other toxic agent, is present during the phase of epithelial cell




proliferation following lung injury.  Oxygen will not produce fibrosis once




reepithelialization is complete.  If it develops, however, fibrosis persists




for up to 6 mo or longer.  The interaction between (>2 and BHT does not occur




because one agent directly enhances the action of the other (Williamson et al.




1978).  Rather, 02 adversely affects a cell population that is different from




the original BHT target cells and which has been called into action to repair




the initial damage.









     It is conceivable that this scenario for the interaction of BHT and 02 in




mouse lung reflects a general principle underlying the development of chronic




lung damage.  Ample experimental evidence documents that many toxic inhalants




and numerous bloodborne agents will produce Type I alveolar cell deaths




followed by Type II cell proliferation.  Several chemicals are also known to




interfere with cell proliferation in the lung (Witschi and Cote 1977b).  In




principle, fibrosis might therefore develop under anv circumstances in which




exposure to a first agent causing alveolar cell death is followed by exposure




to a second cytotoxic agent, provided a critically ordered sequence of




exposure is observed.  It does not seem likely that the temporal relationships




needed to produce fibrosis (according to our proposed mechanism) always occur.




However, if the right conditions are met, the resulting lesion will likely




persist.  Although single episodes might easily go unnoticed, repeated




episodes would eventually result in diffuse and recognizable lung damage.
                                     130

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE     	Witschi






CONTINUING STUDIES









     Attempts have begun to verify our hypothesis with studies of agents other




than BHT and 02.  Thus far, we have not been able to produce fibrosis by




exposing animals first to 2 ppm ozone (O3) followed by 70% O2.  There are




three possible explanations:  (1) our proposed mechanism for pathogenesis of




lung fibrosis is not as universal as assumed;  (2) the cell kinetics following




exposure to 03 are different  from those following exposure to BHT, and our




timing of 02 exposure following O3 was wrong; or (3) the initial lesion




produced by exposure to 03 was too small and comparable only to lesions




produced by low doses of BHT  (50 mg/kg), which have not produced fibrosis.




The third possibility needs to be studied in more detailed experiments; such




studies could well lead to information on no-effect or negligible levels of 03




exposure.  On the other hand, we have found that most animals treated with 400




mg/kg BHT will die if subsequently exposed for 48 h to 1 ppm 03.  The




survivors develop extensive fibrosis.  Animals with acutely damaged lung are




thus more susceptible to 03.  This illustrates that the adverse effects of a




common oxidant on a diffusely damaged lung may be much more serious than the




effects on a healthy organ.   Finally, it will be possible to study the effects




of airborne toxic agents on animals in which diffuse interstitial fibrosis has




been previously produced by exposure to BHT and 02.  These studies will permit




examination of the effects of oxidants in an animal model which mimics a not




uncommon form of human lung disease, chronic interstitial fibrosis.
                                      131

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE	Witschi
REFERENCES
Adamson, I. Y. R., D. H. Bowden, M. G. Cote, and H. P. Witschi.  1977.  Lung
     injury produced by butylated hydroxytoluene.  Cytodynamic and biochemical
     studies in mice.  Lab. Invest., 36:26-32.

Brody, A. R.  1980.  Personal communication.

Haschek, W. M., K. R. Meyer, R. L. Ullrich, and H. P. Witschi.  1980.
     Potentiation of chemically induced lung fibrosis by thorax irradiation.
     Int. J. Radiat. Oncol. Bio. Phys., 6:449-455.

Haschek, W. M., and H. P. Witschi.  1979.  Pulmonary fibrosis—A possible
     mechanism.  Toxicol. Appl. Pharraacol., 51:475-487.
                                        /^ *
Hirai, K. I., H. P. Witschi, and M. G. Cote.  1977.  Electron microscopy of
     butylated hydroxytoluene-induced lung damage in mice.  Exp. Mol. Pathol.,
     27:295-308.

National Research Council.  1980.  Report of Panel on Evaluation of Hazards
     Associated with Maritime Personnel Exposed to Multiple Cargo Vapors.
     Washington, D.C., National Academy of Sciences (in press).

Williamson, D., P. Esterez, and H. P. Witschi.  1978.  Studies on the
     pathogenesis of butylated hydroxytoluene-induced lung damage in mice.
     Toxicol. Appl. Pharmacol., 43:577-587.

Witschi, H. P., and M. G. Cote.  1977a.  Inhibition of butylated
     hydroxytoluene induced mouse lung cell division by oxygen:  Time-effect
     and dose-effect relationships.  Chem.-Biol. Interactions, 19:279-289.

Witschi, H. P., and M. G. Cote.  1977b.  Primary pulmonary responses to toxic
     agents.  CRC Grit. Rev- Toxicol., 5:23-66.

Witschi, H. P., W. M. Haschek, K. R. Meyer, R. L. Ullrich, and W. E. Dalbey.
     1980a.  A pathogenetic mechanism in lung fibrosis.  Chest, 78:3953-3998.

Witschi, H. P., W. M. Haschek, A. J. P. Klein-Szanto, and P. J. Hakkinen.
     1980b.  Potentiation of diffuse lung damage by oxygen:  Determining
     variables.  Am. Rev. Resp. Dis. (in press).
                                     132

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE                Witschi
WORKSHOP COMMENTARY
D. L. Coffin;  I gathered from  the paper  that  there  seems to be a correlation
of this fibrosis with the degree of  epithelial reaction.  In view of the
apparently strong interaction between macrophages and  fibroblastic
hyperplasia, have you examined  the status of the macrophage as an alternate
means of mechanism?

H. P. Witschi;  That's a very good point. No, we haven't looked very much at
the macrophages.  One of the reasons is that we were able to see only
comparatively few macrophages in the damaged lungs.  Dr. Haschek performed the
histology and can confirm this.

     I do not think, in this particular instance, that macrophages are
involved.  Rather,  I think  that Q£ prevents epithelialization, which results
in the formation of fibroblasts.  There are some experimental studies showing
exactly the  same thing.  Some very convincing  results  were reported by
Terzaghi  [Terzaghi, M., P.  Nettesheim, and M.  L. Williams.  1978.
Repopulation of denuded tracheal grafts with normal, preneoplastic and
neoplastic epithelial cell  populations.   Cancer Res.,  38:4546-4553].  Denuded
tracheal grafts, when implanted, filled up with fibroblasts in no time.
Inoculation  with epithelial cells of a certain critical number resulted in no
fibrosis.

Question;  Did you  try altering the  order in which you gave BHT and 03, to see
if the 03 set up a  system for BHT damage?

H. P. Witschi;  No, we have not done this yet.

Question;  You mentioned that you thought this lesion  might disappear after 6
mo to 1 yr.  If it's fibrosis,  why would  you expect  it to disappear?

H. P. Witschi;  What persist are  increased levels of hydroxyproline, which
probably  indicate collagen. Interestingly enough, the change in
histopathology at 6 mo is not very impressive, although the lung still has a
large amount of collagen.

Question;  Could this be just a .matter of turnover:  you have more of a
turnover  in  your damaged lung than in the undamaged  lung?  If so, it isn't
fibrosis.  It's just that more  turnover results in more residual
hydroxyproline.

H. P. Witschi;  No, we have increased synthesis and  diminished degradation.

Question;  So you had total hydroxyproline increase?

H. P. Witschi;  Yes.
                                      133

-------
12.  TOXICOLOGICAL INTERACTIONS IN CHRONIC LUNG DISEASE	Witschi


Comment;  I don't quite understand.  Let me put it this way:   If you were to
do, say, connective tissue stains, would that correspond to your
hydroxyproline?  In other words, you're saying that you don't have increased
collagen tissue but that you do have increased hydroxyproline.  That's
confusing.

H. P. Witschi;  It's merely a question of semantics.  What is measured after
hydrolyzing the lung is hydroxyproline, or an index for collagen.   I think
~90% of all lung hydroxyproline is in collagen.
                                    134

-------
                 13.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
            BY THE RESPIRATORY UNIT,  OAK RIDGE  NATIONAL LABORATORY

                   PART  1.   RAT TRACHEAL TRANSPLANT  SYSTEM

                                 Ann C.  Marchok

                                Respiratory Unit
                         Oak  Ridge National Laboratory
                               Post  Office Box Y
                              Oak Ridge,  TN  37830
INTRODUCTION AND BACKGROUND



     One overriding  theme  of the  research effort in the Respiratory Unit at

Oak Ridge National Laboratory  (ORNL)  is the study of the morphogenesis of lung

cancer.  A major part of this  effort  is directed towards developing and

utilizing experimental animal  model systems to identify agents which might

enhance tumor induction or progression in the respiratory tract.



     Currently, three approaches  are  used.  The first approach is that of

inhalation experiments in  which animals pretreated with carcinogens are

exposed to irritant  gases.  The tracheal washing technique developed at ORNL

by Schreiber et al.  (1975) and further developed there by Yarita et al. (1978,

represents a second  approach.  With this system, we plan to precondition the

respiratory epithelium before  exposure to aerosols.  It should be emphasized

that this model is particularly amenable to studying any interrelationship


                                      135

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM	Marchok






between the nutritional state of the animal and susceptibility.  Studies with




both of these systems are discussed by Dalbey (Chapter 14 of this volume).









     A third approach is the rat tracheal transplant system developed in our




laboratory by Kendrick et al. (1974).  The rat trachea was chosen as a model




because of its structural similarity to the human bronchus.  This report




discusses current and proposed studies with the rat tracheal transplant, most




of which fall under the EPA umbrella.








     An inherent limitation in systems in which the carcinogen is introduced




into the respiratory tract by inhalation or the more common intratracheal




injection is the unpredictability of the site of tumor development.  It is




also impossible to determine either the carcinogen dose or the duration of




carcinogen exposure for any given site of the respiratory tract.  Both of the




model systems developed in our laboratory (i.e., the tracheal washing




technique and the tracheal transplant) have the advantage that the target site




for the carcinogen as well as for the cocarcinogen is well defined.  An




accurate dose of the agent(s) can be delivered, the site of tumor development




can be predicted (Griesemer et al. 1977; Nettesheim et al. 1977; Pal et al.




1978), and multiple exposures of the same or different agents can be carried




out.
                                     136

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM	Marchok






CONTROLLED DELIVERY OF THE TEST AGENT(S)









     Obviously, delivery of the agent  is a key issue  in this model system.




Ideally, one would like to test very low levels of substances and have a high




degree of control over the rate and duration of exposure.  The first tumor




induction studies were carried out with polycyclic hydrocarbons (PCH) released




from beeswax pellets.  The rate of PCH release was determined by measuring the




material remaining in the pellet at different time intervals following




placement in the transplant.









     B. Pal of our laboratory has developed an in vitro assay system to




pretest agent release rates prior to placement of pellets in transplants.




This method consists of  shaking the carcinogen pellets in a flask containing




fetal  calf serum at 37°C.  The serum is changed at time intervals and the




carcinogen released into the serum is  determined by radio assay.  As shown in




Figure 13-1, the release rates in vitro"and _in vivo are




concentration-dependent  (i.e., as the  concentration in the pellet diminishes




with time, the amount released per time diminishes) .  The rate of in vivo




release from pellets containing 100 yg PCH is faster  (100% released in 2




weeks) than that from pellets containing ,1000 yg PCH.  At low concentrations




of PCH, in vivo and jLn vitro release rates are similar.  At high PCH




concentrations, in vivo  rates are markedly slower than in vitro rates.









     These studies prompted exploration of other possibilities for altering




and reducing PCH release.  The first to be explored was the effect of adding







                                      137

-------
13.   RAT TRACHEAL TRANSPLANT SYSTEM
                                                                       Marchok
                              100 {A
                                       10    20
                                      TIME (days)
30
Figure 13-1.  Cumulative release of benzopyrene (BP) from beeswax pellets.
              A, pellets containing 100 yg BP.  o, in vitro release; •, i.n
              vivo release in tracheal grafts.  Each point represents the
              mean observation from 4 pellets _in vitro ± S.D. and 6 pellets
              in vivo ± S.D.  B, pellets containing 1000 yg BP.  o, ±n vitro
              release; •, jln vivo release in tracheal grafts.  Each point
              represents the mean observation from 3 pellets in vitro ± S.D.
              and 12 pellets in vivo ± S.D.  C, pellets containing 1000 yg BP
              implanted s.c.  Each point represents the mean observation from
              6 pellets ± S.D.  From:  Pal et al. (1978).
                                     138

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM	       Marchok






cholesterol or cholesterol laurate to achieve beeswax:cholesterol  proportions




ranging from 1:1 to 1:9.  As seen in Figure  13-2, cholesterol was very




effective in retarding BP release.  At a beeswax:cholesterol  concentration of




1:9, the average amount of BP released in vitro within  1 week from pellets




containing 100 ug BP was ~13% as compared to ~90% released during the same




time from pure beeswax pellets.  A similar retardation  of BP release was




observed in the jln vivo studies.









     Our current effort follows similar lines.  Pal is  attempting to modify




the cholesterol and is pursuing many other kinds of substances, including




various polymers and biodegradable materials, as matrices to control the




release of agents having a wide range of chemical properties.  The goal is to




examine the inductive or cocarcinogenic effects of such agents on the




respiratory epithelium.  Since the model is  still under development, we




currently use commonly known carcinogens and cocarcinogens.  We already have




an indication that slow release of a particular carcinogen dose is much more




tumorigenic than rapid release.  One reason  for this, most likely, is a




reduction in the initial toxic effect.









INCREASED SPECIFICATION OF END POINTS








     The common end point in testing agents  for induction and promotability of




cancer is actual development of a tumor in animal survival studies.  There is




a  great need for end points that are less costly and time-consuming.  Our




laboratory is attempting to reduce this end  point in time and actually use






                                     139

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM
                                                                       Marchok
                           100-
                               0
 10     20
TIME (days)
30
Figure 13-2.  Cumulative release of BP from pellets with modified matrix, at
              100 yg BP per pellet.  A, _in vitro studies,  o, beeswax pellets;
              A, beeswax:cholesterol laurate pellets at a ratio of 1:9; Q,
              beeswax:cholesterol pellets at a ratio of 1:1; H, beeswax:
              cholesterol pellets at a ratio of 1:3; I, beeswax:cholesterol
              pellets at a ratio of 1:9.  Each point represents the mean of 2
              to 5 observations ± S.D.  B, in vivo studies in tracheal grafts.
              o, beeswax pellets; <», beeswax:cholesterol pellets at a ratio
              of 1:3; •, beeswax:cholesterol pellets at a ratio of 1:9.  Each
              point represents the mean of 4 to 6 observations ± S.D.  From:
              Pal et al. (1978).
                                     140

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM	  	     Marchok


numbers and kinds of lesions identified by ways other than and/or in addition

to morphology-  The design of these experiments is as follows:



     1.  Expose transplants to 200 yg dimethylbenzanthracene (DMBA).

     2.  At set times, sample transplants and cut into 2 x 3 ml
         explants.

     3.  Place into organ culture.  After 24 h, collect media and
         prepare the exfoliated cells for cytopathologic diagnosis.
         Fix explant in one set for pathologic diagnosis.  Compare.
         Place second  set of explants into outgrowth culture to
         determine other markers of cellular alterations.



     In studies funded by the National Cancer Institute, we are beginning to

 follow the fate of cell populations from specific lesions.  The properties to

 be determined are:



     1.  Rate of epithelial outgrowth

     2.  Maintenance of primary cultures in suboptimal media
         (increased jLn vitro growth capacity)

     3.  Focal morphological changes in the primary cultures

     4.  Subculturability of the primaries into cell lines

     5.  Markers of transformation in the cell lines:

         (a)  growth in agar

         (b)  formation of tumors

     6.  Correlation of lesion severity as determined
         morphologically to the time of appearance of markers
         of transformation
                                      141

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM	Marchok


     The basic methodology for these studies has been described previously

(Marchok et al. 1977, 1978).  So far, we have observed the following:
     1.  A very early marker of transformation is an increased
         capacity to survive and grow in vitro.

     2.  The brief in vivo carcinogen exposure "initiated" some
         cells, and the processes initiated in vivo proceeded
         in vitro, leading to the emergence of some neoplastic
         cell populations.

     3.  Since the primary cell cultures and most cell lines were
         not malignant when first tested, but some became so later,
         progression of the neoplastic disease must have occurred
         in vitro.

     4.  There were dose-related effects:

         (a) More primary cultures and cell lines could be
             established from tracheas exposed to a high
             (640 vg DMBA) than a low (165 yg DMBA) dose.

         (b) There were differences in the in vitro "latency"
             periods (i.e., the time cells are maintained before
             becoming oncogenic).

         (c) Tumor induction time (i.e., the time from cell
             inoculation to development of a palpable tumor)
             correlated in most cases with j.n vitro latency.
             Tumor induction times were 9 to 60  d for cell lines
             with short in vitro latency periods and from 100 to
             250 d for cell lines with long jLn vitro latencies.

         (d) The combined in vitro-in vivo tumor latency time
             was similar to tumor induction times found in the
             tracheal transplants in vivo.
     With this experimental approach,  we should achieve the following:
     1.   Direct determination of the fate of specific lesion
         types in terms of:
                                     142

-------
13.   RAT TRACHEAL TRANSPLANT SYSTEM          	 	Marchok


         (a)  acquisition of the neoplastic state

         (b)  tumor latency

     2.  Identification of other markers  (end points) for
         determining the tumorigenic potential of cells

     3.  Identification of cell populations at risk in the
         respiratory system of experimental animals

     4.  Use of these end points to test  the carcinogenic
         potential of environmental agents on the respiratory
         system

     5.  Application of the same criteria to the cytopathology
         and to biopsies of human  tissue
     Looking ahead, it  should  be  emphasized that  this approach can be utilized

 regardless of the  form  of  exposure or type of  agent(s) tested.



 REFERENCES
 Griesemer, R. A.,  P.  Nettesheim,  D.  H.  Martin,  and J. E. Caton, Jr.  1977.
     Quantitative  exposure of grafted rat tracheas to
     7,12-dimethylbenz(a)-anthracene.  Cancer Res.,  37:1266-1271.

 Kenrick, J., P.  Nettesheim, and A.  S. Hammons.   1974.   Tumor  induction in
     tracheal grafts:  A new experimental model for  respiratory carcinogenesis
     studies.  J.  Natl.  Cancer Inst., 52:1317-1325.

 Marchok, A. C.,  J. C. Rhoton, R.  A.  Griesemer,  and P. Nettesheim.   1977.
     Increased in  vitro growth capacity of tracheal  epithelium exposed in vivo
     to 7,12-dimethylbenz(a)anthracene.  Cancer Res., 37:1811-1821.

 Marchok, A. C.,  J. C. Rhoton, and P. Nettesheim.  1978.  In vitro development
     of oncogenicity in cell lines established  from  tracheal  epithelium
     preexposed  in vivo to 7,12-dimethylbenz(a)anthracene.  Cancer  Res.,
     38:2030-2037.

 Nettesheim, P.,  R. A. Griesemer,  D.  H.  Martin,  and J. E. Caton, Jr.  1977.
     Induction of  preneoplastic and neoplastic  lesions  in grafted rat tracheas-
     continuously  exposed  to benzo(a)pyrene.  Cancer Res., 37:1272-1278.


                                      143

-------
13.  RAT TRACHEAL TRANSPLANT SYSTEM        	Marchok


Pal, B. C., D. C. Topping, R. A. Griesemer, F. R. Nelson, and P. Nettesheim.
     1978.  Development of a system for controlled release of benzo(a)pyrene,
     7,12-dimethylbenz(a)anthracene, and phorbol ester for tumor induction in
     heterotopic trachea1 grafts.  Cancer Res., 38:1376-1383.

Schreiber, H., K. Schreiber, and D. H. Martin.  1975.  Experimental tumor
     induction in a circumscribed region of the hamster trachea:  Correlation
     of histology and exfoliative cytology.  J. Natl. Cancer Znst.,
     54:187-197.

Yarita, T., P. Nettesheim, and M. L. Williams.  1978.  Tumor induction in the
     trachea of hamsters with N-nitroso-N-methylurea.  Cancer Res.,
     38:1667-1676.
WORKSHOP COMMENTARY
D. L. Coffin;  In the work that showed an interaction between DMBA and
asbestos, how was the asbestos applied to the implanted tracheas?

A. C. Marchok;  The asbestos was released from gelatin pellets.

D. L. Coffin;  And the DMBA was in the beeswax?

A. C. Marchok;  Yes.  They were exposed sequentially.

D. L. Coffin;  Which came first?

A. C. Marchok;  The DMBA came first as the initiator; asbestos was evaluated
as a promoter.

D. L. Coffin;  How long after DMBA was the asbestos administered?

A. C. Marchok;  The gelatin pellets containing asbestos were inserted into the
tracheas 4 weeks after the start of DMBA exposure.
                                     144

-------
                14.  OVERVIEW OP CURRENT AND PLANNED RESEARCH
            BY THE RESPIRATORY UNIT, OAK RIDGE NATIONAL LABORATORY

     PART 2:  TRACHEAL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS

                               Walden E. Dalbey

                               Respiratory Unit
                        Oak Ridge National Laboratory
                              Post Office Box Y
                             Oak Ridge, TN  37830
INTRODUCTION



     This report provides an overview of current Oak Ridge National Laboratory

studies on inhalation of oxidant gases.  Three main areas are discussed:   (1)

irritant gases as cofactors in nitrosamine-induced tumorigenesis, (2)

modulation of tumor incidence using the trachea! washing model for tumor

induction, and (3) the role of frequency and duration of oxidant gas exposure

in pulmonary toxicity.



     Since man is rarely exposed to doses of chemical carcinogens large enough

to result in tumors by themselves, investigations of factors which could

influence tumor incidence are of great practical importance.  Such potential

enhancement is the focus of the first two studies described below.
                                     145

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS	Dalbey






IRRITANT GASES AS COFACTORS IN NITROSAMINE-INDUCED TUMORIGENESIS









     In this study, subcutaneous administrations of diethylnitrosamine (DEN)




were performed to induce tumors in the respiratory tracts of male Syrian




golden hamsters.  This use of a systemic carcinogen was designed to decrease




the variability in tumor response that is apparently inherent with




intratracheal instillation of carcinogens.








     Hamsters received 10 weekly injections of 0.5 mg DEN.  In addition,




groups of animals were exposed to either nitrogen dioxide (N<>2) or




formaldehyde (HCHO) for life.  These gases were chosen as model lower and




upper respiratory tract irritants, respectively.  The irritant treatment was




further subdivided so that irritant exposures began at one of two times in




relation to DEN treatment:  (1) 48 h prior to each weekly DEN injection, so




that the respiratory epithelium was actively proliferating during the time of




carcinogen treatment, or (2) 2 weeks after the final DEN injection.  Irritant




gas exposures were given weekly for the lifetime of each animal.  Ancillary




data indicated that such weekly exposures would result in epithelial necrosis




and proliferation subsequent to each exposure analogous to that occurring




after the initial exposure.  In other words, the animals would demonstrate no




adaptation with weekly exposures.









     After each animal's death, the incidence of respiratory tract tumors was




evaluated by means of a clearing technique rather than histology.  Fixed




tissue was stained with Wright1s stain and rendered semitransparent by






                                     146

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT  INHALATION EXPERIMENTS
Dalbey
dehydration and eventual  transfer  into methyl  salicylate.  Areas of dense cell




aggregation (tumors) stained  darkly and  were readily visible under a




dissecting microscrope.   Observations  of tumors at this subgross level




correlated well with subsequent microscopic evaluation, except that several




tumors found at the subgross  level would not have been observed during routine




histological procedures.   There were no  false  positive or negative tumor




identifications during the histological  evaluation.  All observed tumors were




adenomas; no invasive tumors  were  observed.  As of this writing, the




experimental data are still tentative.   We have not yet fully evaluated the




incidence of tumors in control  animals.









     We obtained data on  tumor  incidence in animals exposed to DEN and either




10 ppm NC>2 or  30 ppm HCHO. The incidence of nasal tumors was low and not




influenced by  irritant exposures.   Most  of the observed tumors were in the




larynx and trachea, with  a smaller percentage  occurring in the lung.  With the




DEN dose used, the percentage of tumor-bearing animals was high, and no




differences in percentage of  tumor-bearing animals were observed between




experimental groups.  However,  the number of tumors per tumor-bearing animal




was significantly increased in  the tracheas of animals that were concurrently




exposed to HCHO and DEN.   Also, there  was a significant increase in total




respiratory tract tumors  in animals concurrently receiving N02 exposures and




DEN treatment, although there was  no significant increase in either the




larynx, trachea, or lung  individually.
                                      147

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS	Palbey






     It appeared from these observations that both irritant gases acted to




enhance tumor incidence when the carcinogen was administered at a time of




epithelial proliferation resulting from irritant gas exposure.   No "promoting




activity" was observed.








     Partly in view of this lack of observed promoting activity, and also in




view of the relative paucity of demonstrated promoting activity in the




respiratory tracts of whole animals, we entered into the second major area of




work:  modulation of tumorigenesis using the tracheal washing technique for




tumor induction.








TRACHEAL WASHING SYSTEM








     Marchok (Chapter 13 of this volume) has discussed the development of the




tracheal washing technique at our laboratory.  The anesthetized hamster is




placed on its back and a small cannula with a double lumen is inserted through




the larnyx.  A buffered solution containing N-methyl-N-nitrosourea (NMU) is




injected via a syringe pump through the outer lumen of the cannula.  The




solution moves 5 mm down the tracheal epithelium before reaching the elongated




tip of the inner tube of the cannula, where suction draws the solution out of




the trachea.  Thus, by placing the cannula tip at a standard location within




the trachea, a particular site of tracheal epithelium can be repeatedly washed




with NMU solution.
                                     148

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT  INHALATION EXPERIMENTS	Daibey






     This model of tumor induction  offers several advantages over




intratracheal instillation or  systemic  administration of carcinogen for




whole-animal studies.  These advantages include:  more quantitative delivery




of the carcinogen to a specific site  of respiratory epithelium; a small,




predictable location of tumor  development to  facilitate histological sampling




of both cancerous and precancerous  lesions; and  induction of carcinomas in




addition to noninvasive lesions.  We  believe  the full potential of this system




has not yet been realized, and we are currently  utilizing it in multifactorial




studies.  The first is a promotion  study  in which the tracheal epithelia of




NMU-treated animals are also exposed  to a known  tumor promoter, croton oil.




An aerosol of croton oil provides a noninvasive  means to repeatedly treat the




epithelium.  We also plan  tumorigenesis studies  involving modification of the




physiological state of the animal  (namely, vitamin A deficiency).  The system




is amenable to studies with other agents  or physiological alterations.









EFFECTS OF INHALED NITROGEN DIOXIDE ON  PULMONIC  LESIONS IN RATS









      A third area of work  in our  laboratory concerns the role of duration and




frequency of exposure  to oxidant  gases  in the severity of resulting pulmonic




lesions.  Much of this work relates to  the biological effects of acute peaks




of oxidant concentration.   So  far,  most of these have involved exposure of




specific pathogen free  (SPF) rats to  NO2-









      As observed  in other  laboratories, we noted a  dose-related  increase in




incorporation of tritiated thymidine into pulmonary DNA after  acute NO2






                                      149

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS	Dalbey






exposure.  The peak incorporation occurred ~24 h after exposure and was linear




with the concentration of exposure.  The limits of detection were rather high




(~10 ppm NO2 for a 5-h exposure) when the biochemical method was used to




determine thymidine incorporation.  Also, we observed an increase in




phospholipids after single NO2 exposures, with a peak in pulmonary




phospholipids at ~48 h after exposure.  The disaturated species of




phosphatidylcholine increased preferentially over unsaturated species,




indicating that the increase in phospholipids may be in part due to increased




biosynthesis.








     We conducted additional experiments on the influence of vitamin A




deficiency on the response to acute NO2 exposures, but the data are incomplete




at this time.  Vitamin A is important in maintenance of the normal epithelium.








     We thought it would be of interest to examine further the relative




importance of exposure duration with single acute exposures to NO2.  The end




points used were thymidine incorporation, amount of phospholipids, and




morphology.  We compared 1- and 5-h exposures.  Unlike the linear increases in




thymidine incorporation and amount of phospholipids in relation to NO2




concentration, the values for these end points appeared to plateau as exposure




duration increased.  Presumably the lungs were adapting during the exposure.




We subsequently found that only one 1- or 5-h exposure to NO2 was sufficient




to adapt the animals to a similar exposure on the following day.  That is,




after the second exposure the animals failed to respond with similar evidence




of cell necrosis or proliferation.






                                     150

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT  INHALATION EXPERIMENTS	Dalbey






     Adaptation to NO2 therefore appears  to  occur  rapidly.  We subsequently




investigated the duration of adaptation.   Animals  were reexposed either 1, 3,




or 7 d after adaptation had been initiated,  and  thymidine incorporation was




monitored subsequent to the second exposure.   Data were returned demonstrating




that adaptation was complete with  a 1-d interval,  intermittent with a 3-d




interval, and nonexistent after 7  d between  exposures.  We can assume,




therefore, that once-a-week exposures  would  result not in adaptation to N(>2




but in occurrence of the same  lung events observed after the initial




exposure.









     To our knowledge, the emphysematous  lesions that have been reported with




chronic exposures of rats were observed under  daily exposure regimes in which




the animals were probably in the adaptive state.   We asked whether similar




emphysematous lesions could be produced with weekly exposures of nonadapted




rats.  SPF male Fischer 344 rats were  exposed  once per week, 5 h/d, to 20 ppm




NC>2.  This high concentration  was  used to ensure an observable effect on the




lungs.  The experiment has not been completed, but some data on pulmonary




function are available.








     Some animals were killed  after 6  and 18 mo  of exposure; additional




animals remain in a recovery period.  Pulmonary  function tests were performed




immediately before killing the animals; we obtained data on pulmonary




compliance, lung volume, pulmonary resistance, nitrogen washout, and flow




volume.  These data did not indicate any  change  in pulmonary function




following N02 exposure.  After the pulmonary function tests, lungs from the






                                      151

-------
14.  TRACHAEL WASHING SYSTEM AND OXIDANT INHALATION EXPERIMENTS	Dalbey






animals were fixed under constant pressure and paraffin sections prepared.  No




change in mean linear intercept was observed after 6 mo of exposure; data on




the 18-mo exposure are not yet available.









     If subsequent data obtained in this study are similar,  the negative




results may represent a significant finding that the emphysematous lesions




previously reported to occur with chronic NO2 exposure do not occur under




conditions of sporadic peak exposures.  Our laboratory is currently planning




similar work with ozone, but these plans are readily adaptable to the needs of




EPA.
                                     152

-------
            15.   EARLY STAGES OF RESPIRATORY TRACT CANCER:   A REVIEW

                                Carol A.  Heckman
                                  C.  C.  Scotta
                                  A.  C.  Olson
                                   F. Snyderb

                                Biology  Division
                         Oak Ridge National Laboratory
                               Post Office Box Y
                              Oak Ridge,  TN  37830
 INTRODUCTION



     About 90%  of all human cancers appear to originate in epithelia.  The

 epithelial linings of several organs,  including the respiratory tract and gut,

 are topologically continuous with the  exterior environment.  These linings

 make up the body's first line of defense•against the entry of noxious

 environmental agents.   Simply by reason of their spatial location, epithelia

 may be particularly susceptible to the toxic and carcinogenic effects of

 environmental agents.   Cumulative effects  and delayed effects are of

 particular concern,  because their detection by epidemiologic methods may not

 be possible until  decades after the initial exposure has taken place.  An

 increased understanding of  delayed effects—particularly of carcinogenicity—•

would be useful for  identifying and controlling suspect agents.
Presently at Department of Nutrition, Harvard University School of Public
 Health, 665 Huntington Avenue, Boston, MA  02115.
bOak Ridge Associated Universities, Oak Ridge, TN  37830.
                                      153

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al_.





     To define changes that are specifically correlated with carcinogenesis,




we need to identify features that distinguish between normal and neoplastic




epithelium at the cellular level.  For technical reasons having to do with the




growth requirements of cultured cells, the features best understood at the




cellular level have been studied in fibroblasts and embryonic mesenchymal




cells.  While these models are far from ideal for studying carcinogensis/ they




have the advantage of yielding rapid and quantitative data for a number of




experimental end points.  These systems have been used to define differences




in the surface structure of the cells (Rapin and Burger 1974; Porter et al.




1973; Borek and Fenoglio 1976), contact inhibition of growth (Aaronson and




Todaro 1968; Holley and Kiernan 1968), anchorage-independence (Jones et al.




1976; Shira et al. 1975), loss of the protein fibronectin from cell surfaces




(Hynes 1974; Yamada and Weston 1975), and increased plasrainogen activator




activity (Ossowski et al. 1973; Unkeless et al. 1973).  All of these




differences have been described in the last 10 to 12 years.









     Until quite recently, many of us who work on carcinogenesis had




considered that at least some of the features that discriminate between normal




and transformed fibroblasts would also discriminate between normal and




neoplastic epithelium.  Such features would provide a "tag" for neoplastic




cells that could possibly be applied to a practical problem:  defining




criteria to use in short-term carcinogenesis assays and in early clinical




detection of cancer.
                                     154

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CAKCER	Heckman et al.






     Unfortunately, the prospect of directly applying what has been learned




from fibroblast models to studies of epithelium now seems less promising.




Recent studies have reported  no changes  in  the amounts of fibronectin on cell




surfaces for bladder or for salivary or  mammary gland epithelium (Wigley and




Summerhayes 1979; Yang et al.  1980).  No difference in plasminogen activator




activity has been found for normal and neoplastic mammary epithelium (Wigley




and Summerhayes  1979).  Both  our studies (Heckman and Olson 1979) and those of




McGrath and Medina on the mammary gland  (Voyles and McGrath 1976; Butel et al.




1977) have shown little consistent difference in contact inhibition of growth.




Our group has also shown the  density of  surface features to correlate poorly




with the tumorigenicity of cell lines from  rat tracheal epithelium (Heckman




and Olson 1979).  Finally, anchorage-independence and lectin-mediated




agglutinability, although somewhat more  promising than the other criteria,




have not been very reliable in discriminating between normal and tumor-derived




human cells (Franks  1979).  These two markers are currently being studied by




many other research  groups.








     A few years ago, when we began to look for markers to use in studies of




progressive changes  in the respiratory tract, we considered that many or most




of the biochemical differences that had  been described for fibroblast




transformation models would relate to their differentiated functions and would




therefore be specific to mesenchymal cells. Recent reports by other




investigators indicate that this may be  the case.  Based on this assumption,




we decided to concentrate our efforts on a  few criteria which had been shown




to discriminate  between normal and neoplastic epithelium.  One of these was






                                      155

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al.





the accumulation of ether-linked lipids, which had been described for whole




tumors by Snyder and Wood (1968) but studied rarely since then.  These lipids,




particularly the alkyldiacylglycerols, are present only in trace amounts in




normal tissues of most organs.  A second marker was described by Montesano and




Sanford (Montesano et al. 1977) for normal and neoplastic cell lines grown in




culture.  These investigators compared several lines and found differences in




the morphology and cytology of cells in colonies.









THE ETHER LIPID MARKER









     Before studying sequential changes in carcinogen-treated respiratory




tract tissues, we wanted to confirm that the lipids occur in tumors from the




respiratory tract but not in normal tissues.  For this, four transplantable




rat tumors obtained after benzo(a)pyrene (BaP) treatment were studied.




Thin-layer chromatography showed the alkyldiacylglycerols to be present in the




squamous cell carcinomas but lacking in the normal lung (Figure 15-1).  All




four transplantable tumors had high levels of alkyldiacylglycerols.  We wanted




to determine how soon after carcinogen treatment this marker could be detected




in respiratory tract epithelium.  To do this, we used a lipid precursor,




1-3H-hexadecanol, specifically incorporated into the ether linkage (Topping et




al. 1978), since mass determinations of the lipids would have required large




numbers of animals for quantification, particularly in the case of normal




epithelium.
                                     156

-------
15.  EARLY STAGES OF  RESPIRATORY TRACT CANCER
                                                                 Heckman et al,
Figure 15-1.
Thin-layer chromatogram of lipids extracted from normal lungs
(Lanes 1-2) and transplantable carcinomas (Lanes 3-6).
      TABLE 15-1.   1-3H-HEXADECANOL  INCORPORATION  INTO TRIGLYCEROLS AND
       ALKYLDIACYLGLYCEROLS OF TRACHEAL  IMPLANT EPITHELIUM (DPM X 10~3)
Treatment
                 Time After Precursor Delivery (d)
                                                          21
 16-week reversal)
                                                        42
untreated
control
DMBA
(4-week exposure)
DMBA
(4 -week exposure,
3.01
±1.94 SE
18.70
±12.94
8.34
±2.86
0.72
±0.36/ SE
7.58
±6.12
4.62
±1.16
0.50
±0.06 SE
5.99
±2.84
2.36
±1.88
0.12
±0.05 SE
0.33
±0.29
0.20
±0.13
0.13
±0.02 SE
0.21
±0.19
-
                                     157

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al.





     The animal model used for studies of the lipid marker was the tracheal




implant (Marchok, Chapter 13 of this volume) after treatment with the potent





carcinogen 7,12-dimethylbenz(a)anthracene (DMBA).  To review the sequence of




morphological changes obtained with the tracheal implant model, after a 4-week




exposure to the carcinogen we observe a squamous metaplastic response mixed




with atrophic epithelium.  By 2 mo after the exposure is terminated, we begin




to  see focal areas of squamous metaplasia with atypia.  Lesions with marked




atypia can be found by 4 mo after termination of the exposure.









     A recent paper dealing with the morphological changes induced by various




polyaromatic hydrocarbons (Lumb and Snyder 1971) demonstrated that the effects




of  the carcinogenic hydrocarbons DMBA and BaP are slow to be reversed in




comparison to those of the noncarcinogenic or weakly carcinogenic




hydrocarbons.  Although the lesions obtained with the two carcinogens differ




in  morphology, they are similar in that they persist in the epithelium.  Thus,




the concept of reversibility appears to be important in this model system.









     Table 15-1 shows the levels of tritium incorporated from 1-3H-hexadecanol




into the alkyldiacylglycerol and triglycerol fractions at various times after




delivery of the precursor to tracheal implants.  The amount of label




incorporated into the normal implants was minimal and declined relatively




rapidly.  The effect of a 4-week exposure to DMBA was striking.  The average




level of labeled alkyldiacylglycerols was 6 times higher than in the normal




epithelium, and the turnover of the label appeared to be slower.  Precursor




incorporation was also studied at 4 mo after termination of carcinogen







                                     158

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER  	Heckman et al.






treatment (i.e., at the time when  atypical  lesions  can  first be detected).




The level of label incorporated at this  time was  2  to 3 times higher than in




the normal epithelium, and  the  turnover  again appeared  slower.









     With the ether lipid marker,  then,  we  succeeded in identifying a




biochemical change that resembles  previously studied morphological changes in




that it is slow to be reversed. The accumulation of ether lipids,




particularly the alkyldiacylglycerols, is also somewhat remarkable in that it




occurs very early after carcinogen treatment.









     We also wanted to understand  the biochemical basis of ether lipid




accumulation.   To study this problem, we cultured normal rat tracheal




epithelium in parallel with cells  from one  of our transplantable tumor lines.




In particular,  we wanted to know whether the turnover of the ether lipids was




impaired in carcinogen-treated  epithelium.   If this were the case, then




absence of the  cleavage enzyme  might provide a more direct marker for




preneoplastic changes.  However, when grown in vitro, the normal epithelium




synthesized levels of alkyl lipids were  equivalent  to those of the tumor




cells.  Incorporation in alkyldiacylglycerols of  labeled palmitate (the




precursor) was  similar  for  the  two cell  types,, as was the rate of turnover




(Scott et al. 1979a).








     The unexpected similarity  of  lipids from normal and neoplastic  cells




grown in vitro  suggested that alkyl lipid accumulation  might be related to




glucose levels, which are high  in  the culture medium and which are known to be






                                      159

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al.






elevated in many neoplastic cells.  In a second series of experiments, we




showed this to be the case.  By growing neoplastic cells from two of our




transplantable cell lines in media supplemented with varied levels of glucose,




we could induce the accumulation of correspondingly high or low levels of the




triglycerides and their ether analogues (Table 15-2).  The effect of glucose




is related to levels of the precursors needed for biosynthesis of the ether




lipids.  The precursor for the glycerol backbone is dihydroxyacetone




phosphate, which is formed as a direct glycolytic intermediate, according to




the Embden-Meyerhof-Parnas scheme.  It is not clear how synthesis of the other




necessary precursors, the long-chain fatty alcohols, might be related to




glucose levels.  Our evidence indicates that their biosynthesis is regulated




by an as yet unidentified product of glycolysis (Scott et al. 1979b).  In any




case, these results suggest the feasibility of a more direct assay relating to




glycolytic metabolism for application to the problem of detecting




preneoplastic changes in the respiratory tract.









THE CELL SHAPE MARKER









     We undertook experiments similar to previously described studies using




liver cells (Montesano et al. 1977) to examine the possibility that




morphological features could serve as markers for early stages of cancer in




the respiratory tract.  The model system was an ^.n vitro system comparable to




the DMBA tracheal implant model.  Two cell lines derived from the tracheal




epithelium after 2 weeks of carcinogen exposure (Marchok et al. 1978) were




used.  Figure 15-2 diagrams the development of oncogenicity in these lines.







                                     160

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER
                                                                 Heckman et al
    TABLE 15-2.  VARIATION  OF  TRIGLYCEROL AND ALKYLDIACYLGLYCEROL CONTENT

                   WITH LEVELS OF GLUCOSE IN CULTURE MEDIAa
Cell
Line

B2-1



BP3-0



Glucose
Concentration
(mM)
0
0.2
1.0
5.5
0
0.2
1.0
5.5
Triglycerol
Content
(yg/mg protein)
4.8
12.3
61.6
109.6
21.1
22.2
22.8
35.6
Alkyldiacylglycerol
Content
(yg/mg protein)
5.7
14.6
41.8
75.5
6.7
9.5
13.0
34.6
aData  from Scott et al. (1979b).
INITIATION TO TIME OF IN VITRO
H VITRO CULTURE CULTURE
y looow
* •
p— T
tOOO WT
f
I                                                          TIME OF
                                                          IN VIVO
                                                          GROWTH
                  ess
                  BP 3-0
                                     —T
 Figure 15-2.
Diagrammatic representation of in vitro derivations and
experimental sampling points  (*) for three rat tracheal

epithelial lines.
                                       161

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al.






The  1000 W line, upon becoming oncogenic, formed tumors in all




immune-suppressed hosts tested.  As positive and negative controls, a cell




line that formed tumors in relatively few sites (165 S) and a highly oncogenic




line from a transplantable tumor (BP 3-0) were used.









     To ensure that we identified changes linked to the oncogenic status of




the  line rather than with the amount of time it was carried in in vitro




culture, a tumor-derived subline of 1000 W was also studied.  In this way, we




could  examine an additional oncogenic 1000 W population that had been




maintained in culture for only a brief time.









     There were no significant differences in kinetic characteristics of cells




in colonies from the early and the tumor-derived 1000 W populations.  For




example, Figure 15-3 plots labeling index versus the number of cells in the




colonies.  In colonies from the early passage of the line, the cells tended to




be flat and regular.  However, in colonies from late passages and from the




tumor-derived subline, the cells tended to be more elongated (although many




were still regular in shape).  A quantitative analysis of the shape criterion




confirmed this visual impression (Heckman and Olson 1979).









     It is important that we identify the cellular mechanism underlying the




changes in cell shape.  Particularly if these changes involve an extracellular




constituent, they could provide a second marker for in vivo changes related to




carcinogenesis.  One way to approach the problem, in terms of the cellular




biology of the model, would be to perturb the cells from nononcogenic







                                     162

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER
                                                               Heckman et  al
                                      LABELING INDEX
           100
           80-
         LJJ
           60
         X
         CO

         CD
         Z


         <
         oc
         O
         a_
         cc
         O
         O
40
         3
         _i
         LU
         O
           20
                          234567

                           POPULATION DOUBLINGS (number)
                                                 8
 Figure  15-3.
   Percentage of cells incorporating 3H-thyniidine in colonies from

   the 1000 W (E) and tumor-derived 1000 W (T) lines.
                                     163

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al^






populations and see if they could be induced to mimic the oncogenic




populations.  To accomplish this obviously requires a more quantitative assay




for cell shape, which we developed using a computerized image analysis method.




Initial tests demonstrated the method's success in discriminating among cell




lines (Olson et al. 1980) (Figure 15-4).  In preliminary studies of the 1000 W




cell line, we learned that differences in cell shape can be detected in single




cells as opposed to colonies.  It is also possible to calculate the mean for




individual cells in a population and to use that value to estimate the length




of time the population has been carried in vitro.  While at present it is not




clear how this particular end point might be applied to problems of in vivo




detection of preneoplastic changes, it remains a promising area for future




development.









ANOMALIES IN RESPIRATORY TRACT CARCINOGENESIS









     The final section of this review discusses some findings that we consider




anomalous in terms of the usual conceptualization of lung cancer etiology.




Hopefully, an awareness of these findings will at least prevent us from




becoming too complacent about our current understanding of carcinogenesis in




the respiratory tract.









     Several years ago, we had the good fortune to collaborate with Dr. Walden




Dalbey in a chronic inhalation study of the effects of tobacco smoke on the




lung.  The studies were performed with Fisher 344 rats; thus, the findings may




only be indicative of the processes which take place in the lungs of larger







                                     164

-------
   1000W



     165S



    2C5



Figure 15-4.
                                                                Jk_
•if
0
0
26
T
0
0
17
™
1
28
4
T
0
7
3
™
49
0
0
T"
0
6
0
                                                                                                         K


                                                                                                         CO







                                                                                                         CO
                                                                                                         H



                                                                                                         1
                                                                                                         o

                                                                                                         o

                                                                                                         I
Hierarchical cluster analysis of cell shape data input for 50-cell samples  from  three

rat tracheal epithelial cell lines.
                                                                                          a
                                                                                          n>
                                                                                          o
                                                                                          m
                                                                                          rt

-------
|5.  EARLY JTAGES OF RESPIRATORY TRACT CANCER	Heckman et al.






animals.  However, the results were useful in determining the fate of




particulates in the lung in a situation of particulate overloading.









     Particulates in the tobacco smoke we studied ranged from 0.1 to 0.6 ym in




actual diameter but were accumulated into larger masses within the




macrophages.  By 3 d after cessation of smoking, nearly all of the




particulates were found within macrophages.  With chronic exposure, the




macrophages were present not only in the alveolar space but also in the




interstitial connective tissue.  They were particularly common in the




connective tissue at the terminal bronchiole and at the pleura.  These sites




suggested a relationship with the patterns of lymphatic drainage in the rat




lung.  However, there was nothing more than this circumstantial evidence to




implicate the lymphatics.









     The major type of lesion found in the lung was typically near the




alveolar duct.  Most frequently, these lesions developed near sites where the




ducts terminated on the pleura.  In serial sections, the alveolar duct lesions




were often adjacent to a branch of the pulmonary vein which had an




accumulation of macrophages in the perivascular adventitia.  Because of this




juxtaposition, it seems likely that perivascular lesions may contribute to the




development of lesions in the parenchyma surrounding the adjacent alveolar




duct.









     At the cellular level, it seems anomalous that the vast majority of deep




lung particulates occur in macrophages while the macrophage does not itself







                                     166

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER     	Heckman et al


seem to be a target cell  for carcinogens in this, organ.  We know that the

macrophage can metabolize BaP,  and that it should be  able to form reactive

metabolites.  At one time, cell biologists also thought  that the alveolar

macrophage did not divide and could not, therefore, be a target for

transformation.  We now know that these cells are capable of division.



     A second apparent anomaly involves the sites in  which macrophages

accumulate in the lung parenchyma.  The most common sites are in the

perivascular adventitia,  yet the cells of the vasculature do not seem to be

target cells, either.



     These questions may be resolved when there is better insight into the

biological and metabolic  characteristics of different cell types in the deep

lung.  Meanwhile, they should serve to remind us that the field of pulmonary

carcinogenesis  still poses a number of unresolved questions.



REFERENCES
 Aaronson,  S.  A., and G. J. Todaro.  1968.  Basis for the  acquisition of
      malignant potential by mouse cells cultivated in vitro.   Science,
      162:1024-1026.

 Borek,  C., and C. M. Fenoglio.  1976.  Scanning electron  microscopy of surface
      features of hamster embryo cells transformed in vitro by X-irradiation.
      Cancer Res., 36:1325-1334.

 Butel,  J.  S., J. P. Dudley, and D. Medina.  1977.  Comparison of the growth
      properties in vitro and transplantability of continuous  mouse mammary
      tumor cell lines and clonal derivatives.  Cancer Res., 37:1892-1900.

 Franks, L. M.  1979.  Personal communication to C. M. Heckman.


                                      167

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al.


Heckman, C. A., and A. C. Olson.  1979.  Morphological markers of oncogenic
     transformation in respiratory tract epithelial cells.  Cancer Res.,
     39:2390-2399.

Holley, R. W., and J. A. Kiernan.  1968.  "Contact inhibition" of cell
     division in 3T3 cells.  Proc. Natl. Acad. Sci. U.S.A., 60:300-304.

Hynes, R. O.  1974.  Role of surface alterations in cell transformation:  The
     importance of proteases and surface proteins.  Cell, 1:147-156.

Jones, P. A., W. E. Lang, A. Gardner, C. A. Nye, L. M. Fink, and W. F.
     Benedict.  1976.  In vitro correlates of transformation in C3H/10T 1/2
     clone 8 mouse cells.  Cancer Res., 36:2863-2867.

Lumb, R. H., and F. Snyder.  1971.  A rapid isotopic method for assessing the
     biosynthesis of ether linkages in glycerolipids of complex systems.
     Biochim. Biophys. Acta, 244:217-221-

Marchok, A. C., J. C. Rhoton, and P. Nettesheim.  1978.  In vitro development
     of oncogenicity in cell lines established from tracheal epithelium
     preexposed in vivo to 7,12-dimethylbenz(a)anthracene.  Cancer Res.,
     38:2030-2037.

Montesano, R., C. Drevon, T. Kuroki, L. Saint Vincent, S. Handleman, K. K.
     Sanford, D. De Feo, and I. B. Weinstein.  1977.  Test for malignant
     transformation of rat liver cells in culture:  Cytology, growth in soft
     agar, and production of plasminogen activator.  J. Natl. Cancer Inst.,
     59:1651-1656.

Olson, A. C., N. M. Larson, and C. A. Heckman.  1980.  Classification of
     cultured mammalian cells by shape analysis and pattern recognition.
     Proc. Natl. Acad. Sci. U.S.A., 77:1516-1520.

Ossowski, L., J. C. Unkeless, A. Tobia, J. P. Quigley, D. B. Rifkin, and E.
     Reich.  1973.  An enzymatic function associated with transformation of
     fibroblasts by oncogenic viruses.  II.  Mammalian fibroblast cultures
     transformed by DNA and RNA tumor viruses.  J. Exp. Med., 137:112-126.

Porter, K. R., G. T. Todaro, and V. A. Fonte.  1973.  A scanning electron
     microscope study of surface features of viral and spontaneous
     transformants of mouse BALB/3T3 cells.  J. Cell. Biol., 59:633-642.

Rapin, A. M. C., and M. M. Burger.  1974.  Tumor cell surfaces.  General
     alterations detected by agglutinins.  Adv. Cancer Res. (G. Klein, S.
     Weinhouse, and A. Haddow, eds.),  20:1-91.

Scott, C. C., C. A. Heckman, P. Nettesheim, and F. Snyder.  1979a.  Metabolism
     of ether-linked glycerolipids in cultures of normal and neoplastic rat
     respiratory tract epithelium.  Cancer Res., 39:207-214.
                                     168

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER     	Heckman et  al


Scott, C. C., C. A. Heckman,  and F.  Snyder.   1979b.   Regulation of  ether
     lipids  and their  precursors in  relation to glycolysis in cultured
     neoplastic cells.  Biochim. Biophys.  Acta.,  575-215-224.

Shira, S., V. H. Freedman,  R. Risser,  and R. Pollack.   1975.   Tumorigenicity
     of virus-transformed cells in nude mice is correlated specifically with
     anchorage independent growth in vitro.   Proc. Natl.  Acad.  Sci.  U.S.A.,
     72:4435-4439.

Snyder, F.,  and R.  Wood.  1968.  The occurrence and  metabolism of alkyl and
     alk-1-enyl ethers of glycerol in transplantable rat  and  mouse  tumors.
     Cancer  Res.,  28:972-978.

Topping,  D.  C., B.  C.  Pal,  D. H. Martin, F.  R.  Nelson,  and P.  Nettesheim.
      1978.   Pathologic changes induced in respiratory tract mucosa  by
     polycyclic hydrocarbons of differing carcinogenic  activity.  Am. J.
     Pathol.,  93:311-322.

Unkeless, J. C.,  A. Tobia, L. Ossowski, J. P- Quigley,  D.  B.  Rifkin, and E.
      Reich.   1973.   An enzymatic function associated with transformation of
      fibroblasts  by oncogenic viruses.  I.  Chick embryo  fibroblast cultures
      transformed  by ovian RNA tumor viruses.  J.  Exp. Med., 137:85-111.

Voyles,  B. A.,  and C.  M. McGrath.  1976.  Markers to distinguish normal and
      neoplastic mammary epithelial cells _in vitro:   Comparison of saturation
      density, morphology and concanavalin A reactivity.  Int.  J. Cancer,
      18:498-509.

Wigley,  C. B.,  and I.  C. Summerhayes.  1979.  Loss of LETS protein  is not a
      marker for salivary gland or bladder epithelial cell transformation.
      Exp. Cell,  Res.,  118:394-398.

Yamada,  K. M.,  and J.  A. Weston.  1975.  The synthesis, turnover, and
      artificial restoration of a major cell surface glycoprotein.   Cell,
      5:75-81.

 Yang, N.-S., W. Kirkland, T. Jorgensen, and P.  Furmanski.  1980.  Absence of
      fibronectin and presence of plasminogen activator  in both normal and
      malignant human mammary epithelial cells in culture.  J. Cell. Biol.,
      84:120-130.
 WORKSHOP COMMENTARY
 E. Hu:  [Inaudible]
                                      169

-------
15.  EARLY STAGES OF RESPIRATORY TRACT CANCER	Heckman et al.


C. A. Heckman:  The study I discussed involved measurements only of the
alkyldiacylglycerols and ether-linked phospholipids.  The gangliosides have
not yet been studied in these model systems.
                                     170

-------
    16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM INHALATION IN THE RAT

                              Nathanial W. Revis
                                   T. Major
                               Walden E. Dalbey

                               Biology Division
                        Oak Ridge National Laboratory
                              Post Office Box Y
                             Oak Ridge, TN  37830
INTRODUCTION



     At present there is widespread interest and concern about the effects of

environmental pollutants upon human health.  Much of this concern stems from

recent experimental studies showing a variety of environmental pollutants to

be carcinogenic.  Information on the relationship of pollutants and

cardiovascular diseases, however, remains to be determined.  Diseases of the

cardiovascular system may be grouped into three major categories:

hypertension, atherosclerosis (including heart attack and stroke), and chronic

heart failure.  Together, these three categories represent the major cause of

death in North America.
                                     171

-------
16.  CARDIOVASCULAR EFFECTS OF OZONK AND CADMIUM  	Revis et al.







BACKGROUND









     The research interests of our group relate to the effects of




environmental pollutants (in particular, energy-related pollutants) in




cardiovascular disease.  In recent studies, we showed that lead and cadmium




(energy-related pollutants) induce hypertension and atherosclerosis in the




White Carneau pigeon.  Similar results were observed in the rat.  These




observations support epidemiologic studies which show that these elements




correlate with hypertension and coronary artery disease (i.e., coronary




atherosclerosis).  We also obtained preliminary results showing that the




chemical carcinogen benzo(a)pyrene induces an increase in the number and size




of aortic atherosclerotic plaques in the pigeon.  These results show quite




clearly that environmental pollutants are involved in cardiovascular disease.









     The major routes of intake of environmental pollutants in humans are the




gastrointestinal tract and respiratory tract.  To obtain the results just




described, pigeons and rats were exposed to the pollutants via drinking water




(i.e., uptake occurred via the gastrointestinal tract).  However, information




on the effects of these pollutants in cardiovascular disease following




inhalation exposures (i.e., uptake via the respiratory tract) remains to be




determined.  With the exception of carbon monoxide (an air pollutant




associated experimentally with atherosclerosis) we know little about the




effects of air pollutants in cardiovascular disease.  Nevertheless, oxidants




such as ozone (03) and the sulfur oxides have been shown to induce reversible




electrical changes in the heart as measured by electrocardiographic methods.







                                     172

-------
16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM      	Revis et al.






In addition, Trams et al.  (1972)  demonstrated that the activity of monoamine




oxidase (MAO) and catechol O methyltransferase  (COMT) in the dog brain is




significantly reduced after exposure to 1 ppm O3 for 18 mo.  These two enzymes




affect blood pressure through their catabolic effect on the neurotransmitter




norepinephrine.  Thus, a decrease in the activity of these enzymes may




increase the half-life of  norepinephrine, ultimately provoking an increase in




blood pressure.









     In previous studies,  we exposed rats to 5  ppm Cd via drinking water for 6




mo.  After  exposure, the animals  were  anesthetized, and the femoral artery




pressure and blood norepinephrine measured.  Prior to killing the animals, %




norepinephrine was injected  into  the femoral vein, and blood pressure and 3fl




norepinephrine levels were measured over a period of 1 h.  The results of this




study showed both blood pressure  and norepinephrine to be significantly




increased in the Cd-treated  rats. Furthermore, the half-life of 3n




norepinephrine was significantly  increased in the Cd-treated animals.  As




previously  noted, the half-life of norepinephrine is in part regulated by MAO




and COMT.   The half-life  is  also  regulated by cellular uptake (both neuronal




and extraneuronal).  We previously showed the activity of these two catabolic




enzymes to  be decreased  in the aorta of rats exposed to 5 ppm for 6 mo.  Thus




the increase in blood pressure observed in the  Cd-treated rats may be the




result of an increase in  the half-life of norepinephrine due to Cd effects on




these two catabolic enzymes.  However, Cd may also affect the uptake of




norepinephrine.  In any  event, a  decrease in the activity of these enzymes can





effect a change in blood  pressure.






                                      173

-------
16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM	Revis et al.


METHODS AND RESULTS



     To expand the observations of Trains et al. (1972), we exposed Fischer

rats as follows:
      (1)  O3, 0.6 ppm, 5 h/d, for 3 d
          (blood pressure recorded on day 4)

      (2)  Cd, 3 mg/m3, for 1 h
          (blood pressure recorded on day 8)

      (3)  Cd, 3 mg/m3, for 1 h
          4 d later:  03, 0.6 ppm, 5 ii/d, for 3 d
          (blood pressure recorded on day 8)
After these exposures, femoral artery pressure, electrocardiographic data, and

tissue concentrations of Cd (i.e, aorta, heart, and lungs) were obtained.



     Table 16-1 shows the effects of these treatments on the cardiovascular

system.  Systolic pressure and heart rate were significantly increased in the

treated animals.  However, diastolic pressure and mean pressure were not

significantly changed.  The relation of maximal velocity of pressure increase

in the femoral artery in the isometric phase of contraction (dpp/dt) was

significantly increased by Cd but not by 03.  Note, also, that the effects of

O3 and Cd were not additive with respect to the changes in systolic pressure,

heart rate, and dpp/dt.  In fact, the changes observed in the Cd plus 03

treated rats more closely resembled those observed in the Cd-treated animals.
                                     174

-------
    TABLE 16-1.   EFFECTS OF CADMIUM AND OZONE INHALATION  ON THE CARDIOVASCULAR SYSTEM IN THE FISCHER RAT
    Group3
                   Systolic Pressure
                       (mm Hg)b
                   Diastolic
                   Pressure
                   (mm Hg)b
                Mean Arterial
                 Pressure
                 (mm Hg)b
                Heart Rate
                (beats/min)b
                               H
                               3
                                                                                                                 M
                                                                                                                 01
                                                                                                                 O
Cd (5)

Cd (5)

03 (5)

Cd +03 (5)
116 ± 22

139 ± 10C

143 ±  9C

139 ±
95 ± 25

81 ±  3

86 ±  4

82 ±  2
102 ± 20

100 ±  7

105 ±  3

100 ±  5
370 ± 80

456 ± 3QC

414 ± 55d

456 ± 40°
2500 ± 400

3300 ± 50QC

2700 ± 200

3050 ± 700°
    aNumber of animals per group  is given in parentheses.

    bMean ± S.E.M.
^h
01   cp < 0.01.

    ap < 0.05.

                                                                                                                 a

-------
16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM	Revis et al.






     The significant increase in the QRS interval and in the amplitude of the




R wave (Table 16-2) suggests that these treatments may have affected the




permeability of the myocardial membranes or the metabolism of the




neurotransmitter norepinephrine.  The increase in heart rate would support the




latter suggestion.  However, an effect on the permeability cannot be ruled




out.









     Table 16-3 shows the tissue distribution of Cd.  As might be expected,




lung Cd was significantly increased (by a factor of 5) following Cd exposure.




Aortic Cd increased significantly, while the concentration of Cd in the heart




did not significantly change following Cd treatment.  In the Cd plus 03




treated rats, Cd was decreased in the lung and increased in the aorta (as




compared to rats treated only with Cd, suggesting that 03 may increase Cd




clearance from the lung.









     These results demonstrate that Cd and 03 have similar effects on the




cardiovascular system.  The results also suggest that the cardiovascular




effects may be ^exerted~~thf ough an effect on norepinephrine metabolism.  With




respect to effects on the cardiovascular system, no interactions between 03




and Cd were evident.
                                     176

-------
                          TABLE 16-2.   EFFECTS OF CADMIUM AND OZONE INHALATION
                      ON THE ELECTRICAL PROPERTIES OF THE HEART IN THE FISCHER RAT
Groupa
                Heart Rate
                (beats/min)b
 PR Interval
    (s)b
 QRS Interval
     (s)b
 Amplitude of
    P Wave
    (mV)b
Amplitude of
   R Wave
   (mV)b
                                                                                                               H
                                                                                                               3
Control (5)

ca (5)

03 (5)

Cd +03 (5)
                  370  ±  80

                  456  ±  30°

                  414  ±  85d

                  456  ±  40C
0.047 ± 0.003

0.047 ± 0.007

0.047 ± 0.001

0.044 ± 0.001
0.021 ± 0.009

0.034 ± 0.003d

0.038 ± 0.0009°

0.039 ± 0.005C
0.0060 ± 0.0004

0.0034 ± 0.0006C

0.0076 ± 0.001C

0.00ft4 ± 0.008°
0.054 ± 0.009

0.070 ± 0.006°

0.067 ± 0.004d

0.069 ± 0.006°
M
W
                                                                                                              O
                                                                                                              N
                                                                                                              O
                                                                                                              p
aNumt>er of animals per group is  given in parentheses.

bMean ± S.E.M.

°p < 0.01.
 d,
 lp <  0.05.
                                                                                                              CO

                                                                                                              (D

-------
16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM	Revlsetal.
   TABLE 16-3.  TISSUE DISTRIBUTION OF CADMIUM IN THE FISCHER RAT FOLLOWING
                         CADMIUM AND OZONE INHALATION
                                         Level of Cd in Tissue
                                           (yig/g dry weight)^
      Groupa                  Aorta              Heart             Lung
Control (5)
Cd (5)
03 (5)
Cd + 03 (5)
0.44 ± 0.09
0.82 ± 0.07°
0.50 ± 0.008
1.29 ± 0.13d
0.24 ± 0.04 1.9 ± 0.7
0.31 ± 0.03 11.3 ± 1.4d
0.29 ± 0.07 2.4 ± 0.8
0.27 ± 0.05 10.3 ± 2.3d
aNumber of animals per group is given in parentheses.

bMean ± S.E.M.

cp < 0.01.

dp < 0.001.



RECOMMENDATIONS FOR FURTHER RESEARCH



     Based on these preliminary studies, our group offers the following

recommendations for future research:
     (1)  Determine if O3 has a dose-dependent effect on the
          cardiovascular system

     (2)  Determine the effects of O3 on norepinephrine metabolism

     (3)  Determine the effects of other oxidants on the
          cardiovascular system

     (4)  Determine if oxidants interact with other air pollutants
          in effecting a change in the cardiovascular system
                                     178

-------
16.  CARDIOVASCULAR EFFECTS OF OZONE AND CADMIUM                  Revis et al,
     (5) Determine the effects of various air pollutants in
         animals with cardiovascular disease
WORKSHOP COMMENTARY
J. L. Whittenberger;   In measuring the effects of 03 exposure on heart rate
and blood pressure, did you measure pulmonary ventilation?  I wonder whether
hypoventilation might  have played a role  in producing the cardiovascular
effects.

N. W. Revis;   The  effects may very well be due to hypoventilation.  They may
as well be attributable to the change in  the pulmonary system rather than the
secondary effect;  I'm  not sure and cannot give a specific answer to your
question.  The results suggest to me  that the effect is not simply an effect
on the sympathetic nervous system required to stimulate the heart rate, thus
affecting systolic pressure.

J. L. Whittenberger;   I was talking only  about the  level of ventilation, not
about the damage  to the lungs.  What  dose of 03 was administered?

W. S. Dalbey;  The dose was 0.6 ppm for 5 h for 3 d in a row.  We measured
pulmonary resistance for  totally other purposes during the O3 exposure, but
the  present animals were  not  examined until the following day.  We didn't do
the  measurements  until ~24 h  after the last day of  exposure.
                                      179

-------
        17.  EFFECTS OF NITROGEN DIOXIDE AND 3-METHYLFURAN INHALATION
                      ON THE SMALL AIRWAYS IN THE MOUSE

                               Wanda M. Haschek

                               Biology Division
                        Oak Ridge National Laboratory
                              Post Office Box Y
                             Oak Ridge, TN  37830
INTRODUCTION



     Small airways, i.e., the smallest bronchi and bronchioles,  are one of the

most vulnerable regions in the lung for damage by many inhaled irritants.

Small airways are a major site of airflow obstruction in chronic bronchitis,

bronchiectasis, and, probably to a large extent, emphysema (Macklem et al.

1971; Hogg et al. 1968).  Despite this fact, few details are available

concerning the response of the epithelial components to injury in this region;

most attention has centered on the alveolar zone of the lung.



     The cell population of the small airways consists of the ciliated cells

(responsible for the movement of small particles up the mucociliary escalator)

and the nonciliated or Clara cells (the progenitors of the ciliated cells).

Although the exact function of the Clara cells is still under investigation,

they are secretory cells and are rich in mixed function oxidases.  The Clara

cells are thus capable of activating drugs and other compounds to highly


                                     180

-------
17.  SMALL AIRWAY DISEASE	Haschek







reactive and toxic metabolites, and  of  inactivating  others.   Small numbers of




goblet cells, responsible  for mucous secretion, are  also present  in the  small




bronchi.









     The ciliated cells  of the  terminal airways are  readily damaged by the




oxidants ozone  (O3) and  nitrogen  dioxide (NO2).   Following a  single exposure




to either 03 or NO2,  the airway epithelium is  rapidly repaired due to




proliferation of nonciliated bronchiolar epithelial  cells which subsequently




differentiate into ciliated cells.   Since 03 and  NC>2 are the  major oxidants




present during  peak traffic hours, their potential contribution to small




airway disease  in man must be considered.









     Exposure of man  to  very high NC>2 concentrations has been reported to




result in bronchiolitis  obliterans.   Bronchiolitis obliterans is  a




particularly severe  form of small airway disease  in  response  to local injury




of the wall.  Initially, cellular granulation  tissue more or  less fills the




bronchiolar lumen.   This tissue then undergoes organization and takes a




polypoid form,  the final fibrosis extending to the musculoelastic and




peribronchial regions.  Partial to complete obstruction of the bronchiolar




lumen occurs.   Bronchiolitis obliterans has also  been reported to follow




exposure to sulfuric  acid, ammonia,  and war gases.   In  a  study of O3 exposure




in rats  (Last et al.  1979), a focal  polypoid thickening of alveolar duct walls




was  observed.   This  lesion was  partially reversible  on  removal from O3.
                                      181

-------
17.  SMALL AIRWAY DISEASE       	Haschek






     Nonciliated bronchiolar epithelial or Clara cells were recently shown to




be the target cells for several toxic furans, including 4-ipomeanol and




3-methylfuran (3MF) (Boyd 1980).  3MF has been identified in city smog and is




believed to be formed by photoxidation from naturally occurring terpenes which




originate from deciduous trees.  The Clara cell activates these compounds to




their toxic metabolites, causing cell death.  With a single exposure to a low




or moderate dose of 3MF, complete regeneration of the bronchiolar epithelium




takes place within a few days; following a high dose, regeneration is not




fully complete even after 3 weeks (Haschek, unpublished observations).









     The study described below was designed to examine whether damage to the




progenitor cell—the nonciliated cell—would interfere with successful




recovery of the injured bronchiolar epithelium.









METHODS









     NO2 was chosen to produce necrosis of the ciliated bronchiolar cells; a




single exposure to 20 ppm for 24 h was used.  Inhalation of 3MF at a dose of




2.5 pi/liter for 1 h was employed to damage the nonciliated bronchiolar cells.




Young adult male Balb/C mice were divided into five groups and treated as




follows:
                                     182

-------
17.  SMALL AIRWAY DISEASE	Haschek












     Group                   	Treatment	




                             Day 1              Day 2              Day 3
A
B
C
D
E
NO2
NO2
N02
-
-
-
3MF
3MF
3MF
3MF
-
-
3MF
-
3MF
      Half the animals in each group were killed on the 6th day; the other half




 were killed on the 10th day.  Animals were killed by cervical dislocation, and




 lungs were fixed in situ by intratracheal instillation of 10% buffered




 formalin.  Lung tissue was processed, embedded in paraffin,  sectioned at 3-4




 ym, and stained with hematoxylin and eosin.
 RESULTS
      A large proportion of the mice exposed consecutively to 3MF died?




 therefore, groups C and E will not be discussed further.   Lungs from animals




 exposed to NO2 alone (group A) showed minimal changes consisting of mild




 hypercellularity around bronchioles and alveolar ducts at 5 d.   By 10 d, the




 lungs appeared virtually normal.
                                      183

-------
17.  SMALL AIRWAY DISEASE	Haschek






     Animals exposed to 3MF only (group D) had marked loss of nonciliated




bronchiolar cells.  In small bronchioles, segmental areas of denudation were




interspersed with squamous and nonciliated cuboidal cells.  In larger




bronchioles, normal ciliated cuboidal and small groups of columnar nonciliated




cells were present.  After 10 d, regeneration was evident:  bronchioles were




lined by varying numbers of undifferentiated cuboidal-to-columnar cells, as




well as some normal ciliated and nonciliated cells.









     Lung changes in animals exposed to W>2 and subsequently to 3MF (group B)




were similar to those seen following exposure to 3MF alone, with the exception




that only small numbers of normal ciliated cells remained in the large




bronchioles.  Scattered small fibrocellular polypoid masses, covered by




squamous to cuboidal epithelium, as well as sessile fibrous thickenings of the




bronchiolar wall, protruded into the lumen.  After 10 d, regeneration was




evident, with many undifferentiated nonciliated cuboidal cells, some




dome-shaped Clara cells, and small numbers of ciliated cells.  Only a few




polypoid lesions remained.









DISCUSSION









     This study demonstrated an interaction between NO2 and 3MF in the small




airways.  While the ciliated cell injury produced by NO2 was virtually




repaired by 5 d, and the effect of 3MF was primarily on the nonciliated




bronchiolar cells, the combined effect of these two agents greatly retarded




regeneration of ciliated cells and produced focal thickening and polypoid






                                     184

-------
17.  SMALL AIRWAY DISEASE	Haschek






lesions in the bronchiolar wall.  The retardation of ciliated cell




regeneration is attributed to damage of the nonciliated progenitor cells by




3MF.  Sustained loss of ciliated cells might severely impair the defense




mechanisms of the ciliary escalator.









     The pathogenesis of the bronchiolar polyps can only be conjectured.  The




polyps seen in this study were similar in both appearance and reversibility to




those produced in the rat by exposure to 03 (Last et al. 1979).  The polyps




produced by 03 were located in the alveolar ducts, whereas in the present




study they were located in the bronchioles.  The difference in location can be




explained by the difference in principal site of injury between these agents.




Since simple epithelial damage is usually repaired without the formation of




such polyps, one may ask whether damage to underlying elements (such as




basement membrane or subepithelial tissue) may play a role in this phenomenon.




In bronchiolitis obliterans, where granulation tissue polyps partially or




completely occlude the bronchiolar lumen, damage to the bronchiolar wall is




implicated.  We produced a lesion similar to that of bronchiolitis obliterans




by intratracheal instillation of a coal liquefaction distillate in the rat




(Haschek et al. 1981).  Extensive necrosis of the airways epithelium allowed




leakage of the distillate (visualized by fluorescence microscopy) through the




mucosa into the underlying tissue and even, in areas, into the surrounding




parenchyma.  Proliferation of the subepithelial connective tissue resulted in




polypoid lesions in the airways and alveolar ducts.  These lesions became




somewhat condensed with time but were still present at 60 d.  Whether such




lesions persist or not may depend on the severity of initial damage, the






                                     185

-------
17.  SMALL AIRWAY DISEASE	Haschek


degree of the proliferative response, or, as in the case of the distillate, on

the persistance of the agent within the lesion.



     In conclusion, despite our recognition of small airway disease as the

underlying cause of airflow obstruction in chronic obstructive lung disease

and our knowledge that small airways may be severely affected by inhaled

oxidants (e.g., N(>2 and 03) as well as naturally occurring toxins (e.g., 3MF),

little is known about the pathogenesis of small airway disease.  The response

of the small airways to various types and severities of injury, as well as to

various combinations of naturally occurring irritants and toxicants, needs to

be examined in detail before an understanding of small airway disease can be

reached.



REFERENCES
Boyd, M. R.   1980.  Biochemical mechanisms in chemical-induced lung injury:
     Roles of metabolic activation.  CRC Grit. Rev. Toxicol., 7:103-176.

Haschek, W. M., M. E. Boling, M. R. Guerin, and H. P. Witschi.  1981.
     Pulmonary toxicity of a coal liquefaction distillate product.  In:  19th
     Annual Hanford Life Sciences Symposium on Pulmonary Toxicology of
     Respirable Particles, Richland, Washington, Oct. 22-24, 1979 (in press).

Hogg, J. C., P- T. Macklem, and W. M. Thurlbeck.  1968.  Site and nature of
     airway obstruction in chronic obstructive lung disease.  New Engl. J.
     Med., 25:1355-1360.

Last, J. A., D. B. Greenberg, and W. L. Castleman.  1979.  Ozone induced
     alterations in collagen metabolism of rat lungs.  Toxicol. Appl.
     Pharmacol., 51:247-258.

Macklem, P. T., W. M. Thurlbeck, and R. G. Fraser.  1971.  Chronic obstructive
     disease of small airways.  Ann. Int. Med., 74:167-177.
                                     186

-------
17.  SMALL AIRWAY DISEASE	            Haschek


WORKSHOP COMMENTARY
Comment;  Putting  fairly toxic materials  into  the  lung  either by inhalation or
by intratracheal administration will  induce more than a mere destruction of
epithelial cells.  The  subsequent organization of  the inflammatory process
(which can be reversible) might be the  source  of such a fibroblast
proliferation.  This would  agree  with Dr. Witschi's observations on the
fibrosis that follows inhibition  of epithelial cell regeneration.  Hitting
small airways so heavily as to induce such  severe  pathologic changes should
not be interpreted as unique to the agent,  but rather to the degree of damage
that is produced.

W. M. Haschek;  I  agree.  I think the airways  have very limited ways in which
they can respond,  and that  the response is  due to  the degree of damage and not
to the agent.

R. P. Sherwin;  There1s quite a distinction between the polyps of your inhaled
agents and the ones produced by the distillate.  One was edema, raising the
bronchiolar  surface.  This  is why we  should be very careful to distinguish
between cellular damage with edema and  spindly epithelial cells as opposed to
true proliferative fibro-collagenous  depositions.  In humans there are many
counterparts to these so-called "benign fibrous polyps."

W. M. Haschek;  I  agree with most of  your comments.  However, it wasn't simply
edema that raised  the surface of  the  bronchiole.   There was cell
proliferation—proliferation of the underlying fibrous  tissue—within these
polyps.

R. P. Sherwin;  I  didn't mean that it was simply edema, but that any fibrous
proliferation may  be a  small part of  the  lesion.   There is a normal turnover
of interstitial cells,  and  a temporary  increase in turnover following injury
may be occurring.  While you may  be right about the fibrosis, should it
disappear, it would be  contrary to the  usual human fibrosis, where destruction
and fibrous  tissue replacement are characteristically persistent.  The key
question is  what absolute amount  of collagen is found.  We must bear in mind
-that several studies of lung fibrosis have  failed  to show increased
concentrations of  collagen.  There is a controversy about this—

H. P. Witschi;  There is no controversy;  that  has  been  resolved.  If you have
a fibrotic lung and take an aliquot only  for analysis,  not only total collagen
but also lung weight has increased.  If you divide collagen by weight, the
"specific activity of collagen,"  so to  speak,  stays the same.  If you measure
collagen per total lung, the value increases.

R. P. Sherwin;  The last statement pertaining  to total  collagen content of the
lung is the  critical point.  Are  we dealing with an absolute or relative
increase of  collagen?   A very simple  and  real  explanation for some if not most


                                      187

-------
17.  SMALL AIRWAY DISEASE	Haschek


of the collagen increase in human lungs with fibrosis is the phenomenon known
as condensation fibrosis.  When sensitive epithelium is lost, the more
resistant interstitial tissues of the alveolar walls become stacked;  i.e., the
alveolar walls collapse and fuse.  Also, alveoli are completely lost and in
part are replaced by bronchioles with normally thicker walls.  The key
question, again, is how to distinguish between relative and absolute collagen
increases.  Is there more absolute collagen?

H. P. Witschi;  Yes.  If you calculate collagen per total lung, there is more.
However, if you just take a few grams of lung, determine collagen, and
calculate collagen for every gram of lung, you do not find more.

R. P. Sherwin;  Ron Crystal of the National Heart, Lung and Blood Institute
(Pulmonary Branch) remarked on this interesting paradox in his own studies
(i.e., a discrepancy between the histological and biochemical data).   At a
meeting we jointly attended, I proposed to him that condensation fibrosis
would be a plausible explanation.  At any rate, there is a problem that has
not been completely resolved.

Comment;  Ron Crystal's problem with increased lung collagen is supposedly in
idiopathic pulmonary fibrosis.  It may or may not be unique to the patients
he's investigated.  Some investigators do see an increase in whole lung
collagen.  Whole lung collagen, as I understand it, is measured by analysis of
hydroxyproline, directed to the total lung.  I think this is what Dr. Witschi
mentioned.  If there is an increase in whole lung hydroxyproline, there is (in
all probability) some form of fibrosis.  Some forms are visible as scars; I
think some are visible with special collagen stains.  What Dr. Witschi
described is diffuse or interstitial fibrosis.  Some of the collagen is not
visible but still apparent as an increase in whole lung collagen.

Question;  How much human lung disease is idiopathic fibrosis?

Comment;  Quite a lot of it.

Comment;  It is a common problem.  There are other kinds of fibrosis, however,
which constitute a very significant portion of the disease.

W. M. Haschek;  The reversibility of lung fibrosis was mentioned.  This is a
very important question.  Human fibrosis does not appear to be reversible.
However, human cases of fibrosis frequently are not detected until a late,
chronic stage.  With bleomycin there is initially fibrosis, but after a year
there is no detectable increase in lung collagen.  Therefore, there is some
indication that very early so-called "fibrosis" may be reversible.  Certainly
this aspect needs investigation.

[Note:  Pickrell (Chapter 24 of this volume) reports a finding of increased
collagen but no histological evidence of fibrosis, and thinning of the
alveolar walls.]
                                     188

-------
          18.   OVERVIEW OF RESEARCH AT LAWRENCE BERKELEY LABORATORY

                               Edward L. Alpen

                         Lawrence Berkeley Laboratory
                           University of California
                             Berkeley, CA  94720
INTRODUCTION



     This report summarizes two oxidants research projects that were conducted

at Lawrence Berkeley Laboratory under EPA sponsorship.  The first project

examined cocarcinogenic effects of nitrogen dioxide and sulfur dioxide in mice

(Dr. White, principal investigator).  The ^second examined the effects of

low-level ozone exposure on serum lipoproteins in guinea pigs (Drs. Lindgren

and Shu, principal investigators).



COCARCINOGENIC EFFECTS OF NITROGEN DIOXIDE AND SULFUR DIOXIDE IN THE MOUSE



     This project was designed to investigate whether the toxic gases nitrogen

dioxide (NO2) and sulfur dioxide (SO2) are tumorigenic by themselves, whether

they promote tumorigenesis, or whether they are in fact cocarcinogenic.
                                     189

-------
18.  LAWRENCE BERKELEY LABORATORY	Alpen






Selection of Model System and Test Gases









     For our model system we employed the lung adenoma initiated by the drug




urethan.  Urethan is carcinogenic in the mouse lung without promotion but can




also be promoted by a number of suitable agents.  The advantage of the urethan




system is that the incidence of lung tumors is lineally dependent upon dose.




Also, urethan1s action is very quick (95% is excreted in ~8 h), so the time of




exposure is very precise.  Finally, the tumors can very easily be counted by




gross methods quite similar to those employed at Oak Ridge National




Laboratory, where the whole lung is dissected.  The nodules in the lung are




counted using a dissecting microscope.









     We chose to test NO2 because it has been shown to influence a number of




Type II alveolar epithelial cells in the mouse lung.  The fact that the




majority of source cells for mouse lung adenomas appear to be Type II alveolar




epithelial cells suggested the possibility of a preconditioning phenomenon




that would increase the number of lung adenomas from urethan or other




appropriate carcinogens.  We chose to also test 802 because there are at least




two reports in the literature indicating that SO2 is a promoting cocarcinogen




(in other words, that SO2 exposure after an initiating event with an




incomplete carcinogen will increase the yield of tumors).
                                     190

-------
18.  LAWRENCE BERKELEY LABORATORY	Alpen






Experimental Protocol









     Randomized CF1 mice were  treated  with  the  test  gases either before or




after exposure to urethan.   Exposures  to NO2  or SO2  were performed in standard




chambers, 24 h/d, except for a very brief break for  removal of feces and




change of food.  Chamber gas levels were continuously monitored and recorded.









     The NO2 exposures  were 20 ppm for 2 d, 10  ppm for 4 d, and 5 ppm for 8 d.




Air  controls were in  the chamber for 6 d.   The  SO2 exposures were:  40 ppm for




3  d, 20 ppm for 6 d,  and  10 ppm for 12 d.   Controls  again consisted of urethan




alone and gas alone.  All  animals survived  these exposure regimens.








     The mice were  serially sacrificed at predetermined times, and the




necessary fixing  and counting of tumors were completed.  Part of each animal's




lung was histologically prepared for the counting of Type II cells.








     The resulting multivariate data were  analyzed using the Cox Relative Risk




Model  for dose-dependent or time-dependent events that are  quantal in nature.




 It is  extremely important, in multivariate studies,  to have independent




 controls  for  each of the treatment groups,  to avoid a multiple  comparison with




a single  control  group.  As noted, our protocol included such controls.
                                       191

-------
18.  LAWRENCE BERKELEY LABORATORY	Alpen






Results;  Nitrogen Dioxide









     The level of adenoma production by ITC>2 alone (without urethan) was




essentially zero (~0.02 adenoma per mouse over the very large sample of




control animals).  Thus NO2 alone had no carcinogenic activity.









     Analysis of the data also indicated no difference between the effects of




NO2 given before treatment with urethan and given after treatment (same




exposure levels) .  In other words, NO2 before urethan was the same as NC>2




after urethan.  There was no significant difference even when the data were




pooled.  NO2 demonstrated neither promotion nor anticarcinogenic effects.




Variation in control values over replications reinforced this assessment:  the




range of control values in the three replications was just about the range of




the data.









     At present, Type II cell counts are not complete for this data set.








Results;  Sulfur Dioxide









     When given without urethan, SO2 (like NO2) demonstrated no effect on the




production of adenomas.









     In contrast to IK>2, however, there were notable differences in the




results of SO2 exposure before and after administration of urethan.  All




concentrations of SO2 depressed the incidence of tumors when given after






                                     192

-------
18.  LAWRENCE BERKELEY LABORATORY	_^_	Alpen






urethan and increased the incidence of tumors when given before urethan.  The




tumor incidence in animals exposed to 20 ppm and 40 ppm was essentially




identical, while the effects of  10 ppm differed markedly for "before" and




"after."









     When SO2 was given before urethan, significant increases in the incidence




of tumors were seen at 20 ppm and 40 ppm SO2.  Such increases in the number of




tumors produced do not qualify as "promotion," since initiation had not taken




place.  The effect may be due to some change in the cellular population which




is ultimately exposed to urethan and in which the tumors are produced.  At the




other exposure level  (10 ppm), there was an insignificant difference from




control values.








     When SO2 was given after urethan (i.e., the standard initiation and




promotion model), we  found essentially the opposite effect.  For all levels of




SO2, there was a significantly suppressed incidence of urethan-produced




adenomas in the  lung.  Such  an effect is termed "suppression" (or




"anticarcinogenic effect").








     Type II cells counts are complete for this data set.  There was no




significant difference from  control values throughout the exposure regimen.
                                      193

-------
18.  LAWRENCE BERKELEY LABORATORY	 Alpen






Ongoing and Planned Studies









     Due to the apparent lack of effect by NO2, we have discontinued all work




with that gas.  An ongoing study with SO2 ties in to the previous work by




employing one exposure level—20 ppm for 6 d—for which data were collected in




the first experiment.  The next group of animals will be exposed to 6.7 ppm




for 18 d, and we will again examine both pre- and post-exposure effects.









     Depending on the outcome of the 6.7 ppm study, we may proceed to a




carcinogen other than urethan.  (Urethan, a carbamate, is a highly artificial




carcinogen that is not known to be carcinogenic in man.)  Host likely, we will




employ benzo(a)pyrene (or another carcinogen), lower the SO2 exposure level,




and increase the exposure time.  We do believe that we are "in the right




ballpark" in terms of exposure criteria.









EFFECTS OF OZONE ON SERUM LIPOPROTEIN CONCENTRATIONS IN THE GUINEA PIG









     Due to the oxidative nature of ozone (03), many interrelated aspects of




lung metabolism involving lipids might easily be influenced by this very




reactive substance.  For example, the Type II cells that proliferate after 03




exposure use exogenous lipids during surfactant synthesis.  Prostaglandins,




which require essential fatty acids for synthesis, also seem to be involved in




O3 toxicity.  Protection against 03-induced peroxidation of unsaturated fatty




acids has been reported by Donovan and Menzel.  In addition, Brown and others




have shown that human deaths due to cardiovascular disease are decreased






                                     194

-------
18.  LAWRENCE BERKELEY LABORATORY	Alpen






during periods of lower urban pollution.  This accumulation of experimental




and epidemiologic evidence led us to consider whether the lipoproteins (which




have a very significant influence on atherosclerosis and cardiovascular




disease) are in fact changed by  exposure  to 03.









     As mentioned earlier, Drs.  Lindgren  and Shu  served as principal




investigators for this study.  Their complete report, "Serum Lipid and Lipid




Protein Concentrations Following Exposure to Ozone," has been accepted for




publication in the Journal of Environmental Pathology and Toxicology.









Experimental Protocol








     Hartley guinea pigs  were exposed  to  03 for 22 d.  The nominal exposure




level was  1 ppm; in retrospect,  we  believe the actual concentration to have




been closer to 0.8 or 0.85 ppm.  As in the N02/SO2 study, standard exposure




chambers were employed.   We monitored  all of the  usual toxicological




indicators, such as body  weight, growth,  food consumption, and mortality.  In




addition,  serum lipoprotein values  were obtained  at the beginning of the




study,  at  the end of the  study,  and for 30 d after exposure.









Results;   Toxicological Indicators








     As expected, we observed a  very marked sex-dependent effect of O3




exposure.  There were  2 deaths in the  male group  but none in the female  group.
                                      195

-------
 18.  LAWRENCE BERKELEY LABORATORY	Alpen






 Other  investigators  (in our laboratory and elsewhere) have noted that  the




 female is much more  resistant than the male for any end point in 03 exposure.









     As predicted from our rat studies, there was a sharp decrease in  food




 intake.  This decrease in food intake was accompanied, surprisingly, by no




 change in body weight.  Body weight was maintained at the control levels




 despite a 35% drop in food intake.  Retrospectively, in light of other 1 ppm




 exposure studies showing (1) a sharp drop in thyroid hormone levels and




 thyroid function and (2) sharp reductions in growth hormone and thyrotropin




 from the pituitary (Clemens and Garcia), it's not surprising that the




 metabolic rate of the animals was sharply reduced.








 Results;  Lipoprotein Levels









     We examined high-density lipoproteins, low-density lipoproteins, very low




 density lipoproteins, triglycerides, and cholesterol.









     There was a very large increase in the cholesterol levels of the exposed




 animals as compared with normal controls.  In males, this increase was 200%;




 in females, 100%.  Comparisons with pair-fed controls (to remove any dietary




 effect) revealed smaller increases in cholesterol:  50% in males and 30% in



 females.









     Triglycerides and high-density lipoproteins were essentially unaffected




compared to either pair-fed or normal controls.  There was, in fact, no






                                     196

-------
18.  LAWRENCE BERKELEY LABORATORY	Alpen





difference between pair-fed and normal animals for these parameters.  On the




other hand, low-density lipoproteins were  increased by  100%  (males) and 50%




(females) in comparison to pair-fed controls.  The very low  density




lipoproteins were increased by  100%  (males)  in comparison to pair-fed




controls.









     The probability values for these observations are  well  below 0.001, so




the  findings are highly  significant.









     Thirty days after the  exposure was  terminated, the changes in




 lipoproteins and cholesterol  were still  evident.  Histologically, the lungs




 had returned to normal.









 Ongoing and Planned Studies








      We recently conducted a few more studies in surviving  animals  from this




 study.  At 8  to 10 mo following cessation of exposure,  the  anJjnals  are




 returning to  normal and probably are not different from normal.  We have no




 intervening values.  Depending on EPA interest,  we will presumably repeat  this




 study at 0.2  ppm.
                                       197

-------
      19.  OVERVIEW OF RESEARCH AT THE UNIVERSITY OF CALIFORNIA - DAVIS

                                Marvin Goldman

                Laboratory for Energy Related Health Research
                           University of California
                               Davis, CA  95616
INTRODUCTION



     This report provides a brief overview of oxidants research within the

Laboratory for Energy Related Health Research at the University of

California - Davis*  Our interpretation of the concepts of scientific

relevance and merit is presented.  Elsewhere in this volume, Dungworth

(Chapter 20) discusses our work with ozone, and Raabe (Chapter 21) describes

our research on rodent and primate reparative and adaptative mechanisms to

coal fly ash and sulfur compounds.



     Our research is not performed in isolation:  we have an interdisciplinary

program that bridges several of the colleges and falls under the aegis of

various organized research units as well as individual scientists.  Our sister

research unit, the California Primate Research Center, is involved in major

research efforts on the effects of particulate and gaseous pollutants in

animal systems.
                                     198

-------
19.  UNIVERSITY OF CALIFORNIA - DAVIS	Goldman





APPROACH TO RESEARCH










     All of our Laboratory's research  takes a basic  toxicological approach.




In our view, however, there is no  such a  thing as a  "toxic agent"—only toxic




levels.  Our goal is to  understand the spectrum of each dose-response curve




and its relevance (if any).  A major problem, of course, is to determine the




proper experimental model  for a given  study.  We choose each model with




judicious care before proceeding to the stage of study design.









     In attempting to extrapolate  from the animal to the human situation, our




Laboratory seeks to understand the nature of the dose-response curve and of




the dosimetry.  In our view, such  an understanding is central to a valid risk




assessment model.  Without knowledge of the nature of the dose-response curve,




the statistical significance of an exposure level versus a nonexposure level




may fall on the "wrong side" of the extrapolation curve and thus be very




misleading.  With regard to the comments  by Alpen (Chapter 6 of this volume),




we feel that an understanding of the mechanism and an understanding of the




dose-response spectrum are parallel requirements; both are essential to a




valid risk assessment model.









CURRENT EFFORTS









     With regard to oxidant air pollutants, our Laboratory's involvement began




in 1974 as part of a major Department  of  Energy project that is primarily




concerned with fossil fuel combustion  (conventional  fuels as well as the new







                                      199

-------
19.  UNIVERSITY OF CALIFORNIA - DAVIS	Goldman





synthetic fuels).  Our research focuses on airborne particulates from coal




combustion.  We employ unique and innovative approaches to collecting samples




in the field.  Laboratory activities include the generation of aerosols for




animal exposure, physical and chemical categorization of samples, and




determination of the sample components that carry biological importance in




either the pristine (as captured) or degraded (following an interaction with




biological membranes)  state.  The latter activity involves a series of




multistage tests.  These extend from the more conventional Ames assays (for




bacterial mutation) through tests of macrophage function and toxicity and




tests of cell transformation (in vivo experimentation followed by in vitro




testing of the transformed or affected cells).  A major portion of our program




relates to progenitor cells rather than end effect cells;  we examine possible




interactions with the immunologic system and attempt to relate hormonal and




immunologic injury to longer-term carcinogenic potential.
                                     200

-------
            20.   PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY

                             Donald L. Dungworth

                           Department of Pathology
                        School of Veterinary Medicine
                           University of California
                               Davis, CA  95816
INTRODUCTION



     Our work on pulmonary effects of oxidants emphasizes the phenomenon  of

adaptation (tolerance).  Two other primary areas of focus are the evolution of

chronic damage resulting from low-level exposure and the effects on lung

growth.



CHOICE OF ANIMAL MODELS



     Frequently there is much discussion about the appropriateness of a

particular animal model for a particular study.  There is no such thing as a

"perfect model"; in choosing a model, the most important issue is:  What

questions will be asked of the model?  We use two primary species:  rats  (for

statistical purposes and cost effectiveness) and monkeys (for structural/

functional similarities to man and a variety of other anthropomorphic
                                     201

-------
20.  PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY	Dungworth







features).  We develop our idea with rats and look to monkeys for the more




definitive studies.









     Regarding lung structural correlates between experimental animals and




man, scanning electron micrographs effectively illustrate that the monkey has




well developed respiratory bronchioles (like man) whereas the rat does not




(Castleman et al. 1975).  The importance of this is that the respiratory




bronchiole of the monkey is the pulmonary region that is most damaged by low




levels of ozone (03).  At 0.2 ppm, damage is essentially limited to this small




airway (Mellick et al. 1977).  We are reasonably confident that the bronchiole




is also where damage caused by high ambient levels of 03 occurs in man.  We




also prefer the monkey because of its usefulness for studies of pulmonary




function.  We can do repeated experiments; also, the difference in chest wall




compliance between the monkey and dog favors the monkey as the species for




definitive work.









EXPOSURE REGIMENS









     Exposure concentrations are 0.2, 0.5, and 0.8 ppm 03.  These




concentrations are measured against the neutral buffered KI calibration.  We




can factor them to the absolute (UV) standard, but for sake of continuity we




use the older calibration.  The duration of exposure depends on the purpose of




the particular study.
                                     202

-------
20.  PULMONARY EFFECTS OF OZONE  IN THE  RAT  AND  MONKEY    	Dungworth






STUDIES OF ADAPTATION DURING CHRONIC  LOW-LEVEL  EXPOSURE









     At 7 d of exposure  in  normal rats,  we  find the approximate morphologic




no-effect level to be ~0.1  ppm 03  (Plopper  et al.  1979).  The effect of




prolonged exposure to various 03 concentrations is dose-related, but also




depends on the balance between adaptation and smoldering irritation.  Our




general hypotheses are commonly  held  in the field.  The  first hypothesis is




that the centriacinar lesion produced by low-level chronic 03 exposure can




cause and/or exacerbate  chronic  obstructive pulmonary disease.  The outcome




depends on the interplay between adaptation and enhancing factors such as




infectious and immunologic  processes.  The  concept is straightforward, but the




mechanisms involved  are  complex  and largely unknown.








     The second hypothesis  is that O3 has a low-level initiation or promotion




capability for neoplasia of pulmonary epithelium.  On current evidence, we do




not believe that  O3  is a significant  carcinogen or cocarcinogen.  It does




produce a larger  pool of proliferating  cells in small airways in the monkey




 (and in the mouse),  however; thus, a  larger cell population may well be at




risk.  This hypothesis requires  further testing.








     The terms "adaptation" and  "tolerance" are often used in different ways.




In the most general  sense,  adaptation is the process by  which tolerance




occurs.  We prefer to use  "adaptation"  when referring to changes in pulmonary




epithelium.  First of all,  this  usage emphasizes the dynamics of the situation




(i.e., what we are trying  to understand).  Secondly, in  O3 toxicology






                                      203

-------
20.  PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY	Dungworth






literature, "tolerance" has been largely preempted by the edemagenic tolerance




model, which is not involved in low-level insult.  There is another reason for




using the term "adaptation."  In general pathology, the term refers to cell




changes in response to an altered environment, thereby implying that the cells




are there to change.  "Adaptation" does not apply to situations in which




maximal damage occurs initially and cells are either killed or incapable of




further response a short time later.  Our use of "adaptation" denotes a




lessening of detectable damage in spite of continuing insult.  In this




context, there is a higher capacity for adaptation in the rat than in the




bonnet monkey.  Here is yet another example of important species differences




in biology.









     The morphologic manifestation of adaptation is well illustrated by




comparing scanning electron micrographs of the rat lung after 7 and 90 d of




exposure to 0.2 ppm 03.  Qualitatively, the lung appears to have returned to




normal by 90 d, although morphometric studies reveal slightly higher than




normal numbers of macrophages.  Comparing 7 and 90 d exposure to 0.8 ppm 03




shows a considerable decrease in the number of macrophages in the lumen, but a




reorganization of the centriacinar region (Boorman et al. 1980).  This is seen




as an intermediate zone resembling a respiratory bronchiole (normally not




present in the rat at this age).









     Because the damage we see is limited to focal (centriacinar) regions of




the lung, an entirely new approach to sampling is needed.  Randomizing counts
                                     204

-------
20.  PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY	Dungworth







on pieces of minced pulmonary parenchyma is not satisfactory.  Methods to




sample specifically but without introducing bias must be devised.









     With respect to adaptation over a 90-d exposure, the most obvious




differences between the rat and the bonnet monkey are the smaller decrease in




intraluminal macrophages in affected respiratory bronchioles seen in the




monkey and the persistent hyperplasia and hypertrophy of cuboidal, nonciliated




bronchiolar epithelial cells also present in the monkey.









     Our main current study is a  12-mo exposure of bonnet monkeys to 0.3 ppm




03, with evaluation of one group of exposed and control monkeys at the end of




exposure and of a second group after a 3-mo recovery period.  Assessment




during the course of exposure is by pulmonary function, including acoustic




response (Jackson and Olson 1980), and by pulmonary lavage parameters.  These




pulmonary function and lavage studies are important both to monitor changes




occurring during the  12-mo exposure and "to provide a basis of comparison for




what might happen in man.  Both types of measurement are feasible in humans




that have been experimentally or naturally exposed to low concentrations of




03.  The principal terminal assessments will be biochemical and morphometric.









OTHER STUDIES









     We have also initiated studies of the effects of O3 on lung growth and




aging in both the rat and monkey.  Morphometric evaluations being developed
                                      205

-------
20.  PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY	Dungworth


by Dr. Walter Tyler will serve as the most important measures in these

studies.



COMPARISON OF EXPERIMENTAL AND EPIDEMIOLOGIC DATA FOR OZONE AND SULFUR OXIDES



     One important consideration in inhalation toxicology is that the

dose-response curve varies both with time and with the chemical agent.  There

is some suggestion that the curves for 03 and sulfur oxides (SOX) differ

considerably.  With ©3, experimentally obtained concentrations in polluted

urban air cause damage which levels off or lessens with time.

Epidemiologically, it is difficult to establish that these levels are in

general harmful to exposed human populations.  With SOX, on the other hand,

levels that are many times ambient levels are needed to cause an experimental

effect, yet the epidemiologic evidence of harm to exposed populations is more

satisfying than for 03.  Perhaps the recruitment of a number of different

modes of subtle damage occurs gradually until a significant lesion develops.

Copollutants in SOX smogs might also enter into this equation.



REFERENCES
Boorman, G. A., L. W. Schwartz, and D. L. Dungworth.  1980.  Pulmonary effects
     of prolonged ozone insult in rats.  Morphometric evaluation of the
     central acinus.  Lab. Invest., 43:108-115.

Castleman, W. L., D. L. Dungworth, and W. S. Tyler.  1975.  Intrapulmonary
     airway morphology in three species of monkeys:  A correlated scanning and
     transmission electron microscopic study.  Am. J. Anat., 142:107-122.
                                     206

-------
20.  PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY  	Dungworth


Jackson, A. C., and D. E. Olson.  1980.  Comparison of direct and acoustical
     area measurements in physical models of human central airways.  J. Appl.
     Physiol.:  Resp. Environ. Exercise Physiol., 48:896-902.

Mellick, P. W., D. L. Dungworth, L. W. Schwartz, and W. S. Tyler.  1977.
     Short-term morphologic effects of high ambient levels of ozone on lungs
     of rhesus monkeys.  Lab. Invest., 36:82-90.

Plopper, C. G., C. K. Chow, D. L. Dungworth, and W. S. Tyler.  1979.
     Pulmonary alterations in rats exposed to 0.2 and  0.1 ppm ozone:  A
     correlated morphological and biochemical study.   Arch. Environ. Health,
     34:390-395.
WORKSHOP COMMENTARY
 D. L. Coffin;  As a pathologist,  I was  quite  impressed with Dr. Dungworth1 s
 presentation.  Many of  these  things have been examined by other investigators,
 but hia group is putting them together  in  a very nice way.

 Question;   Is it neutral buffered KI  or unbuffered KI that you use for 03
 characterization?

 D. L. Dungworth;  Neutral buffered KI is used to calibrate Dasibi photometric
 analyzers.   The Dasibi  analyzers  are  used  for routine monitoring and the data
 are handled by a computer.  We also send the  Dasibi analyzers away for
 occasional  checks against absolute UV photometric standards.

 Question;   How were tissues prepared  for scanning electron microscopy?  Most
 of the  specimens [shown on slides during the  oral presentation] were devoid of
 any mucous  layer in the airways,  for  example.

 D. L. Dungworth;  They  were fixed by  intratracheal perfusion of modified
 Karnovsky's fixative  at 25 cm fluid pressure.

 Question;   For obtaining the  information you  presented, how much better is
 that than,  say, optical microscopy of those sections?

 D. L. Dungworth;  The scanning electron micrograph affords the ability to
 examine a large block of tissue rapidly and also provides high resolution with
 large depth of field.  This means that  one can examine  in detail large amounts
 of airway and alveolar  surface.

 Question;   You made many statements about  cell identification and the like.
 However, you're dealing with  a topographic analysis of  those cells.
                                      207

-------
20.  PULMONARY EFFECTS OF OZONE IN THE RAT AND MONKEY	Dungworth

D. L. Dungworth;   The scanning electron micrographs were used purely to
highlight our studies.  We routinely correlate findings by conventional light
microscopy, scanning electron microscopy,  light microscopy of large 1-ym
sections, and transmission electron microscopy of specifically selected
regions of pulmonary parenchyma.
                                     208

-------
           21.   BIOLOGICAL EFFECTS OF FLY ASH FROM COAL COMBUSTION

                                  Otto Raabe

                Laboratory for Energy Related Health Research
                           University of California
                               Davis, CA  95616
INTRODUCTION



     The Laboratory for Energy Related Health Research has maintained a

special interest in the airborne particles released from power plants of

various types.  Our interest extends to particulates from combustion of coal,

oil, synthetic fuels, and coal-oil mixtures.  This report discusses an

EPA-supported study of the fly ash emitted in combination with sulfur dioxide

from coal-burning power plants.  The study is a collaborative effort with the

California Primate Research Center; the large team of investigators is drawn

from the fields of biochemistry, pathology, respiratory physiology, and so on.

This author's primary interest and involvement are in the dosimetric and

exposure phases; the following report emphasizes these aspects but also

touches on some of the salient biological results obtained to date.
                                     209

-------
21.  BIOLOGICAL EFFECTS OF FLY ASH	Raabe







SAMPLE COLLECTION









     In some cases, we directly sample the particles leaving a smoke stack.




More frequently, a sample is taken from the hopper of an abatement system




(e.g., an electrostatic precipitator).  For laboratory exposures, the




collected materials are re-aerosolized (see "Exposure Techniques," below).









SAMPLE CHARACTERIZATION









     Prior to biological testing, we perform thorough characterization of both




the untreated and the re-aerosolized fly ash.  Because the particles'




aerodynamic behavior is the primary characteristic controlling deposition in




the respiratory airways, we give close attention to the parameters that are




implicated in aerodynamic behavior.  We look very carefully at the size and




shape of the particles.  Comparisons of untreated versus re-aerosolized fly




ash indicate no significant differences in these parameters.









     The distribution of electrostatic charge on the particles can also be




quite important.  In most of our experiments, we reduce the charge to




Boltzmann equilibrium before animals are exposed.









     Chemical characterization is a major focus.  For this, we employ atomic




absorption analysis, neutron activation analysis, and par tide-induced X-ray




analysis.  Basically, we find that no two particles have exactly the same




chemical composition.  Even two particles of the same size will typically show







                                     210

-------
21.  BIOLOGICAL EFFECTS OF FLY ASH	Raabe






different chemical compositions, particularly with  respect to the trace metals




present on the particle surface.  Certain elements  (e.g., zinc, chromium, and




arsenic) tend to be concentrated in  particles of  smaller size due to a primary




association with the  surface.  Comparisons of untreated versus re-aerosolized




fly ash indicate no significant differences in chemical composition.









     Whether the particles are deliquescent or hygroscopic also has a marked




effect on behavior and on particle  surface chemical reactions.









BIOLOGICAL TEST SYSTEMS









     Our program assesses the biological activity of these materials using




short-term bioassay systems  (e.g.,  the Ames assay)  as well as inhalation




exposures in whole animals.   Whole  animals include  the mouse, rat, monkey, and




beagle  dog.  With regard  to  whole-animal experimentation, almost all current




work with fly ash is  performed  in rodents; these  less expensive species




suffice for the necessary range-finding experiments.









EXPOSURE TECHNIQUES








     For inhalation  studies  we  generate fly ash using the Wright dust feed




mechanism.  For several decades this mechanism has  been used to expose animals




to particles of relatively  insoluble materials.  Our Laboratory has modified




the system by installing  a  cyclone  separator at the exhaust  so that larger




particles  (>2.5 ym in aerodynamic  size) are not an  appreciable part of the






                                      211

-------
21.  BIOLOGICAL EFFECTS OF FLY ASH	Raabe






exposure aerosol.  Prior to inhalation experiments,  the sample material is




always size classified to avoid particles that are outside the respirable




range for the experimental species.  After this initial size separation, the




material is loaded as a packed dust cake into the Wright dust feed mechanism.




The feeder provides a continuous flow of aerosol as well as a second size




separation.  Only particles having an aerodynamic equivalent size of <2.5  ym




pass into the exposure chamber.  A Crypton 85 discharge unit assures that the




exposure aerosol is reduced to Boltzmann equilibrium with respect to




electrostatic charge.








     Fairly large exposure chambers (4 m3) are available at the California




Primate Research Center.  Rodents are typically arranged in a monolayer




configuration in these chambers.  To date, the maximum exposure concentration




has been 200 mg/m3.









     A key point about these experiments is that the particles are small




enough to be readily inhaled, resulting in fairly large deposition in the




lower respiratory tract.  At the same time, the conditions are not "dusty" in




the sense that the animals are covered with fly ash or the chamber is covered




with fly ash.  Aerodynamically, the particles are quite stable in this size




range.
                                     212

-------
21.  BIOLOGICAL EFFECTS OF FLY ASH	 	      Raabe






PRELIMINARY RESULTS









     In early screening with  the  Ames  assay (TA-98  strain of Salmonella), the




mutagenicity of fly ash collected from the  stack of a coal-burning power plant




varied with particle  size.  In general,  the larger  particles showed a lower




number of revertents  per  milligram  of  material  used in the test.  The two




smaller size groups (median size  of ~2.2 and ~3.5 ym) tended to have higher




numbers of revertents in  this test.









     When we examined the effect  of temperature on  three strains of




Salmonella, we found  that heating the  samples to >250°C resulted in




disappearance of  most of  the  mutagenic activity. Efforts to explain this




phenomenon are continuing.  Obviously, in the course of combustion fly ash is




heated  far in excess  of 250°C.  Probably the as yet unidentified mutagens are




formed  as the fly ash is  released from the  stack.








     More recently, extensive studies  on fly ash collected from the hoppers of




electrostatic precipitators  (same power plant)  showed no mutagenic activity at




any particle size. Thus, there is  a very definite  biological difference




between fly ash released  from the stack and fly ash collected in the abatement




system.








     With regard  to whole-animal  studies, one long-term inhalation exposure (4




mg/m3,  8 h/d, 180 d)  resulted in  no remarkable  biological changes.
                                      213

-------
21.  BIOLOGICAL EFFECTS OF FLY ASH        	  Raabe


CURRENT STUDIES



     Presently we are involved in studies of combinations of sulfur dioxide

(50 ppm) with fly ash.  In the first experimental series, animals were exposed

for ~14 d; there were not many remarkable biological effects.  We plan to

repeat these experiments at longer time periods.



WORKSHOP COMMENTARY
G. Rausa;  A recent article in Science states that the mutagenie activity of
fly ash comes from methyl sulfate; the association with temperature appears to
be about the same.  Do you have any technique for ascertaining whether those
results are correct?

O. Raabe;  One of the coauthors of that paper, Lee Hansen, has been in our
Laboratory for about six months on sabbatical leave.  I don't believe he has
been able to find any methyl sulfate in our aerosols.

G. Rausa;  Do you have any conjecture as to what it might be?

O. Raabe;  We are working quite diligently to identify the mutagen; it is a
very important project to us.

Question;  What about polycyclic hydrocarbons?  At a conference a couple of
summers ago, Paul Morrow pointed out that benzopyrene, which is the surrogate
for the rest of the polycyclic hydrocarbons, comes out of the stack in a
volatile state and then condenses on the particles as the plume cools.  Not
finding it in the hopper and finding it on the true fly ash would seem to fit
with that.

O. Raabe;  To be mutagenie, benzopyrene must be activated.  In our results, no
activation effect is apparent.  So it may be something "from" benzopyrene, but
it isn't "just" benzopyrene.

     The problem with our mutagenic fly ash is that it is very rich in organic
materials, and polycyclic aromatic hydrocarbons are not a major part of that
organic constituency.  Dr. Kimball of our Laboratory has been working with
this material;  she finds, I believe, that there is so much organic material
(in comparison to polycyclic hydrocarbons) that it becomes very difficult to
identify polycyclic hydrocarbons as the culprit agent.
                                     214

-------
21.  BIOLOGICAL EFFECTS OF FLY ASH	 	Raabe


Question;  Was the ash studied by the Utah group collected way down where it
was cool, or was it more reflective of the stack?

0. Raabe;  I think they collected it somewhere downstream, outside the power
plant.  Ours was collected from  the base  of the smoke  stack.  Thus, there is
quite a difference.

Question;  How close  is the  geometric surface to the gas absorption surface on
these spheres?

0. Raabe;  That's a good question; unfortunately,  I don't have the data with
me.  Our measurements show that  the gas absorption surface is greater than the
geometric surface, but it's  not  remarkably greater.  For the very large
particles, there are  quite a few hollow spheres and spheres that are tightly
packed with little particles. These features are  not  seen in particles of the
small respirable range, however. The small particles  have a rough surface but
no obvious pores.
                                      215

-------
  22.  OVERVIEW OF RESEARCH AT THE INHALATION TOXICOLOGY RESEARCH INSTITUTE

                               Joe L.  Mauderly

                   Inhalation Toxicology Research Institute
           Lovelace Biomedical and Environmental Research Institute
                             Post Office Box 5890
                            Albuquerque, NM  87115
INTRODUCTION



     This report presents a brief overview of the Inhalation Toxicology

Research Institute.  Included are descriptions of the Institute's resources

and of the current research program.  Chapters 23 through 26 of this volume

present more detailed information on four projects relating to oxidant air

pollutants.



OPERATION AND FUNDING



     The Inhalation Toxicology Research Institute is operated by the Lovelace

Biomedical and Environmental Research Institute for the Department of Energy

under the Assistant Secretary of the Environment.  The parent foundation is

the Lovelace Medical Foundation.  Most (currently, 80%) of the Institute's

funding comes from the Department of Energy, but several projects are funded

by other governmental agencies.  Funding from these other agencies (EPA, the
                                     216

-------
22.  INHALATION TOXICOLOGY RESEARCH  INSTITUTE   	Mauderly







Nuclear Regulatory Commission, and the  Department  of Defense) is made possible




through interagency agreements with  the Department of Energy.









FACILITIES AND STAFF









     The Institute is located on a 138-acre site on Kirtland Air Force Base




south of Albuquerque, New Mexico.  Facilities  include ~215,000 ft2 of floor




space; the current replacement value is somewhere  on the order of $20 million.




All  support  services  are provided onsite;  the  Institute is essentially




self-contained.









     There is maintenance housing for ~15,000  small animals; the exact number




housed, of course, depends  on the mix between  mice and larger species.  At




present, there are  facilities for housing ~2,000 dogs.  The Institute




maintains a  breeding  colony of beagle dogs; mice,  rats, and Chinese hamsters




are  also bred.   The current production  capability  is ~4,000 mice and ~2,000




rats per month,  and ~400 dogs per year.  Actual production depends on research




needs at any given time.









      In general, the  Institute  at full  strength employs ~250 people.  The




present total staff of 227  includes ~50 doctoral-level personnel and a mix of




disciplines: ~50%  in the life  sciences (biology  or medical-related degrees),




~25% in biophysics and biomedical engineering, and ~25%  in the physical and




engineering  sciences  (including mathematical modeling and risk assessment).
                                      217

-------
22.  INHALATION TOXICOLOGY RESEARCH INSTITUTE	        Mauderly







RESEARCH CAPABILITIES









     For several years the Institute has maintained strong capabilities in the




area of inhalation exposure to both radioactive and nonradioactive materials




and aerosols.  There is also considerable analytical capability/ not only in




chemistry but also in radioanalysis.









     Another Institute strength is the diversity of biological end points that




can be evaluated.  Capabilities encompass not only the clinical end points




(e.g., physical examinations, radiography, clinical chemistry, and hematology)




but also organ function (e.g., respiratory function, mucociliary clearance,




and lung defense mechanisms).  In other words, it is possible to evaluate




organ function as well as whole body function.









     Deposition and retension of inhaled materials (including aerosols,




vapors, and gases) continue to be important end points in determining critical




doses to tissues.  In the area of immunology, there is a strong program




focusing on the impairment in immune mechanisms caused by inhaled




environmental pollutants (particularly in the lung-associated lymph nodes) and




on determination of the immune pathways involved.









     Mutagenesis and carcinogenesis testing capabilities, of course, are




related, and one of the Institute's major strengths is the ability to go from




in vitro mutagenicity studies to whole-animal carcinogenicity studies.




Current protocols involve bacterial and mammalian cell mutagenicity testing






                                     218

-------
22.  INHALATION TOXICOLOGY RESEARCH  INSTITUTE	Mauderly







followed by long-term  studies  of  cancer  development  in animals.  In the field




of cytogenetics, there is a  group looking directly at the genetic




transformations that take place in mammalian cells and scoring chromosome




damage.









     Mortality, morbidity, and pathology are common  end points in any




long-term animal study. The Institute employs  not only gross morphology and




light microscopy,  but  also transmission  and scanning electron microscopy.




Quantitative morphometric measurements are also commonly used.









     The Institute's capabilities in evaluating biochemical end points are




particularly strong.   One  focal area involves lung lipid chemistry, Type II




cell metabolism and injury,  and lung surfactant studies.  Another strong area




is the  biochemistry of lung  connective tissue.









CURRENT PROJECTS









     Approximately 50% of  the  Institute's current research program relates to




nuclear power  production.   Studies involve both the  fission products and the




transuranics  such  as plutonium and other alpha  emitters.  Both short- and




long-term studies  are ongoing in  these areas.









     The  Fossil Fuel program evaluates effluents from stationary sources:




coal combustion and coal gasification.  In the  area  of coal combustion, both
                                      219

-------
22.  INHALATION TOXICOLOGY RESEARCH INSTITUTE	Mauderly







conventional combustors and fluidized bed combustors are undergoing




evaluation.









     In the Mobile Sources program, the main effort currently relates to the




study of diesel exhaust.  A five-part program begins with characterization of




the exhaust material and extends through animal life-span studies.









     The Solar program is the only area involving studies that are not




primarily inhalation-oriented.  Because of the possibility for heat transfer




fluids to enter the potable water supply. Institute researchers are studying




the toxicology of these materials.  Studies focus on the ingestion and contact




toxicology of heat transfer fluids.









     The Conservation program* s main effort relates to the study of fibers




(both natural and man-made) used in insulating processes.









     The spectrum of materials that can be analyzed ranges from solid




particles to droplets, vapors, and gases.  Materials currently analyzed




include conventional combustor and fluidized bed combustor fly ash, diesel




particulates, ammonium sulfate (as either a solid particle or droplet),




sulfuric acid, and irritant or oxidant vapors and gases.  Most of the current




studies are oriented towards single agents.  As the need is recognized, the




Institute will undertake studies of combinations of these materials, bearing




on the problem of what happens when ambient mixtures are inhaled.
                                     220

-------
22.  INHALATION TOXICOLOGY RESEARCH INSTITUTE    	Mauderly






     Studies can also progress in the opposite direction.  For example,




efforts in the diesel exhaust study started with diluted raw exhaust but now




involve some of the individual components.









     In subsequent chapters  of this volume, other authors from the Inhalation




Toxicology Research Institute report results of studies that have been




concluded, review ongoing efforts, and  emphasize proposed directions for




future research.  Henderson  (Chapter 23)  discusses a project in which the




primary goal was to develop  and  demonstrate the usefulness of cellular damage




indicators for lung injury.  The techniques developed  in this recent, exciting




work show much promise  as early  detectors for lung cell injury and death.








     Pickrell  (Chapter  24) outlines a program on inhaled nitrogen dioxide that




includes  both  short-  and long-term studies.  Silbaugh  (Chapter 25) discusses a




program to examine the  structural and functional changes associated with




inhaled sulfates, again including both  short- and long-term studies.








     Finally, Wolff  (Chapter 26) reviews studies primarily related to the




clearance of inhaled  particles  in the lung and  the effect of inhaled sulfates




on the pattern  (velocity or  rapidity) of clearance of  inhaled materials.  This




is actually a  study of  lung  defense mechanisms.
                                      221

-------
     23.  CELLULAR (IN VIVO) AND BIOCHEMICAL CHANGES FOLLOWING INHALATION
     OF ACID SULFATES:  RAPID SCREENING TESTS TO DETERMINE THE PULMONARY
                   RESPONSE TO COMBINED POLLUTANT EXPOSURES

                             Rogene F. Henderson

                   Inhalation Toxicology Research Institute
           Lovelace Biomedical and Environmental Research Institute
                             Post Office Box 5890
                            Albuquerque, NM  87115
PROBLEM



     A major lack in the information required to regulate air pollutants is

knowledge of how various pollutants interact to produce synergistic, additive,

or negating effects in the exposed populace.  To expose experimental animals

to single component aerosols on a long-term basis is expensive; adding double

or triple components increases the cost even more.  One approach to the

problem is to use rapid in vitro screening tests, which cost less and yield

results quickly.  However, for studying pollutant interactions, an in vitro

approach does not take into account the interactions that affect site of

deposition, clearance, retention, and (consequently) pollutant dose to tissue.

Epidemiology assesses real-world pollutant mixtures but cannot isolate the

effects of individual components of those mixtures.  What is needed is a rapid

bioassay that forms a bridge between the _in vitro studies and the long-term _in

vivo and epidemiologic studies.
                                     222

-------
23.  PULMONARY EFFECTS OF POLLUTANT COMBINATIONS	Henderson






APPROACH









     Our laboratory uses bioassays that involve short-term _in vivo  inhalation




exposures followed by rapid evaluation of pulmonary response by a variety of




indicators of early lung damage.  Our past use of  these  bioassays to measure




pulmonary response to single pollutants is summarized below.









Lavage Fluid Analysis








      Previous work  showed  that analysis of saline  lung washings of  exposed




animals can be  used to  detect an inflammatory response in the lung  (Henderson




et al. 1978a,  1978b,  1979a, 1979b).   Extracellular lactate dehydrogenase in




 the airway fluid indicates increased cell membrane permeability;  lysosomal




 enzymes indicate phagocytic activity? increased sialic acid indicates  an




 irritant response in the upper airway? increased soluble protein  in the




 airways indicates damage of the capillary-alveolar barrier;  and an  influx of




 neutrophils is consistent with an inflammatory response.  Biochemical  and




 cytological changes in lavage fluid correlated well with morphological




 indicators of an acute inflammatory response in Syrian hamster  lungs  exposed




 to a nonionic surfactant (Henderson et al. 1978a)  and to metal  salts




 (Henderson et al. 1979a, 1979b).  The most sensitive indicator of the




 multifocal lung damage (terminal bronchiolitis) characteristic of oxidants  was




 an increase in the polymorphonuclear cells in  lavage fluid from hamsters




 exposed to nitrogen dioxide  (NO2) (DeNicola et al.  1979).  Rats exposed to




 sulfuric acid (H2S04) mist showed an increase  in  the sialic acid content of






                                      223

-------
23.  PULMONARY EFFECTS OF POLLUTANT COMBINATIONS	Henderson






the lavage fluid as well as speeding of mucociliary clearance (Henderson et




al. 1978c).








1^C-Thymidine incorporation in Lung Cells









     Pulse labeling of Chinese hamsters with 3H-thymidine demonstrated an




increased uptake of radiolabel in the lungs of animals exposed to N©2 (Hackett




1979).  Tissue oxidizer analysis of 3H retention by the lung could be used as




a rapid screen for cellular damage.  In animals showing the greatest response,




autoradiographic techniques could be applied to portions of the tissue to




determine the actual site of damage.








Lipid Synthesis









     One indicator of cellular damage is an increased synthesis of membrane




lipids during repair processes,  fln increased uptake of ^C-palmitate into




membrane lipids was detected in lung cells from animals exposed to 5 ppm NO2




for 8 h (Pfleger and Rebar 1978).  The same label would also detect increased




synthesis of the surfactant lipid, dipalmitoyl lecithin.  This lipid is




synthesized by the Type II epithelial cell, which is known to proliferate in




response to the most common type of lung injury—loss of the Type I epithelial




cell.
                                     224

-------
23.  PULMONARY EFFECTS OF POLLUTANT COMBINATIONS       	Henderson






PLANNED RESEARCH









     The bioassays described above have been  used  to  detect pulmonary response




to single pollutants.  We propose  to  extend their  use to detect the effects of




combinations of pollutants.









     We plan to use  the  Charles  River Hartley guinea  pig because of our




previous experience  with this animal's response to H2S04 (Silbaugh et al.




1978).  Animals will be  exposed  to four combinations  of NO2 and H2S04 (5 ppm




N02 plus 1.0 mg/m3 H2SO4;  1  ppm  NO2 plus  1.0  mg/m3 H2S04? 5 ppm NO2 plus 10




mg/m3 H2S04; 1 ppm NO2 plus  10 mg/m3  H2SO4> and to the same levels of the




individual  components.   Control  animals will  be exposed to room air.




Exposures will be for 6  h on 2 subsequent days.  The  animals will be evaluated




at 48 h after initiation of  the  exposure, which has been identified as the




time of peak response to NO2 (DeNicola et al. 1979; Hackett 1979; Pfleger and




Rebar 1978).  The end points to  be measured include thymidine and palmitate




uptake into lung  tissue  as well  as lavage fluid parameters.  The lavage fuid




parameters  will include  lactate  dehydrogenase, acid phosphatase, sialic acid,




soluble protein,  glutathione peroxidase and reductase, fibrin degradation




products, and total  and  differential  cell counts.  In addition to these end




points, we  will perform  light microscopic evaluation  of slices of deep lung




tissue, trachea,  larynx, and nasal turbinates. In animals shown by the screen




to be most  affected  by exposure, autoradiographs of the 3H-thymidine labeled




tissue and  scanning  electron micrographs  of the tissues will be obtained.
                                      225

-------
23.  PULMONARY EFFECTS OF POLLUTANT COMBINATIONS	Henderson


     Upon completion of this study, we will examine the effects of combined

exposure to H2S04 (1 mg/m3) and fly ash (5 mg/m3) from fluidized bed

combustion of coal.  The final experiment will examine the effects of combined

exposure to all three pollutants:  NO2 (5 ppm), H2SO4 (1 mg/m3), and fly ash

(5 mg/m3).



     This research should aid in validating the screening system; the higher

pollutant levels have been included to meet this validation need.  The major

result of the experiment will be to determine if any synergistic effects occur

in combined acute exposures to low levels of the described pollutants.  This

determination should contribute to the groundwork for any regulatory decisions

based on combined effects from pollutant mixtures.



ACKNOWLEDGMENTS
     This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCES
DeNicola, D. B., A. H. Rebar, and R. F. Henderson.  1979.  Early indicators of
     pulmonary damage in Syrian hamsters exposed to N02«  In:  Annual Report
     of the Inhalation Toxicology Research Institute, 1978-1979.  Publication
     LF-69, Lovelace Medical Foundation, Albuquerque, New Mexico, pp.
     524-528.
                                     226

-------
23.  PULMONARY EFFECTS OF POLLUTANT COMBINATIONS	Henderson


Hackett, N. A.   1979.  Proliferation of lung and airway cells  induced by
     nitrogen dioxide.  J.  Toxicol. Environ.  Health,  5:917-928.

Henderson, R. F., E. G. Damon,  and T.  R.  Henderson.   1978a.  Early  damage
     indicators  in  the lung.   I.   Lactate dehydrogenase activity in the
     airways.  Toxicol. Appl.  Pharmacol., 44:291-297.

Henderson, R. F., B. A. Muggenburg, J.  L. Mauderly, and W. A.  Tattle.  1978b.
     Early damage indicators  in the lung.  II.   Time  sequence  of protein
     accumulation and  lipid loss in the airways of beagle dogs with beta
     irradiation of the lung.   Radiat.  Res.,  76:145-158.

Henderson, R. F., R. K. Wolff,  A. H. Rebar, D.  B.  DeNicola, and  R.  L.  Beethe.
     1978c.  Early  indicators of lung damage from inhaled sulfuric  acid  mist.
     In:  Annual Report of  the Inhalation Toxicology  Research  Institute,
     1977-1978.  Publication  LF-60, Lovelace Medical  Foundation,  Albuquerque,
     New Mexico, pp. 352-355.

Henderson, R. P., A. H. Rebar,  and D.  B.  DeNicola.  1979a.  Early damage
     indicators  in  the lung.   IV.  Biochemical  and cytologic response  of the
     lung to lavage with metal salts.   Toxicol. Appl.  Pharmacol., 57:129-135.

Henderson, R. F., A. H. Rebar,  J. A. Pickrell,  and G.  J.  Newton.  1979b.
     Early damage indicators  in the lung.  III.  Biochemical and cytological
     response of the lung  to  inhaled metal salts.  Toxicol. Appl. Pharmacol.,
     50:123-136.

 Pfleger, R. C.,  and A. H.  Rebar.  1978.  Pulmonary alveolar Type II cells and
     macrophage  damage following an NO2 inhalation exposure.   In:   Annual
     Report of the  Inhalation Toxicology Research Institute, 1977-1978.
     Publication LF-60, Lovelace Medical Foundation,  Albuquerque, New Mexico,
     pp. 419-424.

 Silbaugh,  S. A., D. G. Brownstein, R. K.  Wolff, and J. L. Mauderly.  1978.
     Airway response of the Hartley guinea pig to sulfuric acid  aerosol. In:
     Annual Report  of  the  Inhalation Toxicology Research Institute, 1977-1978.
     Publication LF-60, Lovelace Medical Foundation,  Albuquerque, New Mexico,
     pp. 360-363.
 WORKSHOP COMMENTARY
 J. D.  Hackney;   In the 3H-thymidine labeling, what was labeled, and how?  Was
 this by autoradiography or was this extracted from the lung?
                                      227

-------
23.  PULMONARY EFFECTS OF POLLUTANT COMBINATIONS	Henderson
R. F. Henderson;  This was by an autoradiographic technique which showed
labeling of epithelial cells.  But autoradiography is not a screening tool.
It is a very slow procedure.  For a screening tool we propose to use the first
step of our autoradiography studies in which a tissue oxidizer is used to
determine which animals have actually taken up label.  The tissue oxidizer
oxidizes the 3H and 14C in the tissues to 3H2O and 14CO2, which can be
collected and counted in a liquid scintillation spectrometer.

D. E. Gardner;  In your studies showing increases in polymorphonuclear cells,
did you look at the functioning of these polymorphonuclear cells, their
phagocytic capabilities, or the macrophages—

R. F. Henderson;  No.  The macrophages appeared activated (that is, they were
enlarged and had large inclusion bodies), but we made no measurements other
than to observe the morphological changes.

D. E. Gardner;  Did you find a decrease in macrophages with the increase in
polymorphonuclear cells following NO2 exposure?

R. F. Henderson;  Both the macrophages and the polymorphonuclear cells in the
lavage fluid increased.
                                     228

-------
                24.  RESPIRATORY TOXICOLOGY OP NITROGEN OXIDES

                               John A. Pickrell

                   Inhalation Toxicology Research Institute
           Lovelace Biomedical and Environmental Research Institute
                             Post Office Box 5890
                            Albuquerque, NM  87115
PROBLEM



     Assessing the health effects of nitrogen oxides (NOX)  provides

information that is necessary in the regulatory process.  Although short-term

inhalation of high concentrations of nitrogen dioxide (NO2) is known to damage

both pulmonary alveolar macrophages and Type I epithelial cells,  the health

effects of inhaling environmental levels of NO2 for long periods  of time have

not been adequately defined.  Little is known concerning the health effects of

inhaling NO2 in combination with other pollutants such as ozone (O3) or fly

ash.  Also, the effects of inhaling other oxides of nitrogen (nitric oxide,

nitrates, nitrites, peroxyacyl nitrate, nitrous or nitric acid, nitrosamines)

have not been studied extensively.



     Lung tissue exposed to a variety of N02 levels seems to adapt to this

insult.  The mechanism and extent of the adaptation and its relation to the

animal's health status are poorly understood.  Indicators for a variety of
                                     229

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES _ Pickrell







changes are required to assess the extent of pulmonary injury, adaptation, and




residual injury.  Biochemical indicators of damage, although quite sensitive,




must be related to cellular and morphological changes to understand lung




injury following NC>2 inhalation.  Of special importance are those parameters




which persist long after pulmonary injury has ceased.  These are the




biomedical determinants of the resulting pulmonary injury.









     The experiments described below were directed toward defining the NO2




dose causing acute effects and the consequence of those effects for




development of chronic pulmonary injury.  Nitrogen dioxide damage to pulmonary




alveolar Type I cells is reflected in cell necrosis, pulmonary edema, an




influx of polymorphonuclear (PMK) leukocytes, and an increased pulmonary




distensibility.  This damage leads to turnover of extracellular matrix,




heightened immune response in the regional lymph nodes from increased access




of antigen to interstitial tissue, and a proliferation of Type II alveolar




pneumocytes.  Most indicators of damage return to normal and the lung seems to




"adapt" to the NO2 injury.  Spindle cell proliferation, continued low-level




inflammation, continued increased pulmonary distensibility, and increased mean




linear intercepts and pulmonary collagen suggest that incomplete pulmonary




adaptation to N©2 exposure may lead to chronic pulmonary injury.
     Future work will define the relation of incomplete adaptation to




development of chronic pulmonary injury from VfO2 levels relevant to




smog-filled urban atmospheres.  The relation of NO2 and 03 to pulmonary injury




will be explored (again at levels relevant to environmental exposures).






                                     230

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES   	Pickrell






Finally, acute studies will assess the toxicities.of other oxides of nitrogen




in comparison to NO2.









APPROACH









     The effect of inhaling IK>2  at several levels was studied in  Chinese and




Syrian hamsters and  Fischer 344  rats.   Chinese hamsters were used in




cytokinetic  studies  so  that the  resulting data could be correlated with data




from parallel analyses  of lung chromosomes in the same species.   Chinese




hamsters were selected  for the lung chromosome analysis because of the




species' simple chromosome pattern.  Syrian hamsters had been determined to be




free of long-term lung  disease in earlier parallel studies,  and were therefore




selected.  Later,  Fischer 344 rats were used because of a longer  lifespan in




comparison to the Syrian hamsters.  Where comparisons between species were




made, IK>2  seemed  to  induce similar series of morphological,  cytological, and




biochemical  events.   The significance of each change is discussed below.








     Exposure of  Fischer 344  rats to 20 ppm NO2 for 48 h led to damaged




alveolar Type I cells (Pickrell et al. 1978).  These changes were reflected by




increased  airway  lactate dehydrogenase (LDH), suggesting pulmonary cell




damage, and  by  increased airway alkaline phosphatase, suggesting  pulmonary




cell necrosis.  Increased airway protein and trypsin inhibitory capacity




suggested  pulmonary edema.  Pulmonary edema in the airways suggested both




increased  vascular permeability and an interruption of the tight  junction of




alveolar Type I cells due to cell damage.  This Type I cell damage led to an






                                      231

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES	Pickrell






influx of PMN leukocytes into the airway and an increased turnover of collagen




of the extracellular matrix by 14 d.  This change reflected collagenolysis and




proteolysis of pulmonary interstitium probably from exposure to airway enzymes




from inflammatory cells (macrophages and PMN leukocytes).  Such exposure was




possible, since damage to alveolar Type I cells had occurred.  Increased




parenchymal alkaline phosphatase and neutral protease were consistent with the




terminal bronchiolitis observed at that time.  Results from other studies in




rats and Syrian hamsters suggested damage to alveolar Type I cells.









     In a second study, exposure of adult male Fischer 344 rats to 26 ppm NC>2




for 24 h increased the number of anti sheep red blood cell plaques formed by




lung-associated lymph nodes (LALN).  This change may have reflected an altered




load to these nodes due to pulmonary Type I cell damage.  In another study,




Syrian hamsters exposed to 12 ppm NO£ for 48 h had a mild necrotizing terminal




bronchiolitis with increased airway PMN leukocytes.  In another experiment,




Syrian hamsters exposed to 5 ppm NO2 for 8 h showed increased quantities of




viable airway granulocytes by 48 h after initiation of exposure, suggesting




damage to alveolar Type I cells and recruitment of inflammatory cells.









     Necrosis of pulmonary alveolar Type I cells led to hyperplasia of




pulmonary alveolar Type II cells (Evans et al. 1977).  In Chinese hamsters




exposed to 15 ppm N(>2 for 24 h, an increased labeling index for 3H-thymidine




was observed by 24 h and persisted to 3 weeks after exposure (Hackett 1979).




Type II cells were observed twice as frequently in the terminal bronchioles as




in other pulmonary areas.  The Type II cell cycle time was reduced from 26 d






                                     232

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES      	Pickrell






to 3 d.  These changes were consistent with terminal bronchiolitis. Type I




cell death, and Type II cell proliferation.   Epithelial  cells lining small




airways and alveoli were more susceptible  to NO2 exposure than were bronchial




and tracheal epithelia (Hackett  1979).  Syrian  hamsters  exposed to 12 ppm NO2




for 48 h showed bronchiolar epithelial hyperplasia.  Fischer 344 rats exposed




to 20 ppm NC>2 for 48 h had terminal  bronchiolitis and  changes suggestive of




Type II cell hyperplasia  (Pickrell et al.  1978).  Both studies were consistent




with damage to alveolar Type I cells in terminal bronchiolar areas leading to




Type II cell hyperplasia.









     Type  II alveolar cell hyperplasia was associated  with alterations in




synthesized lipids probably destined for cell membranes.  Cells from an




"enriched  Type II cell fraction" of  lungs  of Syrian hamsters exposed to 15




 (Pfleger et al.  1980), 5, or  1 ppm for 8 h had  increased incorporation of




radiolabel into  unsaturated phosphatidyl choline, reflecting increased




production of membrane lipid.  These changes were consistent with increased




production of cell membrane lipid by Type  II cells  (exposed to 1 ppm N02 for 8




h),  a  step which must precede hyperplasia  of Type II alveolar pneumocytes.








     When  rats were  exposed to 20 ppm N02, most indicators of damage returned




to normal  by 2 mo after  initiation of exposure  (Pickrell et al. 1978).  The




lung seemed to "adapt" to continued  exposure.  Return  of airway granulocytes,




tissue alkaline  phosphatase and  acid protease,  and  tissue LDH to control




levels by  2 mo after initiation  of  exposure suggested  that parenchymal tissue




inflammation was no  longer present.   Return of  airway  LDH and alkaline






                                      233

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES _ Pickrell






phosphatase to control levels suggested that parenchymal cell damage and




necrosis were no longer occurring.  Finally/ return of airway collagen to




control levels suggested that disruption of the extracellular matrix had




ceased.  This was consistent with a reduction in Type I alveolar cell necrosis




and an "adaptation" of the lung to continued NC>2 exposure.  Lungs of Syrian




hamsters exposed to 12-22 ppm NC>2 had returned to normal by 21 d after




initiation of a 48-h exposure.  By 7 d after exposure of Fischer 344 rats to




26 ppm NO2 for 24 h, the ability of LALN to form plaques following challenge




with sheep red blood cells had returned to the control level, suggesting that




the antigen load to LALN had returned to normal.  This was consistent with




cessation of damage to Type I alveolar cells and with lung adaptation to
     Several parameters remained altered (relative to control animals) in




Fischer 344 rats exposed to 20 ppm NC>2 for 2 mo.  These changes suggested that




any "recovery" to that exposure was incomplete by 10 mo after stopping




exposure.  Chronic pulmonary injury had presumably occurred.  Pulmonary




distensibility at 20 cml^O was increased relative to controls at 6.5 mo but




not at 12 mo after initiation of NC>2 exposure.  This morphometric parameter




slowly adapted, suggesting that pulmonary inflammation or alveolar distension




lessened.  Persistence of increased neutral protease throughout the study




suggested that chronic low-level inflammation continued 10 mo after N(>2




exposure ceased.  The increase of mean linear intercept observed 2-12 mo after




initiation of exposure was not accompanied by histological evidence of septal




destruction or emphysema.  The thinning of walls and increased alveolar size




( histologically within normal limits) observed 12 mo after initial NC>2






                                     234

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN  OXIDES      	Pickrell






exposure were consistent  with increased  linear  intercept and may have




represented a reduction in elastic recoil.   The degree  to which elastin




quantity or metabolism were altered was  not determined.  By 14 d after




initiation of NC>2  exposure, alveolar walls  were subtly  thickened by apparent




spindle cell proliferation suggestive of early  fibrosis, although collagen




content was normal.   Later, collagen content was increased from 2-12 mo after




initiation of NC>2  exposure, although no  histological  evidence of fibrosis was




present.  One may  speculate that these changes  were consistent with collagen




replacement of  alveolar  elastic  tissue,  loss of elastic recoil, and increased




mean linear intercepts.   If the  replacement was sufficiently diffuse, no




histological evidence of  wall destruction might have  been noted.  Certainly




these changes  suggest that incomplete "adaptation"  to pulmonary injury from




N02 exposure may lead to chronic pulmonary  injury not resolved 10 mo after




cessation of exposure.  Such  chronic pulmonary  injury might occur even in the




absence of  septal  destruction.   The relation of pulmonary "adaptation" and




chronic pulmonary injury to lower levels of IK>2 with  intermittent "spike"




 levels relevant to concentrations encountered in smog should be of




considerable  interest.









 PLANNED RESEARCH








      Future work assessing the toxicity of inhaled NOX  will bear on  important




 scientific  questions relevant to the goals of EPA.   Further work with NO2 will




address the still unanswered questions of (1) a minimal effects or low-risk




 level, (2)  adaptation to intermittent NO2 spikes simulating urban levels, and






                                      235

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES       	Pickrell






(3) the effects of NO2 in older animals or in animals with preexisting




disease.  The most important and relevant question is the mechanism, extent,




and consequences of adaptation to intermittent spikes of NO2 superimposed upon




a 3 to 5 fold lower maintenance level of NC>2«








     A second important area for which there is little information is the




interaction of NO2 with such other pollutants as 03,  peroxyacyl nitrate, fly




ash, nitric oxide, and oxides of sulfur.  The most frequent interaction of NO2




is with 03; these pollutants frequently coexist in smog.  Both have




considerable potential for producing chronic pulmonary injury.  The pulmonary




system seems to at least partially "adapt" to either pollutant.









     Another important scientific need is an assessment of the relative




toxicities of other oxides of nitrogen, peroxyacyl nitrate, nitrosamines,




nitrates and nitrites, and nitric and nitrous acid.  Peroxyacyl nitrate is a




potent oxidant.  Preliminary reports suggest that it has the same approximate




relative toxicity as ©3 and greater relative toxicity than NC>2, although




ambient atmospheric conditions seem quite low.  Nitrosamines are known




carcinogens, but ambient atmospheric conditions appear considerably lower than




even those of peroxyacyl nitrate.









     The mechanism, extent, and consequences of pulmonary adaptation to NOX




will be studied using a maintenance level of 1 ppm NO2 for 24 h/d, 5 d/week.




A spike of 5 ppm N02 will be superimposed on the 1 ppm level in a second group




of animals.  This spike will occur 2 times/d, to simulate urban pollution






                                     236

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES            	Pickrell






conditions.  These groups  will be compared to a third group of animals exposed




to 5 ppm NC>2 5 d/week and  to  a control group housed in exposure chambers.




Sacrifices will occur from 2  weeks to 2 yr after the initiation of  exposures.




The study will employ ~250 Fischer 344 rats.









     End points will relate to the general areas of pathology,  morphometry,




physiology, and biochemistry. The pathological assessment will include




histopathology and electron microscopy (both scanning and transmission




electron microscopy).   Morphometric end points will include pulmonary




distensibility  (displaced volume of lung fixed at 24 cmH2O/kg body  weight) and




mean linear intercept.   Internal surface areas will be calculated.   Pulmonary




physiology will  include measurement of static compliance in excised lungs and




measurement of  other respiratory functions in selected animals.  Biochemical




measurements  will include lipoperoxidation, glutathione peroxidase  and




 reductase, acid and alkaline phosphatase, LDH and glucose-6-phosphate




 dehydrogenase,  protein, trypsin inhibitory capacity, protease activity as a




 function of pH,  and collagen and elastin metabolism.








      The effect of inhaling NC>2 and an associated pollutant, 03, is an




 important scientific question.   Fischer 344 rats will be used to establish  a




 dose response and interaction.   The end points to be studied are described




 above.  Six  groups will be exposed to NO2, O3, or both:  (1) 1 ppm NO2, (2) 5




 ppm N02,  (3)  0.25 ppm 03, (4) 1  ppm O3, (5) 1 ppm NO2 plus 0.25 ppm 03, and




 (6) 5  ppm N02 plus 1 ppm O3.   An equal  sized group of controls will be housed




 in chambers  and exposed to ambient filtered air.  Each group will contain -140






                                      237

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES  	Pickrell


rats.  Serial sacrifices will occur from 1 mo to 3 yr after initiating

exposure.  Approximately 60 rats will be retained for lifespan studies.



     Assessement of the relative toxicities of other oxides of nitrogen will

be accomplished in short-term acute experiments.  These dose-response studies

will employ at least two levels, one quite high.  Responses will be observed

in days to weeks as opposed to years (except for known carcinogens, for which

observations may continue for up to 1 yr).  The relative toxicity information

will be combined with estimates of environmental concentrations in an effort

to predict relative hazard.  The compound with the greatest relative hazard

will be selected for a long-term toxicity study.  We consider peroxyacyl

nitrate and nitrosamines to be the most relevant pollutants for initial

assays.  Peroxyacyl nitrate is reported to be a potent oxidant:  preliminary

work suggests its toxicity to exceed that of NC»2 and to approach that of 03.

To date, peroxyacyl nitrate has received little attention because its ambient

concentrations appear to be extremely low.  Nitrosamines are known to be

potent carcinogens.  However, their ambient concentrations appear to be orders

of magnitude below even those of peroxyacyl nitrate.  Methods of detecting and

quantifying peroxyacyl nitrate and nitrosoamines continue to be improved, and

care must be taken in interpreting measured atmospheric levels.



ACKNOWLEDGMENTS
     This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
                                     238

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES	Pickrell


under Contract Number EY-76-C-04-1013  and conducted in facilities fully
accredited by the American Association for Accreditation of  Laboratory Animal
Care.
REFERENCES
 Pickrell, J. A.,  A.  H.  Rebar,  E.  G.  Damon,  R.  L.  Beethe,  R.  C. Pfleger, and C.
     H. Hobbs.   1978.   Exposure to NC>2 as a model for producing  emphysema:
     Early results.   In:   Annual Report of  the Inhalation Toxicology Research
     Institute,  1977-1978.  Publication LP-60, Lovelace Medical  Foundation,
     Albuquerque, New Mexico,  pp. 428-432.

 Evans, M. J., L.  J.  Cabral-Anderson, and G. Freeman.   1977.   Effects of NO2 on
     the  lungs of aging rats.   II.  Cell proliferation.  Exp. Mol. Pathol.,
     27:366-376.

 Hackett,  N.  A.   1979.   Proliferation of lung and airway cells induced by
     nitrogen dioxide.   J. Toxicol.  Environ. Health,  5:917-928.

 Pfleger,  R.  C.,  A. H. Rebar, and J.  A. Pickrell.   1980.  Biochemical effects
     of nitrogen dioxide on pulmonary alveolar Type II cells and pulmonary
     alveolar macrophages from Syrian hamsters.  Toxicol. Appl.  Pharmacol., in
     press.
 WORKSHOP COMMENTARY
 P. E. Morrow;   You mentioned something about the antigenic burden of  the  lymph
 nodes.  Would you expound on that?

 J. A. Pickrell:  Dr. David Bice of our laboratory performed a study on
 pulmonary immunology and NO2.  Rats were exposed to 26 ppm NO2 for 24 h.  Use
 of sheep red blood cells to stimulate the lung immune response in regional
 lymph nodes led to increased plaque formation in LALN.  The increase  in plaque
 formation peaked at 2 d, was reduced at 3 d, and had returned to normal by  7 d
 after exposure.  His interpretation was that the increased load to the LALN
 was due to interruption of the Type I alveolar pneumocyte barrier, that the
 reduction represented Type II cell proliferation and phagocytosis °f
 peroxyacyl nitrate, and that normal pulmonary architecture had been restored
 by 1 week after exposure.

 P. E. Morrow;   What does that have to do with the Type I cell barrier?


                                      239

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES 	Pickrell


J. A. Pickrell:  The sheep red blood cells were introduced into the lung and
the lymph nodes were stimulated.  Bice interpreted this to mean that the
epithelial integrity of the lung was interrupted; I think he had some
histopathology to back this up.  He concluded that more sheep red blood cells
were therefore available to the lymph nodes and that the lymph nodes were
simply responding to an increased load in a normal immunologic manner.

P. S. Morrow;  I follow the antigenic stimulation concept but I don't think
the Type I integrity follows.

     There is another point that I wish to clarify.  In all of the studies in
which you've done thymidine incorporation and have looked for proliferative
changes and so on, you did not mention the Clara cell in these rodents.  I
thought that this was a very prominent feature of small airway effects:  that
there would be damage in that area and there would be Clara cell
proliferation.  But it hasn't been mentioned; instead, the discussion focused
on Type II cells and basal cells.  Did you look specifically at the Clara
cell?

J. A. Pickrell;  It was looked at by Dr. Nora Hackett of our laboratory; the
predominant response she saw was a Type II alveolar pneumocyte proliferation.

P. E. Morrow;  Isn't the Clara cell alleged to be the progenitor cell in the
small airway epithelium?

J. A. Pickrell;  Yes.

R. F. Henderson;  I don't believe Dr. Hackett distinguished the Clara cell,
but merely indicated nonciliated airway epithelium.

Question;  Did Dr. Hackett differentiate what you're calling "Type II" from
the Clara cell?

J. A. Pickrell;  She differentiated Type II cells from nonciliated and
ciliated airway lining cells.

Question;  What is your understanding of the ambient concentrations of 03 and
N(>2?  You mentioned 1 and 5 ppm as "approaching ambient concentrations."

J. A. Pickrell;  In the Los Angeles Basin, 63 might approach 0.3 ppm to 0.5
ppm; NC>2 seldom exceeds 1 ppm and is usually ~0.5 ppm.  The Los Angeles Basin,
I believe, displays one of the higher concentrations in the country.  Does
that answer your question?

Comment;  Yes.  My comment would be that 5 ppm is not anywhere near ambient
concentration.
(
J. A. Pickrell;  That's correct.  1 ppm would be.
                                     240

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES       	Pickrell


Comment;  I don't believe that there are any monitoring data anywhere close to
1.5 for NO2—

J. A. Pickrell;  I believe in some of the NO2 reports I've seen it has
approached 1, but you're right:   it's more nearly 0.5.

D. E. Gardner;  I'm  sure you're  familiar with Dr. Gus Freeman's work with NO2
in which rats and monkeys were exposed for a lifetime.  Freeman very
methodically described many of the effects that you have reported, not only
with NO2 alone but also with NO2 plus 03  (related to what you're proposing).
Is the purpose of your studies to confirm his?  What have you added to
Freeman' s studies?

J. A. Pickrell;  I don't think he used spikes of NO2 to simulate urban
conditions.

D. E. Gardner;  I'm  referring to your NO2-O3 study.

J. A. Pickrell;  Freeman's NO2-(>3 study, if  my recollection is correct, used
somewhat higher levels than ours.

D. S. Gardner;  The  lowest level was 2 ppm.

J. A. Pickrell;  Freeman's NO2  level was 2 ppm, but the NO2-03 study was done
at a somewhat higher level, if  I remember correctly.

Comment;   The 03  level was  0.6 ppm.

 S. V. Dawson;   In  view of your  use of high and  low levels, what do you
consider  to  be  high and  low levels for the peroxyacyl nitrate and nitric acid
exposures  that  you're planning?   Also, which type of peroxyacyl nitrate will
be used,  and how will you obtain it?

 J. A. Pickrell;   I haven't  decided what  type of peroxyacyl nitrate will be
used.

      When they're  detected at all, peroxyacyl nitrate levels are in the ppb
 range.   The  effects  levels  that have been reported so far are in the ppm
 range.   Therefore,  we propose to start where effects levels have been reported
 and work down  toward ambient levels.   As our high  level, we will use a
 concentration  just below the  concentration at which effects have been
 reported.   For  our low level,  we will  use a level  an order of magnitude
 lower.

 D. E. Gardner;   Have you begun to generate and  monitor  peroxyacyl nitrate?

 J. A. Pickrell:   No.
                                      241

-------
24.  RESPIRATORY TOXICOLOGY OF NITROGEN OXIDES	Pickrell
D. E. Gardner;  Our recent experience suggests that you may encounter real
problems.  Peroxyacyl nitrate is not very toxic.  In most model systems it
presents real problems.  You might talk with the Illinois Institute of
Technology Research Institute [Life Science Division, 10 West 35th St.,
Chicago, IL  60616]; they did the work for us, and talking with them might
give you a head start in this area.

J. A. Pickrell;  Thank you.
                                     242

-------
        25.  ACUTE-CHRONIC LOSS OF LUNG FUNCTION FOLLOWING INHALATION
                               OF ACID SULFATES

                              Steven A. Silbaugh

                   Inhalation Toxicology Research Institute
           Lovelace Biomedical and Environmental Research Institute
                             Post Office Box 5890
                            Albuquerque, NM  87115
PROBLEM



     The existence, nature, and importance of health effects resulting from

the inhalation of acid sulfate aerosols (alone and in combination with other

fossil fuel combustion byproducts) are being addressed.  This project provides

information relevant to the setting of sulfate and fine particulate

standards.



APPROACH



Toxicity of Sulfuric Acid Aerosols in the Guinea Pig



     We conducted dose-response studies of the mortality of Hartley guinea

pigs exposed for 8 h to sulfuric acid (H2SO4) particles of either 0.4 or 0.8

Um in diameter.  A total of 96 animals were divided into groups and exposed to

43, 83, or 109 mg/m3 of the 0.4-um aerosol or 21, 32, or 43 mg/m3 of the


                                     243

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES	Silbauqh






0.8-ym aerosol.  Marked differences in lethality of the two aerosols were




observed.  The LCso for the °«8-Vm aerosol was ~30 mg/m3, while an LC50 for




the 0.4-ym aerosol was not reached even at 109 mg/m3.









Effects of Sulfuric Acid Aerosols on Pulmonary Function in the Guinea Pig









     Using Hartley guinea pigs, we completed studies of the acute respiratory




function effects of inhaled 1.0-pm 112804 particles.  A total of 57 animals




were exposed for 1 h to concentrations of 0, 1.2, 1.3, 14.6, 24.3, or 48.3




mg/m3; respiratory patterns and breathing mechanics were measured during




exposure.  The observed effects differed from the graded dose-response




relationship previously reported by Amdur et al. (1978).  Functional responses




reflecting airway constriction were either absent (nonresponsive animals) or




overwhelming (responsive animals).  The proportion of responsive to




nonresponsive animals increased with exposure concentration, but the magnitude




of pulmonary function change was similar for responsive animals regardless of




concentration.  Our results suggest that the Hartley guinea pig reacts to




inhaled H2SO4 with an essentially all-or-none airway constrictive response,




and that there is considerable variation in the concentrations at which




different animals develop the response.









Animal Strain Differences and Adaptation to Acute Sulfuric Acid Exposure









     Twelve guinea pigs of each of the following strains were exposed to 30-36




mg/m3 of 1.0-ym ^804 aerosol:  Charles River Hartley, Camm Hartley, Camm






                                     244

-------
25.  LUNG FUNCTION LOSS FOLLOWING  INHALATION OF ACID  SULFATES	Silbaugh


English Shorthair, and NIH  Strain  13.   Half of  the  exposures involved a

gradual increase  to  the above concentration over  a  1-h period,  followed by a

1-h exposure at a steady  concentration.  The remaining exposures were

conducted for  1 h with an immediate change from clean air  to the desired

concentration.  The  degree  of labored breathing during exposure was  scored

visually-   The NIH Strain 13 guinea pigs developed  more  severely labored

breathing than the other  strains,  but these animals were later  discovered to

have  respiratory  infections.  All  strains had less  labored breathing when

H2SO4 concentrations were increased gradually rather  than  suddenly,  suggesting

partial adaptation.
 Sulfuric Acid and Nitrogen Dioxide Induced Alterations
 in the Guinea Pig* 8 Sensitivity to Histamine Aerosol
      Hartley guinea pigs exposed for 1 h to an ^804 aerosol of 1  urn in mass

 median aerometric diameter (MMAD) or to nitrogen dioxide (NO2>  gas were

 examined for alterations in airway responsiveness to inhaled histamine.

 Concentrations of H2SO4 ranged from 4 to 40 mg/m3; concentrations  of NO2

 ranged from 7 to 146 ppm.  One group of animals exposed to filtered room  air

 served as the control.  Animals were exposed to stepwise increased

 concentrations of ~0.6-wm histamine aerosol 2 h prior to pollutant exposure

 and at 0, 2, and 19 h after exposure.  The histamine concentration required to

 produce a 50% decrease from pre-challenge dynamic lung compliance was used as

 a measure of airway sensitivity.  All animals exposed to NO2 exhibited an

 increase in histamine sensitivity immediately after exposure; the magnitude of
                                      245

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES	Silbaugh


this increase was concentration dependent.  We observed a dramatic return of

the sensitivities of most animals toward base-line values at 2 h after

exposure; however, the sensitivities of several animals remained elevated at

19 h after exposure.  Animals exposed to 1*2804 exhibited major increases in

histamine sensitivity only if labored breathing developed during exposure.  We

concluded that both NC>2 and 112804 alter airway sensitivity in the guinea pig,

but by apparently different mechanisms.



PLANNED RESEARCH
Alterations in the Guinea Pig* s Sensitivity to Histamine
Following Acute Nitrogen Dioxide Exposure at 5 ppm
     The above results demonstrated that major alterations in the guinea pig's

airway responsiveness to inhaled histamine could be produced by 1-h exposures

to 40 ppm NO2»  Airway-sensitizing effects were noted at lower concentrations,

but more information is needed at concentrations that are nearer ambient

levels.  This study will examine the acute alterations in the guinea pig's

histamine responsiveness induced by 1-h exposures to 5 ppm N02«  Two groups of

animals will be exposed:  one to room air (n = 10) and one to 5 ppm NO2 (n =

10).  One control and one NO2 animal will be exposed on each exposure day.

Histamine responsiveness will be measured before and immediately after

exposure, using methods similar to those described in the previous section.
                                     246

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES	Silbaugh






Pollutant Interaction Studies;   Nitrogen Dioxide and Sulfuric Acid









     Several  epidemiologic studies have suggested a positive correlation




between  air pollutant levels and frequency of asthmatic attacks.   Although




H2SO4 produces an asthma tic-like response in the guinea pig, it does so only




at concentrations that are far above ambient concentrations; the  lowest




concentration at which we have observed this asthmatic-like response in




healthy guinea pigs is 15 mg/m3.  If H2SC>4 does act to promote airway




c.onstrictive responses in human populations, other predisposing factors are




probably prerequisite.  If airway sensitivity to H2SO4 is, like histamine




 sensitivity, increased by oxidant gas exposures, then NC>2 might act as a




 predisposing factor.  In this study, we will determine if an NC>2  exposure




 known to produce altered histamine sensitivity in the guinea pig  also results




 in increased airway  sensitivity to H2SO4.  An initial study will  use high




 concentrations of NO2 and H2S04 to determine whether such an interaction




 exists.  For each of three  exposure days, one group of guinea pigs will be




 exposed for  1 h to 40 ppm NO2.   A second group of animals will be exposed for




 1 h to  filtered room air.   Immediately  after exposure, both control and




 N02-exposed  animals  will  be randomly assigned  to cages within an H2S04




 exposure chamber.  Animals  will be exposed  for 2 h  to an H2SO4 concentration




 of 20 mg/m3  and  1.0  ym  in MMAD.  We will use excised lung volume as an




 indicator of airway  constrictive response,  since our previous experience has




 indicated that the lungs  removed from  animals  which develop an airway




 constrictive response are consistently hyperinflated.   Animals that die during




 H2S04  exposure will  be  removed from the exposure chamber through a






                                       247

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES	Silbaugh


pass-through box; their lungs will be excised immediately and lung volume

measured by water displacement.  Animals that survive exposure will be

sacrificed immediately after exposure with an euthanasia solution; the lungs

will be excised and lung volume measured.  If this preliminary study indicates

a positive interaction, studies will be conducted at lower NO2 and H2S04

concentrations.  Combination as well as sequential exposures are of interest

and will be examined in subsequent studies.
Pulmonary Effects of Chronic Exposure to Sulfuric Acid, Fly Ash, and
Their Combination on Normal and Elastase-Treated Guinea Pigs
     This study will examine the response of normal and impaired

 (elastase-treated) guinea pigs to long-term exposure to 112804, fly ash, and

 their combination.  The study is designed to model "sensitive" segments of the

 human population:  of various laboratory animals, the guinea pig may possess

 airway responses most similar to those of the human asthmatic population;

 elastase treatment will impose an impairment with structural and functional

 similarities to human emphysema.  We plan to conduct lung function and

 biochemical tests plus morphological evaluations that have been carefully

 selected to detect subtle pulmonary effects.



     Five groups of 60 animals each will be exposed for 6 h/d, 5 d/week, for 3

yr.  Of each group of 60 animals, 30 will be treated with porcine pancreatic

elastase ~3 weeks prior to exposure.  The remaining 30 will be untreated.

Present plans are to expose the five groups of animals to (1) filtered room
                                     248

-------
25.  LUNG FUNCTION LOSS FOLLOWING  INHALATION OF ACID SULFATES	Silbaugh



air, (2) low-level H2S04  (in  the range of  0.5-1 mg/m3),  (3) high-level H2SO4

(in the range of  1-5 mg/m3),  (4) fly ash alone (one level), and  (5) high-level

H2S04 in combination with fly ash.  The  H2S04 aerosol will be ~0.4 urn in MMAD.

The fly ash will  be a well characterized sample obtained from a  fluidized bed

coal combustor.




     Ten normal and 10  elastase-treated  animals will be  randomly selected from

each exposure group.  The airway histamine sensitivity of each selected animal

will be measured  prior  to exposure and after  1, 3, and 6 mo and  1, 1.5, 2,

2.5, and 3 yr of  exposure.  Pulmonary function measurements, including

measurements of dynamic compliance and total pulmonary resistance, will be

obtained prior to each  airway sensitivity  measurement.   This group of animals

will be sacrificed at 3 yr after  the start of exposure.   Half of the remaining

normal and elastase-treated guinea pigs  in each exposure group will be

 sacrificed after  1 and  2 yr of exposure.  Other evaluations will be performed

on these animals  (serially or prior to  sacrifice), including measurements of

mucociliary clearance and lung function  measurements  sensitive to small airway

 dysfunction.  The nature and schedule  of these evaluations will  depend on the

outcome of preliminary  feasibility studies.  At sacrifice, one lung from each

animal will be processed for histopathologic  and  morphometric evaluation.  The

other lung will be washed for analysis of  cytoplasmic and lysosomal enzymes

present in the airways.  Cytological assays will  also be performed on airway
                           *
 fluid samples.  The  tissue of the washed lung will be analyzed for collagen


 and elastin content  and enzymatic activity.
                                      249

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SITUATES	Silbaugh
ACKNOWLEDGMENTS
     This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCE
Amdur, M. O., M. Dubriel, and D. A. Creasia.  1978.  Respiratory response of
     guinea pigs to low levels of sulfuric acid.  Environ. Res., 15:418-423.
WORKSHOP COMMENTARY
0. Raabe;  Could you give us more details about the chemical and physical
characteristics of the fly ash aerosols?

S. A. Silbaugh;  The aerosols will come from a fluidized bed combustor at the
Morgantown Energy Research facility-  If requested, I can provide the details
of trace element analysis.

O. Raabe;  What is the size distribution, and so on?

S. A. Silbaugh;  The size distribution is ~3 ym MMAD, I believe.  Maybe Dr.
Henderson can comment on that.

R. F. Henderson;  The sample we ran previously was 3 ym in MMAD, with a
geometric standard deviation of 1.7.  In contrast to your fly ash, it's a low
temperature, highly porous, and very polydispersed ash.

J. D. Hackney;  In the histamine responsiveness studies you used dynamic
compliance measurements, correct?                 •

S. A. Silbaugh;  Yes.

J. D. Hackney;  Did you also look at resistance?
                                     250

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID  SULFATES	Silbaugh


S. A. Silbaugh;  Yes, we looked at resistance, but we found  (along with the
group of Douglas, Dennis, and the late Dr. Bouhuys at Yale)  that the best
indicator of the airway constrictive response in the  histamine-exposed guinea
pig is a decrease in dynamic compliance.  Resistance  changes are more
variable.

     As soon as the compliance drops by 50%, the animal is removed from the
histamine; we don't want any residual effects caused  by the  spasm itself.  All
I can say is that the resistance at that time is variable, and we haven't
followed it beyond the time at which the compliance drop appears.

P. E. Morrow;  You stated that your H2S(>4 aerosol was 0.4 ym in aerodynamic
size.  How do you vary the concentration and keep that droplet size constant?
Have you done it over the range of 0.1 to 100 mg/m3?

S. A. Silbaugh;  We have a system whereby we can keep the particle size
constant.  Dr. Wolff has been more involved with that than I have.  Dr.
Carpenter (Inhalation Toxicology Research Institute)  has worked with a system
in which dry nitrogen is used to pick up sulfur trioxide, which is then mixed
with humid air.  Varying the dilutions of the sulfur  trioxide and the water
vapor as well as the amount of aging permits us to control the particle size
of the aerosol.

R. K. Wolff;  We have no problem in varying the concentration and keeping the
particle size constant.  We have generated 0.4-ym particles  at concentrations
ranging from 100 ug/m3 to 100 mg/m3.

Question;  What is the droplet concentration at 100 mg/m3?

R. K. Wolff;  The concentration is ~5 x 106 particles/cm2—the coagulation
limit.  I'll allude to this a little later.

Question;  On all of your function measurements in the guinea pig, were you
able to precharacterize your responses (i.e., determine preexposure factors
relating to individual animal sensitivity to H2SO4)?

S. A. Silbaugh;  We did some work with that.  In one  study,  we had two
exposure levels that resulted in something like 40/60% responders/
nonresponders.  We took those two exposure groups and looked for preexposure
characteristics that would tend to segregate animals. It does appear that the
responsive animals tend to have higher resistance values and lower compliance
values prior to exposure.  There also appear to be differences in the
relationships between, for instance, transpulmonary pressure and resistance.
So there do appear to be detectable differences that  are related to whether an
animal tends to be responsive or nonresponsive.

     In an article published a few years ago, Dr. Mary Amdur also reported
something like this.  Amdur also observed that animals with  higher resistance
values tended to be more reactive when exposed to various aerosols.


                                     251

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES _ Silbaugh
Comment;  We're doing K^SC^ aerosol studies in humans and we use the sulfur
trioxide system of aerosol generation.  We've found that [inaudible] and
humidity in our chamber determine the particle sizes; we're running 0.1 to
0.3.  The concentration we're using is 100 mg/m3.

     In your studies of ^804 alone, you used 10 to 40 mg/m3.  Have you
decided on the concentrations to be used in the fly ash/H2SO4 studies?

S. A. Silbaugh;  We have not yet decided on the H2SO4 concentrations, but
we're thinking along the lines of 0.5 to 1 mg/m3 for the low level and
somewhere between 1 and 5 mg/m3 for the higher level.  We can only pick one
level for the fly ash; we're thinking of 5 mg/m3.

Question;  Have you permitted the aerosols of acid and fly ash to age
together, to examine any possible interactions?

S. A. Silbaugh;  No, this would be a preliminary part of the study.  The main
study would not start until later in the fiscal year, and we have quite a bit
of developmental work yet to do.  That would be one part of the developmental
work.

Question;  Regarding the mechanics of the exposure facility, what provisions
would be required?  Have you any way to pool the two as a single atmosphere,
or to provide a means for the aging to occur?  Do you think there is any
advantage to it?

S. A. Silbaugh;  That is not the realm of my expertise.  Dr. Wolff, would you
care to comment?

R. K. Wolff;  It would depend on how much aging you wanted.  We could fairly
easily go to a few minutes of aging in a relatively large chamber.  To go
beyond that would involve considerable space and expense.  It could be done if
it were deemed worthwhile.

Question;  What plans have you for varying humidity?

R. K. Wolff;  Previously we varied humidity from 40% and 80%.  In the planned
studies we will probably use 40 to 50% relative humidity.  We have not seen
striking relative humidity effects.  That is another issue we can still
address .

M. J. Wi ester;  The data you showed did not indicate very much of an increase
in hi stamina response up to very high concentrations of N(>2 —

S. A. Silbaugh;  Are you referring to both N<>2 and ^804?

M. J. Wiester:  I'm not sure.
                                     252

-------
25.  LUNG FUNCTION LOSS FOLLOWING  INHALATION OF ACID  SOLFATSS	Silbaugh


S. A. Silbaugh;  The NO2  response  is  quite variable.  We're  still looking at
different ways of analyzing  this.   For  instance,  if you consider a particular
animal to be an outlier,  it  may be that a straight line relationship is not
the best relationship:  it may be  that  with more  animals at  higher
concentrations we'd see another kind  of response.  When NO2  concentrations are
expressed as logarithms,  the straight line fit does become slightly better.

M. J. Wiester;  What I'm  saying is that you don't increase your sensitivity of
the response at the ambient  level—

S. A. Silbaugh;  This  is  what I have  outlined  for a future study.  We need to
expose more animals at levels lower than 7 ppm to really get any information
regarding that region.  The  only area where we can say for sure that there is
a  response is the region  of  40 ppm or so.  With more  animals we should be able
to find out.

M. J. Wiester;   [inaudible]

S. A. Silbaugh;  We  found no correlation of H2SO4 concentration and histamine
sensitivity for the  nonresponsive  animals.  There was a trend at the 2-h point
for  histamine sensitivity to decrease with increasing H2SO4  concentration.
Dr.  Wolff may be able  to  shed some light on the  implications of these results
for  future studies.  The  indications  that 112804 may stimulate mucous
production raise the possibility of a mechanism by which increasing 1^804
concentration can actually desensitize  animals.

D. E. Gardner;   In  your NO2 studies,  did some  of  the  animals die?

 S. A. Silbaugh;  No.   We  obtained  our animals  from Charles River in six
 shipment groups  over the  course of the  study.  The shipment  groups reported
here all  responded  fairly uniformly.   One  shipment group (five animals) had no
 response at  concentrations of up to ~120 ppm.  We contacted  the breeder but
could not detect any preshipment differences.  Our guinea pigs may actually be
 less sensitive than rats  to NO2«

 M. Goldman;  The way the  data are plotted on a log  scale, aren't you really
 looking at a two-phase reaction?   Anything between 50%  and 500% is normal, and
 then you have  a  few there—

 S. A.  Silbaugh;   Right.

 M. Goldman;   Drawing a line with regard to concentrations, it  seems to me that
 you have a  uniform response with sensitive animals  [inaudible].

 S. A.  Silbaugh;  With NO2?

 M. Goldman;   Right.   In other words,  anything over  -20  gave  about  the  same
 response.
                                      253

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES	Silbaugh
[Simultaneous discussion]

M. Goldman;  If you try to scale it, you pick up the same data linearly and
you have what's here.

S. A. Silbaugh;  If these points are omitted, I think what you're saying is
true.

M. Goldman;  I was looking at all of the studies in which you're using this
end point.  It seems to me that you've just got one group of sensitive
animals.  The other groups are not sensitive.  There1s no concentration
effect.

T. Crocker;  I'm struck by the way in which you're using the relationship of
NC>2 to the histamine responsiveness.  Heretofore, the histamine response has
been used to assist in identifying very low level effects of NC>2; here you're
getting a histamine effect and then trying to poison it with a very high level
of NO2-  I don't quite understand the rationale for that approach, but it
doesn* t seem to me to improve our understanding of the responsiveness of the
respiratory tract at very limited concentrations.

S. A. Silbaugh;  We plan to do those exposures at lower levels.  Our histamine
challenge procedure is a recent development, the initial purpose of which was
to simply quantitate a dose-response relationship for NOj.  We wanted to check
our system, you might say.  These exposures have served several purposes.
First, we did find some information concerning the dose-response relationship.
We may be "poisoning it" (as you say) at higher levels, but we did want to see
responses in the initial exposures.  We planned to go down from there.

     We agree that the low concentration range is the important range to
study.  One thing we did obtain is an indication of a wide variation in
individual animal responses.  One is thus more encouraged to go back to lower
levels and do a large number of animals.  We may see no responses in quite a
few animals, but there may be animals that are sensitized at very low levels.
Starting out with the lower levels might have caused us to become discouraged
early, because we wanted to see an effect in this system.

     Also, we didn't know what to expect with the H2SO4 exposure, and we felt
that the NO2 exposures served as a positive control since others have
demonstrated a sensitization to airway constricting agents with exposure to
NO2-

T. Crocker;  Yes, but you didn't use NO2 in the range of 1 to 3 ppm, which is
near the limit of that gas' effectiveness in producing sensitization and where
histamine makes the measurement of that sensitization more relevant (or at
least more detectable, I should say).  The issue, really, is that if you go to
very high concentrations of NO2, you may be poisoning the very receptor
capability of the system that subsequently also responds to histamine.  I'm
concerned that we may be overusing the toxic power of NO2 when what we're


                                     254

-------
25.  LUNG FUNCTION LOSS FOLLOWING INHALATION OF ACID SULFATES	Silbaugh
really looking for is the ability to identify an increase in sensitivity.
[inaudible]  N<>2 in the concentrations you're dealing with would be expected
to wipe out a large part of the cell population, especially near the alveoli/
the [inaudible] of the alveoli junction.  N(>2 in these concentrations would
also affect cells elsewhere in the respiratory tract, and might very well
eliminate, in fact, the animals that respond.

S. A. Silbaugh;  I see your point.

D. E. Gardner:  I think Dr. Crocker's comments are very valid and pertinent,
especially in regard to the discussion yesterday [Chapter 6 of this volume] on
the relevance of high doses and how they can perhaps be misleading when you're
going to  look at low levels of concentration.
                                      255

-------
    26.  LONG CLEARANCE MECHANISMS FOLLOWING INHALATION OF ACID SOLFATES

                               Ronald K. Wolff

                   Inhalation Toxicology Research Institute
           Lovelace Biomedical and Environmental Research Institute
                             Post Office Box 5890
                            Albuquerque, NM  87115
PROBLEM



     Sulfur oxidation products result from the burning of fossil fuels and are

widely present in urban atmospheres.  Acid sulfates inhaled in the urban

atmosphere may have deleterious health effects, and definitive information is

needed to set air quality standards.  Sulfuric acid (H2SO4) mist may be the

most irritating of these products, at least with respect to pulmonary function

effects (Amdur 1971; Amdur et al. 1975); thus, the study of potential health

effects is of considerable interest.



     The mucoclliary clearance system provides a sensitive indicator of lung

injury following exposure to irritant gases and aerosols.  Alterations in

mucociliary clearance have been observed in man, donkeys, and dogs for H2SO4

mist exposures at industrial threshold limit values and below (Leikauf et al.

1979; Schlesinger et al. 1978; Wolff et al. 1978, 1979a, 1979b).  Impairments

in this important lung defense mechanism may contribute to the development of
                                     256

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES	Wolff






chronic lung disease or the exacerbation of existing disease.  Impairment of




mucous clearance might result in  increased residence times of toxic materials




in the lung, thereby increasing susceptibility to bacterial  infection.




Inhibition of mucous clearance could  also lead to the plugging of smaller




airways, followed by atelectasis  and  a  chain of events contributing to chronic




obstructive pulmonary disease.









     The project reported here explored dose-response relationships between




 impairment of tracheal mucous clearance and acid sulfate concentrations in




 acute and chronic exposures of dogs,  rats, and guinea pigs.  Acute studies




 were carried out in dogs  for reasons  of anatomic similarity  to humans.  Other,




 smaller animals were used in efforts  to identify a  chronic animal model that




 best mimics the human  response.









 APPROACH








     Methods have been developed to measure tracheal mucous  clearance in the




 awake  state  in  dogs,  rats, guinea pigs, and rabbits.  Microliter quantities of




 radiolabeled material  are instilled in the tracheas of  these animals using




 halothane anesthesia.   The animals are allowed to  regain  consciousness  in




 restrainers  and the movement of labeled material is measured using external




 detection (gamma camera or slit-collimated Nal scanner).
                                      257

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES                Wolff
Dogs
     Eight beagle dogs were exposed for 1 h to concentrations of 1.0 and 0.5




mg/m3 H2S04 mist.  The particles were 0.9 pm in mass median aerodynamic




diameter (MMAD) with a geometric standard deviation (Og) of 1.4 as measured by




cascade impaction.  Temperature was 74°F and relative humidity was 80%.




Tracheal mucous velocities were measured 1 week prior to exposure and also 0.5




h, Id, and 1 week after exposure.  The method was to anesthetize each dog




with halothane, insert a fiber-optic bronchoscope into the trachea, turn off




the anesthetic, and deposit a 10-yl droplet containing ~20 yCi of 99mTc




macroaggregated albumin (MAA) as the dog started to regain consciousness.




Subsequently, gamma camera scintiphotos were taken at 1-min intervals for 25




min in the awake dog to measure the velocity of the labeled material moving up




the trachea.  Measurements were made of discrete spot velocities and of the




velocity of the leading edge of moving activity.









     For the 1.0-mg/m3 exposures, tracheal mucous velocities were




significantly depressed after 0.5 h (26% reduction, p < 0.05), 1 d (40%, p <




0.01), and even after 1 week (30%, p < 0.05).  Velocities returned to the




control range after 5 weeks.  Also, a much greater proportion of material




remained at the initial deposition site 1 week after exposure than in the




control preexposure experiments (47% vs. 14%).









     The results for the 0.5-mg/m3 exposures were somewhat different.




Nonsignificant increases in clearance were seen after 0.5 h (35%) and  1  d






                                     258

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES	Wolff






(8%).  However, a statistically significant depression  in clearance was




observed 1 week after exposure (35%, p < 0.05).  These  observed abnormalities




indicate impairment of an important lung defense mechanism for a prolonged




period following acute exposures to relatively low levels of H2S04 mist.









     Subsequent exposures of the same 8 dogs to an H2S(>4 aerosol of smaller




size (0.3 ym MMAD) at levels of 1 mg/m3 and even 5 mg/m3 showed no significant




effects on mucous clearance.  Sham experiments were carried out under the same




exposure and measurement schedule described above.  Velocities were very




consistent; there were no significant differences at any time.









     Thus, despite the relatively small difference in the size of these two




aerosols, a wide disparity in effects was observed.  The same pattern was




observed in acute toxicity studies using guinea pigs:   0.8-0.9 ym MMAD aerosol




was estimated to be 5 times more toxic than 0.3-ym aerosol.  The studies




reported in this paper indicated a difference of at least a factor of 5, since




no effect was observed at 5 mg/m3 for the 0.3-ym aerosol compared to the




decided effects at 1.0 mg/m3 for the 0.9-ym aerosol.









     Our accumulated data indicate that upper airway reflex mediated




bronchoconstriction is the predominant mechanism of action of 112804 mist.




Therefore, if more material is deposited in upper airways, a greater effect




may be elicited.  Deposition studies carried out at the Inhalation Toxicology




Research Institute confirm that there is greater upper  airway deposition of




0.8-0.9 ym MMAD H2SO4 aerosols compared to 0.3-0.4 ym MMAD aerosols.






                                     259

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES	Wolff
Rats
     Groups of rats were exposed to 1, 50, and 100 mg/m3 H2S04 for 0.5 h and




to 1, 10, and 100 rag/m3 H2S04 for 6 h.









     Trachael clearance was measured in four males and four females in each




exposure group 1 week prior to exposure and 1 d, 1 week, and 3 weeks following




exposure.  (The 3-week measurement was omitted for the 0.5-h exposure.)  Rats




were anesthetized with 5% halothane to a level of deep anesthesia sufficient




to allow intratracheal instillation of 99mTc MAA.  The rats were placed in




plastic restraining tubes, allowed to regain consciousness, and driven past a




slit-collimated Nal detector to produce a profile scan of the labeled material




in the trachea.  Profile scans were taken atO, 2, 4, 6, 8, 10, 15, 20, 30,




40, 50, and 60 min after initial instillation.  The percentage of activity




remaining at the instillation site after 1 h was the principal measure of




clearance.  Tracheal mucous velocities were also determined by measuring the




movement of the leading edge of material.  The two types of measurements




correlated well but the retention measurements were more consistent.




Two-tailed paired t-tests were used to detect statistically significant




changes in clearance.









     Speeding in clearance was observed following the 0.5-h exposure.  No




changes were observed for the control group, but dose-related changes were




observed for the exposed groups.  Significant speeding of clearance was




observed after both 1 d and 1 week following the 100-jag/m3 exposure.   For  the






                                     260

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID  SOLFATES	Wolff






50-mg/m3 exposure, a  slight  depression of  clearance was  observed at  1 d, while




moderate speeding was observed  at  1 week.  For  the 10-mg/m3 exposure, a




speeding in clearance was  observed only at 1 week after  exposure.









     Following 6-h exposures, dose-related speeding in clearance was again




observed, with the most striking results at the highest  exposure level (100




mg/m3).  Clearance was  enhanced for the 100-mg/m3 exposure at  1 and 3 weeks




following exposure; for the  10-mg/m3  exposure at  1 d; and for  the  1-mg/m3




exposure at 3 weeks.  No significant  changes were seen in the  control group,




and clearance results were highly  reproducible.  Clearance effects were




consistent for the two  sexes except at 1 d after  the  100-mg/m3 exposure.  At




that time speeding was  observed in females but  not in males.








     Biochemical response in the lavage fluid and in  the lung  tissue was




minimal.  In lavage  fluid, the  only detectable  response  to any level of




exposure at  1 d  was an  elevation in sialic acid,  a marker for  acid mucous




glycoproteins.   As for  clearance,  this elevation  occurred at the 100-mg/m3




level  in females only.   Sialic  acid levels were always increased either




preceding or in  coincidence  with increases in mucous  clearance.  Scanning




electron micrographs  also indicated increased mucous  secretions in the




trachea.








      Speeding  in mucous clearance  has been observed  in dogs, man,  and donkeys




but only during  or  soon after low-level exposure  «1  mg/m3)  (Leikauf et  al.




1979;  Schlesinger et al. 1978;  Wolff  et al.  1978, 1979a, 1979b).   The speeding






                                      261

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION  OF ACID  SULFATES
  Wolff
in clearance that was observed in rats  after both  0.5- and 6-h exposures to


H2SO4 mist occurred at levels of up to  100 mg/m3—100 times the industrial


threshold limit value.  This speeding in  clearance was somewhat surprising,


and the reasons for it are not clear.   Figure  26-1 conceptualizes the observed


disparities between the rat and the other extensively studied species.


Speeding was observed in rats at all levels; in the other species, however,


clearance was generally increased at low  levels but depressed at higher levels


(>1 mg/m3).  These results are quoted for aerosols in a similar size range


(0.5-0.9 ym MMAD).
     200
  IU
  >
  CO
  D
  O
  U
                                                        RAT		
                                                     I
                                1                     10

                             H2 SO4 CONCENTRATION (mg/m3)
100
    Figure 26-1.   Mucous  velocity by species following exposure to 112804.
                                     262

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION  OF ACID  SULFATES	Wolff






PLANNED RESEARCH









     Results from  the  rat model  do not appear  to  agree well with results from




human or dog experiments.   Therefore,  the rat  may be  a poor model for chronic




exposure studies.  However, the  guinea pig has been demonstrated to show




pulmonary function responses to  ^804  similar  to  those of humans.




Accordingly, we have  initiated mucous  clearance studies with guinea pigs.




Recently obtained  base-line data show  mucous clearance patterns similar to




those in rats  and  rabbits.   In these studies,  tracheostomies were used to




deposit the  labeled markers. We are starting  to  investigate clearance in




guinea pigs  using  instillation via the intratracheal  route.  This has been




difficult because  of  the guinea  pig's  sensitivity to  laryngospasm.  However,




this problem has been solved and studies  are about to begin.  The advantage of




intratracheal  instillation is that animals can be used as their own controls




(as demonstrated in the previous dog and  rat studies).









Acute Exposures








     For guinea pigs,  we plan  1-h exposures to ^804  at levels of <1 mg/m3




with particle  sizes of 0.3-0.4 ym and  0.8-0.9  ym  MMAD.  Following completion




of these studies,  we  plan 1-h exposures to (NH4)2SO4  and NH4HS04 at levels of




<1 mg/m3 and particle sizes of  0.3-0.4 ym MMAD.   Also,  1-h exposures to 5




mg/m3 fly ash  and  5 mg/m3 fly ash + 1  mg/m3 H2SO4 will  be carried out.  These




studies will be a  prelude to chronic exposures.   For  dogs, we plan  1-h




exposures to (1*114)2804 and NH4HS04 at  <1  mg/m3.






                                      263

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID SULFATES	Wolff


     For all acute exposures, if no effects are initially noted at 1 mg/ra3, we

will conduct more extended or repeated exposures at this level.



Chronic Exposures



     We plan chronic exposures to H2SO4 and H2S(>4 + fly ash according to the

schedule outlined under project RPIS Number 4054, Acute-Chronic Loss of Lung

Function Following Inhalation of Acid Sulfates.  Mucociliary clearance

measurements performed on the same schedule will serve as pulmonary function

measurements in normal and elastase-treated guinea pigs.



ACKNOWLEDGMENTS
     This research was supported in part by the U.S. Environmental Protection
Agency via Interagency Agreement Number EPA-IAG-D5-E61 under U.S. Department
of Energy (DOE) Contract Number EY-76-C-04-1013 and in part by the U.S. DOE
under Contract Number EY-76-C-04-1013 and conducted in facilities fully
accredited by the American Association for Accreditation of Laboratory Animal
Care.
REFERENCES
Amdur, M. O.  1971.  Aerosols formed by oxidation of sulfur dioxide—Review of
     their toxicology.  Arch. Environ. Health, 23:459-468.

Amdur, M. 0.,  J. Bayles, V. Urgo, M. Dubriel, and D. W. Underbill.  1975.
     Respiratory response of guinea pigs to sulfuric acid and sulfate salts.
     Presentation at:  Symposium on Sulfur Pollution and Research Approaches.
     Durham, North Carolina, Duke University Medical Center, May 27-28.

Leikauf, 6., D. B. Yeates, K. Wales, and M. Lippmann.  1979.  Effect of
     inhaled sulfuric acid mist on tracheobronchial mucociliary clearance and
                                     264

-------
26.  LUNG CLEARANCE FOLLOWING INHALATION OF ACID  SULFATES                Wolff
     respiratory mechanics  in healthy nonsmokers  (abstract).  Am. Rev. Resp.
     Dis., 119:227.

Schlesinger, R. B., M.  Lippmann,  and R.  E.  Albert.   1978.  Effects of
     short-term exposures to sulfuric acid  and ammonium sulfate aerosols upon
     bronchial airway function  in the donkey.   J. Am.  Assoc.  Ind. Hyg.,
     39:275-286.

Wolff, R. K., M. Dolovich,  G. Obminski,  and M.  T. Newhouse.   1978.  Effect of
     TLV  levels of  SO2  and  H2SO4  on bronchial  clearance in exercising man.
     Arch. Environ. Health, 33:24-32.

Wolff, R. K., B. A. Muggenburg, and S. A. Silbaugh.   1979a.   Effect of
     sulfuric acid  mist on  tracheal mucous  clearance in awake beagle dogs
      (abstract).  Am. Rev.  Resp.  Dis., 119:243.

Wolff, R. K., S. A.  Silbaugh, R.  L. Carpenter,  and J.  L. Mauderly.  1979b.
     Toxicity of 0.4  and 0.8 ym sulfuric acid  aerosols in guinea pigs.  J.
     Toxicol. Environ.  Health,  5:1037-1047.
 WORKSHOP COMMENTARY
 Comment;   You examined the clearance of particles over a fairly short period.
 In the first phase of clearance, you did have some residual  particles; of
 course, these disappear radioactively-  Is anybody looking at the  long-term
 clearance of residual particles?

 R. K. Wolff;  Currently, a number of studies at the Inhalation Toxicology
 Research Institute are investigating long-term clearance of  particles.  We may
 do some longer-term clearance studies in connection with those outlined here.
                                      265

-------
                27.  OVERVIEW OF CURRENT AND PLANNED RESEARCH
                     BY THE INHALATION TOXICOLOGY BRANCH

                              Donald E. Gardner

                  Health Effects Research Laboratory, MD-82
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     This report presents an overview of current and projected research on

oxidants by the Inhalation Toxicology Branch (Environmental Toxicology

Division, Health Effects Research Laboratory, Office of Research and

Development, U.S. Environmental Protection Agency, Research Triangle Park,

North Carolina).  Elsewhere in this volume, Graham (Chapter 28) and Miller

(Chapter 29) discuss some of these studies in greater detail.



STAFF, FACILITIES, AND BUDGET



     The Inhalation Toxicology Branch (ITB) is staffed by ~22 permanent,

full-time employees and ~13 part-time employees.  The Branch is divided into

three sections:  Pharmacology and Microbiology, Physiology, and Inhalation

Exposure and Dosimetry.
                                     266

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner







     Each Section employs individuals with a wide .variety of scientific




backgrounds.  For example, in the Microbiology and Pharmacology Section there




are several microbiologists, pharmacologists, and a number of biologists;




there are also entomologists who are working in the area of unconventional or




biological pesticides  (i.e., using viruses, bacteria, and protozoans as




pesticides).  The Physiology Section includes individuals with expertise in




cardiovascular disease, physiology, biochemistry, and organic chemistry.  The




Inhalation Exposure  Section is composed of inhalation engineers, electronic




technicians, physical  scientists, and a biostatistician.








     In  addition to  these individuals who are directly employed by ITB, our




program  is supported by ~22 in-house contractor personnel from Northrop




Services, Inc. - Environmental Sciences.  The majority of these individuals




provide  direct support for the inhalation exposure and engineering needs of




the Branch.  They are  responsible for maintaining the exposure facilities,




which includes generating and monitoring a large number of chemicals of




interest to EPA.  Northrop also maintains individuals trained in the areas of




physiology, microbiology, and pharmacology who perform research in certain




areas of our base program.








     Comparing the allocation of funds for 1979 and 1980 indicates how our




research program can shift from one year to the next.  In fiscal year (FY)




1979, 41% of our total budget was devoted to the energy program, whereas in FY




1980 there was no money to either continue or start any studies in that area.




In FY 1980 there was a significant increase (5% to 21%) in criteria research






                                     267

-------
27.  INHALATION TOXICOLOGY BRANCH        	Gardner






fvoiding, and a large jump (19% to 51%) in non-criteria pollutant research.  We




are also involved in one transportation program whose objective is to compare




the carcinogenicity of diesel exhaust and various diesel extracts with roofing




tar, cigarette smoke condensate, and coke oven emissions.  This work is being




done under a grant and represents ~15% of our total budget.  There is also a




small program (5%) in pesticides and toxic substances.









     For FY 1980, ~35% of our total budget is allocated to in-house programs;




the balance (~65%) is funded extramurally under grant programs, cooperative




agreements, or contracts.








GOALS FOR CRITERIA POLLUTANT RESEARCH








     In order to make our research most useful for the setting of criteria, we




incorporate certain broad objectives in our research planning.  First, we're




interested in making animal data more relevant to man.  Others have stressed




that "a mouse is not a man" (e.g.:  Chapters 4, 5, and 7 of this volume).




What we want and need to do is to develop a scientifically sound data base




that can be used to develop modeling systems that will help us extrapolate




animal data to humans.  For example:  If a mouse receives a certain




concentration of a pollutant gas at its nose and subsequently shows a certain




health effect, what concentration would have to be at a man's nose to show a




similar effect, assuming the same kinds of target organs or target tissues




were affected?  Such information would greatly improve the usefulness of




animal toxicology data.






                                     268

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner






     Secondly, we are also interested in confirming and understanding many of




the important toxicological studies that presently comprise the data base for




establishing a scientifically sound standard for ozone (03) and nitrogen




dioxide  (NO2) and other  criteria pollutants.









     Our program is  interested  in  interactions.  It's time to get away from




studying single pollutants.  Other chapters in this volume demonstrate that




there  are  already many studies  examining the effects of such combinations as




03 with  NO2; sulfuric acid (H2SO4) with 03; and H2SO4 with NO2.  As time goes




on, we will expand these mixture studies to include other gaseous and




particulate pollutants (such as hydrocarbons, sulfur dioxide, etc.).








     It  is imperative that we continue to identify any new effects or end




points having potential  utility in human studies.  For example, during the




past few years, we have  established a very comprehensive program specifically




devoted  to the study and development of new models for measuring in small




animals  various sensitive pulmonary functional parameters similar to those




measured in humans.  Our program continues to seek noninvasive techniques that




will hopefully provide indicator systems useful to both the epidemiologists




and to the clinical  investigators.








     Lastly, we are  interested  in  developing new animal models that mimic the




special  human subpopulation that is most sensitive to environmental chemicals.




A number of new animal models of human diseases are now available for




application by toxicologists in environmental studies.






                                     269

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner







POLLUTANTS STUDIED









     To date we have investigated the effects of exposure to 03, NO2, and




combinations of 03 with NO2» with particulates, etc.  Some work has been




conducted under contract to determine the health effects of exposure to




peroxyacetyl nitrate.  We continue to study the effects of complex mixtures,




including the combination of 03, SC>2/ and trans-2-butene.  Our research




program also considers pollutants that are beyond the scope of this volume




(e.g., pesticides, toxic substances, non-criteria pollutants).









MODEL SYSTEMS IN USE









     Biological indicators of health effects include both interaction with




infectious microorganisms (e.g., bacteria, virus, Mycoplasma) as well as




specific alterations in a number of specific host defenses.  Included here are




functional and morphological alterations in alveolar macrophages, various




humoral and cell-mediated immunities (including B and T cell transformation),




and antibody titers.









     Other studies investigate the effects of various environmental chemicals




on upper respiratory clearance mechanisms.  In this model system we examine




changes in ciliary function by measuring the beating rate and frequency, and




we attempt to relate these changes to histological abnormalities of the cilia.
                                     270

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner






     J. O'Neil, M. J. wiester, and J. A. Raub of our Branch have developed a




battery of pulmonary functional tests that should be very useful in measuring




various changes in respiratory rate and other functional parameters in small




animals.









     With 03 and N(>2, current studies model the pulmonary deposition and the




fate and absorption of  these gases in the lung.  Although we are primarily




interested  in  the lung  as the target site, we investigate other organ and




target tissues for any  extrapulmonary effects that might result from inhaling




these  environmental chemicals.









     Animal models are  employed to measure cardiovascular effects of 03, NO2/




and other compounds of  interest to EPA.  The end points include identification




of lipid profiles, EKG  measurements, and determination of other physiological




parameters.








     Another model focuses on pentobarbital-induced sleeping time.  Early




results indicate a significant sex difference:  females are much more




susceptible to the actions of NC>2 and 03 than males.  Graham (Chapter 28 of




this volume) expands on these studies.








     When indicated, we complement all our studies with examination of various




lung and serum biochemical end points and/or performance of necessary




hiatopathology.
                                      271

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner






     In addition to our in-house research efforts, we are collaborating with




investigators at the University of North Carolina in a study of how the




microorganism Mycoplasma pneumonias interacts with normal systems and of how




airborne pollutants may alter the microbial action upon the upper airway.




This model system is discussed further in the following section.









MODEL SYSTEMS UNDER DEVELOPMENT








     Several model systems are presently under development.  In our acute




infectivity system, bacteria are given after exposure to a gaseous pollutant.




In a normal animal or an animal that is not affected by the pollutant, the




bacteria are all rapidly removed or killed within 6-8 h; the animal's lungs




become normal again (i.e., free of the inhaled bacteria).  This very sensitive




model is excellent for mimicing acute infectious diseases but fails to




resemble a chronic infectious disease.









     A chronic respiratory model is being developed by E. Hu of our Branch.




This model uses the bacterium Mycoplasma pneumoniae.  In these studies, the




animals exhibit a low-grade infection which mimics chronic bronchitis.  We




plan to soon examine the effects of various pollutants using this chronic




respiratory disease model.









     A viral model system similar to our acute bacterial infectivity system




but employing viruses instead of bacteria is being developed by M. J. Selgrade




of our Branch.  This project investigates the effects of environmental






                                     272

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner






chemicals on the reactivation of latent viruses and the interaction of viruses




and bacteria.









     There is also interest in better defining the susceptible human




subpopulation.  One  study at Duke University will attempt to determine the




effects of pollutants on the young developing lung.  This study will resemble




the work by Dungworth (Chapter 20 of this volume) but will involve exposures




to NO2 and to 03 for longer periods of time in hope of correlating the




morphological effects with various pulmonary functional changes.  Under this




cooperative agreement study, we plan to perform detailed morphometrics as well




as identification of various biochemical changes in the lung.









     There is also a project whose objective is to examine individuals with




glucose-6-phosphate  dehydrogenase (G6PD) deficiency and to determine how they




respond to 03 assault.  The investigators have bred G6PD-deficient mice and




sheep and are attempting to find out if they are more susceptible to oxidants




than normal animals.









     An"animal model of emphysema is being developed.  Animals which are




exposed to elastase  develop emphysema-like lesions; our program is attempting




to correlate and quantitate the amount of emphysema using pulmonary functional




techniques.  Elastase-exposed animals will be exposed to various pollutants




and their responses  compared to proper controls.  This model mimics an




individual in the population who has emphysema and is exposed to a pollutant.
                                     273

-------
27.  INHALATION TOXICOLOGY BRANCH    	Gardner






To complement this study, we plan to breed and expose a strain of mouse (the




blotchy mouse) that is genetically prone to develop emphysema.









     Finally, there is interest in how diet deficiency may alter




susceptibility to 03 and to N(>2«  The biological end points include various




barrier functions, the production of edema, the breaking of tight junctions in




the upper airway, and similar biochemical and morphological indicators of




damage.









STUDIES TO BE REPLICATED









     There are some important studies that we plan to replicate and/or extend.




One of these is a study by Bartlett et al. (1974) which examined the effects




of oxidant pollutant (0.2 ppm) on the young developing lung.  These authors




concluded that the developing lung is more susceptible to 03 than the adult




lung.









     Using a very sensitive model, Sherwin (Chapter 36 of this volume)




detected edema in the lung at a very low level (0.4 ppm) of NO2«  We plan to




extend and replicate this study.









     Some years ago, 03 was found to accelerate aging (Stokinger 1957).  There




is a need to repeat this study to better define those results.  Unfortunately,




the funds are not available for FY 1980.
                                     274

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner






     Finally, it was recently reported that exposure to 0.2 ppm NC>2 causes a




decrease in metabolism of prostaglandin E  (Menzel et al.  1976; Menzel 1980).




These results also demand replication to better define the mechanism of action




and possible consequences.









     All of these animal studies represent important components of the




existing data base (and therefore of the criteria document).  For that reason,




they have  high priority for  funding.









COLLABORATION WITH OTHER INVESTIGATORS








     In addition to our in-house efforts, we participate in a considerable




amount of  collaborative work with other investigators in the Research Triangle




area.  Within EPA's Health Effects Research Laboratory, we have joined the




Experimental Biology Division (N. Chernoff) in a search for possible




teratogenic effects of NO2*  A similar collaboration (with L. Reiter)  seeks to




demonstrate behavioral effects of exposure to 03 and N02-  With S. Sandhu




(Genetic Toxicology Division) we are searching for mutagenic effects of N©2




exposure.








     There is also collaboration with the University of North Carolina, Duke




University, and North Carolina State University.  Topics of these studies




include immunologic effects  of 03 and N02» effects of oxidants on upper




respiratory tissue (scanning electron microscopy), and pharmacokinetic effects




of 03 on some xenobiotic agents.






                                     275

-------
27.  INHALATION TOXICOLOGY BRANCH	Gardner
REFERENCES
Bartlett, D., Jr., C. S. Faulkner, and K. Cook.  1974.  Effect of chronic
     ozone exposure on lung elasticity in young rats.  J. Appl. Physiol.,
     37:92-96.

Menzel, D. B.  1980.  Pharmacological mechanisms in the toxicity of nitrogen
     dioxide and its relation to obstructive disease.  In:  Nitrogen Oxides
     and Their Effects on Health (Lee, S. D., ed.).  Ann Arbor Science, Ann
     Arbor, Michigan.

Menzel, D. B., W. G. Anderson, and M. B. Abou-Donia.  1976.  Ozone exposure
     modifies prostaglandin biosynthesis in perfused lungs.  Res. Comm. Chera.
     Pathol. Pharmacol., 15:135-147.

Stokinger, H. E.  1957.  Evaluation of the acute hazards of 03 and oxides of
     nitrogen.  Am. Med. Assoc. Arch. Ind. Health,  15:181.
                                     276

-------
    28.   SOME SPECIFIC STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH

                               Judith A. Graham

                  Health Effects Research Laboratory, MD-82
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     This report discusses in more detail some of the projects overviewed by

Gardner (Chapter 27 of this volume).  All are projects of the Inhalation

Toxicology Branch (Environmental Toxicology Division, Health Effects Research

Laboratory, Office of Research and Development, U.S. Environmental Protection

Agency, Research Triangle Park, North Carolina).  It is not the intent of this

report to present data, but rather to indicate the focus and scope of these

studies.



CHRONIC NITROGEN DIOXIDE EXPOSURE USING MOUSE INFECTIVITY MODEL



     One major planned study is a chronic nitrogen dioxide (NO2> exposure.  We

recently built an in-house exposure facility suitable for NO2 (or any gas).

Currently, we are conducting some shorter-term studies to ensure that the

system will work when we actualy begin the chronic exposure.
                                     277

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham






     The driving force behind this study is some previous work with the mouse




infectivity model.  In this model system, a mouse is exposed to a pollutant




and then challenged with an aerosol of live Streptococcus pyogenes.  Mortality




is followed for a holding period in clean air.  This model system is one of




the most sensitive model systems available for ozone (03) and NC>2 animal




inhalation toxicology.








     Previous NC>2 studies examined concentrations no smaller than 0.5 ppm.  At




0.5 ppm, 90 d of exposure were required to observe effects.  We are, of




course, interested in investigating concentrations that are closer to ambient




levels.  Therefore, it will be necessary to extend the time of exposure.




Prior work also showed that, at least for NO2 and the infectivity model,




accurate interpretation of results requires that the chamber exposure patterns




match ambient exposure patterns.  Therefore, we will perform more research




using a base-line NC>2 concentration upon which peak N(>2 concentrations are




superimposed.  In the planned study, we will expose the animals to 0.2 ppm




continuously (except for short periods of cleaning the chamber).  Then the




animals will receive a peak exposure of 0.8 ppm, 5 d/week, 1 h/d in the




morning and 1 h/d in the afternoon.









     A number of parameters will be examined.  Besides the Streptococcal




infectivity model, several mouse pulmonary function tests are now available




and will be applied.
                                     278

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham






PENTOBARBITAL-INDUCED SLEEPING TIME  IN THE MOUSE









     Although over the years  it has  been assumed that gases affect primarily




the lung, there is increasing interest in the extrapulmonary effects of gases




and other pollutants.  There  have been recent indications that extrapulmonary




effects occur upon exposure to oxidizing pollutants.  One of the model systems




used in our laboratory is pentobarbital-induced sleeping time in the mouse.




In this particular model system, the animal is exposed to a pollutant,




pentobarbital is  injected, and sleeping time is measured.









     In our early studies, concentrations as low as 0.1 ppm 03 and 0.25 ppm




NO2 caused significant increases in  pentobarbital-induced sleeping time.




Further evaluation of this model led to some interesting results.  For




example, there appeared to be a difference between the toxicity of NO2 and O3.




As the concentration of 03 was decreased (e.g., from 1 ppm to 0.1 ppm), an




increasing number of days exposure (at 3 h/d) was required to observe




increased pentobarbital-induced sleeping time.  A single 3-h exposure to 1 ppm




03 produced the effect, whereas at 0.1 ppm it was necessary to repeat the




exposure (3 h/d)  for 16 d to  observe about the same level of effect.  However,




with NO2 the effect could be  seen after the first day of exposure:  a 3-h




exposure to 5 ppm NO2 or to 0.5 ppm  N02 demonstrated the effect.  Thus, in




this model system (at least for acute exposures) N02 appears to be more toxic




than 03.  Of course, this conclusion differs from the vast body of




toxicological literature on the two  gases.  Our results suggest a different
                                      279

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham






mechanism of action between 03 and NC>2, and this is something that we want to




follow up.








     To make sure that our results were not unique to pentobarbital, further




examinations employed a variety of drugs (hexobarbital, thiopental, and




zoxazolamine).  We found that our results were indeed not unique to




pentobarbital.









     We were also interested to determine the biological significance of these




effects.  With regard to criteria documents, the concentration is important




but the effect is of prime importance.  Knowledge of the significance of that




effect (e.g.:  Are human beings at risk if this effect is observed in animals,




or even in humans?) is a very, very important question.  This model system is




particularly useful because effects are observed at lower concentrations.









     Disappearance curves for pentobarbital in the blood were performed to




determine (1) if there were effects on either phase 1 or phase 2 clearance,




and (2) if the effects were primarily on hepatic metabolism.  Generally




speaking, changes in pentobarbital-induced sleeping time were attributed to




quantitative changes in xenobiotic metabolism.









     Once we complete the kinetic studies described above, we plan to employ




another model system has already been developed.  This model system,




antipyrine kinetics, is based on the fact that antipyrine is completely




metabolized by the cytochrome P-450 system.  In humans, antipyrine is an






                                     280

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham






indicator of P-450 metabolism  by the  liver.  We plan to test this model in




animals and perhaps use the results to provide some guidance to workers




engaged in human studies.  Perhaps similar studies can eventually be performed




in humans and then related to  the whole  array of effects that can be observed




using the more  invasive techniques with  animals.  Besides looking at some of




these whole-animal models, we  want to examine in more detail some of the




events that may be occurring in the liver with the cytochrome P-450 system.




Thus, we will also run a  rather typical  evaluation of enzyme activities




 (O-demethylase, N-demethylase, and hydroxylase).








      Sex sensitivity studies were part of our initial evaluation of this model




 system.   (Given that EPA  is charged with protecting the most sensitive segment




 of the population, sex sensitivity is an appropriate focus of research.)  We




 began with female mice and the effect was observed.  Then came the question:




 Is there a sex sensitivity, or does this effect only occur in the mouse?  (An




 effect that occurs only in the mouse  might not be extrapolatable to man.)  We




 found that the effect occurs  in the rat  and  hamster as well as the mouse; in




 all these  species, the female was more  susceptible.  There are known sex




 differences in P-450 metabolism in the rat,  but these major differences do not




 apply to other species.   Thus, the  sex difference we observed represents more




 than classical differences in P-450 metabolism.
                                      281

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham







SENSITIVITY TO BRONCHOCONSTRICTING AGENTS









     Our program includes & large battery of studies within the category of




lung metabolism and pharmacological regulation of the lung.  The major




stimulus behind this body of research is the increase in airway resistance




observed in humans after exposure to O3 and NO2.  This particular effect in




humans is one of the major reasons behind regulation of these pollutants, and




we want to better understand its mechanism of action.









     Many current human clinical studies involve pharmacological perturbation




with carbachol, acetylcholine, or other parasympathomimetic agents to see if




the pollutant increases sensitivity to these bronchoconstrictors.  Questions




have been raised as to the biological relevance of increased sensitivity to




bronchoconstrictors and as to how EPA should treat this information from a




regulatory standpoint.  Our group is conducting some mechanistic studies that




are designed to yield an improved understanding of the toxicological events,




thereby giving a better understanding of the potential seriousness of these




events in man.









     We are conducting a number of studies with 03 alone and with NO2 alone.




After we define the effects with these gases, we plan to expose the animals to




the same gases in combination.  Early studies by Menzel (see Gardner, Chapter




27 of this volume) focused on HO2 effects on the metabolism of prostaglandins,




particularly PGE2«  Concentrations of NO2 as low as 0.2 ppm (3 h) inhibited




metabolism of the bronchoconstrictor; the effect was maximal at 18 h post







                                     282

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham






exposure (i.e., there was a delayed effect).  The study used an isolated




ventilated perfused lung (IVPL), so the effect observed was probably on the




epithelial cells of the capillaries in the IVPL preparation.  We are




interested in expanding this study because the prostaglandins affect not only




airway tone but also vessel tone, and changes in these parameters might




influence not only airway resistance but also the relationship between




ventilation and perfusion in the lung.  Also, there are a number of other




possible effects.  Hopefully, expanding this study will lead to a better




understanding of the mechanism involved in airway resistance changes.









     Other compounds (e.g., histamine and SRS-A) are also very involved in




bronchoconstriction, and we plan to measure those compounds as well.  All are




related to arachidonic acid metabolism and products from the associated




reactions.








     Also, the lung is the main site of the converting enzyme activity that




takes angiotensin 1 to angiotensin 2 (the most potent vasoconstrictor in the




body).  We have accomplished some preliminary studies indicating possible




effects on these systems.  We plan to extend these studies and see if changes




in the reflex system will be reflected in changes in cardiovascular function.









     Planned studies that have not yet begun include an examination of cyclic




AMP and cyclic GNP; these are very important regulators of various aspects of




cell function.  Also planned are studies of antioxidant metabolism.  Most of




this work will be done by M. Mustafa (University of California - Los Angeles).






                                     283

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham






Many of Mustafa's studies will examine combinations of O3 and N02,  something




in which we are particularly interested in light of his previous reports of




effects at concentrations as low as 0.1 ppm (and also at low concentrations of




NO2).  We need to know more about combinations of effects and recovery from




those effects*









STUDIES TO IDENTIFY SUSCEPTIBLE POPULATIONS








     With regard to what segments of the human population are more  susceptible




and might need more protection, there are some obvious lines to be  followed:




male vs. female, very young vs. young vs. older, and so on.   These  are very




large classifications.  In spite of many years of toxicological studies with




03 and NO2, it still remains unclear as to whether the young are more or less




susceptible.  Our group wants to improve the state of this knowledge; so far,




the only solid plans are for 03.  However, there is a need for such work with




NO2, also.









     The planned study will use rats.  Animals will be exposed to 0.25 ppm,




0.12 ppm, and 0.08 ppm O3.  A set of animals will be exposed from 0 to 6 weeks




of age (when alveolarization occurs).  Also, a group of young adult animals




will be exposed.  A control group of young adults will be exposed to 03 as




well as air.









     Morphology and extensive morphometric procedures will be the prime




parameters used to evaluate age susceptibility.  Pulmonary function will be






                                     284

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham


inserted into the system as soon as the animals are large enough to be

measured.  It should be possible to do a reasonable battery of pulmonary

function studies when the animals reach ~20 g.



WORKSHOP COMMENTARY
R. K. Wolff;  Did you mention that you were going to use 03, NO2» and some
inhaled particulate?

J. A. Graham;  As part of our Branch's inhaled particle program, we're
interested in doing studies on the combined effects of particles plus gases.
We have not yet decided precisely what particles; the decision will be based
upon data from ongoing studies.

J. L. Whittenberger;  The theme of this meeting is scientific research; a
secondary theme is relationship to standard-setting.  Gardner [Chapter 27 of
this volume] gave as one of the major objectives of your Branch the
confirmation and extension of important findings.  I agree that that's a very
important objective.  It was mentioned that you'd like to confirm and extend
the Bartlett et al. (1974) study  [see Gardner, Chapter 27 of this volume, for
details of reference].  The importance of that study was recognized in 1977,
if not earlier.  Would you be willing to give a mini case history of why the
Bartlett study has not yet been confirmed and extended?

D. E. Gardner;  My guess is that it has mainly been a matter of appropriate
funds and priority.  In reviewing the documents, it becomes evident that this
is an important study when one considers the whole package of animal
toxicology programs.  Now that we have some very sophisticated morphometric
techniques, we think we can not only reproduce the study but also extend it
and make it better.

J. L. Whittenberger;  EPA didn't have to do that study with its own funds;
there's NIEHS "just down the road."  Why didn't they repeat it?

J. A. Graham;  Essentially it comes down to funds and also to the study's
relationship to criteria documents.  When a criteria document is evaluated by
a large number of health scientists, the gaps really become immediate.  That's
why it's a good exercise to read the criteria documents even if you're not
involved in the regulatory process:  you see where the research is needed.

J. L. Whittenberger;  When you say it's a matter of "funds," are you telling
me that somebody in Washington screwed up?
                                      285

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham
D. E. Gardner;  We hope to correct this.

J. A. Graham;  Dr. Gardner [Chapter 27 of this volume] discussed the
relatively low level of funding for oxidants research in fiscal year 1979 as
compared to fiscal year 1980.  There is a major difference.

D. E. Gardner;  If we had the money to fund the projects of every researcher
attending this workshop, all those projects would be very interesting.  But
there is a limited budget, and with this limited budget we have to answer to
"the guy that's paying the bill."  That's why we can only fund projects that
the customer says are most useful in the regulatory setting.

J. L. Whittenberger;  So you're saying that the Program Office doesn't
consider this particular study to be very important?

J. A. Graham;  No, indeed they do consider it to be important.  There was
simply a financial limitation last year.  I'm sure everybody in this room
could propose 10 excellent studies to add on to those outlined here by Dr.
Gardner and me.  But we simply can't do it all at once.  You're going to be
asking the same question about a different study four or five years from now.

D. E. Gardner;  Part of the objective of this workshop plus the "straw man"
document [Miller et al. 1979; see Chapter 2 of this volume for details of
reference] is to bring some of these elements together.  For example, Dr.
Dungworth's work [Chapter 20 of this volume]  will fit in very nicely.  Maybe
bringing all of these together will help solve some of our problems.

J. L. Whittenberger;  Remember that 88% of environmental health research is
not funded by EPA.

Question;  Can you provide a better explanation of how the decisions were
made?  In answer to Dr. Whittenberger, you pointed out that one pressure on
you is the necessity to respond to your client.  What client is interested in
the mechanistic work you were talking about;  what client considers that top
priority?

J. A. Graham;  The Oxidant Research Committee of the Office of Research and
Development.

D. E. Gardner;  As described by Jones [Chapter 4 of this volume], all of our
work goes through research committees made up of clinicians, epidemiologists,
chemists, OAQPS staff, lawyers from the Office of General Counsel, and so on.
Every time we have a package that we want to "sell," we present it to these
committees.  The committees then prioritize it.  To the best of our ability,
we present what we think is appropriate.

Question;  [inaudible]
                                     286

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH                Graham
J. A. Graham;  Are you asking if a study done in rats can be extrapolated to
humans?  I think Miller [Chapter 29 of this volume] addresses the quantitative
aspects of that question.  As a generalization, we feel that if an effect can
be demonstrated in a number of animal species then it is quite possible that a
similar type of effect occurs in humans.  We don't know yet at what
concentration that effect would occur in humans.  But if young rats are more
sensitive than middle-aged rats, it's quite possible that a similar effect
occurs in humans.  The results of such a study would provide guidance for
future epidemiologic studies or even human clinical studies (if they were
ethically possible).

Comment;  How can one use cyclical exposures (especially with a base-line
exposure underneath that) as information in the standard-setting process?

J. A. Graham;  EPA has two major information needs with respect to NO2? these
relate, respectively, to a short-term standard and a long-term standard.
Should each exist and, if so, at what concentration?  The prime drive behind
this study is to answer questions relating to the short-term standard.

     At present, EPA has an annual standard for NO2 which permits the
occurrence of cyclical peaks.  The question is:  Are cyclical peaks safe for
the public health, or do these commonly occurring peaks add an additional
burden for which control measures are needed?  Our planned study will expose
animals to these patterns.  One control will be the base-line concentration of
NO2 itself.  Then we will have base line plus peak.  Let's say that we add an
incremental 20-30% effect to the system.  That might also apply to humans.
The ratio we have chosen for this particular chronic study (0.2:0.8) is the
typical urban ratio that occurs in the ambient environment.

D. E. Gardner;  Those experimental concentrations may be slightly higher than
concentrations in the natural environment, but the ratio is consistent.

Question;  When you go about setting a standard, then, you need some of the
information to help you set long-term standards and some for short-term
standards?

D. S. Gardner;  Yes.

Comment;  In several of the studies [described in this volume], the rat
appears to be an outlier, both with respect to other rodents and with respect
to other species.  Why are you beginning another rat study?

J. A. Graham;  The age susceptibility study will be done in rats because the
Bartlett et al. (1974) study was done in rats.  We want to replicate and
extend that work.  Also, a lot of base-line morphometric information is
available for rats, and it will probably be possible to share some controls
with other studies.
                                      287

-------
28.  STUDIES PLANNED BY THE INHALATION TOXICOLOGY BRANCH	Graham
     I think the whole question of species sensitivity to 03 and N©2 (or to
any pollutant, for that matter) is a very difficult issue.  At this time, in
terms of pulmonary effects, I don't know which species is more or less
sensitive.  At this workshop we heard an excellent presentation [Chapter 11 of
this volume] on some of the differences that exist, let's say, in circulating
anti-oxidant enzymes.

     I think the rat has been used by Mustafa and Freeman.  In terms of the
lung, Mustafa found effects at 0.1 ppm 03 for rats maintained on normal
Vitamin E levels.  For rats with excess Vitamin E levels, the effect was not
seen until 0.2 ppm.  So I think the rat is sensitive,  at least in terms of
lung effect, and lung effects are primarily what we will examine in this
study.

     Once again it's a question of limited funds.  I agree that it would be
ideal to follow up these effects in several animal species.
                                    288

-------
              29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY

                                Fred J- Miller

                  Health Effects Research Laboratory, MD-82
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     There is an acknowledged need to better assess target organ exposure

levels and the comparability of the animal toxicological data base for

standard-setting purposes.  The current standards for oxidant gases are based

mainly upon human studies.  All of the animal toxicological data (which are

95% of the data available) come in "through the back door":  "These effects

are observed in humans and, by the way, some effects are also observed in

animals."  In ethical consideration of what cannot be done experimentally with

humans, and in view of the need to understand mechanisms, researchers must

provide answers that will allow EPA to make better use of the animal

toxicological data base.



     The problem is partially one of dosimetry:  How do we correlate an effect

observed in an animal toxicological study with the exposure concentration

required to deliver the same amount of pollutant to the target site in man?
                                     289

-------
29.  BIOMATHBMATICAL MODELING OF OXIDANT TOXICITY	Miller






We must improve our determinations of the quantities of pollutants that reach




given levels in the respiratory tract and that are associated with given




effects.








     This report describes some preliminary attempts to biomathematically




model the deposition of ozone (03).  Biomathematical modeling of deposition is




an expanding field, and one of increasing interest to EPA.  At present, the




technique is not utilized to the greatest possible extent.  Fuller utilization




will require more and better biochemical and physical data.  Thus, another




goal of this report is to outline some areas in need of further research.




Many of the needed studies are not specific to 03, but would apply to modeling




the deposition of any gas.








NASAL PHARYNGEAL REMOVAL









     It is exceedingly difficult if not impossible to measure the specific




amount of pollutant that reaches a given level in the lung.  It is certainly




possible, however, to experimentally measure nasal pharyngeal removal.  Nasal




pharyngeal removal of 03 in humans represents a major piece of missing data




that would enable us to put into perspective the animal studies and the




effects observed in human exposures.  This information would permit various




manipulations to relate the likelihood of exposure relationships to




probability of effect.
                                     290

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY	Miller






     In one particular study (Miller et al. 1979) we looked at nasal




pharyngeal removal in rabbits.  This was also done in guinea pigs.  Both




species showed 50% removal of the pollutant over a concentration range of 0-2




ppm.  These results for 03 satisfy such other models as that of Aharonson et




al. (1974) for penetration of soluble vapors into the nasal mucosa.  In guinea




pigs, if the concentration exceeded 2 ppm, there was disproportionate removal




by the nasal pharyngeal cavity (an increase of 15%) and the similarity of




deposition was no longer maintained.









GAS TRANSPORT









     The major components for analyzing gas transport are (1) convection,




which can be separated into advection and eddy dispersion, and (2) diffusion




in both the axial and radial directions, which is a function of the flows in




various areas in the lung.  To accurately model the amount of pollutant




removed at each level of the lung also requires the incorporation of any




chemical reactions in addition to morphometric data.  Fortunately, the




morphometric data are starting to become available.









MODELING PROCESS








     These major components are brought together in a differential equation




that represents gas transport in the lung.  The change in concentration per




unit time is a function of convection, axial and radial diffusion, and the




source terms.  For a given pollutant, then, one needs to know these values in






                                     291

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY	_	Miller



order to definitively solve the equation.  As discussed below, we have


completed preliminary modeling of O3.  However,  much work remains to be done


to improve the data base.




MODELING ABSOLUTE DOSAGE:  SOME PRELIMINARY RESULTS




     In our initial 03 modeling, we used a reaction scheme that looks at the


attack of 03 on free fatty acid carbon-carbon double bonds.  Also incorporated


were Mudd et al. (1969) data showing the reactions of 03 with the free amino


acids.  Available data on lipids obtained from human tracheobronchial


secretions were also used.




     After this initial work, we found data on the free amino acid
                                                •*

concentrations in tracheobronchial secretions (Kohler et al. 1969).  The


results discussed below were obtained in updated modeling which incorporated


these new data.  This illustrates that we can always refine our estimates of


dose by incorporating additional biochemical data.




     With respect to dose for each airway generation in the human lung, the


model predicts that the seventeenth generation—the first generation of


respiratory bronchioles—will receive the maximum dose of 03.  For low


concentrations, the model predicts no penetration of 03 through the mucous


layer to conducting airway tissue; as the concentration increases, penetration


into these other generations occurs.  However, for a given concentration  it  is


still the respiratory bronchioles that are hardest hit.  Proceeding distally,



                                     292

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY	Miller






the predicted dose decreases.  In terms of the available histopathological




data for primates and rats (Dungworth et al. 1975; Stephens et al. 1973),




these results correlate well for main site of injury.









     We also obtained modeling results for rabbits and guinea pigs.  The




available morphometric data for these species are not exactly comparable to




humans, but there are similarities in terms of tracheobronchial bronchioles,




alveolar ducts, sacs, etc.  The junction between the last bronchioles and the




alveolar ducts is again hardest hit in terms of 03 dose.  Also, the conducting




airway dose again falls off as one proceeds distally.  The shapes of the




curves are quite similar; there are differences in the absolute magnitude of




the  predicted dose.









      In guinea pigs, higher concentrations are required to predict dosages to




the  conducting airway tissue.  Once again, the terminal bronchiole, alveolar




duct, and respiratory bronchial areas are predicted to receive the maximum




dose.









     We plotted maximal dose as a function of inhaled tracheal concentration




for  man, rabbits, and guinea pigs.  In our first plots, these curves started




to become linear above  100 yg/m^.  When we incorporated the new data on




tracheobronchial free amino acids that are available to react with and deplete




O3 before it can strike conducting airway tissue,  it became necessary to reach




~200 ug/m3 to observe a linear relationship between the respiratory




bronchiolar dose and the  inhaled tracheal concentration.






                                      293

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY
     By accounting for nasopharyngeal removal, this type of information




enables us to put into perspective the exposure concentrations and effects




observed in animal studies.  It becomes apparent, for example, that a lower




tracheal concentration is required to deliver a given dose in man as compared




to the rabbit or guinea pig.  Thus, our modeling data are useful in planning




experiments to fill gaps in the existing data base.  (The 03 criteria document




probably reflects the most extensive toxicological data base for any




pollutant, yet many gaps remain.  Very few studies have looked at a




commonality of end points in several species in terms of dose-response curves,




etc.)









     Many human data have been obtained in studies in which subjects exercised




while being exposed to 03 or nitrogen dioxide (NO2) (for example:  Bates et




al.  1972; Hackney et al. 1975).  Evaluating these studies with our model shows




that, as tidal volume increases, the model predicts a tremendous increase in




uptake by respiratory airway tissue.  Due to the transit times involved, a




relatively constant amount is removed by conducting airway tissue and a




decreasing amount is depleted by the mucous.  Assuming 750 and 200 yg/m-* as




the maximum and minimum concentrations, respectively, we can see the vast




importance of exercise level on the resulting dose.  In many of the human




studies, exercise levels that approximately doubled the tidal volume (~1000




ml) were used.  Assuming a concentration range of from 200 to 750 yg/m-*, the




model prediction ranges from removal of 45% of the 03 to delivery of ~80% to




respiratory bronchioles ( compared to the base line) .  Such a prediction
                                     294

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY	Miller






correlates with increased pulmonary function effects that are indicative of




decreasing airways.









     Because EPA is charged with protecting individuals at any level of




activity and under any exposure conditions, it would be very helpful to know




the exposure level required to yield 400 yg/m3 at the human trachea.




Unfortunately, that information is not currently available.  At any rate, if




we compare tidal volumes of 500 ml and 1750 ml while the subjects are resting




at an 800-yg/m3 trachea level and exercising at a 400-yg/m3 level,




respectively, the model predicts the same maximal dose to the respiratory




bronchioles.  There are millions of adults jogging and children playing; what




are the effects and what are the safe exposure levels in view of such activity




levels?  There is certainly justification, in the Los Angeles Air Basin, for




air pollution alerts and for restrictions on children's play during those




episodes.  Mathematical modeling of dosage permits us to make meaningful




comparisons of experimental results and to better calculate safe exposure




levels, thereby helping to determine whether such restrictions are justified




for other locations and environmental conditions.









QUANTITATIVE BIOCHEMICAL DATA ON THE MUCOUS LAYER:  A MAJOR RESEARCH NEED









     Current modeling efforts focus on the sensitivity of dose to mucous




production.  We are in the process of establishing bounds for
                                      295

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY		Miller



     The biochemical data required to model a gas greatly influence the amount


of useful information that can be obtained.  Fortunately, information is


available for O3; O3 is so reactive that one can consider instantaneous


irreversible reactions with many biological substances.   Unfortunately, this


is not the case for NO2.  Reactions with NO2 are much slower.   With sulfur


dioxide (SO2) there are different reactions to consider,  again requiring a lot


of information.




     To accurately predict how much gas will reach the respiratory bronchial


and alveolar regions requires quantitative biochemical data on what is


available in the mucous to react with the gas.  Such data would enable us to


put species differences into perspective in terms of absolute  dosage.  Most of


the available animal data, however, are qualitative in nature.  For example,


much of the work on mucous secretions has involved staining techniques:  we


know that something is present bat not in what quantity.   Some quantitative
                                       •*

human data are available (Lewis 1971; Kohler et al. 1969).




     The thickness of the mucous layer is one parameter that can impact upon


the probability of finding effects.  Luchtel (1976) completed a study in


rabbits.  There is controversy about the thickness of the mucous layer in


rats.  We also need better data on regional production and transport of


mucous.




     Effective axial diffusivity relates to the effective diffusion down the


airway.  Past modeling has incorporated only molecular diffusion; the correct



                                     296

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY	Miller


term is in fact ~2000 times greater  in  the  first few  generations  of  the

conducting airways.  Some work on a  human model has been  completed by  Scherer

et al. (1975).  For animals, the needed data  are (once again)  lacking.   Also

lacking are the diffusion coefficients  in mucous and  surfactant.   Thus far  in

our modeling, we have been forced to substitute the value for  water.



REFERENCES
Aharonson, E. F.,  H.  Menkes,  G.  Gurtner,  D.  L.  Swift,  and D.  F. Proctor.
      1974.   Effect of respiratory airflow rate  on removal of  soluble vapors by
      the nose.   J.  Appl.  Physiol.,  37:654-657.

Bates,  D. V., G. M. Bell, C.  D.  Burnham,  M.  Hazucha, J.  Mantha, L. D.
      Pengelly,  and F. Silverman.  1972.   Short-term effects of ozone on the
      lung.   J.  Appl.  Physiol.,  32:176-181.

Dungworth, D. L.,  W.  L.  Castleman,  C.  K.  Chow,  P- W. Mellick, M.  G. Mustafa,
      B. Tarkington, and  W. S. Tyler.   1975.   Effects of  ambient levels of
      ozone on monkeys.  Fed.  Proc., 34:1970-1974.

Hackney, J.  D., W.  S. Linn, D.  C. Law, S. K. Karuza, H.  Greenberg, R. D.
      Buckley, and  E.  E.  Pedersen.  1975.   Experimental studies on human health
      effects of air pollutants:   III.   Two-hour exposure to ozone alone and in
      combination with other pollutant gases. Arch. Environ.  Health,
      30:385-390.

Kohler, H.,  P.  Klossek,  H. Aurich, and H. Strobel.  1969. Der gehalt des
      menschlichen  bronchialskretes an freien aminosauren. In:  Folia
      Bronchologica in Zeitschrift fur Erkrankungen der Atmungsorgane, pp.
      259-265.

Lewis,  R. W.  1971.   Lipid composition of human bronchial mucus.  Lipids,
      6:859-861.

Luchtel, D.  L.   1976. Ultrastructural observations on the mucous layer  in
      pulmonary  airways (abstract).  J. Cell. Biol., 70:350a.

Miller, F. J.,  C.  A.  McNeal,  J.  M. Kirtz, D. E. Gardner, D.  L. Coffin, and D.
      B. Menzel.  1979.  Nasopharyngeal removal  of ozone  in rabbits and guinea
      pigs.   Toxicol., 14:273-281.
                                      297

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY		Miller


Mudd, J. B., R. Leavitt, A. Ongun, and T. T. McManus.  1969.  Reaction of
     ozone with amino acids and proteins.  Atmos. Environ., 3:669-682.

Scherer, P. W., L. H. Shendalman, N. M. Green, and A. Bouhuys.  1975.
     Measurement of axial diffusivities in a model of the bronchial airways.
     J. Appl. Physiol., 38:719-723.

Stephens, R. J., M. F. Sloan, M. J. Evans, and G. Freeman.  1973.   Early
     response of the lung to low levels of ozone.  Am. J. Pathol., 74:31-58.
WORKSHOP COMMENTARY
Question;  Do you know the competence of your predictions in terms of error
limits?  Are the estimates really so precise that you can tell the difference
between humans and animals?

F. J. Miller;  We plot our results on a log scale.  Thus, the increments
represent two- and three-fold differences.

Question;  You obtained these results from how many studies, for example, on
humans?

F. J. Miller;  Reaction components are readily available from two papers:
Lewis (1971) and Kohler et al. (1969).  Again, this is a function of the
sensitivity analysis:  If the input data are off by a factor of 25, we're in
trouble; if they're off by a factor of 10, we're okay.  As better data become
available, reiteration of the analysis builds confidence in the model's
predictions.

Question;  Do you intend, in your future work, to give some indication of
confidence in the lines you produce?  Do you have plots for that?

F. J. Miller;  Solving the differential equation does not yield the confidence
limits.  One can change the variables in the model:  frequency, tidal volume,
breath-holding, and so on.  One can look at all those things.  I have just
given you a smattering of them.

Question;  Do you have an indication, all along the way, of confidence in what
you put in?

F. J. Miller;  Well, the differential equation applies to gas transport.
That's not a unique equation but, rather, the equation that would apply to any
coordinate system.  We make simplifying assumptions for analyzing the
tracheobronchial airways and the rest of the lung.  There's nothing "magical"
about the first equation.
                                     298

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY                       Miller
Comment;  Some of the problem here is in going from small animals to large
animals.  Recognizing that effects in humans are ultimately what we're after,
are there (or would you like to see) some large animal studies that would make
some of these measurements experimentally and allow you to confirm your models
as you go along?

F. J. Miller;  We need that kind of information.  The limiting factor is the
morphometric data.  I understand that some data on monkeys are starting to
become available.  The usefulness of this modeling approach is limited by the
availability of morphometric data; beyond that, of course, there is no limit
in terms of its application.

J. L. Mauderly;  I have three comments.  First, with respect to your last
statement, I think there are some fairly detailed morphometric data available
that range from human down through rodents.

     Secondly, I am very glad to see the growing recognition that it1 s not the
air concentration but the inhaled material that causes the effect.  We
physiologists have been saying this for years.  The message is finally coming
through that it's not ventilation but alveolar ventilation that's important,
and that's a simple relationship that we've been working with for decades.

     Finally, Graham  [Chapter 28 of this volume] alluded to some studies by
age and said that you would be mentioning more about this...

J. A. Graham;  The main questions are whether we should use rats or another
species, and whether data on rat age susceptibility can be related to humans.

J. L. Mauderly;  Yes, that's what I was looking for.  My comment is that we
are learning more and more with respect to lung structure and function with
age in these different experimental animals, but right now we know almost
nothing about the senescent lung in most of these species.  We do know quite a
bit about the dog in terms of morphology, morphometry, and pulmonary function.
Very recently, quite a lot of data have been produced on rodents (rats and
hamsters; particularly rats).  What we see is not very encouraging.  We know
almost nothing about primates.

     Now, certainly we know quite a bit about primate morphology and
morphometry and pulmonary function.  A group at Davis has done a lot of work
in this area, but almost nothing on the aged primate and on the time course of
senescence.  This is a "glass house" situation because we're doing and
planning a lot of studies in rats.

     I've said on a number of occasions that the reason for this is that the
life span of the average beagle dog is longer than the political life span of
the average politician.  This is probably about the only reason we're working
on rats.  Statisticians encourage us to use rats, too.
                                      299

-------
29.  BIOMATHEMATICAL MODELING OF OXIDANT TOXICITY	      Miller


     We know at the present time that the pattern of senescence in lung
structure and function in the dog is very similar to that in man, even though
the lung type is different.  We know just from preliminary data that the
pattern is markedly different in rodents (certainly in the rat):   there is a
growing lung throughout the life span, and this must impact on carcinogenesis
and the development of chronic lung disease.  And we're skirting around this
issue and ignoring it and proceeding as though rats are men—which we know
they're not.

     We also know, from some recent data, that lung function may peak at
mid-age in the rat (~18 mo) .  Even one group at 27 mo—very close to the
expected life span for the Fischer rat—had lung function that was still very
good with respect to young adult normals.  In contrast, human lung function
peaks in young adults and begins a slow decline that is progressive and almost
linear throughout the adult life span.  That's not the case in rodents; in
fact, we don't even know yet if there is a terminal decline in rodent lung
structure and function.

     This is a comment rather than a question.  Actually, it1s a sermon.  And
I think it's something we've got to be very much aware of.

F. J. Miller;  I wouldn't deny the need to use other animal species.

J. L. Mauderly;  The point is that we don't have this information about the
dog (and perhaps the primate) .  We must continue to fund at least some studies
in animals known to have similar patterns of change with age.

C. A. Heckman;  Aging studies at Oak Ridge National Laboratory are examining,
in a preliminary way, changes in the aging of the rat lung.  So far, the
computer data indicate that there is probably more cellular septum and greater
septum thickness in the rat lung.  This may be because the physiological and
functional measures are not quite as sensitive in these smaller animals.
There may be some more morphological changes in larger animals that can be
picked up by functional measurements.

     Also, we've found some differences in the way the epithelium from the
trachea grows jln vitro in aging rats as compared to younger rats.  Let me
emphasize that this is all preliminary work.
                                     300

-------
 30.   OVERVIEW OF CURRENT AND PLANNED RESEARCH BY THE HUMAN STUDIES DIVISION

                                Robert E. Lee

                  Health Effects Research Laboratory, MD-51
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     The Human Studies Division (Health Effects Research Laboratory, Office of

Research and Development, U.S. Environmental Protection Agency, Research

Triangle Park, North Carolina) is a new entity.  A reorganization late in 1978

combined the then existing Population Studies Division, which focused on

epidemiologic studies, with the Clinical Studies Division, which was centered

mainly in our Human Research Facility in Chapel Hill, North Carolina.  These

two organizations were combined to create the new Human Studies Division.



     The main intent of the reorganization was to enable better coordination

and information exchange between our epidemiologic and clinical programs.

Both programs are now under a single Director.  Hopefully, the net result of

the reorganization will be better interaction between these two very different

research fields.
                                     301

-------
30.  HUMAN STUDIES DIVISION
EPIDEMIOLOGIC PROGRAM









     Previous EPA-supported oxidant studies have focused mainly on the Los




Angeles area.  Studies have involved the measurement of irritation symptoms in




various cohorts, including athletes and schoolchildren.









     For the past several years we have funded a study with Copley




International to quantify the acute health responses measured by respiratory




disease from various patterns of exposure to nitrogen dioxide, ozone, and




suspended sulfates and nitrates.  The investigators have followed daily




reporting of acute respiratory disease symptoms for a period of 26 weeks in




300 families in each of 4 communities.  This study should be completed in June




1980.









     For the past three years we have funded an effort at Loma Linda




University (Loma Linda, California) to replicate a study of cytogenetic




effects of photochemical oxidants in college freshmen in the Los Angeles




Basin.  This study was designed:  (1) to determine if there is an increased




risk of chromosomal aberration in peripheral lymphocytes among young adults in




an area of high ambient photochemical air pollution, (2) to determine if such




aberrations persist upon emigration from the polluted area, and (3) to




determine if new immigrants develop the chromosomal aberrations.  We expect a




report on this study in February 1980.
                                     302

-------
30.  HUMAN STUDIES DIVISION	      Lee


     Chapman (Chapter 34 of this volume) describes an ongoing air pollution

study in the Texas Gulf Coast area that is designed to assess the effects of

air pollution on local asthmatics1 symptoms and lung function, and on

performance symptoms and physiology in local residents who exercise vigorously

and repeatedly in a standardized way.



     Dawson (Chapter 39 of this volume) describes an integrated set of human

and animal studies planned by the Air Resources Board of the State of

California.



CLINICAL PROGRAM



     Our clinical research activities with photochemical oxidants are

described by Haak, Hazucha, and Ginsberg (Chapters 31, 32, and 33 of this

volume).  These studies are conducted in our $8-million Human Exposure

Facility located in Chapel Hill, North Carolina.



WORKSHOP COMMENTARY
 Question;  We  learned earlier  [Chapter 28  of this volume] that certain animal
 studies will be replicated by  EPA.  What key human clinical studies mentioned
 in the 03 and  W>2  criteria documents will  be replicated?
 R.  S. Chapman;   I would guess  that  the Orehek et al.  (1976) study  [see Chapter
 33  of this volume for details  of  reference] will be replicated.

 R.  E. Lee;   I think  future studies  will  depend  in  some measure on  the
 discussions  at  this  workshop.   That's one of the reasons we asked  you to come.
                                      303

-------
                  31.  HUMAN PULMONARY ADAPTATION TO OZONE

                             Edward D. Haak,  Jr.

                  Health Effects Research Laboratory, MD-58
                      Office of Research and  Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC   27711
INTRODUCTION



     Animal and human studies have established that short-term exposure to low

levels of ozone (03) produces detrimental effects involving multiple organ

systems and cell types.  In most cases, the health impact of these changes

(particularly the long-term consequences) is very unclear.  We do not know

which of the observed detrimental effects has the greatest impact on health,

nor for that matter do we know which of the effects is the most sensitive

indicator of detrimental exposure.  We do know that the detrimental effects

include decreased immuno-integrity of circulating lymphocytes, changes in

biochemical parameters, decreased pulmonary mucociliary clearance, and

decrements in standard pulmonary function tests.



     This report describes a series of studies—the OZADAPT series—that was

designed to examine the effects of sequential 03 exposures on young healthy

nonsmoking men.  In all, OZADAPT involved five studies and a total of 90 young
                                     304

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE	Haak






healthy nonsmoking male subjects.  These subjects were examined for pulmonary




function measures as well as biochemical and immunologic parameters.









     A full presentation of the complex OZADAPT series cannot be given here.




This report presents some of the pulmonary function results and compares these




findings across the five separate studies, with particular attention to the




question of "adaptation" in terms of pulmonary function.









PROTOCOL









     Identical screening procedures were used to select the subjects for all




five studies.  Table 31-1 indicates the mean values and variation of height,




weight, and age for each group of subjects.  Comparisons of demographic data




revealed no significant differences among the groups.  Any subject with a




history of allergy or  asthma was rejected, but no skin tests were performed.








     As shown in  Table 31-2, each OZADAPT study consisted of five consecutive




study  days (Days  1 through 5 = Monday  through Friday).  Day 1 was always a




control day; exposures (4 h/d) were performed on Days 2 through 5.  OZADAPT I




and IV were control studies in which subjects received only air; OZADAPT II,




III, and V involved exposures  to 0.4 ppm 03.  Pulmonary function data were




obtained at three times each day:  GI  (base  line), C%  (2 h), and 04  (4 h).




The pulmonary function data reported in this paper are for forced expiratory




volume
                                      305

-------
 31.   HUMAN  PULMONARY ADAPTATION TO  OZONE
                                                                           Haak
TABLE 31-1. HEIGHT, WEIGHT,

Parameter OZADAPT
height (cm) I
II
III
IV
V
weight (kg) I
II
III
IV
V
age (yr) I
II
III
IV
V
AND AGE OF

Mean
181.2
183.6
181.4
180.3
180.6
76.49
76.55
77.55
71.61
72.79
25.24
24.83
25.89
25.03
24.16
SUBJECTS
Standard
Deviation
6.30
6.04
5.84
8.99
7.08
7.05
9.72
11.41
9.64
8.88
2.58
2.34
1.96
2.83
2.63
 RESULTS AND DISCUSSION
     We observed significant differences in mean pulmonary function for




OZADAPT II  (light exercise) versus OZADAPT III (heavy exercise).  The mean




FEV-) data from OZADAPT II indicated a much lower decrement at the light  (35




liters/min) exercise level.  The mean response diminished to an insignificant




level by-Day 5, the fourth consecutive 03 exposure day.  In contrast, the mean




FEV-i data from OZADAPT III showed a much larger decrement at the heavy




exercise (57 liters/min) level; subjects did not return to base-line values by




24 h after Day 2, the first 03 exposure day.  The response to the second




exposure (Day 3) was similar in magnitude of mean decrease.  By the fourth day






                                     306

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE
                                                                          Haak
                      TABLE 31-2.  EXPERIMENTAL PROTOCOL
OZADAPT
I




II




III




IV




V




Day
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Exposure3
Air
Air
Air
Air
Air
Air
°3
03
03
03
Air
°3
03
03
°3
Air
Air
Air
Air
Air
Air
03
03
03
°3
Treadmill
Exercise'3
light
light
light
light
light
light
light
light
light
light
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
heavy
none
none
heavy
heavy
 a4  h/d.   03  concentration =0.4 ppm.

 bLight treadmill  exercise =  4  mi/h, 0%  grade,  15 min, 35  liters/min
  ventilation.   Heavy treadmill exercise =  4 mi/h,  10% grade,  15 min, 57
  liters/min  ventilation.
                                      307

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE
of consecutive 03 exposure (Day 5), however, the mean decrement had diminished




to little more than the variability in the control values.








     Therefore, on the basis of mean decrement in FEVi across groups, one may




conclude that consecutive exposures are associated with a progressive




diminution in observed decrement.  However, a consideration of individual




subject values across the week points up the great importance of individual




sensitivities, even among "normal" vigorous young men.








     Figure 31-1 displays three individual patterns that were observed.  (All




subjects received identical treatment, exercise, etc.)  Subject #3 was barely




responsive across all exposure days.  Subject #7, on the other hand, was very




responsive on the first and second exposure days, but his reactivity was




greatly diminished by the third and fourth exposure days.  Subject #8 was




equally reactive on the first and second exposure days; more importantly, his




responsiveness persisted.  This clinical observation of wide variance among




"normal" young men prompted us to undertake the additional study (OZADAPT V),




which was designed to further characterize the responsiveness of these




subjects.









     In OZADAPT V, mean FEVi responses were minimal for Days 2 and 3 (rest).




There was a large response on Days 4 and 5 (heavy exercise).  Decrements on




Days 4 and 5 were very similar to those observed on Days 2 and 3 in OZADAPT




III.  Unfortunately, we could not continue the OZADAPT V exposures for a fifth




and sixth day to look for "adaptation."






                                     308

-------
         120.00r
                                                                                                                             U)
U>
o
VD
         110.00*
          100.00-
       >"  90.00 •
           80.00.
           70.00
           60.00
Day 1


 Air
                        -- Subject #3
                        ••• Subject #7
                        — Subject #8
                                               Day 2
            Day 3
     Day 4


      03
               8      8
           8
           CO
o
q
CO
8      8
                                                in     co      r«»     oo      <

                                                  EXPOSURE  DAY CONDITION
8
o
Day 5


  03
                                                                                               CN     CO
                                                                                                            o
         o
         q
         in
                                                               I

                                                               I
                                                                                                                             i-3

                                                                                                                             8
                                                                                                                             H
                                                                                                                             i
                                                                                                                             SI
                      Figure 31-1.   Response variability among three  subjects in OZADAPT  III.

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE		Haak






     The results of OZADAPT V were not unexpected, but they are extremely




important.  Whether an individual responds depends in part on his inherent




sensitivity and also on his exercise challenge or provocation level.









     We decided to define "significant responsiveness" (in the clinical sense)




as a decrease of >20% in FEV^ or maximal midexpiratory flow rate (MMEF), even




though we could statistically define a meaningful decrement at a much lower




level.  We designated the 20% level in order to permit a comparative




examination of all five studies.  Our goal was to identify on an individual




basis (not a mean basis) subjects exhibiting evidence of significant effect




and the persistence of that effect across all four exposure days.









     As shown in Table 31-3, none of the total 45 subjects in the light




exercise control study (OZADAPT I) and the heavy exercise control study




(OZADAPT IV) showed a 20% decrease in FEV-j or MMEF under any measurement




condition on any study Day-  In view of the strictness of our criteria, this




observation is not surprising.









     Table 31-4 indicates the number of responders (as defined by the same




criteria) for each of the four consecutive exposure days in each of the three




03 exposure studies (OZADAPT II, III, and V).  Remarkably, 4 of 15 subjects in




OZADAPT II (light exercise) met these strict criteria on the first day of




exposure.  Even though the responders decreased in number with repeated




exposure, 2 subjects persisted in responsiveness through the fourth exposure




day.  Statistical comparison with the control group (OZADAPT I) was not







                                     310

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE	Haak







       TABLE 31-3.  SIGNIFICANT RESPONSIVENESS*  AMONG  CONTROL  SUBJECTS
OZADAPT
I



IV



Day
2
3
4
5
2
3
4
5
Number of
Significant Responders
0
0
0
0
0
0
0
0








aDefined  as a  decrease of >20%  in FEV^  or MMEF.









significant because of the small number of subjects.   In  OZADAPT  III, 7 out of




15  subjects were initial responders.   Once again we observed  a decrease in




number  of responders with continued exposure,  although 2  subjects were




responsive even after the fourth exposure day.  In OZADAPT V, there were very




few responders on the first and second exposure  days  (rest).  On  the third and




fourth  exposure days (heavy exercise),  the numbers of responders  were very




similar to those in the first and second exposure days of OZADAPT III.  From




this, it  is quite apparent that the ventilation  level of  exercise challenge is




an  important provocative stress that may evoke individual sensitivity to 03




exposure. Note also that, even with this challenge,  one  third of the OZADAPT




V subjects did not respond in terms of our strict criteria.
                                      311

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE
   TABLE 31-4.  SIGNIFICANT RESPONSIVENESS3 AMONG SUBJECTS EXPOSED TO OZONE
OZADAPT Day

II 2
3
4
5
total

III 2
3
4
5
total

V 2
3
4
5
total
Number of
Significant Responders

4/15
2/15
1/15
2/15
4/15 (27%)

7/15
7/15
4/15
2/15
8/15 (53%)

1/15
3/15
9/15
7/15
10/15 (66%)
Probability
(I vs. II:)
0.010
0.111
0.341
0.111
0.010
(III vs. IV:)
0.003
0.003
0.050
0.241
0.001
(V vs. IV:)
0.500
0.112
<0.001
0.003
<0.001
aDefined as a decrease of >20% in FEV<| or MMEF.









CONCLUDING REMARKS









     Investigators are only beginning to understand the many factors which




affect pulmonary function during short-term acute exposures to 03 and other




oxidants.  Among these factors are:  concentration of the gas, ventilation




(exercise) level, duration of exposure, proximity of pulmonary function




testing to exposure peaks, attitude, previous exposure experience, humidity,




temperature, and many others.
                                     312

-------
31.  HUMAN PULMONARY ADAPTATION TO OZONE	 	Haak


     Our OZADAPT studies stress that individual  sensitivity is also very

important.  The fact that the OZADAPT subjects were not "high-risk" subjects

underlines the need to investigate the presumably large population subsets

that are apparently intolerant of 03 exposure.



WORKSHOP COMMENTARY
R. S. Chapman;  Has there been any  subsequent  follow-up on OZADAPT subjects
who seemed not to "adapt" like the  group means, with respect to possible
atopic tendency not uncovered in  the  initial examination?

E. D. Haak;  Are you asking  how many  of our responsive people were discovered
(say, by skin testing)  to be truly  allergic, and how many of the nonresponsive
were found to be not allergic?

R. S. Chapman;  That would be a subquestion.   Having discovered that some of
the subjects were indeed not "adapting," did you do any follow-up skin or
other testing to reveal what we might call an  "occult allergic tendency"?

E. D. Haak;  We have not gone back  and done any skin testing on those
individuals who appeared to  be persisting in their responses to 03 exposure.
As a matter of fact, we haven't gone  back and  done any follow-up testing of
any kind on any OZADAPT subjects.   Rather, we  have been involved in an
analytical appraisal of the  study.  We might be able to go back and do some
follow-up testing.  Unfortunately,  our subject population is a transient
student population; many of  them  have graduated and would not be available for
testing.
                                      313

-------
    32.  OZONE-INDUCED HYPERREACTIVITY AS MEASURED BY HISTAMINE CHALLENGE
                          IN NORMAL HEALTHY SUBJECTS

                               Milan J. Hazucha

                        Division of Pulmonary Diseases
                            Department of Medicine
                              School of Medicine
                         University of North Carolina
                            Chapel Hill, NC  27514
INTRODUCTION



     Haak (Chapter 31 of this volume) has demonstrated that 0.4 ppm ozone (03)

significantly decreases pulmonary performance in human subjects who are

exercising.  Even under resting conditions, 0.6 ppm 03 for 1 or 2 h results in

a significant impairment in group mean pulmonary function.  However,

examination of each individual's performance under such conditions reveals

considerable variation in the degree of response to 03 exposure.



     In previous studies, we observed that some subjects show little change

even at 03 concentrations as high as 0.8 ppm, while other, apparently more

sensitive subjects show decreases of up to 50-60% in maximal midexpiratory

flow rate following 2 h of exposure with exercise (Bates and Hazucha 1973).

Clearly, such a wide spectrum of individual variability is obscured if effects

are evaluated only as group mean responses.  We also noted that the duration
                                     314

-------
32.  OZONE-INDUCED HYPSRREACTIVITY	Hazucha







of effect is only a matter of hours:  we were not able to detect by




spirometric means any residual effects at 24 h post-exposure (Bates and




Hazucha 1973).  Recently, however, Golden et al. (1978) demonstrated that O3




exposure can lead to airway hyperreactivity that persists for several days




(and sometimes weeks) even in subjects who exhibit no continuing effects on




pulmonary mechanics.









     Consequently, the  study reported here attempted to answer two basic




questions:   (1) Does 03 exposure  alter airway reactivity as assessed by a.




standard bronchial challenge technique using histamine aerosol (Chai et al.




1975)7  (2)  Is there any  correlation between individual sensitivity to 03 and




degree of airway reactivity?









PROTOCOL









     The subjects were  14 healthy individuals having no history of allergies




or asthma.   A 3-d protocol was employed.  On Day 1, subjects received a 2-h




air (control) exposure.  On Day 2,  subjects received a 2-h exposure to 0.6 ppm




03. Each exposure period included two  15-min periods of treadmill exercise (4




mi/h x 10%  grade x 15 min = 1 mi  walk).  On Day 3, the subjects returned for




24-h follow-up testing.









     A battery of pulmonary function tests  (PFT's) was performed prior to each




exposure, 2 h post-exposure, and  24 h post-exposure.  The PFT's included:




functional  residual capacity  (FRC)  and  airway resistance (RAW) acquired in a







                                     315

-------
32.  OZONE-INDUCED HYPERREACTIVITY	  Hazucha







plethysmograph, several measures of forced vital capacity (FVC), vital




capacity (VC), forced inspiratory volume (FIV), and maximal voluntary




ventilation (MW) measured by a dry seal spirometer.  During exposure to air




or 03, FVC and RAW were measured at 1 and 2 h of exposure, 10 min after the




preceding exercise period.  In the post-exposure 2-h recovery period, FVC was




monitored at 30-min intervals.









     The PFT battery was always followed by a complete histamine challenge




test.  Histamine aerosol was generated in a De Vilbiss #65 ultrasonic




nebulizer.  The aerosol was passed through an impactor chamber in order to




narrow the particle size distribution to a mass median aerometric diameter of




2.0 ± 1.6 ym.  Seven different concentrations of histamine were generated:




0.0 (saline), 0.3, 0.6, 1.25, 2.5, 5.0, and 10.0 mg histamine salt/ml aerosol.




Briefly, the challenge test consisted of inhaling 5 normal tidal breaths




(800-1000 ml/breath) of each concentration at 5-min intervals.  One FVC and




one RAW measure were obtained 2 min after inhalation of each concentration.









RESULTS









     Figure 32-1 shows the mean response and recovery of 14 subjects to 03 as




determined by changes in forced expiratory volume (FEV^).  The broken line




shows the mean control (air exposure) data (±  1 S.D.) while the solid line




represents the mean 03 exposure data.  Controls showed no statistically




significant changes over the exposure and recovery periods; the maximum




absolute difference between mean FEVi values was only 270 ml.  In contrast, 03







                                     316

-------
                                                                                                         OJ
                                                                                                         N)
                                                                                                         M
                                                                                                         I
            5-
                                                         A
                                                         A
                                                         A
                                                  A
                                                  A
w
a
U)
            2-
                                                                                                    T
                                                                                       H
                                                                                       <
                                                   time (hours)
                                                                                                        a
                                                                                                        u
    Figure 32-1.
Mean FEVj^ in  14 human subjects during and after exposure to air and O3.  Solid line =
2-h exposure  to 0.6 ppm 03; broken  line = 2-h exposure to air  (control).

-------
 32.  OZONE-INDUCED HYPERREACTIVITY	      Hazucha






 inhalation progressively lowered mean FEV"i.   After 2  h of breathing 03, mean




 PEV-i had dropped by 23%, a significant decrease (p <  0.001).   During the 2-h




 recovery period, mean FEV^j returned slowly towards the pre-exposure level and




 at 2 h was almost the same as the control.









      The RAW response (Figure 32-2)  was similar but opposite  in  direction.




 After 2 h of O3 exposure,  mean RAW increased  by 45% (p <  0.001).   At the end




 of the recovery period,  mean RAW was only  7%  above the pre-exposure value (not




 statistically different).   These observations are  in  accord with  previously




 published results from our laboratory as well as other research centers.




 Implicit in these dynamic  lung function changes is the production by 03  of




 significant upper airway bronchoconstriction.   Although a  2-h recovery period




 appeared sufficient for  subsidence of the effects  on  a group mean basis,  some




 of the exposed subjects  continued to exhibit  minor residual effects.








      Although effects on pulmonary mechanics  (on a group mean basis)  had




 virtually disappeared at 2 h post-exposure, analysis  of the histamine




 dose-response curves  reveals evidence of airway hyperreactivity.   Figures 32-3




 and 32-4  show the linear regressions  obtained for  FEV} and RAW histamine




 dose-response data, respectively.  For  simplicity,  only linear regression




 lines  for each condition (pre-exposure, 2 h post-exposure, or 24  h




post-exposure)  are plotted;  individual points and  standard deviations are not




shown.  The 2-h post-exposure regression is represented as a solid  line.  In




Figure 32-3, note the considerable (but not statistically significant)




differences in both intercept and slope of FEV-| at  2 h post-03 versus the






                                     318

-------
                                                                                                      u>
                                                                                                      N)
          2-
      u
      
-------
UI
(O
                                                                                                     Ul

                                                                                                     (O
                                                                                                     M



                                                                                                     Z
                                                                                                     D
                                                                                                     M

                                                                                                     !*>
                                                                                                     fl
                                                                                                     fl
        0.0
                                           468

                                 HISTAMINE CONCENTRATION (mg/ml)
   Figure 32-3.   Mean FEV^ following histamine challenge in 13 human subjects before and after exposure

                 to  air and O3.  A = 2-h exposure to air (control);  O3 = 2-h exposure to 0.6 ppm O3.

                 0 = pre-exposure;  2 = 2 h post-exposure; 24 = 24 h  post-exposure.
a
0)
N
c
n
a-

-------
U)
                                                                                                        w
                                                                                                        O
                                                                                                        25
                                                                                                        §
                                                                                                        S
                                                                                                        M
                                           468
                                   HISTAMINE CONCENTRATION (mg/ml)
                                                       10
    Figure 32-4.   Mean RAW following histamine challenge in  13 human subjects before and after exposure
                  to  air and CK.  A = 2-h exposure to air (control)? O3 = 2-h exposure to 0.6 ppm O3.
                  0  = pre-exposure;
  2-h exposure to air (control)? O3 = 2-h exposure to 0.6 ppm O3.
2 = 2 h post-exposure;  24 = 24 h post-exposure.
                                                                                                       EC
                                                                                                       D)
S-

-------
32.  OZONE-INDUCED HYPERREACTIVITY		Hazucha







other times and conditions.  However, in the RAW regression lines (Figure




32-4), this post-C>3 slope is increased compared to the other conditions; thus,




the same histamine concentration elicited a greater response when administered




2 h after exposure to 03.  Although the differences in intercept are not




statistically significant, the 2-h post-O3 slope is significantly (p < 0.05)




elevated, reflecting greater airway reactivity.









DISCUSSION









     Recent studies by Lee et al. (1977) in dogs and Holtzman et al. (1979) in




human nonsmokers have demonstrated increased bronchial sensitivity immediately




after 03 exposure.  These authors state that such hypersensitization results




from exposure of bronchial irritant receptors caused by damage to the




epithelium.  In both studies, pretreatment with atropine and vagal cooling




blocked this hyperirritability; therefore, the investigators concluded that




the bronchomotor response is mediated via vagal cholinergic pathways.




Unfortunately, the histamine challenges employed in these two studies were




done at the end of exposure, not allowing for a complete recovery from 03.




For this reason, the airway "sensitization" observed in these studies is not




necessarily indicative of true hyperirritability.  It is possible that the




increase in sensitivity (bronchoconstriction) was related to a persisting




O3~induced bronchoconstriction to which the histamine effects were added.




That is, the decrease in the airway radius induced by histamine was




superimposed on the (>3-induced bronchoconstriction, resulting in a predictably




greater effect on airway resistance than if the 03 effect were not present.







                                     322

-------
32.  OZONE-INDUCED HYPERREACTIVITY	Hazucha







However, Habib et al.  (1979)  explored the variability of airway response  to




histamine and concluded  that  base-line bronchomotor tone is  not a major




contribution to the  increase  of resistance.   In our own studies, we allowed




pulmonary function to  return  to the pre-exposure level, in order to avoid this




confounding of measurements with double intervention (03 and histamine).  Even




though FVC, FEV-j, and  RAW values reached control levels, however, the airways




remained hypersensitive.  Such increased sensitivity may last up to several




weeks in some subjects (Golden et al. 1978).









     Closer examination of individual responses in our study showed a wide




spectrum of sensitivity following 03.  Some subjects did not appear to  respond




either  to histamine  or to 03, while a few subjects reacted strongly to  both




challenges.  On reviewing the individual responses to both 03 and histamine




 (considered separately), the  subjects appear to fall into two distinct




categories based  on  type of response:  reactors (decrease in FEV-j of >20% and




increase in RAW of >100% compared to control values) and nonreactors (the




remaining  subjects).  When the same separation criteria are  applied to  the




histamine dose-response data, subjects who responded to O3 seem to have been




more reactive to  histamine.  This reactivity was further enchanced by O3




exposure.  Despite incomplete data  analysis,  the correlation between response




to O3 and  enhanced response to histamine is striking.









     Several plausible explanations for this apparent interdependence of




histamine  reactivity and 03 response have been explored by other investigators




 (Golden et al.  1978).   It is possible that 03-induced peripheral airway







                                      323

-------
32.  OZONE-INDUCED HYPERREACTIVITY	Hazucha


obstruction causes maldistribution of the aerosol, with a greater proportion

going to the larger airways and consequently eliciting increased response.

Ozone damage of epithelial cells with consequent exposure of vagal sensory

receptors is another possible mechanism (Holtzman et al. 1979).  Such denuded

receptors may be more sensitive to inhaled histamine.  Accordingly, some

investigators (Orehek et al. 1977) have proposed the analysis of data in terms

of sensitivity and reactivity to histamine.  Our data have not yet been

analyzed in this manner; therefore, it is difficult to make an in-depth

evaluation of various mechanisms contributing to the magnification of

response.



     Regardless of mechanism, these studies demonstrate that 0.6 ppm 03 will

not only impair dynamic lung function but also increase at least transiently

the sensitivity of the tracheobronchial tree.  Such hypersensitivity may last

well beyond restoration of lung function as determined by spirometry.  The

high correlation between 03 reactivity and magnitude of histamine response

leads us to conclude that the histamine bronchial challenge can be used as a

screening test for 03 reactivity, and might also predict reactivity to other

oxidants.



REFERENCES
Bates, D. V., and M. Hazucha.  1973.  The short-term effects of ozone on the
     human lung.  In:  Proceedings of the Conference on Health Effects of Air
     Pollutants (Assembly of Life Sciences, National Research Council,
     National Academy of Sciences).  Report prepared for the Committee on
                                     324

-------
32.  OZONE-INDUCED HYPERREACTIVITY _ __                      Hazucha
     Public Works, U.S. Senate.   S.  Res.  135,  Serial No. 93-15.  U.S.
     Government Printing Office,  Washington, D.C., pp.  507-540.

Chai, H., R. S. Farr, L. A. Froehlich, D. A. Mathison,  J. A. McLean, R. R.
     Rosenthal, M. D. Sheffer,  II,  S. L.  Spector, ana R. G. Townley.   1975.
     Standardization of bronchial inhalation challenge  procedures.  J. Allergy
     Clin. Immunol., 56:323-327.

Golden, J. A., J. A. Nadel, and H.  A. Boushey.   1978.   Bronchial
     hyperirritability in healthy subjects  after exposure to ozone.  Am. Rev.
     Resp. Dis.,  118:287-294.

Habib, M. P., P.  O. Pare, and L.  A.  Engel.   1979.  Variability of airway
     responses to inhaled histamine in normal  subjects.  J. Appl. Physiol.:
     Resp. Environ. Exercise Physiol., 47:51-58.

Holtzman, H. J. ,  J. H. Cunningham,  J. R.  Sheller, G. B.  Irsigler, J. A. Nadel,
     and H. A. Boushey.   1979.  Effect of ozone on bronchial reactivity in
     atopic and non-atopic  subjects. Am. Rev.  Resp. Dis.,  120:1059-1067.

Lee, L.-Y., E. R. Bleecker, and J.  A. Nadel.   1977.  Effect of ozone on
     bronchomotor response  to  inhaled histamine aerosol in  dogs.  J. Appl.
     Physiol.:  Resp. Environ.  Exercise Physiol., 43:626-631.

Orehek, J., P. Gayrard, A.  P.  Smith, C. Grimaud, and J.  Charpin.  1977.
     Airway response to carbachol in normal and asthmatic subjects.  Am. Rev.
     Resp. Dis.,  115:937-943.
 WORKSHOP COMMENTARY
 J.  D.  Hackney;   Were the subjects exercising or resting?

 M.  J.  Hazucha;   The subjects were exercising twice:   during the  last  15 min of
 the first hour  and during the last 15 min of the second hour of  exposure.

 J.  D.  Hackney;   To follow up on your dose-response findings, did you  test the
 subjects'  responsiveness to histamine 24 or 48 h later?

 M.  J.  Hazucha;   Yes, we tested responsiveness at 24 h post-exposure.   There
 still  appeared  to be increased responsiveness to histamine, but  the
 correlation between 03 response and response to histamine was smaller.

 J.  D.  Hackney;   And at 48 h?

 M.  J.  Hazucha;   We didn't do any testing at 48 h.
                                      325

-------
32.  OZONE-INDUCED HYPERREACTIVITY
                                                                       Hazucha
R. K. Wolff;  Did you make any measurement of the actual inhaled dose of
histamine as opposed to the concentration in the nebulizer?

M. J. Hazucha;  Although we know how much of the aerosol was inhaled, we
really don't know how much of it was retained.  We believe that if the subject
inhaled approximately the same amount of aerosol at each dose, most likely the
same amount of aerosol was retained and thus the doses were comparable.  I
wish we could do such measurements.  It is possible, but we lack the technical
means at present.

R. K. Wolff;  Did you compare the histamine and the methacholine bronchial
challenges?

M. J. Hazucha;  We did.  In another study (using exactly the same protocol) we
challenged subjects with methacholine.  The magnitude of response seemed to be
about the same as with histamine, and we were again able to identify
responders and nonresponders.  The data are not yet completely analyzed so I
cannot give you exact figures.

Question;  Do you plan to use 0.25 ppm 03 in your system?

M. J. Hazucha;  Yes, in our forthcoming 03 threshold study.

Comment:  Are you referring to 03~plus-N02 studies?  We have plans to look at
a study similar to the Orehek study, to try to confirm or deny those
observations.  We will also look at 0.1 ppm 03 to confirm or deny the report
of De Lucia and Adams.  This should follow on the heels of the study discussed
by Ginsberg [Chapter 33 of this volume] .  We expect to start with a low-level
03 exposure which should feature the same points made by Haak [Chapter 31 of
this volume].

S. V. Dawson;  Did you see an apparent increase in the variance of the results
under 03 exposure?  This is something that I see recurring fairly often.  It's
not the mean but actually the variance that's changing, suggesting possibly
that some kind of a control system is becoming unstable.  A random effects
model—a standard statistical model which isn't used much in biology—examines
whether variance is significantly changed from one regime to another.  Have
you looked at that in your 03 studies?

M. J. Hazucha;  No, we have not yet statistically analyzed these variances.
We are indeed planning to do in-depth statistical analyses of the variances of
single and group data.  Specifically, we will look at correlations between
responders and nonresponders; we will evaluate the two group responses within
and between days.

     The increased variance which appeared during exposure and diminished
during recovery was caused, we believe, by the spread of responses induced by
03.  Some of the subjects reacted considerably, while some didn't respond at
all.  Your suggestion that we analyze these variances is very well taken.
                                     326

-------
32.  OZONE-INDUCED HYPERREACTIVITY	                              Hazucha
M. Goldman;  In relation to Dr. Dawson's question on variability, [Figures
32-1 and 32-2] indicate the number of subjects to be ~14 in the beginning but
only ~9 during recovery.  This may be an explanation for the variability
change.  Were the same number of  individuals measured throughout exposure and
recovery?

M. J. Hazucha;  Yes.  Fourteen subjects were studied from the very beginning
to the very end.  In  some conditions, however, we were unable for technical
reasons to recover  data.  That's  why [the  figures]  show only 9 of the 14
subjects included in  the recovery period.
                                      327

-------
    33.  RESPONSE OF NORMALS AND ASTHMATICS TO LOW-LEVEL NITROGEN DIOXIDE

                               Joel F.  Ginsberg

                  Health Effects Research Laboratory,  MD-58
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park,  NC  27711
INTRODUCTION



     The oxides of nitrogen are important constituents of air pollution;

nitrogen dioxide (NO2> is particularly important.   Our interest in defining

the adverse effects of low-level NO2 exposure was  stimulated by the increased

use of diesel fuel in transportation and power plants.



     High levels of NO2 are known to cause inflammatory lung disease in humans

and are implicated as carcinogens in animals.  Lower levels, such as those in

the ambient air, obviously need further evaluation in man, particularly in

sensitive individuals (i.e., those most likely to  be at risk).  A recent

French study (Orehek et al. 1976) suggested that 0.1 ppm NO2 may be harmful to

asthmatics.  This finding is at variance with previous work showing no harmful

effects at levels of up to 1 ppm in normal human subjects.  In the Orehek

study, 20 mild to moderately severe asthmatics, aged 17 to 44, were exposed

for 1 h to 0-1 ppm NO2 and air at weekly intervals.  The protocol also
                                     328

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	Ginsberg






included nonspecific bronchial challenge with carbacholine, a synthetic analog




of the naturally occurring neurotransmitter acetylcholine.  Two bronchial




challenges were performed in each subject.  In terms of specific airway




resistance (SRAW), N(>2 exposure caused the dose-response curve to shift to the




left, and lower threshold doses of carbacholine were required to cause




increases in SRAW.  From this it was concluded that very low levels of NO2 can




adversely affect some asthmatics.  This conclusion suggests that stringent




regulatory standards, particularly a 1-h standard, deserve consideration.








PROTOCOL









     In an effort to corroborate these findings, we designed a similar study




that.is now in progress.  Eventually, 15 normals and 15 asthmatics will have




been exposed to 0.1 ppm N02 and to air (each for 1 h) in a randomized,




double-blind, crossover design.








     Our asthmatics range from asymptomatic to mildly severe.  Where they are




asymptomatic, the diagnosis is based on a history of reversible wheezing or




dyspnea that is clearly seasonal and not associated with upper respiratory




tract infection.  All of our asthmatic subjects have at least one positive




cutireaction to a battery of allergens common to our area.  Our normals have




no symptoms or family history of atopy, and all have negative skin tests to




the same battery of allergens.  All subjects are nonsmokers.
                                      329

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	Ginsberg







     Subjects abstain from all medications for at least 48 h, and all




theophylline-containing foods are withheld for 20 h.  Asthmatics are not




studied within 4 weeks of bronchodilator use,  and none are taking steroids.




No subject is studied within 4 weeks of symptoms of an upper respiratory tract




infection.









     Bronchial challenges are performed as recommended by Chai et al. (1975)




modified for an ultrasonic aerosol generator (De Vilbiss #65).  We use




twofold-increasing concentrations of acetyl methacholine (hereafter referred




to as "methacholine") for seven doses.  The normals receive a first dose of




0.31 mg/ml and a maximum dose of 20 mg/ml.  The asthmatics receive a first




dose of 0.07 mg/ml and a maximum dose of 5 mg/ml.  Five tidal-sized




inhalations with a 4-s breath hold are used to ensure a high percentage of




particle retention.  Within 5 to 6 min of each dose, airway resistance (RAW)




and thoracic gas volume (TGV) are measured plethysmographically by the




technique of DuBois et al. (1956).









     Our major departures, then, from the Orehek study are:  (1) inclusion of




normals; (2) skin testing to exclude occult atopy in the normal group and to




ensure that all the asthmatics share a reagenic mechanism; (3) less clinically




severe asthma in the asthmatic subjects; (4) bronchial challenge with




methacholine (which has a shorter duration of activity due to its affinity for




the enzyme acetylcholinesterase) instead of carbacholine; and (5) more




bronchial challenges (including pre-exposure and 24 h post-exposure) at




varying concentrations.







                                     330

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	Ginsberg






     Bronchial challenge produces dose-response curves which can be analyzed




for the threshold dose of methacholine (defined as the dose causing a 100%




increase in SRAW) and for the slope of the line connecting responses at doses




above the threshold.  While the mechanisms are not clear, changes in these




parameters are gaining favor with investigators as a reflection of the adverse




effects of oxidant pollutants.  Of course, we must remember that




methacholine-induced bronchospasm does not reproduce an asthmatic attack




(which is certainly more complex than simple smooth muscle contraction).




However, methacholine-induced bronchospasm may simulate asthma closely for a




very short period of time (e.g., 15-30 min).  It is also important to remember




that a methacholine-induced increase in SRAW does not necessarily relate to




increased morbidity among the asthmatic population.









PRELIMINARY RESULTS








     We have obtained and plotted some preliminary data for four asthmatics




and four normal subjects.  In Figures 33-1 through 33-5, specific airway




resistance (RAW x TGV/1000) (SRAW) is plotted linearly on the ordinate,  and




the methacholine dose is plotted logarithmically on the abcissa.  The




logarithmic dose plot allows us to stretch the early part of the curve that




would otherwise (for asthmatic subjects) go up very rapidly and be compressed




at the left-hand side of the graph.  The term "PDAR" appears in some figures




and is used to describe bronchial challenge.  The PDAR base line represents




the first challenge given to each subject.  The PDAR 4 line indicates another




bronchial challenge performed in every subject at least 5 d after the






                                     331

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	 Ginsberg







base-line bronchial challenge.  (A minimum of 5 d was chosen because of




compatibility with previous work and good reproducibility of bronchial




challenge at 7-d intervals.)  The PDAR 5 line represents a challenge performed




24 h after PDAR 4; PDAR 6 is a challenge performed 24 h after PDAR 5.




Exposure to either air or NC>2 occurred immediately before the second of these




three PDAR's.  To summarize, the PDAR base line represents a dose response at




some time distant from the other three, which are dose responses taken at 24-h




intervals with an exposure to either air or NC>2 occurring just before the




second of the three.









     Figure 33-1 represents an asthmatic subject exposed to air and challenged




with 2.5 and 5 mg/ml methacholine.  Note that the base line appears shifted to




the right compared to the pre-exposure PDAR 4 curve.  The PDAR 5 curve is




similar in appearance.  Arrows mark each threshold dose.  The threshold doses




and the slopes of the lines beyond the thresholds are similar for PDAR 4 and




5.  The PDAR 6 curve, however, has a somewhat different appearance, with a




higher threshold and a reduced slope—in short, a shift to the right.  This




trend toward a depression of reactivity on the third of three consecutive days




seems to be real and may reflect a repeat treatment effect of the methacholine




challenge.  To assume that bronchial responsiveness is similar from week to




week, then, may not be valid.









     Figure 33-2 (two normal subjects) displays only the dose responses




obtained immediately after air and NC>2, 1 week apart (i.e,., two of the total




seven bronchial challenges done in each subject).  Note that both subjects






                                     332

-------
                                                                                                            U)
                 30
CO
                 25  -
                 20  -
     Figure 33-1.
                  15  -
QC
CO
                  10 -
                   5 -
      	 BASELINE
      	PDAR FOUR
      	PDAR FIVE
      	PDAR SIX
                        10-1
                    1QO. 5          100

                            LOG (DOSE LEVEL)
                                                   10° -5
101
Specific airway resistance following three bronchial challenges  in an asthmatic subject
exposed to air.
                                                                                                           I
                                                                                                           to
                                                                                                           w
                                                                                                            §
                                                                                                            CO
                                                                                                            o
                                                                                         o
                                                                                         CO
                                                                                         §
                                                                                         2
                                                                                         H

                                                                                                           n
                                                                                                           i
                                                                                                           i

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE
                                                     Ginsberg
                  30
                  25 •




                  20




               <  15 H
               cc
               in


                  10 -




                  5 •
                          AIR
                      1Q-1      1Q-0.5    1QO     1Q0.5

                                  LOG (DOSE LEVEL)
30




25 -




20 -
               CO
                 10 •
                  5 -
                       — AIR

                       	NO,
                      10-1     1Q-0-5    i00      1Q0.5     1Q1


                                  LOG (DOSE LEVEL)
Figure 33-2.  Specific airway resistance following two post-exposure

              bronchial challenges in two normal subjects exposed to air

              and NO2*
                                     334

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	Ginsberg






have thresholds of >5 mg/ml.   However, the slopes, of  the  lines beyond the




thresholds differ:  the lower  subject obviously shows a much  increased slope.




Other workers have suggested that these parameters (dose  threshold and slope)




characterize a given subject but do not necessarily correlate with one




another; i.e., a high threshold is not always  associated  with a reduced slope,




as demonstrated by the subject in the lower  panel of  Figure 33-2.









     Another important aspect  of Figure 33-2 is the apparent  lack of pollutant




effect on base-line SRAW.









     Figure 33-3  (two asthmatic subjects) again depicts only  the dose




responses obtained immediately after air  and NO2» 1 week  apart.  These




subjects display  lower thresholds than the normals in Figure  33-2; in each




case, the threshold is <1  mg/ml.  Also, both have steep slopes.








     The plots  in Figure 33-3  (at least those  in the  lower panel) again




suggest  no  significant pollutant effect on airway responsivity.  There is some




suggestion  of a pollutant-associated shift to  the right in the upper panel.








     Figure 33-4  presents the  full spectrum  of seven  dose responses obtained




in the asthmatic  subject of Figure 33-1.  This subject's  dose responses should




probably be examined week by week by comparing, in this case, PDAR 4 to PDAR




1,  both  of  which  are pre-exposure dose-response curves.   For  the week of air




exposures  (upper  panel),  this  subject's dose-response curves  are for some




reason shifted  to the  left of  the base  line.  Therefore,  the  effect of






                                      335

-------
 33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE
              Ginsberg
                              10-0-5     100     100.5
                                 LOG (DOSE LEVEL)
101
                                       10°
                                  LOG (DOSE LEVEL)
Figure 33-3.   Specific airway resistance following two post-exposure
              bronchial challenges in two asthmatic subjects exposed to
              air and NO2«
                                    336

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE
                                                         Ginsberg
                30-
cc
                25-


                20-


                15



                10


                  5
      —— BASELINE
      —••PDAR FOUR
      -•— .PDAR FIVE
      . — .PDAR SIX
                    „./
                     10-1      1Q-0.5     10°       100-5

                                   LOG (DOSE LEVEL)
                 30



                 25-



                 20
               |l5-
               DC
               CO

                 10-
          -BASELINE
          .PDAR ONE
          PDAR TWO
          .PDAR IHREE
                    ™     ••••^^^•^
                                        10°       10°5
                                   LOG (DOSE LEVEL)
                                           101
                                            101
  Figure  33-4.   Specific airway resistance following seven bronchial challenges
                in an asthmatic subject exposed to air and N02.
                                       337

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	Ginsberg






pollutant is best examined by comparing bronchial challenges done 24 h before




and immediately after exposure (in this case, PDAR 1 and PDAR 2 in the lower




panel).  Unfortunately, this subject is not a particularly good example.




Still, comparing pre- and post-exposure curves for a given week probably gives




a better reflection of the effect of pollutant than comparing only




post-exposure curves (as in the Orehek study).








     Figure 33-5 shows the mean responses of the four asthmatics and four




normals.  Again, this depicts only post-exposure curves which, as we have just




discovered, may not be the best way to examine the data.  Nevertheless, these




mean values for both groups show large between-group differences in both




threshold and slope; the asthmatics obviously have lower thresholds and




increased slopes.









     Although these data would have to be considered quite preliminary, the




mean pre- and post-exposure values suggest no effect of N(>2 in either normals




or asthmatics.  The apparent NO2~associated rightward shift in asthmatics may




represent a repeat treatment effect of methacholine.









PRELIMINARY CONCLUSIONS









     We conclude that there may be week-to-week variability in parameters of




response to bronchial challenge in a group of mildly symptomatic asthmatics.




Comparisons of responses after air and pollutant should probably be made




between pre- and post-exposure curves rather than on the basis of






                                     338

-------
                                                                                                               u>
w
w
VO
                    35
                    30-
                    25-
                            	  NORMALS-AIR


                            	  NORMALS-NO2


                            	ASTHMATICS-AIR


                            	ASTHMATICS - NO2
                    20-
I
cc
   15-1
                     10-
                           10-1
                      10-0.5         1QO


                               LOG (DOSE LEVEL)
1QO-5
                                                                                         .0

                                                                                    S=  0.73
                                                                                                               01


                                                                                                               I
                                                                                                               0}
                                                                                                               M
                                                                                             z
                                                                                             o
                                           I


                                           n
                                                                                                               a
    Figure 33-5.   Mean specific airway resistance following two post-exposure bronchial challenges in

                   four normal and four asthmatic subjects exposed to air  and NO».
                                                                                            3

                                                                                             i


                                                                                             i

-------
33.  RESPONSE OF NORMALS AND ASTHMATICS TO NITROGEN DIOXIDE	Ginsberg


post-exposure curves only.  A confounding variable in this type of analysis

may be a repeat treatment effect of methacholine.  Our preliminary data

suggest that a 1-h exposure to 0.1 ppm 1K>2 has no demonstrable effect on

airway responsiveness in either normals or mildly symptomatic asthmatics.
REFERENCES
Chai, H., R. S. Farr, L. A. Froehlich, D. A. Mathison, J. A. McLean, R. R.
     Rosenthal, M. D. Sheffer, II, S. L. Spector, and R. G. Townley.  1975.
     Standardization of bronchial inhalation challenge procedures.  J. Allergy
     Clin. Immunol., 56:323-327.

DuBois, A. B., S. Y. Botelho, and J. H. Comroe, Jr.  1956.  A new method for
     measuring airway resistance in man using a body plethysmograph:  Values
     in normals and in patients with respiratory disease.  J. Clin. Invest.,
     35:327-335.

Orehek, J., J. P. Massari, P. Gayrard, C. Grimaud, and J. Charpin.  1976.
     Effect of short-term, low-level nitrogen dioxide exposure on bronchial
     sensitivity of asthmatic patients.  J. Clin. Invest., 57:301-307.
WORKSHOP COMMENTARY
P. E. Morrow;  It's important to do the bronchial challenge before and after
exposure; at least that has been our experience.

     How might you explain a residual effect of methacholine that persists
from day to day?  With carbachol, which is established to be a longer-acting
drug, we certainly see no effect from a previous administration even over a
matter of hours.  I'm surprised that there might be a residual effect from
methacholine on a day-to-day basis.  Do you have any thoughts about this?

J. F. Ginsberg;  That's a good question.  We really don't have an answer; we
were equally surprised.  We are currently looking at methacholine shelf life;
the fact that we make up methacholine fresh daily may have something to do
with the decreased responsiveness that we see.  If there is some type of
adaptation to methacholine, it may possibly be due to a longer persistence on
receptors by more active metabolites.
                                     340

-------
      34.   HEALTH EFFECTS OF AIR POLLUTANTS IN THE TEXAS GULF COAST AREA

                              Robert S. Chapman

                  Health Effects Research Laboratory, MD-54
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                      Research Triangle Park, NC  27711
INTRODUCTION



     The 1977 Clean Air Act Amendments mandate that EPA perform a study of the

nature, transport, and effects of pollutants in the Gulf Coast area.  EPA's

Health Effects Research Laboratory at Research Triangle Park, North Carolina

is sponsoring the biomedical portion of this project; a companion effort by

the Environmental Sciences Research Laboratory (Research Triangle Park) will

examine pollutant sources and transport.



     This report describes the planned epidemiologic studies, emphasizing the

background work that led to their design.  Also described are some of the

various scientific "shoals" that we hope to avoid.  No data are reported, for

this project remains in the planning and coordination stages.
                                     341

-------
34.  TEXAS GULF COAST STUDY		       Chapman


BACKGROUND



     The specific congressional charge for this work occurs in Section 403(d)

of the Clean Air Act Amendments of 1977:
          (d) The Administrator of the Environmental Protection Agency
     shall conduct a study of air quality in various areas throughout
     the country including the gulf coast region.  Such study shall
     include analysis of liquid and solid aerosols and other fine
     particulate matter and the contribution of such substances to
     visibility and public health problems in such areas.  For the
     purposes of this study, the Administrator shall use environmental
     health experts from the National Institutes of Health and other
     outside agencies and organizations.
Various legislative support documents prepared by relevant congressional

committee staff broaden the scope of this effort to oxidants as well as

aerosols.  Background material also suggests that this work focus on the city

of Houston, Texas.



     We perceive a dominant attitude in the Houston area that it is

unreasonable to develop national regulations from data that have been

collected primarily in one geographic area.  (With respect to oxidants, of

course, this area is Southern California.)  Although this attitude is

sometimes expressed in a more heated than enlightening fashion, there is

underlying validity in the complaint.  As more studies are conducted and more

data become available, it becomes very apparent that a high degree of

circumspection is required whenever one generalizes results gathered in one
                                     342

-------
34.  TEXAS GULF COAST STUDY	Chapman







area to other areas.  Therefore, we believe  that  is  is  scientifically




reasonable to undertake  studies in the  Texas Gulf Coast.









     In view of these factors, we have  resolved not  to  second-guess which




pollutants to assess.  Rather, we intend to  determine the range of pollution




exposure in the Gulf Coast,  the concentrations, the  temporal distributions,




etc. before we irreversibly  commit ourselves to embark  on any specific study.









     Since Congress appears  to want a rather well defined package of studies,




we feel that the  project demands a certain balance between short- and




long-term ambient pollution  exposure.  Similarly, we consider ourselves well




advised to deal with a spectrum of health in the  population (insofar as




resources permit); specifically, we will examine  individuals who are




considered healthy as well as those who are  sick. Finally, although the




studies will be tailored to  the local situation in the  Texas Gulf Coast, we




have selected  study parameters similar  to those already studied in Southern




California.  We thus hope to enable maximum  comparability of results from the




two geographic areas.









PLANNING STUDY









     Our  first effort towards fulfillment of the  congressional charge was to




enter  into a contract with Radian Corporation  (Texas)  assisted by the




Southwest Research Institute (Texas)  for an  assessment of the kinds of studies




(in both the air  quality and biomedical areas)  that  would be most useful.   The







                                      343

-------
34.  TEXAS GULF COAST STUDY	 _.__	_.	Chapman






biomedical portion of the resultant planning document recommended five or six




different kinds of studies as "highest priority-"









     First, the planning document recommended that we assess the influence of




local air pollution on the development of lung cancer in the area.  We




consider this to be a quite reasonable recommendation:  over the past 15 years




or so, lung cancer rates (at least for white males) have practically doubled




in Harris County, the county in which Houston is located.  The current annual




incidence in white males is estimated at ~70 per 100,000 population, compared




to a national incidence of ~40.








     The planning document also recommended that we assess whether the local




ambient air has any mutagenic potential.  Because no actual human effects




would be assessed, such a study does not seem particularly appropriate for the




biomedical arm of this rather circumscribed congressional charge.  Perhaps in




connection with a study of cancer some mutagenicity assessment would be




appropriate; opinion is divided on that point.









     Thirdly, the planning document recommended that EPA perform a study of




maximally exercising individuals who exercise repeatedly and in a standardized




fashion.  This seems to be a quite reasonable recommendation, and we have




planned such a study.  The background for this work comes from both




experimental and epidemiologic studies.  There is now a great deal of




experimental evidence that oxidant (at least, ozone) effects become more and




more discernible as the level of exercise increases.  On the epidemiologic






                                     344

-------
34.  TEXAS GULF COAST STUDY	Chapman






side, the very interesting study almost  15 years ago of high school




cross-country runners in San Marino, California (Wayne et al. 1967) remains




one of the most convincing of all air pollution epidemiologic studies.  The




Wayne et al. study has certainly been stressed in the preparation of oxidant




criteria documents and standards.  Therefore, a study of maximally exercising




people fulfills several of our preliminary criteria.









     Another recommendation of the planning document was that EPA undertake an




ongoing health surveillance system,  i.e., longitudinal tracking over several




years of the occurrence of acute respiratory illness in children.  We do not




feel that this type  of effort fits comfortably into the program that Congress




intended:   (1) Such  a study is rather open-ended? (2) a careful study to




relate disease incidence to local air pollution would of necessity be very




time-consuming and very, very expensive  (consuming all the resources that




Congress appropriated for the entire effort).








     Finally, the planning document  recommended that EPA undertake serial




studies of  asthmatics, whose symptoms, physiology, and/or daily fluctuation in




use of medication might be correlated to short-term ambient pollution




exposure.   We consider this to be an interesting recommendation, largely




because there are again both experimental and  epidemiologic underpinnings for




the work.   A great deal of the experimental  underpinning comes from Dr. Jack




Hackney, whose work  with reactive and unreactive normal subjects as well as




chronic asthmatics suggests that persons with  reactive airways or wheezing
                                      345

-------
34.  TEXAS GULF COAST STUDY		   Chapman







tendency may be unusually sensitive to experimental and/or ambient oxidant




exposures (Hackney et al. 1975a, 1975b; Linn et al. 1978).  For reasons




discussed below, we propose to do the asthma assessment not in children but in




adult-onset asthmatics.









     To summarize, the three recommended studies that EPA will most likely




pursue are:  (1) a study of lung cancer in the area; (2) a serial assessment




of asthmatic symptoms, simple pulmonary physiology, and medication use; and




(3) a serial assessment of the performance, physiology, and (probably)




symptoms of people exercising vigorously in the field in a standardized and




repeated fashion.









     We originally contemplated a fourth study that was not specifically




recommended by the planning document:  a cross-sectional study in which




questionnaires concerning persistent respiratory symptoms and relevant




covariates would be distributed to adults in areas of differing pollution




exposure.  In EPA's hands, this general strategy has proven workable as well




as useful from a regulatory standpoint.  Previous cross-sectional studies on




sulfur oxides and particulates have demonstrated an internal consistency of




results (at least, qualitative consistency) that tends to reinforce the




findings of other EPA studies.  For the time being, limitations of financial




resources and manpower preclude the inclusion of a cross-sectional




questionnaire study in the Gulf Coast project.  Should one of the three




planned studies prove unfeasible, however, we would certainly reconsider this




type of effort.







                                     346

-------
34.  TEXAS GULF COAST STUDY            	Chapman






FEASIBILITY ASSESSMENT









     Prior to collecting any data, we  intend  to complete a thorough




feasibility ascertainment  for  each of  the  three planned studies.  Toward this




end  (and toward scientific assistance  throughout the remainder of the




project), we are assembling an extramural  advisory group that will formally




convene from time to time, and with whose  members the investigators will




maintain frequent informal contact.  The disciplines to be represented in the




group are epidemiology, biostatistics, air quality measurement, historical




estimation of air quality, pulmonary medicine and physiology, and clinical




allergy.  Hopefully, the advisors will help us avoid any severe methodologic




or interpretive problems.









Lung Cancer Study








     The proposed study of lung cancer will receive extensive feasibility




assessment.  The venue  for the proposed effort is Harris County, Texas, the




county  in which Houston is located.  Currently, 700 to 750 new cases of lung




cancer  are estimated to occur  in Harris County each year, the majority in




white males.  Female rates are much lower  (~15 per 100,000 population).




However, we may be  well advised to examine females, since they may be less




likely  to undergo occupational exposure in the petroleum and petrochemical




industries.
                                      347

-------
34.  TEXAS GULF COAST STUDY	Chapman






     A few years ago, MacDonald (1976)  geographically charted the incidence of




lung cancer in Harris County.  She found a correspondence between




concentration of incidence and either (1) proximity to the Houston Ship




Channel (a petroleum and petrochemical  industrial hub) or (2) location in a




downwind path from the Channel.  However, Marmor (1978) pointed out that




MacDonald did not have the opportunity  to measure and assess socioeconomic and




occupational variables.  She did not quantitate smoking differences, nor did




she specifically consider population mobility (certainly a crucial factor in




rapidly growing Harris County).








     Henderson and associates recently  completed a case-control study of lung




cancer in Los Angeles (Pike et al. 1979).  Preliminary observations (Menck et




al. 1974) resembled those by MacDonald:   lung cancer cases appeared to cluster




in an industrial area of the Los Angeles region.  Following a more careful




case control study, however, all of the discernible effects in this geographic




area were observed to result from differential occupational exposure.




Henderson was unable to turn up any effects resulting from ambient exposure




alone.









     Our feasibility effort for the cancer study will be complex, particularly




with respect to air quality estimation extending back over the past few




decades.  We hope to be able to geographically characterize the county in




terms of high, intermediate, and low pollution classifications for individual




pollutants and for combinations of pollutants.  We plan to assort cases and




comparison subjects into these zones and then, after adjusting for appropriate






                                     348

-------
34.  TEXAS GULF COAST STUDY	Chapman






covariates, determine whether the proportion of cases in high pollution areas




differs from the proportion in the control group.  Obviously, the success of




this case-control study will directly depend on the confidence with which we




can construct past exposure estimates for Harris County.  Should




reconstruction prove unsuccessful, we will have to choose between adapting the




study design or abandoning the study altogether.









     Another important feature of the lung cancer feasibility study will be to




ascertain the potential cooperation of  the local medical community.  The study




cannot succeed in the absence of very good relations with the pulmonary




clinicians in the area.








     Our feasibility study will also seek to evaluate the existing tumor




registry capacity.  How quickly will we be informed when a new lung cancer




case occurs?  A lack of rapid turnaround would require us to build our own




surveillance system into  the study.  Some well established tumor registries




take, at best, ~12 to 18  mo to inform an investigator of a specific new case.




For  lung cancer, this is  far too  slow:   the  1-yr mortality for newly diagnosed




cases of lung cancer is ~85%.








     In summary, successful completion  of the lung cancer feasibility




study—particularly the estimation of air quality for previous years—will be




a  considerable task both  scientifically and  politically.  In our opinion,




however, the rise in lung cancer  rates  as well as their current absolute




magnitude  justify this effort.






                                      349

-------
34.  TEXAS GULF COAST STUDY	       Chapman






TENTATIVE PROTOCOL









     Performance of the feasibility and full-scale studies will proceed under




a cooperative agreement executed in late 1979 between EPA and the University




of Texas School of Public Health at Houston.  Principal investigators from the




University of Texas will be Dr. Patricia Buffler and Dr. Reuel Stallones, Dean




of the School of Public Health.  This author will serve as principal




investigator (biomedical) for EPA.  All parties anticipate the effort to be a




cooperative agreement in more than name only.









Lung Cancer Study








     Should our feasibility efforts predict favorably for success of the lung




cancer study, we plan to collect data from cases and comparison subjects over




the entire three-year period of 1981-1983.   If the whole project succeeds, a




final report will be issued by mid-1984.








Asthma Study









     Our ultimate goal in the asthma study will be to follow ~60 adult-onset




asthmatics on a daily basis for 6 mo.  We plan to distribute to these subjects




some device by which each can easily measure one or more parameters of lung




function each study day.  The device used may be a Mini Wright Peak Flow Meter




or perhaps an instrument such as the Vitalor, which yields a permanent tracing




of the forced vital capacity maneuver.






                                     350

-------
34.  TEXAS GULF COAST STUDY         	Chapman






     Our reason for concentrating on adult-onset asthmatics is that we want to




consider the important covariate of fluctuation in medication use.  To date,




this covariate has not been formally assessed in serial studies of asthmatics;




therefore, available studies are open to considerable questions of




interpretation.  Warnings to doctors about the dangers of nebulizer use in




children have appeared in the pediatric literature.  Some sobering stories of




nebulizer overdose and paradoxical nebulizer effects in children have




convinced us that childhood asthma, despite its theoretical advantages, may




not be an appropriate field of  study in this project.  We welcome further




comment on the relative appropriateness of childhood and adult asthma.








     The "nebulizer chronolog"—an instrument to measure nebulizer use—is not




fully tested but may be useable by the time our study starts (probably in the




spring of  1981).  The nebulizer chronolog records the time of day at which




each squirt of the nebulizer occurs; thus, we will have at least the daily




number of  squirts (although it's unlikely that we'll be able to tell whether




one squirt is as big as another).








     Assuming the feasibility study forecasts success for the asthma project,




we expect  to issue a report by  mid-1982.









Exercise Study








     With  respect to the  exercise  study,  it is not possible  in  this report  to




outline a proposed schedule.  The  schedule  will depend  integrally on  the






                                      351

-------
34.  TEXAS GULF COAST STUDY	    Chapman






number of subjects that are ultimately selected, on their frequency of




standardized exercise, on the range of pollution exposure at the study site,




and on the complexity of that exposure.








     In this author's opinion, the most desirable subjects would be "dedicated




adult joggers" who strive to exercise vigorously each and every day, who run a




standard distance, and who perhaps even try to improve their time each day.




It is not certain that we can obtain such a group, largely because of the




absolute requirement for standardized exercise on every occasion.  Perhaps no




informal group of adult joggers, regardless of personal dedication, would




fulfill this requirement.  For this reason, we will canvass the area not only




for adult subject groups but also for possible high school and college groups.




We have already determined (to our dismay) that the high school cross-country




schedule will not permit replication of the Wayne et al. (1967) study




mentioned earlier.  This is because high school cross-country meets tend to




occur at different locations within cities and even within tri-county areas.




Such constant changes of exercise location might exert important unmeasured




effects on the results.









     We hope to produce our report on the exercise study shortly after the




report on asthma (i.e., mid-1982 or perhaps a little later).
                                     352

-------
34.  TEXAS GULF COAST STUDY	Chapman
REFERENCES
Clean Air Act Amendments  of  1977.   Public  Law 95-95, August 7,  1977.

Hackney, J. D., W.  S.  Linn,  J.  G.  Mohler,  E.  E.  Pedersen, P. Breisacher, and
     A. Russo.  1975a.  Experimental  studies  on  human health effects of air
     pollutants:  II.   Four-hour exposure  to  ozone  alone and in combination
     with other pollutant gases.  Arch.  Environ.  Health, 30:379-384.

Hackney, J. D., W.  S.  Linn,  D.  C.  Law, S.  K.  Karuza, H. Greenberg, R. D.
     Buckley, and E. E. Pedersen.   1975b.   Experimental studies on human
     health effects of air pollutants:   III.   Two-hour exposure to ozone alone
     and in combination with other pollutant  gases.  Arch. Environ. Health,
     30:385-390.

Linn, W. S., R. D.  Buckley,  C.  E.  Spier, R. L. Blessey, M. P. Jones, D. A.
     Fischer, and J. D. Hackney.  1978.  Health  effects of ozone exposure in
     asthmatics.  Am.  Rev. Resp. Dis., 117:835-843.

HacDonald, E. J.   1976.   Air pollution,  demography, cancer:  Houston, Texas.
     J. Am. Med. Women's  Assoc., 31:379-395.

Marmor, M.   1978.   Air pollution and  cancer in Houston, Texas:  A causal
     relationship?  J.  Am. Med. Women's  Assoc.,  33:275-277.

Menck, H. R., J. T. Casagrande, and B. E.  Henderson.   1974.  Industrial air
     pollution:  Possible effect on lung cancer.  Science, 183:210-212.

Pike, M. C., J. S.  Jing,  I.  P.  Rosario,  B. E. Henderson, and H. R. Menck.
     1979.  Occupation—"Explanation" of an apparent localized  excess of lung
     cancer in Los  Angeles County. In:  Energy  and Health (SIAM-SIMS
     Conference Series, No.  6).  Society for  Industrial and Applied
     Mathematics, Philadelphia, Pennsylvania, pp. 3-16.

Wayne, W. S., P. F. Wehrle,  and R. E. Carroll.  1967.  Oxidant  air pollution
     and athletic performance.   J. Am. Med. Assoc.,  199:901-904.
 WORKSHOP  COMMENTARY
 Comment;   We have heard a lot [at the Research Planning Workshop on Health
 Effects of Oxidants]  about the animal infectivity model.  I haven't heard you
 describe  any planned studies to investigate human susceptibility to infection.
                                      353

-------
34.  TEXAS GULP COAST STUDY        	Chapman
Are you planning any, or is that something that will be available only from
animal toxicology?

R. S. Chapman;  I'd love to do a study like that on a community level.  But,
frankly, I'm not sure that it's doable right now.  An epidemiologic study
dealing with pollution and respiratory infection would require careful
assessment of specific etiologic agents over time; this would be a very
difficult prospect in itself.  It would require a group of unusually willing
subjects, because (in my opinion) one would have to do serologic testing at a
fairly short interval.  For instance, there's evidence now that the immune
period for respiratory syncytial virus in children may be as short as 3 mo.
This means at least a couple of bleedings per year—a rather ticklish prospect
on an epidemiologic plane.  I'm disheartened to say, though, that a bigger
deterrent is the fact that such a study would need to be long-term (a decade
of data collection, at least).  Frankly, I don't think the present EPA funding
structure could realistically absorb such a long-term study.

Comment;  We see so much of these infectivity models in terms of the criteria
document; I would think it would be important to~

R. S. Chapman;  I agree with you one hundred percent.  There have been some
indirect assessments along this line in self-administered cross-sectional
questionnaire studies in which mothers were asked whether children had certain
specific syndromes diagnosed by a doctor over a given period of time.  Because
they rely on the mother's recall, such studies aren't (theoretically, anyway)
ideal.  More prospective assessments of acute respiratory illness in all
family members have been and are being performed by EPA; Lee [Chapter 30 of
this volume] mentioned one such study.  These studies allow one to compare
communities with respect to gross incidence of upper and/or lower respiratory
illness, perhaps with an allergic overtone.  I don't think they allow the
analysis to become any more specific than that.  Since these studies are
performed on a daily basis, time series statistics might conceivably be
developed to allow some shorter-term inference.  However, I feel that
including some assessment of community experience with specific etiologic
agents would be a highly desirable adjunct to this kind of work.

Question;  With respect to your asthmatic pulmonary function study, will you
draw upon the data that were collected in the Houston Area Oxidants Study?

R. S. Chapman;  Those data may be helpful in locating potential subjects.
Before we start, we hope to know where virtually every adult-onset asthmatic
in the area lives so that we can focus on a couple of residential clusters
(ideally, just one cluster), in order to minimize our aerometric requirements
for ambient measurement.  Other than that and some of the initial recruitment
methods, I'm not sure that the Houston Area Oxidants Study data will be
particularly useful to us.  If those data are carefully reanalyzed, their
usefulness may increase.
                                     354

-------
34.  TEXAS GULF COAST STUDY	Chapman


Comment;  Everything you've described has involved human pulmonary effects or
animal experimentation that includes extrapulmonary effects.  Do you plan to
look at any human extrapulmonary effects?

R. S. Chapman;  The only possible nonpulmonary work would be some simple
cardiovascular physiological measurements (perhaps heart rate, blood pressure,
and/or electrocardiogram) on subjects in the exercise study.

Question;  What about blood chemistry?

R. S. Chapman;  Probably not, although the door isn't entirely closed.  We
must make certain decisions with respect to our financial resources.  For one
thing, we won't be able to rely on  ambient measurements alone; we will
probably have to do some indoor measurements before we're through.  (Whether
those are performed in the homes of subjects or in similar homes nearby has
not been decided.  We're asking a lot of the subjects already.)  Blood
chemistries, desirable as they  might be, probably won't be possible just
because of resource limitations.

Question;  Was Houston involved in  the third National Cancer Survey?

Comment;  Yes.

Question;  Have you compared Houston with other cities with respect to lung
cancer?  Was it always higher?

R. S. Chapman;  No; of the six-county area, Harris County has the second
highest incidence rate.  Galveston  has a somewhat higher rate.

Comment;  I was more  interested in  a comparison to the other cities examined
in the National Cancer Survey.

 [Comments inaudible]

R. S. Chapman;  One of the reasons  we have to find some relatively low
exposure areas within Harris County is to provide grounds on which to assess
pollution effects.  If there is homogeneous exposure throughout the county,
there's no point in doing a case-control study.

     Some of the available data (admittedly, incomplete) suggest that there is
some kind of exposure gradient  over the years.  Whether we can estimate that
gradient is another question entirely.

Question;  Do I understand correctly that you intend, in the exercise study,
to obtain hourly values  for sulfate and nitrate?  Can that be done?

R. S. Chapman;  If we gave you  that impression, we either weren't thinking
very clearly or didn't express  ourselves very clearly in writing.  I don't see
any theoretical deterrent to gathering hourly filter samples for those


                                     355

-------
34.  TEXAS GULF COAST STUDY		Chapman


substances.  I don't see much point in going finer than that; hourly values
themselves may be difficult to obtain.  I'd appreciate comment on this point.
Do hourly measurements of nitrate make any sense?  I think they're doable, but
do they make any sense?  Is the distortion-to-real-information ratio so high
that the hourly measurement is going to be "garbage"?  Or does the ratio stay
constant regardless of the sampling interval?

Comment;  It's a variable measurement demanding care in technique.

J. D. Hackney;  Do I understand correctly that you intend to study adult-onset
asthmatics?

R. S. Chapman;  Yes, but I can be talked out of it!  I have a nebulous
suspicion that there may be something different about adult-onset asthmatics
in comparison to childhood-onset asthmatics who remain asthmatic as adults.  I
have no concrete evidence.  Do you think that my concern holds any water at
all?

J. D. Hackney;  I just wanted clarification.

Question;  Are you speculating that the adult-onset asthmatic is less likely
to be an atopic asthmatic and more likely to be a perennial nonseasonal
asthmatic?

R. S. Chapman;  In all candor, my thinking wasn't that refined.  It is highly
unlikely that we will have more than 60 subjects stay with us throughout the
course of this work.  Therefore, we might be well advised to impose a certain
homogeneity on the group.  I could certainly be convinced that it would be
best to choose adult asthmatics with childhood onset rather than adult-onset
asthmatics.  On the other hand, if there's absolutely no reason for that, we
could certainly widen the field of acceptance.

Question;  In the exercise study, how do you plan to perform all the pulmonary
function assessments "at once" (when they all "pile up on you")?

R. S. Chapman;  That's a good question—one that we haven't really worked out
yet.  The number of testing technicians ultimately hired will probably depend
largely on the number of subjects.  My own feeling is that no single group
will contain more than ~15 subjects.  It seems to me that three testing
technicians could test this number of people spirometrically in a rather short
period of time.  I would not expect a single subject (especially a trained
one) to stay with the testing technician for more than 3 min.

     But another question arises:  Would we be well advised to forget about
testing them immediately post-exercise?  Would we be smarter to test them,
say, 20 min post-exercise after they've had a shower?  That would give us a
standardized interval for testing; testing immediately post-exercise,
especially after an exhausting routine, might not be very informative.
                                     356

-------
34.  TEXAS GULF COAST STUDY	Chapman


M. Goldman;  With respect to the lung cancer study, isn't change in population
makeup a crucial facet?  There is probably a 20- to 30-year latent period for
lung cancer.  Therefore, if there ^s an environmental association, those at
risk may have received it at another location.  Secondly, it's very difficult
to remove the overwhelming effect of cigarette smoking both in women and in
men.  If we do this, there may be very little left.  This is especially true
when we try to determine what the "petrochemical" environment was 20 years ago
and what it is at this time.  In sum, are the problems of an intracounty
assessment surmountable?

R. S. Chapman;  That is a succinct  statement of our own concern.
                                      357

-------
             35.  OVERVIEW OF RESEARCH AND REGULATORY ACTIVITIES
                    OF THE CALIFORNIA AIR RESOURCES BOARD

                                John R. Holmes

                        California Air Resources Board
                                1102 Q Street
                             Post Office Box 2815
                            Sacramento, CA  95812
INTRODUCTION



     This report surveys the research and regulatory functions of the Air

Resources Board (ARB) of the State of California.  Chapters 36 through 39 of

this volume present details of some of the oxidant health effects studies.



BACKGROUND



     The California State Government has sponsored air pollution research on a

substantial scale for the past 10 years or so.  This author heads a program

which actually began in the University of California as Project Clean Air.

About 1970, that program was shifted from the University to ARB and expanded

to include investigators from outside the University of California system.



     Our charge is to do applied research on problems that are unique to and

critical in California.  Over the years, we have tried to adhere fairly


                                     358

-------
35.  CALIFORNIA AIR RESOURCES BOARD	   Holmes






closely to that mandate.  The program was  developed  in much the same way that




EPA's program is now being  developed:   client  relationships were established




with all the various ARB divisions and branches.   Input  from these clients




allows us to propose, on a  yearly basis, a research  program that responds to




their needs.









     Year-to-year  variability  in the funding level sometimes makes it




difficult to plan  with  certainty. Funding has ranged from ~$1.5 to ~$3.5




million a year.  We operate on a zero-base arrangement:  each year ARB puts




together a set of  projects  and takes it through the  budget process from the




Governor's Office  to the Legislature.   Appropriation is  on a




project-by-project basis.








     For the budget year beginning in July 1980, we  hope to have ~$4 million




for research within the State  of California.  The  Governor has proposed a




fairly healthy increase for us this  year?  what the Legislature will  do with




this proposal remains to be seen.









RESEARCH PROGRAM








     The ARB research program is divided into  seven  different categories; each




is discussed below.  Most  of the reports originating from this program are




available through  the National Technical Information Service in Washington,




D.C.  Available  from our office in Sacramento  is  a short report containing




one-page summaries of all  ARB projects conducted  over the past two years.






                                      359

-------
35.  CALIFORNIA AIR RESOURCES BOARD	Holmes






Effects of Air Pollution









     Research into the effects of air pollution represents a major part of our




program.  Health effects research has accounted for ~20% to ~25% of the total




research budget over the past five years; the fraction allocated to health




effects research in the proposed 1980 budget is somewhat larger.  A number of




problems remain to be resolved, and ARE will need better and better




information to establish California air quality standards in the years ahead.









     The other important area is research into effects on vegetation; ~10% of




our budget is devoted to this category.  Agriculture is the State's largest




industry; the value of crops produced in 1978 was ~$12.5 billion, with another




$2 billion generated by the forest products industry.  Thus, the State of




California is very concerned about limiting damage to crops and forests.








Economic Impacts of Air Pollution Control









     In the last couple of years, ARB has become very involved in estimating




the economic impacts of control options.  One of our contractors is working to




incorporate air pollution control in a statewide input-output economic model.




This model is designed to consider economics at the stage when control




measures are first proposed.  When the model is up and running, we'll be able




to test various proposed control measures for economic impacts on directly




affected industries.  Beyond that, the model will permit us to determine how




the costs of air pollution control filter down through our State's economy.






                                     360

-------
35.  CALIFORNIA AIR RESOURCES BOARD	Holmes






Emissions Inventory and Control Technology









     The third category of our research program—emissions inventory and




control technology—receives the  largest  fraction of funding.  About 40% of




our budget is devoted  to  this research, which  includes studies to improve




emissions inventories  for stationary sources and cars, feasibility studies on




the kinds of control equipment that  might be adopted for use in California




(e.g., scrubbers and ammonia injection as a means to control nitrogen oxides




emissions from large sources), and so on.









Atmospheric Processes








     About 15% of our  budget is  committed to a catchall category called




"atmospheric processes."  We've  sponsored a great deal of smog chamber work by




James Pitts' laboratory at the University of California - Riverside.  These




studies are designed to sort out some of  the intricacies of photochemical smog




formation and to determine the fates of various organic molecules present in




the air (solvent hydrocarbons, pesticide  hydrocarbons, etc.).








     Also funded in this  category is a great deal of work in the field.  For




example, ARB sponsors  tracer studies and  airborne measurements to establish




source-receptor relationships and track pollutant movements between the




State's various air basins or air quality control regions.  Results from this




work are calling into  question a fundamental assumption of our State law—the




assumption that the State can be divided  into  separate and independent air






                                      361

-------
35.  CALIFORNIA AIR RESOURCES BOARD	    Holmes







basins or sheds.  We now know that a great deal of pollution is transported




from such locales as the San Francisco Bay Area to the Great Central Valley,




and from the Los Angeles area down to San Diego and out into desert areas.




So, what happens in the way of controls upwind is a very important determinant




of what happens in air basins downwind.  Our field studies establish these




relationships.









Air Quality Modeling









     The fifth category, air quality modeling, entails the development of




models that simulate emissions and such atmospheric processes as chemical




transformation and dispersion.  There are regional models in various stages of




development for San Diego, Los Angeles, Sacramento, and Fresno.  The Bay Area




model was developed independently of our support.









     Although air quality models seem to be the wave of the future, our




experience shows them to be a mixed blessing.  In principle, such models can




provide very precise information on the benefits of various control measures.




But air quality models are also "data hungry":  To exercise a model properly




and with confidence requires a great deal of air quality data, meteorological




data, emissions data, and so forth.  Thus, although ARB is making progress in




this area, we have also discovered that it takes longer than expected to




reduce theory to practice.
                                     362

-------
35.  CALIFORNIA AIR RESOURCES BOARD	Holmes






Meteorological Forecasting









     Because air pollution can reach very high levels in such areas as Los




Angeles and the Central Valley, meteorological forecasting is an important




part of our research program.  We need to be able to forecast the occurrence




of severe episodes of photochemical smog in Los Angeles and of the trapping of




smoke from agricultural burning in the Valley during winter months.  When




employed by research meteorologists, advanced statistical techniques can be




used to very handily forecast the occurrence of such emergencies at least 24




and possibly as long as 36 hours in advance.









Measurement and Measurement  Methods








     Finally, ARB sponsors a fair amount of work on air pollution measurement




and measurement methods.  These efforts focus on ambient levels.  For example,




we've worked with the California Institute of Technology and with the




University of California  - Berkeley to develop methods to survey acid




precipitation.  A report  on  acid rain in California will be issued within the




next two months.








     The laboratory of James Pitts at the University of California - Riverside




has developed sophisticated  measurement techniques for vapor-phase ammonia, a




compound that is very important in the formation of photochemical aerosols due




to its neutralization of  sulfuric and nitric acids.  Understanding ammonia is




important to understanding the nature of the particulate burden in many areas






                                     363

-------
35.  CALIFORNIA AIR RESOURCES BOARD
                                                                        Holmes
of the State as well as for dealing with the problem of visibility




degradation.









CALIFORNIA AIR POLLUTION STANDARDS








     For more than a decade, the State of California has promulgated its own




ambient air quality standards and other standards to limit community exposure




to various pollutants and combinations of pollutants.  These California




standards fall into three distinct categories:  ambient air quality standards,




alert levels, and standards for hazardous substances.








     Ambient air quality standards are, of course, levels that are considered




to be safe over the long term for the general population.








     Alert levels are specific ambient levels at which extraordinary measures




are taken.  For example, the State has established alert levels for ozone that




require special action on the part of industry and, in some cases, the




population in general.  The best known is the school alert level for ozone:




when ozone rises to 0.20 ppm and that concentration is expected to persist for




an hour or more, schools and colleges are notified.  Most of our medical




advisors believe that children, especially, should curtail their outdoor




physical activities when ozone reaches this level.









     Recently the State has considered standards for hazardous substances.




This is a relatively new area for the Federal Government as well as for us.






                                     364

-------
35.  CALIFORNIA AIR RESOURCES BOARD	   Holmes






These hazardous substance  standards  are  somewhat  different from our ambient




standards or alert levels  in that they are  targeted specifically at the




vicinities of sources  (factories, etc.).  For  example, about two years ago ARE




adopted a hazardous substance standard for  vinyl  chloride.  In California,




vinyl chloride is not  a  ubiquitous pollutant—it's  localized to three or four




individual sources in  the  Los Angeles Basin.   Our standard provides that vinyl




chloride levels will be  monitored in the vicinity of each plant (i.e., at the




property line or thereabouts).   The  rationale  for this approach is that most




of the emissions from  such facilities do not emanate from a stack or any




particular large source; rather, they are "fugitive emissions" (principally,




leaks and accidental releases).   It  is very difficult to write an emissions




regulation that addresses  fugitive emissions in an  effective way, so ARB




adopted an air quality standard for  the  vicinity  of each vinyl chloride




facility.  Thus, each  facility  is  free to figure  out the best way to reduce




its  own emissions  to a level that  is safe for  the surrounding community.  This




makes the job of ARB easier,  too:   it's  simpler to  write this type of




regulation as compared to  a detailed engineering  specification type of




regulation.








     The ARB procedure for adopting standards  is  somewhat different  (and in




some ways simpler)  than EPA's procedure.  Responsibility is divided between




ARB, which actually adopts the standard, and the  State  Health Department,




which is required  under State law to advise ARB on  health-related standards.




The  State Health Department, in turn, is advised by a committee  of physicians




appointed by the California Medical Association.






                                      365

-------
35.  CALIFORNIA AIR RESOURCES BOARD	Holmes







      In contrast to the private or closed-door method of EPA (in which the




Administrator proposes standards on the basis of staff recommendations),




California standards are adopted via a public hearing process.  In this




author's opinion, there are virtues to both systems.  In the California




system, the ARB staff does a great deal of the groundwork prior to the




hearings, and the hearings can be lengthy.  However, the Board itself—the




group of individuals who must ultimately make the decisions—is exposed to a




much wider range of viewpoints than in the Federal system.









     Our system also differs from the Federal system in that there are no




rigid timetables for attainment of standards.  The ARB views standards more as




ultimate objectives of the air pollution control program.  The State statutes




do require "reasonable progress" toward attainment of standards, but there are




no rigid timetables.  As technology develops and as economic conditions




permit, ARB and the local air pollution control officials implement emission




controls to move ahead toward our standards.  This gives us a fair degree of




flexibility that is not present in the Federal system.









     In California law there are no distinctions between primary or




health-related standards and secondary or welfare-related standards.




Standards adopted by ARB are assumed to be protective of both human health and




welfare.  There are only a couple of standards that are strictly




welfare-related:  the standard for the hydrocarbon ethylene (a specific




toxicant for certain kinds of plants), and the State visibility standard,
                                     366

-------
35.  CALIFORNIA AIR RESOURCES BOARD	Holmes


which has turned out to be our most  stringent standard, the toughest  to meet,

and the one that we've made  least progress  toward  attaining.



WORKSHOP COMMENTARY
Question;  Do  the  budgetary numbers that you quoted represent the budget of
your research  program,  or is that the total  budget  of  ARE  (which, I presume,
involves a great deal of expenditure for administrative costs and all the
noninvestigative functions)?

J. R. Holmes;   Those figures are for actual  extramural research projects.
Those are the  dollars that we "shovel out the door," so to speak.

Question;  How does the yearly uncertainty in funding  levels affect your
ability to get other people to do work related to ARE  goals?

J. R. Holmes;   Well, it's certainly a problem.  Some of the projects that we
would like to  do almost require a guarantee  of continued funding (for example,
prospective  epidemiologic work).  We've talked to the  Legislature staff and
tried to convince  them  that we ought to have this sort of  funding capability,
but  so  far we  haven't met with success.  The Legislators simply don't want to
commit  funds over  that  period of time.
                                      367

-------
                 36.  LUNG INJURY AND DEPLETION IN THE MOUSE
                    FOLLOWING EXPOSURE TO NITROGEN DIOXIDE

                              Russell P.  Sherwin

                           Department of  Pathology
                              School of Medicine
                      University of Southern California
                              2025 Zonal  Avenue
                            Los Angeles,  CA  90033
INTRODUCTION



     This report summarizes some preliminary animal experimental data which we

consider highly relevant to the question of human adverse health effects

resulting from exposure to oxidants.  Lung tissue of mice exposed to nitrogen

dioxide (NO2) was analyzed for hyperplasia of Type II cells and protein

leakage.



BACKGROUND



     Preliminary studies of human lungs from the Los Angeles Medical

Examiner-Coroner's Office and the Los Angeles County-University of Southern

California Medical Center suggest that a progressive depletion of lung tissue

occurs in everyone.  The rate at which this occurs and the extent of the

depletion in the well population are presently unknown.  However, the crude
                                     368

-------
36.  PULMONARY EFFECTS OF NITROGEN DIOXIDE  IN THE MOUSE	Sherwin






indicators available (in particular,  "naked eye" measurements of emphysema,




and the frequency of bronchiolitis and  structural disruption by microscopic




study) point to a serious underestimation of the extent of destructive lung




disease in the general population.









     The underestimation of  lung disease is largely due to the tremendous




amount of lung reserves (e.g., ~50 billion  Type II cells and ~30 billion Type




I cells in ~300 million alveoli).  Substantial numbers of these lung cells can




be lost in the absence of overt  clinical symptoms or signs, and routine




pulmonary function tests may not become clearly positive until as much as half




of the lung has been irreversibly altered or destroyed.  Routine autopsies are




not much more accurate:  criteria for recognition and quantitation of the




various kinds of emphysema fall  far  short of a uniform definition and




application, and this affects most of the autopsies carried out in major




universities and hospitals.








      In essence, the major problem of air pollution is a disturbance in the




microecology of the human body,  i.e., the diverse cell societies not only of




the lung but of other organs as  well.   As with the visible ecology, vanishing




species of life is an early  and  serious part of the adverse effect.  With




respect to the lung, the Type I  cell is of  special concern:  the ultrathin




lining it provides for the alveoli is critical to gas exchange.  Thus, the




first target of our research effort  was a quantitation or inventory of the




Type  I cell population.
                                      369

-------
36.  PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE 	Sherwin






METHODS AND RESULTS









Inventory of Type I Cell Population








     The methodology we used to count Type I cells is an indirect one, based




on the now well established fact that loss of Type I cells is manifested by a




replacement or hyperplasia of Type II cells.  First, we developed a means to




single out the Type II cell for quantitation (Sherwin et al. 1967).




Subsequently, mouse whole lung sections were processed for lactate




dehydrogenase (LDH) responsiveness.  We found LDH positive granules of the




Type II cell to be densely packed around the nucleus, whereas LDH positive




granules of macrophages and other lung cells were widely dispersed throughout




the cytoplasm and relatively weakly stained.  It should be noted that




intra-alveolar macrophages are, for the most part, removed by frozen section




processing procedure (i.e., intra-alveolar gelatin from perfusion-inflation is




washed out during processing and carries macrophages with it).  Furthermore,




interstitial macrophages are not detected by the gray value setting since they




have extremely thin and elongated cytoplasmic processes within the




interstitium, and this minimizes the LDH response.









     An image analyzer with a highly sophisticated detection system and field




editor provided a very practical means to obtain accurate large-volume




measurements not only of the number of Type II cells but of their sizes as




well (Margolick et al. 1973; Sherwin et al. 1973).  The number of Type II




cells was adjusted to a base line of number of alveoli as reflected in the






                                     370

-------
36.  PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE	Sherwin






amount of alveolar wall area detected in the same field by the image analyzer.




Additional measurements obtained at the same time were:  numbers of Type II




pneumocytes according to size distributions (i.e., by diameters of >8 ym, >10




ym, and >12 ym) , internal surface area of the alveoli, and linear intercepts




of the alveolar walls.  Computer processing of the data permitted large-scale




statistical analysis and diverse kinds of data analysis (Sherwin et al.




1979a).  Macrophage quantitation and total alveolar counts are currently in




progress.









     Most recently, we  studied  the effects of 0.34 ± 0.02 ppm NO2 on




Swiss-Webster  adult male mice.  A total of 120 mice were equally divided into




control and exposed groups.  Exposed animals received NO2 for 7 h/d, 5 d/week,




for 6 weeks.   The  lungs of these animals were inflated with gelatin to




approximate the volume  of full  lung expansion, and frozen sections were




processed for  the  LDH reaction. Eight whole lung sections were obtained from




the left lung  of each animal and numbered according to location in a sagittal




plane running  medially  (from the lung hilum) to  the lateral lung peripheral




area.  Four fields were quantitated for each lung section, for a total of 32




fields per animal. The data obtained are summarized in the tables that




follow.








     Table 36-1 displays the data  from each field treated independently and




analyzed.  The findings are:
                                      371

-------
36.   PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE  MOUSE
Sherwin
            TABLE 36-1.   LUNG TISSUE DATA (FIELDS ANALYZED SEPARATELY)
                    FOR MICE EXPOSED  TO NITROGEN DIOXIDE VS. AIR
Standard
Exposure Mean Deviation
Number of Type II Cells (>8 pm)b
N02 314.35 145.46
air 288.05 130.04
Area of Type II Cells (>10 ym)
N02 3129.09 2885.15
air 2722.58 2265.38
Alveolar Wall Area (no sizing) Minus
N02 39166.40 21265.98
air 38044.79 21759.30
Alveolar Internal Surface Area
N02 2747.92 1231.38
air 2622.36 1180.57
Linear Intercepts of Alveolar Walls
NO2 8229.40 3749.41
air 7786.04 3526.56
Alveolar Wall Area (>10 ym) * Number
NO2 82.84 28.82
air 88.61 34.43
Type II Cell Area * Number of Type II
N02 10.91 4.94
air 10.64 3.99
Internal Surface Area
N02 6.88 0.96
air 6.86 1.04
Standard T
Error Value

3 . 58 5 . 39
3.30

71.01 4.42
57.34
Area of Type II Cellsc
523.37 1.48
550.91

30.30 2.95
29.88

92.25 3.45
89.29
of Type II Cells (>10 ym)<3
0.71 -5.16
2.87
Cells
0.12 1.70
0.10

0.02 0.54
0.03
Freauency
Distribution

3211

3210

3209

3211

3210

3207

3206

3210
Probability
(two-tailed) a

p + 0

p -»• 0

0.14(NS)

0.003

0.001

p + 0

0.09

0.59
 aNS » not significant.

 bper lung field.

 cFor wall area with sizing of >10 vm minus area of Type II cells (>10 urn),  p = 0.441  (US).

 ^Alveolar wall area * number of Type II cells = inverse of number of Type  II cells per lunq field
 adjusted to amount of lung tissue (or number of alveoli); 1 unit «• 3.2 ym2.  AH combinations of two
 wall areas divided by three Type II cell measurements (>8 ym, >10 ym, >12  urn) provided p values + 0.
 All Type II cell measurements (>8 ym, >10 ym, >12 ym) provided p values +  0.
                                            372

-------
36.  PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE	Sherwin


     1.  The numbers of Type II cells  (>8 pm, >10 ym, and >12 yra in
         diameter) were increased for  the exposed group (p * 0).

     2.  The mean areas per field of the Type II cells were greater for
         the exposed group (p > 0).

     3.  The value, alveolar wall area T number of Type II cells per
         field, was greater for the control animals (p •»• 0), and
         indicated a a greater number  of Type II cells for the exposed
         animals since this is an inverse of the number function.  Also,
         the wall area base line was not significantly different between
         groups, indicating that the ratio difference is a function of
         the number of Type II cells.
     Table 36-2  displays  the  results of a two-way nested analysis based on 60

pairs of animals using a  mean field value for each animal.  We found an

increase in Type II  cells for the  exposed group, but this increase was not

statistically  significant,  apparently because of the great variability between

animals.  Also,  the  analysis  assumed a normal distribution (e.g., no high

responders).



     Again using animal rather than field comparisons, we ranked the animals

according to upper quartile distribution and subjected the ranking to

chi-square analysis  (Table 36-3).  A statistically significant (p < 0.025)

difference in  the  number  of Type II cells was noted when the number of Type II

cells was adjusted for alveolar wall area.   (This adjustment compensates for

variations of  inflation of the lung during  processing.)  Without the

adjustment, the  difference was still apparent but only at the borderline of

statistical significance  (p < 0.1).
                                      373

-------
w
BLE 36-2. LUNG TISSUE

Measure

Number of
Type II Cells
(>8 vim)

Number of
Type II Cells
(>12 pm)
Number of
Type II Cells
(>10 um)
Type II Cell
Area


Alveolar Wall Area


Alveolar Wall Area
Type II Cell
Area
Alveolar
Internal
Surface Area
DATA (TWO-WAY NESTED ANALYSIS) FOR MICE EXPOSED



A(G)
G
Residual

A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual

A(G)
G
Residual
A(G)
G
Residual
A(G)
G
Residual
aAnimal nested within group; slide
^NS = not significant.





Distribution
Frequency

112
1
3067

112
1
3067
112
1
3067
112
1
3067

112
1
3067
112
1
3067
112
1
3067
and field are



Mean
Square

327,573
323,856
9,939

169,835
122,024
5,148
230,859
217,392
7,101
107,837,392
91,516,928
3,242,773

10,420,183,040
58,261,504
202,329,168
4,300,115,968
26,804,224
89,261,632
3,270,255,616
675,282,944
69,220,512
considered replicates.



TO NITROGEN DIOXIDE VS. AIR
Frequency
Ratio Probability13

32.96 p * 0
0 . 99 NS


32.99 p * 0
0.72 NS

32.51 p + 0
0 . 94 NS

33.25 p -»• 0
0.85 NS


51.50 p -»• 0
0.06 NS

48.17 p -»• 0
0.01 NS

47.24 p -v 0
0.21 NS





W
ffl
V
1
5
Q
t>
a
a
**
"fl
w
•>
i
o
m
3
d
I
o
•H

H
4
H
z
1
§
•J
"A
ft
M




01
tr
(D
|

-------
36.   PULMONARY EFFECTS OF NITROGEN  DIOXIDE  IN THE MOUSE
                                                       Sherwin
  TABLE 36-3.  LUNG TISSUE DATA  (CHI-SQUARE ANALYSIS  OF CONTINGENCY TABLES)
                  FOR MICE  EXPOSED TO NITROGEN DIOXIDE  VS.  AIRa
         Number of Type II Cells (<8  pm)

                  L      U     Total
          N02     38
          air     47
       Total
                  35
19
10

29
 57
 57

114
       X  = 3.75 (0.05 < p < 0.1)b
                              Area of Type II Cells
NO2
air
Total
2
L
37
46
83

U
18
11
29

Total
57
57
112

                               2.63 (NS)
                Alveolar Wall Area
                                                   Alveolar Internal Surface Area
N02
air
Total
2
L
44
41
85

U
13
16
29

Total
57
57
114

NO2
air
Total
2
L
41
44
85

U
16
13
29

Total
57
57
114

             0.42  (NS)
                                                       0.42 (NS)
        Linear Intercepts of Alveolar Walls

                   L      U    Total
N02
air
Total
2
X = 1.16
1
40 17
45 12
85 29

(NS)

57
57
114



                                                        Alveolar Wall Area *
                                                   Number of Type II Cells (>10
                                           0
                                                     NO2    48
                                                     air    37
                                                   Total
                                                            85
                                           9
                                          20

                                          29
                                        Total

                                          57
                                          57
                                                                          114
                                                   X  = 5.60 (0.01 < p < 0.025)c
                                                    1
         particular  group and membership in the
         (p > 0.1).

        bSame for Type II cells of >10 ym and >12 ym.

        GFor Type II cells of >8 ym:  0.05 < p < 0.1.
                                           375

-------
 36.  PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE	Sherwin






 Protein Content of Lung Tissue Following Exposure to Nitrogen Dioxide









     In other studies, we used a molecular probe, horseradish peroxidase




 (HRP), to measure the protein content of lung tissue from N02~exposed mice.




 In brief, intermittent exposures of Swiss-Webster adult male mice to NC>2 at




 levels as low as 0.4 ppm for periods of 3 and 6 weeks resulted in a greater




 content of HRP in the lungs of exposed animals as quantitated by




 polyacrylamide gel electrophoresis (Sherwin et al. 1979b).  In three




 independent experiments, with two test periods per experiment, exposed animals




 showed greater mean HRP values for five of six periods (upper quartile




 analysis).  Using a two-factor analysis of variance alone, statistically




 significant differences were found in three of six periods.  Ultrastructural




 studies of animals injected with the HRP probe showed HRP to have free access




 to the basal lamina of the lungs of both control and exposed animals; studies




 of the electron micrographs (incomplete at this writing)  have shown no




 differences in distribution.  Pinocytotic vesicles of both endothelium and




 epithelium appear to play major roles in the transport of HRP from the




 capillary to the alveolar lumen as part of the bidirectional protein transport



 mechanism.









 DISCUSSION









     While the findings reported here are preliminary, they are in line with




earlier reports showing both Type II cell hyperplasia and protein leakage at




higher levels of NO2 exposure.  Other studies are in progress to confirm and






                                     376

-------
36.  PULMONARY EFFECTS OF NITROGEN DIOXIDE IN THE MOUSE   	Sherwii


elaborate upon these  preliminary results (see Addendum, below).  The question

of reversibility  is of special concern.   It should be noted that Type II cell

hyperplasia and protein  leakage are very early findings and common

denominators  in diverse  kinds of human lung disease.
REFERENCES
Margolick,  J.  B.,  S.  P.  Azen, and R. P.  Sherwin.   1973.  An image analyzer
     quantitation  of  Type II pneumocytes.  Am.  Rev.  Resp. Dis., 108:704-707.

Sherwin,  R. P.,  S. Winnick, and R. D. Buckley.   1967.  Response of lactic acid
     dehydrogenase-positive alveolar cells in the lungs of guinea pigs exposed
     to nitric acid.   Am. Rev. Resp. Dis., 96:319-323.

Sherwin,  R. P.,  J. B. Margolick, and S.  P. Azen.   1973.  Hypertrophy of
     alveolar  wall cells secondary to an air pollutant.  A semi-automated
     quantitation.  Arch. Environ. Health, 26:297-299.

Sherwin,  R. P.,  K. Kuraitis, and V. Richters.  1979a.  Type II pneumocyte
     hyperplasia and hypertrophy in response to 0.34 ppm nitrogen dioxide:  An
     image analyzer computer quantitation (abstract).  Fed. Proc., 38:1352.

Sherwin,  R. P.,  D. Okimoto, D. Mundy, and J. Bernett.  1979b.  Clearance of
     horseradish peroxidase in the lungs of mice exposed to an ambient level
     of nitrogen dioxide (abstract).  Lab. Invest.,  40:280.
 ADDENDUM
      At the time of review of the draft Proceedings, data from another  study
 had again shown hyperplasia and hypertrophy of Type II cells.   In  this  study,
 mice were exposed to 0.3 ppm NO2 for 6 h/d, 5 d/week, for 6 weeks, with tests
 at 4 and 10 weeks post-exposure.  Also, comparisons of the three test periods
 showed impaired mitochondrial growth (size) .  The mitochondrial and Type  II
 cell alterations persisted to the 10-week post-exposure period.
                                      377

-------
           37.  PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE

                              Steven M. Horvath

                      Institute of Environmental Stress
                           University of California
                           Santa Barbara, CA  93106
INTRODUCTION



     This report summarizes some preliminary results of human clinical

exposures to ozone (03) performed at the Institute of Environmental Stress at

the University of California - Santa Barbara.  Results of some of the

individual studies have been published elsewhere (Folinsbee et al. 1978, 1980;

Gliner et al. 1980; Horvath et al. 1979).



     A special focus of our work is the pulmonary functional response to 03 in

combination with exercise.  Another focus is the habituation response to 03

(and individual differences in that response).  Finally, recent studies

examine some effects of 03 on the central nervous system.



PULMONARY EFFECTS OF OZONE IN COMBINATION WITH EXERCISE



     Exercise at a fixed level of ventilation had dramatic effects on

pulmonary function following 03 exposure.  At 0.5 ppm 03, we observed marked


                                     378

-------
37.  PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE       	Horvath






pulmonary function decrements immediately following 15 min of exercise




(regardless of the time at which the exercise occurred).  Variations in the




amount of decrement  reflected individual sensitivities to 03.  Prolongation of




the exercise  to  30 or 60 min resulted in higher decrements.  In addition to




exercise, environmental temperature was implicated in the degree of pulmonary




function decrement.









     Other  studies examined the capacity for exercise following 03 exposure.




We observed a 10% decrement in maximal exercise capacity following a 2-h




exposure to 0.75 ppm 03 during which subjects exercised intermittently.  A




slight reduction in  the 03 concentration combined with omission of the




intermittent  exercise resulted in no decrement, however.








      In all such studies, it is very important to directly measure and control




the  subject's level  of ventilation.  Attempting to infer ventilation from




heart rate, etc. (as in some studies) is not adequate.  We performed various




03 exposures  (0, 0.1, 0.3, and 0.5 ppm) at various subject ventilation levels




 (10,  30, 50,  and 70  liters/min) and confirmed that forced vital capacity




 (FVC), forced expiratory volume (FEV^, and other parameters decrease markedly




and  consistently with the level of ventilation, and more so with the level of




inspired 03.








      Knowing  the subject's ventilation level during periods of exercise and




rest allowed  us  to  calculate the effective dose in each experimental trial.




Knowing the effective dose permitted us, in turn, to construct straightforward






                                      379

-------
37.  PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE	 Horvath






equations that predict pulmonary function decrement as a function of three




variables:  ventilation, concentration, and exposure.  We are confident of the




validity of these equations:  all are significant at p < 0.01.









PULMONARY HABITUATION TO OZONE IN COMBINATION WITH EXERCISE









     Habituation exposures were performed under a protocol similar to that




described by Haak (Chapter 31 of this volume).  However, our subjects




practiced for 2 weeks prior to exposure, allowing us to obtain base-line




pulmonary function measures.  Exposures to filtered air and 03 (0.2, 0.35, and




0.5 ppm) extended for 3 or 5 d.  Relative humidity was maintained at 45%, to




simulate the environment of a hot summer day in Los Angeles.








     Evaluations of several pulmonary function parameters (including FEV<| and




FVC) revealed a general pattern:  a maximum decrement on the first or second




day of  exposure followed by a return to near-normal on the third or fourth




day.  Despite this general pattern, however, individual subjects displayed




considerable variability in habituation.  For example, one subject showed a




dramatic fall in FEV^ on the first day of exposure but had fully recovered by




the second day; no further change was seen.  A less sensitive subject showed a




slight  change in FEVj on the first day of exposure but no decrement




thereafter.









     At present, we are following the highly sensitive subjects by having them




return  to the laboratory at various time intervals.  Our goals are to






                                     380

-------
37.  PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE	Horvath


determine the length of  time for which habituation persists and to identify

the mechanism by which habituation occurs.



PSYCHOPHYSIOLOGICAL STUDIES



     Studies of subjective,symptoms in O3-exposed subjects showed a marked

correlation between the  number of symptoms  and the environmental temperature.



     In  studies of vigilance task performance, O3~exposed subjects showing

significant pulmonary function decrement in the  first hour of exposure showed

a  gradual  decline in  the ability to detect  signals during the second hour of

exposure.   The most recent studies suggest  a correlation between vigilance

performance decrement and 03 concentration.



     Preliminary studies of 03 effects on the electroencephalogram (EEC)

suggest  a  decrement in the usual 8-10 Hz range.



REFERENCES
 Folinsbee, L. J., B. L. Drinkwater, J. F. Bedi, and S.  M.  Horvath.   1978.  The
      influence of exercise on the pulmonary function changes  due to  exposure
      to low concentrations of ozone.  In:  Environmental Stress:  Individual
      Human Adaptations (L. J. Folinsbee, J. A. Wagner,  J.  F.  Borgia, B. L.
      Drinkwater, J. A. Gliner, and J. F. Bedi, eds.).  Academic Press, New
      York, pp. 125-145.

 Folinsbee, L. J., J. F. Bedi, and S. M. Horvath.  1980.  Respiratory responses
      in humans repeatedly exposed to low concentrations of ozone.  Am. Rev.
      Resp. Dis., 121:431-439.
                                      381

-------
37.  PULMONARY AND PSYCHOPHYSIOLOGICAL EFFECTS OF OZONE          	Horvath
Gliner, J. A., S. M. Horvath, R. A. Sorich, and J. Hanley.  1980.  Psychomotor
     performance during ozone exposure:  Spectral and discriminant function
     analysis of EEG.  Aviat. Space Environ. Med., 51:344-351.

Horvath, S. M., J. A. Gliner, and J. A. Matsen-Twisdale.  1979.  Pulmonary
     function and maximum exercise response following acute ozone exposure.
     Aviat. Space Environ. Med., 50:901-905.
WORKSHOP COMMENTARY
R. K. Wolff;  Regarding the temperature effects, does the subject's
ventilation go up at a higher temperature?  Why do you think temperature has
that effect?

S. M. Horvath;  Body temperature goes up.  At a higher temperature ventilation
increases, so we lower the work rate in order to maintain the ventilation
constant.  In our view, to look at the effects of 03 on the pulmonary
functions of man requires at least a constant ventilation.  If you don't have
a constant ventilation, all you're looking at is variability.  As a
consequence, there is less variability in our obtained data simply because we
spend so much time looking at and measuring the ventilation.  We know the
maximum work capacity of each subject and can predict within 5% what level of
work will maintain a certain ventilation.  We continually monitor the
ventilation and lower or raise the work rate to maintain the ventilation.

Question;  Does your "effective dose" refer to absorbed 03 or inspired 03?

S. M. Horvath;  Inspired 03.  We were originally interested in measuring how
much 03 is really taken up.  The exposure level does not tell you what amount
actually gets into the lungs.  If the Air Resources Board can provide the
funding, we will obtain that information.  In the meantime, we still don't
know how much actual 03 is taken up by the individual.  We do know how much is
presented, and the prediction equations are based entirely on that.
                                     382

-------
                           38.   EPIDBMIOLOGIC  STUDIES
              OF OXIDANT  HEALTH EFFECTS  IN THE LOS ANGELES AREA

                                  Roger Detels

                            School of Public Health
                            University of California
                            Los Angeles, CA   90024
INTRODUCTION



     Since  1972, the University  of California - Los Angeles (UCLA) and the Los

Angeles County Lung Association  have collaborated on studies of the

relationship between chronic  exposure to various types and levels of air

pollution and lung function test performance.  The primary objective of these

studies is  to determine  if changes in lung function test performance over time

can be correlated to residence in areas exposed to high and moderate levels of

photochemical oxidants.  No previously reported studies have attempted a

similar correlation in a large,  free-living population; thus, our project may

be definitive in establishing the magnitude of such correlation (if any).

Coinvestigators for the  project  include this author as well as Dr. Stanley

Rokaw, Dr.  Frank Massey, Dr.  Donald Tashkin, and Mrs. Anne Colson.



     To date, we have completed  base-line testing in four areas and retesting

in one of those four areas.   This report presents some preliminary results of
                                     383

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES	 Detels






a cross-sectional study completed for the years 1972-1977.  There are many




problems associated with the cross-sectional study design; these can be




resolved, in part, by completing longitudinal follow-up of the cohorts already




established.









STUDY SITES








     Four study sites were selected to provide a range of pollution exposures.




Glendora, located in the East San Gabriel Valley, is exposed to high levels of




photochemical oxidants.  Burbank, in the East San Fernando Valley, is exposed




to moderate levels of photochemical oxidants.  Lancaster, located two mountain




ranges north of downtown Los Angeles, is exposed to low levels of all ambient




air pollutants.  The Long Beach site is exposed to low levels of photochemical




oxidants but relatively high levels of sulfur dioxide, particulates, and




(presumably) hydrocarbons.








     These four sites were also selected for contiguity to monitoring stations




of the Southern California Air Quality Management District, which can provide




continuous measurements of levels of selected air pollutants.  Monitoring




stations are located immediately adjacent to or within the four study sites




with the exception of Glendora, where the monitoring station at Azusa (~2-3 km




upwind of the study site) is used to estimate exposures.









     The monitoring data we compared consisted of average annual means of




daily maximum hourly average concentrations of selected air pollutants over






                                     384

-------
38.  EPIDEMIOLOGIC STUDIES  IN LOS ANGELES	  	Detels







the six study years.   The Azusa station had the highest  levels of




photochemical oxidants.   Lancaster showed oxidant levels that were higher than




originally expected but  lower than the levels in Burbank and Azusa.  Long




Beach had the lowest  levels of oxidants but the highest  levels of sulfur




dioxide.  Nitrogen dioxide  levels were highest for Long  Beach and Burbank.




Particulates were high in Azusa.  Particulates were not  measured in Long




Beach, but isopleths  indicate that Long Beach had the  highest particulate




levels.  Hydrocarbon  measurements were a problem because they were not made in




Long Beach until 1977.  We  had hoped that one result of  the planned joint




ARB/EPA studies would be better measurements of hydrocarbons and particulates.









PULMONARY FUNCTION TESTING









Subject Recruitment








     At the  beginning of each test period, articles concerning the research




program appeared in  the local newspapers, and letters  of introduction were




sent to residents of the target areas sequentially so  as to permit lung




function  testing within one week of notification.  Telephone contacts and




construction of household rosters were done by individuals recruited from the




communities  in which the testing was being carried out.








     The  percentage of households enumerated ranged from 84%  in  Lancaster to




98% in Glendora.  The proportion of residents completing lung function testing




ranged from 70% in Burbank to 79% in Lancaster; the actual number  completing






                                      385

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES	Petals






lung function testing ranged from 3403 in Glendora to 4509 in Lancaster.  An




additional 1 to 8% of residents of the four areas completed a respiratory




questionnaire only.  The demographic characteristics of the nonrespondents




were similar among the four study areas.









     The prevalence of residents with a history of asthma, bronchitis, or




emphysema ranged from 10 to 13% in the four communities and was highest in




Lancaster, the study area exposed to the lowest levels of pollutants.








     Lung function testing was completed only on residents who were seven




years of age and older.  We felt that children younger than seven years of age




would not be able to follow the procedures.  Also, it was necessary that each




subject be able to walk up the steps to the mobile lung research laboratory.




Individuals who were so infirm as to not be able to walk up to the laboratory




could not be tested, although we were very willing to assist them if possible.









Testing Protocol









     Prior to pulmonary function testing, subjects underwent an interview




schedule administered in a separate trailer.  Lung function tests were




administered in a mobile lung research laboratory.  These tests included:




forced expiratory spirometry with computer recording of the entire flow volume




curve; the single breath nitrogen test, including the change in level of




nitrogen between 750 and 1250 cm3 of expired air as well as the closing volume




fraction; and body plethysmography, including determination of thoracic gas






                                     386

-------
38.  EPIDEMIOLOGIC STUDIES  IN LOS  ANGELES	Detels






volume, functional residual capacity,  and specific airway resistance.  (We




were quite surprised at the very low refusal  rate for plethysmography:  <1%.)




Each spirometric  flow volume curve was recorded directly onto tape so that we




could  select  best breath, best  two breaths, etc. using computer algorithms.




The results of  the other two tests were recorded by hand.








Quality Assurance









     Several  strategies were used to determine the reliability of the lung




function  testing:  (1) every tenth participant was immediately retested;  (2)




100 residents of  each area were retested three times over the course of a




year;  (3) a  3% probability sample of individuals completing lung function




testing was  reexamined using more intensive techniques at the UCLA pulmonary




function  laboratories.  In addition, instruments were calibrated and




recalibrated several times during each test day.  Reliability for the




spirometric  indices  was excellent; for the plethysmographic indices—fair; and




for the single breath nitrogen indices—poor.









PRELIMINARY RESULTS








     We have completed a preliminary analysis of the  following parameters:




cough  and sputum, any symptom,  forced expiratory volume  (FEV,), peak flow, and




closing volume fraction.  In order to minimize the number of variables between




study  groups, the results presented here are limited  to  white residents who




did not have Spanish surnames and who had never changed  residence or






                                      387

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES	Detels






occupation because of a respiratory problem.  Further, these results are only




for individuals who were either current smokers or who had never smoked, and




includes only individuals who were 25 to 59 years of age.  Analyses of results




for younger subjects, older subjects, and former smokers are ongoing.









     Among residents who had never smoked, the worst test values were most




frequent in Glendora, the area exposed to the highest levels of photochemical




oxidants.  The frequency of worst test values among participants from Burbank




(the area exposed to moderate levels of oxidants) was low and similar to the




frequency among residents of Lancaster (the area exposed to low levels of all




pollutants).  The second highest frequency of worst test values occurred among




residents of Long Beach (the area exposed to relatively high levels of sulfur




dioxide, particulates, and possibly hydrocarbons).









     Among current smokers, the distribution of worst test values was similar




to the distribution among never-smokers.  Two of the tests that presumably




measure predominantly small airway function (AN27E-n_i9C-n and the maximum flow




at low lung volumes) correlated with smoking but not with residence.




Differences in results (any test) between the best and the worst study areas




were greater than differences (same test) between never-smokers and current




smokers.









     The results of the spirometric tests in the Burbank participants were




closer to those in the Lancaster participants than in the Glendora




participants, suggesting the possibility of a nonlinear dose-response curve or






                                     388

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES	Detels







a threshold value for oxidants lying somewhere between the levels occurring in




Burbank and the  levels occurring in Glendora.   Another possible explanation is




that more affected residents of the Burbank study area had moved away before




the program started.   This possibility will be resolved by longitudinal




follow-up of  the established cohorts.









      In  summary, the highest frequency of worst test values occurred among




participants  in the  area exposed to the highest levels of photochemical




oxidants, and the  lowest frequency of worst test values occurred among




participants  in the  area exposed to the lowest levels of all ambient air




pollutants.









 DISCUSSION








      These  results should not be viewed as conclusive evidence of a




 relationship between chronic exposure to photochemical oxidants and impairment




 of lung function.   Although most of the biases we have been able to identify




 would tend to decrease observed differences in comparison  to true differences




 between the communities, cross-sectional studies cannot account for all of the




 variables which may confound these comparisons.  Thus,  it  is essential that we




 pursue the  original objective of these studies:  to observe whether changes in




 lung function test performance are commensurate with the  levels of the various




 pollutants  measured concurrently over the five-year test  interval.
                                      389

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES                               Detels
WORKSHOP COMMENTARY
Comment;  Case-study epidemiology is always a tough thing, and you left
yourself a good caveat.  Still, with regard to oxidants, should you be
measuring people where they live or where they work?  Shouldn't you be looking
at people who work outdoors—telephone linemen, ditch diggers, and so on—to
get some idea of what exposure to those ambient concentrations means?  Spizer
and Ferris and others have found very large differences between ambient levels
of photochemical oxidants and levels of photochemical oxidants indoors where,
unfortunately, most of us spend our busy days.

     And how about the correlation with income that is seen in the
epidemiology of cancer in large cities (high income, low cancer)?  Does this
reflect the fact that people who don't have much money always live down by the
bayou, down in the valley, next to the railroad tracks, close to the factory?
It's never clear to me which is cause and which is effect, but this is
something for you to consider here.  I suppose that you have considered it.

R. Detels;  Yes, we have.  First, with regard to socioeconomic status, I'm
afraid I didn't make it clear that we matched these communities on mean
income, on cost of housing, and on racial distribution.  That's why they are
predominantly white.  If we had taken any other kind of a mix it would have
been impossible to match them and have different air pollution exposures.

     With regard to place of residence vs. place of employment, we took down
histories of where they live and where they work, how much time is spent in
commuting, etc., and we'll be looking at that.  However, Lancaster is ~60-70
mi from downtown Los Angeles.  Therefore, the majority of residents do not
make that commute.  The same is not true for the people from Long Beach,
Burbank, and Glendora, which are the three polluted areas.  People from
Glendora only have "one way to go," and that's to reduce their pollutant
exposure by commuting.  Therefore, it is our impression that the biases that
would be introduced by commuting to an area of different pollutant exposure
would tend to diminish the observed differences from the true differences.
However, we will be looking at subgroups on the basis of commuting patterns.

     We also take occupation into consideration.  We know that the
distribution of occupations is similar in these four communities.

Question;  What do ambient values have to do with the inside of a Lockheed or
Douglas plant?

R. Detels;  I can't answer that.  But I think that if one sees differences in
breathing parameters between communities, then one has to hypothesize
differences—occupational or other exposures—if one is unwilling to accept
residence as the important variable.  We found no differences in occupations
that could account for these differences.
                                     390

-------
38.  EPIDEMIOLOGIC STUDIES  IN LOS  ANGELES	Detels


     We haven't  looked at the difference between .indoor and outdoor exposures.
Frankly, that's  difficult in a  residential  type of study.  The object of this
study is to look at whether breathing ability  is associated with air pollution
levels in the area of  residence.   We realize that this objective has some
shortcomings; if we see no  difference then  we  can't say much.  However, if we
see a difference, the  argument  against a residential effect is a little harder
to make.

Question;  Would you explain the kinds of exposure data that were available to
you?

R. Detels;  Data were  available for a seven-year period from continuously
operating stations of  the Southern California  Air Quality Management District.

Question;  And you had access to the exposure  data from that time up to the
current time?  At what frequency?

R. Detels;  The  Southern California Air Quality Management District's
monitoring is continuous.  The  Air Resources Board and the Southern California
Air Quality Management District have kindly provided us with their measurement
ti»>es so that we can  select any parameter we want.  We picked the mean of the
daily maximum.   However, one can  select other  parameters.  Dr. Yuji Horie of
the Technical Services Division has looked  at  exposure in terms of the number
of days over the air  quality standards for  these areas.  That is another way
of trying to document pollution differences.   Using that technique, the
differences are  even more dramatic.  We have data dating back to the early
1960's.

Question;  Have  you done any analysis in terms of length of residence?

R. Detels;  We have that information.  We deliberately picked communities
having a high proportion of individuals who had lived there five years
previously.  We  did not divide  it into individual years of exposure in that
community.

     When you stop  to think about it, if somebody has lived in Burbank for 5
years but spent  the previous 20 years in downtown New York City, how does he
compare to somebody who has spent 5 years in Burbank and the previous 20 years
in Apple Valley, California?  I don't know. We have the information, but we
don't know how to get at that.

Question;  For your follow-up,  do you plan  to  repeat all the same
measurements?  We have a concern,  for example, in terms of the single breath
nitrogen test.   You indicated that there was very poor reliability.

R. Detels:  That's  correct.  First of all,  it's a gratuitous measurement.  The
instrumentation  is  on the test unit.  To take  it requires an extra 3-4 min of
the participant's time and little else. Second of all, we did see •"
association with closing volume in the four areas and we dxd see a correlation


                                      391

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES                   	Detels
with cigarette smoking.  If we're going to look at these tests to see whether
they're reasonable predictors (one of the subobjectives of this study), it's
important to see that the test itself is consistent in an individual over a
five-year course.  If we find somebody with a low closing volume (which we
think may be an early sign of respiratory disease) in 1973 and then we find
him with a decreased FEV-| but a good closing volume five years later, I would
question whether the closing volume is a very good predictor.  I would think
that the closing volume measurement must at least be consistent with what we
saw in 1973.

     If we had to go out and buy all of that instrumentation from scratch in
order to do the retesting, there would be a very reasonable question raised.
But since it's gratuitous and we can get that consistency estimate, I think we
would be somewhat remiss not to do it.

Comment;  That's my point.  It may be consistent but it may not be reliable.

R. Detels;  In my opinion, in order to be consistent it must be reliable.  It
may not be sensitive, it may not be specific, but it must be reliable.

F. J. Miller;  What percent of response did you get in your follow-up?

R. Detels;  This is one of our urgent concerns.  In the follow-up for Burbank,
we can account for ~84% of the cohorts.  For ~16% we have demographic
information and base-line test results but no retest information.  Our problem
is that people moved out of Burbank.  Unfortunately, we can't bring the mobile
laboratory to these people because they are scattered over a wide area.
Therefore, we have to ask them to drive several hours, if they live in
Southern California, to be retested.  Retesting of individuals who have moved
further away is completely unfeasible.  Most of our refusals were, in fact,
from people who had moved.

     To date, we have completed ~75% of the testing in Lancaster.  At this
point, it appears that the response rate will be better than in Burbank.

     The response rate in the remaining two areas will be better because we
initiated the annual follow-up about a year and a half after the original
Lancaster cohort was done.  Unfortunately, for two years we had no follow-up
in Burbank; this may be part of our problem.  I want to emphasize that, the
longer we delay, the more people will move out and the lower the response rate
will be.

F. J. Miller;  Are they moving from the area or are they just not responding,
to avoid the bother of testing?  Also, you can minimize your differences by
the fact that they choose to move out of the polluted area.

R. Detels;  We send a letter to each individual who has moved away.   If there
is no response, we send a second letter.  Finally, we telephone the
individual.  One of the major questions we ask is:  "Why did you move?"  So,
                                     392

-------
38.  EPIDEMIC-LOGIC  STUDIES  IN LOS ANGELES     	______^_ Detels


if they moved  for respiratory reasons,  we have that  information.  In addition,
we will compare the lung function test  parameters  at base line for those who
moved out vs.  those who remained.  We want to determine if there are
differences between these two groups.  We did that using FEV1? there were no
large differences.   We  are  also looking at the demographic characteristics,
since we have  complete  questionnaires on all  subjects.

     Thus, we  have  better documentation of the characteristics of our retest
nonrespondents than most studies do.  We know a  lot  about their respiratory
physiology at  base  line, and we know  a  lot about their  demographic
characteristics.

Comment;  Are  there sufficient numbers  of people who move from an area because
of respiratory problems to  make it worthwhile to do  a correlation of their
response?  In  other words,  let's assume that  certain people can adapt to
photochemicals in the area  and others cannot.  Those who cannot adapt, move;
they would be  a subgroup that might be  very interesting to study.

R. Detels;  That's  true. As long as  they move within the general Southern
California area, we have that potential.  We  could either bring the laboratory
to them or bring them to the laboratory.

W. Frietsch;   Could you briefly describe this study's origin?  What were the
major objectives?

R. Detels;  There were  two  reasons the  study  came  about.  First, I moved into
the Los Angeles area about  1971 and decided that if  I was going to breathe it,
I might as well study it.  About that time, the  National Heart Lung Blood
Institute issued a  request  for proposal for a population study of areas
exposed to air pollution.  We responded with  a proposal to look for the
changes in lung function parameters in  groups of people exposed to different
levels of photochemical oxidants and  other pollutants over a five-year period.
That remains the major  objective.  Now, we have  a.  whole host of subobjectives.
But I want to  emphasize that it was not designed as  a cross-sectional study.
We were well aware  of the pitfalls of the cross-sectional study.  Still, that
remains the best way to form a population-based  cohort.

F. J. Miller;  From all the caveats you have  mentioned  (in terms of analyses
that could be  done, etc.),  my general reaction is  that  there are a couple of
years of analyses to perform before you can really consider an expanded study
or something on the order of "goals."  It appears  that  you have a tremendous
amount of data that still need to be  interpreted and put in perspective for an
effective study.

R. Detels;  You're  absolutely right.  Given the  best of all possible worlds, I
would have tested all these communities at the same  time.  Then I would have
"put them in the icebox" and said, "Don't anybody  move; it's going to take me
two years to go through the analyses  I'd like to do."  Such analyses would not
be associated  with  the  ultimate objective, and none  would overcome the


                                      393

-------
38.  EPIDEMIOLOGIC STUDIES IN LOS ANGELES            	               Petals
problems of a cross-sectional study.  But such analyses would involve very
interesting cross-sectional studies:  comparisons of different areas,
comparisons of different tests, comparisons of different age groups,
comparisons of different analytical strategies, and so on.  The problem is
that subjects continue to move.  The longer we delay the fewer people will be
available for retesting.

     The second problem is that two measurements taken over five years and
used to estimate mean annual change cannot be compared to a mean estimated
from two measurements taken over seven years.  The actual rate of change in
lung function parameters may not be constant over time.  Such a comparison
assumes that the rate of change is linear.  I'm almost sure that it is in fact
not linear.  For instance, one of the groups in which we're most interested
are those who are progressing from young adulthood to adulthood.  Their
changes in lung function are certainly not linear.  And I'm sure there are
other groups in which the changes are not linear.

     Limited funding has forced us to make compromises.  When cuts were
necessary, those cuts were made in analysis.  You cannot cut the collection of
field data.  Without collection of data there is no study.

     We have an opportunity to do all the analyses, but if we do only the
analyses we'll lose the cohort.  If we lose the cohort, we're stuck with
another cross-sectional study and all of its attendant problems.  So, although
I would prefer to stop and do two or three years of thorough analyses, I don't
think we have that option.  We did not have that option in the past because we
had to focus on the collection of quality data in which we could have
confidence.  Had we diverted the funding from collection of data to analyses,
I'm not sure that we would have obtained anything worthwhile to analyze.
                                     394

-------
          39.  PROPOSED SCIENTIFIC PROGRAM OF THE  COOPERATIVE STUDY
              OF  OXIDANT HEALTH EFFECTS IN THE LOS ANGELES AREA

                               Stanley V.  Dawson

                               Research Division
                         California Air Resources Board
                                 1102  Q Street
                              Post Office  Box  2815
                             Sacramento, CA 95812
INTRODUCTION



     This report describes a program to further our knowledge of oxidant

effects on human health in the Los  Angeles area.  The primary aim is to reduce

uncertainty about health implications of ambient air quality standards.

Funded jointly by the California Air Resources Board (ARB) and EPA, this

program consists of  an integrated set of human and animal studies to be

conducted cooperatively over the next five years.  Table 39-1 summarizes these

various studies.   Most of the initial work will use established methods, but

innovative approaches will be considered in  the program's evolution.



EPIDEMIOLOGIC STUDIES



     The key question of how variously polluted atmospheres relate to measured

health effects in populations is being addressed in a large population study
                                      395

-------
                                   TABLE 39-1.  COMPONENT INVESTIGATIONS
                                                                                            U)
                                                                                            vo
      Atmosphere
      Group Studied
        End Point(s)
                                                                   Investigator,
                                                                    Institution
                                                                                                                      8
ambient
0.3 ppm 03
pollutant mixtures
(oxidant + sulfurous)

0.5 ppm 03
2.0 ppm 03

0.2-1.2 ppm 03
0.2-0.8 ppm 03
census tracts (4)
emergency room walk-in's
asthmatics

human sample
exercising rats
exercising dogs

young dogs
young rats
young monkeys

monkeys
lung function, cruestionnaire
chest complaints
medication, lung function

lung function
clearance + morphometry
lung function + morphometry

terminal lung function
   + morphometry

extensive morphometrv
   (ongoing function)

macrophage motility
   (ongoing structure)
                                                                                       R. Detels
                                                                                       UCLA
J. Hackney
Rancho Los Amigo Hospital

T. Crocker
UCI

R. Phelen
UCI

W. Tyler
UCD

L. Schwartz
UCD
                                                                                                                      M
                                                                S
                                                                H
                                                                z

-------
39.  OXIDANT HEALTH EFFECTS  IN LOS ANGELES
by R. Detels and associates  at  the University of California - Los Angeles




(UCLA)  (see Chapter  38  of  this  volume).   In this study, one full round of




pulmonary function testing and  questionnaires has been completed in four




census  tracts  (4000-6000 population  each, aged 7 and up) located in variously




polluted regions in  the Los  Angeles  area.  Follow-up studies of individuals




originally tested have  been  completed (84% retention rate) in Burbank (high




oxidant, medium sulfur  dioxide  (802))  and nearly completed in Lancaster (low




oxidant, lowest 802).   Follow-up studies  will begin in Long Beach (lowest




oxidant, highest 802) in  spring 1980 and  in Glendora (highest oxidant, medium




SO2)  in 1981.









      The initial part of  this study, a cross section of the four census




tracts, has already  yielded  some useful results.  Pulmonary function test




results were more below normal  in the more "smoggy" areas, but the




relationship between test results and pollutants was not simple.  Because of




the potential  for different  population conditions in different areas, such




cross-sectional studies are  thought  to be of limited validity in establishing




the relationship between  mixed  pollutants and health effects.  The serial or




follow-up study of pulmonary function decrement in given individuals over time




will  permit a  more direct assessment of pollutant effects on populations in




the four tracts.  Preliminary results of  the follow-up study show greater




pulmonary function decrement in Burbank than in Lancaster, an area of




substantially  less air  pollution of  every measured species.
                                      397

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	Dawson






     The UCLA group plans two additional smaller-scale epidemiologic studies.




Tentatively, the first will involve a reporting network for chest complaints




by emergency room "walk-in's."  The other will be a comparative asthma panel




study in districts that have distinctly different oxidant levels but are




matched in other ways.  The use of medication dispensers that can provide




assured counts and the use of quantitative self-testing will be explored.









THE INTEGRATED PROGRAM








     In developing an integrated program that starts with these epidemiologic




studies, we sought projects that might answer questions not addressed in




purely epidemiologic studies, and we sought to add new exposure information to




sharpen the epidemiologic studies themselves.  One limitation of epidemiologic




studies is the lack of confidence that one can place on the effect of a




particular level of a particular pollutant, while a strength of the Betels




cohort study is the determination of actual rate of lung function loss.  An




exposure chamber study, on the other hand, can provide a clear response to a




particular level of pollutant, but the long-term significance of that




immediate response is open to question.  Our integrated program proposes to




directly link these two types of studies by drawing samples for an exposure




study from Detels1 well characterized epidemiologic cohorts subsequent to




follow-up tests of pulmonary function.









     As a further aid in interpreting the epidemiologic studies, four animal




studies are planned.  Using rats and dogs under carefully controlled exposure






                                     398

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	Dawson






conditions, the  first study will examine the lung damage and impairment




attributable to  two important environmental influences:   exercise and




pollutant mixtures (both key factors in Los Angeles).   The second and third




animal  studies will address questions for which the Detels data are not yet




analyzed:   namely, effects on the very young.  Experiments on dogs, rats, and




monkeys will  examine  the extent of lung damage due to  ozone (03) exposure




during  the  development of the lung, with special reference to the stage of




alveoli formation. Lung damage will be assessed morphometrically in all three




of the  animal studies—a difficult approach in humans, although a comparative




post-mortem study on human adults is a future possibility.  The fourth animal




study will  explore a mechanism that is considered to be a likely key to the




accelerated loss of lung function that may occur in polluted regions of Los




Angeles as  well as to the damage observed in terminal  bronchioles of animals




exposed to as little as 0.2 ppm O3.  The hypothesis is that low levels of O3




 impair alveolar macrophage motility and that stagnant  macrophages release




 enzymes which attack the unciliated bronchiolar epithelium.  The approach of




 the study is to lavage (wash) and then measure the motility of macrophages




 from certain lung segments of O3-exposed monkeys.  The lavage technique




 carries future potential for assessment of the human lung condition relative




 to pollution exposure.
                                       399

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	     Dawson






IMPROVEMENT OF POPULATION EXPOSURE ESTIMATES









     Some program resources have been set aside for exploring ways to improve




exposure estimates in the Detels study.  This work will be conducted early in




the overall program.









     The first aspect of this project is to ensure that the basic data from




the monitoring stations are as comprehensive and complete as possible.  A few




standard pollutants were not actually measured during the study,  though all




standard pollutants are now scheduled for routine monitoring in the two




remaining tracts.  In addition, there is an important opportunity to measure




fine particles in all four stations near the tracts.








     This effort involves two further aspects.  One is the question of how the




actual exposures of individual subjects might have differed from the values




reported at the monitoring stations.  Measurements and/or modeling of local




sources and indoor variations may be helpful in answering this common




epidemiologic question.  The other question is of how individuals may have




been temporarily affected by pollutants at the time of the pulmonary function




measurements.









HUMAN LABORATORY EXPOSURES









     The primary exposure of subjects from epidemiologic cohorts will be




performed by J. D. Hackney and associates at Rancho Los Amigos Hospital in






                                     400

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES
Downey, California.   Fifty subjects from the Burbank cohort will  be  exposed to




0.3 ppm 03 for ~2  h.   The preliminary design calls for half of  these subjects




to be drawn  from the  group that experienced the most pulmonary  function loss




and for the  other  half to be drawn from the group that experienced the least




loss.  Testing of  the immediate response to 03 will determine the significance




of any correlation with loss of function over five years.









     Should  a significant positive correlation between long-term  pulmonary




function  loss and  short-term O3 sensitivity be found, two  contrasting




explanations could be offered.  One explanation would be that lungs  which are




a priori  most responsive to O3 insult in the short term are also  most




susceptible  to  long-term function loss.  The other would be that  existing




pulmonary function loss predisposes the lung to short-term O3 sensitivity.




Either  explanation would carry importance in standard-setting,  and future




studies  (e.g.,  similar studies of groups with less O3 exposure  or longitudinal




studies of individuals with high and low short-term 03 sensitivity)  might be




able to distinguish between them.  Related studies by Hackney and others




support the  possibility of detecting relatively small but  significant O3




responses and intergroup differences in response.  However, the likelihood of




successfully detecting differences between groups showing  different  long-term




pulmonary function loss is difficult to predict.  Even if  the attempt to




distinguish  between groups is not successful, considerable additional




information  on responses to "worst case ambient" 03 levels will have been




obtained  on  a group of Southern California residents who,  though  still
                                      401

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	     Dawson






self-selected, bear a well defined relationship to a typical middle-income




community population.









     Because of the phenomenon of desensitization or adaptation to 03




exposure, it seems likely that the clearest test of immediate 03




responsiveness of these subjects would be performed relative to an external




atmosphere that is as free of pollutant as possible.  Thus, special




precautions will be taken to reduce effects of environmental background:  we




intend to work with our subjects during seasons of low smog and to estimate




their environmental pollutant exposures.  In subsequent years, it may be




possible to arrange for a portable exposure chamber for the two most distant




census tracts.  Ultimately, it may still be worth transporting subjects to a




cleaner environment (e.g., Santa Barbara) for several days of tests.  General




animal studies of the phenomenon of desensitization and of the correlation




between immediate 03 responsiveness and rate of pulmonary function loss during




chronic exposure would provide useful correlates to these human studies.









     Subjects who participate in the human laboratory exposures could be




measured in various new ways with no additional inconvenience to them.  A




proportion of program resources support pilot studies of innovative ideas that




appear likely to yield results.  With A. Richters at the University of




Southern California we are considering exploratory studies on the blood




samples that will (in any case) be obtained.  Leukocyte population counts and




leukocyte function tests of both T cells and B cells might reveal differences




between groups.  In addition, with J. F. Mead at UCLA we are considering






                                     402

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	__	Dawson


exploratory studies of  exhalate  to measure pentane concentration as a marker

of peroxidation in the  lung lining.



ANIMAL LABORATORY STUDIES



     A study to examine the effect of  exercise on dogs and rats exposed to

pollutant mixtures for  short  periods is planned by T. Crocker and associates

at the University of  California  - Irvine  (UCI).  That laboratory has developed

exposure chambers which permit careful control of the aging of atmospheric

pollutant mixtures.   In the initial tests, the atmospheres will consist of:



      1.  Toxic hydrocarbons produced by photochemical reactions,
         including formaldehyde, acrolein, or peroxyacetyl nitrate
         alone at 0.1-2 ppm

      2.  03 alone at  0.2-0.6  ppm

      3.  Particulate  ammonium sulfate  at  1 mg/m3 plus SO2 at 0.5-5 ppm

      4.  Combinations of (1.) and  (2.) and of  (2.) and (3.)



      The hydrocarbons and SO2 are  water  soluble; hence, combined

hydrocarbon-particulate or SO2-particulate atmospheres will be studied at low

and  high humidities.   Beagle  dogs  will be exercised  on a refrigerated

treadmill; pulmonary  function will be monitored  by face mask, esophageal

balloon, expired gas, and carotid  blood gases.   Rats will also be exercised;

the  post-exposure ability to  clear inhaled particles from the lung will be

measured.  Detailed lung morphometric measurements will  include

characterization of lesions and of cell  replication.


                                      403

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	Dawson






     Studies of lung development are planned initially for 03 exposures only.




R. Phelan at UCI will expose beagle puppies to 2.0,  0.5,  and 0 ppm O3 for 4




h/d for 5 d at an age of active development, based on a protocol successfully




developed for 1.0 ppm.  The facility for housing the dogs maintains an




exceptionally pure environment.  After development is complete, the lungs will




be subjected to pulmonary function tests and detailed morphometry using




special methods for preparing the microscopic sections.  Mean linear  intercept




will be measured.  W. Tyler at the University of California - Davis (UCD)




plans to augment and continue rat and monkey studies that are in progress.




The rat studies are both episodic (0.8 ppm 03 for 3  d during various  stages of




development) and chronic (1.2, 0.8, and 0.2 ppm).  Measurements by the most




advanced morphoraetric methods will include internal surface area, capillary




volume and surface area, and mean thickness of the air/blood barrier.  The




monkey studies are chronic, involve an exposure level of  0.8 ppm 03,  and start




at 6 mo of age? measures of pulmonary function and extensive morphometry will




be completed (as in the rats).









     L. Schwartz of UCD will perform lavage of alveolar macrophages on healthy




monkeys exposed to 0.2-0.8 ppm 03.  The macrophages produced by lavage will be




examined in a number of ways; the primary approach will be to measure




motility-  The response to lung lining material and chemotactants will receive




attention.  Data on macrophage structural alteration (obtained in ongoing




studies) will be compared to the observed functional changes.
                                     404

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES
                                                                        Dawson
OVERALL ASPECTS



     An unusual  feature of this program is the inclusion of a  statistical

working group both as an aid to investigators and as a management mechanism.

Nine distinguished statisticians from Stanford University, the University of

California -  Berkeley, and UCLA have agreed to join a Design and Evaluation

Committee coordinated by L. Breiman, an independent statistical consultant

based  in  Santa Monica.  Operating under very specific guidelines, the

Committee will be responsible for the following functions:
      1.   To critically review all experimental design for  statistical
          integrity and power, including design modification at the
          proposal stage and as necessary thereafter to ensure
          statistical integrity and power;

      2.   To critically review the statistical analysis of  all
          experimental data;

      3.   To actively assist in the analysis of all data;

      4.   To develop a continuing evaluation of the statistical
          implications of all results bearing on oxidant
          standard-setting.
      In addition, the ARB program management staff will be responsible not

 only for coordinating the investigations but also for synthesizing the

 results.  Some of this synthesis—especially that involving extrapolation of

 animal results to man—may employ mathematical modeling approaches as well as

 biological concepts.  The most useful overall contribution of the synthesis

 may be to obtain, in connection with inputs from the statisticians,  a clearer
                                      405

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES	v   	Dawson


definition of the degree of confidence that can be attached to statements of

oxidant effects at ambient levels.



WORKSHOP COMMENTARY
Question;  Could you outline the approximate resources that would go to each
of the studies?  Also, how did you arrive at that emphasis in terms of meeting
study objectives?

S. V. Dawson;  To answer the last part of your question, we looked at the
merits of each study and the relevance of each in combination with the others.
To answer the first part, with respect to [Table 39-1], the Schwartz, Tyler,
and Phelan studies are each funded in the range of $70,000 to $90,000 a year;
the Crocker study is funded at $200,000; the Hackney study is funded at
$150,000; and the Detels study is funded at $530,000.

Comment;  In my understanding, the original question to be answered by the
study concerned the health effects that occur in Los Angeles, since in Los
Angeles the standards (particularly for 03) are fairly routinely exceeded.
After all the negotiations, is that still the major objective of the study?
Does the design really focus in on that question?

S. V. Dawson;  I would certainly claim that we're looking for health effects
caused particularly by levels of 03 and oxidant in Los Angeles.  All of these
projects were screened for relevance to that question.  It's true that, in
terms of pollutant, four of the studies address only 03.  In another exposure
experiment, 03 is added on to other things (the sulfurous compounds and the
rest of the total oxidants).  We have had trouble, perhaps, distinguishing
between 03 effects and oxidant effects.  This is at least a step in that
direction.

Question;  Were the studies assembled in sequential order?  Did you really
look at the major objective and design a study to achieve that, or did you
more or less pick up pieces of different projects and assemble them?

S. V. Dawson;  From the statements of interest we received, we assembled those
pieces that were most relevant to our objective and that were of sufficient
scientific merit.  Several of the submissions were substantially modified; we
went pretty far in that direction to actually shape the thing but, as you
know, investigators have certain interests and you can only move them a bit.
To a certain extent, one has to take what one can get in the real world.

Question;  What documentation for your presentation exists now?
                                     406

-------
39.  OXIDANT HEALTH EFFECTS  IN  LOS  ANGELES
                                                                        Dawson
S. V. Dawson;  The basic  scientific program is outlined  just as presented
here.

J. A. Graham;  I have  three or four general questions.   First, I don't
understand the rationale  for doing two  studies on  the young until results are
available from at least one of them. Obviously, it's nice to work with a
number of animal species,  but I don't see  the  rationale  for doing all of this
work with O3 and no  work  with NO2,  for  example.  I don't see the rationale for
duplication without  prior data.

S. V. Dawson;  Your  point is well taken, but the investigators do have some
interesting results.  Phelan has some very interesting results with that 4%
shift.  Tyler has some very significant results on the alveolar-volume-to-
body-weight ratio, and his proposal is  to  do detailed morphometry to look at
that.  So, as far as rats and dogs go,  the investigators are producing some
very interesting results  on 03.  But the Committee is also very intrigued with
the monkey study.  The monkey lung is more like the human lung and,  if the
hypotheses about damage are right,  the  monkey  should be  very important for
looking beyond the rat and dog.

J. A. Graham;  I guess that should be included in  the more detailed proposal,
for the purpose of comparing results with  those from extensive studies in this
area by Dr. Freeman.

     My second comment concerns the 03  concentrations.   I don't see how
studies at 2 ppm or  even  1.2 ppm are going to  be highly  relatable to the rest
of the studies:  those are very high concentrations compared to ambient
levels.

S. V. Dawson;  First of all, Phelan is  the only one that has suggested 2 ppm.
I argued that you really  only need a couple of pairs of  animals at 2 ppm to
get sufficient power in your test.   This would indicate  the direction in which
the line is going, thereby giving you a better calibration for the very
difficult lower 03 level.

J. A. Graham;  That  assumes that the mechanism is  the same at the high
concentration.  In the modeling by Miller  [Chapter 29 of this volume], the
tissue dose of O3 was  found to be dependent upon concentration;  at higher
doses the O3 molecule  may hit the pulmonary tissue; at lower doses,  the
molecule may not hit the  tissue.

S. V. Dawson;  I absolutely agree with  you.

J. A. Graham;  My last question is:   When  you  refer to macrophage motility,
are you talking about  chemotaxis?

S. V. Dawson;  Yes.  This is a standard migration  test that Les Schwartz has
published.
                                      407

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES                              Dawson
J. A. Graham;  For 03 in the monkey, why should that parameter—chemotaxis—
be any more sensitive than all the parameters already studied and reported
relative to the macrophages?

S. V. Dawson;  Well, I don't think it's likely to be a more sensitive
parameter, but it ties to the mechanism of damage.  It tests the hypothesis
that the respiratory bronchiole becomes tied up by all of this.  One could
argue, perhaps, that finding an effect in that area at 0.2 ppm 03 supports
that hypothesis.

J. A. Graham;  Those studies aren't going to be done on humans.  In terms of
trying to correlate the epidemiologic, human clinical, and animal studies, I
don't see how macrophage chemotaxis directly relates to humans.

S. V. Dawson;  Well, the monkey is as close as we can get.  It's hard to do
that morphometry on the human.

P. E. Morrow;  In what species do you plan to do the pentane exhalation study?

S. V. Dawson:  Humans.

P. E. Morrow;  How do you propose to do that, inasmuch as the dietary
component (lipid diet) and also the level of oxidant are so significant in
what you would apparently measure?  The active level of oxidant breathed and
dietary component of lipid affect the breakdown.  How do you deal with that in
the human population?

S. V. Dawson;  When a subject is exposed to 03, there are really two different
aspects.  First, if the subject comes from a polluted area, does he produce
more pentane, reflecting more oxidant damage to his lungs?  Secondly, when
given 03, does he produce more pentane before or after the experiment?

Comment;  As I understand it, the use of pentane and isopentane and the
relationship between them can help overcome some of the issues of metabolic
production in the background.  Ethane has also been used.

Comment;  With pentane, the method of detection is much more precise.  The
ongoing levels have to be active? there1s no "memory" of what the individual
breathed yesterday.

Question;  There is some published work on pentane and ethane exploration with
presumably oxidizing cytotoxins but not respiratory pollutants.  Wouldn't it
be useful to have some animal work on this as well?

S. V. Dawson;  Yes, that's absolutely valid and we will explore it.  One of
the suggestions is that we might do such work at UCD.
                                     408

-------
39.  OXIDANT HEALTH EFFECTS  IN LOS  ANGELES	^^     Dawson


Question;  In the walk-in  chest  complaints study,  how do you plan to correct
for the self-selection bias? Secondly,  what kinds of objective measurements
do you plan to analyze for?   What are the end points?

Comment;  Of necessity,  we will  use the  somewhat gross end point of a
self-selected complaint.  The frequency  of such complaints in areas of widely
differing pollution levels and yet  subject to the  same barrage  of publicity
about the episodes was thought to be a way to look at real effect.  We have
already done some feasibility studies on emergency room visits; the study
could probably be more finely tuned by the use of  data on respiratory therapy
services or the medications  administered to people coining there.  (We exclude
nonrespiratory complaints.)   The concept is to set up a network and have it in
place so that, when acute  health advisory levels are  reached, the frequency of
visits during those periods  can  be  compared to other  seasons or years.

Question;  Where  [in  Table 39-1]  is the  leukocyte  function test to which you
referred?

S. V. Dawson;  I don't know. That's tentative;  we're just exploring it as a
pilot study that might possibly  be  funded.

A. P. Altshuller;  Could you or  Dr. Hackney expand somewhat on  the directions
the human experimental work  will take beyond what  has been done to date?

Comment;  As far as we know, all previous work has been on very highly
self-selected individuals—laboratory technicians,  college students,
etc.—that were nonrepresentative of general populations.  In our case, we
certainly don1t have  any means of forcing people from the general population
to go into the study, but  at least  we are sending  out letters asking the
general population, individual by individual,  to join our study.  That, as far
as I know, is completely new. Also, I think that  the 0.3 ppm level is well
worth getting additional data on.

A. P. Altshuller;  The study will be conducted at  that level only?

Comment;  Yes, that's the  plan at least  for the first year.

J. D. Hackney;  The point  I  would make is the following:  The epidemiologists
will present us with  some  subjects  who are asserted to be sensitive to
photochemical pollutants in  view of a more rapid decrement in forced
expiratory volume over five  years.   These subjects will be compared with
subjects showing almost  no decrement. It may be that the subjects showing a
very large decrement  are also sensitive  to O3 over a  short-term 2-h exposure;
such results would be exciting.   There would be at least two possible
explanations;   (1) the O3  somehow caused the rapid decrement; (2) having a
decrement is itself a predisposing  factor for sensitivity to short-term 03
exposure.  Either one of those explanations would  be  of interest.  That s the
positive side.
                                      409

-------
39.  OXIDANT HEALTH EFFECTS IN LOS ANGELES                              Dawson
     On the negative side, being able to do the study will depend on Dr.
Detels1 identification of these people.  He already has them but the question
is whether they will volunteer in sufficient numbers.

     Dr. Detels would do all the matching.  He would send us the subjects and
we would test them.  At the end, we'd try to sort it all out and see if there
was some correlation between epidemiologic decrement and short-term clinical
response.
                                     410

-------
                             40.  PANEL DISCUSSION

                           Timothy Crocker, Chairman
                              Aubrey P. Altshuller
                                 Donald C. Borg
                               Robert S. Chapman
                             William Frietsch,  III
                                 Marvin Goldman
                                 John R. Holmes
                                 Fred J. Miller
                             James L. Whittenberger
T. Crocker;   Of  the possible items of agenda for  this Panel Discussion, it is

my impression that most of those present favor  some sort of scientist-

administrator exchange.  In my view,  such an exchange is potentially very

fruitful  in  that we might receive some information about the character of

future scientific review and program  selection  activities.  In a certain

sense, we stand  at the beginning of a phase  of  investigative effort.  As

researchers,  we  all want this effort  to be relevant and meritorious, but we

need to understand the intended mechanism for judging the quality and

relevance of  our work.



     To begin, Dr.  John Holmes, Director of  the Research Section of the

California Air Resources Board (ARE),  will describe the cooperative agreement

with EPA  under which ARB will conduct scientific  review.
                                      411

-------
40.  PANEL DISCUSSION
J. R. Holmes;  I certainly concur with the point made at the outset of this




meeting:  It's very important, in selecting and managing research projects of




this kind, to bear in mind that we need scientific excellence and regulatory




relevance.  In developing a program, we must consider both aspects.









     In ARB, we have developed a joint committee consisting of members from




the ARB Research Advisory Committee as well as from an Air Quality Advisory




Committee of the California State Department of Health.  These individuals are




scientists who are also involved (either peripherally or directly) in making




recommendations that serve as a basis for air quality standards.  In my




opinion, the panel did an outstanding job in assembling a package of studies




that are both scientifically excellent and relevant to what we view as the




problems in setting an oxidant standard both nationally and in California.  At




this point, none of the individual studies are "cast in stone"; there is still




time to hone and polish and make them even more relevant and scientifically




sound.  The investigators and certainly ARB are very interested in any




comments from this group; we've received a number already.








     In my opinion, before seeing the project through to completion, we need




to  set up a system of scientific guidance that differs from the systems




currently used by EPA and ARB.  In the management portion of our proposal to




EPA, we outlined plans for a joint technical committee to consist of




representatives from the State of California (2), EPA  (2), the National




Laboratories (1), and the scientific community at large (2).  This committee




is  large enough to incorporate a number of different disciplines and yet  small






                                     412

-------
40.  PANEL DISCUSSION
enough to permit  efficient discussions and limited travel cost.   I think this




approach will work out well.   But I would appreciate any comments or




additional guidance on how technical review of the project should be carried



out.









W. Frietsch;  We  in EPA also have a review group that is looking  at this




project (as well  as others) strictly from a planning viewpoint [see Chapters 2



and 3 of this volume].









T. Crocker;  My understanding is that Dr. Paul Altshuller is involved with a




separate Oxidant  Research Committee within EPA's general approach to the




criteria air pollutants.  This Committee develops strategy plans and documents




and will be responsible, as I understand it, for a mission statement.  I may




be overstating  the level of supervision that the Committee will undertake,  but




I know that Dr. Altshuller has given some thought to potential developments




and to the level  of guidance that may be forthcoming from this Committee.




Would you care  to speak to this point, Dr. Altshuller?








A. P. Altshuller;  Well, perhaps a bit of explanation is in order.  During




congressional deliberations several years ago,  the problem of the relevancy




and responsiveness of EPA in-house research to the regulatory needs of the




Agency came to  a  head.  A fairly drastic change in the distribution of




resources was under consideration; the Deputy Administrator asked that we be




given a year to look into alternatives.  We proposed a pilot Research




Committee program; our proposal was accepted.  The pilot program was gradually






                                      413

-------
40.  PANEL DISCUSSION
expanded and today includes essentially all of EPA's research; the Oxidant




Research Committee is just 1 of 16 Committees covering air, water, radiation,




pesticides, etc.









     The individual Committees have levels of responsibility for resources




ranging from $10 million to $60 million.  The Oxidant Research Committee's




responsibility is in the ~$30 million range.  Obviously, the Committee cannot




(and was not designed to) perform in-depth peer review for each small segment




of a $30-million program.  Rather, the purpose of each Research Committee is




to interface between those responsible for research and those responsible for




regulatory development and enforcement.  Thus, each Committee is largely




composed of individuals with technical backgrounds:  personnel from EPA's




regulatory divisions, from the Office of Enforcement, from regional offices,




and from the Office of Research and Development (ORD).  Each Committee is




co-chaired by a representative from ORD (in the case of the Oxidant Research




Committee, Ken Berry) and a representative from the lead program office.  For




oxidant research, the lead program office is the Office of Air Quality




Planning and Standards.









     The Committees serve as a mechanism for expression of concerns about




scientific relevance, responsiveness, and the timeliness with which outputs




from various research programs become available.  The obvious question is:




How can we assure scientific quality?  In part, this depends on the competence




and the dedication of the technical personnel within EPA.  It also depends on




a peer review mechanism.  Dr. Steven Gage, Assistant Administrator for






                                     414

-------
40.  PANEL DISCUSSION
Research and Development,  has been very concerned about the peer review




mechanism for both  in-house and external research.  We  are revising the




research grant program to  return to a peer review procedure resembling that of




the National Institutes of Health (NIH) but which will  also include a




relevancy review.   At present, we don't have all the answers on how to




integrate these  two aspects, but both are vital.  Regardless of whether a




research program involves  pass-through money in another agency, agency




agreements, contracts, or  what have you, the same general question—how to




achieve a scientific product that will be accepted by the scientific community




and  also usable  in criteria and standards development—remains a vital issue.









Question;   Has any significant peer review mechanism been set up?








A. P.  Altshuller;  A specific mechanism for research grant peer review is




being set up.   It will consist of a group of four committees that will




periodically review grant proposals for both scientific acceptability and




relevancy.   Because of certain congressional and other  limitations on the




movement of funds, we have a number of "pigeonholes" where we have to "put the




money," so  to speak, and we will fill these up with proposals until the




funding limit is reached.   To do this, we'll work down  a list of projects




scored for  merit as well as relevancy.  In a sense, it's a tripartite system:




scientific  merit, relevancy, and the amount of money available for the





subcategory.








Comment;  You still haven't answered the question.






                                      415

-------
40.  PANEL DISCUSSION
A. P. Altshuller;  Would you care to enlarge on your question?









Question;  Aside from research grants, is any peer review system under




consideration?








T. Crocker;  Dr. Altshuller, I believe the question refers to the oxidant




program that is the subject of this workshop.  You're talking about an




extramural grant review program that may not be related to our particular




focus.








A. P. Altshuller;  Well, it would be related to all grants, including those




concerned with oxidant air pollutants.  We're just working on possible




mechanisms for review.  A question which does come up is:  To what extent can




the other body of expertise available to EPA on a continuing basis—namely,




the Science Advisory Board (SAB) and its subcommittees—participate?  Just




last week, we had some discussions with the Director of SAB, Richard Dowd; we




discussed some preliminary proposals to involve the Clean Air Scientific




Advisory Committee in more in-depth review of criteria pollutant research,




with particular emphasis on oxidants.  The question is:  Where do we go from




there, when we get down to the question of reviewing program components that




represent a million, half a million, or a quarter of a million dollars?  This




would demand large numbers of people who are willing to commit some




significant fraction of their time.  Very likely we would end up with well




over 100 individuals just in the area of health.  These  100 individuals would




have to be willing to commit a certain fraction of their time to interact in






                                     416

-------
40.  PANEL DISCUSSION
peer review of  a wide variety of projects in health, modeling, ecosystems,




control technology,  etc.   ftnd this has not been worked out.









Comment:   The details of  the review system have all been set down by Dr. Gage.




If you're interested in obtaining those details, I suggest that you contact




Dr. Marlin at EPA Headquarters.









A. P.  Altshuller;  Having talked quite a bit with Dr.  Marlin, I'm not sure




that the  array  of detailed information is really there.  The mechanics may be




there  for the research grants and for the cooperative  agreements.  As for the




various other categories, it is probably a little premature to say that we




have all  the  answers.








F. J.  Miller;  SAB will review the EPA in-house projects on March 20.  SAB




will also perform scientific review of the Department  of Energy (DOE)




transferred projects.  So mechanisms have been set up.  If your question was




whether the DOE-EPA projects will be reviewed, the answer is "Yes."









D. C.  Borg;  With regard to Dr. Altshuller's point, it does take a lot of




people to carry out peer  review (a la NIH, in any case).  What's more, as




thorough  and  reassuring as that sort of review may be,  there are certain




disadvantages.   It grinds with extreme slowness, to say the very least.  Other




agencies  (perhaps less well known to this audience) carry out peer review in a




somewhat  different fashion.  For years, the National Science Foundation (NSF)




has proceeded with peer review (in the physical sciences) much as do editors






                                      417

-------
40.  PANEL DISCUSSION
for peer review journals:  namely, asking for outside written opinions and




then presenting a consensus of that opinion to an internal review board.  In




the life sciences area, there are also panels.  Those who have served on the




NSP panels know that they don1t have as much time to spend on an individual




grant application as does an NIH study section.  The NSF panels perform




something of an intermediary function:  they are much more advisory than the




NIH study sections.  In the physical sciences, however, NSF has managed to do




a fairly creditable job without all of that.  Thus, I would remind EPA that




the NIH way is not the "only" way to achieve peer review.








A. P. Altshuller;  Yours is a rather interesting comment, for this reason:




The EPA research grant/cooperative agreement review system that was in place




from 1973 until 1979 was very similar to the NSF system.  Extramural reviewers




were asked to individually review each project and to submit written comments.




A single in-house scientific reviewer also assessed the project.  A judgment




was made by technical management within the individual Laboratory.  (Later,




that judgment added up to whether or not the project was funded.)  It's




precisely this sort of system that was criticized, resulting in a return to an




NIH-type system.









     One of the principle criticisms was that we were involving too few




investigators.  Critics complained that we tended to be satisfied—once we had




found a group of investigators who could do the job in a specific




discipline—to "ride along" with those investigators in other disciplines.




I'm not sure that was a valid criticism; in all honesty, I don't know where






                                     418

-------
40.  PANEL DISCUSSION
the additional  investigators would have been.  As. far as I could discover, we




had sufficient  resources to fund essentially all of the competent




investigators.   In some areas it may have been a specious criticism; in




others,  it may  have been a relevant criticism.









M. Goldman;   As someone who is not overly familiar with the EPA  system of




review,  I've still not received an answer for a question I jotted down on my




way to this  workshop:   How do you equitably separate the problem of relevancy




(which I assume to be an intra-agency one)  from that of scientific merit?  At




this  workshop,  I hear what seems to be a plea for an equitable peer review




system that  will take care of both the intra- and extramural programs




regardless of how they are handled administratively.   In other words, the




science—regardless of who" s doing it—should be equally good, and a mechanism




for assuring this must exist.  This relates,  perhaps,  to a mechanism to




provide  continuity of support for longer-term programs.   If high-quality




science  is to be attracted to these programs, the issues of peer review and of




how information will be delivered to the various clients are important parts




of the overall  spectrum.  From what I've heard here,  progress is being made




but the  system  is not yet complete.  -Phis business of "extramural versus




intramural program methods" leaves me and others with some confusion.








T. Crocker:   Your comments fairly accurately summarize the situation in terms




of the structure and plan for Agency supervision.  It's moving but it hasn't





jelled.
                                      419

-------
40.  PANEL DISCUSSION
     With respect to extramural grant review, Dr.  Whittenberger chairs a




special SAB Subcommittee that has been involved in [the redirection of Theme 1




of the Energy Health Effects program; see Chapter  3 of this volume].  That




body, of course, serves at the request of EPA; still,  in the sense that it




consists of extramural scientists, the Subcommittee provides something of an




extramural peer review function.  Perhaps Dr. Whittenberger would comment on




possible expansion of the Subcommittee1s mission to meet some of the issues




that have been raised.








J. L. Whittenberger;  Our Subcommittee was not asked to consider scientific




merit.  We were asked to assist in the planning process in cooperation with




EPA scientists and administrators.  We have raised questions about scientific




merit but do not consider it our responsibility to exercise the kind of




judgment that we would perhaps like to exercise.  We're not constituted to do




that.  In our last meeting, we did discuss the need for peer review for




scientific merit as well as for relevance, and suggested that SAB might take




responsibility for setting up subcommittees that would be more properly




constituted to carry out such peer review.  This discussion occurred at the




end of our meeting; we have taken no action to implement it.









     I think it would be worthwhile to respond to  Dr. Alpen's comments




[Chapter 6 of this volume] and to some of the comments by Dr. Goldman




illustrating the need for a much better understanding, on the part of




scientists outside EPA, of the constraints under which EPA operates as a




regulatory agency.  Dr. Altshuller mentioned some of the items on my






                                     420

-------
40.  PANEL DISCUSSION
checklist; I'll  touch on some that he didn't mention.   These comments will




express my point of view as an individual,  not as an SAB representative.









     With regard to Dr.  Alpen's comments [Chapter 6 of  this volume], when I




wear my "Harvard hat" I  agree with everything he says about toxicology and the




relevance of different kinds of studies in  respect to toxicologic analysis of




any obnoxious compound.   When I put on my "EPA hat," however, I am forced to




adopt a very different concept of what is "relevant."   One must never forget




that research conducted  within a regulatory agency is very different from




research conducted outside a regulatory agency.  EPA's  research program is




driven by program offices:  primarily, the  Office of Air, Noise and Radiation;




the Office of Water and  Waste Management; and the Office of Toxic Substances.




When EPA was established, research and development were put in one office that




was supposed to respond  to the program needs of other parts of the Agency.




This arrangement has continued to present certain problems.








     We must remember that the program offices are driven by the specific




requirements of legislation.  Sometimes EPA is undeservedly blamed for




situations  for which it  is not responsible.  The fact that scrubbers are




required  in  power plants that can use low-sulfur coal does not reflect




perversity on the part of EPA but, rather,  legislation  that was written so as




to keep people employed in Appalachia.  Another example is the Toxic




Substances Control Act,  an act that many consider almost impossible to




administer.
                                      421

-------
40.  PANEL DISCUSSION
     Several years ago, the program offices complained about ORD research:




they said that the scientists wanted to "do their thing" without responding to




the needs of the program offices.  Congress became upset and, as Dr.




Altshuller said, demanded that something be done.  Dr. Gage set up a task




force to examine how other agencies handle this problem.  The end result was




the creation of the Research Committees.








     The Research Committees have certain advantages and certain limitations.




One problem is the need to support long-term basic research as well as more




immediate regulatory-related research (i.e., research to provide certain




"answers" within six months or a year).  Somehow, the long-term research needs




seem to "get lost" before they are ever funded.  This is due not only to




limitations of the Research Committee approach but also to limitations




inherent in the legislative process.  Congress recognizes EPA's need for basic




research; the Authorization Subcommittee encouraged Dr. Gage to plan a




so-called Anticipatory Research Program.  But Dr. Gage has had trouble




obtaining funding for that program because the Appropriations Committee




doesn't look at things the way the Authorization Subcommittee does.









     Dr. Gage has been able to set up so-called Centers of Excellence in




universities or consortia of universities.  In designating a Center of




Excellence, EPA adds some relatively long-term funding to orient an already




strong program to EPA needs.
                                     422

-------
40.  PANEL DISCUSSION
     On the  subject that is paramount in this Panel  Discussion—peer




review—I'd  like  to supplement some of the previous  comments on EPA's




"credibility problem."   From Mr.  Costle on down,  there is a great deal of




concern within  EPA about peer review of all research activities (both




intra- and extramural).   As Dr. Altshuller said,  EPA originally had a peer




review mechanism  which worked well in many parts  of  the Agency.  It did not




work as well in other parts of the Agency,  and those places may have had more




influence on Dr.  Gage.   Thus, he  made the decision to set up the centrally




administered Grants Program.  This program,  which is spelled out in the




Federal Register, is now getting  underway.   The first committee meeting is




scheduled for next month and the  notifications will  be made, I believe, in




April.









     That takes care of  a certain part of EPA-funded research, but what about




the cooperative agreements, contracts,  and so on?  I don't believe those




mechanisms have really been developed yet,  although  I know that Dr. Gage is




very interested in putting such systems in place.








     Some of the  problems that I've referred to very briefly were the subject




of a study by the National Academy of Sciences reported in 1977.  In 1978-9,  I




participated in a study, mandated by Congress, in which we looked at EPA




health effects  research  from the  point of view of relevance to program needs




as well as quality.  After receiving our report,  Mr. Costle asked the




Environmental Health Advisory Committee (an SAB committee) to assume an




ongoing responsibility for peer review of EPA's major health effects






                                      423

-------
40.  PANEL DISCUSSION
laboratories.  This will encompass the intramural and extramural research that




is not covered by the centrally administered Grants Program.









     I'd be glad to answer any questions about peer review.









R. S. Chapman;  I'm concerned, because the essential focus of workshops such




as this, and of peer review generally, tends to be the review of scientists




within EPA by scientists outside EPA.  For two reasons, I think that such a




focus is not very effective in resolving limitations in EPA's scientific




credibility.  First, there is already substantial agreement between intramural




and extramural scientists on the standards required to ensure scientific merit




in research projects.  Second, instead of this constant droning about review




of scientists by scientists, what we need is an educative function by




scientists both within and without EPA to all levels of the administrative and




legislative structure under which we operate.  Such a function, if given




"teeth" and if perpetuated, would be far more useful than this constant




reshuffling of various ways to have scientists critique the work of other




scientists, which just isn't "getting the baby washed," to be perfectly honest




with you.  In my opinion, we are today under the same administrative and




legislative pressures to do work that1s against our better judgment as we have




been ever since EPA was formed.  And so I believe that education "up the line"




will be a lot more useful and fruitful than yet another mechanism by which




scientists can review scientists.  Within the Agency, we're quite inclined to




agree with scientists who criticize the merits of our work; we tend to say,




"Yes, we agree with you."  When our critiques are carried into the






                                     424

-------
40.  PANEL DISCUSSION
administrative area, however,  our better judgment tends to be paired against




the conflicting pressure of "The mandate is here now, the money is here now,




the money won't be here if you don't think the work can be done now."  The




former pressure always loses out to the latter.  Until the former pressure has




more  authority, I think that this kind of exercise is, to a considerable




extent,  rather futile.









J. L. Whittenberger;  I understand your feelings and I agree that the program




of which you are a part was subjected to a very critical and in many ways




unfair review.  Personally, I felt that the CHESS program (one of the programs




that took place over a number of years in the unit to which Dr. Chapman




belongs) was a very important program; it should have been constructively




 criticized but continued.  One of the reasons we don't today know more about




 the effects of oxidants on the health of people in Los Angeles is the




 dismantling of that program.








      On the other hand,  I agree  strongly with  Dr. Goldman that scientists




 anywhere ought to be  subject  to  critical review by their peers.   In visiting




 many EPA laboratories,  I have encountered  an  almost uniform desire to be




 critically reviewed.   Many scientists feel that their work has to stand the




 test of peer  review not only for publication  in scientific journals but also




 for  introduction as evidence in court.  They  want that evidence to stand up.




 So,  I hope you'll be  patient with the attempts to extend consistently applied





 peer review within EPA.
                                       425

-------
40.  PANEL DISCUSSION
R. S. Chapman;  I agree completely.  I'm certainly not in favor of less peer




review for EPA scientists.  I'm in favor of widening the perspective of




discussions like this (which, I might say, always come at the end of a meeting




instead of at the beginning or middle) to consider perhaps two more aspects:




(1) the entry of peer review at a higher administrative level, and (2) giving




the working scientists some tools for meaningful implementation of their ideas




coupled with the ideas of the peer reviewers.  I'm certainly not arguing for




less peer review and scientist-to-scientist contact; rather, I'm arguing for




more effective use of it.









J. L. Whittenberger;  I certainly agree.  I encountered that same feeling in




many of the EPA laboratories.  The scientists desire far more feedback, both




from SAB and from the scientists who reviewed their work.








T. Crocker;  I think what Dr. Chapman is advocating might go even to the level




of peer review at upper administrative levels of the agencies and of the




Office of Management and Budget.  In other words, who tells those gentlemen




how much to assign to a specific project in a specific year?









J. L. Whittenberger;  I would extend your comment to Congress as well.









T. Crocker;  In other words, it doesn't do us "bench people" a lot of good to




chew at each other* s quality when the decisions about relevance and resources




are made far beyond our reach.  I think Dr. Chapman is asking how we might




reach that level.  Possibly, increased communication between administrators






                                     426

-------
40.  PANEL DISCUSSION
and scientists would help.  But high-level administrators are not present at




this workshop.
                                       427

-------
                              MAILING ADDRESSES:
                           AUTHORS AND DISCUSSANTS
Edward L. Alpen
Lawrence Berkeley Laboratory
University of California
Berkeley, CA  94720

Aubrey P. Altshuller
Environmental Sciences Research Laboratory, MD-59
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NO  27711

Donald C. Borg
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY  11973

Robert S. Chapman
Health Effects Research Laboratory, MD-54
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC  27711

David L. Coffin
Health Effects Research Laboratory, MD-70
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC  27711

Daniel L. Costa
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY  11973

Timothy Crocker
University of California
Irvine, CA  92717
                                     428

-------
AUTHORS AND DISCUSSANTS
Walden E. Dalbey
Respiratory Carcinogenesis
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN  37830

Stanley V. Dawson
Research Division
California Air Resources Board
1102 Q Street
Post Office Box 2815
Sacramento, CA  95812

Roger Detels
School of Public Health
University of California
Los Angeles, CA  90024

Robert T. Drew
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY   11973

Donald L. Dungworth
Department of Pathology
School of Veterinary  Medicine
University of California
Davis, CA 95616

William  Prietsch,  III
Energy Effects Division, RD682
Office of Research and Development
U.S.  Environmental Protection Agency
Washington,  DC   20460

Donald E. Gardner
Health Effects Research  Laboratory,  MD-82
Office of Research and Development
U.S.  Environmental Protection Agency
 Research Triangle Park,  NC  27711

Joel F.  Ginsberg
 59 Whispering Pines Drive
 Sugar Mountain,  TN  37377
                                      429

-------
AUTHORS AND DISCUSSANTS
Marvin Goldman
Laboratory for Energy Related Health Research
University of California
Davis, CA  95616

Judith A. Graham
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC  27711

Edward D. Haak, Jr.
Health Effects Research Laboratory, MD-58
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC  27711

Sonja Haber
Medical Department
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY   11973

Jack D. Hackney
Environmental Health Services
Rancho Los Amigos Hospital
University of Southern California
7601 East Imperial Highway
Downey, CA   90282

Wanda M. Haschek
Biology Division
Oak Ridge National Laboratory
Post Office  Box Y
Oak Ridge, TN  37830

Milan J. Hazucha
Division of  Pulmonary Diseases
Department of Medicine
School of Medicine
University of North Carolina
Chapel Hill, NC  27514

Carol A. Heckman
Biology Division
Oak Ridge National Laboratory
Post Office  Box Y
Oak Ridge, TN  37830
                                     430

-------
AUTHORS AND DISCUSSANTS
Rogene P. Henderson
Inhalation Toxicology Research Institute
Lovelace Bioiaedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM  87115

John R. Holmes
California Air Resources Board
1102 Q Street
Post Office Box 2815
Sacramento, CA  95812

Steven M. Horvath
Institute of Environmental  Stress
University of California
Santa Barbara, CA   93106

Edward Hu
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle  Park, NC   27711

Junichi  Iwai
Medical  Department
Brookhaven National Laboratory
Associated Universities,  Inc.
Upton, NY   11973

Michael  H. Jones
Strategies and Air Standards Division, MD-12
Office of Air Quality Planning  and  Standards
U.S.  Environmental Protection Agency
Research Triangle Park, NC   27711

Robert  E. Lee
Health  Effects Research Laboratory, MD-51
Office of Research and Development
U.S.  Environmental Protection Agency
Research Triangle Park, NC   27711

T. Major
 Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge,  TN  37830
                                      431

-------
AUTHORS AND DISCUSSANTS
Ann C. Marchok
Respiratory Carcinogenesis
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN  37830

Joe L. Mauderly
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM  87115

Fred J. Miller
Health Effects Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC  27711

Samuel C. Morris
Biomedical and Environmental Assessment Division
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, NY  11973

Paul E. Morrow
Radiation Biology and Biophysics
School of Medicine and Dentistry
University of Rochester
Rochester, NY  14642

A. C. Olson
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN  37830

John A. Pickrell
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office Box 5890
Albuquerque, NM  87115

Otto Raabe
Laboratory for Energy Related Health Research
University of California
Davis, CA  95616
                                     432

-------
AUTHORS AND DISCUSSANTS
Gerald Rausa
RD681
Office of Research and Development
O.S. Environmental Protection Agency
Washington, DC  20460

Nathaniel W. Revis
Biology Division
Oak Ridge National Laboratory
Post Office Box Y
Oak Ridge, TN  37830

Karen M. Schaich
Medical Department
Brookhaven National Laboratory
Associated Universities,  Inc.
Upton, NY   11973

C. C. Scott
Department of Nutrition
Harvard University School of Public Health
665 Huntington Avenue
Boston, HA   02115

Russell P.  Sherwin
Department of Pathology
School of Medicine
University of  Southern California
2025  Zonal  Avenue
Los Angeles, CA  90033

Steven A.  Silbaugh
Institute of Environmental Stress
University  of  California
Santa Barbara, CA   93106

F.  Snyder
 Oak Ridge Associated Universities
Oak Ridge,  TN   37830

 Beverly E.  Tilton
Environmental  Criteria and Assessment Office, MD-52
Office of Research and Development
U.S.  Environmental  Protection Agency
 Research Triangle Park,  NC  27711
                                      433

-------
AUTHORS AND DISCUSSANTS
James L. Whittenberger
Department of Physiology
School of Public  Health
Harvard University
665 Huntington  Avenue
Boston, MA   02115

M. Jean Wiester
Health Effects  Research Laboratory, MD-82
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC  27711

Hanspeter Witschi
Biology Division
Oak Ridge National Laboratory
Post Office  Box Y
Oak Ridge, TN  37830

Ronald K. Wolff
Inhalation Toxicology Research Institute
Lovelace Biomedical and Environmental Research Institute
Post Office  Box 5890
Albuquerque, NM  87115
                                      434
                                                      t US. GOVERNMENT PRINTING OFFICE: 1M1 -757-064/0221

-------