United Starts                                •    -   -J$:
           Environmental Prottction                            tCAO-CIM-OOT
           Agency	"	                       December, 1991
EPA      Research and
           Development
             DRINKING WATER CRITERIA DOCUMENT FOR
             POLYCYCLK AROMATIC HYDROCARBONS (PAHS)
          Prepared for
             OFFICE OF WATER
          Prepared by
          Environmental Criteria and Assessment Office
          Office of Health and Environmental Assessment
          U.S. Environmental Protection Agency
          Cincinnati, OH  45268

-------
                                  DISCLAIMER
    This   document   has   been  reviewed   1n  accordance   with  the   U.S.
Environmental  Protection  Agency's  peer and  administrative  review  policies
and approved for publication.  Mention  of  trade names  or  commercial  products
does not constitute endorsement or recommendation  for use.
                                      11

-------
                                   FOREWORD


    Section  1412  (b)(3)(A)  of  the Safe  Drinking  Hater Act,  as  amended  In
1986, requires  the  Administrator of  the  Environmental  Protection Agency  to
publish  maximum  contaminant  level  goals  (MCLGs)  and  promulgate National
Primary  Drinking  Water  Regulations  for   each  contaminant,  which,  In  the
judgment of  the Administrator,  may have  an  adverse effect on public  health
and  which  Is known  or  anticipated to  occur  In public  water systems.   The
MCLG  Is  nonenforceable  and  1s set  at  a level at which  no known or  antici-
pated  adverse  health  effects   In .humans  occur  and  which  allows  for  an
adequate margin of  safety.   Factor's considered 1n  setting the MCLG  Include
health effects data  and  sources  of exposure other  than drinking water.

    This document provides   the  health effects  basis  to be considered  1n
establishing the  MCLG.  To achieve  this objective,  data  on pharmacok1net1cs,
human exposure, acute and  chronic  toxkUy to  animals  and humans, epidemi-
ology and mechanisms of toxldty are evaluated.  Specific  emphasis  1s  placed
on  literature  data  providing dose-response  Information.   Thus,  while  the
literature search and evaluation performed In support  of  this document  has
been  comprehensive,  only the  reports  considered most pertinent In  the  deri-
vation of  the MCLG  are  dted 1n the document.  The comprehensive  literature
data  base  In support of this document  Includes  Information  published  up  to
1985;  however,  more  recent  data  may  have  been  added  during  the   review
process.

    When adequate health effects data exist, Health  Advisory  values for less
than  lifetime  exposures   (1-day,   10-day  and  longer-term,  -10X   of   an
Individual's lifetime) are Included 1n  this document.   These values are  not
used  In  setting the MCLG,  but  serve  as  Informal  guidance to municipalities
and  other  organizations when  emergency  spills or  contamination  situations
occur.
                                       Tudor  Davles,  Director
                                       Office of Science and Technology

                                       James  Elder,  Director
                                       Office of Ground Water  and Drinking
                                       Water
                                      111

-------
                             DOCUMENT DEVELOPMENT
RHa Schoeny, Document Manager
Environmental Criteria and Assessment Office, Cincinnati
U.S. Environmental Protection Agency
Authors

Rita Schoeny
Glenn R1ce
Patricia A. Murphy
Linda S. Erdrelch
Larry Fradkln
Annie Jarabek
Environmental Criteria and
  Assessment Office, Cincinnati
U.S. Environmental Protection Agency
  Ellen O'Flaherty
  Dept. of Environmental Health
  University of Cincinnati
  Medical Center
Scientific Reviewers

William Bruce Pelrano
Environmental Criteria and
  Assessment Office, Cincinnati
U.S. Environmental Protection Agency

Y. Pate!
Office of Drinking Water
Washington, DC

Roy Albert
Department of Environmental Health
University of Cincinnati
  Medical Center
3223 Eden Avenue
Cincinnati, OH 45267-0056

Daniel Krewskl
13 Lansfleld Way
Ottawa, Ontario
Canada K2G 3V7
  David Warshawsky
  University of Cincinnati
  3223 Eden Avenue
  Cincinnati, OH 45219

  Alexander Wood
  Hoffman-LaRoche
  340 Klngsland Street
  Nutley, NO 07110

  Edmond LaVole
  Rutgers University
  College of Pharmacy
  Department of Pharmaceutical
    Chemistry
  Plscataway, NJ 08855
  Editorial Reviewers

  Judith Olsen
  Environmental Criteria and
    Assessent Office, Cincinnati
  U.S. Environmental Protection Agency
Document Preparation

Technical  Support  Services  Staff:
Office, Cincinnati
Environmental  Criteria  and  Assessment
                                      1v

-------
                            TABLE OF CONTENTS
                                                                     Page
 I.  SUMMARY	     1-1
II.  PHYSICAL AND CHEMICAL PROPERTIES	    II-l
     ACENAPHTHYLENE	    II-7
     ANTHRACENE		    II-7
          Production and Use	    II-7
          Occurrence .  . . .	    II-7
     BENZ[a]ANTHRACENE	    II-8
          Production and Use	    II-8
          Occurrence	    II-8
     BENZO[a]PYRENE	    II-8
          Production and Use	    II-8
          Occurrence	    II-8
     BENZO[b]FLUORANTHENE	    II-8
          Production and Use	    II-8
          Occurrence	 .    II-9
     BENZO[k]FLUORANTHENE. 	    II-9
          Production and Use	."	    II-9
          Occurrence	    II-9
     BENZO[g,h,1]PERYLENE. . 	    II-9
          Production and Use	    II-9
          Occurrence	    II-9
     CHRYSENE	    11-10
          Production and Use	    11-10
          Occurrence .	    11-10
     DIBENZ[a,h]ANTHRACENE . .  .  ,	    11-10
          Production and Use	    11-10
          Occurrence	    11-10
     FLUORANTHENE	    11-10
          Production and Use	 .    11-10
          Occurrence	    11-10

-------
                          TABLE  OF  CONTENTS  (cent.)
                                                                      Page
      FLUORENE	    11-11
           Production and Use	    11-11
           Occurrence	    11-11
      INDENO[l,2,3-cd]PYRENE	    11-11
           Production and Use		    11-11
           Occurrence	    11-11
      NAPHTHALENE	    11-12
           Production and Use	    11-12
           Occurrence	    11-12
      PHENANTHRENE	    11-13
           Production and Use	'.  .  .  .    11-13
           Occurrence	    11-13
      PYRENE	    11-13
           Production and Use	    11-13
           Occurrence	    11-13
      SUMMARY	    11-13
III.  TOXICOKINETICS	III-l
      ABSORPTION.,	III-l
           Oral	III-l
           Inhalation	III-6
           Dermal	   111-13
      DISTRIBUTION	   111-15
           Oral	111-16
           Inhalation	  .....   111-17
           Dermal 	  	   111-19
      METABOLISM	  .  .  .	111-20
           Benzo[a]pyrene as a Model  of  PAH Metabolism.  	   111-22
           Metabolism of Nonalternant  PAHs.	III-28
           Other Toxlflcatlon Pathways	111-28
           Comparative Metabolism 	  .  .   111-33
                                     v1

-------
                         TABLE  OF  CONTENTS  (cent.)

                                                                     Page
     EXCRETION 	  111-35

          Oral	111-38
          Inhalation	,	111-39
          Dermal  	 	  Ill-41

     SUMMARY 	  Ill-42

IV.   HUMAN EXPOSURE	   IV-1
             (To  be provided by the Office of Drinking Water)

 V.   HEALTH EFFECTS IN ANIMALS 	    V-l

     ACUTE ORAL TOXICITY	.V-l

          Acenaphthylene	    V-l
          Anthracene .  . .  . k	 .  .  .  .    V-l
          Benz[a]anthracene	    V-2
          Benzo[a]pyrene	    V-2
          Benzo[b]fluoranthene 	    V-2
          Benzo[k]fluoranthene	    V-2
          Benzo[g,h,1]perylene .  .  .  .	    V-2
          Chrysene	    V-2
          D1benz[a,h]anthracene	    V-3
          Fluoranthene	    V-3
          Fluorene	    V-3
          Indeno[l,2,3-cd]pyrene  	    V-3
          Naphthalene	    V-3
          Phenanthrene		    V-5
          Pyrene	    V-5

    .ACUTE TOXICITY BY OTHER ROUTES	    V-5

          Acenaphthylene	    V-5
          Anthracene	  .  . -.	    V-6
          Benz[a]anthracene. .	    V-7
          Benzo[a]pyrene	    V-7
          Benzo[b]fluoranthene 	    V-8
          Benzo[k]f1uoranthene .  .  .	    V-8
          Benzo[g,h,1]perylene	    V-8
          Chrysene	    V-8
          D1benz[a,h]anthracene	    V-9
          Fluoranthene	    V-9
          Fluorene	    V-9
          Indeno[l,2,3-cd]pyrene	    V-10
          Naphthalene.	 	    V-10
          Phenanthrene	  .    V-l 3
          Pyrene	    V-14

-------
                    TABLE OF CONTENTS (cent.)

                                                                Page
SUBCHRONIC AND CHRONIC ORAL TOXICITY	    V-15

     Acenaphthylene  	    V-15
     Anthracene	    V-17
     Benz[a]anthracene	    V-18
     Benzo[a]pyrene  .	    V-18
     Benzo[b]fluoranthene 	    V-19
     Benzo[k]fluoranthene 	    V-19
     Benzo[g,h,1]perylene 	    V-19
     Chrysene	    V-20
     D1benz[a,h]anthracene	    V-20
     Fluoranthene	    V-20
     Fluorene	    V-21
     Indeno[l,2,3-cd]pyrene 	    V-22
     Naphthalene	    V-23
     Phenanthrene.,.	    V-28
     Pyrene	%	'.  . . .	    V-28

SUBCHRONIC AND CHRONIC TOXICITY BY OTHER ROUTES .......    V-29

     Acenaphthylene	    V-29
     Anthracene	    V-30
     Benz[a]anthracene	    V-30
     Benzo[a]pyrene  	    V-30
     Benzo[b]fluoranthene 	    V-31
     Benzo[k]fluoranthene	 .    V-32
     Benzo[g,h,1]perylene	    V-32
     Chrysene	    V-32
     D1benz[a,h]anthracene	    V-32
     Fluoranthene	    V-33
     Fluorene	    V-34
    . Indeno[l,2,3-cd]pyrene .  .	    V-34
     Naphthalene	    V-34
     Phenanthrene 	    V-34
     Pyrene	    V-34

TARGET ORGAN TOXICITY . . .  .	    V-35
CARCINOGENICITY, ORAL .	    V-38

     Acenaphthylene	    V-38
     Anthracene	    V-39
     Benz[a]anthracene	    V-39
     Benzo[a]pyrene  . .	 . .	    V-41
     Benzo[b]fluoranthene 	    V-58
     Benzo[k]fluoranthene 	 	    V-58
     Benzo[g,h,1]perylene ..... 	 . 	    V-58
     Chrysene .•	    V-58
     D1benz[a,h]anthracene	    V-58
     Fluoranthene 	    V-61

-------
                         TABLE  OF  CONTENTS  (cont.)

                                                                     Page
          Fluorene	    V-61
          Indeno[l,2,3-cd]pyrene 	    V-63
          Naphthalene.	    V-63
          Phenanthrene 	    V-64
          Pyrene .	    V-64

    'CARCINOGENICITY, OTHER ROUTES .	    V-64

          Acenaphthylene	    V-64
          Anthracene	    V-65
          Benz[a]anthracene	    V-65
          Benzo[a]pyrene 	    V-69
          Benzo[b]fluoranthene and Benzo[k]f1uoranthene	    V-83
          Benzo[g,h,1]perylene 	    V-90
          Chrysene	    V-91
          D1benz[a,h]anthracene	    V-95
          Fluoranthene	 .    V-98
          Fluorene	    V-102
          Indeno[l,2,3-cd]pyrene 	    V-103
          Naphthalene	    V-107
          Phenanthrene	    V^llO
          Pyrene	    V-lll

     REPRODUCTIVE/TERATOGENIC EFFECTS	    V-113

          Anthracene	    V-113
          Benz[a]anthracene	    V-113
          Benzo[a]pyrene	    V-113
          Chrysene	    V-120
          Fluoranthene	    V-120
          Naphthalene	    V-121
    /
     MUTAGENICITY/6ENOTOXICITY ...  	  .....    V-123
     SYNERGISH AND/OR ANTAGONISM 	 ......  	    V-123
     SUMMARY	    V-163

VI.  HEALTH EFFECTS IN HUMANS	   VI-1

     INTRODUCTION		   VI-1
     CLINICAL CASE STUDIES	   VI-3

          Oral .	   VI-3
          Other Routes	   VI-5

     EPIDEMIOLOGIC STUDIES	   VI-7

          Oral	   VI-7
          Other Routes	   VI-7

     SENSITIVE POPULATIONS 	 	   VI-14
     SUMMARY	   VI-15
                                    1x

-------
                           TABLE  OF  CONTENTS  (cont.)

                                                                       Page
 VII.  MECHANISMS OF TOXICITY	VII-1

       MECHANISMS INVOLVED IN PAH CARCINOGENICITY	VII-1

            Structure Activity Relationships 	   VII-1
            Tissue Specificity of PAH Metabolism 	   VII-9
            Genetic Control of PAH Metabolism	VII-10
            Other Pathways Involved 1n Activation of PAH 	   VII-14
            PAH Involvement In Carcinogenic Processes	VII-15

       MECHANISMS INVOLVED IN NONCARCINOGENIC ENDPOINTS	VII-16

            Hemolymphatlc System . .  .	   VII-16
            Cardiovascular System	VII-19
            Pulmonary System 	  .   VII-20
            GastrolntesUnal System	VII-21
            Gonads	   VII-21
            Endocrine System . .  . .	VII-22
            Integumentary System 	 	   VII-23
            Visual System	VII-23

       PAH-INDUCED IMMUNOTOXICITY	VII-26
       SUMMARY	VII-28

VIII.  QUANTIFICATION OF TOXICOLOGIC  EFFECTS 	   VIII-1

      'INTRODUCTION	VIII-1
       NONCARCINOGENIC EFFECTS 	  .  	   VIII-6
       SHORT-TERM STUDIES IN ANIMALS  	   VIII-7
       LONGER-TERM STUDIES IN ANIMALS.	   VIII-7

            Acenaphthylene 	   VIII-8
            Anthracene	   VIII-8
            Benz[a]anthracene	VIII-9
            Benzo[a]pyrene 	   VIII-9
            D1benz[a,h]anthracene. 	 	   VIII-10
            Fluoranthene 	   VIII-10
            Fluorene 	  	   VIII-11
            Pyrene 	 	   VIII-11

       QUANTIFICATION OF NONCARCINOGENIC EFFECTS  	   VIII-12

            Derivation of 1- and  10-Day Health  Advisories	VIII-12
            Derivation of Longer-Term HA 	   VIII-15
            Assessment of Lifetime Exposure and Derivation
            of DWELs 	   VIII-15

       CARCINOGENIC EFFECTS	  .   VIII-21

-------
                        TABLE OF CONTENTS (cont.)

                                                                    Page
     CLASSIFICATION OF  PAHs  AS  TO  POTENTIAL FOR HUMAN
     CARCINOGENICITY .  .	VIII-23

          Acenaphthylene  	  VIII-23
          Anthracene	,  .	VIII-23
          Benzo[g,h,1]perylene  	  VIII-23
          Fluoranthene  .  .  	  VIII-24
          Fluorene 	 ...  VIII-24
          Naphthalene	.'.  .	VIII-24
          Phenanthrene  	.'	  VIII-24
          Pyrene 	  VIII-25
          Benz[a]anthracene.  .	  VIII-25
          Benzo[a]pyrene  	  VIII-25
          Benzo[b]fluoranthene  .  	 .....  VIII-26
          Benzo[k]fluoranthene  	  VIII-26      v
          Chrysene . .^	  VIII-26
          D1benz[a,h]anthracene	'	  VIII-27
          Indeno[l,2,3-cd]pyrene  	  VIII-27

     QUANTIFICATION OF  CARCINOGENIC EFFECTS	VIII-27

          Benz[a]anthracene	VIII-28
          Benzo[a]pyrene  	  VIII-28
          U.S.  EPA (1980d) Approach	. .  . .	VIII-29
          Clement  Associates  (1988)	VIII-31
          Krewskl  and Murdoch (1990)  	  VIII-38
          Clement  Associates  (1990a)  	  VIII-39
          U.S.  EPA (1991b)	VIII-49
          Brune et al.  (1981	VIII-52
          Chouroullnkov et al.  (1967)	VIII-54
          D1benz[a,h]anthracene	VIII-62

    'SPECIAL CONSIDERATIONS.	  vin-63

          Toxldty Equivalence  Factor Approach  for  PAHs In
          Group B2	VIII-63
          Interactions  with  Other  Chemicals.  .  . 	  VIII-77

     EXISTING GUIDELINES, RECOMMENDATIONS AND STANDARDS	VIII-79
     SPECIAL GROUPS AT  RISK	  VIII-80

     SUMMARY	VIII-81

IX.  REFERENCES	    IX-1
                                   x1

-------
                               LIST OF TABLES
No.                                Title                              Page
 II-l   Priority Pollutant Polycycllc Aromatic Hydrocarbons
        Found 1n the Environment	    II-2
 II-2   Selected Physical and Chemical Characteristics of
        Important PAHs	    II-3
III-l   Examples of Metabolism of PAHs to Biologically Active
        Forms by Various Enzyme Systems ..."	•. •  •  •   111-30
  V-l   Oral Carclnogenlclty Testing of Benz[a]anthracene
        Administered by Gavage to Hale B6AF1  Mice .	     V-40
  V-2   Tumor Incidence 1n "Atrium of Stomach" Following Gavage
        Administration of Benzo[a]pyrene	     V-44
  V-3   Incidence of Tumors Observed After a  Single  Gavage Treatment
        of Mice with Benzo[a]pyrene 1n Polyethylene  Glycol	     V-45
  V-4   Carclnogenlclty of Oral Benzo[a]pyrene In Sprague-
        Dawley Rats	     V-46
  V-5   Incidence of Forestomach Paplllomas and Carcinomas 1n
        Male and Female CFW Mice Administered Benzo[a]pyrene In
        the Diet	     V-50
  V-6   Carclnogenlclty of Benzo[a]pyrene Administered 1n the
        Diet to Male and Female Swiss Mice. .	     V-51
  V-7   Induction of Forestomach Tumors 1n Ha/ICR Mice Fed
        Dietary Benzo[a]pyrene	     V-55
  V-8   Oral Carclnogenlclty of D1benz[a,h]anthracene In DBA/2
        Mice	     V-60
  V-9   Carclnogenlclty Testing of Fluorene Administered 1n the
        Diet to Female Buffalo Rats	     V-62
  V-10  Dermal, Injection and Implantation Carclnogenlclty
        Assays of Anthracene	     V-66
  V-11  Carclnogenlclty of Benzo[a]pyrene to  Male Syrian
        Golden Hamsters by Inhalation 	     V-72
  V-l2  Carclnogenlclty of Benzo[a]pyrene Administered by
        Intratracheal Instillation to Syrian  Hamsters 	     V-74
  V-l3  Sarcomagenlc Activity of Benzo[a]pyrene 1n Male C3H
        Mice Following a Single Subcutaneous  Injection	     V-79

-------
                           LIST OF  TABLES  (cont.)
No.                               Title                              Page

  V-14  SENCAR Mouse Skin Tumorlgenesls,  Benzo[a]pyrene-
        Tumor Initiation ...................  ...    V-81

  V-15  Carclnogenlclty of Benzofluoranthenes  by  Implantation
        1n Rat Lungs ........................    V-84

  V-16  Cardnogenldty Assay of  PAH  In Newborn Mice ........    V-85

  V-17  Tumor Initiating Activity of  Benzofluoranthenes 1n
        Crl:CD-l  Mice ...... .................    V-87

  V-18  Tumor Initiating Activity of  Benzofluoranthenes 1n
        Crl:CO-l  (ICR)BR Mice ............... '.  .  .  .    V-89

  V-19  Sarcomagenlc Activity of  Subcutaneously Injected
        D1benzo[a,h]anthracene and Benzo[a]pyrene In  Female
        NMRI Mice ........................  .    V-99

  V-20  Incidence of Flbrosarcomas 1n Mice  Associated with
        Subcutaneous Injections of D1benz[a,h]anthracene ......    V-100

  V-21  Cardnogenldty of Indeno[l,2,3-cd]pyrene upon
        Implantation In Rat Lungs .................    V-104

  V-22  Assay of  Pyrene In Newborn Mice  ......  ........    V-112

  V-23  Embryotoxldty and Malformations  1n Swiss Mice
      '  Exposed by Intraembryonal Injection of Benzo[a]pyrene
        and Derivatives . . ............  ........    V-117

  V-24  Results of Short-Term Tests of Polycycllc Aromatic
        Hydrocarbons ........................    V-124

  V-25  Cocarclnogenlc Activity of Various  PAHs with
        Benzo[a]pyrene 1n Mouse Skin ................    V-156

  V-26  Cocarclnogenlc Activity of Various  PAHs with
        Benzo[a]pyrene on Mouse Skin ................    V-157
  V-27  Co-administration of Pu02 and Benzo[a]pyrene
        by Inhalation .......................     V-162

 VI-1   IARC Determinations of PAH Cardnogenesls  Based  on
        Human Data .........................     VI-9

VII-1   Reactivity Indices for Polycycllc  Hydrocarbons.  ......    VII-8

-------
                            LIST OF TABLES (cont.)


                                    Title

         Immunotoxlclty of Benzo[a]pyrene In Mice after
         Short-Term Exposure 	  VIII-13

VIII-2   Comparative Effects of PAH'on Immune Functions In B6C3F1
         Mice	VIII-14

VIII-3   Incidence of Tumors 1n Mice Treated Intragastrlcally
         with Benzo[a]pyrene 	  VIII-30

VIII-4   Forestomach Tumors 1n Mice Fed Benzo[a]pyrene 	  VIII-32

VIII-5   Variable Exposure Data Used to Estimate Parameters
         1n the Ingestlon Dose-Response Model for Benzo[a]pyrene .  .  VIII-35

VIII-6   Forestomach Tumors In Mice Subjected to Variable
         Exposures of Benzo[a]pyrene 	  VIII-40

VIII-7   Historical Control Incidence Data for Forestomach Tumors
         1n Strains of Swiss (Webster) Bred Mice 	  VIII-42

VIII-8   Maximum Likelihood Parameter Estimates for Two-Stage
         Model with Saturation of Growth Rate Function 	  VIII-44

VIII-9   Forestomach Tumors 1n Mice Subjected to Exposures of
         Benzo[a]pyrene for at least 70 Days	VIII-45

VIII-10  Predicted vs. Observed Tumor Incidence and Data Used to
       '  Calculate the Dose-Response Model 	  VIII-51

VIII-11  Incidence to Forestomach Only and Total Contact Site
         Tumors 1n Sprague-Dawley Rats Exposed to Benzo[a]pyrene
         by Gavage or In the Diet	VIII-53

VIII-12  Slope Factors for Humans Based on Benzo[a]pyrene
         Feeding Studies 	 	  VIII-56

VIII-13  Cancer Potency Estimates for Excess Risk of 10~5 from
         Lifetime Exposure Based on Oral Exposure Data for
         D1benz[a,h]anthracene 	  VIII-64

VIII-14  Potency Indices for the Carcinogenic PAH Compounds
         Based on Skin Painting Data	VIII-65

VIII-15  Ranking of PAHs Based on Estimates of Potency In
         Skin Painting Bloassay	  VIII-67

VIII-16  The Carcinogenic Potency of Various PAHs Measured In Two
         Different Animal Bloassay Systems 	  .  VIII-68

VIII-17  Summary of Relative Potency Estimates for Indicator PAHs.  .  VIII-72


                                      x1v

-------
                            LIST OF TABLES (cent.)


 No.                                Title                              Page

VIII-18  Summary of PAHs Relative Potencies	VIII-74

VIII-19  Relative Potencies for PAH and PAH-Conta1n1ng Mixtures. .  .  VIII-76

VIII-20  Comparison of Carcinogenic Potency of PAH Using
         Relative Tumor Dose 	  VIII-78
                                      xv

-------
                               LIST  OF  FIGURES

No.                                  Title                            Page


 II-l   Chemical Structure of Selected PAHs	  .  II-5

III-l   Metabolism Schematic for PAHs 	   111-23

III-2   Metabolism of Benzo[a]pyrene	111-24

III-3   Mechanisms of Enzymatic Activation of Benzo[a]pyrene
        to 7,8-D1ol-9,lO-Epox1des 	   111-27

III-4   The Positions of Bay-Regions. .	111-29

III-5   Cyclic Scheme of Benzo[a]pyrene D1one/D1ol  Involvement
        1n Redox Coupling 	   111-32
                                    xv1

-------
                             LIST OF ABBREVIATIONS

 AHH                      Aryl hydrocarbon hydroxylase
 B[a]P                    Benzo[a]pyrene
 BUN                      Blood  urea  nitrogen
 DMSO                     D1methylsulfox1de
 DNA                      Deoxyrlbonuclelc acid
 DWEL                     Drinking water equivalent  level
 GI                       Gastrointestinal
 GLC                      Gas  liquid  chromatography
 GSH                      Glutathlone
 HA                       Health advisory
 HDL                      High density Upoproteln
 HPLC                     High performance liquid chromatography
 l.m.                     Intramuscular
 l.p.                     Intraperltoneal
 1.v.                     Intravenous
 LDL                      Low  density Upoproteln
 LOAEL                    Lowest-observed-adverse-effect level
 LOEL                     Lowest-observed-effect level
_MFO                ..      Mixed  function oxldase
 NOAEL                    No-observed-adverse-effect level
 NOEL                     No-observed-effect level
 PAH                      Polycycllc  aromatic hydrocarbon
 PEG                      Polyethylene glycol
 q *                      Potency slope for carcinogenic risk estimate
 RfD                      Reference dose
 RNA                      R1bonucle1c add
 s.c.                     Subcutaneous
 SGOT                     Serum  glutamlc oxaloacetlc transamlnase
 SGPT                     Serum  glutamlc pyruvlc transamlnase
 TPA                    •  12-o-Tetradecanoyl phorbol-13-acetate
 UV                       Ultraviolet
                                     xv11

-------
                                  I.   SUMMARY

    This  document  Includes   Information  on  15  polycycllc  aromatic  hydro-
carbons (PAHs)  Included among  those  compounds  Identified  as  priority (water)
pollutants by the U.S. EPA.

    PAHs  are  a  class  of  diverse  compounds  that  are  formed  during  the
Incomplete  combustion of  organic  material.  They are ubiquitous and  enter
the environment  from many sources.   Generally  PAHs  are not very  soluble  1n
water.  Solubility  decreases  greatly with  Increasing  molecular  weight.   The
                                                             •
log of  the  octanol/water  partition  coefficient  (P)  also Increases  rapidly
with  Increasing  molecular  weight.  This  Increases adsorption  to partlculate
matter and  exposure  of  the compounds to mlcroblal degradation.   However,  1t
has been  found  that  PAHs with  more  than  four aromatic rings  are  less  amen-
able  to mlcroblal action.  The  larger P values  also  Increase the probability
of  bloaccumulatlon.  , Volatilization  does  not  appear  to be  a  significant
route of exit from aquatic systems due to  the low vapor pressures of  PAHs.
       /
    As  they are  highly  I1p1d  soluble,  PAHs readily  pass   through  cellular
membranes.   However,  the  rate of absorption  Is Increased when  the  PAHs  are
present 1n  an  oil carrier vehicle.   This 1s particularly true  for  oral, and
dermal routes of exposure.

    The  highly  I1p1d soluble  nature  of  PAH  compounds  results  1n  their
distribution throughout  the  body  In fatty tissues.   The  primary  sites  of
storage In  the  body  have been found  to be  similar In  a  variety  of mammalian
species.  These  Include  the  kidneys, liver and  fat  with some  accumulation
occurring 1n the spleen,  adrenals and ovaries.

04380                               1-1                               03/21/89

-------
     PAH  compounds  are metabolized  by the MFO  system of  enzymes  associated
with  cytochrome  P-450.   The  liver  has  the highest activity  for  this  system
although  MFO activity 1s  found 1n other  organs  such as  kidneys,  lungs  and
skin.  PAH  coumpounds have  been found to Induce Increased levels  of activity
as well as  synthesis of Isozymes of cytochrome P-450-assoc1ated enzymes.

    The  major  routes of  excretion for  PAH  compounds are  hepatoblHary  and
urinary.   Some evidence  for  mammary  gland excretion also exists.   Although
PAHs  present  In  the body  tend to be  present  In  fatty  tissues,  available
evidence  does  not  Indicate extensive  bloaccumulatlon  of  PAHs  1n  these  or
other  tissues.

    The  primary  focus of  research on biologic  effects  of PAHs has been  on
their  carclnogenlclty.   There  1s  generally  a  lack  of  research data  on  the
noncardnogenlc  toxic effects  of  oral  exposures  to  PAH compounds.   What
little research  has been conducted has  centered on  three or four  compounds
                                                                            •
.In  this  large class.   Target organs  are  diverse,  probably due to  the wide
distribution   of   PAH  compounds   throughout   the  body.   Toxlclty  centers
primarily  on hematopoletic  and lymphold systems.  Immunosuppresslon measured
1n  various experimental  systems  has  been observed  following exposure  to  a
number of PAHs.   Nonoral exposure  to PAHs has also  been observed  to  effect
changes  1n  lymphold and hematopoletic systems.

     There   Is  an extensive  data  base   on  the  carclnogenlclty  of  selected
PAHs.   Most studies,  however,  employed  dermal.  Inhalation  or  subcutaneous
rather than oral exposure.   Overall  there 1s  a great  deal  of  variation  1n
carcinogenic  potential  among  this  class of  compounds.   PAHs  may  produce

04380                               1-2                              09/24/90

-------
 tumors  In  the  vicinity where  they are  Introduced;  that  Is,  oral  exposure
 produces   stomach  tumors  and  Inhalation  produces  lung  and  upper  stomach
 tumors.   Distant  site tumors may form  dependent  on  systemic distribution of
 metabolites  and/or  metabolism at target tissues  removed  from site of Intro-
 duction.   Host  PAHs  have  been  observed to produce  genotoxlc effects 1n one
 or  more  test systems.  There are data Indicating that some  PAHs can serve as
 promoters  or  cocardnogens.

     Reproductive  and  teratogenlc  effects  have not been well studied 1n these
 compounds.  There 1s evidence, however,  that  Ingestlon  of benzo[a]pyrene by
 pregnant   mice  results   1n  reproductive deficits  1n  the  F,  generation  and
 that this  compound produces  ootoxldty.

     While  very  little Information exists on  the  effects  of specific PAHs on
 humans,  there  are  numerous  reports Unking  exposure to  environmental  and
 occupatlonally  generated  PAH-conta1n1ng  mixtures  to human  health  effects.
.These mixtures  Include  coal tar, soots,  coke oven  emissions  and cigarette
 smoke./

     The  ability  of  a  PAH .to  Induce  carcinogenic  responses  depends  on Its
 distribution to target  organs,  the  presence of potentially reactive areas 1n
 Its  structure  and   Us  potential  for  transformation  to  reactive  electro-
 phlles.   This last  factor  1s species and  tissue dependent, and also, to some
 extent,   a  function  of  1nduc1b1l1ty  of  the   cytochrome   P-450-assoc1ated
 enzymes.
 04380                               1-3                              09/24/90

-------
    Target tissues  for  PAH  toxlclty  other  than cardnogenesls are generally
those  engaged  1n active  DNA  synthesis.  Mechanisms  Involving PAH-mediated
DNA-damage or  suppression  of  DNA  synthesis  have  been  proposed for  some
target organs.  Naphthalene exposure 1s associated with anemias and cataract
formation; this latter  effect  1s not  seen with  other PAHs.

    There  were no  data  suitable  for  calculation  of  1- or  10-day  health
advisories for  any  PAH described  1n this document.   For all PAHs  In  this
document classified as B2, probable human carcinogen,  data were Insufficient
for  calculation  of  DWELs.   Subchronlc studies  (90-day  gavage  exposure  In
CD-I mice) were used as the bases for the following DWELs:  anthracene, 10.5
mg/i;  f luoranthene,  1.4  mg/i;  fluorene,   1.4  mg/l;  pyrene,   1.0  mg/l.
Although there Is a  90-day  study on  acenaphthylene,  only frank effects were
reported;  data were,  thus,  not appropriate  for  derivation of criteria.
Several studies have been evaluated as the basis for a  DWEL for naphthalene;
consensus, however,  has not  been reached as to  the critical study.

    Evaluation of cardnogenldty data prompted the following  classification
of  these  PAHs:  Group  D, not  classifiable  as to human  cardnogenldty  --
acenaphthylene,  anthracene,   benzo[g,h,1]perylene,  fluoranthene,  fluorene,
naphthalene,  phenanthrene and pyrene; Group B2, probable human carcinogen —
benz[a]anthracene,    benzo[b]fluoranthene,   benzo[k]fluoranthene;   benzo[a]-
pyrene, chrysene, d1benz[a,h]anthracene  and 1ndeno[l,2,3-cd]pyrene.

    Dose-response  data  for   benzo[a]pyrene  cardnogenldty   were  used  to
derive an  upper  bound estimate  of  the slope  of  the  dose-response curve at
low  doses.   This  evaluation,  based  on  the  linearized multistage procedure,


04380                               1-4                              09/24/90

-------
resulted  In  an  upper-bound  slope factor of 11.5 mg/kg/day.   This  use  of  the
multistage model was  Inadequate  because  the experimental  high-dose data  were
excluded  from  the analysis  and  a single  exposure time was  assumed.   Other
experimental data  and  modeling  techniques were  used.   Nine  risk estimates
were  calculated from three  different   studies  In  two  species   of  outbred
rodents.   Several  different models  and  data  -sets  were selected.  All  nine
                                      *
slope factors spanned less  than  one order of magnitude;  four estimates  were
selected  from these nine.   The data,  from  which  these estimates are derived,
are  considered   to  be  less-than-optlmal,  but  acceptable.   These  selected
studies have several  commonalities,  Including mode of  administration,  tumor
sites,  tumor  types and  the presumed  mechanisms  of  action.   The  data  sets
could  not be combined  prior to modeling (the  preferred approach)  because
they  employed  significantly dissimilar  protocols.  The range of  these  four
estimates  1s 4.5-9.0.  Each estimate  Is  based on a low-dose  extrapolation
procedure  and   entails  the  use   of   multiple  assumptions  and  default
procedures.  The geometric  mean  from four slope factors, each  considered  to
be of  equal  merit, .was used to  calculate  a  single oral slope  factor  of  5.8
per (mg/kg/day).  Using standard assumptions for human  body weight and water
consumption, a  drinking  water   unit  risk of  1.7E-4  per mg/L was  derived.
Concentrations   of  benzo[a]pyrene corresponding  to lifetime  risks of  10~s,
10~6   and  10~7   were  determined   to   be   6xlO~2,   6xlO~a   and   6xlO~«
    , respectively.
    An analysis  of the  relative  carcinogenic potency  of  12 PAH  compounds,
using  benzo[a]pyrene  as  the  basis  of  comparison,   was  made  using  the
two-stage model  applied  to a  recent compilation  of  animal  cardnogenesls
data In which  each experiment  tested both benzo[a]pyrene  and one  or  more of
the other compounds.   A  peer  review panel suggested that  different  criteria

04380                               1-5                              10/24/91

-------
be used  to  select data  for  a  final  relative potency determination  and  that
additional studies be  analyzed  1n  order to extend the  11st  of  PAH compounds
for which potencies  are evaluated.  Until  this  1s done,  no recommendations
can  be  made  concerning the  quantitative  cancer  risk   of the  other  PAH
compounds.
04380                               1-6                              10/24/91

-------
                     II.   PHYSICAL AND  CHEMICAL PROPERTIES

    Polycycllc aromatic hydrocarbons (PAHs) are a  class  of  diverse compounds
that  are  formed  during  the  Incomplete  combustion of  organic material.   A
large number  of  PAHs  have been Identified In the  environment,  most  of  which
derive from  anthropogenic  sources '(Santodonato et  a!.,  1980).  As  a  group,
PAHs  are  ubiquitous  In the atmosphere,  water  and  soil.  They  enter aquatic
systems from wastewater,   urban  stormwater runoff,  leaching  from soil,  and
wet and dry  deposition from the atmosphere  (Kveseth et  al.,  1982;  Andren and
Strand, 1981).  Some PAHs  have been  found  1n  surface  and grqundwater as well
as finished  drinking water.   This  document focuses on  those  PAHs  Identified
as priority  (water)  pollutants by the  U.S.  EPA (U.S.  EPA,  1980a-d).   Table
II-l  lists  these  PAHs; some  of  their  physical  and chemical  characteristics
are given 1n  Table II-2.   The structural  formulas  are shown  schematically 1n
Figures II-l and II-2.

    The persistence of PAHs  In the  aquatic environment  1s a  direct function
of their  physical  and  chemical properties.  These  govern  the susceptibility
of  the compounds  to  various degradation  processes,  Including  photolysis,
volatilization,  and  sedimentation  and  mlcroblal   degradation  (Callahan  et
 \
al.,  1979).   To  begin  with,  many.of  the PAHs have low  vapor  pressures and,
thus, low volatility.  Consequently,  volatilization may not  be Important 1n
the  removal  of PAHs  from water.  Second,  the  solubility of PAHs  In  water
tends  to  be  low,  which   Increases  their  susceptibility  to  adsorption onto
aquatic partlculate matter.   This.  1n turn, suggests that  sedimentation and
ultimately,   mlcroblal  degradation,  constitutes  the primary  removal  process.
04390                               II-l                             05/22/91

-------
                           TABLE 11-2
Selected Physical  and Chemical Characteristics of Important PAHsa
Physical and Chemical Characteristics
o
o








i— i
i— <
CO

IXJ
co
Compound
Acenaphthylene
Anthracene
Benz[a]anthracene
22nzo[a]pyrene
Benzo[b]f luoranthene
Benzo[k]f luoranthene
Benzo[g,h,1]perylene
Chrysene
01benz[a,h]anthracene
F luoranthene
Fluorene
Indeno[ 1,2, 3-cd Jpyrene
Naphthalene


CAS
Number
208-96-8
120-12-7
56-55-3
50-32-8
205-99-2
207-08-9
191-24-2
218-01-9
53-70-3
206-44-0
86-73-7
193-39-5
91-20-3


Molecular
Height
152.20
178.23
228.28
252.32
252.32
252.32
276.34
228.28
278.36
202.3
166.21
276.34
128.16


Melting
Point
CC)
92
218
167
178
168
215.7
278
255-256
266.6
111
116-117
162.5-164
80.2


Boiling
Point
CO
265-275
342
435
310-31 2b

480
550
448

375
295

218


Vapor Pressure Solubility
(torr) In H20 at 25*C Density
1xlO~» to 1x10"* 3.93 0.899
1.95xlO"« 0.045, 0.073 1.25
5.0xlO"» 0.014. 0.009 1.274
5.0x10"* 0.0038 1.35
1xlO~" Insoluble
9.59x10"" Insoluble
IxlQ-ioC 0.00026
1x10"" 0.002 1.274
1x10"" 0.0005 1.282
1xlO'»c 0.265 f
1xlO"» 1.68-1.98 1.203d
-1x10""
4.9x10"»« 30-40 1.145


Log Octanol/
Hater Partition
Coefficient
4.07
4.45
5.61
6.04
6.57
6.84
7.23
5.61
5.97
5.20
4.18
7.66
3.37



-------
                                  TABLE II-1
              Priority  Pollutant  Polycycllc  Aromatic Hydrocarbons
                           Found  In  the Environment*
             Acenaphthylene               Chrysene
             Anthracene                   D1bent[a»h]anthraeene
             Beni[a]anthracene            Fluoranthene
             Benio[a]pyrene               Fluorene
             Benio[b]fluoranthene         Indeno[l,2,3-
             Benio[k]fluoranthene         Naphthalene
             Benio[g,h,1]perylene         Phenanthrene
                                          Pyrene
•Source: U,S, IPA, liSOd

-------
                                                                         TABLE  II-2  (cont.)
CO
lO
o
Physical and Chemical Characteristics
Compound
Phenanthrene
Pyrene
CAS
Number
85-01-8 '
129-00-0
Molecular
Weight
178.23
202.26
Melting
Point
(°C)
100
156
Boiling
Point
CC)
340
385
Vapor Pressure
(torr)
6.8x10-"
6.85x10"'
Solubility
In H20 at 25'C
(rag/i)
1.00. 1.29
0.14. 0.132
Density
0.98f
1.271
Log Octanol/
Water Partition
Coefficient
4.46
5.32
          ^Source: Smith  et  al.. 1978; Weast,  1983;  Cleland and  Klngsbury.  1977;  Davis et al.,  1942;  May et al.,  1978;  IARC.  1973.  1983;  U.S.  EPA.
           1980c; Hansch and Leo. 1985
          °0eterm1ned at 10 mm Hg
          ^Determined at 25*C
          ^Determined at 0°C
          eDeterm1ned at 19.8°c
          'Determined at 4°C
co

-------
 Benzlalanthracene
          Chrysene
    Benzolalpyrene       Benzolklfluoranthene
Benzolblfluoranthene     Dibenz[a,h]anthracene
           Indenol 1,2,3-cd]pyrene

                     FIGURE II-l
              Chemical  Structure of Selected PAHs
04390
II-5
07/27/90

-------
    Acenaphthylene
      Anthracene
  Benzo[g,h,i]perylene
     Fluoranthene
           Fluorene
         Naphthalene
                                Phenanthrene
             Pyrene
                   FIGURE II-l (conl.)
              Chemical Structure of Selected PAHs
04390
II-6
07/27/90

-------
Those PAHs  composed  of four or  more  aromatic rings (therefore, having  high
molecular  weight)  are less  amenable  to mlcroblal  degradation;  they are
environmentally  more   stable   compounds.    Third,  the   log   octanol/water
partition  coefficients  of  the  PAHs also  tend to  be  high, and  there  1s  a
positive   correlation   between  these   coefficients   and   bloconcentratlon
potential.   Therefore,   1f  PAH   concentrations   Increase  quickly   enough,
toxldty could occur In aquatic  life.    PAHs,  however, do  tend  to  be  rapidly
metabolized and  excreted.  Fourth, the  previous  characteristics  seem  to  be
most  representative  of the so-called  carcinogenic  PAHs  (see Chapter  VIII),
which  Implies  that  such  PAHs  are more likely  to  persist. 1n the  aquatic
environment, yielding  longer-term  exposures   In  comparison with less  potent
PAHs.  The  production,  use and occurrence 1n  water  for  each of the  15  PAHs
described In this document follows.

Acenaphthylene
    Pertinent data regarding the production,  use and occurrence of acenaph-
                                                                           •
thylene could not be located In the available  literature.

Anthracene
    Production  and  Use.    In   1981,  U.S.  Imports  of  anthracene totalled
21,000 kg,  down  sharply from  510,000  kg Imported  In 1979.   Separate  data  on
U.S.  exports were  not  available (IARC,  1973).   Anthracene has been  used  as
an  Intermediate  1n dye production.  It.has also  been used In  smoke screens,
scintillation counter crystals  and  organic semiconductor  research.

    Occurrence.  Anthracene occurs ubiquitously  as  a  product_of  Incomplete
combustion;  U   also  occurs 1n fossil  fuels.   It  has  been   Identified  1n
04390                               I1-7                             05/22/91

-------
 surface   water,  tap  water   (1.1-59.7   ng/l),   wastewater  (1.6-7.0
 and dried  sediment  of  lakes  (30-650 yg/kg)  (IARC, 1973).
 BenzFalanthracene
     Production  and Use.  There  1s  no commercial  production  or  known use of
 •this  compound.

     Occurrence.    Benz[a]anthracene  occurs   ubiquitously   1n  products  of
 Incomplete combustion;  H  1s  also  found  1n  fossil  fuels.   It  has  been
 Identified 1n surface  water,  tap water  (0'.4-10.7 ng/l),  rainfall  (3.2-12.3
 ng/i),   subterranean   water  (0-1.3   ng/i),    wastewater   (0.5-4.9  yg/a).
 sludge   (230-1760   yg/kg)  and  freeze-drled  sewage  sludge  (0.62-19 mg/kg)
 (IARC,  1973).

 BenzoFalpyrene
     Production  and Use.  There  Is  no commercial  production  or  known use of
'this  compound.
       /
     Occurrence.    Benzo[a]pyrene  occurs   ubiquitously  1n products  of Incom-
 plete combustion;  It also occurs In fossil fuels.   It has  been  Identified In
 surface water   (0.2-13,000  ng/i),   tap   water  (0.2-1000  ng/i),  rain  water
 (2.2-7.3  ng/i),  subterranean  water  (0.4-7  ng/l),  wastewater  (0.001-6000
 yg/l),     sludge   (3-1330   yg/kg)    and    freeze-drled    sewage   sludge
 (540-13,300 vg/kg) (IARC,  1973,  1983).

 Benzofblfluoranthene
     Production   and Use.   There  Is  no commercial  production  or known use of
 this compound.
 04390                               II-8                             04/16/91

-------
    Occurrence.   Benzo[b]fluoranthene  occurs  ubiquitously  1n  products  of
Incomplete combustion;  1t  also  occurs 1n fossil  fuels.  It  has  been  Identi-
fied  In  surface  water  (0.6-1.1  ng/l),  tap  water  (0.4-5.4  ng/l),   rain
water   (4.4-14.6   ng/l),   subterranean  water  (0.6-9.0   ng/l),   wastewater
(0.04-23.7 ng/l) and sludge (510-2160 vg/kg)  (IARC,  1973).

BenzoFklfluoranthene
    Production and  Use.   There  1s no  commercial  production or known use  of
this compound.

    Occurrence.   Benzo[k]fluoranthene occurs  ubiquitously as  a product  of
Incomplete combustion;  1t  also  occurs 1n fossil  fuels.  It  has  been  Identi-
fied  1n surface  water  (0.2-0.8 ng/l),  tap water  (1-3.4  ng/i), rain  water
(1.6-10.1  ng/l),   subterranean   water  (1-3.5  ng/l),  effluent   discharge
(0.01-8 yg/l) and sludge (150-1270 vg/kg) (IARC,  1973).

6enzo[q.h.1lpery1ene
    Production and  Use.   There  1s no  commercial  production or known use  of
this compound.

    Occurrence.   Benzo[g,h,1]perylene  occurs  ubiquitously  In  products  of
incomplete combustion;  It also occurs In considerable  amounts  1n coal  tar
and  1s   an  Important  component of  gasoline  engine  exhaust.   It  has  been
Identified  1n  surface  water  (0.3-28.5  ng/l),   tap   water  (0.8-7.1 ng/l),
rain  water  (2.3-10.8  ng/l),  subterranean  water   (0.7-6.4  ng/l),  waste-
water   (0.4-2.8   yg/l),   sludge   (200-1220   pg/kg),   freeze-drled  sewage
sludge  samples  (400-8700 yg/kg)  and  dried  sediments  from  lakes  (1-1930
vQ/kg)  (IARC, 1973).
04390                               II-9                             03/22/89

-------
Chrysene
    Production and Use.   There Is no  commercial  production or known use  of
this compound.

    Occurrence.   Chrysene  occurs  ubiquitously   and   In   approximately   the
same concentration as  benzo[a]pyrefie  1n  products of  Incomplete  combustion.
In  addition,  Chrysene  and,  preferentially,  related structures  (methyl-sub-
stituted and  partially hydrogenated  chrysenes)  occur  1n higher  concentra-
tions than most  of  the PAHs 1n fossil  fuels  such as  crude oil and  lignite.
Chrysene has  been Identified  1n  surface water  (7.9-62.0 ng/l),  wastewater
(0.732-6.44  ng/t),  freeze-drled   sewage  sludge  (780-23,700   yg/kg)   and
sediments (40-240 yg/kg)  (IARC, 1983).

D1benz[a.h]anthracene
    Production and Use.   There 1s no  commercial  production or known use  of
this compound.

    Occurrence.   D1benz[a,h]anthracene  occurs ubiquitously as  a  product  of
Incomplete   combustion;  H  also   occurs   1n   fossil   fuels.    It  has   been
Identified  1n wastewater  (IARC, 1983).

Fluoranthene
    Production and Use.   There 1s no  commercial  production or known use  of
this compound.

    Occurrence.    Fluoranthene   occurs   ubiquitously    1n    products    of
Incomplete   combustion;  It  also   occurs   1n   fossil   fuels.    It  has   been

04390                               11-10                             12/31/90

-------
Identified  In  surface  water  (4.7-6.5 ng/l),  tap  water  {2.6-132.6  ng/l),
rain   water   (5.6-1460   ng/l),    subterranean   water   (9.9-100.0   ng/l),
wastewater  (0.1-45  yg/l),  sludge  (580-4090  yg/kg),   freeze-drled  sewage
sludge  (610-5160  yg/kg)  and  dried   sediment   of   lakes   (13-5870  yg/kg)
(IARC, 1983).
                                    *
Fluorene
    Production and Use.   There 1s  no  commercial  production  or known  use  of
this compound.

    Occurrence.   Fluorene  occurs  ubiquitously   In  products  of  Incomplete
combustion; It also  occurs  In  fossil  fuels.  It  has  been  Identified 1n  sur-
face  water  (4.1-102.1   ng/l),  tap water  (4-16  ng/l)  and  sewage  sludge
(0.61-51.60 mg/kg) (IARC, 1973).

Indenon .2.3-cdlpyrene
    Production and Use.   There 1s  no  commercial  production  or known  use  of
this compound.

    Occurrence.   Indeno[l,2,3-cd]pyrene  occurs  ubiquitously  1n products  of
Incomplete combustion; It also  occurs  In  fossil  fuels.  It  has  been Identi-
fied  1n  surface  water  (0.2-0.5  ng/l),  tap water  (0.3-4.8  ng/l),  rainfall
(0.2-8.7  ng/l),  subterranean   water   (0.2-5.0   ng/l),   wastewater  (0.01-15
yg/l),  sludge   (470-1200   yg/kg),  freeze-drled   sewage  sludge   samples
(300-7400   yg/kg)  and  dried  sediment  from  lakes   (1-2070  yg/kg)  (IARC,
1973).
04390                               11-11                             03/22/89

-------
Naphthalene
    Production  and Use.   Domestic  production  capacity  of  naphthalene  was
estimated  to  be 660  million  pounds  annually as  of  January 1,  1984.   Total
domestic  consumption  of naphthalene  for  1985 has been  estimated to  be  540
million   pounds;   this  amount   Includes   Imports   of  8  million   pounds.
Exportation  volumes  are believed  to be  1n  the  range of  5  million  pounds
annually.  Major applications  of  naphthalene Include use as an  Intermediate
1n   the   production   of  phthallc   anhydride  (55%  of  consumption),   the
Insecticide carbaryl  (20X),  B-naphthol  (8%), synthetic tanning  agents (6%),
surfactants  (5%),  miscellaneous  organic  Intermediates  (2%),  and  use as  a
moth repellant (2%) (Chemical  Economics  Handbook,  1981).

    Occurrence.   Naphthalene   Is   the  most  abundant  single  constituent  of
coal  tar  (Schmeltz et  al.,  1978).   It  Is  released  1n  the environment  via
Industrial  gaseous and  partlculate  emissions,  aqueous  waste   streams,  and
through consumer uses.

    Naphthalene has been  detected 1n Industrial  effluents  (up  to 32  mg/i),
municipal  wastewater   treatment  plant  effluents  (22 vg/l),  ambient  river
water  (2.0   yg/J.),   seawater,  drinking  water  (1.4  yg/l),   well   water,
and groundwater (U.S. EPA, 1980c;  Shackelford and Keith,  1976;  Eganhouse and
Kaplan,  1982;  Oeslderl  et al.,   1984).   Stuermer  et al.  (1982)  detected
naphthalene '  1n   groundwater   samples   collected   near   underground  coal
gasification  sites  at  concentrations  of  380-1800  yg/l  15  months  after
gasification  activity  had  ended.   Pankow  et  al.  (1984)   reported  mean
dissolved  naphthalene  concentrations  of  11  and   72   ng/l   1n  rainwater
samples  collected  1n  semlrural and  residential locations, respectively,  In
Oregon.

04390                               11-12                            04/16/91

-------
Phenanthrene
    Production and  Use.   There Is no  commercial  production or known use  of
this compound.  Its derivative, cyclopentenaphenanthrene, has  been  used as  a
starting  material  for   synthesizing  bile  adds,   cholesterol  and   other
steroids.

    Occurrence.  Phenanthrene  1s  present  1n  products of Incomplete  combus-
tion;  H  also occurs  1n  fossil  fuels.   It  has  been Identified In  surface
water   (0-1300  ng/i),   tap   water   (3.1-90  yg/l).  wastewater  (70   yg/a)
and dried sediment 1n  lakes  (140-274  yg/kg) (IARC, 1973).

Pyrene
    Production and  Use.   There Is no  commercial  production or known use  of
this compound.  Pyrene  from  coal-tar  has been used as  the  starting material
for the synthesis  of benzo(a)pyrene.

    Occurrence.   Pyrene  occurs  ubiquitously  1n  products   of  Incomplete
combustion; .H also occurs  1n  fossil  fuels.   It  Is found 1n  relatively high
quantities  1n  coal-tar.    Pyrene has  been   Identified   In   surface  water
(2.0-3.7  ng/i),  tap  water   (1.1  ng/i),  rain  water  (5.8-27.8  ng/i),  sub-
terranean  water   (1.6-2.5  ng/l), wastewater   (0.00023-11.8   yg/l),  sludge
(900-47,200  yg/kg)  and   dried  sediment  from  lakes  (7-3940  yg/kg)  (IARC,
1973).

Summary
    PAHs  are  a  class   of  diverse   compounds   resulting  from Incomplete
combustion.  They are ubiquitous pollutants  and  are found 1n ambient air,
soil and aquatic  systems.
04390                               11-13                           12/31/90

-------
    PAHs are  characterized  by low  volatility,  very slight water  solubility
and a propensity for adsorption to  participates  1n  aquatic  systems.   PAHs  of
four  or "more  rings  (characteristic  of  the  carcinogenic  PAHs)  are  less
amenable to  mlcroblal  degradation  than are smaller  compounds, making  them
environmentally  stable   and,  therefore,   Increasing   the  potential   for
longer-term exposure.  The  high  log octanol/water  partition  coefficients  of
PAHs  Indicate  a  propensity  for   bloaccumulatlon.   However,   animal  data
Indicate that PAHs  tend to be rapidly metabolized and excreted.

    With the exceptions of  anthracene  and naphthalene there are  no  reported
commercial  productions  of  any PAH  described 1n  this  document.
04390                               .11-14                            12/31/90

-------
                             III.   TOXICOKINETICS

    Essentially no  human pharmacoklnetic  data  are available  from which to
develop a  coherent  picture  of  PAH absorption, distribution and elimination.
The experimental  data  base consists  almost  entirely of  studies  In rodents
(mice, rats and hamsters) with  occasional  studies  In larger mammals.  While
there  are  certainly quantitative  differences among  these species  (1n the
relative Importance of  bile and urine as  routes  of excretion, for example)
the  qualitative  picture  that  emerges appears  to  be  much the  same across
species.   PAH   kinetics^, are   characterized  by   rapid,  capacity-limited
metabolism  to  both biologically  active  and  Inactive  metabolites.    Both
Increased  levels  of  enzymes responsible  for  the metabolism  of  PAHs  and
Isozymes of  these metabolic enzymes  are Induclble by  chemical  Inducers of
the 3-methylcholanthrene  class.   Excretion  of chemically stable metabolites
Into  urine  and bile 1s  rapid.   Differences  1n  disposition among different
PAHs   are   associated   at   least  1n   part  with  differences   1n  their
l1pophH1cH1es.
      •
Absorption
    Oral.  Fractional absorption  of  PAHs from the  GI  tract 1s dependent on
the specific hydrocarbon  Ingested.   In  general,  the more llpophlllc members
of  this  class must  be  solublllzed  1n  water before  they  can  be absorbed.
PAHs  with  some  Intrinsic  hydrophlllc  character  may   be absorbed  to   some
degree even In the absence  of emulslflers such as  bile  salts.  Absorption of
all PAHs,  however,  1s  expected to be Influenced  by the properties of other
chemicals  coadmlnUtered or  coabsorbed.    These  Include,  of course,  the
constituents of the diet (Chang,  1943; Modlca et al., 1983).


04400                                III-l                           04/24/91

-------
     Chang's  early  (1943)  work  Illustrates   the  dependence  of  fractional
 absorption on  the  compound  Ingested.   While absorption  of  naphthalene by the
 rat  was  complete,  and  that  of  acenaphthene  and  phenanthrene  nearly  so,
 anthracene,  chrysene,  d1benz[a,h]anthracene,   benzo[a]pyrene  and  3-methyl-
 cholanthrene were  only  5-50% absorbed  either  from  the diet or from a suspen-
 sion  In  starch solution.  Absorption of  benzo[a]pyrene  was -60% from a stock
 diet  containing  1%  benzo[a]pyrene,  and  ~40% when  the  benzo[a]pyrene  was
 given  as the  starch  suspension.   Hodlca  et  al.  (1983)  similarly  observed
 that  chrysene  was much less  efficiently absorbed  than  either benzanthracene
 or trlphenylene by young Tasted rats given the compounds orally as emulsions.

    The  dependence  of PAH absorption on the vehicle  1n which It Is  adminis-
 tered  has been well documented.   A  group of  coordinated studies conducted 1n
 the  early 1950s  (Setala  and  Ekwall.  1950;  Ermala  et  al., 1951; Ekwall  et
 al..  1951; Setala,  1954)  examined the  effects  of single solvents.  Benzo[a]-
 pyrene  was  administered to  mice and cats  by  stomach tube, 1n solution or 1n
-a  suspension  of  natural   fats   (olive  oil,   arachldls oil);  natural  fats
 emulsified  In  bile;  bile alone;  long-chain  fatty  acids and  alcohols (oleyl
 alcohol,  olelc add);  polyethylene  glycols;  and  synthetic  emulslflers  such
 as  the  Triton ethers, among other vehicles.   Animals were fasted  for  at
 least  12 hours  prior  to  benzo[a]pyrene  administration and  throughout  the
 experiment,  so  that  the effect  of  the  solvents  could  be  studied  1n  the
 absence  of  food.   Benzo[a]pyrene was absorbed  well  1n  the forestomach Irre-
 spective of  the vehicle.  Benzo[a]pyrene In  natural  fats or other UpophlUc
 solvents did not penetrate  the  stomach  wall,-which  1s not surprising since
 04400                               II1-2                           09/26/90

-------
 fat Itself Is not  absorbed  1n the stomach under  normal  conditions.   Hydro-
 phobic  solvents, as  well  as  the  hydrophoblc  PAHs,  are unable  to  penetrate
 the protective mucous layer  lining the stomach.

     Olelc  add and  oleyl  alcohol,  predominantly Upophlllc but with  hydro-
 phlUc  groups,  effected a  slight  penetration  of  benzo[a]pyrene  Into  the
 glandular  stomach  wall.   Greater  penetration  was associated with administra-
 tion of  benzo[a]pyrene   In  vehicles  with  both  hydrophlUc and  UpophlUc
 properties;  penetration  was greatest  when  the hydrophlllc character  of  the
 solvent  was  dominant  (Efcwall   et  al.,  1951).   Thus, a  solvent capable  of
 solub1!1z1ng   benzo[a]pyrene  1n   aqueous  solution  1s   required  for   Us
 absorption  1n  the stomach.

     The  effective  solublUzers studied  by this  group  of  Investigators  are
 not present  naturally  1n foods to  any significant  extent,  and  H 1s unlikely
 that much PAH absorption  occurs  In  the  stomach at  all under normal  condl-
                                                                            •
-tlons.   However, once the  PAHs have  entered  the small Intestine, they  are
 solublUzed  by the  bile salts In concert  with  fatty acid anlons  and mono-
 glycerldes  (Laher  and Barrowman,  1983),  and  are absorbed  In both  the small
 and large  Intestines  (Ermala et al., 1951).

     Working  with a  group  of  PAHs of Increasing ring number and  correlated
 decreasing  aqueous  solubility, Rahman et  al.  (1986)  showed that  the presence
 of  bile salts 1n  the rat  Intestine  1s  essential  for optimum absorption  of
 anthracene,  7,l2-d1methylbenzanthracene,  and  benzo[a]pyrene   but  not  for
 absorption   of  2,6-d1methylnaphthalene   or   phenanthrene,  whose   aqueous
 solubilities  are 2-3  orders  of magnitude  greater  than  those  of  the other
 three  hydrocarbons.   When  the bile  was  diverted   from   the  Intestine  by

 04400                                II1-3                           09/26/90

-------
 bile duct  cannulatlon,  benzo[a]pyrene  was absorbed  only  22% as  well  as U
 was  1n  the presence  of  normal  amounts  of  bile.    In  addition,  fractional
 absorption  of  the 4- and  5-r1nged  compounds  7,12-dlmethylbenzanthracene and
 benzo[a]pyrene  was  not  as  great  as   that   of  the smaller  hydrocarbons,
 particularly  In  the  absence of  bile.   Thus,  coadmlnlstratlon  of benzo[a]-
 pyrene  with  food  1s  expected  to  enhance  Us  absorption  as  a  result of
 activation  of the natural  mechanisms  for absorption of dietary llplds.

     There 1s no  convincing evidence that  the absorption of  PAHs occurs by
 mechanisms  other  than  passive  uptake.   Rees et  al.  (1971) proposed  the
 possibility that uptake  might  Increase  exponentially  as   the  amount of
 benzo[a]pyrene  In  the  GI  tract Increases.   This  suggestion  was  based on
 studies   of  the  entry  of benzo[a]pyrene  Into  everted sacs  of  rat  small
 Intestine.   Transfer  of  benzo[a]pyrene  from sac   tissue  to the  enclosed
 medium ^as  proportional to  the  concentration  1n the sac  tissue.  Uptake by
 the  sac  tissue  was not  affected by metabolic  Inhibitors,  Indicating  that
                                                                             •
'uptake  was  not  an energy-requiring  process.   However,  the  amount  bound to
 the  sac  tissue  Increased exponentially  with Increasing  Incubation  medium
 concentration  >200  yH,   suggesting  multilayer  adsorption   to   the  tissue
 surface  at  high  concentrations.  That  such an adsorption/absorption process
 might  be relevant to absorption  In vivo  was suggested  by  the  observation
 that the amount  of  benzo[a]pyrene  found  1n retroperltoneal  fat  and mammary
 tissue  of young  female  rats  given  benzo[a]pyrene Intragastrlcally In sesame
 oil  18  hours earlier appeared to  be exponentially  related to the dose.  The
 In  vitro conditions employed by Rees  et al.  (1971),  however,  were grossly
 unphyslologlc,  and "multilayer  adsorption" cannot be a  significant component
 of  benzo[a]pyrene absorption jri vivo.   Furthermore,  the U» vivo  observation
 of     exponentially     Increasing     tissue     concentrations     Is     not

 04400                                 II1-4                          09/26/90

-------
substantiated by  the results  of  other studies.   Bock  and Dao  (1961)  found
that the concentration of  3-methylcholanthrene  In  mammary  tissue and  mammary
fat of  female rats  was  proportional  to the  Intubation  dose, while Modlca  et
al. (1983) calculated that the area under  the  tissue  concentration curve  for
chrysene was  less  than  proportional  to the  dose 1n blood, liver, brain  and
parametMal adipose  tissue,  and  roughly proportional  to the dose In  mammary
tissue.  Thus, there Is no evidence for greater  than  proportional absorption
with dose.

    Uptake of PAHs  1s  partly  Into thoracic  duct lymph but principally  Into
                                                            •
portal venous blood  (Laher et  al.,  1984).   According to Rees et al.  (1971),
10-20X of  a  10  mg  IntragastMc  dose  of  benzo[a]pyrene entered  the  thoracic
duct lymph of rats.  Daniel  et al. (1967)  found  5% of the  radlolabel  from an
1ntragastr1c  dose   of   8.4   yg  of   14C-labeled   d1benz[a,h]anthracene   In
thoracic lymph during the first 24 hours after administration to rats.   Reid
(1977) recovered 18% of an absorbed dose of  benzo[a]pyrene 1n olive  oil  from
thoracic lymph of rats  given  benzo[a]pyrene Intraduodenally.

    Gastrointestinal absorption  of  PAHs  1s  rapid, as  would  be anticipated
for  compounds  utilizing  absorption  mechanisms   designed  for uptake   of
nutrients.   The concentration  of  benzo[a]pyrene  peaked  In  the liver  ~1  hour
after dosing  In  female  rats  Intubated  with 22.8 mg of  benzo[a]pyrene,  while
the blood  concentration  peaked at 1.5-2  hours, the  lag  reflecting  passage
from liver to blood  (Modlca  et al., 1983).   Excretion of metabolites  1n bile
also follows  rapidly, the  rate generally  peaking by  2  hours Irrespective of
the PAH administered (Rahman et  al.,  1986).   It Is Interesting  to note that
1n general the fraction of total  metabolites excreted In bile rather  than 1n
urine  Increases  as  the PAHs Increase  In  size  and decrease 1n  solubility.

04400                                II1-5                            09/26/90

-------
 Radlolabel  from  i«C-d1benz[a,h]anthracene  given  by  Intubation  continued  to
 appear   In  thoracic  lymph  of  two  rats -for  at least  24 hours,  peaking  at
 around  3-4 hours  (Daniel  et al.,  1967).   That some of  this  radlolabel was
 probably  associated   with  metabolite  reabsorptlon  Is  suggested  by  the
 observations  that radlolabel did not  peak 1n  the  blood  plasma  of these two
 rats  until  ~7 hours after dlbenzanthracene administration,  and  that much of
 the   radlolabeled material   extracted  from  plasma,  urine,  and  bile  was
 metabolized   d1benz[a,h]anthracene.    Rees  et  al.   (1971)  reported  that
 benzo[a]pyrene   peaked   In   the  thoracic  lymph  of  rats   3-4  hours  after
 administration  of 10  mg by Intubation, but >the fluorescence-based analytical
 technique  used  by  these Investigators would not  have  distinguished between
 benzo[a]pyrene  and  Us  metabolites.

 Inhalation
    The  absorption  of  PAHs  from  the lung  has  been  Investigated  1n  both
 Inhalation  studies  and  1n  experiments  1n  which the PAHs were given by Intra-
-tracheal  administration.  Both pure PAHs  and PAHs adsorbed on  particles of
 various  sizes and chemical composition have  been  used.   Adsorption of a PAH
 onto  ultraflne  particles   (<1 v,m)  from  which  1t  can  be  released  Into
 biologic  fluids has been  shown to  Increase both  the lung retention time of
 the  PAH and  Us carclnogenldty.   The nature of  the  adsorbent particle Is
 critical:   adsorbents  that  do not  readily release adsorbed chemicals are not
 as  likely  to  enhance   tumor  yield and  other biologic  effects  1n experi-
 mental  animals  as  are  adsorbents  with  lower affinity  for  the  carcinogen
 (Bevan  and Worrell,  1985).   In  addition  to adsorption  onto  partlculates,
 factors  such as particle or  aerosol   size  and size distribution,  the test
 species'  airway anatomy,  metabolism   and  defense  mechanisms,  as  well  as


 04400                               III-6                           09/26/90

-------
properties of the PAH Itself, will Influence the penetration, deposition and
retention of an Inhaled PAH and the probability and routes of Us subsequent
absorption.  Other chemicals,  Including  those  adsorbed to partlculates, may
also Influence the disposition  and biologic  activity  of PAHs taken Into the
lung.                                                                    ,  H
                                   •
    Kotln  et  al.  (1959)   reported   a   vehicle  effect  on  Intratracheally
Instilled   benzo[a]pyrene.     Rad1olabeled-14C-benzo[a]pyrene    (25   yg/0.3
ml  vehicle)  was  administered  to VMstar  male  rats   either  as  an  aqueous
suspension with distilled  water  or  1n  a solution  with tMethylene glycol.
The  esophagus  was llgated  In  each  to prevent  benzo[a]pyrene  from entering
the  GI   tract  by mucodllary  escalator  clearance and  swallowing.   At  24
hours,   38X of  the  administered  aqueous suspension  dose  persisted  In the
lung, whereas  only  17% could  be recovered at  4  hours  1n the  lungs  when
administered In  tMethylene glycol.   The rapid  mobilization was attributed
to   the  greater  speed  of  solublUzatlon  of   the   fine  particles  from
trlethylene  glycol  vs.   the slower  speed  of  solublUzatlon of  the  larger
crystals that form 1n an aqueous suspension.

    The role of particles as  carriers of PAHs  and enhancers of their pulmo-
nary activity has been  the  subject of a number of  Investigations.  Coadmln-
Istratlon of  PAHs with fine  adsorbent  particles was  shown  nearly  30  years
ago  to  result   1n   a   higher   Incidence  of   lung   tumors  In  rats  than
administration of the PAH alone (Pylev et al., 1969;  Safflottl et al.,  1965.
1968).    Coadmlnlstratlon  by Intratracheal Instillation  of  a  suspension  of
14C-benzo[a]pyrene with  carbon  black or asbestos  (Pylev  et al.,  1969)  or
by Intratracheal  Instillation of  particles of  ferric  oxide, aluminum oxide,


04400                                II1-7                           09/26/90

-------
or carbon  coated  with benzo[a]pyrene (Henry and  Kaufman,  1973)  resulted  1n
prolongation  of  radlolabel  retention  1n  the  hamster  lung.   However,   an
Increase In  benzo[a]pyrene  retention  was  not necessarily correlated with  an
Increase In  tumor 1gen1dty.   Farrell  and  Davis  (1974) showed  that carbon
black was  a  more effective  tumor  promoter than  ferric  oxide 1n the golden
hamster lung; aluminum oxide was only minimally active as a  promoter  In this
system.   In   general,  the  probability of  tumor  occurrence  was  shown   to
Increase as particle size range decreased  when either carbon  black  or ferric
oxide was  used as  the adsorbent  (Farrell and Davis,  1974;  Henry et  a!.,
1974).   In the absence  "of  carrier  particles,  large (77%  <42  ym,  0% <10
ym)  benzo[a]pyrene  particles were cleared less  rapidly from the  lung and
were  more   tumorlgenie  than  small  (77X  <5.2  ym,  3%  <1.3  urn)  particles
(Feron et al.,  1980).  Henry et al.  (1975)  showed that benzo[a]pyrene had  to
be  physically  adsorbed  on  the  ferric   oxide  to  be most  effective  as  a
promoter.

    Henry  and  Kaufman  (1973) suggested that the  ability of  the  hydrocarbon
to be eluted  from  Us partlculate  adsorption  sites might  be an  Important
determinant  of  Us   biologic  activity.   Creasla  et  al.  (1976)  demonstrated
that  the  rate  of elutlon  of benzo[a]pyrene  from  Us  adsorption  sites  on
carbon  particles  In  the  mouse  lung  was  greater   for  small  (0.5-1.0 ym)
particles  than  for   large (15-30  ym)  particles.   Benzo[a]pyrene  adsorbed  to
the  larger  particles  was  cleared with  the  particles  themselves,  with  a
half-time  for elimination of 4-5 days.  However, while the  smaller  particles
themselves   were  cleared  more  slowly  Ui/o^   days),  the  benzo[a]pyrene
adsorbed  to  these   particles  was  cleared  more  quickly  (t,/2=36  hours).
04400                                II1-8                           09/26/90

-------
This  Indicates  that  the  benzo[a]pyrene was  being eluted  from the  smaller
carbon particles at  a rate of 15% per  day.   Thus,  for  at least 4 days,  15%
of the Initial dose  of benzo[a]pyrene was  free  to  Interact  with the  respira-
tory  tissue.   In  the  absence  of carbon particles,  50% of a  benzo[a]pyrene
dose was cleared from the lung within  1.5 hours and >95% was  cleared within
24  hours.    These   results  established  that  a   combination   of  prolonged
retention  time  and  biologic availability  of the  adsorbate  are Involved  In
the cocardnogenlc effect  observed for  partlculate and  benzo[a]pyrene 1n  the
lung.

    Based  on  their   studies  with  model   phosphol1p1d  bllayer  membranes
(Lakowlcz  and Hylden,  1978;  Lakowlcz  et  al., 1980)  and  with mlcrosomal
preparations  (Lakowlcz and Bevan, 1979),  the authors  suggested that cocar-
dnogenlc  particles  facilitate the uptake  of adsorbed chemical carcinogens
by cell membranes.   Later  work by the  same  group  of Investigators (Bevan  et
al.,  1981;  Bevan  and Worrell, 1985) supports  this mechanistic hypothesis.
Chang and Hart (1983), who demonstrated that  the chrysotlle  form of  asbestos
enhanced uptake  of   benzo[a]pyrene  Into human  dermal  Hbroblasts  1_n vitro,
proposed that cocardnogenlclty  Is  mediated  1n  part  by  enhanced  cellular
proliferation.  It  should be  noted that  binding  to cellular  DNA was also
enhanced by  the presence  of asbestos  1n  this  study,   so  that the  proposed
mechanisms  of enhanced uptake  and enhanced cellular proliferation are  fully
compatible  and not  mutually  exclusive.   Other  in  vitro studies (Eastman  et
al., 1983), using hamster  trachea1 epithelial cells  Incubated  with benzo[a]-
pyrene -and  asbestos, showed enhanced benzo[a]pyrene uptake and ONA  alkyla-
tlon 4 days post-treatment, which was  attributed to the  presence of asbestos.
04400                                II1-9                            09/26/90

-------
     Absorption   and   distribution  of  benzo[a]pyrene  following  Inhalation
 exposure  of rats  to  the compound  alone  or after  adsorption  onto particles
 has  been  studied by  Investigators  at  the  Inhalation  Toxicology  Research
 Institute  (Mitchell  and  Tu,  1979;  Mitchell,  1982,  1983; Sun  et  al.,  1983,
 1984).  Mitchell  (1982,  1983) reported the disposition  of  an  aerosol  of 500
 yg   3H-benzo[a]pyrene/i,  1-2  ym  mass  median  diameter,  Inhaled  by  rats
 for  <1  hour.   Clearance  of  radlolabel  from  the  respiratory  tract  was
 blphaslc.   Fifty  percent  of the  radlolabel  was  cleared within  2-3  hours;
 radlolabel  remaining  after  this  time was cleared  much  more slowly,  the
 half-life  being  25-50  hours,  depending on location 1n the lung.  The stomach
 and  small  Intestine  contained  higher  concentrations of  benzo[a]pyrene  0.5
 hour after  termination  of exposure  than  any  other  tissue,  and  feces  con-
 tained  ~10 times  the  amount of  radlolabel  found In urine during the first
 day  following exposure.   The observation  that  the amount of  radlolabel  In
 the  stomach and  small  Intestine  was  at  Us  maximum  Immediately following
 termination  of  exposure suggests  mucodllary  clearance  and  swallowing  of
.Inhaled  material.   The  presence  of  radioactivity  1n  other   soft  tissues
 (e.g.,-kidney  and liver) 0.5 hour  after  exposure Indicates rapid absorption
 and  distribution  of the  benzo[a]pyrene  from Its  Initial site of deposition.

     The  concentrations  of  organic-soluble radlolabel,  water-soluble  radio-
 label  and covalently-bound radlolabel were determined  1n  the  lung at 0.5, 6
 and  24 hours  postexposure  (Mitchell,  1982, 1983).   The 2-3-hours clearance
 half-life  appeared to represent loss of organic-soluble radlolabel; after 24
 hours,.BOX of total lung radlolabel  was  covalently bound to macromolecules.
 Inviting  the speculation that  H Is this  bound radlolabel  that persists 1n
 the  lung  with a  half-life of  25-50  hours.


 04400                                 111-10                         02/07/91

-------
    Similar  Inhalation  studies  demonstrated  that  pyrene  was  also  cleared
from  the  rat  lung  rapidly,  although  not 'as  rapidly  as  benzo[a]pyrene
(Mitchell and  Tu,  1979).  The pyrene  aerosol  had a mass  median  diameter  of
0.45  ym.   The  amount  of  pyrene  left In  the  lung  24 hours after  exposure
was 69X  of  that  remaining 0.5 hour  after  exposure;  at  2 days,  H  was  5%.
Covalent binding was not measured.

    Sun  et  al.   (1983)  examined  the deposition,   retention  and  excretion
patterns  of  the  radlolabel   associated  with  3H-benzo[a]pyrene  coated  on
ultraflne  (0.1  tint  median diameter)  particles  of  gallium  oxide,  Ga?0_.
The purpose of this study was  to determine whether association  with ultra-
fine particles affects the disposition of organic air  pollutants.   Rats were
exposed  by  Inhalation  to  the  aerosols  (3500  ng/i)  for  30  minutes,  and
tissues and excreta were collected for 16 days after the exposures.  Studies
with  pure  aerosols  (600  ng/i)  of   3H-benzo[a]pyrene  of   the  same  median
diameter  were  also '.conducted  for  comparison,  with  the  exposure  period
extended to 50 minutes 1n  order  to Insure  that total  deposition  was similar
to that observed  In  the  particle Inhalation experiments.   About  28% of each
aerosol  had been  deposited  1n   the  respiratory  tract  at the  end of  the
exposure period.

    Initial  clearance  from the  lung  was  rapid,  with  half-lives  of  <1  hour
for both aerosols.   During the subsequent slow excretion  phase,  lung levels
of  radlolabel   resulting from  exposure   to  benzo[a]pyrene-coated  particles
were 4-10 times  higher than  those resulting from exposure to  pure  benzo[a]-
pyrene aerosols,  although the  half-lives,  had  they been  calculated,  would
probably  not   have been  greatly  different.   Lung clearance  of  the  pure


04400                                III-ll                           07/29/90

-------
benzo[a]pyrene aerosol  was  also very  similar  to  that  reported by  Mitchell
(1982)  for  aerosol  concentrations  800 times higher.  Of  particular  Interest
1s  that  levels  of  radlolabel  In the  stomach  Increased  with  time  after
exposure  to coated  particles.  Thirty-five  percent of  the  benzo[a]pyrene
Initially deposited In  the lung was recovered  from the  stomach 2 hours  after
exposure to benzo[a]pyrene-coated particles, while no more  than  ~0.5X of  the
amount  of  pure  aerosolized  benzo[a]pyrene (Initially) deposited 1n  the  lung
was found  1n  the  stomach at  any time  following  exposure.   Thus, exposure  to
a  benzo[a]pyrene-coated  particle   rather  than  to  a   pure   benzo[a]pyrene
aerosol  resulted  In a  very  marked shift in the pattern of  lung  clearance,
from  direct  absorption  Into  blood   to  clearance  by  mucodllary action
followed  by 1ngest1on.  Consequently,  of the  total  radlolabel excreted  1n
urine and feces through  day  16, -15X  was  recovered from the urine  of animals
exposed  to  the  pure aerosols and only ~8X from  the  urine of  animals exposed
to the coated aerosols.

    It  should  be  noted that  pure aerosols of 3H-n1tropyrene,  as  well  as
gallium  oxide  aerosols  coated with  3H-n1tropyrene,   were   studied  1n   a
parallel  set  of  experiments,  with  results  that  were different  from  the
benzo[a]pyrene results.   Thus, It   1s  not advisable to generalize  retention
and  absorption  data  for  a  particular  PAH  to  predict  the  quantitative
behavior of other PAHs or nHropyrenes under similar conditions.

    A  possible  effect of dlesel exhaust particles  on  retention of  PAHs  by
the  lung  Is  currently  of  some  concern  (Tyrer  et al.,  1981).   A  study
reported  by  Sun  et  al.  (1984)   showed  that   the  retention  and  excretion
patterns  of 3H-benzo[a]pyrene-coated  dlesel  engine  exhaust   particles  with


04400                                II1-12                          04/24/91

-------
a   mass   median  diameter  of   -0.14   pm  (4-6  pg/l)   were   qualitatively
similar to  those  recorded  for  the  gallium oxide study.   However, the amounts
of  radlolabel  retained  1n  the  lung   during   the  second,  slow  phase  of
clearance were  much higher 1n  the  dlesel exhaust study  than  1n the gallium
oxide study.  Host of this radlolabel was present as unchanged compound.

    Dermal.   PAHs  can be  absorbed  percutaneously.   The  rate  and  extent  of
absorption  are  strongly  dependent  on   the  size  and  configuration of  the
hydrocarbon  molecule,  and  are  also dependent  on the  concentration of  the
hydrocarbon applied to the skin.

    Heldelberger  and  Weiss (1951)  showed that  14C-benzo[a]pyrene  dissolved
1n  benzene  was absorbed  from  a single  application  site  on the  shaved back
skin of  mice and  excreted  principally   In the  feces.   Fifty percent of  the
radlolabel  from  a  63.7  yg/cm2 dose  of  benzo[a]pyrene  was  lost   from  the
site of application by  1  hour; only 6X  remained after 7  days.   D1benz[a,h]-
anthracene  was  only very slowly lost from the application site  and  was  not
      t                                     -                 •
detected  1n the  feces.   By contrast.  Bock  and  Burnham  (1961)  showed that
7,12-dlmethylbenzanthracene and  benzo[a]pyrene  1n a 95": 1 mixture of benzene
and mineral oil reached similar  concentrations  1n shaved  back  skin  of mice 2
hours after cutaneous  application of -400
    Sanders et  al.  (1984)  confirmed  that benzo[a]pyrene and  7,12-dlmethyl-
benzanthracene are rapidly absorbed across the  skin  of  mice.   Six percent of
a  dose  of  125 yg  benzo[a]pyrene/cm2  In  acetone had  disappeared from  the
application site  by  1  hour  and 40%  by 24 hours, with 7% remaining at  the
site  after  7  days.   Eighty  percent  of  a  5.4 yg  7,12-dlmethylbenzanthra-

04400                                111-13                          11/12/91

-------
cene/cm2  dose  had disappeared  by  24 hours, and  96% by 1 week.   Absorption
was dose-dependent;  Increasing  the  7,12-dlmethylbenzanthracene  dose  from 5.4
to  56  and  to  515  vg/cm2  reduced absorption  during  the  first  24  hours
from 82%  to  71X and  to  33X,  respectively;  decreasing the benzo[a]pyrene dose
from  125  to  12.5  and  to   1.25   vg/cm3  Increased  absorption  during  the
first 24  hours  from  41% to 83X and  to 82X, respectively.  Use  of  muzzles  to
prevent grooming  of  the treated area reduced the fraction of  the  dose found
1n  the stomach  and  Us  contents   to  one-third  of  the amount  found  when
muzzles were not  used;  this  fraction, 0.5-1.5X of the administered  dose, was
so  small  that  1t  had  no   significant   Impact  on  measurement  of  either
absorption or excretion.

    Molecular   size    (number  of   rings)   and    structure   affected   the
concentrations  of  12   different   hydrocarbons.    Including   phenanthrene,
anthracene,  benz[a]anthracene  and   3-methylcholanthrene,  studied  In  mouse
skin  under  the  conditions  used  by Bock and  Burnham  (1961).   While  this
                                                                           •
measurement  reflects  both rate of  uptake by the skin and  rate of  transfer
      /
Into the  systemic  circulation,  1t does  Indicate that  skin penetration occurs
and that  the rate of the overall  dermal  absorption process 1s  dependent  on
molecular size and configuration.   It was  also determined that  the  nature of
the  solvent  makes a  difference 1n the absorption  of PAHs.   Small  amounts
(<5X)   of  mineral  oil  added  to  benzene   resulted  1n  greater   dermal
absorption.  However, the absorption was  not  as great when  a mixture of 50X
mineral oil was used as the solvent.

    Fifty  percent   of   a  single    application  of   9.3  yg  anthracene/cm2
(dissolved  1n  71  yl  of  a   1:7 mixture   of  hexaneracetone)  to  the  shaved


04400                                111-14                          11/12/91

-------
back skin of  female  Sprague-Dawley  rats  was absorbed 1n 5 days, as measured
by the appearance 1n urine,  feces and other  tissues  (Yang et  al., 1986).

Distribution
    Elimination  of  benzo[a]pyrene  and  of  7,12-dlmethylbenzanthracene  from
blood  Is  blphask,  the first  phase being  exiremely  rapid.   About  90%  of
either chemical was lost In the  first  10-30  minutes  following  administration
of an  Intravenous dose  to  mice or rats  (Heldelberger and Weiss, 1951; Kotln
et al.,  1959;  Lo, 1964; Iqbal  et al., 1979).  The  remainder  was  lost  more
slowly.

    Clearance  of metabolites   accounts  for  the  second,   slower   phase  of
decline of  radlolabel  1n the  blood.  Excretion of  metabolites 1n bile and
urine  1s  preceded by  a period  during which  the  liver and kidney  contain
relatively  large fractions  of  the  dose.   This  period  1s brief.   By  90
minutes following  Intravenous  Injection of  14C-benzo[a]pyrene 1n  mice, the
liver  contained  only  1.36% of  the   radlolabel, and  72% had already  entered
the  bfle  (Heldelberger  and Weiss.  1951).    D1benz[a,h]anthracene  was  less
rapidly metabolized; 90 minutes after an Intravenous dose was  given to mice,
89% of the radlolabel  was still In the liver  (Heldelberger and  Weiss,  1951).

    Selective accumulation does  not  occur  In target tissues of PAH carcino-
genic  action.    Depending   on   the   I1poph1l1c1ty  of  the   PAH, significant
accumulation 1n  body  fat 1s  likely to  occur;  early authors attributed the
Induction of mammary tumors by  PAHs  to the  fact  that the mammary gland  1s  In
Intimate association  with   fat  (Bock and  Dao,  1961; Daniel  et al.,  1967).
Bock and  Dao  (1961)  found  Uttle phenanthrene  1n  the perlrenal and  mammary


04400                                111-15                          05/13/91

-------
fat of rats given  the hydrocarbon by  stomach  tube  24  hours  earlier.   Concen-
trations  of  7,12-dlmethylbenzanthracene  and 3-methylcholanthrene were  some-
what  higher,  and benzo[a]pyrene was  present  In the  grestest  concentration.
These  concentrations  correlate with  the relative  I1poph1l1c1ty of  phenan-
threne,  7,12-dlmethylbenzanthracene,   and  benzo[a]pyrene  (Rahman   et   al.,
1986).

    Oral.  When  radlolabeled  PAHs  are  administered   Into  the  GI  tract  by
gastric  Intubation  or  by  feeding,  first-pass  metabolism destroys the  sharp
Initial  drop   In radlolabel  characteristic of  Intravenous  administration.
Instead, radlolabel peaks  after several  hours;  this radlolabel  Is associated
with metabolites.  The timing of the  peak depends  on  the particular  PAH, and
1s expected to depend  on the dose  as  well.   It has been  seen  to occur  at ~7
hours  for d1benz[a,h]anthracene and 15 hours  for  7,12-dlmethylbenzanthracene
In rats (Daniel et al.,  1967).

    When a specific  analytical  technique (cyclohexane extraction of  adipose
tissue'followed  by  gas  liquid chromatography) 1s used for  the determination
of unchanged  PAH,  the peak  time occurs  earlier:   at -1.5  hour  for  benz[a]-
anthracene 1n  rats  (Hodlca et al.,  1983),  and  at  ~1  hour for benz[a]anthra-
cene and chrysene and 2.5-3  hours for trlphenylene 1n rats  (Bartosek et al.,
1984).   The  concentrations of  these  PAHs  In  the  liver   peak  shortly before
their  peaks 1n the blood  (Hodlca et  al., 1983; Bartosek et al., 1984).  For
this  reason,  the rate  of  appearance  In  the  bile also  tends to peak  early
(1-2 hours) for  most PAHs  (Kotln et al., 1959;  Rahman et al., 1986).  As far
as can be determined from published  data,  distribution   from  the blood Into
peripheral  tissues  follows  established  principles:   well-perfused  tissues


04400                                111-16                           05/13/91

-------
establish a dynamic  steady  state  with the blood early, while  adipose  tissue
exchanges more  slowly.   Peak adipose  tissue  concentrations occurred at  2.5
hours  for  chrysene,  ~4  hours  for  benz[a]anthracene  and   8   hours   for
trlphenylene (Bartosek et  al.,  1984), and at  3 hours  for  benz[a]anthracene
1n another study (Modlca  et al., 1983).

    No kinetic  evidence  for  this peripheral  distribution  1s  found In  the
concentrations  of PAHs In the blood after a single  dose,  which can be  fit by
a  blexponentlal  equation with  one  term  representing  uptake  and the  other
loss  (Modlca  et  al.,  f983; Bartosek et  al., 1984).  Half-lives  In  rat
tissues appear  to  be about  the  same  for  benz[a]anthracene and chrysene:   1
hour  1n  blood,  0.8  hour  1n liver, 2.5  hours  1n brain,  5  hours  In mammary
tissue, and  14 hours  In adipose tissue  (Modlca et  al.,  1983;  Bartosek  et
al., 1984).

    Inhalation.   As  has  been  discussed  earlier (see  Absorption  Section),
the  pattern  of distribution  following a pulmonary exposure  could  resemble
the  pattern  associated   with  Intravenous  administration   or  the pattern
associated with  oral exposure,  or  could  be   Intermediate  between the  two.
Kotln  et  al.  (1959)  Instilled  14C-benzo[a]pyrene   Intratracheally  1n  tr1-
ethylene glycol  Into rats  whose  esophagi had  been  Ugated to prevent  entry
of benzo[a]pyrene Into the  GI  tract.   Except   for the high  Initial pulmonary
concentration,   the  pattern  of  tissue radlolabel was  similar  to that  seen
after  subcutaneous or  Intravenous administration.   One hour after Instilla-
tion.  37%  of  the radlolabel  was 1n  the  Intestine,  1.3% In  the  liver,  and
43.2% remained  1n the lung.
04400                                111-17                          05/13/91

-------
    The study of  Kotln  et  al.  (1959) demonstrates that pulmonary  absorption
1s  rapid.   Mitchell  (1982)  found  radlolabel  In  soft  tissues  by 0.5  hour
after  a  1-hour  Inhalation exposure  of rats  to 3H-benzo[a]pyrene.   Radio-
label  1n  testls,  brain and  kidney  peaked  at  ~6  hours  and  was  generally
cleared from  these  tissues  1n  ~1   day.   At  the  0.5-hour  time point,  the
radlolabel  In  liver and kidney  wa?  largely polar  metabolites  of  benzo[a]-
pyrene (Mitchell,  1983).   In  every  salient point, this distribution  pattern
and timing  resemble the pattern and  timing of  distribution of  an Intravenous
dose.  It can be  concluded that an  Inhaled aerosol of  pure benzo[a]pyrene Is
distributed essentially like  a  comparable  Intravenous  dose  save for  the
Initial high  pulmonary  concentration.   Inhaled  pyrene  aerosol  also  behaved
much  like   an  Intravenous  dose  In   rats,  although a  substantial  amount  of
pyrene was  found  In  the  stomach 0.5 hour after exposure  (Mitchell and  Tu,
1979).

    Clearance  from the  lung  Into the GI tract  Is  not  as  rapid  a  process as
absorption  Into  blood.  Only when PAHs  are Inhaled adsorbed  to partlculates
       /
does  their  residence  time  In  the lung  allow significant  transfer  to  the 61
tract.  Sun et  al.  (1983) showed  that during  the  first  2  hours after  an
Inhalation  exposure  to   3H-benzo[a]pyrene   adsorbed  onto   gallium   oxide
particles,  radioactivity 1n  the stomach Increased.  In addition,  radlolabel
In  the livers and kidneys  of  these  rats was 5-6 times  higher at  0.5  and 24
hours  than  radlolabel  1n  the  liver  and kidneys of  the  rats  Inhaling  pure
benzo[a]pyrene aerosol.  Although other tissue  measurements  were  not made,
these  observations strongly   suggest  that  the  distribution  pattern  of  an
Inhaled  PAH  adsorbed  onto  partlculate material  will   have  some  of  the
characteristics  of distribution associated with oral exposure.   Inhalation


04400                                Ill-IB                          05/13/91

-------
studies  1n  rats  exposed   to  benzo[a]pyrene  adsorbed  onto  dlesel  engine
exhaust particles (Sun et al.,  1984) led to the same conclusion.

    Tyrer  et  al. (1981) examined  the effect  of  dlesel exhaust exposure  on
the  disposition   of  1AC-benzo[a]pyrene  administered  Intratracheally   1n
gelatin  solution  (20  »il   of   0.2X  gelatin  and  0.5  v9/vl  of   benzo[a]-
pyrene)  to A/J mice.  Mice were  exposed to raw  dlesel  exhaust (6  mg/m3  of
particles)  8   hours/day, 7  days/week  for 9  months  prior  to  Intratracheal
Instillation  of  benzo[a]pyrene.    Autoradlography  sections  were  processed
from mice  killed  at  2.24  and  48  hours post-Instillation.   Qualitatively
there were  no  obvious  differences  between mice exposed  to  both  dlesel  fumes
and benzo[a]pyrene or  to benzo[a]pyrene  only,  possibly because  of  the  large
1nter1nd1v1dual  variance In  expectorate  1ngest1on.   Within  2  hours  after
placement  1n  the lungs,  benzo[a]pyrene was  found circulating In the  blood-
stream  and may  have  entered  the  esophagus  and stomach  by swallowing  of
material  cleared  by  the   mucodllary  escalator.    14C-benzo[a]pyrene  was.
cleared  from   the  blood  by  the kidneys  and  then  excreted.   By  24  hours,
3H-rad1oact1v1ty  accumulated  In  the  stomach,   lower  GI   tract,   kidneys,
bladder and slightly 1n the  lung.

    Dermal.    Mitchell   and  Tu   (1979)   found   trace   amounts  of   pyrene-
equlvalent  fluorescence  In  kidney,  liver  and trachea of rats  24 hours  after
a  single  application of  pyrene  to  the unshaved  back.  No  pyrene or metabo-
lites were detectable In the lung.   The GI tract  contained  1.4X of  the dose.

    Six  days   after  application  of  anthracene to  the  shaved  back skin  of
rats, Yang  et  al.  (1986) recovered a total of 1.3% of  the  applied  dose from

04400                                111-19                          05/13/91

-------
15  (unspecified)  selected  tissues.   At  this  time,  29.1% of the applied dose
had been recovered In the urine and 21.9% In  the  feces.

    To summarize,  a  number of  studies  with  distribution data Indicate that
1) detectable  levels  of PAHs  can  be observed 1n  most  Internal  organs from
minutes  to  hours  after  various   routes  of   administration;  2)  adipose  and
mammary tissues are significant distribution  sites where PAHs may  be  accumu-
lated, stored  and  slowly  released; and  3) after Inhalation exposure,  the GI
tract contains relatively high  levels of  PAH  or  metabolites as the result of
swallowing  unmetabollzed  PAH  from mucoclHary  clearance,  or  derived from
hepatoblHary excretion of metabolites.

Metabolism
    In the  past,   the  relative lack of  chemical reactivity for  tumorlgenlc
PAHs  has  been puzzling In  light  of their dramatic biologic effects.   Early
attempts  to explain  the  carc1nogen1c1ty of  various  PAHs utilized  physico-
chemical   calculations  (Pullman and  Pullman,  1955).   These early  hypotheses
were  based  on the assumption  that  those regions of  the molecule  favoring
substitution  or addition  reactions would preferentially react with  critical
cellular  target   sites  to  Initiate a   carcinogenic  transformation.  This
concept, however,  did not prove successful for PAHs.

    More  recently H  has  been shown  that PAHs are  metabolized  by  enzyme-
mediated  oxldatlve mechanisms  to  form  reactive electrophlles  (reviewed  1n
Conney,   1982;  Gelboln,   1980;   IARC.   1983;  Pelkonen   and  Nebert,  1982;
Santodonato  et a!.,  1981;  Zedeck,  1980).   For many of  the PAHs,  certain
"bloactlvated" metabolites  are  formed   having  the capability  for  covalent


04400                                111-20                          05/13/91

-------
Interaction  with  cellular  constituents  {I.e.,   RNA,   DNA.   proteins)  and
ultimately leading to tumor formation.

    The obligatory Involvement of metabolic activation for  the  expression  of
PAH-lnduced carclnogenesls  has  prompted the Investigation of  PAH  metabolism
In numerous animal models and human  tissues.   From  these  studies  has  emerged
an understanding of the general  mechanisms  Involved  In PAH blotransformatlon.
It 1s  now  known  that PAHs are metabolized  by  the cytochrome  P-450-dependent
mlcrosomal mixed-function oxldase (HFO)  system, often designated  aryl  hydro-
carbon hydroxylase.  The-activity of this enzyme system  1s  readily Indudble
by  exposure  to  PAH  and   1s  found  1n  most  mammalian  tissues, although
predominantly In the liver.  The MFO system 1s  Involved  1n  the metabolism  of
endogenous  substrates  (e.g.,   steroids)  and  the   detoxification of  many
xenoblotlcs  (Nebert  et al., 1981).  Paradoxically,  however,  the  MFO  system
also  catalyzes  the  formation  of reactive  epoxlde  metabolites from  certain
PAHs, possibly leading to carclnogenesls 1n experimental  mammals.

    The route by which PAHs and other  xenoblotlcs enter  the  body  may  deter-
mine  their  fate  and organ  specificity.  A compound absorbed from the lungs
may  bypass  the "first  pass"  effect 1n  the liver  and  reach the  peripheral
tissues In  high concentrations.  The unique  enzymes In  these  extra-rhepatlc
tissues may  differ  In terms of  activity and  specificity, from those  In the
liver.  Thus,  extra-hepatic metabolism may  be extremely  Important   In the
b1oava1labH1ty of a chemical  to different parts  of the body and 1n  target
tissue variability.   Enzymes  capable of metabolizing PAHs  are found  In the
liver,  lung,  kidney,  adrenals,  testes, thyroid, skin,  small  Intestine and
sebaceous  gland  1n  a  variety of  species   Including  human, baboon,  monkey,


04400                                II1-21            .             05/13/91

-------
rat, mouse,  hamster,  guinea pig,  rabbit  and  dog.  Embryonic  tissue  of  rat,
mouse, hamster and chick also possess activity (Zedeck,  1980).

    Benzo[a]pyrene  as  a  Model  of  PAH  Metabolism.   A  discussion  of  the
metabolism of PAHs  1n mammalian  species,  Including  humans,  may be approached
by  examining 1n  detail the  chemical  fate  of  the  most representative  and
well-studied  compound  1n  the alternant  PAH  class,  namely  benzo[a]pyrene.
The metabolism  of benzo[a]pyrene,  subject  of voluminous research,  1s  shown
schematically 1n  Figure III-l.   Only  the  most  Important  pathways will  be
presented  1n  this discussion that  correlates  with  Figure  I.II-2.   This  1s  a
summary and  the  reader  Is  referred  to  Gelboln (1980),  Pelkonen  and  Nebert
(1982), Yang  et al.  (1978), Zedeck  (1980)  and Thakker et  al.,  (1985,  1988)
for primary sources.

    A  monooxygenase  first  Introduces an  oxygen atom  Into any  of  several
positions  of  the molecule  to  produce oxides  or primary "simple" epoxldes.
This Initial  oxygenatlon 1s  catalyzed by one  of  a  number of  different  forms
      /
of  P-450  (MFO).  The epoxldes  then  undergo  spontaneous  rearrangement  to
phenols. Another  pathway for  the  epoxldes 1s  the  reduction  back  to  parent
benzo[a]pyrene.    The  formation  of qulnones  through  the 6-phenol  and  6-oxo
radical  1s less  well characterized  and  1s  discussed briefly 1rv  the  Other
Tox1f1cat1on  Pathways  Section.   Qulnones can  also  be  produced nonenzymatic-
ally  by  I1p1d   peroxldatlon  and   aerobic  oxidation.   The  enzyme  epoxlde
hydrolase  can further metabolize  epoxldes to  dlhydrodlols.  The glutathlone
transferases catalyze conjugation of epoxldes  with glutathlone.
04400                                II1-22                          05/23/91

-------
                  (Endoplasmic

                     Reticulum)
Cytochrome P-450
Mixed-Function Oxidase (MFO)
              Glutathione
  BaP — SQ -^	
petoxification    Transferase
  Products)      (Cytosol)
                           BaP OXIDES
         BaP PHENOLS
                                  Epoxide
                                  Hydrase
                                  (Endoplasmic
                                  Reticulum)
Sulfates
glucuronides
                                I
                                           BaP Quinones
                     BaP DIHYDRODIOLS (Proposed Proximate Carcinogens)
                   MFO
     UDP Glucoronosyl Transferase
        (Endoplasmic Reticulum)
          BaP OIOL EPOXIDES
          (Proposed ultimate
            Carcinogens)
       hU>Soluble Conjugates

       (Detoxification Products)
                              FIGURE ni-i

                      Metabolism Schematic for PAHs

                         Source: U.S. EPA, 1980d
04400
111-23
                                                               04/24/91

-------
o
o
                               CM

                             [OlQ)
      9-OH
                          ®&j
                           '-<* > *.
               3-OH
 'o^\\
3.6-0  \\        ,
     IG-PHENOXY]
     L RADICAL J^-j.  CM  .1
                     6-OH
                                          7-OH
                   T  9, K) epox    9. K)-diol
                                                                    M
                                                                   »« iJCJH
                                                                      n
                                     foTol
                                    LQlOT
                                                                                   4,5-tfiol
                                                                                         9,»fpn
                      7.8-epox
                                                                       ?;8-diol
-^ [?(8-4ol-9,IO e(J
   r    •    n
   [?,§,9,IO-lelfol|
                          	T~
                          CONJUGATES
                 —I	
                  BOUND MACROMOLECULES
                         DNA
                         RNA
                         PROTEIN
ro
       FIGURE  III-2

Metabolism of Benzo[a]pyrene
       •

  Source:  U.S.  EPA, 1980d

-------
    Epoxldes, dlhydrodlols,  phenols  and qulnones  are generally regarded as
"primary" metabolites  of benzo[a]pyrene,  which  undergo further metabolism.
Dlhydrodlols and phenols can be substrates for the  MFO,  and  another  position
of the molecule can become epoxldated  (see following  discussion  of pathway).
Dlhydrodlols, phenols  and qulnones  can also  be  conjugated  with  glucuronlc
add  1n  reactions  catalyzed by  UDP-glucuronosyltransferase  or  with  sulfate
1n  reactions  catalzyed  by  sulfotransferase.   Dlhydrodlols  can  be  dehydro-
genated to catechols by a soluble dehydrogenase.

    The  metabolites  of  benzo[a]pyrene and  other  PAHs  (both  alternant  and
nonalternant) are  often categorized on  the  basis  of solubility 1n  various
laboratory  extraction  protocols  (U.S.   EPA,  1980d).   Those metabolites  that
can be extracted from an aqueous  Incubation mixture using  an organic  solvent
Include the phenols, dlhydrodlols, qulnones and hydroxymethyl  derivatives of
PAHs  having aliphatic  side  chains.   Epoxldes are  Included  1n this  group
although these are rather labile.  The conjugation  products, Including  those
of glutathlone, glucuronldes and  sulfates, remain  In the  water phase  after
extraction.    Identification  and  quantification   of   metabolites  1n  these
fractions provide  researchers  with  Insight  on the  predominant pathways 1n
the metabolism of  a particular  PAH.

    The  pathways   leading  to  the 7,8-d1ol-9,lO-epox1des  of  benzo[a]pyrene
have  been  a focus  of  research  since   1t  was  suggested that  these   are  the
predominant  DNA-blndlng species  In  cell  culture  (S1ms and Grover,  1974).
For example the major  adduct  Identified  1n  human,  hamster,  rat  and  mouse
endometMal   DNA,   exposed  In  vitro to benzo(a)pyrene,  was  antl-benzo(a)-
pyrene-7,8-d1ol-9,lO-epox1de  (Kulkarnl  et  al., 1986).   The amount   of  this


04400                                111-25                          05/23/91

-------
adduct  formed  In hamster  and  mouse DNA was  similar;  human  levels were  3-4
times  higher and rat levels ~5-7  times  lower.   The quantity of B[a]P  bound
In humans was 7  times the  amount  bound  1n  hamsters,  while  rats  and mice were
4  and  2 times  higher  than hamsters,  respectively.   This  study also  showed
that differences exist  1n  the  proportions  and types of  other adducts  formed
t>y these species.

    Although the 4,5-ox1de  1s the  most mutagenlc  metabolite  1n  bacterial
systems relative to  the  7,8-and  9,lO-ox1des,  other data from assays such  as
DNA   binding   studies,   cell   transformation  assays  and   cardnogenlcHy
bloassays support  the  7,8-d1ol-9,lO-epox1des  as ultimate  carcinogenic  forms
of  benzo[a]pyrene  (Pelkonen  and  Nebert,  1982;  IARC,  1983;  Conney,   1982;
Kulkarnl et  al.,  1986;  Thakker  et al.,  1988).   The metabolic  activation
scheme  for  the  formation  of  the  7,8-d1ol-9,10-epox1de 1s  shown   1n  Figure
III-3.   The  7,8-d1ol  formed   from  benzo[a]pyrene   1s  the  trans  Isomer  and
there  are four  possible  stereolsomeHc 9,10-epoxldes derived  from  this.   The
structure and nomenclature of  these d1ol-epox1des  are also  shown  1n  Figure
III-3.  The two  Isomers  (antl  and syn). also referred  to  as  d1ol-epox1des  I
and  II, are  racemlc and   have  been  synthesized   In  order   to study  their
stereochemistry..  D1ol-epox1de II  (£y_£-1somer)  has  been  found  to be more
chemically  reactive  than   the d1ol-epox1de  I   (Gelboln,  1980).   Both  are
unstable In  aqueous  media and are  hydrolyzed  nonenzymatlcally to  tetrols.
In the presence of NADPH or NADH they are nonenzymatlcally  reduced  to trlols.

    The  Idea   that   7,8-d1ol-9,lO-epox1de  metabolite  1s  the  most  probable
ultimate carcinogen  gave rise  to  the "bay-region"  hypothesis  (JeMna et al.,
1978)  for  the  metabolic toxlflcatlon of  PAHs.   The theory was based  on  the


04400                                111-26                          05/23/91

-------
r\>
o
^
v.
r>o
*.
•x.
(O
                        Metabolic formation of the 7,8-dlol-9,10-epoxldes of benzo(a)pyrene
                                                                                o«
                                                                  P-450
                                                                                   H

                                                                        (*)-BP 7,8-dlol-9,10-epoxld«-2
                                 (*)-BP 7,8-oxlde
                                                  (-)-BP 7,8-dlhydrodlol
                            P-450
                                                                                   H

                                                                       (-)-BP 7,8-dlol-9,10-«poxlde-1
             Bonzo(a)pyr«n«
                            P-450
                                                                  P-450
                                                                             7,8-dlol-9,10-epoxlde-1
                                 (-)-BP 7,8-oxlde
                                                  (*)-BP 7,8-dlhydrodlol
                                                                              /
                                                                              OH
                                                                                  OH
                                                                             Potential
                                                                             Reactions

                                                                               Trlols
                                                                              Binding
                                                                                  to
                                                                           Macromolecules
                                                                              Tetrols
                                              (-)-BP 7,8-dlol-9,10-epoxlde-2



                                FIGURE  III-3


Mechanisms  of  Enzymatic Activation of Benzo[a]pyrene  to 7,8-D1o1-9,lO-Epox1des


                   Source: Adapted^pm Yang  et al., 1978

-------
assumption that the unusually high chemical reactivity of such d1ol-epox1des
and their high susceptibility to attack by nucleophlles can be attributed  to
their electronic  properties.   The bay-regions  of  four PAHs, benz[a]anthra-
cene, benzo[a]pyrene, chrysene and  phenanthrene,  are shown  In Figure  III-4.
The  Implications  of  the  bay-region  hypothesis for  the  mechanisms whereby
                                  *
PAHs may  produce cancer  are discussed  1n  Chapter  VII.  Table  III-l   lists
examples  of   the  enzymes  Involved   In  the  toxlflcatlon  of  PAHs  from the
production of dlols, phenols  and  qulnones, and  dlol-epoxldes.

    Metabolism of Nonalternant  PAHs.  The cardnogenlcHy -of the  alternant
PAHs may  be  related to  the  Intrinsic  cardnogenlcHy  of  the bay-region
dlol-epoxldes and the extent to  which  the parent  compounds  are metabolic-
ally converted  to  the  bay-region  dlol   epoxldes   (Thakker  et  al.,   1988).
However.  H   should  be  noted  that  there  1s  Increasing  evidence  that the
bay-region theory of activation does not  appear to  be  the principal route  of
activation for nonalternate  PAHs  such as benzo[b]fluoranthene and  benzofk]-
fluoranthene and that other  mechanisms of  activation are  Involved  for  those
     /
PAHs that  are devoid of  a  bay-region,  such  as  1ndeno[l,2,3-cd]pyrene and
benzo[k]fluoranthene (Am1n et al.,  1982; Geddle et al.,  1982; Hecht et al.,
1980; LaVole et  al.. 1980; R1ce et al., 1985a,  1986. 1987a,b).

    Other  Toxlflcatlon  Pathways.   In addlton  to  the HFO  mediated  formation
of  dlol-epoxldes  and  other oxygenated products,  other routes of metabolism
have been  shown  for  PAHs.

    One-electron oxidation of  PAHs  can  also  form  electrophlUc metabolites
that can  bind to macromolecules  (CavalleH  and  Rogan,  1983).   The  1nter-


04400                                II1-28                          11/12/91

-------
 BENZ [aJANTHRACENE
          BENZO[a]PYRENE
     CHRYSENE
        PHENANTHRENE
                         FIGURE III-4



                   The Positions of Bay-Regions



                   Source: JeMna et al., 1978
04400
111-29
05/13/91

-------
                                  TABLE III-l

          Examples of Metabolism of PAHs to Biologically Active Forms
                           by Various Enzyme  Systems*
          PAH
  Toxlflcatlon
    Enzymes
Biologically Active
Intermediate-
 Benzo[a]pyrene
 Benz[a]anthracene
.Chrysene
 D1benz[a,h]anthracene
monooxygenase

monooxygenase,
epoxlde hydrolase

monooxygenase

monooxygenase,
epoxlde hydrolase

monooxygenase,
epoxlde hydrolase

monooxygenase,
epoxlde hydrolase

monooxygenase,
epoxlde hydrolase
4,5-oxlde

7,8-d1ol-9,lO-epox1de


6-oxorad1cal

3,4-d1ol-l,2-epox1de
8,9-d1ol-lO,ll-epox1de

!,2-d1ol-3,4-epox1de


3,4-d1ol-l,2-epoxlde


[I0,ll]-d1ol-[l2,l3]-epox1de
 *Source:  Adapted  from Pelkonen  and  Nebert,  1982
 04400
         111-30
                    05/13/91

-------
mediates  In  these  pathways  are  radical  cations.   Benzo[a]pyrene tox1f1cat1on
by  free-radical  metabolism  (see  Figure  III-2)  occurs at  position 6.   The
6-hydroxybenzo[a]pyrene  Is  unstable   1n  solution   and   rapidly  undergoes
radical  formation   to  the  6-oxobenzo[a]pyrene.   This  radical  may  Interact
with macromolecules  or be  converted  to  qulnones  (Jeftlc  and  Adams,  1970).
PAH  qulnones  have  been   proposed  to  be  biologically  active  because  of
activity  1n  oxidation/reduction cycles  Involving  qulnone,  hydroxyqulnone and
molecular  oxygen  (Lesko et  al.,  1978).  The  peroxides and  oxygen  radicals
formed  during these  cycles may  be responsible  for  the  ultimate  cellular
Injury.   This oxidation/reduction  cycle 1s  shown  1n Figure  III-5.   PAHs,
being  photoreactlve,  can   absorb  visible  light  and  become  photooxldlzed.
This photooxldatlon  can result  In the  formation of  qulnones,  dlhydrodlols
and phenols (see Chapter 7).

    Marnett et al.  (1978)  and  Marnett  and Reed (1979)  demonstrated  oxygena-
tlon  of  benzo[a]pyrene catalzyed  by  prostaglandln  H  synthesis-dependent
oxidation  of  aractildonlc   add  metabolism  pathway.    Benzo[a]pyrene  thus
activated  has  been shown  to form  covalent  bonds  with DNA and  1s mutagenlc.
Arachldonlc  acid  has been  shown  to be  released  upon  membrane pertubatlon.
This metabolic pathway  1s  particularly Intriguing  for  Instances  such  as
occur  1n  the  lung with  co-exposure to  PAHs  and  Irritants  or  partlculates.
These  Irritant  chemicals   may  affect  membrane  physiology  such  that  the
metabolism of  the  PAH  by  prostaglandln synthetase could be  enhanced locally
1n the  extrahepatlc  tissue. For  example, Uarshawsky et al.  (1984)  examined
the  effects  of Fe^O. .on   tne  rate °f  metabolism of  benzo[a]pyrene  1n  the
lungs  of  male white New Zealand  rabbits.   Benzo[a]pyrene  was administered
Intratracheally to an  Isolated  perfused lung  preparation  with  and  without


04400                                II1-31                           05/13/91

-------
         NAD
     NADH
BWPHYDROQUINONE      .BfrJP  QUINONE
                   CELLS
       DNA BREAKAGE
        CELL DAMAGE
                      FIGURE III-5

     Cyclic Scheme of Benzo[a]pyrene D1one/D1o1 Involvement 1n Redox Coupling

                  Source: Lesko et a!., 1978
   04400
111-32
04/24/91

-------
FepOg.   After   180   minutes   the  Fe203  pretreated  rabbit  lungs  showed
an  Increase  of   benzo[a]pyrene  metabolism   over   the   benzo[a]pyrene-only
treated rabbit  lungs.   The benzo[a]pyrene metabolites, namely dlhydrodlols,
were particularly enhanced by  pretreatment with Fe.O..

    Hydroperoxlde-dependent   epoxldatlon   of   the   7,8-dlol   has    been
characterized   1n   several   other   systems,   such   as   mlcrosomal   llpld
       s
peroxldatlon  (D1x  and  Marnett,   1983),  hematln catalyzed  decomposition of
fatty  add  hydroepoxldes  (Dlx  et  al.,  1985)  and  Upoxygenase  catalyzed
epoxldatlon (Hughes et al., 1989).

    The  tox1f1cat1on  of  benzo[a]pyrene  by  hydroxymethylatlon  catalyzed by
"hydroxymethyl synthetase" has also been suggested, but the presence of  this
pathway has not been confirmed (Rogan et  al.,  1980).

    Both  y-rad1at1on   and  UV-llght  convert  benzo[a]pyrene  to   reactive
                                                                           •
forms, which  bind  covalently  to  macromolecules  and are  mutagenlc, but  the
significance of these mechanisms  1n mammalian  species  Is  not  known  (Pelkonen
and Nebert,  1982).   Metabolites  could  also  disturb  DNA structure without
binding  covalently  by   slipping  1n   between  the  planes  of  the   helix
(Intercalation).   PAH  tetraols  have  been   shown  to   do   this   In   vitro
(Geadntov et al.,  1980).

    Comparative Metabolism.   One  factor  that affects  the  delicate balance
of  toxlflcatlon/detoxlflcatlon   1s  the   tissue  site  that  catalyzes   the
formation  of  chemically  reactive metabolites.  The degree  of  Induction or
Inhibition of  enzyme systems  by  exogenous  chemicals  1s known  to  vary  for
04400                                111-33                          05/13/91

-------
 specific  organs;  for  example,  liver  1s  generally  more  Indudble than  1s
 lung.   The stability  of  reactive metabolites  Is  also an  Important  factor.
 If  the  biologic  half-life  1s   sufficiently  long,   the  metabolite  may  be
 transported  to  other  tissues  for  action or  further metabolism.   Several
 reactive  metabolites  1n  one  tissue may  compete  for  sites.   Differences  1n
 site of administration will also  affect metabolism.

    Freudenthal  et al.  (1978)  examined  ttie metabolism of  benzo[a]pyrene  by
 lung  mlcrosomes  Isolated  from  humans,  rhesus  monkeys  and  Sprague-Dawley
 rats.   Human samples  were normal  lung  tissues  of  subjects  with  pulmonary
 tumors.   The  metabolites  of  14C-benzo[a]pyrene  were  analyzed  by  HPLC.
 Large  Individual variation 1n  amounts  of metabolites produced  among  monkey
 and  human samples were  seen,  probably  due to  the genetic  heterogenldty  of
 the  test  subjects.  Human subjects also  had  a wide  range of  age, were  of
 both  sexes and  had  dissimilar environmental  contacts.   Qualitative differ-
 ences  between  the types  of metabolites  produced by  humans  as compared with
                                                                             •
-those  produced  by the other  animal species were  less dramatic.  This study
 demonstrated  that  the  validity  of  comparison across  species  may  not  be
 compromised  1f  caution   Is  paid  to  the considerations   of   age,  gender,
 exposure  to  Inducing agents and other factors  (U.S. EPA, 1980d).

    That  species  differences  1n metabolism must  be  considered In carcino-
 genic  potency,  however.   Is demonstrated  for  chrysene by  the  work of Weston
 et  al.  (1985).   They  used short-term cultures  of rat  skin (In which chrysene
 Is  not a strong  carcinogen),  mouse skin  (1n which  chrysene  does cause a
 carcinogenic  response) and human skin.   For  each skin type a major pathway
 leading  to  DNA  adduct  formation Involved formation  of  chrysene bay-region


 04400                                 II1-34                          05/13/91

-------
d1ol-epox1des    (chrysene-1,2-d1ol-3,4-epox1de).    Mouse    skin,    however,
released  larger  quantHHes of free dlhydrodlols  to  the medium than  either
rats  or  humans.   Moreover,  there  was  greater  chrysene  adduct formation  1n
mouse skin.

fxcretlon
    HepatoblHary excretion and elimination  In the  feces  1s the  principal
route  by which  metabolites of  PAHs are  excreted.   Early  work  by  Peacock
(1936),  Chalmers  (1940),  and Chalmers  and K1rby (1940)  established  that  the
fluorescent material appearing In the bile of  rabbits, guinea  pigs,  rats  and
fowl  following  Intravenous  Injection of  either colloidal benzp[a]pyrene  or
colloidal anthracene was  not the administered hydrocarbon but a derivative.
Excretion In  the bile  was  well  established  within 15 minutes of  Injection,
and only  a small  amount of  fluorescence appeared  1n  the  urine of rabbits  and
rats.  Almost  none of  the  excreted material  was unchanged  benzo[a]pyrene.
Chalmers  and K1rby  (1940) and  Berenblum and  Schoental  (1942)  found  that only
IX  of  a  subcutaneous  dose  of  60 mg benzo[a]pyrene was  eliminated  unchanged
1n  the'feces  of  rats  or mice.  Kotln et al.  (1959) confirmed  this  fraction,
and  noted  1n  addition  that  all   elimination of  unchanged  benzo[a]pyrene
occurred within the first  10-15 minutes  following  an  Intravenous dose.

    Metabolism 1s  the  rate-determining  step  for  excretion  Into  the bile.
Induction  of metabolism  with  3-methylcholanthrene  enhanced the  rate  and
extent of excretion of  metabolites of benzo[a]pyrene 1n rat bile  (Iqbal  et
al.,  1979).    Pre-treatment  of  rats  with  either  3-methylcholanthrene   or
benzo[a]pyrene Itself also  Increased the  rate of  excretion  Into the bile of
metabolites  of   subsequently  administered   benzo[a]pyrene.    There  was,


04400                                111-35                          05/13/91

-------
however,  no  such  pretreatment  effect  on  excretion  when  the  metabolites
themselves  were  given  Intravenously,  demonstrating that  the  effect of  the
Inducers  1s  on  metabolite formation and  not on metabolite transfer  to  bile
(Schlede et al., 1970).

    Enterohepatlc  circulation  of  metabolites,  occurs,   and   a  portion  of
                                       <
urinary excretion of metabolites can be attributed  to  this  source.   Kotln et
al. (1959) observed that  In rats with  biliary  fistulas,  urinary excretion of
benzo[a]pyrene metabolites was  reduced  from 7-14X to 2-4% of  the dose.   The
existence of  an  enterohepatlc circulation suggested by  this observation was
confirmed by  Chlpman  et  al.   (1982),  who collected  biliary  metabolites  of
benzo[a]pyrene and Introduced  them  Into the  duodena  of b11e-cannulated  rats.
following which the metabolites appeared In both bile and urine.

    A  very  small  fraction  1s  excreted  1n  pancreatic juice.   Iqbal et  al.
(1979) observed  that  during  the Initial 2-hour  period following Intravenous
administration of  benzo[a]pyrene  to rats,  0.03X of the  dose was excreted In
pancreatic  juice while 39% was excreted   1n  the bile.   In  view of  the low
metabolic potential of  pancreatic mlcrosomes In  vitro, the  authors  suggested
that metabolites of  benzo[a]pyrene  1n  the  pancreas may have  originated  In
the liver and been  transferred to the pancreas by the systemic  circulation.

    Systemlcally absorbed benzo[a]pyrene  or  Us  metabolites  (Heldelberger
and Weiss,  1951; Kotln et al.. 1959; Sanders  et al.,  1984) and systemlcally
absorbed  7,12-dlmethylbenzanthracene   or   Us  metabolites  are   also   not
eliminated 1n expired air (Lo, 1964; Sanders et al., 1984).
04400                                111-36             .             05/13/91

-------
    The  fractions  of  14C-benzo[a]pyrene  and  14C-7,12-d1methylbenzanthra-
cene-assodated  radlolabel  excreted  by  rats  In urine  and bile  have been
quantHated.  Thirty-eight percent  of an Intravenous dose  of  benzo[a]pyrene
of unspecified size was excreted 1n bile  by  2  hours;  urine  was not monitored
1n  this  study  (Iqbal et  al.,  1979).   Kotln  et  al.  (1959)   found  39%  of
Intravenous doses  <134 yg In  the  "bile  at  3  hours,  and as much  as  96X  1n
the bile by 14 hours.  In rats with biliary  fistulas, a  maximum of only 3-4%
of radioactivity was  recovered  1n  the urine after a  24-hour period.   Intact
rats had a urinary excretion of 7-14%  of the administered dose,  suggesting
that enterohepatlc circulation serves as a secondary  source of radioactivity
that 1s excreted  1n  the  urine.   The amount  1n the bile  was not proportional
to dose  at doses  of  >150 yg;  only  32% of  a  400 yg dose  was  recovered  1n
the  bile   at  24  hours.   This  amount  (0.32x400 yg = 128  yg)  Is  equal   to
that found In the bile  14  hours  after  the 134  yg  dose (0.96x134 yg =  129
yg),   Illustrating  the  limited   capacity   of  the   rat   to   metabolize
benzo[a]pyrene.
                   \                                   •
    Lo' (1964) administered  3  mg  doses  of  l4C-7,l2-d1methylbenzanthracene
Intravenously to three rats,  and reported that 10-16% of the  radlolabel  had
been recovered In  the  urine  and 18-44% 1n the feces by  17  days.   Sanders  et
al.  (1984) found  38%  of recovered  radlolabel from  an  Intravenous  dose  of
1.04 yg  1n  rat  feces  by 7  days,  and  26% 1n urine.   These  observations
Indicate that  the  urine  Is  a more Important route of excretion for metabo-
lites of 7,12-dlmethylbenzanthracene than for metabolites of benzo[a]pyrene.

    Additional data on relative  utilization  of urine and feces  as routes  of
excretion are summarized  1n  the following Dermal  Section.


04400                                II1-37                          05/13/91

-------
    Oral.   Because  most  of  an  absorbed  oral  dose  of  a PAH  reaches  the
liver  directly  via  the  portal   circulation  before  reaching  the  systemic
circulation,  hepatic  metabolism  and  biliary  excretion  may  determine  the
relationship  between  administered  dose  and  systemic  dose.    It  1s  to  be
expected that at reasonable doses a greater  fraction  of an oral  dose than of
an Intravenous  dose  would be excreted  In  the feces.   In  fact,  low  doses of
orally  administered  PAHs  could,  1n  principle,  be  fully  metabolized  and
excreted  1n  the  bile  without  reaching  the  systemic  circulation  at  all.
Twenty-four  hours  after  administration  of  50 yg  pyrene  to  rats,  24  yg
was  found  In the GI  tract and  none  In  the liver,  kidney, lung  or  trachea
(Mitchell  and  Tu,  1979).   However,  controlled  studies   of   elimination
following  oral  doses  of  PAHs  have generally  utilized  large amounts  of  the
hydrocarbons, so that substantial fractions  are  not metabolized  In the first
pass.   Interpretation  of  biliary excretion  data  following an oral dose 1s
further complicated by enterohepatlc circulation.

    Rahman  et  al.   (1986)  administered   five  radlolabeled  PAHs  Including
benzo[a]pyrene  and   7,12-d1methylbenzanthracene  In   1.0  mg  doses  to  rats
Intraduodenally, and  cumulative  excretion  of radlolabel  1n  bile and urine
after  24  hours  was  measured  and  expressed as a fraction of  administered
radlolabel.   For benzo[a]pyrene  these  figures were 6% for  urine and 25% for
bile,  for  7,l2-d1methylbenzanthracene  they  were  3%  for   urine  and 25%  for
bile.   These data  for  benzo[a]pyrene  are consistent with  the corresponding
values  obtained following  Intravenous  administration of high doses  (see
preceding  discussion),  suggesting  that  hepatic capacity  to metabolize  the
absorbed PAHs on the first pass had been greatly exceeded.
04400                                111-38                          05/13/91

-------
    Modica  et  al.  (1983) gave  emulsions  of benz[a]anthracene, chrysene and
trlphenylene to  rats  by stomach  tube.   Seventy-two hours later,  cumulative
fecal elimination  was  6% of  the  benzanthracene  dose of  22.8  mg,  3% of the
trlphenylene dose  of  22.8  mg, 38% of the  chrysene  dose of 22.8 mg, and 41%
of  the  chrysene  dose  of  11.4  mg,   Indicating  that even  these  relatively
massive doses are absorbed  surprisingly  well.

    Elsele  (1985)  examined  the  uptake and  distribution  of tracer levels  of
14C-naphthalene 1n  laying  pullets, swine  and  dairy cattle.   Each group  of
animals received either  a  single oral  dose or  dally oral doses of  naphtha-
lene for 31 days.  The distribution of  naphthalene  and/or  Us  metabolites  In
various tissues  was measured  24 hours  after  the  last dose.  Considerable
species differences  In the  distribution of naphthalene  and/or Us metabo-
lites were  observed.   However,   In  each of  the  exposed animals  naphthalene
and/or  Us  metabolites reached  the  systemic  system and  distributed to all
tissues examined.

    Inhalation.  Elimination  of  an Inhaled  dose of a  PAH follows  Us  lung
clearance pattern, that  1s,  the  fraction that  Is systemlcally  absorbed  from
the lung  Is  excreted  1n bile and  urine,  while  the fraction that  Is cleared
from the  lung  by  mucoclHary action  and  swallowed  Is  subject  to  GI absorp-
tion.   In  this  case,  however, the amounts  reaching the duodenum may be too
low to  approach  the capacity of  the  liver to metabolize  them on the  first
pass.   In  support of  this  supposition,  Pylev  et  al.  (1969)  observed  that
excretion of radlolabel  In  feces and urine of  hamsters given  Intratracheal
Instillations  of   8H-benzo[a]pyrene   alone,   or   3H-benzo[a]pyrene-coated
carbon or asbestos particles, followed  the same general time  course as  loss


04400                                II1-39                          05/13/91

-------
 from  the  lung,  and  that  for  a  period of  36  days  elimination  via  urine
 remained at  ~80X of that  1n  feces.

     The  most detailed  and precise  studies  1n which  excretion  was measured
 following  Inhalation  exposure  are  those  from  the   Inhalation  Toxicology
 Research  Institute,  1n  which  rats  were  exposed by  Inhalation  to  pure
 3H-benzo[a]pyrene aerosols  and  to  3H-benzo[a]pyrene  adsorbed  on  gallium
 oxide  particles, and excretion  of  radlolabel was  followed  for  time periods
 <16 days.   Using  pure  aerosols   of  3H-benzo[a]pyrene  (500  yg/l)  with  a
 mass median  diameter  of 1-2  ym,   Mitchell  (1982) observed  that radlolabel
                                                             •
 was excreted both 1n  the feces and 1n  the  urine,  with radioactivity 1n the
 feces  ~10 times  that In  the urine  during the first 24 hours after exposure.
 Subsequently (Sun et al.,  1983),  this experiment  was  repeated  with a lower
 concentration   of  3H-benzo[a]pyrene,   600   ng/i,   and  was   extended   by
 exposures  to gallium  oxide particles  coated with  3H-benzo[a]pyrene.   Both
 aerosols had  the same  mass median diameter,  -0.1  \an  as  In  the previous
•study.   The  pure aerosol  was,  as before, eliminated In urine and  feces, with
 the total  amount of  radlolabel  eliminated 1n feces through the  sixteenth day
 after  exposure  about  6 times the amount eliminated In the urine.  Since the
 aerosol  concentration  In this  study was  1/800  that  used  In  the. previous
 study,  H must  be concluded that  the elimination  pattern 1s Independent of
 aerosol  concentration  at least  up  to  ~500  »ig/l.   That 1s,  these aerosol
 concentrations  do not  stress  hepatoblllary  metabolism and  excretion mecha-
 nisms.   Based on a comparison  with the  work of Kotln et al. (1959), this Is
 a  reasonable conclusion.   Kotln et al.  (1959)  found that oral  doses <134
 vg/rat did not  exceed the maximum metabolic  capacity of the animals.
 04400                                111-40                           05/13/91

-------
    When  benzo[a]pyrene  adsorbed  to  gallium  oxide  particles  (3500  ng/i)
was  Inhaled,  the  total  amount of radlolabel eliminated  1n  the feces through
the  sixteenth  day  following exposure was ~12 times  the  amount eliminated 1n
urine,  demonstrating  the shift 1n  excretion  pattern caused by  the  presence
of the particles.

    When benzo[a]pyrene was presented  to  rats as  an  aerosol  coated on dlesel
exhaust  particles  (Sun et  al.,  1984), the amount excreted  1n  feces  by  the
twenty-sixth day following  exposure was  ~5 times the  amount  excreted In  the
urine.                         •                             .

    Dermal.   Since only  an  Insignificant  amount of  radlolabel  appears  In
the  GI  tract as a result  of  grooming by  rats  whose  back skin was painted
with  PAHs  (Mitchell  and Tu,  1979; Sanders  et  al.,  1984),   the  cumulative
fractions  of an absorbed  cutaneous dose eliminated  1n  urine and  1n  feces
should  be  comparable with  the cumulative fractions  found after Intravenous
administration.    Excretion   of   radlolabel   from   14C-benzo[a]pyrene,
      /
l4C-7,!2-d1methylbenzanthracene      and       14C-anthracene,       absorbed
percutaneously, has  been  examined.   The  results  are similar  to the results
of the Intravenous  studies discussed above.

    Thirty-five percent  of  the radlolabel  associated  with a  cutaneous  dose
of  12.5  yg  14C-benzo[a]pyrene/cm2  (22.5  yg  total   dose)   was   found   In
the feces at  24 hours and 80% at 7 days, with -10* 1n  the urine  at 7 days,
by which  time nearly 'all  the benzo[a]pyrene  had been absorbed (Sanders  et
al.,  1984).    7,12-dlmethylbenzanthracene  metabolites were   excreted  to  a
greater  extent  1n  the urine than benzo[a]pyrene  metabolites.  After 7  days,


04400                                111-41                          05/23/91

-------
30%  of   the   radlolabel   associated   with   a  cutaneous  dose  of  5.4  yg
l4C-7,l2-d1methylbenzanthracene/cm2  (1.1  yg   total   dose)  was   recovered
1n the urine  and  only 62% 1n the feces (Sanders et al.t 1984).  Yang et  al.
(1986)  found  that of  the 52% of  the  absorbed  radlolabel  from a  cutaneous
dose of   9.3  yg  a4C-anthracene/cm2,   29% had  been  excreted  1n  the  urine
and 21% 1n the feces  after 6 days.

    Interestingly, these  fractions  appear to be dose-dependent.   Increasing
the  benzo[a]pyrene  dose  from 1.25 to  12.5  to 125  yg/cm3 resulted  1n  a
shift  In   the  excretion  pattern  at 24  hours  In   favor  of the  feces.  In
contrast.   Increasing  the  7,l2-d1methylbenzanthracene  dose  from 5.4 to  56 to
515  yg/cm2 decreased  the  fraction of  the  dose excreted  1n  both  feces  and
urine,  although   fecal  excretion  was  less  markedly  affected  than  urinary
excretion  (Sanders et al., 1984).   These shifts were greater than could be
accounted  for  by  the reduction 1n  percentage  absorbed at  the higher doses,
encouraging the  speculation  that  differential  capacity  limitations  of  the
enzyme systems responsible for  the formation of different  metabolites  might
result In  a relative  decrease 1n production  at  high doses  of metabolites  for
which the urine Is the favored route of excretion.

Summary                                                           .
    The PAHs are  a Upophlllc class of xenoblotlcs  that  are readily absorbed
across cellular membranes.  Major  routes of environmental exposure to PAHs
are  the following: the GI tract  by contaminated food  or water;  the lungs by
Inhalation of aerosols.or  by  hydrocarbon-adsorbed particles; and  the  skin by
direct  contact.    Once   absorbed,  the   PAHs  are   rapidly  and  widely
distributed.   The differing  availability  of   the  PAHs may depend  on  the


04400                                II1-42                          05/13/91

-------
chemical  form   In  which   exposure   occurs   (for   example,   hydrophlllc   or
Upophlllc    solution),    varying    Intestinal    absorption,     metabolic
transformation  or  removal  rates.   PAHs  can  be observed  1n most  Internal
organs  from  minutes  to  hours  after  administration  by  various  routes.
Adipose and mammary tissues are significant  distribution  sites  1n  which  PAHs
may be stored and slowly released.

    PAH  metabolism  Is  affected  by  the  route  of  administration.   PAHs
administered  through  the  lungs or through  1.p.  or  1.v.  Injection  may  avoid
first pass metabolism  In the  liver.   The  blood will  distribute  the PAHs  Into
peripheral   tissues.    This   distribution   will  rapidly   result   1n   the
establishment   of  a   steady-state  concentration   between   the   blood   and
well-perfused   tissues.    Adipose  tissues,   however,  will  exchange   the
deposited PAHs  more  slowly  and will  accumulate  and  slowly  release PAHs.   On
the  other  hand, oral  administration  of  PAHs results  In metabolism by  the
liver  before   systematic   distribution  can  occur   via   the   bloodstream;
                                                                            •
consequently, PAH  metabolites will appear  In the bloodstream  several  hours
after administration.

    PAHs can  be metabolized by enzyme-mediated  oxldatlve mechanisms  to  form
reactive  electrophlles.  The cytochrome  P-450  dependent  mlcrosomal  mixed
function oxldase (MFO) system (arylhydrocarbon  hydrolase)  Is  Induclble  by
PAH  exposure.   These   enzymes  are  found  1n many  mammalian  tissues;  MFO
enzymes of the  liver are the  best  studied and .have been found to be the most
abundant.  Similar  types  of  metabolites  are produced by mammalian tissues;
however, the  proportions  of  the  phenols,  dlols and  epoxldes  produced  may
differ  among   the  various   tissues.   The   mechanism  of   the  dlolepoxlde


04400                                111-43                          05/13/91

-------
metabolic  pathway  for  PAH tox1f1cat1on  and  the  bay-region  hypothesis  are
accepted.   Reactive  epoxldes,  formed  by MFO mediated  PAH metabolism,  have
the ability  to  Interact with and  thereby  alter  DNA,  RNA and proteins.   PAH
metabolites are  most  commonly conjugated  with  glucuronlc acid,  glutathlone
or sulfate.  They are excreted by  the  GI  tract after  hepatoblHary  excretion
or  the  swallowing  of mucus  material  cleared  from  the  respiratory  tract.
Nonenzymatlc mechanisms  of biologic PAH alteration  also  occur.
04400                                II1-44                          05/23/91

-------
                          IV.   HUMAN EXPOSURE
Text to be provided by the Office of Drinking Water,
                                 IV-1                             08/30/85

-------
                        V.  HEALTH EFFECTS IN ANIMALS

    The general  and  specific  health effects  associated  with PAH Ingestlon,
and  to  a  lesser  degree,  Inhalation,  peritoneal  and dermal  exposures,  are
examined  1n  this chapter.  Much  of  the  research  on the  health  effects of
selected PAHs has focused on  their  potential  as skin carcinogens, genotoxlc
agents and on their  Inhalation  effects  (IARC, 1983).  Studies of Individual
PAHs administered by  the oral  route are  generally lacking.

Acute Oral ToxIcUv
    Acenaphthylene.   An  English  abstract of  a paper  by  Knoblock  et  al.
(1969) reported  that acenaphthylene administration "to the stomach" resulted
1n an LD5Q of 3  g/kg  for rats  and  2.2 g/kg for mice.

    Anthracene.    The abstract  of  a  Russian  study  Indicated  that  single
oral doses of 1.47 or 2.44 g/kg of commercial grade  anthracene or 17 g/kg of
pure  anthracene  were not  lethal  to mice (Nagornyl, 1969).   Toxic effects
reportedly .Included  fatlgabllUy;  adynamia;  hlstologlc  hyperemla  In  the
kidney, liver, heart and  lungs; "Upld dystrophy"  In  the liver; and leuko-
cytosls with  neutrophllla.

    Systemlcally  administered  anthracene (50  mg/mi  corn  oil   by  gavage)
followed  by  UV   Irradiation of  the  skin  for  1  hour, 2  hours after dosing,
produced  keratUls   of  the  exposed skin In  mice  (Dayhaw-Barker  et   al.,
1985).. This  effect  reportedly was  less  pronounced 1n mice  exposed  to UV
light only and was not  evident In  vehicle  controls.
04420                                V-l                             12/31/90

-------
    Dally Intragastrlc administration to five male  Wlstar  rats  of 100 mg/kg
anthracene  1n  olive  oil  for  4 days  produced  a   nonsignificant  (3-fold)
Increase  1n  mean  liver cytosollc aldehyde  dehydrogenase activity_£T6rr6nen
et.al.,  1981).  Neither  treatment-related  effects on aldehyde dehydrogenase
activity  1n  the  liver  mlcrosomes  or  postmltochondrlal fractions  of small
Intestinal mucosa  nor  effects  on liver/body weight ratios were observed.

    Benz[a1anthracene.  Pertinent  data  regarding  the  acute oral  toxldty
of benz[a]anthracene could not be located 1n  the available  literature.

    BenzoTalpyrene.    Pertinent  data regarding  the  acute  oral   toxldty  of
benzo[a]pyrene could not be located  1n  the available literature.

    Benzofblfluoranthene.    Pertinent   data   regarding   the   acute  oral
toxldty  of   benzo[b]fluoranthene  could not be  located  In the available
literature.

    BenzoFklfluoranthene.    Pertinent   data   regarding   the   acute  oral
toxldty  of   benzo[k]fluoranthene  could not be  located  In the available
literature.

    Benzorg.h.Hperylene.    Pertinent   data   regarding   the   acute  oral
toxldty  of   benzo[g,h,1]perylene  could not be  located  In the available
literature.

    Chrysene.   Pertinent   data  regarding   the  acute   oral  toxldty  of
chrysene could not be  located  In the available literature.


04420                                V-2                             09/21/90

-------
    DlberrzTa.hlanthracene.     Pertinent   data   regarding   the   acute  oral
toxldty  of  d1benz[a,h]anthracene  could  not  be  located 1n  the available
literature.
    Fluoranthene.   the  oral   LD5Q  for  male  Carworth-Hlstar  rats   exposed
to fluoranthene 1s -2000 mg/kg (Smyth  et  al.,  1962).
    Fluorene.   K1zer  et  al.   (1985),   1n   an   effort   to   Identify  enzyme
changes as  early  Indicators  of hepatocardnogenesls,  examined the effect of
several xenoblotlcs, Including fluorene, on  both the  activity and amount of
hepatic mlcrosomal epoxlde hydrolase.  As part  of this  study, male Holtzman
rats were fed a diet containing 0.06% (10.5  mg/kg/day) fluorene  for 3 weeks.
No hepatotoxldty, as measured by SGOT activity, was  observed.   Furthermore,
fluorene  exposure  failed  to  result  1n an appreciable  elevation of  the
activity  of  mlcrosomal  epoxlde hydrolase.   However,  a  4-fold Increase over
controls  of  epoxlde hydrolase  antigen  was   detected  by  Immunoassay.    After
comparing the  activity and  amount  of hepatic  mlcrosomal  epoxlde hydrolase
Induced by  feeding  the  various xenoblotlcs  with their reported  carcinogenic
potential,  the  authors  concluded   that Induction   of  mlcrosomal  epoxlde
hydrolase 1s not a "key change" leading  to malignancy.

    IndenoH ,2.3-cdlpyrene.   Pertinent   data   regarding   the   acute   oral
toxUHy  of 1ndeno[l,2,3-cd]pyrene  could  not  be located  1n the available
literature.
    Naphthalene.   The  LDcn  for  oral   exposure   to   naphthalene   has   been
    —	           ou
determined for  several  species.  The LD5Q  values  for naphthalene  dissolved
04420                                V-3                             10/08/91

-------
In peanut  oil  for male  and female  Sherman  rats were  2200  and 2400 mg/kg,
respectively  (Galnes,  1969).   In  two  other  studies,  the LD5Q  values for
rats   were   1780   (Toxicology  Data  Bank,  n.d.)  and  9430  mg/kg  (U.S.  EPA,
1980c); the  strain  and sex of  the animals was  not  specified.  For male and
female  CD-I  mice, the  acute oral LD™ values  of  naphthalene  1n  corn oil
were   533  and 710  mg/kg, respectively  (Shopp  et a!.,  1984).   In  a recent
study  (unpublished)  conducted  by  Mallory et  al.  (1985a),   the  acute  ID.-
values  of  naphthalene  1n corn  oil  for male  and female Sprague-Dawley rats
were  reported  to  be  2009  and  3310  mg/kg,  respectively.   An  LD5Q  of 353
mg/kg  was  determined  for  CD-I  mice  In a study In  which  Plasterer  et al.
(1985) treated mice at  doses ranging  from  125-2000 mg/kg/day  for  8  days.
    Although  cataract  formation  1n  rats  following  oral  administration  of
naphthalene  has  been  known for  many years  (FHzhugh and  Buschke,  1949),
recent studies have  shown  that ocular changes can  result from a  single  dose
of  naphthalene,  van  Heynlngen  and  Plrle  (1967)  found  that  lens  changes
developed  In  the  eyes  of rabbits after  a  single gavage dose of  1000  mg/kg.
In CD-I mice,  oral  doses of >400 mg/kg  for males and  >600 mg/kg  for  females
resulted 1n  ptosls with  clear, red secretions around  the eyes within  1  hour
of dosing  (Shopp et al.,  1984).

    Ikemoto  and  Iwata  (1978)  reported  that  oral administration  of  naphtha-
lene  (1  g/kg)  to male  and female albino  rabbits for  2  consecutive  days
resulted In  cataract formation.   Occurrence  of cataracts was accompanied  by
a decrease 1n sufhydryl content In both soluble and Insoluble lens protein.
04420                                V-4                             05/14/91

-------
    Rao and  Pandya (1981) orally  administered male  BHD  and AR albino  rats
1000 mg/kg  naphthalene  each  day for  10  days.   Significant  Increases  In  the
relative  liver  weight  (p<0.01),  and 1n  liver  aniline hydroxylase  activity
and I1p1d  peroxldatlon  (p<0.001) were  observed 1n  the  treated animals.   A
moderate  (nonsignificant)  Increase  1n llpld  peroxldatlon was  also  observed
In the eyes.  No treatment-related changes  In  sulfhydryl  content  or  alkaline
phosphatase activity were observed In the liver, kidneys  or eyes.

    Yamauchl et  al.  (1986) reported  that a  single  oral  dose  of  napthalene
(1000 mg/kg) to male Wlstar rats resulted  In a  significant Increase  1n serum
I1p1d peroxide levels (p<0.05) beginning on the 4th  day after  administration
and continuing  through  the  20-day  observation period.   This Increase  was
paralleled by a  significant  (p<0.05)  decrease In GSH  content  In  the lens of
the exposed  animals.    Zuelzer  and Apt  (1949) reported  a  hemolytlc  effect
when naphthalene was administered to dogs 1n their  diets.
                                                                            •
    Phenanthrene.  'Pertinent   data   regarding   the   acute  oral  toxlclty  of
phenant'hrene could not be located 1n the available  literature.

    Pyrene.   Pertinent  data  regarding  the acute  oral  toxlclty of  pyrene
could not be located In the available literature.

Acute Toxlclty By Other Routes
    Acenaphthylene.   In   a   study  published  1n   Russian   (Rotenberg  and
Mashblts,  1965),   acenaphthylene  was administered  to white  rats  Intratra-
cheally 1n a sunflower  oil solution  or  by blowing  acenaphthylene  powder Into
the  trachea.   The  dosing schedule  used  was  not  provided.   The  pulmonary
tracts of animals  sacrificed 1 month  after the experiment began showed signs

04420                                V-5                             05/14/91

-------
of tracheobronchHIs and  hyperemla, edema  and  necrosis  of  the epithelium 1n
the trachea and bronchi with  the  formation  of  ulcers.   No further details of
this study were available.

    In a  corn oil gavage study  acenaphthylene was administered  dally  to 10
CD-I mice/sex/group  at 0,  125,  250,  500,  750 and  1000 mg/kg/day.  In  the
1000 mg/kg/day group  the mean body  weights were  significantly reduced  1n
both sexes at  the  end  of  the  first week.   Dose-related clinical symptomology
was noted  1n  groups  receiving 500, 750  and 1000 mg/kg/day;  these symptoms
                        •*
Included  languid  behavior,  prostration,  decrease  1n  body  temperature  (cold
to  touch)  and  toxic   effects  to  the  eyes.   These  eye  effects  Included
enlargement,  drying  out and eventual  crusting  over.   In the  1000  mg/kg/day
group  eyes were  adversely affected   1n  40% of  the  males  and  50% of  the
females.   The survival rate  1n  the 750  mg/kg/day group was  90%  among  the
males  and  80% among  the  females;  1n  the 1000 mg/kg/day group  survival  was
significantly  reduced  In  both sexes  at  40%.   Gross  pathology from  the  750
and  1000  mg/kg/day  groups  Indicated  that  the   liver,  stomach,  eyes  and
     /
subcutaneous  tissues  were affected by the  treatment  (Hazelton Laboratories
America Inc., 1989a).

    Anthracene.  Mice  treated 1.p.  with anthracene  were found  to  have  an
LD50 of  >430  mg/kg  bw (Salamone, 1981).   In a  study  by Gerarde  (1960),
mice were  given  1.p. Injections  of 500 mg/kg  bw/day  for 7 days.   Of the 10
mice treated,  9  survived.  The ID,.-  (skin  Irritant dose) for  the  mouse  was
found to be 6.6xlO~4 mmol/ear  (Brune et al., 1978).

    In a corn  oil  gavage  study anthracene was  administered  dally for 14 days
to 10  CD-I  mice/sex/group at 0,  125,  250,  500, 750 and  1000  mg/kg/day.   No

04420                                V-6                             11/12/91

-------
treatment-related  differences  were  observed   between  the  groups  1n  body
weight   gain,   mortality   and   gros.s   pathologic   examination   (Hazelton
Laboratories America, Inc., 1989b).
                                -\
    Benzfalanthracene.   Pertinent  data  regarding  the  acute nonoral  toxlc-
1ty of benz[a]anthracene could not be located \n the available literature.
    BenzoTalpyrene.   The  mouse  L05Q  (1.p.)  for  benzo[a]pyrene  1s  ~250
mg/kg   (Salamone,   1981).    The  ID™  for   skin   Irritation  In  mice   1s
5.6xlO"5 mmol/ear (Brune et al., 1978).
    A single  1.p.  dose of  10  mg benzo[a]pyrene produced a reduction  In  the
growth rate  of  Immature rats  (Haddow  et  al.,  1937).  He 11 man et  al.  (1984)
studied  the  acute  toxldty of  Injections  of  benzo[a]pyrene  In  C57B1  male
mice.   Groups of  7-10  animals  were  Injected  1.p. with  1.12x10~4  mol/kg
benzo[a]pyrene 1n corn oil.  Two hours before  sacrifice  animals  were Inject-
ed with  8H-thym1d1ne  to determine  the  rate of DNA  turnover.  A significant
      /
decrease  (40X)   In  8H-thym1d1ne  Incorporation  Into  the  thymus,   spleen,
small Intestine and testls  occurred  48 hours after  benzo[a]pyrene treatment,
which was Indicative  of  decreased  DNA synthesis 1n  those  organs.  There  was
a stimulatory effect  on  thymldlne  Incorporation In  the  liver  48 hours after
the Injection of benzo[a]pyrene.

    Robinson  et al.  (1975) showed  that  "responsive" mice (those  capable  of
producing Increased levels  of  cytochrome  P-450 mediated enzymes as  a  conse-
quence of  PAH exposure) had  reduced survival  time following a  single  1.p.
dose of 500 mg/kg bw benzo[a]pyrene.


04420                                V-7                              11/12/91

-------
    Wojdanl et  al.  (1984)  reported that  target  cell  killing by lymphocytes
from two mouse  strains  was  decreased  following a single 1.p. Injection of 5
or  50  mg  benzo[a]pyrene/kg.   This study  Is  described  1n  the  Target Organ
Toxldty Section.

    Subcutaneous Injections of benzo[a]pyrene  In  corn  oil  at doses of 5, 20
or  40  mg/kg  1n  female B6C3F1 mice  produced  a  dose-related suppression of
antibody production to both T-cell Independent and T-cell dependent antigens
(White and Holsapple,  1984).
                                                            •
    Benzo[b]fluoranthene.    Pertinent  data   regarding  the  acute  nonoral
toxldty  of  benzo[b]fluoranthene  could  not   be  located  1n  the  available
literature.

    BenzoTklfluoranthene.    Pertinent  data   regarding  the  acute  nonoral
toxldty  of  benzo[k]fluoranthene  could  not   be  located  1n  the  available
                                                                           •
literature.

    Benzofq.h.Uperylene.    Pertinent  data   regarding  the  acute  nonoral
toxldty  of  benzo[g,h,1]perylene  could  not   be  located  1n  the  available
literature.

    Chrysene.    The  LD50  (1.p.)   for  chrysene was  found  to be >320 mg/kg
bw  1n  mice  (Simmon  et al.,  1979).   A  single  l.p.   Injection  of  7.5 mg
chrysene 1n sesame  oil  produced  no toxic  effects  1n  AKR/J or C57B1/60 mice
during a 20-day observation period or  upon necropsy.
04420                                V-8                             03/20/91

-------
    D1benz[a.h]anthracene.   One  or  two  Intraperltoneal  administrations  of
3-90 mg/kg  bw d1benz[a,h]anthracene  1n  sesame oil produced  a reduction  1n
the growth  rate  of young Lister  strain  hooded rats.   This persisted for  at
least 15 weeks (Haddow et al..  1937).

    Fluoranthene.   The   24-hour   CDC_   value   after   dermal  exposure   to
                                     5U
fluoranthene was found to be 3180 mg/kg  In  rabbits  (Smyth  et  al.,  1962).   No
Information  was  reported concerning  target  organs   or   specific  cause  of
death.  As part  of the same  study,  Smyth  et al.  (1962) observed no mortality
1n  six  male  and six  female albino  rats exposed  to  concentrated  vapors  of
fluoranthene for 8 hours.

    Haddow  et al.  (1937) examined  the  effect  of various  PAHs.  Including
fluoranthene, on body growth  In  hooded rats of  the Lister strain.   A  single
1.p.  Injection   of  10 mg/kg fluoranthene  dissolved   In  sesame  oil  had  no
adverse effect on body weight gain over a 24-day  observation period.

    Aso'kan  et al.  (1986)  examined the  Induction of  cutaneous and hepatic
monoxygenase  activities  by  fluoranthene.   Twenty-four  hours  after a  single
topical application of 10 mg/kg  fluoranthene to the backs of  Sprague-Dawley
rats,  significant  Increases 1n  aryl  hydrocarbon  hydroxylase,  7-ethoxyreso-
rufln   0-deethylase   and  7-ethoxycoumarln  0-deethylase   activities   were
observed 1n the  skin and  liver.

    Fluorene.    Pertinent  data  regarding  the  acute  nonoral  toxldty  of
fluorene could not be located 1n  the available literature.
04420                                V-9                             03/20/91

-------
    IndenoH,2.3-cdlpyrene.   Pertinent  data  regarding  the  acute  nonoral
toxlclty  of  1ndeno[l,2,3-cd]pyrene  could not  be  located  In  the  available
literature.

    Naphthalene.   Ir1e  et  al. (1973)  studied  the effects of  naphthalene  1n
mice  (strain  not  specified)  following  s1ng>e  subcutaneous   Injections  at
doses  of  650-1348 mg/kg.   Vigorous  tremors  were noted  In  the mice  for  3-4
days  following  dosing.   The LD5Q  value  was  calculated to be  969  (891-1053)
mg/kg.
                                                             »
    The  24-hour  LD5Q  value of  naphthalene  In  Swiss-Webster .mice was  380
(350-413) mg/kg following  1.p. Injection  (Warren  et  al.,  1982;  Shank  et al..
1980).  Those mice that died did  so  within 24  hours; survivors were observed
for an  additional 6  days.   The target organ was  Identified  by  Warren et  al.
(1982)  as  the lungs, 1n which  naphthalene  caused a  dose-dependent Increase
In  bronchlolar  epithelial necrosis  at  doses  >200 mg/kg.   The  pulmonary
damage  and  lethality  resulting from  naphthalene  administration  were markedly
inhibited by  prior  treatment  with plperonyl butoxlde and enhanced by  prior
treatment with  dlethyl maleate.   This  supports  the  view that P-450-dependent
metabolism of naphthalene  1s  responsible for the observed toxldty and that
glutathlone  plays an  Important  role  In  the  detoxification   of  the  lung-
damaging metabollte(s).

    Reid et -al. (1973)  gave napthalene  dissolved 1n sesame oil to C57B1/6J
mice  by the  1ntraper1toneal  route  and   found  coagulatlve  necrosis  of  the
bronchlolar  and  bronchial  epithelium at a  dose  of 600  mg/kg.    Controls
received sesame oil alone,  and   no  adverse  effects  were reported for  this
group.  The size of the treatment  groups  was  not stated.

04420                                V-10                            10/24/91

-------
     Mahvl  et  al.  (1977) administered  naphthalene  In corn  oil  Intraperlto-
 neally  to C57B1/6J mice.   Two  groups  of 63 mice received corn  oil  alone  or
 remained  untreated.   Groups  of 21  mice each  were  given 67.4,  128 or  256
 mg/kg.   Three animals from each dosage  group  were  sacrificed  at 10 minutes,
 1   hour,  6  hours,  12   hours,  24  hours,   48   hours,  and 7 days  following
 treatment.   Lung  tissue  was  rapidly fixed  and examined by  light,  scanning
 electron,  and  transmission  electron microscopy.  No  changes  were  noted  1n
 either  control  group.   Minor  bronchlolar  epithelial  changes  were  noted  1n
 the group receiving  67.4 mg/kg.   Mice  In  the higher dose  groups  developed
 necrosis   of  secretory  nonclHated  bronchlolar   cells  and  the  adjacent
 dilated  cells.   Epithelial  structure  returned to  normal within 7  days  1n
 all cases.

     Tong  et  al.   (1982)  found  hlstologlc  changes  In  the lungs  of  C57B1/6J
 mice treated  1.p. with  225  mg/kg  naphthalene.  One  day after  dosing,  the
 Clara  cells  In the terminal bronchioles were pyknotlc, and hypereoslnophlUc
'nuclei  appeared to be detaching from the bronchlolar wall.   Three days after
 dosing,  some  surfaces  appeared  to  be  completely   denuded  of  Clara  cells,
 whereas  other  surfaces  appeared to  have  Immature  Clara  cells  scattered
 drcumferentlally.  Five days  after treatment, there was still  evidence  of
 Incomplete   recovery;  by  8  days,   most  of  the terminal   bronchioles  were
 reeplthellaHzed;  and  by  15  days, mature  Clara  cells were  common,  but
 recovery  was evidently  not  complete.

     O'Brien  et  al.   (1985)  Investigated differences  1n naphthalene-Induced
 toxldty  subsequent  to  1.p.   treatment  of male  Swiss   T.O.  mice  and  male
 Wlstar-derlved rats.   In mice doses  >200 mg/kg resulted  In damage  to  the


 04420                                V-ll                            03/20/91

-------
nondllated  bronchlolar  epithelial  cells, and  doses  >400 mg/kg resulted  In
damage  to cells  1n  the  proximal  tubules  of  the  kidney.   Rats  were  more
resistant  to  the cytotoxlc effects of  naphthalene,  as doses of  naphthalene
as high  as  1600 mg/kg caused no  detectable pulmonary  or  renal  damage.   This
species  difference  In  toxldty  was  reflected  by  a  larger  depletion  of
                                  *
nonproteln  sulfhydryls  1n the  lung and  kidney  of  the mouse  than 1n  those
organs 1n the rat.

    It previously  had been  reported  that the  8-hour  Inhalation LC5Q  value
for naphthalene was 100 ppm  (Union  Carbide, 1968).  However,  Buckpltt  (1985)
suggested that  this  value  may be too low.  He estimated  that 1n 8  hours  the
body  burden  would  be <30  mg/rat,  or  ~150-200 mg/kg.  This concentration  1s
less  than  the  oral  or  1.p.  ID™  values  for  rats.   Fait  and  Nachrelner
(1985) reported that  exposure  of  male  and  female VMstar  rats  to  78  ppm
naphthalene  for 4  hours  resulted  In  no mortalities, nor  any  lung,  liver,
kidney,  or  nasal  passage  abnormalities. . In an unpublished Inhalation  study
with  male Swlss-Wesbster  mice,  no deaths  were noted  following  nose-only
exposures  to 90  ppm for 4 hours.   Lung  lesions  however,  were reported
(BuckpHt, 1985).

    van Heynlngen and P1r1e  (1967)  treated one  rabbit  Intravenously with  300
mg of  a  dlhydrodlol metabolite of  naphthalene  1n divided doses over  3  days
and noted retinal  lesions.   They  also  noted lens changes  In  four  rabbits
dosed externally with eye drops of  the  same compound  (IX  dissolved  In  water)
over a period of 2-5 days  for a  total  of 40-70 mg  per  rabbit.

    Acute  ocular   Irritation  was  noted  1n  2/6  New  Zealand  white  rabbits
receiving no postdose eye  rinse after  24 and 48 hours of exposure  to  0.1  mg

04420                                V-12                           11/12/91

-------
 naphthalene  (Mallory et al., 1985c).  This  response  Included  slight  1r1t1s,
 moderate  redness  and  slight swelling and  discharge.  All  animals  appeared
 normal  by 72  hours  postdoslng.  No  positive  response was  noted  In  rabbits
 (three) receiving  a  postdose rinse.

     No  deaths  occurred  when 2500  mg/kg of naphthalene  was applied  to  the
 skin of  male  and female  Sherman rats  (Galnes,  1969).  The  application  of
 2000 mg  naphthalene/kg  (dissolved  1n acetone)  to the skin of  New  Zealand
 white  rabbits  did  not  cause  mortality;  the  LD   ' was,  thus, >2000  mg/kg
 (Mallory  et  al., 1985b).  These  studies suggest that  naphthalene  may not  be
 as  readily absorbed  through  the skin as  It 1s through the Intestinal mucosa.

     Naphthalene (moistened with  2  mi of acetone)  was found  to  be slightly
 to  moderately  Irritating  to the skin of male  and female  New Zealand white
 rabbits  30-60 minutes  postdoslng  (Mallory  et al.,  1985d).  Dermal Irritation
 was still evident up  to 5 days  after test  material  application.   Flssurlng
'of  the  skin  was also noted.
       /
     Naphthalene (100X)  did not  cause  delayed  hypersensUlvUy  In  Hartley
 guinea  pigs  (Mallory et  al., 1985e).

     Phenanthrene.    The  LD™ for mice  (1,p.)  1s  700 mg/kg  bw  (Simmon  et
 al., 1979).

     Yoshlkawa  et al. (1985) Investigated effects  of 1.p.  exposure to phenan-
 threne,  pyrene and  some of their oxidized  products  1n Sprague-Dawley rats.
 As  part  of  this experiment, groups  of  three  males  each  were Injected with
 04420                               V-13                            03/20/91

-------
either  3.0  ml/kg  saline,   3.0  ml/kg  DMSO  or  150  mg/kg  phenanthrene  1n
DMSO.   An  additional   set  of  animals  was  similarly  treated  on  a  second
occasion.   Blood  was obtained  at  24 or  72 hours post-treatment by  cardiac
puncture  and  the  following  measurements  taken:   aspartate  amlnotransferase
(AST),  alanlne amlnotransferase,  -r-glutamyl  transpeptldase  (G6TP),  lactic
dehydrogenase,  glucose, b1!1rub1n,  BUN and  creatlnlne.  Gross  observation
was  made  of   organs.    Phenanthrene  treatment   resulted  In  a  significant
elevation  of  serum  AST and  GGTP by  comparison  with  both  controls.   GGTP
returned  to control  levels  by 72 hours.  Livers  of animals  killed  at either
24 or  72  hours  were described as congested with  a distinct'lobular pattern.
Kidneys were reported to be somewhat  smaller In size  and congested.
    Pyrene.    The   Intraperltoneal  LD5om   (dose  lethal   to  half   the
animals 1n 7  days)  for B6C3F1 mice was found to be  514  mg/kg  pyrene and the
Intraperltoneal  LD5Q...  (dose  lethal  to  half the  animals 1n  4 days)  was
678 mg/kg  bw  (Salamone. 1981).   The growth rate  of young (46-48 days  old)
Lister rats  was  not affected by  Intraperltoneal  administration of  10 mg of
pyrene 1n sesame oil (Haddow et al., 1937).
    Yoshlkawa et  al.  (1985) administered 150  mg/kg pyrene to male  Sprague-
Dawley rats as described  In  the  preceding section.   This  resulted 1n a small
but significant elevation In both serum AST and blllrubln.  At  both  24  and
72 hours sacrificed rats  were observed  to have minimal  congestion and swell-
Ing of livers.

    Yoshlkawa,  T.,  W.   Flory,   L.P.   Ruhr  et  al.   (1987)  treated male
Sprague-Dawley rats with  a  single 1.p.  Injection of 150 mg/kg  pyrene In DMSO

04420                                V-14                            11/12/91

-------
and  evaluated  serum  chemistry  at  24  and  72  hours  for   Indications   of
hepatotoxldty.   No  statistically  significant  differences  were  found  for
AST,   alanlne   amlnotransferase,   sorbHol   dehydrogenase,   GGTP,   lactate
dehydrogenase, glucose,  BUN  or  creatlne  between control rats receiving DMSO
and  rats  administered pyrene.  Furthermore,  no hlstopathologlc  alterations
were noted 1n animals necropsled 72 hours after treatment with pyrene.

Subchronlc and Chronic Oral ToxIcUy
    Bloassays  for  carc1nogen1c1ty  are  reported  1n  this  section only when
they refer to health endpolnts other than tumor Incidence.  '

    Acenaphthylene.   Knobloch  et  al.  (1969)  reported  on  the  effects  asso-
ciated with Ingestlon of both acenaphthylene and acenaphthene.  One group  of
seven  rats was given an  oral dose of 0.6 g/kg bw of acenaphthylene 1n  olive
oil for 40 days.   A  second group of  seven  rats was given an  oral  dose  of  2.0
g/kg bw  of acenaphthene 1n  olive  oil for  32 days.   Both PAHs  yielded  the
following physiologic effects:  "considerable" body  weight loss,  changes  1n
      /
the peripheral blood  pattern, changes 1n renal  function, and Increased  serum
amlnotransferase  activities.   Additional  effects  associated  with acenaph-
thene  only  Included mild  morphologic  damage  to the  liver and kidney, lung
changes consisting of mild  bronchitis,  and  localized  Inflammation of  perl-
bronchial tissue.  Since the source of this Information 1s an abstract, more
precise quantification of the health effects  1s not possible.

    In a Russian study (Rotenberg and Mashblts,  1965),  acenaphthylene  In  oil
was  administered  orally to  white  mice  at a  dose 1/10 the  ID,..  [LD5Q=1760
(range of  1100-2800)  mg/kg] every  other  day  for  2  months.   Treated mice


04420                                V-15                           11/12/91

-------
showed a  significant lag 1n  weight  gain as.compared with  controls.   H1sto-
pathologlc examination  of  organs showed  signs  of stasis In  the  parenchyma-
tous  organs   and  albuminoid  degeneration of  the  liver.   The  most  severe
changes were  observed 1n the  lungs,  which showed  hemorrhage with  destruction
of  the Interalveolar septa  and focal  bronchial   pneumonia.   Purulent  fod
were observed  1n  Isolated  cases, and bronchogenlc lung  cancer was  diagnosed
1n one mouse.  Further details of this study were not provided.

    In a  study conducted  by Hazelton Laboratories America, Inc.  (1989C)  for
U.S.  EPA acenaphthylene  was  administered  to  CD-I   mice   (20/sex/group)  by
gavage at  dosage levels of  0,  100,  200, or 400  mg/kg/day for  at  least  90
days.  Effects examined Included  mortality, clinical  signs, body  weights,
food   consumption,   opthalmology,   hematology,   clinical   chemistry,   organ
weights,  and  gross  and  hlstopathology.   There  was no Increase 1n  mortality
among  males  tested;  however,  the Incidences of treatment-releated  deaths  1n
females  were  15, 25 and 40X for  low-,  mid- and high-dose  groups,  respec-
tively, compared with no  deaths 1n female controls.  No significant  changes
In  mean  body  weights,  body  weight  gains,  or  food  consumption were  found.
Statistically  significant  (p<0.05)   treatment-related   hematologlc  effects
Included decreased  erythrocyte count 1n  all male dose   groups and  high-dose
females;  decreased hemoglobin  and hematocrlt In mid- and high-dose  males  and
high-dose females;  Increased  platelet  counts   1n  mid-  and high-dose  males;
and  Increased   leucocyte  and  segmented neutrophll   counts  In   high-dose
females.    Treatment-related    clinical    findings   Included   significantly
Increased cholesterol* and  albumin  In   high-dose  males and  all  groups  of
treated females, and Increased  total  protein In mid- and high-dose  males  and
all three female treatment groups.  Nean  absolute  and relative  liver  weights


04420                                V-16                            11/12/91

-------
 were  significantly Increased 1n  all  treatment groups of  both  sexes,  accom-
 panied  by  a  dose-related  Increase  In gross  pathologic  findings,  Including
 enlarged,  dark, mottled,  prominent retlcular pattern or pale areas.

    An  Increase 1n the  Incidence and  severity of centMlobular hepatocellu-
 lar  hypertrophy was  seen  1n  all female  treatment  groups and  1n high-dose
 males.   Individual  cell  necrosis  was  observed  In  high-dose  females.   No
 significant changes  In mean absolute  or  relative  kidney  weights were found,
 but  gross  examination  revealed  a  treatment-related  Increase  In  the occur-
 rence  of  granular,  pitted,  rough,  mottled,  or  small appearance.   Nephro-
 pathy,  Increased  Incidence and  severity of renal  tubular  dilatation, epithe-
 lial  hyperplasla  of  the  collecting ducts,  slight  hyperplasla  of the transi-
 tional  epithelium  In  the  renal  pelvis,  and renal  tubule mlcroconcretlons
 were  observed  In  the kidneys  of all  treated  females.   An  Increase  In the
 Incidence  of renal  tubule regeneration was  seen 1n high-dose  males.   Mean
 absolute  and relative  ovary weight decreases  1n  mid-  and high-dose females
'were  accompanied  by  a  slight Increase In Incidence and degree of Inactivity,
 and   fewer  and  smaller  corpora   lutea  1n  the  high-dose  group.   A  small
 Increase  In  the  number   of  grossly  observed ocular  opacities  was observed,
 but  opthalmoscoplc examination  revealed no treatment-related ocular lesions.
 Based  on  liver  and kidney changes and deaths  In females  the LOAEL was deter-
 mined  to  be  100 mg/kg/day;  no  NOAEL  could be determined since  100 mg/kg/day
 was  the lowest  dose  given.

     Anthracene.   In a  chronic bloassay  for  carcinogenic  effects  (Schmahl,
 1955),  a  group  of 28  BDI  and  BDIII  rats  received anthracene  In the diet,
 starting  when  the  rats were  ~100  days  old.   The dally dosage  was  5-15


 04420                                V-17                            03/20/91

-------
mg/rat, and the experiment was terminated when a  total  dose  of  4.5  g/rat was
achieved on  the  550th experimental day.  The  rats  were observed until they
died, with some living more than 1000 days.  No  treatment-related effects  on
Hfespan or gross and histologic  appearance of tissues were observed.  Body
weights were  not mentioned,  and  hematologlc  parameters  were  not  measured.
No chronic LOAEL could be determined from this  study.

    In a 90-day subchronlc toxldty study,  the U.S. EPA (1989a)  administered
anthracene to  groups  of  20 male  and  female  CD-I  (ICR)BR  mice by  gavage.
Dose  levels  were 0,  250, 500  and 1000  mg/kg/day.   Criteria  evaluated for
                                                             »
compound-related effects  were  mortality,  clinical signs, body weights, food
consumption,   opthalmology,  hematology,  clinical chemistry,  organ  weights,
organ-to-body  weight  ratios,  gross   pathology  and   hlstopathology.    No
treatment-related effects  were noted;  therefore,  the NOAEL determined from
this study Is 1000 mg/kg/day.
                                                                           •
    BenzTalanthracene.    Pertinent   data   regarding    the   subchronlc  and
chronic  oral  toxldty  of  benz[a]anthracene  could  not  be  located  In the
available literature.

    Benzofalpyrene.    Aplastlc  anemia,   and   ultimately death,  have- been
linked  to subchronlc  oral exposures  to  benzo[a]pyrene (Robinson et al.,
1975).  Strains of mice used  In the experiment had  been classified  as either
"responsive"   or  "nonresponslve",  based   on the  strain's  susceptibility  to
Induction  of  cytochrome  P-450 and  associated -enzymes  by   PAHs.   Treatment
groups,  consisting  of  30  animals/strain,  were  fed  a laboratory diet   ad
libitum  that  had  been soaked  1n  corn  oil containing benzo[a]pyrene; the


04420                                V-18               .             03/20/91

-------
estimated  oral  dose  was  -120  mg/kg/day.    Responsive  and  nonresponslve
control groups,  each  consisting of 30 animals, were  fed  the same diet that
had  been  soaked  In  unadulterated  corn  oil.   In three  responsive strains
(C57B1/6, C3H/HeN, BALB/cAnN)  fed benzo[ajpyrene, the  following numbers of
mice/group died  over  a 180-day period:  2/30,  3/30  and 1/30, respectively.
This was  by  comparison  with one  iflouse  1n  the  responsive  strain  (C57B1/6)
control group.   Among  the nonresponslve  strains  (AKR/N,  DBA/2),  all of  the
mice In  the  treatment groups  died with  at  least  half  the  deaths  occurring
within 15  days.  Only  two mice  died  In  the  nonresponslve (DBA/2) control
group  over  the same period  of  time.   The high mortality  among  the treated
nonresponslve mice was  attributed to  pancytopenla, which led  to death from
hemorrhaglng  or  overwhelming Infection.    The  nonresponslve benzo[a]pyrene-
treated mice  also experienced a  significant  Increase In  relative  Hver-to-
body weight ratios.

    Benzofblfluoranthene.   Pertinent   data  regarding  the  subchronlc   and
chronic oral  toxldty  of  benzo[b]fluoranthene  could  not  be  located 1n  the
available literature.

    Benzo[k]f1uoranthene.   Pertinent   data  regarding  the  subchronlc   and
chronic oral  toxldty  of  benzo[k]fluoranthene  could  not  be  located 1n  the
available literature.

    Benzorg.h.Hperylene.   Pertinent   data  regarding  the  subchronlc   and
chronic oral  toxldty  of  benzo[g,h,1]perylene  could  not  be  located In  the
available literature.
04420                                V-19                            03/20/91

-------
    Chrysene.   Pertinent  data  regarding  the  subchronlc  and  chronic  oral
toxldty of chrysene could not be  located 1n  the available  literature.

    D1benz[a.h]anthracene.   Pertinent  data   regarding   the   subchronlc  and
chronic oral  toxldty  of  d1benz[a,h]anthracene could not  be located In the
available literature.

    Fluoranthene.   The U.S.   EPA   (1988)  evaluated  the  oral  toxldty   of
fluoranthene  1n  a subchronlc  bloassay.   For 13 weeks  male  and female CD-I
mice (20/sex/group) received either 0, 125,  250 or  500  mg/kg/day of  fluoran-
thene  dissolved   1n  corn  oil.   Baseline blood  evaluations  were  determined
during  the first  week of  study   from  an additional  30  animals/sex/group.
Body weights, food consumption and  clinical  signs of toxldty  were monitored
at  regular Intervals  during   the  experimental  period.  At  the end of the
study  period  the  animals   were   sacrificed,  submitted  for   autopsy,  and
hematologlc and serum chemistry evaluations were performed.

    All treatment  groups  exhibited  Increased  salivation  (never  exceeding 10%
of  the  population  of any   treatment  group).   There  were  dose-dependent
Increases  1n  pigment accumulation  In  the liver and mild nephropathy charac-
terized by  the presence of  multiple fod of  tubular regeneration.   A small.
but  statistically significant  (p<0.01).  Increase   In  relative liver weight
was observed  In  mice receiving  125  mg/kg/day of fluoranthene.   Statistically
significant  (p<0.05)  changes  1n  mice  receiving   250 mg/kg/day  Included
Increases  1n  SGPT and  absolute  and relative  liver   weights,  as  well   as
decreases  1n  packed cell  volume  and  red blood cell numbers  (females  only)
and  albumin/globulin  ratios.   Statistically  significant  (p<0.05)  observa-


04420                                V-20                            03/20/91

-------
tlons  In mice  receiving 500  mg/kg/day  Included  Increases  1n  SGPT,  serum
globulin  and  absolute and  relative  liver weights,  as  well as decreases  1n
packed cell  volume  (females only) and albumin/globulin ratios.  Since  there
was  no  dose-related Increase  1n  clinical signs  (I.e.,  salivation) and  the
changes  In  kidney and  liver  hlstopathology  observed at  125 mg/kg/day  were
not  considered  adverse  by  the  U.S.   EPA   (1991a),   this  dose   level   1s
considered  the  NOAEL.   Based  on  hematologlc  alterations,  Increased  SGPT
levels, and changes  In  kidney  and liver  hlstopathology 1n  animals  receiving
250 mg/kg/day, the U.S.  EPA (1991a) considered this dose level as  the LOAEL.

    Fluorene.   Wilson  et  al.   (1947)  provided  anecdotal  reports  on  gross
and  hlstologlc  appearance of organs  of  rats  exposed to fluorene 1n the diet
as  part  of  an  oncology  study.  They observed  significant decreases 1n  the
rate of growth among albino rats  consuming 0.5  and 1.054 fluorene  In the diet
for  105  days.   Furthermore, liver weights were Increased  In  rats  receiving
>0.25X fluorene,  spleen  weights  were decreased In  all  treated animals  and
testes weights  were decreased  1n the  high-dose   rats.   Neither  numbers  of
      /
animals nor  any organ weights  were published.

    The U.S. EPA  (1989a) conducted a subchronlc toxlclty  study 1n  which CD-I
mice  (25/sex/group)  were exposed for 13  weeks  via gavage to 0, 125, 250  or
500  mg/kg/day  fluorene  suspended  1n corn oil.   Parameters  used   to  assess
toxlclty  Included food  Intake,  body  weight,  clinical  observations,  hemato-
logy  and   serum  chemistry  and   gross   and   hlstopathologlc  examinations.
Increased salivation,  hypoact1v1ty  and  urine-wet  abdomens   In  males  were
observed  In  all  treated animals.   The  percentage  of  mice exhibiting  hypo-
activity  was  dose-related.    In   mice  exposed  at  500  mg/kg/day,  labored


04420                                V-21                             11/12/91

-------
respiration,  ptosls  (drooping  eyelids)  and  unkempt  appearance  were  also
observed.   A  significant  decrease  1n erythrocyte  count  and hematocrlt  was
observed  1n females  treated  with 250  mg/kg/day fluorene  and  1n males  and
females  exposed  to  500  mg/kg/day.    Decreased  hemoglobin concentration  and
Increased  total   serum  blUrubln levels  were  also  observed  1n  the  500
mg/kg/day group.   Decreases  In  ery'throcyte count,  hematocrH and  hemoglobin
concentration  were  all   observed at  125  mg/kg.  These  effects,  although
apparently  dose  dependent, were  not  statistically  significant  at 125  mg/kg
by  comparison  with controls.  A  significant decreasing  trend  1n BUN  and  a
significant  Increasing  trend  1n total  serum  blUrubln were  observed  for
high-dosed  males  and  females.    A  dose-related Increase  1n relative  liver
weight  was  observed  1n   treated  mice;  a  significant  Increase  In  absolute
liver  weight  was  also  observed  In  the  mice  treated with >250 mg/kg/day
fluorene.  A significant  Increase 1n absolute and relative  spleen  and  kidney
weight was  observed  1n males  and  females  exposed to 500  mg/kg/day and  males
at  500  mg/kg/day,  respectively.   Increases  1n  the  absolute  and  relative
liver and spleen weights  1n the high-dose males  and females  were  accompanied
by  Increased amounts  of  hemoslderln  In the  spleen  and Increased  numbers  of
Kupffer  cells  of   the   liver.    No  other  hlstopathologlc  lesions   were
observed.   Using   the data  from U.S.  EPA  (1989a),  the U.S.  EPA   (1991a)
Identified  a   LOAEL  of   250   mg/kg/day   for   hematologic  effects;   the
corresponding NOAEL 1s 125 mg/kg/day.

    IndenoM,2.3-cdlpyrene.   Pertinent  data  regarding  the  subchronlc  and
chronic oral toxIcHy  of 1ndeno[l,2,3-cd]pyrene  could  not be located  1n  the
available literature.
04420                                V-22                            10/24/91

-------
    Naphthalene.   Shopp  et  al.  (1984)  conducted  a  14-day  and  a  90-day
study  on  groups of  male and  female CD-I mice  administered naphthalene  1n
corn oil by gavage.  In  the 14-day  study,  six  groups  of  male and  female  mice
(40-112/group) were  given  doses of 0,  27,  53  or 267 mg/kg/day;  the  highest
dose was  one-half  the   LD5Q  for male  mice.   Male mice demonstrated  lower
survival rates  than females,  apparently  due  to  the  aggressive  behavior  of
group-housed male  mice;  however, the  mortality  1n the  high-dose groups  of
both male  and  female  mice was 5-10%  higher  than  In  the control  groups.
There  was  a  significant  decrease (7-13%)  1n body weight 1n male and  female
mice receiving  the  high  flose.   The high-dose males exhibited a 30% decrease
1n thymus weight, while  females  exhibited a decrease  In  spleen  weight  and  an
Increase  1n  lung  weight.    Gross   pathology  but   not   hlstopathology  was
performed.   No  biologically  relevant changes  were  noted 1n  treated  animals
for  the following  measures:   hematology,  clinical  chemistry,  hexabarbltal
sleeping  time,   or  Immune  function  (humoral   Immune  response,   lymphocyte
responsiveness,  popliteal lymph node response,  and bone marrow function).

    For the 90-day study,  five  groups  of  112  male  and 112  female  mice  were
given  doses of  0,  5.3,  53 or  133 mg/kg/day. A high mortality was seen among
all groups  of  male mice, but appeared  to  be  due to the  aggressive behavior
of group-housed male mice.

    No  significant  effects  on body weight were  noted for  males  or  females.
A significant decrease In the absolute  weight  of the  brain, spleen  and liver
was  noted  for  females,  receiving  133  mg/kg;  however, organ-to-body  weight
ratios  were  significantly different  only for  the  spleen.   Of  the  changes
noted  In the  clinical  chemistry data,  the Increase 1n blood protein content
1n males and  females receiving  53  or 133  mg/kg, the decrease In BUN  In all

04420                                V-23                            05/14/91

-------
treated  female  groups, and  the decrease  In calcium  1on concentrations  In
males receiving 53 or  133 mg/kg were  considered  to  be  treatment related.   No
significant changes were noted  1n  hematology,  1n HFO activity,  or  1n  Immune
function for  either  sex.   Hlstopathology data were not presented and  H  1s
not known 1f naphthalene caused bronchlolar  lesions.

    In  a  subchronlc  oral   toxlclty  study  performed  for the  NTP  (1980b),
naphthalene 1n  corn  oil was administered by gavage to male and  female F344
rats (10/sex/dose) at dose levels of  0,  25,  50,  100,  200 or  400 mg/kg/day, 5
days/week for 13 weeks. 'At 400 mg/kg,  two males  died  during the first week;
this treatment dose caused diarrhea,  lethargy, hunched  posture  and  roughened
halrcoats  1n  rats of  both sexes.   A significant  (I.e.,  >10%)  decrease  1n
body weight gain  was  observed among  males and  females  at 200  and  400 mg/kg
and In females at 100 mg/kg.   Food  consumption  was not  affected.

    All  the rats  In  the study were necropsled and  comprehensive hlstopatho-
loglc examinations were performed  on rats from  the 0 and 400  mg/kg  groups.
Hlstopathologlc examinations  of the  kidneys  and  thymus  were  performed  on
rats from  the 200 mg/kg group  (according  to the hlstopathology  tables;  the
100 mg/kg  group according  to  the text).  The authors stated that lesions  of
the kidney  1n males  and thymus  1n  females   of the  400 mg/kg group may have
been compound-Induced,  and  that no eye  lesions  were found.  The Incidences
of lesions of kidney and thymus were, however, very low.   The renal lesions,
which  did  not  occur   1n  females,   were  observed at  Incidences  of 0/10  In
controls,  2/10  In the 200  mg/kg   group  and 1/10  In   the  400  mg/kg  group.
These renal lesions  consisted of focal  cortical  lymphocytlc  Infiltration  or
focal  tubular  regeneration 1n  the  two  200  mg/kg males and  diffuse  tubular
degeneration  1n  the  one 400  mg/kg male.  Lymphold depletion  of the  thymus

04420                                V-24                            05/14/91

-------
occurred  1n 2/10 females of the  400  mg/kg group,  In none of  the  control  or
200  mg/kg  females, and  In  none  of  the  males of  these  groups.   Hematologlc
analyses  revealed marginal  decreases  1n hemoglobin and hematocMt  1n  males
and  females of  the  400  mg/kg  group,  and  a moderate Increase In the number  of
mature  neutrophlls  and  a decrease  In the number  of  lymphocytes  1n  males  of
the  400  mg/kg  group,  relative   to  controls.  No  hematologlc changes  were
observed  at the lower dosages.

     In  a  similar  study, naphthalene  was administered 1n corn  oil  by gavage
at  0,  12.5, 25,  50,  100  or 200  mg/kg/day,  5   days/week,  to B6C3F1  mice
(10/sex/dose)  for 13 weeks  (NTP, 1980a).   Seven  mice (three males  and two
females  of  the 200  mg/kg  group, one female of the  25 mg/kg group  and one
control  male)  died during  the second,  third and  fourth weeks of  the  study
from gavage  trauma   or  accident.   Transient signs  of toxlclty  (lethargy,
rough halrcoats and decreased  food  consumption)  occurred at  weeks  3-5 1n the
200  mg/kg  groups.   All treated  groups  of  male  mice  gained  somewhat  more
                                                                             •
'weight  than did  control males.   Dose-related decreases In  body  weight  gain
were se'en 1n females, but  were not  statistically  significant.   All  the mice
were  necropsled   and   comprehensive   hlstopathologlc  examinations   were
performed on the  mice from the 0 and 200 mg/kg groups.  No compound-related
lesions  were observed  1n  any  organs,   Including  kidneys,  thymus,  eyes and
lungs.    Hematologlc   analyses,  performed   on  all   groups,   revealed  no
significant, compound-related changes.

     FUzhugh and  Buschke (1949)  noted  the formation of cataracts,  within  3
weeks  of  treatment,  In rats  fed diets  containing 2% naphthalene or  one  of
several  naphthalene  derivatives.   The  effects  of pigmentation on  cataract


04420                                 V-25                            03/20/91

-------
formation 1n rats  and  rabbits  have been summarized by  van  Heynlngen  (1979):
Albino rats  do  not readily convert naphthalene  to  1,2-dlhydroxynaphthalene,
which  auto-ox1d1zes  to form  1,2-naphthoqulnone  (the naphthalene  metabolite
known  to  bind  to the  lens  of  the eye), possibly because polyphenol  oxldase
1s found  only In  plgmented  tissues.   Secondary effects occur  In  the  retinas
of  both  albino  and  plgmented  rats.   For  rabbits,  pigmentation  Is  not  as
Important as a  modifier  of  toxldty.   Rather, the depletion of antloxldants
1s a  critical  step In rabbits  since  the  reserve of ascorbic add and  other
antloxldants Is  considerably less than 1n  the rat.

    The  critical  nature  of the  depletion  of  antloxldants 1n  rabbits,  as
compared  with   the  Importance   of  pigmentation,  was  demonstrated  by  van
Heynlngen  and   P1r1e   (1967)  In  a  gavage  study  In  which  naphthalene  (1
mg/kg/day) was  administered to  Dutch  and  two strains of albino rabbits.   In
more  than  half  of the treated   rabbits,  lens  opacities and degeneration  of
the retina  were observed.  This  occurred  concomltantly with a depletion  of
ascorbic  acid  1n the  aqueous  and vitreous  humours.   Some  of those  rabbits
that  received 10  or more  doses  of naphthalene showed a general yellowing  of
the eye  fluids  and yellow  or brown  cortical areas 1n  the  cataractous  lens,
suggesting  the  presence  of 1,2-naphthaqulnone and  1,2-dlhydroxynaphthalene,
respectively.   Considerable variation exists among rabbits  1n  their  response
to  naphthalene.   Cataract  formation  1n  .rabbits  was  not  noted  following
topical  application   of   a  10X  solution  of  naphthalene  In  oil  or   l.p.
Injection of 500 mg/day for  60 days (Ghettl and Mar1an1, 1956).

    The  effect  of  pigmentation  on   the  development  of  cataracts  1n  rats
exposed  to  naphthalene was substantiated  by Koch et  al.  (1976)  using  five


04420                                V-26                             03/20/91

-------
strains  of  Rattus  norveglcus  of  different  pigmentation.   Groups  of  six
animals of each strain were  gavaged with  either  0  or  1000 mg/kg naphthalene,
dissolved  1n  liquid  parafflne,  every second  day for  75  days.   No  opacities
or  lens abnormalities  were  observed  1n  any of  the  control  groups.   All
animals of  the  plgmented strains  (E3, BOE,  OA)  developed zonular  cataracts
between 16  and  28 days.  Among  the  albino strains (Wlstar,  Sprague-Dawley)
only some  of  the  animals developed changes 1n the eyes.   These changes were
less pronounced In the albino strains and  occurred after  longer latencies of
32-61 days.

    Cataract  formation  1n  C57B1/6J and DBA/2N mice was evaluated  by  Sh1ch1
et  al.  (1980)   and  Sh1ch1   and  Nebert   (1982).   The  C57B1/6J  mice  are
"responsive"  to  the  Induction  of  AHH activity while  the  DBA/2N  mice  are
"nonresponslve"  to  the  Induction  of  AHH  activity  (primarily  P..-450  enzyme
activity  that 1s  believed  to  be  Involved   1n  the  tox1f1cat1on of  PAHs).
Groups  of  15  mice were  fed  laboratory chow  ad libitum that had  been  soaked
for  at  least 24  hours  1n  corn oil  containing  5 or  10 mg/ma  naphthalene.
The  feeding   regimen  was continued  for  60  days.   Dally  Ingestlon was  not
calculated by the authors.   A concomitant  dose of B-naphthoflavone  was  given
twice weekly  as  an Inducer  of  AHH activity.  A 6.7X  Incidence  1n  cataract
formation  was observed  In  C57B1/6J  mice at  each  dose.   In  addition  to
cataract formation, tlsssue  degeneration  1n  the chorold, ciliary  body,  and
1r1s occurred.  In  support of the theory  that the mechanism of naphthalene-
Induced  cataract  formation  Involves   Us  metabolism  by  P,-450 enzymes  to
toxic  Intermediates  with  subsequent  binding  to lens  tissue,  no  cataracts
were observed 1n DBA/2N mice.
04420                                V-27                            10/08/91

-------
    Phenanthrene.   Pertinent  data  regarding  the  subchronlc   and   chronic
oral  toxldty  of  phenanthrene  could not be located  1n  the  available  litera-
ture.

    Pyrene.  Pyrene  was  fed at a  concentration  of 2000 mg/kg diet to  young
male  rats  for  100  days.   An  Inhibition  of growth  was  observed, which  was
reversible upon  addition  of cystlne or methlonlne.   The authors noted  that
livers  of  treated  animals  (Including  rats  consuming benzo[a]pyrene  or
3-methylcholanthrene 1n the  diet)  were enlarged  and had a  fatty appearance
(White and White, 1939).

    The  U.S.  EPA  (1989b)  conducted a  90-day subchronlc toxldty  study  1n
which groups  of male and  female CD-I  mice (20/sex/group) were  gavaged  with
either  0,  75,  125  or  250 mg/kg/day  pyrene  1n  corn  oil  for  13  weeks.
Parameters used to  assess  toxldty Included  body and  organ  weights,   food
consumption,   mortality,   hematology  and   serum   chemistry  and  gross   and
hlstopathology.   Statistically significant  (p<0.01) decreases  In absolute
kidney weights were observed In males of all  treatment  groups and 1n  females
receiving  250  mg/kg/day.    Statistically  significant  (p<0.01)  decreases  1n
relative liver  weights  were observed  1n  males receiving either 125  or  250
mg/kg/day  and  1n  females   receiving 250 mg/kg/day.  Nephropathy, character-
ized  by  the  presence of multiple  foci  of  renal  tubular  regeneration,  often
accompanied by  Interstitial  lymphocytlc Infiltrates  and/or  foci  of  Intersti-
tial  flbrosls, was observed  1n 4,  1, 1  and 9 male mice  1n  the control,  low,
medium and high dosage groups,  respectively.   SlmlHar  lesions were  seen  In
2,  3,  7 and  10 female mice 1n  the 0,  75, 125  and 250 mg/kg/day  treatment
groups, respectively.  The kidney  lesions  were described as minimal  or  mild


04420                                V-28                            10/08/91

-------
In  all  Instances.   Statistically significant (p<0.01) decreases  1n  erythro-
cyte  numbers,  hematocrHs  and hemoglobin levels were  observed 1n male  mice
receiving  250  mg/kg/day.   Statistically significant  (p<0.01) Increases  1n
absolute liver weights  were  observed  1n females receiving 250 mg/kg/day  and
In  relative  liver  weights  of females  receiving either 125 or 250 mg/kg/day
                                  •
and In males receiving  250 mg/kg/day.   Based on nephropathy,  accompanied  by
changes  1n  absolute  and  relative  kidney   weights,   the  U.S.  EPA   (1991a)
1dent1-  fled  125  mg/kg/day  as  the  LOAEL;  the  corresponding NOAEL  1s  75
mg/kg/day.

Subchronlc and Chronic Toxldty By Other Routes
    Acenaphthylene.   In  a  study reported by Rotenberg and Hashblts  (1965),
white  rats were  exposed  to acenaphthylene  dust  at  0.5-1.25 mg/m3  for  4
hours/day  for  4  months.  After  3 weeks  of  exposure,  a delay  In  weight  gain
and a tendency toward  decreased blood pressure were  observed.   Hlstopatho-
loglc  examination  revealed  various  degrees  of  malignancy 1n  the  lungs  of
almost all treated rats.   Focal  bronchitis and  per1bronchH1s  with bronchlo-
      *•
Hzatlon of  the alveolar  and metaplasia of the  bronchial  epithelium  were
observed 1n  the  mildest cases.   Advanced cases showed  desquamatlon of  the
bronchial  and  alveolar  epithelium, paplllar growths  1n  the epithelium  and,
1n  three  rats,  Isolated  regions  of   carcinoma  1n the  form  of  strands  of
epithelial cells.  Further  details of  this study were  not  provided.

    In a  study by Reshetyuk  et  al.  (1970)  published   1n Russian, -100 white
male  rats  were exposed  to  vapors  of  acenaphthylene  at  a concentration  of
18+2.5  mg/m3,  4   hours/day,  6  exposures/week   for  5 months.   In   exposed
rats,  reflexes  of the  upper airways  were  altered and  an  Increase 1n  the
concentration  of  nucleic acids  In  the liver was  observed.   H1stopatholog1c

04420                                V-29                            11/12/91

-------
examination of the  lungs  revealed  aspedflc  pneumonia  as  the  major  pathology
of  Inhalation exposure  to acenaphthylene.   Changes  observed  In  the  lungs
Included  desquamatlon  of  the cells  In  the alveolar  epithelium and  focal
bronchitis  accompanied  by  hyperplasla  and  metaplasia   of   the  bronchial
epithelium.   No  signs  of malignant growth  were observed   1n  this  study.   No
.further details of this study were available.  .

    Anthracene.   Pertinent  data  regarding  the  subchronlc  and chronic  non-
oral toxldty of anthracene could not  be located 1n the available  literature.

    Benzfalanthracene.    An  early  study  (Hoch-Llgetl,  1941) demonstrated
lymphold  system  effects  as  a  consequence  of  benz[a]anthracene  Injection,
Mice received  a  total  dose of 10 mg  In  weekly  s.c.  treatments  for  40 weeks.
Lymph  glands  removed at weekly  Intervals  showed treatment-related  Increases.
In retlculum cells and Iron accumulation.

    Benzo[a]pyrene.  Male  white  carneau  pigeons  were given weekly  1njec-r
tlons  1n  the  pectoral  muscle of 0.1, 10 or  100 mg/kg  benzo[a]pyrene In corn
oil {Revls et  al.,  1984).   Controls were either not  treated or were Injected
with corn oil.   Four  pigeons/group  were cannulated  at  3 months  and at  6
months for recording of  blood pressure  and  sampling of blood for determina-
tion of  plasma cholesterol  and  Upoprotelns.   Atherosclerotic plaques  were
counted  and  sizes  measured.   By  6  months  pigeons  treated  with  10  or  100
mg/kg/week benzo[a]pyrene were  observed to  have  Increased LDL  protein  and
cholesterol  (p<0.01),   HDL  protein  and  cholesterol   (p<0.01)  and  plasma
cholesterol   (p<0.05).    This  was   by   comparison  with  corn  oil-treated
controls.   Benzo[a]pyrene  treatment   did  not  produce  changes  1n  various


04420                                V-30               -             03/20/91

-------
cardiovascular  parameters  Including systolic  and  diastollc blood  pressure,
arterial pulse,  left  or  right  ventricular or central venous pressure,  heart
rate or relative heart weight.

    Among  pigeons  treated  with   benzo[a]pyrene   coronary artery  plaques
ranging  In  size  from  0.6-0.85  mm  were  observed.   Incidence  of  coronary
artery  plaques  was  5/24 for benzo[a]pyrene-treated birds by comparison  with
1/24  for  combined corn  oil  and untreated  controls.   It was  noted  1n  this
study   that   benzo[e]pyrene  administered   1n   the   same  fashion   as   was
benzo[a]pyrene did not Induce plaque formation.

    Penn  and Snyder  (1988) tested  benzo[a]pyrene,  anthracene  and  dlbenz-
[a,h]anthracene  to  determine  whether  artherosclerotlc  plaque formation  1s
related to mutagen1c1ty or carcinogenic potency  1n chickens.   In  their  study
male  White Leghorn chickens  received  weekly  Intramuscular Injections  with
one of  the test compounds  for  16 weeks.   The dosage  and  number of each  group
varied  as  follows:   benzo[a]pyrene at  40  mg/kg  with 6  chickens,  anthracene
at  20  mg/kg  with 6  chickens,  and d1benz[a,h]anthracene  at 20 mg/kg with  5
chickens.  At the end of the Injection period, the  cockerels were killed and
the aortas Isolated.  The  number of  plaque-containing aorta segments/chicken
was   6.8,    4.3  and   6.2   for    the    benzo[a]pyrene-,   anthracene-   and
d1benz[a,h]anthracene-treated  groups,   respectively.    The largest   plaque
volumes were noted 1n the animals  treated with d1benz[a,h]anthracene.

    Benzofblfluoranthene.   Pertinent  data   regarding   the subchronlc   and
chronic nonoral toxldty of benzo[b]fluoranthene could not  be  located 1n the
available literature.
04420                                V-31                             10/24/91

-------
    Benzo[k]fluoranthene.   Pertinent  data  regarding   the   subchronlc   and
chronic nonoral toxldty of benzo[k]fluoranthene could not be  located  1n  the
available literature.

    Benzorq.h.npervlene.   Pertinent  data  regarding   the   subchronlc   and
                                  «
chronic nonoral toxldty of benzo[g,h,1]perylene could not be  located  1n  the
available literature.

    Chrvsene.  Pertinent  data regarding  the  subchronlc  and  chronic  nonoral
    ^^^•Mirf^^^^^^H^H            ^
toxldty of chrysene could not be located 1n the available literature.

    Dlbenzfa.hlanthracene.   Chronic  exposure   to  d1benz[a,h]anthracene  has
been  associated  with  gross  changes  1n  the  lymphold   system  (Hoch-L1get1,
1941).  Weekly s.c.  Injections of  d1benz[a,h]anthracene (0.5 ml of a 0.05X
colloidal  solution  1n  IX gelatine)  were  given  to  a total of  40  albino  male
and female mice for  40 weeks.   The lymph glands were removed for examination
from  2 mice/week  over   the  study  period.   Splenectomles were ddne  during
weeks  21-30.   Among  the effects  noted were an  Increase 1n  retlculum  (stem)
cells,  accumulation  of  Iron,  reduced   lymphold  cells and  dilated  lymph
sinuses.   Reduction  of  lymphold  cells was more  pronounced among d1benz[a,h]-
anthracene-treated mice  by  comparison with anthracene and benz[a]anthracene
treated animals.   Moreover,   the  weight  of the spleens  1n  the  treated  mice
were significantly lower than the spleen weights In the controls.

    H1stolog1c  examination  of  spleens  from  d1benz[a,h]anthracene-treated
animals showed d1m1n1shment  of lymphold  and retlcular  elements.  Livers were
pale  and  soft and  showed evidence  of fatty degeneration and  deposition of


04420                                 V-32                            10/08/91

-------
Iron  In  Kupfer  cells.   Adrenals  were  marked  by  Iron  deposition  In  the
cortical  zone  retlculaMs,  and  kidneys  showed  signs  of  degeneration  of
tubules and Malph1g1an bodies.  There was  some  degeneration  of  spermatogenlc
cells  and two  female mice  had  "many  big  corpora  lutea  1n  the  ovaries"
(Hoch-L1get1,  1941).

    Lasn1tzk1  and  Woodhouse  (1944) conducted  a more detailed  study of  the
effects on the  lymphatic  system of long-term exposure  to  d1benz[a,h]anthra-
cene.   Subcutaneous  Injections  (0.278  mg/1nject1on)  were given to male  rats
5  times weekly  for several  weeks.  The  lymph nodes  1n  the  treated  rats
underwent hemolymphatic  changes,  Including  the appearance of  extravascular
red blood cells  In  the lymph  spaces and large plgmented  cells.

    Malmgren  et al.  (1952) reported that 2-  to  6-month-old  homozygous  strain
C  mice  (3-4  mice/group)  receiving three  subcutaneous  Injections of  either
50, 100 or 400 mg/kg/1nject1on of d1benz[a,h]anthracene spread  evenly  over  a
12-day  period,  which Included the  5  days following  antigen challenge,  had
      /
reduced serum antibody levels.

    Fluoranthene.    Limited data  are available  concerning  the  toxic  effects
of  fluoranthene produced  by  repeated  administration.   These  consist  of
reports of  mortality  produced  1n  mice  by  repeated dermal  application  or
subcutaneous  Injection.

    Shear   (1938)   administered  four  doses,   each  consisting   of  10  mg  of
fluoranthene  In glycerol,  by subcutaneous  Injection  to  strain  A mice.   Six
of 14 mice survived for 18  months.


04420                                V-33                            10/08/91

-------
    Barry et  al.  (1935) applied a  0.3X solution of  fluoranthene  In  benzene
twice  weekly  to  the  Interscapular  region  of mice.   Mortality ranged  from
60-70% after 6 months and 70-90X after 1 year.

    Hoffman et  al.  (1972)  applied 50  yl  of a  1%  fluofanthene solution  1n
acetone 3 times weekly  for  12 months to the  backs of  Swiss  albino mice.   No
mortality had resulted after 15 months.

    Fluorene.   Pertinent  data regarding the  subchronlc and  chronic  nonoral
toxlclty of fluorene could not be located 1n the  available literature.

    Indeno[1.2.3-cdlpyrene.    Pertinent   data  regarding  the  subchronlc  and
chronic nonoral  toxlclty of  1ndeno[l,2,3-cd]pyrene  could  not be  located  1n
the available literature.

    Naphthalene.    A chronic  Inhalation study  of  naphthalene  1n mice  has
recently been conducted by  the National  Toxicology Program  (NTP, 1991).   The
      f
exposure phase  of the  2-year  Inhalation study  has  been completed, but  the
Mstopathology data  and the  final  report  are not  yet available.   No  other
chronic toxlclty data were found.

    Phenanthrene.    Pertinent  data  regarding the   subchronlc  and  chronic
nonoral  toxlclty  of  phenanthrene  could not  be located   In  the  available
literature.                         .

    Pyrene.    Pertinent  data  regarding   the  subchronlc  and  chronic  nonoral
toxlclty of  pyrene could not be located  In  the available literature.


04420                                V-34                            10/08/91

-------
Target Organ ToxUUy
    The target, or preferred, organs for the toxic action of PAHs tend to be
diverse.  This  1s a  consequence  of  two  propensities of  the  PAHs.  First,
there  1s  the  preferential  association  for  normally  proliferating tissues
such as bone marrow, lymphold organs, gonads and Intestinal epithelium (U.S.
EPA, 1980d). Second, PAHs are distributed extensively throughout the body 1n
general.

    The toxldty  of  various PAHs  seems  to center on the hematopoletlc  and
lymphold  systems   of   different  animal  species.   Robinson et  al.  (1975)
observed  hemopoletlc  and  lymphold  effects  of  benzo[a]pyrene  1n  mice  (of
certain   strains),   and   Lasnltzk!  and   Woodhouse   (1944)  found  dlbenz-
[a,h]anthracene to affect the lymph  nodes  of  rats.   While the hematopoletlc
and  lymphold  systems were  also  shown to  be  affected 1n  dogs  treated  with
naphthalene (Zuelzer and  Apt,  1949),  the  two major  target  organs  for  this
PAH  are  nondllated  bronchlolar  epithelial   cells   (Clara  cells)   and  eye
tissue.   Cataracts  have  been   shown   to  develop   1n  several  species  of
laboratory  animals  following exposure  to  naphthalene by  routes  other  than
Inhalation  (FUzhugh  and Buschke,  1949;  Shlchl  et  al.,  1980;  Sh1ch1  and
Nebert, 1982).

    Immunotoxlc effects as  a  consequence of PAH exposure  have  been studied
by a  number of researchers.  Malmgren et  al.  (1952) first reported reduced
hemolysln tHres  to  sheep red blood  cells  (SRBC) 1n mice exposed to dlbenz-
[a,h]anthracene and  benz[a]anthracene.   Subsequently, 1t  was  reported  that
exposure  to carcinogenic levels  of  benzo[a]pyrene  resulted 1n  a  depressed
Immune  response  to  SRBC that  persisted  for  90  days  (Stjernsward,  1966,


04420                                V-35                             10/08/91

-------
1969).   Both  humoral  and  cell  mediated  immunity  have  been shown  to  be
                          k/k
depressed  1n  C3H/Anf(H-2 '  )  mice  Injected  with  150 wg  benzo[a]pyrene/kg
bw  during  days  11-17  of gestation.   Production  of plaque-forming  colonies
against SRBC  (as  measured by the Jerne plaque assay)  was  depressed  severely
from  1-4 weeks of age and  persisted from  5-18 months of  age.  An  Ijn  vitro
graft vs. host  assay also showed  1mmunodepress1ve  effects.   Lymphocytes  from
Immature mice exposed to  benzo[a]pyrene (1-4 weeks) were severely  limited  1n
their ability to show mixed lymphocyte responses  (Urso  and  Gengozlan,  1984).

    B6C3F1  mice  were  treated s.c. dally  for 14 days with  5>, 20 or 40  mg/kg
benzo[a]pyrene  (Blanton   et  al.,   1986).   Polyclonal   antibody responses  to
I1popolysacchar1de  (IPS)  and purified protein  derivative  (PPD) measured  by
Jerne plaque assay were decreased  50-66%  after 7 days  exposure.   Exposure  of
B6C3F1 mice to  s.c. Injections  of  40 mg/kg  for 7 or 14 days  resulted  1n a  73
and 98X suppression of the  T-cell-dependent antibody response,  respectively,
as measured 1n a similar  fashion.
                  \
      s
    Similar results were  reported by Dean  et al.  (1983)  for female  B6C3F1
mice  Injected  s.c.  for  14  days  with 5,  20  or 40 mg/kg  bw.  Exposure  to
benzo[a]pyrene  resulted  1n  decreased numbers of  IgM   and  IgG  plaque-forming
cells In response to  SRBC antigen and reduction 1n IgM plaque-forming  cells
1n  response  to IPS.  In  this  assay  cell-mediated  Immunity (as measured  by
delayed cutaneous hypersensUlvlty  to keyhole limpet  hemocyanln), allograft
rejection  and  susceptibility  to  LIsteMa  monocytogenes  was  unaffected  1n
benzo[a]pyrene treated*mice.
04420                                V-36                            10/08/91

-------
    A comparative  study  of  the ability of PAHs  to  depress  antibody  response
was  undertaken by  White  et  al.  (1985).   PAHs  were  administered  s.c.  1n
either  single  or  multiple doses  to  B6C3F1  and DBA/2 mice.  IgM  response  to
SRBC was  measured  1n a  hemolytlc  plaque  assay.  Single exposures of  B6C3F1
mice  to  1   mmol/kg  d1benz[a,h]anthracene   and  benzo[a]pyrene  resulted  1n
depression of  antibody response.   Fourteen  days of exposure to a  dally dose
of  160  ymol/kg of  the  following  PAHs also  resulted  1n suppression  (-60%)
of   the   IgM  response:    benz[a]anthracene,   d1benz[a,h]anthracene   and
benzo[a]pyrene.   Neither  anthracene  nor  chrysene  exposure   significantly
affected  response.   Immunosuppresslon by  benzo[a]pyrene was  observed to  a
greater extent  1n  DBA/2  mice.  The  B6C3F1  mouse strain Is  highly Indudble
for aryl hydrocarbon hydroxylase;  the DBA/2 mouse strain 1s  not.

    The Immunocytotoxlc effects of benzo[a]pyrene were  studied by  Wojdanl  et
al.  (1984).   Elght-week-old,  Inbred  C3H/FCUM  and  C57B1/6CUM mice In  groups
of  six  animals were Injected  1.p. with  P-815 tumor target  cells.   This  was
followed  10  days  later  by 1.p. Injection with  either  0, 0.5,  5 or  50 mg/kg
      /
bw  of  benzo[a]pyrene In  corn  oil.   After  24 hours splenic lymphocytes  and
peritoneal exudate  lymphocytes were  collected and  assayed for  target  cell
binding and  target  cell  killing effects.   A consistently  decreasing rela-
tionship was  noted  between the dose  of  benzo[a]pyrene and  both  the binding
and killing  of  target  cells  for splenic and  peritoneal  lymphocytes.   At  the
two  highest* dosages  of   benzo[a]pyrene   (5  and  50 mg/kg  bw),   significant
decreases 1n  the  percent of lymphocytes  binding to target  cells  or killing
target  cells  existed as  compared   with  the  controls.   This   study  also
compared  lymphocyte function  of  animals  treated with  3-methylcholanthrene
04420                                V-37                            10/08/91

-------
and  the weak  carcinogen  benzo[e]pyrene.   Lymphocytes  from  benzo[e]pyrene-
treated  animals  were  essentially  similar  to  those  from  control  animals
regarding  target  cell binding and cytotoxlc effects.   The authors  Indicated
that  lymphocyte mediated  Immunity  may be  Inhibited by  PAHs  and  that  this
Immunosuppressent effect can contribute to their carc1nogen1c1ty.

    The effect  of  PAH exposure on Interferon production  In  vivo  was  studied
by  Griffin et  al.  (1986).  Female C3H  mice were  treated  l.p.  with  0.046,
0.46  or 4.6  mg benzo[a]pyrene 1n corn oil.  At  12,  24, 48,  72 or  120 hours
animals were  Injected  with Sendal  virus  to  Induce Interferon  production.
Eight  hours  after  the  virus was  Introduced mice were bled by  cardiac punc-
ture  and  serum was tHred  for  Interferon by a  cytopathlc effect Inhibition
assay  1n   mouse L929  cells.   Mice  receiving  4.6  mg  (180 mg/kg  bw)  were
significantly depressed  In  their ability  to produce  Interferon  at 12,  48 and
120  hours  after  benzo[a]pyrene  treatment.   At the 48-hour challenge  mice
receiving  the  lower  0.46  mg dose  were  also  significantly  Impaired  with
respect to their Interferon production.

Carc1noqen1c1ty. Oral
    Acenaphthylene.   Pertinent  data  regarding  the  oral carc1nogen1c1ty  of
acenaphthylene  could  not  be  located  1n  the  available  literature.   In  an
abstract Knobloch  et  al.   (1969)  reported  on  the  effects  associated  with
Ingestlon  of  acenaphthylene.  An  oral  dose of 0.6 g/kg  bw of  acenaphthylene
1n olive oil  administered  for 40 days to a group  of seven rats  yielded the
following  physiologic  effects:   "considerable" body weight loss, changes  1n
the peripheral blood  pattern, changes  1n  renal  function,  and  Increased serum
amlnotransferase  activities.   No  carcinogenic  effects  were  reported;  the
mice were dosed for a short period of  time.

04420                                V-38                            10/08/91

-------
    Anthracene.   Studies   of   orally  administered  anthracene   have   been
negative for carcinogenic  effects.   Druckrey  and  Schmahl  (1955)  administered
a diet  containing anthracene  1n oil, 6 days/week to 28 BOI  or BDIII  rats  of
unspecified  sex  for  a   period of  78  weeks.    The  total  dose  was  4.5 g
anthracene/rat.  No  tumors  as  a consequence of treatment were found  to have
                                   •
developed.   Similarly,  1n an  unpublished  report,  Schma'hl and Relter  (n.d.)
administered a  total  dose  of  4.4 g of anthracene orally  to  31 rats during a
33-month study.  Again, no tumors were reported  to have  occurred.

    Benzfalanthracene.   Klein  (1963)  1n  a   study  of  ma-le  B6AF1/J   mice
provides evidence of the  carcinogenic  potential  of  Ingested  benz[a]anthra-
cene.   A  3X solution  of  the  compound 1n  Methocel-Aerosol  O.T.  was admin-
istered by gavage at the rate of  3  doses/week  for  5  weeks.   Control  animals
received oral  doses  of  Methocel-Aerosol  O.T.  alone.  After  340-440 days and
at 547-600 days, the animals were  assessed  for  tumor  development.   The  Inci-
dence of pulmonary adenomas and  hepatomas was Increased at  both  assessments;
the  hepatoma   Incidences  at  547 -days  were  higher  than  at  437  days.   No
      /
statistical treatment of the data  was  reported.   The  findings  are  summarized
1n Table V-l.

    The Klein  (1963) paper  cites an  earlier study by  White  and Eschenbrenner
(1945)  wherein 2/6   rats  receiving  benz[a]anthracene In  the diet developed
multiple hepatomas.

    A single gavage  dose of 0.5  mg benz[a]anthracene  1n mineral  oil  produced
no tumors  1n 13 mice after 16 months.  Multiple  gavage administration,  8  or
16 treatments  at 3-7  day  Intervals  resulted  1n  forestomach paplllomas  1n


04420                                V-39                            10/08/91

-------
                                  TABLE V-l

   Oral Cardnogenlclty Testing of Benz[a]anthracene  Administered  by  Gavage
                             to Male B6AF1 Mice*
Duration
Administered Doseb of Study
(days)
0.5 ml of vehicle 444
0.5 ml 437
3% solution
0.5 ml of vehicle 547
0.5 ml 547
3X solution
Target
Organ
lung
liver
lung
liver
lung
liver
lung
liver
Tumor
Type
adenoma
hepatoma
adenoma
hepatoma
adenoma
hepatoma
adenoma
hepatoma
Tumor
Incidence
(X)
10/38
0/38
37/39
18/39
7/20
2/20
19/20
20/20
(26)
(0)
(95)
(46)
(35)
(10)
(95)
(100)
aSource: Adapted from Klein, 1963

bBenz[a]anthracene  In  Methocel-Aerosol  or  vehicle only.
 treated 15 times (3 treatments/week for 5 weeks).
                       All animals  were
04420
V-40
05/14/91

-------
2/27 treated mice  compared  with  0/16 mineral oil-treated controls  (Bock  and
King, 1959).

    Benzofalpyrene.   A  relationship  between  the  1ngest1on   of  benzo[a]-
pyrene and  the  development  of  tumors has been documented  1n several  studies
1n mice.   Both  benign  and  malignant tumors of  the  forestomach as a  conse-
quence of  oral  benzo[a]pyrene treatment  have been  reported by a  number  of
authors (Hartwell, 1951;  Shublk  and  Hartwell,  1957,  1969; Thompson and  Co.,
1971; Tracor/JUco, 1973a,b; Wattenberg,  1972,  1974).
                        *•
    Berenblum and  Haran  (1955)  examined  tumor  Induction 1n the  forestomach
of  male   C3H  and  Swiss  mice  using  a  number   of  carcinogenic  substances,
Including   benzo[a]pyrene.   Mice  were   starved  for  18  hours   prior   to
treatment,  which  consisted  of 0.3 ma of a 0.5X benzo[a]pyrene  solution  1n
PEG-400 administered  by  stomach  tube on  a weekly basis.   No  concurrent
controls   were  reported.  In  one part of  this   study,  treatment  was   for  30
weeks  to   C3H  mice,   resulting  1n  the  formation  of  one papllloma  and  16
carcinomas  of the  forestomach.   In a second experiment, Swiss  mice were  fed
only milk and water for  3 days prior  to  the customary 18-hour  fasting period
to eliminate  the  effects of undigested  food 1n the  stomach at the  time  of
dosing.   Under  these  conditions, 85X (17/20)  of  the  animals  had  tumors  of
the  forestomach.   In  neither experiment   was  any  tumor  of  the  glandular
mucosa of the stomach  found.

    As. part of  an experiment  Investigating tumor  promotion  by dtrus  oils,
albino mice  (sex  not  stated)  were  administered  a  single  oral  gavage  of
either 0,  12.5,  50  or  200  vg  benzo(a)pyrene  1n  polyethylene  glycol  400


04420                                V-41                           10/08/91

-------
after  food had been  withheld  overnight.   The Intestinal  tracts  of one-half
of  the surviving mice  1n  each group were  examined  280 days post-treatment.
The  Incidences  of  benign forestomach tumors were  0/9,  2/10, 0/9  and 5/11  1n
the  control,  low-, mid-  and high-dose groups,  respectively (Pierce,  1961).
The  remaining  mice  were  examined  upon  natural  death  or  569  days  after
treatment.   The Incidences  of  benign  forestomach  tumors were  0/17,  3/17,
0/19   and  8/17   1n    the   control,   low-,   mid-  and   high-dose  groups,
respectively.   The total number  of  tumors found  1n  each  group was 0,  3,  0
and  27,  respectively.  No  carcinomas were  observed  and  no  tumors  at  other
sites  were noted by the authors  (Field and Roe, 1965).

    As  part  of a study  on  the  effect  of  hormonal  state  on  PAH  tumor
Induction,  virgin  female  BALB/c/Ch/Se  mice were  gavaged twice  weekly  with
almond   oil   solutions  of   benzo[a]pyrene   (B1anc1f1or1   et  al.,  1967).
Treatment  was for  15  weeks  resulting 1n  a  total  dose of  15 mg.  Treatment
groups  consisted  of  -25  each  Intact and ovarlectomlzed  animals  receiving
benzo[a]pyrene  alone  or  1n  conjunction  with  500-1000  yg/l  esterone  1n
the  drinking  water.   No concurrent untreated  controls  were reported.   Among
Intact  animals  receiving  benzo[a]pyrene  alone  forestomach tumors  (5/25),
mammary tumors  (2/25) and lung tumors (no  Incidence reported) were observed.

    Fedorenko and  Yansheva  (1967) administered benzo[a]pyrene In  trlethylene
glycol  by  gavage  to  the "antlum  of the stomach"  of CC5_  mice  (sex  not
specified).   The authors used  this  solvent based on Us purported ability to
enhance  the carcinogenic effects  of  hydrocarbons.   The  experimental  period
was  stated  to be 19 months.   As  H  was  also  stated  that  the mice were  kept
until  natural death,   there  Is some  confusion as  to the  length  of exposure


04420                                V-42                            10/08/91

-------
and observation periods.  Gavage doses of  100, 0.01,  0.1,  1  and  10 mg/an1ma1
were  given  10  times.   Incidence  data  were   not   reported   for  solvent
controls.  Tumors were found  In the  liver, mesentery,  peritoneum and  stomach
(Table V-2).

    Roe et al. (1970)  Investigated the use of  sodium  cyclamlde  saccharin and
sucrose  on  the  carcinogenic  activity  of  benzo[a]pyrene  1n  female  Swiss
mice.  A  single  gavage dose  of 50  yg benzo[a]pyrene  1n  polyethylene  glycol
(PEG)  was  given  and  the  animals  were  killed  after  18  months.    Tumor
Incidence 1s given In Table V-3.

    As part  of a  study  on  the effects  of  caffeine  1n  tumor   Induction  1n
rats,   Brune   et    al.    (1981)   administered   0.15   mg/kg/treatment   of
benzo[a]pyrene  to  Sprague-Dawley  rats  for   2  years.   Benzo[a]pyrene  was
administered to three  groups  by 1.5X caffeine  gavage  and  to two  other  groups
through diet.  Untreated and  caffeine gavage (5  times/week)  groups were used
as  controls  (32  rats/sex/group).    Three  different  gavage treatments  were
administered to 32  rats/sex/group;  1n  the first treatment  rats  were  gavaged
with  benzo[a]pyrene  5  times/week   for  a  total annual   dose  of  39  mg/kg
benzo[a]pyrene, 1n  the second rats  were gavaged every  third day for a total
annual dose  of  18  mg/kg benzo[a]pyrene and 1n the  final  every ninth day for
a total annual dose  of 6  mg/kg benzo[a]pyrene.   Rats  that died  spontaneously
or  were  killed  when  moribund  underwent hlstologlc  examination  of  various
organs.   The  combined  Incidences   of  benign  and  malignant   tumors  were
significantly  different  from  controls  (p<0.05,  test)  for the  5  times/week
and the  every  third day gavage groups and at  p<0.10  for  the every ninth day
gavage group.  (For other results  see Table V-4.)


04420                                V-43                            10/08/91

-------
                                  TABLE V-2
           Tumor Incidence 1n "Atrium of Stomach" Following Gavage
                      Administration of Benzo[a]pyrene*
Total Benzo[a]pyrene Dose (mg)                     	Tumor  Incidence
                                                   Carcinoma       Papllloma

             0.01                                     0/16            0/16
             0.10                                     0/26            2/26
             1.00                                     0/24            5/24
            10.00                                    11/30           12/30
           100.00                                    16/27            7/27

*Adapted from Fedorenko and Yarisheva,  1967
04420                                V-44                            10/08/91

-------
                                  TABLE V-3

         Incidence of Tumors Observed After a Single Gavage Treatment
              of Mice with  Benzo[a]pyrene  1n  Polyethylene Glycol*
Tumor Incidence
Treatment Forestomach
Papnioma Carcinoma
PEG 2/65 0/65
PEG * 20/61 1/61
0.25 mg
benzo[a]
pyrene
Lung Liver Malignant
Lymphoma
15/65 5/65 3/65
18/61 9/61 0/61
*Adapted from Roe et al., 1970
04420
V-45
10/08/91

-------
o
*fc
4k
o
                                                        TABLE V-4
                              Carc1nogen1c1ty  of  Oral  Benzo[a]pyrene  1n  Sprague-Dawley  Rats3
Mode of Application
Untreated control
Gavage caffeine control
Gavagec (5 times/week)
Gavagec (every 3rd day)
Gavagec (every 9th day)
Diet (5 days/week)
Diet (every 9th day)
Total
Dose B[a]P
(mg/kg)
0
0
39
18
6
39
6
Fores tomach Tumor
Benign
2/64
3/64
14/64
25/64
11/64
9/64
1/64
Incidence0
Malignant
0/64
0/64
0/64
1/64
1/64
0/64
0/64
Total Tumors
3/64
6/64
14/64
26/64
13/64
10/64
3/64
CO
10
aSource: Adapted from Brune et al., 1981
°Comb1ned Incidence for males and females.  Effective number of animals was assumed to be 64.
C1.5X aqueous caffeine plus benzo[a]pyrene

-------
    Two  other  groups   of  32  rats/sex  were  fed  benzo[a]pyrene  either   5
times/week  [total  annual dose  39 mg/kg  benzo[a]pyrene  or  every ninth day
[total annual dose  of  6 mg/kg benzo[a]pyrene.  Rats that died  spontaneously
or  were  killed  when moribund  underwent  hlstologlc examination  of  various
organs.   The  combined  Incidences  of  benign  and malignant   tumors   were
significantly different  only 1n  the group  fe'd  benzo[a]pyrene  5  times/week
(p<0.05) (see Table V-4).

    Demi et a "I.  (1983)  dosed five groups of  four  female Sprague-Dawley  rats
with  various  combinations of  benzo[a]pyrene  and polychlorjnated  blphenyls
(PCBs) dissolved  1n olive oil by  means  of  gastric Intubation for  12 weeks.
After 12 weeks, the  livers were examined  to  determine  the  number  and  area  of
enzyme-altered  Islands.  Rats  treated  with  PCBs  before  Intubation   with
benzo[a]pyrene and promoted afterwards with PCBs  showed a 3-fold  Increase  1n
the  number  of  enzyme-altered  Islands  present.   The  total  area  of  these
Islands was also  3 times larger  than  those  found 1n the other test  groups.
PCB or benzo[a]pyrene alone  appeared to produce no significant changes, but
the  pretreatment  with  PCB along  with  the  administration  of benzo[a]pyrene
followed by PCB  caused an alteration  1n  benzo[a]pyrene  metabolism that led
to the formation, of enzyme-altered foci.

    As part  o/  t. study of   Inhibition  of  tumor formation by organoselenlum
compounds^ CD-I mice (25/group) were gavaged with  benzo[a]pyrene.  Corn oil
solutions  deHveMng  1  mg  benzo[a]pyrene/0.2  ml were  administered  twice
weekly for  4  weeks.  Controls were  treated  with corn  oil  only.   Nice  began
treatment at  9  weeks of age and  were killed  at  28 weeks.  At this  time  no
control animals had  developed forestomach papHlomas, whereas 85X (17/20)  of

04420                                V-47                           10/24/91

-------
benzo[a]pyrene  treated  animals  developed an average  of  S.3+.2.9 tumors/mouse
(El-Bayoumy, 1985).

    Robinson et al.  (1987)  administered  benzo[a]pyrene to  A/J  mice by gavage
as a positive control on  a  study  of coal tar paints.   Forty female mice were
gavaged with  0.25 mg benzo[a]pyrene  1n  2.2 ma 2% emulphor  twice  weekly for
8 weeks (total  dose  6 mg).   Animals were killed at 8 weeks and were examined
for lung  adenomas and  forestomach tumors.   8enzo[a]pyrene  treatment produced
lung adenomas  1n  61%  (22/36)  of mice with an average  of  1.42^0.40 tumors/
mouse  that  was  significantly Increased by  comparlslon with  controls (29% or
11/38   Incidence,  0.32+0.09  tumors/mouse).    No  forestomach  tumors  were
observed  In controls.   Benzo[a]pyrene treatment  resulted  1n  92%  Incidence;
67% of animals had paplllomas and 61% carcinomas.

    As part of  a  dietary  carclnogenesls  study,  benzo[a]pyrene  (1.5 mg, twice
weekly  for  4  weeks)  was  administered  by gastric  Intubation to 25-30 female
ICR  mice/group .(Benjamin  et  al.,  1988).   Three  groups  Intubated  with
benzo[a]pyrene  had different  feeding  regimens.  The  first  two groups  were
      /
administered  a  basal   diet  supplemented   with   20%  soy  sauce  and  ±0.05%
nitrite.   The  third  group  received  only  the basal  diet  and water.   The
feeding regimens  were  started 1  week  before the  Initial  gastric  Intubation
and were  continued throughout the  experiment.   After 21 weeks  on study all
mice were  sacrificed.   The  tumor  Incidences of the mice 1n  the three groups
were similar  (91%);  however, mice  receiving the soy supplemented  diet  and
nitrite supplemented water  had a  significantly  lower  number  of neoplasms per
mouse  (3.4) than  the other  two groups (5.2 and  4.0 for the group receiving
no supplementation and  the  group receiving only  the soy  supplemented diet,
respectively).  In a follow-up experiment, a fourth  group  was added;  this

04420                                V-48                            10/24/91

-------
group was fed  the  basal  diet  and  received  nitrite supplemented water.   After
the 4-week  benzo[a]- pyrene dosing,  the  feed and water of  some  groups  were
changed  to  allow  further  evaluation  of  the  effects.    This   experiment
Indicated that the  Inhibition of  neoplasla  by soy sauce and nitrite probably
occurred  during  tumor  promotion.   In a  subsequent  study soy  sauce at  a
dietary concentration of 20% was  found  to  produce a  significant reduction In
forestomach  neoplasms  In benzo[a]pyrene-dosed  female  ICR   mice  (Benjamin et
al., 1991; see also Benjamin et al., 1989).

    Dietary  benzo[a]pyrene  at various  doses  was  administered  to  mice (Neal
and Rlgdon,  1967;  Rlgdon and Neal,  1966,  1969).   These findings  are  summa-
rized 1n Tables V-5 and V-6.

    Using male and female CFW-SwIss  mice.  17-180 days old, Neal  and  Rlgdon
(1967) found that  a  dose-response relationship existed between the Incidence
of  stomach  tumors   (paplllomas and  carcinomas) and  long-term,  oral  exposure
to benzo[a]pyrene  1n the diet.  Animals were  fed  a diet containing .0-250 ppm
      /
of benzo[a]pyrene  for <197  days.   No tumors  were  found In  the control group
of 289 animals or  1n the groups  treated with  1, 10 or  30 ppm benzo[a]pyrene.
The Incidence  of  tumors, however,  Increased  between the 40 ppm  and 250 ppm
benzo[a]pyrene dosages (see Table. V-5).  In a  second  experiment  by Neal and
Rlgdon (1967)y mice were fed  100  or  250 ppm benzo[a]pyrene  for 1  day with no
          >-**;,•..,, -
gastric  tuaort developing within  105 days.   However,  SOX  of mice  fed  5000
          :$vs**-v£.
ppm benzp{»]pyrene for  1 day did have gastric tumors  upon examination  -113
          ••'*  _  r
days later.  Groups of mice were also given  food containing  250 ppm  benzo-
[a]pyrene for  1,   2, 4  or 30  days.   Upon  sacrifice 77-104 days  later,  the
tumor  Incidences were  0/10 (1 day),  1/9  (2  days),  1/10 (4 days)  and 26/26
(30 days).   It was  concluded  that the development of gastric tumors In these

04420                                V-49                             10/24/91

-------
                                   TABLE  V-5

        Incidence of Forestomach Paplllomas and Carcinomas 1n Hale and
           Female CFW Mice Administered Benzo[a]pyrene 1n the Diet*
Dose 1n
Diet
(ppm)
0
1
10
20
30
40
45
50
100 '
250
Total
Consumed
(mg)

0.48
4.48
8.88
13.32
17.76
19.8
21.4-29.4
39.2-48.8
70-165
Duration of
Treatment
(days)

110
110
no
110
no
no
107-197
98-122
70-165
Duration
of Study
(days)
70-300
140
140
226
143-177
143-211
141-183
124-219
118-146
88-185
Tumor
Incidence
(X)
0/289 (0)
0/25 (0)
0/24 (0)
1/23 (5)
0/37 (0)
1/40 (3)
4/40 (10)
24/34 (71)
19/23 (83)
66/73 (90)
*Source: Adapted from Neal and Rlgdon, 1967
04420
V-50
10/08/91

-------
   o*t
 250+
                                  TABLE V-6

                Carc1nogen1c1ty of Benzo[a]pyrene Administered
                  In the Diet to Male and  Female Swiss  M1cea

Oose
(ppm)
Duration of
Treatment1* Target Organ Tumor Type
(days)
Tumor
Incidence
(X)
38-210+
80-140
stomach
lung
hematopoletlc
system

stomach
lung
hematopoletlc
system
papl1loma/carc1noma
adenoma
leukemia
papllloma/carclnoma
adenoma
leukemia
 2/175   (1)
33/151  (19)
 0/175   (0)
69/108  (64)
52/108  (48)
40/108  (37)
250*
250*
1000*
1000*
72-99
147-196
73-83
127-187
stomach
lung
stomach
lung
stomach
lung
stomach
lung
papllloma/carclnoma
adenoma
papl 1 lomas/cardnoma
adenoma
papllloma/carclnoma
adenoma
papllloma/carclnoma
adenoma
12/52 (23)
26/52 (50)
9/13 (69)
10/13 (77)
5/9 (56)
7/9 (78)
13/13 (100)
3/13 (23)
aSource: Adapted from Rlgdon and Neal, 1966*, 1969+

bThe duration of the treatment 1s equal to the duration of the study.
04420
                       V-51
                                           10/08/91

-------
mice Is  Influenced both  by  the  amount  of  benzo[a]pyrene consumed per day and
the number of days It 1s fed.

    In  another   set  of  studies,  Rlgdon  and  Neal  (1966,  1969) also  found
evidence  of  an  association between  chronic  dietary  exposure  to  benzo[a]-
pyrene and tumor  Incidence  In  the stomach and  lung, as  well  as  the develop-
ment of  leukemia In mice (see  Table  V-6).   Rlgdon and Neal  (1966)  fed  male
and female Swiss mice dietary concentrations  of benzo[a]pyrene of  0, 250 and
1000 ppm for different  periods  of  time.  The  Incidence of  stomach  tumors
(paplllomas and carcinomas) was related to both dose and length  of exposure.
                                                            •
All the  mice In the 1000 ppm group were  found to have  gastric  tumors  after
86  days  of  benzo[a]pyrene  consumption.    A  similar   set  of  relationships
between tumor Incidence, dose and time held  fairly well  for lung adenomas as
well.   Rlgdon and Neal  (1969) also found  that the occurrence of  leukemia was
related to 1ngest1on of  benzo[a]pyrene.   Of mice  fed  250 ppm benzo[a]pyrene
In their diet over an 80-140 day  period,  69/108 (63.9X) developed paplllomas
or carcinomas of the stomach.  Similarly,  lung adenomas occurred  1n  52/108
(48.IX) mice fed diets  with 250  ppm  benzo[a]pyrene.  Finally,  40/108  (37%)
of the  treated  mice developed  leukemlas.   The ability of  dietary  benzofa]-
pyrene  to  produce mouse  lung  adenomas was  also  confirmed  by  the work  of
Hattenberg and Leong (1970)  and Wallenberg (1974).

    Previously,  a single oral  dose  of 100 mg  benzo[a]pyrene  1n  the diet was
shown  to  produce mammary tumors  In 8/9  female Sprague-Dawley rats  (Hugglns
and Yang, 1962).  Sprague-Dawley  rats  of  both  sexes were treated  dally  with
2.5 mg  benzo[a]pyrene.   Forestomach and  esophageal  paplllomas  developed In
3/40 rats (61bel, 1964).


04420                                 V-52                             10/08/91

-------
    Chouroullnkov et al.  (1967)  administered  diets  containing  benzo[a]pyrene
dissolved 1n olive oil  to  two  groups  of  -80 male albino mice for  a period of
14  months.   Based on  an assumed  food  consumption  of  5 g/day,  the  authors
calculated a  total  dietary dose  to  be  8 mg.   One  group of treated  animals
was permitted water ad_  libitum while  the  second drank  a 3% ethanol solution.
Controls  consisted  of  40 mice  given the  standard diet  supplemented  with
olive oil and a  group of  81  fed  the  untreated diet  only.  Forestomach tumors
(paplllomas) were  found 1n  5/81  surviving animals receiving  benzo[a]pyrene
and  water and   1n  8/81  treated  mice  given   the  3%  ethanol  solution.   No
gastric tumors were observed among controls.

    As part  of  a study  of  the effects  of phenolic antloxldants  and  ethoxy-
quln  on  PAH carclnogenldty,  benzo[a]pyrene  In sesame  oil  was  administered
1n  the diet  to  female  Ha/ICR and A/HeJ mice  (Hattenberg,  1972).   The Ha/ICR
mice  received 0.4 mg/day for 28 days before  being  returned  to  a  normal  diet
for 27 weeks  or 1.26 mg/day  for 28  days  before being  returned  to a normal
diet  for  14  weeks.   Forestomach  tumors  were  observed   1n  11/20  low-dose and
13/19 'high-dose  animals.  Data  on untreated control   Ha/ICR  mice were  not
reported.   The   A/HeJ  mice consumed  0  or  4.8  mg  benzo[a]pyrene/day for  2
weeks, beginning at  age  9  weeks, and were  killed 10 weeks after the  last
treatment.   All  treated  (12/12)  but no  control  (0/12)  animals  developed
forestomach tumors.

    In a similar  study of  sulfur-containing  compounds,  female  Ha/ICR  and
A/HeJ mice  consumed  diets containing 300 ppm benzo[a]pyrene  1n  sesame  oil
(Wattenberg, 1974).  The  Ha/ICR  mice  were maintained  on the diet from age 9
weeks to  15 weeks and observed  until  29  weeks at which time they were killed
and examined  for forestomach tumors.  These  were observed  In  8/20  animals;

04420                                V-53                            10/08/91

-------
no untreated  controls  were reported.  A/HeJ mice  started  on treatment at 9
weeks  of  age and  remained  on the  diet  for 25  days.   On days  7  and 21 of
treatment  animals  were gavaged  with 3  mg  benzo[a]pyrene 1n  0.5  ma  sesame
oil.   At 31 weeks  the mice were  killed and  evaluated  for pulmonary adenomas.
These were present  1n  100X (12/12)  of the animals with a mean number  of 7.8
tumors/mouse.  No untreated controls were reported.

    THolo et al.  (1977) studied production of forestomach tumors 1n mice as
related to IndudbllKy of aryl  hydrocarbon hydroxylase.  Female Ha/ICR mice
9 weeks  of age  were  fed  diets  containing  5X  corn oil or 5%  corn  oil with
benzo[a]pyrene to  constitute  0.2 or  0.3  mg/g  diet.  Treatment  was continued
for 12  weeks  at  which point  the animals  were  killed  and stomachs only were
examined hlstologlcally  for   tumors.   Results  are summarized  1n Table V-7.
Tumors were of the squamous  papHloma type.  Gross observation  of glandular
stomach, lung and  liver  revealed  no tumors   1n either control  or   treated
animals.
                                                                          •
    Further   evidence  of  cardnogenlclty  associated  with  1ngest1on  of
benzo[a]pyrene 1s  found 1n the work of McCormlck et al. (1981).  A group of
20 Inbred  virgin female LEM/Ha1  rats received  a single IntragastMc dose of
50 mg  benzo[a]pyrene  1n  sesame  oil,  while  a second  group received the same
total  dose  1n 8 weekly  fractions  of 6.25  mg.   Mammary carcinoma Incidence
after 90 weeks was 77X In the 50 mg  benzo[a]pyrene single dose group and 67X
1n the fractionated  dose  group.   Mammary  tumors were  observed 1n  SOX of
untreated rats.

    Adrlaenssens  et al. (1983) Investigated the effect of dietary butylated
hydroxyanlsole,  a  phenolic  antloxldant,  on the  formation of benzo[a]pyrene

04420                                V-54                           10/08/91

-------
                                  TABLE V-7
  Induction of Forestomach Tumors In Ha/ICR Mice Fed Dietary Benzo[a]pyrenea
Dose
(mg/g diet)
0
0.2
0.3
Tumor Incidence
0/9
6/9
9/9
Tumors/Mouse
0.0
1.8
4.0
Carcinogenic
Indexb
0.0
121.9
400.0
aSource: Trlolo et al., 1977
bPercentage of tumor-bearing mice x mean number of tumors/mouse
04420                                V-55                            10/08/91

-------
 metabolHe-DNA and  -protein  adducts  1n  the  lung,  liver  and  forestomach of
 female  A/HeJ mice.  The  mice were fed  benzo[a]pyrene  (from 2-1351 ymol/kg)
 In   Identical  diets  to  which  butylated  hydroxyanlsole  (5  mg/g  diet,  ~1
 g/kg/day)  was  or was not added.  Adduct formation, thought by the  authors to
 be   a  necessary,  but   not   a  sufficient,   step  1n  the  development  of
.benzo[a]pyrene  Induced  tumors,  was  examined. 48  hours  after  Ingestlon of
 benzo[a]pyrene.   The major  DNA adduct Identified In each tissue at  each  dose
 was    the   (+)-7,8-d1hydroxy-9,lO-epoxy   7,8,9,10-tetrahydrobenzoa)pyrene:
 deoxyguanoslne;  other  adducts,  Including 7,8-d1hydroxy-9,lO-epoxy  7,8,9,10-
 tetrahydrobenzo(a)pyrene:deoxyguanos1ne   and   (-)-7,8-  d1hydroxy-9,lO-epoxy
 7,8,9,10-tetrahydrobenzo(a}pyrene:deoxyguanoslne, were  Identified.

     Formation  kinetics  of  the  major  adduct  1n  lung and  liver  ONA  from
 animals  on the control  diet showed a sigmoldal curve;  forestomach  adduct-DNA
 complexes  exhibited no saturation over  the levels  tested.   As the  benzo[a]-
 pyrene  dose approached  0,  the  dose-response  curves  became linear; however,
 1n   the  three  organs  examined,  no  threshold dose  appeared to  exist below
 which benzo[a]pyrene metabolite adducts  were not  observed.

     Dietary butylated   hydroxyanlsole  treatment  Inhibited  the   formation
 benzo[a]pyrene metabolite adducts to forestomach,  lung and liver  DNA over a
 wide dietary benzo[a]pyrene  range.   Adduct  formation  In  DNA  from the fore-
 stomach  of  butylated  hydroxyanlsole-treated  animals was  45X  lower than  the
 control  group, but demonstrated  the  same  binding kinetics as the  animals In
 the   control  diet.  The  maximum Inhibition   of  lung  and liver  DNA  adduct
 formation  1n  butylated  hydroxyanlsole-treated animals was  68  and 82% lower
 than the control group, respectively.  As the benzo[a]pyrene dose  approached


 04420                                V-56                            10/08/91

-------
these values declined  to  40  and  55X,  respectively.   Butylated  hydroxyanlsole
treatment also diminished the curvilinear nature of the dose-response curve.

    The  level  of benzo[a]pyrene  metabolites  binding  to  DNA  was -1%  of  the
amount   of   binding  to  proteins  1n   the  three  organs  examined.    The
dose-response  curves  for benzo[a]pynene  metabolites  binding  to  lung  and
liver   proteins   was   parallel   for   both   the   control   and   butylated
hydroxyanlsole-treated  groups.    Inhibition  of  metabolite binding  was  not
dose  dependant  1n  these   organs.   No  consistent   effects   of   butylated
hydroxyanlsole were noted 1n  benzo[a]pyrene metabollte-forestomach  protein
binding kinetics.

    Hamsters have also been  observed  to  develop  paplllomas and  carcinomas  of
the alimentary tract  In  response to gavage or dietary exposure to  benzo[a]-
pyrene  (Dontenwlll   and   Mohr,   1962;  Chu  and  Malmgren,  1965).    Chu  and
Halmgren  (1965)  fed male Syrian  hamsters  diets  containing 500  mg  benzo[a]-
pyrene/kg food or 500  mg benzo[a]pyrene plus 5 g vitamin A palmltate  or  5 g
      /
vitamin A palmltate  only.   This  diet was  provided  4  days/week  and  standard
diet the  remaining  3 days.  Animals  consumed  -5 g food/day,  thus  receiving
~10  mg  benzo[a]pyrene  and/or   100 mg  vitamin A  palmltate/week.    All   10
animals  fed  vitamin A free  diet died 1-4 weeks  after the beginning  of  the
assay,  whereas  those consuming  vitamin  A  survived  <5  months.    Hamsters
consuming benzo[a]pyrene with  vitamin  A  lived  longer (that  1s,  from 6-14
months)  than  animals  fed   benzo[a]pyrene  and  no  vitamin A.   Nine  of   13
treated animals developed tumors  described by  the authors  as  "cancer"  1n  the
forestomach  (9)   and  Intestine  (2).   Vitamin A  1n  the  diet eliminated
Intestinal tumors and  decreased the  severity  of the  forestomach lesions  to


04420                                V-57                            10/08/91

-------
lesions  described  as  papHlomas  and dyskeratoses.  Tumor  Incidence  In the
group  receiving  only  vitamin A  was  0/27  for  esophageal  lesions;  the
Incidence of forestomach tumors was not  described.

    BenzoTblfluoranthene.   Pertinent  data  regarding the  oral carcinogenic-
                                   •
Hy of benzo[b]fluoranthene could  not be located  1n the available  literature.

    BenzoTklfluoranthene.   Pertinent  data  regarding the  oral cardnogenlc-
Ity of benzo[k]fluoranthene could  not be located  1n the available  literature.

    Benzofg.h.1]perylene.   Pertinent  data  regarding the  oral cardnogenlc-
Hy of benzo[g,h,1]perylene could  not be located  1n the available  literature.

    Chrysene.   Pertinent   data  regarding  the   oral   carclnogenlcHy  of
chrysene could not be located 1n the available literature.

    D1benzTa.hlanthracene.    Mice   (strain  not  specified)  were  fetl  a  diet
      /
with  added  d1benz[a,h]anthracene.    After  5-7  months  exposure   the  total
received doses of 9-19 mg  resulted  1n the  Induction of forestomach  tumors In
7/22 survivors at  1  year (Larlnov  and Soboleva,  1938).

    Lorenz and Stewart (1948) administered strain A mice 0.4 mg d1benz[a,h]-
anthracene/day 1n an aqueous mineral oil suspension given In place  of drink-
Ing water.   At  406  days exposure,  2 squamous cell carcinomas and  11 papll-
lomas  of the  forestomach  were observed   (Lorenz  and  Stewart,  1948).   The
authors  also  conducted  experiments 1n which C57B1, C3H,  DBA/2 and A strain
04420                                V-58                            10/08/91

-------
mice were  exposed to  d1benz[a,h]anthracene  In an  olive  oil water  emulsion
provided  to  the  animals  1n  Heu  of  drinking  water  (Lorenz  and  Stewart,
1947).    The DBA/2 mice  developed  tumors  Including pulmonary  adenomatosls,
that Snell  and Stewart (1962a)  concluded were  similar  to  adenomatous lesions
found  1n humans.  Snell  and  Stewart  (1962a,  1962b)  were  thus prompted  to
                                  •
undertake a similar  experiment 1n  mice.   Groups  of 21 male and female  mice
of  the  DBA/2  strain  were  given  0.2  mg/ml  d1benz[a,h]anthracene  In  an
aqueous  olive  oil  emulsion ad  libitum  1n  place of drinking water.   Twenty-
five male  and 10 female control  animals  received the olive oil emulsion  1n
Heu of drinking  water.   Neither  treated nor  control  mice  tolerated  the
olive  oil  vehicle.    Animals   lost  weight   after  a  few  weeks   exposure,
eventually  becoming   emaciated and  dehydrated.   Hales  were  estimated  to
receive  a  dally dose  of  0.85  mg/day while females  received 0.76 mg/day  of
d1benz[a,h]anthracene.   Duration of the experiment was 279 and 237  days  for
male and female mice  In  the  experimental group  and 351  and  226  days  for
controls.   Treated  mice  developed   pulmonary   adenomatosls,   alveologenlc
carcinomas, mammary tumors (females only), precancerous growths of  the small
      /
Intestine and  hemang1oendothel1omas of  the pancreas,  mesentery  and  abdominal
lymph nodes.   A  pulmonary adenomatosls 1n a  male mouse was the only lesion
observed 1n either male  or  female  control animals.  Tumor  Incidences   from
the Snell  and Stewart (1962a)  study  are  compared  with  those  of the Lorenz
and Stewart (1947) study  1n Table V-8.

    A series of assays was done wherein mice of  several  strains were gavaged
twice weekly  with  preparations of 0.5X d1benz[a,h]anthracene In almond  oil.
After a  15-week  treatment  period,  the total  dose  was  15  mg/anlmal.   Mammary
carcinomas  were observed 1n  1/20 BALB/c females and  In 13/24 pseudo-pregnant


04420                                V-59                             11/12/91

-------
o
4»>
4k

O
                          TABLE  V-8

Oral Cardnogen1c1ty of D1benz[a,h]anthracene 1n DBA/2 M1cea
Duration Duration
Sex Feeding Study
(days) (days)
M 177 281




F 158 265





M 279 279




F 237 237





aSource: Adapted from Snell and
^Compound was administered as an
cNo tumors were observed 1n
pulmonary adenomatosls was the

Doseb
(mg/day)
0.56




0.48





0.85




0.76






Tumor Type

Pulmonary adenomatosls
Alveblogenlc carcinoma
Carcinoma, small Intestine
Precancer, small Intestine
Hemang1oendothe11oma
Pulmonary adenomatosls
Alveologenlc carcinoma
Carcinoma, small Intestine
Precancer, small Intestine
Hemangloendothelloma
Mammary card nomad
Pulmonary adenomatosls
Alveologenlc carcinoma
Carcinoma, small Intestine
Precancer, small Intestine
Hemangloendothelloma
Pulmonary adenomatosls
Alveologenlc carcinoma
Carcinoma, small Intestine
Precancer, small Intestine
Hemangl oendothel 1 oma
Mammary carcinoma**
Tumor
Incidence0

4/10
8/10
3/10
6/10
4/10
3/9
7/9
0/9
5/9
2/9
3/9
14/14
14/14
0/14
2/14
10/14
13/13
10/13
0/13
4/13
6/13
12/13

Reference

Lorenz and
Stewart, 1947



Lorenz and
Snell, 1947




Snell and
Stewart, 1962a
.


Snell and
Stewart, 1962a




Stewart, 1962a
aqueous olive oil emulsion provided 1n 11eu of drinking
10 female
only tumor
\}} and Stewart (1962a) study.
°KO mammary carcinomas were observed 1n the
control mice 1n the Lorenz
observed 1n 25 male and 10
male m1ct~
and Stewart (
female control

water.
1947) study. One
mice used 1n fjie


-------
BALB/c  females  (B1anc1f1or1   et   al.t  1976).   A  single  1.5  mg  dose  of
d1benz[a,h]anthracene  In PEG-400  produced  forestomach  paplllomas  In  2/42
male Swiss mice after 30 weeks (Berenblum  and Haran,  1955).

    Fluoranthene.   Pertinent   data  regarding  the  oral  cardnogenlclty  of
fluoranthene could not be located  1n the available  literature.

    Fluorene.   The   carcinogenic   potential  of  fluorene  was   studied  by
Wilson et al.  (1947)  and Morris et al. (1960).  In the Morris et  al.  (1960)
study, female  buffalo rats  were administered 0.05%  fluorene.  1n  their  diets
containing either  3% added  corn oil  or propylene  glycol.  This  resulted  1n
the consumption of either 4.3 mg/day  of fluorene for  ~6 months or 4.6  mg/day
for -18 months.  The  Incidence  of  tumors  1n  the  treatment  and  control  groups
was essentially  the  same (Table  V-9).  The  authors  of the study  described
fluorene as "slightly carcinogenic."

    HUson et  al.  (1947) studied  the  effect on tumor development  1n  albino
rats  of  exposure  to various  concentrations of  fluorene In  the diet  over
various periods  of  time.  One set of  rats was exposed  to  several  concentra-
tions  (number  not specified)  ranging from  0.062-1.OX  fluorene  1n  the  diet
for 104 days while a  second set received  either  0.125,  0.25  or 0.5% fluorene
1n  the diet for  453 days.    Animals  of the  short-term group maintained  on
diets  with  fluorene  concentrations of  0.5 and 1.0%  experienced  significant
decreases 1n  their  rate of growth..  In other aspects  they appeared normal.
The Internal organs  of  rats  exposed for 104  days were  essentially normal  1n
appearance and histology.  Livers  of  rats  consuming the  0.25% and  higher
dose  diets   In the  longer  study  were  significantly heavier  than  normal.


04420                                V-61                             10/24/91

-------
t\>
o
                                      TABLE  V-9


Carclnogenlclty Testing of FlUorene Administered In the Diet to Female Buffalo Rats3
Dose or Duration of
Exposure Treatment
0 mg/day NA •
4.6 mg/dayb 18.1 months
(range, 4.1-19.2)
0 mg/day NA
4.3 mg/dayb 6.1 months
(range, 5.0-6.2)
aSource: Morris et al., 1960
••Reported dally Intake of 0.05X dietary
NA = Not applicable
o
o
CD
Duration of Study
15.5 months
(range. 9.4-19.9)
19.0 months
(range, 5.1-20.1)
13.9 months
(range. 7.8-18.2)
10.2 months
(range. 8.2-10.7)

fluorene



Vehicle or
Physical State
diet with 3X added
corn oil
diet, above plus
fluorene In corn oil
diet with 3X added
propylene glycol
diet, above plus
fluorene added In
corn oil

t



Target
Organ
uterus
uterus
adrenals
pituitary
Inguinal
region
uterus
uterus
R-E system
pituitary
kidney
pituitary
R-E system
kidney
ureter





Tumor Type
adenocarclnoma
flbroeplthellal polyp
cortical adenoma
adenomas
fibroma
flbrosarcoma
carclnosarcoma
granulocytlc
leukemia
chromophobe adenoma
adenoma
chromophobe adenoma
granulocytlc leukemia
squamous-cell
carcinoma
squamous-cell
carcinoma





Tumor
Incidence
(X)
1/18 (6)
2/18 (11)
5/18 (28)
6/18 (34)
1/18 (6)
1/18 (6)
1/18 (6)
1/18 (6)
4/18 (22)
1/18 (6)
2/18 (11)
1/18 (6)
1/11 (6)
1/11 (6)






-------
Spleens of all treated animals weighed less -than  normal as  did  testes  of  the
highest dosed  rats.   In  the longer-term exposure group,  squamous  metaplasia
of the bronchial epithelium was noted  In three  rats  while one rat  exposed to
0.125X  fluorene  1n  the  diet  had developed  a  small  benign  kidney  tubular
adenoma.   Total  number  of  antmals  treated  was  not  Indicated,   nor  was  a
control group described.

    Indenon.2.3-cdlpyrene.   Pertinent  data   regarding   the  oral  carclno-
genldty  of  1ndeno[l,2,3-cd]pyrene  could  not  be located  1n  the  available
literature.

    Naphthalene.   There  1s only  very  limited  Information  available on  the
carcinogenic  potential  of  naphthalene  following   oral   administration   to
laboratory animals.

    Schmahl  (1955)  reported that  naphthalene administered 1n  food was  not
carcinogenic  1n  rats  (In-house  strains  BDI  and BDIII).   Naphthalene  was
      /
dissolved 1n oil and given 6 times/week  1n  food.  The  dally dose was between
10 and 20  mg.  After reaching  a total  dose of 10 g/rat (food Intake was  not
reported), treatment  was  stopped and  animals  observed until  death,  between
700 and 800 days  of  age.

    Tsuda  et al.  (1980)  administered a  single  gavage  dose   of  100  mg/kg
naphthalene  1n  corn oil  to a  group of  10 young adult  F344 rats  (sex  not
specified) at  12  hours after partial  hepatectomy.   A  vehicle control  group
of 10  rats was  Included.   At  2  weeks after  surgery,  2-acety1am1nof1uorene
was added  to  the  diet  at 200 ppm  to Inhibit  proliferation  of "nonreslstant"


04420                                V-63                            10/08/91

-------
hepatocytes.   After  1 week  of dietary  2-acetylamlnofluorene,  a single  2.0
ma/kg  dose of  carbon tetrachloMde  was given  to necrotlze  "nonreslstant"
hepatocytes and permit proliferation  of  "resistant" hepatocytes.  Feeding of
2-acetylam1nofluorene continued for  1 week, followed by  a basal diet for  1
week.   The rats were  then  sacrificed and  Hvers  were  sectioned and  hlsto-
chemlcally  examined  for  the number  and size  of  y-glutamyl  transpeptldase
(GGT)pos1t1ve  fod.   These  foci contain  cells  that are  "resistant"  to  the
necrotlzlng  effects  of  carbon  tetrachlorlde   and   to   the  proliferation-
Inhibiting  effects  of  2-acetylam1nofluorene and are considered  to  represent
an  early   stage 1n  the process  of neoplastlc  transformation.'  Neither  the
number  nor the size  of  6GT  fod  appeared  to  be  Increased 1n  naphthalene-
treated rats compared with  vehicle controls.   The  use of GGT as a  biochemi-
cal marker  of  preneoplastlc  foci  1s generally accepted  (Hendrlch and  PHot,
1987).

    Phenanthrene.   A  single  oral   dose  of 200  mg  phenanthrene dissolved In
sesame  oil  was  administered  to ten, 50-day-old, female Sprague-Dawley rats.
      /
No mammary  tumors were produced within 60 days.  In a positive  control group
of  700 animals given  20  mg  of 7,l2-d1methylbenz[a]anthracene  administered
under  the  same conditions,  mammary  tumor  Incidence  was  100X (Hugglns  and
Yang, 1962).

    Pyrene.  Pertinent  data  regarding  the  oral  carclnogenldty  of   pyrene
could not  be located 1n the  available  literature.

Carc1noqen1c1ty. Other Routes
    Acenaphthylene.   Hartwell  (1951)  cited  a study wherein 20 mice  (strain
not specified)  were skin painted with 0.25X acenaphthylene  In benzene.  At  6
04420                               V-64                           10/08/91

-------
months  13 animals  were alive;  7  were  alive  at  1  year.   No  tumors were
observed.

  .  Rotenberg and  MashbHs  (1965)  reported  "various  degrees of malignancy"
1n  the   lungs  of almost  all  of  an unspecified number  of  rats  exposed  to
acenaphthylene  dust  at  0.5-1.25  mg/m3   for   4   hours/day  for   4  months.
Further  details  were  not  provided.   In another study (published In  Russian)
with  male white rats,  hlstopathologlc  lesions   Including  hyperplasla and
metaplasia of  the  bronchial  epithelium,  but  no  signs  of  malignancy, were
reported  following  Inhalation  of  acenaphthylene  vapors 4t   18  mg/m3,  4
hours/day, 6 days/week for  5 months  (Reshetyuk  et al.,  1970).

    Anthracene.   The  carcinogenic  potential of anthracene  has  been  tested
by  skin  application with  and  without  UV  radiation 1n mice,  1n skin Initia-
tion-promotion assays with  mice,  by s.c.  or 1.p.  routes 1n  rats,  by Implan-
tation Into the  lungs of rats, and  by  Implantation  Into  the  brain  or eyes  1.n
rabbits.  These  studies are  summarized  1n Table  V-10.   The results of the
      /
skin painting bloassays for both  complete  carclnogenldty and for  Initiating
activity  do  not  provide   evidence of  carclnogenldty,  but  contradictory
results  were  obtained  when anthracene was applied  to skin  together with
exposure  to  UV  radiation.  The   Injection  and Implantation studies  do not
provide  evidence of carclnogenldty, but these  results cannot be regarded  as
conclusive due to Inadequacies In experimental  design  (e.g., small number  of
animals, limited number of exposures.  Inadequate controls).

    Benzfalanthracene.   Shlmkln   and   Stoner   (1975)   demonstrated  negative
results  1n the  strain A mouse lung adenoma assay after  a single 1.v.  Injec-
tion of  0.25 mg benz[a]anthracene  and  an observation period of 6 months.

04420                                V-65                             11/12/91

-------
o
Jfc
4k

O
                                                                            TABLE V-10

                                              Dermal, Injection and Implantation Carctnogenlclty Assays of Anthracene
Route
Skin
Species/
mouse/NR
No.VSex
100/NR
Purity
MR
Treatment Duration
dose and number of appUca- NR
tlons not specified; 40X solu-
tion In lanolin
Effects/Comments
No skin tumors; 45 and 6 mice sur-
vived >6 months and >160 days.
respectively
Reference
Kennaway,
1924a,b
Skin




Skin


Skin




Skin



Skin




Skin



Skin
                 mouse/NR      41/NR
                 mouse/NR      SO/NR
                 mouse/Swiss   5/F
                 mouse/NR
                 mouse/NR
                 mouse/NR
                 mouse/NR
                               44/NR
                               44/MR
                               100/NR
                               NR/NR
NR





NR


NR





NR




NR





NR



NR
o
CO
)2 Skin
                 mouse/Skh:    24/m1xed
                 halrless-1
NR
                                                        dose and  number  of applied-        life  (133
                                                        tlons not specified; unsped-      days  average)
                                                        fled solutions  In water,
                                                        benzene or sesame oil

                                                        dose not  specified; 0.3X solu-     732 days
                                                        tlon In benzene  twice weekly

                                                        dose not  specified; 10X solu-      life
                                                        tlon In acetone  3 times/week       (10-20 months)
                                                        on the back
dose not specified; 5X solu-      life
tlon In petroleum jelly-olive     (11  months)
oil 3 times/week on the ear

treatment as above but with       life
ultraviolet radiation (>320 nm)   (10  months)
for 40 or 60 minutes, 2 hours
after skin application

treatment as above but mice       life
received ultraviolet radiation    (10  months)
for 90 minutes

dose and number of appUca-       NR
tlons not specified; 10X solu-
tion 1n petroleum Jelly-olive
oil followed by unspecified
exposure to ultraviolet
(320-405 nm) alone or with
visible light
0 or 4 iig 1n methanol  once        38 weeks
dally, followed by ultraviolet
(>290 nm) for 2 hours, 5 days/
week
                                                    No skin tumors;  d1benz[a.h]anthra-
                                                    certe was tumorlgenlc In same study
No skin tumors; 34 and 16 alive at
6 months and 1 year, respectively

No skin tumors; benzo[a]pyrene
Induced high Incidences of skin
paplllomas and carcinomas under
the same conditions

No skin tumors; l/44*a11ve after
11 months; primary report not
available

No skin tumors; 5/44 alive after
7 months
No skin tumors; 7/100 alive after
7 months
"High Incidence* of skin tumors.
Including many carcinomas, was
observed after 5-8 weeks; no tumors
In controls treated with anthracene,
ultraviolet or ultraviolet with
visible light; primary report not
available; unusually short latency
and Inadequate hlstopathology
reporting noted by IARC (1983)

Incidence of skin tumors not sig-
nificantly Increased In treated
group; survival was 20/24 and 19/24
In controls and treated groups
                                      Pollla. 1939
                                                                                          Bachmann
                                                                                          et al..  1937

                                                                                          Wynder and
                                                                                          Hoffmann.
                                                                                          1959a
Hlescher. 1942



Hlescher. 1942




Hlescher, 1942



Heller. 1950
Forbes et al.,
1976

-------
                                                                         TABLE V-10 (cont.)
     Route
Species/     No.VSex
                                         Purity
          Treatment
                                                  Duration
                                                                                                   Effects/Comments
                                                                                            Reference
    Skin
mouse/S       20/NR
                                        NR
    Skin
              mouse/CO-1    30/F
   Skin
              mouse/CRI:     20/F
              CD/1(ICR)BR
chromato-
graphic-
ally
purified
NR
two applications (0.3 ml of
0.5X solution 1n acetone per
application) with a 30-mlnute
Interval 3 times/week for a
total of 20 applications (30
rag/animal total); IB weekly
applications of croton oil In
acetone (0.3 mt) consisting of
1/0.17X, 2/O.OB5X and 15/0.17X
croton oil, beginning 25 days
after the first anthracene
application; controls received
the same treatment with croton
oil only.
single application of 10
(1782 ng) In benzene followed
1 week later by 5 umol TPA
twice weekly for 34 weeks;
controls received the same
treatment with TPA only

ten applications (0.1 mg In
0.1 ml acetone per application)
applied every other day.  Ten
days later 0.25 tig TPA In 0.1
ml acetone was applied 3 times/
week for 20 weeks
                                  25 weeks
                                                                         35 weeks
                                                                         25 weeks
                                                                                          Incidence of skin  tumors  not  sig-
                                                                                          nificantly Increased  In treated
                                                                                          groups; survival was  19/20 and
                                                                                          17/20 1n the control  and  treated
                                                                                          groups, respectively
                                                                                                       Salaman and
                                                                                                       Roe.  1956
                                                    Incidence  of  skin  paplllomas  was
                                                    4/2B (14X)  In treated  and 1/30 (3X)
                                                    In controls
                                                    No significant  Increase 1n  skin
                                                    tumor  Incidence In  treated  groups
                                                                                                                               ScMbner. 1973
                                                                                                                               LaVole et al..
                                                                                                                               1985
s.c.
o
CD
s.c.
  vS.C.
                 rat/NR        10/NR       NR           weekly injections of 2 ml of      >18 months
                                                        0.05X suspension In water for
                                                       . life (103 mg/an1mal maximum
                                                        total dose)


                 rat/Wlstar    5/NR        NR           5 mg In sesame oil, once          10 months
                                                        weekly for 6 or 7 weeks
              rat/BOI       10/NR       highly       20 mg In unspecified  oil  once      life
              and BDIII                 purified     weekly for  33 weeks
                                                    No subcutaneous  sarcomas;  survival     Boyland and
                                                    was 7/10  after  12 months and 8/10     Burrows,  1935
                                                    after  18  months;  d1benz[a,h]-
                                                    anthracene  was  tumoHgenlc under
                                                    the same  conditions

                                                    No tumors;  examinations apparently     Pollla,  1941
                                                    Included  viscera; d1benz[a,h]-
                                                    anthracene  Induced subcutaneous
                                                    tumors  In 2/5 similarly treated rats

                                                    Injection site  tumors  (Hbromas        Schmahl,  1955
                                                    with sarcomatous  areas) In 5/9
                                                    rats, mean  latency,  26 months; no
                                                    controls, but rats treated'simi-
                                                    larly with  naphthalene In  oil did
                                                    not develop local tumors

-------
                                                                         TABLE V-10 (cont.)
«•
ro
o



<
Ch
CO
Route Species/ No.VSex Purity
1.p. rat/801 or 10/NR
BOIII
Cerebral rabbi t/NR 5/NR
cortex
Implant
Cerebellar rabbi t/NR 2/NR
Implant
Optic rabbi t/NR 2/NR
Implant
Lung Implant rat/Osborne- 60/F
Mendel
'Numbers In treatment and control
highly
purified
NR
NR
NR
refined
recrystal-
11 zed
(If used) group(s)
Treatment Duration
Effects/Comments
20 mg In unspecified oil once life Tumor In one rat (spindle-cell
weekly for 33 weeks (mean -2 years) sarcoma In abdominal cavity); no
control group
10 mg (1 rabbit) or 20 mg 4.5 years
(4 rabbits) pellets
12 mg pellets 4.5 years
4 or 5 mg pellets 4.S years
0.5 mg In 0.05 ml warm soft 55 weeks
1:1 beesway: trlcaprylln
unless specified otherwise.
No tumors; survival was 4/5 after
4 years and 2/5 at 4.5 years
No tumors; survival was 2/2 at
4.3 years
No tumors; survival was 2/2 at
4.5 years
No tumors In treated or control
animals

Reference
Schmahl, 1955
Russell, 1947
Russell. 1947
Russell. 1947
Stanton
et al.. 1972

    NR  = Not  reported;  s.c. =  subcutaneous;  1.p.  *>  Intraperltoneal
o
00

-------
    As part of a  large  study  of  PAHs  and  their  nitrated derivatives,  newborn
CD-I mice  were  treated 1.p.  with  a  total of 2800 ymol  benz[a]anthracene  In
OMSO.   Liver  tumors  were  not  observed  1n  treated  female  mice,  but  31/39
males  developed  Hver  tumors  of which  25/39 were  carcinomas.   This was  a
significant Increase  compared with  controls.  By contrast,  female mice  were
                                   •
observed to  have a  significantly  Increased  Incidence  of  lung tumors  (6/32
compared  with 0/31)  while  the males  did  not  (6/39   compared  with  1/28)
(H1slock1 et al., 1986).

    Benz[a]anthracene  1s'  a well-documented  complete skin carcinogen.   Both
the  parent compound  and  the 3,4-d1hydrod1ol  and   the  3,4-d1o1-l,2-epox1de
have produced tumors  on mouse skin  (IARC,  1973; Santodonato et al.,  1981).
By contrast neither  Oonryu rats  nor  Syrian  golden hamsters  developed  tumors
after  topical  benz[a]anthracene application (Tawflc,   1965;  Shublk  et  al.,
1960).

    Subcutaneous  administration  of  benz[a]anthracene 1n tr1capryl1h  to  mice
resulted In  Injection site sarcomas.   Incidences of sarcomas 1n C57B1  mice
observed 9 months  after being given  graded doses of benz[a]anthracene  were
as follows:  0.05 mg, 5/43; 0.2  mg,  11/43;  1.0  mg,  15/31;  5.0 mg,  49/145;  10
mg, 5/16 (Stelner and Falk, 1951; Stelner and Edgecomb, 1952).   Klein (1952)
showed  that  Intramuscular  Injection of  albino mice derived from  strain A
with  benz[a]anthracene  In  a  1  or  3% combination with croton oil  produced
Injection site Hbrosarcomas and hemangloendothellomas.

    Benzofalpyrene.   Benzo[a]pyrene  Is  known to  produce tumors when  admin-
istered  by  Inhalation or  Intratracheal  Instillation.   It 1s generally  more


04420                                V-69                             10/08/91

-------
effective as  a  lung carcinogen when accompanied by a  resplrable  participate
or a  cocarclnogenlc gas.  Laskln et al.  (1970)  exposed 21 rats  to  10  mg/m3
benzo[a]pyrene  for  1 hour/day  for  1 year.   Two animals  developed  squamous
cell  carcinomas  1n  this  group.   When  rats  were exposed  to  10 ppm  S0«  for
an  additional  6  hours/day,  carcinoma  Incidence  was  5/21.    No  animals
receiving   SO.   only  developed   tumors.    Intratracheal  Instillation   of
benzo[a]pyrene  accompanied  by   India   Ink   resulted   1n   a   dose-dependent
Increase  1n  lung  tumors  1n  rats  (Yanlsheva,  1971).  Rats  that  received
14C-benzo(a)pyrene   along   with    carbon    black   or   asbestos   through
Intratracheal Instillation had  a  higher  incidence of  lung  tumors  than  those
receiving only this PAH (Pylev et al.,  1969).

    Intratracheal  Instillation  studies  of  partlculate  and  benzo[a]pyrene In
hamsters  have also  shown Increased  Incidences  of  respiratory  tumors.   An
Increased  Incidence of   respiratory  tract  tumors was reported  1n  Syrian
golden hamsters  that had been administered  benzo[a]pyrene  along  with ferric
oxide particles  (Safflottl  et  al.,  1965,  1968).   In  a similar  experiment,
FarreVl  and Davis  (1974)  showed that carbon  black  and  ferric  oxide were more
effective  tumor  promoters  than aluminum  oxide  when  these  three types  of
particles were bound to  benzo[a]pyrene.  The percentage of respiratory  tract
tumor-bearing hamsters  1n  each group  was  49,  49 and 22%  for   the  carbon
black, ferric oxide and aluminum oxide  groups,  respectively.   Intratracheal
Instillation  of radlolabeled  benzo[a]pyrene-coated carbon,  aluminum  oxide
and  ferric  oxide  resulted  In an. Increase  1n  the  retention time  of  the
radlolabel 1n the hamster lung when  compared with  administration  of only the
radlolabeled benzo[a]pyrene (Henry and  Kaufman, 1973).
04420                                V-70                            10/08/91

-------
    Hamsters  have  provided a  useful  model  for  the  study of  benzo[a]pyrene
lung  cardnogenlcHy.   Concurrent   exposure  with  sesame  oil  resulted  1n
trachea!  paplllomas  and  carcinomas  (Mohr,  1971).  Saff1ott1  et al.  (1972)
conducted  several   assays  wherein   a  saline  suspension  of  benzo[a]pyrene
coated  Fe-O-  particles  were   repeatedly  Instilled  In  hamster  treacheas.
In one  assay  the  lowest dose employed, 0.25 mg  benzo[a]pyrene/week,  Induced
respiratory  tumors  1n  10/88 hamsters.   At  the  high-dose,   3 mg/week,  the
Incidence was 34/57 with  multiple tumors observed  In  some animals  (Saff1ott1
et al.,  1972).  Using the same model, Sellakumar and Shublk  (1972)  reported
a 3054  Incidence  of respiratory tract  tumors  1n animals  receiving 20  weekly
treatments of 0.5 mg benzo[a]pyrene.

    In  research  by  Thyssen et  al.   (1981),  groups  of  24 hamsters  Inhaled
benzo[a]pyrene at  concentrations  of  2.2, 9.5  or  46.5 mg/ma,  4.5  hours/day,
7 days/week  for  10 weeks  followed by an exposure for  3 hours/day, 7  days/
week for a maximum of  675 days.   Neither control  animals nor  animals  receiv-
ing the  lowest  dose developed  respiratory or  upper  digestive  tract  tumors.
      /•
Above  2.2 mg/m3  benzo[a]pyrene,  the  Incidence  of  respiratory  and  upper
digestive  system  tumors  Increased  with  dose.   Table   V-ll  summarizes  the
findings.

    A  later   report   Indicated  that   concurrent  exposure of  Syrian  golden
hamsters  to  SO. and   benzo[a]pyrene-coated  sodium chloride  resulted  1n  an
enhancement of  tumor  response  and  a  decreased  latency  period (Pauluhn  et
al., 1985).

    Feron  and Kruysse (1978)   and   Ketkar  et  al.  (1978) also studied  the
Incidence  of  respiratory  tract   neoplasms   1n  male  and  female  hamsters

04420                                V-71                             10/08/91

-------
                                  TABLE V-ll

                     CardnogenlcHy of Benzo[a]pyrene to
                 Male Syrian Golden Hamsters  By Inhalat1ona»b
Dose
(mg/m3)
0.0
2.2
9.5
46.5
mg/Hamster
0
29
127
383
Duration of
Treatment0
(weeks)
96.4
95.2
96.4
59.5
Target Organ
respiratory tract
upper digestive tract
respiratory tract
upper digestive tract
respiratory tract
upper digestive tract
respiratory tract
upper digestive tract
Tumord.
Incidence
(X)
0/27 (0)
0/27 (0)
0/27 (0)
0/27 (0)
9/26e (35)
7/26e (27)
13/25f (52)
14/25^ (56)
aSource: Thyssen et a!., 1981

^Exposure was  for  4.5 hours/day for  the  first  10 weeks, 3 hours/day  there-
 after  for  7 days/week as  NaCl  vapor (>99X of  the particles had diameters
 between 0.2 and 0.5 yM) 1n air.
cThe duration of the study 1s equal  to the duration  of  the  treatment

^Tumors were paplllomas, papillary polyps, and squamous cell  carcinomas

e3  nasal  cavity, 8  laryngeal,  1  tracheal,  6 pharyngeal  and 1  forestomach
 tumor

fl  nasal  cavity,  13 laryngeal,  3 tracheal,  14 pharyngeal, 2 esophageal and
 1 forestomach tumor
04420
V-72
10/08/91

-------
associated  with Intratracheal  administration of  benzo[a]pyrene.  Ketkar  et
al. (1978)  found the mean  survival  time  of  both  male and female hamsters was
positively  related  to  the  dose of  benzo[a]pyrene  administered  as  weekly
Instillations In bovine  serum albumin (Table V-12).  For male  hamsters,  the
Incidence of respiratory tract  tumors tended  to  Increase with  dose except at
the highest  level  where  the  tumor Incidence  rate declined slightly.   For the
female  hamsters,  tumor  Incidence was also  Increased by  benzo[a]pyrene,  but
an  Inverse  relationship  held between dosage  and tumor  Incidence.   Why  this
Inverse relationship was observed Is not  readily  apparent.

    Kobayashl (1975)  treated 32 male  and 28  female Syrian golden  hamsters
with Intratracheal  Instillations of 1  mg of benzo[a]pyrene  1n  saline  weekly
for  30  weeks.    The  control   group  (20   mice/sex)   was   administered  an
equivalent  dose of  saline.   After  60   weeks  the mice  were sacrificed  and
complete necropsies  performed.   Survival rates appeared  to  be  equivalent 1n
the benzo[a]pyrene-treated  and  control  groups  (both  sexes combined).   The
Incidences of respiratory tract  tumors -were  11/26  1n treated males and 14/26
1n  treated  females.   The  majority  of these  tumors  were  found  1n  the
peripheral  areas  of the  lung.  No Incidences  of  respiratory  tract  tumors
were found 1n the control mice.

    Stenback and Rowland  (1978)  studied  the role of  talc and  benzo[a]pyrene
In respiratory  tumor formation.  Two groups  of 24  Syrian golden hamsters/sex
received either 18 Intratracheal  Instillations  of 3 mg  of  talc  1n saline or
3 mg benzo[a]pyrene an'd 3 mg talc In  saline  weekly.   Control groups  received
saline  or  remained untreated.   Hamsters  were  autopsled  upon  spontaneous
death.  Animals treated  with talc  and benzo[a]pyrene had a  shorter  Hfespan


04420                          .      V-73                            10/08/91

-------
                                  TABLE  V-12
                CarclnogenlcHy of Benzo[a]pyrene Administered
               by  Intratracheal  Instillation to Syrian Hamsters3
Duration of
Doseb Treatment0
(mg/week) (weeks)
0 41
0.10 40
0.33 24
1.0 10
Incidence of Respiratory Tract Neoplasms^
(X)
Hales
0/29
5/26
7/29
6/27
(0)
(19)
(24)
(22)
Females
•0/30
12/30
10/28
6/20
(0)
(40)
(36)
(30)
aSource:  Ketkar et al.f 1978
^Benzo[a]pyrene   (97X   pure)  was   delivered
 Controls received vehicle only.
            1n   bovine  serum   albumin.
cHean survival time.   Survival  time was  also equal  to  study  duration.
^Carcinomas, adenomas, adenocarclnomas and  paplllomas  were reported.
04420
V-74
10/08/91

-------
(52  weeks)  than  the  untreated controls  (55 weeks), and  these deaths were
primarily attributed to pulmonary dysfunctions caused by either  Interstitial
flbrosls  or  neoplastlc  Involvement  of  the  respiratory  system.   Only  the
hamsters  treated   with   the  talc   and  benzo[a]pyrene   mixture   displayed
paplllomas,   squamous  cell  carcinomas,  and  undlfferentlated  tumors  of  the
lung,  larynx  and  trachea  (Incidence  33/48).    Talc  appeared  to  have   no
carcinogenic activity alone,  but  Increased  the cardnogenldty  of  benzo[a]-
pyrene.

    In an Intratracheal Instillation  study,  male  Syrian  golden  hamsters were
administered benzo[a]pyrene,  arsenic (as  arsenic trloxlde)'  or  arsenic  and
benzo[a]pyrene together.  They were  Introduced with  3 mg  of  charcoal  carbon
dust  carrier.    The control  group   received only  the  carrier  dust.   The
animals were dosed  weekly.   The  benzo[a]pyrene  group (50  hamsters)  received
0.44 mg/exposure  (~6  mg/kg),  and the arsenic group  (67  hamsters)  received
0.25  mg/exposure  (~3  mg/kg).   The  group  receiving   both  chemicals   (90
hamsters)  received  0.45  mg  and 0.23 mg   of  benzo[a]pyrene  and  arsenic*,
respectively.  The  50-week  survival  rates   of the  control,  benzo[a]pyrene,
      /
arsenic  and  combined groups were  40,  54,  52  and  41X,   respectively.   The
100-week survival rates of the control, benzo[a]pyrene,  arsenic, and  arsenic
and  benzo[a]pyrene  combined  groups   were  13, 14, 25  and  13X,  respectively.
The  Incidences of carcinomas  of  the larynx, trachea, bronchll or  lungs were
0/53,  3/47,  17/40  and  25/54  In  the control, arsenic,  benzo[a]pyrene,  and
arsenic and  benzo[a]pyrene  combined groups, respectively  (Pershagen  et al.,
1984).

    Male  Sprague-Oawley  rats were  exposed to  either   [sH]-benzo[a]pyrene
mlcrocrystals,    [8H]-benzo[a]pyrene   mlcrocrystals    In   Iron   oxide    or

04420                                V-75                             10/24/91

-------
chrysotlle  asbestos,  or  [3H]-benzo[a]pyrene  absorbed   to   Iron  oxide  or
chrysotlle  asbestos  participates by  Intratracheal  Instillation.   Rats were
sacrificed at  times  ranging  from 2 minutes to 72 hours  after  administration
and   the   distribution  of   the  radlonucUde   examined.    Absorption  of
benzo[a]pyrene  administered  as   mlcrocrystals   was  rapid.   Addition  of
asbestos decreased the absorption rate of  the mlcrocrystals, and addition of
Iron  oxide had. little  or  no effect .on absorption.  Partlculates  did not
appear  to  Increase   the  cardnogenldty  of   benzo[a]pyrene  by   particle
enhanced  uptake or  Increased retention  time 1n the  lungs  (Bevan  et al.,
1988).

    Two studies  by Feron and co-workers  showed benzo[a]pyrene Instilled In
saline  suspension  produced  respiratory  tumors  In  hamsters.   Feron et al.
(1973) reported that male  Syrian  golden  hamsters receiving a  total  adminis-
tered dose of 0, 3.25, 6.5,  13, 26 or  52 mg benzo[a]pyrene had the following
Incidences of  respiratory  tract  tumors:   0/29,  3/30,  4/30,  9/30,  25/29 and
26/28,  respectively.    Feron and  Kruysse  (1978)  found  a   positive   dose-
response relationship  for  tumor  Incidence  1n  male and female  hamsters  given
      <
a  total  dose  of  0,  18.2 or 36.4  mg of  benzo[a]pyrene In  saline  1ntra-
tracheally  for  52  weeks.   The  Incidence  of  respiratory   tract  tumors
corresponding  to   these  doses was  0,  13.8  and 63.3X  In  males,  while  1n
females  the   corresponding   tumor   Incidences   were  0,  11.1  and  29.2X.
PapHlomas and  carcinomas  of the  trachea  and pulmonary  adenomas  were most
often observed.

    Benzo[a]pyrene, 1  mg/mi  1n phosphate  buffer solution,  was administered
Intratracheally once a week  for  15 weeks to 8-week-old  female  Syrian  golden

04420                                V-76                            10/24/91

-------
hamsters.  Treated  animals  and controls survived 443  and  614 days,  respec-
tively, after the Initial Instillation.  No control animals were observed  to
develop  lung  or trachea! tumors  whereas  6/13  treated  animals had  trachea!
tumors (2 malignant, 4 benign) (Yamamoto et  al., 1985).

    Benzo[a]pyrene  has  been shown  to  produce,  tumors at  various  sites by  a
number  of modes  of  administration.   Intraperltoneal   Injection  of  newborn
BLU:Ha(ICR) mice with  benzo[a]pyrene produced  lung  adenomas.   A dose  of  280
vg/mouse  resulted  1n  a 94X tumor  Incidence  (Busby  et al., 1984).   Individ-
ual Swiss mouse  fetuses  were  treated  with a single 1.p. Injection  of benzo-
[a]pyrene  or  derivatives  In  1  yi  trloctanoln and  acetone.  Doses  of  0,
0.4,  4.0, 9.9  or  19.8 vmol/fetus  were given on  day 15  of Intrauterlne
growth and animals  were carried to term, weaned  and  kept  until 12 weeks  of
age.  At  this  time they were  killed and  examined for lung adenomas.   Inci-
dence of adenomas 1n  the control group  was 4/37 and  In  the treated  groups  at
the stated dosages  were  1/39,  10/42,  10/38 and 12/31.  A  racemlc mixture  of
benzo[a]pyrene  7,8-d1ol-9,lO-epox1des   was   more  tumorlgenlc  producing   an
Incidence of  27/37 at a dose of 4.0 ^mole/fetus  (Rossi  et al., 1983).

    Male Wlstar  rats  were given  an 80 mg/kg bw IntrapeMtoneal Injection  of
benzo[a]pyrene 18 hours  after  partial  hepatectomy (Dock et al., 1988).   The
rats were then  placed on a  specific diet schedule that consisted of  2 weeks
of  a  basal  diet,   followed  by 2  weeks of  a   diet  supplemented  with  0.02%
2-acetylam1nofluorene  (2-AAF).   A  single  Injection  of carbon  tetrachloMde
(2  ml/kg bw) was  administered midway  through  the  2 weeks.   The   rats were
then administered  a basal  diet  for 2  weeks.   The  benzo[a]pyrene  and  2-AAF
04420                                V-77                            10/08/91

-------
produced  a  significant  number of  enzyme-altered fod  1n the  regenerating
liver.  There was a  50%  reduction  1n  the  metabolism  of benzo[a]pyrene In the
proliferating liver.

    Benzo[a]pyrene  1s  carcinogenic  when  administered s.c.  to  mice,  rats,
hamsters, guinea  pigs and  some primates  (IARC, 1983).   Bryan  and  Shlmkln
(1943)  attempted  to establish  dose-response  curves  for Induction  of tumors
following s.c.  administration  1n mice.   Tr1capryl1n solutions (0.25  or  0.5
ml) were  Injected  once  only  Into  the right  axilla  of male C3H  mice,  and
animals were observed throughout their  lifetime.  No controls were  reported,
but  several  no  effect   doses  were observed  (Table V-13).   In  addition  to
Injection  site  sarcomas,  newborn  mice  administered  benzo[a]pyrene  s.c.
developed hepatomas  or  lung adenomas  (Pletra  et al., 1961;  Roe and  Waters,
1967. Toth and Shublk,  1967; Grant  et al., 1968).

    Peralno et  al.  (1984) treated newborn  (day  1)  Sprague-Dawley  rats  by  a
single  1.p.  Injection of  0.59 vmole  benzo[a]pyrene/kg bw.   At  day  21  the
animals were weaned  and  placed on  a SOX  casein  diet containing  0.05X pheno-
barbltal, a  known  promoter  of hepatic  neoplasms.   Animals  were   killed  at
Intervals up  to ~500 days,  and their  livers  were  examined  for  hlstocheml-
cally detectable foci of altered  hepatocytes  as well  as for  hepatic  tumors.
Hepatic  tumor   Incidence 1n  females  was -57%  and  ~37X   In males.   Focus
Incidence had reached 100% 1n both  males and females  by day 200.

    Benzo[a]pyrene Is among the most potent and  best documented  of  skin car-
cinogens.  It 1s routinely  used as a positive control  1n  skin painting bio-
assays of other materials.  Skin tumors have been  produced by benzo[a]pyrene

04420                                V-78                            10/08/91

-------
                                  TABLE  V-13

           Sarcomagenlc Activity of Benzo[a]pyrene 1n Male  C3H Mice
                  Following a Single Subcutaneous  Injection*
Dose (mg)
0.00195
0.0078
0.0156
0.031
0.062
0.125
0.25
0.5
1.0
2.0
4.0
8.0
Tumor Incidence
Number Tumor -Bear Ing
Animals/Number
Effective Animals
2/81
0/40
0/19
0/16
4/20
15/19
14/21
19/19
18/20
19/19
16/19
20/21

Percentage
2
0
0
0
20
79
67
100
90
100
84
95
*Source: Bryan and Sh1mk1n, 1943
04420
V-79
10/08/91

-------
1n mice,  rats,  rabbits and to a  lesser  extent  1n guinea pigs.  In mice  the
tumoMgenlc dose  1s  dependent  upon the solvent used  for  delivery  and  on  the
strain  of mice  (IARC,  1973).   Benzo[a]pyrene Is  both  an  Initiator  and  a
complete  carcinogen.   For  example,  Nesnow et al.  (1986)  studied  the  effects
of  dermal treatment  with benzo[a]pyrene  on 7-  to  9-week-old SENCAR  mice.
The benzo[a]pyrene  was applied as  a single topical  treatment 1n 0.2 ml  of
acetone and as five  dally  doses of  2 mg  each.   One  week after treatment,  2.0
vg of  the tumor promoter  TPA  was administered topically twice weekly.   The
results are presented 1n Table V-14.

    Perelra et al.  (1979)  applied  [3H]benzo[a]pyrene  to shaved female  HA/ICR
mice;    the   mice   were   sacrificed  7   hours    later.    [8H]Benzo[a]pyrene
metabolite conjugated  epidermal  DMA, Isolated  from these treated mice,  was
found   to contain   two  major  benzo[a]pyrene-DNA   adducts.   The   maximum
concentrations  of  both  aducts   occurred  7  hours  after  a  single  dermal
application.   Benzo[a]pyrene-DNA  adduct  formation occurred  1n  proportion  to
dose at doses  several  orders  of magnitude below  doses  that  normally  yield a
carcinogenic  response.
      f                                                    r

    In an analysis  of  two  separate  sk1n-pa1nt1ng  experiments using benzo[a]-
pyrene  1n acetone,  Lee and O'Neill (1971) showed that  the  Incidence  rate  of
both tumors and  Infiltrating  carcinomas could  be described by the  equation
 2       k
d   (t-w)   where  t  Is   time  from  first  application,  w   and  k  are  dose
Independent constants and d Is the  applied dose.  In  the first experiment  75
female albino mice/group (16  total groups)  received  either 6, 12, 24  or  48
yg  of   benzo[a]pyrene/week  1n either  2  applications/week,  4  applications/
week and  2 groups that were administered  3 applications/week.   One group was
04420                                V-80                            10/24/91

-------
ro
o
i
03
                                                        TABLE V-14
                           SENCAR Mouse  Skin  Tumor1genes1s,  Benzo[a]pyrene  -  Tumor  Initiation3
Dose
(Mg/mouse)
0
0
2.52
2.52
12.6
12.6
50.5
50.5
101
101
Sex
M
F
M
F
M
F
H
F
M
F
No. Mice
Surviving
37
39
40
39
40
37
39
40
38
38
Mice with
Pap1l1omasb
(X)
8
5
45
31
73
57
100
75
95
97
Paplllomas/
Mouse1*
0.08
0.05
0.50
0.44
1.8
1.1
5.8
2.8
10.2
7.9
Mice with
Carcinomas
(X)
5
0
5
5
20
23
25
20
30
25
Carcinomas
per Mouse0
0.05
0
0.07
0.05
0.20
0.23
0.25
0.20
0.33
0.25
o
CO
aSource: Nesnow et al., 1986
bScore at 6 months
cCumulat1ve score after 1 year

-------
dosed on  MWF  and the  second  on  TWF.   In the second experiment,  four  groups
of  40 albino  female  mice  were  treated  with either  1,  3,  9 or  27  vg  of
benzo[a]pyrene every fourth day.

    Several electrophlUc  metabolites  of benzo[a]pyrene  are skin  carcino-
gens.  These  Include  dihydrodlols, dlol-epoxldes  and  some  phenolic  deriva-
tives (Gelboln, 1980).

    Benzo[a]pyrene  has also been  reported  to be carcinogenic when  adminis-
tered by  the  following routes:  1.v.;  transplacentally;  Implantation  1n  the
                                                            »
stomach wall,  renal  parenchyma,  and brain;  Injection Into the renal  pelvis;
and vaginal painting (IARC, 1983).

    Groups of  partially hepatectomlzed male  Sprague-Oawley  rats were given
either  a   single  30  mg benzo[a]pyrene  1.p.  Injection  24  hours  after  the
partial  hepatectomy  (10  rats),   repeated  Intragastrlc  Injections   of 4  mg
benzo[a]pyrene for 6 days  (15  rats) or  a  single  olive  oil Injection 24 hours
after .the  partial hepatectomy (10 rats  In  the  control group)  (KUagawa  et
a!.,  1980).    All  groups  were  fed  a  diet  containing  0.05X phenobarbltal
beginning  2   weeks   after  the  partial  hepatectomy.   The  animals were
sacrificed at  52 weeks  and  were  examined  for  hepatic tumors.   In both  groups
receiving benzo[a]pyrene, many rats died  of abdominal wall sarcomas  prior  to
52  weeks.   In* fact  0/10  rats  Injected  with  benzo[a]pyrene  survived  until
week 52.  Of the  rats  that  received Intragastrlc  Injections,  6/7 had hepatic
tumors and  0/9  In  the control  group  had hepatic  tumors.   It was  reported
that  there  were  many  enzyme-altered  foci  In the  livers of  rats  receiving
Intragastrlc benzo[a]pyrene Injections.


04420                                V-82                            10/08/91

-------
active as  the parent compound, whereas  the  9,10-dlol  of  benzo[J]fluoranthene
was less active  than  Its  parent  compound (see Table  V-15).  The  8,9-d1ol  of
benzo[k]fluoranthene was  not tumorlgenlc.   The  9,10-dlol of  benzo[b]fluor-
anthene  1s  a potential precursor  to a  bay region dlhydrodlolepoxlde.   The
benzo[j]fluoranthene 9,10-dlol could be  metabolized to d1ol-epox1de  with  the
epoxlde  ring In a  four-sided "pseudo  bay   region."   Taken  together,  these
results  suggest  that  the  formation  of  bay  region  dlhydrodlol epoxldes  may
not be the major activation mechanism In benzofluoranthene tumor 1 genesis.

    This  work   Is  corroborated ,by  the  reports  of  Amln  et  al.  (1985a,b).
Initiation/promotion  protocols  using   benzo[b]fluoranthene   and  benzo[k]-
fluoranthene were  carried out In  Crl:CD-1(ICR)BR  albino female  mice.   Each
Initiating  compound was  applied  every other  day In  10 doses  as  acetone
solutions  to the shaved  backs  of 20  mice/treatment  group.   Ten days  after
completion  of   the  Initiating  treatment promotion  was   begun using  thrice
weekly  applications  of  2.5  yg  of TPA   (dissolved  In  0.1   ml  acetone).
This was continued  for  30 weeks.  Results  summarized  1n  Table V-18  Indicate
that   both   benzo[b]fluoranthene  and  benzo[k]fluoranthene   can   serve   as
Initiators of cardnogenesls.  Benzo[b]fluoranthene,  however.  Is  more potent
producing  the  same  Incidence of  tumors with  1/100   the  dose necessary  to
Initiate tumors with benzo[k]fluoranthene.

    Injection  site  sarcomas  were observed  1n  18/24 survivors  of  a  total
group  of 16  male and  14 female  strain  XVIInc/Z  mice  given three  s.c. Injec-
tions  of  2.6 mg benzo[-b]fluoranthene over  a period  of  2  months  (Lacassagne
et al., 1963).
04420                                V-83                            05/14/91

-------
                                 TABLE V-15
      Cardnogenldty of Benzofluoranthenes by Implantation  1n  Rat  Lungs3
Treatment
Untreated
Pelletb
Benzofb]-
fluoranthene
Benzo[k]-
fluoranthene
Benzo[g,h,1.]-
perylene
Dose
(mg)
Total ,
—
—
0.1
0.3
1.0
0.16
0.83
4.15
0.16
0.83
4.15
Median
Survival
Time
(weeks)
118
104
110
113
112
114
95
98
109
114
106
Ep1dermo1d
Carcinomas
0/35
0/35
0/35
1/35
9/35
0/35
3/31
12/27
0/35
1/35
4/34
Sarcomas
0/35
0/35
1/35
2/35
4/35
0/35
0/31
0/27
0/35
0/35
0/34
Tumor
Incidence
0/35
0/35
1/35
3/35
13/35
0/35
3/31'
12/27
0/35
1/35
4/34
aSource: Deutsch-Wenzel  et  al.,  1983
bRefers to beeswax and  trloctanoln  Implanatatlon medium
04420
V-84
10/08/91

-------
                                                           TABLE  V-16
                                         Carc1nogen1c1ty Assay of PAH 1n Newborn M1cea
I
oo
en
O
OO
Tumor Inc1denceb
Compound


DMSO

Benzo[a]pyrene

Benzo[b]f luoranthene

Benzo[J ]f luoranthene

Benzo[k]f luoranthene

Indenofl ,2,3-cd]pyrene

Total
Dose
(iimol)



1.1

0.5

1.1

2.1

2.1

Sex


H
F
M
F
M
F
H
F
H
F
H
F
Hepatic Tumors

Adenomas
0/17
0/18
9/17
0/14
6/15
0/17
8/21
0/18
2/16
0/18
0/11
0/9

Hepatomas
1/17
0/18
4/17
0/14
2/15
0/17
3/21
0/18
1/16
0/18
0/11
0/9

Total
1/17
0/18
13/17C
0/14
8/1 5C
. 0/17
11/21C
0/18
3/16
0/18
0/11
0/9
Lung
Adenomas

0/17
0/18
14/17C
9/14
2/15
3/17
11/21C
4/1 8d
1/16
3/18
1/11
0/9
aSource: LaVole et al., 1987
^Denominator denotes animals surviving until at least 35 weeks.
cp<0.005
dp<0\05

-------
controls were  run (Wynder and  Hoffman,  1959b).   As part of  the  same study,
20  Swiss  mice were  treated  with benzo[k]fluoranthene.  No  tumors  developed
1n animals painted with  the  0.1% solution,  but  skin paplllomas were observed
1n  2/20  mice  treated  with 0.5X  benzo[k]fluoranthene.   By  contrast  to  this
study,  no significant   Increase  1n  tumor  Incidence  was  observed  when  40
female  NMRI   mice  were  given   skin  applications  of  3.4,   5.6  or  9.2  yg
benzo[k]fluoranthene  2   times/week  for  their   lifetime.    No   effect  on
mortality was noted as a consequence of this treatment (Habs  et al.,  1980).

    A single application of  11  mg benzo[k]fluoranthene did not Induce tumors
1n 20 Swiss mice  1n  a  63 week study.   When this Initiating Hose was  followed
by  promoting  treatments with croton resin,  18/20 animals  developed papll-
lomas and 5/20 carcinomas (Van DQQren et al.,  1966).

    The  tumor  Initiating activities  of  benzo[b]fluoranthene,  benzo[j]fluor-
anthene  and  benzo[k]f!uoranthene and  three  of their  dlhydrodlols  (9,lO-d1-
hydro-9,lO-d1hydroxybenzo[b]fluoranthene,   9,lO-d1hydro-9,lO-d1hydroxybenzo-
[Jjfluoranthene   and   8,9-d1hydro-8,9-d1hydroxybenzo[k]fluoranthene)   were
evaluated after application  to  the shaved backs of  Crl:CO-l  mice (LaVole et
al.,  1982a).   Each  compound was  applied  1n  acetone  solution  (0.1 ma,)  to
the  backs  of  20  animals/group.   Controls  received  acetone alone.   Three
Initiating dose  levels,  10,   30  and 100 yg for benzo[b]fluoranthene and 30,
100 and  1000 yg  for the  other  two compounds  (10 doses,  every  other  day),
were  used followed  by  2.5   yg  TPA  (3  times weekly  for 20 weeks)  (Table
V-17).   This   study  demonstrated  that   of  the compounds  tested,  benzo[b]-
fluoranthene  was  the most  potent tumor  Initiator  followed by benzo[J]fluor-
anthene.   Benzo[k]fluoranthene  also showed  tumor  Initiating  activity but was
not  a complete  carcinogen.   The  9,lO-d1ol of  benzo[b]f!uoranthene was  as

04420                                V-86                            10/08/91

-------
                                  TABLE V-17
       Tumor  Initiating Activity  of  Benzofluoranthenes  1n Crl:CD-l M1cea
Compound
Benzo[b]f luoranthene


B[b]F-9,lO-d1ol


Benzo[ j ]f luoranthene


B[J]F-9,lO-d1ol

-\
Benzo[k]f luoranthene


B[k]F-8,9-d1ol


Acetone
Total
Initiating .
Dose
(yg)
100
30
10
100
30
10
1000
100
30
1000
100
30
1000
100
30
1000
100
30
-
Percent
Skin Tumor -
Bearing
An1malsb
80
60
45
95
63
26
95
55
30
84
20
5
75
25
5
10
10
15
0
Skin
Tumors/
Animal
7.1
2.3
0.9
8.8
3.8
1.0
7.2
1.9
0.6
4.5
0.3
0.1
2.8
0.4
0.1
0.4
0.1
0.1
0
Other
Tumors
0
"|C
id
0
0
ld
lc
0
0
0
le
0
ld
0
0
0
0
0
0
aSource:  LaVole  et  al.,  1982a.
 times/week for 20 weeks.
Promoting  treatment  was  2.5  >ig TPA,  3
bSk1n tumors were predominantly squamous cell paplllomas
cMa!1gnant lymphoma
dEndometr1al carcinoma of uterus
eLung adenoma
04420
 V-87
10/08/91

-------
active as  the  parent  compound,  whereas  the 9,lO-d1ol  of benzo[J]f1uoranthene
was  less active  than  Us parent compound  (see  Table  V-17).   The 8,9-d1ol  of
benzo[k]fluoranthene  was not tumorlgenlc.   The 9,lO-d1ol of  benzo[b]fluor-
anthene  Is a  potential  precursor  to a  bay region dlhydrodlolepoxlde.   The
benzo[j]fluoranthene  9,lO-d1ol  could  be  metabolized to d1ol-epox1de with the
                                  •
epoxlde  ring  In  a  four-sided  "pseudo  bay  region."   Taken   together,  these
results  suggest  that  the  formation  of  bay  region dlhydrodlol  epoxldes  may
not be the major activation mechanism In benzofluoranthene tumor 1 genesis.

    In  another  skin   painting   assay benzo[b]fluoranthene> 1n  acetone  was
applied to the skins  of  20  female  CD-I mice/group  for a total dose of either
0,  1.0  or  4.0  ymol   (a  total  of 10  subdoses  were applied every  other  day)
(Weyand  et  al.,  1990).   Ten  days  after  the   final   dose,  2.5  yg  of
tetradecanoylphorbol  acetate was applied 3 times/week for 20  weeks.   In the
acetone control  group,  2/20 mice  developed  skin tumors;  the  average number
of  skin  tumors/mouse  was  0.1.   In  both  benzo[b]fluoranthene  groups  20/20
mice developed skin tumors; the average number  of skin tumors/mouse was 8.5
and 11.0 In the low- and high-dose groups,  respectively.

    This work  1s  corroborated  by the  reports of  Am1n  et   al.  (1985a,b).
Initiation/promotion  protocols  using   benzo[b]fluoranthene   and   benzo[k]-
fluoranthene were  carried out  1n  Crl:CD-l(ICR)BR  albino female mice.   Each
Initiating  compound   was  applied  every  other  day 1n 10 doses as  acetone
solutions  to  the shaved  backs  of 20 mice/treatment  group.    Ten  days  after
completion  of  the  Initiating  treatment  promotion  was  begun  using  thrice
weekly  applications  of  2.5  wg  of  TPA  (dissolved  In  0.1  ml  acetone).
This was continued for  30 weeks.  Results summarized  1n  Table V-18 Indicate
04420                                V-88                            11/12/91

-------
                                  TABLE  V-18
   Tumor Initiating Activity of Benzofluoranthenes  1n  Crl:CD-l(ICR)BR  M1cea
Compound1*
Acetone
Benzo[b]fluoranthene
Acetone
Benzo[b]fluoranthene
Acetone
Benzo[k]fluoranthene
Total
Initiating
Dose
(ymol)
—
40
100
—
40
100
—
4000
Percent
Animals
with Tumors
10
45
95
5
42
53
0
37
Skin Tumors/
Animal
0.2
0.9
3.3
0.1
0.5
0.9
0
. 0-7
aSource: Am1n et al., 1985a,b
bEach  treatment  Is  presented  with   Us   concurrent   control.    Promoting
 treatment  consisted of  thrice  weekly  applications  of  2.5  yg  TPA  for  30
 weeks.
04420                                V-89                            10/08/91

-------
that  both   benzo[b]fluoranthene   and  ,benzo[k]fluoranthene  can  serve  as
Initiators of cardnogenesls.  Benzo[b]fluoranthene, however.  Is more  potent
producing  the  same  Incidence  of  tumors  with  1/100 the  dose necessary to
Initiate tumors  with benzo[k]fluoranthene.

    Injection site  sarcomas were  observed  In 18/24  survivors of  a  total
group of  16  male  and  14 female strain XVIInc/Z mice given  three s.c.  Injec-
tions of  2.6 mg benzo[b]fluoranthene over  a  period of 2 months  (Lacassagne
et al.,  1963).

    Benzofg.h.l.lperylene.    Benzo[g,h,1]perylene   did   not  'produce   a  sig-
nificant  Increase  1n  tumor  Incidence  (see Table  V-15)  when  Implanted as
beeswax trloctanoln pellets  1n lungs  of  female  OH rats  (Deutsch-Wenzel et
al.,  1983).   Although a  few pulmonary  tumors were observed  1n rats  after
Intrapulmonary Injection, the data were considered  Inadequate  for evaluation
(IARC, 1983).

    No  Increased  Incidence  of  tumors  was  reported  from two  sk1n-pa1nt1ng
bloassays  of benzo[g,h,1]perylene  conducted  In female Swiss  or Ha/ICR/mll
Swiss mice  (Hoffman  and  Wynder,  1966; Wynder and  Hoffman,  1959b).   Three
Initiation promotion  assays for  skin  tumorlgenesls 1n  mice  were  likewise
negative.   Benzo[g,n,1]perylene  did  not  produce  Injection site  tumors 1n
either of two studies  wherein mice  were exposed subcutaneously  (IARC,  1983).

    Van  DQQren  et  al.  (1973)  reported  some evidence  that  benzo[g,h,1]-
perylene served as a  cocardnogen with benzo[a]pyrene when  both  were  applied
simultaneously to the  skin of ICR/Ha  Swiss mice.


04420                                V-90                             10/08/91

-------
    Chrvsene.   As  part of  a  study of  nitrated PAHs, chrysene  1n DMSO  was
administered 1.p. to CD-I mice on  days  1,  8  and 15 of age.   Tumor  Incidences
were  recorded  after 1  year.   A  total  dose  of  2800  nmol/mouse resulted  In
hepatic and  lung tumors In 41 and 21X  of  the males, respectively.  A  total
dose  of  700 nmol  Induced  liver  tumors In  29% of  the  males; females  were
unaffected (W1slock1 et al.. 1986).

    A  series  of early  skin  painting   assays   conducted  1n  mice  produced
variable  results.   It was  noted  that  1n  these  bloassays  the chrysene  was
likely  to  be  contaminated  with  methylchrysene  or  other  materials  not
detectable by the technology then available (IARC, 1983).

    As  part  of  a  study  of  tobacco constituents,  female  Swiss  mice  were
treated  by  brushing  a   IX acetone  solution  of  chrysene  on  the  skin  3
times/week.  No  solvent control was reported, but  45X of  the treated animals
developed  paplllomas  and  40% were observed  to  have  carcinomas (Wynder  and
Hoffman, 1959a).
      /
    "Specially purified" chrysene  was applied  topically to  20 male  C3H mice
as either a  decahydronaphthalene  solution  or as  a  50:50  mix  with  n-dodecane.
Applied alone,  chrysene Induced  a papllloma 1n  1/12  mice  at 76  weeks.   In
combination  with  n-dodecane 5/19 mice  were  observed  to have  paplllomas  and
12/19 bore carcinomas at 49 weeks (Norton and Christian,  1974).

    Chrysene was assayed  In  a mouse skin Initiation/promotion assay  using
ICR/Ha  Swiss mice.   Paplllomas  were Induced  In  5/20  animals treated  with
croton  resin only  and In  16/20  mice  receiving  croton  oil  preceded by  a
single application of  1 mg  chrysene  1n  acetone  (Van DQQren  et al., 1966).  A

04420                                V-91                            10/08/91

-------
second study using pure TPA  as  the  promoter  yielded  similar  results.   Female
CD-I  mice received  a single  topical application  of  1  mg  of  chrysene  In
acetone,  followed  by  25  yg TPA  (1n 0.1  ml acetone)  2  times/week.   At  30
weeks,  3%  of   the  promoter-only  mice  were observed  to  have  papHlomas,
whereas   73X   of  the  chrysene-treated  animals  developed   these   growths
(ScMbner, 1973).

    Further  evidence  of  chrysene's  potential   as   an  Initiator  of  skin
cardnogenesls  comes  from studies  by Hecht  et al.  (1974)  and Levin  et  al.
(1978).   In the  first  study, 20 female Swiss  mice (Ha/ICR/M1l) were given 10
dally  treatments  with 100 yg  chrysene  In acetone.   After  'a 10-day  resting
period,  2.5  yg  TPA  In acetone  was  applied  3 times/week for  20 weeks.   A
group  receiving  chrysene  only  was  observed  at 72 weeks to have a carcinoma
Incidence  of  4/11 as  compared with  an  Incidence of  11/18  (carcinomas  and
paplllomas) 1n the chrysene plus TPA group (Hecht  et  al.,  1974).

    Levin et al. (1978) used female CD-I mice, 30/group,  In  an assay  wherein
a  single  topical  application  of  a  0.4,  1.25   or  4.0  ymol  solution  of
chrysene  1n tetrahydrofuran:DMSO  (95:5)  was  followed by twice weekly  appli-
cations  of TPA (200  pi  of  a   16  ymol  solution  In  acetone).   A  second
group  receiving TPA  alone  had  a  papllloma  Incidence  of  7X (0.07  tumors/
mouse).   Tumor  Incidence  for the low, medium and high chrysene dose  groups
receiving TPA  were  the following:  25,  43 and 52X with 0.32, 0.97 and  1.45
tumors/mouse,  respectively.

    Three  Initiation  promotion  studies  1n  female CD-I   mice also  were
positive  for chrysene  Initiating  activity.  Wood  et  al.  (1979) reported  an
Incidence  of  21/30  for  observation  of  paplllomas  In  mice  given  a  single

04420                                V-92                           10/08/91

-------
application  of  200  vl  of  a  2  ymol  chrysene  solution  In  acetonerDMSO:
ammonium hydroxide  (1000:100:1),  followed by  twice  weekly applications TPA
(200  pi  of  16   ymol  1n  acetone).    Equivalent  doses  of  the   bay  region
3,4-epoxy-l,2,3,4-tetrahydrochrysene  and  Us  precursor  metabolite  l,2-d1-
hydrochrysene  produced  essentially  the  same  papllloma  Incidence.  Control
animals treated with  acetone  followed  by  TPA had a tumor  Incidence of  1/30.
These  data  support  the  bay-region  theory  of  cardnogenesls  that suggests
epoxldes on  saturated angular benzo  rings  which form  part  of a  bay-region
are particularly  susceptible  to  undergoing  ring opening to an electrophHlc
carbonlum Ion.

    A  second  experiment  by Wood  et  al.   (1980)  employed  chrysene of higher
purity  (98X).  Thirty mice were treated  with  200  yl  of an acetone solution
of  this  chrysene  preparation  and  subsequently with  TPA  (200  yi  of  16
ymol  solution)  2 times/week  for  25  weeks.   The acetone  plus TPA controls
were  observed  to  have a tumor  Incidence  of  4X (0.04 papHlomas/mouse); the
chrysene plus TPA group Incidence was SOX  (2.16 tumors/mouse).
      /
    In  a  third  study, Rice et al. (1985b)  Initiated  25 CD-I  mice by appli-
cation  of  100 yg  chrysene 1n  acetone every  other  day  for  a  total  of 10
times  (1.0 mg  total  dose).   This was followed by thrice weekly  applications
of  TPA beginning  10  days after  completion  of the  Initiation  phase and
continuing  for  20 weeks.  All  animals thus treated  survived,  and 92% were
observed to bear tumors.

    Sencar mice were  likewise sensitive  to chrysene  Initiation.   Paplllomas
were  observed  1n  21/29  mice  treated once with 2  ymol  chrysene followed by


04420                                V-93                            10/08/91

-------
twice  weekly  TPA  treatments  (2  yg/treatment).    TPA  only  animals  had  a
papllloma Incidence of 3/30 (Slaga et al.,  1980).

    Early  studies  with  chrysene  reported  no  Injection  site  tumors  as  a
consequence of  1.m.  or  s.c.  administration  (Bottomly and Twort,  1934;  Barry
and Cook,  1934;  Shear and Lelter,  1941).   Small  numbers of sarcomas  at  the
site of chrysene  Injection were  reported for  C57B1  male  and  female mice when
trlcaprylln or  arachls  oil was  used as the vehicle  (Stelner and  Falk,  1951;
Stelner, 1955; Boyland and S1ms,  1967).  PolUa (1941) reported no tumors to
be  Induced  In  rats receiving repeated  Injections  of chrysene  1n water  or
sesame  oil,  while  Barry  and  Cook (1934) reported  four  sarcomas   1n 10  rats
repeatedly  Injected with  2-6 mg chrysene  compared with  two  sarcomas  In
solvent control rats.

    Perinatal exposure of mice to chrysene  has resulted 1n tumor  Induction.
Male and  female Swiss mice were Injected  s.c. with  100  vg  chrysene  1n  PEG
on  the  day  of birth and  the  next 2  days.   After 70 weeks tumor  Incidences
were the following:   one  Injection of PEG caused liver  tumors 1n  5/20 males
and  3/21  females  and lung  tumors  1n  2/20  males  and  3/21  females;  three
Injections of  PEG  caused liver  tumors  In 10/30  males  and 0/15  females  and
lung tumors  In  4/30 males and 1/15 females;  chrysene caused liver tumors In
13/27 males and  0/21  females  and lung tumors  1n  2/27 males and 1/21  females
(Gi-over et al., 1975).

    In  a  similar  experiment,  Swiss Webster  BLuVHa(ICR)  mice  received  l.p.
Injections of  a  DMSO  solution  of  chrysene  on days 1,  8 and 15 of  life.
Total administered  dose was  1.4 vmol.   By  38-40 weeks  treated  animals  had

04420                                V-94                            10/08/91

-------
developed  lung  tumors (5/24 males,  2/11  females)  and  hepatic tumors  (6/24
males), and  one lymphosarcoma was  observed 1n  a  male mouse.   DMSO-treated
animals developed  only lung tumors  (2/21  males,  7/38  females)  (Buenlng  et
al.,  1979a).   The  same  protocol  was  applied  1n  a  graded  dose  bloassay
wherein the  three  1.p. Injections were of  0.2,  0.4 and 0.8  pmol  repurlfled
chrysene.   There was  no Increase  1n  pulmonary  tumors, but  22% of the treated
male mice  developed  hepatic  tumors  compared with no hepatic  tumors  observed
1n the DMSO controls (Chang et  al, 1983).

    D1benz[a.hlanthracene.   Pulmonary administration  of  d1benz[a.h]anthra-
cene has Induced lung  adenomas 1n mice  (Andervont,  1937;  Rask-N1e1son,  1950,
Kuschner et  al.,  1956).   Intratracheal Instillation of a  d1benz[a,h]anthra-
cene suspension 1n protein blood  substitute with powdered  India Ink  resulted
1n  the development   of  squamous  cell  carcinomas   (Yan1sh1va  and  Balenko,
1966).  Single  1.v.   Injections  of  a colloidal  dispersion of  this  material
were also  tumorlgenlc 1n  a  dose-dependent  fashion  In strain A mice (Heston
and Schnelderman,  1953).
      /
    Kennaway (1930) was the  first of  many researchers  to  report Induction of
skin tumors  by  d1benz[a,h]anthracene.   In  another  study,  a 0.2%  solution  of
d1benz[a,h]anthracene  1n  acetone/benzene  was  painted  twice  weekly on  the
skin  of  Swiss  mice  (38  yg/dose).   Skin   tumors   were  observed  1n  16/20
animals (Lljlnsky et  al., 1965).  Van OQQren et  al.  (1967) topically exposed
ICR/Ha Swiss mice  3  times/week to acetone d1benz[a,h]anthracene solutions at
the following concentrations:  0.001, 0.01  and 0.1X.  Tumor  Incidences were
1/30  (1  carcinoma),  43/50  (39  carcinomas)   and  39/40  (32  carcinomas),
respectively.  The authors noted  that latency was also dose related.


04420                                V-95                             10/08/91

-------
    D1benz[a,h]anthracene was  compared  with benzo[a]pyrene as  to  potency  1n
 Inducing  skin  tumors  1n Swiss  mice.   Repeated  skin exposure  to a  0.001%
 acetone solution of  benzo[a]pyrene  Induced  paplllomas  1n  43%  of the mice and
 carcinomas  1n  3%.   The  same  d1benz[a,h]anthracene exposure  resulted  In  a
 papllloma  Incidence  of  30%  and  a  carcinoma  Incidence of 30%.  Exposure  of
flilce  to  a  0.01%  solution  of  either  compound  resulted 1n  papllloma  and
                                      *
 carcinoma  development  In >90%  of the animals.  Latent periods  for Induction
 were similar for both carcinogens (Wynder and Hoffman,  1959a).

    D1benz[a,h]anthracene has  been  reported to Initiate  skin  tumor develop-
 ment  In  mice at doses as low as 0.02  pg  given  once (Klein, .1960).   Other
 tumors  such  as paplllomas and pulmonary tumors may also arise as a  conse-
 quence  of skin exposure to d1benz[a,h]anthracene  (Buenlng  et al.,  1979b).
 Virgin  C3H  (Jax)  mice  were  painted twice weekly  with  0.25%  d1benz[a,h]-
 anthracene 1n  benzene  (thlophene-free)  for their lifetime.  A  50% Incidence
 of mammary tumors was  observed 1n the control  animals  as  compared  with 10/11
 1n the treated mice  (Ranadlve and Karande, 1963).

    In one study hamsters appeared  to be somewhat more resistant  to the skin
 tumor!genie  properties  of   d1benz[a,h]anthracene.   A group  of  10  Syrian
 golden  hamsters received 20  applications  of  0.2% d1benz[a,h]anthracene  In
 mineral oil  over a  10  week  period.    At  50  weeks, 5  animals survived,  but  no
 tumors were observed (Shublk et al., 1960).

    Several assays have  been reported wherein  d1benz[a,h]anthracene has been
 administered s.c. or 1.m.  Tumors  at Injection sites  have been reported  1n
 rats, guinea  pigs  (low  Incidence),  pigeons  and unspecified  fowl  (Roussy  et
 al.. 1942;  Shabad,   1938; Shabad and UMnson,  1938;  PMchard et  al.,  1964;

 04420                                V-96                            10/08/91

-------
Peacock, 1935).  Mice  are  the  animals  most studied 1n this  regard.   Reports
Included  those  by  Hartwell  (1951),  Shublk  and  Hartwell   (1957,  1969)  and
Thompson and Co. (1971).

    The effective dose for s.c. Induction of tumors was  established  by  Bryan
and  Shlmkln  (1943).   TMcaprylln  solutions  of  d1benz[a,h]anthracene  were
administered once  to  groups  of >19 C3H mice  so as to deliver  the  following
doses:  0.0019,  0.0078,  0.016, 0.03,  0.06,  0.12, 0.25,  0.5,  1,  2,  4  or  8
mg.   Incidences  of Injection  site sarcomas  were as  follows:   2/79,  6/40,
6/19,  16/21,  20/20,  21/23,  19/21,  20/21,  22/22,  19/19,   17/20  and  16/21,
respectively.  It was  noted  that  the  lowest  effective dose  for. d1benz[a,h]-
anthracene was 0.0019 mg while that for 3-methylcholanthrene was >0.0039 mg.
The  authors  also observed that  the average  latent  period  for  d1benz[a,h]-
anthracene  sarcoma  development was 3.7 months  as compared  with 2-5  months
for  3-methylcholanthrene  and 3 months  for  benzo[a]pyrene.   Similar  results
were reported by Oobrovolskala-Zavadska'ia  (1938) and  Lettlnga (1937).
       /
    In an  abstract, Platt et al.   (1983) reported  that  d1benz[a,h]anthracene
showed  stronger  carcinogenic responses when administered  s.c.  to NMRI  mice
than  did   the  following  metabolites:   3,4-d1ol,  5-phenol,   5,6-ox1de.   The
l,2-d1ol  and  5,6-d1ols  of  d1benz[a,h]anthracene  were reported  to  be  non-
carcinogenic by this treatment  regimen.

    Newborn mice  (general purpose/NIH) were  exposed  by  s.c.  Injections  of
d1benz[a,h]anthracene   doses   between  0.003  and  6.7   yg/mouse.    A  dose-
related Increase  In  sarcoma  Incidence was noted for  doses  of  >0.08  yg, and
an  Increase In  lung   adenomas  was observed  for  doses  >0.2  yg (O'Gara  et
al., 1965).

04420                                V-97                             10/08/91

-------
    Female NMRI mice  were  observed to develop Injection  site  sarcomas  after
one  s.c.  treatment  of as   little  as  2.35  yg  d1benzo[a,h]anthracene  1n
trlcaprylln (Pfelffer,  1977).  No  vehicle controls were  reported  but  a  group
of."non-carcinogenic" PAHs tested  concurrently were  observed  to  have  sarcoma
Incidences  of  4-13X.   Table V-19  compares sarcoma  Induction for  dlbenzo-
[a,h]anthracene and benzo[a]pyrene.«

    Lubet et al.  (1983a) found  that  s.c.  Injections  of d1benz[a,h]anthracene
were  associated  with flbrosarcoma  development  1n  mice, but  only for  some
strains.  Four strains  of mice used  Included two,  C3H/HeJ and C57B1/6J, that
respond  to  3-methylcholanthrene treatment  by  Increased  levels and types  of
hepatic  enzymes  Including AHH.   Two strains,  AKR/J  and  DBA/2J, were  non-
responders.  Groups  of  30 animals  were Injected with a single dose of 150
vg  d1benz[a,h]anthracene  1n  0.05   ml  trloctanoln  and  observed  for   9
months.   A  control group  for each  strain, consisting  of  10 animals  each,
received  a  s.c.  Injection of 0.05 ml  trloctanoln alone.   A summary of the
findings  Is given  1n  Table V-20.   The tumor Incidence  1n the treated  animals
varied  between  0  and 80%,  depending on  the  strain.  Tumor  Incidences  were
higher  1n the C3H  and C57B1  mice,  which also were  readily Indudble for AHH.
Likewise, the average latency period (In days)  for  flbrosarcoma  development
varied  with the strain  and  tended to be  Inversely correlated  with the  tumor
Incidence rate.   The  authors concluded  that,  as for  benzo[a]pyrene, the  Ah
receptor  was  Involved  1n the process of  tumor  Induction by  s.c.  Injection
for d1benz[a,h]anthracene.

    Fluoranthene.    Fluoranthene  was  first  tested  for carcinogenic  activity
more  than  5  decades ago  (Barry et  al.,  1935).   The results  from  that


04420                                V-98                            10/08/91

-------
                                  TABLE V-19

    Sarcomagenlc  Activity  of  Subcutaneously  Injected D1benzo[a,h]anthracene
                    and  Benzo[a]pyrene  1n Female NMRI Mice*
Compound
D1benzo[a,h]anthracene





Benzo[a]pyrene





Dose (tig)
2.35
4.7
9.3
18.7
37.5
75.0
3.12
6.25
12.5
25.0
50.0
100.0
Tumor Incidence
37/100
39/100
44/100
56/100
65/100
69/100
9/100
35/100
51/100
57/100
77/100
83/100
*Source: Pfelffer, 1977
04420
V-99
10/08/91

-------
                                  TABLE V-20

        Incidence  of  Flbrosarcomas  1n Mice Associated with Subcutaneous
                     Injections of D1benz[a,h]anthracenea
Strain
C3H/HeJ
C57B1/6J
AKR/J
DBA/2J
Treatmentb
D1benz[a,h]anthracene
Control
D1benz[a,h]anthracene
Control
01benz[a,h]anthracene
Control
D1benz[a,h]anthracene
Control
Tumor
, Incidence
24/30
0/10
16/30
0/10
0/30
0/10
1/30
0/10
X
80
0
33
0
0
0
3
0
Average
Latency
(days)
165
0
242
0
0
0
230
0
aSource: Lubet et al., 1983a

bAn1mals In  the d1benz[a,h]anthracene group  received  a single  Injection  of
 150  yg  of  d1benz[a,h]anthracene  In  0.05  ml  of  trloctanoln.    Control
 animals received an Injection of 0.05 mi trloctanoln alone.
04420
V-100
10/08/91

-------
Investigation, and from several studies conducted since that  time,  Indicated
that fluoranthene had virtually no activity as  a complete  carcinogen.

    Suntzeff  et  al.  (1957) administered  a 10%  solution  of  fluoranthene  In
acetone  3  times weekly  by topical  application to CAP,  Jackson,  Swiss  and
MUlerton mice.  No tumors were found by 13 months.

    Wynder and Hoffmann (1959a) administered a  0.1X solution  of  fluoranthene
In acetone to  the backs of  20  female Swiss (Mlllerton)  mice 3 times/week  for
life.  No tumors were found.

    Hoffmann  et  al.  (1972)  administered  50  vl  of   a   IX  fluoranthene
solution to  the backs of  20 female  Swiss-Albino  Ha/ICR/MUl mice 3  times/
week  for  12  months.    All  treated  mice  survived,   and   no  tumors  were
observed.  As  part  of  the  same study,  30 mice received 0.1  mg  fluoranthene
1n  50  yl acetone every  second day  for  a total of 10  doses; 10 days  later
promotion with  2.5X  croton oil 1n  acetone was started and continued for  20
       /
weeks.   A single papllloma was  noted 1n 29 surviving mice.

    Morton and  Christian  (1974) administered  50  mg  fluoranthene 1h decalln
or In decal1n:n-dodecane (50:50) to  the backs  of  15 male  C3H mice.  The mice
were treated 2 times/week for 82 weeks.   No skin tumors  were observed.

    Van  Ouuren and  Goldschmldt  (1976)  administered  40  yg  fluoranthene  1n
acetone-  3  times  weekly for 440 days to  female  ICR/Ha  Swiss mice.  No skin
tumors  were observed.
04420                                V-101                            10/08/91

-------
    Barry  et  al.  (1935)  administered four  doses  of  10  mg fluoranthene  In
glycerol  by  subcutaneous  Injection  to  strain A mice.  Six out  of 14  mice
survived for 18 months; no tumors were found by 19  months.

    Shear  (1938)  administered  four doses of  10  mg fluoranthene  1n  glycerol
by subcutaneous  Injection  to strain A mice.   Six out  of  14 mice survived for
18 months; no tumors were found by 19 months.

    Fluoranthene was  tested  by Busby et  al.  (1984)  1n a mouse lung adenoma
assay.   Newborn  male  and  female  mice  of Swiss-Webster  BLU:Ha strain  were
Injected  IntraperHoneally with  a total  dose of 700  yg  (163  mg/kg) or  3.5
mg (815  mg/kg)  fluoranthene  In three Installments  on days 1, 8 and  15  (1/7,
2/7 and  4/7 of  the dose, respectively).  H1stopatholog1c examination of  the
lungs  of the  mice  sacrificed  at 24 weeks  showed  a  significant  Increase  1n
the high-dose group 1n the  total  Incidence  and number/mouse (28/48 or  58%,
1.08  tumors/mouse)  of lung  tumors  compared with  vehicle  controls (5/55  or
9%, 0.09 tumors/mouse).  Tumor response 1n the low-dose group (10/51 or  20%,
0.24  tumors/mouse)  was not  statistically  significant.  A positive  response
In a  lung adenoma assay  1n the absence  of  corroborating  studies Is generally
considered Insufficient evidence  of carc1nogen1c1ty.

    Fluorene.   Studies of  fluorene as a  complete  mouse  skin carcinogen  and
as a  cocardnogen  with  3-methylcholanthrene  were negative or  Inconclusive
(Kennaway, 1924a; Rlegel et al., 1951).  IARC  (1983)  considered both reports
Inadequate for evaluation.
04420                                V-102                           10/08/91

-------
    Fluorene was  also  found to be  Inactive  both  as a tumor  Initiator  (with
TPA  as  a  promoter)  and  a complete  carcinogen  1n  skin-painting  bloassays
performed  1n  female  Ha/ICR   mice  by  LaVole  et  al.  (1979,   1981b).    No
Injection  site  tumors  were produced  1n 10  strain  A  mice  after seven  s.c.
Injections  of  10 mg  fluorene  1n  glycol.   The  study was  terminated  at  18
months (Shear,  1938).

    IndenoH,2.3-cdlpyrene.    Indeno[l,2,3-cd]pyrene   was   administered   to
female OH  rats  by Implantation  1n the  lung  of beeswax-tMoctanoln pellets
containing  0.16,  0.83  or  4.15 mg  of the compound.   Results summarized  In
Table  V-21  Indicate  a  dose-dependent  Increase  In  keratlnlzed epldermold
carcinomas capable of  Invading the extrapulmonary chest wall.  There was  no
Increase  In ple1omorph1c  sarcomas.   A  calculated  potency  from this  assay
relative to  that  of  benzo[a]pyrene (=1.00)  was 0.08  (Deutsch-Uenzel et  al.,
1983).

    By contrast to these  results,  1ndeno[l,2,3-cd]pyrene was  not tumorlgenlc
      /
1n newborn  CD-I  mice treated  1.p.  on days  1,  8  and  15 of life  (total  dose
2.1 vmol, see Table V-16) (LaVole et al.,  1987).

    Groups  of  20  female  Swiss  albino Ha/ICR/MH  mice were given  topical
applications of  1ndeno[l,2,3-cd]pyrene  prepared as dloxane or acetone  solu-
tions.  Dloxane  preparations  of 1ndeno[l,2,3-cd]pyrene at  concentrations  of
0.05  or  0.1X did not  Induce  skin  tumors whereas benzo[a]pyrene at  the  same
concentration produced -tumors In  90X of the  treated  mice  In 7 months.   By
contrast,  acetone solutions  of  1ndeno[l,2,3-cd]pyrene produced skin  tumors
In a  dose-related fashion. No tumors were observed  In animals  painted with


04420                                V-103                           10/08/91

-------
                                 TABLE V-21

   CardnogenlcHy of Indeno[l,2,3-cd]pyrene upon  Implantation 1n Rat Lungs3

Treatment


Untreated
Pelletb
Indeno[l,2,3-cd]-
pyrene


Dose
(rug)

—
—
0.16
0.83
4.15
Median
Survival
Time
(Weeks)
118
104
116
109
92
Tumor

Ep1dermo1d
Carcinomas
0/35
0/35
3/35
8/35
21/35
Incidence

Sarcomas

0/35
0/35
1/35
0/35
0/35


Total

0/35
0/35
4/35
8/35
21/35
aSource:  Deutsch-Wenzel et al., 1983
      t
^Refers  to beeswax  and trloctanoln Implantation medium
04420
V-104
10/08/91

-------
the 0.01 and 0.05X solutions.  The 0.1X  treatment  Induced  six  paplllomas  and
three carcinomas  beginning at 9  months; the  0.5X concentration of  Irideno-
[l,2,3-cd]pyrene  resulted  1n seven paplllomas  and five carcinomas with  the
first tumor  appearing at  3  months.   This paper also  reported that a  total
dose  of  250 yg  1ndeno[l,2,3-cd]pyrene  delivered  1n 10  applications  was  a
sufficient  Initiating  dose when  followed by  croton oil promoting  treatment
(Hoffman and Wynder,  1966).

    Indeno  [l,2,3-cd]pyrene  was  not  shown to  be  a complete carcinogen  when
applied  to  the skin  of'female NMRI mice twice a week for their  lifetime.
Total doses  applied  1n acetone were  0,  3.4,   5.6  or  9.2 mg/an1mal (Habs  et
al., 1980).

    Solutions  of  4.0  yrnol  1ndeno[l,2,3-cd]pyrene   and   two   fluoridated
metabolites,    2-fluoro1ndeno[l,2,3-cd]pyrene     and     8,9-d1fluoro1ndeno-
[l,2,3-cd]pyrene,   were  applied   to  the shaved   backs  of  30  female  CD-I
                                                                            •
mice/group  every  other  day   for  a total  of  10   doses.   Mice treated  with
benzo[a]pyrene and  acetone  served  as  positive and negative  controls.   Ten
days  after   the  last Initiating dose  12-0-tetradecanoylphorbol-13-acetate
(TPA) was applied to  the backs of the  mice  as a promoter 3 times/week for 20
weeks.   At  the end  of the  promotion  period  1ndeno[l,2,3-cd]pyrene and  TPA
treatments  Induced  tumors  1n  72X  of   the  mice  (2.1  tumors/mouse),  the
2-fluoro1ndeno[l,2,3-cd]pyrene and TPA  treatments  Induced  tumors  In  76X of
the  mice (3;9 tumors/mouse)  and  the  8,9-d1fluoro1ndeno[l,2,3-cd]pyrene  and
TPA treatments Induced tumors 1n  40X of  the  treated mice (0.6 tumors/mouse).
At  the end  of  the Initiation period,  DNA from the skins  of  5  mice/group was
Isolated.   [32]P-postlabel1ng analysis  of  the hydrolyzed  DNA   showed  that


04420                                V-105                           10/08/91

-------
1ndeno[l,2,3-cd]pyrene and 2-fluoro1ndeno[l,2,3-cd]pyrene each  form a  single
major DNA  adduct  when analyzed by  thin-layer  chromatography.   These adducts
have different  retention  behaviors.   8,9-01fIuoro1ndeno[l,2,3-cd]pyrene  does
not  have  this  adduct; the authors  speculated  that this could  be  one  factor
1n the variance of  the tumor  Incidences  of  the three compounds  (R1ce et  al.,
1990).                                         •

    As part  of  a  study of Initiating capability of  Us  major metabolites  1n
mouse skin,  1ndeno[l,2,3-cd]pyrene was  applied to  the shaved  backs  of  20
Crl:CD-l(ICR)BR female mice.  Acetone solutions were  applied, every other day
for  10 days  for a total  Initiating dose of 1  mg.  This  was  followed 10  days
.later by  thrice weekly TPA applications  (0.0025X  1n 100 yl acetone)  for  20
weeks.  Tumor  Incidence was  essentially 100X.   Indeno[l,2,3-cd]pyrene-l,2-
dlol  and   -l,2-ox1de  treatment  both  resulted  1n 80%  tumor  Incidence  by
comparison  with  8-hydroxy  (~25X)  and  the acetone-treated  controls  (~5X)
(R1ce et al., 1986).
      /
    An  earlier  Initiation-promotion  bloassay  performed   by   Rice  et   al.
(1985c)  showed  a  pronounced  dose-response  relationship.  Following the  same
protocol  described above, there was an 80%  tumor Incidence  In mice receiving
a  total  Initiating dose of  1 mg  1ndeno[l,2,3-cd]pyrene with  an  average  of
about four  tumors/mouse  after  22  weeks of  promotion.  However, when  the
total  Initiating,  dose  was   decreased  to  100  or  300  yg,  the  number  of
tumor-bearing mice was below statistical  significance.

    Injection  site sarcomas  were  reported  In 10/14 male  and 1/14  female
XVIInc/Z mice  administered  three  Injections at 1  month Intervals  of  0.6  mg
1ndeno[l,2,3-cd]pyrene 1n olive oil (Lacassagne et  al.,  1963).

04420                                V-106                           10/24/91

-------
    Naphthalene.    In   an   Inhalation   study,   naphthalene   (scintillation
grade,  purity  >99%) vapor  was administered  at 0  and 10  ppm  to 75  B6C3F1
mice/sex/dose for 6  hours/day,  5  days/week for  103 weeks  (NTP,  1991b).   Two
groups  of  75 B6C3F1 mice/sex  were administered  30 ppm naphthalene for  the
same amount  of  time as  the 0 and  10  ppm groups.   In each  group  25  mice/sex
were designated  for  hematology  evaluation;  5 mice/sex were to be sacrificed
after 14  days  and 3, 6,  12  and  18 months.   Only the  14-day  hematology  mice
were  sacrificed  1n  the control  male  mice  because  of high mortality.   When
compared with survival  In exposed  male groups,  survival was decreased  1n the
male  control  group;  the decrease  was  the result   of  wound  trauma  from
fighting  among  animals   caged  together.   Survival  was  equivalent  In  all
female  groups.   Final   body  weights  were equivalent  In  all  groups and  no
clinical observations could be attributed directly to naphthalene.
                                                               *

    There  was   a significant   Increase   1n  the  Incidence  of  alveolar  or
bronchlolar  adenomas  1n  the  high-dose females by  palrwlse  comparison  with
controls.  The  Incidences were  5/68 (7X), 2/64  (3X)  and 28/134  (21X)  1n the
      t
control,  low-  and  high-dose groups,  respectively.   The  recent  historical
control  Incidence  for  females  1n  this  laboratory  1s  91/1166 (7.8X,  range
0-16X)  and  the  recent  historical  control Incidence for females  specifically
In Inhalation studies Is  39/466 (8.4%, range 0-12%).   The  first  Incidence of
alveolar or bronchlolar adenomas was reported on  experiment day  714, 736 and
656  In  the control,  low-  and   high-dose  groups,  respectively.   In  the
high-dose  females there  was a  single Incidence  of alveolar  or  bronchlolar
carcinoma;  this  was not  observed  In  the other  groups.   There  was also  a
statistically   significant   Increase   In  the   Incidence   of   respiratory
epithelial cell  hyperplasla  In the naphthalene-treated groups.   Because the


04420                                V-107                           10/24/91

-------
 Incidence  of  alveolar or bronchlolar  adenomas  was  statistically significant
and  above  historical  control   ranges,   NTP  judged  the  Increased  adenoma
 Incidence  to be directly related to naphthalene exposure.

    There  was  a   statistically  significant  Increase  In  the   Incidence  of
respiratory  epithelial  cell  hypefplasla  1n   the  naphthalene-treated  male
groups.  There  was no significant  Increase  In  the Incidence of alveolar  or
bronchlolar  adenomas  or  carcinomas  (either  separately or  combined).   The
 Incidences of alveolar or bronchlolar  adenomas  or  carcinomas  (combined) were
7/69  (10%),  17/69  (25%)  and 31/135 (23%)  In  the  control,  low-  and high-dose
groups,  respectively.  The  male  control  Incidences  were 94/478  (19.7%, range
10-30%)  1n recent  Inhalation  studies  and 229/1172  (19.5%,  range  6-42%)  In
all  carcinogen  studies conducted  recently 1n  this  laboratory.  As  of this
writing  the U.S. EPA has not evaluated this study.

    Adklns et al.  (1986)  exposed groups of 30  six- to  elght-week-old female
A/J  strain mice by  Inhalation to  naphthalene at concentrations of  0,  10  or
30 ppm,  6  hours/day,  5 days/week  for  6 months.  After  the 6-month exposure
period,  excised lungs were examined  for  tumors.   Naphthalene  exposure  did
not  significantly  Increase  tumor  Incidence,  but  did cause a  statistically
significant  Increase (p<0.05) In. the number of  adenomas  per  tumor-bearing
mouse lung.  There was no dose-response.

    Schmahl (1955)  administered  naphthalene by  1.p. Injection  to  a  group  of
10  rats  (In-house  strains   BDI  and BOIII)  with  another  group  of  10 rats
serving as controls.  The dally  dose was  20 mg/rat  and  Injections were given
weekly  for  40  weeks.   Animals  were  observed  until  spontaneous  death.
Hlstologlc examination offered no evidence of  carcinogenic  effects.

04420                                V-108                           10/08/91

-------
    Boyland et  al.  (1964) Implanted  naphthalene  Into the bladder  of stock
Chester  Beatty  mice  (23)  In an  effort  to  determine  the   suitability  of
naphthalene as  a  potential  vehicle  for cardnogenldty  testing.   The mice
were observed for  30 weeks.  Tumor Incidence was as low as when paraffin wax
was  used  and  lower  than  with  cholesterol.  Naphthalene  was judged  to  be
Inert but  to  have no advantage over  cholesterol  as  a base for Implantation
pellets.

    Coal tar-derived naphthalene that contained -10%  unidentified Impurities
was  tested  for  cardnogenldty  by  Knake  (1956).    White,  rats   (40,  sex
unspecified) were given  seven subcutaneous  Injections of  500 mg/kg naphtha-
lene In  sesame  oil  at 2-week Intervals.   Lymphosarcomas  were found 1n 5/34
surviving  rats  at  18 months  (14.7%),   whereas  vehicle  controls  had  a  2%
Incidence of  these  tumors.   Mice  (25,  Inbred black)  were painted with 0.5%
naphthalene 1n  benzene  5 days/week  for life.   Four  treated  mice developed
leukemlas In contrast to  0/21 vehicle controls;  the  untreated control Inci-
dence was 0.4%.  The value of these studies for assessing cardnogenldty 1s
very limited  due to  the  presence  of  potentially  carcinogenic  Impurities.
Moreover, the vehicle 1n  the  mouse  study has  been shown to cause leukemlas,
and the site of Injection 1n the rats study was painted, prior to Injection,
with carbofuchsln, a known carcinogen.

    Kennaway  (1930)  reported  that naphthalene was not carcinogenic In skin
painting studies  1n mice.   The  concentration,  purity, dosing  regimen, and
other details were  not provided.  The  reaction product  of  naphthalene and
aluminum trichloride  was  reported to be carcinogenic, but  the  product was
not Identified.

04420                                V-109                           10/08/91

-------
    Schmeltz  et  al.  (1978)  tested  the carcinogenic  activity of  benzo[a]-
pyrene  and  naphthalene  In  female  ICR/HA  (Sprague-Dawley) mice.   A 100  vi
solution containing  0.25% naphthalene and 0.003%  benzo[a]pyrene  was painted
on  the shaved  backs  of  30  mice 3  times/week  for  78 weeks.   Naphthalene
Inhibited  benzo[a]pyrene-1nduced  tumors;  -42%  of  the  mice  treated  with
benzo[a]pyrene alone  had skin  tumors  whereas -20%  of mice had  skin  tumors
when naphthalene and benzo[a]pyrene were administered together.

    Phenanthrene.   Data  on  phenanthrene's potential  as  a  skin  carcinogen
have  been  summarized   by IARC .(1983).    Phenanthrene  did .not   serve  as  a
complete carcinogen  In two studies  1n  mice  that were  Incompletely  reported
(Kennaway,  1924b;  Roe  and  Grant,  1964).   Phenanthrene -was  reported to  be
active  as  an Initiator  1n  female CD-I mice  1n  one study  (ScMbner,  1973).
Thirty-five  weeks   after  Initiation  with  10 ymol  phenanthrene and  twice
weekly  treatments   with  5  ymol  TPA,  12/30  mice  developed  paplllomas  as
compared  with   0/30  TPA-treated  controls.     Phenanthrene.  however,   was
Ineffective as an Initiator of  skin  tumorlgenes 1s  In Swiss Ha/ICR,  albino, S
       *•
and CD-I mice  (LaVole  et al., 1981a; Roe, 1962; Salaman and  Roe, 1956;  Wood
et al., 1979) and  Inactive  as a  promoter  In  mice  of unspecified  strain  (Roe
and Grant,  1964).   It was  noted  by Wood et  al.  (1979) that  30  female  CD-I
mice given  one  topical  application  of 10 ymol  phenanthrene or  the 1,2- or
3,4-d1hydrod1ol   followed  by  twice  weekly applications  of 16  ymol  TPA  (In
200 yi acetone)  were  observed  to have  papllloma  Incidences 2-4  times  that
of  background.   These   Incidences  were  not   significantly  Increased  by
comparison with  controls  because of  the  small  number  of  animals  tested and
the spontaneous  tumor Incidence.
04420                                V-110                           10/08/91

-------
    Subcutaneous administration  of phenanthrene  to  mice did  not  result  In
treatment-related  Increases  In  tumor   Incidence  (Stelner,  1955; Roe,  1962).
Likewise,  neither  s.c.  nor  1.p.  treatment  of  neonatal  mice  resulted  In
significant  tumor  Induction (Grant and  Roe,  1963; Buenlng  et al.,  1979a).
While the  1,2-d1ol-3,4-epox1des  of phenanthrene are mutagenlc  to  Salmonella
typhlmuMum  and  to  mammalian  (V79)  cells,  they  did not  Induce  pulmonary
tumors 1n newborn mice (Wood et  al., 1979;  Buenlng et  al., 1979a).

    Pyrene.   Intratracheal  Instillation  of   pyrene   with   Fe_03   particles
did  not  result  1n  Increased numbers  of respiratory  tumors. In male  Syrian
golden hamsters (Sellakumar and  Shublk, 1974).

    As part  of  a  study of nitrated PAHs (W1slock1 et  al., 1986),  a  dimethyl
sulfoxlde  solution of  pyrene was  administered  1.p.  to newborn CD-I mice  on
days  1,  8 and  15  of age.   Total  administered doses  were  200, 700 or  2800
nmole.  Animals were  weaned, separated  by  sex  and observed without  further
                                                                           •
treatment  to  1  year.   A  small  but significant Increase 1n  liver  carcinomas
      /
was observed  In the mid-dose male  mice only (Table V-22).  The Incidences  of
total  liver   tumors  (adenomas  and carcinomas),   lung tumors  or   malignant
lymphomas were not significantly elevated 1n treated animals.

    Tests  of  pyrene  as  a complete skin carcinogen  and as  an Initiator  of
carclnogenlcHy 1n mice  have been negative or  Inconclusive  (Badger et  al.,
1940;  Roe and  Grant,  1964; Norton  and  Christian,   1974;  Van  Duuren and
Goldschmldt,   1976;  Salaman and  Roe,  1956;  Scrlbner,  1973).  Co-adminis-
tration of pyrene and benzo[a]pyrene  to  the backs  of  ICR/Ha  mice resulted  1n
an  enhancement  of   the  benzo[a]pyrene  tumor1gen1c1ty  (Van  Duuren  and
Goldschmldt,  1976;  Goldschmldt  et al., 1973).

04420                               V-lll                            10/08/91

-------
                                  TABLE V-22
                       Assay of Pyrene In Newborn M1cea
Tumor Incidence
Sex
H
F
M
F
M
F
H
F
Total
Dose
(nmol)
0
0
200
200
700
700
' 2800
2800
Liver
Adenomas
7/73
0/65
0/29
0/31
0/25
0/49
2/14
0/18
Liver
Carcinomas
0/73
0/65
0/29
0/31
3/25b
0/49
1/14
0/18
Lung
Adenomas
3/73
1/65
1/29
1/31
2/25
5/49
1/14
1/18
Lung
Carcinomas
2/73
1/65
0/29
0/31
0/25
0/49
0/14
0/18
aSource: Wlslockl et al., 1986
bS1gn1f1cantly different from controls (p<0.05).  Fisher exact test
04420
V-112
10/08/91

-------
    Pyrene  did  not  produce  tumors  1n  Jackson  A mice  Injected  s.c.   and
observed 18 months (Shear and LeUer,  1941).

Reproductlve/Teratoqenlc Effects
    Evidence  of   deleterious  reproductive  effects,  documented  for  PAHs  In
general and benzo[a]pyrene In particular, has teen equivocal.  This  evidence
1s considered here.

    Anthracene.   A  total  of   8  mg  anthracene/mouse  was   administered  as
dally s.c. doses  or  In  a single oral dose during the last week  of  gestation
to  dams  of   the  following  strains:   BALB/C,  C3H/A  and C57B1  X  CBA  FI.
Fetuses were  removed and  the  kidney  cells  established  1n  culture.  These
cells  exhibited  enhanced  plating  efficiency  as  well  as some  hyperplastlc
changes  by comparison  with  fetal  cells  obtained  from untreated  animals
Indicating the ability of anthracene to pass through the  placenta  (Shabad et
al.t 1972).

    BenzFalanthracene.    Wolfe   and  Bryan  (1939)   reported  fetal  death  and
resorptlon In  two pregnant rats as a  consequence  of  s.c. Injection of  5  mg
benz[a]-anthracene beginning  with  day 1 of  gestation.

    Benzofalpyrene.   Rlgdon   and   Rennels  (1964)  conducted  two  series  of
experiments In  rats  (strain not  specified)  to ascertain possible  reproduc-
tive consequences  of dietary benzo[a]pyrene.   In  series  one, eight  females
and an unspecified number of males were fed  laboratory  chow  to which 1 mg/kg
benzo[a]pyrene had  been added, and  a control  group  of  six  females and  an
unspecified number of males  were  fed a standard diet.  Treated  females  were


04420                                V-113                            10/08/91

-------
mated with  control  males,  and control females were mated  to benzo[a]pyrene-
treated males.   Vaginal  smears were  taken  during a 28-day  period  beginning
with  the   first  day  of  benzo[a]pyrene  feeding.   The  authors  observed  no
treatment-related  effects  on  the  estrus  cycle.   Three normal  pregnancies
were  reported  for  the  control   females,  and  five  benzo[a]pyrene-treated
females also  became  pregnant.   Of the treated-females,  only one delivered a
total of  four  pups at day 23.  Two  of  the  four  pups were stillborn,  one of
which  was  grossly  malformed.    One pup  left  with  the  dam   died  3  days
postpartum, presumably of  starvation.  Of the dams  not delivering, one  was
found upon autopsy to carry four dead fetuses.

    In  the  second  series 6 control  male  and  female rats were mated as  were
males and 7  females  fed benzo[a]pyrene  as   described  above.   The  pregnant
control  rats  had   apparently  normal  pregnancies.   Two  benzo[a]pyrene-fed
females of  seven mated became  pregnant.  On  autopsy.  It was found  that  one
dam carried four dead fetuses and that fetal  resorptlon had occurred  In  the
other.  While these  data  suggest a  reproductive effect of benzo[a]pyrene,
the report was unclear as to  specifics  of experimental  design and length  and
timing  of  the feeding period  for  treated males  and females.  In  this  study
the  1  mg/kg  dose  appears.to  be  the  LOAEL   with  no  NOAEL  for  reproductive
effects 1n rats.

    Rlgdon and Neal  (1965) conducted a series of  four  experiments  on  repro-
ductive effects In Swiss mice  fed  diets containing benzo[a]pyrene at concen-
trations  of  0, 250,  500 or  1000  ppm.   The  animals  were  given the dietary
benzo[a]pyrene for various  time  spans  during mating,  gestation and  post-
partum.    Blue  fluorescence.  Indicative  of the presence  of benzo[a]pyrene or

04420                                V-114                           10/08/91

-------
Its metabolites,  was demonstrated  In  the Internal  organs  of  treated  mice.
No  teratogenlc,  embryotoxlc,  fetotoxlc  or other  reproductive effects  were
observed  In the  treated animals.   Fertility  of  male  mice  was  apparently
unaffected.  Males fed the 500  ppm  diet  for  30 days  were shown to have sperm
present In the lumen of testlcular tubules.

    In a study of CD-I strain  pregnant mice,  however,  HacKenzle and Angevlne
(1981) found adverse  reproductive effects In the offspring of  animals  dosed
with  benzo[a]pyrene.   Benzo[a]pyrene 1n  corn  oil  at  doses of  0, 10,  40  or
160 mg/kg  bw was  administered  by gavage  on days 7-16  of ges.tatlon to groups
of  30  or  60 dams.   At  these doses  benzo[a]pyrene was not  toxic  to  dams  or
fetuses.  Pregnancies'were carried  to term and  pups  permitted to  nurse until
weaning.   Number  of  pups  and  gross  abnormalities  were  scored on  day  1
postpartum, and on day  4 pups  were  weighed and sexed.   At  this time  Utters
were culled to 8  pups for breeding  studies.   At 6  weeks of age  10 F,  males
and  10 F,  females  were sacrificed for. hlstologU  examination of  repro-
ductive  organs.   At  7  or  8  weeks the  remaining  F,  mice  were bred  with
       /
untreated  animals.   Females,  both  those  treated  \n  utero and  those  mated
with  F-  males,  were  sacrificed on days  14-19  of  gestation,  and  numbers  of
Implants,  fetuses and resorptlons  were  recorded.   Gross abnormalities  were
also noted.

    Reduced fertility and reproductive  capacity  were  also observed  In  the
F.  mice.   The FI  males  exposed  to 10  mg/kg  bw  in  utero  showed a  marked
reduction  In  gonadal weight,  although  no effect  on body weight  was  found.
All control males (45/45) were fertile (sired at least  one litter) and as a
group  Impregnated 80% of the females to  which  they  were exposed.  Fertility
04420                                V-115                           10/08/91

-------
rates  were 20/25,  3/45 and  0/20  for  the  10,  40  and  160  mg/kg male  dose
groups,  respectively,  and  the percent  of  Impregnated females was 52,  7  and
OX, respectively.

    In  the  control  females  the pregnancy rate was  35/35;  the pregnancy rate
for the  female  groups  exposed  in  utero  was  23/35,  0/55 and 0/20 for  the 10,
0 and  160 mg/kg dose  groups,  respectively.   In this study a 10 mg/kg dose Is
the LOAEL with  no NOAEL for a reproductive effects  In mice.

    To  examine  embryotoxldty,  fetal  Swiss  mice were  mlcrolnjected  In  utero
on  day 10, 12  or  14  of  gestation with benzo[a]pyrene,  benzo[a]pyrene-4,5-
oxlde,   a    racemlc    mixture   of    7B,8a-d1hydroxy-9o,lOa-epoxy-7,8,9,lO-
tetrahydrobenzo[a]pyrene (compounds  thought  to  be carcinogenic  metabolites
of  benzo[a]pyrene)  or  6-methylbenzo[a]pyrene  (Barb1er1  et al.,  1986).   The
vehicle was 1:1  tr1octanon1n:acetone and the  mice were administered  the test
compounds at  doses  of  0.4-16  nmol/embryo.   An additional group  was  treated
transplacentally with  47.5 nmole benzo[a]pyrene  on  day 10,  12 or  14*  On  day
18, the  dams  were sacrificed  and  Inspected for Implantation  sites,  as well
as  the  number of dead  and  live  fetuses; live fetuses were  further  examined
for gross  malformations.   Results,  given  1n  Table V-23,  show  that  1ntra-
embryonal administration of  benzo[a]pyrene  and  benzo[a]pyrene-4,5-ox1de  did
not significantly  Increase the  Incidence  of  fetal malformations;  however,
the 7,8-d1ol-9,lO-epox1de  caused  significant  Increases  In the Incidence  of
fetal    malformations.    Specific   malformations    Included   exencephaly,
thoracoschlsls,  gastroschlsls, phocomella and  edema.   6-Methylbenzo[a]pyrene
also  Increased   the  Incidence of  terata;  transplacental  administration  of
benzo[a]pyrene,  however, did  not.


04420                                 V-116                            10/24/91

-------
                                  TABLE  V-23
           EmbryotoxIcHy and Malformations 1n Swiss Mice Exposed by
          Intraembryonal Injection of Benzo[a]pyrene and Derivatives3
Treatment
Compound
Control5
Benzo[a]pyrene
Benzo[a]pyrene-4,5-
oxlde
Benzo[a]pyrene-7,8-
d1ol-9,lO-epox1de
6-Methy1benzo[a]pyrene

Day
10
12
14
10
12
14
10
12
14
10
12
14
10
12
14
Dose
nmole/Embryo
0.5 (yl)
1 (P*)
2 (vl)
4
8
16
4
8
16
0.4
2
4
4
8
16
Viability Malformations
No. Live/ No. Malformed/
No. Treated Total No. Alive
53/69
30/36
59/82
18/41C
42/78c
28/45
25/28
21/43C
51/66 [sic]
7/46d
59/1 llc
37/58
20/366
13/27C
39/65
11/53
2/30
4/59
6/18
6/42
4/28
8/25
4/21
5/51
7/7d
36/59
-------
     LeGraverand et al.  (1984) found the extent of embryotoxlclty and terato-
 genlclty of benzo[a]pyrene to be related both to the route of administration
 and to  the affinity  of  a  receptor  (AHH)  for  agents  that  Induce hepatic
 cytochrome  P-450  response.   Benzo[a]pyrene  (120  mg/kg/day)  was  orally
 administered to pregnant  mice  from day 2 to  day 10  of  gestation.  The mice
 were either  heterozygous or  homozygous for  the low-affinity  receptor  for
 Inducers  of cytochrome   P-450  (the  high-affinity  receptor   1s  dominant).
 Pharmacoklnetlc studies  Indicated that embryos developing 1n dams homozygous
 for  the   Iow-aff1n1nty   receptors   were   exposed   to   higher   levels   of
 benzo[a]pyrene.  Within this  group of embryos, receptor  homozygotes  had an
 Increased Incidence of  IntrauteMne  toxlclty  and  malformation when compared
 with receptor   heterozygotes.  Heterozygous  dams  exhibit an enhanced ability
 to metabolize  benzo[a]pyrene  In  their Intestines and  liver;  therefore,  the
 developing  embryos   receive   less   benzo[a]pyrene.    Consequently,   less
 Intrauterlne toxldty  and  fewer  malformations  were  seen.   Intraper1ton1al
 administration   of  benzo[a]pyrene  caused    the  exact  opposite  effects,
.Indicating the Importance of  both the administration route and  genotype of
 the exposed animal  1n the  potential for a toxic response.

     Swartz and Mattlson (1985) showed that  acute  exposure to benzo{a]pyrene
 had an adverse, albeit  transient,  effect on  oocyte  follicle  growth,  ovula-
 tlon and  formation of  corpora  lutea.  Five female  C57B1/6N  mice/group were
 treated  with a single  1.p. Injection of 0,  1, 5, 10, 50,  100 or 500 mg/kg bw
 benzo[a]pyrene  In  corn  oil.   One  week post-treatment numbers  of corpora
 lutea  were significantly reduced In mice treated with  5-500  mg/kg benzo[a]-
 pyrene.   By week 4 this  depression was evident only In the mice  In the two
 04420                                V-118                           10/08/91

-------
highest  treatment  groups.   Ovarian  toxldty  1n the  highest  dose group  was
Indicated not only by absences of  corpora  lutea  but  also  by lack  of  signs of
folllculogenesls.

    Taklzawa  et   al.   (1984)   established  ED5Q  doses  for   small   oocyte
destruction  by  benzo[a]pyrene 1n  several  mouse  strains.   Treatment was  by
1ntraovar1an  Injection  of  0,  0.1, 1.0,  10 or 30 yg/ovary.  Two  weeks  post-
treatment mice  were  killed and ovarian  tissue examined h1stolog1cally.   The
ED5Qs  for  various  strains  were  the   following:   C57B1/6N,  ED5Q  =  3.38
vg/ovary; DBA/2N = 36.14; C57B1/6J x  DBA/2JF1  = 8.27.

    Pregnant C3H/Anf mice were treated 1.p. at either  11-13 or  16-18 days of
gestation with  100 or  150  mg benzo[a]pyrene/kg bw.   Body weights of progeny
were  comparable between control  and treated  groups.   Offspring  of  treated
animals,  however,  were observed  to  have a severe  supresslon  of  Immune
function as  measured by the  ability  to  elicit an antibody  response  to  sheep
red  blood  cells.   This  supresslon  of  response  persisted  Into  adulthood
Indicating  permanent  1mmunosuppress1on  when  prenatal  exposures  occur  (Urso
and Gengozlan, 1980).

    Benzo[a]pyrene  has  been   shown  to  be  a   transplacental   carcinogen
producing liver tumors  In HA/ICR,  strain A and  C57B1  mice and lung tumors 1n
the Ha/ICR mice (Bulay, 1970;  Bulay and Wattenberg 1971; Nlkonova. 1977).

    Female  A strain  mice were  subcutaneously administered 150 mg/bw benzo-
[a]pyrene 1n  sunflower  oil on  the  18th and 19th  day  of pregnancy (Turusov et
al.,  1990).   The  offspring were mated 1n  a  brother-sister  system to create
04420                                V-119                           10/08/91

-------
the   F2-F5   generations.    Controls   receiving   only  sunflower   oil   were
maintained.  The  offspring  were  sacrificed at 1 year  of  age.   When compared
with  controls,  there was  a statistically significant  Increase 1n  the  lung
adenoma  Incidence 1n  both  the  males  and females  of the  first  generation.
The lung  tumor  Incidences  1n Fl  females were 42/78  and 12/78  1n  the exposed
and control  mice, respectively, and  1n Fl  males  the Incidences  were  52/67
and  6/75 1n  the  exposed  and control  mice, respectively.   The   F2  females
exhibited  a  statistically   significant  Increase  1n  lung  tumor  Incidences
also; these were  20/91 and  10/95 1n  the female  F2  descendants  of  the treated
dams  and  1n the  controls,  respectively.  This  Increased Incidence  of  lung
tumors  was  not  seen  1n   the   F2  males  or  1n  later  generations,  but  a
statistically  significant   Increase  1n  tumor  multiplicity was  seen In  the
F2-F5 generations.

    Rabbit   progeny   have   also  been   shown  to   develop   tumors   after
transplacental  exposure   to  benzo[a]pyrene  during  the  last   third   of
preganancy (BenlashvlH, 1978).
      f
    Chrysene.   Mallard  duck  eggs  were  painted   with   10  yS,  of  a   0.1X
solution of chrysene  1n  an  unspecified petroleum hydrocarbon  reported  to be
of "relatively  low embryotox1c1ty"  (IARC,  1983).   Embryotoxlclty  and terato-
genlc effects among ducklings were  observed (Hoffman and Gay,  1981).

    Fluoranthene.   Irvln  and  Martin  (1987)  reported,   1n an  abstract,  a
developmental study  1n  which a  single  1.p.  Injection of  fluoranthene  (dose
unspecified) was  given  to  pregnant C57/B6 mice  on gestatlonal days  6.  7,  8
or  9.   A  gestation-dependent   Increased  rate  of  embryo  resorptlon  was
observed.

04420                                V-120                           10/08/91

-------
    Naphthalene.   In what  appears  to  be  two  reports  of  the  same  study
(Plasterer et al.,  1985;  Booth  et al., 1983), single oral doses  {300  mg/kg)
of naphthalene were administered dally for 8  consecutive days  to  50  pregnant
CD-I mice beginning on day 7 of gestation.  This  dose was  estimated  to be  at
or just below the maximum tolerated dose  for  acute  lethality.   A  significant
Increase In maternal lethality (p<0.05) and a decrease In  mean maternal  body
weights as well as  the number of  live  pups per  Utter  (p<0.05) on postpartum
day 1 were  noted  when  compared  with the controls.  There  was  not  a  concomi-
tant Increase 1n  dead  pups.   There were no effects on pup survival  and  mean
body weights.   No gross  congenital abnormalities were detected In  the  pups
although the method used  to  examine the pups was  not reported.

    Hardln et al.  (1981)  administered naphthalene  1.p.  (395 mg/kg)  In  corn
oil to pregnant Sprague-Dawley rats on day 1  of  gestation.  Dally Injections
continued  through   day   15.    No  treatment-related  effects  on  maternal
toxlclty, fetal  toxIcHy,  or  teratogenesls were  reported.
                                                                            •
       /•
    In a  pilot  range-finding  study, 20 artificially Inseminated  New Zealand
white rabbits  (at  least  24  weeks of  age and weighing  4-5  kg) were  orally
dosed with naphthalene (In IX methylcellulose vehicle) at  50-1000 mg/kg from
gestatlonal days 6-18.  Maternal  lethality and/or abortion were Increased  at
doses  of  >630  mg/kg,  but  no   data  were  collected.    No  differences  1n
reproductive  parameters  were  noted,   and no  malformations or fetal  death
occurred at the lower dose levels  (Na1smith and  Matthews,  1985).

    In  the main  study by  Nalsmlth  and  Matthews   (1986),  18 artificially
Inseminated  New  Zealand  white rabbits  per  group  were   orally  dosed  with
naphthalene (1% methylcellulose vehicle) at 0, 40,  200 or  400  mg/kg/day from

04420                                V-121                           10/08/91

-------
gestation  days  6-18.   Dams were  at  least 24 weeks  of  age (exact ages  were
not specified).  Maternal body weights and body  weight  gains  were comparable
among  all  test  groups and  controls.   Food  consumption  of  high-dose  (400
mg/kg)  animals  was  significantly   greater   (p<0.05)  than  controls  during
gestation  days  7-15,  23-25 and  27-29.   Pharmacotoxlc signs  observed  during
the  study  Included  decreased   activity,  dyspnea,  weight  loss,  cyanosis,
salivation,  and  loose  stools   or   diarrhea;  these  signs  occurred  In  an
apparent  dose-related  manner.    Gross   examination  of  dams  and  controls
Indicated  no  differences  1n  the following reproductive measures:  number  of
corpora  lutea,  total  number of  Implantations,  viable or  nonvlable  fetuses,
pre- or   postlmplantatlon   loss,   fetal  body   weights,  and   fetal   sex
distribution.   The malformations  and  variations  were  equally  distributed
among groups, and  no  dose-related trends  were apparent.   These  malformations
Included  2 Incidences  of  fused  sternebrae   and  an  Incidence  of  umbilical
hernlatlon from  3 different Utters 1n  the  control  group, an Incidence  of
vlceral  malformations  1n  the  mid-dose  group   and  3  Incidences  of  fused
                                          - •                                  •
sternebrae  from 3  different  Utters 1n  the  high-dose   group.   The  study
authors  concluded  that  oral   administration  of   naphthalene  to  pregnant
rabbits did not  evoke a teratogenlc  effect.   The U.S.  EPA (1987a)  concluded
that the  teratogenlc  potential  could not  be  adequately assessed.  The  data
were considered  Incomplete because of lack of Information  on the methods  of
fetal sacrifice and of visceral  and  skeletal  examinations.

    Matorova (1982) reported that naphthalene administered by gavage  (0.015,
0.15 and  1.5  mg/kg) on a  chronic basis  (duration not reported)  to  pregnant
female  albino  rats  was  associated  with adverse  effects  on reproductive
function  and  development  of progeny.   These  effects  Included  slow  fetal
development,  an  Increase  In  number   of hemorrhages  and  bleeding  of  Internal

04420                                V-122                           10/08/91

-------
 fetal  organs, and a  reduced  viability.   The  reported  threshold  for effects
 was  0.075  mg/kg.   A  lack  of  Information  on  protocol  design,   tests  for
 significance,  and  experimental  data  renders  these  results  difficult  to
 Interpret.

 flutaqen1c1ty/Genotox1c1tv
     The  PAHs have long  been  recognized as  having  mutagenlc  and other geno-
 toxlc  effects.   The literature contains many reports of short-term Ui vitro
 and  IJN  vivo  tests  of  the  genotoxldty   of  these  compounds  and  of their
 metabolites.   Because  of the  extensive scope of these reports, their  results
 are  presented 1n tabular form  (Table  V-24).

 Synerqlsm and/or Antagonism
     Because PAHs rarely occur  1n Isolation  from each  other  In the environ-
 ment,  1t Is  Important  to  understand  the potential  health effects  associated
 with  mixtures   of  PAHs.  Much  research  has  focused   on  the  promoting or
'Inhibitory  effect  of   noncarclnogenlc  PAHs,  such   as   pyrene,  on  the
 carcinogenic  potential  of  known  carcinogens,  often  benzo[a]pyrene.   The
 route  of administration Is typically  nonoral  (dermal,  subcutaneous  Injection
 or  Inhalation)..  Some  of   the   reports   on  cocardnogenlc  activity  are
 considered In this  section.

     Falk et  al. (1964)  conducted a  series of experiments  with  C57B1  male
 mice  (3-4 months old)  to  assess  the potential Inhibitory effects  of  phenan-
 threne  and  other PAHs  considered noncarclnogenlc on  the tumor 1gen1c1ty of
 d1benz[a,h]anthracene.  Groups of 30  animals  received  single  s.c.  Injections
 of  various  dosages  of  d1benz[a,h]anthracene alone  and 1n combination  with

 04420                                V-123                           10/08/91

-------
                                                                              TABLE V-24



                                                    Results of Short-Term Tests  of Polycycllc Aromatic Hydrocarbons
to
Test
BACTERIA
Mutation



MAMMALIAN CELLS
Cell transformation
BAC1ERIA
DNA damage


Mutation
Organism (Assay)

Salmonella typhlmurlure
(reverse mutation, his)

S. typhlmurlum (reverse
mutation, his) (taped-
plate assay)
S. typhlmurlum (forward
mutation. BAG*)

Syrian hamster embryo cells
(morphologic changes)

Escherlchia coll
(2P_LA)
E. coll
(uvrA. recA. lexA, polA)
Bacillus subtllls
(reel
S. typhlmurlum
(reverse mutation, his)
Exogenous Activation*
' System

Rat liver S9 (Aroclor)
Rat liver S9
(3-methylcholanthrene)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor
or phenobarbltal)

None

Rat liver S9 (none)
Rat liver S9. (Aroclor)
Rat liver S9 (Aroclor)
Rat liver $9 (Aroclor or
3-methylcholanthrene)
Results Reported
ACENAPHTHYLENE
Negative
Negative
Negative
Positive

Negative
ANTHRACENE
Negative .
Negative.
Potency
(A minimal
Inhibitory
concentration/
nmole)' <0.0009
Negative
Negative
Comments

Tested up to 250 pg/mt In
strains TA1537 and TA1538
Tested up to 3 pmol/plate In
strains TA98 and TA100
Tested up to 50 pg/plate In
strains TA98 and TA100
Strain TM677 tested at
1 pmol/mt

1-50 pg/mt

Tested up to 250 pg/mt
Minimal Inhibitory concentra-
tion >100 pg/well In all
strains with and without S9
Tested at 62 pg/well
Tested up to 1000 pg/plate In
strains TA1535. TA1537,
TA1538. TA98 and TA100
Reference

Gatehouse, 1980
Florin et al..
1980
Bos et al..
1988
Kaden et al.,
1979

Tu et al., 1986

Rosenkrantz and
Polrler, 1979
OeFlora et al..
1984
McCarroll
et al.. 1981
McCann et al.,
1975; Simmon,
1979b; La Vole
et al., 1979;
Salamone et
al.. 1979; Ho
et al.. 1981;
Florin et al.,
1980

-------
TABLE V-24 (cont.)
(V) 	
o
Test
Nutation (cont.)




<
«J •
ro


FUNGI
Recombination
MAMMALIAN CELLS
ONA damage
o
t\>
vo
Organism (Assay)
S. typhlmurlum
(reverse mutation, his)





S. typhlmurlum (reverse
mutation, his) (taped-
plate assay)
$_._ typhlmurlum
(forward mutation, 8AG*t)
S. typhlmurlum (forward
mutation. ARA")

Saccharomyces cerevlslae
(mltotlc recombination, 03)

Primary rat hepatocytes
(unscheduled DNA synthesis)
HeLa cells (unscheduled
DNA synthesis)
Exogenous Activation*
System
Hamster or rat liver S9
(Aroclor)
•°Co gamma Irradiation
of anthracene
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor or
phenobarbltol)
Rat liver S9 (Aroclor)

Rat liver S9 (Aroclor)

None
Rat liver S9 (phenobarbltal
or 3-methylcholanthrene)
Results Reported
Positive
Negative
Negative
Negative
Positive
Negative
Negative
Negative
Positive

Negative

Negative
Negative
Comments
In strain TA100 from 50-400
tig/plate
In strains TA98. TA1535,
TA1537 and TA1538
In strain TA1537 at 0.22 mM
Tested 1n TA1535. TA1537.
TA1538, TA95 and TA100
TA97 at 5 and 10 pg/plate
Tested In TA98 and.TAlOO at up
to 200 pg/plate
Tested up to 50 pg/plate In
strains TA98 and TA100
Tested up to 225 nmol/ml
In strain BA9 and 0.5-3 pmol

None

Tested up to 1 pg/ml
Tested up to 100 pg/ml
Reference
Carver et al..
1986
Gibson et al..
1978
Selxas et al..
1982
OeFlora et al.,
1984
Sakal et al.,
1985
La Vole et al..
1985
Bos et al.,
1988
Kaden et al..
1979
Dorado and
Pueyo, 1988

Simmon. 1979a

Williams. 1977;
Probst et al.,
1981
Martin et al.,
1978; Martin
and McDermld,
1981

-------
                                              TABLE V-24  (cont.)
o
iv> Test
o
DNA damage (cont. )


Mutation

rvj


Chromosome effects

o
o
CO
to
Organism (Assay)
F344 rat tracea epithelial
organ culture (unscheduled
DNA synthesis)
Chinese hamster ovary cells
(DNA synthesis)
Primary rat hepatocytes
(unscheduled DNA synthesis)
Chinese hamster V79 cells
(forward mutation. 6TGR)
Mouse lymphoma L5178Y cells
(forward mutation, TFTR)
Human lymphoblastold TK6
cells (forward mutation,
TFTR)
Rat liver epithelial cell line
ARL18 (forward mutation 6TGR)
Chinese hamster V79 cells
(forward mutations; OUAR,
6TGR)
Mouse lymphoma L5178Y cells
(forward mutation. TK-)
Chinese hamster ovary cells
(forward mutation. HGPRT)
Chinese hamster D6 cells
(sister -chromatld exchange;
breaks)
Rat liver epithelial ARL-18
cells (sister -chromatld
Exogenous Activation*
System
None
None <
None
Rat liver S9 (?)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
None
Liver and lung cell mediated
systems for OUAR, kidney
and bladder cell mediated
systems for 6TGR
None
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
None
None
Results Reported
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Positive
Negative
Negative
Negative
Comments
Tested from O.lxlO"* to
10xlO'« M
Tested 1000 pg/ml
Tested at lx!0~» M
, Tested up to 125 pg/ml
Tested up to 100 nmol/mi
Tested up to 200 nmol/mt
Tested at 10'» and 10"«
Tested at 3 pg/mi
Up to 500 pg/mi
Tested at 6-15 pg/ml
Tested at 3.5 pg/ml
Tested up to 1 pmol/mi
Tested up to 1 pmol/mi
Reference
Ide et al..
1981
Garrett and
Lewtas. 1983
Williams
et al.. 1989
Knapp et al.,
1981
Amacher and
Turner, 1980;
Amacher et al.,
1980
Barfknecht
et al.. 1981
Ved Brat
et al.. 1983
Langenbach
et al.. 1983
Mitchell
et al.. 1988
Oshlro et al.,
1988
Abe and Sasaki,
1977
Tong et al..
1981b
exchange)

-------
                                                                             TABLE  V-24 (cont.)
0
CO
to
Test
Chromosome effects
(cont.)




Cell transformation




BACTERIA
ONA damage

Organism (Assay)
Rat liver epithelial cell
line ARL18 (slster-chromatld
exchange)
Chinese hamster bone marrow
cells (111 vivo slster-
chromatld exchange)
Rat hepatocy.tes (ONA single
strand breaks)
Mouse lymphoma LS178Y/TK+/-
cells (ONA strand unwinding
assay)
Human peripheral lymphocytes
(sister -chromatld exchange)
House BALB/3T3 cells
(morphologic changes)
Guinea pig fetal cells
(morphologic changes)
Syrian hamster embryo cells
(morphologic changes)
C3H/10T1/2 cells mouse
embryo Hbroblast
(morphologic changes)
Mouse embryo C3H10T1/2 cells
(morphologic changes)

E. coll (polA)
E. coll
(uyrA. recA. jejcA, polA)
Exogenous Activation* Results Reported
System
None Negative
None . Negative
None Negative
Rat liver S9 (Aroclor) Positive
Intact rat hepatocytes Negative
None Negative
None Negative
None Negative
None Negative
None Negative
None Negative
BENZralANTHRACENE
Rat liver S9 (none) Negative
Rat liver S9 (Aroclor) Negative.
Potency
(A minimal
Inhibitory
concentration/
nmol) >0.0001
Comments
Tested at 10"» and 10~« M
450 mg/kg was Injected 48 and
24 hours prior to test
Tested at 0.003. 0.03 and
0.3 mM
Tested at 4.74x10"* H to
9.48x10"' M
Tested at 10"« to 10'* M
Tested at 10~« to 10~* M
Tested at 10 pg/mi
Tested at 0.5 yg/mi
Tested up to 50 pg/ml
Tested at 4 PH
Tested up to 30 pg/ml

Tested up to 250 pg/ml
Minimal Inhibitory concentra-
tion >1000 pg/well with and
without S9 In all strains
Reference
Ved Brat
et al.. 1983
Roszlnsky-
Kocher et al . ,
1979
Slna et al..
1983
Garberg et al..
1988
Undahl-
Klessllng
et al.. 1989
DIPaolo et al. .
1972
Evans and
DIPaolo. 1975
Plenta et al..
1977
Peterson
et al.. 1981
Lubet et al..
1983b

Rosenkrantz and
Polrler. 1979
Deflora et al..
1984

-------
                                                                           TABLE  V-24 (cont.)
            Test
rs>
oo
o
oo
Organism (Assay)
Exogenous Activation*
       System
Results Reported
Comments
                           threne; pyrene; benzo[k]-
                           fluoranthene; benzo[g,h,1]-
                           perylene; anthanthrene;
                           coronene; chrysene; benzo-
                           [J]fluoranthene; benzo[b]-
                           fluoranthene; Indeno-
                           [l,2,3-cd]pyrene; dlbenz-
                           [a.hjanthracene; clophen
                           ASO; benzo(c)phenanthrene,
                           benzo[b]naphthal[2,3-d]-
                           thlophene; benzo[bj-
                           naphthol[1,2-d]th1ophene;
                           trlphenylene; benzo[b]naph-
                           thol[2.1-djtheophene)
                                                                                            benzo[J]fluor-  •
                                                                                            anthene, clophen
                                                                                            ASO, benzo[b]fluor-
                                                                                            anthene, Indeno'-
                                                                                            [1.2.3-cd)pyrene.
                                                                                            d1benz[a.h]-
                                                                                            anthracene
                                                 by different PAHs.  the weak
                                                 carcinogen benz[a]anthracene
                                                 exhibited mutagenlc activities
                                                 comparable to those of benzo-
                                                 [ajpyrene In strain TA100
Reference
Mutation E. coll (expression of
lac Z gene under control
of SOS gene. sfIA)
S. tvphlmurluni (expression
of lac Z gene under control
of urn! gene)
S. typhlmurluni
{reverse nutation, his)















Rat liver S9 (Aroclor)

Rat liver S9 (Aroclor)


Rat liver S9 (Aroclor)




•
•°Co gamma Irradiation
of benz(a)anthracene


Rat liver S9
(me thy Icholan threne)



Rat liver S9 (f luoranthene;
benzo[e]pyrene; phenan-
Posltlve

Weakly positive


Positive

Negative



Positive

Negative

Positive

Positive


Positive when rats
were treated with
In strain PQ37 up to 100 mM

Tested at 2.4 pg/mi


20 yg/plate In strain TA100

Strain TA1535



In strains TA98. TA1537 and
TA1538 at 250 yg/plate
Tested 1n strain TA1535 up
to 250 pg/plate
In strains TA98 and TA100
up to 3 iimoles/plate
Strains TA97. TA98, TA100.
TA1537, TA1538. Potency
(revertants/nmole)=12
In the presence of various
liver S9 preparations Induced
von der Hude
et al.. 1988

Nakamura
et al., 1987

McCann et al.,
1975; Coombs
et al., 1976;
Simmon, 1979b;
Salamone
et al.. 1979
Gibson et al . ,
1978


Florin et al . ,
1980
DeFlora et al. ,
1984

Norpoth et al. ,
1984

-------
                                                                             TABLE V-24 (cont.)
f\>
o
              Test
                             Organlsn (Assay)
                                   Exogenous Activation*
                                          System
                               Results  Reported
          Comments
   Reference
       Nutation (cont.)
                        S.  typhlmurluro
                        (reverse mutation,  his)
 o
 CD
FUNGI

Recombination


INSECTS

Nutation



NANHALIAN CELLS

ONA damage
                               S. tvphlmurlum (reverse
                               mutation, his) (faped-
                               plate assay)

                               S. tvphlmurlum
                               (forward mutation, 8A6R)
                               S. cerevlslae
                               (mltotlc recombination, 03)
                               Drosophlla melanogaster
                               (lethals, vlslbles. bobbed
                               mutants)
Chinese hamster ovary cells
(ONA synthesis)

Primary rat hepatocytes
(unscheduled DMA synthesis)
                                Rat liver  S9                   Positive
                                (3-methylcholanthrene)

                                Rat hepatocytes  Intact  and      Positive
                                homogenized (Aroclor)
                                                               Rat liver S9 (Aroclor)         Positive
                                                               Rat liver S9 (polychlorl-      Positive
                                                               nated blphenyl or pheno-
                                                               barbltal)

                                                               Hamster liver S9 (Aroclor)     Positive
                                                        Nouse.  rat  or  pig  S9           Negative
                                                        (Aroclor)

                                                        Rat  liver S9  (Aroclor)         Negative
                                                        Rat  liver  S9  (Aroclor          Positive
                                                        or phenobarbltal)
                                Rat liver  S9 (Aroclor)          Negative
                                None                           Positive
None                           Negative
                                                               None                           Positive
                                                   0.07  iimol/plate  In  strain
                                                   TA100

                                                   Suspension  Incubation  TA100
                                                   weaker response  with Intact
                                                   cells

                                                   In  strains  TA98  and TA100
                                                   at  50 pg/plate

                                                   In  strains  TA98  and TA100
                                                   at  SO pg/plate
                                                                                   In strain TA100 at 10-50
                                                                                   wg/plate

                                                                                   In strain TA100 at 10-50
                                                                                   pg/plate

                                                                                   In strains TA9B and TA100
                                                                                   at 50 jig/plate
                                                                                   65 nmol/mt In strain TN677
                                                   None
                                                   Administered  by mlcrolnjectlon
Tested 1000
                                                                                                           100 nmol/mt
                                  Bartsch et al..
                                  1980

                                  Utesch et al.,
                                  1987
                                                                                                                                             Bos et al..
                                                                                                                                             1988

                                                                                                                                             Ito et al..
                                                                                                                                             1988
                                                                                                                                             PhlUpson and
                                                                                                                                             lonnldes, 1989
                                                                                      Bos  et  al..
                                                                                      1988
                                                                                      Kaden  et  al.
                                                                                      1979
                                  Simmon. 1979a
                                  Fahmy and
                                  Fahmy, 1973
Garrett and
Lewtas. 1983

Probst et al.,
1981

-------
                                                                              TABLE  V-24 (cont.)
09
O
o
CD
IO
Test Organism (Assay)
DNA damage (cont.) HeLa cells (unscheduled
ONA synthesis)
Primary F344 rat trachea!
epithelial organ culture
(unscheduled ONA synthesis)
Primary F344 rat hepatocytes
(unscheduled ONA synthesis)
Mutation Chinese hamster V79 cells
(forward mutation, OUA*)
Chinese hamster V79 cells
(forward mutation, 6T6R)
House lymphoma LS178Y cells
(forward mutation, TFT")
Rat liver epithelial cells
(forward mutation. 6T6")
Chromosomal effects Chinese hamster ovary cells
(slster-chromatld exchange)
Simian virus transformed
Chinese hamster ovary cells
(selective DNA amplification)
Bone marrow cells of Long-
Evans rats (chromosomal
aberrations)
Cell transformation .Syrian hamster embryo cells
(morphologic changes)
Mouse prostate C3HG23 cells
(morphologic changes)
Exogenous Activation*
System
Rat liver S9
(3-methylcholanthrene)
None
None
Syrian hamster embryo cell
feeder layer
Rat liver S9
(3-methylcholanthrene)
Various
None
None
None
None
None
None
Results Reported
Positive
Negative
Positive
Positive
Positive
Positive
Negative
Positive
Negative
Negative
Positive
Positive
Comments
Up to 100 pmol/ml
Tested at 0.1x10'* to
10xlO"« M
Tested at lx!0'« M
44 nmol/ml
46 nmol/ml
Small Increase at
40 nmol/ml
Up to 100 nmol/ml
2 tig/ml
Tested at 0.14 pinole
lr\ vivo, rats received 25,
50 or 100 mg/kg
0.1 pg/mi
10 pg/ml

Reference
Martin et al.,
1978
Ide et al..
1981
Williams
et al.. 1989
Slaga et al..
1978
Krahn and
Heldelberger.
1977
Amacher et al..
1980; Amacher
and Turner.
1980
Tong et al. .
1981a'
Pal. 1981
Pool et al..
1989
Ito et al..
1988
Plenta et al. .
1977; OlPaolo
et al., 1969.
1971
Marquardt and
Heldelberger,
1972

-------
                                                                            TABLE  V-24  (cent.)
o
-("• ........
4k
o Test
Cell transformation
(cont.)
Organism (Assay)
House C3H/10T1/2 cells
(morphologic changes)
Exogenous Activation* Results Reported Comments
System
N
None Negative Tested up to 100 nmol/ml
Reference
Nesnow and
Heldelberger .
1976
      BACTERIA

      DNA damage
i
to
      Nutation
GO
vo
E. coll
(uyrA. recA.  lexA. polA)
E. coll  (expression of
lac  Z gene under control
of SOS gene,  sf1A)

S. typhlmurlum  (expression
of lac Z gene under control
of urou C gene)

'S. typhlmurlum
(reverse mutation, his)
Rat liver S9 (Aroclor)
                                                              Rat liver  S9 (Aroclor)
                                                              Rat liver  S9 (Aroclor)
Rat liver S9 (Aroclor)
                                                              •°Co gamma Irradiation
                                                              of benzo(a)pyrene

                                                              •°Co gamma Irradiation
                                                              of benzo(a)pyrene

                                                              Rat liver  S9 (Aroclor)
                                                              Rat liver  S9 (Aroclor)
                                Rat or hamster liver S9
                                (Aroclor)
BENZOfalPYRENE

   Negative.
   Potency
   (A minimal
   Inhibition
   concentration/
   nmole) <0.0002

   Positive
   Positive



   Positive


   Positive


   Negative
                                                               Negative
                                                               Positive
                                                               Positive
                                                                                             Positive
Minimal Inhibitory concentra-
tion >BOO pg/well In all
strains with and without S9
                                                    In  strain  PQ37  up  to 100 mM
                                                    Tested  at  1
In strain TA100 up to
1000 pg/plate

In strains TA98, TA1537 and
TA1S38 from 50-100 wg/plate

In strain TA153S up to 100
ng/plate

Strain TA1535
Strain TA97, TA98. TA100,
TA1537, TA1538.  Potency
(revertants/nmole) = 18S

Tested at 2.5 |ig/plate In
strains TA98 and TA100.
Positive only with metabolic
activation

Tested at 8 nM 1n strain
TA100.  Selenium at nontoxlc
concentrations Inhibited the
mutagenlclty of benzo[a]pyrene
In both the rat liver and
hamster liver systems
DeF lora et al.,
1984
                                                          von der Hude,
                                                          et al., 1988
                                                                                      Nakamura
                                                                                      et  al.,  1987
Andrews et al.,
1978

Gibson et al..
1978
                                                                                      DeFlora  et  al.,
                                                                                      1984
                                                                                      Lofroth et al..
                                                                                      1984
                                                                                      Teel,  1984

-------
                                                                           TABLE V-24 (cont.)
             Test
     Organism (Assay)
   Exogenous Activation*
       ,.  System
Results Reported
                                                                                                                           Comments
Reference
      Nutation (cont.)
      FUNGI

      Recombination
1,    INSECTS
GO
^    Chromosomal effects
      Nutagenldty



      PLANTS

      Chromosome damage


      MAMMALIAN CELLS

      ONA damage


o
•v
o
CD
S. typhlmurlum
("reverse nutation, his)
                              S.  tvphlimirlum
                              (^forward mutation.  BAG")

                              S.  typhlmurlum
                              (forward mutation.  ARAR)
S. cerevlslae
(mltotU recombination, 03)
D. roelanoqaster
(whole chromosome gain)
                              D.  melanogaster (sex
                              linked recessive lethal
                              nutation)
                              Tradescant Ions clone 4430
                              (tetrad analysis)
                              F344 rat trachea epithelial
                              organ culture (unscheduled
                              ONA synthesis)

                              Primary BALB/C mouse
                              epidermal cells benzo[a]-
                              pyrene-ONA adduct formation)
Rat liver S9 (Aroclor)         Positive
                                                             Rat hepatocytes Intact and     Positive
                                                             homogenized  (Aroclor)
                                Rat liver  S9 (Aroclor)          Positive


                                Rat liver  S9 (Aroclor)          Positive
Liver mlcrosomes               Negative
(Aroclor)
None                           Negative
                                None                           Negative
                                None                           Positive
                                None               .            Positive
                                None                           Positive
                     Strains TA97, TA98 and TA?nn
                     positive at 1 tig/plate

                     Suspension Incubation; str
                     TA100; weaker response wll
                     Intact cells

                     In strain TM677 at 4 MM
                                                    In strain BA13 at a minimum
                                                    dose of 0.77 nMol
                     None
                     Larvae and adult flies exposed
                     1n feed.  Oocytes and oogonla
                     examined for chromosome gain

                     Fed at 2500 and 50.000 ppm
                                                       Dorado and
                                                       Pueyo, 1988
                                                                                                                                                   Simmon, 1979a
                                                                                                                                                   Fabian and
                                                                                                                                                   Matoltsy. 1946
                                                                                      Valencia
                                                                                      et al.. 1989
                                                    Minimum effect level 12.6 ppm     Sandhu et al..
                                                                                      1989
                                                    Tested at concentrations that     Ide et al..
                                                    ranged from 0.1x10'* to           1981
                                                    lOxlO'* M

                                                    Increase In benzo[a]pyrene-DNA    Nakayama
                                                    adduct formation with PAH con-    et al., 1984
                                                    centratlons In range of 50-250
                                                    nmol

-------
                                                                            TABLE  V-24  (cont.)
o •
Jfc - -
4k
o Test Organism (Assay) Exogenous Activation* Results Reported Comments
System
OKA damage (cont.) Chinese hamster ovary cells Adult rat hepatocytes Positive
(DMA single strand breaks)
Permeablllzed human dlplold- None Negative
flbroblasts (to which E. coll
DNA polymerase was .added)
(DNA strand breaks)
Primary rat hepatocytes None Positive
(unscheduled ONA synthesis)
Mouse lymphoraa L5178/TK*/- Rat liver S9 (Aroclor) Positive
Dose and exposure time depen-
dent Increases noted at IxlCT4
and 5xlO~» H
Tested at 1 yH for 30 minutes
Tested at lxlO~« M
Tested at 5xlO"« to 50xlO~« M
Reference
Yang et al. ,
19B4
Snyder and
Matheson. 1985
Williams
et al.. 1989
Gar berg et al..
                              cells (ONA strand-unwinding
                              assay)
      Hutatlon
Chinese hamster
cel
6Tfi
                              cells (forward mutation.
                                 Ri
CO
CO
      Chromosomal effects
o
o>
                              Chinese hamster ovary cells
                              (forward mutation. HGPRT)

                              Mouse lymphoraa L517BY
                              (forward mutation. TK~)
Mouse Tymphoma LSI78 cells
(forward mutation. TK~)

Mouse embryo cells
C3H/10T1/2 (ONA single
strand breaks)

Rat liver epithelial cells
RL-12 (s1ster-chromat1d
exchange)
Human peripheral lymphocytes
(slster-chromatld exchange)
                              Chinese hamster epithelial
                              liver cells (slster-
                              chromatld exchange)
Rat liver S9 (Aroclor)         Positive



Rat liver S9 (Aroclor)         Positive


Rat liver S9 (Aroclor)         Positive



Rat liver S9 (Aroclor)         Positive


None                           Negative



None                           Positive




Intact rat hepatocytes         Positive

None                           Positive

None                           Positive
Tested up to 6
                                                                                   Tested at 0.5-4.0
                                                                                   Tested at  5xlO"«  to
                                                                                   10xlO~» mol/l
                                                                                                                 Tested from 3.2-10
Tested up to 10 pg/ml



Tested at 1.6xlO~* M to
1.6x10'' M.  Cell line can
activate procarclnogen to
genotoxlc form

Tested at 10"« to 10"* M

Tested at 10~« to 10"6 M

Tested at 0.11-1 vg/mi
                                                                                                                      1988
Reclo and Hsle.
1984
Oshlro et al..
1988

Wangenhelm and
Bolcsfoldl.
1988

Clay and Cross.
1990

Lubet et al..
1983b
                                                                                                                                                   Murlson et al.
                                                                                                                                                   1984
Llndahl-
Klessllng
et al.. 1989

OeSalvIa
et al.. 1988

-------
                                             TABLE V-24 (cont.)
ro Test
o
Chromosomal effects
(cont.)
Cell transformation
i

BACTERIA
Mutation
i
3
MAMMALIAN CELLS
Mutation
BACTERIA
Mutation


5 BACTERIA
CD
^ Mutation
Organism (Assay)
Simian virus transformed
Chinese hamster ovary cells
(selective ONA amplification)
House BalbC/3T3
(morphologic changes)
C3H/10T1/2 mouse embryo
flbroblast cells
(morphologic changes)
Syrian hamster embryo cells
(morphologic changes)

S. typhlmurluni
(reverse mutation, his)


DPI-3 epithelial cell line
(6-TG)

S. typhlmurlum •
(reverse mutation, his)



S. typhlmurlum
Exogenous Activation* Results Reported
System
None Positive
None Positive
None Positive
None Positive
BENZOfblFLUORANTHENE
Rat liver S9 (Aroclor) Negative
Rat liver S9 (Aroclor) Positive

No data. Negative
BENZOfklFLUORANTHENE
Rat liver S9 (Aroclor) Positive
Rat liver S9 (Aroclor) Positive
Rat liver S9 (Aroclor) Positive
BENZOfq.h.nPERYLENE
Rat liver S9 (Aroclor) Positive
Comments
Tested at 0.01 iimol
At 10 and 20 pg/ml
Tested at 4 pM
Tested at 1-5 pg/ml

Tested up to 100 tig/ml In
strain TA100
At 100 pg/p1ate ID strain
TA100. 7 nmol/plate In strain
TA98, 0.5 pmol In TA100
,
Tested up to 4 pM .'
i
'
Tested up to 100 pg/plate In
strain TA100
5 wg/p1ate 1n strain TA98
From 10-50 pg/plate 1n strain
TA100

At 20 pg/plate In strain TA100
Reference
Pool et al.,
1989
DIPaolo et al..
1972
Peterson
et al.. 1981
Tu et al.. 1986

Nossanda
et al., 1979
LaVole et al..
1979; Hermann.
1981; Amln
et al., 1985b

Dooley et al..
1981

LaVole et al..
1980
Hermann et al.,
1980
Meyand et al..
1988

Andrews et al..
(Yeverse mutation,  his)
1978; Mossanda
et al.. 1979

-------
TABLE V-24 (cont.)
g Test
Nutation (cont.)









MAMMALIAN CELLS
< ONA damage
j^
S BACTERIA
ONA damage






Nutation




_,
o
o
CO
10


Organism (Assay)
S. typhlmurlum
(reverse nutation, his)






S. typhlitmrluro
(forward mutation, 8AGR)

Chinese hamster ovary cells
(DNA synthesis)

E. coll (polA)



S. -typhlmurlum (expression
of lac Z gene under control
of umu C gene)
S. typhlmurlum
(reverse mutation, his)







(


Exogenous Activation*
System
N
Rat liver S9 (Aroclor)

•°Co gamma Irradiation '
of benzo(g,h,1)pyrene
None

Rat liver S9 (Aroclor)

Rat liver S9 (Aroclor)


None noted


Rat liver S9 (none)



Rat liver S9 (Aroclor)


Rat liver S9 (Aroclor-)


Rat liver mlcrosomes
{Aroclor ) ,







Results Reported
Positive

Positive

Negative

Positive

Positive


Positive

CHRYSENE
Negative

Negative

Positive


Positive
*

Positive




Positive

Positive

Comments
In range 0.1-1000 pg/plate In
strain TA98
In strains TA98. TA1537 and
TA1538 at 400 pg/plate
In strain TA1535 at concentra-
tions up to 800 pg/plate
At 10 pg/plate 1n TA97 and TA98

At 72 nmol/ml In strain TM677


At 1000 pg/ml DNA synthesis
at 3X of controls

Tested up to 250 pg/mi

None

Tested at 15 pg/mi


At 10 pg/plate In strain TA100


At 125 nmol/plate In strain
TA100; dose-response seen
with Increasing concentrations
of mlcrosomes

None

None

Reference
SaTamone
et al.. 1979
Gibson et al . .
1978


Sakal et al..
1985
Kaden et al . ,
1979

Garrett and
Lewtas, 1983

Rosenkrantz and
PolMer, 1979
Lelfer et al..
1981
Nakamura
et al.. 1987

McCann et al..
1975; LaVole
et al.. 1979
Wood et al..
1977



Dunkel and
Simmon, 1980
Toklwa et al. ,
1977

-------
                                             TABLE V-24 (cont.)
4k
o Test
Nutation (cont.)




«c
i
— ,j
CO




^
o FUNGI
w
^v
<£> Recombination
Organism (Assay) Exogenous Activation*
System
S. typhlmurlum Rat liver S9
Treverse nutation, his) (3-methykholanthrene)

Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Hamster and rat S9
(Aroclor)
None
Rat liver S9 (Aroclor)
.Rat liver S9 (Aroclor)
•°Co gamma Irradiation
In air

S. typhlrourlum (reverse Rat liver S9 (Aroclor)
nutation, his) (taped-
plate assay)
S. typhlmurluro Rat liver .59 (Aroclor)
(forward mutation, 8AGR)

S. .cerevlslae Rat liver S9 (Aroclor)
Results Reported
r
Positive
Negative
Positive
Positive
Positive
Negative
Positive
Negative
Positive
Positive
Negative
Negative
Positive

Negative
Comments
At 1.5 (jmol/plate 1n strain
TA100
At 1.5 (jmol/plate In strain
TA98
Identified In highly mutagenlc
fraction of raw gas condensate.
Nutagenlclty assessed 1n
strain TA98
At 5 vg/plate, strains TA97.
TA98 and TA100
In strain TA100 at 5 ng/plate
In strains TA98 and TA100
In strain TA100 at 100 pg/plate
In strain TA98 at 100 »g/plate
In strains TA98 and TA100 at
50 pg/plate
In strains TA98. TA1537 and
TA1538 at 1000 ug/plate
In strain TA1535 up to 1000
vg/plate
In strains TA98 and TA100 at
50 vg/plate
At 45 nmol/ml In strain TN677

None
Reference
Florin et al. .
1980

Benson et* al. ,
1984
Sakal et al.,
1985
Carver et al.,
1986
Glatt et al.,
1986
Glatt et al.,
1986
Bos et al..
1988
Gibson et al..
1978

Bos et al.,
1988
Kaden et al..
1979

Simmon, 1979a
(mltotlc recombination,  03)

-------
                                                                                TABLE V-24 (cent.)
ro
O
CO
oo

ID
Test
HAHHALIAN CELLS
DNA damage


Nutation
•
Chromosomal effects



Cell transformation


Organism (Assay)

Primary rat hepatocytes
(unscheduled DNA synthesis)
F344 rat trachea epithelial
organ culture (unscheduled
DNA synthesis)
Primary rat hepatocytes
(unscheduled ONA synthesis)
Chinese hamster V79 cells
(forward mutation OUA",
BAG")
V-79. Chinese hamster cells
(forward mutation. 6-TGR)
Chinese hamster cells
(sister chromatld exchange)
House oocytes
(chromosome aberrations)
Hamster spermatogonla
(chromosome aberrations)
Hamster bone marrow cells
Syrian hamster embryo cells
(morphologic changes)
House prostate C3HG23 cells
(morphologic changes)
House prostate H2 cells
(morphologic changes)
Exogenous Activation*
System
V

None
None
None
Syrian hamster embryo cell
feeder layer
Rat liver S9
None
in vivo (gavage)
in vivo (gavage)
(phenobarbltal)
In vivo (gavage)
(phenobarbltal)
In vivo (gavage)
(phenobarbltal)
None
None
Tested both with and without
rat liver
Results Reported

Negative
Negative
Negative
Negative
Negative
Negative
Positive
Positive
Positive
Negative
Positive
Negative
Negative
Comments

Tested up to 100 nmol/ml
I
Tested from 0.1x10"* to
10xlO~« H
Tested at concentrations up
to lxlO"« H
Tested up to 10 wg/mi
Tested from 2.5-80 vq/mt
Tested from 2.5-80 jig/rot
900 mg/kg
Weak positive at 450 mg/kg
Increase not significant at
total dose of 900 mg/kg

Tested at 10pg/mt
Tested up to 10 yg/mi
Tested from 10-40 ug/mt
Reference

Tong et al..
1981a
Ide et al..
1981
Williams
et al.. 1989
Huberman and
Sachs. 1976
Glatt et al..
1986
Roszlnsky-
Kocher et al..
1979
Easier et al..
1977


Plenta et al. .
1977
Marquardt and
Heldelberger.
1972
Glatt et al..
1986

-------
                                             TABU  V-24  (cont.)
2 Test
o
BACTERIA
ONA damage

Nutation


i
CO
CO

PLANTS
Chromosome damage
HAHHALIAN CELLS
ONA damage
o
V.
INJ
V.
• f*
Organism (Assay)
Exogenous Activation*
System
Results Reported
Comments
Reference
DIBENZfa.hlANTHRACENE
E. coll (recA) .
B. subtllls (recA)
S. typhlrourlum
(reverse mutation, his)

•

S_._ typhlmurlum
(forward mutation. 8AGR)

Tradescantla clone 4430
(tetrad analysis)

Human foreskin epithelial
cells (unscheduled DNA
synthesis)
Hela cells (unscheduled
DNA synthesis)
Syrian hamster embryo cells
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat or guinea pig liver
(Aroclor) or (3-methyl-
cholanthrene) rat liver S9
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
•°Co gamma Irradiation
of d1benz[a,h]anthracene
Rat. hamster, pig or human
S9 (Aroclor)
Rat liver S9 (Aroclor)
•
None

None
3-methylcholanthrene
None
Positive
Positive
Positive
Positive
Positive
Negative
Positive
Positive

Positive
B
Positive
Positive
Negative
At 25 pg/well
Minimal Inhibitory concentra-
tion for rec* 50 pg/well;
for rec- 12 pg/well
At 5 pg/plate 1n strain TA100
At 12 nmol/plate In strain TA98
In strain TA100 up to 1000
pg/plate
In strains TA98. TA1535 and
TA1537
In strain TA100 at 10-50
pg/plate
At 75 nmol/mt In strain TM677

Minimum effective concentration
was 12.5 ppm

In range 1-100 pg/mt
At 100 pmol/ml
Tested up to 20 pg/ml
Ichlnotsubo
et al., 1977
McCarroll
et al.. 1981
McCann et al.,
1975; Andrews
et al., 1978
Baker et al.,
1980; Hermann,
1981
Andrews et al. ,
1978
Gibson and
Smith. 1979
Phlllpson and
lonnldes. 1989
Kaden et al..
1979

Sandhu et al..
1989

Lake et al..
1978
Martin et al..
1978
Casto. 1979
(unscheduled  DNA  synthesis)

-------
                                                                              TABLE V-24 (cont.)
CO
to
o
oo
Test
DMA damage (cont.)




Nutation





Cell transformation















BACTERIA
DMA damage




Organism (Assay)
Primary rat hepatocytes
•{unscheduled ONA synthesis)
F344 rat trachea epithelial
organ culture (unscheduled
ONA synthesis)
Chinese hamster V79 cells
(forward mutation, OUA",
8A6«)
Chinese hamster V79 cells
(forward mutation. 6T&")

Syrian hamster embryo cells
(morphologic changes)




Mouse C3H10T1/2 cells
(morphologic changes)


Nouse prostate C3H cells
(morphologic changes)

Nouse prostate C3HG23 cells
(morphologic changes)


B. subtllls (rec A)


S. typhlmurlum (expression
of lac Z gene under control
Exogenous Activation*
v System
None

None


Syrian hamster embryo cell
feeder layer

Rat liver S9
(3-methylcholanthrene)

None





None

None

None


None



With or without rat
liver S9 (polychloMnated
blphenyl)
Rat liver S9 (Aroclor)

Results Reported Comments
Negative Tested up to 100 nmol/mi

Negative Tested from 0.1x10'* to
10xlO'« N

Positive At 1 tig/ml


Positive At 56 nmol/mt

.
Positive At 0.5-10 ng/ml





Positive At 20 pg/mt

Weakly positive At 10 and 30 pg/mi

Positive At 10 iig/mi


Negative . Tested up to 10 wg/ml


FLUORANTHENE
Negative At 10 rag/well


Negative Tested up to 167 ug/ml

Reference
Probst et al.,
1981
Ide et al..
19B1

Huberman and
Sachs. 1976;
Huberman, 1978
Krahn and
Heldelberger,
1977
OlPaolo et al..
1969; Plenta
et al.. 1977;
Casto et al..
1977; Casto.
1979
Reznlkoff
et al.. 1973
Lubet et al..
1983b
Chen and
Heldelberger .
1969
Narquardt and
Heldelberger.
1972

Klnae et al..
1981

Nakamura
et al.. 1987
                               of  umu  C  gene)

-------
                                                                             TABLE  V-24  (corit.)
o
4fc


O
              Test
     Organlsn (Assay)
   Exogenous Activation*
          System
Results Reported
                               Comments
                                    Reference
       Nutation
L. typhlmurlum
(reverse mutation,  his)
I
wj
*•.
o
                               S. typhlmurlum  •
                               (reverse mutation,  his)
o
CD
                               S. typh1mur1um
                               (reverse mutation, his)
                               (taped-plate assay)
S. typhlmurlum
^forward mutation, 8AGR.)
Rat liver S9 (Aroclor)
                                                               Rat  liver  S9  (Aroclor)
                                                               Rat  liver  S9  (Aroclor)
                                                               Rat  liver  S9  (Aroclor)
                                                               Rat  liver  S9  (Aroclor)
                                With or  without  rat
                                liver S9 (polychlorlnated
                                blphenyl)

                                With or  without  rat
                                liver S9 (polychlorlnated
                                blphenyl)

                                Rat llvor S9 (Aroclor)
                                Rat liver S9
                                (3-methylcholanthrene)

                                Rat liver S9 (Aroclor)
                                                               Rat liver  S9  (Aroclor)
                                                               Rat liver  S9  (Aroclor)
Negative



Negative


Weakly positive


Negative



Positive


Negative



Positive



Positive


Negative


Positive


Positive




Positive
Unknown concentration In
airborne partlculate materials
(strain TA98)

Strains TA98 and TA100
                                                                                   Unknown concentration  In
                                                                                   strain TA9B

                                                                                   Tested up  to  1000  pg/plate  In
                                                                                   strains TA98.  TA100, TA1535.
                                                                                   TA1537 and TA1538

                                                                                   At  5  pg/plate  In strain TA98
                                                    Strains TA98 and TA1537  at
                                                    100 pg/plate
                                                    Tested up to 200 pg/plate 1n
                                                    strain TA100

                                                    Tested In strains TA98 and
                                                    TA100 at 3 ymol/plate

                                                    Tested up to 500 nmol/mt
                                                    In strain TA97

                                                    Tested up to 100 pg/bottom
                                                    plate In strains TA98 and
                                                    TA100.  Results negative with-
                                                    out the addition of S9 mix.

                                                    Lowest concentration to show a
                                                    significant mutation rate was
                                                    5 pM.  Not mutagenlc at up to
                                                    50 pH without the addition of
                                                    S9 mix.  Strain TM677
                                                      Toklwa et al. .
                                                      1977
                                                                                     LaVole et al.,
                                                                                     1979

                                                                                     Epler et al..
                                                                                     1978

                                                                                     Salamone et al.,
                                                                                     1979
                                                      Hermann et al.,
                                                      I960

                                                      K1nae et al.,
                                                      1981
                                                    Strain TA100 at 100 ng/plate     Klnae et al.,
                                                                                     1981  .
                                                      LaVole et al..
                                                      1982b

                                                      Florin et al.,
                                                      1980

                                                      Hera and
                                                      Pueyo. 1988

                                                      Bos et al..
                                                      1988
                                                      Rastetter
                                                      et al., 1982

-------
                                                                        TABLE V-24 (cent.)
         Test
     Organism (Assay)
   Exogenous Activation*
          System
Results Reported
          Comments
   Reference
  Hutatlon  (cont.)
  MAMMALIAN CELLS

  Hutatlon
  Chromosomal effects





  BACTERIA

  DNA damage



  Nutation
S. typhlmurluro
(forward mutation, BAG")
                          S.  typhlmurluro
                          (forward mutation. Ara")
Human lytnphoblast cell line
AHH-1 (forward mutation to
6GT«)

Chinese hamster ovary cells
(differential cytotoxldty
of repair-deficient strains)
Chinese hamster ovary cell
(forward mutation 6GTR)

Chinese hamster ovary cells
(sister chromat1d exchange)

Chinese hamster epithelial
cell strain
S. typhlmurlum (expression
of lac Z gene under control
of umu C gene)

S. typhlrourlum
(reverse mutation, his)
Rat liver S9 (Aroclor)
                                                          Rat liver 59 (Aroclor)
                                Rat liver S9 (Aroclor)
MFO Induced by treatment
with B-naphthaflavone
                                                          Rat liver S9 (Aroclor)
                                                          Rat liver S9 (Aroclor)
                                                          Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
None (using metabollcally
competent liver cells)
Rat liver S9 (Aroclor)
•°Co gamma Irradiation
of fluorene
o
o
CO
Positive


Positive


Positive




Negative



Positive



Negative


Weakly positive


Positive


Negative


FLUOREME

Negative



Positive       •'


Negative
Strain TN677 at 1 pg/mt           Kaden et al.,
                                  1979

Tested up to 10 gg/mt In          Babson et al.,
strain TM677                      1966

Tested up to 500 nmol/mt          Hera and
In strain BA9                     Pueyo, 1988
Tested at 10-100 iiN, 48-hour      Crespl and
exposure                          Thllly. 1984
                                                    Strains  UV4  and UV5 deficient      Hoy  et  al..
                                                    In  nucleotlde excision repair      1984
                                                    Strain  EM9 defective 1n DNA-
                                                    strand-break rejoining

                                                    Tested  at 20 wg/mt                LI.  1984
Tested at 9. 18 and 36 Mg/ml      Pallttl et al..
                                  1986

Tested at 25, 35 and 45 wg/mt     DeSalvIa
                                  et al.. 1988
Tested up to 16.7
In strains TA98 and TA1538 up
to 1000 yg/plate

In strains TA1535 and TA1537
up to 1000 pg/plate
Nakamura
et al'., 1987
Gibson et al..
1978

-------
                                                                            TABLE  V-24  (cont.)
ro
             Test
     Organism (Assay)
   Exogenous Activation*
          System
Results Reported
Comments
Reference
      Mutation (cont.)
S. tvphlmurluro
(reverse mutation, his)
      MAMMALIAN CELLS

      DMA damage
      Mutation
                              S. typhlmurluni (reverse
                              mutation, his) (taped-
                              plate assay)

                              S. typhlmurlum
                              (forward mutation, 8AGR)
Primary rat hepatocytes
(unscheduled DMA synthesis)

Primary mouse or hamster
hepatocytes (unscheduled
DNA synthesis)

Permeablllzed human dlplold
flbroblasts and E. coll DNA
polymerase (DNA strand breaks)

Primary rat hepatocytes
(unscheduled DNA synthesis)

Mouse L5178Y TK+/- lymphoma
cells (forward mutation, TK~)
                              Chinese hamster ovary cells
                              (forward mutation, HGPRT)
Rat liver S9 (Aroclor)          Negative
                                                              Rat  liver  S9  (Aroclor)          Negative
                                                              Rat  liver S9  (3-MC  or  TCDD)     Negative
                                                              Rat  liver  S9  (Aroclor)          Negative
                                Rat liver  S9 (Aroclor)          Negative
                                Rat liver  S9 (Aroclor  or        Negative
                                phenobarbltal)
None                           Negative
                                                             •None                           Negative
                                                              None                           Negative
                                                              None                           Negative
Rat liver S9 (Aroclor)         Negative

None                           Negative

Rat liver S9 (Aroclor)         Positive
                                Rat liver S9 (Aroclor)          Negative
                     Tested up to 1000 |ig/plate In
                     strains TA1535, TA1537, TA98
                     and TA100
                                                                                    Tested  at  up to 250 ug/plate
                                                                                    strains TA97.  TA98 and TA100

                                                                                    Tested  up  to 1504 nmol/plate
                                                                                    In  strain  TA100

                                                                                    In  strains TA98 and TA100 up
                                                                                    to  50
                                                    In strains TA98 and TA100 up
                                                    to 50 |ig/plate
                                                    Tested up to 300 nmol/ml In
                                                    strain TM677
                     Tested up to 10 nmol/ml


                     Tested up to 1 pM



                     Tested at 3 mM for 2 hours



                     Tested at IxlO'4 M


                     Tested up to 30 iig/mi

                     Tested up to 60 pg/ml

                     Tested at 5.8x10"" to
                     7.78xlO~« mol/l


                     Tested, at 1-20
                        McCann et al..
                        1975; La Vole
                        et al.. 1979.
                        1981b

                        Sakal et al.,
                        1985

                        Pahlman and
                        Pelkonen. 1987

                        Bos et al..
                        1988
                                                       Kaden et al.,
                                                       1979
                        Probst et al.,
                        1981

                        McQueen et al.,
                        1983
                                                                                      Snyder and
                                                                                      Matheson. 1985
                        Williams
                        et al.. 1989

                        Oberly et al.,
                        1984
                        Wangenhelm and
                        Bolcsfoldl,
                        1988

                        Oshlro et al.,
                        1988

-------
                                                                            TABLE V-24 (cont.)
              Test
     Organism (Assay)
   Exogenous Activation*
          System
                               Results Reported
          Comments
   Reference
      CHROMOSOME  EFFECTS
       BACTERIA

       Mutation
       BACTERIA

       ONA  damage



       Mutation
o
CD
Mouse lymphoma cells (DMA
strand breaks)

Chinese hamster cell line
(chromosomal aberrations)
S. typhlmurlum
(reverse mutation, his)
Rat liver S9 (Aroclor)
                               Positive
S. typhlmurlum (expression
of lac Z gene under control
of umu C gene)

S. typhlmurlum
(reverse mutation, his)
Liver S9 mix source unknown    Positive


                        INDENOfl.2.3-cd1PYRENE

Rat liver S9 (Aroclor)         Positive
                                                              Rat  liver S9  (Aroclor)
Rat liver S9 (Aroclor)
                               Positive


                             NAPHTHALENE

                               Negative
With or without rat liver S9   Negative
(Aroclor)
                                                              With or without rat liver S9   Negative
                                                              Rat  liver S9                   Negative
                                                              (3-methylcholanthrene)

                                                              Rat  liver S9  (Aroclor)         Negative
                                                              With or without rat liver S9   Negative
                                                              (Aroclor)
                                Rat or hamster S9
                                (Aroclor)

                                Rat liver S9 (20-100
                                lit/plate) (Aroclor)
                               Negative


                               Negative
At 0.15 MM both with and
without metabolic activation.

25 pg/mi
                                                    At  20  |ig/plate  for  TA100 and
                                                    2 pg/plate for  TA9B

                                                    At  -3  pg/plate  TA100
Tested at 6.3 pg/ml
                                                    Tested up to  1000 pg/plate  In
                                                    strains TA98.  TA100.  TA1535
                                                    and TA1537

                                                    Tested up to  300' pg/plate In
                                                    strains TA1535.  TA1537.  TA1538.
                                                    TA9B.  TA100

                                                    Tested at 3 pinol/plate In
                                                    strains TA98  and TA100

                                                    Tested up to  1.6 mN 1n
                                                    strain TA1537

                                                    tested In two  DNA-repalr defi-
                                                    cient  strains,  TA100  and TA98.
                                                    and two strains  which have
                                                    full DNA repair  capacity:
                                                    UTH8414 and UTH8413

                                                    Tested up to  10  ing/plate In
                                                    strain TA100

                                                    Tested up to  1000 pg/plate
                                                    1n  TA98
Garberg et al..
1988

Hatsuoka
et al., 1987
                                  LaVole et al..
                                  1979; Hermann
                                  et al.. 1980
                                  Rice et al..
                                  1985a
Nakamura
et al.. 1987
                                  McCann et al.
                                  1975
                                                                                                                      Godek et  al.,
                                                                                                                      1985
                                                                                                                      Florin et  al..
                                                                                                                      1980

                                                                                                                      Selxas et  al..
                                                                                                                      1982

                                                                                                                      Connor et  al..
                                                                                                                      1985
                                  Mortelmans
                                  et al.. 1986

                                  Narbonne
                                  et al.. 1987

-------
                                                                              TABLE  V-24 (cortt.)
Test
Nutation (cont. )
Organism (Assay)
S. typhlnurlum
(reverse nutation, his)
Exogenous Activation*
System
Rat liver S9 (20-100
til/plate) (Aroclor)
Results Reported
Positive
Comments
Tested up to 1000 ng/plate
In TA1S35. Naphthalene gave
Reference
Narbonne
et al.. 1987
        MAMMALIAN CELLS

        DNA damage
ro

10
                                S. typhlmuMuni
                                (reverse mutation, his)
                                (taped-plate assay)
         Chromosomal  effects
Rat hepatocytes
(unscheduled ONA synthesis)
Bone marrow of CD-I mice
(mlcronuclel Induction)

Chinese hamster ovary cells
(slster-chromatld exchange)
                                                                With or without rat liver
                                                                S9 (polychlorlnated
                                                                blphenyls)

                                                                Rat liver S9 (Aroclor)
                                                                With or without rat liver
                                                                S9 (Aroclor)

                                                                Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)




Rat liver S9 (Aroclor)



Rat liver S9 (Aroclor)




None



None


Rat liver S9
Negative



Weakly positive




Negative


Negative


Negative




Positive



Negative




Negative



Negative


Positive
                                                                                   >2 times the control value of
                                                                                   revertants at 100 pg/plate
                                                                                   with 20 pi S9/plate and at
                                                                                   16          et al.. 1985
pg/ml were extremely cytotoxlc.

Tested at 250 mg/kg (the MTO)     Sorg et al.,
In corn oil                       1985

Positive with and without rat     NTP. 1991b
liver S9 when tested at 15 and
27 tig/ml

-------
TABLE V-24 (cont.)
o 	
J>
•i* Test
ro
O
Chromosomal effects
(cont.)
Cell transformation

BACTERIA
DMA damage

<
i
-t»
Ul




o
«x
ro
4k
vO
Organism (Assay)
Chinese hanster ovary cells
(chromosomal abberatlons)
Fischer rat embryo cell line
(morphologic changes)
AKR leukemia. virus-Infected
Swiss mouse embryo cell line
(morphologic changes)

B. subtlUs (rec)
E. coll (polA+1
S. typhlnmrlum (expression
of lac Z gene under control
of umu C gene)
S. typhlmurlum
(reverse mutation, his)





Exogenous Activation*
System
Rat liver S9
None
None

Rat liver S9 (Aroclor)
Rat liver S9 (none)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver mlcrosomes
(Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat or hamster S9
(Aroclor)
Results Reported
Positive
Negative
Negative
PHENANTHRENE
Negative
Negative
Negative
Positive
Negative
Negative'
Negative
Negative
Positive
Positive
Positive
Comments
Tested at 45 and 67.5 vq/ml
Tested up to 100 pg/mi
Tested at 5 pg/mi

Tested at 125 Mg/well
Tested up to 250 |ig/mi
•
Tested up to 8.3 |ig/mi
At 12 vg/plate In strain TA100
Tested up to 50 nmol/plate In
strain TA100
Tested up to 50 pg/plate In
strains TA1535. TA1537. TA98
and TA100
Tested up to 200 pg/plate In
strain TA100
Tested at 0.28 mN 1n strain
TA1537
Positively Identified In a
highly mutagenlc fraction of
raw gas condensate
At 5 ng/plate TA97
Tested from 50*200 t>g/plate
In strain TA100
Reference
NTP. 1991b
Freeman et al. ,
1973
Rhlm et al..
1974

McCarroll
et al., 1981
Rosenkrantz and
PoUler. 1979
Nakamura
et al.. 1987
Oesch et al . ,
1981
Mood et al.,
1979
McCann et al . ,
1975
La Vole et al..
1981a
Selxas et al.,
1982
Benson et al. .
1984
Sakal et al..
1985
Carver et al..
1986

-------
                                                                              TABLE V-24  (cont.)
r\>
o
o
CD
Test
DNA damage (cont.)
FUNGI
Recombination
HAHNALIAN CELLS
ONA damage



Nutation

Chromosomal effects
Cell transformation


Organism (Assay)
»
S. typhlmurlum
(forward mutation. 8AGR)

S. cerevlslae
(mltotlc recombination, 03)

Human foreskin epithelial
cells (unscheduled DNA
synthesis)
Primary rat hepatocytes
(unscheduled ONA synthesis)
F344 rat trachea epithelial
organ culture (unscheduled
ONA synthesis)
House skin (effect of co-
application with benzo[a]-
pyrene on formation of ONA
adducts)
Chinese hamster V79 cells
(forward mutation. 8AGR,
QUA") ,
Human lymphoblastotd TK6
cells (forward mutation,
TFT«)
Chinese hamster ¥79-4 cells
(sister -chromatld exchange,
aberrations)
House prostate C3HG23 cells
(morphologic changes)
Syrian hamster embryo cells
(morphologic changes)
House BALB/3T3 cells
Exogenous Activation*
System
Rat liver S9 (Aroclor or
phenobarbltal)

Rat liver S9 (Aroclor)

None
None
None
None
Syrian hamster embryo cell
feeder layer
Rat liver S9 (Aroclor)
Syrian hamster embryo cell
feeder layer
None
None
None
Results Reported
Negative

Negative

Negative
Negative
Negative
Negative
Negative
Positive
Negative
Negative
Negative
Negative
Comments
Tested up to 300 nmol/ml In
strain TH677

None

Tested up to 400 pg/ml
Tested up to 100 nmol/ml
Tested at 0.1x10"* to
10x10"* H
Average 17X decrease 1n forma-
tion of benzo[a]pyrene-DNA
adducts upon co-application
with benzo[a]pyrene
Tested at 1 pg/ml
Tested at SO nmol
Tested up to 10 pg/mi
Tested up to 10 pg/mi
Tested up to 40 pg/mi
Tested up to SO pg/mi
Reference
Kaden et al..
1979

Simmon. 1979a

Lake et al..
1978
Probst et al..
1981
Ide et al..
1981
Rice et al..
1984
Huberman and
Sachs. 1976
Barfknecht
et al.. 1981
Popescu et al..
1977
Narquardt and
Heldelberger,
1972
Plenta et al. .
1977
Kakunaga, 1973
                                (morphologic changes)

-------
                                                                            TABLE V-24 (cont:)
0
CO
Test
Cell transformation
(cont.)

BACTERIA
DNA damage



Nutation




Organism (Assay) Exogenous Activation*
System
S
Guinea pig fetal cells None
(morphologic changes)
Mouse embryo C3H/10T1/2 None
flbroblast cells

E. coll Various
Tree. eolA*..E°l*. uvrA)
B. subtllls (rec-J Various
None
S. typhlmurlum (expression Rat liver S9 (Aroclor)
of lac Z gene under control
of urau C gene)
S. typhlmurlum *°Co gamma Irradiation
(reverse mutation, his) of pyrene
Rat liver S9 (poly-
chlorinated blphenyl)

Rat liver S9 (Aroclor)
Various
I
Results Reported
Negative
Negative
PYRENE
Negative
Negative
Negative
Negative
Positive
Positive
Negative
Negative
Positive
Comments
Tested a 1 5 pg/mi
Tested at 4 MM

International collaborative
program (consensus view of
participants)
International collaborative
program (consensus view of
participants)
No difference In Inhibition
at 100 pg/plate
Tested up to 167 pg/ml
Tested at 160 pg/plate In
strains TA9B. TA1537 and TA1538
At 100 pg/plate with activa-
tion 1n strain TA1537
At 100 pg/plate with and
without activation In strains
TA98 and TA100
Tested up to 1000 pg/plate In
strains TA1535. TA1537. TA98
and TA100
At 25 pg/plate In strain
TA1537; consensus view of
Reference
DIPaolo. 1975
Peterson
et al.. 1981

Ashby and
K1lby, 1981
Ashby and
Kllby. 1981
K1nae et al.,
1981
Nakamura
et al.. 1987
Gibson et al. ,
1978
Klnae et al..
1981

NcCann et al.,
1975; La Vole
et al.. 1979;
Florin et al..
1980; Ho
et al.. 1981
Bridges et al. .
1981
                                                              Rat liver S9 {Aroclor)
Positive
participants In International
collaborative program

Positive at 2 pg/plate.
TA1537 TA97; higher doses
TA100
Hatljasevlc and
Zelger. 1985

-------
                                                                            TABU  V  ?4  (cont.)
             Test
                             Organism (Assay)
                                   Exogenous  Activation*
                                          System
Results Reported
          Comments
   Reference
      Nutation (cont.)
CD
FUNGI

Nutation



INSECTS

Nutation


NANNALIAN CELLS

ONA damage
oo
10
                        S.  typhlimirlum
                        (reverse nutation,  his)
                              S. typh1mur1um (reverse
                              mutation, his) (taped-
                              plate assay)

                              S. typhlmurlum
                              ("forward imitation. 8AGR)

                              S. typhlmurlum
                              (forward mutation, ARAR)
S. cerevlslae and
Schlzosaccharomyces pombe
(different genetic endpolnts)
                              0. melanogaster (sex-
                              linked recessive lethals)
Human foreskin epithelial
cells (unscheduled ONA
synthesis)

Primary rat hepatocytes
(unscheduled DNA synthesis)

HeLa cells (unscheduled
ONA synthesis)

Human flbroblast cell line
W138 (unscheduled DNA
synthesis)

F344 rat trachea epithelial
organ culture (unscheduled
DNA synthesis)
                                Rat  liver  S9  (Aroclor)
                                                              Rat liver  S9  (Aroclor)
                                                        Rat liver  S9  (Aroclor)
                                                        Rat liver  S9  (Aroclor)
                                                        Rat liver  S9  (Aroclor)
                                                              Various
                                None
                                                              None
                                                              None
                                                              Rat liver S9
                                                              (3-methylcholanthrene

                                                              Rat liver S9 (Aroclor)
                                                              None
Positive


Negative


Negative



Positive


Positive




Negative





Negative




Negative



Negative


Negative


Positive



Negative
At 2 ng/plate TA97. 4
wg/plate TA9B

Tested up to SO pg/plate In
strain TA98

Tested up to 50 ug/plate In
strain TA98
                                                                                    At  140  nmol/ml  In  strain  TN677
                                                                                    The  minimum positive  dose  In
                                                                                    strain  813 was  0.25
Consensus view of participants
In International collabratlve
program
800 dig/kg fed 1n diet for up
to 72 hours
Tested up to 400
                                                                                                            Tested up to 500 mol/mt
                                                                                                            None
                                                                                                            At  7.2
                                                                                                            Tested  from O.lxlO"6  to
                                                                                                            10x10"* N
Sakal et al.,
1985

Bos et al..
19B8
                                  Kaden et al.,
                                  19/y

                                  Dorado and
                                  Pueyo. 1988
de Serres and
Hoffman. 1981
Valencia and
Houtchens, 1981
Lake et al..
1978
                                  Probst et al..
                                  1981

                                  Martin et al..
                                  1978

                                  Robinson and
                                  Mitchell. 1981
                                  Ide et al..
                                  1981

-------
TABLE V-24 (cont.)
0
-*».
j».
o




<
1
U3





n/12/91
Test Organism (Assay)
ONA damage (cont.) Mouse skin (effect of co-
appllcatlon with benzo[a]-
pyrene on DNA adduct formation)
Primary F344 rat hepatocytes
(unscheduled DNA synthesis)
Mutation Mouse lymphoma L5178V cells
•(forward mutation, TFTR)
Mouse lymphoma L517BY cells
Mouse lymphoma L517BY cells
(forward mutation. TK~)
Mouse lymphoma LSI 78V cells
(forward mutation. TK~)
Chinese hamster ovary cells
(forward mutation. HGPRT)
Chromosomal effects Rat liver epithelial ARL18
cells (slster-chromatld
exchange)
Chinese hamster ovary cells
(slster-chromatld exchange)

Chinese hamster V79 cells
(slster-chromatld exchange)
Rat liver RLj cells .
(aberrations)
Chinese hamster liver
epithelial cells (slster-
chromatld exchange)
Human peripheral lymphocytes
(slster-chromatld exchange)
Exogenous Activation*
System
None
None
Rat liver S9 (Aroclor)
None
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
Rat liver S9 (Aroclor)
None
Rat liver S9 (Aroclor)

Syrian hamster embryo cell
feeder layer
None
None
None
Intact rat hepatocytes
Results Reported
Positive
Negative
Positive
Negative
Positive
Positive
Negative
Negative
Positive
Negative
Positive
Negative
Negative
Negative
Positive
Comments
Co-application with benzo[a]-
pyrene resulted 1n average In-
crease In level of DNA adducts
of 56X
Tested up to 1E-4 M
At 10 |ig/ml
Tested up to 128 pg/ml
Tested from 5.9-9 pg/mi
Tested at 1.5xlO"» to
2.02xlO'» M
Tested from 2.5-15 pg/ml
Tested up to 1 pmol/mt
Between 19 and 300 pg/mi
Tested up to 100 pg/ml
Tested at 10 pg/ml
Tested up to 100 pg/ml
Tested at 5-45 ng/ral
Tested at 10'4 to 10'6 M
lested at 10~4 M
Reference
Rice et al..
1984
Williams
et al.. 1989
Jotz and
Mitchell. 1981
Mitchell
et al.. 1988
Mitchell
et al.. 1988
Wangenhelm and
Bolcsfoldl.
1988
Oshlro et al. .
1988
Tong et al. .
1981b
Evans and
Mitchell. 1981
Perry and
Thomson. 1981
Popescu et al .
1977
Dean. 1981
DeSalvIa
et al.. 1988
Llndahl-
Klessllng
et al.. 1989

-------
                                                                              TABLE  V-24 (cont.)
c\>
o
 en
 O
Test
Chromosomal effects
(cont.)
Cell transformation



•\
Organism (Assay) Exogenous Activation*
System
Mouse lyaphoma cells (DNA Rat liver S9 (Aroclor)
strand breaks)
Syrian hamster embryo cells None
(morphologic changes)
Mouse prostate C3H cells Mouse embryo flbroblast
(morphologic changes) feeder layer
Mouse BALB/C-3T3 cells None
(morphologic changes)
Guinea pig fetal cells None
(morphologic changes)
Syrian hamster embryo cells None
(morphologic changes)
Results Reported
Positive
Negative
Negative
Negative
Negative
Negative
Comments
Tested at 0.5xlO~4 and
5xlO"« M
Tested up to 20 pg/mi
Tested at 1 gg/mi
Tested at 20 jig/ml
Tested at SO pg/mi
Tested at 10 pg/mt
Tested at 1-100 tig/mi
Reference
Garberg et al..
1988
DIPaolo et al.,
1969; Plenta
et al.. 1977;
Casto. 1979
Chen and
Heldelberger,
1969
DtPaolo et al..
1972; Kakunaga.
1973
Evans and
DIPaolo. 1975
Tu et al.. 1986
        •Compound In parentheses refers to type of Inducing treatment administered to animal from which the exogenous activation system was prepared.
 o
 CO

-------
other  PAHs  1n an  ethylene glycol,   trlcaprylln  or  ethyl  laurate vehicle.
The  mice  were then  observed  for  18 months  to  determine  the  Incidence of
sarcomas.    Survival  was  90-100X  during  the  critical  period  of   tumor
formation,  the  4th  to  llth month.   The authors  found  that both  dose and
Interval between administration of the carcinogen  and  noncardnogen affected
the  degree  of observed  Inhibitory response.  Phenanthrene  had  substantial
Inhibitory  effects  on  the  production  of sarcomas by  d1benz[a,h]anthracene,
particularly  at  lower  dosages.   At  275  yg of  the  carcinogen,  concomitant
administration  of  a  24:1 molar  ratio  of   phenanthrene   to  d1benz[a,h]-
anthracene  (1n ethyl laurate) yielded ~50X reduction 1n  the  percentage of
tumor-bearing  animals  as  compared  with  administration   of  d1benz[a,h]-
anthracene  alone.   When   the  dosage   was  decreased  to  60   yg  of  the
carcinogen,  a  similar  molar  ratio  of phenanthrene to d1benz[a,h]anthracene
was  shown   to  yield a  comparable reduction  In  tumor development.   It was
found   that   either  concomitant   administration   of    carcinogenic   and
noncardnogenVc PAHs  or  administration  of the noncardnogen either  2  days
prior  to   or  after  the   carcinogen  administration  resulted   1n  maximum
      /
reduction 1n  the  percentage of  tumor-bearing  animals.  When the  vehicle was
changed   to   trlethylene  glycol,    phenanthrene   In    combination    with
d1benz[a,h]anthracene  had  a   substantial  promoting   effect,   approximately
doubling the percentage of tumor-bearing  animals.

    In  the   same   study   benzo[a]pyrene  was  also  tested  for  carcinogenic
effects 1n  combination  with  other  noncardnogenlc,  weakly  carcinogenic and
potent  carcinogenic hydrocarbons  1n a   trlcaprylln  vehicle (Falk  et  al.,
1964).. The chosen ratios of  putative antlcardnogen to carcinogen approxi-
mated  those  that  occur   1n  the  environment.   PAHs,  Including benzofa]-
fluorene,   chrysene,  benzo[k]fluoranthene,   perylene,   and  a   mixture  of

04420                                V-151                          10/08/91

-------
anthracene, phenanthrene  and  pyrene, had  substantial  Inhibiting effects on
the  ability  of  benzo[a]pyrene  to  produce Injection  site  sarcomas.   Other
PAHs,  Including  acenaphthylene,  fluorene,  anthracene, benzo[g,h,1]perylene,
1ndeno[l,2,3-cd]pyrene and coronene,  had  no such  Inhibiting effects.

    By  contrast,  Pfelffer (1973,  1977)  found no  Inhibitory effects  of 10
noncardnogenlc PAHs  on  benzo[a]pyrene  or d1benz[a,h]anthracene cardnogen-
1c1ty.  Groups of 100 NRHI female mice were given  single  subcutaneous  Injec-
tions  of various dosages  of benzo[a]pyrene and d1benz[a,h]anthracene, and 10
noncardnogenlc  PAHs,  separately and  1n  combination.  In p.5  mi  tMcapry-
I1n.   The  PAHs,  which   the  authors  considered  noncardnogenlc,  Included
benzo[a]anthracene,    phenanthrene,   anthracene,    pyrene,     fluoranthene,
chrysene,  benzo[g,h,1]perylene,  perylene,  benzo[e]pyrene  and coronene.  The
animals were examined weekly  for  the development of sarcomas for 114 weeks.
Relationships   were  found  to  exist  between   Increasing  dose  and  tumor
development for  benzo[a]pyrene,   d1benz[a,h]anthracene,  benzo[a]pyrene  plus
                                                                          •
d1benz[a,h]anthracene,  and  all  12  PAHs  administered  as  mixtures.   No
treatment-related Increases In sarcoma development were  observed for any of
the  10 noncardnogenlc  PAHs  when  administered  separately.   Interestingly,
the  noted  dose-response  curves  for  combinations   of   PAHs   most  closely
resembled   the  curve   for   
-------
     Pott  et al. (1977) conducted comparable  experiments  on  the carcinogenic
effects  of automobile exhaust condensates  (AEC)  1n mice.   Female  NRHI  mice
1n  groups  of  ~88 animals  were administered single subcutaneous Injections of
various  dosages of  AEC  and  benzo[a]pyrene,  separately and  1n combination,
which  had been  dissolved  or suspended  In trlcaprylln.  AEC  contained  both
.PAHs and  non-PAH substances.  When AEC was  present  with  benzo[a]pyrene, the
Incidence  of  tumors  (most of which were  sarcomas)  was  less  than when benzo-
[a]pyrene  was administered  alone.   This  Inhibitory effect  of AEC  was  par-
ticularly  obvious  at the higher dosages  of  benzo[a]pyrene,  but a decreasing
relationship  between tumor  Incidence  and level  of  AEC was  observed  at all
dosages  of benzo[a]pyrene.  In a  second series of  experiments,  Pott  et al.
(1977)  sought  to assess  the  effects  of a PAH fraction and non-PAH substances
prepared  from the AEC.   These were  Injected separately and  1n  combination
with benzo[a]pyrene.  Mice Injected  1n this  set of experiments were observed
for  a  year.    The  PAH-conta1n1ng  fraction   of  AEC  was  tumorlgenlc.   When
assayed 1n  combination with other AEC  fractions, tumor1gen1c1ty was reduced.

    The   relative  proportions  of  PAHs  found  1n  automobile  exhaust  gas
condensates  were determined  1n  relation  to benzo[a]pyrene.   The  PAHs  were
divided  Into  two  groups;  the  first   group  was composed of benzo[a]pyrene,
d1benz[a,h]anthracene,  benz[a]anthracene  and  benzo[b]fluoranthene  and  the
second  group  phenanthrene,   anthracene,  fluoranthene,   pyrene,  chrysene,
benzo[e]pyrene and  benzo[g,h,1]perylene.   The  concentrations  used  1n  each
dose group were determined by the  relative  concentration of benzo[a]pyrene.
The  benzo[a]pyrene   concentrations   selected   for   the   first  dose  group
(benzo[a]pyrene,   d1benz[a,b]anthracene,  benz[a]anthracene   and  benzo[b]-
fluoranthene)  were 1, 1.7 and  3.0 yg  benzo[a]pyrene and the  benzo[a]pyrene


04420                                V-153                           10/08/91

-------
concentrations selected  for  the  second  dose  group (phenanthrene,  anthracene,
fluoranthene,  pyrene,  chrysene,  benzo[e]pyrene  and  benzo[g,h,1]perylene)
were  1,  3,  9  and  27 yg  benzo[a]pyrene  (benzo[a]pyrene  was not  Included  1n
the  group).   A third  dose  group was created  by combining the compounds  In
groups one  and two;  the relative benzo[a]pyrene  concentrations selected  for
this  dose group  were 1,  1.7  and 3.0 yg benzq[a]pyrene.   All  compounds  were
dissolved 1n acetone.   Four control  groups that  received only  benzo[a]pyrene
at  doses  of 1,  1.7  or  3.0  yg  1n acetone and a solvent  control  group  were
also  employed.  The  shaved  skins of 100 female  NHRI  mice/group were treated
twice/week  until  natural  death  or  development  of  an  application  site
carcinoma (Schmahl  et al., 1977).

    The Incidence of  application site  sarcomas 1n the solvent control  group
was  1/81;  no  other   tumors  were reported.   In  the  benzo[a]pyrene  control
groups the  carcinoma Incidences  were  10/77, 25/88 and  43/81  In  the  1,  1.7
and  3 yg  benzo[a]pyrene  groups,  respectively.  The  carcinoma   Incidences
for    the    first     dose    group   (benzo[a]pyrene,    d1benz[a]anthracene,
benz[a]anthracene and  benzo[b]fluoranthene)  were  25/81,  53/88 and  63/90  1n
the  1,  1.7  and  3  yg  benzo[a]pyrene  groups, respectively.   The  carcinoma
Incidences   for    the   second   dose    group   (phenanthrene,   anthracene,
fluoranthene,  pyrene,  chrysene,  benzo[e]pyrene  and  benzo[g,h,1]pery1ene)
were  1/85,  0/84, 1/88  and 15/86  1n the 1,  3,  9 and  27  yg  benzo[a]pyrene
groups, respectively.   In the  third group  (combination  of  groups one  and
two)  the carcinoma Incidences  were  44/89,  54/93 and 64/93 1n  the 1,  1.7  and
3  yg  benzo[a]pyrene  groups,  respectively.  The  reported   papHloma  and
sarcoma Incidences  of all groups did not exceed 5X (Schmahl et  al.,  1977).
04420                                V-154                           10/08/91

-------
    The  cocardnogenlc  potential  of  various  substances found  1n cigarette
smoke was studied by Van Ouuren et al.  (1973) and  Van  Duuren and  Goldschmldt
(1976).  In  the  1973 study, female  ICR/Ha  Swiss  mice  1n  groups  of  50 were
given  repeated  dermal   applications   of  5  yg  benzo[a]pyrene   1n  0.1  ml
acetone  3  times/week either alone  or  with  pyrene,  benzo[g,h,1]perylene or
benzo[e]pyrene.  Control groups  received  either no  treatment  or  treatments
of  acetone,  benzo[g,h,1]perylene,   pyrene   or  benzo[e]pyrene  alone.   The
experiment ran for 52 weeks.  Cocardnogenlc activity was  shown by the  three
PAHs  (Table  V-25).   No  skin tumors were  found  In  the control  groups.  When
benzo[a]pyrene was applied with the Individual noncarclnogens,  the number of
mice with paplllomas  Increased  1.5-2.6 times over  the number  observed when
benzo[a]pyrene was applied  alone.   Likewise,  the  total number  of  paplllomas
was 2-6  times  higher when  the noncarclnogens were present as  compared with
benzo[a]pyrene alone.   Moreover,  the number of days  from first  application
to  appearance of  the  first tumor  In  the multiple  treatment  groups was
similar to that 1n the benzo[a]pyrene  group  alone.

    In'a  second  study,  Van  Duuren and  Goldschmldt (1976)  assessed both the
cocardnogenlc   and   tumor-promoting   activity   of    benzo[g,h,1]perylene,
fluoranthene, benzo[e]pyrene and  pyrene 1n  conjunction with benzo[a]pyrene.
Again, female  ICR/Ha  Swiss  mice  In  groups of 50 were  used.  Various dosages
of  the  above noncardnogenlc  PAHs  alone   and  1n  conjunction  with  5  yg
benzo[a]pyrene were  applied 1n 0.1  ml of  acetone or  1n  DHSO  3  times/week
for <440 days.  Pyrene,  fluoranthene and  benzo[e]pyrene  were found  to be
potent  cocardnogens  for   benzo[a]pyrene,   while   benzo[g,h,1]perylene was
determined  to  be "moderately cocardnogenlc".   Pyrene and f luoranthene also
proved to be weak tumor-promoting agents when  an  Initiating dose of benzo-
[a]pyrene was used.  The results  of  the study are presented 1n Table V-26.

04420                                V-155                          10/08/91

-------
                                  TABLE  V-25

  Cocardnogenlc Activity of Various  PAHs with Benzo[a]pyrene 1n Mouse Sk1na
                              (1 Year Exposure)
  Carcinogen0
Cocarclnogen
  Dose of
Cocarclnogen0
Np. of Mice with
 PapHlomas/No.
  of Survivors
  at 52 Weeks
Benzo[a]pyrene
Benzo[a]pyrene
Benzo[a]pyrene
Benzo[a]pyrene
—
Benzo[e]pyrene
Benzo[g,h,1]perylene
Pyrene
—
15
21
12.
13/42
34/39
20/37
27/41
aSource: Van Duuren et al.t  1973

DBenzo[a]pyrene  (5 yg)  and/or  cocardnogen  was  applied  as  acetone  solu-
 tions  3  times/week  for 52  weeks.   There were  no  tumors  In  the  groups
 treated only with acetone,  benzo[e]pyrene.  benzo[g,h,1]perylene or pyrene.
04420
                 V-156
                          10/08/91

-------
                                                     TABLE V-26

                     Cocarc1nogen1c Activity of Various PAHs with Benzo[a]pyrene on Mouse Sk1na

Carcinogen''


Benzo[a]pyrene

—
Benzo[a]pyrene
Benzofajpyrene
—
Benzo[a]pyrene
Benzo[a]pyrene
—
Benzo[a]pyrene
—
—
Benzo[a]pyrene
Benzo[a]pyrene
Benzo[a]pyrene

Cocarclnogen


Acetone
Acetone
Benzo[e]pyrene
Benzo[e]pyrene
Benzo[e]pyrene
Benzo[g,h,1]perylene
Benzo[g,h,1]perylene
Benzo[g,h,1 jperylene
Fluoranthene
Fluoranthene
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene
Dose of
Cocarclnogen
(v9)

._
--
15
15
5
21
21
7
40
40
40
12
40
12
4
Duration
of
Experiment
(days)
368
440
368
368
368
368
368
368
440
440
440
368
440
368'
368
Days to
First
Papllloma

251
210
--
246
249
—
222
238
--
99
—
—
229
186
250
Mice with
Paplllomas/
Total Paplllomas

14/16
16/26
0
33/79
24/33
0
20/39
19/31
0
39/126
0
0
35/66
26/42
12/14
    aSource: Van DQuren and Goldschmldt, 1976

    &Benzo[a]pyrene  (5 pg/0.1  ml  acetone or  DMSO)  was  applied 1n  the same  solution  as  the Cocarclnogen,
     3 times/week.
o>

-------
    R1ce et al.  (1984)  also  found  fluoranthene  and  pyrene to have cocardno-
genlc  potential  when  combined  with benzo[a]pyrene.   When female  CD-I  mice
were  given dermal  applications of  150  yi  [3H]  benzo[a]pyrene and  either
fluoranthene  or  pyrene,  the occurrence  of  DNA adduct  formation  (after  24
hours)  was 66 and  56X,  respectively,  above the  level  found  for  benzofa]-
pyrene  alone.   Phenanthrene  co-application  resulted  1n  an  average  17X
decrease and formation of [3H]benzo[a]pyrene-DNA adducts.

    Slaga  et  al.  (1979) examined  the  effects  of several  PAHs  on  skin  tumor
Initiation  by   7,l2-d1methylbenz[a]anthracene   (DMBA)   and   benzo[a]pyrene.
Either  benzo[e]pyrene,  pyrene or  fluoranthene  1n acetone  was  applied  to the
backs  of  female  CD-I mice  (30/group)  5 minutes before  Initiation  with DMBA
or  benzo[a]pyrene.   Beginning 1 week  after  Initiation  mice received  10  v9
of  12-0-tetradecanoylphorbol-13-acetate  (TPA) 1n acetone  twice weekly  for  30
weeks.   The  Incidences of  both  paplllomas and   carcinomas  were  observed
weekly  and removed  at  random  for  hlstologlc  verification.   When  applied
before    Initiation,    pyrene   and    benzo[e]pyrene    slightly    enhanced
benzo[a]pyrene Initiation whereas  fluoranthene  had  a  marginal  effect.   These
same  PAHs  had an  Inhibitory effect  on  7,l2-d1methylbenz[a]anthracene  skin
tumor   Initiation.    When   pyrene  and  fluoranthene  were   applied   after
benzo[a]pyrene Initiation, they  had no tumor  promoting effects.

    Huang  et  al.  (1986) Investigated the ability of  benzo[a]pyrene phenolic
metabolites to Interfere with the mutagenlclty and tumor1gen1c1ty  of  benzo-
[a]pyrene  or  the putative  ultimate carcinogen,  benzo[a]pyrene-7,8-d1ol-9,!0-
epoxlde.   Of  12  Isomerlc phenolic metabolites,  3-hydroxy-benzo[a]pyrene was
the  most   potent   antagonist  of  7,8-d1ol-9,lO-epox1de  mutagenlclty  for


04420                                V-158                           10/08/91

-------
Salmonella  typhlmuMum  strain  TA100;  3  ymol  reduced mutagenlcHy  by 50%.
The  3-hydroxy-benzo[a]pyrene  also  decreased m1crosome-med1ated mutagenlcHy
of benzo[a]pyrene and benzo[a]pyrene-7,8-d1ol for TA100.  Likewise, mutagen-
lcHy of  the 7,8-d1o1-9,lO-epox1de for mammalian  cells  (V79) was decreased
by 50%  upon addition  of 8  yM 3-hydroxybenzo[a]pyrene.   A  2500  ymole dose
of  3-hydroxybenzo[a]pyrene  was  applied  topically  to  the   shaved  backs  of
female CD-I mice 5 minutes before addition  of a  tumor-Initiating  dose  of  200
vmole of  benzo[a]pyrene-7,8-d1ol-9,!0-epox1de.   This was followed  by  16-20
weeks of  promoting  treatment with  TPA.   At the  end of  this  time  a  72-78%
reduction 1n numbers of  tumors/mouse  and  a 42-55% decrease. 1n Incidence  was
observed by  comparison  with  animals  treated  with the 7,8-d1ol-9,lO-epox1de
and  TPA  alone.   A  less  dramatic  reduction (35-41%  decrease  1n  Incidence,
40-52%  Inhibition  In number  of tumors/animal)  was  observed when benzo[a]-
pyrene was the Initiating agent.

    The enhancing or Inhibiting action  of non-PAH materials on PAH cardno-
genlclty has  also  been a  subject  of study.  El-Bayoumy (1985)  showed that
two  synthetic organoselenlum compounds,  p-methoxybenzeneselenol  and benzyl-
selenocyanate,  Inhibited  forestomach  tumors 1n  mice treated with benzofa]-
pyrene.   As described 1n the  oral cardnogenUHy section, 9-week-old  female
CD-I  mice were fed  a standard diet for  1  week.   At the  beginning of week 2,
25 mice/group  were  fed  diets  containing  varying  levels  of  the  above  two
compounds or  their  nonselen1um-conta1n1ng  congeners.  A third pair  of com-
pounds,  phenothlazlne and phenoselenazlne,  were also  tested.   Beginning with
week 3, mice were gavaged  twice weekly for 4 weeks with 1 mg  benzo[a]pyrene
1n corn  oil.   The  test  diets were continued for  1  week post-treatment  and
then standard diets  were  fed until  week 28 when the mice were killed.  Mice


04420                                V-159                          10/08/91

-------
receiving benzo[a]pyrene and standard diet had a  forestomach  tumor  Incidence
of 85X.  This  was  decreased  to 25X In the high-dose  p-methoxyphenol-treated
group.  Benzylselenocyanate and methoxybenzeneselenol produced  Insignificant
reductions  1n  tumor Incidence but  significant  reductions 1n  the number of
tumors/mouse.

    As noted  1n  the nonoral  carc1nogen1c1ty section, the pulmonary  cardno-
genlcHy of  benzo[a]pyrene can be  greatly  enhanced by co-administration of
partlculate  material.   Pershagen  et  al.  (1984)   studied   the  pulmonary
cardnogenlcHy  of  benzo[a]pyrene  alone  and  1n combination  with arsenic
trloxlde  (ASpO-)   1n   male   Syrian  golden  hamsters.    The  animals   were
divided  Into  four  groups:    ASpCL,  benzo[a]pyrene,   Aso°3  Plus  benzo-
[a]pyrene.  and vehicle  controls.   At each  of  15 weekly  Instillations,  3
mg/kg  of  arsenic  and/or  6 mg/kg  of  benzo[a]pyrene  was  administered.    All
groups received  a  carrier  dust (charcoal carbon),  which  Increased the  lung
retention of  arsenic.   Carcinomas  of  the larynx,  trachea,  bronchi or  lungs
were  found  1n  3/47,   17/40   and   25/54  animals  examined   1n   the  As.Og,
benzo[a]pyrene  and  As-O.   plus  benzo[a]pyrene   groups,  respectively.   No
respiratory tract  carcinomas were  found  1n  the  53 controls.  The Incidences
of  pulmonary  adenomas,  paplllomas  and   adenomatold  lesions  were markedly
higher In the  arsenic  group than  In  the  control  group (p<0.01).  There  was
also  some evidence of  a positive  Interaction between arsenic and  benzo[a]-
pyrene In relation  to  adenomatos1s  lung  tumors.   The authors noted  that  the
presence  of   the   carrier  dust  prevented  rapid  lung  clearance  of   the
As90..  They  proposed- that the  arsenic  may have had an adverse effect on
  £ 
-------
    Met1v1er  et  al.   (1984)  studied  the effects  of  different  levels  of
                    /
Plutonium  oxide  (PuO?)  and  benzo[a]pyrene  on  lung cardnogenesls  1n SPF

Wlstar rats In  an effort  to determine  the  cocardnogenlc effects  of  Ionizing

radiation  and  benzo[a]pyrene.  Eight  different  experimental  groups  Inhaled

PuO^  at   four  dose  levels  with  and without  benzo[a]pyrene  (given  In two

doses of  5 mg by  Intratracheal  Instillation).  The results are  summarized 1n

Table V-27.   They  show  that  the  Incidence  of  lung tumors  Increased  as  a

function  of  both  PuCL dose  and benzo[a]pyrene  exposure.   The  authors felt

that a multiplicative relative risk model  fH the observed  effects.



    Jones  et  al.  (1984) Investigated  the  effect  of  Ia,25-d1hydroxychole-

calclferol  [1,25-{OH)_D3] on  the  transformation  of cells  pretreated with

benzo[a]pyrene and  benzo[e]pyrene.  Treatment of  Syrian hamster embryo  cells

with  benzo[a]pyrene  for  3  days  followed  by   treatment  with   1,25-(OH)2D3

for  4  days Increased  transformation of  the cells.   Benzo[e]pyrene,  not  a

complete  carcinogen  by Itself,  Induced cell  transformation when  followed by
                                                                           •
1,25-(OH) 0, application.
         b 0
      /


    Rahlmtula et  al.   (1977)  examined  the capabilities  of nine  antloxldants

(ascorbate, butylated  hydroxyanlsole,  butylated  hydroxytoluene,  ethoxyquln,

glutathlone, NNN'N'-tetramethyl-p-phenylened1am1ne dlhydrochlorlde,  nordlhy-
                                                           #
droguarlarotlc add, propyl gallate and pyrogallol)  to  affect benzo[a]pyrene

hydroxylatlorr by  rat  Hver  mlcrosomal  MFCs.   All of the antloxldants  tested

reduced the bacterial  mutagenldty of  benzo[a]pyrene 1n the presence of rat

liver mlcrosomes  and  cofactors.   They  also found that several  synthetic and

naturally  occurring  flavones, when  Incorporated  In  the diet  (3-5 mg/g) or

applied  to  the  skin,  caused  an  Increase  1n  benzo[a]pyrene  hydroxylase
04420                                V-161                            10/08/91

-------
                                                       TABLE V-27


                               Co-administration of Pu02 and Benzo[a]pyrene by Inhalation*
No. of Animals
Pu02 Initial
Lung Burden
(Bq)
0
220
630
6300
0
220
630
6300
Benzo-
[a]-
pyrene
(mg)
0
0
0
0
2x5
2x5
2x5
2x5
No. of
Animals
89
89
30
19
38
29
22
19
Median
Survival Time
(days)
864
820
798
345
675
444
480
330
Median Lifetime With Pulmonary
Dose Malignancies
(By)
0.0
3.3
9.4
76.3
0
2.9
8.5
75.4
Fatal
0
4
6
1
7
15
14
18
Incidental
0
13
8
5
3
2
2
1
With Fatal
Benign Tumors
0
0
0
0
2
10
0
0
       'Source: Met1v1er  et al.,  1984
o
CO

-------
activity  In  the  small   Intestine  and  skin,  respectively.   In  addition,
pulmonary   adenoma   formation   resulting   from  oral   administration   of
benzo[a]pyrene   was   totally  prevented,  and   skin  tumors   Initiated   by
benzo[a]pyrene  application  to  mice  were  significantly reduced  (>50X)  by
treatment  with  the   synthetic  flavbne,  B-naphthoflavone.   Pulmonary  tumor
formation was  also reduced 50% by  Incorporation of  the  naturally  occurring
flavone, quercetln pentamethyl ether, Into the diet.   Sullivan  et al.  (1978)
demonstrated  that  BHA,  BHT,  phenothrazlne,  phenothrazlne  methosulfate  and
ethoxyquln  all   can  reduce   the   quantitative  yield  of   benzo[a]pyrene
metabolites In Incubations with rat liver mlcrosomes.

Summary
    Studies of the general and  specific  health effects 1n animals associated
with exposure  to PAHs tend  to  be limited 1n  the  number of PAHs considered
and  the  route of  administration.   A wide  variety of general  toxic  effects
have been  linked  to  exposures.   Effects on  the hematopoletlc and  lymphold
systems seem to  predominate,  such ,as anemia,  hemolysls,  leukemia  and  lympho-
cyte  toxlclty.  Equally  Important  to   the  nature  of  effects,  the  dosage
eliciting  such  effects  appears  to vary  by  route  of  administration;  for
example,  for  compounds  like  fluoranthene,  acute  toxlclty  (LD5Q) occurs  at
a lower dosage for oral versus other routes  of exposure.

    Some PAHs.are  known  to be carcinogens for  animals.  In  terms  of specific
health  effects,-  relatively  potent  carcinogens  like   benzo[a]pyrene  and
d1benz[a,h]anthracene  as  well   as  moderate-to-strong   cocardnogens  like
fluoranthene and pyrene  are  found  among the PAHs.   Dose-response  relation-
ships  for  oral exposures have  been  documented for  some  of  the carcinogenic
04420                                V-163                           10/08/91

-------
PAHs.  The  carcinogenic  effect  of some PAHs when administered  to animals  by
other routes  such  as  subcutaneous  and  IntraperKoneal  Injections,  Inhalation
and skin-painting has also been documented.

    Environmental PAH exposures are thought to occur to mixtures rather  than
Individual   compounds.    Experimental   PAH  'mixtures   research   assessing
carcinogenic  activity  has  focussed  on  nonoral  routes  of  administration;
these  Include  subcutaneous  Injection,  dermal  application  and  Inhalation.
When  studied  1n  combinations.  Inhibition,   promotion   and  cocardnogenlc
effects have  been  reported for  PAHs.   PAHs'  have also been  reported  to  act  as
Initiators  of  cardnogenesls   when  followed  by  repeated  treatments  with
non-PAHs  such as  plutonlum  oxide.    Mixtures,  such as  automobile exhaust
condensate, that contain  a  variety of  PAHs  as well as other  compounds,  have
also been shown to be carcinogenic 1n  animals.

    Reproduct1ve/teratogen1c effects have not  been evaluated for most  PAHs.
However,  there  1s  some  evidence suggesting  that Ingestlon  of  benzo[ajpyrene
reduces fertility  and reproductive capacity  and  has  deleterious effects  on
the offspring of mice so exposed.

    Several  PAHs  have   been  extensively  studied   for   genotoxlc   effects.
Benzo[a]pyrene  seems  clearly  to  be   mutagenlc   In   both  prokaryotlc  and
mammalian cells  1n the  presence of an activation  system and causes various
chromosomal effects  1n  mammalian  cells.  Benzo[g,h,1]perylene, chrysene and
d1benz[a,h]anthracene are mutagens In several systems,  and the latter  also
damages the DNA of  both prokaryotlc  and  eukaryotlc cells.   Benz[a]anthra-
cene,  normally  a  weak  carcinogen, can be  activated  to  exhibit  mutagenlc

04420                               V-164                           10/08/91

-------
activities comparable to benzo[a]pyrene.  For the remainder of the compounds
under review, there are either too few data to draw firm conclusions, or the
results   have   been  mixed.   Most   of  these   compounds   have  exhibited
genotoxldty 1n some systems.
04420                                V-165                           10/08/91

-------
                         VI.   HEALTH  EFFECTS  IN  HUMANS
Introduction
    Human beings are likely to  be  exposed  on a  dally basis to various levels
of PAHs  1n  the  ambient environment.   Primary vehicles of  exposure  among the
general  population  Include Inhalation  of  mainstream and  sldestream  tobacco
smoke  and  exhaust  products  from the  combustion  of  commonly  used  fuels.
Certain  occupational   groups   have  the additional  burden  of  exposure  to
extremely high levels of complex mixtures  that  contain PAHs  as well as other
toxic  and  tumorIgenlc  compounds.  Those  considered at  high risk  for  these
exposures  Include  steelworkers,   coke  oven  workers,  gas  workers,  foundry
workers,  aluminum  reduction  plant  workers,  roofers,   chimney  sweeps,  and
possibly auto mechanics.   Scrotal  cancer  among chimney  sweeps  1s  considered
to  be  the  first and  oldest   Identified  occupational  cancer  (Pott,  1775).
These  occupational  exposures   occur  almost  exclusively  via  Inhalation  or
direct dermal contact.   It 1s  from  this  body of scientific evidence  that  a
causal  relationship has  been   clearly established  between  PAH-conta1n1ng
mixtures and  several  specific  cancer  sites   (IARC,  1984,  1985; WHO,  1988).
These  studies are briefly  summarized  at   the end of this chapter,  although
their  direct  relevance  to  oral  Ingestlon   of  PAHs  1n  drinking  water  1s
unclear.

    PAHs  do occur  In  Ingested  media  such  as  water  and  foods,  but  these
exposures have  not been well-characterized  or  evaluated.   PAHs  have  been
Identified  In  smoked   foods,  charbrolled   meats,  raw vegetables,  shellfish,
and  refined fats  and  oils.    To  study  ep1dem1olog1cally  the  relationship
between  any  specific   nutrient   and   a   specific   adverse   event   Is  very
difficult,  however,  and  extreme   caution   must   always  be   exercised  In


04430                                VI-1                             08/09/91

-------
Inferring causality.  Howson et al.  (1986) point out  that,  although  an  early
study  Implicated  1ngest1on of  PAHs via  charcoal  broiled  and  smoked  foods
with  an  Increased risk of  gastric  cancer, subsequent experimental  research
showed that  PAHs  were not  good  candidates for carcinogens of  the  glandular
stomach and  therefore probably  were only  Indirectly  associated with  gastric
cancer.

    Cigarette smoking 1s strongly and consistently associated with  cancer  at
several sites Including the lung, larynx,  oral cavity, esophagus  and  bladder
(Schottenfeld and  Fraumenl,  1982;  Maclure and MacMahon, 1980).   In  addition
to n1trosam1nes and  other  chemicals, 34  PAHs,  some  of which are recognized
as carcinogens, have  been  Identified 1n mainstream and sldestream  cigarette
          i
smoke  (IARC, 1983; Uynder,  1988; Appel  et  al., 1990).   Inhalation of  tobacco
smoke  1s  also  considered  to  be   a  major  risk  factor  for cardiovascular
disease.   However,  the  specific  relationship  of  any  of  the  compounds  1n
tobacco smoke to the  complex etiology of  this  disease 1s  unclear.  Albert  et
al. (1977) demonstrated 1n  chickens that direct Injection  of PAHs  [DHBA and
       /
B[a]P] Into  the  aorta could Induce atherosclerotic  lesions.   In humans,  H
1s  likely  that  other  constituents  of  tobacco  smoke,   primarily carbon
monoxide  and nicotine, also  are  Involved 1n  the  disease  process and may
possibly  be  the  main  culprits, not   PAHs  (Uynder.  1988).   This  will  be
briefly   addressed   later   1n   the  review   of   more  recently  published
occupational studies  (Hansen,  1983, 1989;  Gustavsson et  al.,  1988),  but  a
complete  review  of the smoking  and human  health  literature 1s  well beyond
the scope of this document.
04430                                VI-2                            08/09/91

-------
Clinical Case Studies
    Oral.   The  only available  case reports  of oral  1ngest1on  of  Isolated
PAHs  are   concerned  with  accidental  Ingestlon  of   naphthalene.    Acute
hemolytlc anemia 1s  the most  frequent  manifestation of  naphthalene  poisoning
1n  humans.   Case reports  have  described the appearance of acute  hemolytlc
anemia after naphthalene Ingestlon  by  children  (Jacobzlner  and Raybln,  1964;
Athreya et  al.,  1961; Gross  et a!.,  1958;  Zlnkham  and Chllds,  1957,  1958;
Haggerty,  1956;  Chusld  and  Fried, 1985;  Bregman,   1954;  MacGregor,  1954;
Abelson and  Henderson,  1951; Mackell  et al.,  1951;  Zuelzer  and Apt,  1949)
and adults (Anz1ulew1cz et al., 1959;  Zlnkham and  Chllds,  1957,  1958; Gldron
and Leurer, 1956).

    The reported mechanisms  and range of  exposure to  naphthalene 1n  these
case studies were:  1)  chewing,  sucking, or  swallowing  of mothballs  (one  to
numerous) as a  single  Incident or  for  periods  up  to 3 months;  2)  Ingestlon
of  toilet  bowl  deodorant  cakes  (pure  naphthalene) by a child over a period
of a year;  and  3)  Ingestlon of  naphthalene-containing deodorant 1n a diaper
pall for  an unspecified  period.    Doses of  naphthalene  were  not  generally
reported In  these  case  studies  because  of  the  poorly defined nature of  the
exposure.    Tests  to  detect  naphthalene derivatives  1n  the  urine  of  the
anemic Individuals  were  negative 1n some cases and  positive In  others.

    Symptoms of  naphthalene  tox1c1ty  that  frequently precede the  diagnosis
of  acute  hemolytlc anemia  1n  persons  of  all  ages  Include mild  to severe
jaundice,   dark  urine  (red,  orange,  or port  wine  colored),  pallor,  and
lethargy  (U.S.   EPA,  1987a).   Severe  Jaundice  Is   often  the   reason  for
hospUaHzatlon,  since  the  jaundice  often develops  before  severe anemia


04430                                VI-3                           08/09/91

-------
 becomes  manifest.   However,  It  1s  clear  that  anemia  and  Jaundice can develop
 In  parallel  as  shown by  a  time-course  study  of hematologlc  changes  1n  a
 16-year-old girl who  had Ingested  about 6  g  of naphthalene  1n a  suicide
 attempt  (Gldron and  Leurer,  1956).   Vomiting and tachycardia  are  occasion-
 ally  observed as precHnlcal  signs  of  naphthalene  poisoning In persons  of
 all  ages.   Precllnlcal  signs  of naphthalene toxldty observed  primarily  1n
 neonates  or  children  Include   anorexia,  cyanosis,   shallow  respiration  or
 apnea,   convulsions,  and  diarrhea.   Precllnlcal  symptoms  of  naphthalene
 poisoning  reported  by children  or  adults Include fever, confusion,  pain  In
 abdominal  or  kidney  region,  pain  at urination,  nausea,  headache,  fainting,
 and  vertigo (U.S. EPA, 1987a).

     Frequent  laboratory  findings Indicative of  severe hemolytlc anemia  after
 naphthalene poisoning 1n  persons  of all ages  Include  depressed hemoglobin,
 hematocrU,  and erythrocyte  count;  elevated   leukocyte  and  retlculocyte
 counts;  erythrocyte anlsocytosls,  polychromatophHla, fragmentation, sphero-
.cytosls,  and  mlcrospherocytosls;  and  occasional hemogloblnuMa  (U.S.  EPA.
 1987a)r   Additional   laboratory  findings  Indicative  of  severe  hemolytlc
 anemia  after   exposure  of neonates  or   children  to   naphthalene  Include  the
 following:  erythrocyte   polkllocytosls   and  mlcrocytosls;  elevated  serum
 blllrubin;  occasional observation  of Heinz  bodies,  nucleated  erythrocytes,
 and   Howell-Jolly  bodies; and   occasional . observation   of  methemogloblnurla
 (U.S. EPA,  1987a).

     In  most  studies  of  persons who have developed  severe  hemolytlc anemia
 after exposure  to  naphthalene,  treatment with  blood transfusions, treatment
 with blood transfusions  plus  alkali therapy,  or observation without either


 04430                                VI-4                           05/15/91

-------
 of  these  treatments  has  led  to  complete patient recovery  with  no observed
 complications.   However,  deaths have been observed after naphthalene-Induced
 hemolytlc  anemia  (U.S.  EPA,  1987a),  but  It 1s  unknown whether  they  were
 directly related  to  the exposure.

     Other  Routes.  In addition  to  the cases of  acute  naphthalene poisoning
 following   oral   exposure,   similar  effects  have  been  reported  following
 exposure   to  naphthalene  by  other   routes:   combined dermal  absorption and
 Inhalation of  naphthalene  vapor by  neonates  (GMgor  et  a!., 1966; Nalman and
 Kosoy,  1964;  Valaes  et al.,  1963;  Dawson et al., 1958;  Cock, 1957; Schafer,
 1951)  and  adults  (Younls  et al., 1957);  Inhalation  of  naphthalene vapor by
 neonates  (Hanssler,  1964;  Irle,  1964);  Inhalation of naphthalene vapor  by a
 child  and  adults  (L1n1ck,  1983); and transplacental ^exposure of  the fetus to
 naphthalene that  had been Ingested  by  the mother (Anzlulewlcz et al.,  1959;
 Zlnkham and  ChUds,  1957,   1958).   These  exposures  occurred  via combined
 dermal  absorption and  Inhalation for a  few days of naphthalene vapor   from
.apparel and bed  clothing  that  had  been  stored  1n mothballs; Inhalation of
 vapor  'from  a  naphthalene-containing  medication; Inhalation  of naphthalene
 vapor  for  several years from excessive  numbers  of mothballs kept  throughout
 the home;  and transplacental  exposure,  for about  3 months,  of fetuses to
 naphthalene Ingested by the  mother.

     Two groups of Individuals  have been  shown  to be especially susceptible
 to  naphthalene-Induced hemolytlc anemia:
     1.  Persons whose erythrocytes are deficient  1n  glucose 6-phosphate
        dehydrogenase  (G6PDH)  or  persons  1n whom  erythrocyte  GSH   Is
        rapidly depleted  by certain  oxldant  chemicals (U.S.  EPA,  1987a).
        The precise mechanism  by which GSH  1s depleted  or a  deficiency
        of  G6PDH  leads  to naphthalene-Induced hemolysls  In these  cases
        Is  not  clear.
 04430                                 VI-5                           05/15/91

-------
     2.  Neonates   (U.S.  EPA,  1987a).   The  sensitivity  of  neonates  to
        naphthalene  1s  explained In part  by the same factors that confer
        sensitivity   to  children  and  adults;  namely,  G6PDH deficiency
        and/or  diminished levels  of  GSH  as described above.  Additional
        naphthalene   sensitivity   1n  newborns   may  be  conferred  by  the
        Immaturity  of   pathways   necessary   for   the  conjugation   and
        excretion  of naphthalene  metabolites  (Valaes  et  al.,  1963).
        Evidence  for the  latter hypothesis 1s  suggested  by the  finding
        that  glucuronlde  excretion by  human   newborn  Infants   Increased
        gradually  during  the  first week  of  life  and  that  the  Initial
        levels  and  the  rate  of  Increase  were  lower   In  the   premature
        Infant  than  1n the full-term  Infant  (Brown and Burnett,  1957).


     Fanburg  (1940)  described  the case of a man who had developed an  allergic

 reaction to naphthalene  from  clothing  that   had  been stored  In  mothballs.

 The reaction  was  an  exfollatlve dermatitis resembling  mycosis  fungoldes.

 The elimination  of naphthalene  from the patient's  environment  resulted  1n

 prompt   recovery,  which  lasted  uninterruptedly  during  a 7-year  period  of

 observation.
     Case  studies  that describe the presence  of  cataracts 1n persons exposed

 to naphthalene by the oral, dermal or Inhalation routes  have been summarized
                                                                            •
-1n an  ambient water  quality criteria  documnt  for  naphthalene  (U.S.  EPA,

 1980c).'  Ghettl and Mar1an1 (1956) associated the occurrence of cataracts  In

 8/21  workers  with  naphthalene  exposure  1n  a  manufacturing  plant.   Other

 cases  of  occupational Instances  of cataract formation have been described  by

 Hollwlch  et  al. (1975).



     Possible preneoplastlc epidermal  changes  1n humans have been associated

 with dermal  exposures  to  benzo[a]pyrene  (Cottlnl  and  Mazzone,  1939;  Klar,

 1938;  Rhoads et al., 1954).  A group of 26 patients was  given dally applica-

 tions  of  a 1% solution of benzo[a]pyrene on protected  (thigh) or unprotected

 (upper  extremity)  skin.   These  Individuals were hospitalized  for  a variety





 04430                                 VI-6                            08/09/91

-------
of conditions  Including pemphigus  vulgaMs, mycosis  fungoldes,  prokeratosls,
psoriasis, xeroderma  plgmentosum,  basal cell  cancer,  squamous cell  cancer,
lupus erythematosls,  syphllUs  (1n various  stages)  or ringworm.   Treatment
was for ~4 months and resulted  in  a series  of  progressive  alterations 1n  the
skin proceeding from erythema and  pigmentation to the  development of  lesions
described  by  the  authors  as   verrucae.    The   skin   changes  and  lesions
regressed within 2-3 months after  cessation of treatment.  The  authors  noted
that the  Individual  with  xeroderma plgmentosum  did  not respond  differently
from the other subjects (Cottlnl and Mazzonl,  1939).   Klar  (1938) and Rhoads
et al.  (1954)  reported  potentially deleterious epidermal  changes 1n  men  who
had experienced accidental dermal  exposures  to  benzo[a]pyrene.

Ep1dem1oloq1c Studies
    Oral.  There  are  no   studies  of  oral  exposures  to  Identified  PAHs  In
the literature.  PAHs are  known to occur In  some foods as  by-products of  the
preparation  process  and  1n   others   as  a   consequence   of   environmental
pollution, but  their role as  dietary  causes  or contributors  to cancer  can
only  be assessed  Indirectly In  epldemlologlc  studies  of nutrition.  This
body of data Is not specific  for PAHs  and beyond  the  scope  of  this document.

    Other  Routes.   Epldemlologlc  studies   Unking   PAHs   to   human   health
effects  have  studied  human  exposures  as   they actually  occurred.    These
studies  can,  therefore,  be  evaluated  to  determine the  health  effects  of
exposure only to mixtures  of PAHs  by  Inhalation  and  dermal  contact.  Occupa-
tional  exposures  to  PAH-conta1n1ng   complex   mixtures  and   70  Industrial
processes that generate PAHs  have been evaluated by the IARC  Working Group.
It 1s not  possible  to determine from these  studies  the effect  of Individual

04430                                VI-7                           05/15/91

-------
 PAHs.   Toxic effects  Include  a variety  of  skin lesions and  noncancer  lung
 diseases  such as  bronchitis.

     The  compiled  case  reports  and  large cohort  studies  of exposures  to
 complex  mixtures  of  PAHs  have shown  strong cause-effect  relationships  In
 that high  rates  of  cancer  can be  associated- with  exposure, dose-response
 gradients  can  be demonstrated and  results  are  consistent across  time and
 geographic  location.  The  data Include  thousands  of case reports spanning
 over 200 years and  cohort  studies of  cancer  Incidence and  mortality  from
 several  countries.  Data-summarized In IARC  (1984)  monographs of polynuclear
 aromatic  compounds  demonstrate cardnogenlclty  to  humans  frpm  Industrial
 exposures  to mixtures  of  PAHs 1n  aluminum production,  coal gasification,
 coke production  and  Iron  and steel  founding.   Bitumens,  coal   tars  and
 derived  products,  shale oils  and  soots  have also been  evaluated  for  their
 human  health effects  (IARC, 1985).   A  detailed evaluation  of  PAH mixtures
 can be  found 1n the document,  "Carcinogen Assessment of Coke  Oven  Emissions"
.(U.S.  EPA,  1984b).   Evidence  of  human cardnogenlclty based  largely  on the
 IARC review mentioned Is  summarized  1n  Table VI-1.   The extent  to which the
 data derived from  these  types  of  exposure  will be  applicable  to drinking
 water  exposures  1s unknown and  creates  the  major  difficulty  for the risk
 assessment.

     Despite the high  levels and large proportion of PAHs,  other  exposures to
 gases  and metals are  always present In these occupational  environments.  For
 example,  coke  oven  emissions contain  other carcinogenic  agents   Including
 arsenic,  beryllium, chromium,  nickel, 2-naphthylamlne and  benzene.  Asbestos
 1s  a possible  exposure  for chimney sweeps (soot).  Cocarclnogens  (compounds


 04430                                VI-8                           05/15/91

-------
                                                                             TABLE VI-1

                                                    IARC  Determinations of PAH Carclnogenesls Based on Human Oataa
           Agent or Process    '   Route''
                  Sites
                 Human*
                    Carcinogenic  Factor*1
                                                   Data Base
Coal
gasification




Inhalation,'
derMl




lung

bladder

skin*

sufficient

Halted

sufficient

Coal tar fro* the destructive
distillation of coal.
Older coal gasification pro-
cesses.
Coal tar fro* the destructive
distillation of coal.
Analysis of 3753 skin tumors, bladder
cancer deaths by occupational group;
cohort studies of gas Industry.



           Coke production      Inhalation,
                                derMl
           Aluminum
           production
           Bltunens
           Coal tars and
           derived products
Inhalation
Inhalation,
dermal

Inhalation,
derMl
lung
skin

kidney

Intestinal
pancreatic

lung
bladder
pancreas

lung
skin

skin
                                                skin
                               sufficient
                               sufficient

                               Halted

                               Inadequate
Hatted
Halted
Inadequate

Inadequate
Inadequate

sufficient
                               Halted
Possible coal tar fume.
Coal tar froa destructive
distillation of coal.
Coal tar froa destructive
distillation of coal.
Coal tar froa destructive
distillation of coal.

Aluminum production Industry.
possibly pitch fume.
Bitumens only.
Coal tars froa destructive
distillation of coal, coal
tar pitches.
Creosotes.
                                                 Analysis of 3753 skin tumors, bladder
                                                 cancer deaths by occupational group;
                                                 cohort studies In coking plant, In steel
                                                 Industry and coke oven workers.  Cohort
                                                 studies In five countries; two case-
                                                 control studies.
Cohort studies In aluminum production
plants In USSR. USA. Canada.  Norway
(21.829 workers In USA).

Cohort study of roofers. Includes coal
tar exposures.

Numerous case reports of occupational and
pharmaceutical exposures; cohort study of
roofers.
to

-------
                                                                       TABLE VI-1 (cont.)
o
j*.
-Oi
CO
Agent or Process
Shale oils
Routeb
Inhalation.
dermal
Sites
skin
Humanc
sufficient
Carcinogenic Factor^
Shale oils.
Data Base
Case reports; occupational mortality;
cohort study of shale oil workers; mule-
skinners cancer.
        Soots                Inhalation.     skin           sufficient     Soot.                              Historic and contemporary case reports
        (chimney origin)     dermal                                                                           from various countries; cohort studies of
                                                                                                              sweeps In four countries.


        *TMs  table  Is summarized from IARC (1984. 1985).

        "Both  Inhalation  and dermal exposures are possible  In most cases.   Skin contact Is most likely for coal  tars and  derivatives,  shale  oils  and
          soots and 1n occupations where these products are used such as textile manufacturing (mule skinners).

        cThe  preamble to IARC  document  states  "The evidence  for  carclnogenlclty from  studies  In  humans  1s Judged  to  fall Into one of  four  groups.
          defined as  follows:  1) Sufficient evidence of carclnogenlclty  Indicates  that  there 1s a causal relationship  between  the  exposure and  human
          cancer;  2)  limited evidence of  carclnogenlclty  Indicates that a causal  Interpretation Is credible,  but  that  alternative  explanations,  such
          as  chance,  bias  or confounding,  could not  adequately  be  excluded;  3) Intdeqmte evidence,  which applies  to  both  positive and  negative
          evidence.  Indicates that  one  of  two  conditions  prevailed:  (a) there  are few pertinent  data,  or (b)  the  available studies, while  showing
          evidence  of association,  do  not exclude chance,  bias or  confounding;  4) Ho  evidence applies  when several  adequate "studies are available
          that  do not show evidence  of carclnogenlclty.

        dpAHs  have been measured In  the occupational setting.  All  processes  listed Included  exposure  to  complex mixtures Including  at  least  seven
          of those PAHs of concern In this document (see Chapter II for occurrence).

        eCase  reports of PAH associated skin cancers Include a large proportion of scrotal cancers.
ro

-------
that   enhance   tumor  production   In   experimental   animals  when   applied
concurrently with  carcinogens) and  tumor promoters,  both  PAH and  non-PAH,
may also be components of the mixtures.

    Data published  since the  IARC  summary continue  to provide evidence  of
the  cardnogenldty  of  PAH  mixtures  from  Inhalation and dermal  exposures.
Hansen  (1983)  conducted  a  5-year mortality  follow-up among  Danish  chimney
sweeps  and  noted a  2-fold  overall  excess compared  with  males employed  In
other  occupations.   When the  excess  was examined by  age  and cause,  It  was
found  that  the high mortality  of  the sweeps 1n  the  older  age group  (45-74
years)  was  exclusively  due  to an excess  number  of  deaths  from cancer  and
1schem1c heart  disease,  whereas the excess  among the younger sweeps  (15-44
years)  was  due to other causes.   Those sweeps dying of cancer or  Ischemlc
heart disease had been occupatlonally exposed for  an  average  of 30  years.

    Auto  mechanics   are  considered  to  be   another   group  at  risk   of
occupational  exposure  to PAH  mixtures  but have  not  been  as   thoroughly
studied as  chimney sweeps.   The recent work of Hansen  (1989)  reports  on  the
10-year mortality among  a cohort of  Danish  auto  mechanics.    Overall, a  21%
Increase  1n  total   mortality   was  observed  1n  relation  to  a   comparable
population of  skilled  workers  presumed to be unexposed  to PAH mixtures.   In
addition to  the moderate Increase 1n cancer mortality,  attributed  primarily
to an  excess  of pancreatic  cancer,  a  statistically  significant Increase  1n
Ischemlc heart disease deaths was  found.

    Gustavsson  et  al.  (1988)   conducted  a study  of  cancer  Incidence 1n  a
cohort  that  comprised nearly  all  Swedish chimney sweeps  employed any  time


04430                                VI-11                           09/25/91

-------
 between 1918  and  1980.   Earlier mortality  analyses  among  the  cohort  had
 shown excess mortality from cancers  of  the  lung,  esophagus,  liver, and from
 leukemia.   Increased mortality  from Ischemlc heart  disease  was also found.
 This  study of  Incident  tumors  found a  significant  excess  of cancers of the
 lung,  esophagus,   and   bladder,  which  contributed   to  an  overall  excess
 Incidence   In  the  cohort.   A  nearly  significant  Increase  of hematopoetlc
 malignancies   that  Included  both multiple   myeloma  and  leukemia  was  also
 observed.   H1stolog1c examination attributed  the  lung cancer Increase to an
 excess  of  squamous  cell  and  und1fferent1ated/small cell carcinomas.   The
 excess  risk   for   the  solid  tumors depended  mainly  on  an  excess  during
 followup for more  than 30 years  from start of  exposure.

    S1em1atyck1  et al.  (1988)  explored potential  occupational exposures as
 risk  factors  for  about  20 cancer  sites  In  a   population-based  Incident
 case-control   study  In   Montreal.    The  study  was  designed  to  generate
 hypotheses  for  further  analytic study.   This  particular  report examined
-associations   with  occupational  exposures   to  10  types  of  exhaust  and
 combustion  products and  the cancer  sites of  Interest.  The 10  exposures were
 classified  as  follows:  four  were   exhaust  products  of  Internal   combustion
 engines,  distinguished by  the  type  of  engine fuel  used (gasoline,  dlesel,
 jet   fuel  or  propane),  and  the remaining  six were products  derived  from
 "nonenglne" combustion  of  coal, coke, wood,  liquid  fuels  (heating  oil,
 kerosene,  naphtha  and  lamp oil), natural  gas and  propane.   Among the many
 relationships  examined  In  the  study, several  are  of particular  Importance.
 A major finding  was the association between squamous-cell  lung  cancer and
 both  dlesel and  gasoline  exhaust.   Five other  relationships  were noted and
 considered  by the  Investigators  to  warrant  further study regarding the role


 04430                                VI-12                            08/09/91

-------
of  PAH-contalnlng  mixtures   1n  the  etiology  of  human  cancer.   The  most
promising  leads  were the following:  1)  the effects of exposure  to gasoline
and dlesel  exhaust on the occurrence of colorectal cancers;  2)  the effects
of exposure  to gasoline exhaust on  the occurrence of kidney  cancer;  3) the
effects  of  exposure  to coal   combustion  products  on  the  occurrence  of
pancreatic cancer  and  possibly  on  nonadenoma lung  cancer;  4)  the effects of
exposure  to  combustion  products   of  liquid   fuels  on  the  occurrence  of
prostatlc  cancer;  and  5) the effects of exposure  to  natural-gas combustion
products on the occurrence of bladder cancer (S1em1atyck1  et al.,  1988).
                         w
    The  findings  from  these  four studies provide  further  Important evidence
that  occupational  exposure  to complex  PAH  mixtures, which  occurs  primarily
via Inhalation or  dermal  contact,  may also result  1n  cancers  at  sites other
than  those  logically expected.  I.e.,  lung  and  skin.  It 1s unclear  at this
time  as  to  what  the  mechanism  of  exposure  to  the target organs  might
Involve.   Gustavsson  et al.  (1988)  propose that,  for chimney sweeps,  PAHs
                                                                            *
probably  enter  the  body  through direct  resorptlon In the bronchi,  or  are
expectorated,  swallowed and  absorbed  through  the gastric  and  Intestinal
mucosa.  It Is likely that both routes of uptake are relevant.

    The  reported  Increase  1n  1schem1c  heart   disease   In  three  of  these
studies  (Hansen, 1983, 1989; Gustavsson  et  al.,  1988)  1s  quite striking when
viewed 1n light of the normal expectation of  a  "healthy worker effect".  The
usual  expectation  1n an  occupational study  of morbidity  and/or mortality 1s
that  there  will  be  a  depression of  the standardized morbidity  ratio (SHR)
for  cardiovascular  diseases.    This  occurs  due   to  pre-select1on  of  the
workforce  for  fitness  for  physical  labor.    Further  studies .specifically


04430                                VI-13                           08/09/91

-------
 designed  to address  the  Issue  must be carried  out  to determine  1f  this  1s
 Indeed  a  true  disease excess and not an artifact of the study methods.

 Sensitive Populations
     Data  to determine what Individuals within the general  population may be
 more susceptible to PAH-lnduced toxlclty  are  generally  lacking.   One factor
 that Increases an  Individual's  risk for  lung cancer 1s  cigarette smoking,
 which  1s  suspected to act  synerglstlcally with  other occupational exposures
 such as  asbestos  and uranium mine  atmospheres.   Although  high  exposure to
 sunlight  and low level  of skin  pigmentation are well documented risk factors
 for  skin  cancer,  no  data exist to  determine  whether Individuals with these
 risk factors are more susceptible  to PAH-lnduced skin lesions.

     Aryl  hydrocarbon hydroxylase  (AHH)  1s  the  term given  to  the enzyme or
 enzyme  systems  Involved  In  the  oxldatlve metabolism  of  PAHs,   especially
 benzo[a]pyrene.    Genetically  determined   Individual   differences  In  the
.ability  to Induce  both  Increased levels  and  Isozymes  of  AHH  have  been
 proposed  as a  possible reason for differences among humans  In  susceptibility
 to  lung  cancer  (for  further  discussion  see Chapters  III and  VII).  Some
 early studies  claimed that persons  with  lung  cancer had higher  InduclblHty
 of  AHH  In cultured  lymphocytes,  but   subsequent   studies   have  not  all
 supported this claim (Perera,  1990).   Busb.ee  et al. (1980) present  evidence
 supporting  the  need  to  analyze  AHH 1n more  than one  tissue  and recommend
 measuring  1ndudb1l1ty  In  alveolar macrophages and  lung tissue  as  well as
 lymphocytes.  More  res.earch  will  be  required to  determine the  sensitivity
 and specificity of this enzyme system as  a  predictor  of  human  susceptibility
 to lung cancer.

 04430                                VI-14                           05/15/91

-------
    Cardnogen-DNA adducts may ultimately prove to be helpful 1n Identifying
those  persons  at  high  risk of  adverse outcome following  exposures  to PAH
mixtures.   It  Is  currently  possible to  detect  PAH-DNA adducts  In several
types of human tissue and sera In  those  known  to be  either  occupatlonally or
environmentally  exposed  (Herbert  et  al.,  1990;   Hemm1nk1  et  al.,  1990;
Perera,  1990).   However, the  role of  adducts- 1n  Inducing  cancer  or  other
adverse outcomes Is still a matter of much  scientific discussion.   Long-term
                                   •'*'"..""'              •
validation  studies are required  1n appropriate  human  populations  before
wide-scale  monitoring   for  these  blomarkers  can  be considered  as a  more
refined and  specific way of predicting  who may or  may not be at higher risk
(I.e., more susceptible) of  a particular outcome.

Summary
    With  the  exception  of   naphthalene,  which  most commonly  causes  acute
hemolytlc anemia,  no  data were  located  directly pertaining to PAH toxldty
to humans by  the oral  route of  exposure.  A large body of  literature exists
on toxldty  and  carclnogenldty  to humans of a variety of  mixtures contain-
ing PAHs.   The  summarized IARC data  clearly demonstrate that Inhalation and
dermal  exposures  to complex mixtures  containing  PAHs result  In lung, skin
and other cancers.  The  most recent  occupational studies continue to provide
compelling  evidence regarding  the  carclnogenldty  of  PAH  mixtures.   The
Increased risk  of  cardiovascular disease 1s noteworthy  and must be studied
further.  In  the  occupational environments  studied,  some  airborne PAHs may
be  cleared  by mucodllary  action and  swallowed.   This  Information  may be
useful  for  determining   the potential carclnogenldty  to humans of Ingested
PAHs.
04430                                VI-15                          05/15/91

-------
    One must  be cautioned  that  using  these data  to  predict  the effects  of
exposure to any single PAH 1s  Inappropriate.   Furthermore,  most 1f not  all
PAH exposures  occurring  to humans are  1n  the form of  a  complex mixture  of
the compounds,  1t  seems  both  unrealistic  and  unnecessary  to  attempt  to
assess  the  potential  role of  each  Individual  compound.   Studies  to  date
clearly Implicate  this chemical  family as  human  carcinogens at  relatively
Intense, high level exposures.  Recent work  has also  more  clearly  delineated
the role  of  PAHs  and benzo[a]pyrene  1n particular  as causal  carcinogenic
constituents  In the  highly complex mixture  known  as  tobacco smoke  (Wynder,
1988).  While  all  the Information summarized 1n  this  chapter  Is  Important
for evaluating  and  clarifying the role  of  PAH mixtures and  their  relation-
ship to cancer and cardiovascular disease 1n  humans,  the direct  relevance of
the  Information  to   the   assessment  of  any  adverse  effect  arising  from
1ngest1on  of PAHs  In drinking  water remains  to be  determined.
04430                                VI-16                           09/25/91

-------
                         VII.  MECHANISMS OF  TOXICITY

    The carcinogenic  potential  of  PAHs has  been  a  subject of study for >50
years.  They  were  among the first  chemical  agents  shown to  be  tumorlgenlc,
and  their  role  1n the  causation  of  human  cancers  has  been a subject of
                                  •
speculation almost  since  Pott's (1775) first description  of  "soot wart" In
chimney sweeps  1n  the 18th century.   Noncarclnogenlc health effects of  PAHs
have  not  been  studied  extensively;  mechanisms  of  PAH  toxlclty apart  from
their ability to Induce neoplasla are,  thus,  not  well  elucidated.

Mechanisms Involved 1n PAH Carc1noqen1c1ty
    Structure  Activity  Relationships.   Many   physico-chemical  and  enzy-
matic  parameters  must  be  considered  with  respect  to  PAH carclnogenldty.
Solubility and  Intracellular localization proximate  to metabolic enzymes are
likely to  be  Important determinants of  the  cardnogenlcUy of a particular
PAH.

    Following the  Identification of  the first carcinogenic hydrocarbon  from
soot, benzo[a]pyrene,  an Intensive effort was mounted to Isolate the various
active components  of  carcinogenic  tars  (IARC,  1973).   It became apparent
that  carcinogenic  PAHs  are structurally  derived  from the  simple angular
phenanthrene  nucleus  (Pullman   and  Pullman,   1955;  Arcos  and Argus,   1974).
UnsubstUuted PAHs  with  fewer  than  four condensed rings that  have  been
tested have so  far  not  shown tumorlgenlc activity.   Furthermore, of the six
possible  PAHs with arrangements with four benzene rings, only  three of these
compounds    are    active   as   carcinogens,   namely   benzo[c]phenanthrene,
benz[a]anthracene  and  chrysene.   The  unsubstUuted penta-  and hexacycllc


04440                                VII-1                           05/07/91

-------
aromatic  hydrocarbons  are  generally   the  most  potent carcinogens  of  the
series.   These   Include  benzo[a]pyrene,  d1benz[a,h]anthracene,   dlbenzo-
[a,h]pyrene,   d1benzo[a,1]pyrene,   d1benzo[a,l]pyrene,  d1benzo[a,e]pyrene.
benzo[b]fluoranthene, and benzo[j]fluoranthene.  Less  potent  carcinogens  are
the  dlbenzanthracenes   and   dlbenzophenanthrenes.   Only  a  few  heptacycllc
hydrocarbons  show  carcinogenic  activity,  Including  phenanthro[2',3':3,4']-
pyrene, peropyrene,  and  d1benzo[h,rst]pentaphene.   There are very  few  known
carcinogenic  hydrocarbons with  more than seven unsubstltuted aromatic  rings
(Santodonato et al.,  1981).

    Active  carcinogens  are  also  found among  those PAHs  containing a  non-
aromatic  ring.   Examples  of   this   type  of  compound   are  cholanthrene,
1,12-ace-benz[a]anthracene,   8,9-cyclopentanobenz[a]anthracene,   6,7-acebenz-
[a]anthracene,     acenaphthanthracene,      l,2,5,6-tetrahydrobenzo[J]cyclo-
pent[f,g]aceanthrylene,  and  "angular"  steranthrene.   All of  these  compounds
contain an Intact conjugated phenanthrene moiety.

    Alfcyl  substitution  at  certain  positions  In the  ring  system of a  fully
aromatic hydrocarbon has been observed  to confer carcinogenic activity  or  to
enhance carcinogenic potency.   Arcos  and Argus (1974)  noted  that monomethyl
substitution  of  benz[a]anthracene  makes these dervatlves  potent  carcinogens
1n mice.  Potency depends on the position of substitution  1n the decreasing
order  position  7 >  6  >  8*12 >  9.  A  further  enhancement  of  carcinogenic
activity 1s  produced by appropriate dimethyl substitution  of benz[a]anthra-
cene.   Carcinogenic  compounds are  produced  by 6,8-d1methyl-,  8,9-dlmethyl-
8.12-dlmethyl-,  7,8-d1methyl-,  and  7,12-dlmethyl-substHutlon.    The   last
04440                                VII-2                           05/07/91

-------
 compound   1s   among   the   most   potent   PAH   carcinogens  Identified   1n
 cardnogenesls  bloassays.    It  has  not  been  shown, however, to  occur as  a
 product   of   fossil   fuel   pyrolysls   or  to   be  a  major   environmental
 contaminant.   Carcinogenic  trlmethyl-  and tetramethyl-benz[a]anthracenes  are
 known,   and   their  relative  potencies  are  comparable   with   the   parent
 7,l2-d1methylbenz[a]anthracene.  Methyl  substitution  1n the angular  ring of
 benz[a]anthracene   tends   to  reduce  the  carcinogenic   potential   of   the
 molecule; 4,5-d1methylbenz[a]anthracene may be an exception In  this regard.

    Alkyl substitution of partially  aromatic  condensed  ring systems may  also
 enhance  the  carclnogenlcHy of  a  compound.  3-Methylcholanthrene, a  highly
 potent carcinogen,  1s the best example of this type.

    Data  derived  largely  from  skin  painting studies  have shown that  the
 carclnogenlcHy  of PAHs  tends  to  decrease  with  alkyl substHuents  longer
 than  methyl,   possibly   due  to  a  decrease   1n  transport   through   cell
 membranes.   Benz[a]anthracene  1s  especially  sensitive  to  decreased cardno-
 genlclty  caused by  the  addition of  bulky substHuents  at the  7-pos1t1on.
 This  observation lent credence  to  the  view  for most polycycllcs that  high
.reactivity  of  the  mesophenanthrenlc region  (called  the  "((-region*)  was  a
 critical  determinant  for  carclnogenlcHy.   Other studies,  however, show  that
 the  K-reg1on  may   not  be  Involved  1n  binding   to DNA,   a step  considered
 Important Vn  the carcinogenic  process  (Blobsteln et al.,  1976; Welnsteln et
 al., 1976; Hoore et al.. 1977).

    Partial  hydrogenatlon  of   the   polycycllc   aromatic   skeleton  has  been
 observed  to  decrease  the  carcinogenic  potency of some  PAHs.  This was noted

 04440                                VII-3                          10/03/91

-------
for  hydrogenated  derivatives  of  benzo[a]pyrene,  benz[a]anthracene,  and
3-methylcholanthrene.   By constrast, the cardnogenldty of d1benz[a,h]anth-
racene,  d1benzo[a,1]pyrene,   and   d1benzo[a,h]pyrene   1s  not  significantly
altered  by  mesohydrogenatlon,  probably  because  extensive  resonance  Is
preserved   1n   the   molecule.   Moreover,   5,6-d1hydrod1benz[a,h]anthracene
actually displayed a  4-fold  Increase  1n  cardnogenlcHy  by comparison with
the parent  unsaturated hydrocarbon  (Arcos  and  Argus,  1974), possibly due to
the hydroph1l1c1ty  and ease  of Intracellular  transport  of Us dlhydrodlol
derivative.

    For  many  years,   Investigators  have  sought  a common  molecular feature
among PAH  carcinogens  that  would  serve to  explain  their  biologic  activity.
The "electronic theory of cardnogenesls"  has relied upon an analysis of the
Influence  of   electron  density  at  specific molecular  regions  to explain
unique  reactivity  with cellular  constituents.   A basic  assumption arising
from  the work of  the  Pullmans and  others  (Pullman and  Pullman,  1955) was
that  a  mesophenanthrenlc  region  ("K-reg1on")  of  high  ^-electron density
and with  a  propensity  for  addition  reactions  was  a  critical  structural
feature  for  polycycllc  carcinogens.   In   an expansion of  this hypothesis,
further  biologic significance  was  attributed to  the concomitant presence of
a  rather  unreactlve  meso-anthracenlc  region ("L-reg1on")  for  high cardno-
genlcHy.   In  addition,  a   region  of  comparatively   low  reactivity,  which
characteristically undergoes   metabolic  perhydroxylatlon  (corresponding  to
the 3,4-pos1t1ons  of  benz[a]anthracene),  had  been  designated  the  M-reg1on.
According  to  the  theory,  only  binding  of  the  K-reg1on to  critical cellular
sites would Initiate  tumor  formation.   Binding at  the L-reg1on could cause
no  tumorlgenlc effect,  while  1nact1vat1on  would  be  produced  by  metabolic


04440                                VII-4                          12/31/90

-------
perhydroxylatlon  In  the  M-reg1on.   The  three  regions  of  reactivity are
Illustrated below 1n the benz[a]anthracene skeleton:
                                               H-reg1on of metabolic
                                               perhydroxylatlon
                                            K-rtglon
The electronic  K-L  theory of carcinogenic reactivity has encountered  numer-
ous Inconsistencies, primarily because  these  relationships were  derived  from
physico-chemical properties of the parent hydrocarbon and gave no  considera-
tion to the biologic effects of metabolites.

    That many  chemical  carcinogens require  metabolism to reactive  electro-
phlUc  forms  for  their  activity  1s generally accepted.   PAHs are  certainly
1n the  class of those carcinogens  requiring  so-called  "metabolic activation"
for binding to critical macromolecules.  The  realization that  this  metabolic
activation  to  reactive  Intermediates  was  a  necessary first  step has  made
possible  an understanding  of  some  of  the  Inconsistencies  encountered  1n
structure-activity theories, such as K-reglon binding, that are  based  solely
on a consideration of the parent compound.

    The  metabolic  processes   undertaken  by  mammals  exposed  to  PAHs  are
described  1n  Chapter  III.   In   general,  they  consist  of  oxidation  steps
catalyzed by  cytochrome  P-450-assodated  enzymes.  Products  Include  epoxlde
Intermediates,  dlhydrodlols,  phenols,  qulnones  and  combinations  of  these.
These  oxidized  forms of  PAHs have  been  shown  to  exhibit various  biologic


04440                                VII-5                           05/07/91

-------
activities.   For  example,  various  benzo[a]pyrene  epoxldes  (Including  the
K-reg1on  epoxlde),  phenols,  qulnones,  dlols  and  dlol-epoxldes  have  been
shown  to  be  mutagehlc   1n  several  mammalian  cell   and  bacterial   assays
(Gelboln, 1980; IARC, 1973; Schoeny et al.,  1985;  Chesls  et  al.,  1984).

    Research 1n recent years has focused attention on  the potential  reactiv-
ity  of  d1ol-epox1de metabolites  of  PAHs,  and their  ease  of conversion  to
trlol carbonlum  Ions.   The "bay-region" theory proposes  that  d1ol-epox1des,
which are more readily  converted  to carbonlum Ions, will be  better  alkylat-
Ing  agents  and thus mutagens  and  Initiators  of  cardnogenesls  (Jerlna  and
Lehr, 1977;  Wood et al.,  1979).   Examples  of  a  bay-region In a  polycycllc
hydrocarbon  are  the regions  between the 10 and 11  positions of  benzo[a]-
pyrene and the 1  and 12 positions  of benz[a]anthracene:
             Bay-region
                                                  9
               /                                       8 r  ,7  u 6
             Benzo[a]pyrene                         Benz[a]anthracene
The  theory  predicts that  dlol-epoxldes 1n  which the  oxlrane oxygen  forms
part of  a  bay-region (such as benzo[a]pyrene 7,8-d1ol-9,lO-epox1de) will  be
more reactive  and hence more  carcinogenic than  d1ol-epox1des  1n which  the
oxlrane  oxygen  1s  not  situated  1n  a  bay-region.   The  unique  structural
feature  of  the  d1ol-epox1des  appears  to  be  that the  epoxlde  Is  on  a
saturated angular benzo-rlng,  which forms part  of  a bay-region on the  PAH.
Perturbatlonal  molecular  orbital   calculations, which  predict   ^-electron
energy changes.  Indicate that the  epoxldes  on  saturated benzo-rlngs  (which
form part of the  bay-region of a hydrocarbon)  undergo ring  opening to  form a

04440                                VII-6                          11/13/91

-------
carbonlum  1on much  more  easily than do nonbay-reglon epoxldes  (Pelkonen  and
Nebert,  1982).    Synthetic   bay-region  d1ol-epox1des  of  benz[a]anthracene,
benzo[a]pyrene,  and  chrysene have been shown  to  be more mutagenlc In  vitro
and/or  tumoMgenlc  than other   dlol-epoxlde  metabolites,  their  precursor
dlhydrodlols,  the  parent  hydrocarbons,   or  other  oxldatlve  metabolites.
Moreover,  quantum mechanical  calculations  provide support  for  the  concept
that  reactivity   at  the  bay-region 1s  highest  for  all  the  dlol  epoxldes
derived from polycycllc hydrocarbons.

    The  bay-region  concept  has  received  enough  confirmation   to  lead  to
                                           i
suggestions that an analysis of theoretical  reactivity  1n  this  manner  may be
useful  In  screening  PAHs as  potential  carcinogens  (Smith et  al.,  1978).
Among  several Indices  of theoretical reactivity examined, the presence  of  a
bay-region for a  series  of  PAHs displayed a high  degree of  correlation with
positive carcinogenic activity (Table VII-1).

    It  1s  possible to predict what would  be  an  ultimate carcinogenic  form
from  the  bay-region theory  and  a consideration  of a PAH structure.   There
1s, however, no way at the  present  time  to  predict whether  a  particular  PAH
will,  1n   fact,  be metabolized  to  a  bay-region  dlol-epoxlde  1n any  given
mammalian  tissue.   For   example,  benzo[e]pyrene  1s  generally  considered
noncardnogenlc.   Neither the  parent compound nor  the  9,10-dlhydrodlol,  the
presumed   bay-region   d1ol-epox1de   precursor,   has  significant  Initiating
potency 1n mouse  skin  (IARC, 1983).  It has  been  shown  that neither cultured
mammalian  cells  nor rat liver mlcrosomes  metabolize benzo[e]pyrene  to  the
9,10-dlhydrodlol;  furthermore,  9,lO-d1hydrod1ol  1s not  metabolized  1n these
systems  to   the   expected   dlol-epoxlde.   When   synthetic  benzo[e]pyrene
9,10-dlhydrodlol  was Injected  1n  newborn  mice,  hepatic  tumors  were observed,
04440                                VII-7                           12/31/90

-------
suggesting  that  the  neonatal  "liver  possessed  some  metabolic  capacity not
extant  1n mouse  skin  or  Yn vitro systems (Buenlng et al.,  1980).  Synthetic
trans  and  ds  9,lO-d1ol-ll,!2-epox1des  produced   hepatic  and  pulmonary
tumors,  respectively,  1n   newborn  mice  (Chang  et  al.,  1981).    Thus,  a
bay-region d1ol-epox1de of  benzo[e]pyrene appears to  be carcinogenic, but  1s
not formed 1_n vivo under  normal  circumstances.

    Various researchers have  attempted  to refine and extend  the K- or  bay-
region  theories  to  allow  for predlctlvlty of  carcinogenic  potential from a
consideration of PAH  structure.   For  example,  the  reactivity Index  used  by
Mohammed  (1983)  established a  correlation  between  the  K-reglon reactivity
and  the bay-region  theory to  determine PAH  metabolic   products.   A  later
development considered contribution of  both  slgma and pi-electronic  systems
when calculating bond  superdelocaHzabllUy (Mohammed, 1985).  The potential
of  PAHs  for  one-electron   oxidation,  as discussed  by Caval1er1  and  Rogan
(1983) 1s likely to  be a  factor In cardnogenlcHy and needs Inclusion  1n a
predictive quantitative structure activity scheme.
      *
    Tissue Specificity of  PAH Metabolism.  The capacity  of mammalian  liver
to metabolize PAHs  Is well-documented.   Other  tissues may  also be  Involved
1n  the  metabolism of  a  specific PAH to reactive forms.   AHH,  the  primary
benzo[a]pyrene metabolizing system, has  been found 1n human liver, placenta,
lymphocytes, monocytes and  lung  macrophages  (Gelboln,  1980).

    Cytochromes  P-450 and associated enzymes are known to be  present 1n mam-
malian lungs at  lower  concentrations  than are generally  found 1n the  liver.
Studies  using  lung mlcrosomal  preparations,  cultured trachea  and  alveolar
macrophages and  Isolated   perfused  lungs have  shown  that benzo[a]pyrene  1s

04440                               VII-9                           12/31/90

-------
metabolized  to  the  same oxidized forms  as  are produced by the  "hver.   What
differs 1s the proportion of classes of metabolites,  the rate  of  metabolism
and the ultimate  tissue distribution  of metabolites  (Moore and  Cohen,  1978;
Santodonato  et  al.,  1981).   Co-administered  particles  or  gases  may  also
Influence  the  measures  (Warshawsky et  al.,  1981;  Schoeny and  Warshawsky,
1983).

    Intestinal mucosa has been  shown  to have MFO activity  and presumably  to
have  the  capacity to metabolize  PAHs.   It has also  been observed that  the
susceptibility of mice  to  Induction of forestomach tumors can be  correlated
with  these  enzymatic   activities.    Generally,   levels   of  benzo[a]pyrene
metabolizing enzymes In  rodent  small  Intestine and colon are rather  low and
not readily Indudble to higher levels  (Santodonato et  al.,  1981).

    Genetic Control  of  PAH  Metabolism.  As noted  1n the preceding sections
and In  Chapter  III,  oxldatlve metabolism  of  PAHs  1s generally  accomplished
                                                                           •
through a  series  of  enzymes associated with cytochrome P-450.   Both  quanti-
tative and  qualitative  changes 1n  these enzymes  can be  Induced  In response
to  exposure  to a  variety  of  agents,   Including the  PAHs  themselves.   PAHs
can,  thus,  be  responsible  for Inducing enzymes  for  their own  metabolism.
Including  activation to  mutagenlc  and  carcinogenic  forms.  The degree  of
1nduc1b1l1ty as well as  the  spectrum  of enzymes  produced  1n response  to a
particular  agent   are   not  only  tissue-specific,  but  also species/strain
specific.    In  particular the  enzyme  system,  loosely  called AHH,  that  1s
Involved In PAH metabolism  (especially benzo[a]pyrene)  has been shown  to  be
under  genetic control.   The genetics  of AHH Induction  has  been  well-charac-
terized In  mouse  strains observed  to  be "responders"  or  "nonresponders"  to

04440                                VII-10                         05/07/91

-------
 Induction  by  PAH (usually 3-methylcholanthrene) (Nebert et  al.,  1981).   The
 expression of  the responsive  trait  Is  due  to Inheritance of an allele of the
 Ah  locus,  which  codes  for a cy.tosollc receptor.   This  receptor  regulates
 expression of  a set of  cytochrome  P-450 and associated enzymes.   When  PAH,
 2,3,7,8-tetrachlorod1benzo-£-d1ox1n  or  other  compounds   are  bound  to  the
 receptor,  the  sets  of  genes  coding  for  the various  metabolic  enzymes  are
 Induced to higher levels  of synthesis.   Responsive  mice  have a high-affinity
 receptor  that  readily  binds  to  a  number  of PAHs  and  1s thus  more  easily
 Induced  1n responders  than  1n  nonresponders,  which  have a  low  affinity
 receptor  (Nebert  et al.,  1982;  Elsen et  al.,  1983).   A  number  of  biologic
 effects  observed  1n test  animals as  a  consequence of  PAH exposure can  be
 shown  to  be  affected  by the  responsiveness  or   nonresponslveness  of  the
 strain.  For example, mouse strains C3H/HeJ  and  C57B1/6J,  both responsive to
 d1benz[a,h]anthracene   Induction   of   AHH,  were  more   susceptible   to
 cardnogenesls  after   s.c.   exposure   to  that  compound  than  were   two
 nonresponslve  strains,  AKR/J  and DBA/2J.  Incidences were  24/30  for  C3H/HeJ
 and  16/30 for  C57B1/63  by   comparison  with  0/30  and  1/30  for AKR/3  and
 DBA/2J, respectively  (Lubet  et al.,  1983a).  Similar  results were  reported
 by Kourl et al. (1983), who also  showed  that among  progeny of  a heterozygous
 responder  (B6D2F1)  x   a  homozygous  nonresponder  (D2)   backcross,   that
 susceptibility  to  subcutaneous  .tumor  formation   by   d1benz[a,h]anthracene
 segregated with responsiveness.

    Route  of   exposure  plays a  part  1n  the   extent  of  biologic  effects
produced by PAHs  In responsive and nonresponslve mice.. When  benzo[a]pyrene
 1s administered  topically, s.c.,  1.p. or  Intratracheally, responsive  mice
are more  likely  than nonresponders to  develop  tumors  or  toxic  responses  at
 the site  of  application.   When benzo[a]pyrene  exposure  1s through the  oral

04440                                VII-11                         08/09/91

-------
route, nonresponslve mice are more likely to develop  leukemia or  bone marrow
toxldty  (Nebert,  1981;  Legraverend  et  al.t  1983).  This  1s  due  to  the
effects  of  first pass metabolism  and excretion of  the  PAH.   Nebert et  al.
(1980)  showed,   for  Instance,  that   after  oral exposure  10-20  times  more
[3H]  benzo[a]pyrene  reaches  the  bone  marrow and  spleens  of  nonresponders
than of responsive mice.

    When fetal toxldty  Is  being  assessed there Is the additional  complica-
tion  of  the dam's responsiveness.   This  will  determine  to some extent  the
dose  of  maternally  administered compound that  Is  delivered.to the  fetuses,
which  will  themselves be  responders or  nonresponders.   For  Instance, when
nonresponder dams  received  oral benzo[a]pyrene  on days 2  and  10 of gesta-
tion,  nonresponder  fetuses  showed  a   greater   toxic  response and  more
malformations than  did  responders.   When heterozygous  responder dams were
                                                •*
similarly  exposed,  there  was  no  segregation  of   toxic  response  with  the
nonresponder allele;  that  1s,  both  nonresponder and responder fetuses were
equally  affected.   In  the  case  of   the  nonresponder dam, no  Induction  of
       /
benzo[a]pyrene metabolism took  place  and  the fetuses received a  larger dose
of  the PAH.  This  allowed  the genetic  differences  In  the embryos  to  be
detected; 1n  this 'instance,  greater  amounts of toxic  benzo[a]pyrene metabo-
lites  have  been  Isolated  from nonresponder fetuses  than  from  responders.
When  the  responsive  dam was  exposed, Intestinal  and hepatic metabolism  of
benzo[a]pyrene was Increased.   Fetuses received less  PAH  than  1n the  above
Instance;  thus   there  was  less  overall  fetal  toxldty,  and  the genetic
differences  among  Individual fetuses  was not  seen.   The  reverse  of  these
effects was  observed  when  dams were treated  Intraperltoneally  (Legraverend
et al., 1983).


04440                                VII-12                           05/07/91

-------
    It  has  been  shown  that ovarian  AHH 1s  Induclble  1n responsive (e.g.,
C57B16/N) but  not 1n nonresponder (e.g., OBA/2N) mouse strains.  Responders
have  likewise  been observed to  be more  sensitive  to oocyte destruction by
PAHs  than nonresponders.   The  susceptibility  to  oocyte killing  by PAHs,
however,  1s   not  Inherited  as  a  simple  autosomal  dominant  trait as  Is
responsiveness.  This suggests  that other factors In addition to ovarian PAH
metabolism are Involved 1n oocyte destruction  by  PAH (Hattlson et al., 1983).

    An  area  of  considerable  uncertainty 'with  regard  to  the  carcinogenic
hazard of PAHs  to humans  Involves the relationship between AHH activity and
cancer risk.   Genetic variation 1n AHH InduclbllUy has been Implicated as a
determining  factor   for   susceptibility   to  lung  and   laryngeal  cancer
(Kellerman et  al.,  1973).  It  was  suggested  that  the extent of AHH  Inducl-
bllUy In  lymphocytes  was correlated  with  Increasing susceptibility to lung
cancer formation.

    Paigen et al. (1977,  1978a,b) examined  the question of genetic  suscepti-
bility  to  cancer  and  concluded  that epldemlologlc  evidence -supports  this
hypothesis.   Moreover,  they were  able   to  show that' AHH  InduclbllUy  1n
lymphocytes segregates 1n the human population as a genetic  trait.   However,
their  studies  failed to  find  a  correlation  between this  InduclbllUy  and
presumed  cancer  susceptibility,  either  among  healthy  progeny  of cancer
patients  or  In  patients  who  had their  cancer  surgically  removed.   It  1s
noteworthy that  previous  Investigations   on AHH  InduclbllUy were  conducted
In persons with active cancer.
04440                                VII-13                          11/13/91

-------
    In an  attempt  to elucidate sources  of  variability 1n human AHH  Inducl-
blllty,  a  study was  undertaken using  cultured monocytes  obtained  from  10
sets  of  monozygotlc  twins  and 17  sets of  dlzygotlc  twins.   Both  benz[a]-
anthracene-lnduced and basal levels of  AHH were determined.   Genetic  factors
were  found to  account  for  50-66% of observed 1nter1nd1v1dual differences  1n
                                   •
AHH 1nduc1b1l1ty.   The authors felt  that  a  relatively  few number of  genes
were  Involved  1n regulation of AHH  Induction  (Okuda  et al.,  1977).   It  would
seem  that  susceptibility  to cardnogenesls  by  PAHs could  be a genetically
determined trait  at the level  of  metabolism.  The  contradictory  nature  of
studies In this area, however, point  to the  fact  that  many  factors regarding
PAH activation  and  subsequent  steps  In  the  carcinogenic  process need  to  be
elucidated.

    Other  Pathways  Involved 1n  Activation  of  PAH.   PAHs are  photoreactlve
compounds.    Visible light  can be  absorbed   by  several   PAHs  at  sufficient
energy levels  to  result  1n  photoox1dat1pn.   Benzp[a]pyrene can  be photo-
oxidized  to  the 1,3-,  6,12- and  3,6-qu1nones as well as  to dlhydrodlols and
phenols'  (Katz  et  al.,  1979;   Gibson  and Smith,  1979).   It  1s  likely  that
qulnone  formation  Is  through  a  phenoxy   radical  and   that  phenol  and
dlhydrodlol  formation  may   also  proceed through  radical  formation  (Inomata
and Nagata,  1972;  Jeftlc   and Adams,  1970; Greenstock  and  Wlebe, 1978).
Benz[a]anthracene  1s  also  known to form qulnones  as a consequence of  light
exposure  1n aqueous media  (Mill et  al.,  1981).

    Caval1er1 et al.  (1985) described a 1-electron PAH oxidation pathway  as
opposed  to  the  2-electron mono-oxygenase  pathway.   This  produces   radical
cations or radicals  depending  on  the  PAH.   In addition to  cytochrome P-450,


04440                                VII-14                   .        11/13/91

-------
hydroperoxldases,  peroxldases  and prostaglandln  synthetase  can  participate
1n  these  types  of   reactions.   PAH  free  radicals  {for  example,   6-oxy-
benzo[a]pyrene  free  radical)  have  been  Implicated In  PAH  binding  to  DNA
(Santodonato et al.,  1981).

    Cobalt  ^-Irradiation  of  PAHs,  Including  benz[a]anthracene,  benzo[a]-
pyrene  and  chrysene,  has  resulted  In   formation  of  mutagenlc  compounds.
Likewise  UV-1rrad1ated  benzo[a]pyrene  was shown  to be mutagenlc  for  Salmo-
nella  typhlmurlum  {Gibson  et  al.,  1978).  The potential  for  visible  light
and  other electromagnetic  radiation  to  activate PAHs  has Implications  for
mechanisms Involved In PAH  skin cardnogenesls.

    PAH   Involvement  1n   Carcinogenic  Processes.   A  description  of   the
hypothetical  mechanisms   purported  to   be  Involved   In  the   carcinogenic
processes 1s beyond the scope  of  this document.   Suffice  It  to  say that  PAHs
may  participate  In many proposed carcinogenic  steps.   The majority of  PAHs
described  1n  this document  are mutagenlc  1n  one  or  more  test systems  as
described 1n Chapter  V.  Many  have been shown to  be Initiators  1n mouse  skin
Initiation-promotion   assays.    PAHs  can,  thus,  generally be  described  as
capable of DNA-b1nd1ng, or of  causing DNA damage  leading  to  mutations, which
could be  Involved 1n  the Initiation  phase  of  cardnogenesls.

    Consideration  must  also  be given  to  the fact  that,  1n  addition  to  the
Initiation  of  resting  cells   by  a  chemical  carcinogen,   a  promotion  phase
Involving cell  proliferation  Is also  Involved  In skin  cardnogenesls  (Yuspa
et al.,  1976).   Promotion 1s  likely  to  be  a  phase common  In  the  carcino-
genic  process  for  most  tissues.   Certain  PAHs  may   function  only   as
Initiators and  have  no  promoting  ability.   It would  appear,  however,  that

04440                                VII-15                           05/07/91

-------
the more potent  complete  carcinogens  also serve as promoters by  Interacting
with  cellular membranes,  altering- genetic  expression,   or  1n  some  other
fashion causing  cell  proliferation.   It  should be noted,  also, that  as  PAHs
have been  shown  to have  various  1mmunosuppress1ve  effects,  they may  Impair
the  body's  capacity  for  Immune  surveillance of  neoplastlc  growths.   PAHs
may, thus,  play a part In  all steps of a  carcinogenic  process.

Mechanisms  Involved In Noncarclnogenlc Endpolnts
    The preferred  target  sites of  PAHs  appear  to  be rapidly  proliferating
tissues  such  as  Intestinal   epithelium,  bone  marrow,  lymphold  tissue  and
gonads.  This has led  Investigators to the  hypothesis  that the  toxic  effects
of PAHs are due  to a  specific  attack  on  DNA of cells  1n  the  DNA synthesis or
S  phase  of the  cell  cycle.   Alterations  1n enzyme activity resulting  from
the direct attack on  DNA  may  also  have Important significance  to  the  adverse
effects  resulting  from  PAH  exposure.   Information  on  certain  PAHs  not
considered   In  the rest of  the document  1s presented here for  Illustrative
purposes.
      /
    Hemolymphatic  System.    7,l2-D1methylbenz[a]anthracene  Is   well   known
for Us effects  on the  hematopoletlc  system.   Female  Sprague-Dawley rats fed
112  or 133 mg/kg  7,12-d1methylbenz[a]anthracene developed pancytopenla  due
to a severe depression of hematopoletlc  and  lymphold  precursors  (Caweln and
Sydnor, 1968).   Maturation  arrest occurred at  the  proerythroblast, promylo-
blast  and  promegakaryocytoblast levels;  no  Injury occurred to  stem cells or
circulating formed  elements.   In another  study, rats  receiving 300  mg/kg
(orally)  and   50 mg/kg  (1.v.) of  7,l2-d1methylbenz[a]anthracene  displayed
extreme  atrophy   of  hematopoletlc  elements,  shrinkage  of lymphold  organs,


04440                                VII-16                          12/31/90

-------
agranulocytosls, lymphopenla  and  progressive  anemia (Philips et al.,  1973).
Similar results  have  been noted  1n  mice  and  rats given  repeated  Injections
of  d1benz[a,h]anthracene.   D1benz[a,h]anthracene  administered  to  mice  1n
weekly subcutaneous Injections for 40 weeks caused an  Increase  In  the  number
of  lymph   gland stem  cells,  an  overall  decrease  1n lymphold  cells,  and
dilation  of  lymphold  sinuses.  The  weights of  the  spleens of mice  treated
with  d1benz[a]anthracene  were  also  significantly  less   than  those of  both
controls  and animals  treated with  anthracene  or  benz[a]anthracene  (Hoch-
L1get1, 1941).

    In a  similar study,  rats  given  subcutaneous  Injections  5 times  weekly
for several  weeks  underwent  changes  1n  lymphold tissue characterized  by  the
presence  of   extravascular  red  blood  cells  1n  the  lymph spaces  and  the
presence  of  abnormal  large  plgmented cells (Lasn1tsk1 and  Woodhouse,  1944).
The  noncardnogen,  anthracene,  did   not  produce  as  dramatic  a  change  In
lymphold  tissue.   These  studies  led Investigators  to believe that  Inhibi-
tion of DNA  replication may  be Involved since  only  very  rapidly  proliferat-
ing hematopoletlc elements were affected.

    Acute hemolytlc anemia 1s  the most  frequent manifestation  of  naphthalene
poisoning 1n  humans and has  been  described 1n  newborn  Infants, children and
adults  (U.S.  EPA,  1980c,   1987a).    Pertinent  Information  regarding  the
mechanism of  naphthalene-Induced  hemotoxldty has  been obtained by examining
two groups of Individuals that have  been shown  to  be  especially  susceptible
to naphthalene-Induced hemolytlc anemia.
04440                                VII-17                          12/31/90

-------
    The  first  group of naphthalene  sensitive  Individuals comprises persons
whose  erythrocytes  are  deficient   1n  glucose  6-phosphate  dehydrogenase
(G6PDH)  or  persons  1n whom  erythrocyte  GSH 1s  rapidly  depleted by certain
oxldant  chemicals  (GMgor  et al.,  1966;  Nalman and  Kosoy,  1964;  Valaes et
al., 1963;  Athreya  et al.,   1961;  Dawson  et al., 1958;  Gross  et al., 1958;
Zlnkham  and  Chllds,  1958).   The precise mechanism  by which GSH 1s depleted
or  a  deficiency of  G6PDH  leads  to naphthalene-Induced  hemolysls  1n these
cases 1s not clear.  A deficiency of  G6PDH  will  decrease  the rate of conver-
sion of  nlcotlnamlde adenlne dlnucleotlde phosphate from  Us oxidized  (NADP)
to  Us  reduced  form  (NADPH).    One hypothesis  for  Increased  naphthalene
sensitivity  1n G6POH-defIdent  Individuals  1s that  the  decreased   avail-
ability  of  NADPH  will  decrease the  conversion of oxidized  glutathlone to
GSH, reduce the rate of conjugation and excretion of  naphthalene metabolites
and  Increase  the  accumulation  of  naphthalene  metabolites  1n the  body.  A
similar  hypothesis  may explain  Increased  naphthalene sensitivity  In Indi-
viduals  1n which erythrocyte GSH  can be  rapidly depleted by certain oxldant
chemicals (Nalman and  Kosoy, 1964;  Kellermeyer  et  al., 1962; Dawson et  al.,
1958; 'Gross  et al.,  1958; Zlnkham and  Chllds,  1958).   Gross  et al.  (1958)
demonstrated a quantitative  correlation  between G6PDH deficiency and  dimin-
ished levels of GSH 1n Infants beyond 55  hours of age;  however, diminished
levels of erythrocyte  GSH were  observed  1n Infants of less than 55 hours of
age  despite  high   levels  of   G6PDH activity.   A  second  hypothesis  for
Increased naphthalene  sensitivity In  G6PDH-def1c1ent  Individuals 1s that the
decreased availability of NADPH will, 1n the presence  of  oxldant metabolites
of naphthalene, allow  the accumulation of  methemoglobln  and products  of Us
further Irreversible oxidation  (Kellermeyer  et  al.,  1962).
04440                                VII-18                          12/31/90

-------
    The second group  of  naphthalene-sensitive  Individuals comprises neonates
(Grlgor et al., 1966; Nalman and Kosoy, 1964;  Valaes  et  al.,  1963;  Dawson et
al.. 1958; Gross  et al., 1958;  Zlnkham  and Chllds, 1958).   The  sensitivity
of  neonates  to naphthalene  1s  explained 1n  part  by the  same factors  that
confer  sensitivity  to children and adults;  namely,  G6PDH deficiency  and/or
diminished levels of  GSH as described above.  Additional naphthalene  sensi-
tivity  1n newborns  may  be conferred by the  Immaturity of pathways  necessary
for the conjugation and  excretion of naphthalene metabolites  (Valaes et  al.,
1963).   Evidence  of the latter hypothesis  1s  suggested  by the finding  that
glucuronlde excretion  by human  newborn  Infants  Increased  gradually  during
the first week of life  and that the Initial levels and  the rate  of Increase
were lower 1n  the premature Infant than  1n the full-term Infant (Brown and
Burnett,  1957).

    Cardiovascular  System.   Smoking   1s  a  known  risk  factor  1n athero-
sclerosis, and PAHs are  a major component of  cigarette  smoke (McGIll, 1977;
Wald et   al.,  1973).   Injections   (1.m.)  of  pure  PAHs Into  chickens  has
resulted  1n development  of  prollferaUve lesions bearing a close  resemblance
to  human  atherosclerotic plaques  (Albert et al.,  1977;  Bond et  al.,  1981).
It  has been  proposed  that  human atherosclerotic  plaques are In  fact  benign
hyperplastlc  lesions  of mutagenlc  origin  (Hartman,  1983).   Majesky  et al.
(1983)  undertook  a. study of PAH metabolism  1n  two pigeon strains: athero-
sclerosis susceptible  White Careneau  (WC-2)  and  atherosclerosis  resistant
Show Racer   (SR-39).   After  treatment  with  an  enzyme  Inducer   (3-methyl-
cholanthrene)  hepatic homogenates  from  the  susceptible  strain  were  more
capable  of  benzo[a]pyrene  metabolism  than  those  from  SR-39;  furthermore
7,8-dlol,  a  mutagen1c/cardnogen1c precursor,  was  Increased.   Assays of
aortic  homogenates  showed that  this tissue  had an even  greater capacity for

04440                               VII-19                          12/31/90

-------
benzo[a]pyrene metabolism  1nduc1b1lKy as  a  function of  strain.   It would
appear  that  the  responslve/nonresponslve concept   may   be   applicable   to
nonrodent species and to noncancer  endpolnts.

    Pulmonary  System.   Various   Investigators   have  observed   that   l.p.
administration  of  naphthalene  to  rodents results   In  selective  pulmonary
bronchlolar  epithelial   cell  (Clara)  necrosis,   but   not  hepatic  or  renal
necrosis (long et al..  1981c, 1982; Warren  et al.. 1982;  Mahvl et  al.r  1977;
Reid  et  al., 1973).   In an  effort to  determine the mechanism of  action,
numerous studies  have  focused  on   the  biochemistry  of naphthalene and  the
covalent binding characteristic  of  Us  metabolites.

    Shank  et al.  (1980) found  that  mice  pretreated with  dlethyl  maleate
prior  to  1.p.   Injection   of  naphthalene  had  three  times  the  level   of
covalently-bound naphthalene metabolites  In lung, liver,   kidney and  spleen.
Studies with 14C-naphthalene  Injected  Into mice  revealed a similar  binding
pattern: binding  was  highest  1n the  lung but  low 1n  the  spleen.   Increased
binding corresponded  to rapid  and  significant depletion  of  GSH 1n lung  and
liver, and   to a  lesser extent 1n  the kidney.   Covalent  binding was  dose-
dependent and  exhibited a  threshold  at  dosages  between  200 and  400 mg/kg.
Warren et  al. (1982)  suggested  that  lung  damage may be  mediated by  P-450
dependent metabolism and GSH depletion.

    Buckpltt  and  Warren (1983) extended  these  studies,  utilizing  a  variety
of metabolic  Inhibitors.  The results  suggested that  some  of  the metabolites
Involved  In  GSH depletion  and  covalent  binding  1n extrahepatlc  tissues
originated  In the  liver.   Buckpltt et al. (1985,  1987)   suggested that  the
difference  1n the  rates of  formation of  specific metabolites  between target

04440                                VII-20                          12/31/90

-------
and  nontarget  tissues, and  1n different  species,  may  reflect  the  stereo-
chemistry of  epoxldatlon  by the  tissue-specific  P-450 Isozymes.  This may,
1n turn, relate to the selective pulmonary necrosis  observed 1n mice.

    Confirmation that  P-450  was  Involved 1n pulmonary necrosis  was obtained
In   the   studies   of  BuckpHt  et   al.   (1986).    Liver  mlcrosomes  from
phenobarb1tol-1nduced mice administered  300 mg  naphthalene/kg 1.p.  exhibited
73X  less  covalent binding  In  the  presence of plperonyl  butoxlde, a  P-450
Inhibitor, than controls.  A similar  degree of Inhibition  also  was observed
with  SKF  525A.   It  was-reported  that plperonyl butoxlde  also  blocked  the
pulmonary Injury exhibited by naphthalene In controls.

    Gastrointestinal   System.   Male  and  female   rats  were observed with
Injury to their Intestinal epithelium  after oral and  Intravenous administra-
tion  of  7,l2-d1methylbenz[a]anthracene   (Philips  et  al.,  1973).    Further
analysis  demonstrated  that   the Incorporation  of  14C-labeled thymldlne Into
DNA  of  small  and  large Intestine,  spleen, cervical  lymph nodes and  other
lymphatic  structures  was  Inhibited  as  much  as  90%  within 6  hours  after
d1methylbenz[a]anthracene  administration.   This finding  Indicates  a  strong
Inhibition of  DNA synthesis and  led  the  authors  to  believe  that  DNA 1n  S
phase  cells   1s  particularly  susceptible to  7,l2-d1methylbenz[a]anthracene
and  presumably to other PAH compounds as well.   It  has been proposed that
somatic cell mutations, such as  can be  Induced by PAH metabolites, may play
a  role  In  the formation   of  gastric  polyps  and  chronic  gastric   ulcers
(Hartman, 1983).

    Gonads.  Testicles  and   ovaries contain  rapidly proliferating  cells  and
thus, may be  especially  susceptible  to  damage by  PAHs.   Severe  testlcular

04440                                VII-21                          12/31/90

-------
damage was  Induced  1n adolescent rats  by  a single  Intravenous  Injection  of
7,l2-d1methylbenz[a]anthracene  (0.5-2.0  mg).   Similar effects were  produced
1n  adult  rats  given  7,l2-d1methylbenz[a]anthracene  orally  (20  mg)   and
Intravenously  (5 mg).   Lesions  Involved  destruction  of  spermatogonla  and
resting spermatocytes, both  of  which  are the only  testlcular  cells  actively
synthesizing  DNA.   The remaining germinal  cells and  Leydlg  cells  were  not
damaged  by  7,l2-d1methylbenz[a]anthracene.   No  testlcular  damage  was  pro-
duced by a single feeding of 100 mg  benzo[a]pyrene  (Ford and  Hugglns, 1963).

    In  female   mice,  7,l2-d1methylbenz[a]anthracene   was   shown   to  cause
destruction of small oocytes and  a  reduction  In  numbers of growing and large
oocytes after  oral  administration.   Mattlson and  Thorgelrsson  (1977)  have
shown  that  destruction  of  primordial  oocytes   In  mice following  treatment
with  3-methylcholanthrene was  correlated  with  the  genetic  capability  for
PAH-lnduced  Increases  In ovarian AHH  activity.   This  Indicates an  apparent
link  between  ovarian  metabolism of  PAH   and   ovatoxldty.   This  link  1s
strengthened by  the observation that the effective  dose to kill  50% of small
oocytes  (ED,-n)  was  less  for   responsive  C57B1/6N  mice   (3.38   tig/ovary)
than  for  nonresponders   DBA/2N   (36.14  yg/ovary).    An F,  generation  from
mating  of  these two  strains  had   an  Intermediate Intraovarlan  Injection
ED5Q of 8.27 yg/ovary.

    Endocrine  System.   Few  data  are available  concerning  the  mechanism  of
toxlclty and the effect of PAHs on  the endocrine system.  Selective destruc-
tion  of  the  adrenal  cortex and  Induction of adrenal  apoplexy by  7,12-dl-
methylbenz[a]anthracene has  been  demonstrated In rats after  a single 1ntra-
gastrlc dose  of  30  mg.  The same amount of adrenal  damage could be produced
by a  5 mg dose  of  7-hydroxymethol-12-methylbenz[a]anthracene, the principal

04440                                VII-22                          12/31/90

-------
7,l2-d1methylbenz[a]anthracene oxldatlve metabolite.   No  adrenal  damage was
noted with other 7,l2-d1methylbenz[a]anthracene metabolites, thus  Indicating
that a specific reactive Intermediate may be responsible for this  phenomenon
(Santodonato et al., 1981).

    Integumentary System.   The Integument  1s  highly  susceptible  to agents
that Inhibit DNA synthesis as  Is evident 1n cancer  patients  receiving radia-
tion  treatment  or   chemotherapy.   Such  Individuals show  signs  of alopecia,
dermatitis and  skin sloughing.  Workers exposed to PAH-conta1n1ng materials
such  as  coal  tar,  mineral  oil  and  petroleum waxes  are known  to develop
chronic  dermatitis  and  hyperkeratosls  (Heuper, 1963;  NAS,  1972a).   It  Is
well  documented  that  the  application  of  carcinogenic  PAHs to  mouse   skin
leads to  destruction  of  sebaceous  glands,  skin ulceratlons,  hyperplasla and
hyperkeratosls  (Bock,  1964).   Sebaceous glands undergo  rapid  cell  turnover
and are the most sensitive  structures  of the skin to PAH-lnduced toxldty.

    Visual System.   Ocular  toxldty,  particularly cataract formation, has
long  been  associated with  naphthalene  administration 1n  rodents  and  other
laboratory animals  as well  as 1n  humans  (U.S. EPA,  1980c, 1987a).

    Oral  administration of naphthalene  1s  believed to result 1n Us metabo-
lism 1n the liver and the metabolites then travel  through  the bloodstream  to
the  eye  where  further  metabolism  takes  place  (van   Heynlngen,  1979).
Evidence  In   rats   and  rabbits  suggest that  l,2-d1hydroxy naphthalene  1s
enzymatlcally  converted  to  l,2-naphthoqu1none which  then  reacts  with eye
proteins,  resulting  1n  damage  (P1r1e  and  van Heynlngen,  1966;  Rees and
Plrle, 1967).


04440                                VII-23                          12/31/90

-------
    van Heynlngen (1979), 1n her review of the literature, hypothesized that
susceptibility  to  naphthalene-Induced  cataracts   Is  more pronounced  In rat
and rabbit  strains with  lightly  plgmented  or  dark eyes, due to the presence
of. polyphenol  oxldase.   This  nonspecific enzyme,  found only  1n plgmented
tissues, catalyzes  the formation  of  melanin from  tyroslne.
                                  •
    van Heynlngen  and Plrle (1967)  suggested  that the  toxic  metabolite Is
1,2-d1hydroxy  naphthalene.   In  gavage  studies   1n  which   naphthalene was
administered dally  to  39 rabbits  at  1  g/kg,   they  detected  1,2-d1hydroxy
naphthalene  and 1,2-naphthoqu1none  1n  the  eyes  and three metabolites  1n
blood   that  could  be   converted   by   different  enzymes  1n   the  eye  to
l,2-d1hydroxynaphthalene.  In more than half  the  rabbits, lens  opacities and
degeneration of  the  retina were  observed.   In addition, 1,2-naphthoqu1none
can  oxidize ascorbic  add  present  1n  the  aqueous  and  vitreous  humors,
resulting  1n  oxalic   add  formation  as  the ascorbic  add  concentration
decreases  (van Heynlngen,  1970a,b).   Although  ascorbic  add   decreases  In
aqueous and vitreous humors, the level 1s maintained  or Increases 1n the eye
lens Itself  (van Heynlgen,  1970a).   Presumably dehydroascorblc acid, formed
by  oxidation   by  naphthoqulnones,   penetrates  the  lens  and  1s  reduced  to
ascorbic add.   Ascorbic add diffuses only slowly from the  lens.  Excessive
depletion of ascorbic add may account for the appearance of calcium oxalate
crystals (Plrle and  van Heynlngen,  1966).  GSH  appears  to  be  maintained at
high  levels 1n  eye  lens 1n  spHe of  extensive oxldatlve reactions  (van
Heynlngen,  1970a).

    van Heynlngen (1970b) found that the albino Wlstar rat has  only about 3X
of  the  concentration of  catechol  reductase   (an enzyme  that  catalyzes the
1nterconvers1on of  qulnones  and dlols) found  In the  rabbit lens.   The rat

04440                                VII-24                          12/31/90

-------
also has  less  ascorbic  add  In  aqueous  humor  than  the rabbit (van Heynlngen,
1979).   This  would result  In  a higher  level  of  1,2-naphthoqu1none.   Thus,
polyphenol  oxldase may  be  the  most  Important  factor 1n  the  rat eye  while
catechol reductase may play a crucial role In ocular toxldty 1n the  rabbit.

    Rao and Pandya (1981)  reported Increased I1p1d peroxldatlon  In  the eyes
of male albino  rats administered 1 g  napthalene  dally for  10 days.  Alkaline
phosphatase showed a  slight  Increase and  aniline hydroxylase activity  was
not detected.    Liver peroxide levels  were elevated  but serum I1p1d peroxides
were not measured.

    L1p1d  peroxides  have  been  suggested  as  a causal  factor  1n  cataract
formation.  Yamauchl et al.  (1986)  Investigated this aspect 1n  relation  to
naphthalene.   Naphthalene  (1  g/kg)  In  acacia oil  was  administered  to  male
Mlstar  rats  dally for  up to  18 days.  GSH  content  1n lens  and serum  and
liver  I1p1d peroxide  levels were  measured  during  Interim  sacrifice.   Serum
peroxide  levels  Increased  significantly on  the  fourth  day and  reached a
maximum on  the seventh  day.   Liver   peroxide  levels  had a  similar  pattern.
GSH content 1n  lenses  decreased to about 64X on the  fourth day and  remained
depressed.  The authors  suggested  that  I1p1d peroxides are  stable enough  to
reach  the  lens and cause  ocular damage.  Microscopic observation Indicated
slight  cataractous changes 1n  some  rats .on  the 14th day  when serum  Upld
peroxide  levels were elevated  (Yamauchl et  al.,  1986).   It  was suggested
that peroxides  may play  a role 1n  cataract formation,  1n addition to  the
role  played  by  1,2-n.aphthaqulnone.   A  decrease   In nonproteln  sulfhydryl
content  1n lens   has  previously  been   associated  with  naphthalene-Induced
cataracts 1n rabbits  (Ikemoto and Iwata,  1978).


04440                                VII-25                           12/31/90

-------
PAH-Induced Immunotoxlclty
    Numerous  Investigators   have  demonstrated  that  carcinogenic  PAHs  can
produce  an  Immunosuppresslve  effect.   This  effect  was  first  observed by
Malmgren  et  al.  (1952)  using  high  doses  of   3-methylcholanthrene  and
d1benz[a,h]anthracene In  mice.   Subsequent studies  established  that  single
carcinogenic doses  of 3-methylcholanthrene,  7,l2-d1methylbenz[a]anthracene,
and benzo[a]pyrene caused a  prolonged  depression of  the  Immune  response to
sheep  red   blood  cells   (Stjernsward,  1966,  1969).   Noncarclnogenlc  hydro-
carbons  such as  benzo[e]pyrene and  anthracene  reportedly had  no  Immuno-
suppresslve  activity.   In  a  review  on  Immunosuppresslon  and   chemical
cardnogenesls,   substantial  evidence  was  presented  to  Indicate  that  the
degree  of   Immunosuppresslon  was  correlated  with  carcinogenic  potency  for
PAHs  (Baldwin,  1973).   Both cell-mediated  and  humoral Immune reactions are
affected by PAHs.

    The  effects  of  three PAHs, 3-methylcholanthrene,  d1benzo[a,h]anthracene
and d1benzo[a,c]anthracene on  the Immune  response  was Investigated  1n  mice
1n  relation to   tumorIgenesls  (Lubet  et  al., 1984).   The  subcutaneous  and
oral  routes were  used  1n  two strains  of mice,  C57B1/6  (B6N)  and  DBA/2N
(D2N).  The B6N strain Is an AHH Indudble  species,  and was more  susceptible
to  tumor formation than was  the nonlnduclble  D2N  strain.  D1benz[a,h]anthra-
cene at doses of 25  or 50 mg/kg produced  an Immunosuppresslon  effect as  mea-
sured  by a modification  of  the Jerne  plaque assay.   This  effect  was  more
pronounced  1n the AHH Indudble B6N mice  than In  the 02N  mice.   In  contrast,
d1benz[a,c]anthracene caused minimal effects  1n either strain.   In  general,
1t  was  noted that the  more  potent carcinogens  produced  greater Immunosup-
presslon than  the  noncardnogens.   The route of  administration  also  Influ-
                                          i
04440                                VII-26                          12/31/90

-------
enced  the  Immunosuppresslve  effects.   Following 1.p. administration of high
concentrations  of  PAHs   striking   1mmunosuppress1on  was  observed  1n  both
strains  of  mice.  When  the  oral  route  was employed,  the nonlndudble AHH
mice (D2N) showed greater susceptibility than did the B6N  mice.  The authors
suggested that  In  AHH 1nduc1ble mice, orally  administered PAHs are rapidly
metabolized and  are  rapidly excreted.   In  the nonlndudble  mice  the  llpo-
ph1l1c PAHs are  absorbed and distributed to  target  organs.  They concluded
that  AHH  1nduc1b1l1ty  1s   an  Important  factor  In  the   Immunosuppresslve
activity of PAHs.

    A  study  by  White  et  al.   (1985)  showed  that  suppression  of  an IgM
response to SRBC In mice  exposed to PAH was a  function  of  the strain respon-
siveness.  Responsive  B6C3F1 mice  did not experience  the  degree of  Immuno-
suppresslon following  subcutaneous exposure as did DBA/2 nonresponslve  mice.
This study further showed a  correlation between a PAH's capacity for  Immuno-
suppresslon and  Us  cardnogenldty.  Anthracene,  chrysene, benzo[e]pyrene
and  perylene  had no  significant effect  on  Immune  response  while benz[a]'-
anthracene, benzo[a]pyrene,  d1benz[a,c]anthracene  and  d1benz[a,h]anthracene
exposure resulted 1n 55-91%  suppression.  Even greater  1mmunosuppress1on was
caused by 3-methylcholanthrene  and  7,l2-d1methylbenz[a]anthracene.

    These results were contradicted to some extent 1n 1n vitro  studies  using
the weak  or  noncardnogen/cardnogen  pair benzo[e]pyrene  and benzo[a]pyrene
(Blanton  et  al.,  1986).   These   workers   found  that benzo[e]pyrene did
suppress certain  Immune  function but required higher  doses than did benzo-
[a]pyrene.  By contrast to the White et al. (1985) study,  anthracene  In this
Instance  caused j_n  vitro  Immunosuppresslve  effects.   These  studies  also
demonstrated that the  1mmunosuppress1on by benzopyrenes 1s  due  to effects on

04440                                VII-27                          12/31/90

-------
various  cell  types and  1s  not  attributable  only to cytotoxlclty.   It  has
been   shown   that   benzo[a]pyrene   has   potent   effects   on  production  of
Interleukln 1  by macrophages In  vitro  (Lyte and B1ck, 1986).

Summary
    The  cardnogenes1s-1n1t1at1ng  potential  of  a PAH  1s  dependent  upon  a
number of  factors:  Us  I1p1d  solubility  and  distribution to target organs,
the presence  of  potentially  reactive  areas of Us  structure and Us poten-
tial for metabolism to  reactive  electrophlUc  forms.  This  latter factor 1s
tissue and species dependent and 1s related to some extent  to the 1nduc1b1l-
Uy of  higher  levels  and particular Isozymes  of  cytochrome  P-450-assodated
enzymes.   Carcinogenic  PAHs  are generally  mutagenlc  and  can  damage  DNA.
These activities are  very  likely to be  Involved  1n  the compound's activity
as  an  Initiator  of  carcinogenic processes.   Many  PAHs  are  complete  skin
carcinogens,  serving as  their  own promoters.   Some  are  active as promoters
or cocarclnogens for  other  Initiating  agents.

    Target tissues  for  PAH-med1ated toxlclty other  than cardnogenesls are
generally actively engaged  In  ONA synthesis.   These  tissues Include hemato-
poletlc and Immune systems, gonadal tissues and  the  lens of the eye.  Mech-
anisms  Involving suppression  of  DNA synthesis or  DNA  damage have been pro-
posed for  PAH-lnduced toxldty to hematopp1et1c  elements,  the GI system and
the  cardiovascular  system.   In  addition  to  the  organ  systems  covered,
PAH-1nduced lesions have been  demonstrated In the lungs,  liver  and kidney.
No  evidence concerning  precise mechanisms of  toxlclty  has  been found, but
many  toxic  endpolnts are  linked to  the  species or strains capability for
ready Induction of PAH metabolizing  enzymes.


04440                               VII-28                          12/31/90

-------
    Naphthalene appears  to  be metabolized In  a  different manner  than  other
PAHs.   Studies  Indicate  that naphthalene  1s  metabolized  1n  the liver  to
1,2-naphthaqulnone.   Administration  of  naphthalene  causes  an elevation  1n
serum Upld peroxide  and  liver  peroxide levels.   It has  been  suggested that
l,2-naphthaqu1none and elevated  serum  llpld  peroxide levels may play a role
1n naphthalene  Induced cataract  formation  1n r-odents.   Naphthalene Injection
has  also caused  bronchlcal  epithelial  cell  necrosis  In  rodents  and  has
caused   acute   hemolytlc   anemia   1n    Instances   of   human   poisoning.
04440                                VII-29                          05/10/91

-------
                 VIII.  QUANTIFICATION OF  TOXICOLOGIC EFFECTS

Introduction
    The  quantification  of  toxlcologlc effects  of  a  chemical  consists  of
separate  assessments  of  noncarclnogenlc   and  carcinogenic  health  effects.
Chemicals  that  do not  produce  carcinogenic effects  are  believed to  have  a
threshold  dose below which no adverse,  noncarclnogenlc  health  effects  occur,
while carcinogens are assumed to act without a  threshold.

    In  the  quantification   of   noncarclnogenlc  effects,  a  Reference  Dose
(RfD),  [formerly  termed the  Acceptable Dally Intake (ADI)]  1s  calculated.
The RfD  1s an  estimate (with uncertainty  spanning  perhaps an order  magni-
tude)  of  a  dally  exposure  to  the  human population  (Including  sensitive
subgroups) that  1s  likely  to be  without  an appreciable risk  of  deleterious
health effects  during a lifetime.   The RfD 1s  derived from  a  no-observed-
adverse-effect   level   (NOAEL),   or    lowest-observed-adverse-effect   level
                                                                           *
(LOAEL),  Identified  from  a  subchronlc  or chronic  study,  and  divided  by  an
uncertainty factor(s)  times  a modifying  factor.   The  RfD  Is  calculated  as
follows:
     RfD =	(NOAEL or LOAEL)	 =
           [Uncertainty Factor(s) x Modifying Factor]   	
    Selection of the uncertainty factor to be  employed  1n  the calculation of
the RfD  1s based  upon  professional Judgment,  while  considering  the  entire
data base  of  toxlcologlc  effects  for the chemical.   In  order  to ensure that
uncertainty  factors  are  selected  and  applied  1n   a  consistent  manner,
04450                                VIII-1                           12/31/90

-------
the  U.S.  EPA  (1991a)  employs  a modification  to the guidelines  proposed  by

the National Academy of Sciences (NAS, 1977, 1980) as follows:


Standard Uncertainty Factors (UFs)

        Use a  10-fold  factor when  extrapolating from valid experimental
        results from studies using prolonged exposure to average healthy
        humans.   This  factor 1s  Intended  to account for  the variation
        1n sensitivity among the members of the human population.  [10H]

        Use an  additional  10-fold  factor when  extrapolating  from valid
        results  of  long-term  studies  on  experimental  animals  when
        results of  studies of  human  exposure  are not available  or  are
        Inadequate.  This  factor  Is Intended to  account  for  the uncer-
        tainty  In  extrapolating  animal  data   to the  case  of  humans.
        [10A]

        Use an  additional  10-fold  factor  when  extrapolating from  less
        than chronic  results on experimental  animals when there  1s  no
        useful  long-term  human  data.   This   factor  1s  Intended  to
        account  for the  uncertainty  1n extrapolating  from less  than
        chronic NOAELs to chronic NOAELs.  [IDS]

        Use an  additional  10-fold  factor  when deriving  an RfD  from  a
        LOAEL  Instead  of  a NOAEL.   This factor 1s  Intended  to  account
        for  the  uncertainty  In extrapolating  from  LOAELs  to  NOAELs.
        [10L]

Modifying Factor (MF)

    •   Use  professional   judgment  to  determine  another  uncertainty
        factor  (MF) that 1s greater  than zero and less  than or  equal  to
        10.   The  magnitude  of  the  MF  depends  upon the  professional
        assessment  of  scientific  uncertainties  of  the  study and  data
        base not  explicitly treated  above, e.g.,  the  completeness  of
        the overall  data  base and the  number  of species  tested.   The
        default value for  the MF  1s 1.
    The  uncertainty  factor  used for  a  specific  risk  assessment  Is  based

principally  upon  scientific  Judgment   rather   than   scientific   fact  and

accounts  for  possible   1ntra-  and   Interspedes   differences.    Additional

considerations not  Incorporated  1n the  NAS/ODW  guidelines for  selection  of

an -uncertainty  factor Include  the use  of a  less than  lifetime   study  for

deriving  an  RfD,  the significance  of  the  adverse health  effects and  the

counterbalancing of beneficial  effects.



04450                                VIII-2                          11/15/91

-------
    From  the  RfD, a  Drinking Water  Equivalent  Level  (DUEL)  can  be calcu-
lated.   The  DUEL  represents  a   medium  specific   (I.e.,  drinking  water)
lifetime  exposure  at  which adverse,  noncardnogenlc health  effects  are not
anticipated to  occur.   The DHEL assumes 100X  exposure  from drinking water.
The DHEL  provides  the  noncardnogenlc health effects basis for establishing
a  drinking water  standard.    For  Ingestlon data,  the  DUEL   Is  derived as
follows:
               DHEL _   (RfD) x (Body weight 1n kg)   _        ^
                     Drinking Water Volume 1n l/day = 	
where:
        Body weight = assumed to be 70  kg  for  an  adult
        Drinking water volume = assumed to be  2 i/day for an adult

    In addition  to  the RfD and  the DWEL, Health Advisories  (HAs) for expo-
sures  of  shorter duration  (1-day, 10-day and  longer-term)  are determined.
The  HA values  are used  as  Informal  guidance  to municipalities  and other
organizations when  emergency  spills or contamination situations occur.  The
HAs are calculated  using an  equation  similar to the RfD and DWEL; however,
the NOAELs  or  LOAELs are Identified from acute  or subchronlc studies.  The
HAs are derived as follows:
                        (NOAEL or  LOAEL)  x (bw)
                          (UF)x(_ l/day)
    Using the above equation, the following drinking water HAs are developed
for noncardnogenlc effects:
    1.  1-day HA for a 10 kg  child Ingesting 1  i .water  per  day.
    2.  10-day HA for a 10 kg child  Ingesting 1 l  water per day.
    3.  Longer-term HA for a  10 kg child Ingesting 1  I  water  per  day.
    4.  Longer-term HA for a  70 kg adult Ingesting 2  i  water  per  day.
04450                                VIII-3                          08/28/87

-------
    The  1-day  HA  calculated  for  a  10  kg  child assumes  a  single  acute
exposure  to  the chemical and  1s  generally  derived from a  study  of <7 days
duration.  The  10-day  HA assumes  a limited  exposure period of 1-2 weeks and
1s generally  derived  from a  study of <30 days duration.  The  longer-term HA
1s  derived  for  both   the  10  kg  child and  a 70  kg  adult and  assumes  an
exposure  period of  ~7  years  (or  10% of  an Individual's  lifetime).  The
longer-term  HA 1s  generally  derived  from  a  study of  subchronlo duration
(exposure for 10X of animal's lifetime).

    The U.S.  EPA categorizes the carcinogenic potential of a chemical, based
on the overall we1ght-of-evidence,  according  to the following scheme:

        Group  A:  Human  Carcinogen.   Sufficient  evidence  exists  from
        epidemiology  studies  to  support  a   causal  association between
        exposure to the chemical  and human cancer.
        Group  B:  Probable  Human  Carcinogen.   Sufficient  evidence of
        carc1nogen1c1ty  1n  animals with  limited  (Group B1)  or  Inade-
        quate (Group 62) evidence In humans.
        Group  C:  Possible  Human   Carcinogen.    Limited  evidence  of
        cardnogenldty 1n animals  1n the absence of human data.
        Group  D:  Not   Classified  as to Human Cardnogenlclty.   Inade-
        quate human and animal evidence of carclnogenlcUy or  for which
        no data are available.
        Group  E:  Evidence   of   Noncarc1nogen1c1ty  for  Humans.   No
        evidence of  carclnogenlcUy  In  at  least  two  adequate  animal
        tests  In  different species  or 1n both  adequate ep1dem1olog1c
        and animal  studies.
    If toxlcologlc evidence  leads  to  the classification  of the  contaminant
as a  known,  probable  or possible  human carcinogen, mathematical  models are
used  to  calculate  the  estimated   excess cancer   risk  associated  with  the
Ingestlon of  the  contaminant  1n  drinking  water.   The  data  used  In  these
estimates  usually   come  from  lifetime  exposure   studies  using animals.
04450                                VIII-4                          12/31/90

-------
In order  to  predict  the risk for humans from animal data, animal  doses  must
be converted  to equivalent human doses.  This conversion  Includes  correction
for  noncontlnuous  exposure,  less than  lifetime  studies and for  differences
1n.s1ze.   The factor  that  compensates  for the  size  difference 1s the  cube
root of  the  ratio  of the animal and human body weights.   It 1s  assumed  that
ttie  average  adult  human  body weight  1s  70  kg  and  that  the  average water
consumption of an adult human 1s 2  a of  water  per  day.

    For  contaminants  with  a  carcinogenic potential,  chemical  levels   are
correlated with a  carcinogenic  risk estimate by  employing a cancer  potency
(unit  risk)  value  together with  the assumption  for  lifetime  exposure  from
1ngest1on of  water.   The  cancer  unit risk Is usually derived from a.linear-
ized multistage model with a 95% upper  confidence  limit providing  a low  dose
estimate; that  1s,  the true risk  to humans,  while not Identifiable,  1s  not
likely  to exceed  the  upper  limit  estimate and,  1n  fact,  may  be  lower.
Excess cancer risk estimates may also be calculated using other models  such
as the one-hit, Welbull,  loglt  and probU.  There 1s little  basis   1n  the
current  understanding of  the  biological  mechanisms  Involved  1n  cancer  to
suggest that any one  of these models  1s able  to  predict risk more  accurately
than any other.  Because each model  1s  based  upon  differing assumptions,  the
estimates derived for each model  can differ  by several  orders of magnitude.

    The  scientific data base used  to calculate and  support  the  setting  of
cancer  risk  rate  levels  has  an  Inherent  uncertainty that  1s  due  to  the
systematic and random errors 1n scientific measurement.  In most cases,  only
studies  using experimental  animals  have  been  performed.  Thus,  there  1s
uncertainty  when   the data  are extrapolated to   humans.   When  developing


04450                                VIII-5                          12/31/90

-------
cancer  risk  rate levels, several  other  areas  of uncertainty exist,  such  as
the  Incomplete  knowledge concerning  the health  effects  of contaminants  1n
drinking  water,  the  Impact  of   the  experimental  animal's  age,  sex  and
species,  the  nature of  the target organ  system(s)  examined and  the actual
rate of  exposure  of the Internal targets 1n experimental  animals  or  humans.
Dose-response data  usually  are  available  only for  high  levels of  exposure
and  not  for  the lower levels  of  exposure  closer  to  where a standard  may  be
set.   When  there  1s  exposure  to more than  one  contaminant,  additional
uncertainty results from a  lack of  Information about  possible  synerglstlc  or
                        ^
antagonistic effects.  Since  PAH  occur  only as mixtures  In.the environment,
no epidemlologic data are found for Individual  PAH.

Noncardnoqenlc Effects
    For  many  of the  PAHs  experimental  data on  noncardnogenlc effects  are
either  nonexistent  or  provide  Insufficient  Information  on which  to  base
criteria  for  drinking water  exposure.   In  general,  existing  data  are  not
suitable for criteria derivation for one or  more  of these  reasons:
      f
        Studies were designed to assess  only carcinogenic  potential
        Studies were designed to determine  only lethal  dose (1059)
        Studies do not give  dose/response data
    •   Studies contain  only  one  dose   level,  at which  severe health
        effects occurred
    •   Studies do not measure chronic exposure
    •   Studies were by other than oral  exposure
        Measure of dose Is  not known
        Sample size 1s too  small
    •   Test animals  used  do not  provide  relevant  models  for  human
        health assessment.

04450                               VII1-6                          09/23/91

-------
     Recently,   the  U.S.  EPA  (1988,  1989a,b,c)  and  Hazelton  Laboratories
 America,   Inc.   (1989c,  1989d)  conducted  90-day  subchronlc  bloassays  for
 acenaphthylene,   anthracene,  fluoranthene,   fluorene   and  pyrene.   These
 studies  examined a  variety of  toxlcologlc endpolnts  and were  of adequate
 design,  thereby providing  sufficient  Information on which to base criteria
 for  drinking water  exposure.   These  studies  were  Insufficient  to evaluate
 potential  carclnogenldty of the above PAHs.

 Short-term Studies 1n Animals
     There  were  short-term  studies  wherein  animals were  treated  by the oral
 route   reported  for  only  acenaphthylene,  -anthracene,  fluoranthene  and
 fluorene.   For  acenaphthylene  there  1s  an  oral  LD5Q for  rats  and  mice
 reported  1n abstract form  (Knobloch et  al.,  1969).   For anthracene there 1s
 a  report  In  abstract form stating that a  single  oral  dose  of  17  g/kg of
 anthracene Is  not  lethal   to  mice  (Nagornyl,  1969).   For  fluoranthene the
 only data  are  from a range-finding study on  more than  300  compounds; the
.study  reported  the  oral  LD5Q for  rats as  2000  mg/kg/day  (Smyth  et  al.,
 1962).'  For fluorene, K1zer et al.  (1985) reported  that 1n rats  fed 10.5
 mg/kg/day  1n the diet for  3 weeks, no  Increases In either SGOT or hepatic
 mlcrosomal epoxlde  hydrolase  activity  were  observed.   Thus, there  are no
 data suitable for derivation  of 1- or  10-day  HAs for  any PAH  discussed In
 this document.

 Longer-term Studies 1n Animals
     No-longer-term  exposures of animals by the  oral route were reported for
 the  following:   benzo[b]fluoranthene,  benzo[k]fluoranthene,  benzo[g,h,1]-
 perylene,  chrysene,  1ndeno[l,2,3-cd]pyrene, phenanthrene  and pyrene.
 04450                               VIII-7                          11/15/91

-------
    Acenaphthylene.   Acenaphthylene  was   administered   to  groups  of  seven
rats  1n  doses of  600 mg/kg bw  for 40 days  or 2000 mg/kg  bw for 32  days.
Effects  were  reported  on  body  weight,  peripheral  blood,  renal  function,
kidney and  liver morphology,  and  bronchitis  was  Induced  (Knobloch et  al.,
1969).   The  report  Is  In  abstract  form; data  are,  thus,  Insufficient  for
derivation of criteria.

    In  a subchronlc  toxldty bloassay  conducted  by Hazelton  Laboratories
America,  Inc.  (1989c) acenaphthylene was administered  to  groups of 20  male
and  female   CD-I  mice  by gavage.   Dose  levels  were 0,  100,  200 and  400
mg/kg/day.   Criteria  evaluated for compound-related effects were  mortality,
clinical  signs,   body weight,  food  consumption,  opthalmology,  hematology,
clinical  chemistry, organ weights,  gross  pathology  and  hlstopatholgy.   Based
on liver  and kidney  changes and  deaths 1n females  the  LOAEL determined  from
this  study   1s  100 mg/kg/day;  no  NOAEL  was  determined.   Due  to the  high
mortality observed 1n females  receiving  100  mg/kg/day the data  provided  are
                                                                            »
considered Insufficient for  derivation of  criteria.
       t
    Anthracene.   Two  bloassays  for carclnogenlclty  of Ingested  anthracene
have been conducted  (Druckry and  SchmShl, 1955;  SchmShl and  Relter,  n.d.).
Neither   study  mentions  noncancer  health  effects,  nor  were  any   tumors
reported.   The  latter  study  has  not  been  published.   In   a  subchronlc
toxldty  study, the U.S. EPA (1989e)  administered anthracene to groups  of 20
male and  female  CD-I  (ICR)BR mice  by  gavage.   Dose levels were 0, 250,  500
and 1000 mg/kg/day.    Criteria  evaluated  for  compound-related effects  were
mortality,  clinical   signs,  body  weights,  food  consumption,   opthalmology,
hematology,   clinical  chemistry,  organ weights, organ-to-body  weight  ratios,


04450                                VIII-8                         11/15/91

-------
gross  pathology  and  Mstopathology.    No .treatment-related   effects   were
noted; therefore, the NOAEL determined  from this study 1s  1000  mg/kg/day.

    BenzTalanthracene.   Klein  (1963)  published a  report of  benz[a]anthra-
cene carclnogenlclty for mice treated  by gavage.  Only  one  concentration  was
administered,  and  noncancer endpolnts  were not  described.   Thus, data  are
Insufficient to derive criteria for benz[a]anthracene.

    Benzo[a]pyrene.   Benzo[a]pyrene was  administered  to mice  at multiple
concentrations  1n  the diet  1n> order  to assess  Us  carcinogenic  potential
(Rlgdon  and Neal,  1966,  1969;  Neal  and  Rlgdon,  1967).   Treatment-related
Incidences  of  tumors of   the  forestomach  and  lung,  and  leukemlas   were
observed.   No  noncancer  health effects  were  reported,  however.   Rlgdon  and
Neal (1965) also conducted a series  of assays  to  determine  1f  dietary benzo-
[ajpyrene  produced deleterious  reproductive  effects.   Oral  benzo[a]pyrene
concentrations  of  250,  500 or  1000  ppm  over various  time  periods up  to
lifetime  showed no  treatment-related  effects, except  lack of weight  gain
      /
related to feed unpalatabllUy.

    NacKenzle  and  Angevlne (1981)  dosed groups  of  30 or  60  pregnant  CD-I
mice with  gavage preparations  of  benzo[a]pyrene to deliver 0,  10,  40 or  160
mg/kg bw/day.   This  was done  only on  days 7-16  of  gestation.  No maternal
toxldty  was   noted,  nor  were  there  signs of  fetal  toxldty.   Pups   were
culled to  8/l1tter and  used  1n  an F,  mating study.   By  days  20 and  42,
F.  animals  exposed  to   benzo[a]pyrene  Ijn  utero  were  observed to   have
decreased  body weights  1n comparison with  controls.    Gonadal  weights  of
treated  animals  were also significantly reduced.  The  testes from  animals
04450                                VIII-9                          07/26/91

-------
exposed  In  utero  to  the  low  benzo[a]pyrene  dose  weighed  ~60X  those  of
controls;  testes  from  the  40  mg  (middle  dose)  group  weighed  ~18X  of
controls.   As most  P..  ]j±  utero exposed  females  had  no  ovaries  or  only
remnants of  ovarian  tissues,  weights  were  generally  not  recorded.   Fertility
was  reduced  among F, treated  animals.   Of control males  100X were  fertile,
20/25 from  the  group exposed to  10 mg,  3/45  from the 40 mg  group,  and  none
of  the  males  treated   In   utero with  160  mg  benzo[a]pyrene/kg/day  were
fertile.  Of treated females, 34X  of the  low-dose group  produced a  Utter;
none  of  the middle- and  high-dose  females produced  a  litter.  This  study,
                        <•
thus, demonstrates a LOAEL  for  gonadal weight decrease and loss of  fertility
of 10 mg/kg/day for benzo[a]pyrene.

    DlbenzTa.hlanthracene.   D1benz[a,h]anthracene  was carcinogenic  to  mice
receiving the  compound  1n an  olive  o1l/dr1nk1ng  water  emulsion  (Lorenz  and
Stewart,  1947;  Snell  and  Stewart,  1962a).    Only  one  concentration  of
treatment  suspension was  administered.    The  only  noncancer health  effect
reported was  dehydration  and emaciation  of animals due to  poor tolerance of
the vehicle.  There are,  thus, no data suitable  for derivation of criteria.

    Fluoranthene.   In a 13-week  bloassay  1n  mice,  the  U.S.  EPA  (1988)
administered either  0,  125, 250  or  500  mg/kg/day of fluoranthene to  groups
of 20 male  or female  CD-I  mice  by  gavage.   Body weights,  food consumption
and  clinical signs of toxldty  were monitored at regular  Intervals  during
the  experimental  period.  At  the end of  the study  period the  animals  were
sacrificed  and  submitted  for autopsy  and hematologlc  and  serum chemistry
evaluations.  A  LOAEL of  250 mg/kg/day based  on statistically significant
(p<0.05) changes 1n SGPT and  absolute and  relative liver  weights, as well as


04450                                VIII-10                          11/15/91

-------
decreases  1n  packed cell  volume  and red  blood  cell  numbers  (females  only)
and  albumin/globulin  ratios, was  Identified by  the  U.S.  EPA (1991a);  the
corresponding NOAEL Is 125 mg/kg/day.

    Fluorene.   Neither  of  two  reports  of bloassays  for oral carcinogenic
potential  of  fluorene   gave Indication   of  an  Increased  tumor  Incidence
(Morris  et  al., 1960; Wilson et  al., 1947).  The  former study  reported  no
effects  related to  treatment.   In  the   latter  study  one  set of rats  was
exposed  to dietary concentrations of 0.062 or l.OX  fluorene  for 104 days  and
a second set  to 0.125,  0.25 or 0.5X fluorene  for  453 days..  The  animals  of
the  short-term  group  appeared  normal 1n  all  respects except  for  treatment-
related  decreases   In  growth rate.   The  longer-term  treated  animals were
observed  to  have  significantly  Increased  liver weights  (two highest dose
groups)  and  decreased  spleen  weights  (all  treated  animals).   While  these
changes  were  described  as  "significant",  no numerical data were  presented.
This study 1s, therefore, unsuitable for  criteria derivation.
                                                                           •

    In' a 13-week subchronlc bloassay  1n mice,  the U.S.  EPA  (1989a)  admin-
istered  either 0, 125, 250  or 500 mg/kg/day  fluorene,  suspended 1n corn oil,
to  groups  of  25 male  and  female  CD-I  mice  by gavage.   Measures used  to
assess toxldty Included  food  Intake,  body  weight,  clinical  observations,
hematology  and  serum chemistry,  and  gross  and  hlstopathologlc  changes.
Using  the  data  from  U.S.  EPA  (1989a),  the  U.S.   EPA  (1991a) Identified a
LOAEL of 250 mg/kg/day for hematologlc effects;  the  NOAEL  1s  125 mg/kg/day.

    Pyrene.   In  a  13-week  subchronlc   bloassay  1n  mice,   the  U.S.  EPA
(1989b)  administered either  0, 75,  125  or 250 mg/kg/day  of  pyrene to  groups


04450                                VIII-11                          11/15/91

-------
of  20 male  and  20  female  CD-I  mice by  gavage.   Criteria  used  to  assess
toxldty  Included  body  and  organ  weights,  food  consumption,  mortality,
hematology  amd  serum  chemistry,  and  gross  and  hlstopathology.   Based on
nephropathy,  accompanied  by  statistically  significant  (p<0.01)  changes 1n
absolute  and  relative  kidney  weights,  the  U.S. EPA  (1991a) Identified  125
mg/kg/day as the LOAEL; the NOAEL is  75 mg/kg/day.

Quantification of Noncardnoqenlc Effects
    Derivation  of 1-  and  10-Day  Health Advisories.   There  were  no  data
suitable  for  calculation of  1-day or 10-day  HAs  for children  for  any  PAH
covered 1n this document.

    A  number   of short-term  exposure  studies  have  been  reported  wherein
various  Immunologlc  endpolnts have been measured.   These are summarized 1n
Tables VIII-1  and VIII-2.  While It Is  Important to consider  Immune  deficits
as  Indicators  of  early  toxldty, 1t   Is  not  appropriate  to  base  health
advisories on  the data presented  1n  these  two  tables.   First,  none of  the
studies  employed an oral  route  of  administration,  thereby  decreasing  the
relevance of  the  data  to  drinking water health advisories.  Secondly,  1t 1s
not clear whether  the  Immune  deficits  observed  are,  1n fact. Indicators of
toxldty or  are normal  adaptive responses to stress.

    For benzo[a]pyrene the data  of MacKenzle  and Anglvlne (1981) could  also
be  considered  as an  Indicator  of noncancer  toxic effects  resulting  from
short-term  exposure  in  utero.   In  view  of  the   fact  that  this   exposure
encompassed  an  entire  critical phase  of the  animal's  Hfespan,  namely  the
fetal  developmental  phase, the data may  be more  suitable  for derivation  of a
lifetime health advisory.

04450                                VIII-12                         11/15/91

-------
                                TABLE VIII-1

      Immunotoxldty of  Benzo[a]pyrene 1n Mice after Short-Term Exposure
Days
Exposure
1
1
1
1
1
1
10
14
Strain Route
CBA l.m.
C3H/ANF in utero
B6C3F1 s.c.
C57 1.p.
C3H l.p.
C3H l.p.
C57B1/6 l.p.
DBA/2 l.p.
B6C3F1 s.c.
B6C3F1 s.c.
LOAEL/NOEL
(mg/kg)
33.3/ND
150/ND
252/ND
5.0/0.5
5.0/0.5
18.0/1.8
25/-
50/25
5.0/-
40.4/ND
Effect*
PFC
PFC
GvH
mixed lymph
number of
spleen cells
target cell
killing
INF
PFC
PFC
AbT1
CPLPS
spleen
weight
PFU
Reference
Stjernsward,
1966
Urso and
Gengozlan,
1984
White
et al., 1985
Wodjanl
et al., 1984
Griffin
et al., 1986
Lubet
et al., 1984
Dean et al.,
1983
White
et al., 1985
*A11  effects   noted  were  decreases  by  comparison  with  controls  In  the
 measurements   Indicated.   PFC   =   plaque-forming  spleen   cells;   PFU  =
 plaque-forming units; GvH  =  graft vs.  host response; mixed  lymph  =  mixed
 lymphocyte  response;   INF   =   Interferon  production;   AbT1   =  antibody
 production  to a  "[-Independent  antigen;  CPLPS  =  cell  proliferation  1n
 response to Upopoly-sacchaMde mHogen.

ND = Not determined as  only  one dose was  tested

- = No NOEL reported
04450
VIII-13
11/15/91

-------
                                 TABLE VIII-2

        Comparative Effects of PAH on Immune Functions In B6C3F1  M1cea
      Compound
   Doseb
(mg/kg/day)
      Effect
Anthracene

Benz[a]anthracene


Benzo[a]pyrene




Chrysene

D1benz[a,h]anthracene
   28.5

   36.5


   40.4
   36.5

   42.7
None

Decreased antibody
forming cells

Decreased antibody
forming cells;
decreased spleen
weight

None

Decreased antibody
forming cells;
decreased spleen
weight
aData from White et al. (1985)

bAnimals were treated s.c. with 160 ymole/kg/day
04450
   VIII-14
          07/27/90

-------
    Derivation of  Longer-Term HA.  There  were no data  suitable for  calcu-
lation of longer-term HAs for children or  adults  for  any  PAH  covered  1n  this
document.

    Assessment of  Lifetime  Exposure  and  Derivation  of  DWELs.   There  were
no data  suitable  for  calculation  of RfDs  for  the  following:   benz[a]anthra-
cene,   benzo[b]fluoranthene,   benzo[k]fluoranthene,    benzo[g,h,1]perylene,
chrysene, d1benz[a,h]anthracene, 1ndeno[l,2,3-cd]pyrene  or phenanthrene.

    Acenaphthylene — Quantitative data  on the  subchronlc  oral  toxlclty  of
acenaphthylene has  been  provided  by  Hazelton  Laboratories  America,  Inc.
(1989c).  Acenaphthylene  .was  administered to  CD-I   mice (20/sex/group)  at
dosage levels of 0, 100, 200  or 400 mg/kg/day. for  at  least 90 days.   Effects
examined Included  mortality,  clinical signs,  body  weights,  food  consumption,
opthalmology, clinical  chemistry,  organ  weights, gross pathology and  hlsto-
pathology.    Treatment-related  effects were  observed  1n  all  dosage  groups.
Due to the high mortality observed  1n all  groups  of  treated females,  the low
dose  Is considered  an  PEL.   Therefore,   this  study  1s  Insufficient  for
deriving criteria  since the  choice  of  mortality  as  the critical effect  1s
not appropriate as the basis for an RfD.

    Additionally,  four  subchronlc  studies  on the  toxldty  of acenaphthylene
were  located  In  the literature:   two oral studies (Rotenberg and Mashblts,
1965;  Knobloch et al.,  1969)  and  two  Inhalation   studies  (Rotenberg  and
Mashblts, 1965;  Reshetyuk  et al., 1970).  However, deficiencies  (no  experi-
mental  controls,   short  duration  and  Incomplete  reporting  of study  design
and/or results)  preclude  the use of any of  these  studies as a basis  for  an
oral RfD.
04450                                VIII-15                         11/15/91

-------
    Anthracene ~ Quantitative  data  on  the  oral   exposure  to  anthracene
reported 1n U.S. EPA  (1989e)  failed  to  demonstrate  treatment-related  effects
to male  or female  CD-I  (ICR)BR mice at  doses up  to  1000 mg/kg/day for  at
least 90  days.   From  these  results, a NOAEL  of  1000  mg/kg/day was  Identi-
fied.  Using this NOAEL, the DUEL 1s derived as follows.
    Step 1  - RfD Derivation
            RfD m (1000 mq/kq/day)  = ^^ mg/kg/day
                        3000         (rounded to SxlO'1 mg/kg/day)
where:
    1000 mg/kg/day = NOAEL reflecting  no  treatment-related  effects  1n
                     mice (U.S.  EPA,  1989e)
    3000           = combined uncertainty  factors:  100  to account  for
                     1ntra- and   Interspedes  extrapolation,  10  for  the
                     use of  a  subchronlc  study for RfD  derivation,  and
                     3  for  the   lack  of reproductive/developmental  and
                     supporting  chronic  toxlclty data
    Step 2 - DUEL Derivation

              nun        OxIO"1  mq/kq/day)  (70  kg)     in ,. mn
              DWEL =               (2 I/day)          =  10'5 mg/l
where:
      /
    3x10'*  mg/kg/day = RfD
    70 kg            = assumed body weight  of an adult
    2 i/day          = assumed volume of water consumed by an adult
    Benzol"aIpyrene — For benzo[a]pyrene an  Indicator of a  noncancer  toxic
effect was  the  decrease 1n  fertility and  gonadal  weight  1n mice  exposed  In
utero.  Dams  were gavaged with  10 mg/kg/day  benzo[a]pyrene In corn oil  on
days  7-16  of  gestation  (MacKenzle  and  Angevlne, 1981).   While  this  does  not
constitute  a  subchronlc  or  chronic exposure 1t does cover the  entire period
of development  most  sensitive to  Insult of the test  population.  The  study
could, thus, be considered for use  In estimating a  longer-term  risk.

04450                                VIII-16                         11/15/91

-------
    For   benzo[a]pyrene,   available   data  show   that   the  appearance   of
neoplastlc  effects  occurs  at  lower  doses than  do  Indicators  of  systemic
toxldty.   For  example a  LOAEL  based on  the  MacKenzle and Angevlne  (1981)
data  Is  10 mg/kg/day,  while  significant  Increases  1n  tumor Incidence  have
been  observed  In mice  receiving 6.5  mg/kg/day benzo[a]pyrene  1n the  diet
(Neal  and  Rlgdon,  1967).   It,   thus,   seems  Inadvisable  at  this  time  to
propose any health advisories  for benzo[a]pyrene based on  noncancer effects.

    Fluoranthene — In  a  13-week subchronlc  bloassay 1n mice  the U.S.  EPA
(1988) Identified a LOAEL  of  125 mg/kg/day of  fluoranthene,  This  LOAEL  was
based  on  statistically  nonsignificant  Increases  1n clinical  signs,  serum
chemistry  and changes  1n liver  and  kidney pathology as well as  significant
Increases  In relative liver weights of male mice.   In a  revaluation  of U.S.
EPA (1988), the  U.S.  EPA (1991a) Identified the 125  mg/kg/day dose  level  as
the NOAEL  and  the 250 mg/kg/day dose  level as  the LOAEL.  The low dose  was
considered  the  NOAEL  as  clinical  signs  (I.e., salivation) were not  dose-
related  effects,  and  changes   1n serum enzymes and kidney  and liver  hlsto-
pathology  were  not  considered  adverse at  125  mg/kg/day.   The 250  mg/kg/day
dose  1s  considered  the  LOAEL  based  on significant  reductions 1n  packed cell
volume and red  blood cell  numbers  1n  females,  and  albumin/globulin  ratios
and significant Increases 1n SGPT and absolute  and  relative liver  weights  In
both  sexes.   Based on  a  NOAEL  of  125 mg/kg/day,  the DWEL  1s  derived  as
follows:

    Step 1 - RfD Derivation
                  »«-                 • -4.17x10-- mg/kg/day
                                          (rounded to 4x1 0'* mg/kg/day)
04450                                VIII-17                         11/15/91

-------
where:
    125 mg/kg/day = NOAEL   reflecting   the   absence   of   dose-related
                    effects In mice (U.S.  EPA, 1988)
    3000          = combined  uncertainty  factors: 100  to  account  for
                    1ntra- and  Interspedes  extrapolation,  10  for  the
                    use of a  subchronlc study  for  RfD derivation, and  3
                    for  the   lack  of  reproductive/developmental   and
                    supporting chronic  toxldty data

    Step 2 - DUEL Derivation

                 DUEL . (4x10-' mq/kq/day) (70 kq) =
                                (2 l/day)
where:
       4xlO~2 mg/kg/day = RfD
       70 kg            = assumed body  weight of an adult
       2 l/day          = assumed volume of water consumed by an adult

    Fluorene — In  a  13-week  subchronlc  bloassay 1n  CD-I  mice  (U.S.  EPA,
1989a),  quantitative  data  were   reported  on  the  adverse  health effects
associated with oral exposure  to  fluorene.  From  these results, the U.S.  EPA
      /
(1991a)  Identified  a LOAEL  of 250 mg/kg/day for hematologtc  effects;   the
corresponding NOAEL 1s 125 mg/kg/day.   Using  this  NOAEL,  the  OWEI  1s derived
as follows.

    Step 1 - RfD Derivation

             RfD m  (125 mq/kq/day)  = 4
-------
    3000          = combined uncertainty  factors:  100  to  account  for
                    1ntra- and  Interspedes  variability,  10 for the use
                    of a subchronlc study for RfD derivation, and 3 for
                    the    lack    of    reproductive/developmental    and
                    supporting  chronic  toxldty data

    Step 2 - DUEL Derivation

                 nuci    (4x10"* mq/kq/dav) (70 kg)
                 DWEL = 	(2 i/day)          = K4 m9/l
where:
    4x10~2 mg/kg/day = RfD
    70 kg            = assumed  body weight of an  adult
    2 a/day          = assumed  volume of water consumed by an adult

    Naphthalene ~ Quantitative  results  on  the  chronic  oral  toxldty  of
naphthalene have  been provided  by Schmahl   (1955).   Groups  of 28  BD I and
BD II rats received naphthalene (estimated dally dose 10-20 mg) 1n the diet,
starting when the  rats were -100 days  old;  the experiment was terminated on
the 700th  experimental day when a total dose of  10  g/rat  was achieved.  No
treatment-related effects  were  observed.   The use of this study to derive an
      /
oral RfD has  been  questioned by the  U.S.  EPA (1989c) because this study was
designed  to  assess  the   cardnogenlcHy  of naphthalene,  small  numbers  of
animals were  used, and  there   1s  uncertainty  about  the actual  dose admin-
istered.

    Subchronlc studies with mice  and rats by NTP (1980a,b) and Shopp et al.
(1984) support a  lower chronic  NOEL  determined  by the Schmahl (1955) study.
In the NTP  (1980a)  study,  mice were  treated by gavage with 0, 12.5, 25, 50,
100 or  200 mg/kg/day naphthalene, 5 days/week for  13 weeks.  Comprehensive
hlstologlc  examination of  the  high-dose   and  control  groups revealed  no

04450                                VIII-19                        11/15/91

-------
 treatment-related  lesions.   Clinical  signs of toxlclty occurred at  or  above
 the  100 mg/kg/day  dose,  but  no  effects  were  observed at  or  below the  50
 mg/kg/day  dose.   No adverse  effects  on mortality, weight  gain,  1mmunolog1c
 effects, clinical  chemistry and hematologlc  parameters  or  organ  weights were
 observed 1n  mice given a 53  mg/kg/day  dose  by  gavage for 90 days  (Shopp  et
 al., 1984).  H1stolog1c examinations were  not'performed  In  this  study.   Sub-
 chronic  NOELs  from  these  studies were  35.7 mg/kg/day  (NTP,  1980a) and  53
 mg/kg/day  (Shopp et al., 1984).

    Derivation of  quantitative data  on  the oral toxldty of naphthalene are
 not  derived  because concurrence  on  the most appropriate  study  on  which  to
 base the calculation has not been  reached (U.S.  EPA,  1989c).

    Pyrene — In  a  13-week   subchronlc  bloassay  1n  CD-I  mice  (U.S.   EPA,
 1989b),  quantitative  data  were  reported  on  the  adverse  health  effects
 associated with  oral  exposure to pyrene.  From these results, the  U.S.  EPA
 (1991a)  Identified  a  LOAEL  of 125 mg/kg/day for nephropathy accompanied  by
 changes  1n absolute and relative  kidney weights; the corresponding  NOAEL  1s
 75 mg/kg/day.  Using this NOAEL,  the  DWEL 1s  derived  as follows.
    Step 1  - RfD Derivation

                   (75 mq/kq/day)  =  ? ^Q_2  mg/kg/(jay
                        3000        (rounded  to  3xlO~* mg/kg/day)
where:
    75 mg/kg/day = NOAEL reflecting the  absence  of nephropathy  In mice
                   (U.S. EPA,  1989b)
    3000         = combined  uncertainty  factors:  100  to  account   for
                   Intra- and  Interspedes variability,  10 for the  use
                   of a subchronlc study for RfD derivation, and 3  for
                   the  lack of  reproductive/developmental  and support-
                   Ing chronic toxldty  data.
04450                                VIII-20                          11/15/91

-------
    Step 2 - DUEL Derivation
            nun    (3x10*2  mq/kg/day)  (70 kg)   ,
            UWtL  =         ,~                 =  '•
                                                (rounded to 1 mg/a)
where:
    0.025 mg/kg/day = RfD
    70 kg           = assumed body weight of an adult
    2 I/day         = assumed volume of water consumed by  an  adult

Carcinogenic Effects
    The majority  of  health effects data  for Individual  PAHs concerns  their
potential  as  carcinogens.    The   data  base  for  these  compounds  consists
entirely of  animal  studies.   There are  numerous  case reports and  epldemlo-
loglc  Investigations  on human  health effects  of PAH-contalnlng  materials.
These,  however,   have  been  reports  on  exposures   to  environmental  or  to
occupational ly  generated  mixtures  containing  PAHs  and other   compounds.
There are no reports of exposures  to  Individual  PAHs with the exception of -a
case  study  of  skin  painting  of  benzo[a]pyrene.    The exposed  subjects
      /
developed verrucae,  wart-Uke benign  lesions  that  regressed upon  cessation
of  treatment  (Cott1n1  and Mazzone,  1939).   This observation Is  significant
1n  that  H  1s  not unlike  the  process  observed 1n animals skin painted with
PAHs.   In  response  to  a carcinogenic  PAH, animals  generally  develop  non-
malignant lesions  (paplllomas)  that may regress upon cessation of  treatment
or  may  progress   to  carcinomas  1f  treatment  Is continued or  followed by  a
promoting treatment.

    As reviewed 1n Chapter V,  much of the  data  on PAH cardnogen1c1ty comes
from skin-painting bloassays, subcutaneous  Injection  studies and to  a  lesser

04450                                VIII-21                         09/23/91

-------
extent  from  experiments  wherein  PAHs  were  administered  Intratracheally or  by
Inhalation.   There  are  comparatively  few  oral  bloassays.  The  relevance  of
data from  skin  painting  bloassays  to  evaluate  the potential  for  Induction  of
nonskln  cancer  1n humans has  been  a  subject of  discussion.  The  example  of
benzo[a]pyrene  may  be   Instructive   1n   this   context.    This   PAH  causes
                                   •
Injection  site  sarcomas   upon  subcutaneous  administration, skin  tumors  when
applied  topically,   lung  neoplasms   when  delivered  Intratracheally  or  by
Inhalation and  forestomach  tumors  In  rats  when  given  orally.    It  appears
that benzo[a]pyrene  produces  neoplastlc growth  at the site of delivery  when
at  least  two conditions   are  met:   that the tissue Is capable of  metabolism
of  benzo[a]pyrene to reactive forms, and  that  the tissue 1s of a type  that
normally  undergoes   some  degree  of  proliferation.    The  requirement   for
metabolism  of  benzo[a]pyrene  1s  not  a  limiting  factor for  formation  of
distant  site tumors; many  metabolites  can be  transported 1n the  blood  and
the majority of tissues assayed show some capacity for PAH metabolism.

    The other PAHs In  this  document have been  studied to  a lesser  extent and
by  fewer exposure routes  than  has  benzo[a]pyrene.  There  are  a  few examples,
however, to  Indicate that when a PAH  1s positive 1n  skin  painting bloassays,
H  will produce tumors when  administered by  other routes.  Benz[a]anthracene
and d1benz[a,h]anthracene, which produce skin tumors  when applied  topically,
also produce neoplasms when  delivered  orally  (Klein, 1963;  Bock  and  King,
1959; LaMnov and Soboleva,  1938;  Lorenz and Stewart, 1947,  1948; Snell and
Stewart,  1962a,  1962b).   It  appears  justified  to  say that  evidence  of
cardno- genlclty from a PAH  skin-painting  bloassay  Is  a cause  for  concern
and  should  not be  Ignored  when  evaluating  a  PAH as   a  potential  human
carcinogen.


04450                                VIII-22                          11/15/91

-------
Classification of PAHs  as  to  Potential for Human Carc1noqen1c1ty
    Assignment of compounds to a classification  based  on their  likelihood  of
producing carcinogenic  effects 1n humans was done  according  to the U.S.  EPA
Guidelines  for  Carcinogen Risk  Assessment  (U.S.  EPA,  1986).   Using  these
criteria  the PAHs  described  1n  this  document may  be  classified  1n  the
following two groups.

    Group D.  Not classifiable as to  human  cardnogenldty.  This  1s  due  to
a  lack  of specific  human  evidence  and  Inadequate animal data  for cardno-
genldty.

    Acenaphthylene.   One  skin painting  blassay  1n mice showed no Increase
1n  tumor  Incidence.   One assay  for  mutagenldty  1n   Salmonella  was  not
positive.

    Anthracene.   Studies  wherein  rats  were administered anthracene  orally
did  not  result  1n  tumor  Induction.   Lung Implantation  and  skin  painting
     /
bloassays,  and   subcutaneous,  Intraperltoneal   and Intracerebral  Injection
likewise have not  shown a tumorlgenlc  effect.   Mutagenldty for  Salmonella
strain TA97 has  been reported.

    Benzorq.n.llpervlene.   A  bloassay by  Intrapulmonary Injection of  rats
was considered Inadequate for evaluation of this  PAH  due to  the presence  of
Impurities;   Two skin-painting  studies  for  complete   cardnogenldty  and
three Initiation-promotion assays  In mice were  negative.  There are data  to
suggest  that  this  compound  may  act  as  a   co-cardnogen  for benzo[a]pyrene
applied   to   mouse   skin.    Benzo[g,h,1]perylene   was   mutagenlc  for   S.
typhlmurlum.
04450                               VIII-23                         09/23/91

-------
    Fluoranthene.   Skin  painting  bloassays  for  complete  carclnogenlcHy
have  not  shown  positive  results.   Initiation/promotion  protocols  wherein
fluoranthene was used  as  the Initiator, as well  as  a  subcutaneous  Injection
study, were  negative.   There 1s some evidence that  fluoranthene  served  as  a
cocardnogen  for  benzo[a]pyrene.   Fluoranthene  produced  an  Increase  1n
Incidence  1n lung adenomas  for  males  and  females  combined  In a short-term jm
vivo  bloassay  generally considered not adequate for evaluation  of  carclno-
genlcHy.  Evidence for genetic toxldty of fluoranthene Is equivocal.

    Fluorene.   Two  oral  bloassays  reported  no  Increase ,1n  tumors  as  a
consequence  of  fluorene treatment.   Both assays  had  deficiencies  limiting
their  usefulness.    IARC  considered  two   apparently  negative  skin-painting
bloassays  to  be  Inadequate  for-evaluation;  a more  recent Initiation/promo-
tion  assay was  not  positive.   Two subcutaneous  administration studies  were
negative.  Genetic  toxicology data are limited but negative.

    Naphthalene.   Bloassays  of  naphthalene by the  oral  route and as a  skin
tumorlgen were negative or  Inadequate for evaluation.   An Inhalation and  an
1ntraper1toneal  Injection study were  negative, and a  subcutaneous  Injection
study was  Inadequate for evaluation.   Supporting data for  genetic  toxldty
are negative but  limited.

    Phenanthrene.   An  assay  1n which rats received  a  single oral  treatment
showed  no  Increase  In  tumor  Incidence.   Treatment  of  mice   topically,
subcutaneously and  1ntraper1toneally  has  not resulted  1n  tumor  Induction.
One study  reported  phenanthrene to be an   Initiator of  skin  tumorlgenesls  1n
CD-I mice  when  followed by  high  concentrations  of TPA.   This was  countered


04450                                VIII-24                          09/23/91

-------
by negative  studies  1n  three other mouse strains.  Phenanthrene and a  dlol-
epoxlde  derivative  were  mutagenlc   for  Salmonella  typhlmurlum  and   human
lymphoblast  cells.   Other  genetic  toxicology testing has not  shown  positive
results.

    Pyrene.   Intratracheal   Instillation  of  pyrene  and   Fe.03   particles
did  not  Induce  tumors  1n  hamsters.   Skin  painting  assays  1n  mice  for
complete cardnogenesls  or  Initiating capacity have been negative  or  Incon-
clusive.   Mice  Injected  either  subcutaneously or Intraperltoneally did  not
develop tumors, but  there  1s  evidence  that pyrene enhances  tumor1gen1dty  of
topically  applied  benzo[a]pyrene   1n  mice.  Both   positive   and   negative
results have been reported  for  assays of mutagenlc  effect; pyrene was  not
shown to transform mammalian cells.

    Group  B2.   Probable human carcinogen.   These  Judgments  were  based  on
sufficient animal evidence  In the absence of  human data for  Individual  PAHs.
      /
    BenzTalanthracene.   Benz[a]anthracene  produced  tumors   1n  mice treated
orally, and  1n  various  mouse strains  treated  Intraperltoneally,  Intrave-
nously, topically, subcutaneously  and  Intramuscularly.  It  was mutagenlc  for
Salmonella.  DrosophUa and  mammalian  cells and produced  DNA damage, SCE  and
morphologic  transformation  1n  cultured  cell.   Benz[a]anthracene was  positive
1n a mouse lung adenoma assay.

    BenzoTalpyrene.   Orally  administered benzo[a]pyrene   was carcinogenic
to three  mouse strains.    It  produced tumors when administered Intratrache-
ally  to  rats and  hamsters.   Benzo[a]pyrene  delivered Intraperltoneally  has


04450                                VIII-25                          11/15/91

-------
 Induced  tumors  1n  mice  and  rats.   It  1s  the best documented  experimental
 skin carcinogen producing  tumors  1n mice,  rats,  rabbits  and  guinea  pigs.   It
was  carcinogenic  when administered subcutaneously  to  mice,  rats,  hamsters,
guinea  pigs  and   some   primates.    Benzo[a]pyrene  has  produced   positive
responses 1n a number of  genetic  toxlology assays  In bacterial  and  mammalian
cells.

    Benzo[b]f1uoranthene.    Exposure  of  rats  by  lung  Implantation  resulted
 1n tumor formation  as did  1ntraper1toneal  exposure of  newborn mice.   A total
of three  skin  painting and  Initiation/promotion  studies 1n. mice were  posi-
tive as  was an assay by  subcutaneous  Injection of mice.   Benzo[b]fluoran-
thene was mutagenlc In a forward assay 1n Salmonella.

    Benzo[k]f1uoranthene.    Lung  Implantation  produced  tumors  In rats,  and
Initiation/promotion  protocols  1n two  mouse strains  resulted  1n  Increased
tumor  Incidence.    IntrapeMtoneal   Injection  1n  newborn  mice   produced
equivocal results.  Benzo[k]fluoranthene was mutagenlc 1n  a  forward  assay 1n
Salmonella.

    Chrysene.   Chrysene  produced  tumors  In   several  mouse  strains  when
applied topically  1n  assays  for  complete skin  cardnogenldty or In Initia-
tion/promotion protocols.   Several  early studies employing  Intramuscular  or
subcutaneous  Injection  of mice  and  rats produced   negative  or  equivocal
results.   Three  studies  wherein  neonatal  mice of two strains  were  exposed
Intraperltoneally  reported  Increased  tumor  Incidence  1n  liver  and  other
sites.   Chrysene produced  mutations In Salmonella and  chromosome  aberrations
and morphologic  transformation 1n mammalian cells.


04450                               VIII-26                         09/23/91

-------
    D1benz[a.h]anthracene.   Mice  of  various   strains  have  been  shown  to
develop  tumors  as a  consequence  of oral exposure  to  d1benz[a,h]anthracene.
Mice  were also  reported  to develop  tumors  after  pulmonary,  Intratracheal,
Intravenous  or  topical treatment.   Injection  site tumors  (Intramuscular  or
subcutaneous)  have  been  observed  In  mice,  rats,  guinea  pigs,  pigeons  and
unspecified  fowl.   Results  of  DNA damage,  mutation  and morphologic  trans-
formation assays have been positive.

    IndenoH .2.3.-cdlpyrene.   Lung  Implantation  of   1ndeno[l,2,3-cd]pyrene
produced  tumors  1n  rats.   Skin painting assays  for complete  carclnogenldty
and  Initiating  ability were  positive  1n  two  mouse  strains,  and  Injection
site  tumors  were reported  after  subcutaneous  exposure of a  third  strain.
Limited data Indicate mutagenldty for Salmonella.

Quantification of Carcinogenic  Effects
    Compounds  classified  as  Group  A,  B  and   C  carcinogens  are  generally
regarded  as  suitable  for  quantitative  risk assessment.   There are  studies
sufficient  to  classify seven  PAHs In  this  document  1n group  B2.  Many  of
these  studies,  however,  do  not  provide a suitable basis   for  dose-response
assessment for the following reasons:
    •   Studies do not give dose/response data.
        Studies contain only one dose level.
        Studies  use  exposure other  than oral  exposure (such  as  skin
        painting).
        Sample size 1s too small.

No  carclnogenldty  bloassays  using the  oral  route  of exposure  have  been
reported   for   benzo[b]fluoranthene,   benzo(k)fluoranthene,   chrysene   and
1ndeno[l,2,3-cd]pyrene.

04450                                VIII-27                          09/23/91

-------
    BenzTalanthracene.   In a  study by  Klein  (1963), male  B6AF1 mice  were
gavaged with a  3X  solution of  benz[a]anthracene  3 times/week for 5 weeks and
observed  for  either 437 or  547  days.   Incidences of both  lung  adenomas and
liver  tumors  were  Increased at  both  observations.  As  there was  only one
dose  of  compound administered, and  exposure  was of  short  duration,  deriva-
tion of a quantitative risk estimate based on this study Is Inappropriate.

    Benzofalpyrene.   Several studies  have reported  Increased  Incidence  of
alimentary  tract  tumors  1n  rodents  as  a consequence of  oral  benzo[a]pyrene
exposure.   In   the  majority of  these,  there  was a  single, gavage  exposure
(e.g., Muggins  and Yang,  1962; HcCormlck  et  al., 1981), or only one gavage
or  dietary  dose was  employed  (Berenblum and  Haran,  1955; G1bel, 1964; Chu
and  Halmgren  1965;  B1andf1or1   et  al.,  1967;  Wattenberg,  1972,  1974;
El-Bayoumy, 1985).   Trlolo et al.  (1977)  observed forestomach  tumor Induc-
tion  1n  female  mice  (9/group) fed 200 and  300 ppm benzo[a]pyrene  1n the
diet.  Treatment was  for a relatively  small percentage  of the  animals' usual
Hfespan;  that  1s,  12  weeks.    Effective  numbers   of  animals  were  not
reported.   In   part  of  the  Wattenberg  (1972)   study,  mice  receiving  two
dietary doses of benzo[a]pyrene were observed  to develop forestomach tumors.
Treatment was for <1 year,  and no concurrent controls  were reported.

    A quantitative  assessment  for oral exposure  to benzo[a]pyrene regarding
Its carcinogenic  effects  can  be  based on  the experiments reported  by  Neal
and Rlgdon  (1967), In which benzo[a]pyrene at  doses ranging between  1 and
250 ppm  1n the  diet  was  fed  to  strain CFW mice  for <197 days.  No tumors
were found  1n  the  control  group nor 1n  the groups  treated with  1,  10 or  30
ppm benzo[a]pyrene.  The Incidence  of  tumors,  however.  Increased between the

04450                                VIII-28                         09/23/91

-------
40  and  250 ppm  benzo[a]pyrene  dosages.   Stomach  tumors,  which were  mostly
squamous  cell  paplllomas but also  some  carcinomas, appeared  with an  Inci-
dence  significantly  higher  than controls  (p<0.001.  Fisher  Exact Test)  at
several doses.   Quantitative  risk  estimates for human cancer  Incidence  were
developed by U.S.  EPA  (1980d) using the linearized multistage  procedure and
more recently  by Clement Associates (1988,  1990a)  using a two-stage  model.
Krewskl and  Murdock  (1990) also applied the  ArmHage-Doll  and a  two-stage
model to these data.   These approaches  appear below.

    U.S.  EPA  (1980d)  Approach.   The  Nedl  and  Rlgdon  (1967)  data,  with
adjustments to approximate dally lifetime doses (Table VIII-3), were used  1n
calculation  of  a quantitative  risk  estimate  for  human  lifetime  cancer
Incidence  by  use  of  the linearized multistage procedure.   Tumor  Incidence
data at  the  highest  three doses were  not  used 1n  the extrapolation  due  to
lack of  fH  to  the multistage model.   [A  discussion of  the  fit  of data  to
the multistage  model  appears  In Human  Health Methodology Appendix to  the
                                                                           •
October 1980 Federal  Register  (45 FR 79379).]
      /
    A  carcinogenic potency factor  for  humans (q,*)  was determined  to  be
11.53  (mg/kg/day)"1.   An estimate  of  the  cancer  risk  from  consuming  1  yg
benzo[a]pyrene/l water  (unit  risk)  could  be  calculated  as follows:
    Unit risk - "•53 (mfl/kq/daY)-ix 2 I/day x 0.001  mq/vq . 3.29xl
-------
                                 TABLE VIII-3

  Incidence of Tumors 1n Mice Treated  Intragastrlcally with Benzo[a]pyrenea»b
Experimental Dose
(ppm diet)
0
1
10
20
30
40
45
50
100
' 250
Transformed Dose
(mg/kg/day)
0.0
0.1
1.3
2.6
3.9
5.2
5.9
6.5
13.0
32.5
Incidence0
No. Responding/No. Tested
0/289
0/25
0/24
1/23
0/37
1/40
4/40
24/34
19/23
66/73
aSource: Neal and Rlgdon, 1967

^Length  of  exposure  =  110  days;  length  of  the  experiment  =  183  days;
 Hfesrpan of mouse = 630 days; average weight  of mouse = 0.034 kg

cThe  Incidences  at the  highest three doses  were not  used  1n the  extrapo-
 lation due to lack of fit of the multistage model.
04450                                VIII-30                         09/23/91

-------
    There  are  substantial  difficulties with  the  procedure  used to  derive  a
risk estimate  from  the  Neal  and  Rlgdon (1967)  data  as  given In Table VIII-3.
Among the factors that render this data set problematic are  the following:
    1.  Exposure  time was  variable  1n  the higher  dose groups  (Table
        VIII-4).   The  exposure   time   was   reported  as  a  range,  and
        Individual animal data are now unavailable.
    2.  Animals were started on test at differing ages.
    3.  A  very sharp  Increase  1n tumor  Incidence was  noted  1n  those
        animals fed 50 ppm by comparison with those fed 45 ppm.
    4.  There was an  apparent leveling off  of  tumor  Incidence among the
        highest dose groups.

    In the U.S. EPA  (1980d)  use  of  the multistage model  the peculiarities  of
the data  set were handled by using one estimate  of exposure  time  (110 days)
and by  discarding data at  the highest three doses as  they did not  fit  the
model.  This precluded  use  of the highest  Incidence  data and  resulted In  the
odd situation of  eliminating a dose group  (50 ppm)  essentially equivalent  to
one (45 ppm) that was used In the modeling procedure.

    Clement  Associates  (1988).    An  alternate  approach  was  proposed   by
Clement Associates  (1988)  1n which  the Neal and  Rlgdon  (1967)  data were  fit
to a  two-stage  dose-response model.   This two-stage model  1s  a special case
of the Moolgavkar and Knudson (1981) and  Hoolgavkar  (1986)  cancer  risk model
as adapted by  Thorslund et  al. (1987)  to  account  for  exposure  to  known
levels of  carcinogens.   According  to  the two-stage model,  the  population  of
cells at  risk  for Induction of  cancer consists of stem  cells.  These cells
can divide,  undergo differentiation  to terminal, nond1v1d1ng  cells,  die,  or
undergo changes  that result  1n  a preneoplastlc  state.   It  1s  assumed that
this  last option  Includes a  mutation or  heritable change at a critical site.
04450                                VIII-31                          11/15/91

-------
trt
C3
OJ
                                                          TABLE  VIII-4

                                         Forestomach Tumors  1n Mice  Fed  Benzo[a]pyrene*
First Subgroup
Exposure
B[a]P ppm
X

0
1
10
20
30
40
45
50
100
250
250
250
250
250
250
250
100
100
5000

Age First
Exposed
toi
300
30
30
116
33
33
31
17
20
18
49
56
49
62
49
91
74
48
98

Age Last
Exposed
tfl
300
140
140
226
143
143
141
124
118
88
50
58
53
67
56
121
81
78
99

Age Last
Observed
tl
300
140
140
226
143
143
143
124 .
118
88
155
162
155 .
168
155
198
182
156
209
Second Subgroup

Age First
Exposed
102




67
101
71
22
24
20








180

Age Last
Exposed
tf2




177
211
181
219
146
185







•
181

Age Last
Observed
t2




177
211
183
219
146
185








294
No. with
Fores tomach
Tumors/No.
of Mice
r/n
0/289
0/25
0/24
1/23
0/37
1/40
4/40
24/34
19/23
66/73
0/10
1/9
1/10
4/9
3/10
26/26
0/10
12/18
17/33
          *Source: Adapted from Neal and Rlgdon  (1967)  by Clement  Associates  (1990a)
VO

-------
The  preneoplastlc  cell  then  has the  options  of division,  differentiation,
death, or further change  to a  fully  transformed  or  cancer  cell.   This  cancer
cell,  1f  permitted  to proliferate,  will  form  the  basis  of a tumor.   In  the
Moolgavkar  and  Knudson  (1981)  model,  these processes  are described  mathe-
matically by postulating  specific exposure-dependent  rates  for cell  changes.
A simplified version of the model can be expressed  1n the following manner:
                                  •
                     Kt)  = H0H-\f C0{v)  [exp(B-D)(t-v)]dv              (8-1)
where
       I(t)  = age-specific cancer Incidence at age t
       MO    = transition rate from stem to preneoplastlc cell
       M-j    = transition rate from preneoplastlc to cancerous  cell
       C0(v) = number of susceptible stem cells per Individual  target
               organ at age v
       B     = birth rate or rate of cell proliferation of preneoplastlc
               cells
       0     = death rate of preneoplastlc cells

It Is  likely  that at  least  some of the  Increased  Incidence Is  attributable
to the  fact that  the  50 ppm  dosed animals were  younger when  exposure  was
begun.   It 1s,  nevertheless,  appropriate  to  use  as  much  of  the data  as
feasible  In calculating  the  quantitative estimate.   It  Is  reasonable  to
assume  that  the biologic  processes  described  by  the above equation  can  be
affected  by  exposure   to  carcinogens   and  that  the  likelihood  of  their
occurrence  1s  a  function  of  exposure  time.    Thorslund  et  al.   (1987)  thus
developed  a  version  of   the  model  that  Incorporated  time  and  exposure
dependence.
04450                                VIII-33                         11/15/91

-------
    In  applying the  two-stage  model  to the  Neal  and Rlgdon  (1967)  data  1n
Table  VIII-5,  some  modifications were made by Clement  Associates  (1988)  and
some  parameters  were estimated.  In  the absence  of  experimental  Information
about  cell  stages and differences  1n  exposure over time,  1t was not possible
to  estimate the  Individual  background and  exposure-Induced  mutation  rates
for preneoplastlc  and transformed  cells  using tumor  rate data.   Nor  could
the relative  transition  rates  that correspond  to each stage  be  Identified.
Two exposure-Induced relative  transition  rates  and a  background  transition
rate  were  estimated from  bloassay  data.    If  these  transition  rates  are
linear  functions of  dose  (which  1s  likely at  low  doses), they  may  be
expressed   as   Mn   =  an  f   &nx   and   Mi   ~   ai   f   &ix   where  a   1s
the background  transition  rate  and 6 1s the  PAH-lnduced  transition rate  per
unit  of  exposure   for  each   stage.   It  was   assumed   that  the  factors
Influencing the background transition  rates  and  the  PAH-lnduced  transition
rates  have  the same  relative   effectiveness for  each  stage.   Under  this
assumption,   BO/<*O   =  B-i/a,    =  s»    tne  relative  transition   rate,   so
that MQ^ = OQO^l+Sx)2 = H(l+Sx)*, where  M =  o0or

    In this application of the  model  H was assumed  that  promotional  effects
result  from multiple molecular  Interactions.   Thus, the growth  rate  of
preneoplastlc   cells   G at  low  doses   1s  virtually  Independent of  exposure
level   so  that  G = 8 - D.   If  this  assumption 1s violated at higher doses,
the shape of  the curve will  have greater  curvature than  quadratic,  and  the
quadratic model will be rejected.

    Finally, as a first approximation H was  assumed that the  number  of  stem
cells   post-maturity,  CQ(v),  Is  relatively  constant  and  may  be taken to  be
unity [I.e..' CQ(v) * CQ * 1).
04450                                VIII-34                         11/15/91

-------
                                 TABLE VIII-5

      Variable  Exposure  Data  Used  to  Estimate Parameters  In the  Ingestlon
                    Dose-Response  Model  for Benzo[a]pyrenea
Dose Age First
(ppm) Exposed (days)
x ts
0
1
10
20
30
40
45
50
100
250
—
30
30
116
33
67
33
101
31
71
17
22
20
24
18
20
Age Sacrificed
(last exposed)
(days)
t
300
140
140
226
143
177
143
211
141
181
124
219
118
146
88
185
Number of Number of
Animals Forestomach Tumors
Exposed
n Observed Expected**
289
25
24
23
37
40
40
34
23
73
0
0
0
1
0
1
4
24
19
66
i
0.000
0.002
0.232
0.875
3.091
5.746
7.560
19.692
14.220
64.259
aSource:  Neal and Rldgon (1967) as adapted by Clement Associates (1988)

blt  was  assumed  that  one-half  the  animals  1n each  group  were  exposed
 during  each  of  the  Intervals reported,  ts  to  t.   x2 =  17.12,  d.f.  =  8,
 pssO.03
04450
VIII-35
12/31/90

-------
    Substituting  these  expressions [G,  M,  S, CQ(v)]  Into  equation 8-1  and
Integrating  yields  a cumulative  hazard  function H(x,t), so  that  the  prob-
ability  that  a tumor will  develop by time  t as a  result  of exposure to  a
level of genotoxlc agent x can be expressed as:
    P(x.t) = 1 - exp [-H(x.t)] = 1 - exp  [-M(USx)2][exp (Gt)  - 1  - Gt]/G2
                                                                        (8-2)
where
       M = background tumor rate parameter
       S = fractional  Increase  1n the  transition  rate  between  cell
           stages per unit dose, assumed  to be the same for  each stage
       G = B - D and 1s  the  exposure-Independent  growth  rate  of preneo-
           plastlc cells
       t = the time (or  age) at which risk 1s evaluated
    The  level  of  agent at  the  target  tissue  was  assumed  to be  directly
related to the administered  dose.  This  was  based on observations  of experi-
ments discussed  earlier  In which  the  rate of formation of the major  benzo-
[a]pyrene  dlol  epoxIde-DNA  adduct was  found to be  linearly related with
respect  to  dose  1n  the  forestomach,  lung,  and  skin (Perelra et  al.,  1979;
      /
Adrlaenssens et al., 1983).

    The  specific  dose-response  model  derived  for   benzo[a]pyrene  (equation
8-2) was  thus  a  restricted form of the  model developed  by  ArmUage and Doll
(1957), Moolgavkar and  Knudson  (1981), and Thorslund et al.  (1987).   It  was
restricted by  assuming  that G  Is  Independent of treatment  (x) and  that  the
two  transition  rates are  linear  functions  of  x with  proportional coeffi-
cients.   The consequence  of these  assumptions   Is  that at  constant   t  the
dose-response function  has only two parameters:
                           P(x)  =  1 - exp-A(l  + Sx)2                    (8-3)
                            2
where A = M [exp(Gt)-l-Gt]/G  and t = age at last observation.
04450                                VIII-36                         09/23/91

-------
    Use  of  this modification  of  the two-stage  model  for quantitative  risk
estimation for benzo[a]pyrene has several advantages.

    1.  At  low doses,  the  model  converges  to  a linear, .nonthreshold
        form.
    2.  Only two parameters  need to be estimated.
    3.  A stable point estimate of risk could be obtained directly.
    4.  The  mathematical  form  of  the  two-stage  model  1s  based on  an
        acceptable theory of cancer Induction.
    5.  The  model   Is   consistent  with  what  1s  known  regarding   the
        mechanism of  tumor  Induction by PAHs,  Including benzo[a]pyrene
        (e.g., exposure to cigarette smoke).
    6.  Data  from  all  dose  groups  reported  1n  Neal  and Rlgdon  (1967)
        could be used In the risk estimation.

    As  Indicated  earlier,  the  animals  1n the  Neal  and Rlgdon  (1967)  study
were  exposed  at various ages  and for  varying  times.   In order  to  describe
the dose-response relationship  for the  1ngest1on  of benzo[a]pyrene and  tumor
Incidence  for  each  exposure  group,  estimates  for   each  of  the   exposure
variables were made as  Indicated  1n  Table  VIII-5.  As  a first approximation,
each mouse was assumed to be exposed to one of  two exposure  patterns:
    (1)  those that were youngest at  first exposure and were  sacrificed
         at  the  earliest age  (e.g.,  for the  group  exposed  to  0.04  mg
         benzo[a]pyrene  1n  Table  VIII-4, were  first  exposed  at  age  33
         days and last exposed at age 143 days), and
    (2)  those that were oldest at  first  exposure and  were  sacrificed
         at  the  oldest  age  (e.g.,  for  the  0.04  mg/group,   were first
         exposed at 67 days  and last exposed  at 177 days).

    One-half  of  the mice In  each exposure  group  were assumed to have  been
exposed to each  exposure pattern.   Using the maximum  likelihood  method,  the
data 1n Table  VIII-5, and the  assumption  that  the tumors are  Incidental,  the
parameters 1n  the  hypothesized risk  equation  were estimated  as  outlined  1n
04450                                VIII-37                         09/23/91

-------
Gart  et al.  (1986).   The  resulting best-fitting model  has  the  parameters
estimates   y<* -»0,   G  =  0.194,   and   YB2  =  0.2142.    This  mathematical
model  gives a  statistically  adequate fH  (I.e.,  the resulting  x2 has  a  p
value  >0.01,  which  1s  the criterion  used for the  multistage model  by  the
U.S. EPA).  This fH 1s shown 1n Table VIII-5.
    The resulting model  1s  a quadratic function of  dietary  concentration at
high  exposure  levels.  This  model  1s assumed  to  be valid over  the longest
exposure  duration  employed  for a group  of animals; that  1s,  t-t =(219-22)/
7=28.143 weeks.  The  parameter  value  for  G' (0.194),  however,  1s Incompatible
with  Increases  1n  age-specific cancer death  rates that have  been  observed.
This  value  for  G Implies that  the rates  Increased at about  the 15th to 16th
powers  of  age  (since  h(t)=exp Gt-1   -at     ).  This  1s greater  than  the
highest  reported rate,  which  1s that  for  human  prostate  cancer  (k=13).
Values  for  k are usually between 4 and  8.  This large value  of  G  suggested
to  the authors  of  Clement  Associates  (1988)  that  benzo[a]pyrene   exerts  a
                                                                           •
promotional effect, which was not Incorporated Into the model.
      t
    To  calculate a  lifetime risk estimate for  humans  H was  assumed  that  a
mouse  consumes   13X  of  Us  weight/day   as  food,  a standard surface  area
adjustment  was  made  for differences  between species and  70  kg  was assumed
for human  body  weight.  The  low  dose linear term for humans  was calculated
to be 3.22 per(mg/kg)/day.

    Krewskl  and Murdoch (1990).   Krewskl and  Murdoch  (1990)  also made  an
attempt to  make greater  use of the available data 1n Neal and Rlgdon (1967)
shown  In  Table  VIII-4.   Appropriate  adjustments  were  made  for start  and
04450                                VIII-38                         11/15/91

-------
duration  of  exposure  and  age at  sacrifice.   Both  the  Arm1tage-Doll and  a
two-stage model adapted  from  Moolgavkar  (1986) with  exponential  expansion  of
preneoplastlc  cells,  were fH  to  the data.   As Is  shown  1n Table  VIII-6,
however, neither model  was  very  successful 1n predicting the  observed  tumor
Incidences.  The authors  suggested  that  this  failure to fH  the  data may  be
due to  an  underlying deviation  of  the theoretical  basis of  the  models  from
reality,  but  may  also  represent  Inadequate  experimental  technique 1n  the
generation of  data  or the use of  the approximate mathematical forms of  the
model.   Alternative  explanations for  the failure  of the  model  to  fit  the
data are the following:
        use of an  Inappropriate  method of adjusting  for differences  1n
        the exposure Interval  within an exposure  group;
    •   lack  of  an  appropriate  exposure-dependent  growth   rate  of
        preneoplastlc cell growth rate expression;
    •   background tumor rates poorly defined  by  the data;
    •   lack  of  data  on  whether   observed   tumors  are  paplllomas  or
        carcinomas.

    Clement  Associates  (1990a).   Hore   recently   a model   was  developed
Incorporating a different assumption  as  to how  to  treat the  exposure groups
containing animals with  different durations of exposure. • This model  differs
from  the  earlier  Clement Associates  (1988)   report  1n  that  both transition
rates  and  the  growth  rate   of  preneoplastlc cells  were  considered to  be
exposure-dependent.   To  this  end a  simple saturation assumption  was  used  to
define  the functional  form  for the  dose-dependence of   the  preneoplastlc
growth rate.

    The  cumulative  hazard  function  for   the two-stage model for  exposure
constant  at  level  x  over  the  Interval  tfl  to t.  for t=t, 1s  discussed
by Thorslund et al.  (1987).

04450                                VIII-39                          11/15/91

-------
o
4k
£»


O
o
-J
•V
ro
                                                          TABLE VIII-6


                          Forestomach Tumors 1n Mice Subjected to Variable Exposures of Benzo[a]pyrenea


Dose
Group

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19





Dose d
(ppm)

0
1
10
20
30
40
45°
50
100
250
250
250
250
250
250
250
100
100
5000




Exposure




Number

Period Time at Sacrifice Number of
t-|-t2 days


20-140
30-140
116-226
50-160
67-177
51-161
20-172
22-132
19-137
49-50
56-58
49-53
62-67
49-56
91-121
74-81
48-78
139-140



T days

300
140
140
226
160
177
163
172
132
137
155
162
155
168
155
198
182
156
252



Animals Exposed

289
25
24
23
37
40
40
34
23
73
10
9
10
9
10
26
10
18
33



Observed

0
0
0
1
0
1
4
24
19
66
0
1
1
4
3
26
0
12
17
X2
d.f.
P
of Animals with Tumors


Predicted

A-D Model
0.0
0.2
1.9
3.6
8.3
11.4
13.0
18.8
13.3
67.0
0.4
0.6
1.3
1.5
2.2
16.1
1.0
5.8
17.9
50.9
16
2xlO~s

M-V-K Model
0.0
0.2
1.4
2.7
6.3
8.8
10.4
23.8
11.1
66.6
0.4
0.8
1.6
1.7
2.5
15.5
1.1
5.6
21.9
44.7
15
9xlO~s
aSource: Adapted from Krewskl and Murdoch (1990)

^Corrected from 0.05 (rounded value) given by authors

-------
    To  fH  the  model  H was assumed that the exposure groups are  made  up  of
two subpopulatlons with different exposure durations of  one  of  two ranges  as
shown  1n  Table  VIII-4.   It was  also  assumed  that the first and  second  sub-
populations  1n  any exposure  group j  are a  proportion  of  the total of  Y*
                                                                            J
and  1-Y«;  Y*  may be  considered  as  an unknown  parameter.   The  propor-
tion of animals responding In exposure group  j*1s:

       P(x,t) = Yj[l-exp-H(x.t01,tfl,t1)]+(l-Yj)[l-exp-H(to2.tf2,t2)]

When Y=! (animals exposed under uniform conditions),

                        P(x,t)=l-exp-H(x,tortfl.t1).

    The data  1n  Table  VIII-4 were fH to the model with the addition of two
zero  dose  groups.  The  control  groups  reported by Neal  and  Rlgdon (1967)
showed no Incidence of forestomach tumors.  Using zero Incidence  data 1n the
model may result  1n  some  cases 1n unstable parameter estimates;  1t was  thus
      /
decided to  Include some  nonzero historical .control  data.  Mice used  by  Neal
and  Rlgdon  (1967) were described  as  "an  Inbred CFW  strain  kept   In  this
laboratory for 8  years".   The  CFW mouse  1s, 1n  fact, an outbred  strain  from
Carworth Farm (CF) and was derived from  Swiss Webster.   The  outbred  stock  1s
still  maintained  by  Charles   River  Breeding  Laboratories  (Crl).    It  1s
believed that the  Inbred  mice  from the  testing  laboratory are  not likely  to
be  genetically  Identical  to other  Inbred  CFW  colonies or  to other  Swiss
Webster mice.   However,  forestomach  tumor  Incidences  for untreated  animals
considered to be  similar  to  the test population  are given 1n Table  VII1-7.
04450                                VIII-41                          09/23/91

-------
                                 TABLE VIII-7

          Historical Control Incidence Data for Forestomach Tumors 1n
                     Strains of Swiss (Webster) Bred Mice
Strain
Swiss
random-bred
CPU
SPF Swiss
(Webster)
SWR/J Swill
/
CFW
Texas colony
Sex
M
F
H
F
H
F
M
F
unknown
Age
(days)
_ _
—
--
700
608
300
Tumor Type
papllloma
--
—
'squamous cell
carcinoma
squamous cell
carcinoma
—
Incidence
1/99
0/99
0/100
0/203
0/28
0/38
2/268
1/402
0/289
Reference
Toth et al.,
1976
Sher, 1974
Prejean
et al., 1973
Rabsteln
et al., 1973
Neal and
Rlgdon, 1967
04450
VIII-42
09/23/91

-------
These  data  Indicate  that  forestomach  tumor   Incidence  was  quite  low  1n
Swiss-derived mice  In  the 1970s.  The  data  chosen for Inclusion were  those
for  SWR/J  mice  reported  by Rabsteln  et al.  (1973).   These  data  have  the
advantages of being from mice of a known average  age  and  of  having  a  nonzero
Incidence of squamous cell  carcinomas  1n both  males and  females.   The  SWR/J
Incidence data are consistent wlth'those  reported  for  the test  population at
300 days of age.

    Two  forms  of  the  model were  used.   The  first  used  the  simplifying
assumption  that  each  of  the  two   subpopulatlons of  the  exposure  groups
contained  half  the  animals  (y^l/2).   The   second   assumption  made  was
that  the  preneoplastlc  cell  growth   rate   reverts   to background  after
exposure; 1n other words,  G*(x)=G(0).   The growth rate was  defined to  be of
a general functional  form:
                                             bx

E
                                         *
where m 1s a given  Integer,  here  taken  to  be 1  as a conservative assumption;
b and C are unknown parameters to be estimated from the data.
      /

    To obtain  maximum  likelihood  estimates of  the unknown parameters,  the
general approach  discussed by Gart  et  al.  (1986)  was used (Table  VIII-8).
No adequate  fit  to the  full  data set was  obtained.   If,  however,  the  data
were restricted to  the  lower dose, longer exposure groups  (presumably  those
most relevant for low-dose continuous human  exposure  estimates)  a reasonable
fH was obtained.  Table VIII-9 presents comparisons  of estimates derived by
Clement Associates (1988, 1990a),  and Krewskl and Murdoch  (1990).
04450                                VIII-43                         11/15/91

-------
                                 TABLE VIII-8

        Maximum Likelihood  Parameter  Estimates  for Two-Stage Model with
                      Saturation of Growth Rate Function
     Parameter
           Physical Meaning
Estimated Numerical
  Values for All
Continuous Exposure
       Data
A


6(0)


S
u(x,t)=P(x,t)-P(o,t)
      - U730
      x=lx!0~s
    product of background
    transition rates

    background growth rate of
    preneoplastlc cells

    relative cell transition rates
    per unit of exposure (mg/g 1n
    food)

    relative Increase of preneo-
    plastlc cell growth rate per
    unit of target dose equiva-
    lent exposure from mg/g 1n
    food

    saturation coefficient
    exposure (mg/g food)

    low-dose risk
   2.4569xlO~10


   1.1995xlO'2


   845.56



   159.32
                                                             16.303
                                         3.00x10"*
                                         2.97xlO~s
                                    Model

                              p(x,t)=l-exp-H(x,t)
H(x,t) =
                               fa(X)
                                      {exp[G(x)t]-G(x)t-l}
                            G(x) = G(0)
                          —1
                          UCxJ
04450
                 VIII-44
           09/23/91

-------
in
o
 en
 —
 vo
                                                           TABLE VIII-9


                     Forestomach Tumors 1n Mice Subjected to Exposures  of  Benzo[a]pyrene  for  at  least 70 Days3
Number of Animals with Tumors

Dose
Group



1
2
3
4
5
6
7
8
9
10
11
12




Dose d
(ppm)



0
1
10
20
30
40
45
50
100
250
0
0



No. of
Animals
Exposed



289
25
24
23
37
40
40
34
23
73
268
402




Observed


Krewskl and Murdoch
Predicted


Clement Associates
A-0 Modelb M-V-K Modelc . M-V-K:In1t1at1on


0
0
0
1
0
1
4
24
19
66
2
1





0.0
0.2
1.9
3.6
8.3
11.4
13.0
18.8
13.3
67.0
—
—
X2
degrees of
P


0.0
0.2
1.4
2.7
6.3
8.8
10.4
23.8
11.1
66.6
—
—

freedom


1988
0.0
0.0
0.2
0.9
3.1
5.7
7.6
19.7
14.2
64.3
—
—
9.94
7
0.192

1990
0.0
0.0
0.2
0.9
3.1
5.8
7.2
19.4
14.2
64.9
2.6
0.4
18.08
9
0.034
M-V-K:
Initiation
and Promotion
0.0
0.0
0.0
0.2
0.7
2.4
3.7
18.8
18.9
66.1
2.0
1.0



 Source:   Clement  AssocHes  (1990a)

 Arm1tage-0o1l  model

"Moolgavkar-Venson-Knudson model

-------
    A difficulty with  the  previous  estimate  by  Clement  Associates  (1988)  was
that  the  value  of  6  was  unreaHstlcally  large.    This   difficulty   was
addressed  by  Clement Associates (1990a) by  deriving  a G,Q.   for humans.   It
was  assumed  that  the  human G_  1s  equal  to the mouse GQ  times a  constant
equal to  the mouse  Hfespan  of 2  years divided by  a  human Hfespan of  76
years.  The  result 1s GQ=0.1152.   To  relate this value  to  a power of  age,
they  solved  for  the  value of k (the number  of  stages  In  a multistage  model)
that  will  cause  the same  relative  Increase  In  the age-specific risk  for  an
Individual at  one-half Hfespan compared  with  full  Hfespan.   For  6=0.1152
and Hfespan =76, Clement  Associates (1990)  found that  k-7.?3.   As  Indicated
In an earlier section, most human values for k  fall between  4 and  8;  for  the
U.S.  population  the  value  for  k  for  stomach cancer  1s ~7.   They  concluded
that the value of G._.  for humans  of 0.1152 1s 1n a  reasonable range.

    A maximum  likelihood  estimate  for  humans was  obtained.   Under an addi-
tive  risk  model  the  risks become a  linear  function  of dose  as  x  approaches
0.   The additional  risk  over  background  was   expressed as u(x,t)=P(x,t)-
P(o,t),  where P(x,t) 1s the probability  of cancer by  age  t given an exposure
level of  x.   As  the exposure  x becomes small,  one can derive  the  approxi-
mation  u(x,)~L(t).x,  where  L(t)  1s   the  low-dose linear  term.   Extensive
application of the  chain rule of derivatives for  the model   1n  Table  VIII-8
resulted 1n the following equation:
             L(t) =        exp[G(0)t] . [2(S-b)+bG(0)t][1-P(o,t)]
    Substituting the maximum  likelihood  parameter estimates In Table  VIII-8
Into this formula provides  a  value of L(t)=29.63 for t=730 days.  To  change
the estimate  Into  units  of mg/kg/day, the  assumption  was made that a mouse
consumes 13X of Us  body  weight per day.

04450                                VIII-46                         11/15/91

-------
Thus, the low-dose linear term for mice
                            =  29.63  x  0.001/0-.13
                            =  0.2278 per(mg/kg)/day
To  obtain  the  low-dose  linear   term  for humans  a  standard  assumption  of
surface area equivalence  between  mice and humans was made  and  the resulting
maximum-likelihood estimate of the low dose linear was found:
                  °-2278   ^           « 2-90 Per(mg/kg)/day
                            V 0.034 kg
    An  approximate  conditional upper  bound was  then  calculated by  Clement
Associates (1990a).  The  low-dose  linear  term was considered to  be  a linear
combination of  exposure-related parameters  b  and S.   By  assuming  that  the
age specific rate function for  controls was  known (that 1s, the parameters A
and G(0)  were  fixed  at  their  maximum likelihood  estimates),  the  low-dose
linear term could be expressed as

                   L(t)=H(o,t)  . [l-P(o.t)]  {2S+b}[G(0)t-2]}

where P(o,t) = l-exp-H(o.t)
      /
               A
and H(o,t) =       {exp[G(0)t]-G(0)t-l}
             G(0)2

The equation  was  solved   using t=730, A=2.4599xlO~»  and   G(0)=0. 11992 . from
Table  VIII-8.    An  upper  95X  bound  on  L(t)  was  obtained by  substituting
values  for  S  (relative  transition  rate)  and  b  (promotional  effect)  that
maximized L(t),  subject to certain constraints.   In  this procedure  the upper
bound on the promotion  parameter, b,  was  reduced  from  Us  maximum likelihood
value  by  about  2-fold.   The  human upper bound  of  the low-dose  linear  term
thus obtained was 5.88 per(mg/kg)/day.

04450                                VIII-47                         11/15/91

-------
    In  U.S.  EPA (1991b) Issue was taken with  the  manner  In  which the linear
slope  and  the upper bound  were  calculated.   The  authors'  Interpretation  of
the above  slope factor  1s  that  1t represents  an  lower bound  rather  than  an
upper  bound;  1t  Is  the slope or  tangent at dose 0.  They further objected  to
the  following:   (1) not all  parameters  were  Included  1n  the process  of
obtaining  the  upper  bound,  but rather  only  two  of  the  dose-dependent
parameters;  (2)  1t  was  not  considered  biologically reasonable  to  require a
more than  SOX reduction 1n the cell growth  rate In  order  to obtain an upper
bound;  and  (3) the  historical  background tumor  Incidence  used was  from a
mouse  strain not used  1n the bloassay.

    To  these authors the Inability to derive an  upper  bound by starting from
the parameter  values  associated  with  the point  estimate  of dose  response
suggests a peculiarity of  the model.  It can be shown  that a SOX decrease  1n
the cell  growth  rate  value can  significantly  alter the  shape  of  the  dose
response model.
                                                                           •
    As  an  alternative,  they  provided  an   upper  bound  by  extrapolating
linearly from the  10X response  point  to the  background  of  an empirically
fitted  dose-response  curve.   It  was  noted  In  cases  wherein  the  fitted
dose-response  curve 1s  not  linear at  low doses  that  the potency  slope  or
upper  bound  can be  defined  as a  secant  from a  point on  the dose-response
curve  to  the  zero  dose  point.   Similar  concepts  and  approaches  have  been
proposed by Krewskl et al.  (1986,  1991)  and  Gaylor and  Kodell (1980).  Their
results Indicate that  potency slopes thus  calculated are  comparable to those
obtained  from  a   linearized  multistage  procedure  for   the  majority  of
compounds  Investigated.


04450                                VIII-48                          11/15/91

-------
    The upper  bound  calculated  1n  U.S.  EPA (1991b) on  the  Neal and Rlgdon
(1967) data as modeled by Clement Associates (1990)  1s  9.0  per(mg/kg)/day.

    U.S. EPA  (1991b).   The  authors of  this report  chose a model to  reflect
the exposure  pattern  In the Neal  and  Rlgdon (1967) study; that  1s,  partial
lifetime   exposure   over  different   (presumably  developmentally   varied)
portions of  their lifetime.   To  this end  they used  a We1bull-type  dose-
response model  that  could accomodate  partial  lifetime exposure.  To derive
this model  H was asssumed that  the hazard  rate  for tumor  occurrence  Induced
by benzo[a]pyrene can be given  by

                           hi(t)  =/(t-s)k-2D(s)ds

where D(s)  = f(d)
           = a-jd^d^*. ..+amdm for some positive  Integer m
             If tfl <  s < t1  and
      D(s)  = 0 otherwise
     and d  = dose given to animal at time s.

The  dose-related hazard  function  h,(t)  Implies  that  the  early  benzo[a]-
pyrene  exposure contributes more to  the  hazard  rate  than  does the  later
exposure.   This  assumption 1s  consistent  with  the observation that a single
dose  of benzo[a]pyrene of  a  certain magnitude  Is  sufficient  to Induce
papHlomas  after a sufficiently lengthy observation  period.
04450                                VIII-49                         12/10/91

-------
The cumulative hazard by time T for background can be given by
                                       T>
                               Ho(T)  «/atk-1dt
                                     = (l/k)aTk
The  cumulative  hazard for T>t  for  the hazard  Induced  by benzo[a]pyrene can
then be expressed as
                       Hi(T)  =-T [(T-t0)k-(T-tT)k].
                               v K- i *K

Thus, the probability  of  tumor  observation' by  time t (given that animals are
exposed  only  to  benzo[a]pyrene  during  the  time  Interval   [tQ,  t,])  has
the form
             P(d,t) = l-exp{-(q0+q1d+...fqmdm)[(t-t0)k-(t-t1)k]},
This  model  was  considered  to  be  equivalent  to  the  multistage model  that
would  be  used when  animals  are exposed  to benzo[a]pyrene for  their  entire
      /
lifetimes.  A  quadratic  model with m=2 was  found to be adequate to  fit  the
Neal and Rlgdon  (1967) data.  The  results  of the modeling are given In Table
VIII-10.   U.S. .EPA  (1991b)  reported that  data  fit  was  adequate for  all
points  except  those  from  the 50 and  100  ppro treatment groups.  They point
out that the poor fit may  be attributable  to biologic  factors rather  than to
statlsltlcal reasons; they note that  the  animals 1n these dose groups  were
started on  test  at  a  younger   age (<20  days)  than were  animals  1n other
groups.
04450                                VIII-50                         11/15/91

-------
                                 TABLE  VIII-10

             Predicted vs. Observed Tumor Incidence and Data Used
                    to Calculate  the  Dose-Response  Modela»b
Dose
(mg/g diet)
0.000
0.001
0.01
0.02
0.03
0.04
0.045
0.05
0.10
0.25
to
(day)
0
30
30
116
50
67
50
20
22
19
tl
(day)
300
140
140
226
160
177
160
172
132
137
t
(day)
300
140
140
226
160
177
162
172
132
137
Observed0 Predicted
Incidence Rate
0/289
0/25
0/24
1/23
0/37
1/40
4/40
24/34
19/23
66/73
(0.00)
(0.00)
(0.00)
(0.05)
(0.00)
(0.03)
(0.10)
(0.70)
(0.82)
(0.90)
0.00
0.00
0.00
0.02
0.04
0.08
0.11
0.30d
0.40d
0.98
aSourc«:  U.S. EPA, 19915

bto  =  age  of mice  (1n days)  when exposure  began, t]  = age  of mice  (1n
 days)   when  exposure was  terminated,   t  = age  of mice  (In  days) when  the
 study  was  terminated.   Data on  the   control  and  the  first nine  low-dose
 groups were  taken  from Table 1  of Nell and Rlgdon  (1967).   Data from other
 groups were  not Included  because  animals 1n  these groups were  exposed  to
 high doses of B[a]P for only very short periods.

C0bserved number of animals with tumors (Incidence rate)

dThe predicted  value  lies outside  the 95X  confidence   Interval  calculated
 from  the  observed  response.   This crude  test  1s used  to  determine  whether
 or not the model reasonably predicts  the observed response.
04450                                VIII-51                         11/15/91

-------
    Parameters were estimated by the maximum likelihood method as follows:
                   q2 = 2.1xlO"5
                   k = 3.13.
                                  • •
    A  slope  factor  at  the 10X  point  (d=  2.4  ppm  or  0.3 mg/kg/day)  was
calculated  by setting  tQ  =  0,  t.  = 730,  and  t  =  730.   The human  upper
bound thus obtained was 4.5 per(mg/kg)/day.
                        V

    Brune  et al.  (1981).  The study by Brune  et al  (1981)  Is  the  only
reported benzo[a]pyrene Ingestlon study  of 2  years  duration.  Benzo[a]pyrene
was administered to Sprague-Dawley rats  1n either the diet  or  by gavage In a
solution  of  1.5X  caffeine.  Effective  numbers  were  not  reported;  1t  was
assumed that  the number of  animals started on test  (32  males and 32 females)
constituted  the  group examined  for  tumors.   Incidence  rates  for males  and
females combined are given In  Table  VIII-11.

    A linearized multistage procedure was used  to  calculate an upper  bound
slope  factor (ql*) .for  benzo[a]pyrene   from these data.   A  body  surface
equivalence  assumption  was used for  Interpedes conversion;  that 1s,  the
animal slope was multiplied by a factor equal  to
where
       U.  = 70 kg (assumed adult human weight)
       W  = 0.4 kg (assumed rat weight)
        a
       and dose 1s expressed 1n mg/kg/day
04450                                VIII-52                         11/15/91

-------
                                TABLE VIII-11
        Incidence to Forestomach Only and Total  Contact  SHe  Tumors  1n
            Sprague-Dawley  Rats  Exposed to Benzo[a]pyrene by Gavage
                                or  In  the Diet3

Dose
(mg/year)
Gavage
0
6
18
39
Diet
0
6
39

Median Survival Time
(days)

102
112 .
113
87

129
128
131
Tumor

Forestomach Only

3/64
11/64
25/64
14/64C

• 2/64
1/64
9/64
Incidence

Total Contact S1teb

6/64
13/64
26/64
14/64C

3/64d
3/64d
10/64d
aSource:  Adapted from Brune et al.,  1981
^Includes forestomach, larynx and esophagus
cData  from this  group  were  not  used  1n  calculation  due to  the  shorter
 survival  time.   No  adjustment  was  done  as  time-to-death  data  were  not
 available.
dThe Incidence of total tumors 1n males  for  the control,  low- and  high-dose
 groups was 3/32, 3/32 and 8/32, respectively.
04450
VIII-53
11/15/91

-------
    Higher  slope factors were  obtained  by U.S. EPA  (1991b)  from the gavage
data  (27.1  per(mg/kg)/day  based  on  total   contact-site  tumors  and  27.5
per(mg/kg)/day  for forestomach  tumors  only)  than  were calculated  from the
dietary  Intake  study  [4.7  per(mg/kg)/day  based  on  total  tumors   and  3.8
per(mg/kg)/day  for forestomach  tumors].   It   1s  not  possible  to  determine
from  the Brune  et  al.  (1981)  paper  whether  the enhanced  tumor  response 1s
due to  the effects of gavage  (potential  Irritation),  to the co-carcinogenic
activity of caffeine,  or to some combination of both factors.

    In  the  dietary, but  not  1n the gavage component of the study, more total
contact-site tumors (larynx,  esophagus  and forestomach)  were  observed 1n the
males  than  1n  the   females.  Although  1t was  possible  to  separate  total
contact-site  tumors   Into  Incidences  for  males and females,  Information 1n
the paper  was  not adequate  for  the same adjustment for  forestomach  tumor
Incidence.   In order  to derive a  potency estimate based  on male  rat  data
only,  the  slope factor  calculated  by U.S. EPA (1991b)  on  the  combined  male
and female data was  multiplied by a factor of 1.5.   The SOX  Increase was
      /
used  to adjust  for   the risk  1n  cancer  risk for  male-only total  tumors.
Slope factors  based  on male  rat  total  contact-site  tumors or  forestomach
tumors only were 7.1  and 5.7 per(mg/kg)/day respectively.

    Chouroullnkov  et  al.  (1967).   Chouroullnkov et  al. (1967)  was  another
relatively  long-term  study  (14 months)  wherein albino mice of  unspecified
strain  were administered a  total  estimated dose  of 8  mg  of benzo[a]pyrene
mixed  'with  olive oil   In the  diet.   The  Incidence of  forestomach tumors was
0/81  for  the  controls and 5/81  for  the treated animals.   As there  was  only
04450                                VIII-54                         11/15/91

-------
one  non-zero  dose the  slope  factor was  calculated  1n U.S.  EPA (1991b)  by
direct extrapolation  from  the observed response of  0.062  at 0.63  mg/kg/day
to the background rate of 0:
                                                                  3
           (0.062/063)  x  (70 kg/0.03 kg)    x (24 months/14 months)
                             = 6.5 per(mg/kg)/day
where (24/14)3 1s a factor to adjust for  less-than-Hfetlme  observation.
Choice of a Quantitative Estimate for  Benzofalpyrene
    Slope  factors  described  In  the  foregoing  text  are   presented   for
comparison In Table VIII-12.   Note  that  the slope factors obtained  from  the
gavage portions of  the  Brune  et  al.  (1981) study are not  Included.  This  1s
because  of  the  uncertainty   as  to   magnitude  of   co-carcinogenic  effect
presented by the  caffeine  vehicle and the potential   Irritation component  of
gavage.  Only results of dietary  studies are considered  1n this  Instance  for
basis of the quantitative  risk estimate.   Also  eliminated  from consideration
at  thfs  point  1s  the  slope factor at dose  0   (maximum  likelihood  estimate)
derived  1n  Clement  Associates  (1988),  In  order to facilitate  comparison
among like estimates (upper bounds).

    As Indicated  1n Table VIII-12, these  estimates  span less than  an  order
of magnitude.  Each  1s  based  on  a Iess-than-opt1mal   but acceptable  data  set
from studies 1n two species of outbred rodents.   Each estimate 1s  based on a
low  (tose   extrapolation  procedure  that   entails   the  use  of   multiple
assumptions and default procedures.
04450                                VIII-55                         11/15/91

-------
                                 TABLE VIII-12
       Slope Factors for Humans Based on Benzo[a]pyrene Feeding Studies
Study
Nell and
Rlgdon, 1967



Brune et al.,
1981



Chouroullnkov
et al., 1967
Slope Factor
per(mg/kg)/day
11.5
5.9
9.0
4.5
4.7
7.1
3.8
5.7
6.5
Comments
Linearized multistage, highest
points dropped (U.S. EPA, 1980d)
Two-stage, conditional upper bound
(Clement Associates, 1990)
Clement two-stage, slope from 10%
response (U.S. EPA, 1991b)
We1bu11-type model (U.S. EPA,
1991b)
Larynx, esophagus, forestomach
tumors, male and female rats
Larynx, esophagus, forestomach
tumors, males only
Forestomach tumors only, male and
female rats
Forestomach tumors only, male rats
only
. Extrapolated from the observed
response.
04450
VIII-56
11/15/91

-------
    The  least  acceptable  of the  estimates Is  that  presented  by U.S.  EPA
(1980d)  of  11.53 per(mg/kg)/day based  on  the  Neal  and Rlgdon  (1967)  data.
This application of the  linearized multistage model does not  account  for  the
variable  times  of  exposure or  the varying  ages of  exposure  of the  dose
groups.  In  fitting the  model  all  data from exposures  >45  ppm were dropped.
In  1987  an Agency work  group,  the  Carcinogen  Risk  Assessment  Verification
Endeavor  (CRAVE),  found  sufficient  reason to  revise   this  estimate 1n  the
light of Improved risk quantUatlon  procedures.   It was decided  at this time
not  to  Include  the   1980   quantitative  estimate  on  the  Integrated   Risk
Information System (IRIS).

    The  use  of  simplified   two-stage  Moolgavkar-Venson-Knudsen models,  has
permitted  the  use  of  more  of  the  data reported  by  Neal  and  Rlgdon  (1967).
The  two-stage  model   1s described  as  being  biologically  based.   It   1s
consistent  with  current   thinking  that   PAHs   act  as   carcinogens   by  a
multistage   process   with  a   small   number   of  stages   and  allows   for
Incorporation of  terms  that may model  other  than  Initiating activity.   In
      /
the  form of the  model  presented by Clement  Associates (1990a),  a term to
permit  modeling data  for  benzo[a]pyrene   acting  as  Us   own  promoter  was
Included.  This 1s not Inconsistent  with  the observation of benzo[a]pyrene's
activity as a  complete carcinogen  1n skin.  The  modified  two-stage model 1s
relatively   simple  requiring   the  estimation   of   a  limited   number   of
parameters.  As applied  1t  1s  linear at low doses as  long  as  the  background
Incidence 1s not  zero.   This application  has,  however, necessitated  the  use
of  control  data  of  -a  historical   nature  from  related  but  not  Identical
strains of mice.
04450                                VIII-57                         11/15/91

-------
    The   simplified   two-stage  model   does  not  accurately  predict  tumor
 Incidence  from high-dose short-term exposures.   It  may be  that  exposure of
 this  sort results  In tumor  Induction  by  mechanisms  sufficiently different
 from  chronic  low-dose exposure that similar models  could  not be expected to
 apply.  A  potential  problem discussed by Clement  Associates  (1990a)  1s that
 the   approximation   for   the  exact   hazard   function  used   to  generate
 mathematical  expressions  may  not  be sufficiently accurate.   They suggested
 that  Improved  accuracy would  be  possible  using the exact model 1f both birth
 and death rates of pre-neoplast1c cells were available.

    The use  of available  forms  of  the  two-stage model for  generating risk
 assessments has  come under some  criticism.  It would  appear  that this type
 of model  1s  best  defined  as  biologically  based when  the  several parameters
 given biologic meaning  can be estimated from  specific  data.   In the case of
 the   Neal  and  Rlgdon   (1967)  study,   the  data   are  Inadequate  to  this
 application.  A proposed future direction  1s  to Investigate  the use of human
                                                                           •
 ep1dem1olog1c  data for  the calculation of estimates  of  parameters  dealing
with   background   transition  rates   and   background   growth   rates   of
preneoplastlc cells.

    For  the   Neal  and  Rlgdon (1967)  data  1t  can  only  be  said that  the
two-stage model  provides  a convenient  curve  fitting tool, and  that  Us  use
 1s not  contralndlcated by  Information  on  potential carcinogenic mechanisms
for benzo[a]pyrene.
04450                                VIII-58                         11/15/91

-------
    The  use of  the Welbull  type model  1n a  curve-fitting  procedure  also
allows for  the consideration of  the  variable exposure  time  of  the  animals  1n
Neal and  Rlgdon  (1967).   A difficulty Is the lack  of  fH  to  the  50 and  100
ppm dose  group data.   In  a test  of this  model  to several other  benzo[a]-
pyrene  data  sets  1t  was  shown  that  1t  accurately  predicted  the   tumor
response  1n  mice  1n many  dietary  studies  but  not  1n  those wherein  adminis-
tration  was  by   gavage.   There  are  two   plausible  explanations  for  this
discrepancy:   (1) benzo[a]pyrene given  by  gavage Induces greater  Incidences
of  gastric  tumors than  does  1ngest1on;  (2) or  the effect  of PAH  continues
after exposure.   The latter  hypothesis  1s  supported by observation  of  tumor
Induction as a consequence of a single dose  of  benzo[a]pyrene.

    The Brune  et al.  (1981)  dietary  exposure  data provide  a more limited
dose range  than  do  those of Neal  and  Rlgdon  (1967).   For  the lower dose  of
the  two  treated  groups  the  Incidence  of   both   total   contact-site  and
forestomach  tumors are equal  to or less than the controls. The data are  not
                                                                           •
reported  as  single  sex  Incidences  for forestomach  tumors.   This  adds  some
      /
additional  uncertainty,  as  1t  appears from  Inspection of the  report  that
there was essentially  no  effect 1n  the  females.   It  should  be  noted  that
this study was done In  rats.

    The Welbull model,  with parameters that fit  the Neal  and Rlgdon  (1967)
dietary exposure  data,  tends  to  underestimate  the  tumor Incidence  observed
1n the gavage studies.

    The Chouroullnkov (1967) study offers the twin  disadvantages of  only  one
non-zero dose and less-than-Hfetlme exposure and observation.


04450                                VIII-59                         11/15/91

-------
    There  1s  little  basis  on  which  to make  a  recommendation  of a  single
slope  factor  from  those  listed  1n  Table  VIII-12.   One way  to present  a
quantitative risk estimate  for  benzo[a]pyrene  Is  as  a  range,  excluding those
estimates  considered  Inappropriate.   As discussed  previously,  use  of  the
11.5 per(mg/kg)/day  estimate  based on  Neal  and Rlgdon  (1967)  1s  not  recom-
mended  as  the  higher  doses  (and  tumor Incidences) were excluded from  the
modeling procedure  nor were duration differences  included.   The  three upper
bound estimates from two different modeling  procedures  applied  to these data
all  have  both   pros  and   cons,  and   represent  reasonable  estimates  from
problematic data.

    It  1s  reasonable  to exclude  the slope  factor of  6.5 per(mg/kg)/day  as
this  was   obtained   from   direct  extrapolation  from   the  observed  data
(Chouroullnkov et al.,  1967).   This  Is  a rather  unusual  approach  and  1s  not
directly comparable  with  the  other calculations.  Furthermore, there  are  no
special circumstances  that  dictate the  desirability  of  choosing this  unusual
approach.  There 1s  considerable uncertainty 1n applying a  linearized  multi-
stage procedure to this data  set:  there 1s  only one non-zero dose point  and
there 1s uncertainty regarding  the duration  extrapolation factor.   The Brune
et  al.  (1981) and  other data  Indicate that  for benzo[a]pyrene  gavage  and
dietary exposure may not be strictly comparable.

    For the  Brune  et  al.  (1981)  data,  the  most appropriate approach  Is  to
use  the  majority  of  the  data.    Total   contact  site  tumors.  Including
esophagus,   pharynx,  etc.,   are  a  reasonable  basis for  comparison with  the
data  obtained  for   mice,   which  developed  only  forestomach  tumors.   The
estimate derived by applying the linearized multistage  procedure  to  the data


04450                                VIII-60                         11/15/91

-------
for   total   contact  site   tumor   for  both   males   and  females   1s   4.7
per(mg/kg)/day.

    The range of the most acceptable data  sets  1s,  thus,  4.5-9.0 per(mg/kg)/
day.  The median Is 6.25 per(mg/kg)/day.

    There  Is  also  precedent  for recommending  a combined  risk estimate  by
using  a geometric  mean  (or other  combination  technique)  to  calculate  a
single  number  from several  slope  factors  of equal  merit.   As  discussed  by
Stlteler  and  Schoeny  (1991)   the  use  of  a  geometric  mean  offers  some
advantages over  other  central  tendency calculations for  values  derived  from
linearized multistage  models.   They further  recommend that, when  possible,
data  sets  be  combined before a  modeling  procedure when  a combined  estimate
1s  desired.    For   the  data  sets  1n  question   this   latter  approach  seems
Impractical 1n  view of  the very dissimilar experimental  protocols  employed.
Vater and Schoeny (1991) have set out  a series  of biologic  considerations  to
be  weighed  when  evaluating data  sets  for  combination.    These  criteria
Include  Judgments  as  to   study  quality,  common  mechanisms   of   action,
similarity of  tumor types  and  sites.  Among  commonalities  of the data  sets
for benzo[a]pyrene  are  the  following:  mode  of administration,  tumor  site,
tumor types,  and presumed  mechanisms  of action.

    There are precedents for using multiple data  sets  from different studies
using more than  one sex,  strain and species; for  example, quantitative  risk
estimates  for  DDT  (U.S.   EPA,   1985)  and  carbon tetrachloMde  (U.S.  EPA,
1984d).
04450                                VIII-61                          11/15/91

-------
    There  1s  less  precedent  for use of  combined  risk estimates based  on  a
mean  of slope  factors  obtained  by differing  modeling  procedures.   In  the
Interest  of  using  more  of  the  available data,  the slope  factor  of  5.79
per(mg/kg)/day, which Is  the geometric  mean  of the four  estimates chosen for
the range,  Is recommended.   A unit risk  for  drinking water  1s  calculated as
follows:
       5.79 per(mq/kq)/day x 2 a/day x 0.001 mq/Pq m K65xlo-4 per
where:
       70 kg   = assumed weight of adult human and
       2 a/day - assumed water consumption of adult human

    Concentrations  of  benzo[a]pyrene  corespondlng  to  lifetime  risks  of
10~s,   10~6   and   10~7   calculated   from   the   above   unit   risk   are
6.0xlO~2,   6.0xlO~»   and  6.0xlO~4   yg/l,   respectively.    At   a   12/04/91
discussion  of  the Carcinogen  Risk  Assessment Verification  Endeavor  (CRAVE)
Work  Group,  the  range,  median  and  geometric  means  were  found  to  be
acceptable.

    DlbenzCa.hlanthracene.  Two  studies  Involving  oral exposure to  dlbenz-
[a,h]anthracene  have been  reported  (Lorenz and  Stewart,  1947; Snell  and
Stewart,   1962a).    The  former   study  was  designed  to  compare   tumor
susceptlbll- 1ty of various mouse strains and  to test  the  suitability of the
olive oil-water emulsions as a vehicle.

    The  Snell  and Stewart  (1962a)  study  reports four  sets  of  tumor  Inci-
dences; alveolar carcinoma 1n male and female mice,  hemangloendothelloma for
04450                                VIII-62                         12/10/91

-------
males  and mammary  carcinoma  for  females.  Exposure  duration was  reported
only as  a range for each group, presenting a difficulty  In  the  use  of these
data for calculation of quantitative risk estimates.

    In the Errata:  PAH Ambient  Water  Quality Criterion  for  the Protection of
Human Health  (U.S.  EPA,  1982),  a midpoint  of the exposure duration range was
used  to  calculate  quantitative risk estimates  using  the  linearized  multi-
stage  procedure.   These are   presented  In Table  VIII-13.   It  should  be
emphasized that  the estimates  1n  this   table were prepared for  purposes  of
comparison  with  risk  estimates  for  other PAHs and  were  not  considered
suitable for calculation of criteria.

Special Considerations
    Toxlclty  Equivalence Factor Approach  for  PAHs  1n Group  B2.   It  would
be  very  useful  In  the hazard  assessment of a  diverse  but related  group  of
compounds  such   as  PAHs  to be able to prepare  quantitative estimates  of
biologic  effects  by comparison with  a  well-studied type compound..   Various
      /
attempts  have been made,  for  example,   to  compare carcinogenic  activity  of
Individual  PAHs  with  that  of  benzo[a]pyrene.    If such  comparisons  were
proven  valid  then  1t   would   become  feasible,   for  example,  to  estimate
carcinogenic  potency of  a  mixture  of Identified  PAHs by  applying a  suitable
model 1f  all activities were  additive.   Another  potential  use would  be  1n
regulating less well-studied compounds by  comparison with the  standard.  The
authors  of   the  Errata:   PAH  Ambient  Water   Quality  Criterion  for  the
Protection of Human Health  (U.S.  EPA, 1982) used  data  sets  for  several PAHs
administered  1n  skin bloassays to  generate  comparative potency  estimates.
Results of these  calculations  based on   skin painting bloassays  are  shown In
Table VIII-14.
04450                                VIII-63                         12/10/91

-------
                                 TABLE  VIII-13

    Cancer Potency Estimates  for Excess Risk  of 10~5  from Lifetime Exposure
            Based on Oral Exposure Data for D1benz[a,h]anthracene*
Spedes/Sex
House/M
Mouse/M
Nouse/F
Mouse/F
Slope Factor
per(mg/kg)/day
0.57
0.75
1.24
2.36
Tumor Type
hemangloendothelloma
alveolar carcinoma
alveolar carcinoma
mammary carcinoma
'Source: U.S. EPA, 1982
04450
VIII-64
09/23/91

-------
                                                                           T/    VIII-14
                                          Potency Indices for the Carcinogenic PAH Compounds Based on Skin Painting Data9
o
en
o






H- 1
1— (
1— 1
1
cr>
en
Compound
Benzo[a]pyrene '
Benzo[a]pyrene
Benzo[a]pyrene
D1benz[a.h]-
anthracene
Benzo[ b]-
f luoranthene
Benzo[k]-
f luoranthene
Chrysene''
Benz[a]anthracene
Indeno[l,2,3,cd]-
pyrene
Ql*
470
435 .
NA
299.62
35.64
NA
0.53
NA
NA
EOio (and 95X C.L.)
2.98x10-*
(8.62x10"*, 5.98xlO"»)
1.43xlO"»
(3.68x10"*, 2.49xlO"»)
NA
6.34x10"*
(3.24x10"*, 9.44x10"*)
5.0x10"»
(2.75xlO"«, 7.25xlO"»)
NA
0.35
(0.23, 0.47)
NA
NA
qi*
152.49
67.62
20.83
292.81
11.57
0.30
0.88
0.28
1.16
edio (and 95X C.L.)
9.33xlD"«
(6.54x10"*. 1.21x10"*)
(sisixlO"*. 2.57x10-*)
1.43xlO-»
(4.24xlO"». 2.44xlO"»)
6.16x10"*
(3.28x10"*. 9.04x10-*)
1.29x10-»
(8.54xlO"». 1.73xlO"»)
No point estimate
(0.35. none)
0.21
(0.34, 1.12)
0.73
(0.34, 1.12)
0.15
(0.08, 0.22)
Reference
Wynder et al., 1957
Wynder and Hofmann. 1959b
Blngham and Falk, 1969
Wynder and Hofmann, 1959a
Wynder and Hofmann, '1959b
Wynder and Hof roan, 1959b
Wynder and Hofmann, 1959a
Blngham and Falk. 1969
Wynder and Hofmann. 1959a
o
IO
CO
aSource: U.S. EPA. 1982; Chu and Chen.  1984
''Since  there  Is only one  dose group  for  chrysene, Ql*  Is calculated  by using  Kap1an-He1er  survival analysis  and the assumption  that the
 control group has a zero response.
NA - Not available
Ql* «  Estimate of  carcinogenic  potency based  on  the  Incidence  data  combined with  the t1rae-to-tumor data.   It  Is the 95X confidence upper
       bound for the linear coefficient In  the multistage  model.
qi* «  Estimate of carcinogenic potency based on the Incidence data  alone.  It 1s the 95X confidence upper bound for the linear coefficient In
       the multistage model.                                                                                                  .
EOjp, " Estimate of carcinogenic potency based on the Incidence data  combined  with the t1me*to-tumor data as the effective dose associated with
       a risk of 10X.
edjQ = Estimate of carcinogenic potency based on the Incidence  data alone as the effective dose associated with a risk of 10X.
NOTE:  For  Ql*  and  qi*.   higher   numbers  Indicate  Increased  potency  whereas   for   ED)Q  and  ed]g.   lower  numbers  Indicate  Increased
       potencies.

-------
    For a  complete  discussion  of  the  caveats  and assumptions Involved 1n the
generation  of  these estimates, refer  to  U.S. EPA  (1982).   The  authors  note
that  the  potency  estimates  1n Table  VIII-14 are  not  directly  comparable.
The q.j*  value 1s the  slope  of 95X upper  limit on  the  linear  coefficient 1n
the multistage  model   (as revised  by  Howe and Crump, 1982)  when  tumor Inci-
dence  data  are  used.   By  contrast  the Q * 1s  the 95X confidence upper-bound
for the  linear  coefficient  1n the multistage model developed by  Daffer et
al. (1980)  and  evaluated at  t=12  months.  This model  Incorporates time-to-
tumor  data for  Individual  animals dying  In  the course  of the  study.   The
ed,Q  1s  the dose level 'corresponding  to  the  10X Incremental  tumor response
when  Incidence  data   are used;  ED,Q  1s  similarly defined  for  use  with
time-to-tumor  data.   The Q * and  q,*  reflect estimates  of  carcinogenic
potency  at  low  doses  outside the  experimental  range.   By  contrast,  the
ED,Q   and   ed,Q  reflect  carcinogenic   potency   within   the  experimental
range, and  presumably  are not  as  model-dependent  as the  other  two Indices.
The authors  of  U.S.   EPA (1982)  ranked  the PAHs based  on  the  potencies
computed by the four methods  as given 1n Table VIII-15.

    It 1s  Interesting  to  compare  these  rankings  with a  set of  rankings based
on unmodeled  data.   The data  1n  Table VIII-16 were  prepared  by Santodonato
(1986)  from published  data  of  Conney (1982).   As  can  be  seen  from  both
tables,  the rank  assigned   to a   carcinogenic PAH  depends on  the  type of
model-based  procedure  applied,  the  carcinogenic   endpolnt  measured,   and
whether nonmodeled data are  used.  What 1s  not apparent, but Is  discussed 1n
U.S.  EPA (1982),  1s  that variations  1n study  protocol  for a given  route and
endpolnt change the overall outcome of the  test  and, thus, potency  estimates
04450                                VIII-66                         09/23/91

-------
171
                                                     TABLE VIII-15
                       Ranking9 of PAHs Based on Estimates of Potency 1n Skin Painting B1oassayb
              Ql1
          Ql*
        "10
         ed10
     Benzo[a]pyrene
     01benz[a,h]anthracene
     Benzo[b]fluoranthene
     Chrysene
01benz[a,h]anthracene
Benzo[a]pyrene
Benzo[b]fluoranthene
Indeno[1,2,3-cd]pyrenec
Chrysene0
Benz[a]anthracenec
D1benz[a,h]anthracenec
Benzo[a]pyrenec
Benzo[b]fluoranthene
Chrysene
01benz[a,h]anthracenec
Benzo[a]pyrenec
Benzo[b]fluoranthene
Benz[a]anthracened
Indeno[l,2,3-cd]pyrened
Chrysene^
     aRank1ng 1s most potent first to least potent last.
     bSource: U.S. EPA, 1982
     c»Compounds with the same superscript are Indistinguishable 1n potency. '
r\i
CO

-------
                                 TABLE  VIII-16

      The Carcinogenic Potency of Various PAHs  Measured 1n Two Different
                           Animal Bloassay Systems3

Benzo[a]pyrene
Benz[a]anthracene
Chrysene
D1benz[a,h]anthracene
Phenanthrene
Skin Study5
( tumor s/ymol)
22.5
0.15
0.36
13.4
0.03
Rank
1
4
3
2
5
Newborn Studyc
( tumor s/iimol)
4.5
0.57
0.26
111.7
0.14
Rank
2
3
4
1
5
aSource: Santodonato, 1986

bM1ce  received  a  single topical  application of  compound  followed  7  days
 later by twice weekly applications of TPA for 16-25 weeks (Conney,  1982).

cM1ce were  given  1.p. Injections  of  1/7,  2/7 and  4/7  of the  total  dose  of
 compound on the first, 8th and 15th days of life,  respectively.  The  animals
 were killed at 22-42 weeks of age (Conney, 1982)
04450                                VIII-68                         09/23/91

-------
and  ranks.   Variations  1n use of  solvents,  animals  of varying  susceptibil-
ity, and  both  length  of exposure and overall study duration  further  compli-
cate attempts to compare carcinogenic potency.

    While  the  absolute  rank will  vary as a  function of the  considerations
outlined above, certain commonalities can be seen.  In general,  three groups
of  PAHs  can be  distinguished.   The most potent  carcinogens  are  Invariably
benzo[a]pyrene  and d1benz[a,h]anthracene.   A  second group  of  carcinogens
having  Intermediate  potency consists  of  benzo[b]fluoranthene and  generally
(but  not   Invariably)  benz[a]anthracene,  chrysene   and   1ndeno[l,2,3-cd]-
pyrene.  The third group  consists  of weak carcinogens or PAHs  that have  not
shown   significant  . cardnogenlcHy   Including   phenanthrene,   pyrene   and
benzo[k]- fluoranthene.

    An alternative, more  quantitative approach to  risk  estimation for PAHs
has  been  proposed by  Clement  Associates  (1988,  1990b) and  other authors.
This entails deriving a numerical  estimate  of the carcinogenic potency of  a
particular  PAH  by comparison with  that  of benzo[a]pyrene.   This method 1s
not unlike  the  generation  of toxldty equivalence factors proposed for risk
estimation  of  chlorinated  dlbenzodloxlns  and dlbenzofurans proposed  by U.S.
EPA (1987c) and recently revised  (U.S. EPA,  1989d).
           £&fx$"        '    '
          .•7-WC •?.,.;••, .
    What i\i^done* .In  this   approach  1s  to  calculate  a   benzo[a]pyrene-
equlvalent^dose* using data  from  studies  (such as  skin painting  bloassays)
generally not  considered  suitable for quantitative risk estimation.  In  the
Clement Associates (1988,  1990b)  reports  only those data sets were  Included
wherein benzo[a]pyrene was  tested concurrently with one  or more  PAH.  This
                                                                i,

04450                                VIII-69                         11/15/91

-------
was  done  In  order  to  account   for   Inter-laboratory  variations,  varying
susceptibility to  carcinogenic  activity of  the  different  test animals,  vary-
ing  metabolic  capacity of  these  animals,  and  differences 1n  protocols  and
endpolnts  measured.   For  each report  considered,  the  comparison was  made
between benzo[a]pyrene  activity and  the activity  of  a  particular  PAH  1n that
same report.

    In  the Clement Associates  (1988)  report all  risk estimates  were  gene-
rated  using  the  two-stage model.  The  low-dose terms or  maximum likelihood
estimates  are  amenable to comparison whereas  It  1s felt  that  comparison  of
upper  bounds  has   less  meaning.   In  calculating  the estimated  transition
rates  from the two-stage  model on which  comparisons were  based,  the  assump-
tion was made  that paplllomas and other  benign tumors are observable clones
of  first  stage cells.   It  was further  assumed that carcinomas  can  develop
from paplllomas, and  thus the observation of either  a  papllloma or carcinoma
serves  as   evidence  of at  least one  transformation or   stage.   In  those
Instances wherein  benign and  malignant  tumors were  combined,  the  simple form
of the model equation was used:

                           P(x) = 1 - exp -  A(HSx)

The  form  of  the  model employing two  stages  was  used  for  data 1n  which
malignant tumors were reported separately:
                           P(x)  = 1  - exp  - A(USx)2
04450                                VIII-70                         09/23/91

-------
    To derive  the potency  for  each PAH  relative  to benzo[a]pyrene,  It  was
assumed  that  the PAH and  benzo[a]pyrene  have similar dose-response  curves,
but    that    H    takes   a   proportionately   larger    concentration    of
nonbenzo[a]pyrene  materials  to   Induce   the  same  response.   The  relative
potency,  Rj, 1s  simply  the  ratio  of  estimated  transition  rates,  with that of
benzo[a]pyrene taken as  1.0.

    The  result of a series of calculations  was  a range  of  relative  potency
estimates.   For   example,   for   benzo[b]fluoranthene  Individual   relative
potencies  from  the  three  data   sets  were  1.067,  0.874 a^d 0.232.   In  a
similar  fashion  the  ranges  of relative potencies presented  1n Table  VIII-17
were generated using data from studies described 1n  Chapter  V.   Using benzo-
[b]fluoranthene  as  an   Illustration, a range of  relative  potencies  spanning
an order  of magnitude was obtained.

    Clement Associates  (1988) proposed   that  selection   of  the  most  appro-
priate relative   potency  be  based   on  a  consideration   of  the  qualitative
      /
differences among the  studies.   Criteria used  1n  selection   of  the  most
suitable  studies  Included the following:
    1.  Relevance  of   route  of  administration  1n   the  bloassay   to
        presumed  human exposure
    2.  Duration  of exposure approximating natural  Hfespan
    3.  Sample size
    4.  Inclusion of a concurrent vehicle control
          '••^-  --':«
    5.  Observance of a dose-response  relationship  consistent with  the
       model  (linear-quadratic)
    6.  Extent to which observed  responses cover the  possible  response
        range
    7.  Absence  of   complicating  variables  such  as  administration  of
        promoting agents

04450                                VIII-71                          11/15/91

-------
                                                                         TABLE VIII-17
                                                    Summary of  Relative Potency Estimates for  Indicator PAHsa
1NJ
CO
Compound
Benzo[ a ]pyrene
Benz[a]anthracene
Benzo[b]f luoranthene
8enzo[k]f luoranthene
Benzo[gh1]perylene
Chrysene
01benz[ah]anthracene
Indeno[ 1 . 2, 3-cd Jpyrene
Pyrene

House Skin Subcutaneous
Carclnogenesls Injection
Into Mice
1.0 1.0
0.145C
0.1678
0.0206
0.0151
0.00441
l.lll 2.82k. 4.501
0.021*. 0.0891

i
Test System
Intrapulmonary Initiation-
Administration Promotion
to Rats0 on Mouse Skin
1.0 1.0

0.140 0.258f. 0.1259
0.066 - 0.022f
0.022 0.0051
0.0409

0.232 ' 0.0741



IntraperHoneal
Injection In
Newborn Mice
1.0
0.057,
0.232.
0.040.

0.125.

0.013^
0.081.

0.524. 0.496
-------
    Based on  application  of  these  criteria  single  relative  potency estimates
were proposed  as  given 1n Table VIII-18.  Only  those  PAHs  classified  as  B2,
probable  human carcinogen, are  Included  1n this  listing  as 1t  1s  Inappro-
priate  to  calculate a quantitative  risk  estimate  for  those 1n group  D,  not
classifiable.

    Other approaches  for  obtaining a  single  estimate  are also feasible.   A
weighted average or some  other  numerical  combination procedure  could be used
In  those  Instances  wherein  a  range  of  relative potencies  1s  available.
Other criteria  than  those described above (or different emphasis  on certain
criteria)  could result 1n choices  other  than  those  1n  Table VIII-18.   It
should  be noted 1n  this  context that  the Deutsch-Wenzel et  al.  (1983) study
was  thought  to  be  less   reliable  than more  standard  skin  painting studies
because of the  unusual route  of  exposure  (lung Implantation  of  wax pellets).
This  experimental  protocol  also may  entail   confounding  variables such  as
trauma  or  Irritation  (which  could  have promoting  effects) or  changes  1n
metabolism of  the Implanted PAH.   It  Is worthy to  note that this  1s the only
assay wherein  1ndeno[l,2,3-cd]pyrene had  an elevated  response  by comparison
      /
with  the  generally  more carcinogenic   PAH   benzo[b]fluoranthene.   A  more
appropriate method  of  study  and relative potency  selection  may be to  employ
an objective method such  as that described by  DuMouchel and Harris (1983).

    In  1990  a  modified   TEF  approach  was   proposed  (Clement  Associates,
1990b).   In  order  to  Increase  the  precision  of   the  relative  potency
estimates,  the  analysis   Incorporated  time-to-tumor   Information,  vehicle
controls from  multiple experiments and historical  controls.  A form  of  the
two-stage model was used  that was  extended  to  accomodate  possible saturation


04450                                VIII-73                         11/15/91

-------
                                 TABLE VIII-18



                      Summary of PAHs Relative Potencies
Compound
Benzo[a]pyrene
Benzo[e]pyrene
Benz[a]anthracene
Benzo[b]f1uoranthene
Benzo[J ]f luoranthene
Benzo[k]fluoranthene
Chrysene
D1benz[a,h]anthracene
Indeno[l ,2,3-cd]pyrene
Relative
Potency
1.0

0.145
0.140
0.061
0.066
0.0044
1.11
0.232
Reference

Deutsch-Wenzel et
Blngham and Falk,
Deutsch-Wenzel et

Deutsch-Wenzel et
Wynder and Hoffman
Wynder and Hoffman
Deutsch-Wenzel et


al., 1983
1969
al., 1983

al., 1983
, 1959a
, 1959a
al., 1983
04450
VIII-74
09/23/91

-------
 of  processes  for  converting PAH to  reactive  metabolites.   Data recommended
 for  this  approach were exclusively  from a series of experiments on Induction
 of  lung epldermold carcinomas by Implantation  of  PAH  1n the lungs of female
 Qsborne-Mendel   rats.   These  data  were  collected by   a   single  group  -of
 Investigators.   Published  data  on  background  rates  of lung  carcinomas  1n
 female  Osborne-Mendel  rats  were  used 1n addition  to  assay vehicle controls
 to  provide  a  non-zero Incidence  for  the  zero dose  used   In  the two-stage
 model.   The relative potencies obtained  by  comparison of maximum likelihood
 estimates  are  given In Table VIII-19.  Relative potencies  for PAH-conta1n1ng
 mixtures were  used  to  test  addUlvlty assumptions.

     Rugen  et al. (1989)  assessed relative carcinogenic potencies of PAH as a
 basis   for  proposing  acceptable  exposure  levels   (AEL)  1n  drinking  water.
 They used  the  published  slope   factor  of  11.53  per(mg/kg)/day  for  benzo-
 [ajpyrene   (U.S.  EPA,  1980d)  and a  10~s risk level  to  determine  a 0.028
 pg/1 concentration as   the AEL.    Relative   potencies  or  relative  tumor
-doses (RTD) were determined as follows:
                RTD
 where:
        d,  =  dosage  of chemical  1
        n,  =  tumor frequency after m months of exposure to d,
        d~  -  dosage  of chemical  2
        n2  =  tumor frequency after m months of exposure to d2
 04450                               VIII-75                         11/15/91

-------
                                 TABLE  VIII-19
           Relative Potencies  for PAH and  PAH-Contalnlng M1xturesa»b
                      Material                     Relative Potency

          Benzo[a]pyrene                                1.0
          Benzo[b]fluoranthene                          0.1228
          Benzo[J]fluoranthene                          0.0523
          Benzo[k]fluoranthene                          0.0532
          Benzo[e]pyrene                                0.0070
          Indeno[l,2,3-c,d]pyrene                       0.2780
          Benzo[g,h,1]perylene                          0.0212
          Anthanthrene                                  0.3160
          Flue gas from coal-fired furnace              0.0542
          Diesel engine exhaust                         0.0028
          Gasoline engine exhaust                       0.0217
          Sldestream cigarette smoke                    0.0030
aRelat1ve  potencies   based   on  maximum  likelihood   estimates   from  lung
 Implantation data.
bTable adapted from Clement Associates (1990b)
04450                                VIII-76                         11/15/91

-------
Direct  comparisons  of dose and  tumor  frequency were made  for  several  skin-
painting assays.   No model was  applied  to the data but  rather  single  doses
were chosen as  the  basis  for  comparison.   Table VIII-20  lists  the RTD values
and data on which they were based.

    One end  use of a  series  of  relative potency would be  1n  the evaluation
of  PAH-contalnlng mixtures  (a  TEF  approach).  This  1s  an  alternative  to
assuming that all  carcinogenic PAHs  are  equlpotent  with  benzo[a]pyrene.   All
the means  of  comparing PAHs  described  1n this chapter  Indicate  that benzo-
[ajpyrene and d1benz[a,h]anthracene  are  of  comparable  potency  as  carcinogens
and the  other  PAHs described  In this  document less active.   In  the  absence
of  synerglstlc   Interactions  risk  estimates   of   PAH   mixtures  based  on
one-to-one benzo[a]pyrene equivalency will be overestimates.

    A  critical  review and  analysis  of  the  above  has not  yet been  done  by
U.S. EPA.  This will  be necessary before a  TEF or  other  comparative approach
                                                                           •
to  the  derivation  of  quantitative estimates for PAH can  be recommended.  At
this time It 1s thus  recommended that  a  quantitative cancer risk  estimate be
proposed only for benzo[a]pyrene.

    Interactions  With  Other  Chemicals.    As   reviewed   In   the  Synerglsm
and/or Antagonism Section  1n  Chapter V,  several  of  the PAHs have  been deter-
mined  to  be affected by  the presence  of  other PAHs or  other  substances.
Much of the research  1n this  area has  focused  on the promotion or Inhibitory
effect of noncardnogens  such as pyrene on a  known  animal  carcinogen,  often
benzo[a]pyrene.   The  route of administration  1s typically nonoral  (dermal,
subcutaneous Injection or  Inhalation).   There  are data for both  enhancement


04450                                VI11-77                         09/23/91

-------
o
.*»
f»

CTI
                                                                            TABLE VIII-20


                                                Comparison of Carcinogenic Potency of  PAH  Using Relative Tumor Dose*




»— 1
1
CD




PAH
Benzo[a]pyrene
D1benz[a.h]anthracene
Benzo[a]pyrene
Benzo[ J]f luoranthene
Benzo[b]f luoranthene
Benzo[a]pyrene
Indeno[ 1 , 2, 3-c , d Jpyrene
Benzo[a]pyrene
Benz[a]anthracene
Benzo[a]pyrene
Benz[a]anthracene
Observation Time
(months)
7
7
8
8
8
10
10
18
18
15
15
Concentration
(X)
0
0
0
0
0
0
0
0
1
0
0
.01
.01
.01
.1
.1
.005
.5
.02
.0 •
.001
.02
Tumor Incidence
(X)
75
45
. 85
65
20
50
30
50
14
33
2.5
Relative
Tumor Dose
1
1.67
1
13.1
42.5
1
167
1
179
1
264
Reference
Wynder
Wynder
Wynder
Wynder
Wynder
Wynder
Wynder
and
and
and
and
and
and
and
Blngham and
Blngham and
Wynder
and
Hoffman
Hoffman
Hoffman
Hoffman
Hoffman
Hoffman
Hoffman
Falk.
Falk.
Hoffman
Blngham and Falk.
. 1959a
. 1959a
. 1959a
. 1959b
. 1959b
. 1959a
. 1961
1969
1969
. 1959a
1969
          *Table adapted from Rugen et al.. 1989
o
(A
co

-------
and,. Inhibition of  biologic  activity  when PAHs are tested as simple mixtures
and as components  of complex mixtures.   The outcomes are dependent upon  such
variables  as  strain and  species,  whether  the  PAHs are  given  concurrently
(tests for  cocardnogenesls)  or applied  sequentially  (tests for  Initiation
and promotion) and the route employed.

    It has been postulated  that PAHs act  by similar mechanisms  as  Initiators
of cardnogenlclty, which allows an  assumption  of simple similar action and
dose addltlvUy.   This assumption, however, does  not  take Into  account  the
data,  albeit  limited,  on  both  PAH  promotional  and  Inhibitory effects.
Presentation of any PAH risk assessment  based on  add1t1v1ty must,  therefore,
be qualified;  theoretically both  under- and  over-estimation of risk could
result from an assumption  of addltlvUy.

Existing Guidelines. Recommendations  and  Standards
    A drinking water  standard  for  PAHs  as  a  class has been developed (U.S.
EPA,  1980d).   The  World  Health Organization European  Standards  for Drinking
      /
Water  recommends   a  concentration  of PAHs  not  to exceed  0.2  yg/i (WHO,
1970).  This recomendatlon  1s based on the composite analysis of six  PAHs  1n
drinking water. .  Fluoranthene,  benzo[a]pyrene,  benzo[g,h,1]perylene, benzo-
[b]flubranthene,  benzo[k]fluoranthene and 1ndeno[l,2,3-cd]pyrene were used.
These six  were not  chosen  on  the basis  of  potential  health  effects,  but
because  they  were considered  useful  Indicators of  the  presence   of  PAH
pollutants (Borneff and Knute,  1969).  A  quantitative risk estimate of 11.53
(mg/kg/day)'1  based  on the data  of Neal  and Rlgdon   (1967) was derived  In
U.S.  EPA  (1980d).   From  this,  the criterion for the  ambient  water  quality
for  PAHs  associated with human  lifetime cancer  risk  of 10~5  was estimated
to be 28 ng/l (U.S. EPA,  1982).
04450                                VIII-79                         09/23/91

-------
    In  the  occupational  environment,  a  Federal  standard has been promulgated
for  coke oven  emissions,  based  primarily  on  the  presumed  effects  of  the
carcinogenic  PAHs  contained  1n  the  mixture  as  measured  by  the  benzene
soluble  fraction  of total partlculate matter (U.S.  EPA,  1984b).  Similarly,
the AC6IH (1988) recommends  a workplace  exposure  limit for coal,  tar pitch
volatlles,  based on  the  benzene-soluble  fraction containing  carcinogenic
PAHs.   NIOSH  has  also  recommended  a  workplace   criterion  for  coal  tar
products  (coal  tar, creosote  and coal  tar pitch),  based on  measurements of
the cyclohexane extractable fraction.   These criteria are summarized below:

           Substance               Exposure Limit	        Agency
      Coke oven emissions     0.15 mg/m3 (soluble           NIOSH (1973b)
                              fraction), 8-hour              OSHA (1985)
                              time-weighted average
      Coal tar products       0.1  mg/m8, 10-hour            NIOSH (1973a,
                              time-weighted average         1977)
      Coal tar pitch          0.2 mg/m8, (benzene           ACGIH (1988)
      Volatlles               soluble fraction), 8-hour
                              time-weighted average
Special Groups at Risk
    There are several  human  subpopulatlons  that  are considered  to  be  at
Increased risks  from exposure to PAHs.  The  largest subpopulatlon  comprises
those who are occupationally exposed  to one or more of the PAHs. . Coke oven
emissions and coal   tar  pitch volatlles  Include PAHs such  as  naphthalene,
acenaphthene  and fluoranthene.   Industries  that expose workers  to  pesti-
cides, coal  tar,  creosote and fossil  fuel  oils  (I.e., steelworkers,  roofers
and auto  mechanics) appear to be at  higher risk.   Persons living  In Indus-
trial .areas,  or  where  exposure  to  auto exhaust  Is greater and  smokers  are
also considered to be at an Increased  risk from  PAH exposure.
04450                                VIII-80                         11/15/91

-------
    It  1s  known  that  PAHs  are  dependent   upon  metabolism  to  reactive
electrophlles for  their  activity  and that Individuals of  a  species,  humans
Included,  vary  1n  their  capacity  to undertake  this  metabolism.   It 1s also
known  that  the   1ndudb1lHy  of  AHH  1n  human  tissues  1s  genetically
determined.    The  degree  to  which  1nduc1b1l1ty  or  nonlndudblHty  of  AHH
figures as a risk factor  for human cardnogenesls, however, Is not clear.
                                  •

    Another  genetic  predisposing  factor may  be the  Individual's capacity to
repair certain  types of  DNA damage  that  can be caused by PAH metabolites 1n
a  fashion  that   1s  error-free and  does  not lead  to fixation  of  mutagenlc
events.

Summary
    There  were no data reported suitable for calculation of 1-day, 10-day or
longer-term health  advisories.   DWELs for  several  PAHs were  determined on
the  basis  of 90-day studies.  They are  the  following:  anthracene,  10.5
mg/l; fluoranthene, 1.4 mg/i; fluorene, 1.4 mg/l;  pyrene, 1.0 mg/l.
      /
    Welght-of-evldence  determinations for  likelihood  of  human cardnogenlc-
1ty were  these:  Group  D,  not  classifiable as  to  human cardnogenldty —
acenaphthylene,   anthracene,  benzo[g,h,1]perylene,  fluoranthene,  fluorene,
naphthalene, phenanthrene  and pyrene; group  B-, probable  human  carcinogen
—  benz[a]anthracene,  benzo[b]fluoranthene,  benzo[k]fluoranthene,  benzo[a]-
pyrene, chrysene, d1benz[a,h]anthracene and  1ndeno[l,2,3-cd]pyrene.
04450                                VIII-81                         11/15/91

-------
    Data  from three  studies wherein  benzo[a]pyrene was  given orally  were
modeled by  several  procedures.   Both a  two-stage and the  linearized  multi-
stage procedures were  used  to calculate upper bound  risk  estimates  from the
data of Neal  and Rlgdon  (1967).  Data  from both  male and female rats as  well
as  various  subsets of  the  published data from  the  dietary portion  of  the
study by Brune et al.  (1981)  were  used.   An extrapolation  was also done  from
the  observed  response  1n Chouroullnkov  et  al.  (1967).   The upper  bounds
judged  to  be   most   acceptable  range   from  4.5   per(mg/kg)/day   to   9.0
per(mg/kg)/day.  A geometric mean  of the  acceptable  upper  bound estimates 1s
5.8  per(mg/kg)/day  or   1.65xlO"4  per   wg/l.   Concentrations   of  benzo-
[a]pyrene  corresponding  to   lifetime  risks  of  10"5,   10~6 and  10~7  were
determined    to    be     6.0xlO~2,     6.0xlO~8    and     6.0xlO~«    vg/l,
respectively.   These  risk  estimates  were  found to  be  acceptable by  the
Carcinogen Assessment  Verification  Endeavor (CRAVE)  Work  Group 12/04/91.
04450                                VIII-82                         12/10/91

-------
                                IX.  REFERENCES

Abe, S. and M. Sasaki.   1977.   Studies  on  chromosomal  aberrations  and sister
chromatld exchanges Induced by chemicals.   Proc.  Jap.  Acad.   53:  46-49.

Abelson, S.M.  and  A.T.  Henderson.  1951.   Moth  ball  poisoning.   U.S.  Armed
Forces Med. J.  2:  491-493.

ACGIH  (American  Conference  of  Governmental  Industrial  Hyg1en1sts).  1988.
Threshold  Limit   Values  and  Biological   Exposure  Indices-  for  1988-1989.
Cincinnati, OH.

Adklns,  B.  Jr.,   E.W.   Van   Stee,  J.E.  Simmons  and  S.L.  Eustls.    1986.
Oncogenlc  response  of  strain  A/J mice  to Inhaled  chemicals.   J.  Toxlcol.
Environ. Health.   17:  311-322.

Adrlaenssens,  P.I..  C.M.  White  and  M.U.  Anderson.    1983.   Dose  response
relationships  for  the  binding  of  benzo[a]pyrene metabolites  to  DNA  and
protein   1n    lung,   liver  and   forestomach  of   control  and   butylated
hydroxyanlsole-treated mice.  Cancer Res.   43: 3712-3719.   (Cited  in  Clement
Associates, 1988)

Albert, R.E.,  M. Vanderlaan,  F.J.  Burns and M. N1sh1zum1.   1977.   Effect of
carcinogens on chicken atherosclerosis.   Cancer  Res.   37:  2232-2234.

Amacher, D.E.  and G.N.  Turner.   1980.   Promutagen  activation by  rodent-liver
postmltochondrlal fractions  1n  the  L5178Y/TK cell  mutation assay.   Mutat.
Res.  74: 485-501.
            *
04460                                IX-1                             10/07/91

-------
Amacher,  O.E.,  S.C.   Palllet,  G.N.  Turner,   V.A.  Ray  and  D.S.   Salsburg.
1980.   Point  mutations  at  the   thymldlne  klnase  locus  In  L5178Y  mouse
lymphoma  cells.   II.  Test  validation  and  Interpretation.   Mutat.   Res.   72:
447-474.

Am1n, S., E. LaVole and  S.S.  Hecht.   1982.  Identification  of metabolites of
benzo(b)fluoranthene.   Carclnogenesls.  3:  171-174.

Am1n, S., K. Hule and  S.S.  Hecht.   1985a.  Mutagenldty  and tumor  Initiating
activity   of  methylated   benzo[b]fluoranthenes.    Carclnogenesls.   6(7):
1023-1025.

Am1n S., N.  Hussaln, G.  Balanlkas,  K.  Hu1e and S.S. Hecht.   1985b.  Mutagen-
ldty  and  tumor  Initiating  activity of  methylated   benzo[k]fluoranthenes.
Cancer Lett.  26:343-347.
                                                                           »
Andervont, H.B.  1937.   Pulmonary'tumors  In mice.   I. The  susceptibility  of
the lungs of albino mice  to the carcinogenic action of 1,2,5,6-dlbenzanthra-
cene.  Public Health Report, Washington,  DC.   52:  212-221.

Andren, A.U. and  J.W.  Strand.  1981.  Atmospheric deposition  of partlculate
organic carbon  and polyaromatlc  hydrocarbons  to Lake Michigan.   In:  Atmo-
spheric  Pollutants  1n  Natural  Waters,  S.J.  E1senre1ch,   Ed.   Ann  Arbor
Science Publishers,  Inc., Ann Arbor, MI.  p. 459-479.

Andrews.  A.M.,  L.H.  Thlbault  and  U. Lljlnsky.   1978.   The relationship
between carclnogenlcHy  and mutagenldty of  some polynuclear  hydrocarbons.
Mutat. Res.  51:  311-318.

04460                                IX-2                            10/07/91

-------
Anzlulewlcz,  J.A.,   H.J.  D1ck  and  E.E.  ChlarulH.   1959.   Transplacental
napthalene poisoning.  Am. J. Obstet. Gynecol.   78:   519-521.

Appel, B.R.,  G.  Gu1rgu1s, I-S. K1m  et  al.   1990.   Benzene,  benzo(a)pyrehe,
and  lead  In  smoke  from tobacco  products  other  than cigarettes.   Am.  J.
Public Health.  80:  560-564.

Arcos, J.S. and M.F. Argus.  1974.   Chemical Induction of  Cancer.   Vol.  IIA.
Academic Press,  New York.  (Cited  1n U.S.  EPA,  1980d)
                                          *
Armltage,  P.  and  R. Doll.   1957.   A two-stage  theory of  cardnogenesls  1n
relation  to  the  age  distribution of  human  cancer.   Br. J.  Cancer.   11:
161-169.  (Cited In Clement  Associates,  1988)

Ashby,  J.   and  B.  Kllbey.   1981.   Summary  report  on  the  performance  of
bacterial  repair,  phage  Induction,  degranulatlon,  and  nuclear  enlargement
assays.  ITK  Evaluation  of Short-Term Tests  for Carcinogens.   Report  of  the
      /
International  Collaborative  Program.  Progress 1n Mutation  Research,  Vol.  1,
F.J.  de  Serres  and J.  Ashby,  Ed.   Elsevler,  North  Holland,  New  York.
p. 33-48.

Asokan, P., M. Das,  D.P.  B1k et al.  1986.  Comparative effects  of topically
applied nitrated arenes and  their  nonnltrated parent  arenes on cutaneous  and
hepatic  drug  and  carcinogen metabolism  1n  neonatal  rats.  Toxlcol.  Appl.
Pharmacol.   86:  33-43."

Athreya, B.H., A.K.  Swain and B.  D1ckste1n.   1961.  Acute hemolytlc  anemia
due to the Ingestlon of naphthalene.   Indian J.  Child Health.   10:  305-308.

04460                                IX-3                             10/07/91

-------
 Babson,  J.R.,  S.E.  Russo-Rodrlguez,  R.V.  Wattley et al.   1986.   Mlcrosomal
 activation   of   fluoranthene  to  mutagenlc  metabolites.    Toxlcol.   Appl.
 Pharmacol.  85: 355-366.

 Bachmann, W.E.,  J.W.  Cook,  A. Dansl et al.   1937.   The  production of  cancer
 by pure hydrocarbons IV.  Proc. Roy. Spc.  London.  123:  343-368.

 Badger,  6.H.,  J.W.  Cook  and C.L.  Hewett  et al.   1940.   The production  of
 cancer by pure hydrocarbons.  V. Proc. R.  Soc. London Ser.  B.   129: 439-467.

 Baker. R.S.U., A.M.  Bonln,  I.  Stupans  and  G.M.  Holder.   1980.   Comparison  of
 rat and guinea pig as  sources  of  the S9  fraction 1n the  Salmonella/mammalian
 mlcrosome mutagenldty test.  Mutat. Res.   71: 43-52.

 Baldwin, R.W.  1973.   Immunologlcal  Aspects  of  Chemical  Carclnogenesls.   In_:
 Advances 1n Cancer Research, G. Klein  and  S.  Welnhouse,  Ed.   Academic  Press,
 New York.  18: 1-74.
      /

 BarbleM 0., E.  Ognlo,  0.  Rossi,  S. Ast1g1ano  and  L. Rossi.   1986.   Embryo-
 toxlclty of  ben7o(a)pyrene  and some of  Us  synthetic  derivatives 1n  Swiss
mice.  Cancer Res.  46: 94-98.

 Barfknecht, T.R.,  B.M.  Andon,  H.G.  Thllly and  R.A. HUes.  1981.  Soot and
mutation 1n bacteria and  human cells.   In: Chemical Analysis  and  Biological
Fate: Polynuclear Aromatic  Hydrocarbons.   5th Int.  Symp.,  M.  Cooke and  A.J.
Dennis, Ed.   Battelle Press, Columbus,  OH.  p. 231-242.
04460                                IX-4                            10/07/91

-------
Barfknecht, T.R.,  R.W.  NalsmHh  and  R.3. Matthews.   1985.   Rat hepatocyte
primary  culture/DMA  repair test.  PH  311-TX-008-85.  5601-56-1   (unpublished
material).  Pharmakon Research  International,  Inc.,  Waverly,  PA.  Submitted
to  Texaco, Inc., Beacon,  NY.

Barry, G.  and  J.W.  Cook.   1934.   A  comparison of the action of some poly-
cycllc aromatic  hydrocarbons  In  producing tumors  of connective  tissue.  Am.
J. Cancer.  20: 58-69.

Barry, G.,  J.W.  Cook,  G.A.D.  Haslewood,  C.L. Hewett,  I. 'Hleger  and E.L.
Kennaway.  1935.   The  production  of  cancer  by pure  hydrocarbons-Part III.
Proc. Royal Soc., London.   117:  318-351.

Bartosek,  I.,   A.  Gua1tan1,   R.   Modlca,  M.  Flume  and  R.  Urso.   1984.
Comparative kinetics of oral benz(a)anthracene, chrysene and tMphenylene In
rats: Study with hydrocarbon mixtures.  Toxlcol. Lett.  23: 333-339.
      /
Bartsch,   H.,   C.  Halavellle,  A.M.  Camus  et  al.    1980.   Validation  and
comparative studies  on  180 chemicals  with  S.  typhlmurlum strains  and  V79
Chinese  hamster   cells  In  the  presence  of  various  metabolizing  systems.
Mutat. Res.  76:  1-50.

Basler,  A.,  B. Herbold,  S. Peter and G.  Rohrborn.   1977.   MutagenlcHy of
polycycllc hydrocarbons.   II.  Monitoring  genetlcal  hazards  of  chrysene in
vUro and In vivo. Mutat.  Res.   48:249-254.   (CUed  1n U.S. EPA, 1984c)
04460                                IX-5                            11/21/91

-------
Ben1ashv1l1,  D.  Sh.    1978.   A  comparative  study  on  the  action  of  some
carcinogens   Inducing   transplacental   blastomogenlsls   1n  rabbits.    Vopr.
Onkol.  24: 77-83.  (Cited 1n IARC, 1983)

Benjamin,  H., J.  Storkson,  P.  Tallas  and  M.  Parlza.   1988.   Reduction  of
                                   •
benzo(a)pyrene-1nduced  forestomach  neoplasms  1n  mice  given  nitrite  and
dietary soy sauce.  Food Chem. Toxlcol.   26(8): 671-678.

Benjamin,  H., J.  Storkson,  P.  Tallas  and  H. Parlza.  1989.   Inhibition  of
benzo(a)pyrene  Induced  forestomach neoplasla 1n mice by nitrite and  dietary
soy  sauce.   Abst.   1439.   Federation of American  Societies  for Experimental
Biology.  73rd Annual Meeting New Orleans,  LA.  March 19-23.  p. A474.

Benjamin, H., J.  Storkson,  A.  Nagahara  and  M.  Parlza.   1991.  Inhibition  of
benzo(a)pyrene-1nduced  mouse  forestomach  neoplasla  by dietary soy  sauce.
Cancer Res.   51: 2940-2942.

Benson', J.M., R.L. Hanson, R.E.  Royer,  C.R.  Clark  and R.F. Henderson.   1984.
lexicological -and  chemical  characterization of the  process  stream  materials
and  gas  combustion   products  of  an   experimental  low-BTU  coal  gaslfler.
Environ. Res.  33: 396-412.

Berenblum,  I. and N.   Haran.   1955.   The   Influence  of   croton  oil  and  of
polyethylene  glycol-400 on cardnogenesls  1n  the  forestomach  of the  mouse.
Cancer Res.  15: 510-516.
04460                                IX-6                            10/07/91

-------
Berenblum,  I.   and  R.   Schoental.   1942.   The  rate  of  disappearance  of
3:4-benzpyrene  from   the   mouse   after  subcutaneous  and   Intraperltoneal
Injection.  Blochem. J.  36: 92-97.

Bevan, D.R. and W.J.  Worrell.   1985.   The b1oava1lab1lHy of benzo(a)pyrene
adsorbed  to  carbon blacks.  In:  Polynuclear  Aromatic  Hydrocarbons: Mecha-
nisms, Methods and Metabolism,  8th Int. Symp., M. Cooke and A.J. Dennis. Ed.
Battelle Press, Columbus, OH.   p.  173-187.

Bevan, D.R., S.C.  Rlemer  and J.R.  Lakowlcz.   1981.   Transfer of polynuclear
aromatic  hydrocarbons  from partlculate  matter  to  membranes   measured  by
fluorescence  spectroscopy.   In;  Chemical  Analysis  and  Fate:  Polynuclear
Aromatic  Hydrocarbons,  5th Int.  Symp.,  M.  Cooke  and  A.J.   Dennis,  Ed.
Battelle Press, Columbus, OH.   p.  603-614.

Bevan,  D.R.,  M.R.  Ulman and  V.  Mack.   1988.   Effects of  partlculates  on
disposition of benzo(a)pyrene 1n  vivo.   Proceedings  of AARC Vol.  29.  March.
      /

B1anc1f1or1, C.,  F.  Caschera,  F.E.  Glornelll-Santllll  and  E.   BucdarelH.
1967.  The  action of  oestrone and  four chemical carcinogens 1n Intact and
ovarlectomlzed BALB/c/Cb/Se mice.   Br.  J.  Cancer.  21: 452-459.

Blngham, E. and H.L.  Falk.  1969.  Environmental carcinogens  - The modifying
effects of  cocardnogens  on the  threshold response.  Arch. Environ. Health.
19: 779-783.
04460                                IX-7                             10/07/91

-------
Blanton,  R.H.,  M.  Lyte, M.J.  Myers  and P.M. B1ck.   1986.   Immunomodulatlon
by  polyaromatlc hydrocarbons  In  mice  and  muMne cells.   Cancer  Res.   46:
2735-2739.

Blobsteln,  S.H.,   I.B.  We1nste1n,  P.   Oansette,  H.   Yag1  and  D.M.   Jerlna.
1976.  Binding  of  K- and non-K-reg1on arene  oxides and  phenols  of  polycycllc
hydrocarbons to polyguanyllc add.  Cancer Res.   36:  1293-1298.

Bock, F.G.   1964.   Early  effects  of hydrocarbons on mammalian  skin.   Progr.
Exp. Tumor Res.  4:126-168.  (CHed 1n  IARC, 1973)

Bock, F.G.  and M.  Burnham.   1961.   The effect  of  experimental  conditions
upon  the  concentration  of  hydrocarbons  1n  mouse  skin   after  cutaneous
application.  Cancer Res.   21: 510-515.

Bock, F.G.  and T.L. Dao.   1961.   Factors affecting the polynuclear  hydro-
carbon level 1n rat mammary glands.   Cancer  Res.   21: 1024-1029.

Bock, F.G.  and D.U. King.   1959.   A study of  the  sensitivity  of  the mouse
forestomach  toward  certain polycycllc  hydrocarbons.   J.  Natl.  Cancer Inst.
23: 833-839.

Bond. J.A.,  A.M.  Gown, H.L.  Yang,  E.P.  Bendltt  and M.R.  Juchau.  1981.
Further  Investigations  of  the capacity  of polynuclear aromatic  hydrocarbons
to elicit atherosclerotic  lesions.  J. Toxlcol.  Environ.  Health.  7:  327-335.
04460                                IX-8                            10/07/91

-------
Booth,  G.H.,  W.S.  Bradshaw and  M.W.  Carter.  1983.   Screening  of priority
chemicals  for  potential  reproductive hazard.   U.S.  Department  of Health,
Education,  and  Welfare,  National  Institute  for  Occupational   Safety  and
Health, Cincinnati, OH.   DHEW  contract no.  210-81-6012.
                                  *
Borneff,  J.  and  H.  Kunte.   1969.   Carcinogenic  substances  In  water  and
soil.   XXVI: A routine  method  for  the determination of PAH 1n water.  Arch.
Hyg. (Berl).  153:  220-229.

Bos, R.P., H.J.C.  PMnsen,  J.G.M.  van Rooy,  F.J. Jongeneelen, J.L.G. Theuws
and  P.Th.  Henderson.    1987.   Fluoranthene,  a volatile  mutagenlc  compound,
present In creosote and  coal tar.   Mutat.  Res.  187: 119-125.

Bos,  R.P.,  J.L.G.  Theuws,  F.J.  Jongeneelen  and  P.Th.  Henderson.   1988.
Mutagenldty of  b1-,  tr1-,  and  tetra-cycllc  aromatic  hydrocarbons  1n the
"taped-plate assay" and 1n the  conventional  Salmonella mutagenldty assay.
Mutat. Res.  204: 203-206.
      /
Bottomly,  A.C. and C.C. Twort.  1934.  The  carclnogenlcHy  of chrysene and
olelc acid.  Am.  J. Cancer.   21:  781-786.

Boyland, E.  and  H. Burrows.  1935.   The  experimental  production of sarcoma
1n  rats and mice  by  a  colloidal aqueous  solution of  l:2:5:6-d1benzanthra-
cene.  J. Pathol. Bacterlol.  41:231-238.

Boyland,  E.  and  P.  Sims.   1967.   The carcinogenic  activities  1n mice of
compounds related to benzo(a)anthracene.   Int.  J.  Cancer.   2:  500-504.


04460                                IX-9                            10/07/91

-------
Boyland.  E.,  E.R.  Busby,  C.E.  Dukes,  P.L.  Grover  and  D.  Hanson.   1964.
Further experiments on  Implantation  of materials  Into  the  urinary bladder  of
mice.  Br. J. Cancer.  18: 575-581.

Bregman,  R.   1954.   Moth ball  poisoning:  A case presentation.   CUn.  Proc.
Child. Hosp. DC.  10: 1-3.

Bridges,  B.A.,  D.B. McGregor  and E. Zelger.   1981.   Summary report on  the
performance  of  bacterial  mutation  assays,   in:  Evaluation  of  Short-Term
Tests  for Carcinogens.   Report  of the International  Collaborative  Program.
Progress  1n  Mutation Research,  Vol. 1,  F.J.  de Serres and  J.  Ashby,  Ed.
Elsevler, North Holland, NY.   p. 49-67.

Brown, A.K.  and H.  Burnett.    1957.   Studies on  the neonatal  development  of
the glucuronlde conjugating system.  AMA J. D1s.  Child.  94:  510-512.

Brune, K., H.  Kalln,  R. Schmidt  and E.  Hecker.  1978.  Inflammatory,  tumor
Initiating and  promoting activities  of polycycllc  aromatic  hydrocarbons  and
dlterpene  esters   In  mouse   skin  as  compared  with  their   prostaglandln
releasing potency in vitro.   Cancer Lett.   4: 333-342.

Brune, H., R.P. Deutsch-Wenzel, M. Habs, S.  Ivankovlc  and  D.  Schmahl.   1981.
Investigation  of  the   tumor1gen1c  response  to  benzo(a)pyrene  In  aqueous
caffeine  solution  applied orally to  Sprague-Dawley   rats.   J.   Cancer  Res.
Clln. Oncol.   102:  153-157.
04460                                IX-10                           10/07/91

-------
Bryan, W.R.  and M.B. Shlmkln.  1943.   Quantatlve  analysis  of dose-response
data  obtained  with   three  carcinogenic  hydrocarbons   In   strain  C^H  male
mice.  J. Natl. Cancer Inst.   3: 503-531.

BuckpHt, A.   1985.   Submission  of naphthalene criteria document  review to
Dynamac  Corporation  (letter  to Dr.  N. Hajjar).   Dec.  16.    (Cited  1n  U.S.
EPA, 1987a)

BuckpHt, A.R.  and  D.L.  Warren.   1983.    Evidence  for hepatic  formation,
export   and   covalent  binding  of  reactive   naphthalene  . metabolites   In
extrahepatlc tissues  In  vivo.   J. Pharmacol. Exp. Ther.  225: 8-16.

BuckpHt, A.R.,  L.S.   Bahnson  and  R.B.   Franklin.   1985.   Evidence  that
1-naphthol 1s  not  an obligate Intermediate 1n  the  covalent binding  and  the
pulmonary  bronchlolar  necrosis  by  naphthalene.   Blochem.  Blophys.  Res.
Commun.  126: 1097-1103.
      /
BuckpHt, A.R.,  L.S.  Bahnson  and  R.B.  Franklin.   1986.  Comparison  of  the
arachldonlc   add  and NADPH-dependent  mlcrosomal  metabolism  of  naphthalene
and .2-methylnaphthalene and  the effect  of Indomethacln on the  bronchlolar
necrosis.  Blochem. Pharmacol.  35: 645-650.

BuckpHt, A.R.,  N. Castagnoll, S.D.  Nelson,  A.D.  Jones  and L.S.  Bahnson.
1967.   StereoselectlvHy  of  naphthalene  epoxldatlon  by   mouse,  rat,  and
hamster  pulmonary,  hepatic,  and  renal  mlcrosomal  enzymes.   Drug  Hetab.
Dlspos.  15: 491-498.
04460                                IX-11                           10/07/91

-------
Buenlng,  M.K.,  W.   Levin,   J.M.  Karle  et  al.    1979a.   Tumor1gen1c1ty  of
bay-region  epoxldes  and  other  derivatives  of  chrysene and  phenanthrene  1n
newborn mice.  Cancer Res.   39: 5063-5068.

Buenlng,  M.K., W.  Levin,  A.W. Wood  et  al.   1979b.   Tumor1gen1city  of  the
dlhydrodlols  of  d1benzo(a,h)anthracene  on mouse  skin  and In  newborn mice.
Cancer Res.  39: 1310-1314.

Buenlng,  M.K., W. Levin,  A.W.  Wood et  al.   1980.  Tumor 1 genie activity  of
benzo(e)pyrene derivatives  on  mouse skin  and  1n newborn ml-ce.  Cancer  Res.
40: 203-206.

Bulay, O.K.  1970.  The  study  of  development of  lung  and skin tumors 1n mice
exposed In  utero to  polycycllc  hydrocarbons.   Acta.  Med.  Turc.    7:  3-38.
(Cited 1n IARC, 1983)
                                                                           •
Bulay, O.H.  and  L.W.  Wattenberg.   1971.   Carcinogenic  effects  of  polycycllc
hydrocarbon  carcinogen  administration  to  mice  during  pregnancy  on  the
progeny.  J. Natl. Cancer. Inst.  46:  397-402.  (CHed In IARC,  1983)

Busbee, 0., T. McLemore,  R.R.  Martin,  N.  Wray,  M.  Marshall and E.  Cantrell.
1980.  High  aryl  hydrocarbon  hydroxylase 1nduc1b1l1ty Is positively  corre-
lated  with  occurrence  of   lung  cancer.   In.:   Polynuclear  Aromatic  Hydro-
carbons: Chemistry and Biological  Effects, A. Bjorseth and A.J. Dennis,  Ed.
Battelle Press, Columbus, OH.  p.  917-934.
04460                                IX-12                           10/07/91

-------
Busby, W.F. Jr., M.E. Goldman, P.M. Newberne. and G.N. Wogan.   1984.  TumoM-
genlcHy of fluoranthene 1n a newborn mouse lung adenoma bloassay.  Cardno-
genesls.  5: 1311-1316.

Callahan,  M.,  M.  SUmak,  N.  Gabel  et  al.    1979.   Water-Related Environ-
mental  Fate  of  129  Priority  Pollutants,  Volume  II.   EPA  440/7-79-029b.
Prepared by U.S.  Dept. of Commerce.

Carver,  J.H., M.L.   Machado  and  J.A.  MacGregor.    1986.    Application  of
modified   Salmonella/mlcrosome   prescreen   to   petroleum-derived  complex
mixtures   and  polynuclear   aromatic   hydrocarbons   (PAH).    Hutat.   Res.
174:247-253.

Casto, B.C.   1979.  Polycycllc hydrocarbons and Syrian hamster embryo  cells:
Cell transformation, enhancement of viral transformation and analysis  of  DNA
damage.  In: Polynuclear Aromatic Hydrocarbons, P.M. Jones and P.  Leber,  Ed.
Ann Arbor Science Publ., Ann  Arbor,  HI.   p. 51-66.
     /
Casto,  B.C.,  N.  Janosko  and J.A.  DIPaolo.   1977.   Development  of a  focus
assay  model  for  transformation  of  hamster  cells  in  vitro  by chemical
carcinogens.  Cancer Res.  37: 3508-3515.

Caval1er1,  E. and E.  Rogan.   1983.   One-electron  oxidation  of aromatic
hydrocarbons  1n  chemical  and biological  systems.   In: Polynuclear Aromatic
Hydrocarbons:   Formation,   Metabolism  and  Measurement,  M.   Cooke  and A.J.
Dennis, Ed., 7th  Int.  Symp.  Battelle  Press,  Columbus,  OH.   p.  1-26.
04460                         .       IX-13                           10/07/91

-------
Cavallerl,  E.,   E.  Rogan,  C. Warner  and  A.  Bobst.   1985.   Synthesis  and
characterization  of   benzo(a)pyrene  and   6-methyl-benzo(a)pyrene   radical
cations.   In:  Polynuclear  Aromatic  Hydrocarbons:   Mechanisms,  Methods  and
Metabolism, M.  Cooke  and A.J. Dennis,  Ed., 8th Int.  Symp.   Battelle Press,
Columbus, OH.  p. 227-236.

Caweln,  M.J.  and K.L.  Sydnor.   1968.  Suppression  of cellular activity  In
the ret1culoendothel1al  system of  the rat by 7,l2-d1methylbenz(a)anthracene.
Cancer Res.  28: 320-327.

Chalmers, J.G.   1940.   Elimination  of  3:4-benzpyrene from the rat.   Blochem.
J.  34: 678-684.

Chalmers, J.G.  and  A.H.M.  K1rby.   1940.   The elimination  of 3:4-benzpyrene
from the animal  body  after  subcutaneous  Injection.   I. Unchanged benzpyrene.
Blochem.  34: 1191-1195.

Chang, L.H.  1943.  The  fecal excretion  of  polycycllc hydrocarbons  following
their administration to the rat.   J. B1ol. Chem.   151: 93-99.

Chang, M.J.W.  and R.W.  Hart.   1983.  The  effect of asbestos on the uptake
and  metabolism  of  benzo(a)pyrene  by normal  human  flbroblast.  In:  Poly-
nuclear  Aromatic Hydrocarbons:  Formation,  Metabolism and  Measurement,  M.
Cooke and A.J. Davis,  Ed., 7th Int. Symp.  Battelle  Press, Columbus,  OH.

Chang,  R.L.,  M.  Levin,  A.W. Wood et  al.   1981.    Tumor 1gen1 dty  of  the
dlastereomerlc bay-region benzo(e)pyrene  9,10-d1ol-l1,l2-epox1des 1n  newborn
mice.  Cancer Res.  41: 915-918.

04460                                IX-14                           10/07/91

-------
Chang,  R.L.  W. Levin,  A.W.  Wood  et  al.  1983.   Tumor1genU1ty  of enantl-
omers  of  chrysene   1,2-d1hydrod1ol  and  of  the  dlastereomerlc  bay-region
chrysene 1,2-d1ol-3,4-epox1des  on  mouse  skin  and In  newborn  mice.  Cancer
Res.  43: 192-196.

Chen, T.T.  and C.  Heldelberger.   1969.   Quantitative  studies on  the mal-
ignant  transformation  of  mouse prostate  cells  by carcinogenic hydrocarbons
1n vitro.  Int. J.  Cancer.   4:  166-178.   '

Chesls,   P.L.,  D.E.   Levin,   M.T.   Smith  et  al.   1984.   MutagenUHy  of
qulnones: Pathways of  metabolic  activation and detoxification.  Proc. Natl.
Acad. Sd.   81: 1696-1700.

Chlpman, J.K..  P.C.  Hlrom,  G.S.  Frost  and  P.  MUlburn.   1982.    Benzo(a)-
pyrene metabolism and  enterohepatlc circulation  In the rat.   In.: Biological
Reactive  Intermediates.   II.  Chemical   Mechanisms  and  Biological  Effects,
                                                                           •
Part A,  R.  Snyder  et  al.,  Ed.   Plenum  Press, NY.   p. 761-768.
      /
Chouroullnkov, I., A.  Gentll and  M.  Guerln.   1967.   Study of the carcino-
genic activity  of  9,lO-d1methylbenzanthracene and  of  3,4-benzopyrene given
orally.   Bull. Cancer.   54:  67-78.   (English translation)

Chu,  M.H.L.  and  C.U.  Chen.   1984.   Evaluation  and estimation of  potential
carcinogenic  risks   of  polynuclear   aromatic  hydrocarbons.   Presented  at
Symposium  on  Polynuclear  Aromatic  Hydrocarbons  1n   the  Workplace,  1984.
International Chemical  Congress  of  Pacific Basis  Societies.
04460                                IX-15                           11/21/91

-------
Chu,  E.W.  and  R.A.  Malmgren.   1965.  An  Inhibitory  effect of  vitamin  A on
the  Induction  of tumors of  forestomach  and cervix 1n  the  Syrian  hamster by
carcinogenic polycycllc hydrocarbons.  Cancer Res.  25: 884-895.

Chusld, E.  and  C.T.  Fried.  1985. Acute hemolytlc  anemia  due to naphthalene
                                   •
Ingestlon.  AHA Am. J. DIs. Child.  89: 612-614.

Clay,  P.  and M.F.  Cross.   1990.  Mlcrowell  mutation assays:  Evaluation of
ethylmethanesulphonate, benzo[a]pyrene  and benzldlne  using the tk  locus 1n
L5178Y mouse lymphoma cells.  Hutagenesls.   5(Suppl.): 45-54-.

Clement Associates.   1988.  Comparative Potency  Approach  for  Estimating the
Cancer  Risk Associated  with  Exposure  to Mixtures  of Polycycllc  Aromatic
Hydrocarbons.  Interim Final Report.   EPA Contract No. 68-02-4403.

Clement    Associates.     1990a.     Ingestlon    dose-response   model    for
benzo[a]pyrene.  EPA Contract No. -68-02-4601.
      /

Clement Associates.   1990b.   Development  of  Relative Potency  Estimates for
PAHs and Hydrocarbon  Combustion Product  Fractions Compared to Benzo[a]pyrene
and  Their   Use   1n  Carcinogenic  Risk  Assessment.    Prepared   for  the  U.S.
Environmental Protection Agency.

Cock, T.C.   1957.   Acute  hemolytlc  anemia 1n the  neonatal  period.   ANA Am.
J. D1s. Child.   94: 77-79.
04460                                IX-16                           10/07/91

-------
Conney,  A.M.   1982.   Induction  of mlcrosomal enzymes  by  foreign chemicals
and  cardnogenesls  by  polycyclk  aromatic  hydrocarbons:  G.H.A.  Clowes
Memorial Lecture.  Cancer Res.   42: 4875-4917.

Connor, T.H., J.C.  Thelss,  H.A  Hanna,  O.K.  Montelth and T.S. Matney.  1985.
Genotox1c1ty  of  organic chemicals frequently found  1n  the air  of mobile
homes.   Toxlcol.  Lett.   25:  33-40.

Coombs,  M.M.,  C.  D1xon  and  A.M.  Klssonerghls.    1976.   Evaluation  of  the
mutagenldty  of  compounds  of  known  cardnogenlcHy,  belonging  to  the
benz[a]anthracene,   chrysene,  and   cyclopenta[a]phenanthrene series,  using
Ames's  test.  Cancer Res.  36: 4525-4529.

Cott1n1, G.B.  and   G.B.  Mazzone.   1939.  The effects  of  3:4-benzpyrene on
human skin.  Am.  J.  Cancer.   37:  186-195.

Creasla, D.A.,  J.K. Poggenburg,, Jr. and  P.  Netteshelm.   1976.   Elutlon of
     /
benzo(a)pyrene from  carbon  particles  In the  respiratory  tract  of mice.   J.
Toxlcol. Environ. Health.  1:  967-975.

CrespV,  C.L.  and W.G. Thllly.   1984.   Assay  for  gene mutation  1n  a human
lymphoblast line,  AHH-1, competent for xenoblotlc  metabolism.   Mutat.  Res.
128: 221-230.

Oaffer, P.Z., K.S.  Crump and M.D. Masterman.  1980.   Asymptotic theory for
analyzing  dose-response  survival  data  with  application  to  the   low-dose
extrapolation problem.   Math.  B1osc1.   50: 207-229.

04460                                IX-17                            10/07/91

-------
Daniel, P.M., O.E. Pratt  and  M.M.L.  Prlchard. •  1967.   Metabolism of  labelled
carcinogenic hydrocarbons 1n rats.  Nature.  215: 1142-1146.

Dawson, J.P., W.W. Thayer and  J.F.  Desforges.   1958.   Acute  hemolytlc  anemia
In  the newborn  Infant  due  to naphthalene poisoning:  Report  of  two  cases,
with   Investigations  Into   the   mechanism of   the   disease.    Blood.    13:
1113-1125.

Dayhaw-Barker,  P.,  C.P.  Sambuco,  P.O. Forbes  and P.M.  Barker,  II.   1985.
Development of an animal model to study the phototoxlc effects  of anthracene
and  UV In  skin  and  occular  tissues.   13th  Annual Meeting  of  the  American
Society   for   Photoblology,   New   Orleans,   LA,  USA,  June   23-27,   1985.
Photochem. Photoblol.  41: 122s.

Dean,  B.J.   1981.   Activity  of  27 coded compounds   1n  the RL1  chromosome
assay.  In:  Evaluation  of Short-Term  Tests for Carcinogens.  Report of  the
International Collaborative Program.   Progress  1n  Mutation Research, Vol.  1,
P.O. de Serres and 0. Ashby,  £d.   Elsevler, North Holland,  NY.   p. 570-579.

Dean.,  J.H.,  M.I.  Luster, G.A.  Boorman,  L.D.  Lauer  and R.W. Leubke.   1983.
Selective  1mmunosuppress1on   resulting from  exposure to  the  carcinogenic
congener of benzopyrene 1n B6C3P1  mice.  Clln.  Exp. Immunol.   52: 199-206.

DePlora,  S.,  P. Zanacchl,  A. Camolrano,  C.  BennlcelH  and G.S.  Badolatl.
1984.  Genotoxlc activity and potency  of 135 compounds 1n  the  Ames reversion
test and 1n a bacterial  DNA-repa1r test.  Mutat. Res.   133(3):  161-198.
04460                                IX-18                 .          10/07/91

-------
Demi,  E.,  0.  Oesterle  and  F.J.  Welbel.   1983.   Benzo(a)pyrene  Inmates
enzyme-altered   Islands   1n  the  liver  of  adult   rats   following  single
pretreatment  and  promotion with  polychlorlnated  blphenyls.   Cancer  Lett.
19: 301-304.
                                   •
De  Salvla,  R.,  R.  Meschlnl,  M.  F1ore  et al.   1988.  Induction  of slster-
chromatld  exchanges  by  procarclnogens  In  metabollcally  competent  Chinese
hamster epithelial liver cells.   Mutat.  Res.   207:  69-75.

de Serres,  F.J.  and G.R.  Hoffman.  1981.  Summary report  oh  the performance
of  yeast  assays.   In:   Evaluation  of  Short-Term  Tests  for  Carcinogens.
Report  of  the  International  Collaborative  Program.   Progress  In  Mutation
Research, Vol. 1, F.J. de  Serres  and J.  Ashby,  Ed.   Elsevler, North Holland,
NY.  p. 68-76.

DesldeM, P.G.,  L.  Leprl,  D.  Helmler,  S. Glannessl and L.  Checch1n1.   1984.
Concentration, separation  and  determination  of hydrocarbons  1n  sea  water.
      /
J. Chromatogr.  284: 167-178.

Deutsch-Wenzel, R., H. Brune, G.  Grimmer, G.  Dettbarn and  J.  Mlsfeld.   1983.
Experimental  studies  1n rat  lungs  on  the carclnogenlclty  and  dose-response
relationships  of  eight   frequently   occurring  environmental   polycycllc
aromatic hydrocarbons.  J. Natl. Cancer. Inst.  71:  539-543.

DIPaolo,  J.A.. P.  Donovan  and R. Nelson.  1969.  Quantitative  studies  of |n.
vitro  transformation  by chemical  carcinogens.   J.  Natl.  Cancer  Inst.   42:
867-874.
04460                                IX-19                           11/21/91

-------
DIPaolo,  J.A.,  P.J.   Donovan  and  R.L.   Nelson.   1971.   Transformation  of
hamster  cells  In  vitro  by  polycycllc   hydrocarbons  without cytotoxldty.
Proc. Natl. Acad. Sc1. USA.   68:  2958-2961.

DIPaolo,  J.A.,  K.  Takano and  N.C.  Popescu.   1972.   QuantHatlon of chemi-
cally  Induced  neoplastlc   transformation of  BALB/3T3  cloned  cell  lines.
Cancer Res.  32: 2686-2695.

D1x,  T.A.  and  L.3.   Harnett.   1983.   Metabolism  of  polycycllc  aromatic
hydrocarbon  derivatives  to  ultimate  carcinogens  during I1p1d peroxldatlon.
Science.  221: 77-79.

D1x,  T.A.,  R.  Fontana,  A.  Panthan  and  L.3.   Marnett.    1985.   Hematin-
catalyzed    epoxldatlon    of   7,8-d1hydroxy-7,8-d1hydrobenzo(a)pyrene    by
polyunsaturated fatty  add  hydroperoxldes.  J.  B1o1.  Chem.   260:  5358-5365.

Dobrovolskala-Zavadsha'ia,  N.   1938.   Les doses   mlnlmes  de  1,2,5,6-dlbenz-
anthracene  capables  de produlre,  en  une seule  Injection  sous-cutanee,  un
cancer chez la souMs.  C.R. Soc.  B1ol. (Paris).   129:  1055-1057.  (Fre.)

Dock, L.,  6. Scheu,   B. Jernstrom,  M. Martinez,   U.  Torndal  and  L.  Erlkson.
1988.   Benzo(a)pyrene metabolism and Induction  of  enzyme-altered  fod  1n
regenerating rat liver. Chem.-B1ol.  Interact.  67: 243-253.

Dontenwlll,  W.  and   0.  Mohr.    1962.   Experlmentelle Untersuchungen  zum
problem  der  carclnomentstehung 1m  Resplratlpnstrakt.    I.  Die  Unterschled-
Hche Wlrkung des Benzpyrens auf  die  EpUhellen  der  Haut,  der Mundhohle und
der Trachea des Goldhamsters.   Z.  Krebsforsch.  65: 56-61.   (Ger.)

04460                                IX-20                          10/07/91

-------
 Dooley,   J.F.,   D.   Warshawsky,  G.  Yerganlan  and  R.T.  Christian.   1981.
 Cell-mediated  mutagenesls of  a  nontumorlgenlc epithelial cell  line (DPI-3)
 with polycycllc  aromatic  and heteroaromatlc  hydrocarbons.   In  Vitro.   17:
 201.

 Dorado,  G.  and  C. Pueyo.   1988.   L-arab1nose resistance test with Salmonella
 typhlmurlum as   a  primary tool for  carcinogen screening.  Cancer  Res.   48:
 907-912.

 Druckrey,   H.  and  D.  Schmahl.  1955.   Cancerogenlc  effect of  anthracene.
 Naturwlssenschaften.   42:  159-160.

 DuMouchel,  W.H.  and J.E. Harris.   1983.   Bayes  methods for  combining  the
 results  of  cancer studies 1n  humans and  other species.  J.  Am. Stat. Assoc.
 78:  293-315.  (CHed 1n Clement Associates, 1988)

-E1sen,  H.J.,  R.R. Hannah, C.  Legraverend,  A.B.  Okey and D.W. Nebert.  1983.
 The  Ah' receptor:  Controlling factor  1n  the Induction  of  drug metabolizing
 enzymes  by  certain  chemical  carcinogens  and other environmental pollutants.
 In:  Biochemical  Actions  of  Hormones,  Vol. X,  M.G. LHwack,  Ed.   Academic
 Press, NY.  p.  227-258.   (Cited 1n Legraverend et al.,  1983)

 Ekwall,  P., P.  Ermala, K. Setala and  L.  SJoblom.  1951.  Gastric absorption
 of 3,4-benzypyrene.   II.  The significance of the solvent for the penetration
 of 3,4-benzypyrene Into the  stomach wall.   Cancer Res.   11: 758-763.
 04460                                IX-21                           11/21/91

-------
 El-Bayoumy,  K.   1985.   Effects  of organoselenlum compounds on  Induction  of
 mouse  forestomach  tumors by benzo[a]pyrene.  Cancer Res.  45: 3631-3635.

 Epler,  J.L., F.W.  Larimer,  T.K.  Rao,  C.E.  Nix  and  T. Ho.   1978.   Energy-
 related  pollutants  In  the  environment:   Use  of  short-term tests  for mutagen-
 1c1ty  1n  the  Isolation and  Identification  of  blohazards.  Environ.  Health
 Perspect.  27: 11-20.

 Ermala,   P.,  K.   Setala  and   P.   Ekwall.    1951.    Gastric  absorption  of
 3,4-benzpyrene.   I.  The-effect  of physiological  processes  on  absorption.
 Cancer Res.  11: 753-757.

 Evans,  C.H.  and  J.A.  DIPaolo.   1975.   Neoplastlc  transformation  of  guinea
 pig  fetal cells  1n culture  Induced  by  chemical carcinogens.   Cancer  Res.
 35: 1035-1044.
                                                                            •
.Evans,  E.L.   and   A.D.  Mitchell.   1981.   Effect of  20  coded chemicals  on
 sister 'chromatld  exchange frequencies  1n  cultured  Chinese  hamster  cells.
 In:  Evaluation  of  Short-Term  Tests for  Carcinogens.   Report  of  the  Inter-
 national  Collaborative  Program.   Progress 1n  Mutation Research,  Vol.  1,  F.J.
 de Serres  and J. Ashby, Ed.  Elsevler North Holland, NY.  p. 538-550.

 Fabian,  G.  and  G.  Matoltsy.   1946.   Test  of  a cancerogenlc substance  1n
 respect   to   the   "non-disjunction"   frequency  of  the   X-chromosomes  In
 DrosophHa.  Nature  (London).  158: 911-912.
04460                                IX-22                           11/21/91

-------
Fahmy,  O.G.  and H.J. Fahmy.   1973.   Oxldatlve activation of benz(a)anthra-
cene  and  methylated derivatives  1n  mutagenesls  and carclnogenesls.   Cancer
Res.  33:  2354-2361.

Fait, D.W. and  D.J.  Nachrelner.   1985.  Napthalene acute  Inhalation  toxlclty
study  (unpublished  material).    Bushy  Run  Research  Center,   Export,   PA.
Submitted to Pharmakon Research International,  Inc. Waverly,  PA.

Falk, H.L., P.  Kotln and  S.  Thompson.  1964.  Inhibition of  carclnogenesls.
The  effects  of   polycycllc  hydrocarbons  and  related  compounds.   Arch.
Environ. Health.  9: 169-179.

Fanburg, S.J.   1940.  Exfol1at1ve dermatitis  due  to naphthalene: Report  of
an  eruption  resembling  mycosis  fungoldes.   AHA Arch.  Oermatol.  Syphllol.
42: 53-58.

Farrell, R.L.  and  6.W.  Davis.   1974.   Effect  of partlculate benzo(a)pyrene
carrier on carclnogenesls 1n the  respiratory tract of hamsters.   In:  Experi-
mental  Lung  Cancer: Carclnogenesls  and  Bloassays,  E. Karbe  and J.F. Park.
Int. Symp.   Sprlnger-Verlag,  New York, Berlin,  Heidelberg, Germany.

Fedorenko,   Z.P. and N.Y. Yanysheva.   1967.  Experimental  reproduction  of
tumors  of  the  antral  part  of  the  stomach   1n  mice by  administration  of
various doses of 3,4-benzpyrene.  Hyg. SanU.   32(5):  168-173.

Feron,  V.J. and A.  Kruysse.   1978.   Effects  of  exposure to  furfural  vapour
1n  hamsters  simultaneously  treated  with  benzo[a]pyrene  and dlethylnltros-
amlne.  Toxicology.  11: 127-144.

04460                                IX-23                            11/21/91

-------
Feron, V.J., D.  DeJong and  P.  Emmelot.   1973.   Dose-response correlation for
the  Induction  of  respiratory-tract  tumors  1n  Syrian  golden  hamsters  by
Intratracheal Instillations of benzo[a]pyrene.   Eur.  J.  Cancer.   9:  387-390.

Feron,  V.J.,  P.O.  Van  Den Heuvel,  H.B.W.M.  Koeter and  R.B.  Beems.   1980.
Significance  of  particle  size  of  benzo(a)pyrene for   the   Induction  of
respiratory tract tumors  1n hamsters.   Int.  J.  Cancer.   25:  301-307.

Field,  W.E.H.  and  F.J.C.  Roe.  1965.   Tumor  promotion  1n the  forestomach
epIthelHum [sic]  of  mice by  oral  administration  of citrus- oils.  J.  Natl.
Cancer Inst.  35(5): 771-787.

FUzhugh, O.G.  and  W.H.  Buschke.   1949.  Production of cataract In rats by
beta-tetralol and  other  derivatives of  naphthalene.   Arch. Opthalmol.  41:
572-582.

Florin,  I.,  L.  Rutberg,  N.  Curvall  and C.R.  Enzell. 1980.   Screening of
tobacco   smoke   constituents   for  mutagenlcUy   using   the   Ames'    test.
Toxicology.   18: 219-232.

Forbes,  P.O.,  R.E.  Davles  and  F.  Urbach.  1976.   Phototoxldty and  photo-
carclnogenesU.   Comparative effects of  anthracene and 8-methoxypsoralen 1n
the skin of  mice.  Food Cosmet. Toxlcol.   14: 303-306.

Ford, E. and C.  Muggins.   1963.   Selective  destruction 1n  testls Induced by
7,12-d1methylbenz(a)anthracene.  J.  Exp.  Med.   118:  27-40.
04460                                IX-24                           10/07/91

-------
Freeman, A.E.,  E.K Helsburger,  J.H.  Helsburger,  R.G. Wolford,  J.M.  Maryak
and R.J. Huebner.   1973.   Transformation of cell cultures  as  an Indication
of  the  carcinogenic  potential  of  chemicals.   J.  Natl.  Cancer  Inst.   51:
799-807.

Freudenthal, R.I.,  S.G. Hundley  and S.M. Cattaneo.   1978.   A  comparison of
the metabolites  of benzopyrene  by  lung mixed  function  oxldase  from  rat,
rhesus  and  humans.   In:   Cardnogenesls   Vol.   3:  Polynuclear  Aromatic
Hydrocarbons, P.M. Jones and R.I.  Freudenthal,  Ed.   Raven Press.

Galnes,   T.B.    1969.   Acute   toxlclty  of  pesticides.   Toxlcol.   Appl.
Pharmacol.   14:  515-534.

Garberg, P., E.  Akerblom and  G.  Bolcsfoldl.  1988.   Evaluation  of  a geno-
toxlclty  test   measuring   DNA-strand  breaks   1n  mouse  lymphoma  cells  by
alkaline unwinding and hydroxyapatUe elutlon.  Mutat.  Res.   203: 155-176.
                                                                           •
Garrett, N.E. and J.  Lewtas.   1983.   Cellular toxldty 1n Chinese hamster
ovary cell  cultures:  1.  Analysis  of  cytotoxldty endpolnts for twenty-nine
priority pollutants.  Environ.  Res.  32:  455-465.

Gart, J.J.,  D.   Krewskl,  P.N.,  Lee,  R.E. Tarone  and  G. Wahrendorf.   1986.
Statistical Methods 1n Cancer  Research.   Vol. III.  The  Design and Analysis
of  Long-Term Animal  Experiments.   IARC  Sc1.  Publ.  No.  79,  Lyon,  France.
(Cited 1n Clement Associates, 1988)
04460                                IX-25                           11/21/91

-------
Gatehouse,  0.   1980.   Mutagenlclty of  1,2 ring-fused acenaphthenes against
S.  typhlmurium  TA1537  and TA1538: Structure-activity  relationships.  Mutat.
Res.  78: 121-135.

Gaylor,  D.  and  R.  Kodell.  1980.   Linear extrapolation  algorithm for  low
dose risk assessment of  toxic  substances.   J. Environ. Pathol. Toxlcol.   4:
305-312.  (Cited 1n U.S.  EPA, 19915)

Geaclntov,  N.E., V.  Ibanez,  A.G.  Gagllano,  H.  Yoshlda  and  R.G. Harvey.
1980.  Kinetics   of  hydrolysis  to  tetraols  and  binding  of  benzo(a)pyrene
7,8-d1hydrod1ol-9,lO-ox1de and Its tetraol  derivatives to  ONA.  Conformation
of adducts.   Blochem.  Blophys.  Res.  Comro.   92: 1335-1342.

Geddle,  J.E.,   S.   Am1n,   K.  Hule  and  S.S.  Hecht.   1987.   Formation  and
tumor1gen1city   of  benzo(b)fluoranthene   metabolites  In   mouse   epidermis.
Carclnogenesls.   8:  1579-1584.

Gelbolri,  H.V.    1980.   Benzo(a)pyrene  metabolism,  activation,  and  cardno-
genesls: Role and regulation of mixed-function oxldases and  related enzymes.
Physlolog. Rev.   60: 1107-1166.

Gerarde, H.U.   1960.   Toxicology  and biochemistry of  aromatic  hydrocarbons.
In.:  Elsevler  Monographs  on  Toxic   Agents,   E.  Browning,   Ed.    Elsevler,
Amsterdam,  p.  240-321.

Ghettl,  G.  and   L.  Marlanl.   1956.   Eyes  changes  due to  naphthalene.   Med.
Lav.  47: 524.
04460                                IX-26                           11/21/91

-------
G1bel,  W.   1964.   Experlmenteller BeHrag  zur  Synkarzlnogenese belm Spels-
erohrenkarzlnom.  Klebsarzt.   19:  268.   (Cited  In  IARC,  1973)

Gibson,  T.L.  and  L.L.  Smith.   1979.   Radiation-Induced oxidation of benzo-
(a)pyrene.  J. Org.  Chem.  44: 1842-1846.
                                     *
Gibson,  T.L.,  V.B.  Smart and L.L. Smith.   1978.   Non-enzym1c activation of
polycycllc aromatic  hydrocarbons as mutagens.   Mutat.  Res.   49:  153-161.

Gldron,  E.  and  J.   Leurer.    1956.    Naphthalene  poslonln'g.   Lancet.   1:
228-231.

Glatt,  H.,  A.  Seldel,  W.  BochnHschek et  al.   1986.   Mutagenlc  and cell-
transforming  activities of   tr1ol-epox1des  as compared  to  other chrysene
metabolites.  Cancer Res.  46: 4556-4565.

Godekv  E.G., R.W. Nalsmith and  R.J. Matthews.  1985.  Ames  Salmonella/micro-
some  plate  test   (EPA/OECD)  (unpublished  material).   Pharmakon Research
International   Inc.,  Waverly,  PA.    Submitted   to  the  Office  of  Toxic
Substances, U.S. EPA, Washington,  DC.   Document No.  86-870000559.

Goldschmldt, B.H., C.  Katz and B.L.  Van  DQQren.   1973.  The cocardnogenlc
activity of non-carcinogenic  aromatic hydrocarbons.  Proc. Am. Assoc. Cancer
Res.  17: 84.  (Abstract)  (Cited  1n  IARC,  1983)
04460                                IX-27                            11/21/91

-------
Grant, G.A.,   R.L.  Carter,  F.J.C.  Roe and H.C. P1ke.   1968.   Effects of  the
neonatal  Injection  of a  carcinogen  on the Induction  of  tumors by the  sub-
sequent application to the  skin of the same carcinogen.   Br.  J. Cancer.   22:
346-358.

Greenstock, C.L. and  R.H. W1ebe.   1978.   Photosensitized  carcinogen degrada-
tion  and   the  possible  role of  singlet  oxygen   1n  carcinogen  activation.
Photochem. Photoblol.   28: 863-867.

Griffin,  G.D.,  E.F.  Calle,  H.D. Morris,  K.Y.  Long,  S.S.  Schuffman and  W.M.
Mitchell.    1986.    Inhibition  of  murlne  Interferon  production  following _1ji
vivo administration of benzo{a)pyrene. J.  Interferon  Res.   6:  115-121.

Grlgor, W.G.,  H. Robin  and J.D.  Harley.  1966.   An  Australian variant on
"full-moon disease."  Med. 0. Aust.  2: 1229-1230.

Gross, R.T.,  R.E.  Hurwltz  and  P.A.  Marks.  1958.   An hereditary enzymatic
defect   1n   erythrocyte   metabolism:   Glucose-6-phosphate    dehydrogenase
deficiency.  J. Cl1n.  Invest.  37:  1176-1184.

Grover, P.L.,  P. S1ms, B.C.V.  MHchley and F.J.C. Roe.   1975.   The cardno-
genldty of polycycllc hydrocarbon epoxldes. 1n newborn mice.   Br. J. Cancer.
31: 182-188.

Gustavsson, P.,  A.  Gustavsson  and C.  Hogstedt.   1988.   Excess of cancer 1n
Swedish chimney sweeps.   Br. 3.  Ind.  Med.   45:  777-781.
04460                                IX-28                            10/07/91

-------
Habs, M.,  D.  Schmahl and  J.  Mlsfeld.   1980.   Local cardnogenlcHy of  some
environmentally  relevant  polycycllc  aromatic  hydrocarbons  after  lifelong
topical application to mouse skin.   Arch.  Geschwulstforsch.   50:  266-274.

Haddow,  A.,  C.M.  Scott and  J.D.  Scott.   1937.   The  Influence of  certain
                                  •
carcinogenic and  other  hydrocarbons  on body  growth 1n  the  rat.  Proc.  R.
Soc. London B.   122: 477-507.

Haggerty,  R.J.    1956.   Naphthalene  poisoning.    New   Engl.  J.  Med.    255:
919-920.

Hansch,  C.  and   A.J.  Leo.   1985.   Hedchem  Project.   Issue  #26.    Pomona
College, Claremont, CA.

Hansen,  E.S.   1983.   Mortality  from cancer  and   1schem1c  heart disease  1n
Danish  chimney   sweeps:  A  five-year  follow-up.   Am.  J.  Ep1dem1ol.    117:
160-164.
      /

Hansen,  E.S.   1989.   Mortality  of  auto  mechanics.   A ten-year  follow-up.
Scand. J. Work  Environ. Health.   15:  43-46.

Hanssler,  H.  .  1964.    Lebensbedrohllche  Naphthal1nverg1ftung  be1   elnem
Saugllng durch Vapor1n-Dampfe [Life-threatening  naphthalene  Intoxication  of
an Infant through vapor 1n fumes].   Dtsch.  Med. Uochenschr.   89:  1794-1797.

Hardln,  B.D.,  G.P.  Bond.,  M.R.  S1kov,  F.D.  Andrew,  R.P.  Bellies and  R.W.
Nlemeler.  1981.   Testing of  selected  workplace  chemicals for  teratogenlc
potential.   Scand. J. Work Environ. Health.  7(suppl. 4):  66-75.

04460                                IX-29                            10/07/91

-------
Hartman, P.E.   1983.  Mutagens:  Some  possible  health  Impacts  beyond  cardno-
genesls.  Environ. Mutagen.  5: 139-152.

Hartwell,  J.L.    1951.   Survey  of  compounds   which  have  been  tested  for
carcinogenic  activity.   Publ.  Health  Ser.  Publ.  No.   149,  GPO,  2nd  ed.,
Washington, DC.

Hazelton  Laboratories  America,  Inc.    1989a.   Dose  range-finding  study  1n
mice  with  acenaphthylene.   HLA Study  No.  2399-128.   Sponsored by  Dynamac
Corporation,  Rockvllle,  *MD  for,  the  Offlc'e  of  Solid  Waste  and  Emergency
Response, U.S. EPA, Washington, DC.

Hazelton  Laboratories  America,  Inc.    1989b.   Dose  range-finding  study  1n
mice  with  anthracene.    HLA  Study  No.   2399-130.   Sponsored  by   Dynamac
Corporation,  Rockvllle,  HD  for  the  Office  of  Solid  Waste  and  Emergency
Response, U.S. EPA, Washington, DC.

Hazelton Laboratories  America,  Inc.   1989c.    Subchronlc  toxlclty  study  1n
mice  with  acenaphthylene.   HLA  Study  No.  2399-129,  sponsored by  Dynamac
Corporation,  Rockvllle,  MD,  for  the  Office   of  Solid   Waste and  Emergency
Response, U.S. EPA, Washington, DC.

Hazelton  Laboratories  America,  Inc.    1989d.   Subchronlc  ToxIcHy  In  Mice
with Anthracene.   HLA Study No. 2399-131, sponsored  by  Dynamac  Corporation,
Rockvllle,  MD  for  th£  Office  of  Solid  Waste  and   Emergency  Response,
Washington, DC.
04460                                IX-30                           11/21/91

-------
Hecht,  S.S.,  W.E.  Bondlnell and  D.  Hoffmann.   1974.   Chyrsene and  methyl-
chrysenes:  Presence  on tobacco smoke  and  cardnogenldty.   J. Natl.  Cancer
Inst.  53: 1121-1133.

Hecht,  S.S.,  E.J.  LaVole,  S.  Amln,  V. Bedenko  and  D.  Hoffmann.  1980.   On
the  metabolic   activation   of  the   benzofluoranthenes.    In:   Polynuclear
Aromatic  Hydrocarbons:  Chemistry  and Biological  Effects,   A.  BJorseth  and
A.J. Dennis, Ed.  Battelle Press,  Columbus,  OH.   p.  417-433.

Heldelberger, C. and S.H.  Weiss.  1951.   The distribution .of  radioactivity
In  mice  following  administration of  3,4-benzpyrene-5-C14  and  l,2,5,6-d1-
benzanthracene-9,10-C14.  Cancer Res.   11:  885-898.

Heller,  W.   1950.    Experimental  study  on tumours  produced  by  light.   2.
Tumours  produced  by  light  through photosens1b1!1sat1on.   Strahlentheraple.
81: 529-548.  (Ger.)  (Cited 1n IARC,  1983)

     /
Hellman,  B., R.  D'Argy  and  S.   Ullberg.   1984.   The in  vivo  uptake  of
trltlated  thymldlne  as  a  potential   short-term  test  of  toxic  effects  of
polycycllc  aromatic  hydrocarbons  In  different  organs.   Toxicology.   29(3):
183-194.

Hemmlnkl,  K.»  K.  Randerath,  M.V.  Reddy  et  al.   1990.   Postlabellng  and
Immunoassay -analysis  of  polycycllc   aromatic  hydrocarbons  —  Adducts  of
deoxyrlbonuclelc acid 1n  white blood  cells of  foundry workers.   Scand.  J.
Work Environ. Health.  16:158-162.
04460                                IX-31                            10/07/91

-------
HendMch,  S.  and H.C.  PHot.   1987.   Enzymes of  glutathlone  metabolism as
biochemical markers during  hepatocardnogenesls.  Cancer Metastasis Rev.  6:
155-178.

Henry,  M.C.  and  D.G.   Kaufman.  1973.   Clearance  of  benzo(a)pyrene   from
hamster  lungs  after  administration  on  coated  particles.    J.  Natl.  Cancer
Inst.  51: 1961-1964.

Henry,  M.C.,   C.D.  Port  and  D.G.  Kaufman.   1974.    Role  of  particles 1n
respiratory  cardnogenesls   bloassay.    In:    Experimental   Lung  Cancer:
Carclnogenesls  and  Bloassays,   E.  Karbe  and  O.F.  Park,  Ed.   Int.  Symp.
Sprlnger-Verlag, New York, Berlin,  Heidelberg, Germany.

Henry,  M.C.,   C.D.  Port  and D.G.  Kaufman.   1975.  Importance  of physical
properties  of  benzo(a)pyrene - ferric oxide mixtures  1n  lung tumor  Induc-
tion.  Cancer  Res.  35:  207-217.

Hera, C. and C.  Pueyo.   1988.   Response  of the L-arab1nose  forward mutation
assay of  Salmonella  typh1mur1um  to  frameshlft-type mutagens.   Mutat.   Res.
203: 39-45.

Herbert, R., N.  Marcus,  M.S.  Wolff  et  al.  1990.   Detection  of adducts of
deoxyrlbonuclelc add  1n white blood cells  of  roofers by 3aP-postlabel1ng.
Scand. J.  Work  Environ.  Health.   16: 135-143.

Hermann, M.   1981.    Synerg1st1c effects  of  Individual  polycycllc aromatic
hydrocarbons  on  the  mutagenldty  of  their  mixtures.    Mutat.   Res.   90:
399-409.
04460                               IX-32                           10/07/91

-------
Hermann, M.,  J.P.  Durand,  J.M.  Charpentler  et al.   1980.   Correlations of
mutagenlc activity with  polynuclear  aromatic hydrocarbon content of  various
mineral  oils.    In.:  Polynuclear   Aromatic   Hydrocarbons:   Chemistry  and
Biological   Effects,   4th  Int.   Symp.,  A.  Bjorseth  and  A.J.   Dennis,  Id.
Battelle Press, Columbus, OH.   p. 899-916.

Heston,  W.E.  and  M.A.   Schnelderman.   1953.   Analysis  of  dose-response  1n
relation to mechanisms of pulmonary  tumor Induction  1n mice.  Science.   117:
109-111.

Heuper, W.C.   1963.   Chemically  Induced skin  cancers  1n man.  Natl. Cancer
Inst. Nonogr.   Vol. 10.

Ho, C-H., B.R.  Clark, M.R.  Guerln,  B.D. Barkenhus, T.K. Boa  and  J.L. Epler.
1981.   Analytical   and   biological   analyses   of   test   materials  from  the
synthetic  fuel   technologies.    Studies   of  chemical   structure-mutagenlc
activity relationships of  aromatic  nitrogen compounds relevant  to  synfuels.
Mutat.,Res.   85: 335-345.

Hoch-L1get1,   C.   1941.   Studies  on  the  changes  In  the lymphold tissue  of
mice  treated   with  carcinogenic  and noncardnogenlc  hydrocarbons.  Cancer
Res.  1: 484-488.=
Hoffman,-U\£4and  N.L. Gay.  1981.   Embryotoxlc effects of  benzo(a)pyrene,
          "'•'•-"'. *T -
chrysene   and   7,l2-d1methylbenz[a]anthracene   In   petroleum   hydrocarbon
mixtures In mallard ducks.   3. Toxlcol Environ.  Health.   7:  775-787.
04460                                IX-33                           11/21/91

-------
Hoffmann,  D.  and E.L. Wynder.   1966.   Beltrag zur cardnogenen Wlrkung  von
Dlbenzopyrenen.  Z. Krebsforsch.  68: 137-149.  (Ger.)

Hoffmann,  D., G. Rathkamp,  S.  Nesnow and  E.L.  Wynder.   1972.   Fluoranthenes:
Quantitative  determination  1n cigarette  smoke,  formation by  pyrolysls,  and
tumor-Initiating activity.  J. Natl. Cancer Inst.   49: 1165-1175.

Hollwlch,  F., A. Boateng  and  B.  Kolck.  1975.   Toxic  cataract.  In:  Cataract
and  Abnormalities  of  the Lens,  J.G.  Bellow, Ed.  Grune and Stratton,  New
York, NY.  p. 230-243.                                      .

Horton,  A.W.  and  G.H.  Christian.   1974.   Cocarclnogenlc versus  Incomplete
carcinogenic   activity   among   aromatic    hydrocarbons:   Contrast   between
chrysenes and benzo[b]tr1phenylene.   J.  Natl.  Cancer Inst.  53: 1017-1020.

Howe,  R.B.  and K.S.  Crump.   1982.   Global  82.   A  computer  program  to
extrapolate quantal animal  toxldty  data  to low doses.  Office of  Carcinogen
      /•
Standards, Occupational Safety and  Helath Administration,  U.S. Department  of
Labor, Contract 41USC252C3.

Howson,  C.P.,  T.  Hlyama  and  E.L.   Wynder.   1986.   The  decline  In  gastric
cancer: Epidemiology of an unplanned triumph.   Ep1dem1ol.  Rev.  8:  1-27.

Hoy, C.A.,  E.P.  Salazar  and  L.H. Thompson.   1984.   Rapid detection  of  DNA-
damaglng agents using repair-deficient  CHO cells.   Mutat.  Res.  130:  321-332.
04460                                IX-34                           10/07/91

-------
Huang,  M-T.,  A.W.  Wood,  R.L.  Chang  et  al.   1986.   Inhibitory  effect of
3-hydroxybenzo(a)pyrene  on  the  mutagenldty  and  tumoMgenldty  of  (±)-76,
8o-d1hydroxy-9a,10a-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene.            Cancer
Res.  46: 558-566.

Huberman, E.   1978.  Cell transformation and mutability of different  genetic
loci  In  mammalian cells by  metabollcally  activated carcinogenic  polycycllc
hydrocarbons.  In.:  Polycycllc Hydrocarbons and Cancer,  Vol.  2.    Molecular
and  Cell  Biology, H.V.  Gelboln  and  P.O.P.  Ts'o, Ed.   Academic  Press, NY.
p. 161-174.

Huberman  E.  and  L. Sachs.   1976.   Mutability  of  different  genetic loci 1n
mammalian cells  by metabollcally  activated carcinogenic polycycllc  hydro-
carbons.  Proc. Natl.  Acad.  Scl.  USA.   73:  188-192.

Hugglns,  C.  and  N.C.  Yang.   1962.   Induction  and  extinction  of  mammary
cancer.  Science.   137: 257-262.
      /'

Hughes, M.F., M.  ChamulHrat, R.P. Mason and T.E.  Ellng.  1989.   Epoxldatlon
of   7,8-d1hydroxy-7,8-d1hydrobenzo(a)pyrene  via   a  hydroperoxlde-dependent
mechanism catalyzed by Upoxygenases.   Cardnogenesls.   10: 2075-2080.
          *&.-•'•>.''.    '"
          ••t'jfc^":-•.•
IARC  (International  Agency for  Research on Cancer).   1973.   Certain  Poly-
          -
cycllc Aromatic Hydrocarbons  and Heterocycllc  Compounds.  Monographs  on the
Evaluation  of  Carcinogenic  Risk  of  the  Chemical to  Man,  Vol.  3.   Lyon»
France.
04460                                IX-35                           11/21/91

-------
IARC  (International  Agency  for  Research on  Cancer).   1983.  Monographs  on
the  Evaluation of  the Carcinogenic  Risk  of  Chemicals  for  Humans.   Poly-
nuclear  Aromatic  Compounds.   Part  1.  Chemical,  Environmental  and  Experi-
mental Data, Vol.  32.  World Health Organization.

IARC  (International  Agency  for  Research on  Cancer).   1984.  Monographs  on
the  Evaluation of  the Carcinogenic  Risk  of  Chemicals  for  Humans.   Poly-
nuclear  Aromatic   Compounds.   Part  3.  Industrial   Exposures   1n  Aluminum
Production, Coal Gasification, Coke  Production,  and  Iron  and  Steel Founding,
Vol. 34.  World Health Organization.

IARC  (International  Agency  for  Research on  Cancer).   1985.  Monographs  on
the  Evaluation of  the Carcinogenic  Risk  of  Chemicals  for  Humans.   Poly-
nuclear  Aromatic  Compounds.   Part  4.  Bitumens,   Coal-Tars   and  Derived
Products, Shale-oils and Soots, Vol. 35.  World Health Organization.

Ichlnotsubo, D.,  H.F.  Mower,  3.  SetUff and M.  Mandel.   1977.   The use  of
rec_-bacter1a for testing of carcinogenic substances.   Mutat.  Res.  46: 53-61.

Ide,  F.,  T.   Ishlkawa   and  S.   Takayama.  1981.  Detection   of  chemical
carcinogens by assay of unscheduled  DNA  synthesis  1n  rat  tracheal  epithelium
In short-term organ  culture.  J.  Cancer  Res. CUn. Oncol.   102: 115-126.

Ikemoto,  F.  and  S.   Iwata.   1978.   Sulfhydryl  contents  of   soluble and
Insoluble  lens  proteins  1n  naphthalene  and traumatic cataracts  In  rabbits!
Opthalmlc Res.   10:  194-201.
04460                                IX-36                           11/21/91

-------
Inomata,  M.   and   C.   Nagata.    1972.    Photolnduced  phenoxy  radical  of
3,4-benzopyrene.  GANN.   63:  119-130.

Iqbal,  Z.M.,  A. Yoshlda and  S.S.  Epstein.  1979.   Uptake  and excretion of
benzo[a]pyrene and Us metabolites by the rat pancreas.  Drug Hetab. Dlspos.
7: 44-48.

Irle, D., T.  Sasaki  and  R.  Ho.  1973.   Acute toxUHy, Inhalation  toxldty
and  skin  Irritation  of  cyclododecane (CD), trlcyclododecane (TCD),  naphtha-
line  (NP) and  para-dlchlorobenzene  (parazol)  (PZ).  Toho  Igakkal  Zasshl,
20: 772-775.

Irle, U.   1964.  Akute hamolytlsche Anamle  durch Naphtha11n-Inhalat1on be1
zwel  Fruhgeborenen und elnem  Neugeborenen  [Acute hemolytlc  anemia caused by
naphthalene  Inhalation  In two  premature  babies and  one  neonate].   Dtsch.
Med. Wochenschr.  89: 1798-1800.

Irv1n,'T.R. and  J.E.  Martin.   1987.  In. vitro and !n  vivo embryotoxlclty of
fluoranthene, a  major prenatal  toxic component  of dlesel soot.  Teratology.
35: 65A.

Ito,  Y.,  N.  Ueda,  S. Maeda et  al.   1988.   Induction of chromosomal aberra-
tions 1n  rat bone marrow  cells and mutations  In Salmonella typhlmurlum by
benz[a]anthracene derivatives.  Mutat. Res.   206: 55-63.

Jacobzlner, H.  and  H.W.  Raybln.  1964.  Naphthalene poisoning.  N.Y.  State
J. Med.   64:  1762-1763.
04460                                IX-37                            11/21/91

-------
Jeftlc,  L.  and  R.N.  Adams.   1970.   Electrochemical oxidation  pathways  of
benzo(a)pyrene.  J. Am.  Chem.  Soc.   92:  1332-1337.

Jerlna,  D.M.   and  R.E.  Lehr.   1977.   The  bay-region   theory:  A  quantum
mechanical  approach  to  aromatic  hydrocarbon-Induced cardnogenldty.  ln_:
Mlcrosomes and Drug Oxidations, Proc. Third International Symposium, Berlin,
July,  1976, V. Ullrich,  I.  Roots,  A. HUdebrandt,  R.W.  Estabrook  and A.H.
Conney, Ed.   Pergamon  Press  Inc.,  Oxford,  p. 709-720.   (Cited  1n  Smith et
al., 1978)

Jerlna, O.M.,  H.   Yag1,  R.E.  Lehr  et al.   1978.   The bay-region  theory of
cardnogenesls  by  polycycllc  aromatic   hydrocarbons.    In.:   Polycycllc
Hydrocarbons and Cancer,  Vol. 1, Environment, Chemistry and Metabolism, H.V.
Gelboln and P.O.P. Ts'o,  Ed.   Academic Press,  NY.  p.  173-188.

Jones, C.A., H.F.  Callaham and E. Huberman.  1984.  Enhancement of chemlcal-
carc1nogen-1nduced cell  transformation  1n  hamster  embryo cells  by (alpha)
25-dehydroxycholecaldferol,  the biologically  active metabolite  of vitamin
D3.  Cardnogenesls.   5:  1155-1159.

Jotz, M.M. and A.D.  Mitchell.  1981.  Effects of 20 coded  chemicals on the
forward mutation  frequency at  the  thymldlne klnase  locus  In  L5178Y  mouse
lymphoma cells,  in: Evaluation of Short-Term Tests for Carcinogens.  Report
of the  International  Collaborative  Program.  Progress 1n Mutation Research,
Vol. 1, F.J. de Serres and J. Ashby, Ed.   Elsevler,  North Holland, New York.
p. 580-593.
04460                                IX-38                           11/21/91

-------
Kaden,  D.A.,  R.A.  HUes  and  W.G.  Thllly.   1979.   MutagenlcHy of soot  and
associated  polycycllc  aromatic  hydrocarbons   to   Salmonella   typhlmurlum.
Cancer Res.  39: 4152-4159.

Kakunaga,  T.   1973.   A  quantitative system  for assay  of  malignant  trans-
formation by chemical carcinogens using a clone  derived  from  BALB/3T3.   Int.
J. Cancer.  12: 463-473.

Katz, M.,  C.  Chan, H.  Toslne et al.  1979.   Relative  rates of  photochemical
and  biological  oxidation  (Ijn  vitro) of  polynuclear aromatic  hydrocarbons.
In.:  Polynuclear  Aromatic  Hydrocarbons,  P.M.  Jones  and  P. Leber,  Ed.   Ann
Arbor Science PubUcatons, Ann Arbor, MI.   p.  171-189.

Kellermann, 6., H. Luyten-Kellerman  and C.R.  Shaw.   1973.  Genetic  variation
of aryl  hydrocarbon  hydroxylase In  human lymphocytes.  Am.  J. Hum.  Genet.
25: 327-331.
      /
Kellermeyer, R.W., A.R.  Tarlov, G.J.  Brewer, P.E.  Carson  and  A.S.  Alvlng.
1962.  Hemolytlc effect of  therapeutic drugs: Clinical considerations  of  the
pr1maqu1ne-type hemolysls.  J. Am.  Ned.  Assoc.  180:  388-394.

Kennaway, E.L.  1924a.  On  cancer-producing tars and tar fractions.   J.  Ind.
Hyg.   5: 462-488.

Kennaway, E.L.  1924b.   On the cancer-producing factor 1n tar.  Br.  Ned.  J.
1: 564-567.
04460                                IX-39                           11/21/91

-------
Kennaway,  E.L.   1930.   Further experiments -on  cancer producing substances.
Blochem. J.  24: 497-504.

Ketkar, M.,  G.  Resnlk, P. Schneider and  U.  Mohr.  1978.  Investigations on
the carcinogenic burden by air pollution  1n man.  Intratracheal Instillation
studies  with benzo[a]pyrene  1n   bovine  serum  albumin   1n  Syrian  hamsters.
Cancer Lett.  4: 235-239.

K1nae,  N.,  T.  Hashlzume,  T.  MakHa,  I.  Tomlta,  I. Klmura  and H.  KanamorV.
1981.   Studies   on  the  toxldty  of  pulp  and  paper mill' effluents  - 1.
Mutagen1c1ty of  the sediment  samples derived  from Kraft  paper mills.  Water
Res. 15: 17-24.

KUIgawa,  T., T. Hlrakawa,  T. Ishlkawa,  N.  Nemoto and  S.  Takayama.  1980.
Induction of hepatocellular  carcinoma 1n rat liver by Initial  treatment  with
benzo(a)pyrene  after   partial  hepatectomy and  promotion by  phenobarbltol.
Toxlcol. Lett.   6:  167-171.

K1zer, D.E., J.A. Clouse, D.P. Ringer  et  al.  1985.  Assessment of rat liver
mlcrosomal epoxlde hydrolase  as  a marker  of hepatocarclnogenesls.   Blochem.
Pharmacol.  34:  1795-1800.

Klar,  E.   1938.   Ubef die  Entstehung  elnes  EpUhelloms  belm Menschen  nach
experlmentellen   ArbeUen  m1t  Benzpyren.   Klin.  Wochschr.    17:  1279-1280.
(Ger.)
04460                                IX-40                           11/21/91

-------
Klein, M.   1952.   Effect of croton  oil  on  Induction of tumors by  1,2-benz-
anthracene,  desoxychollc add,  or  low  doses  of  20-methylcholanthrene  1n
mice.  J. Natl. Cancer Inst.  13:  333-341.

Klein, M.   1960.   A  comparison  of the  Initiating  and promoting actions  of
9,!0-d1methyl-l,2-benzanthracene   and   1,2,5,6-dlbenzanthracene    1n    skin
tumorlgenesls.  Cancer Res.   20:  1179-1183.

Klein, M.   1963.   Susceptibility  of  strain  B6AF/J  hybrid Infant  mice  to
tumorlgenesls  with  1,2-benzanthracene.,  deoxychollc  add  and  3-methylchol-
anthrene.  Cancer Res.  23:  1701-1707.

Knake, E.   1956.  Ueber  schwache Geschwulsterzeugende  Wlrkung  von Naphthalln
und  Benzol  [Weak  tumor  producing  effect   of   naphthalene   and   benzene].
Vlrchows  Arch.  Pathol.  Anat. Physlol.   329:  141-176.   (CUed  1n  U.S.  EPA,
1987a)

Knaap,' A.G.A.C., C.  Goze and J.W.I.M.  Simons.  1981.  Mutagenlc activity  of
seven coded samples  of  V79  Chinese hamster  cells.   In.:  Evaluation  of  Short-
term  Tests  for  Carcinogens.   Report  of  the  International   Collaborative
Program.    Progress   1n  Mutation Research,  Vol.   1,  F.J.  de   Serres  and  J.
Ashby, Ed.   Amsterdam, Elsevler,  North Holland,   p.  608-613.

Knobloch,  K.,  S.  Szendz1kowsk1   and  A.   Slusarczyk-Zalobna.   1969.   The
Investigations  of  acute and  subacute  toxic  action  of  acenaphthene and
acenaphthylene.  Hed. Pracy.  20:  210-222.   (English abstract)
04460                         .       IX-41                            10/07/91

-------
Kobayashl, N.  1975.  Production  of  respiratory  tract  tumors In hamsters by
benzo(a)pyrene.  Gann.   66:  311-315.

Koch, H.R., K. Doldl and  0.  Hockwln.  1976.  Naphthalene cataracts  1n rats.
Association of eye  pigmentation  and cataract development.   Doc. Ophthalmol.
48: 293-303.   (Cited In  U.S.  EPA,  1987a)

Kotln,  P.,  H.L.  Falk and R.  Busser.   T959.  Distribution, retention,  and
elimination  of C14-3,4-benzpyrene  after  administration  to mice  and rats.
J. Natl. Cancer Inst.  23: 541-555.

Kourl, R.E., G.H. Connolly,  D.W.  Nebert  and R.A. Lubet.  1983.  Association
between  susceptibility  to  d1benzanthracene-1nduced  flbrosarcoma   formation
and the Ah locus.   Int.  3. Cancer.  32: 765-768.

Krahn,  D.F.  and  C. Heldelberger.   1977.    Liver  homogenate-medlated muta-
                                                                           •
genesis 1n Chinese hamster V79 cells by polycycllc aromatic  hydrocarbons and
aflatoxlns.  Mutat.  Res.  46: 27-44.

Krewskl,  D.  and  D.J.   Murdoch.   1990.   Scientific  Issues  In  Quantitative
Cancer Risk  Assessments,  S.H.  Moolgavkar  and D.L. Thompson, Ed.   BUkhauser
Boston Inc. New York.   (In press)

Krewskl,  D.,  D.  Murdoch  and A.  Dewanjl.   1986.  Statistical  modeling and
extrapolation  of  cardnogenesls   data.  In:   Modern Statistical  Methods In
Chronic  Disease  Epidemiology,  S. Moolgavkar  and  R.  Prentice,  Ed. W1ley-
Intersdence, New York,  NY.   p.  259-282.   (Cited  1n U.S.  EPA. 1991b)


04460                                IX-42                            11/21/91

-------
Krewskl, D., J. S1em1atyck1, L. Nadon, R. Oewar and M.  Gerln.  1989.  Cancer
risks  due  to  occupational  exposure to  polycycllc  aromatic  hydrocarbons; a
preliminary  report.    In:   Genetic   Toxicology  of  Complex  Mixtures,  M.D.
Waters, S. Nesnow, 3. Lewtas, M.M. Moore and F.B. Daniel,  Ed.  Plenum Press,
NY.                                                          .         '

Krewskl, D., D. Gaylor  and M.  Szyszkowlcz.   1991.  A model-free approach  to
low dose  extrapolation.  Environ.  Health Perspect.   90: 279-285.   (Cited  1n
U.S. EPA, 1991b)

Kulkarnl, M.S., K. Galloway, M.F. IMgaray and D.G. Kaufman.  1986.  Species
differences  1n the  formation  of  benzo(a)pyrene-DNA  adducts 1n  rodent  and
human endometrlum.  Cancer  Res.   46:  2888-2891.

Kuschner, M.,  S.  Laskln,  E. Crlstofano  and  N.  Nelson.  1956.   Experimental
carcinoma  of  the  lung.   Iri:  Proc.   3rd  Natl.  Cancer   Conf.,   Detroit.
L1pp1ncott, Philadelphia, Montreal,   p.  485-495.
       /
Kveseth, K.,  B.  Sortland  and  T.  Bokln.  1982.   Polycycllc  aromatic hydro-
carbons In sewage, mussels  and  tap water. Chemosphere.   11:  623-639.

Lacassagne,  A.,  N.P.  Buu-Hol,  F.  Zajdela,  D.  Lav1t-Lamy  and  0.  Chalvet.
1963.   ActlvHe  cancerogene  d'hydrocarbures  aromatlques  polycycllques  a
noyau fluoranthene.  Un. Int.  Cancer Acta.   19: 490-496.  (Fre.)

Laher,  J.M.  and  3.A.   Barrowman.   1983.  Polycycllc  hydrocarbon  and  poly-
chlorinated  blphenyl solub1!1zat1on  In  aqueous solutions  of mixed  micelles.
Llplds.  18: 216-222.

04460                                IX-43                           11/21/91

-------
Laher,  J.M.,  M.W.  Rlgler,  R.O.  Vetter,  J.A.  Barrowman  and  J.S.  Patton.
1984.   Similar  b1oava1lab1lHy and  lymphatic  transport  of  benzo(a)pyrene
when  administered to  rats  1n  different  amounts of  dietary fat.   J.  L1p1d
Res.  25: 1337-1342.

Lake, R.S.,  M.L.  Kropko,  M.R.  Pezzuttl.,  R.H.  Shoemaker  and H.O.  Igel.  1978.
Chemical  Induction of  unscheduled DNA  synthesis  1n  human  skin  epithelial
cell cultures.  Cancer Res.  38: 2091-2098.

Lakowlcz,  J.R. and  D.R.  Sevan.   1979.    Effects  of  asbestos,  Iron  oxide,
silica  and  carbon  black  on  the  mlcrosomal  availability  of  benzo(a)pyrene.
Biochemistry.  18: 5170-5176.

Lakowlcz, J.R. and  J.L. Hylden.   1978.   Asbestos-mediated membrane uptake of
benzo(a)pyrene observed by fluorescence spectroscopy.   Nature.  275: 446-448.

Lakowlcz, J.R., D.R.  Bevan  and S.C. Rlemer.   1980.  Transport of  a carcino-
gen, benzo(a)pyrene,  from partlculates to  I1p1d  bllayers.  Blochem. Blophys.
Acta.  629: 243-258.

Langenbach,  R.,   C.  H1x,   L.   Oglesby  and  J. Allen.   1983.   Cell-mediated
mutagenesls  of Chinese hamster  V79 cells  and Salmonella  typh1mur1um.   Ann.
N.Y. Acad. Sd.  407:  258-266.

Larlnov,  L.F.  and N.G.  Soboleva.   1938.    Gastric  tumors  experimentally
produced  In  mice by  means  of  benzopyrene   and  dlbenzanthracene.   Vestn.
Rentgenol. Radio!.  20: 276-286.


04460                                IX-44                            10/07/91

-------
Laskln, S., M. Kuschner and R.T. Drew.  1970.  Studies  In pulmonary carclno-
genesls.   In:  Inhalation  Cardnogenesls,   U.S.   Atomic  Energy  Commission
Symp.,  M.G.   Hanna,  P.  Netteshelm  and  J.R.  Gilbert,  Ed.   Series  No.  18.
p. 321-351.

lasnltzkl,  A. and  D.L.  Woodhouse.   1944.   The  effect  of l:2:5:6-d1benz-
anthracene on the lymph-nodes  of the rat.   J.  Anat.  78: 121-129.

LaVole, E.J., E.V. Bedenko, N.  Hlrota, S.S. Hecht  and 0. Hoffmann.  1979.  A
comparison  of  the  mutagenlcHy,   tumor-1n1t1at1ng   activity  and  complete
cardnogenldty  of  polynuclear  aromatic  hydrocarbons.   1ft:   Polynuclear
Aromatic  Hydrocarbons,  P.W.  Jones   and  P.  Leber,  Ed.   Ann  Arbor  Science
Publishers, Ann Arbor,  HI.  p. 705-721.

LaVole,  E.J., S.S.  Hecht,  S.  Am1n,  V.  Bedenko and  D.  Hoffmann.   1980.
Identification of  mutagenlc  d1hydro
-------
LaVole,  E.J.,  S.   Am1n,  S.S.  Hecht,  K.  Furuya  and  D.  Hoffmann.    1982a.
                                                    <,
Tumour   Initiating  activity   of   dlhydrodlols   of   benzo(b)f1uoranthene,
benzo{j)fluoranthene and benzo(k)fluoranthene.   Cardnogenesls.   3: 49-52.

LaVole,  E.J.,  S.S.  Hecht,  V.  Bedenko  and D.  Hoffmann.   1982b.  Ident1f1-
                                   •
cation  of  the mutagenlc  metabolites of  fluoranthene,  2-methylfluoranthene,
and 3-methylfluoranthene.  Cardnogenesls.  3:  841-846.

LaVole,  E.J.,  D.T.  Coleman,  J.E.  Rice,  N.G.  Geddle and D. Hoffmann.   1985.
Tumor-Initiating  activity,   mutagenldty,   and  metabolism  of  methylated
anthracenes.  Cardnogenesls.  6: 1483-1488.

LaVole,  E.J.,  J.  Braley,  J.E.  R1ce and  A.   Rlvenson.   1987.    Tumor1gen1c
activity  of  non-alternant  polynuclear   aromatic   hydrocarbons   1n   newborn
mice.  Cancer Lett.  34: 15-20.

Lee, P.M. and J.A.  O'Neill.  1971.   The  effect  both  of  time and  dose  applied
on  tumor  Incidence rate  1n  benzopyrene  skin  painting  experiments.   Br.  J.
Cancer.  25: 759-770.

Legraverend, C., D.E.  Harrison,  F.W. Ruscettl  and  D.U. Nebert.   1983.   Bone
marrow  toxlclty  Induced by  oral benzo[a]pyrene:  Protection  resides  at  the
level of the Intestine and liver.  Toxlcol.  Appl. Pharmacol.   70:  390-401.

Legraverend, C., T.H.  Guenthner  and  0.  W. Nebert.   1984.  Importance  of  the
route of administration for  genetic  differences  1n benzo[a]pyrene-1nduced  1.n
utero toxlclty and teratogenldty.   Teratology.   29:  35-47.

04460                                IX-46                          10/07/91

-------
Lelfer, Z., T.  Kada,  H.  Handel,  E. Zelger, R. Stafford and H.S.  Rosenbranz.
1981.   An  evaluation of   tests   using  ONA  repair-deficient  bacteria   for
predicting genotoxldty and cardnogenldty.   Mutat.  Res.   87:  211-297.

Lesko,  S.A.,   R.J.   Lorentzen  and  P.O.P.  Ts'o.   1978.    Benzo(a)pyrene
metabolism:  One-electron  pathways  and  the  role  of nuclear  enzymes.    In:
Polycycllc  Hydrocarbons  and Cancer:  Environment,  Chemistry and  Metabolism,
Vol. 1, H.V. Gelboln and  P.O.P. Ts'o,  Ed.   Academic  Press,  Inc.,  New  York.

Lettlnga,   T.W.   1937.  De  Cardnogene wlrklng  van  klelne doses  l,2,5,6-d1-
benzanthraceen.  MO Thesis,  Amsterdam.  (Cited 1n  IARC,  1973)

Levin,  W., A.U.  Mood,  R.L.  Chang et  al.   1978.   Evidence  for  bay-region
activation  of  chrysene !,2-d1hydrod1ol  to an  ultimate carcinogen.   Cancer
Res.  38:  1831-1834.

L1,  A.P.   1984.   Use of Aroclor  1254-lnduced  rat  liver  homogenate 1n  the
      /'
assaying  of  promutagens  1n Chinese hamster  ovary cells.   Environ.  Mutagen.
6: 539-544.

L1J1nsky,   W.,  H.  Garcia,  B. Terraclnl and U. Safflottl.   1965.   Tumorlgenlc
activity  of hydrogenated  derivatives  of  d1benz(a,h)anthracene.   J. Natl.
Cancer Inst. 34: 1-6.  (Cited 1n IARC, 1973)

L1ndahl-K1essl1ng,  K.,  I.  Karlberg  and  A.  Olofsson.  1989.   Induction  of
sister-chromatid   exchanges  by   direct   and  Indirect  mutagens  1n  human
lymphocytes, co-cultured  with  Intact rat  liver  cells.   Mutat.  Res.   211:
77-87.

04460                                IX-47                            11/21/91

-------
L1n1ck,  M.   1983.    Illness   associated  with  exposure  to  naphthalene  1n
mothballs - Indiana.  Morbid.  Mortal. Weekly Rep.   32:  34-35.

to, S.M.   1964.   Distribution  of labeled DMBA 1n rat  tissues  and  organs  and
Its excretion from the body.  Voprosy Onkologl.   10:  41-44.

Lofroth, G., R. Toftgard, L. Nllsson,  E.  Agurell  and J.A.  Gustafsson.   1984.
Short-term  bloassays  of  nltro  derlvates  of benzo[a]pyrene  and  perylene.
Carclnogenesls.  5(7): 925-930.

Lorenz, E.  and  H.L.  Stewart.   1947.  Tumors of the  alimentary tract  Induced
1n mice by  feeding  olive  oil emulsions containing carcinogenic hydrocarbons.
J. Natl. Cancer Inst.  7:  227-238.

Lorenz, E.  and  H.L.  Stewart.   1948.  Tumors of the  alimentary tract  1n mice
fed  carcinogenic  hydrocarbons  In   mineral-oil  emulsions.   J. Natl.  Cancer
Inst.  9: 173-180.
       /
                                           r
Lubet,  R.A.,  G.M.  Connolly,  D.W.   Nebert  and R.E.  Kourl.   1983a.   Dlbenz-
[a,h]anthracene-1nduced subcutaneous tumors  1n mice.   Strain  sensitivity  and
the role of carcinogen metabolism.   Carclnogenesls.   4(5):  513-517.

Lubet,  R.A.,  E.  K1ss,  M.M.  Gallagher,  C. Dlvely,  R.E.  Kourl  and  L.M.
Schechtman.  1983b.   Induction of  neoplastlc  transformation   and  DNA  single
strand .breaks  1n  C3H/10T 1/2  clone 8  cells by polycycllc hydrocarbons  and
alkylatlng agents.  J. Natl. Cancer Inst.   71(5):  991-997.
04460                                IX-48                           11/21/91

-------
Lubet, R.A., M.J. Brunda, B. Lemalre, 0.  Dansle,  D.W.  Nebert  and  R.E.  KouM.

1984.  Pqlycycllc hydrocarbon Induced 1mmunotox1c1ty 1n mice: Role of  the  Ah

locus.   ITK   Mechanisms,   Methods   and  Metabolism: . Polynuclear   Aromatic

Hydrocarbons.   8th   Int.  Symp.,  M.  Cooke  and  A.J.  Dennis,  Ed..  Battelle

Press, Columbus, OH.  p. 843-855.



Lyte, M.  and P.H.  B1ck.   1986.  Modulation  of  1nterleuk1n-l production  by

macrophages  following  benzo[a]pyrene   exposure.   Int.  J.   Immunopharmacol.

8(3): 377-381.


MacGregor,  R.R.   1954.  Naphthalene  poisoning from the  1ngest1on of  moth-

balls.  Can. Med. Assoc. J.   70: 313-314.



Mackell, J.V.,  F.  Rleders,   H.  Brleger  and  E.L.  Bauer.   1951.   Acute  hemo-

lytlc anemia due to  1ngest1on of  naphthalene  mothballs: I.  Clinical  aspects.

Pediatrics.  7: 722-728.


      /
Mackenzie,  K.M.  and D.M. Angevlne.   1981.    Infertility  1n mice exposed  in

utero to benzo[a]pyrene.  B1ol.  Reprod.   24:  183-191.



Maclure,  K.M.   and   B.  MacMahon.   1980.   An ep1dem1olog1c  perspective  of

environmental cardnogenesls.  Epldem.  Rev..  2: 19-48.



Mahvl,  D.,  H.  Bank  and R.  Harley.    1977.   Morphology  of  a  naphthalene-

Induced bronchlolar  lesion.   Am. J.  Pathol.   86:  559-566.
04460                                IX-49                           10/07/91

-------
Majesky,  M.W.,  H-Y.L.  Yang, E.P.  BendHt  and  M.R.  Juchau.  1983.   Cardno-
genesls  and  atherogenesls:  Differences  1n monooxygenase  1nduc1b1l1ty and
b1oact1vat1on  of  ben2o[a]pyrene  In  aortic and  hepatic  tissues of  arthero-
scleros1s-suscept1ble versus resistant pigeons.   Carclnogenesls.   4:  647-652.

Mallory, V.T., R.W. Na1smith and  R.J. Matthews.  1985a.  Acute  exposure  oral
toxlclty  82  EPA/OECD  (unpublished   material).   Pharmakon  Research  Inter-
national,  Inc.,  Waverly,  PA.   Submitted  to Texaco,  Inc.,  Beacon,  NY.   PH
422-TX-002-84. 5601-56-1.  (CHed In  U.S. EPA,  1987a)

Mallory,  V.T.,  R.W.  Nalsmlth  and R.J.  Matthews.   1985b.   Acute  exposure
dermal  toxlclty  83  EPA/OECD   {unpublished  material).   Pharmakon  Research
International,  Inc.,  Waverly,   PA.   Submitted  to Texaco,  Inc., Beacon, NY.
PH 402-TX-002-84.  5601-56-1.  (CHed  1n  U.S. EPA, 1987a)

Mallory,  V.T.,  R.W.  Nalsmlth   and  R.J.  Matthews.   1985c.   Rabbit  eye
Irritation   study   (wash)   (unpublished   material).    Pharmakon   Research
      /
International,  Inc.,  Waverly,   PA.   Submitted  to Texaco,  Inc.,  Beacon, NY.
PH 402-TX-009-84.  5601-56-1.  (CHed  1n  U.S. EPA, 1987a)

Mallory,  V.T.,  R.W.  Nalsmlth  and R.J.  Matthews.   1985d.   Primary  dermal
Irritation  study   1n  rabbits   83/EPA   (unpublished  material).   Pharmakon
Research  International,   Inc.,   Waverly,   PA.   Submitted  to  Texaco,   Inc.,
Beacon,  NY.  PH 420-TX-013-84.  5601-56-1.  (CHed 1n U.S. EPA,  1987a)
04460                                IX-50                           10/07/91

-------
Mallory,  V.T.,  R.W.  NalsmHh  and  R.J.  Matthews.   1985e.   Delayed contact
hypersensHWHy In  guinea  pigs  (unpublished material).  Pharmakon  Research
International,  Inc.,  Waverly,  PA.   Submitted  to Texaco,  Inc.,  Beacon, NY.
PH 424-TX-001-84.  5601-56-1.  (CHed In U.S.  EPA,  1987a)

Malmgren, R.A., B.E.  Bennlson  and T.W. McKlnley,  Jr.   1952.  Reduced  anti-
body  tHers  In mice  treated with  carcinogenic  and cancer  chemotherapeutk
agents.  Proc. Soc. Exp. B1ol.  Med.   79: 484-488.

Marnett, L.J. and  G.A.  Reed.   1979.  Perox1dat1c  oxidation of tenzo(a)pyrene
and prostaglandln  biosynthesis.  Biochemistry.   18:  2923-2929.

Marnett, L.J., G.A. Reed and D.J. Dennlson.  1978.   Prostaglandln  synthetase
dependent    activation    of   7,8-d1hydro-7,8-d1hydroxy-benzo(a)pyrene    to
mutagenlc derivatives.  Blochem.  Blophys.  Res.  Comm.  82:  210-216.

Marquardt, H. and  C.  Heldelberger.   1972.   Influence  of  "feeder cells" and
      /
Inducers  and  Inhibitors  of  mlcrosomal  mixed-function  oxldases  on  hydro-
carbon-Induced  malignant  transformation  of  cells   derived  from  C3H   mouse
prostate.  Cancer  Res.  32:  721-725.

Martin,  C.N.  and  A.C.  McDermld.   1981.   Testing of 42 coded compounds for
their ability to Induce unscheduled DNA repair synthesis  1n  HeLa cells.  In:
Evaluation of Short-term Tests for Carcinogens.   Report of the International
Collaborative  Program.    Progressive   Mutation  Research,  Vol.   1,  F.J.   de
Serres and J. Ashby, Ed.  Amsterdam,  Elsevler,  North  Holland,  p. 533-537.
04460                                IX-51                            10/07/91

-------
Martin,  C.N.,  A.C.  McDermld  and  R.C.   Garner.   1978.   Testing of  known
carcinogens  and  noncardnogens  for their ability  to  Induce unscheduled DNA
synthesis In HeLa cells.  Cancer Res.   38: 2621-2627.

Hatljasevlc, Z. and E.  Zelger.  1985.  Mutagenldty of pyrene 1n  Salmonella.
Mutat Res.  142: 149-152.

Hatorova, N.I.  1982.   Materials  to substantiate maximum permissible concen-
trations  of  naphthalene and chloronaphthalene 1n reservoir water.  [English
translation]  Gig. Sanlt'.   11:  78-79.

Matsuoka, A., T.  Sofunl,  S.  Sato, N.  Mlyata and M. Ishldate, Jr.  1987.  Ijn
vitro clastogenldty of nltropyrenes  and nHrofluorenes.   Mutat. Res.  182:
366-367.

Mattlson, D.R.  and S.S.  Thorgelrsson.   1977.  Ovarian metabolism  of poly-
cycllc  aromatic  hydrocarbons   and associated  ovotox1c1ty  In   the  mouse.
      /
Gynecol. Invest.  8:  11.

Mattlson, D.R.,  K.  Shlromlzu and  M.S. Nightingale.   1983.  Oocyte destruc-
tion by polycycllc aromatic hydrocarbons.  Am. 0. Ind. Med.  4: 191-202.

McCann,   J.,.  E.   Choi,  E.  Yamasakl   and  8.N.  Ames.    1975.   Detection  of
carcinogens  as  mutagens  1n  the  Salmonella/mlcrosome  test:  Assay  of  300
chemicals.  Proc.  Natl.  Acad. Sd. USA.  72:  5135-5139.
04460                                IX-52                           10/07/91

-------
McCarroTI,  N.E.,  B.H.   Keech  and  C.E.  Piper.   1981.    A mlcrosuspenslon
adaptation  of  the  Bacillus  subtnis   'rec'  assay.   Environ.  Mutagen.   3:
607-616.

McCormlck, D.L.,  F.J. Burns and R.E. Albert.  1981.  Inhibition of benzo[a]-
pyrene-lnduced mammary  carclnogenesls  by retlnyl  acetate.   J. Natl.  Cancer
Inst.  66: 559-564.

McGIll,  H.C.   1977.   Atherosclerosis:  Problems   1n  pathogenesls.   Athero-
sclerosis Rev.   2:  27-65.  (Cited  1n Hartman,  1983)

McQueen,  C.A.,   D.H.  Krelser  and  G.M.  Williams.   1983.   The  hepatocyte
primary  culture/ONA  repair  assay  using  mouse  or  hamster  hepatocytes.
Environ.  Mutagen.   5:  1-8

Met1v1er, H., J. Wahrendorf and R. Masse.   1984.   Multiplicative effect of
Inhaled  plutonlum  oxide and benzo(a)pyrene on  lung  carclnogenesls  1n rats.
      /
Br. J. Cancer.   50: 215-221.

Mlescher, G.   1942.  Experimental  studies  on  the formation  of  tumours by
photosensH1zat1on.   Schwelz.  Med.   Wochenschr.   72:   1082-1084.    (Ger.)
(Cited 1n IARC,  1983)

Mill,  T.,  H.R.  Mabey,  B.Y.  Lan  and  A.   Baraze.    1981.   Photolysis  of
polycycllc aromatic hydrocarbons  In water.   Chemosphere.   10:  1281-1290.

Mitchell, C.E.   1982.   Distribution and retention of benzo(a)pyrene 1n  rats
after Inhalation.  Toxlcol. Lett.   11:  35-42.

04460                                IX-53                           10/07/91

-------
Mitchell,  C.E.   1983.  The  metabolic  fate of  benzo[a]pyrene In rats  after


Inhalation.  Toxicology.  28: 65-73.





Mitchell,  C.E.  and K.W. Tu.  1979.  Distribution,  retention  and elimination


of  pyrene  1n  rats   after  Inhalation.   J.   Toxlcol.  Environ.  Health.   5:


1171-1179.                            .



              i

Mitchell,  A.D., C.J.  Rudd and W.J.  Caspary.   1988.   Evaluation  of the L5178Y


Mouse   lymphoma   cell   mutagenesls  assay:   Intralaboratory   results   for


sixty-three  coded  chemicals  tested  at  SRI   International/   Environ.  Mol.


Mutagen.   12(Suppl. 13): 37-101.





Modlca,  R.,   M.  Flume,  A.   Gualtanl  and  I.   Bartosek.   1983.   Comparative


kinetics  of  benz(a)anthracene,  chrysene and trlphenylene 1n  rats after  oral


administration.  Toxlcol. Lett.   18: 103-109.



       t

Mohammad,  S.N.   1983.   Ind.  J. Chem.   A22:   639-641.   (Cited   1n  Mohammad,
      /

1985)





Mohammad,  S.N.   1985.  MINOO/3 calculations   of  .carcinogenic activities  of


polycycllc hydrocarbons.  Ind.  J. Blochem. Blophys.   22:  56-63.





Mohr,  U.   1971.   Kanzerogenese  durch D1athyln1trosam1n.  Untersuchungen zur


Dlaplazentaren Wlrkung.  Fortschr.  Med.   89:  251-253.  (Ger.)





Moolgavkar, S.H.   1986.   Cardnogenesls modeling: From molecular biology  to


epidemiology.  Ann. Rev. Public  Health.   7: 151-169.






04460                                IX-54                           10/07/91

-------
Moolgavkar, S.H. and A.G.  Knudson.   1981.  Mutation  and  cancer:  A model for
human cardnogenesls.   0.  Natl.  Cancer  Inst.  66: 1037-1052.

Moore,  B.P.  and G.M.  Cohen.   1978.   Metabolism  of  benzo(a)pyrene  and Us
major metabolites  to ethyl  acetate-soluble  and water-soluble metabolites by
cultured rodent trachea.   Cancer  Res.   38: 3066-3075.

Moore, P.O., M.  Koreeda,  P.G. Wlslockl  et  al.   1977.  In vitro reactions of
the  dlastreomerlc   9,lO-epox1des  of   (t)  and  (-)-trans-7,8-d1hydroxy-7,8-
d1hydrobenzo[a]pyrene  with  polyguanyllc  add  and evidence  for  formation of
an  enantlonmer  of  each  dlastereomerk  9,lO-epox1de  from  benzo[a]pyrene In
mouse  skin.   IJK  Drug  -Metabolism  Concepts,   D.M.  Jerlna,  Ed.   American
                                  \
Chemical   Society,   Symp.  Series   No.   44,   American   Chemical   Society,
Washington, DC.  (Cited 1n Santodonato  et al.,  1981)

Morris,  H.P.,  C.A. Velat,  B.P.  Wagner,  M.  Dahlgard  and F.E.  Ray.   1960.
Studies  of cardnogenldty 1n  the rate of  derivatives  of  aromatic  amines
related to 0-2-fluorenylacetam1de.   J.  Natl. Cancer Inst.  24: 149-180.

Mortelmans  K.,  S.  Haworth, T.  Lawlor, W. Speck,  B. Talner and  E.  Zelger.
1986.   Salmonella  mutagenldty  tests:  II. Results  from  the testing  of 270
chemicals.  Environ.  Mutagen.  8(Suppl.  7): 1-119.

Mossanda,  K.,  F. Poncelet,  A. Fouassln and  M.  Mercler.  1979.   Detection of
mutagenlc  polycycllc  aromatic  hydrocarbons  1n African  smoked  fish.    Food
Cosmet. Toxlcol.  17: 141-143.
04460                                IX-55                           11/21/91

-------
MuMson,  G.L.,  S.D.  Hall  and  A.M.  Lee.    1984.   Metabolic  activation  of
procarclnogens  to  genotoxlc  products  In  cultured  rat  liver  cells.   Expl.
Cell B1ol.  52:  355-360.

Nagornyl, P.A.  1969.  Comparative  study of  toxldty  In pure and commercial
anthracene.   G1g.  Tr. Prof.  Zabol.  -13(5):  59-61.

Nalman, J.L. and M.H. Kosoy.  1964. Red cell glucose-6-phosphate dehydrogen-
ase deficiency  — A  newly recognized  cause of  neonatal jaundice and kernlc-
terus In Canada.  Can. Med.  Assoc.  J.   91:  1243-1249.

Nalsmlth, R.W.  and   R.J.  Matthews.   1985.   Dose-range  finding-developmental
toxldty  study  1n   rabbits  (unpublished  material).   Pharmakon  Research
International, Inc.  Uaverly, PA.  Submitted to Texaco, Inc., Beacon, NY.  PH
329DR-TX-001-85.  5601-56-1.

NaUmHh, R.W.  and   R.3.  Matthews.   1986.   Developmental  toxldty  study 1n
rabbits'  (unpublished material).   Pharmakon  Research  International,  Inc.,
Waverly,  PA.   Submitted  to  Texaco,  Inc.,  Beacon,  NY.   PH  329-TX-001-85.
5601-56-1.

Nakamura,  S.,  Y.   Oda,  T.  Shlmada,   I.   Ok1  and  K.  Suglmoto.   1987.
SOS-lndudng  activity of  chemical  carcinogens  and  mutagens  In Salmonella
typh1mur1uro TA1535/pSK 1002:   Examination  with  151  chemicals.  Mutat. Res.
192: 239-246.
04460                                IX-56                           11/21/91

-------
Nakayama,  J.,  S.H. Yuspa  and M.C.  PolMer.   1984.   Benzo[a]pyrene  --  DNA
adduct  formation  and  removal  In mouse  epidermis  j_n vivo  and  Ui  vitro:
Relationship of  DNA binding  to  Initiation of  skin cardnogenesls.   Cancer
Res.  44: 4087-4095.
                                   •
Narbonne,  J.F.,   P.  Cassand,  P.   Alzleu,  P.. Gro'Her, G.  Mrllna  and  O.P.
Calmon.   1987.    Structure-activity  relationships   of  the  N-methylcarbamate
series 1n Salmonella typhlmurlum.   Mutat.  Res.   191: 21-27.

NAS  (National  Academy  of  Sciences).   1972a.   Biologic  effects  of  atmo-
spheric pollutants: Partlculate polycycllc organic matter.   Washington,  DC.

NAS (National Academy of Sciences).   1972b.  Chemical  reactivity of PAHs  and
azaarenes.  In: Polycyllc Organic Matter.  Natl. Acad. Sd.. Washington,  DC.
p. 63-81.  (Cited In IARC,  1973)

NAS  (National  Academy  of  Sciences).   1977.   Drinking  Water  and  Health.
      /
Vol. 1, p. 19-63.

NAS  (National  Academy  of  Sciences).   1980.   Drinking  Water  and  Health.
Vol. 3, p. 25-65.

Neal,  J.  and R.H.  Rlgdon.   1967.   Gastric tumors In mice fed  benzo[a]pyrene
— A quantitative study.   Tex. Rep.  Blol.  Med.   25:  553-557.
04460                                IX-57                           10/07/91

-------
Nebert,  D.W.    1981.   Genetic  differences  In  susceptibility  to  chemically
Induced  myelotoxldty  and leukemia.   Environ.  Health Perspect.   39:  11-22.
(Cited 1n Legraverend et a!., 1983)

Nebert,  D.W.,  N.M.  Jensen,  R.C.  Levitt  and  J.S.  Fellon.   1980.    Toxic
chemical depression of the bone marrow and  possible  aplastlc anemia explain-
able on  a genetic  basis.  CUn. Toxlcol.  16: 99-122.   (Cited  1n Legraverend
et al.,  1983)

Nebert,  D.W.,  H.J.  E1sen, M.  Neg1sh1,  M.A.  Lang and  L.M.  Hjelmeland.   1981.
Genetic  mechanisms  controlling  the Induction of polysubstrate  monooxygenase
(P-450)  activities.  Ann. Rev. Pharmacol.  Toxlcol.   21:  431-462.

Nebert,  D.W.,  M.  Neg1sh1, M.A.  Lang,  L.M. Hjelmeland and  H.J.  E1sen.   1982.
The  Ah  locus,  a  multlgene  family necessary for  survival  In  a  chemically
adverse  environment:  Comparison with  the  Immune  system.   Adv. Genet.   21:
1-52.  (Cited 1n Legraverend et al., 1983)

Nesnow,  S.  and C.  Heldelberger.   1976.   The effect  of modifiers of  mlcro-
somal  enzymes  on  chemical oncogenesls  1n cultures  of C3H mouse cell  lines.
Cancer Res.   36: 1801-1808.

Nesnow,  S.,  H. Bergman and T.J. Slaga.  1986.   Comparison  of the tumoMgenlc
response of SENCAR and C57BL/6 mice to  benzo[a]pyrene  and the  1nterexper1-
mental variability over a three-year period.  Environ.  Health Perspect.   68:
19-25.
04460                                IX-58                           10/07/91

-------
Nlkonova, T.V.   1977.   Transplacental action  of  benzo[a]pyrene and  pyrene.
Bull. Exp. B1ol. Med.  84:  1025-1027.   (CHed  1n  IARC,  1983)

NIOSH  (National  Institute   for  Occupational  Safety  and  Health).    1973a.
Recommended Standards for Occupational Exposure to Coal Tar  Products.   NIOSH
Publ. No. 78-107.  U.S.  DHEW.

NIOSH  (National  Institute   for  Occupational  Safety  and  Health).    1973b.
Recommended  Standards   for  Occupational  Exposure to  Coke  Oven  Emissions.
NIOSH Publ.  No. 73-11016.   U.S.  DHEW.                        -

NIOSH  (National  Institute   for   Occupational  Safety  and   Health).    1977.
Registry  of  toxic effects of  chemical  substances.  Volume  II.   NIOSH  Publ.
No. 78-104-B.  U.S. DHEW.

Norpoth,  K.,  A.  Kemena, 0.   Jacob  and C.  Schumann.  1984.  The  Influence  of
18  environmentally  relevant  polycycllc  aromatic  hydrocarbons  and  Clophen
      /
A50, as  liver monooxygenase Inducers,  on the mutagenlc  activity of  benzo-
[a]anthracene In the Ames  test.   Carclnogenesls.   5(6): 747-752.

NTP  (National  Toxicology  Program).   1980a.  Unpublished  subchronlc  toxldty
study:  Naphthalene  (C52904),  B6C3F1  mice.   Batelle  Columbus  Laboratory,
Columbus, OH.  Subcontract no.  76-34-106002.

NTP  (National  Toxicology  Program).   1980b.  Unpublished  subchronlc  toxldty
study: Naphthalene  (C52904), Fischer  344  rats.  Batelle Columbus Laboratory,
Columbus, OH.  Subcontract no.  76-34-106002.


04460                                IX-59                           10/07/91

-------
NTP  (National  Toxicology  Program).    1991a.   National  Toxicology  Program
Chemical Status Report, 04/03/91.  Research Triangle Park, NC.

NTP  (National Toxicology  Program).   1991b.   Toxicology and  Cardnogenesls
Studies  of Naphthalene  In  B6C3F1 Mice  (Inhalation  studies).   NTP  TR  410.
NIH Publ. No. 91-3141.

Oberly,  T.J., B.J.  Bewsey and  G.S.  Probst.   1984.  An  evaluation of  the
L5178Y  TK+/-  mouse  lymphoma  forward  mutation  assay  using  42  chemicals.
Mutat. Res.   125: 291-306.

O'Brien,   K.A.F.,   L.L.   Smith  and  G.M.  Cohen.    1985.   Differences   1n
naphthalene-Induced  toxldty  1n  the  mouse and  rat.  Chem.-B1ol.  Interact.
55: 109-122.

Oesch, P.,  M. Bucker and H.R.  Glatt.   1981.   Activation of phenanthrene  to
                                  i
mutagenlc  metabolites and  evidence  for  at least  two  different  activation
pathways.  Mutat. Res.  81:  1-10.

O'Gara,  R.W., M.G.  Kelly,  J.   Brown  and  N.  Mantel.   1965.   Induction  of
tumors  In  mice  given  a  minute  single   dose  of  d1benz(a,h)anthracene  or
3-methylcholanthrene  as  newborns.  A dose-response  study.   J. Natl.  Cancer
Inst.  35(6): 1027-1039.

Okuda,  T.,  E.S.  Vesell,  V.  Plotkln  et  al.   1977.   Interlndlvldual  and
1ntra1nd1v1dual variations 1n aryl hydrocarbon hydroxylase  1n  monocytes  from
monozygotlc and dlzygotlc twins.   Cancer Res.   37:  3904-3911.


04460                                IX-60                           11/21/91

-------
OSHA  (Occupational  Safety and  Health  Administration).   1985.   Occupational
Standards and Permissible Exposure Limits.   29  CFR  1910.1000.   p.  655.

Oshlro, Y., P.S. Balwlerz and C.E. Piper.   1988.   Evaluation  of  the  division
arrest method  of the  CHO/HGPRT mutation  assay.   J. Ap.pl.-  Toxlcol.   8(2):
129-134.

Pahlman,   R.  and 0.  Pelkonen.    1987.   Mutagenlclty  studies  of different
polycycllc aromatic  hydrocarbons:  the  significance of enzymatic  factors  and
molecular structure.  Cardnogenesls.   8: 773-778.

Palgen, B., H.L. Gurtoo,  J. Hlnowada et al.  1977.   Questionable relation of
aryl  hydrocarbon  hydroxylase  to  lung-cancer  risk.  New Engl.  J. Med.  297:
346-350.

Palgen, B., E.  Ward,  K.  Steenland, L.  Houten, H.L.  Gurtoo and  J.  Hlnowanda.
1978a.  Aryl hydrocarbon  hydroxylase  1n cultured lymphocytes of  twins.   Am.
J. Hum.- Genet.   30:  561-571.

Palgen,  B.,  H.L. Gurtoo, E. Ward et  al.  1978b.   Human  aryl  hydrocarbon
hydroxylase  and cancer  risk.  In:  Polynuclear Aromatic  Hydrocarbons:  2nd
International  Symposium  on Analysis,  Chemistry  and Biology:  Cardnogenesls
—  A  Comprehensive  Survey,   Vol.  3,  R.I.  Freudenthal  and  P.W. Jones,  Ed.
Raven Press, NY.  p. 429-438.
04460                                IX-61                            11/21/91

-------
Pal,  K.   1981.    The   Induction  of  sister-chromatid  exchanges  In  Chinese
hamster ovary  cells  by K-reg1on epoxldes and some  dlhydrodlols  derived  from
benz[a]anthracene,  d1benz[a,c]anthracene  and d1benz[a,h]anthracene.   Mutat.
Res.  84: 389-398.

PalHtl, F., R. Cozzl,  M.  Flore et  al.   1986.   An \n vitro and In vivo study
on  mutagenlc  activity  of   fluoranthene:   Comparison  between  cytogenetlc
studies and HPLC analysis.  Mutat. Res.   174: 125-130.

Pankow, J.F.,  L.M.  Isabelle  and W.E.  Asher.  1984.   Trace  Organic  compounds
1n  rain.   1.  Sampler  design and  analysis  by adsorption/thermal  desorptlon
(ATD).  Environ.  Sc1. Technol.  18:  310-318.

Pauluhn,  J.,  J.   Thyssen,   J.  Althoff,  G.  Klmmerle  and  U.  Mohr.   1985.
Long-term  Inhalation  study  with  benzo[a]pyrene  and  SO-  1n  Syrian  golden
hamsters.  Exp. Pathol.  28:  31.  (Abstract)
      /
Peacock,  P.R.    1935.    Studies  of  fowl   tumours  Induced  by  carcinogenic
agents.  Am. J. Cancer.  25:  37.  (Cited 1n  IARC.  1973)

Peacock, P.R.   1936.   Evidence regarding  the  mechanism  of  elimination  of
1,2-benzpyrene,  1,2,5,6-dlbenzanthracene,   and  anthracene  from  the  blood-
stream of Injected animals.  Br. J.  Exptl. Pathol.  17: 164-172.

Pelkonen,  0.   and  D.H.  Nebert.  1982.   Metabolism  of polycycllc  aromatic
hydrocarbons:  Et1olog1c  role   In  carclnogenesls.    Pharmacol.   Rev.    34:
189-222.
04460                                IX-62                           11/21/91

-------
Penn,  A.  and  C.   Snyder.    1988.   Ater1osclerot1c  plaque  development  1s
promoted   by   polynuclear   aromatic   hydrocarbons.    Cardnogenesls.    9:
2185-2189.

Peralno,  C.,  E.F.  Staffeldt,  B.A. Carnes,  V.A.  Ludeman,  J.A. Blomqulst  and
S.D. Vessel1nov1tch.   1984.   Characterization of  h1stochem1cally  detectable
altered  hepatocyte  fod  and  their relationship  to hepatic  tumorlgenesls  1n
rats treated  once  with d1ethyln1trosam1ne  or  benzo[a]pyrene within one  day
after birth.  Cancer Res.   44: 3340-3347.

Perelra,  M.A.,   F.3.   Burns   and  R.E.  Albert.   1979.   Dose  response  for
benzo[a]pyrene adducts 1n  mouse  epidermal  DNA.  Cancer Res.  39:  2556-2559.
(Cited In Clement Associates,  1988)

Perera,  P.P.  1990.   Molecular  epidemiology: A  new tool  1n  assessing  risks
of environmental carcinogens.   Cancer 3. Cl1n.   40(5):  277-288.
      /
Perry,   P.E.  and E.J.  Thomson.    1981.   Evaluation of  the   sister  chromatld
exchange  method  1n  mammalian cells  as a  screening system  for  carcinogens.
In:  Evaluation  of  Short-term Tests  for Carcinogens.   Report of the  Inter-
national  Collaborative Program.   Progress  1n  Mutation  Research,  F.J.  de
Serres  and J. Ashby, Ed.   Elsevler/North Holland, New York.   p.  560-569.

Pershagen  G.,  G. Nordberg and  N.E.  Bjdrklund.   1984.  Carcinomas  of  the
respiratory  tract  1n  hamsters given arsenic trloxlde  and/or benzo(a)pyrerie
by the pulmonary route.  Environ  Res.  34:  227-241.
04460                                IX-63                           10/07/91

-------
Peterson, A.R., J.R. Landolph, H. Peterson, C.P.  Spears  and  C.  Heldelberger.
1981.   Oncogenlc  transformation  and mutation  of C3H/10T1/2  clone 8  mouse
embryo flbroblasts by alkylatlng agents.   Cancer Res.   41:  3095-3099.

Pfelffer,  E.H.    1973.   Investigations  on  the  carcinogenic  burden  by  air
pollution In man.  VII. Studies on  the oncogenetlc  Interaction  of  polycycllc
aromatic hydrocarbons.   Zbl. Bakt. Hyg.,  I.  Abt. Or1g.  B.  158:  69-83.

Pfelffer,  E.H.   1977.   Oncogenlc   Interaction  of   carcinogenic   and   non-
carcinogenic polycycllc  aromatic  hydrocarbons  1n mice.   In:  A1r  Pollution
and Cancer 1n Han, V. Mohr et a!., Ed.   IARC,  Lyon,  France,   p.. 69-77.

Philips, F.S.,  S.S.  Sternberg and H. Harquardt.   1973.   In vivo  cytotoxlclty
of polycycllc hydrocarbons.   In.:  Pharmacology and the Future of Han.   Proc.
5th Int. Congr. Pharmacol., 1972,  San Francisco, CA.   2:  75-88.

PhllUpson,  C.E.   and  C.   loannldes.    1989.   Metabolic  activation   of
polycycllc aromatic  hydrocarbons  to  mutagens  1n  the Ames  test by  various
animal species  Including man.  Hutat. Res.   211: 147-151.
         R.J.,   J.A.  Polley  and  W.B.   Lebherz,  III.   1977.   Horphologlcal
transformation  of  early passage golden  Syrian  hamster embryo  cells  derived
from  cryopreserved  primary  cultures  as  a reliable  An.  vitro  bloassay  for
Identifying diverse carcinogens.  Int.  3. Cancer.   19: 642-655.

Pierce,  W.E.H.   1961.   Tumour -promotion  by  lime  oil  1n   the  mouse  fore-
stomach.  Nature (London).   189: 497-498.

04460                                IX-64                           11/21/91

-------
Pletra, G., H.  Rappaport and  P.  Shublk.   1961.   The effects of carcinogenic
chemicals 1n newborn mice.   Cancer.   14: 308-317.  (CHed 1n IARC, 1973)

P1r1e,  A.  and  R.  van  Heynlngen.   1966.   Reactions  of  1,2-d1hydroxynaph-
thalene and of  1,2-naphthaqulnone with  eye  constituents.   Blochem.  J.   100:
70p-71p.                              .        "

Plasterer,  M.R.,  M.S.  Bradshaw,  G.M. Booth,  M.W.  Carter,   R.L.  Schuler  and
B.D.  Hardln.    1985.   Developmental  tox1c1ty  of   nine  selected  compounds
following prenatal exposure 1n the mouse:  Naphthalene, p-n1trophenol, sodium
selenlte,  dimethyl  phthalate,  ethylenethlourea,   and  four  glycol  ether
derivatives.   J. Toxlcol. Environ. Health.   15: 25-38.

Platt,  K.L.,  E.H.  Pfelffer,   H.R.  Glatt  and F.  Oesch.    1983.   Bacterial
mutagenldty  and  cardnogenldty of  potential  metabolites  of d1benz[a,h]-
anthracene.  J.  Cancer  Res.  CUn.  Oncol.  105(2): A23.
      /
PolUa, J.A.   1939.   Investigations on the possible  carcinogenic  effect  of
anthracene  and  chrysene and   some  of  their  compounds.    I.  The effect  of
painting on the skin of mice.   J.  Ind. Hyg.  Toxlcol.   21: 219-220.

PolUa,  J.A.   1941.  Investigation  on  the  possible  carcinogenic  effect  of
anthracene and  chrysene  and  some  of  their  compounds.   II.  The  effect  of
subcutaneous  Injection   1n  rats.   J.  Ind.  Hyg.   Toxlcol.    223:  449-451.
(Cited In IARC,  1983)

Pool,  B.L.,   A.O.  Yalklnoglu, P.  Klein and  J.R.  Schlehofer.  1989.   DNA
amplification In genetic  toxicology.   Hutat. Res.  213: 61-72.

04460                                IX-65                  .         10/07/91

-------
Popescu,  N.C.,  D.   Turnbull  and  J.A.  DIPaolo.   1977.    Sister  chromatld
exchange  and  chromosome  aberration  analysis  with  the  use  of  several
carcinogens  and  noncardnogens.    Brief   communication.    J.  Natl.  Cancer
Inst.  59: 289-293.
                                  *
Pott,  P.   1775.  Ch1rurg1cal  observations  relative  to  the  cancer  of the
scrotum.   London.   Reprinted 1n  National  Cancer  Institute Monograph  1973.
10: 7-13.

Pott,  F.,  R.  Tomlngas  and J.  Mlsfeld.   1977.   Tumors  In'mice  after   sub-
cutaneous   Injection  of automobile  exhaust  condensates.    ln_:  A1r Pollution
and Cancer In Man, U. Mohr  et al.,  Ed.   IARC,  Lyon,  France,  p. 79-87.

Prejean, J.D., O.C.  Peckham,  A.E.  Casey,  D.P. GHswold,  E.K.  Welsburger and
J.H. Welsburger.  1973.  Spontaneous tumors 1n Sprague-Dawley rats and  Swiss
mice.  Cancer Res.  33: 2768-2773.
      /
PMchard,  R.W., J.W. Eubanks and C.C. Hazlett.   1964.  Age  and breed effects
on Induction of flbrosarcomas by d1benz(a,h)anthracene 1n pigeons.  J.  Natl.
Cancer Inst.   32:  905-915.

Probst, G.S.,  R.E.   McMahon,  I.E.  H111,  C.Z. Thompson,  J.K. Epp  and   S.B.
Neal.   1981.   Chemically-Induced  unscheduled  ONA synthesis  1n  primary rat
hepatocyte cultures:   A  comparison  with  bacterial  mutagenldty  using 218
compounds.  Environ. Mutagen.   3:  11-32.
04460                                IX-66                           11/21/91

-------
Pullman,  A.  and  B.  Pullman.   1955.   Electronic  structure  and carcinogenic
activity  of  aromatic  molecules —  New developments.   Adv.  Cancer  Res.  3:
117-169.

Pylev,  L.N.,  F.J.C.  Roe and  G.P.^ Warwick.   1969.   Elimination  of radio-
activity  after  Intratracheal   Instillation  of  trHlated  3,4-benzopyrene 1n
hamsters.  Br. J.  Cancer.  23:  103-115.

Rabsteln, L.S., R.L.  Peters  and G.J.  Spahn.   1973.   Spontaneous  tumors and
pathologic lesions 1n  SWR/J mice.   J.  Natl.  Cancer  Inst.  50: 751-758.

Rahlmtula, A.D.,  P.K. Zacharlah  and  P.J.  O'Brien.   1977.   The  effects of
antloxldants  on the metabolism and mutagenldty of benzo(a)pyrene J[n vitro.
Blochem. J.   164:  473-475.

Rahman, A.,  J.A.  Barrowman and A.  Rahlmtula.   1986.   The  Influence of  bile
on  the b1oava1lab1l1ty  of  polynuclear  aromatic  hydrocarbons  from  the rat
      /
Intestine.  Can. J.  Physlol. Pharmacol.   64:  1214-1218.

Ranadlve, K.J.  and  K.A.  Karande.   1963.   Studies  on l,2:5,6-d1benzanthra-
cene Induced  mammary cardnogenesVs In mice.   Br.  J. Cancer.  17:  272-280.

Rao,   6.S.   and  K.P.   Pandya.    1981.    Biochemical   changes  Induced  by
naphthalene after  oral  administration  In albino rats.  Toxlcol.  Lett.  8:
311-315.
04460                                IX-67                           11/21/91

-------
Rask-N1elsen,  R.   1950.   The  susceptibility of  the  thymus, lung,  subcuta-
neous  and  mammary  tissues  1n  strain  Street mice  to  direct application  of
small doses of four different  carcinogenic hydrocarbons.  Br. J.  Cancer.   4:
108-116.

Rastetter,  W.H.,  R.B.  Nachbar,  S.  Russo-RodMguez and  R.  Wattley.   1982.
Fluoranthene:  Synthesis  and  mutagenlclty  of  four  dlol  epoxldes.   J.  Org.
Chem.  47: 4873-4878.

Redo,  L.  and  A.W.   Hs1e.   1984.    Glucuronlde  conjugation   reduces  the
cytotoxldty  but  not  the mutagenlclty  of benzo[a]pyrene  1n  the  CHO/HGPRT
assay.  Teratogen. Carcinogen.  Mutagen.  4: 391-402.

Rees, J.R. and A. P1r1e.   1967.  Possible  reactions of  l,2-naphthaqu1none  1n
the eye.  Blochem. J.   102: 853-863.

Rees,  E.D.,  P. Handelstam, J.Q.  Lowry and H.  Llpscomb.   1971.   A  study  of
the mechanism  of  Intestinal absorption of  benzo(a)pyrene.  Blochem.  Blophys.
Acta.  225: 96-107.

Reid,  J.B.   1977.  Benzo(a)pyrene absorption from the  Intestinal   tract  of
the  rat  with emphasis on  lymphatic  uptake  and  tissue distribution.   Ph.D.
Dissertation  submitted  to the Graduate  School  of  the  College  of  Medicine,
University of Cincinnati,  1n the Department of Environmental  Health.

Reid, W.D.,  K.F.  Ilett,  J.M.  Gllck  and G.  Krishna.   1973.   Metabolism and
binding of  aromatic  hydrocarbons 1n the  lung:  Relationship to  experimental
bronchlolar necrosis.   Am. Rev. Resplr.  D1s.   107: 539-551.

04460                                IX-68                            10/07/91

-------
Reshetyuk,  A.L.,  E.I.   Talaklna  and  P.A.  En'lakova.   1970.    Toxlcologlc
assessment  of  acenaphthene  and   acenaphthylene.    61g.   Tr.   Prof.   Zabol.
14(6): 46-47.  (Cited 1n U.S.  EPA,  1987b)

Revls, N.W., R.  Bull, D. Laurie and C.A. Schiller.   1984.   The  effectiveness
of  chemical  carcinogens  to  Induce  atherosclerosis  1n   the  white  Carneau
pigeon.  Toxicology.  32: 215-227.

Reznlkoff,  C.A.. J.S.  Bertram,  D.W.  Brankow  and  C.  Heldelberger.   1973.
Quantitative  and qualitative  studies  of  chemical  transformation of  cloned
C3H  mouse  embryo  cells  sensitive  to  postconfluence  Inhibition  of  cell
division.  Cancer Res.  33:  3239-3249.

Rh1m, J.S.,  O.K.  Park,  E.K. We1sburger and J.H. Welsburger.  1974.   Evalua-
tion of an In vitro assay system for  carcinogens  based on  prior Infection  of
rodent cells  with nontransformlng  RNA tumor virus.    J.  Natl.  Cancer  Inst.
52: 1167-1173.
      /
Rhoads,  C.P.,  W.E.  Smith,  N.S.  Cooper  and  R.D.   Sullivan.   1954.   Early
changes 1n  the skins  of several  species,  Including  man, after  painting with
carcinogenic materials.   Proc. Am.  Assoc.  Cancer Res.  1:  40.

R1ce,  J.E.,  T.J.  Hosted,  Jr.  and  E.J.   LaVole.    1984.   Fluoranthene  and
pyrene enhance benzo[a]pyrene — DMA adduct formation  In vivo 1n  mouse skin.
Cancer Lett.  24: 327-333.
04460                                IX-69                           10/07/91

-------
R1ce,  J.E.,  D.T. Coleman,  T.J.  Hosted, Jr.',  E.J.  LaVole,  D.J.  McCaustland
and  J.C.  Wiley, Jr.  1985a.   Identification  of  mutagenlc  metabolites  1n
Indeno-  [1,2,3-cd]pyrene formed  Ui vitro  with rat  liver enzymes.   Cancer
Res.  45:  5421-5425.

R1ce,  J.E.,  G.S.  Makowskl,   T.J.   Hosted,   Jr.  and  E.J.  LaVole.   1985b.
Methylene-brldged bay  region  chrysene  and  phenanthrene  derivatives  and their
keto-analogs:  Mutagenldty  1n  Salmonella  typhlmurlum  and  tumor-Initiating
activity on mouse skin.  Cancer Lett.  27:  199-206.

R1ce,  J.E.,  D.T. Coleman,  T.J.  Hosted,  E.J. LaVole,  D.J.  McCaustland  and
J.C. Wiley.   1985c.  On  the  metabolism, mutagenlcUy, and  tumor-Initiating
activity  of  Indeno  [1,2,3-cd]  pyrene.   In:  Polynuclear  Aromatic  Hydro-
carbons: Mechanism,  Methods and Metabolism,  M.  Cooke  and A.J.  Dennis,  Ed.
Battelle Press, Columbus, OH.   p. 1097-1109.

R1ce, J.E.,  T.J. Hosted, Jr., M.C.  DeFlorla, E.J.  LaVole, D.L.  Fischer  and
J.C.  Wiley,   Jr.    1986.    Tumor-Initiating  activity  of  major   .In.  vivo
metabolites  of  1ndeno[l,2,3-cd]pyrene  on  mouse  skin.   Carc1nogenes1s.   7:
1761-1764.

Rice, J.E.,  N.6.  Geddle  and E.J. LaVole.  1987a.  Identification of metabo-
lites  of  benzo(J)fluoranthene   formed  In  vitro In  rat  liver  homogenate.
Chem.-B1ol. Interact.  63: 227-237.

R1ce, J.E., E.H. Weyand, N.G. Geddle,  M.C. DeFlorla and E.J.  LaVole.  1987b.
Identification of tumorlgenlc  metabolites  of benzo(j)fluoranthene  formed  1n
vivo 1n mouse skin.   Cancer  Res.   47: 6166-6170.

04460                                IX-70                            11/21/91

-------
R1ce, J.E., E.H. Weyand, C. BurMll and E.J.-LaVo1e.  1990.  Fluorine probes
for  Investigating the mechanism of activation of 1ndeno(l,2,3-cd)pyrene to a
tumorlgenlc agent.   Carclnogenesls.   11(11):  1971-1974.

Rlegel,  B.,  W.B. Wartman,  W.T.  H111  et  al.   1951.   Delay  of methylchol-
anthrene  skin   cardnogenesls  In mice  by  1,2,5,6-dlbenzofluorene.   Cancer
Res.  11: 301-306.

Rlgdon,  R.H.  and  J. Neal.   1965.   Effects of  feeding  benzo[a]pyrene  on
fertility, embryos  and young mice.   J. Natl.  Cancer  Inst.   34: 297-305.

Rlgdon, R.H. and J.  Neal.   1966.  Gastric  carcinomas  and pulmonary adenoma
1n mice fed benzo[a]pyrene.   Tex.  B1ol. Med.   24: 195-207.

Rlgdon,  R.H.  and  J. Neal.   1969.   Relationship  of  leukemia  to  lung and
stomach tumors  1n mice  fed  benzo[a]pyrene.   Proc.  Soc.  Exp. Blol. NY.  130:
146-148.

Rlgdon,  R.H.  and  E.G.   Rennels.   1964.   Effect  of  feeding  benzpyrene  on
reproduction 1n the  rat.  Exper1ent1a.   20:  224-226.

Robinson, D.E. and A.D. Mitchell.  1981.  Unscheduled DNA  synthesis  response
of human flbroblasts, WI-38 cells, to 20 coded chemicals.  In:  Evaluation of
Short-term Tests for  Carcinogens.  Report of  the International  Collaborative
Program.   Progress  on  Mutation  Research,  Vol.  1,  F.J.  de  Serres  and  J.
Ashby, Ed.  Elsevler, North-Holland,  New  York.  p. 517-527.
04460                                IX-71                            11/21/91

-------
Robinson, J.R., J.S. Felton,  R.C.  Levitt,  S.S.  Thorgelrsson  and  D.W.  Nebert.
1975.   Relationship between  "aromatic hydrocarbon  responsiveness"  and  the
survival  times  In  mice  treated with  various  drugs  and environmental  com-
pounds.  Mol. Pharmacol.   11: 850-865.

Robinson, M., R. Laurie, R.  Bull and J.  Stober.  1987.   Carcinogenic  effects
1n A/J mice  of  partlculates  of  coal  tar  paint used 1n potable water  systems.
Cancer Lett.  34:  49-54.

Roe,   F.J.C.    1962.    Effect  of  phenanthrene  on   tumour-1nH1at1on   by
3,4-benzopyrene.  Br.  J.  Cancer.  16:  503-506.

Roe,  F.J.C.  and G.A.  Grant.   1964.   Tests of  pyrene  and  phenanthrene  for
Incomplete  carcinogenic  and antlcarclnogenlc   activity.   Br.  Emp.   Cancer
Campaign.  (Abstract)   (CUed In IARC,  1983)

Roe,  F.J.C.   and  M.A.  Waters.   1967.  Induction  of  hepatoma  In  mice  by
carcinogens  of  the polycycllc  hydrocarbon  type.    Nature  (London).   214:
299-300.

Roe,  F.J.C.,  L.S.  Levy and  R.L.   Carter.   1970.   Feeding studies on  sodium
cyclamate,   saccharin  and   sucrose  of  carcinogenic  and   tumour-promoting
activity.  Food Cosmet. Toxlcol.  8:  135-145.

Rogan,  E.G.*  R.U.  Roth,   P.A.   Katomskl-Beck,   J.R.   Laubscher  and  E.L.
Cavallerl.   1980.   Non-enzymat1c  ATP-med1ated   binding  of  hydroxymethyl
derivatives  of  aromatic hydrocarbons  to  DNA.   Chem.  B1ol. Interact.   31:
51-63.

04460                                IX-72                           10/07/91

-------
Rosenkrantz, H.S.  and  L.A. PolMer.  1979.   Evaluation  of  the mutagenldty
and  DNA-mod1fy1ng  activity of  carcinogens  and noncarclnogens  1n mlcroblal
systems.  J. Natl.  Cancer  Inst.   62:  873-892.

Rossi, L., 0. Barb1er1, M. Sangulnetl, A. Stacclone, L. Ferrerl Santl and L.
Santl.   1983.    Carcinogenic  activity  of  benzo[a]pyrene  and  some  of  Us
synthetic derivatives  by  direct  Injection  Into the mouse  fetus.  Cardno-
genesls.  4: 153-156.

Roszlnsky-Kocher,  G.,  A.   Basler  and G.  Rohrborn.  1979.   MutagenlcHy of
polycycllc  hydrocarbons.    V.  Induction of  sister-chromatid  exchanges  Jji
vivo.  Mutat. Res.   66: 65-67.   (Cited 1n U.S.  EPA,  1984c)

Rotenberg, lu.  S. and  F.D. Mashblts.  1965.  Toxlcologlc aspects  of acenaph-
thylene.  G1g.  Tr.  Prof. Zabol.   9(9): 53-54.   (Cited In U.S.  EPA, 1987b)

                                                                            •
Roussy,  G.,  M.   Guerln  and P.  Guerln.   1942.  Act1v1te comparee des trols
prlnclp'aux  hydrocarbures   synthetlques   cancerlgenes.   Bull.   Cancer.   30:
66-73.  (Fre.)

Rugen,  P.J., C.D.  Stern and S.H.  Lamm.   1989.  Comparative  carclnogenlcHy
of  the PAHs as  a  basis for  acceptable  exposure  levels (AELs) In drinking
water.  Reg. Toxlcol.  Pharmacol.   9(3):  273-283.

Russell,  H.  1947.   An unsuccessful  attempt  to Induce  gllomata  In  rabbits
with cholanthrene.   J.  Pathol.  Bacterlol.  59:  481-483.
04460                                IX-73                           11/21/91

-------
Saff1ott1,  U.,  F.  Cefls,  L.H.  Kolb  and  P.   Shublk.   1965.   Experimental
studies  of  the  conditions  of  exposure  to  carcinogens  for  lung  cancer
Induction.  J. A1r Pollut. Control Assoc.  15:  23-25.

Safflottl, U.. F. Cefls and  L.H.  Kolb.   1968.   A method for the experimental
Induction of bronchogenlc carcinoma.  Cancer  Res.  28:  104-124.

Safflottl,  U.,  R.   Hontesano,   A.R.  Sellakumar  and  D.G.  Kaufman.   1972.
Respiratory  tract  carclnogenesls  Induced   1n   hamsters  by  different  dose
levels  of benzo(a)pyrene  and  ferric  oxide.   J.  Natl.  Cancer  Inst.   49:
1199-1204.

Sakal,  H.,  D. Yoshlda  and  S.  M1zusak1.  1985.   MutagenlcUy  of  polycycllc
aromatic  hydrocarbons  and qulnones on  Salmonella typhlmurlum  TA97.   Hutat.
Res.  156: 61-67.

Salaman,  H.H.  and F.J.C.  Roe.    1956.   Further  tests  for  tumour-Initiating
activity:  N,N-d1-(2-chloroethyl)-p-am1nophenylbutyr1c  acid  (CB1348)  as  an
      /
Initiator  of   skin  tumour  formation  In the  mouse.    Br.  J.  Cancer.   10:
363-378.

Salamone,  M.F.   1981.   Tox1c1ty  of   41  carcinogens   and  noncardnogenlc
analogs.  Prog. Mutat.  Res.  1: 682-685.

Salamone, H.F.,  J.A.  Heddle and  M. Katz.   1979.  The mutagenlc activity  of
thirty  polycycllc aromatic  hydrocarbons (PAH)  and  oxides  1n  urban  airborne
participates.   Environ. Int.  2: 37-43.


04460                                IX-74                            10/07/91

-------
Sanders,  C.L.,  C. Skinner  and  R.A.  Gelman.  1984.   Percutaneous  absorption
of   [7.10-14C]benzo[a]pyrene   and   [7,12-1-»C]d1methylbenz[a]anthracene   1n
mice.  Environ. Res.   33: 353-360.

Sandhu, S.S.,  T.  Ma,  Y.  Peng and X. Zhou.   1989.   Clastogenldty  evaluation
of  seven  chemicals   commonly  found  at  hazardous   Industrial  waste  sites.
Mutat. Res.  224: 437-445.

Santodonato,  J.   1986.   Syracuse  Research  Center,  Syracuse,  NY.   Personal
communication to L. Fradkln, U.S. EPA,  Cincinnati,  OH, July 28.

Santodonato, J.,  0. Basu and P.M. Howard.   1980.  ^Multimedia human  exposure
and  carcinogenic  risk  assessment for  environmental  PAH.   ITJ:  Polynuclear
Aromatic  Hydrocarbons.   Chemical and Biological Effects,  M. Cooke  and  A.J.
Dennis, Ed.  Battelle Press, Columbus.  OH.  p. 435-454.

Santodonato,  J.,  P.   Howard  and  D.  Basu.  1981.    Health  and  ecological
assessment  of   polynuclear  aromatic   hydrocarbons.    3.   Environ.   Pathol.
      /
Toxdol.  5: 1-364.

Schafer,  W.B.   1951.   Acute hemolytlc  anemia related  to  naphthalene:  Report
of a case 1n a newborn Infant.   Pediatrics.  7:  172-174.

Schlede,  E.,  R.  Kuntzman and  A.H.  Conney.   1970.   Stimulatory  effect  of
benzo(a)pyrene  and phenobarbltal pretreatment  on   the biliary  excretion  of
benzo(o)pyrene metabolites In the rat.   Cancer Res.   30: 2898-2904.
04460                                IX-75                           10/07/91

-------
Schmahl,  0.   1955.   Examination  of the carcinogenic  action  of naphthalene
and anthracene 1n rats.  2.  Krebsforsch.  60:  697-710.

Schmahl,  0.  and Relter.  n.d.   Unpublished  research data.   (CHed  1n U.S.
EPA, 1984a)
                                  •
Schmahl,  D.,  K.G.  Schmidt  and H.  Habs.   1977.  Syncardnogenlc  action  of
polycycllc hydrocarbons  In  automobile  exhaust  gas   condensates.   IARC Sc1.
Publ. Iss.  16: 53-59.

Schmeltz,  I.,  J.  Tosk, J. Hllfrlch et  al.   1978.   Bloassays  of naphthalene
and  alkylnaphthalenes  for  cocardnogenlc  activity.   Relation to  tobacco
carclnogenesls.  in:   Carc1nogenes1s  — A  comprehensive  survey.    Vol.  3,
Polynuclear  Aromatic   Hydrocarbons,  P.W.  Jones and  R.I.  Freudenthal,  Ed.
Raven Press,  New York, NY.  p.  47-60.   (CHed  1n U.S. EPA,  1987a)

Schoeny,  R.   and  D.   Warshawsky.   1983.   MutagenlcHy   of  benzo(a)pyrene
metabolites generated  on the  Isolated  perfused lung  following participate
exposure.  Teratogen.  Carcinogen.  Mutagen.  3: 151-162.

Schoeny,  R.,  T.  Cody, M. Radlke and  D.  Warshawsky.   1985.  MutagenlcHy of
algal  metabolites  of  benzo(a)pyrene  for  Salmonella  typhlmurlum.  Environ.
Mutagen.  7:  839-855.

Schottenfeld,  D. and J.F. Fraumenl, Jr., Ed.   1982.   Cancer Epidemiology and
Prevention.  U.B. Saunders Co., Philadelphia,  PA.
04460                                IX-76                           10/07/91

-------
Scrlbner,  J.D.   1973.   Brief communication:  Tumor  Initiation by apparently


noncardnogenlc  polycycllc  aromatic  hydrocarbons.    J.  Natl.  Cancer  Inst.


50: 1717-1719.





Selxas,  G.M.,  B.M. Andon,  P.G.  Holllngshead  and  W.G.  Thllly.   1982.   The


aza-arenes  as  mutagens  for   Salmonella  typh1mur1um.   Mutat.   Res.    102:
                                  •

201-212.





Sellakumar,  A.  and P.  Shublk.   1972.   Carclnogenlclty  of  7H-d1benzo(c,g)-


carbazole  1n  the  respiratory  tract  of hamsters.  J. Natl. Cancer Inst.   48:
                                          »

1641-1646.





Sellakumar,   A.   and   P.   Shublk.    1974.    Carclnogenlclty  of   different


polycycllc  hydrocarbons  1n  the respiratory  tract  of  hamsters.   J.  Natl.


Cancer Inst.  53: 1713-1719.





Seta'la, K.   1954.  Experimental chemical cardnogenesls  and  the  Influence of


solvents.  Nature (London)   174:  873-875.
      /




Setaia*. K. and P.  Ekwall.   1950.  Penetration of benzpyrene  Into the  stomach


wall of mouse.  Science.  112: 229-231.   (Cited  1n  Ermala et  al., 1951)





Shabad, L.H.  1938.  Experimental tumours 1n guinea pigs.  Arch.  B1ol.  Nauk.


51: 112-119.
04460                                IX-77                           10/07/91

-------
Shabad,  L.M.  and J.P.  Urlnson.   1938.   Alterations  In  the liver of  guinea
pigs,  following  administration  of a chemically pure  carcinogenic  substance,
1,2,5,6-dlbenzanthracene.  Arch.  B1ol.  Nauk.   51: 105-111.  (CHed  1n  IARC,
1973)

Shabad,  L.H.,  J.D.  Soroklna, N.I.  Golub and  S.P.  Bogovskl.  1972.   Trans-
placental effect  of  some chemical compounds  on  organ cultures of  embryonic
kidney tissue.  Cancer Res.   32:  617-627.

Shackelford,  W.M.  and  L.H.  Keith.   1976.   Frequency of  organic  compounds
Identified  1n water.   U.S.  Environmental  Protection Agency, Environmental
Research Laboratory,  Athens,  GA.   EPA-600/4-76-062.   NTIS PB-265470/5.

Shank, R.C.,  L.P.  Barrows  and A.R. Buckpltt.  1980.  Comparative  metabolism
of  hydrazlne  and  naphthalene.   A1r Force Aerospace  Research   Laboratory,
Wright-Patterson A1r  Force Base,  OH.   AFAMRL-TR-80-103.   NTIS AD-A092710/3.

Shear, M.3.   1938.   Studies 1n  cardnogenesls.   V. Methyl derivatives of
1,2-benzanthracene.   Am. 0.  Cancer.  33: 499-537.

Shear, M.G. and  J. Lelter.   1941.  Studies 1n cardnogenesls.  XVI. Produc-
tion  of  subcutaneous  tumors  In  mice  by miscellaneous polycycllc  compounds.
J. Natl.  Cancer Inst.   2: 241-258.

Sher,  S.P.    1974.   Review  article.    Tumors 1n control  mice:   Literature
tabulation.  Toxlcol.  Appl.  Pharmacol.   30:  337-359.
04460                                IX-78                           10/07/91

-------
Sh1ch1, H.  and  D.W.  Nebert.  1982.   Genetic  differences  In drug metabolism
associated with ocular toxUHy.   Environ.  Health  Perspect.   44:  107-117.

Sh1ch1, H., H.  Tanaka,  N.M.  Jensen and D.W. Nebert.  1980.  Genetic differ-
ences  1n cataract and  other  ocular abnormalities  Induced by paracetamol and
naphthalene.  Pharmacology.   20:  229-241.

Sh1mk1n, M.B. and  G.O.  Stoner.   1975.  Lung  tumors In mice: Application  to
carclnogenesls bloassay.  In.: Advances In Cancer Research,  G.  Klein and  S.
Welnhouse, Ed.  Raven Press,  NY.   12:  1-58.

Shopp,  G.H.,  K.L. White,  Jr.,  M.P.   Holsapple  et  al.   1984.   Naphthalene
toxldty In CD-I mice:  General toxicology and  Immunotoxlcology.  Fund.  Appl.
Toxlcol.  4: 406-419.

Shublk, P.  and  J.L.  Hartwell.   1957.  Survey of compounds which have  been
tested  for  carcinogenic  activity.   Public  Health  Service Publ.  No.   149:
Suppl. ;,  GPO, Washington,  DC.   (Cited 1n  IARC, 1973)

Shublk, P.  and  J.L.  Hartwell.   1969.  Survey of compounds which have  been
tested  for  carcinogenic  activity.   Public  Health  Service Publ.  No.   149:
Suppl. 1,  GPO, Washington,  DC.   (Cited 1n  IARC, 1973)

Shublk,  P.,  G.  Pletra  and  G.  Delia  Porta.    1960.    Studies   of   skin
carclnogenesls In the Syrian golden hamster.   Cancer Res.   20: 100-105.
04460                                IX-79                           11/21/91

-------
S1em1atyck1, J., M. Gerln,  P.  Stewart,  L.  Nadon,  R.  Dewar  and  L.  Richardson.
1988.  Associations between several sites  of cancer  and  ten  types of exhaust
and combustion products.  Scand.  J. Work Environ.  Health.   14:  79-90.

Simmon,  V.F.   1979a.   l
-------
Slaga, T.,  L.  Jecker,  W. Bracken and C.  Weeks.   1979.   The effects of weak
or non-carcinogenic  polycycllc  hydrocarbons  on  DMBA and benzo(a)pyrene skin
tumor Initiation.  Cancer Lett.   7:  51-59.

Slaga, T.J.,  G.L.  Gleason,  6.  Hills  et al.  1980.   Comparison  of the skin
tumor-1n1t1at1ng  activities  of  dlhydrodlols and  d1ol-epox1des  of various
polycycllc aromatic hydrocarbons.   Cancer  Res.   40:  1981-1984.

Smith, I.A., G.D. Berger, P.G. Seybold and M.P. Serve.   1978.  Relationships
between  cardnogenlcHy  and  theoretical   reactivity  Indices  In  polycycllc
aromatic hydrocarbons.   Cancer  Res.   38: 2968-2977.

Smyth,  H.F.,  C.P.  Carpenter,  C.S.  Well,  U.C.  Pozzanl  and  J.A.  StMegel.
1962.   Range-finding toxlclty  data:  List  VI.   Ind. Hyg.  Assoc.  J.    23:
95-107.

Snell,  K.C.  and  H.L.  Stewart.   1962a.   Pulmonary adenomatosls  Induced   1n
      /
DBA/2 mice by oral administration of d1benz[a,h]anthracene.  J.  Natl. Cancer
Inst.  28: 1043-1049.

Snell, K.C.  and H.L. Stewart.   1962b.   Induction of pulmonary  adenomatosls
1n DBA/2 mice  by  the oral administration  of d1benz[a,h]anthracene.  Acta.
Un. Int. Cancer.  19: 692-694.

Snyder,  R.D.  and D.W.  Matheson.  1985.   Nick  translation  - A new  assay  for
monitoring  DNA  damage and  repair In cultured  human  flbroblasts.   Environ.
Mutagen.  7: 267-279.


04460                                IX-81                            11/21/91

-------
Sorg,  R.M.,  R.W.  Nalsmlth and R.G. Matthews.  1985. Mlcronucleus  test  (MNT)
OECD  (unpublished material).   Pharmakon  Research  International  Inc.,  Waverly
PA.  Submitted to Texaco, Inc. Beacon, NY.  PH309A-TX-007-85.   5601-56-1.

Stanton, H.F.,  E.  Miller, C.  Wrench  and R. Blackwell.   1972.   Experimental
Induction  of  epldermold carcinoma  In the  lungs  of  rats by cigarette  smoke
condensate.  J. Natl. Cancer Inst.  49(3): 867-877.

Stelner, P.E.   1955.   Carclnogenldty  of multiple chemicals  simultaneously
administered.  Cancer Res.  15: 632-635.

Stelner, P.E.  and J.H.  Edgcomb.   1952.   Carclnogenldty of 1,2-benzanthra-
cene.  Cancer Res.  12: 657-659.

Stelner, P.E.  and H.L.  Falk.  1951.   Summation  and  Inhibition effects  of
weak  and  strong carcinogenic  hydrocarbons:  1,2-Benzanthracene,   chrysene,
1,2,5,6-dlbenzanthracene and 20-methylcholanthrene.   Cancer  Res.  11:  56-63.
      /

Stenback,  F.  and  J.  Rowland.   1978.   Role  of  talc  and benzo(a)pyrene  1n
respiratory tumor formation.   Scand. J.  Resp.  D1s.   59:  130-140.

StHeler,  U.  and  R.  Schoeny.    1991.   A  statistical  test  for  evaluating
compatibility of  two  or more  studies  In cancer  risk assessment.   Presented
at  the Society  of  Toxicology  Annual Meeting,  February 25-March  1,  1991,
Dallas, TX.
04460                                IX-82                          11/21/91

-------
Stjernsward,   J.    1966.    Effect   of   noncardnogenlc   and   carcinogenic
hydrocarbons  on  antibody-forming cells  measured  at  the  cellular  level  j£
vitro.  J. Natl.  Cancer Inst.  36:  1189-1195.

Stjernsward,   J.    1969.    Immunosuppresslon   by  carcinogens.    Ant1b1ot.
Chemother.  15: 213-233.  (Cited 1n  IARC,  1973)

Stuermer,  D.H.,  D.J  Ng and  C.J.  Morris.   1982.  Organic  contaminants  1n
groundwater  near   an   underground  coal  gasification  site  In  northeastern
Wyoming. Environ.  Sd. Technol.  16:  582-587.

Sullivan,  P.O., L.M.  Calle,  K.  Shafer and M.  Nettleman.   1978.   Effect  of
antloxldants  of   benzo(a)pyrene free  radicals.   In.:  Polynuclear  Aromatic
Hydrocarbons:  2nd  Int.  Symp.  on Analysis,  Chemistry, and Biology  (Carclno-
genesls -  A  Comprehensive  Survey:  Vol. 3), R.I.  Freudenthal and  P.M.  Jones,
Ed.  Raven Press,  New York.  3: 1-8.
      /
Sun, J.D., R.K. Wolff.  G.M.  KanapHly  and  R.O.  McClellan.  1983.  Effects  of
particle association  on the  biological  fate  of  Inhaled organic  pollutants.
In:  Polynuclear  Aromatic  Hydrocarbons:  Formation, Metabolism and Measure-
ment,  M.  Cooke   and  A.J.  Dennis,   Ed.,   7th   Int.  Symp.   Battelle   Press,
Columbus, OH.

Sun,  J.D..  R.K.  Wolff,  G.M.  Kanapllly  and  R.O.  McClellan.   1984.   Lung
retention  and metabolic  fate of   Inhaled  benzo(a)pyrene  associated  with
dleseT exhaust particles.  Toxlcol.  Appl.  Pharmacol.   73:  48-59.
04460                                IX-83                           11/21/91

-------
Suntzeff,  V.,  A.B.  Cronlnger,  E.L.  Hynder,  E.V.  Cowdry  and  E.A.  Graham.
1957.  Use of  sebaceous-gland  test  of  primary  c1garette-tar  fractions and  of
certain noncardnogenlc polycycllc hydrocarbons.   Cancer.  10: 250-254.

Swartz, W.J. and  D.R.  Hattlson.   1985.   Benzo(a)pyrene  Inhibits  ovulatlon  1n
C57BL/6N mice.   Anatom.  Rec.  212:  268-276.

Taklzawa, K.,  H.  Yag1, D.H. Jerlna  and  D.R. Matt 1 son. 1984.  Hurlne  strain
differences  1n ovotoxldty  following  IntraovaHan  Injection with  benzo(a)-
pyrene,   (+)-(7R,8S)-ox1de,   (-)-(7R,8R)-d1hydrod1ol,   or   (+)-{7R,8S)   -
d1ol-(9S, !OR)-epox1de-2.   Cancer Res.   44:  2571-2576.

Tawflc,  H.N.   1965.   Studies  on ear-duct  tumors 1n  rats.   II.  Inhibitory
effect of  methylcholanthrene  and  1,2-benzanthracene  on  tumor  formation  by
4-d1methyl-am1no-st1lbene.   Acta Pathol. Jap.   15(2):  255-260.   (Cited  1n
IARC, 1973)

Teel, R.W.   1984.   A  comparison  of  the effect of selenium  on the  mutagenlc-
      /•
Uy and metabolism  of  benzo[a]pyrene 1n  rat and  hamster  liver S9  activation
systems.   Cancer  Lett.   24: 281-289.

Thakker,  D.R., H. Yagl,  W.  Levin,  A.W.  Wood, A.H. Conney  and D.H.  Jerlna.
1985.  Polycycllc aromatic  hydrocarbons:  Metabolic activation  to  ultimate
carcinogens.    Ln:  B1oact1vat1on of   Foreign  Compounds,  M.U.   Anders,  Ed.
Academic  Press, NY.   p. 177-242.
04460                                IX-84                           10/07/91

-------
Thakker,  D.R.,  W.  Levin,  A.W.  Wood,  A.M. Conney, H.  Yag1  and D.M.  Jerlna.
1988.     Stereoselectlve    blotransformatlon    of   polycycllcal   aromatic
hydrocarbons to  ultimate  carcinogens.   ITK  Drug  Stereochemistry: Analytical
Methods and  Pharmacology,  I.W.  Walner and D.E.  Drayer,  Ed.  Marcel  Dekker,
Inc., New York.  p. 271-296.

Thompson, J.I.  and Co.   1971.   Survey of compounds  which  have been  tested
for  carcinogenic   activity.      Public  Health   Service   Publ.  No.   149:
1968-1969.  GPO, Washington, DC.  (CHed In  IARC,  1973)

Thorslund, T.W., C.C.  Brown and 6. Charnley.  1987.  Biologically motivated
cancer risk models.  Risk Anal.   7: 109-119.

Thyssen, J., J. Althoff, G. Klmmerle and  U. Mohr.  1981.  Inhalation  studies
with benzo[a]pyrene  1n  Syrian golden hamsters.   J.  Natl.  Cancer Inst.   66:
575-577.

Toklwa,  H.,  K. MorHa,  H. Takeyoshl,  K. Takahashl  and  Y. Ohnlshl.   1977.
     /
Detection of mutagenlc  activity In partlculate air pollutants.  Mutat.  Res.
48: 237-248.

Tong,  C.,  M-.F. Laspla,  S. Telang  and G.M.  Williams.   1981a.  The  use  of
adult  rat  liver cultures  In the  detection  of the  genotoxldty of  various
polycycllc aromatic hydrocarbons.  Environ.  Mutagenesls.   3:  477-487.

Tong,  C.,  S.V. Brat  and  G.M.  Williams.   1981b.   S1ster-chromat1d exchange
Induction by  polycycllc aromatic  hydrocarbons  1n an Intact  cell  system  of
adult rat-liver epithelial cells.  Mutat.  Res.   91: 467-473.

04460                                IX-85                          10/07/91

-------
long,  S.S.,  Y. Hlrokata,  M.A.  Trush et  al.   1981c.   Clara cell  damage  and
Inhibition  of  pulmonary  mixed-function  oxldase  activity  by  naphthalene.
Blochem. Blophys. Res. Commun.  100: 944-950.

Tong,  S.S.,  M.C.  Lowe, M.A.  Trush  et  al.   1982.   Bronchlolar  epithelial
damage and Impairment  of pulmonary  mlcrosomal  monooxygenase  activity In mice
                                  •
by naphthalene.  Exp. Mol.  Pathol.  37: 358-369.

Torronen, R.,  U.  Nouslalnen and  0.  Hannlnen.   1981.   Induction  of  aldehyde
dehydrogenase  by  polycycllc  aromatic  hydrocarbons  1n   rats.   Chem.  Blol.
Interact.  36: 33-44.

Toth, B. and P. Shublk.  1967.  Cardnogenesls  1n  AKR mice Injected at birth
with benzo(a)pyrene and d1methyln1trosam1ne.  Cancer Res.  27:  43-51.

Toth,  B.,  L.  Mallck  and  H.  Shlmlzu.   1976.   Production of  Intestinal  and
other  tumors  by  !,2-d1methylhydraz1ne  and  dlhydrochlorlde 1n  mice.  I.  A
light microscope study of colonlc  neoplasms.  Am.  J.  Pathol.   84: 69-86.

Toxicology Data Bank [Data base],   n.d.   U.S.  Department of  Health and Human
Services, National Library of Medicine, Bethesda,  MD.   TDB no.  0184.

Tracor-Jltco.   1973a.    Survey  of   compounds   which  have  been  tested  for
carcinogenic   activity.  Public   Health  Service  Publ.   No.  149:  1961-1967.
GPO, Washington. DC.  (Cited 1n  IARC, 1973)
04460                                IX-86                           10/07/91

-------
Tracor-JHco.   1973b.   Survey  of  compounds  which  have  been  tested  for
carcinogenic  activity.    Public  Health  Service  Publ.  No.  149:  1970-1971.
GPO, Washington. DC.  (Cited 1n IARC,  1973)

TrVolo, A.J.,  6.E.  Aponte  and O.L.  Herr.   1977.   Induction of aryl  hydro-
carbon  hydroxlase   and  forestomach  tumors  by  benzo[a]pyrene.   Cancer  Res.
37: 3018-3021.

Tsuda, H.,  G.  Lee  and E. Farber.  1980.  Induction  of  resistant hepatocytes
as  a  new  principle for  a  possible  short-term in vivo  test  for  carcinogens.
Cancer Res.  40: 1157-1164.

Tu,  A.,  W.  Ha "Mo well,  S.  Pallotta  et  al.    1986.    An   Inter-laboratory
comparison  of  transformation  1n  Syrian hamster  embryo  cells with model  and
coded chemicals.  Environ. Hutagen.   8: 77-98.

Turusov,   V.S.,   T.V.   Nlkonova  and  Y.u.D.   Parenov.   1990.    Increased
multiplicity  of lung  adenomas  1n  five  generations of  mice  treated  with
      /
benzo(a)pyrene when pregnant.   Cancer  Lett.   55:  227-231.

Tyrer,  H.W..   E.T.  Cantrell,   R.  Horres,  I.P.  lee,  W.B. Pelrano and  R.M.
Banner.  1981.  Benzo(a)pyrene metabolism In  mice exposed  to dlesel  exhaust:
1. Uptake and distribution.  Environ.  Int.  5: 307-311.

Union Carbide  Corporation.  1968.  Naphthalene safety data sheet.   (Cited In
U.S. EPA,   1980c)
04460                                IX-87                           10/07/91

-------
Urso, P.  and  N.  Gengozlan.  1980.  Depressed humoral  Immunity  and Increased
tumor Incidence  In mice following ^n  utero exposure  to  benzo(a)pyrene.   J.
Toxlcol. Environ. Health.  6: 569-576.

Urso, P. and  N.  Gengozlan.   1984.  Subnormal  expression  of cell-mediated and
humoral   Immune  responses  In progeny  disposed  toward a  high   Incidence  of
tumors  after  in  utero  exposure  to benzo[a]pyrene.   J.  Toxlcol.   Environ.
Health.   14: 569-584.

U.S. EPA.   1980a.  Ambient Water  Quality  Criteria  Document for  Acenaphthene.
Prepared by the  Office  of  Health  and Environmental Assessment,  Environmental
Criteria and Assessment  Office, Cincinnati, OH  for the Office of Water Regu-
lations  and Standards, Washington. DC.   EPA 440/5-80-015.  NTIS PB 81-117268.

U.S. EPA.   1980b.  Ambient Water  Quality  Criteria  Document for  Fluoranthene.
Prepared by the  Office  of  Health  and Environmental Assessment,  Environmental
Criteria and Assessment  Office, Cincinnati, OH  for the Office of Water Regu-
lations  and Standards, Washington, DC.   EPA 440/5-80-049.  NTIS PB 81-117608.

U.S. EPA.   1980c.   Ambient Water Quality Criteria Document  for Naphthalene.
Prepared by the  Office  of  Health  and Environmental Assessment,  Environmental
Criteria and Assessment  Office, Cincinnati, OH  for the Office of Water Regu-
lations  and Standards, Washington. DC.   EPA 440/5-80-059.  NTIS PB 81-117707.
04460                                IX-88                           10/07/91

-------
U.S.  EPA.   1980d.  Ambient  Water  Quality Criteria Document  for  Polynuclear
Aromatic  Hydrocarbons.   Prepared by  the  Office  of Health  and  Environmental
Assessment, Environmental Criteria and  Assessment  Office,  Cincinnati,  OH for
the   Office  of   Water   Regulations   and   Standards,   Washington,   DC.    EPA
440/5-80-069.   NTIS PB 81-117806.

U.S.  EPA.  1982.  Errata:  PAH  Ambient  Water  Quality  Criterion  for  the
Protection of  Human  Health.   Prepared  by  the Office  of Health  and Environ-
mental Assessment, Environmental Criteria  and Assessment  Office,  Cincinnati,
OH for the Office of Water Regulations and Standards,  Washington,  DC.

U.S.  EPA.   1984a.  Health Effects  Assessment  for  Polycycllc  Aromatic  Hydro-
carbons  (PAH).   Prepared by  the Office of Health and  Environmental  Assess-
ment,  Environmental  Criteria and Assessment  Office,  Cincinnati, OH for the
Office of Emergency and Remedial Response, Washington, DC.   EPA 540/1-86/013.
NTIS PB86-134244/AS.

U.S.  EPA.  1984b.  Carcinogen  Assessment of Coke  Oven  Emissions.   Office of
Health' and Environmental  Assessment,  Washington, DC.   EPA  600/6-82-003F.
NTIS PB84-170182.

U.S.  EPA.   1984c.  Health  and  Environmental  Effects Profile  for  Chrysene.
Prepared by the  Office  of Health and Environmental Assessment,  Environmental
Criteria and Assessment  Office,  Cincinnati, OH for the Office of Solid Waste
and    Emergency    Response,    Washington,    DC.     EPA/600/X-84/186.     NTIS
PB88-131123/AS.
04460                                IX-89                           10/07/91

-------
U.S.  EPA.   1984d.    Health  Assessment  Document  for  Carbon  TetrachloMde.
Prepared  by  the  Office  of  Health and Environmental Assessment,  Environmental
Criteria and Assessment Office, Cincinnati, OH.  EPA 600/8/82-001F.

U.S.  EPA.   1985.    The Carcinogen  Assessment  Group's   Calculation  of  the
Cardnogenldty  of  Dlcofol  (Kelthane),  DDT,  DDE and ODD  (TOE).   Prepared  by
the  Office  of   Health  and Environmental  Assessment,  Carcinogen  Assessment
Group,  Washington,   DC for   the   Hazard   Evaluation   Division,  Office  of
Pesticide  Programs,  Office of  Pesticides  and Toxic Substances, Washington,
DC.

U.S.  EPA.   1986.   Guidelines  for  Carcinogen  Risk   Assessment.    Federal
Register.  51(185):  33992-34003.

U.S.  EPA.   1987a.   Summary   Review   of   Health   Effects   Associated   with
Naphthalene:  Health  Issue Assessment.  Office  of Health  and  Environmental
Assessment,  Environmental  Criteria  and Assessment  Office,  Research  Triangle
Park, NC.  EPA/600/8-87/055F.   NTIS PB88-172374.
      /

U.S.  EPA.   1987b.   Health Effects  Assessment for  Acenaphthylene.   Prepared
by the Office of Health and Environmental  Assessment,  Environmental  Criteria
and  Assessment   Office,  Cincinnati, OH  for  the  Office  of  Solid Waste  and
Emergency Response,  Washington, DC.   EPA/600/8-88/011.   NTIS PB88-179510.

U.S.  EPA.   1987c.   Interim Procedures  for Estimating  Risks Associated  with
Exposures  to Mixture'  of  Chlorinated   D1benzo-£-d1ox1ns  and  -dlbenzofurans
(CDDs and CDFs).   Risk Assessment Forum.  EPA/625/3-87/012.  NTIS PB89-125041.


04460                                IX-90                            10/07/91

-------
U.S.  EPA.   1988.   13-Week  House  Oral  Subchronlc  Toxldty  Study  with
Fluoranthene.  Prepared  by Toxldty Research  Laboratories,  LTD.,  Muskegon,
MI for the Office of Solid Waste,  Washington, DC.

U.S.  EPA.    1989a.   Mouse  Oral  Subchronlc  Toxldty  Study  with  Fluorene.
Prepared  by Toxldty  Research  Laboratories,  LTD.,  Muskegon,  MI  for  the
Office of Solid Waste,  Washington,  DC.

U.S.  EPA.   1989b.   Mouse  Oral  Subchronlc  Toxldty  of  Pyrene.   Study
conducted by Toxldty Research  Laboratories, Muskegon, MI  for  the  Office of
Solid Waste, Washington,  DC.

U.S.  EPA.    1989c.   October  RfD  Workgroup  Meeting.   -Memorandum  from Mark
Greenberg to Bill Pepelko, U.S.  EPA.  October 17.

U.S.  EPA.   1989d.   Interim Procedures for  Estimating  Risks Associated with
Exposure  to Mixtures  of  Chlorinated  D1benzo-p-d1ox1ns  and  -dlbenzofurans
      /
(CDDs and CDFs) and 1989  Update.  Risk  Assessment  Forum.  EPA/625/3-89/016.

U.S.  EPA.    1989e.   Subchronlc  Toxldty  1n  Mice  with  Anthracene.   Final
Report.   Hazel ton  Laboratories America,  Inc.   Prepared  for  the  Office of
Solid Waste, Washington,  DC.

U.S.  EPA.    1991a.   Integrated Risk  Information  System   (IRIS).   Online.
Office  of  Health and  Environmental  Assessment,   Environmental  Criteria  and
Assessment Office,  Cincinnati,  OH.
04460                                IX-91                           11/21/91

-------
U.S.  EPA.   1991b.   Quantitative Risk Assessment of  Ingested  Benzo[a]pyrene.
Human   Health   Assessment   Group,   Office  of   Health   and   Environmental
Assessment, Washington, DC.  Final  Draft.

Utesch, D., H. Glatt and  F.  Oesch.   1987.   Rat  hepatocyte-medlated  bacterial
mutagenldty 1n  relation  to  the  carcinogenic  potency of  benz(a)anthracene,
benzo(a)pyrene  and  twenty-five  methylated  derivatives.   Cancer  Res.   47:
1509-1515.

Valaes, T.,  S.A.  Dox1ad1s and  P.   Fessas.  1963.   Acute hemolysls  due  to
naphthalene Inhalation.  J. Pedlatr.  (St  Louis).   63:  904-916.

Valencia,  R. and  K.  Houtchens.  1981.  Mutagenlc  activity of 10 coded  com-
pounds  1n  the  DrosophUa  sex-Hnked  recessive  lethal test.   In.: Evaluation
of Short-term Tests  for Carcinogens.  Report of the  International Collabora-
tive Program.  Progress 1n  Mutation  Research,  Vol.  1, F.J. de Serres and  J.
Ashby, Ed.  Elsevler, North Holland,  NY.   p.  617-659.

Valencia,   R.,  3.M.  Mason  and S.  Zlmmerlng.   1989.   Chemical mutagenesls
testing 1n Drosophlla.  VI.  Interlaboratory comparison of  mutagenldty tests
after treatment of larvae.  Environ.  Molec. Mutagen.   14: 238-244.

Van  DQQren,  B.L.  and  B.M.  Goldschmldt.   1976.   Cocardnogenlc  and tumor-
promoting  agents   1n  tobacco  carclnogenesls.   J.  Natl.   Cancer  Inst.   56:
1237-1242.
04460                                IX-92                           10/07/91

-------
Van DQuren, B.L.,  A.  Slvak,  A.  Segal,  L. Orris and L. Langseth.  1966.  The
tumor-promoting agents  of  tobacco  leaf and  tobacco smoke  condensate.   J.
Natl.  Cancer Inst.  37:  519-526.

Van Duuren,  B.L., L. Langseth,  B.M.  Goldschmldt et  al.  1967.  Cardnogen-
1cUy  of  epoxldes,   lactones  and  peroxy  compounds.   VI.    Structure  and
carcinogenic activity.  J. Natl.  Cancer  Inst.   21: 623.

Van Duuren, B.L.,  C. Katz and B.M. Goldschmldt.  1973.  Brief communication:
Cocarc1nogen1c agents In  tobacco carclnogenesls.  J. Natl. Cancer Inst.  51:
703-705.

van Heynlngen, R.  1970a.   Ascorbic add 1n the lens of  the naphthalene-fed
rabbit.  Exp.  Eye  Res.  9: 38-48.

van Heynlngen,  R.  1970b.  Effect of  some cyclic  hydroxy  compounds  on the
accumulation of ascorbic  add by the  rabbit  lens ±n vitro.   Exp.  Eye Res.
      /
9: 49-56.

van  Heynlngen.  R.   1979.   Naphthalene cataract   1n  rats  and rabbits:  A
resume.  Exp.  Eye  Res.  28:  435-439.

van Heynlngen,.  R. and A.  P1r1e.   1967.  The  metabolism  of naphthalene and
Us toxic effect on the  eye.  Blochem.  J.  102:  842-852.

Vater,  S.  and R.  Schoeny.   1991.  Biological  considerations  for  combining
carcinogenic quantitative estimates.   Presented at the Society of Toxicology
Annual Meeting, February 25-March 1, 1991,  Dallas, TX.

04460                               IX-93                  .         11/21/91

-------
Ved  Brat,  S., C.  long,  S. Telang  and  6.M.  Williams.  1983.   Comparison  of
sister  chromatld  exchange  and  mammalian  cell  mutagenesls  at  the hypoxanthlne
guanlne  phosphorlbosyl  transferase  locus   1n  adult  rat  liver  epithelial
cells.  Ann. N.Y. Acad. Sd. 407: 474-475.

von  der  Hude.  W., C.  Behm,  R.  Gurtler  and A. Basler.  1988.   Evaluation  of
the SOS chromotest.  Mutat. Res.  203:.81-94.

Wald, N., S. Howard, D.G.  Smith  and K.  Kjeldsen.   1973.   Association between
atherosclerotic  diseases  and carboxyhaemoglobln  levels  1n tobacco  smokers.
Br. Hed. 0.  1: 761-765.

Wangenhelm,  J.  and  G.  Bolcsfoldl.   1988.   House lymphoma L5178Y  thymldlne
klnase locus assay of 50 compounds.   Mutagenesls.   3:  193-205.

Warren, D.L.,  D.L. Brown,  Jr.  and A.R.  Buckpltt.  1982.   Evidence  for  cyto-
chrome  P-450  mediated  metabolism 1n  the bronchlolar  damage by naphthalene.
Chem. B1ol. Interact.  40:  287-303.
      /

Warshawsky,  D.,  R.  Nlemeler  and  E. Blngham.   1981.  Influence  of  sulfur
dioxide  on  metabolism  and  distribution   of  benzo(a)pyrene   In   Isolated
perfused rabbit lung.  0.  Toxlcol. Environ.  Health.  7:  1001-1024.

Warshawsky,  0..  E.  Blngham  and R.W.  Nlemeler.   1984.   The  effects  of  a
cocardnogen,  ferric  oxide,  on  the metabolism of  benzo(a)pyrene  In  the
Isolated perfused lung.  J. Toxlcol. Environ. Health.   14:  191-209.
04460                                IX-94                           10/07/91

-------
Wattenberg,  L.W.   1972.   Inhibition  of  carcinogenic  and  toxic  effects  of
polycycllc  hydrocarbons  by phenolic antloxldants  and  ethoxyquln.  J.  Natl.
Cancer Inst.  48:  1425-1430.

Wattenberg,  L.W.   1974.   Inhibition  of  carcinogenic  and  toxic  effects  of
polycycllc  hydrocarbons  by  several  sulfur-containing  compounds.   J.  Natl.
Cancer Inst.  52:  1583-1587.

Wattenberg,  L.W.  and  J.L. Leong.    1970.   Inhibition  of the carcinogenic
action of benzo(a)pyrene by flavones.   Cancer Res.   30:  1922-1925.

We1nste1n,  I.B.,  A.M.  Jeffrey, K.W.  Jennette  et al.  1976.   Benzo[a]pyrene
dlol epoxldes as Intermediates 1n nucleic add binding  \n_ vitro and \n_ vivo.
Science.  193: 592-595.

Weston,  A.,  R.M.   Hodgson, A.J.  Hewer,  R.  Kuroda and  P.L.  Grover.   1985.
Comparative  studies  of the metabolic  activation of chrysene  1n  rodent  and
human skin.  Chem-B1ol. Interact.   54:  223-242.
      /

Weyand E.H.,  N. Geddle,  J. R1ce, A.  Czech,  S.  Am1n and  E.J.  LaVole.   1988.
Metabolism  and  mutagenlc  activity  of benzo[k]fluoranthene and  3-, 8-,  and
9-fluorobenzo[k]fluoranthene.   Carclnogenesls.   9:  1277-1281.

Weyand,  E.H.,  S.   Patel,   E.J.   LaVole,  B.  Cho  and  R.G.  Harvey.    1990.
Relative  tumor  Initiating  activity  of benzo(a)fluoranthene,  benzo[b]fluor-
anthene,  naphtho[l,2-b]fluoranthene and  naphtho[2,l-a]fluoranthene on  mouse
skin.  Cancer Lett.  52:  229-233.


04460                                IX-95                           10/07/91

-------
White,  F.R.  and  A.B.  Eschenbrenner.   1945.   Note  on   the  occurrence  of
hepatomas  1n  rats following the  1ngest1on  of  1,2-benzoanthracene.   J.  Natl.
Cancer Inst.  6: 19-21.

White, K.L.,  Jr.  and M.P. Holsapple.  1984.   Direct  suppression  of In-vitro
                                  •                                  ^™^~ ^^^_»
antibody  production  by mouse  spleen cells by  the  carcinogen  benzo(a)pyrene
but not by  the  noncardnogenlc congener  benzo(e)pyrene.   Cancer  Res.  44(8):
3388-3393.

White,  J.  and  A.  White.    1939.   Inhibition  of  growth of 'the  rat  by  oral
administration  of  methylcholanthrene,  benzpyrene, or  pyrene, and  the effects
of various dietary supplements.  J.  Blol. Chem.  131: 149-161.

White, K.L.,  Jr.,  H.H.  Lysy  and  M.P.  Holsapple.  1985.  Immunosuppresslon by
polycycllc  aromatic  hydrocarbons.    A  structure-activity  relationship  In
B6C3F1 and DBA/2 mice.  Immumopharmacology.  9: 155-164.

WHO  (World Health Organization).   1970.   European  Standards  for  Drinking
Water, 2nd ed., Revised.

WHO (World  Health Organization).   1988.   Emissions of heavy  metal  and  PAH
compounds  from  municipal   solid  waste Incinerators:  Control  technology  and
health effects.   Report  on  a WHO  meeting, Florence,  October  12-16,  1987.
WHO, Copenhagen.  67 p.

Williams, G.M.   1977.   Detection of chemical  carcinogens  by unscheduled  DNA
synthesis 1n rat liver primary cell  cultures.   Cancer Res.   37:  1845-1851.


04460                                IX-96                           11/21/91

-------
Williams, G.M., H. Mori and C. McQueen.  1989.  Structure-activity relation-
ships 1n  the  rat  hepatocyte  DNA-repalr  test for 300 chemicals.  Mutat. Res.
221: 263-286.

Wilson,  R.H.,  F.  DeEds  and A.J.  Cox,  Jr.   1947.   The carcinogenic activity
of  2-acetam1nofluorene.   IV.  Action of related compounds.   Cancer  Res.  7:
453-458.

Wlslockl,  P.6.,  E.S.  Bagan,  A.Y.H.  Lu  et al.   1986.   Tumor 1gen1dty  of
nitrated  derivatives  of  pyrene,  benz[a]anthracene, chrysene  and benzo[a]-
pyrene 1n the newborn mouse assay.  Cardnogenesls.  7:  1317-1322.

Wojdanl, A., H. Attarzadeh, G. Wolde-Tsadlk and L.J. Alfred.  1984.  Immuno-
cytotoxldty  effects  of polycycllc  aromatic  hydrocarbons  on  mouse lympho-
cytes.  Toxicology.  31: 181-189.

Wolfe,  J.H and  W.R.  Bryan.   1939.   Effects  Induced  1n  pregnant  rats  by
Injection  of   chemically  pure  carcinogenic agents.   Am.  J.  Cancer.   36:
359-368.  (Cited In IARC,  1983)

Wood, A.W.,  W. Levin,  D. Ryan  et al.   1977.   High mutagenldty  of  meta-
bollcally  activated  chrysene  1,2-d1hydrod1ol.   Evidence   for   bay  region
activation .of chrysene.   Blochem.  Blophys.  Res.  Commun.   78: 847-854.

Wood,  A.W.,   R.L.   Chang,  W.    Levin  et   al.    1979.    Mutagenldty  and
tumorIgenlclty  of  phenanthrene   and  chrysene  epoxldes  and  dlol epoxldes.
Cancer Res.  39: 4069-4077.

04460                                IX-97                           11/21/91

-------
Wood,  A.W.,  W.  Levin,  R.L.  Chang  et al.   1980.   HutagenlcHy and  tumor-
Initiating  activity   of   cydopenta(c,d)pyrene  and   structurally   related
compounds.  Cancer Res.   40:  642-649.

Wynder,  E.L.    1988.   Tobacco  and  health:  A  review  of   the  history  and
suggestions of public  health  policy.   Public Health Reports.  103(1):  8-18.

Wynder,  E.L.  and D.  Hoffman.   1959a.  A  study  of  tobacco cardnogenesls.
VII. The role of higher polycycllc hydrocarbons.  Cancer.  12: 1079-1086.

Wynder,  E.L.  and 0.  Hoffman.   1959b.   The  carclnogenlcHy of  benzofluor-
anthenes.  Cancer.  12: 1194-1199.   (CHed 1n IARC,  1973)

Wynder,  E.L.  and 0.  Hoffman.   1961.   The carclnogenlcHy  of  d1benzo[a,l]-
pyrene.  Nature (London).   192:  1092-1093.

Wynder,  E.L.,  L. Fritz  and  N.  Furth.   1957.   Effect  of  concentration  on
      „'                                '
benzopyrene 1n  skin  cardnogenesls.   J.  Natl.  Cancer  Inst.  19:  361-370.
(CHed 1n Chu  and Chen, 1984)

Yamamoto, A.,  A.  Hlsanaga  and  N.  Ish1n1sh1.   1985.   Comparative study  on
the carclnogenlcHy of N-n1trosod1ethylam1ne and benzo[a]pyrene to the  lung
of  Syrian golden  hamsters  Induced  by  Intermittent  Instillations  to  the
trachea.  Cancer Lett. 25: 271-276.

Yamauchl, T.,  S. Komura and  K.  Yag1.   1986.   Serum I1p1d  peroxide  levels  of
albino rats administered naphthalene.  Blochem. Inter.  13:  1-6.

04460                               IX-98                           11/21/91

-------
Yang, S.K.,  J.  Deutsch  and H.V. Gelboln.  1978.  Benzo(a)pyrene metabolism:
Activation  and  detoxification.    JJK  Polycycllc  Hydrocarbons  and  Cancer:
Environment, Chemistry and Metabolism, Vol. 1, H.V. Gelboln and  P.O.P. T'so,
Ed.  Academic Press, NY.  p.  205-231.

Yang, K.H.,  C.G. Shin,  S.Y.  Choe  and D.H.  Kim.   1984.   Detection  of DNA
single-strand  breaks  Induced  by  procardnogens  1n  Chinese  hamster  ovary
cells cocultured  with  rat  hepatocytes.   J.  Toxlcol.  Environ.  Health.  13:
133-143.

Yang, J.J.,  T.A.  Roy  and C.R.  Mackerer.   1986.   Percutaneous absorption of
anthracene 1n the rat:  Comparison of in vivo  and In vitro  results.  Toxlcol.
Ind. Health.  2(1):  79-84.

Yanlsheva,  N.Ya.  1971.  The  substantiation of the  maximum permissible  con-
centration  of  benzo(a)pyrene  1n  the  atmosphere  of  settlements.   Gig. 1.
SanH.  37: 87-91.  (Rus.)

Yanlsheva,  N.Ya.  and  N.V.  Balenko.   1966.   Experimental  lung cancer  caused
by  Introduction  of  various  doses  of  1,2,5,6-dlbenzanthracene.    G1g. 1.
SanH.  31: 12.  (Cited  1n  IARC, 1973)

Yoshlkawa, T., L.P.  Ruhr, W. Flory, D. Glamalva, O.F. Church  and W.A.  Pryor.
1985.   Tox1c1ty  of polycycllc  aromatic  hydrocarbons.  1.   Effect  of phen-
anthrene,  pyrene  and  their  ozonized  products on  blood chemistry  1n rats.
Toxlcol. Appl.  Pharmacol.   79:  218-226.
04460                                IX-99                           11/21/91

-------
Yoshlkawa.  T.,  W.  Flory,  L.P.  Ruhr  et  al.   1987.  Toxlclty  of  polycycllc
aromatic  hydrocarbons.   II.  Effect  of  NO^-nHrated  phenanthrene  and  pyrene
on blood chemistry 1n rats.  Vet. Hum. Toxlcol.  29: 25-30.

Younls,  0..  Th.  Platakos,  A.  Veltsos  and S.  Vafladls.   1957.   Intoxication
de  1'homme  par  vole  aerlenne  due  a  la  naphtallne.   Etude  cllnlque  et
experlmentale  [Clinical   and   laboratory  study   of   cases   of  naphthalene
Intoxication].  Arch. Inst. Pasteur Hell.   3: 62-67.

Yuspa,  S.H.,  H.   Hennlngs  and  U.  Safftottl.   1976.   Cutaneous  chemical
cardnogenesls: Past, present and future.   J. Invest.  Dermatol.   67:  199-208.

Zedeck, M.   1980.  Polycycllc  aromatic  hydrocarbons.  A  review.   J.  Environ.
Pathol. Toxlcol.  3: 537-567.

Zlnkham, U.H.  and B.  Chllds.    1957.   Effect of  vitamin K and  naphthalene
metabolites  on  glutathlone metabolism  of erythrocytes  from normal  newborns
and  patients with  naphthalene  hemolytlc  anemia.   ANA  J.  01s.  Child.   94:
420-423.

Zlnkham, U.H.  and  8. Chllds.  1958.   A defect of  glutathlone  metabolism m
erythrocytts  fro*  patients  with  a  naphthalene-Induced  hemolytlc anemia.
Pediatrics.  22: 461-471.

Zuelzer, H.U. and  L. Apt.   1949.  Acute  hemolytlc  anemia due  to naphthalene
poisoning.  J. Am. Ned. Assoc.   141:  185-190.
04460                                IX-100                          11/21/91

-------