-------
                                                                           is by
  334-
  335'
 336-
                                                  ,„ carcinogen^,   j.  Natl


 33a-
Cancer Inst. , 52(5): 1583-1587, 1974
                                                       compounds.  J.  Natl
                                                             on mouse
"'
341.  Shamberger, R.J.  Antioxidants and cancer   TTT   c i  •
     antioxidants decrease carcinnn«n 
-------
343. Wattenberg,' L.W. and J.L. Leong.  Inhibition of the carcinogenic action
     of benzo[a]pyrene by flavones.  Cancer Res., 30:1922-1925, 1970.

344. Sullivan, P.O., L.M. Calle, K. Shafer, and M. Nettleman.  Effect of
     antioxidants on benzo[a]pyrene free radicals.  In:  Carcinogenesis,
     Vol 3:  Pblynuclear Aromatic Hydrocarbons.   P.W. Jones and R.I. Freudentnal
     (eds.).  New York, Raven Press.   1978.

345. Benson, A.M.,  R.P. Batziner, S-Y. L. Ou,  E.  Bueding, Y-N, Cha   and
     P  Talalay.  Elevation of hepatic glutathione S-transferase activities
     and protection against mutagenic  metabolites of benzo[a]pyrene  by dietary
     antioxidants.   Cancer Res., 38:4486-4495,  1978.

346. Nettesheim,  P., C.  Snyder, M.L. Williams,  M.V.  Cone, and J.C.S.  Kim.
     Effect of  vitamin  A on  lung tumor induction in  rats.   Proc. Amer. Assoc.
     Cancer Res.,  16:54,  1975.

347 Cone   M V.,  and P.  Nettesheim.   Effects  of vitamin A on 3-methylchol-
     anthrene-induced  squamous  metaplasia and early  tumors  in the  respiratory
     tract of rats.  J.  Natl.  Cancer Inst.,  50:1599-1606, 1973.
 348.
 349.
     Chu  E.W.,  and R.A.  Malmgren.   An inhibitory effect of vitamin A on the
     induction  of tumors  in the forestomach and cervix in the Syrian hamster
     by carcinogenic polycyclic hydrocarbons.   Cancer Res., 25:885-895, 1965.

     Smith  D M , A.E.  Rogers, and P.M. Newberne.  Vitamin A and benzo[a]-
     pyrene carcinogenesis in the respiratory tract of hamsters fed a semi-
     synthetic diet.  Cancer Res., 35:1483-1488, 1975.

350  Nettesheim, P., and M.L. Williams.  The influence of vitamin A on the
     susceptibility of the rat lung to 3-methylcholanthrene.  Int. J. Cancer,
     17:351-357, 1976.

351  Wattenberg, L.W.  Inhibition of carcinogenic effects of polycyclic hydro-
     carbons by benzyl isothiocyanate  and related compounds.  J.  Natl. Cancer
     Inst., 58(2):395-398, 1977.

352. Czygan, P., H. Greim, A. Garro, F. Schaffner, and  H.  Popper.  The effect
     of dietary protein  deficiency  on  the ability of  isolated hepatic  micro-
     somes to alter the  mutagenicity of a primary and a secondary carcinogen.
     Cancer  Res., 34:119-123,  1974.
                                          6-274

-------
  354' SSsr
  355'
 356'
      '"
     Natl.  Acad. Sci.  USA, 73(3)- 950-954     6chemcals-   d,scussion.  Proc
                                        N
National Institutes of Health  Pub?ip^?Ih ?at1?nal C
Education and Welfare^ Be?hesda' Md  2M14      ""'
                                                           Institute,

                                                         Dept'  Of Health.
359'
362'

                                  6-275

-------
365. Wislocki, P.G., A.W. Wood, R.L. Chang, W. Levin, H. Yagi, 0. Hernandez,
     P M  Dansette, D.M. Jerina, and A.M. Conney.  Mutagenicity and cyto-
     toxicity of benzo[a]pyrene arene oxides, phenols, quinones, and dihydro-
     diols in bacterial and mammalian cells.  Cancer Res., 36:3350-3357, 1976.

366. Wood, A.W., W. Levin, A.Y.H. Lu, H. Yagi, 0. Hernandex,  D.M. Jerina,  and
     A H. Conney.  Metabolism of benzo[a]pyrene  and benzo[a]pyrene derivatives
     to mutagenic  ..roducts by highly purified hepatic microsomal enzymes.
     J. Biol. Chem., 251(16):4882-4890,  1976.

367. Wislocki, P.G., A.W. Wood,  R.L. Chang,  W.  Levin, H.  Yagi,  0. Hernandez,
     D M. Jerina,  and  A.H. Conney.  High mutagenicity and toxicity of  a diol
     epoxide  derived from benzo[a]pyrene.   Biochem.  Biophys.  Res. Commun.,
     68(3):1006-1012,  1976.

368. Thakker, D.R., H.  Yagi, A.Y.H.  Lu,  W.  Levin,  AH   Conney  anj  D.M  Jerina.
     Metabolism  and benzo[a]pyrene:   Conversion of (+)-trans-7,9-dihydroxy-
     7 8-dfhydrobenzo[a]pyrene to highly mutagenic 7,8-diol-9,10-epoxides-.
     Proc. Natl.  Acad.  Sci.  USA, 73(10):3381-3385,  1976.

369. Oesch,  F.,  P. Bentley,  and H.R.  Glatt.  Prevention of benzo[a]pyrene-
      induced mutagenicity by homogeneous epoxide hydratase.   Int.  J.  Cancer,
      18:448-452, 1976.

 370.  Levin,  W.,  A.Y.H. Lu,  D.  Ryan, A.W. Wood, J.  Kapitulnik, S.West  M.-T.
      Huang,  D.R. Thakker, G.  Holder, H. Yagi, D.M. Jerina, and A.  H.  Conney.
      Properties of the liver microsomal monoxygenase system  and epoxide hydrase:
      factors influencing the metabolism and mutagenicity of  benzo[a]pyrene.
      In:  Origins of Human Cancer, Cold Spring  Harbor Laboratory, Cold Spring
      Harbor,. New York.  1977, in press.

 371. Glatt, H.R., K. Vogel, P. Bentley, and F.  Oesch.  Reduction of benzo[a]-
      pyrene mutagenicity by dihydrodiol dehydrogenase.   Nature 27_7:3i9-.^u,
      1979.

 372. Glatt, H.R., and  F. Oesch.  Phenolic  benzo[a]pyrene metabolites  are  mutagens.
      Mutation Res., 36:379-384,  1976.

 373. Wood, A.W.,  R.L.  Chang, W.  Levin,  P.E.  Thomas, D. Ryan, T.A. Stoming,
      D  R. Thakker, D.M. Jerina,  and A.H.  Conney.   Metabolic  activation of
      3-methylcholanthrene and  its  metabolites  to  products mutagenic to
      bacterial  and mammalian  cells.   Cancer Res.,  38:3398-3404,  1978.

 374. Malaveille,  C.,  H.  Bartsch,  B.  Tierney,  P.L.  Grover, and  P. Sims
      Microsome-mediated mutagenicities of the dihydrodiols  of  7,12-dimethyl
      benz[a]anthracene:  high mutagenic activity  of the  3,4-dihydrodiol.
      Biochem. Biophys. Res.  Commun.,  83(4):1468-1473,  1978.
                                          6-276

-------
  37S-
                 hydrocarbon,
 377-
      hydrocarbons.  Pl.
                                               dffferent
  Acad. Sci. USA, 73:607-611
                                                                       of
                                              maml»al1an «"«.  Proc. Natl

382'
           „eseasepo,ycc
     aflatoxins.  Mutation Res. ,46^27-44^977.
 the discussion of the develoDino D«

 application of DMB*.   KuruIe^.T ? Jlc!):
                                                         -  s-
                                                         -   -      >
                                                °Var'an tumor's ' ^specially,

                                                           8 ^"^"
               sagas' i
 79:161-198, 197S.
                                                        1n «»flnancy:

                                                    hydrocarbons.   Hereditas,
    cells.   Cancer Res.3in?54-4"31976!'Cln09en"tranSf0rmed 9uinea P1^
human lymphocytes.   Nature
                                                 9?6
                                                       exchanfles 1" cultured
                                    6-277

-------
    Human Genet.,  36_:361-264,  1977.

                                    ^
    1977.


    New York.  1978.
                      o^^
     S2?iatlles in normal  and xeroderma pigmentosum  human. fnbroblasts.
     Mutation Res., 43:117-138, 1977.
392-
     R. Montesano, H. Bartsch, L. Tomatis
     No. 12, Lyon.  1976.  pp. 15-24.
1975.


2Si ^r^ev^nt's^t
tribulaSon of a modern Sisyphus
1975.
                                                          ' f ^no^s
                                         Can.  J.  Genet. Cytol ., 17:471-
       :231-236, 1976.
      Biophys. Res. Commun. , 72(2): 732-738, 1976

       Hereditas, 72:311-318, 1972,
                                          6-278

-------
   398-
     Res:, 28:1738-1752  1968




400. stlch, H.F.  Chromosomes of tumor cells   I
                                                 -
                                                <<«>ethylbenz[a]anthracene.   Cancer
  401-
 405-
                          renu
       .  Natl. Cancer Inst. , 50(2)f497-?Jf , \m d'methylbenzta]anthracene.





                                       ivCV°??c0T;  '""«=»• rt«Me. of primary

                                            '   hydr0carbons in Syrian .hamsters.
      Natl. Cancer Inst.,
                                            ,97,
 408-
                                                                       and
                                                           concentration by


                                                                   e
410'
*"'
                                        6-279

-------
•«•
     cyclic aromatics.   Cancer
     ^^M.
     hydrocarbon carcinogenesis.  Cancer Res. , 33:302 309,
 «"•
      33:2354-2361, 1973.
      p.  549-559.
      American Environ.  Mutagen  Soc. ,
                                                        -
                      a
      Appl . Pharmacol . , 23_: 288-325 ,  1 972.
                                       o^o^
       Mutation Research, 51:311-318, 1978.

                                          ^

                                                       mutaaen1c potency ^
  422  Tang, T. , and H.A. FHedmn.   Carcinogen activation by human 11 vsr  enzymes

       In the toes mutagenicity test.   Mutation Res., 46:387 394, 1977.




  423. Selkirk, O.K.'  Diverge of metal »ljc  «*J»*"» systemS f°r

       mutagenesis assays.  Nature 270:604-607, 1977.











        New York 10595,  unpublished report, 1978.
                                           6-280

-------
              KCh\r:  LSSiM,?; RBhrPOm-   ^n-'ty of polycycMc
       Mutation Res., 66:65-67  ]9?9   Slster-ct"-™atid exchanges In vivo.^ Y   C
              n~yM;
                        omouse
            section.   J.  Rept,oc(e
                                                                 Effect of
 430

    '
                               neurn

      .  Soc.  Exptl. Biol.  Med   ?|58l-86,
                                                           °f

                                                            Pr°9eny  (3499«-
'  effe'rts  ofR7;12C:SineBthyiben
  tlves  in tne Sprague-DLey



'  from'the embry

      by 2-methyl-  , 2- bis-(3-
433-
                                                '
                                 AR   Uie' and B  F
                                                                  Protection
         1975.
                                     Jr.
                                                       -~
                                                    New York, Academic
                                            °f
                                                            Conference
                                              «' «- occupational  cancers.
                                    6-281

-------
439. Kennaway, E. L. , and N.M. Kennaway   A ^her study jf the incidence of
     cancer of the lung and larynx.  Brit. J. Cancer, 1-260 298, iy«-/.

440. Henry, S.A., N.M. Kennaway, and E.L. Kennaway   The incidence of cancer
     of the bladder and prostate in certain  occupations.  J. Hyg. , 3J.125
     137, 1931.

441  Kuroda,  S.  Occupational pulmonary  cancer  of generator gas workers.
     Ind. Med.  Surg. ,  6:304-306, 1937.

442. Kawai, M. ,  H.  Amanoto,  and K.  Harada   Epldemlologlc  study of  occupa-
     tional lung cancer.   Arch.  Environ.  Health,  14.859 864,  iyb/.

443. Bruusgaard, A.   The occurrence of certain forms of ca ncer among employees
      in gasworks.   Tid.  for den Norske Laegeforen,  79:755-756, 1959.

 444  Reid,  D.D., and C.  Buck.   Cancer in coking plant workers.  Brit. J. Ind.
      Med.,  13:265-269, 1956.

 445. Lloyd, J.W.  Long term mortality study °f steelworters.   V. Respiratory
      cancer in  coke plant workers.  J. Occup. Med., 13(2). 53 68, 1971.
 446. Doll, R.  The causes of death among gas "°^s ^special reference
      to cancer of the lung.  Brit. J.  Ind. Med., 9:180  187,  19W.

 AAV  nnll  R   R E W  Fisher,  E.J. Gammon, W. Gunn, G.O.  Hughes,  F.H. Tyrer,
      and W  Wilson   Mortality of gas  workers with  special  reference to
      Sneers  of Se lung and bladder,  chronic bronchitis,  and pneumocomosis.
      Brit. J.  Ind. Med., 22:1-12, 1965.

 448 Doll   R  M  P  Vessey,  R.W.R.  Beasley,  A.R.' Buckley, E.G.  Fear, R.E.W.
 4   Fisher?" E.J  Gammon,^. Gunn,  G.O.  Hughes, K.  Lee, and B  Norman-Smith
      Mortality of gasworkers - final report of  a prospective study.  Brit. J.
       Ind.  Med.,  29:394-406,  1972.
       cology, Dayton, Ohio, October 13-15, 1976.
   «•
        19:524-529,  1959.
453. Hammond, E.C., I.J. Sell toff, P. L
     of benzpyrene and cancer in man.  Ann. .NY Acad
                                                                     Inhalation
                                                                  'lt> 1^4,  is/o.
                                           6-282

-------
                                                         *»—•* Group's
  455'
 '  l:"a-3R,,
                             pollution and lung cancer
                                                                      „ and

                                            «"1- compounds.  Arch.  Environ.



                                                  tooacco
463-
464.  Mancuso  T F   and F i
                                                        among  native white,
.  Cancer
                                      2437™'
                                                                  V.H. Handy

                                                                  °f "6W
                                                       C— cticut State
Cancer Inst. ,  28:947-1001,  1962
                                                              mortality as
                                                 ,  white  males.  J. Natl.
                                      6-283

-------
  XSSS: ti
  Inst. , 32:803-838, 1964
   1033-1042,  1954.
    tion by benzpyrene and other substances
                                          .   Brit. Med. J. , 2.923 939,
    8:181-198, 1954.

    547,  1969.

    Environ.  Health,  15:237-248,  1967.
     card?orespiraor;  symptoms among migrants and nat ve-born
     United States.   Natl. Cancer Inst. Monograph No.  19, 1966.

477. Eastcott, D.F.   The epidemiology of lung cancer in New Zealand.
     1: 37-39, 1956.
                                                                    Lancet,
     20:79-105,  1958.
479. Dean 6.   Lung cancer  among white South Africans.  Brit.  Med.  J.,  2:852-
     857, 1959.
480. Dean, G.  Lung cancer in South Africans and British immigrants.   Proc.
     Roy. Soc. Med., 57:984-987,  1964.
                                        6-284

-------
  481
  482

                                    °f the  '""• '"
                                                                and Wales.
      deposli a'nd
 with particular
 vanadium, and
                                to  3-
                                                  K
                                                  b™"^itis, and pneumonia
 cr«esmo                     oc1"?
 of causation.   Brit.  J.  Can«" |J:59?-623  ?966S1

                                                                   hWothesi
49°'
                                                •- P«l-n.W cancer.   Arch.
                                                                  pollution
493'
Assoc. ,  166:1294-1308,
                                                           cause-  J- Am. Med.
                                       6-285

-------
495  Hitosugi, Ml  Epidemiological study of lung cancer.  Inst.  Public Health
     Bull., 17:237-256, 1968.
(ds.),   international  Agency
                                                                  f
                                 for Research on Cancer, Lyon,  France, 1977
     press) p.  457-609.
     Geochim. et Cosmochim. Acta. ,  39:
 AQQ   Rn™eff  J   F  Selenka  H. Kunte,  and A. Maximos.   Experimental studies
 4"'  ^formation of polycyclic aromatic  hydrocarbons in plants.  Emnron.
      Res., 2:22-29, 1968.
      Cosmochim. Acta., 40:1141-1143, 1976.
 •"•
      Naturwissenschaften. ,  62(11): 536- 537, 1975.
 «•
  -•
      513-528, 1966.
       Exp. Biol. and Med. ,  150:49-51, 1975.

       Interact., 10:265-268,  1975.
•  ss-s-.^
  Microbiol., 32(1): 95-101,  1976.
                                                       zr i-rar s
  507. Poglazova,  M.N., and M.N.  Meisel.  LocalIratlon  of benz(a)pyrene  in
       bacterial  cells.  Mikrobiologiya, 40(6):1050-1053,  is/i.
                                         6-286

-------
508-
   "
                                             e
                 .  Hyg.  Abt.  1,  155(2)68-74! 197 K

                        e^^                         Inhibition of sterol  and
      .  Antibiot.,  25(6):365-8;  ?972          cell-free systems  of yeast.
   ' butoxide'on'forma?ion oTrt
     of Staehyjococcus aureus.



5"'  Por^systeVas e^te'd
         .,  3rd,  pp.  193-207,  1972
                                                           and P
                                                                    carotenoids
                                                           ™ elect-n trans-
                                                           Workshop Conf . ,

   and concentration.  Proc. lnt. spa".'
   benzo[a]pyrene.  Biofiz. Zhivoi
                         (Eng.
   Water Pollut. Contr. Fed. , llo-2029
   Biol., 17:201-208, 1972.
                                                                       ]973
                                                    '  and
                                                               .   Res,stance
                                                     actlwated  =l"1ge.  j.
                                                                 and
                                                by marine fish.  Marine
                                    6-287

-------
519. Southworth, G.R., J.J.  Beauchamp,  and B.K  Schmieder   B1oacci«.ulat1on
     potential of polycyclic aromatic hydrocarbons in Daphnia £ulex.
     Water Re., 12:973-977,  1978.
     binding anSSa^nlclt^HSppe-s'eyUr's Z. Physio!. Chem. Bd. 357, S.
     1019-1067, 1976.

522  Ahokas  J T., 0. Pelkonen, and N.T. Karki.  Metabolism of polycyclic
     hydrocarbons by a  highly active aryl hydrocarbon hydroxylase system in
     the  liver of a trout  species.  Biochem. Biophys. Res. Commun., 63(3):
     635-641, 1975.

 523  Arffmann  E., and  B.C.  Christensen.  Studies on the  newt test  for car-
 523' cinogen?city:   I.   Benzo[a]pyrene,  dibenz[a,h]anthracene and 3-methyl-
     cholanthrene. Acta Pathol. Microbiol.  Scand.,  52:330-342,  1961.

 524  Balls  M   Benzpyrene-induced tumours  in  the clawed  toad,  Xenopjjs laeyjs.
      Experientia, 20(3):143-145,  1964.

      Raddina  S G-   T  Mill, C.W.  Gould, D.H.  Liu,  H.L. Johnson,  D.C.  Bomberger,
      and C V* Fojo'   The Environmental  Fate of Selected Polynuclear Aromatic
      Hvdrocarbons   Prepared by Stanford Research Institute,  Men!o  Park,
      California, under Contract No. 68-01-2681.  U.S  ^ironmental
      Protection Agency.  Washington, D.C. Publ. No. EPA-560/5 750 009.

 *9S  Bridbord  K   and J.G. French.  Carcinogenic and mutagenic risks associated
 O&D. DnuDuiu, i\. , auu W.M. •               ^ ,              \/niirlpar* Aromatic

      ^droSSsI^W. J§L anSCRnr?rrud4nt0hai l^^l^T^
      York,  1978.

      Mazumdar  S   C   Redmond, W.  Sollecito,  and N. Sussman.  An epidemiological
      S  of'exposure to coal-tar-pitch volatiles among coke  oven workers.
      APCA J., 25(4):382-389,  1975.

  528. Menck, H.R., and  B.E.  Henderson    Occupational differences  in rates  of
       lung cancer.   J.  Occup.  Med., 18(12):797-801,  1976.
                  c                   uon
       s?udies9wilh benzo[a]pyrene in bovine serum albumin in Syrian hamsters.
       Cancer Letters, 4:235-259, 1978.
                                           6-288

-------
                                       1' "'
   531.
                                                   '
  thrace e S.^Tz-epox^ran?^

  elo'flo?-^""6 3'«-d"W*«««l 1"
                                                                          ''
                                                        ?1?s*ere?ler1c  benz[a]an-

                                                         "3"1116" °f
  532'
                       on mou   -
 Nati.  Cancer Inst.,



i,  J.M.  Karle,  D.M.
                          on mouse ski'n   ex™«,
       3.4-dllwdrodlol.  Proc.  Natl.  Acad.  ST5.I!!!!.1.
                                                                          of
 anthracene diol epoxides and
                                                                    of
                                                                      .  O.H.  Kane,
      an ulttat. carcinogen
                                               W3-l
      is an
epoxides.  Cancer Res     n699-
                                n699-lW
                                                             diols and diol-
of tha optical enantiomersof the   aneeomeric
9,10-epoxides.  Cancer Res., 39:67-7?,
                                                                   e 7,8-dlol-
               hydrod,o,s
54°-

                                       6-289

-------
 , REPORT NO.
 EPA-600/ 9-7 9-008
i.TITLEANDSUBTIILb
 Health  Assessment Document for Polycyclic
 Organic Matter                   ___
   UTHOBM Joseph .Santodonato'-DrlTThilip Howard  Dipak
'su  and Sheldon  Lande, Syracuse Research Corp., Dr.

     ^M^S^^^^
 Syracuse Research  Corporation
 Merrill  Lane
 Syracuse, N.Y. 13210
                                                   13 TYPE OF REPORT AND PERIOD COVERED
                                                    'Final	_	'
                                                   14. SPONSORING AUENCY CODE

                                                    EPA/600/00    	
Research Triangle rant, n.v>. ^/"-	i	.	;      .	
^SUPPLEMENTARY NOTES ORNL = Oak Ridge National  Laboratory, Oak Ridge, TN.  37830
                 CMMV = [ajpyrene
                                         uspended particulace matt
                                         nobile sources
                                         tationary sources
                                         iza arenes
                                         mi no arenes
                                         19 SECURITY CLASS (1
                                         UNCLASSIFIED	
                                         20. SECURITY CLASS (Thispage/

                                         UNCLASSIFIEI
  COSATI Field/Group
    1)6A~
    06C
    06F
    06J
    06P
    06T
    07C
21. NO. OF PAGES

22. PRICE
             • B... ^-77!
                                       6-290

-------

-------

-------

EPA-600/9-79-008'
  December 1979
                                    HEALTH
                            ASSESSMENT
                              DOCUMENT
                   FOR  POLYCYCLIC
                  ORGANIC  MATTER

                                     (PREPRINT)
                     U.S. ENVIRONMENTAL PROTECTION AGENCY
                      Office of Health and Environmental Assessment
                             Office of'Research and Development
                       Environmental Criteria and Assessment Office
                      Research Triangle Park, North Carolina 27711

-------

-------
                                   EPA-600/9-79-008
                                   December 1979
HEALTH ASSESSMENT
          DOCUMENT
   FOR POLYCYCLIC
   ORGANIC  MATTER
                (PREPRINT)
                    bv
          Joseph Santodonato, Dr. Philip Howard,
          Dr. Dipak Basil, and Dr. Sheldon Lande
            Syracuse Research Corporation
                 Merrill Lane
                Syracuse, NY 13210
               Dr. James K. Selkirk
            Oak Ridge National Laboratory

                 Dr. Paul Sheehe
             State University of New York
       V.S. ENVIRONMENTAL PROTECTION AGENCY
        Office of Health and Environmental Assessment
           Office of Research and Development
         Emironmental Criteria and Assessment Office
         Research Triangle Park, North Carolina 27711

-------
                                    PREFACE
     This document was written in response to Section 122 of the Clean Air Act
Amendments of 1977, P.L. 95-95, as noted under 42 U.S.C. Section 7548.  The
purpose of this report is to decide whether or not polycyclic organic matter
(POM) emissions into the ambient air cause or contribute to air pollution which
may reasonably be anticipated to endanger public health.
     The document constitutes an in-depth review of the original published
literature and the current scientific position with regard to knowledge of
polycyclic organic matter.   Chemical and physical properties, atmosphere
forms, fate and transport,  measurement techniques, ambient levels, toxicology
and occupational and epidemiological data on POM's have been reviewed.
     The Agency is pleased to acknowledge the efforts of all persons and
groups who have participated in preparing this document.

-------
                          CONTRIBUTORS AND REVIEWERS
     The Syracuse Research Corporation (SRC) and its consultants were the major
authors of this document.  Dr. Robert Bruce served as the EPA Project Director.

                                  SRC Staff:

                            Mr. Joseph Santodonato

                               Dr. Philip Howard

                                Dr. Dipak Basu

                               Dr. Sheldon Lande


                               SRC Consultants:

                             Dr. James K. Selkirk
                               Biology Division
                         Oak Ridge National Laboratory
                             Oak Ridge, TN   37830

                                Dr. Paul Sheehe
                       Department of Preventive Medicine
                         State University of New York
                              College of Medicine
                             Syracuse, NY   13210


     The following people from.the Environmental Research Center, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina, served on the
EPA Task Force on POM:

Mr. Michael Berry
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Research Triangle Park, NC   27711

Dr. Robert M. Bruce
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Research Triangle Park, NC   27711

-------
Mr. Joseph  Bumgarner
Environmental Monitoring Systems  Laboratory
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Dr. Neil Chernoff
Health  Effects  Research Laboratory
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Mr. Robert  Faoro
Office  of Air Quality Planning and Standards
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Mr. Douglas Fennel 1
Environmental Criteria and Assessment Office
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Dr. Robert Morton
Health  Effects  Research Laboratory
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Mr. Robert Jungers
Environmental Monitoring Systems  Laboratory
U.S. Environmental  Protection Agency
Research Triangle Park, NC .  27711

Mr. Justice Manning
Office of Air Quality Planning and Standards
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Dr. David McKee
Environmental Criteria and Assessment Office
U.S. Environmental  Protection Agency
Research Triangle Park, NC   27711

Mr. Thomas McMullen
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Research Triangle Park, NC   27711
                                    IV

-------
 Mr.  James  Mulik
 Environmental  Sciences  Research  Laboratory
 U.S.  Environmental  Protection  Agency
 Research Triangle Park,  NC   27711

 Dr.  Steve  Nesnow
 Health  Effects  Research  Laboratory
 U.S.  Environmental  Protection  Agency
 Research Triangle Park,  NC   27711

 Dr.  Shabeg Sandhu
 Health  Effects  Research  Laboratory
 U.S.  Environmental  Protection  Agency
 Research Triangle Park,  NC   27711

 Dr.  Eugene Sawicki
 Environmental Sciences Research  Laboratory
 U.S.  Environmental  Protection  Agency
 Research Triangle Park,  NC   27711

 Mr.  James  R. Smith
 Health  Effects  Research  Laboratory
 U.S.  Environmental  Protection  Agency
 Research Triangle Park,  NC   27711

 Ms.  Beverly Tilton
 Environmental Criteria and Assessment Office
 U.S.  Environmental  Protection  Agency
 Research Triangle Park,  NC   27711
     The following, people served as consulting contributors and reviewers in
the preparation of this document:

Dr. Ian T.  Higgins
Professor of Epidemiology
School of Public Health
University of Michigan
Ann Arbor,  MI   48109

Dr: Dietrich Hoffmann
Chief, Division of Environmental Carcinogenesis
Nay!or Dana Institute for Disease Prevention
American Health Foundation
Dana Road
Valhalla, NY   10592

-------
Dr. Peter W. Jones
Program Manager
Analytical and Environmental Chemistry Division
Battelle Columbus Laboratories
505 King Avenue
Columbus, OH   43201

Dr. Edmond J. LaVoie
Head, Section of Metabolic Biochemistry
Nay!or Dana Institute for Disease Prevention
American Health Foundation
Dana Road
Valhalla, NY   10592

Dr. David F.S. Natusch
Professor of Chemistry
Colorado State University
Fort Collins, CO   80523

Dr. Herbert S. Rosenkranz
Professor and Chairman
Department of Microbiology
New York Medical College
Valhalla, NY   10595

Dr. Warren Winkelstein
Professor of Epidemiology
Department of Biomedical and Environmental Health Sciences
School of Public Health
University of California
Berkeley, CA   94750

-------
                                    ABSTRACT

     This  document was  prepared  in  response to  the  Clean  Air  Act  Amendments
of 1977 and constitutes an  in-depth scientific  review  and assessment  of the
literature on polycyclic organic matter  relative to the health and ecological
effects associated with exposure to these complex organic molecules,  either
as pure compounds or as ROM-containing effluents from  industralized sources.
It does not cite every published article relating to POM  since a Multimedia
Health Assessment Document  for POM,  J. Environ. Pathol. Toxicol.  (in  press,
1980), has been prepared for EPA by Syracuse Research Corporation and covers
exposure from air, food, and water  sources.
     In addition to the chapters devoted to the health and ecological  effects
associated with exposure to POM, there are additional  topics discussed which
are necessary to support these effects from a qualitative and quantitative
viewpoint.   These include separate chapters which discuss chemical and
physical  properties,  atmospheric forms, atmospheric fate and transport,
measurement techniques,  and ambient atmospheric levels.
                                     vn

-------
                              TABLE OF CONTENTS
                                                                           Page
1.  EXECUTIVE SUMMARY	    1-1
    1.1  DEFINITION AND FORMATION	    1-1
    1.2  ANALYTICAL METHODS	    1-2
    1.3  AMBIENT CONCENTRATIONS	    1-3
    1.4  HEALTH AND ECOLOGICAL EFFECTS	    1-4

2.  INTRODUCTION	    2-1
    2.1  REFERENCES	    2-3

3.  PHYSICAL AND CHEMICAL DATA	    3-1
    3.1  DEFINITION AND FORMATION	    3-1
    3.2  STRUCTURE AND PROPERTIES	    3-4
         3.2.1  Chemical Structure	    3-4
         3.2.2  Physical Properties	   3-27
         3.2.3  Atmospheric Forms of POM		   3-28
    3.3  ATMOSPHERIC CHEMISTRY	   3-35
         3.3.1  Photooxidation	   3-35
         3.3.2  Oxidation and Nitration	   3-41
         3.3.3  Atmospheric Stability	   3-45
    3.4  SUMMARY AND CONCLUSIONS	   3-45
    3.5  REFERENCES	   3-48

4.  SAMPLING AND ANALYTICAL METHODS FOR POM	    4-1
    4.1  INTRODUCTION	    4-1
    4.2  SAMPLING	    4-3
         4.2.1  Mobile Sources	    4-4
         4.2.2  Tobacco Smoke.	    4-6
         4.2.3  Stationary Sources	    4-6
                4.2.3.1  Emissions from stacks	    4-6
                4.2.3.2  Fugitive emissions from road-way dust	   4-10
         4.2.4  Ambient Air	   4-11
                4.2.4.1  Collection of total  air particulate matter	   4-12
                4.2.4.2  Distribution of particle size	   4-14
    4.3  DESORPTION OF POM FROM COLLECTION MEDIA	   4-16
         4.3.1  Extraction	   4-16
                4.3.1.1  Solvent extraction	   4-17
                4.3.1.2  Mechanical disruption	   4-19
                4.3.1.3  Hydrofluoric acid dissolution	   4-19
         4.3.2  Thermal Stripping	   4-21
         4.3.3  Vacuum Sublimation..	   4-22
    4.4  CLEAN-UP AND SEPARATION	   4-23
         4.4.1  Solvent Partitioning	:	   4-24
                4.4.1.1  Methanol-water	   4-24
                4.4.1.2  Tetramethyl uric acid in methanol	   4-25
                4.4.1.3  Acetonitrile	   4-25
                4.4.1.4  Dimethyl formamide	   4-25
                4.4.1.5  Nitromethane	   4-25
                4.4.1.6  Dimethyl sulfoxide	   4-25
                                       vm

-------
     4.4.2  Column Chromatography.
4.5
4.6
4.7
       4.4.2.1  Alumina.
       4.4.2.2  Silica gel	
       4.4.2.3  Florisil	
       4.4.2.4  Cellulose acetate	
       4.4.2.5  Gel filtration	
4.4.3  Paper Chromatography	
4.4.4  Paper and Thin-layer Electrophoresis..
4.4.5  Thin-layer Chromatography	
4.4.6  Gas Chromatography	
       4.4.6.1  Packed column	
       4.4.6.2  Capillary column	
4.4.7  High Pressure Liquid Chromatography...
DETECTION	,	
4.5.1  Flame lonization and Electron Capture.
4.5.2  U.V. Absorption	
4.5.3  Luminescence Analysis	
       4.5.3.1  U.V. excited luminescence emission	
       4.5.3.2  Shpol'skii effect	
       4.5.3.3  X-ray excited optical luminescence (XEOL).
       4.5.3.4  Sensitized fluorescence	
       4.5.3.5  Synchronous luminescence spectroscopy	
4.5.4  Mass Spectrometry	
4.5.5  Other Techniques	
ANALYTICAL METHOD USED BY EPA FOR NASN MONITORED SAMPLES..
DIFFICULTIES IN POM MONITORING AND ANALYSIS	
4.7.1
            Sampling Errors	
            4.7.1.1  Nonisokinetic sampling	
            4.7.1.2  Non-incorporation of sampling efficiency...
            4.7.1.3  Evaporative losses and reactive conversion.
     4.7.2  Losses During Transportation and Storage	
     4.7.3  Errors in Analytical Procedures —	
            4,7.3.1  Losses during desorption process	
            4.7.3.2  Column and thin-layer chromatographic and
                    storage 1osses	
            4.7.3.3  Losses in evaporative concentration step...
            4.7.3.4  Errors due to incomplete resolution	
     4.7.4  Uncertainty in the Number of Parameters to be
              Monitored	
4.8  ECONOMIC CONSIDERATION IN THE SELECTION OF THE ANALYTICAL
     TECHNIQUE	
4.9  SUMMARY AND CONCLUSIONS	:	
4.10 REFERENCES	,
AMBIENT LEVELS	
5.1  AMBIENT ATMOSPHERIC  LEVELS	,
5.2  SUMMARY AND CONCLUSIONS	,
5. 3  REFERENCES	
4-27
4-28
4-28
4-29
4-30
4-30
4-31
4-31
4-31
4-34
4-34
4-35
4-36
4-39
4-39
4-40
4-40
4-40
4-42
4-43
4-44
4-44
4-45
4-46
4-46
4-48
4-48
4-48
4-49
4-49
4-51
4-53
4-53

4-53
4-53
4-53

4-55

4-56
4-59
4-63
 5-1
 5-1
5-39
5-42

-------
6.  HEALTH EFFECTS.
    6.1
    6.2
    6.3
    6.4
    6.5
   6.6
   6.7
PHARMACOKINETICS	
6.1.1  Absorption...
6.1.2  Distribution.
6.1.3 .Elimination..
METABOLISM	
         6.2
         6.2
         6.2
         6.2
         6.2.5
         6.2.6
       Metabolism of POM in Animals	
       Metabolic Bioactivation of Benzo[a]pyrene	
       Tissue-Specific Metabolism of POM	
       Comparative Metabolism of POM in Animals and Man.
       Actions of Enzyme Inducers	
       Intracel1ular Bindi ng of POM	
TOXICOLOGY	
6.3.1  Acute Exposures	
6.3.2  Subchronic and Chronic Exposures	
6.3.3  Effects on the Immune Systems	
CARCINOGENESIS	
6.4.1  Structure-activity relationships	
       Chemical Reactivity and Carcinogenicity....-	
       Skin Carcinogenesis and Induction of Local
           Sarcomas	
           4.2
           4.3
                6.
                6.
                6.
                6.
                6.
            ,1
            .2
            .3
6.4.3.
6.4.3.
 .4.3.
 .4.3.4
 .4.3.5
 .4.3.
 .4.3.
6.4.3.8
6.4.3.9
            .6
            .7
                 Benzo[a]pyrene	
                 Benz[a]anthracene	
                 7,12-Dimethylbenz[a]anthracene	
                 3-Methylcholanthrene	
                 Dibenz[a,h]anthracene	
                 Dibenzo[a,h]pyrene	
                 Dibenzo[a,i]pyrene	 6
                 Dibenzo[c,g]carbazole	 6
                 Benzo[c]phenanthrene	 6
       6.4.3.10  Benzo[b]f 1 uoranthene	 6
       6.4.3.11  Benzo[j]fluoranthene	 6
6.4.4  Respiratory Tract Carcinogenesis	 6
6.4.5  Carcinogenesis in Newborn Mice	 6
6.4.6  Tumors in Other Tissues	 6
       6.4.6.1   Oral exposure to POM	*... 6
6.4.7  Carcinogenicity of POM Mixtures	 6
6.4.8  In Vitro  Carcinogenesis Studies	 6
6.4.9  Dose-Response Models in Carcinogenesis	 6
6.4.10 Synergistic and Antagonistic Interactive Effects	 6
MUTAGENESIS	 6
6.5.1  In Vitro  Studies	 5
       6.5.1.1   Microbial systems		 6
       6.5.1.2   Somatic cells in culture...	 6
6.5.2  In Vivo Studies	 6
       6.5.2.1   Effects in somatic tissues	 6
       6.5.2.2   Effects in germinal  tissues	 6
6.5.3  Correlation of Mutagenicity with Carcinoqenicity	 6
REPRODUCTION AND TERATOLOGY	                      5.
HUMAN STUDIES	!!!.'.'!!!!."!.'!!. 6-
6.7.1  Occupational Studies	',,', 6'
6.7.2  Community Studies	'. g-
 6-1
 6-1
 6-1
 6-3
6-14
6-28
6-29
6-37
6-40
6-50
6-54
6-60
6-66
6-67
6-70
6-73
6-75
6-78
6-83

6-85
6-85
6-92
6-92
6-97
6-98
6-99
-100
-101
-101
-102
-103
-103
-111
-113
-115
-119
-126
-133
-140
•150
 150
•150
 156
 164
 164
 166
 170.
 178
 186
 187
 202

-------
            6.7.2.1  Urban-rural comparisons	  6-204
            6.7.2.2  Migrant studies	  6-209
            6.7.2.3  Regression analyses	  6-209
            6.7.2.4  Retrospective and prospective analyses	  6-219
6.8  ECOLOGICAL EFFECTS	  6-223
     6.8.1  Effects on Microorganisms and Algae	  6-223
     6.8.2  Effects on Aquatic Organisms and Amphibians	  6-228
6. 9  SUMMARY AND CONCLUSIONS.	  6-232
     6.9.1  Absorption, Distribution, and Excretion	  6-232
     6.9.2  Metabolism and Metabolic Activation		.  6-233
     6.9.3  Toxicology	  6-237
     6.9.4  Mutagenesis	  6-238
     6.9.5  Carcinogenesis	  6-240
     6.9.6  Reproduction and Teratology	  6-243
     6.9.7  Human Studies	  6-243
     6.9.8  Ecological Effects		  6-245
6.10 REFERENCES	  6-246
                                  XI

-------
                                LIST OF FIGURES
Figure
Page
3-1   Mechanism of benzo[a]pyrene formation 	    3-3
3-2   Accepted nonr nclature for benzo[a]pyrene 	    3-6
3-3   Absorption specturm of benzo[a]pyrene in ethanol 		   3-29
3-4   Size distribution of particles containing benzo[a]pyrene	   3-31
3-5   Percent mass of benzo[a]pyrene in relation to stage number for
        fine sampling sites in Toronto, Canada .	   3-33

5-1   Composite quarterly averages for BaP at 32 urban and 19 nonurban
        NASN stations	    5-4
5-2   Benzo[a]pyrene—seasonality and trends (1966 to 1975) in the 50th
        and 90th percentiles for 34 NASN urban sites	   5-19
                                               o
6-1   Clearance of radioactivity derived from [ H]3,4-benzopyrene in
        the 1 ung	   6-21
6-2   Clearance of radioactivity derived from [ H]3,4-benzopyrene in
        liver	,	   6-23
6-3   Excretion of radioactivity dervied from [ H]3,4-benzopyrene in
        feces 	,	   6-24
6-4   Excretion of radioactivity derived from [ H]3,4-benzopyrene in
        urine 	3	   6-25
6-5   Clearance of radioactivity derived from [ H]3,4-benzopyrene in
        kidneys 	3	   6-26
6-6   Blood, level of radioactivity derived from [ H]3,4-benzopyrene	   6-27
6-7   Enzymatic pathways involved in the activation and detoxification
        of BaP	   6-30
6-8   Oxidative metabolites of benzo[a]pyrene 	   6-32
6-9   Probable metabolic pathways of 7-methylbenz[a]anthracene 	   6-36
6-10  Stereochemical course of metabolism of (+) and (-)-BaP 7,8-dihy-
        drodiol	   6-39
6-11  Comparative metabolism of benzo[a]pyrene by lung microsomes from
        rat, rhesus monkey, and human	   6-52
6-12  Log-log plot of the cumulative incidence of skin carcinomas per
        mouse versus time	   6-86
6-13  Incidence of cancer per rat after weekly topical application of
        either 20, 100,  or 50 ug DMBA in 1.0 ml acetone	   6-95
6-14  Respiratory nodule incicence in hamsters given BaP by intratracheal
        instillation	  6-108
6-15  Respiratory nodule incidence in hamsters given BaP by intratracheal
        instillation with and without Fe203	  6-109
6-16  The inhibiting effect of hydrocarbons found in polluted urban air
        and cigarette smoke in concentrations approximating those
        occurring naturally when injected in combination with BaP	  6-142
                                    xii

-------
                                 LIST OF TABLES
 Table
 3-1   Polycyclic aromatic hydrocarbons	    3-8
 3-2   Aza and inn no arenes	   3-19
 3-3   Miscellaneous polycyclic organic  matter	   3-26
 3-4   Percent destruction of polycyclic aromatic  hydrocarbons  under	
         atmospheric conditions	    3-39
 3-5   The half lives of BaP,  BbF,  and BkF  under various'reaction	
         conditions.	        3-42
 3-6   Comparison of PAH content of air  solids before  and  after storage!!!   3-44
 4-1   Recovery of POM from adsorbent sampler  method	    4-8
 4-2   Amounts of POMs collected on glass-fiber filter and polyurethane
         foam.	   4-15
 4-3   Comparison of the extraction yield of POM with  different solvents!!   4-18
 4-4   Comparison of HF extraction  with  Soxhlet extraction	   4-20
 4-5   Efficiencies  of solvent-partition purification  procedure	'.   4-26
 4-6   Recovery of  PAH by linear elution adsorption chromatography..         4-29
 4-7   Recovery of  PAH by Sephadex  LH-20 chromatography	   4-30
 4-8   Recovery of  various  PAH  during thin-layer chromatography	!!!!!!   4-32
 4-9   Amounts  of aza-heterocyclic  compounds in a coal-tar-pitch polluted
         air sample  separated by TLC	   4-33
 4-10   HPLC conditions and  separation mechanisms used  for  PNA analysis!!!!.  4-37
 4-11   Percentage losses  of POM from filter paper due  to variation of
         storage  time	   4_52
 4-12   Comparative cost,  throughput, and other  relevant data for'four	
         commonly used analytical methods	   4-57
 5-1   Influence  of  occupational  and other factors upon benzo[aipyrene
         exposure.	   5_2
 5-2    Annual average  ambient benzo[a]pyrene concentrations at  NASN urban'
         stations	   5.5
 5-3    Annual average  ambient benzo[a]pyrene concentrations at  NASN
         nonurban stations	   5-11
 5-4    Annual average  ambient benzo[a]pyrene concentrations at NASN urban'
         stations . •.	  5_^3
 5-5    Polycyclic aromatic compounds in the air of selected cities!!!!!!!!  5-16
 5-6    Annual average  concentration  of PAH compounds in the air over
         greater  Birmingham, Alabama, 1964 and 1965	  5-17
 5-7    Concentration of polycyclic aromatic hydrocarbons in Norway
        aerosols	;	  5-25
 5-8    Concentration of ten polycyclic aromatic hydrocarbons  in Ontario
        city air	  5_26
 5-9    Polycyclic organic matter in ambient air	  5-27
6-1    Distribution and rate of elimination of radioactivity  after" "	
         intravenous injection of   C-benzo[a]pyrene into rats	   6-5
6-2   Distribution and rate of elimination of radioactivity  after
        intratracheal instillation of   C-benzo[a]pyrene into rats	   6-6
6-3   Effect of molecular structure on  hydrocarbon localization in
        mammary gl and and fat	;	;...	   5-7
6-4    Levels of orally administered 3-methylcholanthrene in  tissues "of"'
        50-day-old virgin female rats	    6-9
                                      xm

-------
Table
Page
6-5   Levels of orally administered 3-methylcholanthrene in breast and
        fat of virgin rats and in milk, breast, and fat of lactating
        rats	   6-10
6-6   Distribution and elimination of radioactivity after instillation
        of  H-dibenzo[c,g]carbazole in Syrian hamsters	   6-11
6-7   Concentration (ng/g) of DMBA, BaP, and MCA in tissues of 21-day
        fetuses, their placentas, and the liver of pregnant rats 3 hours
        after administration of the compounds in a dose of 200 mg/kg by
        gastric tube	   6-13
6-8   Distribution of radioactivity in rats with biliary fistulas after
        intravenous injection of   C-benzo[a]pyrene in plasma solution...   6-16
6-9   Amounts of B(a)P detected in hamster lungs following intratracheal
        instillation.	   6-19
6-10  BaP-hydroxylase activity in different organs of the rat after
        methyl cholanthrene and phenobarbitone pretreatment	   6-42
6-11  Summary of major organic solvent-soluble and water-soluble
        metabolites following the short-term tracheal organ culture of BP
        or its metabolites	'	   6-44
6-12  Comparison of benzo[a]pyrene metabolite patterns in liver and
        intestinal microsomes from MC-treated rats	   6-47
6-13  Activation of benzo[a]pyrene by colon and liver S9 fractions	   6-49
6-14  Metabolite percentages of BP metabolites from rat, rhesus, and
        human lung microsomal assays	   6-53
6-15  Carcinogenic compounds in descending order of potency	-..   6-77
6-16  Skin tumors in mice treated with benzo[a]pyrene and derivatives	   6-88
6-17  Summary of the skin tumor initiating activities of benzo[a]pyrene
        and its metabolites	   6-91
6-18  Induction of sarcoma by benzo[a]pyrene	   6-93
6-19  Comparative carcinogenicity of polycyclic hydrocarbons and related
        compounds, measured by induction of lung tumors (LT) in strain
        A mice	'	  6-105
6-20  Induction of respiratory tract tumors in Syrian golden hamsters by
        intratracheal instillation of POM	  6-106
6-21  Induction of respiratory tract tumors in rats and mice	  6-110
6-22  Carcinogenicity of POM by oral administration to various mammals...  6-116
6-23  Gastric tumors in mice fed benzo[a]pyrene	  6-118
6-24  Relationship of leukemia to lung_and stomach tumors in 108 mice fed
        benzo[a]pyrene	  6-120
6-25  Classification of test groups	..  6-122
6-26  Tumor incidence resulting, by the end of the 114th week, from a
        single subcutaneous application of test substances	  6-123
6-27  Doses (|jg) applied in dermal administration experiments, in
        relation to benzo[a]pyrene	  6-124
6-28  Findings at the site of application of PAH to mouse skin	  6-125
6-29  Hamster embryo cell transformation produced by several polycyclic
        hydrocarbons and their derivatives	  6-129
6-30  Effect of antioxidants on benzo[a]pyrene metabolism as assayed by
        Salmonella typhimurium (strain TA98) mutagenicity in the
        presence of rat liver microsomal fraction and NADPH...."	  6-146
6-31  Comparison of inherent mutagenic activity of thirty BaP derivatives
        in Salmonella typhimurium TA98 and in Chinese hamster V79 cells..  6-153

                                       xi v

-------
 Table

 6-32   Mutagenicity  of  carcinogenic  hydrocarbons  to  Chinese  hamster  V79
         cells  co-cultivated with  Syrian  golden hamster  embryo  cells	 6-157
 6-33   Induction  of  ouabain- and 8-azaguanine-resistant  mutants by
         different chemical carcinogens	 6-158
 6-34   Induction  of  temperature-resistant mutants  in Chinese  hamster'ovary
         temperature-senstitive cells by  different chemical carcinogens... 6-159
 6-35   Mutagenic  response of the X chromosomes in  Drosophila	         6-167
 6-36   Mutagenic  response of the BB  loci  in Drosophila..		 6-169
 6-37   Agents producing early fetal  deaths and/or  preimplantation losses"
         significantly  beyond control limits but requiring further
         verification because of lack of  internal  consistency	 6-171
 6-38   Comparison of mutagenicity in Ames assay with carcinogenicity of
         polynuclear hydrocarbons	 6-173
 6-39   Induction of  sister-chromatic exchanges by  intraperitoneai	
         injection of POM in Chinese hamsters	 6-176
 6-40   Effect of 3-methylcholanthrene on pregnancy in mice	1	 6-179
 6-41   Data on mouse embryos treated with 0.01 ml  of 3-methylcholanthrene. 6-180
 6-42   Effect of benzo[a]pyrene administration to pregnant mice on
         pulmonary adenoma formation in the progeny	 6-183
 6-43   Effect of benzo[a]pyrene administration to pregnant mice on'skin'  '
         papilloma formation in the progeny	 6-184
 6-44   Summary of epidemic!ogical  evidence of carcinogenicity.........'....• 6-189
 6-45   Summary of observations on observed deaths  and relative risks of
         death from all  cancers, respiratory cancer,  and cancer of the
         digestive tract	 6-192
 6-46  Average annual age, race, and sex specific  mortality rates'per	
         100,000 population for cancer of the lung in the United States,
         1948-49	 6-206
6-47  Age-adjusted cancer-incidence rates per 100,000 population',  by	
        primary site and sex for urban and rural  areas	      6-207
6-48   Ratios of age-adjusted mortality rates,  1950 to 1969,  among white"
        males in petroleum industry counties to  those in control  counties
        by cancer site	 6-211
6-49  Ranking of the cities investigated by the  values of'the atmospheric
        air pollution indices  and death rate in the  population from
        cancer of the 1 ungs and bronchi	           6-214
6-50  Polycyclics content (ppb) of dried microalgae  as a fiinction'of" "'
        geographical location	  6-224
6-51  Effect of aromatic hydrocarbons in marine bacteria	  6-227
                                       xv

-------

-------
•..-.   .                        1.  EXECUTIVE SUMMARY

      This document was prepared in response to the Clean Air Act Amendments of
 1977 which add Section 122 to the 1970 Act requiring the Administrator to decide
 whether atmospheric emissions of polycyclic organic matter (POM) may reasonably
 be anticipated to endanger public health.   This document reviews POM data on
 chemical and physical properties, atmospheric' forms, atmospheric fate and trans-
 port, measurement techniques, ambient atmospheric levels, toxicology, and occupa-
 tional health and epidemiology.   Another document [Multimedia Health Assessment
 Document for Polycyclic Organic Matter, J.  Environ.  Pathol.  Toxicol.  (in press,
 1980)] has been prepared for EPA by Syracuse Research Corporation and covers
 POM exposure from air, food, and water sources.
 1.1  DEFINITION AND FORMATION
      The two POM chemical  groups most commonly found in ambient air are polycyclic
 aromatic hydrocarbons (PAH), such as  the well  known  carcinogen benzo[ajpyrene
 (BaP) and the PAH nitrogen analogs.   In addition,  a  small number of oxygen-
 containing POM have been detected in  ambient air.  The major environmental
 sources  of POM appear to be the  combustion  or pyrolysis of materials  containing
 carbon and hydrogen.  'Although hundreds of  individual  compounds could be formed
 under various combustion or pyrolysis conditions  or  from various carbon-emitting
 sources,  only about 25 parent PAH and 32 nitrogen  and oxygen analogs  have been
 quantitatively analyzed so far in ambient air.  The  failure  to detect POM may be
 attributable to the low concentrations  of some compounds and a lack of an appropriate
                                        1-1

-------
quantitative analytical method for these agents (the presence of over
100 individual PAH in ambient air has been documented, but quantitative data
are not available).
     It is generally agreed that POM compounds are associated with suspended
particulate matter from both mobile and stationary sources, principally
respirable particles.  Many of the compounds are susceptible to oxidation or
photochemical reactions in laboratory experiments, but this susceptibility
varies considerably among individual compounds, depending upon the physical
form of POM tested and the reaction conditions used.  However, available
monitoring data suggest that many POM compounds associated,with particulate
matter are probably stable in ambient air for several days.  A recent study in
Norway has demonstrated that at least 20 PAH associated with particulate
matter are stable enough in the atmosphere to travel from England or France to
Norway.
1.2  ANALYTICAL METHODS
     A number of methods are available for sampling POM from various sources
and for the analysis of collected samples for quantification of individual
compounds.  POM in ambient air are usually collected as particulate matter on
filters.   Some of the more volatile POM may not be efficiently collected, but
the collection efficiency for POM (including BaP) has not been determined.
Because of the large number of individual compounds in the POM family,
careful fractionation and separation procedures are necessary to
ensure accurate quantisation.  Distinquishing between individual
compounds is important because slight alterations in
                                        1-2

-------
chemical  structure modify carcinogenic  activity,  and  thus  alter  the potential
health hazard.
     There are several relatively  simple,  inexpensive analytical procedures
that are  capable of reliably measuring  several  individual  POM compounds.
However,  as the number of compounds to  be  measured increases, the cost and
time of analysis increase dramatically.  It is  apparent that not all of the
individual POM compounds can be monitored  on a  routine basis because of the
cost and  labor that would be required.  However,  the  current practice of
monitoring only BaP seems equally  unacceptable.   Since BaP levels are not
necessarily representative of the  carcinogenic  activity of particulate
fractions collected from ambient air, in the future a multi-component
analysis  of POM compounds would be more desirable.  The actual components
selected  should be easy to determine analytically and should be repre-
sentative of those POM air pollution fractions which demonstrate carcino-
genic activity.
1.3  AMBIENT CONCENTRATIONS
     While POM compounds have been detected in ambient air, the most
extensive data available are on BaP.   The following trends have been noted
for BaP:   the median urban value declined from 3.2 ng/m3 in 1966 to 0.5 ng/m3
in 1975,  an 80 percent decrease; levels in the winter were higher than in
the summer, probably due to greater coal consumption;  the levels in urban
cities with coke ovens were 40 to 70 percent higher than in cities  without
coke ovens, but this is thought to be due in part to higher heating and
industrial emissions in those cities.   Some correlation  between the amount
                                        1-3

-------
of BaP and the amount of other PAH has been noted,  but there are several
exceptions.   These exceptions suggest that the relative amount of each
individual POM.compound depends upon local combustion or pyrolysis sources.
     The accuracy of the absolute values of BaP is  unknown because of the
lack of information on the collection efficiencies, oxidation during col-
lection, or degradation in the time between collection and analysis.
However, the observed trend of overall decreasing BaP levels is believed to
be essentially valid, even if absolute levels are difficult to determine.
Nevertheless this trend cannot be used to imply that the concentration of
other POM is declining or that the carcinogenic activity of POM fractions
of air particulates is reduced.  For the most part, monitoring data for the
other POM compounds have been collected only once at a limited number of
sites.
1.4  HEALTH AND ECOLOGICAL EFFECTS
     The major human health concern over exposure to POM is the possible
development of cancer.  It is well established that POM-containing extracts
of particulate air pollutants are carcinogenic when painted on the skin of
rodents or injected into newborn mice.  Although it cannot be unequivocally
stated that any of the POM are human carcinogens, several of these com-
pounds are among the more potent animal carcinogens known to exist.
     POM gain ready access to the body's circulation either by inhalation,
ingestion, or contact with the skin.  In their parent form, however, POM
produce no major adverse effects.  Instead, these compounds must first be
metabolized by enzyme systems=of the body to produce chemically reactive
                                        1-4

-------
intermediates which are then capable of inducing cancer.   As would be
expected, the alteration of enzymic activity involved in the formation of
reactive POM metabolites can have a marked effect on the carcinogenic
process.  This enzyme activity can be either enhanced or depressed by
exposure to noncarcinogenic chemicals (such as food additives and pollu-
tants), drugs, and naturally occurring agents in the diet.   Moreover, a
wide variation exists in the drug-metabolizing enzyme activity of humans;
and it may be that the degree of this activity is a determinant for the
susceptibility of humans to carcinogenic POM.  The body also possesses
mechanisms to repair carcinogen-induced damage, and the individual capacity
for such processes is another likely determinant of susceptibility.
     Numerous agents have also been identified which either promote or
inhibit the carcinogenicity of POM when simultaneous exposure occurs, and
their mechanism of action is as yet undetermined.  Because of the diversity
of human lifestyles and the multitude of chemicals to which humans are
exposed that may either enhance or diminish their susceptibility, human
response to carcinogenic POM is expected to vary considerably.  Moreover,
the risk of carcinogenic POM exposure to individuals who are in danger of
developing cancer from other causes (e.g., radiation, chemicals, viruses)
has never been fully evaluated.  It is for these reasons that animal studies
may not accurately reflect the cancer risk to humans of POM exposure.
     In studies with animals, an apparent "threshold" level for POM exposures
is often seen, below which no carcinogenic response is produced.  This may
be due to the fact that body defense mechanisms are capable of deactivating
                                        1-5

-------
certain minimum amounts of a carcinogenic substance.   On the other hand,  in
experimental situations it is not surprising to observe deviations from the
predicted number of responses at very low doses, simply because limited
numbers of animals are generally employed.  Furthermore, it has been shown
in animal studies that, even at levels which do not produce tumors, BaP can
still interact with critical cellular constituents (e.g., DNA) in a dose-
related fashion.  Such interaction is regarded to represent an initial
event in malignant transformation.  Therefore, no one can presently say
whether there exists a safe level for human exposure to POM.  However, in
real-life situations most scientists agree that the effects of carcinogenic
POM will most likely be a continuous function of dose, no matter how
smal1.
     Whereas exposure to POM in occupational situations can clearly be
associated with increased lung cancer development, epidemiological evidence
in community settings does not permit a definitive conclusion.  This is
primarily because of the overriding impact that cigarette smoking has on
POM exposure compared to inhalation of POM-containing particulate air
pollutants.  Nevertheless, when smoking habits are taken into considera-
tion, urban residents have a two-fold risk for the development of lung
cancer in relation to residents in rural environments.  Since no other
variables (e.g., lifestyle, medical care, etc.) can adequately account for
this difference, most investigators feel it is reasonable to attribute at
least some of this lung cancer excess among urban dwellers to the higher
relative concentration of POM in city air.  Moreover, it has been shown
                                        1-6

-------
that, in some cases, cities having the highest levels of ROM-containing pollu-
tants in the air also report a higher rate of lung cancer mortality than other
cities.   Even among migrant groups who leave these polluted environments, an
excess of lung cancer still appears while they are residing in an adopted
country.  Thus, it is evident that lifetime exposures to POM may not be nec-
essary to result in an increased cancer risk.
     Due to the relationship between molecular events involved in carcino-
genesis and mutagenesis, it is not surprising that numerous POM produce
mutations in certain animal and bacterial cells in culture.  However, POM
have not been conclusively shown to produce mutations in either animals or man
which can be passed from one generation to the next.   Furthermore, there is no
evidence to indicate that POM in the air may be producing mutations in lower
organisms in the environment, or that POM are presently disturbing the
ecological balance in any way.   POM are apparently not bioaccumulated in any
species, nor are they transferred through, the food chain.   On the other hand,
the high sensitivity of certain fish and amphibians to carcinogenic chemicals
suggests that specific adverse ecological effects may result from low levels
of contamination.
     In conclusion, it is evident that the risk of lung cancer development by
exposure to POM is real.  The magnitude of that risk to the general popula-
tion, however, cannot be accurately determined based on our present knowledge.
It is apparently much less than the lung cancer risk associated with cigarette
smoking, since POM exposure from smoking is far greater than that from inhal-
ing polluted air.   The involvement of POM in producing cancers of other sites
                                        1-7

-------
(e.g., stomach, bladder) is also a possibility which has not been thoroughly
explored in human populations.   Nevertheless, in conjunction with smoking and
the other foreign chemicals (carcinogenic and noncarcinogenic) to which man
is exposed, it is likely that the absolute risk of POM exposure at any concen-
tration is magnified in real-life situations.  Thus, in adopting the most
conservative approach for evaluating the human health hazard of POM, most
scientists conclude that although the risk of cancer by exposure to low doses
of POM is probably very small,  it will not be possible to completely elimin-
ate the risk unless all exposures are prevented.
                                        1-8

-------
                              2.  INTRODUCTION


                          123
     A number of documents  '  '  have reviewed polycyclic organic matter (POM)

 in considerable detail, but a considerable amount of information has been

 developed since then which needs to be evaluated in determining whether the

 release of POM into the atmosphere will endanger public health.  The Clean Air

 Act Amendments of 1977 amended Part A of Title I of the 1970 Act by adding

 Sectipn 122 as a new section entitled, "Listing of Certain Unregulated Pollu-

 tants."  The new section reads as follows:

          Not later than one year after date of enactment of this section
     (two years for radioactive pollutants) and after notice and oppor-
     tunity for public hearing, the Administrator shall review all  avail-
     able relevant information and determine whether or not emissions of
     radioactive pollutants (including source material, special nuclear
     material, and by-product material), cadmium, arsenic and polycyclic
     organic matter into the ambient air will cause, or contribute to,
     air pollution which may reasonably be anticipated to endanger public
     health.   If the Administrator makes an affirmative determination
     with respect to any such substance, he shall simultaneously with
     such determination include such substance in the list published
     under section 108(a) (1) or 112(b) (1) (A) (in the case of a sub-
     stance which, in the judgment of the Administrator, causes, or
     contributes to, air pollution which may reasonably be anticipated to
     result in an increase in mortality or an increase in serious irre-
     versible, or incapacitating reversible, illness),  or shall include
     each category of stationary sources emitting each substance in
     significant amounts in the list published under section lll(b) (1)
     (A), or take any combination of such actions.

     In an effort to comply with the legislated mandate, this publication has

been prepared as part of the basis to allow the Administrator to decide

whether or not POM emissions cause or contribute to air pollution which may
                                        2-1

-------
reasonably be anticipated to endanger public health.   Given that the above
determination is positive, this document will assist the Administrator in
determining which regulatory mechanism is the most appropriate for POM.
     In order to accomplish this objective, this document has reviewed the
following information on POM:  chemical and physical properties, form in
ambient air, formation, fate and transport, measurement techniques, ambient
levels, toxicology, and occupational and epidemiological information.  The
status of control technology for POM has not been considered in this report.
     The scientific literature has been reviewed through 1977 and the early
part of 1978.  Because POM includes literally hundreds of individual com-
pounds, all the papers ever published relating to POM in the environment and
their effects could not be cited.  Emphasis was placed upon those chemicals
and chemical groups which had the highest potential for affecting'public
health.  For this reason two chemical groups -- the polycyclic aromatic hydro-
carbons (PAH), including benzo[a]pyrene (BaP), and the aza arenes — were
focused upon because of their association with carcinogenesis.
                                        2-2

-------
2.1  REFERENCES
1.
2.
3.
Particulate Polycyclic Organic Matter.
Washington, D.C.  1972, 361 pp.
                                             National Academy of Sciences.
Scientific and Technical Assessment Report on Particulate Polycyclic
Organic Matter (PPOM).  U.S. Environmental Protection Agency
Washington, D.C.    Publication No. EPA-600/6-75-001.   1975.

Preferred Standards Path Report for Polycyclic Organic Matter.  U.S.
Environmental Protection Agency.   Office of Air Quality Planning and
Standards.  North Carolina.  1974.
                                       2-3

-------

-------
                         3.   PHYSICAL AND CHEMICAL DATA

  3.1   DEFINITION  AND  FORMATION
       Polycyclic  organic matter  (POM) can  include many chemical  groups as  indi-
  cated by the following  list developed by  an EPA  Task  Force  in 1975:]
                Polycyclic aromatic  hydrocarbons  (PAH) (see  Table 3-1)
                Aza arenes (arenes containing a ring nitrogen) (see Table 3-2)
                Inn no arenes (ring nitrogen with  a hydrogen) (see Table 3-2)
                Carbonyl arenes  (see Table 3-3)
                Dicarbonyl arenes (quinones) (see Table 3-3)
                Hydroxy carbonyl arenes
                Oxa arenes and thia arenes
                Polychloro compounds
                Pesticides (e.g., aldrin, chlordane,  DDT)
 Chemically, any organic compound that contains two or more rings could be
 considered a POM.  However,  of major concern are  the carcinogenic polycyclic
 aromatic hydrocarbons (PAH),  such as benzo[a]pyrene,  and their nitrogen
 analogs, aza and  imino  arenes,  which are formed during organic combustion
           O C C         '      •      '
 processes.  >  >    Because of  their common sources, their  existence in urban
 air,  and the  considerable  experimental data  on their  carcinogenic effects,
 the PAH  and the nitrogen analogs have received the most  attention,  and
 therefore this  report has  focused  on these POM.   However,  some of the
.other arenes  (for example, 7H-Benz(de)anthracen-7-one) have  been  detected  in
 urban atmospheres.3   The latter  compounds  will be considered in this report
 even  though a limited amount of  data is  available.
                                     3-1

-------
     POM may be formed in most combustion or elevated temperature processes
which involve compounds containing carbon and hydrogen.  The amounts and types
of individual chemicals that are formed are dependent upon the starting hydro-
carbons and the conditions of combustion or pyrolysis.  Although the exact
intermediates that are postulated vary considerably, free-radical paths are
commonly suggested at the high temperatures attained in the flame front (500 to
800°C) or under pyrolysis conditions.  Badger  first postulated the synthetic
route to POM formation outlined in Figure 3-1.  His pyrolysis studies were
conducted by passing the hydrocarbon vapor in nitrogen through a silica tube at
700°C.  Although the use of nitrogen atmospheres has been criticized as
lacking relevance to actual combustion,  the reducing conditions are similar
to those of the oxygen-deficient environments of a flame, and the data are in
                                                                   o
good qualitative agreement with the POM combustion products formed.   For
example, Boubel and Ripperton  found that benzo[a]pyrene (BaP) is produced
during combustion even at high percentages of excess air, but that the
amount of BaP is greater at lower percentages of excess air.  Lending support
                                                    Q
to the postulated route to POM, Badger and Spotswood  pyrolyzed toluene,
ethylbenzene, propylbenzene, and butyl benzene and obtained the highest
yields of benzo[a]pyrene with butyl benzene.  Badger and Spotswood  also
found that when 1,3-butadiene is pyrolyzed with pyrene at 700°C, no increase
in the yields of benzpyrenes is observed, which suggests that Diels-Alder
type reactions are probably not important.
     More recent studies tend to confirm most of the mechanism proposed by
       4                     10
Badger.   Crittenden and Long   determined the chemical species at various
flame heights of oxy-acetylene and oxy-ethylene flames.  Compounds identified
                                     3-2

-------
  BaP
Figure 3-1.   Mechanism of benzo[a]pyrene formation.
                            3-3

-------
suggest that the C~ species react to form C«, Cg, and €„ species, and that
reactions involving styrene and phenylacetylene are probably important in
the formation of polycyclic aromatic hydrocarbons.   Also, a C,QH-,Q species
was detected in the gases of both flames and, although its identity is
uncertain, it may correspond to the Cg-C. species (butylbenzene, phenylbuta-
                                                 4
diene, or related radicals) postulated by Badger.
     With the Badger step-wise sequence mechanism,  various free radical
intermediates, and aliphatic and aromatic fuels, one could form hundreds of
POM.
     By far the most significant source of atmospheric POM is pyrolitic or
combustion processes, although some investigators have suggested that some
polycyclic aromatic hydrocarbons in the environment may be synthesized by
plants and microorganisms   and that part of the PAH could be from natural  -
                   12                                                   13
combustion sources.    However, recent studies of aqueous sediment cores
suggest that the concentrations detected correspond relatively well with
energy production from various fuels, which indicates that anthropogenic
combustion may be the major source of PAH even in aqueous media.
3.2  STRUCTURE AND PROPERTIES
3.2.1  Chemical Structure
     The nomenclature of POM compounds has suffered from considerable ambi-
guities in the past due to different peripheral  numbering systems.  For
example, carcinogenic benzo[a]pyrene was named "3,4-benzpyrene" by American
scientists and "1,2-benzpyrene" by European workers, while the noncarcino-
genic isomer, benzo[e]pyrene, was named "1,2-benzpyrene" by American scientists
and "4,5-benzpyrene" by European scientists.  Thus, 1,2-benzpyrene could be
                                     3-4

-------
 the carcinogenic or noncarcinogenic isomer, depending upon where a scientist
 worked.
      The currently accepted nomenclature is that adopted by the Internationa]
 Union of Pure and Applied Chemistry (IUPAC) and by Chemical Abstracts Service.2
 Of major importance are the following rules that determine the orientation from
 which the numbering is assigned: '
      1.   The maximum number of rings lie in a horizontal row;
      2.   As many rings as possible are above and to the right of the horizontal
           row;
      3.   If more than one orientation meets these requirements, the one with
           the minimum number of rings at the lower left is chosen.
 The carbons are then numbered in a clockwise fashion, starting with the first
 most counterclockwise carbon which is not part of another ring and is not en-
 gaged in a ring fusion.   Letters are assigned in alphabetical  order to faces
 of rings,  beginning with "a" for the side between carbon atoms 1 and 2 and
 continuing clockwise around the molecule; ring faces common to two rings are
 not lettered.   Thus,  benzo[a]pyrene would have a benzene ring  fused to the
 "a" bond of the parent pyrene structure (Figure 3-2).
      The following  tables  provide lists of the POM that in most instances
 have been  detected  in  air.   However,  some compounds (e.g.,  7,12-dimethyl-
 benz[a]anthracene and  3-methylcholanthrene)  have been included in  the table
 even though they are  synthesized compounds  which are not formed during com-
.bustion and are not present in ambient air.   These synthetic compounds were
 included because they  are  used as model  compounds  in experimental  carcinogen-
 esis studies and have  provided fundamental  information  concerning  mechanisms
 of tumor formation.  The chemical  names,  synonyms,  structures,  numbering
                                      3-5

-------
benzo[a]pyrene
pyrene
 Figure 3-2.  Accepted nomenclature for benzo[a]pyrene.
                          3-6

-------
system, and acronyms used throughout the report, along with their relative

carcinogenicity, are presented.  The tables (3-1, 3-2, and 3-3) are divided

into polycyclic aromatic hydrocarbons (PAH), aza and imino arenes, and

miscellaneous other POM that have been detected in air.  The structures are

presented in the same way as those in the NAS report, Particulate Polycyclic
               2
Organic Matter,  in that aromatic rings are shown as plain hexagons and a

methylene group as CH2>  The indications of carcinogenicity, taken from the

NAS report, were obtained from several Public Health Service (PHS) surveys of
                                     "I Q-93
compounds tested for carcinogenicity.        In some instances, indications of

carcinogenicity were obtained from individual references; in those cases the

references are provided.  The indications of carcinogenicity which are a

summary of all routes correspond to the following simple code:

                         not carcinogenic
                    +    uncertain or weakly carcinogenic
                    +    carcinogenic
        ++, +++, ++++    strongly carcinogenic

For the most part this system of indicating carcinogenicity is very approximate.

For more detailed treatment of the available data see Chapter 6.   Additional

compounds besides those listed in the NAS report were added in this report

when they have been detected in air.   Older nomenclature are included in

parentheses, and names with an asterisk indicate disagreement with standard

numbering.
                                     3-7

-------
                       a
                    ..  o
                    01  CJ
                    e  o
                    H
                                                                                   u  o  o      -
                                                                                   0)  p- p.     ^)
                                                                                   o.          u
                                                                                   O  00 00     3
                                                                                   S  c  c     n
                                                                                   O.-H -H     U
                                                              I
                                                                        3-8

-------
                       (U
                    01  C
                    C  a)
                       -
                   A  co
                   1-1  I
                   O  CM
                                                                      X
                                                                        CM
o
a.
                                                                                                             -* VO
                                                                                                             00  •
                                                                                                             CO CM
                                                                                                                 O
                                                                                                                 o
                   o  e
                   d  a)
                   d)  o
                   o  ra
                   •U  C
                   C  cd
                   Q  N
                   c  vo
                   O)   -
                  •H   i
                  O  CM
                                                                      CM
                                                                     O
                                                                            +
                                                                            +
       vO
       vo
       CM

        I '
       VO
       vo
       CM
                                                                                                                 vo
                                                                                                             O  rH
                                                                                                                 CO
                                                                                                                 CM
                                                                                                                CO

                                                                                                             CM CM
                                                                                                             CM   •
 O
CJ
  I
co
 d)

3
                      s
                  gj  O
                  C!  O
                  iu  cd
                  U  M
                  td -a
                  •— •  I
                  C   «
                  d) vo
                  •H CM
                  a   «
                                                                      CM
                                                                      IN

                                                                    CJ     +
                  O
                  o.

                  I
                                       01
                                       M
                                       3
                                      4J

                                       O
                                       3
                                       M
                                                                           a
                                                                           0!
                                                                           60
                                                                           o
                                                                           c
                                                                           td
                                                                           CJ
O
M
                                                                                         •H  c  a
                                                                                         AJ -H -H
                                                                                         V"  O  O
                                                                                         
                                                                                         O  60  00
                                                                                         1-1  c  e
                                                                                         CUi-l -H
       td  

          o  o
          jj CO
                                                                         3-9

-------












•






CO
•y*
o
CO
C£
«c
o
CD
OS
o
>-
-T*
1 i
O
»— 1
[_

2g
§
•a:
0
i— <
o
>-
o
1
o
o.


*
^3"
0
O
1
CO

OJ
JQ
(2
f_












rl m
*


o

35
3
U 1 P4

Vr>
.
r- «







^-%
e e
QJ Q)
•s -H
c a /~\

o a) / \— /
C- Cu / \

os \ y—\
•-• 0) \ / \ CM
° "? \_/ pT1 ,
C *cr co -f- fii
0 •> i-IO
\^»











ai ci 	 m
•g "" \C **/*
e ) / o
«» / \ . ^
g oA «A 	 * *,.
£ ^V^ /H- A H , *
X O 3 -H S
1 a a s 2
6u o to u
W pt4 U <






VD m
r-l CM




OO
CM CO
r*. •
CO CS
m
n \o
C^ CSJ
r*.
en en

\o
o O
r-». -
cn  • OJ
o CM  4-1 O
*J -H -H
MOO
OJ P. C. 11
D. M
O 60 60 3 "
U C C3 « G
C.-H -H (0 U <1)
<-) i-l 0) >
,^-| M ^-| VJ 6 60 tH
CO QJ O P* COO
O S CQ t-4 CO
w o X
>> &. • OJ
3-10

-------
                     .

 C  •
 Cl) vO
                                                    + 1
 O
 o:
 CJ
 o


 o
 Q.
 0)   14

§ §
                  o &
                  N  I
                  C CM
                                              EC
                                                r-*
                                                r-H
                                              u
                I
               CTl
               CO
 lO
                 1
                 I
                 I
c
0)
00
o
                                                                                 QJ 4J 4J

                                                                                 tH C C
                                                                                 4J -H 5
                                                                                 J^ O O
                                                                                 « 0.0.


                                                                                 O 00 00
                                                                                 u c c
                                                                                        c
                                                                                     (J  (1J

                                                                                     OOr-l
                                                                                     o  o
                                                                                    ,J  w
                                                                        3-n

-------
                     e  B
                     C3  C9
          §
          T-H
          PM
                                                                                  O CO
                                                                                  i-l CT>
                                                                                  •-I CO
                      CJ  91
                      C  C
                      2  01
 O
CJ
CO

 o
                      a 01
                      c B
                      0) Q)
                                                                                •H c e
                                                                                4J -H -H
                      •n
                      B
                      3
      00
 rt    o
i-H    B
 3    -rt
 B    2
&    3
                  I
                 O  CO 00
                 u  c  c
                 (2.^1 -H
                    4J rH
                                                                                                    ui  0)
                                                                                                       >
                                                                                                    COr-j
                                                                        3-12

-------
                                   0-  to
o
CQ
££
•f.
o
O
on
o
CD
a:


»5
OO
                                                     3-13

-------
                01
                g
                o
                m

                o
                N


                <0
                                                          OO
                                                                                (0

                                                                                (1)

                                                                                0
                            t-H .0
                            -* a
                            co 0)
              a)

              g
              O



              I
                                                n)
§


4J

O
               •S

              .§
               •H
                4J
                O.
                p

                co  e


               *:
                O  O
                C  

                OOr-l
                o  o
                 X


                 I
                                                           3-14

-------
                                                                              _- O
                                                                              8 ^
                                                                              sf rt
                                                                              -* rn
                                                                              \O  .
                                                                              tn -a-
 ,>.£,
 C. P. £.
i—• N N
JO C C



 N I  t
 c c-* *a-
  K
CQ i-t CO
                                 a ^
                   t-s
                   o. a) 41
                   2 c ti
 I
CO
 a>
                    ^ f
                                                                  o o
                                                               •H e c
                                                               4-1 -H vH
                                                                O 60 00
                                                                H C C
                                                                a. -H -H
                                                            3-15

-------
                       U B  01
                    a  c a)  e
                    e  a n  «
                    01  U >» W
                    14  X W  >*
                    X Q..C  CU
                    O. M O  N
                    .S C-—l C
                                                         _    +   p,

                                                         "U  +   |
                                                          csl      e5
                                                         0+5
                                                                                      t*. c*i  a)
                                                                                      rH  •  C
                                                                                      en tn  a>
o
o
a.
 o
CJ



•H
U
•H
01
DO
O
c

o




R
rj
0
a « -
a) u *J
-H C C
U *H *H
U O O
0) O. C.
P*
O 00 60
14 C C
p. i-l i-t
U iH
iH t-4 -H
A (U O
a X oq

n
e

o
§
a
0)
S.
f-t
o



c
u 
c! q o
* ,3 w

X
rrt
                                                                3-16

-------
                                                                    co
                                                                                                           o
                                                                                                        o o
                                                                                                           CM
                                                                                                          I vO
                                                                                                           co  rd
                                                                                                        i 01
                  P. h
                 •—• (>-,
                 •X3 (X
                  O (U
                  l   c
                 CO (U
                  • rH
                 CM  >>
                  - C
                 i-H  O

                  O  O.
                  C  I  .
                   C -H
                 •H ra -a
                                                  CM
                                                  CM
                                                 U     I
 o
CJ
CO

 0
                                                      o
                                                      c
                                                      3
I
o
                                                                                   •H C  C
                                                                                   •M *H -H
                                                                                   H O  O
                                                                                   
-------
              o
             cj
                                                           CM
                                                          O
                                                                              oo  «     tJ
                                                                              Q) 4J  4-1     O
                                                                             •H  a  a    H
                                                                              JJ «H -H
                                                                              VJ  O  O      -
                                                                              
-------
                      a) CD
                      55
                      o  o
                      01  (S
                     1—1 N

                     Jig
                      N  .0
                      C  I
                        -
 n  o  o
 0)  0. CL
 p.
 o  oo oo
 u  e  c
 O.-H  -H
    4J  rH
rH rH  -H
                                                                                                     O
                                                                                                     H
                                                                                                     0)
                                                                                                     >-l
                                                                                                     3
                                                                                                     10
                                                                                                     in
                                                                                                     CD
                                                                                o
                                                                                a
                    E  00 rH
                    coo
                      (J  CO

                    x"
                    tl)
                    E
                                                                    3-19

-------
                    01
                "O  h
                •H  U

                 l-l  03
                 U  N
                 CS)  C
                1—1 0)
                 N VD

                 C   •
                .0  I
                •H -3-
                                                           "U.
                                                            CM
UJ

UJ


-------
                         It
                      
                     >-! M
                     o oj
                     o u
                    i— i N
                     id c
                     0)
                    CO
z
  T*™


SC~
  vC
  i--

O
CM

co


 O)


s
 ITJ
                    id
                   CJ
                  J
      c
      01
      oo
      o
      c
      •rt
      o
      t~l
      cd
      u
I
O
                                                                                          ID   «  -
                                                                                          
-------
CO
LU
               ca o
a
CVJ
 0)
               ^^ c
               J51
                                                               •H C  C
                                                                O 60 60
                                                                o 53
                                                   3-22

-------
                 
-------
                  Ot
                  c
                  cd
                  B
                  01
CO
LU

UJ
OH


 JD
 
-------
                      .1
                      C  3
                      cu  o-
                      j=  o
                      *J  en
                      e -H
                      rH <*>
                      <4-(  •»

                      CO (N
                          I

                         g
                         (U
                         (1)
                         c
                         •H

                         S
 O

 > o
                      D. J=
                      nt  P.
                      N  x-<.
                      <  C
    o
    N
    C
    0)
    CQ
 £1
 O
OJ
 i
co

 o>
                        •O
                        •H
                      O .0
                      3 CM
4!  §
 I   CU
si- 73

   3
    I

    I
   in
                     I
                     o
                     c.

                     I
                                               3
                                               e
                                               H
 O
 C
•H
 O
 h
 CO
CJ
                                                                  w   *>  «
                                                                  CU  4J  4J
                                                                  •H  C  C
                                                                  4-1  -H  -H
                                                                  MOO

                                                                  Cu
                                                                  o  oo so
                                                                  M  C  C!
                                                                  D.-H  -H
                                                                  co  a)  o
                                                                  CJ  S  03
                                                                 •C
                                                                 CU
                                                                                                   O
                                                                                                   E-i
o
o.
CO
         U  CU

      e M ,3
      coo
         H-l CO
                                                                    3-25

-------
            Cl
            B
            0] x^
            u o
            « C
            b O
            f D
            U .6
            C u
            O C
            Sg
            N A
oi
UJ
1
O
             k
O

CJ
g
co
§
             I
3
UJ

CO
 CO
  I
 CO

 (U
 (O
             I
             s
00
o   e
5   -I
u   o
                                                       U-S.5
                                                       MOO
                                                       
-------
 3.2.2  Physical Properties
      Most of the POM compounds are high melting/high boiling point solids that
 are extremely insoluble in water.55  With the exception of the basic nitrogen
 heterocyclics (e.g., quinolines, acridines), the major POM compounds do not
 dissociate and, therefore, are found in the neutral  fractions of extracts of
 particulate material.5'6
      Only limited physical data are available on the numerous POM compounds.
 The information that is available is tabulated in Tables  3-1, 3-2,  and 3-3.
 Melting points  are available  for most compounds  and  most  values  are consider-
 ably over 100°C (phenanthrene [101°C]  and benzo[c]phenanthrene [68°C]  are
 unusual exceptions).
      Vapor pressures  of the pure  chemical  are  available for nine  PAH.
 Depending upon  the  ring size  and  molecular weight of the  compounds, the vapor
 pressure  at 25°C of the pure  compound  varies  from 6.8 x 10"4  Torr for phen-
 anthrene  (3 rings and 14 carbons) to  1.5  x 10"12  Torr for coronene  (7 rings
 and  24 carbons).24'26   This property  has  considerable impact  on the amount
 of PAH that remain  adsorbed on particulate matter  in the atmosphere and
 retained  on particulate  matter during  collection of air samples on glass fiber
 filters.   In one study  the following compounds experience the indicated
 losses at  the various times when filtered air is drawn over particulate
matter at  1.2 cubic meters per minute:1'27  (1) tetracyclic or larger PAH, 2
hours - no  loss; (2) tetracyclic PAH, 24 hours - some loss;  and (3) penta-
cyclic or  larger PAH, 3 weeks  - no loss.  The actual  retention of POM on
particulate during collection  is likely to depend upon temperature,  face
                                     3-27

-------
velocity of the air with respect to the filter, and the adsorption character-
istics of the individual POM.
     The ultraviolet absorption spectra are also available for many compounds
(Table 3-1).  Most of the PAH absorbed light strongly at wavelengths longer
than 300 nm.28  The absorption spectra for benzo[a]pyrene (Figure 3-3) are
typical.  Because these compounds absorb light at wavelengths found in sun-
light (>300 nm), it is possible that they may photochemically react by direct
excitation  (see Section 3.3.1).  However, the available spectra were taken for
PAH in organic solvents, while PAH in the environment are usually found
associated with particulate  matter and adsorbed PAH may have considerably
different absorption properties.
3.2.3  Atmospheric Forms of  POM
     Relatively little  information is available on the exact form of the
various  POM in the atmosphere.   Because  of  the high melting points  and  low
vapor pressures of most POM, the compounds  are generally  considered to  be
associated  with particulate  matter,  either  as  pure material or  adsorbed to
                          2
other particulate matter.
      Some information  is available on  the relationship of POM to particle
 size, and this  has  particular significance  with  respect to respiratory  tract
 entry,  retention,  and  deposition.   The aerodynamic size of particles  will
 determine how much  and where the particles  are deposited.   Relatively large
 particles (> diameter  of 10  to 15 pm)  are deposited in the nasal and  oral  cavi-
 ties.   Smaller particles will pass into the tracheo-bronchial  regions.   The
 actual  deposition efficiencies within the respiratory tract will vary consider-
 ably.   They are dependent upon the dimensions and configurations of the
                                      3-28

-------
CO
n
CM
              200
  300


WAVELENGTH, nm
400
    Figure 3-3.  Absorption spectrum of benzo[a]pyrene  in ethanol-28
    a, 4030 (3.60); p, 3845 (4.44), 3635 (4.36),  3470  (4.10), 3300


                                        274° (4-50h  "'• 2655 (
                                 3-29

-------
individual air path, the pattern and depth of respiration,  and the  character-
                        29
isties of the particles.
     The particle size also affects the residence time of particulate matter.
Urban aerosols appear to have a residence time without precipitation of 4 to
40 days for particles less than 1 urn in diameter and 0.4 to 4 days for par-
                              30
tides 1 to 10 (jm in diameter.
     Tebbens et a!.31 have separated particulates into different size frac-
tions before extraction.  They found a constant weight of BaP per unit weight
of carbonaceous particles for all particle sizes.
     DeMaio and Corn32  collected particles with a two-stage elutriator.  The
 first stage collected 50 percent of particles with a  diameter  of 4.6 urn and
 100  percent of 6.5  urn particles.  The  second stage, which  consisted  of a
 glass fiber filter  paper,  collected 100  percent of all particles greater than
 0.3  urn.   The  amount collected in the two stages varied considerably  for the
 six  compounds analyzed  and depended upon the sampling period.   Only  5  percent
 of the  BaP was detected in stage one,  while  more  than half the pyrene  was  found
 in stage one.   Kertesz-Saringer and coworkers33  used a four-stage  Casella
 cascade impactor backed up with a  glass fiber filter and determined the BaP on
 various size fractions at various  times (September 11 to September 24; October
 15 to November 14;  and November 21 to December 16) of the year in Budapest.
 Their results (Figure 3-4) agree with those of DeMaio and Corn32 in indicating
 that the great majority of BaP is  in the respirable size range.  In winter,
 when the concentration of BaP is the highest, the pollutant is associated more
 with smaller particles.  Also, about half the weight of atmospheric BaP is
 associated with particles of radii less than 0.15 [jm.
                                       3-30

-------
Ill
99
98
97
96
95

90
LU
oc  80
LU
CL
CO
2  70
   60
   50
   40
    30
                                                                      i i
     0.1
                                                    • 9/11 to 9/24
                                                    O 10/15 to 11/14
                                                    A 11/21 to 12/16
                                                    D AVERAGE
                                                     I    i   i  i  i  i
                                    1.0
                              RADIUS(r),Mm
10
     Figure 3-4.  Size distribution of particles containing benzo[a]pyrene.
                                                                           33
                                      3-31

-------
     Katz and Pierce34'35 have conducted a detailed study of the relationship
of PAH to relative size of particulate matter in Toronto, Ontario.   Partic-
ulates were collected at five locations by a five-stage Anderson Hi-Vol
cascade impactor.  The size ranges (urn) for the five stages were:  1 > 7.0;
2 = 3.3 to 7.0; 3 = 2.0 to 3.3; 4 = 1.1 to 2.0; and 5 < 1.1.  Twenty PAH
compounds were monitored, but since the data for the other compounds were
similar to data for BaP (Figure 3-5), only a few data on other isomers were
presented.  A seasonal' relationship to submicron BaP particulates was  found,
with the highest mass fractions of submicron BaP particles  (~ 60 percent total
aerosol mass) found during winter.  At two locations, 40 to 60 percent of the
PAH was associated with submicron particles; at another  location only  6 to
14 percent PAH was thus associated.   Differences were also-found between the
fractional mass  of submicron  BaP particulates  samples taken at 50 feet and at
300 feet above the ground at  different  locations.   Similar  size  distribution
curves were  also found for two  oxygen heterocyclic PAH.  These results
demonstrate  that BaP  and  PAH  are  associated  primarily with  small, respirable
particles.
      In  elevated-temperature  combustion systems,  substantial  amounts of  POM
probably exist in the vapor  state.   However, they appear to condense or  adsorb
on other particles rapidly once they are away from the high-temperature  areas.
Thomas et a!.36 determined the BaP in deposited soot in the upper portion  of
the glass chimney of a combustion system and compared it to the BaP in soot
 collected on a filter.   No  difference was detected, suggesting that BaP
 rapidly adsorbs onto particulate matter as soon as it leaves the combustion
 area.  Thomas et a!.35 also  determined the amount of BaP from the combustion
                                      3-32

-------
 E
 o
 o
 o


 1
 a.
LU
O
ec
100


 50


 30





 10



  5
                                                                MacDONALD

                                                                 CARTIER
s
100



 50


 30
   10
o
cc
LLJ
O.
                 BATHURST-S
                                                       .COLLEGE-S
                                                      COLLEGE-W
                                       STAGE NO.
          Figure  3-5.   Percent mass of benzo[a]pyrene in relation tc
          stage number for five  sampling sites  in Toronto,  Canada.3'
                                       3-33

-------
system collected on a filter (maintained at 80° to 90°C) or collected in a
water and cold trap.  Even at the elevated temperature of the filter, 96.3
percent of the total BaP was found on the filter.  Sublimation of BaP off the
particulate trap was ruled out.
     Krstulovic et  a!.37 have placed polyurethane plugs behind a standard
glass fiber filter  (1.0 |jm retention with 98 percent efficiency) to  collect
particles that pass through.  The authors refer  to these particles as being  in
the  gas phase because they act  as gas  or vapor in that  they  follow fluid
streamlines and  are subject  to  Brownian movement.  Concentrations of several
individual  compounds  in  air  samples  taken  in  Rhode Island  were  greater  on the
"foam plug than  on the filter.   For  example,  the  concentration of
 dibenz[a,c]anthracene in Providence was 3709 ng/1000 m3 on-the foam  plug but
 806 ng/1000 m3  on the filter.   These results are consistent with those  of
 Pierce and Katz,34'35 which indicated that sizable amounts of PAH will  be
 associated with particles less than 1 urn in diameter.
      Natusch and Tomkins38 have developed a mathematical description of the
 process of adsorption of PAH vapors on  fly ash particles.   Their calculations
 indicated that PAH would almost certainly exist as vapors at temperatures
 (>150°C) encountered in the stack systems of  fossil-fueled power plants.
 However.at ambient temperatures (-10°  to 30°C), "essentially quantitative
 adsorption  [to  particulates] would  be expected."38   Their  theoretical  study
 also  suggests that since  "the  extent of adsorption  of  PAH is proportional to
 the frequency  of collision  with the available surface  area,  then adsorption
 will  result in  small  respirable particles  being preferentially enriched (per
  unit mass) in  potentially carcinogenic PAH
,,38
                                       3-34

-------
3.3  ATMOSPHERIC CHEMISTRY
     The POM compounds emitted to the atmosphere have been shown to have
considerably different stability.  Unknown, however, is whether the degradation
products are more or less toxic.   Recent evidence by Pitts and coworkers suggests
that in some instances (e.g., nitrogen dioxide reaction with benz[a]pyrene to
form 6-nitrobenzo[a]pyrene and a mixture of 1-nitro and 3-nitrobenzo[a]pyrene)
the products are more mutagenic.
     The atmospheric stability of different POM compounds is dependent upon
such factors as molecular structure, amount of available light, and presence
of oxidizing pollutants.    Also,  because POM in the atmosphere are mostly
found in the atmosphere adsorbed to particulate matter, such factors as
particle size (affecting rate of deposition and surface area exposed), porosity
of the particle, and adsorption factors are also likely to be important to
environmental stability.   The rates that have been measured vary considerably
depending upon the experimental conditions.
3.3.1  Photooxidation
     Most investigators have used light sources that simulate sunlight (>300 nm).
However, the form of the POM being photolyzed has been much more varied.
Solutions of compounds are one of the most convenient forms for photolysis
studies, but the relevance of solution photochemistry to photolysis of POM
adsorbed on parti cul ate matter is probably remote.   Kurats'une and Hirohata40
photolyzed 13 PAH in cyclohexane and dichloromethane with sunlight and with
fluorescent lamps filtered through glass to provide wavelengths longer than
300 nm.  They found that naphthacene was the most unstable of the PAH tested
and that benzo[a]pyrene and alkyl derivatives of benz[a]anthracene were labile
to light.  The following compounds were stable under the experimental
                                     3-35

-------
conditions:  phenanthrene, fluorene, pyrene, fluoranthene, chrysene,  benz[a]-
anthracene, and benzo[k]fluoranthene.  In another study by Kuratsune,   photol-
ysis of BaP in benzene (light >280 nm and oxygen was present) allowed the
isolation of 6,12-, 1,6-, and 3,6-benzo[a]pyrenequinones.
     Because POM in the environment are usually found adsorbed to partic-
ulate matter, many investigators have studied the photolysis of POM natur-
ally occurring on particulates from combustion, POM coated on some solid
particles, and POM coated in thin films on  a smooth surface.  Typical of
                                                                       42
photolysis studies of pure POM adsorbed on  solids is the work of Inscoe
who examined photodecomposition of  PAH during thin layer chromatography.
She exposed 15 PAH adsorbed on four different adsorbents  (silica gel G,
aluminum  oxide G, cellulose powder,  and acetylated cellulose) to ultraviolet
light  and ordinary room light.  No  changes  were  noted  for phenanthrene,
chrysene, triphenylene,  and picene, but pronounced changes were detected for
the other 11  compounds  when silica  gel and  aluminum oxide were used;  less
dramatic  changes were found with  the other  two  adsorbents.   Exposure to
ordinary  room light  gave slower  but similar results.   The 11 compounds
that were photosensitive were anthracene,  naphthacene, benz[a]anthracene,
 dibenz[a,c]anthracene,  dibenz[a,h]anthracene pyrene,  benzo[a]pyrene, benzo[e]-
                                                            42
 pyrene, perylene, benzo[ghi]perylene, and coronene.   Inscoe   also identified
 1,6-pyrenedione and 1,8-pyrenedione as photoproducts  of pyrene.   As  will be
 discussed later, some of these compounds appear to be fairly stable  when
 adsorbed on natural  particulate matter and, therefore, the above results may
 be only  indicative of what occurs during thin layer chromatography.
                                      3-36

-------
                        ft O
      Andelman and Suess   attempted to simulate photodecomposition of PAH  in

 water by adsorbing BaP on calcium carbonate and irradiating aqueous suspensions

 with white fluorescent light.   BaP was found to be  unstable under  laboratory

 illumination  with 80  to 90 percent lost in  17 hours in  15  to 50  percent

 dioxane-water;  but it is apparently somewhat dependent  upon the  liquid system

 since BaP is  stable for 17 hours  in 100 percent dioxane or 50 percent methanol/

 water.

      A couple of  investigators  have examined the effect of some  common air

 constituents  on the photochemistry of  adsorbed  POM.  Jager and Rakovic44

 adsorbed  pyrene and BaP  to fly  ash and alumina  from an  acetone solution and

 irradiated the adsorbed  compounds  in a quartz tube  filled  with 10 percent S0?

 in air  at room temperature and  at  60°  to 70°C.  From irradiation of pyrene at

 room  temperature, the authors were  able to isolate  many compounds containing

 sulfur, including 1-pyrenesulfonic  acid (15 percent); from BaP,  benzo[a]-

 pyrene-4-sulfonic acid was  detected.  Whether this  type of reaction occurs in

 ambient air samples is unknown since the experimental concentration of SO,, is

much higher than is normally present in ambient air.

     Geacintov   studied the photolysis of 20 PAH adsorbed on solid poly-

 styrene fluffs in the presence of oxygen and nitric oxide.   He found neglig-

 ible photochemical degradation compared to quenching by the paramagnetic

gases, 02 and NO.   However, the PAH were efficient photocatalysts for the

formation of singlet oxygen, a reactive oxidizing agent which may react with

the PAH.
                                     3-37

-------
     Perhaps the earliest study of POM photochemistry was conducted by Falk
and coworkers.45  They noted that the amount of BaP in ambient air compared
with other PAH was greater than one would expect on the basis of amounts
emitted and thought this could be explained by a difference in atmospheric
stability.  They studied the stability of two forms of PAH:  "pure" (dissolved
in solvent and applied to a filter) and "adsorbed" (combustion particles
collected on filter-sample divided into control and exposed sample).  The
samples were exposed to air and to a  synthetic smog (1 hour of exposure
equivalent to approximately 100 hours of exposure  to  natural  smog) with and
without irradiation.  Their results  are presented  in  Table 3-4.   For  most
compounds,  destruction  of the  "pure"  samples  is essentially  similar in  dark or
light.  However,  BaP  stability is significantly different with  and without
light.  Also,  the lower rate  of reaction  of the adsorbed form of BaP  suggests
that adsorption may provide some protection from  photooxidative processes.
The results for pure unadsorbed chrysene  with light are  difficult to  explain.
 Falk et al.47 concluded that the three-ring compounds disappear rapidly and
 that differences in atmospheric stability can explain why the ratio of pyrene
 to BaP is reversed following atmospheric residence.  Also, adsorption on
 particulate matter seems to have a stabilizing effect compared to coating on a
 filter.          .    .
      Tebbens et al.31 also used the  filter paper procedure of Falk et al.
 to study photolysis of BaP by sunlight or simulated  sunlight.  With BaP
 deposited on various filters  from an ether solution, different  filter material
                                                                      31
 resulted in different  losses  (60 to  92 percent loss).   Tebbens  et al.   also
 studied the chemical modification of BaP and perylene in smoke  passed through
                                       3-38

-------













CO
0
CO
o:
- I-H
—i a; .
O LU
Q- :c
CL.
Lu oo
0 0
0 
.a
(O
H—










+j
o
a
in
c
o
"O
Q)
.a
S-
o
in
~a

OS.
i— i.
JZ
c
en
o

co



s_
•P x:
co oo
^_

c
^- S-

I^J





i.
^ 00
5
3

^ j_
x:
£Si
















J« i oo en co r- r-~ i !
* ^^ ^*^ ' i









^n i i — «^f- o r^. o i i
•— 11




*

' 1 ff)  O 1 r^ LO LT3
i i oo to LO i oo Ji:









^S^^OJ ! 000





^ OO C\J r— CSl | ^Z






? 5 . 5 S ° ! ooo





^ooSl!^0 j ooo





O)
c:
OJ
11 ilia
S; S- -C £ £ -^
5:5 -PI — ii — i jc a> a>
cc 
4->
££>>'- a> a> cu ox:
•=ca.Q.LL.a3cacaoo
3-39

-------
a 22-foot dynamic flow chamber (Pyrex pipe surrounded with fluorescent lamps)
by comparing the difference between irradiated and non-irradiated samples.
They found that irradiation caused disappearance or modification of 35 to
65 percent of the original content and that oxygen, although its presence
increases the rate of photodegradation, is not necessary for photoalterations.
     Thomas et al.36 used a dynamic flow-through chamber that was coupled to
a furnace and controlled lighting having one-quarter the intensity of an
average July noonday sun.  They measured the BaP/soot  at the entrance and
exit of the chamber.  A 58 percent decrease in the ratio was noted with the
light  turned on  compared to when the  light was off (retention time of 40
minutes).  Thomas  et al.35 also examined  the effect  of particle  size  by
coating glass beads (1 g  of either 220 urn or 28  urn diameters) by evaporating
the solvent.  The  beads were  then  exposed to  sunlight in  flasks  and agitated
 hourly to expose new  surfaces.  BaP  on small  beads decreased 1  percent  in  12
 hours; on large beads,  1  percent  in  7 days.   The authors  concluded that the
 60 percent decrease in the 40 minutes retention time of the BaP/soot  in the
 dynamic chamber may be related to the much greater surface area of the soot
 exposed to light.
       Lane and Katz48 have recently concluded that the previous experiments
 were  not representative of actual atmospheric conditions, since the oxidant
 levels used were often too high and  fluorescent lamps are poor simulators of
 sunlight.  They used a dynamic flow  reaction chamber  (20-liter glass jar) with
 0  levels of zero to 2.28 ppm and irradiation from a  Quartz!ine  lamp.  They
  o
 studied the disappearance of 500 ng  of BaP, benzo[b]fluoranthene  (BbF), and
 benzo[k]fluoranthene (BkF),  distributed  in a thin dispersion in  a petri dish.
                                       3-40

-------
 The results presented in Table 3-5 show that BaP reacts rapidly with ozone
 compared to BbF and BkF, even in the absence of light.   This is particularly
 significant because this type of nonphotochemical  oxidation could occur
 during the normal  24-hour high-volume procedure used for sampling ambient air.
 The authors also found that the rate of decomposition slowed in time,  suggest-
 ing that the surface-exposed layer reacted quickly but  hindered the  decomposi-
 tion of subsurface material.   However,  it should be kept in mind that  these
 results are based  upon PAH  coated in a thin film on glass.   Whether  these
 are the same results  that would be obtained with PAH adsorbed on soot  is
 unknown.   Falk  et  al.46  have  shown that soot adsorption  can have a stabilizing
 effect  on  BaP.
      Relatively few studies have  attempted  to identify the  photooxidative
 products.   As indicated  earlier,  Masuda and  Kuratsune41  and Inscoe42 have
 identified  quinones of BaP and  pyrene.   This  is  consistent  with  the monitoring
 results of  Pierce  and  Katz,49 who  tentatively identified five polycyclic
 quinones in air  samples  in Toronto.  The seasonal increase  of polycyclic
 quinones during  the summer is compatible with a photochemical mechanism.
 3.3.2  Oxidation and Nitration
     As indicated  in some studies  in the previous section,  light is not
 necessarily required for the atmospheric decomposition of POM.   Falk et
al.   found that for many PAH (excluding BaP) the amount of loss of an
 individual compound coated on a filter, was the same in air in dark as in
 light.  Also, the amount of loss of PAH adsorbed on soot was very high in a
synthetic smog that had a high oxidant concentration.  In addition,  Lane and
    no
Katz   found that BaP reacted rapidly in reasonably low ozone concentrations
                                     3-41

-------
        Table 3-5.   THE HALF LIVES  OF  BaP,  BbF,  and  BkF UNDER VARIOUS
                              REACTION CONDITIONS4B
         Reaction -conditions
                                Half lives,  hrc
Irradiation
Ozone
concentration (ppm)
BaP
                                                      BbF
BkF
None

Quartz! ine
Q500T/CL
lamp
0.19
0.70
2.28
O.-O
0.19
0.70
2.28
0.62
0.4
0.3
5.3
0.58
0.2
0.08
52.7
10.8
2.9
8.7
4.2
3.6
1.9
34.9
13.8
3.3
14.1
3.9
3.1
0.9
aThe half life was determined from the linear regression analysis of the
 initial linear portion of each decomposition curve.
                                       3-42

-------
 (0.19 to 0.70 ppm) without irradiation and that the addition of irradiation had
 little effect on the rate.  However, the other two PAH (BbF and BkF) were
 not nearly as susceptible to ozone oxidation.   Thus, it appears that POM
 will be susceptible to air oxidation without light irradiation and that the
 reactivity is likely to vary for different structures.
                  39
      Pitts et al.    have demonstrated that BaP deposited on glass  fiber
 filters will  react with 03 and PAN to form oxygenated products and with NO-
 to form nitro derivatives when these gases are drawn through the filters.
 Exposure of BaP  to 1  ppm N02  containing traces of  nitric acid results  in
 formation of  compounds  that were assigned  the  structure  6-nitro-,  1-nitro-, and
 3-nitro-BaP (the latter two isomers  were unresolved  by TLC).   Both the  6-
 nitro-BaP and the  mixture of  1-nitro-acid  3-nitro-BaP were  direct  acting
 frameshift mutagens  in  the  Ames  assay.  With BaP exposed  to  03, PAN, or  smog,
 oxidation products  including  hydroxy-,  dihydroxy-  and dihydrodiol  derivatives
 were detected.   Perylene  under similar  experimental conditions (1  ppm N0?)
 also formed a  nitro derivative,  3-nitroperylene, which was a directly active
 mutagen  in the Ames reversion assay  even though the parent compound, perylene,
 shows no  mutagenic activity with or without activation.  All of these re-
 actions were carried out on PAH deposited on glass-fiber filters.   Whether
 these compounds would react similarly in the form found in the atmosphere
 (adsorbed to particulate matter) is unknown.
            27
     Commins   also provided suggestive evidence of nonphotochemical decompo-
 sition of PAH  by collecting PAH particulate matter from air and comparing the
results of immediate analysis  to results from analysis after 1 year of
storage in a sealed envelope (Table 3-6).   Significant loss of lower molecular
                                     3-43

-------
                o

                U)

                O)

                D
                n.
                 LO
 S- CM
 3   T3
 CO   C
 CO
 0)
        LO    00    r—
        CM    CM    •—
        CM    CO    c—
                                 r—    O
                                             CM    CM
                                             LO    «*
                                             CM    •—
 S-
 D.
 s_
 o

 t
      o
      o.
o
u
               X
               co
               CO
                         en

                           O
                           X
                           en
                     LO
en

  o
  r—
  X

  LO

  LO
                                                               o

                                                                i
                                            CM

                                             I
                                    >>
                                 C           5-
                                 Q)     fl)     dj
                                 t-     =     a.
                                 •>.    cu    i—i

                                       j=    x:
                                 cu    -P     o>
                                 i—i    C    '—i
                                 o     re     o
                                 N    -C     N
                                 e    -p     c:
                                 a>     c     cu
                                 03    
-------
  weight PAH was noted but little loss with the higher molecular weight com-
  pounds was found.  This loss could be explained by evaporation of the lower  .
  molecular weight compounds rather than by oxidation.
  3.3.3  Atmospheric Stability
       The above results would suggest that many POM oxidize or photqdegrade
  at a significant rate under atmospheric conditions.   However, no satisfactory
  quantitative rate data are available.   In contrast to the results reported
  above,  Lunde and Bjrfrseth50 published suggestive evidence that many POM  are
  stable  enough in the  atmosphere  to  travel  long distances.   They monitored air
  samples  in  Norway which  had different trajectories.   They identified 20
  different PAH and determined that samples  with trajectories  from  western
  Europe contained  about 20 times  more  PAH than  samples with trajectories from
  northern Norway or stationary air from southern  Norway.   The  PAH  found in  high
  concentrations under appropriate wind trajectories were phenanthrene, anthra-
  cene, methylphenanthrene/-anthracene, fluoranthene, dihydrobenzo[a&b]fluorenes,
  pyrene, benzo[a]fluorene, benzo[b]fluorene, 1-methylpyrene, benzo[c]phenan-
  threne, benz[a]anthracene, chrysene/triphenylene, benzo[b&k]fluorathene,
 benzo[e]pyrene, benzo[a]pyrene,  perylene, indeno[l,2,3-cd]pyrene, benzofghi]-
 perylene, anthanthrene, coronene.  These results suggest that the above com-
 pounds are relatively stable in  the  atmosphere in the  form in which they  are
 naturally present.
.3.4  SUMMARY AND  CONCLUSIONS
      The POM chemical  groups most commonly found in ambient air are polycyclic
 aromatic  hydrocarbons  (PAH), such as benzo[a]pyrene, and their nitrogen
 analogs,  aza-  and imino-arenes.   Other related  compounds such as carbonyl
 arenes and dicarbonyl  arenes (quinones)  are less  commonly  detected.
                                     3-45

-------
     The major environmental sources of POM appear to be from .combustion or
pyrolysis processes which use materials containing carbon and hydrogen.
Current theory suggests a free-radical, step-wise, sequence mechanism which
with various fuels could result in the formation of hundreds of POM.
     Most of the POM are high melting/high boiling point solids that are
very insoluble in water.  At 25°C, the available vapor pressures of the pure
compounds vary for the individual compounds from 10"  Torr (3 rings) to 10
Torr (7 rings).  The PAH are strong adsorbers of ultraviolet light of wave-
lengths from 200 nm to 400  nm.
     It is generally agreed that most  POM  compounds  are  associated with sus-
pended particle matter, with a  large portion found with  particles smaller  than
1 pm.  However, two- or three-ring  compounds may  be  found partially  in  the
vapor phase.  Whether  POM  are condensed  into discrete particles after cooling
or  are adsorbed on  surfaces of  existing  particles  is still  unknown.  Urban
aerosols  appear to  have  a  residence time without  precipitation of 4  to  40  days
for particles  less  than  1  urn in diameter and 0.4  to  4 days  for particles  1 to
 10  urn  in  diameter.   Thus,  the atmospheric  lifetime of a POM particle will  be
 closely related  to its particle size.
      Experimental  chemical and  photochemical  reactivities  of POM  on  particu-
 late matter vary considerably from half-lives  of less than a day  to  several
 days.   Particularly interesting are the different reactivities noted for the
 various compounds studied and the indication that reaction rates  of compounds
 on smaller particle sizes are likely to be faster than  those of larger particles.
 Light irradiation considerably accelerates the decomposition of some compounds
 but not others.   Also, various experimental conditions  have given contradic-
                                      3-46

-------
 tory  results.  One  report  has  indicated  that oxidation by ozone  is the most
 important decomposition process for BaP  in air, but many other reports have
 documented the photochemical susceptibility of BaP.  However, monitoring data
 suggest that at least 20 PAH are stable  enough in the form found in the
 atmosphere that they can travel long distances with particulate matter.   Thus,
 although an exact residue time of POM is difficult to estimate from the avail-
 able data,  many of the POM compounds are stable enough to remain at signifi-
 cant concentrations  during transport from the source to where they are inhaled.
 Although the  smaller particles  will  have a longer  atmospheric residence  time,
 they will probably also  react faster because  of the greater  surface area
 exposed to oxidants  and  light.  All  the  available  information does support
 previous conclusions that at  least the outer  layer of some POM adsorbed on
 particles is  relatively  reactive.  However, this may mean that there is an
 inner  layer of POM that would be relatively stable and perhaps available for
 elution by biological fluids following inhalation.   The relatively long
 stability of several  PAH based  upon monitoring data is consistent with this
possibility.   Only limited information is available on the products of decom-
position or their toxicologic properties.  Quinones appear to be common
atmospheric degradation products of PAH.
                                    3-47

-------
3.5  REFERENCES

 1    Scientific and Technical Assessment Report on Particulate Polycyclic
     Orqanic Matter (PPOM).  U. S. Environmental Protection Agency.  Washington,
     D. C.  Publication No. EPA-600/6-75-001.  1975.

 2   National Academy of Sciences, Committee on Biological Effects of Atmospheric
     Pollutants.  Particulate Polycyclic Organic Matter.  Washington, D. C.,
     1972.

 3   Sawicki, E.  Airborne carcinogens and  allied compounds.  Arch.  Environ.
     Health.  14(1):46-53, 1967.

 4   Badger  G. M.  Mode of formation of carcinogens  in human environment.
     Natl. Cancer Inst.  9:1-16,  1962.

 5.  Hoffman, D., and E. L. Wynder.  Chemical  analysis  and carcinogenic bio-
     assays of  organic particulate pollutants.   In:   Analysis, Monitoring,
     and  Surveying:  Air Pollution,  Vol. II,  2nd Ed.  A.  C.  Stern (ed.).
     New  York and London,  Academic Press,  Inc.   1968.   p.  186-247.

 6.  Hoffman, D., and E. L. Wynder.  Organic  particulate  pollutants-chemical
     analysis and bioassays  for carcinogenicity.   In:   The  Effects oT  Air
     Pollution:  Air Pollution, Vol.  II, 3rd Ed.   A.  C.  Stern  (ed.)  New York
     and  London, Academic  Press.   1977.  p. 361-455.

 7.  Boubel,  R.  W., and  L.  A.  Ripperton.   Benzo[a]pyrene Production during
     controlled combustion.   J. Air  Pollut. Contr.  Assoc.   13:553-557, 1963.

     Badger,  G.  M.,  and  T.  M.  Spotswood.   The formation of aromatic hydro-
     carbons at high  temperatures.   Part IX.,  The pyrolysis of toluene, ethyl-
     benzene,  propylbenzene and butylbenzene.   J.  Chem. Soc.  4420-4427, 1960.

      Badger, G. M.,  and T. M.  Spotswood.   The formation of aromatic hydro-
      carbons at high temperatures.   Part XI.   The pyrolysis of buta-1,3-diene
      and of buta-1,3-diene with pyrene.   J. Chem. Soc.   4431-4437,  1960.

      Crittenden, B.  D.,  and R. Long.  The mechanisms of formation of pplynuclear
      aromatic compounds in combustion systems.  In:  Carcinogenesis.   Vol.  I.
      Polvnuclear Aromatic Hydrocarbons:   Chemistry, Metabolism,  and Carcino-
      genesis, R. I.  Freudenthal and P. W.   Jones (eds.).  New York,  Raven Press,
      1976.

 11   Suess  M.  J.   The environmental load  and cycle of polycyclic aromatic
      hydrocarbons.   Sci. Total Environ., 6:239-250, 1976.

 12.  Blumer, M., and W.  W. Youngblood.  Polycyclic aromatic hydrocarbons in
      soils and recent sediments.  Science, 188:53-55,  1975.
 8.
10.
                                       3-48

-------
23.
24.
25.
 13'  =ln^-R\V R' u- LaF13™es and J. W. Farrington.  Sedimentary polycyclic
      aromatic hydrocarbons:  The historical record.  Science, 198:829-831,


 14.  Dipple  A.   Polynuclear aromatic carcinogens.  In:  Chemical Carcinogens,
      ton^'D  C , 1976:}'PP  245-3?4S         American Chemical Society, Washing-


 15.  Patterson,  A.  M   L.  T.  Capell, and D.  F.  Walker.   The ring index.   A

      SocieJy/aSon! Tc" 19^427^

 16.  Capell,  L.  T. ,  and D.  F.  Walker.   Supplement I (1957-1959)  to the rina
      index.   American Chemical  Society,  Washington, D.C.  1963.


 1?"
 18.
            '  i'  T:'  and°-.F-  Walker.   Supplement II (1960-1961) to the ring
              American Chemical Society, Washington, D.C.   1964.

       n*s  !r  T-'  and,uD-.F-  Walker.   Supplement III (1962-1963) to the ring
      index.   American Chemical Society, Washington, D.C.   1965.
 19.   Hartwell,  J.  L.,  and P.  Shubik.   Survey of compounds which have been treated
      for carcinogenic  activity.   Public Health Service Publication 149 (2nd
      ed.J.   Washington,  D.  C.,  U.S.  Government Printing Office, 1951.   583 pp.

 20.   Shubik,  P., and J.  L.  Hartwell.   Survey of compounds which have been tested
                ogenic  activity.   Supplement  1.   Public Health Service Publication
              Washington,  D.C.,   U.S.  Government Printing Office,  1957.   388 pp.

 21.   Shubik,  P., and J.  L.  Hartwell.   Survey of compounds which have been tested
      |°T C,arcl1lno(|?nic  activity.   Supplement  2.   Public Health Service Publication
      i^y-z.   Washington,  D.C.,  U.S. Government Printing Office,  1969.   655 pp.

 22.   Tracor Jitco, Inc.   Survey  of compounds  which  have been  tested  for car-
      cinogenic activity  (1970-1971).   Public  Health Service Publication 149
      Washington, D.C., U.S. Government Printing Office.   1974   1667 pp
     Tracor Jitco, ^Inc.  Survey of compounds which have been tested for car-
     cinogenic activity (1972-1973).  Public Health Service Publication 149
     Washington, D.C. , U.S. Government Printing Office.  1976.  1638pp.
     Romh' S' V/'w^T'. C' W' Gou1d' D" H' Liu> H- L- Johnson, D. C.
     Bomberger  and C. V. Fojo.  The environmental fate of selected polynuclear
     aromatic hydrocarbons.   Stanford Research Institute, Menlo Park  Calif
     Environmental Jrotection^Agency, Washington, D. C.  Contract No.' EPA-68-
                        -              C' Pupp'  The vaPor Pressures and enthalpies
                 on of five polycyclic aromatic hydrocarbons.  Can. J. Chem. ,
     2£(.4J: PO/-563, 1974.
                                     3-49

-------
26.  Pupp, C., R. C. Lao, J. J. Murray, and R. F. Pottie.  Equilibrium vapor
     concentrations of some polycyclic aromatic hydrocarbons arsenic tnoxide
     and selenium dioxide, and the collection efficiencies of these air pollutants.
     Atmos. Environ., 8(9):915-925, 1974.

27   Commins, B. T.  Interim report on the study of techniques for determination
     of polycyclic aromatic hydrocarbons in air.  Natl.  Cancer Inst. Monograph
     No. 9, 225-233, 1962.

28   Clar, E.  Polycyclic Hydrocarbons.  New York and  London, Academic Press.
     Berlin-G6ttingen-Heidel berg, Springer-Verlag.  Volumes  1 and 2.  1964.

29.  National Academy of Sciences.  Subcommittee on Airborne Particles.   Air-
     borne Particles. . Washington, D.C.  1977.

30.  Eswea, N.A. and M. Corn.  Residence time of particles  in urban  air.
     Atmos. Environ., 5:571-578,  1971.

31.  Tebbens, B. D., J. F.  Thomas, and M.  Mukai.   Fate of arenes incorporated
     with  airborne  soot.  J. Amer. Ind.  Hyg.  Assoc.,  27:415-422, 1966.

32.  DeMaio,  L., and M. Corn.  Polynuclear aromatic  hydrocarbons associated
     with  particles  in  Pittsburgh air.   J.  Air.  Pollut.  Contr.  Assoc.,  lj>:67-
     71,  1966.

33.  Kertesz-Saringer,  M.,  E.  Meszaros,  and T.  Varkonyi.  Technical  Note on the
     size  distribution  of benzo[a]pyrene containing  particles  in urban  air.
     Atmos.  Environ., 5:429-431,  1971.

34.  Katz, M.,  and R. C.  Pierce.   Quantitative  distribution of  polynuclear
     aromatic hydrocarbons  in  relation to  particle size of urban particulat.es.
     In:   Carcinogenesis, Vol.  1, Polynuclear Aromatic Hydrocarbons:   Chemistry,
     Metabolism and Carcinogenesis,  R. I.  Freudenthal and P. W.  Jones (eds.).
     New York,  Raven Press.  1976.

 35.  Pierce, R.  C., and M.  Katz.   Dependency of polynuclear aromatic hydrocarbon
     content on size distribution of atmospheric aerosols.  Environ. Sci.
     Techno!.,  9(4):347-353,  1975.

 36.  Thomas, J.  F., M.  Mukai,  and B.  D.  Tebbens.  Fate  of airborne benzo[a]pyrene.
      Environ. Sci.  Technol.,  2(l):33-39, 1968.
 37.
 38.
Krstulovic, A. M., D. M. Rosie, and P. R. Brown.  Distribution of some
atmospheric polynuclear aromatic hydrocarbons.  Amer. Lab., p. 11-18,
1977.

Natusch, D. F. S., and B. A. Tomkins.  Theoretical consideration of the
adsorption of polynuclear aromatic hydrocarbon vapor onto fly ash in a
coal-fired power plant.  In:  Carcinogenesis, Vol. 3, Polynuclear Aromatic
Hydrocarbons.  P. W. Jones and R. I. Freudenthal (eds.).  Raven Press, New
York.  1978.  pp 145-153.
                                      3-50

-------
 39.  P tts, J. N. Jr.,  K  A.  Van  Cauwenberghe,  D.  Grosjean,  J.  P.  Schmid,  D.  R.
       ltZ',W'  L- Belser' G*  B'  Knuds°n,  and  P.  M.  Hynds.   Atmospheric reaction
      of polycyclic aromatic  hydrocarbons:  Facile  formation  of  mutagenic nitro-
      derivatives.  Science,  in  press,  1978.

 40.  Kuratsune, M., and T. Hirohata.   Decomposition  of  polycyclic  aromatic
      hydrocarbons under laboratory  illuminations.  Natl.  Cancer Inst.  Monograph
      INO .  y j I I /"" I^oj  1962.

 41.
 43.
 44.
 45.
46.
47.
48.
49.
50.
51
                               __): 805-81  .

42.  Inscoe, M. N.  Photochemical changes in thin-layer chromatograms of poly-
    . cyclic, aromatic hydrocarbons.  Anal. Chem., 36:2505-2606,  1964.

     Andelman, J.  B., and M. J. Suess.  The photodecomposition of 3,4-benz-
     pyrene sorbed on calcium carbonate.  In:  Organic Compounds in Aquatic
     Environments  by S.  D. Faust and J. V. Hunter, p. 439-468, 1971.

     Jager  J.   and M.  Rakovic.  Sulphur-dioxide-induced qualitative changes
     in polycyclic aromatic hydrocarbons adsorbed on solid carriers   J  Hva
     hpidemiology, Microbiology, and Immunology, J8(2):137-143, 1974.

     Geacintov, N   E.   Reactivity of Polynuclear Aromatic Hydrocarbons With
     Up and NO in  the Presence of Light.  Prepared under Grant No.  R801393  U S

     650/ir-74-0?0   P1973Ctl"°n Agen°y'  Washin9ton' D-c-  Publication No.  EPA- '  '


                                                                    IV.
Falk  H  L. , I. Markul, and  P.  Kotin.  Aromatic  hydrocarbons.   IV.   Their
Tate following emission into atmosphere  and  experimental  exposure  to
washed air  and synthetic smog.  A.M. A. Arch.  Ind.  Hlth. ,  13:113-17,  1955.

Falk  H. L   P. Kotin, and_A. Miller.  Aromatic  polycyclic  hydrocarbons  in
Km ™oai^n indlcators of carcinogenic hazards.   Intl.  JY Air  Poll?,
^.^ui— ^uy,  iybu.
hrK           - Katz'  The Photomodification of benzo[a]pyrene,
benzo[b]fluoranthene  and benzo[k]fluoranthene under simulated atmospheric
££?£?' A-C ^6,°I P°11utants i"n the Air and Water Environments
Part 2.  I. A. Suffet (ed.), New York, Wiley-Interscience.  1977.

Pierce, R  C. , and M.  Katz.   Chromatographic isolation and spectral analysis

                           1" ^ al>  °11Ut

                                     3-51

-------
52.
53.
54.
55.
Hecht, S. S., W. E. Bondinell, and D. Hoffman.  Chrysene and methyl-
chrysenes:  Presence in tobacco smoke and carcinogen!city.  J. Nat.
Cancer Inst., 53:1121-1133, 1974.

Hirao, K., Y. Shinohara, H. Tsuda, S. Fukushima, M. Takahashi, and N. Ito.
Carcinogenic activity of quinoline on rat liver.  Cancer Res., 36:329-
335, 1976.

Shear, M. J., and  J. Leiter.  Studies in carcinogenesis.  XVI.   Produc-
tion of  subcutaneous tumors in mice  by miscellaneous polyclic  compounds.
J. Nat.  Cancer  Inst., 2:241-258,  1941.

May, W.  E., S.  P.  Wasik, and  D.  H. Freeman.  'Determination  of  the
solubility  behavior of  some polycyclic aromatic hydrocarbons  in  water.
Anal. Chem. 50:997-1000, 1978.
                                       3-52

-------
                 4.  SAMPLING AND ANALYTICAL METHODS FOR POM

 4.1  INTRODUCTION
      In vjvo bioassays have demonstrated that the greatest carcinogenic activ-
 ity of organic pollutants in air is associated (1) with the neutral  fraction
 containing both polynuclear aromatic compounds (PAH) and their neutral
 nitrogen analogs,  the aza-arenes,  and (2)  with the basic fraction containing
 the basic aza- and imino-arenes.   Therefore,  it is necessary to monitor these
 three  classes  of compounds  in  environmental  samples  both for assessment of
 carcinogenic potential  and  enforcement of  any  pollution discharge standard
 that may be  set to control  their emission.   Excluding  the  natural  sources,
 some of  the  contributing  sources of POM  in the  air are  from  mobile sources
 entering the atmosphere as  vehicular  and aircraft  emissions  and  from station-
 ary  sources  that include  home heating, power plants  utilizing fossil fuels,
 refuse burning, road abrasions, and industrial processes.  The monitoring of
 these POM compounds can be divided  into four distinct steps:  (1) sample
 collection;  (2) desorption of the collected compounds from the collection
 media;  (3) clean up and separation of the desorbed compounds; and (4) detec-
 tion and analysis for quantisation.
     The objective(s) of sampling must be defined clearly before the initia-
tion of sample  collection.  Three primary objectives in the consideration of
POM sampling include determination  of (1) airborne concentrations of total
                                   4-1

-------
suspended particulate matter, (2) distribution of particle size associated
with the POM, and (3) chemical composition of the particles and their
levels in the environment.  In addition to these objectives, the selection of
sampling site(s) and the implementation of appropriate sampling strategy play
important roles in the acquisition of relevant data desired from ambient air
sampling.
     POM can be present in the air samples in three forms:  (1) adsorbed on
foreign particulate matter, (2) condensed POM in suspension with air, i.e.,
aerosol form, and (3) vapor phase.   Evidently, POM from different sources will
be present  in all the three forms, although their proportion will depend on
the  nature  of the source.  Any collection method for monitoring POM  should
take these  factors  into consideration.  The errors in most  sample collection
methods arise from  the inability  to  collect either wholly or partly  all three
forms  of POM, particularly the volatile components.
     The desorption of the collected compounds is no  less problematic.  The
particulate adsorbed part of  the  POM is bound firmly  to the medium.   This
often  leads to  incomplete and unreproducible desorption.  The  selection of
both the method of  desorption and the nature of  solvent  (whenever used) play
an important role in the  recovery of POM  obtained from this step.
     The lack of clean  up and separation  of  POM  in  environmental samples
poses  a serious hindrance to a better understanding of the  composition.   Yet
 separation of hard-to-separate isomers and the  trace constituents  is quite
 important.   Small differences in the structure  of these  compounds may cause
 great differences in the biological  activities.   Also, due  to  the  specificity
 of biological reactions,  sometimes a trace component may produce more adverse
                                    4-2

-------
 biological  effects than the predominant components.   Therefore,  the selection
 of the clean up and separation procedure will  not only depend on the nature of
 the sample  but also on the objective of the analysis.
      With regard to the method of detection, any technique chosen must be
 sensitive and selective to provide interference-free  quantisation of the  trace
 components  as well  as  the  predominant ones.
      Despite the greatest  of precautions,  there  are some  unavoidable losses,
 both  during collection and analysis  of POM.  The losses during analysis
 alone may vary considerably depending on  the nature of sample, method  used,
 and the experience  of  the  analyst.   To secure quantitative  data,  internal
 standards must be  used from the beginning  of the analysis.  This  is  usually
 done  by spiking the  sample  with radioactive compounds such  as C14-benzo[a]-
          14
 pyrene, C  -benzo[a]anthracene, or some other nonisotopic compounds  not present
 in  the original  sample.  A  number of  compounds including o-terphenyl, bi-
 benzyl, and benzo[b]chrysene, have been used for the latter purpose.
      The  instability of the  POM resulting  from volatility, photosensitivity,
 and chemical transformation/degradation dictate  that special precautions be
 used  in handling these samples.  Losses and chemical transformation during
 sampling  and separation caused by these factors  should be avoided whenever
possible.   The  state-of-the-art of the four individual steps required for POM
monitoring  is discussed in the following sections with the above mentioned
difficulties in mind.
4.2  SAMPLING
     The POM sampling methods can be divided into four groups:   (1) mobile
sources, (2) tobacco smoke (3) stationary sources, and (4) ambient air.
                                   4-3

-------
4.2.1  Mobile Sources
     The most important mobile source of POM in the air is vehicular
exhaust.  The sampling methods for POM from this source are treated in this
section.
     Certain vehicular parameters play important roles in the characteristics
of POM emissions from this source.  It is important to ascertain and define
these parameters even before initiating any sampling procedures.  Performance
characteristics of the engine, its condition at the time of experimentation,
and  the  fuel characteristics influence the nature and amount of POM emissions.
The  effects of performance characteristics, such as engine load, speed, cycle
of operation, fuel-to-air ratio,  and  operating temperature, have been demon-
strated  by Spindt.1  Gross2  has  shown the effect of accumulation of deposits
in the  combustion chamber of the engine on the POM emission.  A number  of
investigators3"7  have  demonstrated the  relationship between POM emission
characteristics  and fuel  composition, namely,  its  aromaticity and  PAH content.
      A variety  of procedures are available  for sampling POM from automobile
 exhausts.   It is not possible,  however, to  discuss  all  these procedures in
 detail.   The general principle of sampling  auto  exhausts and  references to a
 few procedures  are  presented here.
      In one procedure,8 the particulate matter and condensable  vapors from
 auto exhausts are collected by a total  condensation trap.   In  another proce-
 dure,9 the exhaust aerosols from the automobile are passed through a glass
 cooler to bring the aerosol to an optimum temperature.   The cooled aerosol is
 then passed through a fiberglass filter and finally through a double filter
                                    4-4

-------
 of silica gel.   The POM collected at different stages of the system are
 individually eluted with organic solvents  for their analysis.
      Recently,  Spindt  used an  air dilution method for the collection  of
 POM from  auto exhausts.   In this procedure,  the hot exhaust aerosol  is
 precooled to 75°  to 170°C by a  precooler and is mixed with clean  and dry air.
 The air is chilled  by passing it through a coil  immersed in a  bath  of  dry ice
 and methanol  to a temperature that gives a diluted sample temperature  of 20°
 to  30°C.   The mixture is  passed through a  pretreated filtering system  consist-
 ing of a  fiberglass  pad placed  ahead  of a  glass  fiber filter.   The  purpose of
 the fiberglass  pad  is to  collect the  bulk  of the large  soot particles  before
 the aerosol  comes in  contact with  the glass-fiber  filter.   The residual  from
 the aerosol  is  finally passed through a Chromosorb-102  trap.   The sample  flow
 rate  is measured  by a dry gas test meter.
      Irrespective of  the collection method used, the  efficiency of collection
must  be known and must be reproducible.   This is hardly the  case for samples
collected  from automobile exhausts.  While some investigators9 have described
the efficiency of removal of particles of certain sizes, they  did not report
the overall efficiency.   By the  use of radioactive tracer, one investigator1
reported the recovery efficiency of BaP from automobile exhausts as low as
6 percent, under no-load condition, but failed to determine whether the loss
was principally during sample collection or the analytical procedure.  It
appears that there is a need for a standardized procedure for the collection
of POM from automobile exhausts  which can  determine the collection efficiency
and reproduce the technique with reasonable accuracy.
                                   4-5

-------
4.2.2  Tobacco Smoke
     POM content in tobacco smoke depends primarily on the
processing methods used for the manufacture of tobacco products, namely, the
nature of the curing and drying process10 and steps involving incorporation
of additives such as flavoring materials.  In addition, substantial amounts
of POM originate from the paper used to roll the tobacco of the cigarette.
In determining the POM content in tobacco smoke, it is important to realize
that mainstream smoke and sidestream smoke do not necessarily have the same
composition.10  The sampling method, therefore, should specify the part of
smoke that has actually been sampled.
     For collection of mainstream smoke, the standard method consists of
using a fully automatic smoking machine where the cigarettes are smoked to a
butt length  of 23 mm under  standard conditions.  These include a puff volume
of 35 ml, puff duration of  2 seconds,  and puff  frequency  of one per minute.
The collection of particulate matter  is  accomplished by an electrostatic
precipitator.10  In some methods, a cold-trap system replaces the  electro-
                                                            12
static precipitator for the collection of  smoke condensates.
      For  collection of sidestream smoke, a wide glass  tube fitted  on the
cigarette holder at the  end opposite  to the smoking side  has been  used.
The  sidestream smoke  from  this  tube  is drawn through  a side-arm and collected
by means  of two  traps  in  series,  packed with glass-wood and cooled to  -70°C.
4.2.3  Stationary Sources
      Emission from stationary sources can  be divided into two  groups:   (1)
 stack emissions  and (2)  fugitive emissions from road-way  dust.
 4.2.3.1   Emissions from stacks—Several methods are available  for  collecting
 POM samples from these sources which include fossil fuel-fired steam  generators,
                                    4-6

-------
 home heating stacks, municipal incineraters, Portland cement plants, and coke-
 oven plants.   Only a few collection methods are described here.
      The EPA Method-5   is designed to determine the particle emissions from
 stationary sources under isokinetic conditions.   The sample is isokinetically
 withdrawn from the stack by means of a heated probe (minimum temperature of
 250°F at the  exit end)  to prevent condensation of vapors and passed through a
 heated (minimum temperature of 225°F) filter holder containing a precondi-
 tioned glass-fiber filter.   The particulate-filtered gases  are subsequently
 passed through four impingers,  the first  two containing  100 ml  water,  the
 third one empty,  and the fourth one containing 200 g silica gel.   All  the
 impingers are  cooled in  an  ice  bath.   The exit gases from the  last impinger
 pass  through a dry test  meter  for recording  the  flow rate.   The flow rate is
 adjusted  by the valves attached to  the air pump(s).
      Doubts have  frequently  been  expressed regarding the collection  of volatile
 organic compounds  by the  EPA Method-5  sampling procedure described above.   The
 adsorbent sampler  method  developed  by  Jones  et al.14 utilizes a Tenax bed to
 adsorb these volatile organic compounds and  reduce the sampling errors.   The
 stack gases, according to this  method, are sampled isokinetically by a sampling
 probe and passed through  a heated filter  as  in EPA Method-5.  After  leaving the
 hot filter, the emissions are cooled in a glass coil  (120 by 0.8 cm) of the
 adsorbent sampler and, then, pass through a Pyrex frit and into a cylindrical
 column of Tenax bed  (7 by 3 cm) containing 12 g of the resin.  The flow rate
 through the adsorbent sampler is typically 14 liters/min.  The cooling coil  and
Tenax adsorbent are maintained at a constant temperature by means of a thermo-
 stated circulating water bath.   The gases leaving the sampler are drawn through
                                   4-7

-------
an aqueous impinger, a silica gel trap, and a dry gas meter by means of a vacuum
pump as in EPA Method-5.
     Laboratory validation studies for the recovery of a few POM from spiked
samples by the above adsorbent sampler method is presented in Table 4-1.
         Table 4-1.  RECOVERY OF POM FROM ADSORBENT SAMPLER METHOD14
POM
Pyrene
Chrysene
Peryl ene
Benzo[ghi Uperyl ene
Coronene
Recovery from
test #1
91+3
90+5
91+4
101+10
80+7
Recovery from
test #2
98+4
92+5
105+5
106+10
92+8
Recovery from
test #3
104+4
106+5
102+6
103+7
100+14
      Although  the absolute  collection  efficiency  is  difficult  to  determine
 under actual field sampling conditions,  a comparison of  the  total  POM  loading
                                                                          14
 between adsorbent sampler method and EPA Method-5 was made by  Jones  et al.
 Field studies  with a 50-hp  oil/gas fired boiler and  effluents  from carbon
                                                                            3
 black manufacturing facilities showed total  POM loading  of 55.2 and  124 ug/m  ,
 respectively,  with adsorbent sampler and 4.2 and 56.5 [*g/m3  with EPA Method-5.
 The higher collection efficiency for the adsorbent sampler is  apparent from
 this study.  In addition, the adsorbent sampler method reduces the sampling
 time by as much as 50 percent and demonstrates superiority in its performance
 for collecting POM from combustion systems burning higher sulfur fuel  at a
 relatively high particle loading.
      As a phased approach to environmental sampling, a Source Assessment
 Sampling System  (SASS) has been developed by EPA's Industrial Environmental
 Research Laboratory.15  Elements  in the  SASS train are designed to separate
 source  samples into  size fractionated particulate, organic vapors, and  inorganic
                                     4-8

-------
 components.   The sample from the stack is introduced into this sampling
 system by a stainless steel  inlet nozzle and a heat controlled stainless steel
 probe at a flow rate of approximately 4 cfm.   The sample from the probe
 enters three cyclones with a backup filter to give four sizing intervals of
 MO urn,  10-3 urn,  3-1 urn,  and less than 1  urn.   The temperature of the cyclones
 and filter is maintained  at  205°C.   After particulate removal,  the hot sample
 gas is cooled to  20°C and passed through  the  XAD-2 sorbent.   The condensate
 passing  through the  XAD-2 is collected separately.
      To  trap the  residual  volatile  inorganic  materials,  the  sample next
 enters three oxidative impingers cooled in an ice bath.   A fourth  ice-cooled
 impinger containing  indicating desiccant  is used  to prevent  moisture  from
 entering the pump.   A continuous sampling time of approximately  five  hours  is
 required for collection of acceptable  sample  size.
      It  should  be mentioned  that the SASS  train has been  used for  a limited
 field  validation study.   In  one  study15 conducted with doped fuel-fired
 boiler,  the  recovery  of four volatile metals was  found to vary between 33% to
 75%.  Another field evaluation study conducted with the SASS train proved that
 the particulate concentrations determined by this method compared very well
with EPA Method-5 in terms of both precision and accuracy.16  Although the
 reproducibility of the SASS train for POM collection was not individually
evaluated in this study, organic materials collected by two SASS trains agreed
well in quantity and composition.
     Recent studies   have demonstrated that the stainless steel gas cooler
and XAD-2 cartridge of the SASS train may cause sample contamination with the
incoming gas stream.   Replacing stainless  steel  with pyrex glass appears to
have obviated this problem.
                                   4-9

-------
     A simple non-isokinetic sampling procedure has been described by Parson
and Mitzner18 in which Pyrex glass tubing containing Tenax resin is directly
inserted into the stack.  A portable pump is used to draw the sample through
the Tenax column at a rate determined by a rotameter.  However, the serious
drawback with this system is its obvious inability to quantitatively adsorb
all the POM at elevated stack temperatures.
     A recent study conducted by Adams et al.19 to characterize the behavior
of different resins used as the sorbent trap showed  that XAD-2 has a greater
volumetric and weight capacity than  Tenax GC resins.  Therefore,  XAD-2 is
preferred as a  sorbent  trap for collection  of  POM.   A modified EPA Method-5
train with a sorbent  trap  operating  for  four hours is capable  of  collecting
all  materials boiling above 190°C when using XAD-2 and  above 240°C when  using
Tenax GC  traps.   Both materials efficiently collect POM,  but the  XAD-2 will
 have a  much  greater capacity  for  the lower boiling compounds.   A  source  sampling
 train using  XAD-2 as  the adsorbent has been described earlier.
 4.2.3.2  Fugitive emissions from  road-way dust—Owing to abrasions of tar and
 bituminous road surfaces,  particles are liberated which contain carcinogenic
 POM.  It is, however, important to distinguish the contribution of vehicular
 traffic towards the overall POM emissions near a highway.  Emission from
 exhausts, burning or spillage of lubricating oils, and tire wear are factors
 arising from vehicular traffic which  cause POM emission in the air.  One way
 to evaluate the  POM emissions arising from abrasion of road materials alone is
 to make comparative measurements  alongside a  highway section of  tar-asphalt
                                                                   20
 and one of concrete, both subjected to an  identical traffic load.    The  POM
  emissions  due to  road abrasion  are  also  dependent  on  seasonal variation.
                                                                         20
                                     4-10

-------
       For collecting  road abrasion samples according to particle  size, an
 Andersen sampler has been used by Ciaccio et al.21  When properly calibrated,
 the sampler can be used to collect particles on stages 2 through 7, repre-
 senting aerodynamic diameters from 0.43 to 7.0 urn which are purported to be
 deposited in the three compartments of the lung.
      The Model 21-000 Andersen sampler, containing stages 0-7 having 1-mil
 thick polypropylene circle on each stage,  is conditioned at a controlled
 temperature and humidity and the polypropylene circles on each stage are
 weighed and placed back on the stages.   The  sampler is assembled and the
 collection  started at an appropriate  rate.   At the end of the sampling period,
 the polypropylene  circles  containing  the collected particles  are  conditioned
 and analyzed  for determining  the  particle  size  distribution.
     For collecting total particulate matter, a Hi-Vol  sampler with  the
 filter holder  containing a flash-fired  glass-fiber filter has  been used.21
 The flow rate  during  sampling can be adjusted to a suitable value for effi-
 ciently collecting  the particulate matter.
 4.2.4  Ambient Air
     The sampling of POM in ambient air is usually performed with the objec-
 tive of determining the particle size distribution and nature and concentra-
 tions of individual components at various points in the environment.   The
 selection of sampling sites play important  roles in sample collection.
Sampling sites are determined to evaluate the following:  (1)  characterize
the rural  or urban background levels,  (2) assess the health hazards  to people
in the  vicinity,  (3) determine source  effects,  and  (4)  establish  transport
                                  4-11

-------
mechanisms.  Reference should be made .to the available sources  »    for
detailed treatment of these aspects in the selection of sites.
     In addition to the selection of a preferable method and site, certain
other factors must be adequately taken into consideration during ambient
sampling.  The height of the sampler intake above ground level and the local
topography and climate each has influence on the data obtained.  For example,
the  influence of summer-like temperature on losses  of benzo[a]pyrene from
airborne  particles has been studied  during real high-volume atmospheric samp-
lings.24   Occurrence  of  seasonal variation  in  specific  surface areas and
                                                            on
densities of suspended particulate matter has  been  observed.    The effects
of wind on the  collection  efficiencies of particulate  matter  has  been  demon-
 strated by Ogden and Wood.26   Other  factors,  such  as sampling rate and time,
 are also important in the  overall  sampling  strategy.  The total  amount of
 samples to be collected usually depends on  the sensitivity of the .analytical
 method intended to be used for the determination of individual  components.
      The  method for collection of POM from ambient air will be divided into
 two sections depending on the objective of the sampling.
 4.2.4.1   Collection of total air particulate matter-Air particulate matter
 is  usually  collected on flash-fired glass-fiber filters using a high volume
 air sampler.  The sampler consists  of three units:  (1) the  face  plate and
 gasket,  (2) the filter  adapter assembly, and  (3) the motor unit.  The  sampler
 must  be  capable of passing environmental air  through  an  approximate 400  cm
 area  of  a clean 20.3 x  25.4  cm glass  fiber  filter  at  a rate  of at least  1.13
 n»3/min.   The glass-fiber  filters  should have  collection  efficiencies  of  at
  least 99 percent  for particles  of 0.3 micron  or  larger in diameter.   The
                                     4-12

-------
  motor must be capable of continuous  operation  for at least 24 hour  periods.
  It is important  that  the sampler  be  properly installed  in  a suitable  shelter
  to protect it from  extreme weather conditions  and debris.   For measuring the
  air flow  rates through the sampling  unit,  it must be  provided with  either a
  calibrated flow  meter (rotameter) or a gauge.  The details  of this  method are
  given  by  Sholtes et al.27.
      On the average, when a high-volume sampler is located  in an urban area,
  it will collect approximately 250-350 mg of particulate matter while sampling
 2000-2400 m3 of air during a 24 hour period.   Of the total  amount of the
 particulate matter,  approximately 10 percent,  that is, 25 mg, will  be the
                  28
 orgamc fraction.    The quantity of organic  fraction generally needed for
 the analysis of POM fraction  depends  on the sensitivity  of  the analytical
 method used.  However, to obtain the  needed amount of organic fraction it may
 become necessary  to  pool  together  organic  fractions of several individual
 high-volume air samples  from  a  single monitoring site.
      POM are present in polluted air  sorbed onto airborne particulate
 matter, usually characterized as primary and secondary.  Primary particulate
 matter  is  found in sizes  between 1 Mm and 20 urn.29  Secondary  particulate
 matter  ranges  in  size  from molecular clusters of the order of  5 nm to particles
 with  diameter as  large as several micrometers.29  Very fine particles that
 cannot be  retained on  fiberglass filters act like a gas or vapor in that they
 follow fluid streamlines and are subject to Brownian motion.  For collection
 of this gaseous phase various  adsorbents have  been used.
      In one of the methods,30  polyurethane foam  has been  used as adsorbent.
An aluminum sampling probe used to collect suspended particulate matter
                                   4-13

-------
consists of a piece of tubing with a grid for holding a glass-fiber filter
and two cylindrical polyurethane plugs (8.5 x 6.0 cm).   The plugs located
behind the glass-fiber filter prevents loss of volatile fraction of POM.  A
similar method has been used by Fox and Staley31 for collecting both particu-
late and gas phase components from atmospheric samples.
     Schuetzle et al.32 designed an apparatus for simultaneous sampling of
gaseous and particulate air pollutants.  An adjustable one-stage impactor has
been used  for sampling'aerosols greater than  1-2 urn  in diameter and is follow-
ed by  a glass-fiber  filter for collection  of  particles less than 1-2 urn  in
diameter.  A properly conditioned Chromosorb-102 support  is used to trap
gaseous air pollutants.
      A sampling system using only an adsorbent has  also been  used.     It is  a
 simple system consisting of  a vacuum pump, a flow measuring and regulating
 device,  and a sample tube containing the adsorbent.   The  sample tubes  are  110
 by 10 mm and contain 4 ml of conditioned Tenax GC  held in position by  plugs
 of si lam" zed glass wool.
      The usefulness of the adsorbent in collecting POM not otherwise retained
 on the glass-fiber filter is shown by Krstulovic et al.30  Table 4-2 shows that
 the amount of POM collected on the adsorbent depends both on the nature of
 POM and the location of  collected sample.
 4.2.4.2   Distribution of particle size—An assessment of  the POM content of
 the polluted atmosphere  with respect to size of particles in an aerosol often
 becomes  necessary.   Knowledge  of the fraction of the  air particulate  matter
 which can cause deposition  in  the various compartments of the  respiratory
  system is of prime importance  since some  of the POM are  proven carcinogens.
                                     4-14

-------
                                               to
                                                                 cr>
                                                                 e\j
 CD
 co
  >    ^
            J;    -1=    Q.    TD
 O)
JC
 a.

DQ
                  c:
                  re

                  o
                                                        to

 «     S-     O    CO
 CON
 «     =3     C    CM
-C    •—     CO
CL.    U_    CQ    i—
                                                                           cu
                                                                           u
                                                                           a>
                                                                           cu
                                                                          •a
                                          o
                                              4-15

-------
There is considerable disagreement regarding the aerodynamic sizes which can
cause deposition of participate matter in thefthree compartments of the lung
(see Section 6).
     Whitby et al.34 have presented evidence that the mass distribution of
atmospheric aerosols is usually bimodal, with one mode occurring below 1.0 Mm
and the  other mode occurring  in the 5 to 15 urn  range.  A dichotomous sampler
for particulates  has been designed to collect and fractionate  samples into two
size  ranges.35  Membrane filters  in the two air paths collect  the  respective
samples.  However,  such a sampling technique  has rarely been applied to
collect POM from  air particulate  matter.
      The experimental  determination  of  particulate  distribution is frequently
                                                      oc
 done by five-stage Andersen Hi-Vol cascade impactors.     The first four
 stages of the sampler comprise the fractionating head,  while the fifth stage
 is a backup filter positioned between the fractionating head and a standard
 Hi-Vol  air sampler.  When operated at a flow rate of 20 cfm, the sampler
 fractionates suspended particulate matter into  five aerodynamic size ranges
 according to certain cut-off diameters.
 4.3  DESORPTION  OF POM FROM  COLLECTION MEDIA
      A  number of methods are available for the desorption of  the  collected
  POM  from the filter and adsorption traps  used  for  sample collection.  Three
  methods most commonly  used are:   (1) extraction, (2) thermal  desorption, and
  (3)  vacuum sublimation.  Each  of these methods has been  individually  discussed.
  4.3.1   Extraction
       In the extraction method, collected POM are desorbed  from the filtering
  medium by means of a solvent.   Depending on the technique  used, the extraction
                                     4-16

-------
 method can be subdivided  into three distinctly different groups:   (1)  solvent
 extraction, (2) mechanic!! disruption, and (3) HF acid dissolution.
. 4-3-1-1  Solvent extraction--The principle behind solvent extraction is to
 dissolve the collected POM from the filtering medium by digesting  in a
 suitable solvent.   The usual method involves refluxing of the filter in a
 Soxhlet extractor for a certain^length of time.   A wide variety of solvents
 have been used for this purpose.  These include acetone,44"47 benzene,6'28'47"49
 cyclohexane,47'49'51"53 methanol,47'49 tetrahydrofuran,19'30'47 methylene
 chloride,28'47'55'56 chloroform,57 pentane,22 carbon disulfide,49 and a
 methanol-benzene mixture.58'59   Potthast  and Eigner60 used  propylene carbonate
 as a solvent  for extraction  of  POM from meat  products.   They claim that the
 POM are more  soluble in this  solvent than any other.   The selection of  a
 particular solvent  or a mixture  of solvents  is based on the  efficiency  of
 extraction, selectivity in dissolution, time  required for extraction, and the
 reproducibility  of  extraction.   Liberti et al.47 carried out the  extraction
process with various  solvents, such as cyclohexane,  benzene,  dichloromethane,
tetrahydrofuran, acetone,  and methanol.  Not  only was the extraction yield
for PAH maximum with  benzene but benzene also showed  selectivity  towards POM
extraction.   The extraction time was found to be reasonable  and a 10-hour
period was found adequate  for complete extraction.   Experiments carried out
with cyclohexane showed a poorer efficiency and a much longer extraction
time.  Based on these data, these authors  recommended benzene as the solvent
for POM extraction.
     Cautreels and  Cauwenberghe49 also  made a comparative study of the various
solvents  for  extraction of organic compounds  from an  artificial  aerosol  model
                                   4-17

-------
which may not be representative for real atmospheric suspended matter.   Their
study of various solvents, e.g., benzene, carbon disulfide, cyclohexane,
hexane, and methanol is presented in Table 4-3.
           Table 4-3.  COMPARISON OF THE EXTRACTION YIELD OF POM
                           WITH DIFFERENT SOLVENTS49
% extraction yield
Compound
Phenanthrene
Fl uoranthene
Pyrene
Benzo[b]fluorene
Benz [a] anthracene
Benzo[a]pyrene
Dibenz [a, h] anthracene
Benzo[ghi]perylene
Benzene
96
99
96
101
102
75
43
39
Carbon
disulfide
98
100
97
99
106
65
39
40
Cyclohexane
46
52
52
31
68
6
Hexane
69
58
62
??a
 a  Below the  detection  limit.
      Although  the yields  of  extraction  of  POM were  found  to be  10 to 20
 percent higher with  methanol  than  with  benzene,  the latter solvent  showed
 more selectivity towards  POM than  methanol.  The quantitative extraction
 time was determined  to be 8  hours.   Both cyclohexane and  hexane proved un-
                                                  49
 suitable for the extraction  of POM.   These authors    also showed that
 larger errors  could  occur during the extraction  of  particulate  matter than
 during the analytical  method itself.
      It should be  mentioned  that due to the nature  of the artificial sample
 the extraction yields  of PAH as reported by the  above authors may have signi-
 ficance only for comparison  of various  solvents. The absolute  extraction
 yields from natural  samples  may vary considerably from these  results.  Various
 laboratories have  reported a much  higher cyclohexane extraction efficiency
 for BaP from glass-fiber filter samples.
                                         50
                                    4-18

-------
  4'3-1'2  Mechanical disruption-The widely  used  Soxhlet extraction  is time-
  consuming, decomposes some of the pollutants, and  lacks good precision.61  In
  order to avoid these errors, a mechanical disruptive technique which makes
  use of a specially designed blender has been used by Bove and Kukreja.62  The
  method has been claimed to be rapid and precise for the extraction of organic
  contents from glass-fiber filters.   A superior procedure has been described
  by Golden and Sawicki.52  In this procedure, the sample is suspended in
  cyclohexane and subjected to ultrasonic waves at room temperature.   Silica
  powder (5 micron size)  is added  to  adsorb polar  co-extractives.   At  the  end
  of sonification,  the extract is  filtered on  a sintered  glass filter.  The
  entire method has been evaluated  by Sawicki  et a!.63
      The  recovery of POM by  the sonification method was found to  be  95 to
  98.2 percent.  A  comparison of the recovery  of aromatic compounds between
  ultrasonic  and Soxhlet method showed that the ultrasonic method extracted 49
  percent of  the total particulate matter collected on glass-fiber filters with
  a  relative  standard deviation of ±1.33 percent compared to 30 percent for the
  Soxhlet extraction with a relative standard deviation of ±26.1 percent of the
  total extractables.  Not only the extraction efficiency but the reproducibil-
  ity of the ultrasonic method was  superior to the  Soxhlet method.
      Disadvantages with  the sonification method lies in  the fact that the
 ultrasonic extraction must be done in  a way that  avoids  an  unacceptably  high
 noise level.
-4-3'1'3  Hydrofluoric acid dissolution-This  technique utilizes the ability
 of  HF acid to  chemically  dissolve  the glass-fiber  filters,  leaving the organic
                                   4-19

-------
residue of the filter paper free to extract quickly with an appropriate sol-
vent.  For samples collected on more than one 8 by 10 inch filter, the extrac-
tion time was about 30 minutes.
     Comparison of the HF method with the 6-hour Soxhlet extraction technique
is given in Table 4-4.
       Table 4-4.  COMPARISON OF HF EXTRACTION WITH SOXHLET EXTRACTION64
weiqht of extractives, mg
Sample number
1
2
3
4
5
6
7
8
9
10
11
12
13
Blank
HF Method
2.7
2.4
2.6
3.0
2.3
2.3
3.4
2.7
3.5
3.0
2.6
2.4
2.5
1.0
Soxhlet Method
3.4
2.0
1.9
3.3
2.6
2.4
2.6
2.2
3.2
3.1
2.1
2.8
2.7
1.0
      Determination of absolute efficiency of extraction for three POM by
 the HF method showed high percent recovery values (97.7 to 100 percent).   It
 was also proven that a Lewis acid such as HF did not cause ring and/or side
 chain rearrangements for alkyl-substituted aromatics.64  The authors suggested
 that the new method should prove highly useful for enriching POM content
 collected on glass-fiber filters.  However, this procedure lacks general
 applicability since some of the collected organics would react with HF and
                                    4-20

-------
 the hazards associated with the handling of highly reactive HF acid are
 considerable.
 4.3.2  Thermal Stripping
      The extractive methods discussed above require that the extract be con-
 centrated and usually only a small aliquot of the concentrate be used for
 analysis, e.g., gas chromatography.   This results in overall reduced re-
 covery and sensitivity.   Gas-phase extraction, however, is more efficient.
 According to this method the POM trapped on the collecting media are therm-
 ally stripped directly into the separating and detecting device.   Since the
 method involves no evaporative concentration,  it eliminates the corresponding
 loss factor.   Another advantage of this  method is that the whole sample col-
 lected (instead of a fraction  of it)  can be used for  a single  analysis,
 thereby increasing the sensitivity of the method and  eliminating the  need for
 collecting large amounts  of sample.
     Burchfield et al.65  used  this method for  the analysis  of  POM from  air-
 borne  particulate  matter.   The  Hi-Vol  filter was  ground to  about  20 mesh in
 a Wiley mill.   The samples  were  then  loosely packed into a  special stripper
 tube and the organic  components  desorbed by a  stream of nitrogen  at 300°C.
 The desorbed compounds were trapped in a cold  column containing the same
 packing that would subsequently  be used for GLC.  Although a minimum of 10
 hours was required, stripping was carried out  overnight.  Due to  the long
 stripping time, variations  in gas flow rates between the tubes were unimportant.
The compounds trapped in the cold column were  finally heated and the compounds
were separated on  the main column.
                                   4-21

-------
                                                     33
     A very similar method employed by Bertsch et al.    consisted of heat
desorption of organic compounds from a porous polymer (Tenax GC) p'hase instead
of a glass-fiber filter-trapped phase.  The desorbed compounds were trapped on
Emulphor ON-870 cooled at dry ice or liquid nitrogen (N2) temperature.  The
advantage of this method was the substantial reduction of stripping time from
10 hours to 15 minutes only.
     A slight modification of the above methods which eliminates the cold trap
utilizes direct injection of the heat-desorbed components onto  a packed column
by way of a gas-sampling valve.18  This method is  simpler than  the other two
stripping methods  since it eliminates  the  special  accessory  equipment needed
for  the cold trap.
     There are  several disadvantages  with  the thermal  stripping methods.   Even
though the recovery of the  adsorbed compounds by thermal stripping methods  may
be  high,  data showing this  effect  are still  lacking.   This  is primarily  due to
the difficulty  in  introducing  internal standards in the system.  That the  high
temperature  used during  the thermal desorption process does  not cause any
 rearrangement  and/or decomposition remains to be established.  Since the
 injection technique to  the final  gas  chromatographic phase cannot be instan-
 taneous,  it is  bound to  broaden some  peaks resulting in resolution problems.
 Finally,  replicate injections of the  same sample is not possible with this
 technique.
 4.3.3  Vacuum Sublimation
      This technique for the extraction of atmospheric pollutants has been
 developed by Japanese researchers.  A more recent apparatus  for extracting
 seven particulate samples by one operation is given by Matsushita et al.
                                                                         66
                                    4-22

-------
  The extraction apparatus used is constructed with a rotary pump, a MacLeod
  manometer,  a connector,  vacuum sublimation tubes, sublimation flasks and an
  electric furnace.   Seven particulate samples are put in seven sublimation
  flasks  connected with their respective sublimation tubes.   These tubes  are
  attached via a common connector to  a manometer and a rotary pump.   After
  cooling the  sublimation  tubes  with  ice water,  these are evacuated  to 0.005-
  0.001 mm Hg.   The sublimation  flasks  are next  heated to 310-315°C.   POM
  sublimed from  the particulates  are  deposited on  the cold part of the sublima-
  tion tube.   At the end of 40 minutes  of sublimation,  the deposited  sublimates
  from the individual tubes are dissolved in a small  amount of benzene.
      The recovery and reproducibility  of the above  method was evaluated66
 with spiked  samples and for three POM  tested the recovery varied between
 97.4 and 98.1 percent and the maximum  coefficient of variation was 2.13 per-
 cent (relative standard deviation).   A comparative study between this method
 and ultrasonic extraction method proved that they both have almost identical
 recoveries for BaP from airborne particulate samples.66  However, studies
.conducted by Monkman et al,69 contradicted these findings.
      The vacuum sublimation technique is faster than Soxhlet extraction  and
 is routinely used by Japanese researchers41'67 for extraction of POM from
 Hi-Vol  glass-fiber filters.   It has  also been recommended by Schultz et  al.68
 as a routine method  for a limited POM analysis.
 4.4  CLEAN-UP AND SEPARATION
      The next step following the desorption of POM from the  collection
 media consists  of clean-up  and  separation of these compounds from undesirable
 impurities and  from one another.  Of course,  the  extent of clean-up  and
                                   4-23

-------
separation will depend on the nature of the collected sample.   This  section
deals with the various methods used for this purpose.  Depending on  the
method employed, the clean-up and separation process can be divided  into:
(1) Solvent Partitioning, (2) Column Chromatography, (3) Paper Chromatography,
(4) Thin Layer Chromatography, (5) Gas Chromatography, and (6) High  Pressure
Liquid Chromatography.  The individual processes are dealt with separately.
4.4.1  Solvent Partitioning
     This  is  one of the oldest methods used for the  isolation of POM as a
group from the bulk of impurities.  According to this method, the POM  in a
certain  solvent are allowed to distribute  in another immiscible solvent which
has  either higher  or  lower distribution  coefficient  for the POM than the
original  solvent.   In the former instance,  the  POM are  extracted in the
second  solvent, and in the latter instance,  the impurities accumulate  in the
 second  solvent leaving the first solvent relatively free  of impurities.  A
 variety of solvents which have been used for solvent partitioning are:   (1)
 methanol-water, (2) tetramethyl  uric acid in methanol,  (3) acetonitrile,  (4)
 dimethyl formamide, (5)  nitromethane, and (6) dimethyl  sulfoxide.
 4.4.1.1  Methanol-water—In this solvent partitioning step originally proposed
 by Hoffmann  and wynder,70 the POM in cyclohexane solution are allowed to
 partition in methanol-water solution.  The polar impurities in cyclohexane
 solution preferentially  distribute themselves in the more polar methanol-water
 phase,  leaving the cyclohexane  phase containing the POM relatively free of
 impurities.   It should be mentioned that  a single partitioning step like this
 is  only a part of several partitioning  or other separatory steps in the overall
 clean-up scheme.  However, methanol-water phase has been  used by a number of
                                     4-24

-------
  other  authors71'73  as  a preliminary  step for  getting  rid of the polar com-
  ponents  from the POM fraction.
  4'4-1-2  Tetramethyl uric acid in methannl-A solvent system incorporating
  tetramethyl uric acid  in methanol as a selective complexing agent has been
  used to  separate several POM.  Distribution coefficients for a number of
  POM in this solvent system have been given by Mold et al.74  However, this
  solvent system lacks wide applicability.
 4.4.1.3  Acetonitrile-Haenni et al.75 determined the distribution of POM
 between n-heptane and acetonitrile and found low distribution  ratios.   These
 authors concluded that  this  solvent is not  very  effective  in concentrating
 POM compounds.
 4-4.1.4  Dimethyl  formamide-This  solvent was  found to be much  more  efficient
 for the removal  of POM  from  heptane phase.75   Consequently, this solvent  has
 been used by other authors9'73 for extracting  POM from eyelohexane phase  as
 well.
 4.4.1.5   Nitromethane—The partition  coefficients of POM between nitro-
 methane and cyclohexane was determined by Hoffmann and Wynder12 and varied
 between 4.4 and 1.65 for a number of  POM tested.   The same partition coeffi-
 cients determined between nitromethane and other aliphatic solvents75'76
 showed values quite similar to that determined by the above authors.   This
 solvent has been used by a number of authors, particularly for  the clean-up
 of automobile exhaust samples.53'55'71'77
4.4.1.6  Dimethyl sulfoxide-The first evaluation of this solvent as  a parti-
tioning medium for POM was  made by Haenni  et al.75  A comparison of four
solvents,  e.g.,  acetonitrile,  nitromethane,  dimethyl  formamide,  and dimethyl
                                   4-25

-------
sulfoxide, showed that the last solvent was much more suitable on the basis
of distribution coefficients and selectivity for POM compounds.75  Comparative
study of the percent recovery between two widely used solvents made by Acheson
      78 ,
et al.   is given in Table 4-5.
     Table 4-5.  EFFICIENCIES OF SOLVENT-PARTITION PURIFICATION PROCEDURE
                                                        78
Compound
  % Recovery
Nitromethane  DMSO
Compound       % Recovery
            Nitromethane  DMSO
 Fluoranthene        48-83
 Pyrene              38-42
 Benz[a]anthracene   50-64
  & chrysene
 Benzo[k]fluoranthene  54-59
 Benzo[ghi]perylene  15.8-45
              92-96    Benzo[a]pyrene
              94-96      & Benzo[e]pyrene
              90-92    Perylene
                       Indeno[l,2,3-cd]-
              91-94      pyrene
              90-91    Coronene
               33-36
               < 10-24
               41-49
95-99
84-90
97-100
               10.0-44    82-93
      From their investigations  Acheson et al.78 conclude  that  DMSO  is  the
                                                               79  80
 best extractive medium for POM  compounds.   Other investigators  '    have
 effectively used this solvent for isolating POM compounds from impurities.
      It should be mentioned that in a complex mixture it  often becomes nec-
 essary to first fractionate the POM extract into acid, basic,  and neutral
 fractions prior to further clean-up procedures.  '     In  this  method,  the  POM
 mixture in a suitable solvent is extracted with an aqueous alkaline solution
 to separate the acidic components.  Subsequent extraction with an aqueous
 acid solution separate the basic components from the neutral fraction.  By
 carefully controlling the acidic and basic strength of the aqueous solution,
 the individual fractions can be further subdivided into groups of compounds
 with varying acidic  and basic strength.  By carrying out the extraction with
                                     4-26

-------
  IN sodium hydroxide or HC1,  Cautreels and Cauwenberghe49 found that on the
  average 99 percent of n-paraffins stayed in the neutral  fraction.   About
  90 percent of neutral  polyaromatics  was  found in the neutral  fraction  and
  10 percent in the  acidic  fraction.   The  yield of other neutrals, e.g.,  phthal-
  ate ester,  was about  96 percent  in the neutral  fraction.  The  basic  fraction
  gave an  average of  75  percent yield  for  the  acidic compounds.  The results
  obtained  from the isolation of some  selected  basic compounds,  however,  showed
  very poor recovery.
 4.4.2  Column Chromatography
      This is the most widely used separation method for the POM.   A large
 variety of adsorbents, such as (1) alumina, (2) silica gel,  (3) Florisil, (4)
 cellulose acetate,  and (5) gel  materials, have been used  for separation.
 Adsorbents with uniform particle  size, 60-80, 80-100, and 100-120 mesh  and
 column  diameter:length ratios  of  at least 1.25 are suggested.   The  laboratory
 temperature should  be  kept reasonably constant and the solvent  used  for
 elution should be of highest purity.   Maximal  separation  can be achieved by
 slowly  and evenly increasing the  polarity of the eluting  solvents.  The
 hydrocarbons are eluted from the  column in the following  order:   aliphatics,
 olefins,  benzene derivatives, naphthalene derivatives, dibenzofuran fraction,
 anthracene  fraction, chrysene fraction, benzpyrene fraction, and coronene
         28
 fraction.    The factors that affect  the  Chromatography of the  organic frac-
tions are:   (1) the  retardation volume, i.e., the volume of eluent passing
through the column per gram of adsorbent before the substance in question
leaves the column;  (2) the volume spread of the eluted substances, i.e., the
volume of eluent in  which the substance is found; and (3)  the volume separation,
                                   4-27

-------
                                                                        po
i.e., the volume of eluent separating one eluent substance from another.
These factors can be affected by three variables:   (1) percentage of water  in
the sorbent, (2) percentage of polar solvent in the solvent mixture, and (3)
possible miscellaneous factors, e.g., the composition and weight of the
organic fraction.
4.4.2.1  Alumina—Alumina is available in various grades, sizes, pH and
activity.   For the separation of neutral POM fractions, Woelm Neutral Alumina
is generally used.  The sample:adsorbent ratio varies between 1:100 and
                                                                         84
1:1000.  The alumina  is deactivated with 13.7 percent water prior to use.
Elution  of  the  adsorbed compounds  is  done by gradually  increasing the polarity
of the solvent  system.  Ether-pentane,85 cyclohexane-ether,46 cyclohexane-
benzene,86  and  benzene-methanol87  are some  of the  solvent systems used  by
various  researchers.   Some  investigators46  have  successfully separated  PAH
 on  very  long alumina column.   However, such separation  procedures  need  auto-
 matic fraction  collectors to collect a number  of eluting fractions.
      The main disadvantage of alumina column separation is that it is  very
 time consuming, sometimes causes decomposition of compounds on the column
 (particularly polynuclear aromatic amines), and good reproducibility is often
 difficult to achieve.
 4.4.2.2  Silica gel-It is difficult to get silica gel with a standardized
 preparation.  Material of 100-200 mesh or Davison grade  12 or equivalent is
 found suitable for chromatographic purposes.28'71  The sample:adsorbent ratio
 varies between 1:50  and 1:500.  Some  investigators used  silica gel with 3 to
                                                                        r-l  -J-\
 5 percent  moisture content,47'57  while others used the dried silica gel   '
                                     4-28

-------
 without deactivation with water.  The elution of adsorbed compounds  is usually

 done by benzene51'57'71 or hexane-benzene.47

      It should be recognized that silica gel is normally used in combination

 with other adsorbents or other separation methods and is rarely used as the

 sole separation process.  A separation method using both alumina and silica

 gel chromatography was used by Liberti et al.47  The average percent recovery

 for 9 PAH compounds was found to be 77.2 percent.   The individual recoveries

 as reported by these authors are listed in Table 4-6.


     Table 4-6.   RECOVERY OF PAH BY LINEAR ELUTION ADSORPTION CHROMATOGRAPHY47
      PAH
                                         % Recovery from alumina
                                         & silica gel  chromatography
 Fluoranthene
 Pyrene
 1-2  Benzanthracene
 Chrysene
 3-4  Benzofluoranthene
 Benzo[a]pyrene
 Benzo[e]pyrene
 Indenofl,2,3-cd]pyrene
 Benzo[g,h,i]perylehe
68.8
82.0
77.6
82.1
82.4
80.5
80.5
90.6
90.3
     Column chromatography with silica gel suffers from all the disadvantages

normally encountered with alumina.  The relatively slow flow rates of solvents

through silica gel causes additional problems.

4.4.2.3  Florisi 1--Column chromatography using Florisil is not as widespread

for the separation of POM as the other two methods already described.

Activated Florisil, 60-100 mesh, have been used for fractiohation of PAH com-
      77 87                .......
pounds  '   and also for isolation of PAH compounds as a group from other
                                   4-29

-------
impurities.79'80' The eluting solvent for this sorbent is usually hexane-
benzene87 or benzene alone.80  The advantage with Florisil column is that
PAH separate rather well, despite a short elution time.   To prevent photo-
oxidation during chromatography, the Florisil column should be protected from
light.
4.4.2.4  Cellulose acetate—Although cellulose acetate is a promising sorbent
for separation of POM, its use as a column chromatographic material is very
limited.  Griest et  al.79 have reported  using this technique for the purifica-
tion  of  PAH.
4.4.2.5  Gel  fi1tration—Wi1k et  al.88 were  the  first to  use this technique for
the separation of POM.   Others  have  since found  the  system to  be valuable for
separation  of PAH from automobile exhausts,14'89 air particulate matter,   '
cigarette smoke,73  and biological samples.90'91   Separation of neutral  aza-
arenes has  also  been achieved  by  this  technique.92   The  separation  is best
achieved isothermally on Sephadex LH-20  (ratio,  about  1:1000)  columns with 2-
propanol as solvent at a flow  rate  of 6-7 ml/min.  Gladen89 determined  the
 recovery efficiencies of 8 PAH by this technique which is shown in Table  4-7.
         Table 4-7.  RECOVERY OF PAH BY SEPHADEX LH-20 CHROMATOGRAPHY
                                                                     ,89
 Compound
% Recovery
 Anthracene
 Phenanthrene
 Pyrene
 Fluoranthene
 Benz[a]anthracene
 Benzo[a]pyrene
 Perylene
 Benzo[g,h,i]perylene
   101
    98
    94
   102
   100
    97
    98
    98
                                     4-30

-------
      Although gel filtration is free from most of the drawbacks of other
 column chromatographic methods, its main disadvantage is that it takes a long
 time (1 to 3 days) for separation.
 4.4.3  Paper Chromatography
      Paper chromatographic methods are used for the separation of individual
 fractions and are usually preceeded by a preliminary separation by column
 Chromatography.   For optimum separation, the commonly used paper is acetyl-
 ated cellulose70'81  and the Chromatography is carried out by ascending tech-
 nique.   A slight modification of the paper by impregnating with 10 percent
 liquid  paraffin  has  also been used for chromatographic separation.82   Develop-
 ing is  done with a variety of solvent mixtures.   These include  toluene-
 methanol-water (1:10:1),  methanol-ether-water (4:4:1), methanol-chloroform
 (3:1),  ethanol-toluene-water (17:4:1),  and ethanol-benzene-water (12:6:1).
      In modern separation techniques  paper Chromatography  is  rarely used.
 There are  several  reasons for this.   The time  required for  separation, diffi-
 culty in getting reproducible papers, inadequate  resolution of compounds, and
 the  non-quantitative nature  of analysis are disadvantages for this technique.
 4.4.4  Paper and Thin-layer  Electrophoresis
      Separation  of basic  aza-arene by this technique was accomplished by
        oo
 Sawicki.    However,  this method has restrictive use for reasons similar to
 the paper chromatographic technique.
4.4.5  Thin-Layer Chromatography
     This technique is quick, inexpensive,  reasonably reproducible, and one
of the better methods for the separation of isomeric compounds.   With  the
availability of in situ scanning technique  (thin-layer scanner),  this  method
                                   4-31

-------
has been used for quantification of POM with reasonably good accuracy.   A
number of materials have been used as sorbent materials for the plates.
These are:  (1) silica,47'76'96'97 (2) alumina,95'99'100 (3) cellulose,83 (4)
cellulose acetate,68'90'93"95 and (5) polyamide.101  A number of modifications
                                                                             1 /\o
including silinazation of silica gel102 and channel thin-layer chromatography
have been proposed.  In the latter procedure, the components which need to be
fractionated-on a thin-layer plate are made to flow into narrow development
channels  in order to prevent the spreading of the chromatographic spots.
     Both PAH  and polynuclear aza-heterocyclic compounds have been separated
by TLC  method.  A  large number  of variations of TLC procedure, such as
variation of developing solvents, use of  a mixture of  sorbents, developing in
one  or  two  dimensions, have  been  used to  affect better resolution of the
components.  The  effect of solvent variations on  cellulose acetate TLC  has
been discussed in detail  by  Woidich  et al.95  Although alumina-cellulose
 (2:1)  has been used as  a  composite adsorbent,98 the  best resolution  is
 obtained when  two-dimensional  TLC on a composite  plate of aluminum oxide-
 40 percent acetylated cellulose (2:1) is  used.104'105  The percent recovery
 of a few isomeric arenes  separated by this method105 is given in  Table  4-8.
     Table 4-8.  RECOVERY OF VARIOUS PAH DURING THIN-LAYER CHROMATOGRAPHY105
Compound % Recovery
BeP 86.0+5.3
BaP 85.2+5.1
BbF 92.1+3.5
BkF 91.2+3.7
Perylene 90.4+4.6
Dibenzo[def, 90.9+3.9
mno] chrysene
Compound % Recovery
Benzo[ghi]perylene 88.7+4.0
Naphtha[l,2,3,4-def] 87.9+5.5
Chrysene
Benzo[rst]pentaphene 91.8+4.2
Dibenzo[b,def] chrysene 89.8+3.9
Naptho[2,18-gra] naphtha- 90.7+5.1
cene
Dibenzo[def,b]chrysene 84.1+6.2
                                     4-32

-------
      A  large  number  of  aza-heterocyclic  compounds  found  in  urban  airborne

 particulates  and  in  air pollution source effluents have  been separated by TLC

 procedure.  The basic fraction from a coal-tar-pitch sample when  subjected to

 a two-dimensional TLC separation on silica gel-cellulose (2:1) with pentane-

 ether (9:1, v/v) and dimethyl formamide-water (35:15, v/v) solvent systems

 yielded the results shown in Table 4-9.

      Two-dimensional  TLC separation has also been successfully used98 during

 the quantification of acridine,  benz[c]acridine5 7H-benz[de]anthracen-7-one

 and phenalen-1-one in airborne particulate samples.


            Table 4-9.   AMOUNTS OF AZA-HETEROCYCLIC  COMPOUNDS IN A
              COAL-TAR-PITCH  POLLUTED  AIR  SAMPLE  SEPARATED BY TLC85
 Compound
 Acridine
 Benzo[f]quinoline
 Benzo[h]quinoline
 Phenanthridine
 Benz[a]acridine
 Benz[c]acridine
Concentration
  in mg/

1000 m3 air
   0.870
   0.420
   0.260
   0.020
   0.200
   0.120
 Compound
~

 Indeno[l,2,3-ij]-
   isoquinoline
 11 H-Indeno[l,2-b]-
   quinoline
 Dibenz[a,h]acridine
 Dibenz[a,j]acridine
Concentration
  in mg/
1000 m3 air
   0.030

   0.190

   0.010
   0.001
     There are several disadvantages with the TLC procedure.  Mass spectrometry

and liquid chromatography, when applied as additional analytical tools for

identification of POM, have demonstrated that the individual spots or bands

are sometimes, in fact, mixtures of two or more compounds.   Also, the TLC pro-

cedure cannot be used when separation of a large number of compounds is re-

quired since TLC resolution is limited.   Furthermore, during TLC procedure
                                   4-33

-------
trace amounts of compounds may photodecompose requiring addition of an
internal standard to correct for the losses.
4.4.6  Gas Chromatography
     The closeness of the chemical structure among some of the POM and
their relatively high boiling points demand the stringent requirements of high
resolution and thermal  stability at high temperature for the chromatographic
column  packings.   Unfortunately, until  recently the available liquid phases
that were thermally  the most stable were barely adequate for the temperature
required to  separate some of the  high  boiling POM.  However, a  large  variety
of column packing materials have  since been used  for  POM separation.   Depend-
 ing on their mode of operation, gas  chromatographic methods used so far can be
 divided into:  (1) packed column;  and (2)  capillary columns,  e.g., surface-
 coated open tubular column and wall-coated open tubular column.
 4.4.6.1  Packed column-A number of packing materials have been used in packed
 column Chromatography.  A gas-solid chromatographic phase using Chromosorb
 101115 and GLC (gas liquid) phases using OV-1,65'106 OV-7,98 OV-25,79 poly-
                 w           98 108    -     ,, 79 cn on 86,108,109 .p™ 57,110
 metaphenoxylene,93  apiezon  [_,98'10b apiezon W,   SE-30,           SE 52,
 Dexsil-300 GCf«,65f 106,111,113,114 Dexsil-400 GC,106 Dexsil-410  GC,106 and
 nematic liquid crystals114'116'117 have found their applications  as the pack-
  ing materials.
       One  of the  superior column  packing materials  available for POM  separation
  is Dexsil-300 or Dexsil-400.   The column  bleeding  characteristic  and resolution
  capability for these packing materials make them highly  suitable for POM  anal-
                                                                     106
  ysis.  The resolution of Dexsil-400 is even better than  Dexsil-300.     The
  separation of both PAH and aza-arenes in samples containing coal  tar has  been
                                     4-34

-------
  accomplished with these packing materials.106  Dong et al.118 have used
  Dexsi1-300 column for the separation and quantification of a number of aza-
  arenes  in  the basic  fraction  of airborne particulate matter.
      Nematic liquid  crystals  with  the general  formula N,N'-b1s(p-substituted
  benzylidene)-a,a'-bi-p-toluidine have been  shown to  resolve  hard-to-separate
  PAH isomers.  Three  p-substituted  compounds, methoxy (BMBT),  hexyloxy  (BHxBT),
  and phenyl  (BPhBT) have been  investigated.114'116'117   However, column bleed-'
  ing at higher temperatures remains a  problem with this  packing material.  For
  routine separation of benzpyrene isomers, a BMBT packed column operated iso-
 thermal ly at 130°C has been found useful.114
 4.4.6.2   Capillary column-These columns are more advantageous than packed
 columns  due to their ability to affect higher resolution.   The number of
 theoretical plates obtained from such columns may vary between 20,000-70,000.119
 Versamid-900,102 SE-30,119 XE-60,119  SE-52,11>53>58,72,119,120,123  S£_54 125
 m-bis-m-(phenoxylphenoxy)-benzene,121  Dexsi1-300,14  Emulphor  ON 870,33
 carbowax-20 polymer,122 OV-1,124 OV-17,9 OV-101,9'73'91  apiezon L,47'76 and
       "I OC
 SP-2100     all  have been used  as packing materials for capillary columns.  The
 most widely used column  for separation of POM is SE-52.  Cantuti et  al.119
 compared several  liquid  phases  and  found that a SE-52  column of 50 m  length
 was most effective for POM separation  and  had a theoretical plate of 40,000.
 Recent investigations by Bjrfrseth125 have  demonstrated that capillary columns
 using SE-54 as the stationary phase have excellent separation efficiency, low
 column bleed, and long-term stability.  Both PAH and aza-arenes have been
 separated with this packing material.125  A 40-m capillary column  with Versamid-
900 as the stationary phase is very effective for the separation of  aza-arenes
(theoretical plate of 60,000).102
                                   4-35

-------
                                      128
     Investigations by Lao et al.127 with Dexsil-300 coating liquid has  shown
that the separation of POM on surface-coated open tubular column is better
than packed columns for the range from fluoranthene to methylchrysene.   How-
ever, the resolution of the later peaks with longer retention times remains
incomplete.
     Analysis of  POM carried out by wall-coated capillary columns, coated
with SE-52  of liquid phase thickness  of 0.27 urn showed excellent reproducibil-
ity, high  sensitivity,  and high resolution  between  individual  isomers in
comparison to regular  packed columns.128  WCOT columns have been claimed
offer  numerous  advantages over packed or porous-layer open  tubular columns.
Some of these are superior resolution, minimum adsorption effect,  and  high
 stability.
 4.4.7.  High Pressure Liquid Chromatography
      The ability of HPLC to separate POM compounds not normally accomplished
 by GLC indicates the great potential of this method for POM analysis.   In
 addition,  HPLC offers many advantages over GLC:   it operates at much lower
 temperatures, fraction collection  is  simple and convenient, the capacity to
 accept sample is high, and  fluorescence  spectroscopic detection demonstrate
 high  sensitivity and  selectivity for POM compounds.
       In HPLC,  the mobile phase modifier  is the most powerful  variable  for
  changing  retention times.   The choice of mobile  phase is dictated by the mode
  of separation  accomplished by the  sorbent.  In  the normal  mode,  generally
  hydrocarbon solvents are used as the mobile phase."  Any solvent that  dissolves
  POM in fair amounts, is miscible with water,  and is transparent at U.V.  absorp-
  tion wavelength (for U.V. or fluorescence detector) can be used as a modifier
  in the reverse phase-L.C.
4-36

-------
     It should be mentioned that solvent programming has proven to be a power-
ful tool  in separating the compounds from each other.  Several researchers
have taken advantage of increased temperature (70 to 80°C) of the column to
obtain a  similar objective.  At higher temperature reduced viscosity of the
mobile phase increases column efficiency and decreases relative retention
times.  The sample capacity is also increased at higher temperatures allowing
larger amounts of sample injection.  Table 4-10 summarizes the different
columns and mobile phases used by various authors for the analysis of POM by
HPLC.   The most versatile columns for POM analysis operate on the principle of
reversed-phase mechanism.   One such column containing u-Bondapak r   has been
applied for the separation of both PAH97 and aza-arenes.144
4.5  DETECTION
     The different methods of detection with their relative advantages and
disadvantages have been individually discussed in the following sections.
4.5.1   Flame lonization and Electron Capture
     When gas chromatography is used for separation and analysis of POM the
usual  method of detection  is either flame ionization detection (FID)11'53'77'
  '    or electron capture detection (ECD).  7'93'108  One advantage with the
EC detector is its greater selectivity towards BaP compared with BeP.148
However,  the high temperature used in the column oven for the separation of
higher boiling compounds requires even higher temperature capability for the
detector.   Most EC detectors do not fulfill  this requirement.   As a result,
most of the recent GC analysis have been performed with FI  detectors.   The
advantage with the flame ionization detector is  its wide dynamic range of linear
response for most POM.
                                   4-37

-------











to
CO

to
f
o.
t
i
i
^
0
u

i
i
i
i


'j
(
1
•i


en
CM
i.
0

Q)
C
0
u
(J
Loo
•r- >— 4
O U-

Adsorptior






u
•f
CM
re
a, c,
C'r-
04*
4* re
C i-
0) u
0.0
l to
C-r-






re
c

in

t
o
CM
r—



^

a.
•a c
t- o
ro-i-
•o-P
C 3
rai—
•p o
co to


0
CO
r—






f
>•
=

•a
a,
to
S- Q)
a) to
> re
O) -C
a: Q.







go*
qj in
io
r^ c,
04*
=rs
u
*e o
o to
CO-r-








t_l
i
•*—
CO
CO
a
o



^x
y
re
3
c:
o
i-
10


2

•
u
o
u_

t
>
=

•o
a,
to
(. a,
a> to
> re
0) J=
on a.






o
c o
O CM
15
a, r-
4* i: o
coa>
_ai IDS:
(-00

«
•§"
c
Q
ffi

3.
CJ~
re
1
o
m
a>
o.
3 00
O I—
oo





oB
•=5
a> re
_E 3-
•5 re
a, a>
to in


CO CO CO CO






• • . •
>•>•>• >
= = ' = =

.° Tg -o l-
4J a, a, i a,
O. to to a, M-
s_ s. a, $- a, o to
o a, 
"- i g T-re
•— £ £ -uc:
•i— a) Q) i cn
u>  tn -^ Q.
O O CD " E
a. o o cM-r-



3T 3=oB 3=

Q) Q ^* QJ O
t^ 4^
•i- -a c *o -r- c *a c

E 3 T3 4^ ^1 ^ •*•* ^ -^
3X 4^0 4JOJO 4->O
< Q> (/) (/> CO S U) CO U)
4-38

CO CO
^~ . f~






• •
> >
=> =>

(. -a
i a> a,
0) 1- to
oi in c. a,
t- c a, to
j: (- 'Hi f
o 4* os a.




c
re
4* O
o o
43 4* ^CM
Q) *^ dC
4J >i- D13: O
1 £ m 1 ^
II 111


1 0)
fr. re
0 X
3 O
^ !—
"»-T-
o in
t-
£>,
c£^
re 4*^
s- 
4**r- tO
a) "o re
4*r- .=
Pa. re
- 0 E
m s- s-
•• a. a>
**> € C5
U)
3
re
x
3C 3
tf t/)
o. a; a>
•a c •-- re

^ 4* E _fl)
re i— 4* c
4* 0 30
to in < u
^•^•1

% $
^ r™





U tl
O O
i^ ,?
u_ u_
c
5 ^
Q. U)
tT i. a,
o a, to
to > re
s s-g.







°CM
i
C -r- 3: •£
u c- z; t-
I-H >^ 00 '^^







0
o
i oo
z. v~
i i
a> re
^ ^
a. o
to u



3= £
^ ^
Q. Q.
T3 C T3 C

C 3 C 3
4* "o 4*"o
to 
-------
            01
            u


            I
            0)
         O      i—
   0)

Ml
  •o

  +>
  o
  o
S:
01
        re
        a. u
 1/3
 =5
                 o
                 3
        "O      -O
        ai       a,
        in       u>
        s- a>     t- a>
        O) in     a> in
           re
                         >  re
                                  (.
                                  o
•o
 at
 in
 s- a>
 a> u>
 > re
 Olf
ce o.
                                    c
                                    01
                                                                         c_
                                                                         o
                                                          t-
                                                          o
                                          S-
                                          ro
                                            •O
                                             at
                                             in
                                             s-  a)
                                             0)  u>
                                                ia
   a.
   s-
   o
   in

  •
                        °cx,
           -0.2
           Ol +J
                                                              Q)  O
                                                              >  in
                                                              0) -U
                                                             O£  03
                                                             ro o


                                                             .-g
                                                             O) Q.

                                                             •!-> O
                                                             ID S.

                                                             » Q.
                                                           O>
                                                           in
                                                           I- fl>
                                                           o> in
                                                           > n)
                                                           our
                                                          0£ O.
Sr
o
                                B-
                                                     O) oi
                                                     o  c
       m <,_

       r— O


          0)
       o-o
          O
 a.
 LU
 to
10
z
o
       £2
                                                       u
                                                    *>.•!-

                                               g
     £01
     0.
  3 I
  in c
  in

  2B
 CL. CD
                                          c
                                          re
                                          x
                                          Q>
                                                t- 01

                                                •r- re


                                                II
                                                01
                                                             re    o

                                                          St.    re
                                                          r>-     O
                                                          o ro    s-
                                                          CO S   Q-
a>

J3
ro
      •o
       «
       N
               o
               £-

               O
               s-
              •p
               a)  re    x
              T3  U    re

               *•-    -s
               U-i-    O
              O  in    r>i
                                                    re
                                                   03
                                                    E


                                                    Ol
                                    2
                                    o
                                    a.
                                    i
                                                            S
                                                            re

                                                            f
                                                                       08

                                                                         ao
                                                                                re
                                                                                Q.
                                                                                re
                                                   o
                                                   o
                                                                                         o
                                                                                         a.
                                                                                        at aj
                                                                                        j^ c
                                                                                        co
                                                        i  i—
                                                        in o
                                                        in c.


                                                       II
              =
              re r
              *>
              co
 i.
 re
 o-t.
   01
 u -u

'si "re
 «•

 o. o>
 in -u
 o re
                               Q- a)
                               in -u
                               o re
                               II
                                O
                  re •—
                  •t-> o
                  co in
t- O
re -i-
T3 -u
c: 3
m t —
+» o
CO in
                                                                       re

                                                                       6
                                                           fi  3
                                                            o. u
                                                            in-r-
                                                            o •<->
                                                            E=  s_
                                                           -S  re
                                                           
                                                         £-0
                                                         re
                                                      •r- 4J

                                                      i— U

                                                          3
                                                          "
                                                4-39

-------
4.5.2  U.V. Absorption
                                                                     9 12 28
     This technique is the most widely used method for POM detection.  '   '  '
44,46,84-86,91  The advantages of this technique include the commercial  avail-
ability of high quality spectrophotometers, the relative insensitivity of the
degree of absorption to trace impurities, and the fact that the absorption
spectrum of a mixture is usually the sum of the spectra of the components.
The  latter characteristic offers a crosscheck of the total concentration  in a
        j
complex mixture determined by other methods.  The disadvantages include the
requirement of two  or more separation  steps for reliable identification and
the  relatively lower detection  sensitivity compared to other detection devices.
      Although the total  spectrum for  each  POM compound is  unique, the indivi-
dual features are not.   Therefore,  identification and quantification  of  a POM
compound  on the basis of only  one  peak is  unreliable.  A computer program
 utilizing the U.V.  spectra has  been developed for rapid screening and simul-
                                                  147
 taneous identification  of a large  number of peaks.
      Incomplete separation of individual compounds  is  a major  problem in
 interpreting U.V. spectra and since the compounds  rarely  completely separate,
 questionable results are obtained with U.V.  detection of  complex mixtures even
 with a programmed spectral analysis.
 4.5.3  Luminescence Analysis
 4.5.3.1  U.V. excited luminescence emission—This technique has become well-
 established as a sensitive and selective technique for POM detection.  '
 68,71,72,84,91,99  Several groups have  found it to be at  least an order  of
                                                       •I OQ 1 AQ
 magnitude more sensitive than  absorption spectroscopy.    '     It is also more
 sensitive (lower detection limit) and less expensive than mass spectral
                                     4-40

-------
  detectors.  Mass spectrometers normally have a nanogram detection limit,150
  although integrated ion-current techniques reduce this limit to the subpico-
  gram range.151
       Since luminescence spectroscopy is nondestructive, individual fractions
  can easily be collected and subjected to further analysis.   Furthermore, this
  method provides additional  selectivity.   While  many interfering compounds may
  absorb light,  only  a few emit in  the region  of  particular  POM compounds.
       Various modifications  of this  method  have  been  used to  increase the
  sensitivity and selectivity of the  method.  Application of low temperature
  fluorescence and phosphorescence for the characterization of  aromatic com-
  pounds with enhanced sensitivity has been  utilized by Sawicki and Johnson.152
  However, this technique  is  rarely used at the present time.  Room temperature
 Phosphorescence measurements have been made without the problems of typical
 low temperature sample matrix preparation, yet maintain good sensitivity and
 selectivity.   Phosphorescence is  increased by adsorption of the sample  on to
 a rigid matrix-like  backing  such  as  filter paper,  thereby quenching non-radia-
 tive vibrational de-excitation,153 and  by the  addition  of a heavy atom  per-
 turber which increases  the spin-orbit coupling between  the  singlet  and  the
 triplet excited  states.154
      The selectivity of fluorescence  detection for POM compounds  has been
 increased by characterization at two wavelength combinations, e.g., at 313/360
 and  365/445 nm as a double fingerprint..155  By means of  a selective excitation
 and/or oxygen-quenching method, POM compounds have been  quantitatively analyzed
without chromatographic separation.156  Identification of compounds showing
minor features  in zeroth derivative fluorescence  spectra has been accomplished
                                   4-41

-------
by first and/or second derivative fluorescence and fluorescence modulation
techniques.31  Since derivative spectroscopy deals with the shape characteris-
tics of the spectral absorption rather than the intensity changes, increased
sensitivity and selectivity are obtained over conventional absorption spectro-
meters.  The curvature characteristics are often quite large and specific to
individual compounds, thus allowing selective analysis of a component in a
rather complex mixture.  In addition to the scanning mode of operation where
a complete spectrum is obtained, the second derivative spectrometer can be
operated  in  a "dwell" mode at a specific wavelength allowing for real-time
monitoring of a selected component.
      In situ fluorescence detection on a thin-layer plate with a scanner  has
                                                                80 90 94  95
also  been used to  increase the detection limit  of  POM  compounds.   '   '   '
Fluorescence spectroscopy using variable wavelength detectors  has been used
to  increase  the selectivity of detection in the HPLC method.   '    Finally,
multicomponent analysis by the use of  rapid scanning fluorescence  spectro-
scopy interfaced with a computer  for data  reduction by algorithm  programming
                  157
has been  studied.
      However,  there are quite a  few complications  with this detection technique.
All emission techniques are  influenced by  repeatability of wavelength settings,
 scatter of light in monochromators, and random fluctuations in source inten-
 sity.  Quenching by oxygen is a serious problem and may be very selective.
 Impurities which introduce intersystem crossing will  enhance phosphorescence
 at the expense of fluorescence intensity.
 4.5.3.2  Shpol'skii effect—Detection of PAH compounds by this technique was
 first demonstrated by Shpol'skii158 in 1952.   The characteristic sharp-line
                                    4-42

-------
  (quasi-linear)  luminescence emission  spectra  obtained by U.V.  irradiation  of
  the  crystalline matrix  of PAH  in  n-paraffin solvents  at  77°K  or  below  have
  been used  for the  identification  of these compounds.   The original n-paraffin
  solvents used for  this  study were n-hexane and n-heptane.  However, most PAH
  compounds  may not  be soluble in n-paraffins.  Therefore, Kirkbright and
  DeLima159  used  10% (v/v)  cyclohexane in n-paraffin as the solvent mixture
  which was  demonstrated  not to disturb appreciably the spectrum obtained.  From
  the study  of ShpoT'skii effect with 23 PAH compounds in several n-paraffin
  solvents,  such as n-hexane, n-octane,  and n-decane,  these authors also con-
 cluded that some PAH produced more well-defined spectra in a particular n-
 paraffin solvent compared to another.
      Although it appears that direct measurement  of  the intensity of  quasi-
 linear luminescence emission is capable  of permitting  the quantitative  deter-
 mination of PAH  compounds  directly, real  samples  pose  several  problems.  The
 effects  of  energy transfer,  the inner-filter effect, and  experimental variables
 all cause error  in  the analysis.   The  use of combined  internal
 standard -  standard additions technique has been  demonstrated  to yield  accept-
 able  results  in  such samples.159  This luminescence emission method is  also
 suitable as a "fingerprinting" technique for qualitatively identifying  PAH in
 a mixture.
 4-5'3'3  x-ray excited optical luminescence (XEOLV-This method is a slight
 variation of Shpol'skii effect in that it uses  X-ray (1 to 10A) to obtain
 quasi-linear fluorescence and phosphorescence-emission  from samples frozen  in
 n-heptane matrix at 90°K.  D'Silva et  al.16°  reported this technique as  a
method for the detection of nanogram levels of  PAH.   The suitability of  this
                                   4-43

-------
method in the analysis of real samples remains to be established.   However,
some of the potential advantages for X-ray over U.V. excitation are:   (i)
freedom from optical cross-talk between the exciting and luminescence radia-
tion; (ii) larger population of higher energy electron levels, thereby reveal-
ing additional lines; and (iii) substantial phosphorescence emission, aiding
further characterization from these lines.
4.5.3.4  Sensitized  fluorescence—The inherent fluorescence of PAH compounds
is greatly enhanced  in the presence of certain sensitizers, thereby decreasing
the fluorescence detection limit of the former compounds by several orders of
magnitude.  This technique has been employed  as  a spot test method for screen-
ing environmental  samples for the presence of PAH.     According to this
method,  naphthalene  as sensitizer is  co-spotted  with the sample spot on a
Whatman  #42 filter paper.  The dried  spot is  exposed to light of appropriate
wavelength and intensity and any sensitized fluorescence emission  is recorded.
On  filter paper,  10  pg of PAH in a  spot can generally be visualized when
treated  with  naphthalene.   Experiments with a few real-life  samples  have shown
that the method is specific for  PAH with  minimum interference from other
compounds.   However, large concentrations of  nitroaromatic compounds  have
been shown  to act as fluorescence  quencher in this  method.
4.5.3.5  Synchronous luminescence  spectroscopy—A recently developed method
 termed synchronous luminescence spectroscopy, where the excitation and
 emission wavelengths are simultaneously varied with a fixed  wavelength differ-
 ence of 3 nm between the two, has  been proposed162 for the routine monitoring
 of PAH in complex samples.  , The spectra obtained from this technique is simpli-
 fied with improved resolution between component peaks.
                                    4-44

-------
  4.5.4  Mass  Spectrometry
       Mass  spectrometry affords a sensitive means of determining the probable
  identity and relative purity of POM chromatographic fractions.  This method
  has gained wide acceptance for analysis of POM compounds 32»33.59,73,112-
  115,120,121  r
              Generally, tfns has been accomplished by low voltage ionization
 methods which produce spectra containing mainly singly-charged,  intact molecu-
 lar ions.
      In POM analysis, the most widely used methods  for sample introduction to
 the mass spectrometer are direct insertion probe32'112'151'163 and d1rect
 interfacing GC-MS method. 106.115,121,127  The sample  ioni2ation  .g ^^
 done by electron impaction and the  separation of  ion  beams  is accomplished by
 either low resolution quadruple  instrument or high resolution magnetic scan-
 ning equipment.   The claimed113 advantage  of  quadruple  instrument is:   (i)
 its initial  relatively low cost;  (ii)  ability for straightforward  maintenance
 and repair;  and  (iii)  extremely high-speed linear mass scan,  simplifying
 system control,  data logging and spectra interpretation.  The  disadvantage of
 the quadrupole method  is its low mass  resolution and low mass  sensitivity at
 higher  masses.   Both high  resolution32'151'163'164 and low resolution106'113'114
 mass spectrometers have been used for  PAH and aza-arene analysis.
     The analysis of complex samples usually produces  numerous spectra and the
 spectrometer generates an enormous amount of data from a single chromatogram.
Therefore,  the data reduction and quantification requires a computerized data
processor with an appropriate programming system.   At  the end of  the GC run,
the computer is  used to plot a reconstructed gas chromatogram of  total  ion
amptitude versus  the spectrum number (ion abundance  chromatogram).
                                   4-45

-------
Identification of these chromatographic peaks is accomplished by plotting the
mass spectrum of a specified peak or by specific ion monitoring (mass frag-
mentography).  t
     One problem with electron impact ionization is that isomeric POM may
give identical mass spectra making unambiguous identification sometimes
impossible.  Kites and Dubay155 found that charge-exchange-chemical ionization
mass spectrometry can be  used to distinguish many isomeric POM.  They have
established  that spectra  of POM have characteristic ratios of the protonated
molecular  ion  to the molecular ion when 5 to 10 percent methane  in argon  is
employed as  the reagent gas.  This technique is not expected to  replace
electron impact GC-MS, but it should be a useful  supplemental tool for
differentiating both  isomeric PAH  and  aza-arene.
4.5.5   Other Techniques
      Fourier-transform NMR spectroscopy56'72 has  been used  for  the elucidation
 of the position  of substitution  in the benzene ring positions  of the POM.
 Other techniques  of detection  utilizing matrix isolation  Fourier-transform
 IR166 and Piezoelectric crystal167 need considerable improvement to  become
 viable methods for POM analysis.
 4.6  ANALYTICAL METHOD USED BY EPA FOR NASN MONITORED SAMPLES
      The following method173 was used for the quantification of BaP from NASN
 monitored samples collected during the period 1966-1972.
      The particulate sample collected on glass-fiber filter was extracted with
 benzene in  a  Soxhlet extractor for at least 6 hours.  The filtered extract was
 dried  in  a  60°C oven.  The residue was dissolved in methylene chloride  (1.0 ml
 for 25-mg residue) and a portion of this solution was spotted on a  scribed and
                                     4-46

-------
  activated TLC plate coated with 250 Mm alumina.   After developing the spots in
  benzene-pentane (2:8),  the BaP spot was scraped  from the plate.   The quantita-
  tive  elution  of BaP from the adsorbent was  accomplished by shaking with
  pentane:acetone (95:5).   The eluant was dried  in a vacuum oven  at 30 to 35°C
  and about  50  torr.   The  residue was dissolved  in concentrated sulfuric  acid
  and the  fluorescence of  this  solution  was recorded at  excitation  and emission
  wavelength of 470 nm and  540  nm, respectively.   The concentration  of BaP in
  the unknown was determined by comparison with standard BaP solutions.
      For samples collected during the period 1973  to the present, the follow-
  ing modified procedure174 was used for the quantification of BaP.
      The quarterly composites of glass-fiber filter were Soxhlet extracted
 with cyclohexane for six hours.   The extract was  concentrated in a Kuderna-
 Danish evaporator at 50°C with a stream of dry nitrogen.   50 Ml  of the con-
 centrate was  spotted on  a channelled 20% acetylated cellulose plate with a
 •nultispotter.   P!ates were developed with 2:1  ethanol.-methylene  chloride
 solvent mixture  and  air  dried.   The  BaP spot on the plate was scanned with  a
 thin layer  plate scanning attachment at excitation and  emission wavelength  of
 388 nm  and  430 nm, respectively.  The concentration was evaluated  from the
 integrated  strip chart area reading  obtained by a digital  integrator.
     Since  the BaP in this procedure  did  not separate from anthanthrene, the
plate was then scanned at  an excitation and emission wavelength of 434 nm and
470 nm, respectively, for the determination of anthanthrene concentration.
The anthanthrene contribution in BaP could be corrected by subtracting 16% of
anthanthrene concentration from the BaP reading (at BaP wavelengths, anthan-
threne was found to be 84% less efficient).
                                   4-47

-------
4.7  DIFFICULTIES IN POM MONITORING AND ANALYSIS
     The data reported on the levels of POM in different sources are suscept-
ible to various kinds of errors and uncertainties at different stages of the
monitoring and analytical phase.  The first source of uncertainty in monitor-
ing POM levels arise from the lack of knowledge regarding the degradation of
these compounds under environmental conditions (see Section 3.3).  The eval-
uation of adverse health effects requires a judicious selection of parameters
that need to  be monitored.  The selection process becomes difficult  in the
absence of  specific information regarding the form of POM that may exist in a
certain atmosphere  as a result  of  environmental  reactions.  Even  in  cases
where  it  has  been  decided  to  monitor  a  specific  form of  POM,  e.g., a selected
group  of  PAH, the  reported values  are subject to substantial  errors.  Such
 errors and uncertainties may  arise during  defining the monitoring strategy,
 sampling procedure, transportation and storage  of samples,  and analytical
 procedures.  Possible sources of error in  each  of the  individual  categories
 will  be discussed in the following sections.
 4.7.1   Sampling Errors
 4.7.1.1  Nonisokinetic sampling—Most ambient sample collections are done
 non-isokinetically:  That is,  as the air containing the particulate matter is
 drawn into the sampler inlet,  the speed and direction of the air are changed.
 The inertia! characteristics of suspended particulates bring about losses  of
 larger particles both by drift from the sampled air stream and by impaction on
 the surface  of the sampler inlet.  Although it is difficult to estimate the
 error arising from nonisokinetic  sampling, such errors may be minimized by
 avoiding eddy current, turbulence, divergence or convergence, and changes  in
 the direction of the sampled air  stream.
                                     4-48

-------
  4'7-1-2   Non-incorporation  of  sampling  efficlencv-The  evaluation of  sampling
  efficiency  is  particularly  difficult  under ambient atmospheric sampling condi-
  tions.  The percent  recovery of a collected sample is,  however, rarely deter-
  mined either during  ambient or source sampling.  The lack of incorporation of
  sampling efficiency  lends considerable uncertainty in the reported POM levels.
  4-7-]-3  Evaporative losses and reactive conversion-Ennprt.inn *f onM Pn
 widely-used glass-fiber filter alone without the back-up adsorbent is subject
 to two kinds of errors.  First, lower-molecular-weight compounds  on adsorbed
 particulate phase may have some equilibrium vapor concentrations  under atmos-
 pheric conditions.   This vapor part will escape collection on the  filter.
 Second,  air passing through  a  filter containing the  collected substances will
 carry away amounts  equal  to  or  less  than the equilibrium vapor  concentration
 because  of desorption from the  adsorbed  substrate.  Quantitative data  on
 losses of  POM  from  soot particles collected on  filter have been reported by
 Thomas et  al.,37  DeWeist  and Rondia,24 Murray et  al.,38  and Rondia.39  Pupp et
   40
 al.   determined  the  equilibrium vapor concentration for a number of POM and
 concluded that  considerable  losses occur during ambient  sampling of compounds
 having an equilibrium vapor  concentration of 0.5 x 10~3  mg/m3 or higher at ambient
 temperature.  Considerable losses are expected with pyrene, anthracene,
 phenanthrene and BaA.  However, the equilibrium vapor concentration of com-
pounds adsorbed to particulate matter may be quite different and lead to much
different conclusions.
     Besides sublimation, the chemical  changes  of POM which may occur on the
sampling probe and on the filter paper require  further investigations.   The
first of the chemical  changes may occur as the  result  of  catalytic  effects  of
                                   4-49

-------
                                                 4?
the probe material used during sampling.   Hermann   studied this effect and
concluded that neither Vycor nor stainless steel used as probe material causes
any catalytic effect in promoting chemical reactions.  However, there are many
oxidants in the air which are drawn through the filter during sampling.  Lane
and Katz43 using simulated atmospheric conditions have demonstrated that POM
coated on a glass surface will be rapidly oxidized by the prevailing oxidants
in the atmosphere.  Their work also indicates that BaP is more prone to oxida-
tion than BbF.
     Since atmospheric POM exist on the surface and  in the interstices of soot
particles in  a multi-layered form, the surface-exposed POM.on the filter may
be oxidized rapidly by the prevailing oxidants  in the atmosphere followed by a
slower penetration reaction of the subsurface POM and POM'adsorbed  in  the
pores of particles.  However, reactions in the  latter two cases can  be expected
to be slower  not  only because the oxidants have to penetrate  to their  surfaces
but  also because  a layer of oxidized material may protect them  from  further
oxidation.  Quantitative data regarding this oxidative conversion of POM
during collection are  not available.  There  is  also  conflicting evidence
regarding the importance of this process.  For  example,  by passing  clean  air
at a rate of  100  liters/min through  filters  containing  airborne particulates
Matsushita  et al.41  found negligible  losses  of  PAH when  air  treatment  was less
than 7  days.   But by passing  air for 3  weeks at a  rate  of  3  liters/min through
 filters  containing particulates,  Commins168  reported 76  percent and 87 percent
 losses  of fluoranthene and pyrene,  respectively.   No significant  losses  were
 observed for high molecular weight compounds,  including BaP.   Sawicki  pointed
 out that he failed to observe significant changes  in pyrene concentrations
                                    4-50

-------
 under similar conditions.168  Whether the lower temperature of the filter and
 the passed air was responsible for the lack of loss was not clear from his
 comments in the above reference.   It is also not clear from these experiments
 whether the losses of PAH are due to oxidative conversion,  purely evaporative
 processes or both.
 4.7.2  Losses During Transportation and Storage
      Another source of error that needs to  be  properly addressed  is  the  loss
 of  POM  during transportation and  possible storage  of the  collected samples
 prior to analysis.   Losses  during this  period  can  occur from  three sources:
 (1)  photooxidation,  (2)  volatilization,  and  (3)  chemical  oxidation.  The  first
 of  these losses  can  be easily prevented by transporting and storing  the sample
 in  the  dark.   Losses  of  POM  due to  the  other two factors  have been noted  by
 various  investigators.   A decrease  in the concentrations  of fluoranthene,
 phenanthrene,  and pyrene by  40 to 60 percent,  and  rearrangement of dibenzo[a,l]-
 pyrene to dibenzo[a,e]fluoranthene  and  BaA to  chrysene  have been noted by
 storing  the  sample for 3 weeks.169  Hermann42  has  noted considerable error due
 to degradation or reaction when storing the sample at room temperature and
 exposing it  to air.  Similar  results have been observed by other investigators;
 the results  of Commins    are shown in Table 4-11.
     Since the samples in Table 4-11 were stored in the dark at room tempera-
ture and were exposed to air, the losses may be due to oxidation,  sublimation,
or both.  By storing the sample in glass bottles not exposed to air,  Hermann42
reported no significant loss of POM.  The storage of samples at cooler temperature
(in a refrigerator or in contact with ice) will further prevent the possibility
of losses due to both volatilization and chemical reaction.
                                   4-51

-------
       Table  4-11.   PERCENTAGE  LOSSES  OF  POM  FROM  FILTER  PAPER  DUE  TO

                          VARIATION  OF STORAGE  TIME168
     Compound
% losses under different storage times'
3 wks. at                    1 yr. at
room temp.                  room temp.
Fluoranthene
Pyrene
Benzo[a]pyrene
Benzo[e]pyrene
Anthranthrene
Benzo[ghi]perylene
Coronene
37
34
5b
+22b
+9°
9

92
88
32
23
21
10
1
? The samples were stored in sealed envelopes.
D These are the % gain (instead of loss) observed.
                                    4-52

-------
 4.7.3  Errors In Analytical Procedures
 4'7-3'1  Losses during desorption process—Several sources of loss are
 possible during analytical treatment of samples.  The first of these takes
 place during desorption of POM from the sampling filter.   The loss during
 this step depends on the nature of POM and the method used for desorption.
 The widely used method of Soxhlet extraction can give poor recovery for some
 POM and the recovery is dependent on the solvent used (see Section 4.3.1.1).
 4-7-3-2  Column and thin-layer chromatographic and storage losses—Analytical
 procedures  involving column chromatography can cause  losses  of POM either
 due to  irreversible adsorption or photoreaction.85 The photoreactivity of a
 number  of POM  in some adsorbed phase is  considerably  greater than  in solu-
 tion.   Similar losses due to photoreaction on  TLC  plates have been  observed by
 Inscoe.      Therefore,  it is absolutely  necessary  that all analytical proce-
 dures be  performed  with  long wavelength  (red)  light of low intensity.   The
 storage of  samples  during analysis should,  likewise, be done  at -4°C and  in
 the  dark  to  avoid losses.
 4'7-3'3   Losses  in  evaporative concentration step—The loss of POM  in evapor-
 ative concentration steps  is another most frequently encountered source of
                      O"|
 error.  Fox and  Staley    reported a recovery of only 81 to 83 percent of BaP
when the  flash evaporation of solvent from extract was allowed to proceed to
dryness.  The loss  from this step can be expected to be even greater for other
POM with  higher volatility, such as pyrene, BaA, anthracene,  and phenanthrene.
4-7-3-4   Errors due to incomplete resolution-The greatest discrepancies in
the reported literature values  for monitored POM (as in 1-Aza-fluoranthene in
Table 5-5) probably arise due to incomplete resolution of  isomeric  compounds.
                                   4-53

-------
In some instances separations of isomers are very important from a toxicological
point of view because two isomers may exhibit different carcinogenic character-
istics, e.g., BaP and BeP.  Some of the hard-to-separate isomeric compounds
are:  BaA,'chrysene, and triphenylene; BbF and BkF; BaP and BeP; benzo[b]chrysene
and indeno[l,2,3-cd]pyrene; benzo[ghi]perylene and anthranthrene; and the
alkyl- substituted products of the respective group of compounds.  Most packed
columns,  including the widely used Dexsi1-300 GC, will not separate these com-
pounds.106  Although the  separation  is  better with Dexsil-400 GC  columns, it
                         106
is by no means complete.IUD  Columns packed with liquid nematic crystals  have
been claimed to accomplish much better resolution.116'117  But column bleed at
higher temperature has restricted the use of this packing material.   Improved
resolution using WCOT columns has been obtained.  However,-using a 50-m
capillary column coated with OV-1, Lunde and Bjrfrseth failed to separate BbF
from BkF.124   Lee et a!.56 used an 11-m capillary column coated with SE-52
and showed that alkyl- substituted derivatives  of many isomers could not be
separated.
     TLC can  provide better  resolution when the mixture  is spotted on the
plates  coated with  composite adsorbents  and developed  in two  dimensions with
appropriate  solvent systems.   The limitation  of the TLC  method lies  in its  in-
ability to separate a  mixture when a multitude of components  are  present.
 Even in cases where separations can be achieved, losses  are  encountered  during
 the quantification procedure no matter whether solvent elution of the TLC spot
 or in situ scanning is used.
      The commonly used modern technique of HPLC separation has the advantage
 of better resolution of the hard-to-separate components already mentioned.
                                     4-54

-------
 The reverse-phase packing material, p-Bondapak C]8, provides a good separation

 of alkyl-substituted components.  In complex samples containing alkyl-substi-

 tuted as well as unsubstituted parent PAH, the use of reversed-phase HPLC

 could yield misleading results.  For example, 1,3,5-trimethyl naphthalene, 1-

 methyl phenanthrene, and fluoranthene elute simultaneously with CH.CN-H^O
                                                                   «J    £.
 mobile phase even though these compounds consist of two, three, and four
                     14.^
 rings, respectively.

      It should be mentioned that in some instances correction for the losses

 during the analytical  procedures have been attempted by the use of internal

 standard(s).   However,  when the quantification  of a number of parameters are

 attempted,  correction  by  the addition of one  or two internal  standards  may not

 be adequate since the  losses are not uniform  and  vary  from compound to  com-
 pound.

 4.7.4   Uncertainty in the Number of  Parameters  to be Monitored

     Lee  et al.   compiled a list of more  than  100 PAH compounds which  have

 been detected  in  airborne particulate matter.  The list  could be considerably

 higher  if other neutral and  basic aza-arenes are  included.  The vastness of

 the POM and the uncertainty  in the number  of parameters  which need monitor-

 ing constitutes one  of the biggest problems.  Certainly  it is necessary to

 determine the  levels of hitherto.undetected compounds which may be present in

 the air in  order to  evaluate their potential health effects.  But a cost-

 effective routine analysis requiring good precision and  accuracy often demand

 that the number of parameters be  restricted to a few selected ones.  There-

 fore,  there is a great need to define a few representative POM which will

measure the carcinogenic potential of atmospheric POM to man.
                                   4-55

-------
4.8  ECONOMIC CONSIDERATION IN THE SELECTION OF THE ANALYTICAL TECHNIQUE
     The cost of POM analysis will depend on the selected method of analysis.
The method selection is, in turn, dependent on the objective of the analysis.
When a method is intended to be used for dosimetric purposes in occupational
settings, it must be capable of real-time monitoring of POM in the event of
a process leak  or other  increase  of the POM content in the working atmosphere.
In addition to  this requirement,  integrating dosimeters are needed for  deter-
mining  the integrated  exposure received by workers and the extent of  contamin-
ation of the  tools  and the  working  areas.   Irrespective of the  economic con-
siderations the commonly used analytical  methods are  time-consuming and do  not
meet these dosimetric  requirements.   Monitoring instruments,  such  as  time-
 resolved fluorescence, correlation spectroscopy, second  derivative spectro-
 meters, and other methods which fulfill  dosimetric requirements,  are  in the
 developing stages and it is difficult to project cost estimates for these
                                                                 171
                                                                     for a
methods.  The reader is referred to a report by Hawthorne et al.
review on this subject.
     On the other hand, several analytical methods are presently available for
the determination of the nature and the level of POM collected from environ-
mental  samples.  The selection of a particular analytical method depends on,
(1) number of parameters to  be determined,  (2) desired accuracy for each de-
termination, and (3) sensitivity of the analytical method which determines the
lowest  detection limit of  the measured parameters.  A comparative  cost estimate,
through-put, and other relevant  data  for  four commonly used techniques is shown
 in Table 4-12.   All  the data in  this  table are estimates with some values taken
 from available literature.31'172  These  costs do not  include the  cost of initial
 survey, sample collection, method validation, etc., which  can be  a considerable
 cost factor in any sampling program.
                                     4-56

-------
TJ
C

















i — i
** i""*
— i 	 i
0. 
— ^
__J
\

5
i—
3
™
3
o
CJ





.
3
3

_5
_l























-^
O
Q.
£
O
U

LO
I^
I
O
•"
in
•a
§
o
Q.
^
O
U
in
CM
o
00



TJ
c.

o
Q.
O
U

CM
r— .
»
•P
•f—
•f—
-Q
re
u
"a.
Q.
,
o re
to "a co
T oo >






in
i— T3
•w- •— co





CO

a.
E
re
to
^^N "O
•P C
in 3
0 0
u s-
re
*~rft f ^*
re s- re
C 3 TD
O -P •^
«r- to
-p cu 





O
CD
cn co
CM
CM r—
4- "^






re i — i —
o re re
•1- S- •!-
•P Q> -P
5*> r~ «^.
1— Q. C
re -r- .r-
c: s-
re co »
Q- to
-P
>> T3 C
-p c co

> -P
•r- r— IO
-p re co
•i- -P >
in •!- c
c O.T-
co re
tO «_3
4-57

-------
     It can be concluded from Table 4-12 that GC-MS-computer technique is
preferred over other methods for non-routine analysis involving a vast number
of parameters.  The initial capital investment for this method is extremely
high, although the recurring expenditure is comparatively low.  For routine
analysis of a dozen or so parameters, HPLC-fluorimetric technique appears  most
promising in view of its cost effectiveness.  The TLC fluorimetric method
should be the method of choice when one or two parameters are intended to
be monitored.  This method needs the least amount of capital  investment and
operator skill for performing the  analysis.
      It should be pointed out that Table 4-12 is not intended to include all
the  available methods, but rather  serves as  a guideline for the evaluation of
cost effectiveness between methods.  It is  also evident that  the selection of
a particular  analytical method  is  rarely based on pure economic considerations
alone, but  primarily depends on the  intended objective of the analysis.  The
vast number of POM in  the  environment  and  the uncertainty in  the number of
parameters  which need  monitoring constitutes one  of the biggest problems in
the consideration of cost  evaluation.   Certainly, monitoring  of BaP  as an
 indicator for environmental  POM pollution  is cost effective;  but the selec-
 tion of this parameter alone for the aforementioned purpose remains  question-
 able.  One of the great needs  in this  regard is  to  define a few representative
 parameters (instead of BaP alone) which will be  indicative  of carcinogenic
 potential of environmental POM to man and restrict  the monitoring to these
 parameters.  This will be one of the most important steps  towards  cost reduc-
 tion in POM analysis.
                                    4-58

-------
  4.9   SUMMARY  AND  CONCLUSIONS
       Polycyclic organic  matter  (POM)  in  environmental  samples  consist  of  a
  vast  array of compounds.  Some  of these  compounds show carcinogenic activity
  necessitating the need for their monitoring.  The monitoring is usually per-
  formed with the objective of determining the particle  size distribution and
  evaluating the nature and concentrations of a few individual components in the
  sample.  This section briefly describes the various methods available for
  sampling POM  from various sources and the analytical methods presently
 available for quantification of data from the collected samples.   Particular
 emphasis has been  placed on the difficulties commonly encountered in the
 various monitoring phases.
      Collection  of various  source  samples is usually performed  by  filtration
 through a suitable medium.  With high  temperature  effluent samples,  the air  is
 passed through a cooling  train prior to passage  through the  filtering medium.
 High-volume  samplers  are  used commonly to collect  total ambient particulate
 matter with glass-fiber filter (99 percent efficient for 0.3-um particles).
 Some  investigators have used additional back-up  filters consisting of adsorp-
 tive media to  collect POM present in the  vapor phase  and to prevent evapor-
 ative  losses of POM.  The high-volume  air samplers do not provide information
 relative to aerosol particle size-weight distribution which is essential
 for the prediction of areas of deposition in man as a result of inhalation
of aerosol.  Cascade or cyclone impactors are often used for this purpose.
Limited information is available on sample recovery from sample collection
procedures.  More research is needed to establish the extent of evaporative
and oxidative losses which may occur during sample  collection.
                                   4-59

-------
     The recovery of POM from the widely used Soxhlet extraction for desorp-
tion of POM from the filtering/adsorbing medium is poor for some compounds
and is dependent on the solvent(s) used.  A method utilizing dissolution by
Bonification yields better results.  Techniques which use sublimation and a
thermal method for desorption of POM from collection medium are promising,
although the latter method is only applicable when-coupled with gas chroma-
tographic analysis.
     The problem of any analytical method for the quantification of POM,
which  usually occur as a small quantity in a large matrix of impurities,  is
three-fold.  First, the compounds  of interest must be adequately separated
from impurities to the extent necessary for their  interference-free detection.
Second, the  POM (many of which are isomeric compounds) must be  adequately
resolved  from each other.  Third,  a high  sensitivity of  detection  is  required
to quantitate small amounts  of POM usually collected from  environmental
samples.
      Separation procedures  involving  solvent partitioning  is  generally  used
for the isolation of  POM from paraffinic  hydrocarbons.   Column  chromatog-
 raphy with alumina,  silica  gel,  Florisil, or,  occasionally,  cellulose,  and
 cellulose acetate,  has  been used not  only for the enrichment of POM but also
 for the resolution of the individual  POM.  However,  separation  is  rarely
 complete and quantitative.   Gel  filtration produces  better resolution but the
 procedure is very time consuming.
      TLC is widely used for resolution of POM, particularly the isomeric
 compounds.  The method is quick and inexpensive but is hardly suitable for
 separations involving a large number of compounds.  Moreover, this method
                                    4-60

-------
 results in some losses of POM.  Gas chromatography offers a good method for
 the separation of POM.  The separation is more quantitative than either column
 or thin-layer chromatography.   With the exception of liquid nematic crystal
 phases, most liquid phases used in GLC do not separate isomeric compounds.
 Even with nematic crystals,  column bleed at higher temperatures has limited
 this packing material  to lower temperature use.   Better resolution is achieved
 with wall-coated open  tubular  capillary columns.
      HPLC  offers several  advantages including higher speed,  higher resolution,
 and lower  operating temperature than gas  chromatography.   The  method is  non
 destructive  and the injected sample can be recovered easily.   This  technique
 is  gaining wide acceptance as  a modern  method for the separation  of POM,
 although the  separation  of isomeric POM from  real-life  samples  may  not be
 complete.
     With  regard to  detection,  flame  ionization and  electron capture  detectors
 are  used when the mode of  analysis  is gas  chromatographic.  Fluorescence
 spectroscopy has, however, become well  established as a sensitive and selec-
 tive analytical technique  for POM when  other  chromatographic procedures are
 used.  Several groups have found it to  be  at  least ten times more sensitive
 than the U.V. method.  It  is also more  sensitive and  less expensive than mass
 spectral detectors.   Mass spectrometers normally have a nanogram detection
 limit, although integrated ion-current techniques reduce this limit to sub-
pi cogram range.
     No single analytical separation procedure to date is capable of providing
complete separation and resolution of the POM fractions.   Therefore, POM is
actually analyzed with  a combination of separation and detection methods.   The
                                   4-61

-------
selection of analytical methods is dependent on the number of parameters to be
monitored and the required accuracy and cost of analysis.   In the past, the
most widely used combined method consisted of column chromatography followed
by TLC and U.V./fluorescence detection.  More recent methods utilize column
chromatography in combination with HPLC and U.V./fluorescence detection.  When
the number of parameters to be monitored is very large, column chromatography
coupled with gas chromatography-mass spectrometry  is used as the method of
analysis.  In order to provide valid quantitative  results, appropriate corrections]
for analytical recovery of the POM should be provided through the addition of
internal  standards.
     The  widely  accepted practice of monitoring benzo[a]pyrene alone,  as an
indicator for  other  POM, is  questionable.   More research  effort  should be
directed  towards defining  a  few  representative  parameters which  would  be
indicative of  carcinogenic potential of POM present in  the  environment.
                                     4-62

-------
 4. 10  REFERENCES
  5.
 8.
 9.
10.
11.
                                                  of Heavy
                        effect of fuel and vehicle variables on polynuclear
       t  hpAA-05 |nd.Pheno1 emission.  Paper No. 720210/ Presented
      *972  pp  20       ^ Engineenn9 Con9ress and Exposition, Detroit, Michigan,
  3'   fm?!can' °; R'' a?d J'-M-  Colucc1-   Polynuclear aromatic hydrocarbon
      emissions from automotive engines.   SAE Trans.  79:1682-1698, 1970

  4.   Griffing  M  E.,  A.  R.  Maler,  J.  E.  Borland,  and R.  R.  Decker   Applvi
      a new method for  measuring benzo[a]pyrene in  vehicle exhaust to the s
                 Fm,-^-  Presented at the  Symposium  on Current Approaches to
      1971.   p.  -|3miSS10ns Contro1-   American Chem.  Soc.,  Los Angeles, Calif.
       ?i  J-'  v-'.Maftrand^ea,  G.  Morozzi,  and S.  Toccaceli.   Carcinogenic
        '                                     11 car operati'ns on      "
      Rinehart, W.  E. , S. A. Gendermalik, and  L.  F. Gilbert.   Fuel  factors  in
      ?oU±rence ^U^ ™5sjons'  Rented at American  Industrial  SJgiene
      LonTerencs, Detroit, Michigan,  Paper No. 127, 1970. pp.  15.
          C°nC'
                         Eftect of aasoline aronatic content on polynuclear

                                                             6°         "
                                                                     air.
Natl. Cancer Instit.  Monograph 9:17-57,  1962.

Grimmer  V. G. , A. Hildebrandt, and H. Bohnke.  Sampling and analysis of
polycyclic aromatic hydrocarbons in automobile exhaSst gas   SdoJ? Und
531-536  ml' PetrOChemie Verelnlgt mit Brennstoff-Chemie, 25:442-447,
tobacco'smoke"
tooacco smoke.
and tobacco smoke.
                                h- -                         in tobacco and
                               The origin of 3:4-benzopyrene found in tobacco
                         Analyst, 85:723-727, 1960.                    tooacco
Carugno  N. , and S. Rossi.  Evaluation of polynuclear hydrocarbons in
cigarette smoke by glass capillary columns.  J. Gas Chromatogr ?? 5:103-106,
                                   4-63

-------
12.   Hoffman, D., and E. L. Wynder.  Short-term determination of carcinogenic
     aromatic hydrocarbons.  Anal. Chem., 32:295-296, 1960.

13   EPA   Method-5.  Determination of particulate emissions from stationary
     sources.  Federal Register, Vol. 36, No. 247, Dec. 23, 1971.

14.   Jones, P. W., R. D. Giammar,  P. E. Strup, and T. B. Stanford.  Efficient
     collection o1 polycyclic organic compounds.  Environ. Sci. Techno!.,
     10(8):806-810, 1976.

15.   Harris, D. B., W. B.  Kuykendal, and  L. D. Johnson.  Development  of  a
     source assessment sampling  system.   Presented at the  4th National Con-
     ference on Energy and the Environment, Cincinnati, Ohio, Oct.  5,  1976.

16.   Smith,  F., E.  D. Estes, and D.  E.  Wagoner.   Field  evaluation of  the SASS
     train and  level-1 procedures.   Research  Triangle  Institute, N.C., To  be
     published.

17   Jones,  P.  W.,  J. E. Wilkinson,  and P.  E.  Strup.   Measurement of  poly-
     cyclic  organic materials and other hazardous organic  compounds in stack
     gases.   EPA  Report  No.  EPA-600/2-77-202, October,  1977.   NTIS  No. PB274  013.

18   Parsons,  J.  S.,  and S.  Mitzner.   Gas chromatographic  method for  concentra-
     tion and analysis  of  traces of industrial  organic pollutants  in  environ-
     mental  air and stacks.   Environ.  Sci.  Techno!., 9(12):1053-1058, 1975.

19.  Adams,  J., K.  Menzies,  and P. Levins.   Selection and Evaluation of  Sorbent
     Resins  for the Collection  of Organic Compounds.  NTIS Report No. PB-268 559,
     April  1977.

20   Waibel   M.  Air and water  contamination by road abrasion.   Zbl.  Bakt. Hyg.,
      1.  Abt.  Orig. B163, 458-469, 1976.

21    Ciaccio, L.  L., R. L. Rubino, and J. Flores.  Composition of organic
      constituents in breathable airborne particulate matter near a highway.
      Environ. Sci. Techno!., 8(10):935-942,  1974.

 22.   Guidelines:  Air Quality Surviel lance Networks.  U.S. EPA, Office  of Air
      Programs, Research Triangle  Park, N.C., Publication  No. AP-98, May 1971.

 23.   Sampling Location Guidelines, U.S.  EPA, Div. of Atmospheric Surveillance,
      Nov. 1972.

 24.   DeWiest, F., and D.  Rondia.  On the validity of determinations  of  benzo[a]-
      pyrene in airborne particles in the summer  months.   Atmos. Environ., J_iKp;-
      487-489,  1976.

 25   Corn, M., T.  L. Montgomery,  and N.  A. Esmen.   Suspended particulate  matter:
      Seasonal  variation in  specific surface  areas and  densities.   Environ. Sci.
      Techno!., 5:155-158, 1971.
                                     4-64

-------
 26.







 27.









 28.









 29.





 30.







 31.









 32.







33.







34.





35.

                                                                          J. Air
      tography coupled with fluorescence techniques.   Anal   Chel !  1(7): §92-998,

                                                          t
     pollutants.  J. Air  Pollut.  Control Assoc. ,  23(8): 704-709?  1973




     vo"atiLWin ^>C;n??a;?' 3n? 5'. Zlatk1s-  The determination of organic
     vo I at lies in air pollution studies:  Characterization of profiles
     J. Chromatogr. Sci., 12(4): 175-182, 1974.                pronies.



     Whitby  K. T.., R.  E. Charlson, W. E. Wilson, and R.  K  Stevens   The

     size of suspended particulate matter in air.'  Science', 183^098-1100, 1974.


     Dzubay, T. G   and R. K. Stevens.  Ambient air analysis with

                       flu°rescence ^ectrometer.   Environ
36'   aJnma+^'skaHd *'  £'  P1?rc6-   Quantitative distribution of polynuclear

     ?n?Mrl^ft±S:K^,1? r?lat!°? to Part1cle size of urban    "9-
                                  4-65

-------
38.  Murray, J. J., R. F. Pottie, and C. Pupp.  Vapor pressures and enthalpies
     of sublimation of five polycyclic aromatic hydrocarbons.  Can. J. Chem.,
     52:557-563, 1974.

39.  Rondia, D.  Volatility of polycyclic hydrocarbon.  Int. J. Air Wat. Pollut.,
     9:113-121, 1965.

40.  Pupp, C., R. C.  Lao, J. J. Murray, and R. F. Pottie.   Equilibrium vapor
     concentrations of some polycyclic aromatic hydrocarbons, AS.Og and SO,,
     and the collection  efficiencies of these air pollutants.  Atmbs. Environ.,
     8:915-925, 1974.

41.  Matsushita, H.,  K.  Arashidani, and T. Handa.   Investigation  on a long
     terra collection  of  polynuclear aromatic  hydrocarbons  in environmental
     air.  Bunseki  Kagaku, 25(6):415-417, 1976.

42  Hermann,  T.  S.   Development of Sampling  Procedures  for Polycyclic  Organic
     Matters and  Polychlorinated Biphenyls.   NTIS No.  PB 243 362, August  1974.

43.  Lane,  D.  A.,  and M.  Katz.   Fate of Pollutants  in the  Air  and Water Environ-
     ments.  Part 2,  I.  H. Suffet  (ed.), Wiley-Interscience, New  York,  1977,
     pp.  137-154.

44  Commins,  B.  T.   Polycyclic  hydrocarbons  in  rural and  urban air.   Int.  J.
     Air Poll., 1:14-17, 1958.

45.  Stanley,  T.  W.,  D.  F.  Bender,  and W. C.  Elbert.   Quantitative aspects  of_
     thin-layer chromatography in  air  pollution  measurements.   In:  Quantitative
     Thin-Layer Chromatography,  New York, Wiley,  1973, pp. 305-322.

46.  Cleary, G.  J.   Discrete separation of  polycyclic hydrocarbons in airborne
     particulates using very long  alumina columns.   J. Chromatogr.,  9:204-215,
     1962.

47.  Liberti,  A., G.  Morozzi,  and L.  Zoccolillo.   Comparative determination of
     polynuclear hydrocarbons  in atmospheric dust by gas liquid chromatography
     and spectrophotometry.   Ann.  Chim., 65:573-580, 1975.

48.   Kertesz-Saringer,  M., and Z.  Morlin.   On the occurance of polycyclic
      aromatic hydrocarbons in the urban area of Budapest.   Atmos. Environ.,
      9:831-834, 1975.

49   Cautreels, W., and K. Van Cauwenberghe.   Extraction of organic compounds
      from airborne particulate matter:  Water, Air, and Soil Pollution, 6:103-110,
      1976.

 50.   Bumgarner, J.   E.P.A., E.M.S.L., A.C.B., S.F.M.C.S.,  Research Triangle
      Park,  N.C.  Personal Communication.
                                    4-66

-------
 51.




 52.



 53.



 54.



 55.



 56.



 57.



 58.



 59.




60.



61.


62.
                                       of atmospheHc


      ^ji«t,^^^,K-.^15^/:^si
 carbons in complex mixture.  J. Chromatogr. Sci, 12^606-612? 1974
hydrocarbons in airborne particulates.  Anal.  Chen,.,  «5l 566-^7? *976

Chakraborty  B. B. , and R.  Long.  Gas chromatographic analysis of
                         1n soot samPles-  En?1ron  Icl
Cautreels, W. ,  and K. Van Cauwenberghe.  Fast quantitative
organic compounds in airborne particulate matter Sy Ss chr
with selective  mass spectrometric detection.  J. Chromatogr
                                                      -  53-264 ,
Potthast K. ,  and G.  Eigner,.  A new method for the rapid isolation
                 hydr°C-bons f™ «W -at proSlStJ?91  "
                  °f aerotox1cants' CRC Critical Reviews in Analytical
                        4-67

-------
63.
64.
65.
66.



67.



68.



69.



70.


71.



72.



73.



74.
Sawicki, E., T. Belsky, R. A. Friedel, D. L. Hyde, J. L. Monkman,
R. A. Rasmussen, L. A. Ripperton, and L. D.  White.  Total participate
aromatic hydrocarbons in air:  Ultrasonic extraction method.  Hlth. Lab.
Sci., 12:407-414, 1975.

Kukreja, V. P., and J. L. Bove.  An enrichment method for polycyclic aro-
matic hydrocarbons collected on glass fiber filters, using hydrofluoric
acid.  J. Environ. Sci. Hlth., A VK8&9):517"524> 1976-

Burchfield, H. P., E. E. Green, R. J. Wheeler, and S. M. Billedeau.  Recent
advances in the gas and liquid chromatography of  fluorescent compounds.
I.  A direct gas-phase isolation and injection system for the analysis  of
polynuclear arenes in air particulates  by gas-liquid chromatography.  J.
Chromatogr., 99:697-708,  1974.

Matsushita, H., K. Arashidani, and H. Hayashi.  Vacuum  sublimation  method
for  extraction of  polynuclear aromatic  hydrocarbons  from airborne particu-
lates.  Bunseki Kagaku, 25(6):412-415,  1976.

Shirayama, H.  Analysis of polycyclic aromatic hydrocarbons  in  atmospheric
pollutants by  a high-speed liquid chromatograph-fluorescence spectrophoto-
meter coupling device.  Nippon Kagaku Kaishi, 1:118-123, 1977.

Schultz, M. J., R. M.  Orheim, and H.  H.  Bovee.  Simplified  method  for  the
determination  of  benzo[a]pyrene  in ambient  air.   Amer.  Ind.  Hyg. Assoc.  J.,
34(9):404-408,  1973.

Monkman, J.  L,  L. Dubois, and C. J.  Baker.   Rapid  measurement  of  polycyclic
hydrocarbons  in  air  by microsublimation.  Pure and  Applied  Chem.,  24:
731-738, 1970.

Hoffmann,  D.,  and E.  L.  Wynder.   On  the isolation and  identification of
polycyclic aromatic  hydrocarbons.  Cancer,  13(5):1062-1073,  1960.

Ledford, C.  J.,  6.  P.  Morie, and C.  A.  Glover.   Separation  of polynuclear
aromatic hydrocarbons in cigarette smoke by high  resolution liquid chroma-
tography.   Tobacco Science,  U:158-160, 1970.

Novotny, M.,  M.  L. Lee, and  K.  D.  Bartle.   The methods for  fractionation
analytical separation and identification of polynuclear aromatic hydro-
carbons in complex mixtures.  J.  Chromatogr.  Sci.,  12:606-612,  1974.
 Harke, H.  P., D.  SchUller, H.  J.  Klimisch, and K.  Meisner.
 of polycyclic aromatic hydrocarbons in cigarette smoke.  Z.
 Unters.-Forsch.,  162:291-297,  1976.
Investigations
Lebensm.
 Mold, J. D., T. B. Walker, and L. G. Veasey.  Selective separation of poly-
 cyclic aromatic compounds by counter current distribution with a solvent
 system containing tetramethyluric acid.  Anal. Chem., 35:2071-2074, 1963.
                                    4-68

-------
 75.
 76.
 77.




 78.



 79.



 80.



 8.1.


 82.


83.



84.
 Haenni, E. 0., J. W. Howard, and F. L. Joe, Jr.  Dimethyl sulfoxide:
 Superior analytical extraction solvent for polynuclear hydrocarbons and
 for some highly chlorinated hydrocarbons.  J. Assoc. Office Agr. Chem.,
 45:67-70, 1962.

 Candeli, A., G. Morozzi, A.  Paolacci, and L.  Zoccolillo.   Analysis using
 thin-layer and gas liquid chromatography of polycyclic aromatic hydrocarbons
 in the exhaust product from a European car running on fuels containing a
 range of concentrations of these hydrocarbons.   Atmos.  Environ., 9:843-849,
 I -/ / O *                                                             "~~

 Hoffmann, D.,  and E.  L.  Wynder.   Respiratory  Carcinogens:  Their Nature and
 Precursors.   In:   Identification and Measurement of Environmental  Pollutants
 Westley, B.  (ed.),  Symposium,  Ottawa,  Ontario,  Canada,  June 14-17  1971
 pp.  9-16.                                                               '

 Acheson, M.  A., R.  M.  Harrison,  R.  Perry, and R.  A.  Wellings.   Factors
 aTtecting the  extraction and analysis  of polynuclear aromatic  hydrocarbons
 in water.  Water  Research,  10:207-212,  1976.
      Griest,  W.  H.,  H.  Kubota,  and M.  R.  Guerin.
      aromatic hydrocarbons  by packed column GLC.
      1975.
                                              Resolution of polynuclear
                                              Anal.  Lett.,  8(12):949-957,
      Basu,  D.  K. ,  and  J.  Saxena.   Monitoring of polynuclear aromatic hydro-
      carbons  in  water.   II.   Extraction and recovery of six representative
      compounds with  polyurethane  foams.   Environ.  Sci.  Techno!,  (in press).

      Dubois,  L., A Corkery,  and J.  L.  Monkman.   The chromatography of polv-
      cyclic hydrocarbons.   Int. J.  Air Pollut.,  2:236-252,  1960.

      Hluchan  E., an.d M. Jenik.  Determination  of airborne polycyclic hydro-
      carbons  by paper  chromatography.   J.  Chromatogr.,  9J:531-538,  1974.
Sawicki, E. , M. Guyer, and C. R. Engel.

                     f P°1ynuclear
                                              Paper and thin-layer  electro-
                                                       c compounds.  J.  Chromatogr.,
     Sawicki, E   W. Elbert, T. W. Stanley, T. R. Hauser, and  F. T. Fox.  Separ
     ation and characterization of polynuclear aromatic hydrocarbons  in  urban
     airborne particulates.  Anal. Chem., 32:810-815, 1960
85.  Sawicki, E   J. L. Meeker, and M.. J. Morgan.  The quantitative composition
     of air pollution source effluents in terms of aza-heterocyclic compounds
          °UClear aromat1c hydrocarbons.  Int. J. Air and Water Pollut.,
                1965.
                                   4-69

-------
ftfi
88



89.


90
 92.


 93



 94



 95




 96.



  97
Sawicki  E   P  R  Atkins, T. Belsky, F. A. Friedel, D.  L.  Hyde, J.  L.  Monkman,
R  A  Rasmussen  L  A  Ripperton, J  E. Sigsby, and L. D.  White.  Tentative
SetL of analysis for polynuclear aromatic hydrocarbons in automobile
exhaust.  Hlth. Lab. Sci., TV-228-239' 1974'

Brass  H  J   W  C  Elbert, M. A. Feige, E. M. Gliek, and A. W. Lington.
Unitld Siates Lei, Lorain, Ohio, Black River Survey:  Analysis for hexane
organic extractables and polynuclear aromatic hydrocarbons.  Office of
Enforcement and General Counsel, U.S.  EPA,  Cincinnati, Ohio, Oct. 1974.

Wilk  M   J  Rochlitz, and H. Bende.   Column chromatography of  polycyclic
aromatic 'hydrocarbons on  lipophilic  Sephadix  LH-20.   J. Chromatogr. , 24:
414-416,  1966.

Gladen,  R.  The determination of carcinogenic polycyclic aromatic hydro-
carbons  in automobile exhaust.   Chromatographia,  5:236-241, 1972.

Tomingas, R. ,  G.  Voltmer,  and R. Bednarik.   Direct fluorometric analysis
of Satic polycyclic  hydrocarbons  on thin-layer chromatograms.  The  Sci.
of the Total  Environ.,  7:261-267,  1977.

 Panalaks  T    Determination and identification of polycyclic  aromatic
 Spoons  insmoked and charcoal -broiled food products  by   igh pressure
 liquid chromatography.   J.  Environ.  Sci.  Hlth., B_TH4).299 315, is/b.

 Hoffmann, D. , G.  Rathkamp, and S.  Nesnow   Quantitative determination of
 9-methyl carbazoles in cigarette smoke.  Anal.  Chem. , 4J.1256  l^by, iyby.

 Harrison  E  K , and C. I. B. Powell.  The determination of polynuclear
 Somali "'hydrocarbons by gas-liquid chromatography.  Ann.  Occup. Hyg. ,
 18:199-206, 1975.

 Klimisch, H.  J. , and E. Kirchheim.  A rapid method for the determination
 of benzo[a]Pyrene, benzo [a] anthracene, and chrysene  in cigarette smoke.
 Chromatographia, 9:119-122, 1976.

 Woidich, H., W.  Pfannhauser, G. Blaicher,  and  K.  Tiefenbacher.  Thin-layer
 chromatographic  separation  and  in situ detection of  polycyclic aromatic
 hydrocarbon:  Studies  using acetylitid cellulose.   Chromatographia, 10:
 140-146, 1977.
  Hunter,  L.   Quantisation of environmental  hydrocarbons  by thin-
  chromatography:   Gravimetry/densitometry comparison.   Environ.  Sci.  Technol
  9:241-246,  1975.

  Dona  M   D. C.  Locke, and E.  Ferrand.   High pressure liquid chromatographic
  Sod for routine analysis of major parent polycyclic aromatic hydrocarbons
  in suspended parti cul ate matter.  Anal. Chem. ,  48:368-372, 1976.
                                4-70

-------
      Sawicki, E. , T. W. Stanley, S. McPherson, and M. Morgan.  Use of gas-liquid
      and thin-layer chromatography in characterizing air pollutants by fluoro-
      metry.   Talanta, 1_3: 619-629, 1966.

  99.  Sawicki  E. , R. C. Corey, A. E.  Dooley, J. B. Gisclard, J. L. Monkman,
      K.  h.  Neligan, and L.  A.  Ripperton.  Tentative method of microanalysis for
      benzo[a]pyrene in airborne parti culates and source effluents.  Hlth  Lab
      Sci. ,  7:56-59, 1970.

 100.  Sawicki  E.,  R. C.  Corey, A. E.  Dooley, J. B. Gisclard, J. L. Monkman,
      R.  E.  Neligan, and L.  A.  Ripperton.  Tentative method of spectrophoto-
      metric  analysis for benzo[a]pyrene in atmospheric particulate matter.
      Hlth.  Lab.  Sci.,  7(suppl. ): 68-71 ,  1970.
Bories, G.  Use of polyamide for the separation of different polycyclic
         hydrocarbons by thin-layer chromatography.  J. Chromatogr. , 130:
          1977.                                                        -
 101.
 102.  Brocco,  D. ,  A.  Cimmino,  and M.  Possanzini.   Determination of aza-hetero-
      cyclic compounds  in  atmospheric dust by a combination of thin-layer and
      gas  chromatography.   J.  Chromatogr.,  84:371-377,  1973.

 103.  Zoccolillo,  L., and  A. Liberti.   Determination  of polycyclic hydrocarbons
      by channel thin-layer chromatography.   J.  Chromatogr., J20:485-488,  1976.

 104.  Chatot,  G.   R.  Dangy-Caye,  and  R.  Pontages.   Determination  of the  correction
      factor in the estimation of polynuclear aromatic  hydrocarbons separated by
      thin-layer chromatography.   Chromatographia,  5:460-463,  1972.

 105.  Pierce,  R. C. ,  and M. Katz.  Determination of atmospheric isomeric poly-
      cyclic arenes by thin-layer chromatography and  fluorescence spectrophoto-
      metry.   Anal. Chem. ,  47: 1743-1748, 1975.

 106.  Lao, R.  C. ,  R.  S.  Thomas, and J.  L. Monkman.  Computerized  gas chromato-
      graphic-mass spectrometric  analysis of  polycyclic  aromatic  hydrocarbons
      in environmental samples.   J. Chromatogr., JJ2:681-700,  1975.

 107.  Lane, D.  A., H.  K. Moe, and M.  Katz.   Analysis  of polynuclear aromatic
     hydrocarbons, some heterocyclics, and aliphatics with a  single gas chroma-
     tograph column.   Anal. Chem., 45:1776-1778, 1973.

 108. DeMaio, L. ,  and M. Corn.   Chromatographic analysis of polynuclear  aromatic
     hydrocarbons with packed columns:  Application to air pollution studies
     Anal. Chem., 38:131-133,  1966.

109. Shushunova,  A. F. , P. E.  Shkodich, N. G. Shkanakin, and  L. Z. H.  Lembik
     bas Chromatographic determination of polycyclic aromatic hydrocarbons in
     sewage.   Gig. Sanit. , 8:61-62,  1975.
                                   4-71

-------
110.  Chatot, G.,  R.  Dangy-Caye, and R.  Fontanges.   Atmospheric polycyclic
     hydrocarbon.  II.  Application of chromatographic coupling in gas phase
     thin-layer chromatography to the determination of polynuclear arenes.
     J. Chromatogr., 72:202-206, 1972.

111.  Schulte, K.  A., D. J. Larsen, R. W. Hornung, and J. V. Crable   Analytical
     methods used in a study of coke oven effluent.  J. Amer. Ind. Hyg. Assoc.,
     36:131-139, 1975.
112.  Greinke, R. A., and  I. C. Lewis.  Development of a gas chromatographic-
     ultraviolet absorption spectrometric method for monitoring petroleum pitch
     volatiles in the environment.  Anal. Chem., 47:2151-2155, 1975.

113.  Lao, R. C., H. Oja,  R. S. Thomas, and J.  L. Monkman.  Assessment  of environ-
     mental problems  using the combination of  gas  chromatography  and quadrupole
     mass spectrometry.   The Sci.  of the Total  Environ., 2:223-233, 1973.
 114. Strup,  P.  E.,  R.  D. Giammar, T.  B.  Stanford,  and  P. W. Jones    In:   Carcino
     qenesis, Vol.  1.   Polynuclear  aromatic  hydrocarbons:   chemistry,  metabolism,
     and  carcinogensis,  R.  I.  Freudenthal and P.  W. Jones  (eds.),  New York,
     Raven  Press,  1976,  pp.  241-251.
 115. Mitera, J.,  J.  Novak,  and J. Mostecky.   The determination of  organic substances
     in the atmosphere.  Sbornik Vysoke Skoly Chemicko Technologicke V Prage,
     Techno!ogie  Paliv,  36:5-48, 1977.

 116. Janini, G. M., K. Johnston, and W. L.  Zielinski,  Jr   Use of  a nematic
     liquid crystal for gas-liquid  chromatographic separation of polyaromatic
     hydrocarbons.   Anal.  Chem., 47:670-674, 1975.

 117. Janini, G. M., G. M.  Muschik,  J. A. Schroer, and W. L. Zielinski, Jr.
     Gas-liquid chromatographic evaluation and gas-chromatography/mass spectro-
      metric application of new  high temperature liquid crystal stationary
      phases for polycyclic aromatic hydrocarbon separations.   Anal. Chem.,
      48:1879-1883, 1976.
 118  Dong,  M. W., D.  C. Locke,  and D.  Hoffmann.   Characterization of aza-arenes
      in basic  organic portion  of suspended particulate matter.  Environ.  Sci.
      Techno!., 1J.: 612-618,  1977.

 119  Cantuti,  V., G.  P. Cartoni, A.  Liberti,  and  A. G. Torri.   Improved  evalua-
      tion  of polynuclear hydrocarbons  in atmospheric  dust  by  gas chromatography.
      J.  Chromatogr.,  V7:60-65,  1965.

      Hitps  R  A   Presented at the  Division of Petroleum  Chemistry,  Inc.,
      AmeHcan'cLicaf^ciety, Chicago,  Illinois,  August  24-29,  1975, pp. 824-828.

  1?1  Oro  J   and  J.  Han.   Application of  combined  chromatography-mass spectro-
      metfyto  the  analysis of aromatic hydrocarbons formed by pyrolysis  of methane.
      J.  Gas Chromatogr.,  5:480-485,  1967.
                                     4-72

-------
 122.  Hill,  H.  H., Jr., K.  W.  Chan, and F. W. Karasek.  Extraction of organic
      compounds from airborne participate matter for gas chromatographic analysis
      J.  Chromatogr. ,  131:245-252, 1977.

 123.  Bartle,  K.  D. ,  M. L.  Lee,  M. Novotny.   An integrated approach to the analysis
      of  air-pollutant polynuclear aromatic hydrocarbons.   Proceedings of the
      Analytical  Division of the Chemical Society,  J3(10):304-307, 1976.

 124.  Lunde, G. ,  and  A. Bjorseth.   Polycyclic aromatic hydrocarbons in lonq-ranqe
      transported aerosols.   Nature,  258:518-519,  1977.

 125.  Bjorseth, A.  Analysis of  polycyclic aromatic hydrocarbons in environmental
      samples by  glass capillary gas  chromatography.   .En:   Carcinogenesis,
      vol. 3, Polynuclear Aromatic Hydrocarbons,  ed.  by  P.  W.  Jones and
      R.  I.  Freudenthal.  Raven  Press,  New York,  1978, pp.  75-84.

 126.  Jones, P. W.  Battelle Columbus Laboratories,  Ohio.   Personal Communica-
      tion.

 127.  Lao, R. C. ,  T. H.  Oja, and L. Dubois.   Application of a  gas  chromatograph-
      mass spectrometer-data processor  combination  to the  analysis of  the poly-
      cyclic aromatic  hydrocarbon  content of  airborne pollutants.   Anal.  Chem. ,
128. Onuska  F. I., A. W. Wolkoff, M.  E. Comba,  R. H.  Larose, M.  Novotny,  and
     M. L. Lee.  Gas chromatographic analysis of polynuclear aromatic  hydro-
     carbons in shellfish on short, wall-coated  glass  capillary columns.   Anal.
     Letters, 9(5): 451 -460, 1976.

129. Szakasits, J. J. , and R. E. Robinson.  Disk conveyor flame ionization
     detector for liquid chromatography.  Anal.  Chem., 46:1648-1652, 1974.

130. Gold, A.  Carbon black adsorbates:  Separation and identification of  a
     carcinogen and some oxygenated polyaromatics.  Anal. Chem., 47-1469-
     1472, 1975.                                                 —

131. May,  WE   S. N.  Chester, S.  P. Cram, B. H. Gump, H. S. Hertz, D. P. Enagonio,
     and S.  M.  Dyszel.   Chromatographic analysis of hydrocarbons in marine sedi-
     ments and seawater.   J.  Chromatogr. Sci., 13:535-540, 1975.

132. Doran,  T. , and N.  G. McTaggart.   The combined use of high efficiency  liquid
     and capillary gas chromatography for the determination of polycyclic aro-
     matic hydrocarbons in automotive exhaust condensates and other hydrocarbon
     mixtures.   J.  Chromatogr.  Sci. ,. 12:715-721 , 1974.

133. Soedigdo,  S. ,  W.  W.  Angus, and J.  W.  Flesher.   High-pressure liquid chroma-
     tography of polycyclic aromatic hydrocarbons and some of their derivatives
     Anal.  Biochem. ,  67:664-668, 1975.                                          ' .
                                   4-73

-------
134.  Selkirk, J. K. , R. G. Croy, and H. V. Gelboin.  Benzo[a]pyrene metabolites:
     Efficient and rapid separation by high-pressure liquid chromatography.
     Science, 184:169-171, 1974.

135.  Karger, B.  L., M. Martin, J. Loheac, and G. Guiochon.  Separation of poly-
     aromatic hydrocarbons by liquid-solid chromatography using 2,4,7-trinitro-
     fluorenone impregnated Corasil I columns.  Anal. Chem., 45:496-500, 1973.

136.  Lochmuller, C. H., and C. W. Amoss.  3-(2,4,5,7-tetranitrofluorenimino)-
     propyldiethoxysiloxane - a highly selective, bonded 7t-complexing phase
     for high-pressure liquid chromatography.  J. Chromatogr., 108:85-93,  1975.

137.  Schmit, J.  A., R. A. Henry, R. C. Williams, and J. F. Dieckman.  Application
     of high speed reversed-phase liquid  chromatography.  J. Chromatogr. Sci.,
     9:645-651, 1971.
138. Strubert, W.  Separation of fused ring aromatics on silica gel:
     of nanogram or ppb amounts.  Chromatographia, 4:205-206, 1973.
detection
139. Lloyd, J. B. F.  Nitrogen heterocycle  and polynuclear  hydrocarbon  fluorescence
     and adsorption effects  in the presence of silica  gel.  Applications  in  high-
     pressure liquid and micro-column  chromatography.  Analyst, j[00:529-539,  1975.

140. Wheals, B. B., C. G. Vaughan, and M. J. Whitehouse.  Use  of  chemically  modi-
     fied microparticulate silica and  selective  fluorimetric detection  for the
     analysis of polynuclear hydrocarbons by high-pressure  liquid chromatography.
     0. Chromatogr., J06:109-118, 1975.

141. Jentoft, R. E., and T.  H. Gouw.   Separation of  polycyclic aromatic hydro-
     carbons by high resolution  liquid-liquid chromatography.  Anal.  Chem.,
     40:1787-1790,  1968.

142. Klimisch, H. J.  Determination  of polycyclic aromatic  hydrocarbons - separa-
     tion of benzpyrene isomers  by  high-pressure liquid  chromatography  on cellulose
     acetate columns.  Anal. Chem.,  45:1960-1962, 1973.

143. Jentoft, R.  E., and T.  H. Gouw.   Pressure-programmed supercritical fluid
     chromatography of wide  molecular  weight range mixtures.   J.  Chromatogr.
     Sci.,  8:138-142,  1970.

144. Dong,  M.,  D.  C.  Locke,  and  D.  Hoffmann.  Separation of aza-arene by,high-
     pressure  liquid chromatography.   J.  Chromatogr. Sci.,  lj5:32-35, 1977.

145. Wise,  S. A.,  S. N.  Chesler, L.  R. Hilpert,  and W. E. May.  Chemically-bonded
     aminosilane stationary  phase for the separation of polynuclear aromatic
     compounds.   Anal.  Chem., 49:2306-2610, 1977.

 146. Goldstein,  G.   Separation of polycyclic aromatic hydrocarbons by liquid
      chromatography on cross-linked polyvinylpyrrolidone.  J.  Chromatogr.,  1_29:
      61-72, 1976.
                                    4-74

-------
 147. Qazi, A.  H. ,  C.  A.  Nau,  H.  Turner,  and G.  T.  Taylor.   Identification of
     carcinogenic  and noncarcinogenic polycyclic aromatic  hydrocarbons through
     computer  programming.  J.  Amer.  Ind.  Hyg.  Assoc. ,  34:554-558,  1973.

 148. Burchill,  P., A.  A.  Herod,  and  R. G.  James.   A comparison of some chroma-
     tographic methods for  estimation of polycyclic aromatic hydrocarbons in
     pollutants.  _In:   Carcinogenesis, Vol.  3:   Polynuclear Aromatic Hydro-
     carbons,  ed. by  P.  W.  Jones  and R.  I.  Freudenthal.  Raven Press,  New York,
     1978, pp.  35-45.

 149. Skoog, D.  A., and D. M.  West.   Principles  of  Instrumental  Analysis.
     New York,  Holt,  Reinhart,  and Winston.   1971.   pp.  222.

 150. Popl, M. ,  M. Stegskal, and J. Mostecky.  Determination of polycyclic aro-
     matic hydrocarbons  in  white  petroleum  products.  Anal.  Chem. ,  47:1947-1950
     1975.

 151. Perry, R.   Mass  spectrometry in  the detection  and  identification  of  air
     pollutants.  Int.  Sym. Identification  Meas. Environ.  Pollut.,  Ottawa
     Ontario,  Canada,  1971, pp. 130-137.

 152. Sawicki,  E. , and  H. Johnson.  Characterization  of  aromatic compounds  by low-
     temperature fluorescence and phosphorescence:   application to  air pollution
     studies.  Microchem. J. , 8:85-101,  1964.

 153. Wellons, S. L. , R. A.  Paynter, and J.   D. Winefordner.    Room  temperature
     phosphorimetry of biologically important compounds absorbed  on  filter
     paper.  Spectrochima Acta, 30A: 21 33-21 40,  1974.

 154. Vo-Dinh, T. , E.  Lue Yen, and J.  D. Winefordner., Heavy-atom  effect on room
     temperature phosphorimetry.  Anal. Chem. , 48(8): 1186-1188, 1976.

 155. Hellmann,  H.   Fluorescence spectra of biogeneous polycyclic  aromatics,
     Z.  Anal. Chem., 278:263-268, 1976.

 156. Heinich, G. , and H. Guesten.  Fluorescence spectroscopic determination  of
    ,. carcinogenic pojycyclic aromatic hydrocarbons in the atmosphere.  Kernforsch
     Karlsruhe (Berlin), 77-94, 1975.

 157. Warner,  I. M. , J. B. Calliss, E. R.  Davidson, and G. D. Christian.  Multi-
     component analysis in clinical  chemistry by use of rapid scanning fluore-
     scence spectroscopy.  Clin. Chem., 22:1483-1492, 1976.

158. Shpol'skii, E.  V. , A. A.  Il'ina, and L. A.  Klimova.  Fluorescence spectrum
          ,  .   . ,  .   .        ,
of coronene in frozen solutions.
1952 (in Russian).
                                           .   .         .
                                       Dokl.  Akad. Nauk.  S.S.S.R. , 87:935-938
                                                                   —
159.  Kirkbright, G.  F. , and C.  G.  DeLima.   Use of the Shpol'skii effect for the
     determination of trace amounts of polynuclear aromatic hydrocarbons   Proc
     Soc.  Analyt.  Chem., Vh 55-60, 1974.
                                   4-75

-------
160.  D'Silva, A. P., G. 0. Oestreich, and V. A. Fassel
     luminescence of polynuclear aromatic hydrocarbon.
     1976.
 X-ray excited optical
Anal. Chem., 48:915-917,
161. Smith, E. M., and P. L. Levins.  Sensitized fluorescence for the detection
     of polycyclic aromatic hydrocarbons.  Draft report for EPA Contract No.
     68-02-2150, T.D. 10302 by Arthur D. Little, Inc., Cambridge, Mass., 1978.

162. Gammage, R. B., T. Vo-Dinh, A. R. Hawthorne, J. H. Thorngate, and W. W.
     Parkinson.  A new generation of monitors for polynuclear aromatic hydro-
     carbons from synthetic fuel production.  In:  Carcinogenesis, Vol. 3:
     Polynuclear Aromatic Hydrocarbons, ed. by P. W. Jones and R. I. Freudenthal,
     Raven Press, New York, 1978, pp. 155-174.

163. Schuetzle, D.  Analysis of complex mixtures by computer controlled high
     resolution mass spectrometry.  I.  Application to atmospheric aerosol
     composition.  Biomed. Mass. Spectrometry, 2:288-298, 1975.

164. Johnson, B. H., and T. Aczel.  Analysis of complex mixtures of aromatic
     compounds by high-resolution mass spectrometry at low-ionizing voltages.
     Anal. Chem., 29:682-685, 1967.

165. Hites, R. A., and G. R. Dubay.   Charge-exchange-chemical ionization of
     polycyclic aromatic compounds.   In:   Carcinogenesis, Vol 3:  Polycyclic
     Aromatic Hydrocarbons, ed. by  P. W. Jones and  R.  I. Freudenthal,  Raven
     Press, New York, 1978, pp. 85-88.

166. Wehry, E.  L., G. Mamantov, R.  R. Kemmerer,  H.  0.  Brotherton, and  R. C. Stroupe.
     Polynuclear aromatic hydrocarbons:  Chemistry, metabolism,  and Carcinogenesis.
     In:   Carcinogenesis, Vol.  I.   R. I. Freudenthal  and P. W. Jones  (eds.),
     New  York,  Raven Press, 1976,  pp. 299-309.

167. Karmarkar,  K. H., and  G. G.  Guilbault.   Detection and  measurement of  aromatic
     hydrocarbons  in the air by a coated piezoelectric crystal detector.   Environ.
     Letters, JO:237-246,  1975.

168. Commins,  B.  T.   Interim Report on  the study of techniques for  determination
     of polycyclic aromatic hydrocarbons in air.   Natl.  Cancer Instit.  Monogram
     No.  9,  1962,  pp.  225-233.

169. Butler,  J.  D.   Air  pollution, smoking, and lung cancer.   Chem.  Br.,  V[(10):
     358-363,  1975.

170.  Inscoe,  M.  N.   Photochemical changes  in thin-layer  chromatograms of poly-
      cyclic aromatic hydrocarbons.  Anal Chem.,  36:2505-2506,  1964.

171.  Hawthorne, A.  R.,  R.  B.  Gammage, and D.  J.  Simpkin.   Assessment of dosi-
      metry requirements  and techniques for measuring polycyclic  aromatic
      hydrocarbons.   ORNL Report No. Conf.  770301-4, 1977.
                                    4-76

-------
     ODM,'    ?'L    S' Thomas> and J- L. Monkman.  In abstracts of the  first
     OKNL workshop on polycyclic aromatic hydrocarbons:  Characterization and
     measurement with a view toward personnel protection.  Compiled by
     R. B. Gammage.  ORNL No. TM-5598, 1976, pp. 17-30.

173.  Clements, J   and F.  J. Toth.   Old benz[a]pyrene analytical method.  E.P.A.,
     E.M.S.L., A.C.B., S.F.M.C.S.,  Research Triangle Park, N.C.

174.  Swanson,  D.,  C.  Morris, R.  Hodgecoke, J. Bumgarner, and R. Jungers.  New
     benz[a]pyrene analytical method, E.P.A., E.M.S.L., A.C.B., -S F M C S
     Research Triangle Park,,N.C.                                     "     '
                                  4-77

-------

-------
                              5.   AMBIENT LEVELS
 5.1   AMBIENT ATMOSPHERIC LEVELS
      A variety of polycyclic aromatic hydrocarbons (PAH),  aza-arenes,  and
 carbonyl-arenas have been detected in ambient air.   Because of its  carcino-
 genic properties,  benzo[a]pyrene (BaP) has  been  the most extensively monitored
 and  has frequently been  used as an indicator  of  PAH.   However,  the  relative
 amount of individual  POM compounds released from different sources  can  vary
 considerably,  but  most are present in ambient air samples.   For example,  auto-
 mobile emissions contain a relatively low amount of BaP  as  compared  to  other
 PAH.   Also,  there  are several other social  or occupational  sources of POM which
 may  result in  larger exposures  to  humans than those  from ambient air.   For
 example,  Bridbord  et al.1  have  tabulated human exposure to  PAH  in terms of BaP
 from a  number  of occupational and  other sources  (Table 5-1).  This treatment
 suggests  that  smoking one pack  of  cigarettes  a day results  in an appropriate
 exposure  of BaP twenty times greater than that of a person exposed to ambient
 air BaP levels (assumed 2 ng/m3 BaP).   The exact amount of BaP intake is
 difficult to determine because of the different modes of inhalation during
 smoking, working, or exposure to ambient air.2  Bridbord et al.1 also con-
cluded "that to the extent that polynuclear aromatic compounds besides  BaP are
also present, BaP represents a poor surrogate..."
                                     5-1

-------
              Table 5-1.  INFLUENCE OF OCCUPATIONAL AND,OTHER
                   FACTORS UPON BENZO[a]PYRENE EXPOSURE1
Factor
Smoking one pack of cigarettes
each day
Coke oven workers
Top side workers
Side and bench exposure
Coal tar pitch worker
Airplane pilots
Transatlantic flights
Domestic cross country
Pmnlm/oa -fn rpst.aurant
BaP Exposure
(pg/day)
0:4
180
70
750
0.93
1.38
0.8
Cigarette
equivalents
(packs/day)
1
450
175
1875
2.3
3.5
2
Source of
data
NAS report 1972
NIOSH data!*
NIOSH data
1962 report
NIOSH data!*
NIOSH data
1975 data
Person living near expressway
  24-hr/day (adverse
  meteorology)

Commuter on an expressway
  2-hr/day (adverse
  meteorology)

Exposure to ambient BaP levels
  8-hr/day
0.02
0.04
0.02
0.05    1976 projection
        from 1974 data
0.10    1976 projection
        from 1974 data
0.05    assumed exposure
        to 2 ng/m
 Unreferenced NIOSH data—presumably the latest data in 1976.
                                      5-2

-------
      Concentrations of BaP are different in various cities and at different
 times of the year.  In general, the concentrations are usually highest during
 the winter months (Figure 5-1), probably due to heating sources.3'4  However,
 there are some exceptions.  Cleveland, for instance, does not follow the high
 winter-low summer concentration pattern.   It has been suggested that this may
 be due to significant industrial  emissions of BaP that would be uniform
 throughout the year.3  The higher BaP concentrations in the winter in most
 cities are not due to an increase in particulate matter,  but rather can be
 attributed to an increase in the  amount of BaP in the particulates.
      From 1966 to 1970,  the benzene-soluble organics (BSD) of quarterly
 composite samples of suspended particulate  matter collected  on  glass-fiber
 filters  (99  percent  efficient  for particles  of 0.3 urn)  by  the 250  National  Air
 Surveillance Network (NASN)  stations were analyzed for  BaP.   These data,  in
 the  form  of  annual arithmetic  averages  for  the  numerous NASN  urban and  non-
 urban stations,  are  presented  in Tables 5-2 and 5-3  for the years 1966  to
 1970.  The quarterly composite average  for all  the stations during the  same
years is presented 'in Figure 5-1.   The  levels from year to year remain  rela-
tively constant and do not necessarily correlate to city size.5  Particularly
interesting is Los Angeles, where high auto emissions do not  result in high
BaP levels.  During the period from 1971 to 1976 a more limited number of NASN
stations (40) were sampled for BaP.  The data from these stations for the
total period from 1966 to 1976 are presented in Table 5-4.
     Other polycyclic organic compounds are frequently found in higher concen-
trations than BaP.  Sawicki4 found that benzo[ghi]perylene (BghiP)  concentra-
tions during the summers of 1958 to 1960 were higher than  BaP levels  in the
                                     5-3

-------
o  2
o
o
D-  1
ca  «
CD
I   I   I   I   I
                           I   I   I   !   I  I   I   I   I   I   I  I
                                           COMPOSITE AVERAGE
                                            URBAN STATIONS
     COMPOSITE AVERAGE
     NONUR3AN STATIONS
       1234123412341234   1234
           1966         1967        1968        1969      '.1970
                         TIME, year and quarter
      Figure  5-1.  Composite quarterly averages  for BaP at,Q
                  32 urban and  19 nonurban NASN stations.
                             5-4

-------
        Table 5-2.
ANNUAL AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS
         AT NASN URBANISATIONS5
     Station
 Alabama
  Birmingham
  Gadsen
  Huntsville
  Mobile
  Montgomery

 Alaska
  Anchorage

 Arizona
  Phoenix
  Tucson

 Arkansas
  Little Rock
  West  Memphis

 California
  Burbank
  Glendale
  Long  Beach
  Los Angeles
  Oakland
  Ontario
  Pasadena
  Riverside
  Sacramento
  San Bernardino
  San Diego
  San Francisco

Colorado
 Denver

Connecticut
 Hartford
 New Haven
                               1966
                     1967
          1968
                                                             1969
                                                   1970
18.5
3.5
3.1
6.5
2.3

2.4
2.7
4.2
2.9
           1.7
           0.6
           1.2
           1.1
           2.3
           2.3
           3.5

3.1

2.3
1.9
2.4
2.7
4.2
2.9
1.7
1.8
1.8
2.6
2.0
1.3
2.5
1.6

1.3
0.8
 2.5
 0.7
 0.9
 2.2
2.4
2.1
1.9
 2.1
 0.7
 0.9
 2.2
                              2.3
1.4
1.4
 2.2
 0.5
 1.1
 2.4
          2.5
2.0
2.1
                                                   0.4
 0.7
 0.6
£.3


2.1
2.7

1.8



1.7
1.1

1.0
2.1
1.3
1.7





1.6
1.5

1.6
2.1
1.8
1.6
0.9
2.3
1.3
1.4
1.0
1.2
1.8
2.9
1.6
2.3
1.9
1.6
0.6

0.8
1.8
0.9
1.4
1.2
1.9
1.0
1.0
1.2
1.0
0.6
0.7
0.7
0.7
0.8
0.7
0.6
                                                  2.2
1.4
1.2
                                        5-5

-------
Table 5-2 (Cont'd).  ANNUAL AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS
                         AT NASN URBAN-STATIONS'
                                 (ng/nT)
Station
Delaware
Newark
Wilmington
District of Columbia
Florida
Jacksonville
Tampa
Georgia
Atlanta
Hawai i
Honolulu
Idaho
Boise City
Illinois
Chicago
Springfield
Indiana
East Chicago
Hammond
Indianapolis
Muncie
New Albany
South Bend
Terre Haute
Iowa
Davenport
Des Moines
Cedar Rapids
Kansas
Kansas City
Topeka
Wichita
1966

1.0
2.2
2.4



1.4
0.2

3.5

3.3
6.8
3.9
10.4
2.4
5.4
2.2


3.2
2.5

1.2
0.8
1967

1.4
2.7
1.9



3.0
0.5

2.4

3.0
5.7
2.5
5.7
1.6
2.6
3.7

2.7
0.8
0.5
0.5
1968

0.9
1.9
1.9

2.9
1.5
1.8
0.6

2.0

3.1
1.1
1.9
2.1
4.1

3.7


1.1
0.7
1.2
0.7
1.0
1969


1.7
4.3

2.3
1.0
1.9
0.6

6.0

3.9
1.3
6.8
3.3
5.2

4.3
3.7
4.0

1.7
0.9

1.1
0.4
0.7
1970

0.4
1.1


1.4
0.5
0.9
0.2

1.1

2.0
0.9
5.3
1.7
2.3

3.7
2.4
2.8

0.9
0.7
0.3
-2.4
0.3
0.5
5-6

-------
    Tab,e 5-2 (Cont'd).  ANNUAL AVERAGE AMBIENT BE^taJPYRENE CONCENTRATIONS
      	-—-— '"ifc^J.1—111 U I_ll£
AT NASN URBAN-STATIONS^
        (ng/nO
Station
• — 	 	
Kentucky
Ashland
Covington
Lexington
Louisville
Louisiana
New Orleans
Mai ne
Portland

Maryland
Baltimore
— • 	 • 	 	 	
1966
	
10.5
3.1
2.5

2.3




9 «
— 	 — 	
1967 1968
	 — 	 _
9.3
1.9 3.6
1.8 3.0
2.1 2.7

1.8 1.6


2.3

O O rt *+
1
1969
— • • —
10.9
4.1
1.9

1.5





—
1970
	
6.7
4.4
1.6

1.1


1.1


 Massachusetts
  Worchester

 Michigan
  Detroit
  Flint
  Grand Rapids
  Trenton

 Minnesota
  Duluth
  Minneapolis
  Moorhead
  St. Paul

 Missouri
  Kansas City
  St. Louis

Montana
 Helena

Nebraska
 Omaha
                       1.7
                                                              2.8
1.5
                                           2.1
                                           1.6
4.7

2.2
1.6
0.7
1.8



2.7
5.4
1.4
2.8

1.3
2.3 "

2.3
0.8
1.3
5.1
0.8
3.4
1.4
2.7
1.1
0.9
1.8
1.8

0.9
1.9
3.9
1.7
1.7
1.6
2.1
1.4
1.0
1.8
1.6
3.3
0.5
1.6
2.6
1.5
0.9
0.8
1.1
0.6
1.6
1.0
1.1


1.0
                                        5-7

-------
  Table 5-2 (Cont'd).   ANNUAL AVERAGE  AMBIENT  BENZO[a]PYRENE CONCENTRATIONS
                           AT MACKI  IIDDAKI  CTATTnMC3
                           AT NASN URBANISATIONS*
                                   (ng/ind)
    Station
1966
1967
                                                  1968
         1969
                                        1970
Nevada
 Las Vegas
 Reno

New Hampshire
 Concord

New Jersey
 Camden
 Glassboro
 Jersey City
 Marlton
 Newark
 Patterson
 Perth Amboy
 Trenton

New Mexico
 Albuquerque

New York
 New York City

North  Carolina
 Charlotte
 Durham

North  Dakota
 Bismarck

Ohio
 Akron
  Cincinnati
  Cleveland
  Columbus
  Dayton
  Toledo
  Youngstown
 1.3
 0.6
 3.0
 0.7
 4.2
 1.2
 2.1

 2.1
 2.2
  2.0


  4.1


  5.7
  4.1
  3.6
  3.1
  2.9
  2.7
  1.8
  7.3
 1.1
 4.6
 1.5
 0.8
 3.5
 1.6
 3.3
 1.9
 2.1
  1.9


  3.9


  6.3
  3.7
  1.9
  2.9
  1.7
  3.7
  1.9
  8.2
1.4
3.1
1.0
1.6
1.2
2.3
1.3
2.1
2.0
1.2
1.0
 1.8
 5.6
 8.0
                      0.9
 3.0
 1.8
 3.0
 2.2
 2.4
 1.8
 5.6
0.7
2.4
1.1
2.7

1.8
1.2
1.2
1.5
1.1
          3.6
 4.9
 3.4
                      1.0
 2.9
 3.8
 2.7
 1.9
 1.5
 9.9
                                         0.6
1.9
1.2
4.7
1.4
1.5
1.2
1.0
1.1
                                         1.1
          3.0
1.9
3.9
                     0.4
 2.6
 2.8
 1.6
 1.5
 1.4
 7.1
                                         5-8

-------
    Table 5-2 (Cont'd).
                                     (ng/m)
      Station
  Oklahoma
   Oklahoma  City
   Tulsa

  Oregon
   Eugene
   Medford
   Portland

  Pennsylvania
  Allentown
  Altoona
  Bethlehem
  Harrisburg
  Lancaster
  Philadelphia
  Pittsburgh
  Reading
  Scranton
  Warminster
  West  Chester
  Wilkes  Barre
  York

 Rhode  Island
  East  Providence
  Providence

South Carolina
  Columbia
 Greenville

Tennessee
 Chattanooga
 Knoxville
 Memphis
 Nashville
                                1966
  1.5
  0.7
  3.3
 2.3
 3.8
 4.9
 2.3

 0.9
.3.6
5.0
8.4

1.7
5.5
           1967
   0.7
   0.6
  2.4
  4.8
  3.5
  1.6
  2.8
          4.2
22.9
 7.0
 1.6
 7.0
                                                    1968
  0.7
  0.8
  8.2
  4.1
 1.2
 2.0
           6.2
          18.6
7.4
9.8
1.3
6.0
                               1969
  0.7
  0.5
  4.1
  2.6
1.8
29.5
2.9

5.9
7.0
2.9
5.2
2.2
1-1

1.8
1.2
18.0
2.1
1.3
2.9
6.3
2.4
6.1
0.9
1.0
1.6
1.9
1.9
22.3
2.0
1.5
4.0
13.8
1.8
7.7
1.0
1.3
1.5
2.0
 1.2
 2.2
          1.3
          7.0
4.2
4.7
0.7
2.8
                                                                        1970
  0.9
  0.8
                                         2.3
                                2.4
                               19.3
                                2.7
                                1.5

                                2.4
                                5.9
                                1.6
                                2.9
                                         1.3
                                         1.2
 1.2
 2.1
                                        3.4
5.5

1.4
3.6
                                       5-9

-------
Table 5-2 (Cont'd).  ANNUAL AVERAGE AMBIENT BENJO[a]PYRENE CONCENTRATIONS
                         AT NASN URBANoSTATIONSv
                                 (ng/nO
Station
Texas
Dallas
Houston
San Antonio
Utah
Ogden
Salt Lake City .
Vermont
Burlington
Virginia
Danville
Hampton
Lynchburg
Norf ol k
Portsmouth
Richmond
Roanoke
Washi ngton
Seattle
West Virginia
Charleston
Wisconsin
Kenosha
Madison
Milwaukee
Superior
Wyomi ng
Casper
Cheyenne
1966 1967 1968
1.4
0.9
0.6 1.4 0.9
0.5 0.8
1.2 0.7 1.0
0.8 0.7
3.2 2.5
2.2 1.5
9.2 8.7
2.8 3.5 4.9
7.7 10.2
5.2
7.5 7.7
2.7 1.8 2.0
3.4 4.6
1.4
1 3
4.1 4.7
3.3
n q
0.5 0.6
1969
2.0
0.6
0.7
0.7
0.5
1.8
0.9
6.3
3.9
3.4
2.2
5.3
1.6
2.6
1.7
4.0
1.6
0 6
0.5
1970
1.9
1 9
1 . f-
1.0
2.5
1.4
0.7
2.7
1.1
.,4,5
1.8
4.9
2.1
6.2
1.5
2.1
1.3
1.1
2.5
1.5
0.4
0.4
                                     5-10

-------
Table 5-3.
ANNUAL «ERAGENA«BIENT BENZOWPVgENE CONCENTRATIONS
                 (ng/m-5)

Station
Arizona
Grand Canyon
Maricopa County
Arkansas
Montgomery County
California
Humboldt Coutny
Idaho
Butte County
Indiana
Monroe County
Parke County
Mai ne
Acadia National Park
Missouri
Shannon County
Montana
Glacier National Park
Nebraska
Thomas County
Nevada
White Pine County
New Hampshire
Coos County
New ~York
Jefferson County
1966 1967 1968 1969 1970

O-3 0.2 0.2 0.2 0 1
0.2 0.5 0.3 O.Y

°-3 0.1 0.2 0.2 0.1

°-4 °-4 0.3 0.5 0.1

0.2 0.1 0.1

0-5 0.5 0.3 0.2
°-9 0.4 0.3 0.4

°-2 - 0.3 0.1 0.2

0.2 0.2 0.2 0.2

0.3 0.4 0.4

°-2 0.2 0.1 . 0.1

°-] . o.i o.i o.i

°-2 0.2 0.2 0.1 0.1

°-2 0.2 0.3 0.2
                                5-11

-------
Table 5-3 (cont'd).
ANNUAL AVERAGE AMBIENT BENZQ[a]PYRENE CONCENTRATIONS
  AT NASN NONURBAN STATIONS3
           (ng/nT)
Station 1966
North Carolina
Cape Hatteras 0.2
Oklahoma
Cherokee County 0.2
Oregon
Curry County 0.1
Pennsylvania
Clarion County 1-5
Texas
Hatagorda County 0.3
Vermont
Orange County 0.9
Virginia
Shenandoah National Park 0.9
1967 1968 1969 1970
0.2 0.1 0.2
0.2 0.2 0.2 0.2
1.1 0.1 0.1 0.1
2.1 1.0 1.2 1.2
0.1 0.2 0.1 0.3
0.3 0.3 0.2
0.3 0.3 0.3 0.2
5-12

-------
^IT*-:.,  ^W^WCMW^U,^    cocMuxnmcoioininfocMoo
-ooooo,  oooocModd d^l  odd dodo d-d d- csi



                                                     "5 «0



                               5-13

-------
i

1
r
 «
•y 1
o 5
H- ( '
a:
^

0
o
o

LU
LU
C£
Q_
1 — I
mo
u-i •
coro
LU en
CO C
r— Z

LU
M tO
ca z:
s: o
P
LU 
<: z
.
T3
TJj
C
O
CJ
v**'
«4-
m

O)
2
to
H-








D(^>I — Ln "sf ^f
?! 0 CD 1 CD 1 1 1 0 O
n co CM CM ( ' | ( in r- '
n CD CD 1 CD 1 1 1 0 r—
4- CM CM m
•«^ | *| *lll *l
T» I CD 1 CD 1 1 1 O 1
•O «3'CM«=|-'— CO«d-CMlO
jj ! CDCDCDCDCDOCDO
™—

^^^ _
CM cn^covDr— ininr^-vo
Cn cnCDCDCDCDCDr— OCO
r- •

r- m^-cM inr^cnco
en i o o i — 101 — O" —
r~
o mcM oocMin i— m
en mi — i < — 01 — ICMCM
r™~

en CM en co ID ID o
Cn "^"l 1 CO CD r— 1 CM ^
r""


co «3- en co CD ID r-~
tO •!! . • • 1 • •
en r»« i i «=J- o CM i •* «*
i»» en in co co
O .11 ... 1 1 1
en CM i i co o i— i i i
r— CM

f^ - ^A* c^ oo w r^^ ^j ^^
O> OOOICVIOCMICO^
r™*



m
5>

& •*£
~ i.
nj TO -Q
^^ . o irt IB*
J^ ^ rrt •* . •
OXTJ >>• -Pro1 (0 > U
*". 5 «?° . z "5 "S * 5
O)r— CO >• tO "3S. 3: C «
0 Z 0 0 CO
O •> "T3 - "-P CO
roos-r-fOr-ccos
+J in o.1*- c: +J ^ s- S
XT O CO O J=. 
£- O
en co
LL. O
u
• s-
o n3
CO
T3 !-
CO CO
in -c:
CO CO
•I— «I—
in (/>
c. c.
(0 (0
J3 -Q
i- s-
ID ^
n) -Q
5-14

-------
  10 cities examined, but the reverse was true during winters in those communi-
  ties that burn coal.
       An early study of the various PAH compounds in urban air was reported by
  Sawicki and coworkers.7  They analyzed eight compounds in seven cities during
  the summer of 1958 and winter of 1959.   They collected their samples on glass-
  fiber filters,  Soxhlet extracted with  benzene,  separated isomers  with column
  chromatography,  and quantitated  the  compounds  from  ultraviolet absorption
  spectra.   Their  results  are presented  in Table  5-5.  Good relative correlation
  between BaP and  the other  PAH, except coronene, was found; the authors
  concluded  that concentrations of  the other compounds could be  predicted from
  BaP data.  The advantage of this  is that costly and laborious  work could be
  avoided if one can calculate other PAH values from BaP values.   Hauser and
  coworkers, in an unpublished study of PAH in Birmingham, Ala.,5 found a
 similar correlation between many of the compounds they examined (Table 5-6).
 Their results  indicated the effect of localized sources on the  immediate sur-
 roundings.
      This  correlation between  the  concentration  of BaP  and other compounds
 does  not always occur.  For example,  a study  by  Kertesz-Saringer and  Morlin8
 found  little or no  relationship between BaP and  other PAH.  They analyzed
 11 PAH in Budapest  air  by collection on glass-fiber filters, Soxhlet  extrac-
 tion with benzene,  and  thin layer  chromatography separation.  Bands of the
 chromatographic plate were selected so that no more than three  compounds were
 found in each band to be analyzed spectrophotometrically.  The  lowest correla-
tion coefficient was found between BaP and chrysene or phenanthrene.   Some
relationship was found between  pyrene  and BaP, but the authors concluded that
                                    5-15

-------
Table 5-5.  POLYCYCLIC AROMATIC COMPOUNDS IN THE AIR OF SELECTED CITIES'
                               (ng/m3)

City
Winter 1959
Atlanta
Birmingham
Detroit
Los Angeles
Nashville
New Orleans
San Francisco
Summer 1958
Atlanta
Birmingham
Detroit
Los Angeles
Nashville
New Orleans
San Francisco
BghiP

8.
18.
33.
18.
17.
7.
7.

' 5.

9
0
0
0
0
3
5

,1
8.3
9.
.5
2.3
3
4
2
.4
.6
.6
BaP

7.4
25.0
31.0
5.3
25.0
4.1
2.3

1.6
6.4
6.0
0.5
1.4
2.0
0.3
BeP

4.7
10.0
23.0
8.1
14.0
6.4
2.9

1.5
5.9
5.3
0.6
1.2
3.1
0.5
BkF

6.
13.
20.
5.
15.
3.
1.

1.
4.
4.
0.
1,
1
0

0
0
0
7
0
9
7

3
6
,9
,5
• 0
•8
.2
P

6.0
17.0
36.0
6.0
30.0
2.3
1.9

0.7
2.1
2.8
0.3
0.6
0.3
0.1
Cor

4.3
3.5
6.4
12.0
4.6
27.0
4.9

2.5
2.4
1.8
2.2
1.3
2.5
1.6
Per

1.1
5.5
6.0
1.6
4.4
0.8
0.3

0.4
2.1
1.7
0.03
0.2
0.4
<0.1
A

0.5
2.2
2.0
0.2
1.8
0.1
0.1

0.2
0.3
0.4
0.0
0.1
0.1
0.02
Total

38.9
94.2
146.4
56.9
111.8
27.6
21.6

13.3
32.1
32.4
6.4
9.2
14.8
5.4
                                      5-16

-------








LU
1—
Z
CO LO
a ID
z en
a_ TD
o to
3: ID
<: en
a. i—
LL. -
o 
Z= C3 00
CD Z E
O M \
S CO
LU a; c
CD I—I \^
LU O£
>i 1 1
U~J
«=t t—
 <^j- ^j £^j
C\J CM CO CO CSJ CSJ •— ^ r-I oj r-^
CM ^CMCMCO _ '^ -_ ^
— CDCM'r^o' r-" o' o' 0* o'

""I ^^^cn oo o r- CM LO
_!• f~. f^ , ' __ * * • • * (
i-- Jsjjf!^ ^I ^ COCO 00

°^ °? "~. Q. "* "*• <^ CM 00 CM
•- 0 ^- CM o' r-' o' CD CD o'

LO CD 00 LO 0 CM ID ID «tf- ^J-
""" OO CM r— f^» CM


"? <^I~!C?U^ °° °° ° «> en
O IV.' U3 LO LO' CO VD 00 l^s! pvl
r- CMr- ^/VUOOf^CM
""I r~Ic!esf'!;1I co r>- CM ID o
£2 °° ^ ^: •* p « CN! r-; CM'




OOOOCMLO-^l- 00 CM 0 en 0
CM i— ' i — ' OO r—


. r— r— ,— r— ,_
€
if 1 ^ f ^ *
II t 1 * 1 3
« s 5 £ £ ™ 1
"- H~f 2. 1 — >

^»
CM
00
o

CO

CD'

CM

oo
CM'


00
en




ID
f"*"


LO




LO
LO




-------
"the concentration of 3,4-benzpyrene[BaP] is a poor indicator for other com-
pounds."
     Gordon and Bryan9 found similar results while monitoring PAH at four
sites in Los Angeles.  They found that the amount of coronene paralleled the
estimated traffic density at the four sites.  At three of the four sites, the
PAH patterns normalized to coronene were similar and resembled patterns of
auto exhaust.  However, at the remaining site (near petroleum refineries and
chemical plants), the PAH pattern, including BaP, was distinctly different.
     An interesting monitoring trend has been developed from the NASN BaP
values for the past ten years.3'10'11  The declining trend for BaP during 1966
to 1970 is depicted in Figure 5-1 and is based upon BaP data at 32 urban and
19 non-urban stations.  As can be seen,  the average BaP concentration decreased
from 3.2 nanograms/m3 in  1966 to 2.1 nanograms/m   in 1970, approximately a
30 percent decrease.10  A recent analysis11 of BaP trends during the ten year
period  of 1966 to  1975 has been based upon  34 urban (24 with coke ovens) and
3 rural sites.   These trends  for the urban  sites  (Figure 5-2) are consistent
with previous results  indicating a  steady decline of BaP.  Even  concentrations
in  the  three remote  rural areas (Grand  Canyon National Park, Arizona;  Acadia
National  Park, Maine;  and Shennandoah National  Park, Virginia)  (Table  5-4)  in-
dicated a downward trend.  The  following limitations to  the  study were noted:
 (1)  small  number of areas were  covered  and  data was only available  from one
 single  monitoring site (located in  center city  business  area);  and  (2) a
 relatively high  number of urban sampling sites  were  located  in  areas where
 coke ovens  were  present.
                                         5-18

-------
10.0
                               DASHED LINE
PERCENTILES OF QUARTERLY
MEASUREMENTS
                                        - 4 QUARTER MOVING AVERAGE
                                         OF PERCENTILE VALUES
                                             PERCENTILE
                                       QUARTERLY MEASUREMENTS
                                          (UPPER CURVES)
              PERCENTILt
     OF QUARTERLY MEASUREMENTS
          (LOWER CURVES)
                                 YEAR
      Figure 5-2  Benzo[a]pyrene - seasonality and trends
                  (1966  to 1975) in the  50th and 90th,
                  percent!les for 34 NASN  urban sites11
                              5-19

-------
     Although a large number of the NASN sites were in areas that had coke
oven emissions, the monitoring stations were not sited for those emissions.
Comparison of the coke oven sites to the non-coke oven sites showed that the
coke oven sites had generally higher (40 to 70 percent) BaP concentrations  and
the downward trend for both types of sites was parallel.   The authors con-
cluded that the higher values could not necessarily be attributed to coke oven
emissions alone since coke oven sites were usually located in more industrial-
ized northern cities where industrial and heating sources may contribute
significant amounts of BaP.  Also noted was a decline of BaP summer levels for
coke sites that was greater than the non-coke sites.
     Examination of Figure 5-2 also illustrates the high winter-low summer
seasonal concentration of BaP.  "This seasonality is most dominant in the
first five years (1966 to 1970) and is much less pronounced for the most
recent years."
     During the ten year period of 1966 to 1975 the BaP concentrations at
urban sites decreased about 84 percent, dropping from a median of the individ-
                                                   o
ual annual average concentration of 3.2 nanograms/m  in 1966 to 0.5 nano-
grams/m3 in 1975 (Fig. 5-2).  The urban sites used for this trend analysis are
indicated in Table 5-4 along with an indication of areas where coke ovens are
located. Nashville, Tennessee was used by  Faoro and Manning   for the trend
analysis but is  not presented in Table 5-4.  The decline  in BaP is believed to
be due  primarily to decreases in coal  consumed  in  family  dwellings or build-
                                                                 11
                                                                     Also
ings in and around the central  business districts of urban areas.
contributing to the decline is  the improved disposal of solid wastes  and the
restrictions on open burning.
                                      5-20

-------
       Because of the importance of the NASN BaP monitoring data to the assess-
  ment of exposure, evaluation of the analytical procedure used is necessary.
  During the total period of 1966 to 1976 all the samples were collected on
  glass-fiber filter (99 percent efficient at 0.3 urn) with high-volume samplers.
  Twenty-four-hour samples were taken  biweekly and sent to a central  laboratory
  where individual  samples were combined by quarter to save time  and  expense.3
  Up  until  1972,  the sample was extracted  with  benzene in a Soxhlet extractor
  for six hours followed by thin  layer chromatography (TLC).   BaP was  removed
  from the TLC plate, dissolved in sulfuric  acid, and measured by fluorescence
  spectroscopy.   Starting  in approximately 1973, a faster method of analysis was.
  used which may  have had  some  affect on the data generated.   This consisted of
  cyclohexane extraction of the filter (cyclohexane was found to be as efficient
  in extraction as benzene), TLC separation, and fluorescence spectroscopy of
 BaP on the plate.   Anthanthrene interferes with the latter procedure but its
 contribution to  the fluorescence spectra can be subtracted using a character-
 istic emission band for anthanthrene.
      The efficiency of  collection and stability of BaP during the analytical
 procedure  has not been  completely examined.  Many  studies  have indicated  that
 BaP  is primarily associated with particulate matter,12'13'14  with a sizeable
 percentage  (up to 60 percent)  in the less-than-one-Mm range.15'16  HOW much
 BaP  might pass through  the  0.3-um glass-fiber filter  is unknown although the
 efficiency of collection  should increase as the filter is loaded with particles.
Also, DeWiest and Rondia17 have provided experimental evidence that sizeable
differences between the BaP collected by high volume samplers in  summer as
compared to winter temperatures will  occur.
                                     5-21

-------
                                                               18
     Stability during collection has been addressed by Commins.     He  collect-
                                                                 o
ed two identical samples of smoke on glass-fiber filters at 1.2 m  per minute
for two hours.  He then drew filtered air through one of the filters for two
hours at the same flow rate.   None of the eight hydrocarbons (including BaP)
                                                      18
had changed significantly.   A similar study by Sawicki   indicated  no  loss
of BaP between identical samples whose only difference was that one set had
twice as much air drawn through the filter.  Thus BaP adsorbed on particulate
matter that is collected on a glass-fiber filter appears to be stable  to an
identical volume of air.  However, if oxidation of the outer layer  of  BaP
occurs during the first sampling period but not during the second period
                                                                   19
because of protection of the inner BaP by the oxidized outer layer,   then  the
values for the above experiments should be the same.  This possibility would
result in lower reported BaP values than the quantities being inhaled.
     Commins18 also examined the stability of BaP on storage of filter paper
or in cyclohexane.  BaP from a smoke sample stored on a glass-fiber filter  in
a sealed envelope for a year lost 32 percent compared to a sample immediately
analyzed.  However, with BaP kept in cyclohexane in the dark for six months
                              1 o
no loss was detected.  Commins   concluded that the filters should  be  extract-
ed as soon as possible.  Since BaP samples are combined into three  month
samples and at times during the early 1970's backlogs of over a year were
encountered, substantial losses of BaP could occur if the samples were not
extracted soon after collection.
     Declining trends for PAH have also been noted in Great Britian.
      on
Leahey   found that the concentration of BaP and BeP in downtown London was
decreasing while coronene was remaining about the same.  The author attributed
                                     5-22

-------
 the reduction of BaP and BeP to reduction in coal use and suggested that
 coronene had remained constant because of its greater proportion in auto
 exhaust which is increasing.   The samples were collected on glass fiber
 filters, sealed in glass tubes at the end of each month, and analyzed at the
 end of a year.   The filters were extracted with cyclohexane, separated by
 column chromatography (alumina), and analyzed by ultraviolet spectrophoto-
 metry.
      Aza-arene  compounds have  recently received considerable monitoring
 effort.   Brocco et al.    and Dong and coworkers22 have  detected  a number of
 new compounds including  quinolines,  azafluoranthene,  azapyrene,  and isoquino-
 lines.   Their results  are included in the tabulation  in Table 5-8.   The con-
 centrations  of  aza-arenes are  usually considerably lower than PAH,  and  the New
 York City monitoring of  Dong et  al.22 are compatible  with previous  data.
 However, the results of  the Brocco et al.21  analysis  of Rome  air  samples  are
 considerably higher than  any previously reported  values.  Many of these com-
 pounds (e.g., the  quinolines) are very volatile and,  therefore, the  different
 values may be due  to the  different collection and  analytical  procedures which
were used.  Both Qong et  al.22 and Brocco et al.21 collected  their samples on
glass fiber filters and then extracted with benzene21 or benzene/methanol22
                     f\-\
 (4:1).  Brocco et  al.    cleaned up their  sample by TLC and analyzed by gas
chromatography.   Dong et al.22 partitioned the aza-arenes into sulfuric acid,
neutralized and extracted them back into chloroform, and analyzed by HPLC.
Qualitative confirmation was accomplished by GC/MS.
     Very few studies have examined the possibility of long term transport of
POM which might  result in POM contamination of areas that are downwind from
                                     5-23

-------
large emission sources.  Lunde and Bjrfrseth26 monitored air samples in Norway
which had different trajectories.  They identified 20 different PAH (see Table
5-7) and determined that samples with trajectories from western Europe contained
about 20 times more PAH than'samples with trajectories from northern Norway or
stationary air from southern Norway.  When the wind was blowing from the right
direction, the POM concentrations  in Norway were  of the same order of magnitude
as  in  downtown London.  This  study is particularly important because  it provides
a list of at least 20 PAH  compounds that are  stable  enough to  be  transported
from the source  to humans  where they will be  inhaled.   At least four  compounds
 (benzo[c]phenanthrene, benz[a]anthracene, benzo[a]pyrene, and indeno[l,2,3-
 cd]pyrene) which have well-resolved peaks are suspected carcinogens.
      Perhaps the most detailed recent study of PAH compounds in atmospheric
 samples was conducted by Katz et al.37  They analyzed ten individual  PAH at
 five different  sites  in Ontario during three month periods in 1975 and 1976.
 Table  5-8 tabulates  the results from two sites along with total  PAH and the
  ratio  of PAH and the BaP  concentration.  These results demonstrate a varying
  relationship between BaP  and total PAH  compounds.
       Table  5-9  tabulates  the various  compounds  for  which monitoring  concentra-
  tions have  been reported.   As many as  100  compounds have been found  in urban
  atmospheres,3'23 but they are not listed in Table 5-9 because quantitative
  information could not be found.
       Table 5-9 illustrates the vast number of POM that have been detected in
  the atmosphere and there are probably many others that have not been quanti-
  tated for  lack of an appropriate  analytical method.  The accuracy of the
  concentrations reported  in Table  5-9 probably varies considerably, especially
  for  the lower  molecular  weight compounds which  are  likely to  sustain  substan-
  tial  losses by volatilization during collection on  glass-fiber  filter in  high-
   volume samplers.
                                        5-24

-------
        Table 5-7.   CONCENTRATION OF POLYCYCLIC AROMATIC HYDROCARBONS
                        IN NORWAY AEROSOLS,  ng/m3.26
Sample Number
Sampling period


PAH


Phenanthrene
Anthracene f
Methyl -phenanthrene/-
anthracene
Fluoranthene
Di hydrobenzo [a&b] f 1 uor-
enes
Pyrene
Benzo[a]fluorene
Benzo[b]fluorene
1-Methylpyrene
Benzofcjphenanthrene
Benz [a] anthracene
Chrysene/Triphenylene
Benzo [b&k] f 1 uoranthene
Benzo[e]pyrene
Benzo[a]pyrene
Perylene
Indeno[l ,2,3-cd]pyrene
Benzo[ghi]perylene
Anthanthrene
Coronene
1
Feb. 20-21
1976

England,
France


4.725


0.661
6.637

0.874
4.864
0.815
0.571
0.147
1.021
0.585
1.756
4.312
1.191
0.965
0.090
1.306
1.142
0.225
0.212
2
Nov. 25-26
1975
Origin of
Northern Eng.
Scotland


1.216
0.278

0.216
3.965

0.363
3.293
0.318
0.149
0.099
0.957
0.740
3.269
4.013
2.635
2.053
0.191
1.920
1.971
0.423
0.183
3
Jan. 25-27
1976
air
Northern
Norway


0.036
0.038

. _
0.171

0.032
0.135
0.021
0.117
0.009
0.038
0.041
0.099
0.083
0.066
0.059
trace
0.062
0.064
0.007

4
Feb. 1
1976

Stationary
air
Southern
Norway
0 146
\J • 1 T^U

0 052
v • \J\JC~
0 324
\J • *Je-T^
0.032
0 286
\J • ^\j\j
0. 026
n 14ft
\J . I T^O
0.009
0 Iflft
\J . I UO
n n?^
\J • w / O
0 194
\J • I ^~
0 4.R4
\J • "U*T
0.135
0.098
0.011
0.144
n i4.n
w • I "V
0.022
0.020
Total Identified PAH
32.099
28.252
                                                      1.108
                                           2.435
                                        5-25

-------
                                              00
                                                               uo
                                                      <£>

                                    o o
                                    cr> o
                                    o   •
                                    CM r—
                                                       ID
                                                       P-.
                                                       a*
                                                                
en
                                                                    cnocSco co^t-oco «5r-
                                                                       10 en
                                                                                      CM
                                                                                               CM
                                                                                                     co r
                                                                                                     CM CM
                                                                                                     <£> O
                                                        ra      10

                                                        c      o\
                                                        o      *—
                                                        s-       a>
                                                        n-       
     ^- CM co o co co en «"~ i

     r^   CM          C3
                                                                                                     (O IO
                                                                                                     O CT>
                                                                 to!
                                                         «n o    i—

                                                        "aJ ?     a>

                                                         co   •     ^

                                                        "
                                                         ••-» T-   «e
                                                         (O (/>
      coco.—»--t-.oococqcMco
      CO r— I
      co i
           • CO
O 00 "=»•
                                                            


                                                         *
                                                          c  a>
                                                          a>  o
                                                          (/> CO

                                                            ••
                                                         t-  C
                                                          4J  O
                                                          (O (O
                                                          Q. O
                                                                                                      CO r—
                                                                                                         ca
                                                                                                          
-------







1 — 1

O O -P
"- J=. ea
•P/O -P U
03 E Ov-
S- -o
•P 0 in C
c: o in «i—
 en r-



f^SSSf^fCSS^^^^ ^^^LOr-^^
2£22§gSggg££ £S^S£§S
-p
•r—
o
S—
O5 +J ' ,_
c (/> u) oj i a)
r- .,- d) .. ^^ ^7 =" Q.4— .(-> ^— ^ 2s i^*s ^ fj [!^ -^- O -j->
m -p s-" « S S S .5: 1 «= « z « • o> •*- -o § 3 y ro I
SS5w>5in>,'0(D •co.co.^i'Sg.f •§ fe^ ««|
-^ t-OZT CQfvi ^f- ^ 5 ,5 ,5 OO033OO3
"SOCQCQ-J — ll/>

_l
!j3
S-
s g-- a
^ CM 5: 05
^D p. • tjD ^3 f— ^ , "^s, CU
i "^i *— ' • « ^3^ ^J
CV1 fv^ * I ^"~ • to f^ ^~^ CO "^ 03
^^ CT5 1 f IQ 1 f^ , ' 1 ^^ QQ *v I .«_

^ 	 ^ ' t— ' ^J <~~ ^~ r— f^^ I ^| ._ 1 00 i~— • O
OA ci o ci UDO ' "**" * ' — ..o.Lflr~To
CD ^ ° "~ CM r— -^- i— oo' C3 O C5
O)
C
u
03
coi -S '
0) C t-
1 	 QJ *—
03 (J (°
_d ("0 -_J
•PS- /°,
^* f" ^— ^
0. -P £
fO c trr
2 <: m



r- CM CS



LO ^o CM
r**"* {Q r^x




>,
?^ f™*
r~~ f0
•P HH
. .

** QJ
4J £= 0
C O Qi
£ « ,
^ ' «
U3 <1)
1 (U i—
5 If
-P fd to
•r- <^-
S ID r—

1
Ij3
i.
03
Q.
O5
"NSS QJ
CJ5 i ^ .
~^ (O r— .
r— • 
-------





»— 1
Z
UJ

Eg
<£-
1—4

UJ
O
»— 1
CD
O
0
O
O
O
Q_


•u
C
o
o
cn
LO

to

CO
CO



f^^
f^^
cn
n™






10
cu
Providence an






CM
O
O
1
LO







CU
f-~
03
(j
fO
i.
£Z
C
to
u
•t
to
N
C
CU
.0
Q

CM



|"^^
r^^
cn
^—






o
S- TJ
•r- C
10 fO
i "TO
S- Z
o





5
ID
0
1
CM
O
CU
o>
u
10
s-

•p
to
CU
c.
0)
s-
jz:
ro
c
0)
o.
1
4J
CD

«* -ID CO 00 CO
CM CM CO



CM r^- LO LO r^
CD r**« i^ f^ f^*
cn cn cn cn cn




o>
c.
•r™*
TJ ^=
C 0 S
03 -P 0
.c: -r- s- to
D.4- OJ
^ » S- TJ tO
.p S- •!- C
01 tO IO fO 0)
3 4J r- U
TJ 1 O) -P C
C •• C O) CU
•r- I— >>UJ O) TJ
(0 tO O.T-
feSEzl-Ss-
OJ O O 3 S-
2E Z Q: CD 0.





O
CM CO
i—^ CO
1 LO 1
ID CM i—
LO CO 1 ' —
o cn o CD
"A o CM r- co











cu
c
O)
-p
c •
10
CD
Q_

U3 CO
CM CM



r^. LO
r^ r**»
cn cn







o
4-
S- TJ
•r- C
fO fO
1 D)
0 O
^y fy^





CO
CO
0
o
1
CM
CD cn
i— O






cu
c
cu
i-
•p
c:
to
c
cu
D-
1j
o
N
c
eu
CQ

ID
CM



p*1^
f^^
cn







o
S-
q-
S- TJ
tO tO
i "01
o





CM
o
CD
1
LO
CO
CD









CU
cu
s-
0
3
C
^
O
N
0)
GO)

CD CO t«D
CM CM CM



cn cn cn







s • i
2 2
M_ M-
S- TJ S- XJ
•r- C v- C
to to to to
i cn i cn
o o o
z o; z





r^ CM
r— CO
r— O
? ?
LO " — 00
CD r-^ CD

W
CU
c
cu
S-
o
3
P-"
^
cu -9
C ^O
0,) <3
o 'o'
3 N
tfZ cu
I 	 , 0
-° ?
|_.-f t-
o -o
N >i
C ' -C
0) ••-
OQ a

CM CM



LO CO
ID VD
cn cn
^— i —






•P c
.T- to
O -Q
S- t-
•P 3
CU /•">
Q to
0) LO
C i— •
in to  cu
CO «=C





CD
LO
en
CD 
-------





C£
J-H
<3C
j_
1 1 [

or)
*flC

;y
HH
UJ
t—
-
o
— 1
o


•
^*v
^o
— •

§
3
T
t
i


















0)
u
C
CD
Ol
ll
^U
§
•r-
-P
(0 
•r- jC CO
+*O .p O
« S 0 v-

•- "O
-> o in c
c 0 in v-
0) 0 CD
U ,_ ,_ oj
C --v, c in
0 O) 3 .1-
O 2.*«— ' 3£
















0)
j— •
5



CM OOCOCM CM CM CM CM CM CM ^ CM £> °



1^5 CD CD Lrt IO f*^. i r^
en m en ^ ^ ^ ^ ^° ^° r^ S r^ S f^ £? S *° ^


c to
o <->,••-
•i- -O r—
•r: -P o TO
 O E ^C t- ^r c_ ^P
^^ i-c cScln'^eSin tu^s m "S,^

(O "^ n» n\ ^ **~ r~" ' f Q^ yj [ ^"^ fli fft i
Q! rt» ' r™" ' ^ ^^ P^i f C p"— ' rt) p • ~~^
llil l!f|
O O 03 13 O S- O O V""'«3- ^ W W ^ — ' o W ^ ^^ w w
^^ ^^ ^^ QU — — 1 O_ C\| ^J |— f_ J_J^ ^. -y» | , ^^ O
•3-








^ CD CD CO O
r~~ CD r~™ ct* i ^^ „/ _^ _ * • « C3
I c?4- »— t i * ^* *** ^ r— ef- •
^^ ' ' 1 CD III i , t ^^ 1
f— r— O CO LO CO CO fi r-*i -i* ^ ' * f1^ r—
oo^-cn.^or-o o o r-'cM'd cvi.dd d dd

•a
CO CU
•P <= c
C 0) a)
o -£= ^z
0 -P 4J
•>— ' C. C.
<0 f5
0) S- s_
C O o
0) 3 3
•P **•" '[
C ' — " . , — , • '
re -Q -1-3
£_ 1 	 1 , 	 ,
0 0 o
3 N |s|
j_ a) ni
                                                    in
                                                                  r- VD

                                                                  CO i—
                                            i-n
                                                   CM
                                                          cn
                        l£) If)
                        r-- r-s
                        CD CT>
                                            O     S-     i.
                                            S-     0»     O)
                                           •P    -P     E
                                            cu /->  c.     e
                                           Q tn -I— m  3  in
                                               <1)  S OJ  in    nj a)
                                            0)  W T-
             wo)
             o  tu
                                              .,,
                                           •1— tO •!-  O T-
                                                  o  u
       in a)
       o en o
Q.T-  Q. C -p
~ -P  E eC  C
              O
                                                         u u  in
                                                                o
                                          •no.0!
                                          r-     CSJ     sj-
                                           1       1      1
                                          •-;     r^     «\J

                                          •—     *—     o
                          |
                         CM
                      CM r-

                      do
   OQ
                    CO
                                          01

                                          O)


                                          C


                                          o
                                         N
 O)
CQ
5-29

-------






h-4

ra
u
3





^
E S- CU
0 CO 4->
• p" if^- tw
FE°'-5
4J O W C
E C3 in «i—
CU CD CU
O r— t— CC
O O> 3 -r^




















CD
Mel
Z

m en
CM




ID CO
ID r-
r— r—






in
CU
r~
O CU
•i— O)
S- C
CO 
0 O/-N
_l (A
C 0)
•i- C r—
•r- Q.
CU tn ro
4-3 eu in
u)'i- ^

o
CM «3-




O CO
I — 1
1 CO
co o
i-^ O







CU

CU

4J
E
CO

o
r— -a
M— —
i — i-P
.M c,

O O
N ' — '
C
CU
CO

CM
CM




f^^
f^^
I—









O
(4—
s- -a
•r- E
CO CO
2E .
S_ T*r
0
Z



CO
CO
o
o
CM
00






0)
C
CU
^=
4-3
E
CO
S-
0
3
q_

^

I_«J
O
N
C
CU
00

CM CM CSl"* CM CMCSJCM00^ CM CM CM CM CM




in in • CM CM r>- un ir> ID m ID co LO CD CM JD P--
S $ gg B: S;S^&;^5;c7;S^ ££ g

CO JZ
•r- U
r— CO
CO CU
r— S-
en cu in 4-> 4->
c E cu in co
..— Ei — 3
,_ 3 a) .
o» ^:
i — i CU +J
O E m
N CU 5;
E S- T
CU >> r-
CO Q-
5-30

-------










ce
i — i
- f^. LO LO l^Lninioi^ ,,->,« « ^. ^
coco CD CD cof"- coco cDt^-rSr^fCf^r^f^
CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD

fg >> >> 4-> .jH
CU S S *o •"
s- oj -P +j i. - "T; j°
18 - § § -sl 1 5
Jl-g^ s^ -5E f5 - ^ 1
3^^cu cucI'Sm §""a^-
' Q-*"~ -i— tO **— JDZS^ m5
O^"^.^"^"^ Cl-LU " S- C+->
c £ fe'G ^ 'G 5 = -5 1 1 1 c | = «« « 5 S
•a 3 .£V .1- r^. £ = ra4H§w+j"~ r?a, ^ ^ °°h
C "O tO tO CU 1 O) *i 	 to cl j t rm 'Tij
c:'~Ito •^ -3. co = ^z. - i— . T • • 0 . CM
1 ?0 CD en •- CM . 0 .
i — co i i LO r»-7 oo r— T" CD '
o r~- co co op- co o CM CD co LO >^- 1^ CD T
00 CM 0 CDCM CDr- CMr-'oo'Sc^do'o'ddcM


cu
s=
cu
S-

Q.
1 1
to

o
M
r~
cu
QQ




^h ^" ^f* CD CM
CM CM CM CM OO







CM co LO r— r*.
co CD CD r— co
en en CD CD CD


o
y-
-p
to c: cu
0) to E Q.
v~' ~8 2 £
to =5 q- 3
O) 4-> LU
•r- . cu cu s- -a
•P CO S- 0) T- c C
•i— . CU S- to tO T-
U ZD -C -P r—
O. to I D) to
• • — s cu to ecu
co ro en o c >>LLJ •!—
• -P tO E O to -»->
ro to t. +J -o s . .1-
"O CU IO C S- Z U
CM > O O





CD
o
cz>
1
CD LO
LO CM 00 LO
CM O 1 CD CM
1 . CM . I
i— LO i— CM CM



cu
c
cu
S-
^J
a.
J 1
cu
1 1
o
N
c
cu
OQ
5-31

-------
                                      CM CM    CM
                                                      CM
                                                               CM
                     00 i— '
                     CM CO i
                                                                                           CO
                                       LO CM
                                       VD l^>
ID
UD
i CM

I cn
ID CD ID
r- r-» r-.
cn cn cn
                                  un
                                  r^-
                                  (XI
                                           O)
 tO        CUE
_Q        fO -P O-
t-       ^=n- t.
 3           O.M-
       • ^ ^^

  .  d)  CD S-  •> S- "O
If)  S-  CD -P S- n- C
  .  cu  s- cn to to to
                                                       -P  3 -P
                                                 o>
                                                    Q. in "O
                                       •i— CO tO D) O
                                                 fO
                                                          n-r-  >>LLJ
                                                              (0  tO
                                                                          in O
                                                                          0) I—
                                                                          -
                       O  O  3   n-       n-
                      o;	l co   3C       DQ
                                                                               -P
                                                 CM   CO i—
                co
          i—    CM
                       I CO
                                                                              3.3 3.3
                                                                                    O5r—    CO  N  O
                        • CM OO
                                0
                                       u
                                             I  CU i-
                                        i— 1—    O
                                                        CM O
                                                                       O O i—    CM
                                         o>
                                         D- CU

                                        S* CO
                                        •r-3 S-

                                        •cT-p
                                         u  c
                                        i — i (0
                                         O .C
                                         N -P
                                         c  c
                                         O)  fO
                                             CU
                                             c
                                             O)
                                             O)

                                             to"
                                             N
                                             c:
                                             (U
 O)
CO
                                    5-32

-------






1 — 1

u
CU
CO
co
c
o
•r—
re cu
o +->
•r- re
t— Q
X}
Q_






£=
0
-IJ
re
u
o




" i
a s-
o cu
*-5n -4-5
re E o
•t-> o in
cu o cu
U r— r—
c^ c
O D5 3
O i*— '














CO
E
re





















/-"•%
T3
a>
-p
re
u
•r-
-o
c
CU
in
*r^




















•sl" to oo i — cn
CNJ CM CM CO



Lf) r — LD UD CO
en en en en en





in
cu
'5 1 |
i- S- «=c
a in s- ~a o
cu T— c in 	 i
c i — re re a>
•i— Q- i — i — i
E i en co
in re . c en in
cu in >,uj c: cu
•4-> re 
s-
Q.
U
CO
CM
r—
' '
O
cc
a>
-o
c
I-H


S C^^S0100^10 ^ « «« •* *••-•* U3
1 ' CMCMCMCMCM CM

.
LO LomiotOLniOLO in ui esuo' CM «- CM r-
2 2 2 2 2 £? 2 °2 ^ ^ ^2 2 ^ ^ en

re
•— >> o
"^ re •— >, s-
CU t. re r— it r-n

C ^/) ( — | i> 4 4
•»-» 1^ rf ^^ -P ^
o "o s_ £~>m cu^curecoE
a'ln'S jJ1 Si ° ° c'w'D::c§ S 3 >, °~'*~
OJ-p- S-^ mo O -Sr^' r: 1^


i i
•^- «r—
•P -)->
re re S
°- °~ o re
ir> _, cncocncu ai i
10 ,° ~sv.4->^v-pLncD
r-^ "c LT> cnrecnreuo . r— < ,
r~"e^r"~ ^SiOCO^CO i 	
co a><-oog^-S^UL^UcM ^-^ rer-- J2 2
•— Q- 00* ^ ^ r--' CD ^ r- 0* CD ^-' £ 0* 0 CD











CU „
cr . __
cy ^~
in ^cu
t* >>
CJ CU
5-33

-------








oi
»— i

u
o
O-


•O
+J
o
o
T
in
0)
fO












0)
0
c
Q)
Q)
'1 T,
0)
C£
C
O
•r-
(0 0)
•r- rtJ
f— a
.0
3
ex.





Location


^^\
i -o
c s- 01
O CD 4-}
•r- JC. C0
4*O 4J 0
(0 £ O «i—
4-5 O 10 C
£= 0 W-r-

CM CO i— CM






m m io in io co







in
0)
'co
0) C
o
ce in
o
in i —i
0)
i— in i
4-5 OJ O)
in D) o i — tn
Q) C 4-> Q. CD
0.0: c e 4->
CO CO O CO «r-
E -O in S- in in
0 3 0 O





10
o
CO
CM
CT> O IO •
r-^ "3- 0 T
1 O CO 1 I—-
CM i — i — co co o
O r— O O O O




-o
c
o

CO

CO

>fc
s-

a.


^. _j. ^.j. ^-^t io io io cMOOcncnoopr— in
CM CM CM CM r- r— CM CO CM CM CM CM CO CM






ir> CM m in CM m in r-~ r*. in io jo in io jo r^
io vo io 1010 r*- r»- r~» &mS

"~

_
o • ._
i- tn
£ 05 Ji
c i
4-> f— 3
4j in c= -o-e -.2 l~
•r-CO«O £=0 §S--r-?«_ C=
g ^ -g 5.^5 5 2 2 Io o
•S^S 3 ^^^S § *- 5 • -f =
Q^.^ .coa>£? »£ ° s--o -^« mS
c».-!i ^feSilnm^ ,£ -5§= fe= S-^
"*^° ^"1^1"^' ' ' ? S 0) o •£ S) u)
SScolocnocT-r-gt. tj >»LU -j-j 0)ts c^ c ^c ^
^^CM"° J re § J ° i 5 lZ^£§5C2m^5cM
CO r~- "5J« -^— ^— . v" ^** * ~~ ^


i l
• .
4-5 4-5
s- i- **•
(0 CO tO
o- cu a.  • csl
IH IO CJj's 05*3 i— CM fO CO
i in «^-in, n.O3.or>» 101 oor«~cMO
00 T CM"^ o o °^ i • • tf» . . • •
CN] CM CO r^CJ S -CO r- rocOCOOOOOCO.-



c:
a>
CO
Q-
•r—
C35
t t
O
NJ
C
0)
QO


in
CM






r-
(O
^~






<~N
r-
c
O 1
en — i «i-
r™" (/5
1 1 S-
^- 0)
^^ in •>— '






CZ)
o
o
p~"















5-34

-------





1 — 1
— Q
-Q
CL.






c
o
»r~*
-P
fO
(J
o



/—•N
•> i "O
c; s_ oj
0 OJ -p
•r- .XT re
4*0 -p u
JO E O T-
S- T3
•p o in c:
c o in •!—
OL) CD O>
U i — i — OJ
c\. c in
O 0) 3 «i-
0 n.v_^ 3:















OJ
E
03
21


LO to LO LO cn
CM CM CM CM


ID ID LO ID CO
ID ID VD ID r~~
cn en en en en
to
> E  c/5
W r— 10 |
in s_ o i a;
OJ oj o  > Q. =5 OJ O. Ol
•r- C i- r— -p E +J
in -r- OJ ro -r- ro -r-
v- ' > > m in in
CM -J ^- .3- ^|-








t — VD CM
•vl" en oo en
i iii
CO CT> <=J- CM
LT> ...
ID <^- O i — O

73
C
O
o
O)
c
O)
^>

CX
!E
O5
i^i—j
o
N
C
OJ
CQ


CM CM CM CM CM CO (


LOCM LO LOCM r~- r--a
i— s_ j_
-P in 3 a_iTi 5
OJ OJ -P >> ij]
•P <-OS-OJJ->l_.,— r-cr
f~" i ^|T f^ ^^ ^^ - ^~" *~~
if) .^soj0"10"0 ! ^^
QJ CO fO CT O C *r- p— >jUJ «r—
4-J • ^_? fO E O fO ro i ^
 "i^ S— O 5c • , _fl
to *^j d) fO c fO U S— ^y u £
CMr— ^ ^Z JjJ .-(.^








00
<^ . CO
~; ' o .
<^ ., 1 CO
To OLO ^!
f10 rO CM CD CM ID ^f"
CM (— CM «i- 0 d ri- ^











OJ
C
OJ
c
o
s_
o
o
                                                       CM
                                                       CM
            CM
            CM
                     cncncncncncncncncncn    cn
                                                                en
                                 
                                                                tQ
                                                               O
                                                                S-
                                                                o
                             en
                          LO
                          o
                                roro
       CD

' CD    O

i i—    CD

i r-i-    CD

 f~^    e^i
          0
          o

          d
                                                     O)
 S-
 u
"=C
                                                              in
                                                              
-------







o;
t_
S
UJ
1— <

^
1— 1

UJ
o

0)
ja
to









o
o
£=
CU
cu
II
»f «•
cu
a:
o
•r—
tO CO
U -P
•r" tO
i— a
,Q
=5
CU



4
Location


^
1 T3
C. S- CD
o eu -p
•i— J^ to
•P/0 -P U
10 E 0-r-
i. -0
4J CD 10 C
C O to «i—
CU 0 CU
O i— r— CU
c->> c w
O OJ 3 «i—
C_> 3-^*^ 3?












CU
Z


CM

,



r-.
r^ i
CD 1







s-
•r—
fO
C
to
s-
3
s-
•I- tO
10 CU
t5£
ja o
za to
CU C
C 0
CU •!—
N -P
C U
CU (0
OJ.
•^ cu
OJ<—
n.-Q
^^^ ^3
o "o
CM OO to
CD CO









cu
•p-
•r-
i-
o
fO
1— 1
to
N
cu
. CO


CO ^ L
CM CM





r- LO
to to
CD CD
"iH i—





s-
'5
c
to
s-
3
3 S-
CU •!—
OJ tO
S- C=
cu to





u_
oo
CQ
OJ
OJLO
CD T
CO r—
rLd















CM CM CM





LO r^- LO






s-
•r-
fO
C
ra
s_
s- s- = s-
tO tO OJ to
to
c. c. s- e
fO (0 CU fO
ja .a > -Q
s- s- <: s-





LL_
00
CQ
OJ
OJ
«* 00
0 00 r— 0
CD CD O O






CU CU
C L-
x» ;5
'^ *C
0 U
CO fO

•1—3 -C
to" nT
M N
C. C
cu cu
5 Q


^(- i— CM r— CM
CM CM CM CM





CM co r~ co r~-
to f^* i*^ ^ Is**
CD CD CD CD CD




OJ
C.
I—
"c "u s- s-
tO -P 'r- 'J-
*D-
10 to o o
3 •*->
^ . f f
-i- i— 0 O
to >- >-
s_ o cu cu
cu o cu o cu
Z Q£ "Z^O£. Z





CD LO
to co
0 0
CD 0
1 1
CM CO
CM CO
CD i — O IO O
,— r- 0 0 0








CU
•I —
o

« g "1
0 •!- CT
M r— i—
fO O >>
o c -c:
^ vr-* ^-9
10 3 CU
cj <=y -z.


CM





CO
p-
CD








cu
o
Q£





co
CD




CU
c
*^~
o
c
«r—
3
CT
r—
-P
CU
(E
to
CM"


CM
CM





r--
CD







s-
*tO
•r—
O
s-
o





00
00
o
i
o
o






to
cu
c
1—
o
c
3
CT
^>>
-P
CU
O


CM
CM





r*-
CD







S-
•r—
fO
J?
o
s-
0
1





CM
CM
CD
CD
O
0








to
cu
c
o
c
•r-
3
CT
LU


CM
CM





CD







5-
'iti
•r-
0
s_
'0
z





o
o
CD
a










to
cu

o
c.
3
CT
O
CO
5-36

-------










O£
J j
1—

ca
-
o
a.

.
•a
c
o
o
s^**'
Ol
1
0)
-Q
re
i—







0)
u
c
at
s_

0)
a;
c
o
•r—
re cu
U -4-J
••- re
r— Q
-Q
a.


c
o
•1—
^_>
re
u
o
— i





^^
i -a
C. S- CD
O CU -P
•i- .e re
-t-*r> +J o
re s o T-
S- T3
+•> O 05 C
c o w ••-
cu o  i r^
O 1 O5



I
o re
***-^ j ^

S- • f~^
're W If
re ^ ^
S- Z3 <1)
Q) /*^s
C C
ai o
N] »r~

.
^ ™ * *^* "^>J LO VNJ C\J CSJ CSi C\l
tM*LO(~»r~>.r^.r>»i^
uoir- icik k P; p; P;
OJICT; OT.OT 5; OT &; S;
*~" I— 1—


t/) U)
UD -P* LD -5* ^ ^ *** "^


•I— '1- -^ -I- •!- ^£ ^ _J^ _x^ v^
re re s~ re re s— s- ?— s_ g_
= =>- CC>- >- ^ >? ^
•Q-QS -Q-Q3; 2: * S •*
^-^-0) J-s-S S g S S
= => z => =3 z: "z. -z. -z. -z.




1 1 Q_
to co
|>g » § § . o «
Q, ^ O i — CO O
•5^ • • v3 o o o d
C^ . ?: ^O ^--^ CD CD
**) en o r— core o i — I— ^ o
o •— o o'-^-f— o' d o' d



 10
^ '- -^
i = o
.E ••- =
fl) ^ 3 •>-
1 §• • §" s-
o 51 g £ «
1 £ - i £ >>
o- "51 T £ -g 5
S | 5 g s 1
's' "= re '5 °° t.
^ "-H c cr s_ ai
S 4- OJ 0 0 J=
CQ r— rt ^ "^




^v
CsJ





r^
r^
O5
"~~



re
+*

if
^
3:
a>
•z.






<\j
U3
•
Q





O5
0)
C
•r—
r—
O
/lisoquin
^=
4->
 J> ^ 1^.
i — r** r>- ^
cri en OT o^
r— t— i — , —



re re re re
4J -M & +?

t t t t
-° £ o o
* 3B 3- 3:
> -r- N N
•P d) 
-------
          0)
          u
          0)


          0)
            U
           £80

          •pom
           COM'
           o> o 

                                                                                o>
                     in
                     r^
                     en
Lf)
r-.
en
                                                re)
                         u
                         O)  (D
                                                   O)
                                 •r-     S-
                                 ns     o
                                                                                S- U)
                                   O>
                        0) V     (0 •!-
                        o> u)    -Q  fo
                        «u         s
                        0) (0
                        > -Q
                                                        O)
                                                     5-  0) -r-
                                                    •i- n- U
                                                                                  •r-     O
                                                                 
                                          .      .
                                     m o> 
-------
  5.2  SUMMARY AND CONCLUSIONS
       Numerous POM compounds have  been  detected  in  ambient  air  (Table  5-9) with
  the most  extensive data  available on benzo[a]pyrene  (BaP).  The  concentrations
  of  BaP are different  in  various cities and at different times  of the year.  In
  general,  the  concentrations are usually highest during the winter months and
  lowest during the  summer (Figure  5-2) which has been attributed  to heating
  sources of BaP during the winter.   The most comprehensive monitoring data on
  BaP are available  from the National Air Surveillance Network (NASN).   BaP data
  are available on the following stations during the indicated years:   1966 to
  1970,  250 stations; 1971  to 1976,  40 stations.   Trend analysis of the NASN
 data indicates that BaP concentrations  have declined considerably from an
 annual median value for urban  sites of  3.2 ng/m3 in 1966  to 0.5 ng/m3 in 1975,
 an 84  percent decrease.   The decline is believed to be  due  primarily  to
 decreases  in  coal consumption  in residential  dwellings  and  buildings  as well
 as to  improved disposal of  solid wastes and restrictions on open  burning.
 Concentrations of BaP  in  urban  cities where coke ovens  are  located were
 considerably  higher (40 to  70 percent)  than the non-coke sites.   However, this
 may  be due to  the fact that most of the coke ovens  are  located  in the 'north
 (BaP from  heating)  and in highly industrialized areas.  Recent  trends on other
 POM  compounds  are  not available.
     The accuracy of the absolute  values of BaP from the NASN data is unknown.
The collection efficiencies of glass-fiber filters with high-volume air
samples has not been determined and the  importance of BaP associated with
particles less than 0.3 urn (which  could  pass through the glass-fiber filter)  is
unknown.   Oxidative loss of BaP during collection is another unknown;  BaP
                                     5-39

-------
degradation may occur between the time the sample is collected and then
extracted.  In addition, there is experimental evidence indicating that recov-
ery efficiencies are considerably different at summer versus winter tempera-
tures.
     Trends for other POM have not necessarily followed the decrease for BaP.
For example, in London the concentration of coronene has remained the same
while BaP was decreasing.  This was attributed to the greater proportion of
coronene  in auto emissions which are increasing in  London.
     Whether BaP can be  used as an indicator  of other POM compounds is not
fully established.  Some investigators-have found good correlations between
BaP and other PAH while  others have reported  little or no relationship.
       4
Correlations between BaP and other POM probably vary in different areas and
are dependent upon the  local POM emission  sources which will provide different
ratios of individual POM.  Because the ratio  of many of the  individual POM
changes in various locations, the use of BaP  as an  indicator of  carcinogenic
potential should be  reevaluated.  A reasonable alternative would appear to  be
some  form of multi-component analysis that would  include  some  of the other
carcinogenic or tumor-promoting  POM.
      A little  information  on the  nitrogen  heterocycles, aza- and imino-arenes,
is available.   In  general,  these  compounds are  found at concentrations 10 to
100 times lower than the PAH,  although  a recent study  in  Rome found 1  to
10 ng/m3  of individual  aza-arene compounds.   The monitoring trends  of  aza-
arene concentrations are unknown.   Whether they are increasing or decreasing
 in concentration  and whether there is a seasonal  fluctuation is also
                                      5-40

-------
 undetermined.  They have been detected near coal conversion plants, but their
 major emission source is unknown.
      A recent study in Norway has demonstrated that at least 20 PAH associated
 with particulate matter are stable enough in the atmosphere to travel  long
 distances.   Four of these compounds,  including BaP,  have been shown to have
 carcinogenic activity.
      In  summary,  the concentrations of at least 25  PAH and 32 aza-arenes or
 oxygen-substituted  POM  have been  determined  in ambient air which  suggests
 that they are  stable enough to come in  contact with  humans, animals, and
 plants.  The concentration  of BaP  has decreased significantly  over  the last 10
 years (84 percent decrease), but data in  the United States on whether other
 POM  compounds have followed this trend are unavailable.  Data in other coun-
 tries have indicated that POM compounds (e.g., coronene), which are found in
 relatively large proportions in automotive emissions, are remaining constant
while BaP levels are decreasing.
                                    5-41

-------
5.3  REFERENCES

1.   Bridbord, K., J. F. Finklea, J. K. Wagoner, J. B. Moran, and P. Caplan.
     Human exposure to polynuclear aromatic hydrocarbons.  In:  Carcinogenesis,
     Vol. 1.  Polynuclear Aromatic Hydrocarbons:  Chemistry, Metabolism, and
     Carcinogenesis, edited by R. I. Freudenthal and  P. W. Jones.  New York,
     Raven Press.  1976.  p. 319-324.
2.   Hoffman, D.  Benzo[a]pyrene in polluted air.  Prev. Med., 1:450-451,  1972.

3.
     Special Report:  Trends in Concentrations of Benzene-soluble Suspended
     Particulate Fraction and Benzo[a]pyrene.  1960-1972.  U.S. Environmental
     Protection Agency, Research Triangle Park, N. C.  Publication No.
     EPA-450/2-74-022.  1974.

4.   Sawicki, E.  Analysis for airborne particulate hydrocarbons:  Their
     relative proportions as affected by different types of pollution.  Nat.
     Cancer Inst. Monograph No. 9, 1962.  pp. 201-220.

5.   Scientific and Technical Assessment Report on Particulate  Polycyclic
     Organic Matter (PPOM).  U.S. Environmental Protection Agency.
     Washington, D.C.  Publication No. EPA-600/6-75-001.  1975.

6.   Manning, J.  Environmental Protection Agency, Research Triangle  Park, N.C.
     Personal Communication.

7.   Sawicki, E., T.  R. Hauser, W. C. Elbert, F.  T.  Fox, and  J.  E. Meeker.
     Polynuclear aromatic  hydrocarbon composition of  the atmosphere  in some
     large  American cities.  Am.  Ind. Hyg. Assoc. J.  23(2):137-143.   1962.

8.   Kert<§sz-Sa>inger, M.,  and Z. Morlin.  On the occurrence  of polycyclic
     aromatic hydrocarbons  in  the urban  area of Budapest.   Atmos.  Environ.
     9:831-834,  1975.

9.   Gordon, R.  J., and  R.  J.  Bryan.   Patterns  in airborne  polynuclear hydro-
     carbon concentrations at  four  Los  Angeles  sites.  Environ. Sci.  Technol.,
     7:1050-1053.

10.  Faoro, R.  B.   Trends  in concentrations  of  benzene-soluble suspended parti-
     culate fraction  and benzo[a]pyrene.   J. Air  Pollut.  Contr. Assoc., 25:638-
     640,  1975.

11.  Faoro, R.  B.,  and J.  A.  Manning.   Trends  in  benzo[a]pyrene (1966-1975),
     preprint.

12.  Thomas, J.  F., M.  Mukai,  and B.  D.  Tebbens.   Fate of airborne benzo[a]-
     pyrene.   Environ.  Sci.  Technol.,  2(1):33-39, 1968.
                                      5-42

-------
  13.
  14.
 15.
 16.
 17.
 18.
 19.
20.
21.
22.
23.
                                                                   sn
                                                           sn?
                                     s:
                               ^
                                                                  2-
1976.



Brocco, D.,  A  Cimmino, and M. Possanzini.   Determination of aza-hetero-

cyclic compounds  in atmospheric dust by a combination of thin-layer and

gas chromatography.  J. Chromat., 84:371-377,  1973.             Y



Dong  M.  W.,  D  C. Locke, and D.  Hoffmann.   Characterization of aza-arenes
in basic  ornamr  nnvtinn nf 
-------
24.  Hoffman, D., and E. L. Wynder.  Chemical analysis and carcinogenic bio-
     assays of organic participate pollutants.  In:  Air Pollution, Vol. II,
     2nd Ed.  A. C. Stern (ed.), Academic Press, New York, 1968, pp. 187-247.

25.  Hoffman, D., and E. L. Wynder.  Organic particulate pollutants - chemical
     analysis and bioassays for carcinogen!city.   In:  Air Pollution, Vol. II,
     3rd Ed., A. C. Stern (ed.), Academic Press, New York, 1977, pp. 361-455.

26.  Lunde, G., and A. Bjrfrseth.   Polycyclic aromatic hydrocarbons  in long-range
     transported aerosols.  Nature, 268:518-519, 1977.

27.  Colucci, J. M., and C. R. Begeman.  The automotive contribution to airborne
     polynuclear aromatic hydrocarbons  in Detroit.  J. Air Pollut.  Control Assoc.
     15:113-122, 1965..

28.  Liberti, A.,  G. Morozzi,  and  L. Zoccolillo.   Comparative determination  of
     polynuclear hydrocarbons  in atmospheric  dust  by  gas  liquid chromatography
     and spectrophotometry.  Annali di  Chimica, 65:573-580,  1975.

29.  Fox,  M.  A., and S. W.  Staley.  Determination  of  polycyclic aromatic  hydro-
     carbons  in atmospheric particulate matter by  high pressure liquid  chroma-
     tography coupled with  fluorescence techniques.   Analytical Chem.,  48:
     992-998, 1976.

30.  Colucci, J. M.,  and  C.  R. Begeman.  Polynuclear aromatic hydrocarbons  and
     other pollutants  in  Los Angeles  air.   In:   Proceedings  of the International
     Clean Air  Congress,  Vol.  2.   Academic  Press:   New York, U.S.A.;  London,
     England.   1971.   pp.  28-35.

31.  Gordon,  R. J.  Distribution of airborne polycyclic  aromatic hydrocarbons
     throughout Los Angeles.   Environ.  Sci.  Techno!., |0:370-373,  1976.

32.  Waller,  R. E., and B.  T.  Commins.   Studies of the smoke and polycyclic
      aromatic hydrocarbon content of the air in large urban areas.  Environ.
      Res., 1:295-306,  1967.

33.   Krstulovic, A. M., D.  M.  Rosie,  and P.  R. Brown.  Distribution of some
      atmospheric polynuclear aromatic  hydrocarbons.  Amer.  Lab. (July 1977),
      11-18, 1977.

 34.   Bender, D. F.  Thin-layer chromatographic separation and  spectrophoto-
      fluorometric identification and estimation of dibenzo[a,e]pyrene.   Environ.
      Sci.  Techno!., 2:204-206, 1968.

 35.   Stanley, T. W., M. T. Morgan, and J. E. Meeker.  Rapid  estimation of
      7-H-benz[de]anthracen-7-one and phenalen-1-one in organic extracts of
      airborne particulates from 3-hour sequential air samples.  Environ. Sci.
      Techno!., 3:1198-1200, 1969.
                                          5-44

-------
36.
37.

                                       5-45

-------

-------
                               6.   HEALTH EFFECTS

  6.1   PHARMACOKINETICS
       There  are  no data available  concerning the pharmacokinetics of  POM  in
  humans.  Nevertheless, it is possible to make  limited assumptions based  on
  the results of  animal studies conducted with several POM, particularly BaP.
  The different patterns of metabolism for POM in human and animal tissues has
  been especially well-studied, and has contributed significantly to an under-
  standing of the mechanisms of POM-induced cancer.
 6.1.1   Absorption
      The demonstrated toxicity of POM by oral  and  dermal  administration indi-
 cates  that  they are  capable  of passage across  epithelial  membranes.1   The
 high  lipid  solubility of  compounds in this  class supports this  observation.
 Early  studies  conducted on the  intestinal absorption  of BaP  confirmed that
 transfer across  the  gut readily occurs and, moreover, that tissue accumulation
 is exponential as BaP concentrations  increase.2  These results  are consistent
with a mechanism involving physical adsorption of BaP to  the intestinal
mucosa as a first.step.  A unilayer is presumably formed  at saturation concen-
trations and a multilayer at higher BaP levels.  Increases in the medium
concentration of BaP above the saturation level would thereby permit an
exponential  transport into the cells by passive diffusion.  The authors
indicated further that the intestinal  transport of  MCA,  DMBA, chrysene, and
                                  6-1

-------
benz[a]anthracene was similar to BaP.   Assuming that the ability of various
organs to absorb BaP (and other POM) also follows an exponential relationship,
relatively small differences in the magnitude of environmental exposures
could be expected to exert a dramatic influence on the body burden of POM.
Such a relationship has been demonstrated for the accumulation of BaP in
               o
adipose tissue.
     The absorption of POM through the lungs has particular relevance to
environmental exposure situations.  However, the major health-related effects
of inhalation exposure involve  local lesions in the respiratory tract;
unfortunately, few investigators bother to measure systemic levels of POM.
Nevertheless, it is known that  BaP administered intratracheally to rats
       *     *
appears  in the tissues with the same pattern of distribution  as when given
parenterally.3  Vainio and coworkers4 recently showed that unchanged BaP
quickly  appears in the perfusion fluid of isolated perfused rat lungs fol-
lowing intratracheal  administration of a 200 nmole dose.  Pretreatment  of rats
with  an  intraperitoneal  injection  of MCA to  induce microsomal enzyme activity
caused an  increase  in the amount of water-soluble metabolites of  BaP appearing
in the perfusate.   An increased covalent binding of  BaP  metabolites to  the
lung  tissue  also  accompanied  the MCA-mediated  induction  of microsomal mono-
oxygenases.   The  significance of tissue-binding  of  POM  metabolites is dis-
cussed in  Section 6.2.6  of  this report.
      Because POM  generally  reaches the  lung under  environmental  conditions  by
 adsorption on carrier particles,  the extent of particle deposition in the
 lungs and the elution of the chemical  from the particle have  an important
                                                                         5
 bearing on its biological  effect.   In this regard,  Creasia and  coworkers
 showed that when BaP is  adsorbed to large carbon particles (15-30 urn) and
                                    6-2

-------
  Instilled into  the  lungs,  50  percent  of both  the  BaP  and  the  carrier  particles
  were  cleared  from the  lungs in  four to  five days.   Little carcinogen  was
  released  from the carbon particles in this case,  and  therefore contact with
  the respiratory epithelium (and carcinogenicity)  was  low.  With smaller
  carbon particles (0.5-1.0 urn), however,  50 percent particle clearance was not
  achieved  until seven days after administration.   In this  case, 15 percent of
  the adsorbed BaP was eluted from the particles and left free to react with
  the respiratory tissues.  In the complete absence of carrier particles,  BaP
 was cleared from the lung at 20 times  the rate of adsorbed BaP.   This  obser-
 vation may explain the difficulty in producing experimental pulmonary  tumors
 with BaP  without the use of carrier particles.   Other  investigators6 confirmed
 that carbon particle size affects  BaP  retention in the lung, but  also  demon-
 strated that BaP retention  was not affected by particle size when  adsorbed  on
 ferric oxide or  aluminum oxide.
     The  elution of  BaP from carbon particles  instilled in the respiratory
 tract  of mice  can be enhanced  if the particles  are instilled during  the acute
 phase  of respiratory infection.7  The fate of the  eluted BaP in the  lung was
 not determined although  the author predicted that  rapid elimination would
 occur, thus  reducing the risk of pulmonary carcinogenesis.
 6.1.2  Distribution
     Regardless of its route of administration, POM, once absorbed, becomes
 localized in a wide variety of body tissues.   The distribution  of radioactivity
derived from 14C-BaP in the rat and mouse was  determined following subcu-
taneous, intravenous, and intratracheal  administration.3  The pattern of     '
                                   6-3

-------
distribution was found to be similar in all  cases,  except for high local
pulmonary concentrations following intratracheal  administration (Tables 6-1
and 6-2).  Concentrations of BaP-derived radioactivity in the liver reached a
maximum within only 10 minutes after injection and represented 12 percent of
the total dose.  Radioactivity in the liver was reduced to one to three
percent of the administered dose within 24 hours.  Similarly, maximum blood
levels of BaP following intravenous injection were reached very quickly, and
radioactivity became barely detectable after 10 minutes.  Minimal tissue
localization of BaP and/or its metabolites occurred in the spleen, kidney,
lung, and stomach; maximum radioactivity derived from labeled BaP was recov-
ered in the bile and feces.  Levels of radioactivity in fat, skin, and muscle
were not determined, nor was the amount of unchanged BaP measured in any
tissue.  Bock and Dao8 later showed that relative to other tissues, unmetabol-
ized BaP was extensively localized  in the mammary gland and general body fat
after a  single feeding of the carcinogen (10-30 mg).  This accumulation of
BaP was  greater than that resulting from administration of 3-methylcholanthrene
(MCA), 7,12-dimethylbenz[a]anthracene (DMBA),  or phenanthrene  (Table 6-3).
In all cases, the level  of carcinogen achieved in the tissue was  directly
related  to  the dose administered and was dependent  upon the  use of a lipid
vehicle.  The  authors suggested that the mammary fat pad may act  as a  car-
cinogen  trap, which slowly releases the unchanged hydrocarbon  to  the glandular
cells.   In  the rat, the  glandular tissue is  a  common site for  POM-induced
breast cancer.
     A more complete  study on the tissue distribution of MCA was  reported  in
1959 by  Dao and coworkers  after  it was shown  that  oral  administration of MCA
                                    6-4

-------
     o



     o
     h— I
     I—
     O
     UJ
    o

    UJ

    «c
    cc.
    Of
    LU
   •— < »o
   >  c/o

   P  Sc
   "  «

   §  o

   S   fe
   —I

 a:


 oo
 1—4

 O
 I
ID
            CT>



•— O f- O VO
                                                    01
                       O *->
                    m w in t.
                    O) E O) 01
                    O O *J > •


                    u_ (/> •- 'Jj :
                                            1
                                             ID

                                             in
                                             01


                                             in -o
                                             O) O)
                                            4-1 S_
                       in o
                       c u
                       ID O)
                       O) ^


                       O -U


                    ^ • i*. Q>
                 O^ O OJ O
                 C i— J= S-
                 3 Q.4-1 OJ
                 —I 00 O Q-
                                                        at
                                                        >
                                                        E

                                                       OS

                                                       O

                                                        C


                                                        ai
                                                        c
                                                        O)


                                                        I
                                                                                O
                                                                                ro
                                                                                 OJ
                                                                                 v>
                                                                                 o
                                                                                o
                                                   6-5

-------
0
«£
_J
—I
I—
Z
*""*
_J
UJ
a:
o
*^
Qa
5
OS
I—
I**"*

OS
UJ
t "
«c
^^ CO
»— "
U. UJ
0 Of
z a.
o i — i
^^ fO
tc o1
Zi^i
r^i
*—* Z
•-< CO

UJ 0
U. r—
Ut O
i


^^
•a:
o
t»<-4

co
OS
i—
CO

o



CM
VO
U •
ra
H-
^
*S c
o
0 *J
•^* 3
10 O
r— I/I
^- r—
4-> O
I/I U
C >>
**" *o>
S-
03 03
4-> C
<4- Ol
10 *-
01 J=
•5 01
1— 1-
^J






0 •
C J=
o
^ e
10 O
r— in
•*• C
J->  3
M- O
3
oi o-
E 10
p

































**"

















CM






^™








































Ol
in
O
•o
•a
u
ai

y
e
1
IO
,!_
O
X























01
•t~

















en ai
CM CMOOOOvor^. CMir>
r^O^ruicocMr^OOvo
r— T r—















<*> n .
CM r— CO CO




SCO
_ f
— — r'J_*oof^OQ,_I
m *f co






*B
|
IO
(j
o
I1
c
IO

in
01
l/l "O .
Ol Ol
4-> S-
— . 01

in o
C U
10 01
41 cn t-
C L-
j= -J: o <->
mloSua.?? SuS
oiEOioiccaiaioiu
CJ o *•* > •|W T3 C r~ ^ ^
S *J C-f- S- •-- 3 O.-M Ol

































ai

U
Ol

E

•
o
c
a>
c
ai
L>

a.
IO
N
c
Ol
1
5

O)
a
in
CM


oi
1/1
a


6-6

-------
Table 6-3.  EFFECT OF MOLECULAR STRUCTURE ON HYDROCARBON LOCALIZATION IN MAMMARY
                                 GLAND AND FATa8
            Hydrocarbon
3-Methylcholanthrene
Benzo[a]pyrene
7,12-Dimethy1benz[a]anthracene
Phenanthrene
Number
of rats
  30
  10
  10
   5
                                                  Hydrocarbon concentration,
                                                          ug/gm + S.E.b
                                                  Mammary fat
 18.0 + 1.6
.29.1  + 4.0
 23.4 + 1.7
  1.8 + 0.1
  Levels of hydrocarbon were determined 24 hours after a
                                   Fat
39.2 +3.9
55.9 + 5.3
39.3 +_ 4.4
 2.8 + 0.3
  +. Standard  deviations.
                                    6-7

-------
produces mammary cancers.  A quantitative analysis of MCA in rat tissues
24 hours after its oral administration demonstrated that the hydrocarbon was
mainly localized in the fatty and breast tissues (Table 6-4).   In virgin
animals, levels of MCA were higher in fat than in breast tissues, whereas in
lactating rats this relationship was reversed.  Moreover, significant amounts
of MCA appeared in the milk at 24 hours after administration (Table 6-5).  On
the other hand, West and Horton10 reported that the transfer of ingested BaP
and MCA to the milk of rabbits and sheep was far less than that in rats; a
phenomenon which is probably common to all ruminants.
      Detailed studies  concerning the tissue distribution of tritiated DMBA
produced similar results as for MCA.11  Twenty-four hours after-receiving a
                           3
single oral dose of 30 mg 3HrDMBA at two levels of radioactivity (26 nd"  and
2600 jjCi), the greatest amount of activity recovered from rat tissues was in
the perirenal fat.  Moderate levels of radioactivity were also detected in
kidney, liver, and mammary tissue.  Radioactivity in the fat and mammary gland
persisted for at least 3 days and represented predominantly unmetabolized
DMBA.  In contrast, most of the radioactivity localized in the liver was due
to the presence of polar DMBA metabolites, some of which were bound to nucleic
acids and cellular proteins.
     The tissue distribution of 3H-dibenzo[g,c]carbazole (DBcgC) following
intratracheal instillation was qualitatively and quantitatively similar to
BaP.12  Except for high  local concentrations of DBcgC in the  lungs, the
greatest amount of radioactivity  was  localized in the intestines of treated
hamsters  (Table 6-6).
                                    6-8

-------
 Table 6-4.   LEVELS OF ORALLY ADMINISTERED 3-METHYLCHQLANTHRENE IN TISSUES  OF
                        50-DAY-OLD VIRGIN FEMALE RATST
       Tissue
Level of 3-methylcholanthrene,
         wg/g tissue3»b
 Breast,  rat  1
 Fat,  rat 1

 Breast,  rat  2
 Fat,  rat 2

 Breast,  rat  3
 Fat,  rat 3

 Brain, pooled
 Lung, pooled
 Kidney,  pooled
 Thymus,  pooled
 Liver, pooled
 Muscle,  pooled
 Heart, pooled
 Spleen,  pooled
 Uterus,  pooled
 Ovaries, pooled
Adrenals, pooled
 Pituitaries,  pooled
             15
             70

             12
             17

             24
             38

              3.5
              2.1
              1,1
              1.4
              0.3
              0.04
             NFC
             NF
             NF
             NF
             NF
             NF
  MCA was given as a single 30 mg dose in 1 ml of sesame oil.  Animals were
  sacrificed 24 hours later and levels of MCA in crude benzene extracts of
  tissue determined by fluorescence.

  Total content of 3-methylcholanthrene was less than 0.1 yg in pooled
  tissues of 3 rats.
  Not found.
                                    6-9

-------
00

^£
cc

g
O
Oi
— „

U*
o

f—
* y^
u.

0
ec

Y—?^
CO (O
ec CO
UJ 1—
.CO OH

1— « z
^^
Ul J—

5?
C£ U
1— —1
e| U.
_1 O

5 ec
_l U.
>••
a: o
UJ eC
1 •
CO (—
to
0 eC
LUl LlWF
a: oi
uj ca
I—
CO •>
*••• V^
z _i
»— < «— i
y ?s^
O
eC Z
f-^
JO
— i z

cc
o
u.
o

co

UJ
^^
UJ
._!

B
LO
1
10

CO

 f"" O* IO f^*
CM C5 CM O •— i—



t— CM

to (/)
s -a
CO CO
tin* (&_
*»*• **^
to i/i i/>
fO  CO ••-> f—
^ fO L* 

to
O CL
LO +J 4->
 Q.
re cn
S- C i/>

C [ * (Q
•r* (O "O
cn -i-»
s_ o r-»

> — J
p- E
§ LO
t- cr>
C CM
(O
•a co

•f*"
C CO
CO
•£ c
cn i-
CO
(/) ^3
IO
S s-
ai
i— CO
o E
o
CO S-
E 0
 O
«<- o
O J=
Q.
f— O
E ^;
f— U
CO
Cr*t
fci
•r- «n

C^ lO
c
O «r-
r—
•o
<4_ QJ
O N
CO P-
c re
CO C
s.  O)
C S-
re v

"o
J= CO
y • 3
i— S- (/>
>,  'P-
4-> A3 4-* •
0) r— -t->
t in o cn
co S- ••-
5 V} P™*
«4- 0 4J
o -c u c
re o
cn^* L* *p"
C CM 4J -t->
•p- X «
^o -o ai +J
QJ ai f-
a> o ai u
M- -t- CX
H— QJ QJ
QJ -p-. N
i— S- C 3.
cn u 
-------
  Table  6-6.
                                     Percent instilled dose recovered at
 Trachea
 Lungs
 Liver
 Kidneys
 Brain
 Intestines (large and small)
 Fat
 Urine
 Feces
  11 + 1.2
  63 + 2.2
1.4 + 0
0.0 ~
0.0
 11 + 1.4
0.0 ~
0.0
0.0
4.0 + 0.35
32 + 4.1
4.6 +0.40
2.0 + 0.36
0.7 + 0.06
31 + 2.6
0.6 +-0.15
0.6 + 0.07
8.2 + 0.12
3.0 + 0.34
18 +2.6
4.3 + 0.15
1.8 + 0.11
0.7 + 0.01
35 +5.8
0.6 + 0.04
3.2 + 0.13
15.6 +1.45
Total % recovered
                                 86.4+5.44     83.7+4.65   82.2+4.23


                            rtflM."9  H-d1benz°tc'93carba2ole suspended with
                                   6-n

-------
     Transplacental passage of POM is a critical  factor in determining the
risk for transplacental toxicity and carcinogenesis resulting from maternal
exposures.  The ability of BaP, DMBA, and MCA to cross the placenta in rats
was investigated following a single intragastric dose of 200 mg/kg of the
hydrocarbon as a suspension in sunflower oil.13  With DMBA, the concentration
of the carcinogen in fetal tissues reached a maximum (1.53 to 1.6 MS/9) 2 to
3 hours after maternal administration.  At 2 hours, the ratio of DMBA con-
centrations in maternal liver, placenta, and fetuses was 10:1.5:1.  Passage of
BaP into the fetus was even greater than for DMBA.  MCA, on the other hand,
crossed the placenta in only trace amounts.  These data are summarized and
compared in Table 6-7.  Intravenous injection of a 25 mg/kg dose of DMBA
produced similar fetal tissue  levels as the intragastric administration of
200 mg/kg DMBA.
      Limited studies have been conducted with human cells to characterize the
                                                                      14
pattern of cellular uptake and distribution of POM.  Ekelman and Milo
exposed cultured human neonatal foreskin fibroblasts to radio!abelled BaP and
DMBA.  Radioactivity was  first concentrated at the cell membrane, then random-
ly distributed  throughout the  cytoplasm, and ultimately (after  12 hours)
localized at the nuclear  membrane or  in the nucleus.  Their  results  further
indicated that  BaP was first bound to  a cytoplasmic receptor protein and
subsequently transferred  to the nucleus.   On the  other  hand, DMBA was absent
from  a charcoal-treated cytoplasmic fraction,  indicating  that protein binding
did  not occur,  and consequently DMBA was  not extensively  localized  in the
nucleus.  The  authors  postulated  that differences in  binding of POM to cyto-
plasmic protein could account  for the apparent resistance of human  cells  to
                                    6-12

-------
<
z to
1 1 i £23

3 «•
Q_
0£. «-H

LU CO
or a
a
"O
CO Q-
LU 5:
co o
2 <~)
LU LU
U. 0=
«c Li-
es 0
1
CM O
U- *~
r>^
1 1 1 t/}OO
CO Z LU
CO •— • CO
Pi?
^"* (->) p™
Z "^ 0
LU C£
«C 1— 1—
o u. co

O CO
z: a: >-
 CO
o
,«:z: CD
O- v^
(o co ^-*.
co en
ct'ec o
ca a: o
S C\J

Z U-
U- < O
o ^
ts -
^"^ LxJ
CD a:
^x. Q-
cn
— ^ 1 1
**- •* O

z; a:
O LU


1 ^™i

^v
1— LU
Z =C
LU 1—
•z. o
O Z
0

O)

»

•2 •







S-
01
— '








O in
a1
S. VI
^3r-
3 C
Z 10


4J in
'e_o

^^ 4->
IO
c t.

e
10 <*-








in
*>
O)
u
IO
oL













in
in
2
4J
U.











o'm
I- in
01 >>
= c



in
4-> in
•*• e
JE 0
^ *<&
c u

>
c
10 M-
o; o

O m

S. in
-Sr^
E ID
Z 10



•4-J in
•^* c
E O
•»~ •»—
r— 4-1

^3 «^
C &.

 o
o *->
C V)

u c
1. »^-
a c
o -o
n


v




I

^_
i
o

in
^™->
O
CO
co




*•




^^
g
1
CO
<3

oo
CM





**"




o
in
.
PO
1

vo

ir^
x^
ro
in





ct
CO

O





*r




O
CM
O
1
O

•
f—
"~"
g
CM




in




_,__
in
i
CM
CM

PO





LO




^>
Q
.
in
i

IO

fw
«!.«»
rv.
CM





d.
IO
CO





1C




CM
CM
o
1
n
O
C5
^•^
0
o




10



CO
o
0
1
in
OJ
u
s-
1 *


o
o
o





U3


^IH,
CM
O
O
o
1
in
O)
o
10

4->

cr,
8
CD




•a:

^r



6-13

-------
many carcinogens.  In support of this view, the authors cited unpublished
evidence that human diploid foreskin cells were successfully transformed by
BaP but not by DMBA.
     Theoretical studies on the uptake of POM by model membrane phospholipid
vesicles suggest a role for particle-mediated membrane uptake of POM as a
factor in carcinogenesis.15'16  Chrysene and BaP were shown to enter phos-
pholipid bilayers much more rapidly when adsorbed on silica or asbestos.  When
adsorbed to asbestos, BaP was transported into membranes more rapidly than
when adsorbed to silica, despite the fact that the asbestos sample had 60-fold
less surface area than the silica  sample.  The significance of these findings
was related to the possible involvement of environmental particulates in
aiding the transport  of POM into cell  membranes and  increasing the effective
concentration of carcinogenic material available  for microsomal activation  in
the cytoplasm.   Evidence was cited regarding the  apparent  interaction of  POM
in cigarette smoke with asbestos  inhalation in the production of  lung cancer
as support for  the  role of particulates  in chemical  carcinogenesis.
6.1.3   Elimination
     As  long ago as  1936,  it was  recognized that  various polycyclic  aromatic
 hydrocarbons were primarily  excreted through  the  hepatobiliary  system and the
 feces.17'18  With the advent of radiotracer methodologies, it became possible
 to study more quantitatively the elimination  of POM.3'19   Patterns  of    C-BaP
 excretion were determined in both intact rats and in rats  with  biliary  fis-
 tulas.3  When BaP was given intratracheally,  or by subcutaneous and intra-
 venous injection, maximum excretion of radioactivity always occurred through
                                    6-14

-------
  the feces (Tables 6-1 and 6-2).  In addition, the bile contained significant
  amounts of radioactivity, which was enhanced by increasing the administered
  dose of BaP until a saturation level was reached at 150 Mg BaP (intravenous
  dose).   Radioactivity was detectable in the bile for 24 hours following treat-
  ment,  and a total of 1 percent of the administered BaP was recovered unchanged.
  The total  recovery of radioactivity in BaP-treated rats having biliary fistu-
  las is  shown  in  Table 6-8.   The importance  of enterohepatic  circulation in  the
  excretion  of  BaP was  indicated by the  fact  that  urinary excretion of radio-
  activity was  reduced  from 7  to 14 percent of  the total  dose  in  intact  rats to
  3 to 4 percent of the  total  dose in rats with biliary  fistulas.
      The influence of  route  of administration on the rate of POM elimination
  Is likely to be an important determinant in quantitative studies.   Following
 the intragastric administration of MCA to rats,  82 percent of the dose is
 excreted in the feces within 24 hours.20  However,  when injected intraperi-
 toneally,  only 30 percent of the dose appeared in the feces after  3  days
 Assuming that  MCA is  readily absorbed through  the gastrointestinal tract (a
 valid assumption  based on the data of Rees and coworkers2),  it was apparent
 that rapid  absorption  and metabolism accounted for  the  lack of appreciable MCA
 retention.  It may thus be postulated that liver  "first-pass" metabolism  as
 well  as biotransformation by  enzymes of the  intestinal  mucosa represent an
 effective means of handling ingested POM's.  On the other hand, POM's reaching
 the circulation by absorption through the lungs or skin will not make an
 initial  passage through the liver and thus are more likely to reach internal
target organs in the parent form.  Since environmental exposures are  very
                                   6-15

-------
o

5
«£
cc
 a:
 tu
 to

                            10
                            4->
                            o
    CO
u.  _,

>-  c/>
 « 0)
OJ tn
^ o
                    u
                 >, at
                 $- •'-5
                 01 C


                 8Z
                 a> o

      —
  >•  o
                            co
            J-



            CM
                    en   oo
                    co   r^
                                                    \o    CM
                                                    cn    co
                               f*.    CO    CO
0»    00   O   CO
i—    i—   CO
                                           ,—    p>-   VD
                                           CM    CM   CO
  o


  1
  l-«
  K-

  CO
                       4-> S-  O
                        (O O) JC
                        t. a.
                        = x
           a>
              ai
   CO


   vo


   (U
                           CO
           O)
           I/I

           a
                      fsj    CTl   O    VD    <*">


                      i*!    co   co    CM    «n
                                       en   vo
                                       f-   CM
                                                       
-------
  important by inhalation,  it cannot be assumed that excretion of POM will  be
  sufficiently rapid to prevent their accumulation'in various  tissues.
       When administered by subcutaneous  injection  or by application  to  the
  skin  of female  mice,  BaP,  MCA,  and DBahA  are  eliminated almost  entirely by
  the feces.19 However,  the rate of disappearance  from  the site  of application
  or injection varies considerably  in  the decreasing  order BaP>MCA>DBahA.   In
  particular,  the elimination of  DBahA when applied to the skin is extremely
  slow, and none of the compound  or  its metabolites are  found in the feces.
  Poor dermal  absorption of DBahA probably accounts for  this observation since
 binding of metabolites to skin proteins was not extensive, nor is DBahA known
 as a potent skin carcinogen.
      Disappearance of BaP from the blood and liver of rats following a single
 intravenous injection  was very rapid.21   The concentration of BaP in the
 blood  one minute after a 10 pg injection was 193 ± 29 ng;  after  five minutes
 the concentration  of BaP in the  blood was  31 ± 1 ng.   Similarly,  in  the liver
 the half-time for  BaP  disappearance was  about  10 minutes.  In  both blood and
 liver, however,  the  initial  rapid  elimination  phase  was followed  by  a slower
 disappearance phase, lasting six hours or more.  In  the same experiment, dis-
 appearance of BaP  from the  brain was  slower than from blood or liver, and  the
 concentration  of BaP in fat increased during the six hour observation period.
 Schlede and coworkers21 concluded that a rapid equilibrium occurs for BaP
 between blood  and liver, and that rapid disappearance from the blood is due
 to both metabolism and distribution into tissues.  This contention is sup-
ported by data22 showing that pretreatment with BaP (which induced microsomal
enzyme  activity)  accelerates both the rate of BaP disappearance fr6m  all
                                  6-17

-------
tissues and the excretion of BaP metabolites into the bile.   The ability of
BaP to stimulate its own metabolism may have important implications for human
situations where lifelong exposure to POM is known to occur.   Furthermore,
in the pregnant rat pretreatment with BaP, MCA, chrysene, benz[a]anthracene,
or DBahA had a marked stimulatory effect on the tn vitro metabolism of a
                                                                    23
subsequent dose of BaP in maternal liver, placenta, and fetal liver.
     Elimination of BaP from the lung is generally regarded to be influenced
by the presence of carrier particles to which the hydrocarbon is adsorbed.
Most investigators have found that BaP retention by the lung is  increased when
administered with particulate material.6'24  Pelfrene25 administered single
intratracheal  instillations  (3  mg) of BaP to hamsters  either alone  or  in
combination with talc  (3  or  9 mg).   He observed  an increased retention time
for BaP  in the lungs of  hamsters  also receiving  talc  (Table 6-9).   The mechanism
for talc-induced retention of BaP was not explored,  although the possibility of
saturation of  the  macrophagic system by  talc was offered as an  explanation for
reduced clearance.
      However,  recent studies have also  indicated that intratracheal instilla-
 tion of BaP  to hamsters in combination  with ferric oxide did not increase
                                                        ?6
 carcinogen retention over that obtained with BaP alone.     Moreover,  it is now
 known that carrier particles are not an absolute requirement for the induction
 of respiratory tract tumors by BaP.27  Rather than affecting retention, it was
 suggested that carrier particles determine tissue localization and thereby
 tumor development.  A crucial factor in determining adequate lung retention of
                                     6-18

-------










H-4
§
_)
£
CD
Z
3
Ul
»—
•5
2.
=
z in
i-« CM
UJ i— i
0 <
£ =1
UJ I— I
0. Z
,«, _j
CO <
Ul
o 5
CO CC
fe 5
1 1
i 2

•
en
i
vo
V
uH
^


















• •
o
i:
Z>
O
^P
to
^
4J
«/» «/>
C 1-
•^ 3
1 O
in
o
Q.

IO
'C
o

W)
4J
g
3







•













CO
VD



m
cn



00





CM






VO






co





o






+J
c
^J
^~*
J_

0

ff
CM
.. I
•r 1
^>
0
co
•
0 +1
vo
*
o
CM

%Q >^
vo en
o CM
r^ tn

in in
CM


^A IhA
^^ c^
i-^ 10
+ 1 +1
in o
en co
CM VO


^_
i^ in
co in
+ 1 +1
r""» P««»
• •
vo en
in in

CM
CO CO
• *
CO CM
+ 1 +1
CO r—
en en
in vo


+
o. n_
*~+ *-^ u
to 10 r—
"co *oa -JS
en en en
SEE
co co co

O

0
CM
.^ i
+ \
CO
in
*5
•«d-
+ |
r«"
•
CM

feft
CM
CO
+ 1
^^
•
00
CO


vo
vo
vo
CM
VO



CM
in
in
vo


co
00*
+ 1
CO
en



a.

IO i—
^— to
CO •*->
en en
co en























•
-a
Q)
^"

£
Q|
^p
W)
 w)
 f
£ -^
0. E
01 ~O
t. 10
O 10
££
«.
*^
, 3 a>
o en
s» to
t. Q)
09 O
01
M O.
SO)
f
*f* ^J

ff| *
vo a!
•
E "^
S.^^
55
10 1-
•O^j
0^
0) 01
en £
«O -i-
i- +J
Is
>
CJ-T-

en a> i/i
•^ •£ C
U. -4-> t-
C r—
E 0)
o-o j=
*r* O) U
+J 4-> B3
10 0 S-
"> *J 10
09 fll ^

^^
^^~-"*~
10 
-------
carcinogens for tumor development is the concommitant exposure to agents which
inhibit ciliary activity.28  The tracheobronchial  epithelium is normally
protected by a mucous blanket and ciliated cells.   These cells propel  foreign
materials captured in the mucous flow upward and out of the respiratory tract
to be swallowed and excreted.  It has been known for many years that irritants
(e.g., S02, polluted air, cigarette smoke) can inhibit ciliary activity, thus
increasing the time available for elution of ROM's from soot particles in the
lungs.  This process can be expected to markedly influence the response to
carcinogenic POM.
     On the other hand, it should not be assumed that particles  cleared from
the  lungs  by ciliary activity are necessarily removed from the body.  These
particles  are  usually  swallowed  and thus present the opportunity for POM
absorption via the  gastrointestinal tract,  a process which  is  known to  result
in tumor  formation  (see Section  6.4).
      A thorough examination  has  been  conducted  on  the  clearance  of  intra-
tracheally instilled 3H-BaP  (5  mg total  dose) from various  tissues  of  the
 hamster.29  The effect of the presence  of adsorbents (asbestos,  carbon black)
was  also  studied.   Elimination  of radioactivity from the lung was found to be
 biphasic; an initial rapid phase lasted about 2 weeks and was not signifi-
 cantly influenced by the presence of adsorbents (Figure 6-1).   During this
 phase, 99 percent of the administered dose was  eliminated from the respiratory
 tract.  In the second (slow) phase of elimination, a significant retention of
 radioactivity became evident only among animals given 3H-BaP together with an
 adsorbent.  This accumulation of 3H-BaP-derived radioactivity in the lung,
                                    6-20

-------
   100
0.01
                                      •	• [3H1  3,4-Benzopyrene alone
                                      °—o t  H]  3,4— Benzopyrene + abestos
                                      •—^ [3 H]  3,4- Benzopyrene + carbon black
                                 14             21
                      Time in days from single intratracheal injection
28
  Figure 6-1.   Clearance  of radioactivity derived from
  [ H]3,4-benzopyrene in  the lung.29
                             6-21

-------
although small, was considered toxicologically significant.   The levels  of
radioactivity in other organs and in blood, urine, and feces were not appar-
ently influenced by co-administered adsorbents, but did appear to follow a
biphasic pattern of elimination (Figures 6-2 to 6-6).   However, blood and
fecal levels of radioactivity dropped off more rapidly in those animals  given
3H-BaP with carbon black.  These results (and those of numerous similar
studies) are presented with the assumption that the pattern of distribution
and elimination of the tritium label is an accurate reflection of the BaP
molecule and/or its metabolites.  In most cases, no attempt is made to iden-
tify metabolites or determine the extent of tritium loss into body water or of
bioexchange of the label.
     Excretion of intratracheal doses of DBcgC in  hamsters proceeds in much
the  same way as for BaP.12  The principal  route of excretion is via the
feces.  Retention of  DBcgC by the respiratory tissues was dependent on the
administration vehicle  (saline or water suspension), but in all cases was less
than for BaP.  Only 6 percent of an intratracheal  dose of DBcgC in saline
remained in the lung  after 18 hours, whereas with BaP, 25 percent of the
administered compound was still present after  18  hours.
                                    6-22

-------
           100
s
S
                                            I  H]  3,4— Benzopyrene alone
                                            (  H]  3.4- Benzopyrene + asbestos
                                            [  HJ  3,4- Benzopyrene + carbon black
                              6                14                21
                             Time in days form single intratracheal injection
23
            Figure 6-2.   Clearance  of radioactivity derived  fro.
            L HJ3,4-benzopyrene in  liver.
                                      6-23

-------
      100
•g
0.

X
C  03
eg  vl
Iff  O

k "^

3  g

*  o
*- o
3
o

1

-------
       10
 X
 V
 g
 •
 o'£
 S E
II
^^  0.1
     0.011—
•   • I H] 3.4-Benzopyrene alone
ff   ° I H] 3.4— Benzopyrene -t- asbestos
•—• [3H] 3,4- Benzopyrene + carbon black
                                     7             13
                           Time in days from single intratracheal injection
                                           36
        Figure 6-4.   Excretion of radioactivity derived from
        [ H]3,4-benzopyrene in urine.
                                  6-25

-------
                             t   t (3 H ]  3,4- Benzopyrene alone
                             a   0 [3 Hj  3,4- Benzopyrene + abestos
                             ^.. ^ [3 HI  3,4- Benzopyrene + carbon black
                           14             21
                 Time in days from single intratracheal injection
                                                        28
Figure 6-5.   Clearance of  radioactivity  derived  from
r3H-j3j4_benzopyrene in kidneys.
                            6-26

-------
        10
   :
 £ £
 •S*
:>*.   o.i
*3 O

g
O


1
cc
     0.01
            1    3
                                                •—• [  HJ 3,4- Benzopyrene alone


                                                0   ° I  H] 3.4— Benzopyrene + abestos


                                                *^+ 13 HJ 3 -4 - Benzopyrene + carbon black
                                              14                 21


                          Time in days from single intratracheal injection
28
           Figure 6-6.   Blood  level of radioactivity derived  from

           [ H]3,4-benzopyrene.

                                      6-27

-------
6.2  METABOLISM
     In the past, the relative lack of chemical reactivity for tumorigenie
ROM's has been puzzling in light of their dramatic biological effects.   Early
attempts to explain the carcinogenicity of various ROM's utilized physico-
chemical calculations.30  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 POM.
     More  recently  it was  learned that  POM  are metabolized via enzyme-mediated
oxidative  mechanisms to form reactive electrophiles.31   For  many of  the  POM,
certain "bioactivated"  metabolites  are  formed having the capability  for  cova-
 lent interaction with  cellular constituents (i.e.,  RNA,  DNA,  proteins) and
 ultimately leading to  tumor formation (see Section  6.2.6).
      The obligatory involvement of metabolic activation for the  expression  of
 POM-induced carcinogenesis has prompted the investigation of POM metabolism
 in numerous animal models and human tissues.   From these studies has emerged
 an understanding of the general mechanisms involved in POM biotransformation.
 It is  now known that POM are metabolized by the cytochrome P-450-dependent
 microsomal mixed-function oxidase (MFO) system, often designated aryl hydro-
 carbon hydroxylase.32'33'34'35  The activity  of this enzyme system  is readily
 inducible by exposure  to  chemicals and is  found in  most mammalian tissues,
 although  predominantly in the  liver.36,37,38,39,40,41,42  Jhe MR) system is
 involved  in  the metabolism  of  endogenous  substrates (e.g.,  steroids) and the
 detoxification of  many xenobiotics.  Paradoxically, however, the MFO system
                                     6-28

-------
  also catalyzes  the formation of reactive  epoxide metabolites  from certain
  POM,  possibly leading to  carcinogenesis in  experimental  mammals31'43"48
  (see  Section 6.4.2).   A second  microsomal enzyme,  epoxide  hydrase,
  converts epoxide metabolites  of POM to vicinal glycols,  a  process which may
  also  play a critical  role in  carcinogenic bioactivation.   Figure 6-7 presents
  a schematic representation of the various enzymes  involved in activation and
  detoxification pathways for BaP.  At present this  also appears to be repre-
  sentative of the general  mechanism for POM metabolism.
      A discussion of the metabolism of POM in mammalian species, including
 man, is best approached by examining in detail  the chemical fate of the most
 representative  and well-studied compound in  the POM class,  namely BaP.   The
 metabolism  of BaP has been extensively studied  in rodents,  and the results  of
 these investigations provide  useful  data which  can be directly compared to
 and  contrasted with the results  of more  limited studies  employing human cells
 and  tissues.  Therefore, separate discussions are presented below based upon
 the  available experimental evidence  regarding POM metabolism in  general, and
 BaP  metabolism in particular,  in both animals and man.
 6.2.1  Metabolism of  POM in Animals
     The metabolites  of POM produced by microsomal  enzymes  in mammals can
 arbitrarily be divided  into two  groups on .the basis of solubility.  In one
 group are those metabolites which can be extracted  from an aqueous incubation
mixture by an organic solvent.  This group consists of ring-hydroxylated
products, such as phenols  and dihydrodiols,49'50 and hydroxymethyl derivatives
of those POM having aliphatic side chains,  such  as 7,12-dimethylbenz[a]-
anthracene51  and  3-methylcholanthrene.52'53  m  addition to  the hydroxylated
                                   6-29

-------
 I
s
g
ca
                           (ENDOPLASMIC
                             RETICULUM)
                      GLUTATHIOIME
CYTOCHROMEP-450
MIXED-FUNCTION OXIDASE (MFO)
         BaP-
               -SG
      (DETOXIFICATION  TRANSFER ASE
       '  PRODUCTS)       (CYTOSOL)
                                    BaP OXIDES
                                                             BaP PHENOLS
                                          EPOXIDE
                                          HYDRASE
                                          (ENDOPLASMIC
                                          RETICULUM)
                                     sulfates
                                     glucuronides I
                                                             BaP QUINONES
                             MFO
                  BaP DIOL EPOXIDES
                 (PROPOSED ULTIMATE
                    CARCINOGENS)
                                    BaP DIHYDRODIOLS {PROPOSED PROXIMATE CARCINOGENS)
       UDP-GLUCURONOSYL TRANSFERASE
            (ENDOPLASMIC RETICULUM)
          H,O-SOLUBLE CONJUGATES
         (DETOXIFICATION PRODUCTS)
                Figure 6-7.  Enzymatic  pathways  involved in  the
                              activation and detoxification of BaP.
                                         6-30

-------
 metabolites  are quinones, produced both enzymatically  by microsomes  and  non-
 enzymatically by air oxidation of phenols.   Labile metabolic  intermediates
 such as epoxides can also be found in this fraction.43'44'45'54
      In the  second group of POM metabolites  are the water-soluble products
 remaining after extraction with an organic solvent.  Many of these deriva-
 tives are formed by reaction (conjugation) of hydroxylated POM metabolites
 with glutathione,  glucuronic acid, and sulfate.55'56  Enzyme systems involved
 in the formation of water-soluble metabolites include glutathione S-transferase,
 UDP-glucuronosyl  transferase,  and sulfotransferases.43'57'58  Conjugation
 reactions  are believed  to represent detoxification mechanisms, although this
 group of derivatives  has not been rigorously  studied.
      The metabolite profile  of  BaP which has  recently been  expanded  and
 clarified  by  the use  of high pressure  liquid  chromatography is depicted in
 Figure  6-8.   This composite  diagram shows  three groups  of positional  isomers,
 three dihydrodiols, three quinones, and several phenols.  The  major BaP
 metabolites found in microsomal incubations are 3-hydroxy-BaP,  1-hydroxy-BaP,
 7-hydroxy-BaP, and 9-hydroxy-BaP.  The BaP-4,5-epoxide  has been isolated  and
 identified as a precursor of the BaP-4,5-dihydrodiol.  Other studies  indicate
 that epoxides are the precursors of the 7,8-dihydrodiol and 9,10-dihydrodiol
 as well.
     Since the resonance properties of POM make ring openings difficult,
 enzymatic attack in the microsomes functions to open double bonds and add an
oxygen-containing moiety, such as a hydroxyl group, to give it more solu-
bility in aqueous media (e.g.,  urine)  and thus facilitate removal  from the
body.   In the  formation  of metabolic intermediates  by oxidation mechanisms,
                                   6-31

-------
[e-PHENOXY
|_ RADICAL
                               9,IO-epor|

                               tetroI]
                         7-OH
        	1	
           CONJUGATES
           GLUTATHIONE
         GLUCURONIC ACID
—T—	
 BOUND MACROMOLECULES
       DNA
       RNA
       PROTEIN
     Figure 6-8.   Oxidative metabolites of benzo[a]pyrene.
                             6-32

-------
  relatively  stable  POM are converted to unstable products  (i.e., epoxides).
  Thus, nucleophilic attack of this reactive intermediate,  through the formation
  of a transient carbonium ion, would be greatly enhanced.  Arylations of this
  type are common to many classes of carcinogenic chemicals.  Therefore, the
  microsomal cytochrome P-450-containing MFO system and epoxide hydrase play a
  critical role in both the metabolic activation and detoxification of many
  POM.
      Various forms of liver microsomal cytochrome P-450 can be isolated from
 animals  treated with different enzyme inducers.59'60'61>62  Moreover,  the
 metabolite profiles of BaP  can be qualitatively altered depending on the type
 of cytochrome P-450 present  in the incubation mixture.40'63  This  observation
 has important implications  in  considering the carcinogenic action  of certain
 POM toward tissues  from  animals  of different  species,  sex, age,  nutritional
 status,  and  exposure  to  enzyme-inducing chemicals.   Limited evidence is  also
 available  indicating  that multiple forms  of epoxide  hydrase  exist among
 animal species, which may also influence  the  pattern of POM  metabolism with
 respect  to carcinogenic bioactivation.62
     The metabolic fate of most  POM has not been studied as  extensively as
 that of BaP.  Nevertheless, several generalizations are evident which apply
 to most unsubstituted and alkyl-substituted polycyclics.   To a limited
 extent, direct attack on saturated carbon atoms may occur  to form, sequentially,
 alcohols, ketones, aldehydes, and carboxylic acids.100  More commonly,  meta-
bolic attack on one or more of the aromatic double bonds (K-region and non-
K-region) leads to the spontaneous formation of phenols or of isomeric dihydro-
diols  by  the intermediate formation of reactive epoxides.100?101   Dihydrodiols
                                   6-33

-------
are further metabolized by the microsomal monooxygenases to yield
diol-epoxides, compounds which are implicated as ultimate biologically
reactive intermediates.  Removal of activated intermediates by conjugation
with glutathione or glucuronic acid, or by further metabolism to tetra-
hydrotetrols, is a key step in protecting the organism from toxic interaction
                         34 102
with cell macromolecules.  '
     With unsubstituted POM, such as benz[a]anthracene and DBahA, oxidation
                                                       ^ ^-u ,1  A- n  34,103-105,530
occurs with rat liver homogenates to produce phenols and dihydrodiols.
Both K-region and non-K-region  (see Section 6.4.2) oxidative metabolites are
formed, which may subsequently  be conjugated with glutathione and excreted in
the urine as mercapturic acids.  In addition, hydration of the epoxide inter-
mediates formed from these compounds may yield  dihydrodiols-.  Benz[a]anthracene
is metabolized by rat  liver microsomes  in the presence of epoxide hydrase
primarily to  its 5,6-  and 8,9-dihydrodiols.530  Small amounts of the  3,4- and
10,11-dihydrodiols are detected, but only trace quantities of phenols are
formed.  It  is the 3,4-dihydrodiol  of benz[a]anthracene that is a proximate
carcinogenic  metabolite, as predicted by the bay-region theory  (see Section
6.4.2).531  The major  metabolite of DBahA  formed  by  rat  liver microsomes  is
the 3,4-dihydrodiol, which  is  also  the  immediate  precursor  of the carcinogenic
diol-epoxide of DBahA  as predicted  by the  bay  region theory (see  Section
6.4.2,  Section 6.4.3.4, and Section 6.4.5).532  Evidence indicates  that  K-region
epoxides are better substrates for  the  formation  of glutathione conjugates
 than non-K-region epoxides.
      Among the simpler unsubstituted ROM's, such as phenanthrene,  no evidence
 for the formation of a K-region phenol  or a conjugate thereof can be detected.
                                    6-34

-------
 The major  phenanthrene  metabolite  is  a  K-region  dihydrodiol which may  also  be
 excreted as a conjugate with  glucuronic or  sulfuric  acid.
       In contrast to the unsubstituted compounds,  POM-bearing alkyl substitutes
 (e.g., MCA, DMBA) are primarily hydroxylated at  the  alkyl side chain.  The
 metabolism of DMBA has  been carefully studied using  rat liver homogenates.106'107
 These studies clearly showed that the main products  of DMBA metabolism are the
 isomeric 7-hydroxymethyl-12-methylbenz[a]anthracene  and 12-hydroxymethyl-7-
 methylbenzanthracene.    In addition, DMBA and its hydroxymethyl  derivatives are
 metabolized into 3,4-  and 8,9-dihydrodiols (non K-region).   In  contrast, a
 K-region 5,6-epoxide is also formed, but instead of being hydrated to the
 dihydrodiol,  it is  conjugated with glutathione.106  Although all  of the polar
 DMBA metabolites have  not yet been identified,  no derivative has  thus far been
 detected which  involves  enzymic oxidation at both the 5,6-  and 8,9-bonds in  the
 same molecule.   Recent studies have implied  that  metabolic  activation of DMBA
 for binding with DNA occurs  through the  generation of a diol epoxide  in the
             1 flft
 1,2,3,4-ring.     Ring hydroxylation at  the  3- and 4-positions of DMBA  (M-regiori)
 had previously been shown  to occur.109  Thus, evidence for the "bay region"
 theory being of  general  applicability  is increasing.
     Metabolism  of the carcinogens  7-  and  12-methylbenz[a]anthracene  is
 qualitatively similar to DMBA.109  The products formed by rat liver homo-
 genates include:  dihydrodiols at the  5,6-(K-region)  and 8,9-(non-K-region)
 positions; hydroxymethyl derivatives;  various phenols; and glutathione con-
 jugates at the K-region.  Figure 6-9 presents a probable scheme for the
metabolism of 7-methylbenz[a]anthracene which would also apply to 12-
methylbenz [a] anthracene .
                                   6-35

-------
                                                             CH2 — OH
                                           NH —Glo
Figure 6-9.   Probable metabolic pathways of 7-methylbenz[a]anthracene
                                                                       109
                                  6-36

-------
       The metabolites of MCA formed by rat liver homogenates, liver micro-
  somes, or a highly purified cytochrome P-448-dependent monooxygenase system
  with epoxide hydrase are all  similar.53'1™  Formation of 1-hydroxy MCA and
  2-hydroxy MCA predominates.   Smaller amounts of the K-region trans 11,12-
  dihydrodiol  and a glutathione conjugate  derived therefrom are formed from
  MCA,  as  well  as trans  1,2-dihydroxy  MCA  and  several  phenols.   The  two major
  dihydrodiols  which  are produced by the action of epoxide  hydrase on  (±)  1-
  hydroxy  MCA are the  diastereometrically  related  trans  9,10-dihydrodiol.
  These  results  led Thakker and coworkers53 to postulate that the ultimate
  carcinogenic metabolite of MCA may be the "bay region" diol epoxides  (9,10-
  diol-7,8-epoxides) derived from 1-hydroxy MCA.
      An  important consideration in evaluating the health hazards of POM is
 whether metabolism in rat liver preparations is  indicative of the pattern of
 POM metabolism in more likely target  organs  (e.g.,  lung)  (see Section 6.2.3).
 Moreover, it is vital to  determine whether metabolism in  rodents mimics
 metabolism in  man (see  Section 6,2.4).  In this  regard, it has been shown
 that the  metabolism  of  benz[a]anthracene,  7-methylbenz[a]anthracene,  and  BaP
 was  qualitatively, similar  in rat liver and lung.111'11* Both  K-region and
 non-K-region dihydrodiols were detected,  which almost certainly  arose  from
 epoxide precursors.
 6.2.2   Metabolic  Bioactivation of Benzo[a]pyrene
     The  role of  metabolic biotransformation in conferring carcinogenic
activity  for POM  has been clearly established using BaP as a model compound.
While it  is known that BaP is metabolized by the  MFO system to various arene
oxides, phenols, dihydrodiols,  and diol epoxides, it is the latter type of
                                   6-37

-------
                                                                          47,71
                                                 .   ...       54,58,64-70
intermediate which has the greatest toxicologic significance.
The BaP-7,8-diol-9,10-epoxides are derived from BaP-7,8-
dihydrodiol and exist as a pair of diastereomers in which the 7-hydroxyl
group is either cis or trans to the 9,10-epoxide moiety (Figure 6-10).   Each
of the diastereomeric BaP-7,8-diol-9,10-epoxides can be further resolved into
a pair of optical enantiomers, the relative amounts of which depend on the
stereoselectivity of the final oxygenation step catalyzed by the MFO system.
     In liver microsomes from rats pretreated with MCA, BaP is stereoselec-
tively metabolized to the  (-)-enantiomer of BaP-7,8-dihydrodiol and to (+)-
BaP-7B,8crdiol-9a,10crepoxide.  These metabolites possess high mutagenic and
carcinogenic activity compared to the parent compound and other BaP metabo-
lites,67j68>72>73 and are  believed to be proximate  and  ultimate carcinogenic
intermediates  of BaP, respectively (see Section 6.4.2 and Section  6.4.3.1).
In particular,  the  (+)-trans-BaP-7,8-diol-9,10-epoxide  was  shown to be the
                                                                       73
most carcinogenic of the four BaP-diol  epoxide isomers  in newborn  mice,   the
                                                                72
most mutagenic of the four isomers in  Chinese  hamster V79 cells,   and  is the
 isomer which stereoselectively binds to the  exocyclic  ami no group  of  guanine
 in double-stranded calf thymus DNA.74  Therefore,  evidence  is very strong to
 indicate that (+)-BaP-7B,8ordiol-9a,10a-epoxide is the major ultimate car-
                                                                 73
 cinogenic metabolite of BaP.  Furthermore, Buening and coworkers    have
 suggested that (+)-BaP-7B,8crdiol-9a,10a-epoxide may be the most potent
 compound ever tested for  the induction of pulmonary tumors  in mice.
      Factors affecting the  direction of BaP metabolism have obvious toxico-
 logic significance in determining its  carcinogenic activity in various
tissues, individuals, and species.   In this regard,  Deutsch and coworkers
                                                                         71
                                   6-38

-------
( _ ) _ BP 7 ,8 OIHYDRODIOL
  OH

( + » - BP 7 ,8 - DIHYDRODIOL
                                      (-) - BP - 7^, 8a- OIOL - 90,100 - EPOXIDE
                                      ( + )-BP-7(3.8a- DIOL - 9a. 10a- EPOXIDE
                                      (+) - BP - 7 a. 30 - DIOL - 9a, ioa - EPOXIDE
                                               OH
                                      (_)_BP-7a.80-DIOL-90.100-EPOXIDE
 Figure 6-10.   Stereochemical  course  of metabolism of  (+)
                 and  (-)-BaP 7,8-dihydrodiol.
                               6-39

-------
and Thakker and coworkers47 have shown how the various forms of cytochrome  P-
450 from rabbit and rat liver microsomes influence the regiospecific and
stereoselective metabolism of BaP.  Different forms of cytochrome P-450 had
marked effects on the conversion of BaP and its (-)-7,8-dihydrodiol to diol
epoxide metabolites which bind to DNA.  The balance of activities among the
various cytochromes vis-a-vis exposure to microsomal enzyme inducers was thus
considered to be an important determinant of carcinogenic susceptibility.
6.2.3  Tissue-Specific Metabolism of  POM
     Metabolism studies have historically focused  on  the liver as the principal
site for  xenobiotic transformation.   Recently, a considerable research  effort
has been  directed  at  the  characterization of  POM'metabolism in those tissues
which  are most susceptible to malignant transformation, particularly the
respiratory tract.  The objective of  this work has been to  correlate tissue-
 specific patterns  of  POM  metabolism with  the carcinogenic activity of  a
 particular substance  in that organ.  In addition,  the actions  of various
 enzyme inducers in altering metabolic pathways have been  examined to determine
 how various xenobiotics can modify the bioactivation of carcinogenic POM.
      The cytochrome P-450 content and MFO activity of the lung are lower than
 in the liver.75  However, the activity of conjugating enzyme systems and
 epoxide  hydrase in the lung is very  low, and thus may be critical in deter-
 mining susceptibility of  this organ  to carcinogenic  POM.    Studies on the
 metabolism  of BaP in the  isolated  perfused rabbit lung revealed that arene
 oxide intermediates  formed by the  oxidative metabolism of  BaP were further
 enzymatically transformed by epoxide hydrase  and  glutathione S-transferases.
                                     6-40

-------
  The removal of arene oxides by hydration and conjugation, however, proceeded at
  a rate which was an order of magnitude less than the overall rate of meta-
  bolism.
       Pretreatment of rabbits with MCA, a microsomal enzyme inducer,  had no
  effect on the metabolism of BaP by microsomal  preparations,  nor did  it increase
  cytochrome P-448 in pulmonary microsomes.   In  the isolated perfused  rabbit
  lung,  pretreatment  with MCA increased only  the  epoxide  hydrase  activity,  and
  consequently the  rate of production  of dihydrodiol  metabolites.
      In contrast  to the metabolic  activity  of the rabbit  lung toward POM,  the
  lungs  of  rats and mice  respond to  pretreatment  by enzyme  inducers with a 4- to
  25-fold increase  in BaP  hydroxylase activity.^  Recent studies by Vahakangas
  and coworkers76^ compared the ^f^ Qf pretreatment ^ ^ ^ ^^^
  bitone on the metabolism of BaP in isolated perfused rat lungs.   Pretreatment
 with MCA enhanced both the rate of BaP hydroxylation and the  activity of
 epoxide hydrase.   This was evidenced by an increase  in measured  rates of BaP
 hydroxylase enzyme activity and enhanced production  of dihydrodiol  metabolites.
 In  addition,  MCA  pretreatment stimulated conjugation reactions involving  BaP
 metabolites.   Accompanying the  enhancement of BaP metabolism  by  MCA pretreat-
 ment was an  increase in  covalent binding of  BaP  metabolites to lung tissue.  A
 similar increase in  covalent binding was produced by exposing rats to cigarette
 smoke,  and thus implicated enzyme induction by environmental agents as being
 toxicologically significant in the  lung.
     Although pretreatment of rats with phenobarbitone slightly increased the
 rate of BaP hydroxylation in rat liver and kidney (Table 6-10),  it had the
opposite effect in the rat lung.76'77  Moreover,  the  activities of epoxide
hydrase  and the conjugating enzymes  in rat lung were  not  enhanced by
                                   6-41

-------
     Table 6-10.
BaP-HYDROXYLASE ACTIVITY IN DIFFERENT ORGANS OF THE
RAT AFTER METHYLCHOLANTHRENE AND PHENOBARBITONE PRE-
TREATMENT7'
Pretreatment
       BaP-hydroxylase (fluorescence units/g x min.)
       Perfused lung       Liver               Kidney
3-Methy 1 chol anthrene
Control
388 + 55*
114 ±31
26982 i 4957*
5527 + 1392
258 t 85*
38 + 13
Phenobarbitone
Control
54 + 9
82 + 19
4872 + 2048
3674 + 909
21+4
11+5
  Significantly different from control  (P<0.05) by Student s t-test.
The values represent means + S.E. of  6  animals  in each group.  The livers
and kidneys were removed from the same  animals  as the lungs for perfusion.
                                      6-42

-------
  phenobarbitone  pretreatment.   Previous  investigators,  however,  have  reported
  epoxide  hydrase to  be  inducible  by phenobarbitone  in the  lungs.54
      Since the  tracheo-bronchial mucosa  is the  respiratory target tissue for
  ROM-induced cancer  in  both animals and man, a number of studies have focused on
  the metabolism  of BaP  in rodent  trachea.  The importance of enzyme inducers in
  determining the  fate and toxicologic significance of POM in the trachea was
  illustrated in  studies by Simberg and Votila.78  They reported that rats
  receiving daily 1-hour exposures to cigarette smoke for one or 10 days
  displayed an increase (290 to 320 percent) in MFO activity in the trachea.
  Furthermore,  a two-fold increase in covalent binding of BaP metabolites to
 nucleic acid and protein fractions  of the trachea occurred 12 hours  after a
 single  cigarette smoke exposure.  No  changes  were detected In the  activity  of
 epoxide hydrase  in  the trachea, and only a slight increase in UDP-glucurono-
 syltransferase in the trachea and lungs  resulted from repeated cigarette  smoke
 exposures.  Enhancement of MFO activity  by cigarette smoke  in  rat  lungs was
 nearly  double  the increase observed in the trachea.
     Moore and Cohen   conducted  an extensive study  on  the metabolism of BaP
 and its major  metabolites in  cultured trachea of rats and hamsters.  The
 predominant ethyl acetate-soluble (Phase  1) metabolites of BaP in the trachea
 of both species  were the 9,10-dihydrodiol  and 7,8-dihydrodiol  (Table 6-11).
 Smaller amounts  of the 4,5-dihydrodiol, monohydroxybenzo[a]pyrenes, benzo[a]-
 pyren-3-yl hydrogen  sulfate, and  tetrols were also found.   Much of the 4,5-
 dihydrodiol and monohydroxybenzo[ajpyrenes produced from BaP was rapidly
 conjugated and thus  removed to the water-soluble metabolite fraction.   This
explains previous observations that  only small  amounts  of 4,5-dihydrodiol  and
monohydroxybenzo[a]pyrenes can be found as organic solvent-soluble  metabolites
                                   6-43

-------
C3
35
O
co
LU
jr-.
cpi^-
cons
«t  •
H-co
Ul LU
_JO
ca ca
=j<:
d£
cos


^e

s§
o
z: a.
cc ca
UJ U.
-10
ca

II



  0)


  JQ
  (O


(/)

J^
•fMt

"o
J3
|


0)
S
3

"o
to

J_
0)

i







to
0)
•r-
1
4J

yf;

o>
2
3

*O
to
i
0)
S
0)
U

f«—
>»
.c
4_>
LU




to
0)
{j
0)
Q.
CO
-a
01
•at-
IS
a. ta
E •+•>

/ s g?
i-»
•\JD
1 r- 0)
10 O C
1 »»i- O)
co ^* "o t«
0 =->
<4-d3 s- a.
o «-o»— *
0) >> as
to c •£= — •
a> a> T- o
•M S--O IM
> i c
en O.CO a>
f» ^_ ^ j^
o c « o
o g»5^

0) 0) T- >>
^O ^Q ^O ^C
i- >> On-
c x s- "o
o o-o i
s- s- >>o
3-0 -Cr—

3 .e "O en
5
i
CO
•> 1
r- 1 O)
0 C C
O ^^ ^>
e X OB
•a o*~*
*r^ >•* O
*T3 JC N
1 O C
CO C OI.Q
• O -Q 0)
-o «
JD - sm-
•«r— «O r—
i—O 3
O i— "to
T-"O U)
•a o a) c
0 S- C 0)
j_-a a) o>
•o >> x- o
^^ct-fe
t-^0— • >>
a i co J=
i tr>«-^
O^Or^

M
en




j_
S
to
E
£




a.
CO
o
s_
-o
^Ih
.c
1
co

«*-
0
•S
(O
•y
T2
O 0)
o ^

a) a.
^j ^^^
»!•• fQ

O O
t. N
3 C
U 0)
3 -Q
3

JD
at
4J
£
3
a>
at
o
s-
•o
<—

r—
>»
CO
1
1
^:
n>
o
N
c
0)
CD-
S.
to
S
ta
•o
(O

1)1*
cl


a.
ca
o
i
CO
                                       I
                                      O
                                                     00
                                         a>      a)
                                     q-  C   M- C   «»-
                                      O  0)    O 0)    O

                                      a>  >>    a) >>    o;
                                     4J  C3.   4-> Q.   4J
                                      (O*~*    itj'—*    10
                                      en ta    01 >
                                        x
                                                IM
                                                 =    l_    §,_
                                                5    Olo      "o
                                                      «,.    «5
                                                                o
                                      o-b    p-p    p-p    p-o
                                ^t   -o x"   -o  x1  -o-o   -o
                                 ta    T~ o    f-o   «^o   -^
                                J2.   ci.    ci.    ct-    c
                                      O •"—    O *^"    O •»—    O •»"
                                      3 T3    3"O    3 "O    3-O


                                      C3      CD      CD      CD
                                                                     M-    O C    U
                                                                           s- o    o>
                                                                     O    3t-    Q.
                                       fl)

                                       at


                                       I
                                       o
                                       IM

                                       0)
                                       ja

                                       x1
                                       o
                                        •»
                                       en
         to
         to
         o

a       ,3

"o      "^


 0)      T-
                                              £
                                             •o
                                              (Q
                                              en
                                               •b

                                              00
                                                       O
                                                      
                                                                       to
                                                                       a.
                                                                       rO

                                                                       O>
                                                                       O
                                                                        O
                                                                        s_
                                                                            s-
                                                                            O
                       D'l-    O
                         E    **""
                         
                                                                            •a
                                                                            o-o
                                                                           J- to

                                                                           JBJ5
                                                                           -t-> ca
                                                                           v- O>
                                                                           ai 3
                                                                           £8

                                                                            *«
                                                                            M -M
                                                                           ^- (O
                                                                                  •o

                                                                                   (O
                                                                                   O

                                                                                  T>
                                                                                    01

                                                                                    0)
                                                                                    U

                                                                                    a>
                                                                                    u
                                                                                    to
                                                                                    a>

                                                                                    o
O  «n   r~
                                                                             O>
                                                                             CO)
                                                                             t- c
                                                                             X O
                                                               01 4J

                                                               ~~  0)
                                                      CO      T-
                                                       to
                                                       i
                                                a>
                                                               to
                                       cl
                                      -o
                                       o
                                        o

                                        o
                                                o
                                               •r—
                                               •O

                                                £


                                               1
                                               •r-
                                               O

                                               O


                                               en*
                                                       -o
                                                        o

                                                       •o
                                                      a

                                                      co
                                                        ««
                                                      r*.
                                                               o

                                                              •a
                                                               o
                                                                i
                                                               LO
                                                                        to

                                                                        S

                                                                        0)
                                                                        •o
                                                                        O)
                                                                        0)
                                                                        •o
                                                                            1-"
                                                                            «JI -a
                                                                                    «=
                                                                                    O
                                                                                   "S
                                                                            •Su   T:
                                                                            •^" «r"  • C
                                                                               t_0 0)
                                                                            0) CO —I'D
                                                                             5li>,o
                                                                                ca .a to
                                                                             to to    r—
                                                                             0)  I  to CO
                   a>  . 0}  0)    0) to
                   £ to E  to o» -i-> -a

                   o) E    JSS^I !o

                     +J r—  CO 3 OX!
                   V) to Jl  «*- r— XI C
                   > 3  r— O CO CO
                   4J lo^  3 tO-tJ -U
                   n-    o  to  i  o> to
                   r- +J tf|  r— O) E „
                   O t=  I  >>-t->   JJ
                   ja o) s-  s- « <" -^
                   CO > 0>  CO •*-> (O •«->
                   ^ r- -t->    0) O>
                   0) O jg  1- U •*= O

                  ct? "xT"      o
                                                 6-44

-------
in cultured human and rodent respiratory tissue.80  Metabolism of the
7,8-dihydrodiol and 9,10-dihydrodiol yielded the corresponding tetrahydrotetrols
via their respective diol-epoxides.  In addition, a major metabolic pathway
for the 7,8-dihydrodiol was to the water-soluble glucuronide conjugate, whereas
almost no glucuronide conjugate was formed from the 9,10-dihydrodiol.
     The presumed importance of the BaP 7,8-dihydrodiol as a proximate car-
cinogen is a primary reason why tissue-specific patterns of metabolism may
ultimately determine target organ susceptibility to BaP-induced cancer.  As
suggested by the work of Moore and Cohen,   an alteration by environmental or
genetic factors of the activities of the conjugating enzymes can affect the
levels of 7,8-dihydrodiol, and consequently the formation of reactive diol
epoxide intermediates.
     Investigation of the metabolism of BaP in rodent trachea has been
extended to include cultured human bronchus as well.  In a series of experi-
ments it was shown that human bronchial mucosa contains MFO activity that is
readily inducible by pretreatment with POM such as benz[a]anthracene and BaP,
although large interindividual variations apparently existed.80  On the other
hand, the activity of epoxide hydrase was not affected by benz[a]anthracene.
Enhanced MFO activity in human bronchial mucosa was reflected in an observed
increase in BaP binding to DNA following pretreatment with BaP or
benz[a]anthracene.  One of the major metabolites of BaP that binds to DNA in
human bronchus was tentatively identified as a diol epoxide enzymatically
derived from the (-)-trans-7,8-dihydrodiol of BaP.  Therefore, these results
in human respiratory tissue support the data from similar studies in animals
regarding the importance of tissue-specific MFO activity in the bioactivation
of POM to carcinogenic intermediates which bind to DNA.  More important, these
                                   6-45

-------
studies strengthen the argument that certain POM can induce bronchial  cancer
in humans.
     Among the non-hepatic tissues investigated for their capability to
metabolize POM, the intestinal mucosa has received considerable attention.
Although the MFO activity of the intestinal tract is low when compared to the
liver, it is nevertheless a site of ROM-induced cancer in animals.  Bioactiva-
tion of POM by intestinal cells would,  therefore, be expected to  occur in all
susceptible species,  this contention was  recently  confirmed by studies  showing
that susceptibility to  BaP-induced forestomach tumors  in mice was correlated
                                     81
with MFO  activity in  the same tissue.
     Another  group of workers82 has examined BaP metabolism by  microsomes  and
 isolated epithelial cells from rat small intestine.  Their'results agreed  with
 those  obtained by previous investigators showing that cytochrome  P-450 content
 and MFO activity are normally low in the intestinal epithelium.   However,  MCA
 displayed a marked stimulatory effect on both cytochrome P-450 content and MFO
 activity.  In contrast to the situation with  rat liver, neither phenobarbital
 nor pregnenolone-16a-carbonitrile caused substantial increases in MFO activity.
 Their results further  demonstrated that the epoxide hydrase activity of rat
 small intestine was  very low  and  non-inducible.  This was evidenced by  the
 fact  that relatively little  dihydrodiol formation  resulted from  the incubation
 of BaP with  microsomes from  intestinal cells  as  compared  to liver cells from
 MCA-pretreated  rats  (Table 6-12).
       Fang  and Strobe!83'84  have recently  demonstrated that rat colon  mucosa
  possesses  a microsomal drug-metabolizing  enzyme system.   They  evaluated the
  ability of colon 9000 x g supernatant (S9) fractions to activate BaP to
                                     6-46

-------
QC
UJ
  CM
toco
OC OO
UJ t—

£s
Q. O
   UJ
I—I LU


O I—
CO  I
«C O
LU a:
z u.
UJ
ce. is)
>- UJ
a. 2:

 (O oo
t—iO
oee
eo

U_ e£
O Z
O 00
OO LU
s: o

s<


CM

 I
IO
             OJ
             (/>
             
             >> OJ
             10 i—
             I—I O
             O J3
             N UJ
             OQ E
                     f^Of— ft i—
                                                      O
                                                      O
                     uo o LD in o o LD

                     i-^ CO r-^ O C3 CM rt
          in o in

          *o «^ evj
             o
             o
                     •a
                     2
                     •a
                     •a

                     o
                     IB -i- O  O
                     JZ -O -i- -i-
                     •M O T3 -O       0)
                        s_ o  o  a> a> c
                     i. -a s.  s.  c c o
                     a >»T3 -a  o o c
                                         a> >, >>
                      o T- .c jr -r- «^ 3 T- o  o
                         i ~a -o
                      > >> O)

 "I  I  •»->
                                             6-47

-------
metabolites which were mutagenic in the Ames Salmonella test system.
Activation of BaP was examined using colon S9 preparation from rats treated
with various enzyme inducers, and the results compared with those obtained
using corresponding liver S9 preparations.  Their results, summarized in
Table 6-13, indicated that an approximate doubling of the mutation rate
occurred with addition to the test system of colon or liver S9 preparations
from uninduced rats.  Pretreatment of rats with p-naphthoflavone induced a
four-fold  increase  in the activation of BaP by the colon, whereas pheno-
barbital/hydrocortisone and  Aroclor 1254  had no effect.  These results
suggest a  role for  POM bioactivation by the MFO system  in colon carcino-
genesis, a theory supported  by  the observation that cultured  human colon can
                                                          OC
activate carcinogens  leading to covalent  protein  binding.     Furthermore,
these  results  indicate that  dietary constituents  having the ability  to
 induce or  inhibit MFO activity  in the  gastrointestinal  tract  may  have a
 strong influence on individual  susceptibility  to  colon cancer.
      In any mammalian tissue, differences in the  quantity and/or  structure
 of reactive intermediates of POM formed by the MFO system are clearly  re-
 flected in the degree of susceptibility to neoplastic transformation.
 Genetic factors affecting the delicate balance of activating  and detoxifying
 enzymes produce differences in  the metabolism  and binding to  DNA of carcino-
 genic POM.85'87  These factors, in combination with diet and simultaneous
 exposure to non-carcinogenic chemicals, must be considered in evaluating the
 carcinogenic potential of POM and in explaining observed differences in
 response among species and  target organs.  In addition, the sex-dependent
 regulation of POM metabolism has been demonstrated for certain rat tissues,
 and thus adds still another dimension to the interpretation of experimental
 results.
88
                                     6-48

-------
                                                         0)
      I
      i— JT  OJ
      -4->  o  S-
i-     O  ia  Q.
•M  O>  IB LU
0)1-

5;  2
o.  2 •<->   • -u
                                                 6-49

-------
6.2.4  Comparative Metabolism of POM in Animals and Man
     An important consideration in evaluating the health hazards of POM is
whether metabolism in various animal tissues and species is indicative of
the pattern of POM metabolism in the target organs of humans.   Moreover, it
is essential to determine whether differences occur in the metabolism of POM
by:   (a) different tissues in the same animal; and (b) different animals of
the same species.
      Numerous  studies  have shown that  qualitative  and  quantitative differences
                                                           .   .      .    34,89-9:
exist in the metabolism of BaP  by different tissues and  animal  species.
For the most part,  however,  interspecies extrapolations  of
 qualitative patterns of POM  metabolism appears to  be  a valid  practice.   On
 the other hand,  marked differences  in patterns of tissue-specific metabolism
 may prevent the reliable extrapolation of data from hepatic to extrahepatic
 (i.e., target organ) tissues.  These differences may also exist in human
 tissues, possibly due to different levels of drug-metabolizing enzymes
 present, and differences in responsiveness to enzyme-inducing agents in
                 94
 specific organs.
       Nevertheless,  it has been adequately  demonstrated  that  human cells
 possess the mechanisms to bioactivate POM  to  carcinogenic intermediates.
 When added to cultures of human skin  fibroblasts,  BaP caused an  induction of
                                                                         95
 MFO  activity, increased cellular proliferation,  and  produced DNA damage.
  On the other  hand,  MCA was  relatively ineffective, while DMBA was capable of
  inducing MFO  activity and cell proliferation, but did not produce DNA damage.
  In addition,  noncarcinogenic POM such as anthracene and phenanthrene had  no
  effect on the human cells.   The fact that neither MCA nor DMBA could be
  detected in the cell nucleus  implies that in human cells the necessary
                                     6-50

-------
  metabolic sequences for bioactivation of POM carcinogens may not operate in
  all  tissues and for all  such compounds in the class.   This may explain the
  failure to induce neoplastic or morphologic transformation with POM treat-
  ment iji vitro  of randomly proliferating human cell  populations.
       Freudenthal  and coworkers96 recently examined  the  metabolism  of BaP  by
  lung microsomes  isolated  from  the rat,  rhesus  monkey, and  man.   Metabolite
  profiles  obtained by high  pressure  liquid chromatography are shown  in
  Figure  6-11.  Their  results  confirmed previous observations regarding  the   '
  existence  of considerable  individual variation in BaP metabolism among
  samples from the  same species.   In addition, it was apparent that qualitative
 and quantitative  inter-species variation also existed (Table 6-14).  Never-
 theless, the qualitative differences between man and the other animal species
 were by no means dramatic, and probably do not compromise the validity of
 extrapolations  concerning POM metabolism.
      In other investigations  of BaP metabolism by human  lung microsomal
 fractions, it was noted  that  a higher percentage  of  dihydrodiol  metabolites
 is  formed in the lung of  humans as compared to the rat."   This  result may
 reflect differences in relative levels  of  epoxide hydrase in human  versus
 rat lung,  and can  be  expected to  influence carcinogenic  susceptibility.  The
 authors  felt that  the relative  insensitivity of the  rat  lung to  POM-induced
 cancer  supports this  conclusion.   However,  large  variations in metabolite
 profiles formed by the 15 human specimens employed in this study indicate
 that generalizations cannot yet be stated regarding the metabolic activity
of human lung.
     The metabolite pattern obtained for BaP in human monocytes  is similar
to that obtained with human liver microsomes45 and human  lymphocytes.97
                                   6-51

-------
        _ 2S
        3 20
tlU

*
              '  1  '
                                                  BP
                             3.6-
                             DIONE

                        i'.8"...    6.12-OIONE
             •9.10-DIOL
         •• °0   M  40  60  80  100  120  140  160  180  200 220

                                FRACTION
           18

           16 -

           14
         S
         i10
          §
          <=  4
I  '  I
               '_ 9.10-DIOL
               3.6-
       7.8-     OIONE   9_
  4,5-  DIOL  1.6-  6,12-OH-BP
            DIONE|\ OIONE

                    s«_
                     '  •  I
             0 o   i '  40   60   80  100  120 140  160  180  200  220
                                 FRACTION
                   20   40  60  80  100  120 140  160  180 200  220 240
Figure  6-11.
Comparative  metabolism of benzo[a]pyrene  by
lung microsomes  from rat, rhesus  monkey,  and
human  '°
                                  6-52

-------
  oo
 oo
 ui
 §
 s
   en
 o_ oo
 03 >-

 u. oo
 Ooo

B-S
UJ CD



CO 2=


£1

 •
*!•

 I


CU

ro
    O
   U3
    n
 to
 a)  cu
 e>0 E
 (C
 01
 o
 (-1
 0)
 Cu  CO
    0)
 01 i-H
    a

 ~  oi
 BJ 4J
 4-1 1-t
 -l
2  O
                    C
                    CO








o
vO


i-H
t-»







en vo
in CM






O vo
en  C
•H O
T3 -»H
C 4-*
•^ CO
C
0) -g
CO )-<
CO 4J
C. 01
Ol -O
CO
E 4-»
0 i-l
• ^4 ^
Cu u-i -H
3 4J
o to o
W CU CD
00 i— 1
O. •
J= E -a-
CJ CO
CO 05 4J
cu u
oo oi
U C i—l
O 3 J3
U-, i-H 3
in
•u c
4) O Uj
•-t 0
O 0)
o -o c
O. ffl O
E 'H
tQ i-l
4J to Q.
co c cu
l-i O U
•H X •
m 4-1 01 u
CO o
0 -H £ CO
CO Li
oo cu «c 4-1
C 4J 4-1 CO
3 CU -H
CO .C O t



















CO
V
•fH
u
cu
Cu
to

c
CU
01

01
,0
^4
01
U-I
U-I
•rt
"O
CO
E
„
B
"
5
o
c

^
3
U-I
o

to
u
•TH
4-1
CO
•H
Vi
0)

u

±4
CO

u

iH
flj
1^
3

u
3
to
0)
E-i
3
                                                6-53

-------
However, in cultured human bronchus (24 hours) and pulmonary alveolar macro-
phages an absence of phenols (i.e., 3-hydroxy-BaP) and paucity of quinones
was observed.98  Instead, a relative abundance of the trans-7,8-dihydrodiol ^
metabolite of BaP was demonstrated.  This result confirms the work of others
and is  noteworthy in  light of  the  possibility that the 7,8-dihydrodiol is
capable of further  oxidative metabolism to  an ultimate carcinogenic  form  of
BaP.   It is not known whether  a longer incubation period would have  changed
 the pattern of metabolite formation.
 6.2.5  Actions of Enzyme Inducers
      The metabolism of most lipophilic xenobiotics is catalyzed by the
 microsomal mixed-function oxidase system.32'113  One of the oxidases in this
  system,  designated aryl  hydrocarbon hydroxylase  (AHH),  is  responsible for
  both the detoxification and metabolic activation of  POM.   AHH  is the term
  used for the cytochrome P-450-dependent monooxygenase systems  that  metabolize
  BaP to fluorescent phenols.   The activity  of AHH can be induced (i.e.,
  enhanced) by substrate and non-substrate xenobiotics,  either in vivo or in
  cultured mammalian cells.61   This enzyme can be found in 90 percent of the
  tissues examined in rats, mice,  hamsters, and monkeys.113  These include
  !iver,  lung, testis, muscle, bone marrow, skin, brain, intestine, placenta^
  and cultured mammary gland,  lymphocytes,  leukocytes, and  monocytes.
  The ultimate  cytotoxicity and/or carcinogenicity of POM may  in large part be
   determined by the level of AHH activity in the  organ of exposure.   The
   important role of AHH in the susceptibility to  POM-induced damage  results
   from the fact that POM must usually be metabolized to reactive intermediates
   before toxicity can be expressed.    In this regard, it has been shown that
                                       6-54

-------
                                         20
                                          38
  cell lines which were highly resistant to the toxic effects of BaP were also
  lacking in AHH activity.113
       For more than 20 years it has been known that ROM's such as BaP, DBahA,
  MCA, and chrysene are potent inducers of AHH in liver microsomes.61   Thus it
  is not  surprising that cigarette smoke has  a marked stimulatory effect on
  AHH in  lung,  liver,  intestine,  and human placenta.   Intragastric administration
  of MCA  increases  AHH activity  of rat  liver,  kidney,  and  intestinal mucosa,
  and intraperitoneal  injection  increases  AHH  activity in  rabbit  bone marrow.
  In  cultured hamster  fetus cells,  Nebert  and  Gelboin114 established that AHH
  activity was  readily  induced by  carcinogenic and noncarcinogenic  POM.
  Maximum enzyme induction was generally produced with a POM concentration of
  10 MM in the culture medium.  AHH inducibility in animals is under genetic
  control, and it can be shown that noninducible strains of mice are also
  resistant to the carcinogenic effect of subcutaneously injected MCA.115
      Ke.llera.an and coworkers116'117 have used human lymphocytes to demonstrate
 that AHH inducibility by MCA may also  be under genetic control  in humans,  and
 further  that the  extent of AHH  induction may correlate with  susceptibility
 to  chemical  carcinogenesis.   These investigators  observed a  virtual absence of
 lung cancer  among  patients having little  capacity for AHH induction.   There-
 fore,  it may be postulated that  genetic  heterogeneity with regard  to micro-
 somal  enzyme induction  might account for  the  apparently inherited  suscepti-
 bility (i.e.,  familial  clustering) for bronchogenic carcinoma.  However,
 difficulty in  reproducibility of  the assay for AHH inducibility in lymphocytes
 and monocytes, as well as large temporal and inter-individual variability,
 have prevented an evaluation  of its genetic control  in humans.36'118'121  At
the present time,  attempts to confirm several aspects of Kellerman's studies
have been unsuccessful.
6-55

-------
     Paigen and coworkers121  have examined the question of genetic  suscepti-
bility to cancer and concluded that epidemiologic evidence supports this
hypothesis.  Moreover, they were able to show that AHH inducibility in lym-
phocytes segregates in the human population as a genetic trait.   However,
their studies failed to find a correlation between this inducibility and
presumed cancer susceptibility, either among healthy relatives of cancer
patients or  in patients who had their cancer surgically removed.  It is
noteworthy that previous  investigations on AHH  inducibility were conducted
in persons with active cancer.
      Recent  studies with  other human tissues  (liver  and placenta)  have pro-
vided important new data  concerning the carcinogen-metabolizing capacity of
 man and its  implications  for cancer susceptibility.   Conney  and coworkers122
 examined individual differences in the metabolism of drugs and carcinogens  in
 human tissues, and have identified drugs which may serve as  model  substrates
 to provide  an indirect index of carcinogen metabolism for man.   The rates of
 antipyrene, hexobarbital, and zoxazolamine hydroxylation in human autopsy
 livers were highly, but  not perfectly, correlated with the rates of BaP
 metabolism.   In  human placenta, an almost perfect correlation was found^
 between zoxazolamine  hydroxylase  activity and  BaP hydroxylase  activity.
 Thus,  metabolism of BaP  and  zoxazolamine by  human placenta  occurs by the same
  enzyme system(s) or by different enzyme  systems under the same regulatory
  control.124  BaP and zoxazolamine hydroxylase activities were  also  shown to  be
  significantly enhanced in placentas obtained from women  who smoked cigarettes.
  The lack of perfect correlations for the hepatic metabolism of BaP and certain
  drugs in many subjects indicated the presence of several monooxygenases in
123
                                     6-56

-------
 human liver which catalyze the oxidative metabolism of these compounds.
 Furthermore, large inter-individual differences exist in the capacity of
 humans to metabolize foreign chemicals both in vitro and jn vivo.  Further
 studies showed that 7,8-benzoflavone markedly stimulated the hydroxylation
 of BaP, antipyrene, and zoxazolamine in human liver samples, but with a wide
 variation in magnitude among different samples.   These results suggested the
 presence of multiple monooxygenases or cytochrome P-450's in the different
               l ?R
 liver samples.      Moreover,  7,8-benzoflavone did not affect the hydroxylation
 of coumarin or hexobarbital,  thereby indicating  the existence of different
 monooxygenases for the metabolism of these  substrates.   Multiple forms  of
 cytochrome P-450  have been  shown  in the livers of rats,  rabbits,  and  mice,
 but not thus  far  in humans.124  More importantly,  however,  MCA  is a potent
 inducer of BaP  hydroxylase  activity in  rats but  does  not  stimulate antipyrine
 hydroxylase,  clearly  suggesting that metabolism  of POM in rodents may be
 regulated  by  different  enzyme systems than in humans.124
      In contrast  to the apparent multiplicity of cytochrome  P-450-dependent
 enzyme systems for the oxidative metabolism of POM  in man, a single epoxide
 hydrase with broad.substrate specificity may be present in human  liver.122'126
 Because the hydration of arene oxides may lead to the formation of dihydrodiol
 carcinogen precursors, the capacity of different humans to metabolize epoxides
 may affect cancer susceptibility.   It is not known, however, if enhanced
 dihydrodiol formation would increase cancer risk or decrease cancer risk.
                     1 ?7
     Thomas and Slaga    did not obtain a correlation of AHH induction with
 skin-tumor-inducing ability in mice for a series  of unsubstituted hydrocar-
bons.   Nevertheless, the highest AHH enzyme  activity was  found in the  epi-
dermal layer of the skin,  which  is the major point of contact with many
                                   6-57

-------
environmental chemicals.   These results may be interpreted to indicate that
a chemical carcinogen may not necessarily induce its own bioactivation, but
instead can be transformed into a reactive intermediate by virtue of in-
creased AHH activity stimulated by other noncarcinogenic compounds.
     Due  consideration must also be given to the fact that,  in addition to
the  initiation  of  resting  cells by a  chemical  carcinogen,  a  promotion  phase
                                                                      I C.O
 involving cell  proliferation  is also  involved  in skin carcinogenesis.
 Therefore, although certain  aromatic  hydrocarbons  are  effective  enzyme 1n-
 ducers,  their bioactivated metabolites may function only as  an initiator
 having no promoting ability.   A potent complete carcinogen,  however, will  be
 transformed not only into a powerful  tumor initiator but will also be able
 to  interact with cellular membranes,  alter genetic expression, and ultimately
 cause irreversible cell proliferation.  These observations  raise certain
 doubts concerning  the validity and/or  reliability  of equating enzyme  induci-
 bility with carcinogenic  potential for chemical agents.   Further  reinforce-
 ment  of  this opinion  has  been provided by  Shulte-Hermann129 who showed that
  cell  proliferation is not a direct result  of  enzyme induction,  even though
  both processes are normally coupled.
       The further possibility that the genetics of AHH inducibility is organ-
  dependent as well as strain-dependent in animals has important implications
  for evaluating susceptibility to POM-induced cancers.130  Most significant
  is the  demonstration that pulmonary AHH may be inducible in all strains  of
  mice, regardless  of  the  inducibility  of hepatic AHH.   Since the respiratory
  epithelium represents  a  primary portal of entry for POM, AHH activity which
  is induced in this tissue  may bear  importantly on susceptibility  to  malignancy.
                                      6-58

-------
       Enzyme  induction  by POM is  not limited to  AHH.   Owens131  recently
  demonstrated that MCA  can induce hepatic  UDP-glucuronosyltransferase  activity
  in certain inbred strains of mice.   This  enzyme  catalyzes  the  conjugation and
  excretion of POM substrates  after they have  first been oxygenated by  AHH.  The
  induction of this transferase activity and that  of AHH was apparently regulated
  by a single  genetic locus.   However, transferase inducibility does not depend
  on AHH levels, but rather  is stoichiometrically related to the concentration
 of a specific and common,cytosolic receptor regulating both enzyme induction
 processes.   Owens further demonstrated that AHH activity can be fully induced
 in certain  mouse strains (e.g.,  by 2,3,7,8-tetrachlorodibenzo-£-dioxin) without
 greatly enhancing the  transferase activity.   Earlier studies had established
 that  chrysene and chlorpromazine  were potent inducers of  AHH activity  while
 having little effect on transferase  activity.132 Subsequent exposure  to
 carcinogenic  POM (i.e.,  MCA)  could lead to maximal oxidative  metabolism but
 little  transferase-catalyzed  removal  of metabolites by glucuronic acid
 conjugation.   This situation  would be exacerbated by  the  fact that metabolites
 of MCA  are incapable of  further inducing the  transferase  activity.  This
 effect  may have considerable  toxicologic significance  in  that highly reactive
 epoxides of POM formed by  the action  of AHH under these circumstances  may not
 be adequately  removed by glucuronidation.   Thus, one must consider the total
 exposure to all environmental  agents and their possible effect on critical
 enzymatic processes before attempting to assess the  toxicologic impact of
exposure to  a specific  POM.
                                   6-59

-------
6.2.6  Intracellular Binding of POM
     Investigators have known for a number of years that,  when applied to the
skin of animals or added to cells in culture, carcinogenic ROM's become cova-
lently bound to tissue constituents.133  The molecular basis for carcinogenic
and cytotoxic consequences of exposure to POM was attributed to the binding
of the compounds  to  nuclear DNA,134'135 although binding occurs to RNA and
protein  as well.136'137   In the  regenerating rat liver, DMBA produced an  inhibi-
tion  of  DNA  synthesis  accompanied by  a  long-lasting  (more than  4 weeks)  DMBA-
DNA  binding.135  In contrast,  benz[a]anthracene, a weak carcinogen,  inhibited
DNA  synthesis only slightly,  and benz[a]anthracene binding  to  DNA  was  greatly
 diminished between 11  and 24 hours after administration.   It was  suggested
 that cells in the S phase of the cell cycle were primarily affected by the
 carcinogen.   Similar effects on DNA synthesis were obtained using single
                                                                _  .   138
 applications of  initiating doses of DMBA or DBahA to the skin of mice.
 The weak tumor initiator, DBacA, had little effect on DNA synthesis.  In mouse
 skin, it was also  shown  that DMBA preferentially binds to nonreplieating
 DNA.139 Moreover,  DNA-bound DMBA moieties  could  still function as templates
 for  further DNA  synthesis.140   This  finding suggested  that  newly  synthesized
 abnormal  daughter DNA may be  genetically altered  by an error-prone  post-
  replicative repair process which leads to mutations and/or carcinogenesis.
  Such a hypothesis supports the somatic mutation theory of carcinogenesis.1
       Binding of a carcinogen to DNA is believed to be a critical  step in
  tumor initiation.  The application of BaP to mouse epidermis resulting in the
  formation of BaP-DNA adducts has been used in examining dose-response rela-
                                                        n A o
  tionships for BaP tumorigenesis.  Albert and coworkers    found that the
141
                                      6-60

-------
 formation of BaP adducts was a linear function of dose between 4 and TOO nmoles/
 mouse.   Since BaP carcinogenesis in mouse skin was a linear function between 16
 and 500 nmoles/mouse,  binding to DNA may allow for extrapolation of the dose-
 response curve down to levels which are unsuitable for carcinogenesis studies
 with limited numbers of animals.   Furthermore,  it allows  one to evaluate the
 biological  effects  of  environmentally realistic exposures to a particular POM.
 Lutz and coworkers143  noted  that over a dose  range of 40  ug/kg to  4 mg/kg
 (approximately 50-5000 nmoles/animals,  given  i.p.) the binding of  BaP to rat
 liver DNA was  linear with doses  up  to about 1 mg/kg.   A nonlinear  dose-
 response relationship  between 1  mg/kg and 2 mg/kg  was  evident,  which  could
 have been caused  by an induction  of microsomal  enzyme  activity.  Above 2 mg/kg,
 microsomal  enzyme activity was further  enhanced, while DNA binding  increased
 only slightly.
      Because polycyclic aromatic  hydrocarbons are  not  highly  reactive chemi-
 cally,  it was  believed that a reactive metabolite was  responsible for the DNA-
 binding phenomenon.   Gelboin144 and Grover and Sims145  independently confirmed
 this  hypothesis by demonstrating that polycyclic hydrocarbons bind covalently
 to DNA in the presence of rat liver microsomes.   The low level of binding for
 BaP to DNA (one molecule per 50,000 to 500,000 nucleotides) suggested the
 possibility of a unique binding site for BaP derivatives on the DNA.144
 Subsequent studies conducted jji vitro with hamster tracheal epithelial cells
 incubated in the presence of 7,8-benzoflavone  (an affector of microsomal
drug-metabolizing enzymes) confirmed that BaP  binding to DNA was dependent on
an intact drug-metabolizing system.146'147  In a related study, the binding of
                                   6-61

-------
BaP and MCA to rat liver DNA In vitro was enhanced in the presence of an
epoxide hydrase inhibitor.148  It is now known, however, that both aryl
hydrocarbon hydroxylase and epoxide hydrase are required for the metabolic
activation and subsequent binding of BaP to DNA.149  These data implied that
active epoxide intermediates are involved  in the binding.of polycyclic car-
cinogens  to DNA,  although the  extent of  binding need not correlate with
carcinogenic  potency.   Others  have  argued, however,  that a  6-oxybenzo[a]pyrene
free radical  may be involved in the binding of BaP to DNA and synthetic
polynucleotides.150-"153  It was suggested that metabolic hydroxylation of
 BaP at position 6 leading to the formation of the 6-oxy BP  radical  may be
                                       154
 only a minor route for binding to DNA.
      The role of K-region arene oxides as reactive  intermediates in the
                                                                    • 4-3, 5o, 1ot)
 metabolism of polycyclic hydrocarbons has been extensively revnewed,
 although the structure of the bound material  was  not  known at the time.
 Further  confusion  resulted  from the demonstration that high  levels  of  non-
 enzymatic  binding  of polycyclic hydrocarbons  to  DNA occurred with In  vitro
 microsomal  systems.156  Nevertheless, extensive  microsome-dependent binding
  of BaP to synthetic polynucleotide acceptors has been shown, which could be
  both inhibited or enhanced by conditions which affect microsomal enzyme
  activity.156'157  Similar studies using DMBA-epoxides revealed that covalent
  complexes formed primarily with poly G,  indicating that guanosine  residues
-  in  nucleic  acids may be preferred targets for reactive arene oxides.
       Systematic studies were undertaken  to elucidate the  relative  role  of K-
   region  epoxides in binding reactions with nucleic  acids.  Blobstein  and co-
   workers159  examined the In vitro  binding of both K- and non K-region arene
                                      6-62

-------
 oxides to poly G  by measuring  changes  in  the  absorption  and  fluorescence

 spectra of the modified  nucleic acid.   Reactivity with nucleic acids was

 found to vary considerably, depending  upon the structure of  the parent

 hydrocarbon and position of the epoxide moiety.  Significant binding to poly

 G occurred with K-region arene oxides  of DMBA, BaP, and MCA  (all strong

 carcinogens) and benz[a]anthracene (a weak carcinogen).   K-region arene

 oxides of the noncarcinogens pyrene and phenanthrene produced no changes in

 the absorption spectra for poly G.  However,  the non-K-region BaP 9,10-, BaP

 7,8-, and phenanthrene 3,4-oxides also produced alterations in poly G as

 measured  by  changes in fluorescence spectra.   Further investigations revealed

 that RNA  adducts  forms when cultured bronchial mucosa is  exposed  to BaP160

 or when BaP  is painted on mouse skin161 are produced when BaP 7,8-diol-9,10-

 epoxides  react at  the  2-amino group of  guanosine.  Whereas  Weinstein and

 coworkers  ° observed  the formation of  an  adduct  only between RNA and (+)-

 7p,8a-dihydroxy-9a,1Oa-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene, Moore and
          •I rn
 coworkers    found that  (±)-7B,8a-dihydroxy-9p,10p-epoxy-7,8,9,10-tetra-

 hydrobenzo[a]pyrene also  reacted with RNA In vivo by cis  and  trans addition.

 The  7^,8a-dio^-96,^Op-ePoxide arises mainly from the  (+)-enantiomer  of BaP

 7,8-dihydrodiol while the (-)-enantiomer produces primarily the 7p,8a-diol-

 9a,10a-epoxide (see Section 6.2.2).  This observation explains why both AHH

 and  epoxide hydrase are required for BaP binding to DNA,  since BaP 7,8-oxide

must be reduced by epoxide hydrase to form BaP 7,8-dihydrodiol, which is

further metabolized to the BaP 7,8-diol-9,10-epoxide by the microsomal

monooxygenases.   Thus,  the K-region BaP 4,5-oxide does not appear to be

responsible for most of the observed covalent binding of  BaP to RNA.   In
                                   6-63

-------
addition, examination of DNA adducts to BaP formed in vitro  with  Syrian
hamster embryo cells,163'164'165 cultured human bronchi,149  BHK 21/C13 cells
in culture,166 or mouse liver microsomes167 corroborated the results obtained
for RNA, and disputed the involvement of a BaP K-region oxide in the binding
reaction.  Further support for the  importance of BaP 7,8-diol-9,10-epoxides
in the  binding of BaP to nucleic acids was provided by studies which demon-
strated that BaP 7,8-dihydrodiol is highly carcinogenic, and both diastereo-
meric BaP  7,8-diol-9,10-epoxides are potent mutagens  (see Sections  6.4 and
6.5).   A diol-epoxide of DMBA was  also shown  to  be  involved in the  binding
                                     nr  108,168
 reaction with DNA  of mouse  embryo  cells.
      Not all  investigators  agree that  polycyclic hydrocarbon binding to  DNA
 represents the essential interaction leading to  mutagenesis/carcinogenesis.
 Sarrif and coworkers159 were able to purify a protein ("h-protein") from C3H
 mouse  liver and skin which could covalently bind to metabolites of MCA.   The
 protein was postulated as a primary target in hydrocarbon-induced carcino-
 genesis,  playing a  similar role to hormone receptors in transporting car-
 cinogens  to  a target site(s)  in the nucleus.  Alternatively it was  suggested
 that the  protein-carcinogen  conjugate may  in itself  act as an abnormal
 derepressor  to  give rise to  tumor formation.  In related studies,  MCA and
 DBahA were found  to bind with high affinity to  a nuclear subfraction from
  cultured mouse embryo cells.170  This subfraction contained 15  percent  of
  the nuclear DNA and represented that fraction of the nuclear chromatin  which
  is transcriptionaTly active.  The specific macromolecule responsible for the
  binding  could have been,  in addition to DNA, either RNA or one of the
                                      6-64

-------
 chromatin proteins  involved  in gene regulation.  In support of the possible
 involvement of polycyclic carcinogens with essential chromatin components,
 it was noted that the weak carcinogen DBacA exhibited minimal binding to
 this same nuclear subfraction.  Other investigators171 incubated BaP with
 isolated calf thymus nuclei in the presence of NADPH and rat liver microsomes
and demonstrated that an uneven distribution of carcinogen in chromatin
occurs.   The data suggested a localization of carcinogen in the  outermost
"spacer"  regions  of ONA in the nucleosome.
                                 6-65

-------
6.3  TOXICOLOGY
     The potential for POM to induce malignant transformation dominates the
consideration given to health hazards resulting from exposure.   Although the
emphasis on carcinogenicity is certainly justified when dealing with public
health issues concerning POM, one must recognize that non-neoplastic lesions
may also result from environmental and occupational contact.  Such effects
can be seen with  subthreshold doses of carcinogenic POM and with those
compounds which possess no tumorigenie activity.
      As  long  ago  as  1937, it was known that  carcinogenic  POM produced
systemic toxicity as manifested by an  inhibition of body  growth  in  rats  and
mice.172 Tissue  damage resulting from the administration of various POM to
experimental  animals is often widespread and severe,  although  selective
organ destruction may  occur (e.g.,  adrenal necrosis,  lymphoid  tissue damage).
•Few investigators, however,  have  attempted to ascertain the molecular  mech-
 anism of POM-induced cytotoxicity.   Nevertheless,  current opinion favors the
 concept that normally proliferating tissues  (intestinal epithelium, bone
 marrow, lymphoid organs,  testis)  are preferred targets for POM,  and this
 susceptibility may be due to a specific attack on DNA of cells in the S phase
 of the mitotic cycle.173  Additional factors which may have an important
 bearing on the adverse effects resulting  from POM exposure are primary and
 secondary alterations in enzyme activity  and immunologic competence.   More-
 over, these  toxicant-induced changes may  play an important role in the
 eventual induction  of neoplasia.
                                     6-66

-------
  6.3.1   Acute Exposures
       Target organs for the toxic action of POM are diverse,  due partly to
  extensive  distribution in  the body and also to the selective attack by these
  chemicals  on proliferating cells.   Damage  to the  hematopoietic  and  lymphoid
  systems  in experimental  animals  is a particularly common  observation.
  Yashuhira174 described severe  degeneration  of  the  thymus  and marked  reduction
  in weight  of the spleen  and mesenteric  lymph nodes of CF1 Swiss and  C57BL mice
  given a single intraperitoneal injection of MCA (0.3-1.0 mg)  between 12 hours
  and 9 days after birth.  Degeneration of young cells in the  bone marrow and
  retardation of thyroid gland development were also noted.   Newborn mice were
 highly susceptible to the toxic effects of MCA, with many animals dying from
 acute or chronic wasting disease following treatment.   Among surviving CF1
 mice, numerous thymomas eventually developed; none were evident, however,  in
 C57BL mice  despite  serious  thymic damage.
      DMBA is well-known for its effects on  the bone marrow and lymphoid
 tissues.  With single feedings (112 or  133  mg/kg B.W.)  to  female Sprague-
 Dawley  rats,  age  50 days, DMBA induced  pancytopenia by  causing a severe
 depression  of hematopoietic and lymphoid precursors.175  Maturation  arrest
 occurred at the proerythroblast levels;  no  injury  to the stem cells  or  the
 formed elements in  the  peripheral blood  was evident.  The  fact that only the
more rapidly  proliferating  hematopoietic elements were vulnerable to attack by
DMBA led the  authors to suggest that inhibition of DNA replication may be
involved in the toxicologic response.
     Philips and coworkers173 provided strong support for the argument that
DMBA-induced cytotoxicity is mediated via an interaction with DNA.   Female
                                   6-67

-------
Sprague-Dawley rats receiving 300 mg/kg B.W.  DMBA orally and male rats re-
ceiving an intravenous injection of 50 mg/kg B.W. DMBA displayed injury to  the
intestinal epithelium, extreme atrophy of the hematopoietic elements, shrink-
age of lymphoid organs, agranulocytosis, lymphopenia, and progressing anemia.
Mortality among rats receiving DMBA by gastric intubation (females) was about
65 percent; the group treated by intravenous injection showed about 25 percent
mortality.  In rats given 50 mg/kg B.W. DMBA intravenously, incorporation of
14C-labeled thymidine into DMA of small and large intestine, spleen,  bone
marrow,  cervical  lymph  nodes, thymus,  and testis was 'significantly inhibited.
This  inhibition was as  high  as 90 percent in several  organs at  6  hours, and
 indicated a  strong inhibition of DNA  synthesis.   Consequently,  the authors
 postulated that  DNA in  S phase  cells  is particularly susceptible to  DMBA
 attack.   This contention probably holds true  for other carcinogenic  POM as
 we! 1.
      Another lesion which is characteristic of that produced by X-rays is  the
 severe testicular damage induced by DMBA in rats.176  Single intravenous
 injections of DMBA (0.5 to 2.0 mg) given to adolescent (25 days of age) rats
 caused transient degenerative changes in the testis which were most evident 38
 to 40 days after treatment.  Essentially the same effects were produced in
 adult rats,  age  60 days, given  DMBA orally (20  mg) and intravenously  (5 mg).
 Lesions  of the testes  were  highly specific and  involved destruction  of sperma-
 togonia and  resting  spermatocytes, both of which are  the only  testicular cells
 actively synthesizing  DNA.  Neither  the remaining  germinal cells nor the
 interstitial cells were damaged by DMBA.   Surprisingly,  no testicular damage
                                     6-68

-------
  was produced by single feedings of BaP (100 mg),. MCA (105 mg),  or 2-aceto-
  aminophenanthrene (40 mg).
       For many years  it has  been known  that the  application of carcinogenic
  polycyclic  hydrocarbons  to  mouse  skin  leads  to  the  destruction  of sebaceous
  glands,  hyperplasia,  hyperkeratosis, and  even ulceration.177  Sebaceous
  glands are  the  skin  structures  most sensitive to  polycyclic hydrocarbons,  and
  assay methods  for detection of  carcinogens have been based on this effect.
 Although a  good correlation can be obtained between carcinogenic activity and
 sebaceous gland suppression for many ROM's (e.g., MCA, DMBA, BaP, DBahA,
 benz[a]anthracene), such an effect is neither necessary nor sufficient for
 carcinogenesis.  However, workers exposed to POM-containing materials such as
 coal tar, mineral  oil, and petroleum waxes are known to show chronic derma-
 titis,  hyperkeratoses, etc.,178'1^ and tne possible signif1cance flf thesg
 skin disorders to  human cancer is  not  known.
      In female animals,  ovotoxicity has been  reported to  result  from  admin-
 istration of POM.   DMBA was  shown  to cause the destruction of  small oocytes
 and  reduced  the  numbers of growing and  large  oocytes  after oral  administration
        i ftn
 to mice.      More.recently it was  reported that destruction of primordial
 oocytes in mice by injection of  MCA was correlated with the genetic capability
 for  POM-induced increases in ovarian aryl  hydrocarbon hydroxylase activity.181
Thus, it  is  apparent that ovarian metabolism of POM and ovotoxicity are linked
and are under genetic control.
     A toxic reaction which is  apparently unique to DMBA is the selective
destruction of the adrenal cortex and induction of adrenal apoplexy in
     1 op
rats.     Adrenal apoplexy,  increased adrenal  gland weight, and increased
                                   6-69

-------
 adrenal  hemoglobin content were  induced in  female  Sprague-Dawley rats by a
 single intragastric dose of 30 mg DMBA.  'The same  amount  of  adrenal  damage
 could be produced by a 5 mg dose of the principal  DMBA oxidative metabolite,
 7-hydroxymethyl-12-methylbenz[a]anthracene.  Other DMBA metabolites  produced
 no adrenal damage, thus indicating that a specific reactive  intermediate may
 be responsible for this phenomenon.
 6.3.2  Subchronic and Chronic Exposures
      Repeated injections  of benz[a]anthracene derivatives to mice and rats
 have produced prominent changes  in the  lymphoid tissue.  Early investigators
 administered DBahA,  benz[a]anthracene,  and anthracene  to mice in weekly sub-
 cutaneous injections for  40 weeks.183   Each animal  received a total dose of
 10  mg,  and 19 to 38 mice  were treated  with each of  the three hydrocarbons.
 Analysis of lymph glands  removed at weekly intervals  showed an increase of
  reticulum (stem) cells and an accumulation of iron  in all treatment groups.
  Lymphoid cells  were reduced and lymph  sinuses dilated in all groups, although
  these effects were more common  in mice receiving  DBahA.   The weights of the
  spleens in mice treated with DBahA were significantly reduced  in  comparison
  to controls and those animals receiving benz[ah]anthracene  or  anthracene.
       A more detailed study of the subchronic effects of DBahA on lymph nodes
  of male  rats was reported  in 1944.184  Subcutaneous injections given five
  times weekly for  several weeks  caused  normal lymph nodes to undergo hemo-
  lymphatic changes.   These  changes  are  characterized by the presence of extra-
'  vascular red blood cells in  the lymph spaces and the  presence of large pig-
                                                               •I QO
  mented cells.   These changes were not observed by  Hoch-Ligeti    in mice, but
   could be produced in rats by BaP and  MCA  in addition to  DBahA.  The non-
                                      6-70

-------
 carcinogen anthracene, on the other  hand,  did not produce  as  dramatic a
 change in the  lymph nodes of rats.
      Since a likely route of environmental exposure to POM is by inhalation,
 there is considerable interest in its effect on the respiratory epithelium.
 In light of the concern over ROM-induced neoplasms of the respiratory tract
 (see Section 6.7), an understanding of early pathological alterations and
 pre-neoplastic lesions in this tissue has particular significance.
      Repeated intratracheal  applications of DMBA or BaP (100 ug in gelatin or
 mineral  oil)  to male and female  hamsters over periods  from 4 to 16 months
 resulted in  significant mortality (60 percent)  before  the appearance of  lung
        -ion
 tumors.     All groups of treated hamsters  commonly  displayed acute pneumonia
 and chronic pneumonitis.   The  latter  condition was associated with  a prolif-
 eration  of large squamoid alveolar cells, which  occasionally  formed tissue
 masses filling  one or  more air spaces.   As  this  proliferating alveolar epi-
 thelium began to destroy  the adjacent reticulin  framework,  the  lesions were
 considered to be malignant.  Damage to the  trachea occurred in practically all
 animals and consisted  of mucosal  and  submucosal  chronic exudative inflammation,
 ulceration, hyperplasia of reserve cells, squamous metaplasia, and  squamous
 cell carcinoma  (in  the DMBA group only).
     In a subsequent study conducted by Reznik-Schuller and Mohr,186 BaP-
 induced damage to the bronchial epithelium of Syrian golden hamsters was
 examined in detail using semithin (1  urn) tissue sections.   Animals were
 treated intratracheally with 0.63 mg BaP (total  dose) dispersed in a solu-
 tion of saline,  dodecylsulfate,  Tris-HCl, and EDTA once weekly for life.
Animals were  serially sacrificed  at weekly intervals  following the first
                                   6-71

-------
month of treatment and semithin sections of the bronchi were examined
microscopically.  In the first animals sacrificed, minimal focal cell pro-
liferation in the area of the basement membrane was evident in the bronchial
epithelium.  By 7 weeks, cytoplasmic vacuolization of both goblet and cilia-
ted cells had occurred.  Epithelial and basal cell proliferation continued
for several weeks and led to the formation of three- to four-layered hyper-
plastic regions by the llth week.  Epithelial cells began to penetrate
through the basement membrane by the 12th week, and within 2 more weeks the
bronchial epithelium began to continuously grow into the  surrounding lung
tissues.  Microscopic bronchogenic adenomata had  developed by the 20th week.
These tumors consisted primarily of ciliated cells and goblet cells, with
only  a few basal  cells present.  The  apparently small  amount of basal cell
proliferation  may have  been  the reason why  squamous  metaplasia  was  not
observed by  the time  the experiment  had ended after  21 weeks.   Squamous
metaplasia and keratinization were found in the trachea,  but  not in the
 bronchi,  after 21 weeks of treatment.   Although these investigators found no
 increase in the number of alveolar macrophages,  others have reported numer-
 ous alveolar macrophage responses in BaP-treated  hamsters as  well  as focal
                                                              -I Q"7_ "1 QQ
 areas of accumulated macrophages containing a yellow pigment.
      Epithelial proliferation and cell hyperplasia in the absence of necrosis
 and/or marked inflammation is a common observation in the tracheobronchial
 mucosa of animals directly exposed to carcinogenic POM.  This phenomenon was
 shown with repeated exposures of DB[c,g]C,  DMBA, BaP, and dibenzo[a,i]pyrene
 .  .    4.    186,188-190
 in hamsters.    '
                                     6-72

-------
   6.3.3   Effects  on  the  Immune  Systems
       Numerous investigators have demonstrated  that carcinogenic  POM  can
   produce an  ircmunosuppressive  effect.  This action was first observed by
   Malmgren and coworkers19' in  ,952 using Mgh doses of Hcfl and ^^ .n
  •"ice.  Subsequent studies established that single carcinogenic doses of MCA
  DMBA, and BaP caused a prolonged depression of the i.une response to sheep
  red Mood cells.^-™  Noncarc1nogen,.c hydrocarbons ^ as  benzo[e]pyrene
  and anthracene  reportedly had  no immunosuppressive  activity.   In  a recent
  review  on  i«nosuppression and chemical  carcinogenesis,  substantia!  evidence
  was presented to indicate that the degree  of  immunosuppression was  correlated
  with carcinogenic potency for  POM. ™  Both ceH-mediated  and humoral immune
  reactions are affected by POM.
      The cellular mechanism of  immunosuppression by POM has been investigated
  by  several groups of researchers in recent years.   Rowland and Hurd195 re-
 ported that spleens from DMBA-treated (1  mg in 0.2 ml  corn oil  by  single
 intramuscular injection) mice are deficient in bone  marrow-derived lympho-
 cytes but not in  thymus-derived lymphocytes.   Since  the marrow-derived cells
 are  short-lived and  quickly replaced  by the rapidly  dividing marrow-derived
 lymphocyte population, it  was suggested that DMBA preferentially attacked
 this population.  Because  immune capacity was  determined only 7 days after
 DMBA treatment, an effect  on the  long-lived thymus-derived cells could
 probably not have been seen.  This explanation was supported by data from
 Yamashita and coworkers196 who analyzed the effects of DMBA on marrow- and
thymus-derived lymphocytes of mice up to>six weeks  after exposure.   In these
studies,  a bimodal immunosuppressive  effect was obtained-a short-acting
                                  6-73

-------
depression of marrow-derived cells and a long-lasting depression of thymus-
derived cells could be seen.
     In contrast to the results presented above, long-term (6 to 9 months)
exposure of guinea pigs to MCA (4 mg every 4 weeks) resulted in no sustained
                                                 197
suppression of cell-mediated or humoral immunity.     These investigators
suggested that immunosuppression resulting from exposure to POM is a function
of its general toxic effect, but does  not persist long enough to be involved
in neoplastic induction.
     The  importance of an  intact  immune system  in resisting the effects of
chemical  carcinogens  is debatable.   Immunosuppression  induced by thymectomy
or treatment with antilymphocyte  serum has  a more dramatic effect  on carcino-
                                                           194
genesis  induced by viruses than that induced by chemicals.      Moreover,  a
dissociation between  the carcinogenic and immunosuppressive  effects of MCA
was  shown at low doses in C3H mice.198  That is,  single  doses  of MCA  less
 than 0.1  mg induced subcutaneous  sarcomas in many treated animals  in  the
 absence of serum antibody responses to sheep red blood cells.   In addition,
 one-time administration of MCA intragastrically (0.2 to 2.0 mg) to C3H mice
 was carcinogenic but  had  no effect on immune responses.
                                     6-74

-------
  6.4  CARCINOGENESIS
       Numerous polycyclic aromatic compounds are distinctive in their ability
  to produce tumors in skin and most epithelial  tissues of practically all
  species tested.   Malignancies are often induced by acute exposures to micro-
  gram quantities  of POM.   Latency periods  can be short (4-8  weeks)  and^the
  tumors  produced  may resemble  human  carcinomas.   Carcinogenesis  studies  in-
  volving POM have historically been  directed at  studies primarily involving
  effects  on the skin  or lungs.  In addition, subcutaneous or intramuscular
  injections are frequently employed to produce sarcomas at the injection site.
  Ingestion has not been a preferred route of administration for the bioassay of
 POM, despite the fact that it is a major route of human exposure (see Section 5.5).
      Concern over potential  human cancer risk posed by POM present  in the
 atmosphere stems  from studies  demonstrating that crude extracts  of  airborne
 particulate matter and automotive exhaust  were  carcinogenic  to animals.1"-204
 Fractions soluble in benzene or benzene-methanol  produced tumors  in mice by
 skin painting  or  subcutaneous  injection, and in  hamsters  by  intratracheal
 instillation.  Both  the aromatic  and oxygenated  neutral subfractions  of  the
 airborne  particulate  were active  as complete carcinogens, and indicated  the
 presence  of numerous  carcinogenic materials, including non-POM.   The  further
 demonstration that the basic fraction of organic pollutants was carcinogenic
 to newborn mice indicated the presence of aza-heterocyclic hydrocarbons.
Since the carcinogenicity of the total  organic particulates and aromatic
neutral subfractions could only partly be explained by the presence
                                  6-75

-------
of BaP, its usefulness as a measure of carcinogenic risk may,  therefore,  be
limited.
     From investigations in which polycyclic carcinogens were painted on the
                                                                  .   205,206
skin of mice has emerged the two-stage .theory of skin carcinogenesis.
The first stage, initiation, results from the ability of a carcinogen to
effect  a permanent change within a cell or cell population following a
single  application.  The measure of carcinogenic potency is often regarded
as the  capacity  for tumor  initiation.  However, some weak or non-carcinogens
can be  active  as tumor  initiators  (e.g., dibenz[a,c]anthracene, 1-methyl-
chrysene,  benz[a]anthracene).   The second  stage, promotion, is a prolonged
process which  does  not  necessarily require the  presence of a carcinogen, but
nevertheless a chemical stimulus must be  supplied  (e.g., by the phorbol
esters contained in croton oil).   A complete carcinogen is one which,  if
 applied in sufficient quantity, can supply both initiating and promoting
 stimuli (e.g., DMBA, BaP).  The formation of skin  tumors by  polycyclic
 hydrocarbons may also be influenced by inhibitors  and accelerators  (cocarcino-
 gens), thus complicating the interpretation of experimental  data.
      The carcinogenic  effects of POM when applied to the skin of animals
 have been known for decades.  In  1939, Iball207 collected the results of a
 series of experiments  to  arrive at a method for comparing the carcinogenic
 potencies of  various polycyclic aromatic  chemicals.  His results, presented
 in Table  6-15,  express tumorigenie potency  in  mouse skin as the ratio of
 percent tumor incidence to the average latency period.  This expression,
 commonly  referred  to as the Iball .index,  is  still  used as a means of  com-
 paring the  relative activity of carcinogens.   An  important data compilation
                                     6-76

-------
 '2«2'nS'
                    !§J
         "§ 0 C w
         £.ij§
                   22"  S32^  iSSSSSz^o^^?^"--
    SS™  S2*  2SZS2222C:'
1 wt *n 03 o* o eo to h
                    =
                      --      «

                                                                         18
                                                                         Q.
                                                                         Q.
                                                                         a.
                                                                         to
                                                                        •o
                                                                         o
                                                                         a.

                                                                         o
                                                                         o
                                 6-77

-------
on agents tested for cardnogenicity has more recently been published by the
U.S. Public Health Service (Publication No.  149) which lists the results of
tests on hundreds of chemicals in numerous species including rodent, avian,
and amphibian.
     Experimental models for respiratory carcinogenesis have major limitations
in that the delivery of carcinogens to the tracheobronchial tree in measured .
amounts and their adequate retention at the target tissue are poorly con-
trolled.  Therefore, the conduct of dose-response studies for lung tumor
induction has been seriously hampered.  Moreover, the possible relevance of
the two-stage theory of carcinogenesis to lung cancer has not been clearly
established.  Much of the bioassay data on POM-induced lung cancer has been
derived from animal model systems employing various modes of administration
(inhalation, intratracheal instillation,  intravenous injection), and the use
of  carrier particles (e.g., ferric oxide) for the delivery of the carcinogen
to  the bronchial epithelium.  Thus, the results obtained from these studies
cannot always be directly compared.  The  most commonly employed method for
the study of POM-induced  lung cancer involves intratracheal instillation of
test material in the Syrian golden hamster.
6.4.1  Structure-activity relationships
     Following  the  identification of the  first carcinogenic hydrocarbon from
soot, BaP, an intensive effort  was mounted to isolate the  various active
                                 208
components of carcinogenic tars.      From the earliest studies  conducted,
the realization emerged that  carcinogenic POM are structurally  derived  from
                                        209
the simple angular  phenanthrene nucleus.     However, unsubstituted POM
with  less than  four condensed rings that  have been  tested  have  not  shown
                                    6-78

-------
  tumorigenic activity.   Furthermore, of the six possible arrangements with
  four benzene rings,  several  of these compounds are active:   benzo[c]Phenanthrene,
  benz[a]anthracene and  chrysene.   The unsubstituted penta- and hexacyclic
  aromatic  hydrocarbons  are  clearly the most potent of the series.   These
  include BaP, DBahA,  dibenzo[a,h]pyrene,  dibenzo[a,i]pyrene,  diberrzo[a,l]pyrene,
  dibenzo[a,e]pyrene,  benzo[b]fluoranthene,  and  benzo[j]fluoranthene. Somewhat
  less potent as carcinogens are the  dibenzanthracenes and  dibenzophenanthrenes.
 Only a few heptacyclic hydrocarbons show carcinogenic activity.  These
 include Phenanthr0-[2',3':3,4']pyrene peropyrene and dibenzo[h,rst]pentaphene.
 Beyond seven unsubstituted aromatic rings, there are very few known carcinogenic
 hydrocarbons.   However, many physical-chemical  and enzymatic parameters must
 be dealt with when speaking of carcinogenic POM.   Factors such as solubility
 with increasing molecular size and intracellular localization to achieve
 metabolic  activation  are  likely to be important determinants of the true
 carcinogenicity of a  particular POM.
      Among the  unsubstituted  polycyclic hydrocarbons containing a nonaromatic
 ring, a  number  of  active  carcinogens are  known.   The most.prominent examples
 of this  type of compound  are  cholanthrene,  11,12-ace-benz[a]anthracene, 8,9-
 cyclopentanobenz[a]anthracene,  6,7-ace-benz[a]anthracene,  acenaphthanthracene,
 l,2,5,6-tetrahydrobenzo[j]cyclopent[f,g]aceanthrylene, and "angular" ster-
 anthrene.  All  of these compounds  retain an intact conjugated phananthrene
 segment.
     The addition of alkyl substituents in certain positions in the ring
system of a fully aromatic hydrocarbon will often confer carcinogenic activity
or dramatically enhance existing carcinogenic potency.  In this regard, Arcos
                                   6-79

-------
and Argus209 noted that monomethyl substitution of benz[a]anthracene can lead
to strong carcinogenicity in mice, with potency depending on the position of
substitution in the decreasing order, 7>6>8=12>9.   A further enhancement of
carcinogenic activity is produced by appropriate dimethyl substitution of
benz[a]anthracene.  Active compounds are produced by 6,8-dimethyl-, 8,9-
dimethyl-, 8,12-dimethyl-, 7,8-dimethyl-, and 7,12-dimethyl-substitution.  The
latter compound is among the most potent POM carcinogens known, although it
has not been shown as a product of fossil fuel pyrolysis.  Methyl substitution
in the angular ring of benz[a]anthracene, however, tends to deactivate the
molecule, although 4,5-dimethylbenz[a]anthracene may be an exception.  Carcino-
genic trimethyl- and tetramethylbenz[a]anthracenes are known, and their rela-
tive potencies are comparable to the parent 7,12-DMBA.  In general, free
radical synthesis of polycyclic hydrocarbons by pyrolysis does not favor alky!
side chain formation.
     Alky! substitution of partially aromatic  condensed ring systems may also
add considerable  carcinogenic activity.  The best example of this type of
                                               209
activation is MCA, a highly potent carcinogen.
     With alky! substituents  longer  than methyl, carcinogenicity tends to
decrease, possibly due to a decrease  in transport through cell membranes.
However,  different positions  in  the  benz[a]anthracene molecule will  vary with
respect to the  effect  of  n-alkyl  substitution  on carcinogenicity.   Benz[a]-
anthracene  is especially  sensitive to  decreased carcinogenicity  caused  by  the
addition  of  bulky substituents  at the  7-position, and is  indicative of  a once
widely-held  view  for most polycyclics  that  high reactivity  of  the  meso-phenan-
threnic  region  (now called  the  "K-region")  was a  critical determinant for
carcinogenicity.209  Current  studies show that the  K-region is  not involved in
                                    6-80

-------
  critical binding to DNA (see Section 6.2.6).  The substitution of highly
  polar groups (e.g., -OH, -COOH) in the 7-position of benz[a]anthracene
  abolishes tuniorigenic activity whereas a wide variety of less-polar sub-
  stituents can enhance activity in position" 7 (e.g.,  -CH2OH, -CH.CH.OH, -
  CH2COONa, -CH2COOCH3,  -CH.OOCC^  -CN,  -CH.CN,  -CHO,  -NH,,  -SH,  -COCCI,, -
  OCH3).                                                                 3
       Recent  studies  have  indicated that methylation of  the  angular  "bay
  region"  (see Section 6.4.2)  benzene ring  not only in  benzfa]anthracene but
  also  in  other four, five, and six-ring  aromatic  hydrocarbons  leads to  a
  significant  decrease, or even to elimination, of the carcinogenic activity
  of the molecule.  Methylation in other positions does not diminish, but  fre-
  quently increases, carcinogenicity.  For example, 7- and 8-methyl-BaP are
  inactive, whereas 2-,3-,4-,5-,6-,ll-,  and 12-methyl-BaP are strong carcinogens.
      Partial  hydrogenation of the polycyclic aromatic  skeleton can generally
 be expected to decrease carcinogenic potency.  This  was  shown  with various
 hydrogenated  derivatives of  BaP,  benz[a]anthracene,  and  MCA.   On  the  other
 hand,  the carcinogenicity  of DBahA,  dibenzo[a,i]pyrene,  and  dibenzo[a,h]pyrene
 is not significantly altered  by meso-hydrogenation.  This may  be due  to the
 fact that  extensive resonance capability  is preserved.   Moreover,  5,6-
 dihydro-DBahA actually displayed a  fourfold increase in  carcinogenicity  in
 comparison to  the parent hydrocarbon,209 possibly due to the hydrophilicity
 and ease of intracellular transport of.its dihydrodiol  derivative.
     Structure-activity relationships for nitrogen-containing heterocycles
have not been as thoroughly investigated as for polycyclic hydrocarbons.209
Nevertheless,  several  generalizations can be derived  from available
                                   6-81

-------
data.  Arcos and Argus209 observed that the benz[c]acridine nucleus is more
likely to yield carcinogenic derivatives than benz[a]acridine.   In addition,
they noted that the general rule of increased activity resulting from methyl
substitution on one of the meso-anthracenic carbons of benz[a]anthracene
also applies to this series (i.e., position 10 in benz[c]acridine or position 9
for benz[a]acridine).  Lengthening of the alky! substituent or addition of a
second methyl group to benz[c]acridine will not increase activity and may,
in fact,  abolish  it.  Dimethyl  substitution of benz[a]acridine, on the other
hand, appears to  enhance carcinogenicity.
      In  the dibenzacridine series,  carcinogenicity  appears to decrease in
the  order dibenz[c,h]acridine > dibenz[a,h]acridine > dibenz[a,j]acridine,
with all compounds being less active than the monomethyl-  and dimethylbenz[a]-
acridines.   Addition of meso alkyl  substituents  will  generally  enhance
carcinogenicity,  but hydrogenation of the aromatic  skeleton  or  addition  of
 further benzene rings will substantially reduce  or  abolish activity.
      Other POM carcinogens containing one or more nitrogen heteroatoms
 belong to the benzocarbazole series.  Most prominent among these is dibenzo-
 [c,g]carbazole.  However, dibenzo[a-,g]carbazole also shows some carcinogenic
 activity.  The activity of dibenzo[a,g]carbazole, however, is considerably
 less than  for the isosteric benzopyridocarbazoles.    Furthermore, although
 dibenzo[a,i]carbazole shows only very weak activity, introduction of a
 second  nitrogen  atom to form  7,8-benzo-pyrido[2',3':1,3]carbazole produces a
 dramatic potentiation of  sarcomatogenic  activity.  While the acridines,
 carbazoles, and  other nitrogen heterocycles  have not been studied in as
 great detail as  polycyclic hydrocarbons, their  importance as potentially
 hazardous  environmental  pollutants  is  likely to  increase  in the  future.
                                     6-82

-------
  Greater emphasis on alternatives to  solid  fuel combustion  (e.g.,  coal  lique-
  faction) is expected to result in greater  emissions of these  nitrogen-
  containing compounds.
  6.4.2  Chemical Reactivity and Carcinogenicity
       For many years,  investigators .have sought a common molecular feature
  among POM carcinogens  which would serve to explain their biological activity
  The  "electronic theory of  carcinogenesis"  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 Pullmans30 and others  was that a  .eso-phenanthrenic
  region ("K-region'') of  high electron density and  with a propensity  for
 addition reactions was a critical structural feature for polycyclic carcinogens
 In expanding this hypothesis, further biological significance was attributed
 to the concomitant presence of a rather unreactive meso-anthracenic region
 ("Legion") for high  Carcinogenicity.  !„ addition, a region of comparatively
 low reactivity which characteristically undergoes metabolic perhydroxylation
 (corresponding to the 3,4-positions of ben2[a]anthracene)  has been designated
 the M-region.  According to the theory,  only binding of the K-region to
 critical cellular sites  would  cause tumor formation; protein binding at the
 L-region causes  no tumorigenic  effect, while inactivation  is produced  by
metabolic perhydroxylation  in the M-region.   The three  regions  of reactivity
are readily distinguished in the benz[a]anthracene skeleton:
                 L-region
                                               M-region of metabolic
                                               perhydroxylation
                                           K-region
                                    6-83

-------
The electronic K-L theory of carcinogenic reactivity has encountered numerous
inconsistencies, primarily because these relationships were derived from
physical-chemical properties of the parent hydrocarbon and gave no considera-
tion to the biological effects of activated metabolites..
     Advances  in recent years have focused attention on the potential  reac-
tivity of diol  epoxide metabolites of  POM, and their ease  of  conversion  to
trio! carbonium ions.  Under the  assumption that diol  epoxides, which  are
more  readily  converted to carbonium  ions, will be better  alkylating  agents  to
produce  carcinogenesis and mutagenesis, the  "bay region"  theory has  been
proposed.31'64'211   Examples of a "bay region" in a polycyclic hydrocarbon
 are the regions between  the 10 and 11  positions of BaP and the 1  and 12
 positions of benz[a]anthracene:
            ,B«y region
            Benzo[a]pyrene
  The  theory predicts that diol  epoxides in which the oxirane oxygen forms part
  of a "bay region" (e.g., BaP 7,8-diol-9,10-epoxide) will  be more reactive and
  hence more carcinogenic than diol epoxides in which the oxirane oxygen is not
  situated in a "bay region."  Experimentally, the "bay region" theory has been
                      n  +u       267,530,531,533,534,536,537 g p 66-68,72,73,538
  confirmed for benz[a]anthracene,                            Dar>
  chrysene,535 and DBahA.532'539   Moreover, quantum mechanical calculations
  were in accord with the concept that  reactivity at the "bay region" is  highest
  for all the diol epoxides derived from polycyclic  hydrocarbons.
                                      6-84

-------
        The bay region concept has received enough confirmation to lead to sug-
   gestions that an analysis of theoretica, reactivity in this manner may be
   useful  in screening POM as potential  carcinogens.2'2  Among severa,  indices
   of theoretical  reactivity examined, the  presence  of a  bay region  for a series
   of POM  displayed a  high  degree  of  correlation with  positive  carcinogenic
   activity.  lf
   6.4.3   Skin Carcinogenesis and  Induction of Local Sarcomas
   6.4.3.1  Benzcjatorene-The abi,ity of BaP and its various metabolic products
   to mniate skin tumors ,„ the two-stage mouse Carcinogenesis system has been
  extensively evaluated in recent years.   Emphasis has been placed on the
  Identification of a  BaP derivative  which  acts  as the principle ultimate
  carcinogen resulting from metabolic activation.  Various  BaP-oxides,  -phenols
  -dihydrodiols, and diol-epoxides were assayed for  tumorigenicity and  compared'
  to  the parent compound  as  a measure of bioactivation.214'221.5^
      The tumor response obtained when BaP alone  is applied weekly to the skin
 of nice is shown in Figure 6-,2.'«  It ,. apparent ^ ^ ^ ^^
 that increasing weekly doses of BaP caused a shortening  of the latency period
 for carcinoma formation.  Furthermore,  it was  determined that the development
 of pap^omas as  a precursor lesion  to  carcinoma  formation occurred only at
 higher BaP  doses  (e.g.,  32 Mg  and 64 pg per week).  At the lower  dose  levels
 (8 Mg and 16  ug per week),  carcinomas appeared de novo without  precursor
 papilloma formation.
     Among the arene oxides of BaP-tested as complete carcinogens by repeated
 top,cal applications to mouse skin (0.4 Mmole once every two weeks for
 60 weeks), BaP 7,8-oxide was shown to be highly carcinogenic.214  « ,ower
doses (0.1 pmole), BaP 7,8-oxide was  not as potent a carcinogen as BaP    The
                                   6-85

-------
 3
 a
             1.4
                                        • 128 M9
                                        • 64M9
                                        A 32 M9
                                        o 16 ug
                                           1000
                               Time (days)
Fiaure 6-12.  Log-log plot of the cumulative incidence of skin
FTgure   \t.  ^^^ per mouse versus time.  BaP was applied
              topically every week at the indicated doses.  Time
              was measured from the first BaP treatment.'^
                              6-86

-------
   K-region BaP 4,5-oxide  was  only  weakly carcinogenic  at  high  doses,  whereas
   BaP  9,10-ox,de was  inactive.   Subsequent  studies  to  determine  if the phenolic
   non-enzymatic rearrangement products of BaP oxides were carcinogenic estab-
   lished that  none of the seven  phenols  of BaP tested  (4-,5-,6-,7-,8-,9-  and
   10-hydroxybenzo[a]pyrene) were active.219  Among the regaining five'possible
   isomeric phenols of BaP, only 2-hydroxybenzo[a]Pyrene was shown to be highly
  carcinogenic at the 0.4 umole dose.22'  However_
  found as  a metabolite  of BaP in biological  systems.
       Following the  demonstration  of the carcinogenicity  of BaP  7,8-oxide to
  -use skin,  Levin at .,.™  demonstrated that Bap  7>8.d1hydrodjol _ whjch ^
  form  in vivo  by the  action of epoxide hydrase on BaP  7,8-oxide, was  even more
  potent than the parent epoxide.   In  fact, it was shown217  that BaP 7,8-
  dihydrodiol was a slightly stronger  skin carcinogen than BaP,   In the same
  study  it was  established that the diastereomeric BaP 7,8-diol-9,10-epoxides
 were either inactive or weakly active as complete carcinogens  on mouse skin.
      A summary of the availab!e information concerning carcinogenicity of BaP
 and its derivatives  to  mouse  skin  is presented  in Table 6-!6.   It  cannot be
 determined with certainty from these data,  however, which products of BaP
 biotransformation may act as  ultimate carcinogens.  The apparent lack of
 carcinogenicity of the BaP T.S-diol-g^O-epoxides to mouse  skin, despite
 abundant evidence concerning their mutagenicity (see Section 6.5), binding to
 nucleic acids  (see Section 6.2.6), and high carcinogenicity to newborn
 •ice,   •   may be explained by poor penetration of the skin.  Since it is a
 highly  reactive e,ectrophile,  a BaP diol  epoxide may readily alkylate nudeo-
Philic sites of any kind and thus have a reduced probability of passing
                                  6-87

-------
CO
UJ
cc
UJ
 a:
 re
 CO
 UJ
 UJ
 UJ
 O
 ce.
  co
  10
  cu
   to
                                                                                                 o     —
                                                                OOOOOOOO«TO
                                                                                                 §
            V S-
                       eefi!SSS2J2s:ffi—»°   oooooooooo^o
                               —     »

                               r—     l/»

                                i     5
            a> at
                                                            oes
oooddooooes
                                                                                                   »  «
                                                                                                   ) JD
—  •-      ^2
O 01  r-      10
   >   a»
                                                                                                  CO l/l   l/l
                                                                                                  „_.!_>   (O
            .0  O
                                                                    u u o  o  o  o
                                                                                        . O. 0.
                                                                                        i O O
                                                                                                   U t—   O
                                   I ,—  *J  in   O

                                   ;  3  .^  =  -a
                                   ' M-   in  O     0.  c ^:  *J
                                   >  o   o *•>   «
                                   .  ^)   O  S   01
                                  at o   *J o   S
                                  i— «     -^
                                                                                                   u. 10  i/*)  re
                                                          6-88

-------
      IS)
                   CU
                   (J
                   c
                   Ol
                                 «s- 00 co
                                            • o o
                                                       CD 03
                                                                               CVJ CSI C\J
                                                                                                  J CM CM
    a.
    >-
    a.

    'as
    i—i
    O
                  V)
               V.  i_

              5  §
               S  5
                                  o o
                                             O co *r eg
                                             ^r c\j CM
                                                                           O O O
                                                                                      CO CM o
                                                                                                                     0.
                                                                                                                     ID
                                                                                                                     Q.
                                                                                                                  £
                                                                                                                  <0
                                                                                                                  V)
                                                                                                                  O

                                                                                                                  J3
                                                                                                         CU


                                                                                                         I
                tn
              V i.
               CV.HO (^
                                             i cr> r*.
                                                                            O O O
eo
C£
 CO
   U  f
                                                                                                            
                                                                                                            >J=   o
                                                                                                           *j eu
                                                                                                           (O -C
                                                                                                           > <->
                                                                                                        in
                CM
          «T»— O

          odd
                CM
           ^"TCD

           dd d
   L. <«-   o
   01 o   £
  *o    •»—
  a.£   £
     -X   ro
     v>   (j

  s- -o  —
  O CO  r-
£
                                                                                                              >
                                                                                                              IO
                                                                                                        a>

                         cu^.'SiS
                        X! •— X x
                   UlufJcOCOCOCOpi'^.1"!

                   T^Tr^r^r^f^To^^;^
                                               o o  o o o
                                               Iflii
                                                7 7 7 7 7
                                                OO CO CO CO OO

                                                r^ r^ r~- r~- t»»

                                                o. Q_ o. a. a.
                                                "2 ^ 'O t3 ro
                                                £3 co co m a
                                                             c
                                                          o£
                                                          ~£
 X
 O CO 1  r —
 S.  . o — — _
T3 r- M
 >l  I  C CO  CU O)


o o o x  x x
 i  C. S. O  O O
   01 -O O.  Q. Q.
      >t CU  CU CU
 •* '~*^  f*
mo  10 r— i— _

 i   •.
                                                                                             o o o
                                                                                             en ro m
     I  CU
    o s-


 >>OM—i
 5_r «•
 O  CO i__i—^
i   • O CM CM CM
^  p«.  rvi
 >*> 1  C Ol CU CU
                                                                                                                 IB
                                                                                                                 3
                                                                                                                 o-
                                                      co
                                                         Gl CU "& '
                                                         Ol 4-1 ^*
? CL ° X  X X
 3 OJ-0 Q. 0. C.
CO  I  >^ CU  CU CU

  C  14-
  S-2   °
  S- *J   >,
 *j  na  4-^
     u  4->
  CU -i-  v>
  i- —  o
  £S  £
    10-0
 S—  *
 1 S

 3'I
ca  >,
i«» S
tn
(_>  in
   ^!
cu  a)
•—  cu
TJ  2
      in  (C

     '«  §

      §5
      U  C
         Q
      in  
-------
intact through the cytoplasm to reach critical  cellular receptors,  which  most
likely are in the nucleus.
     The possibility that susceptibility to BaP-induced skin cancer may be
related to age was examined by Peto and coworkers.222  The incidence of skin
cancer in mice resulting from regular BaP application (20 Mg) starting at
10, 25, 40, or 55 weeks of age was related to duration of exposure and not
to age at the start of treatment.  This suggests that the induction of the
carcinogen-activating enzyme  system  does  not change with  increasing age.
      The  activity of BaP and  its  derivatives as tumor  initiators in the  two-
 stage mouse skin carcinogenesis  system has  received  considerable attention.
 The rationale for investigating the tumor-initiating capabilities  of  BaP
 derivatives stems from the knowledge that hydroxylated intermediates  may be
 better substrates for activation to a proximate or ultimate carcinogen.
 Moreover, it would not be expected that all carcinogenic POM metabolites
 should be equally effective as tumor initiators,  since different tissues or
 species may have greater or lesser capacity to activate a specific molecule.
      Among the  various metabolites of BaP investigated for tumor-initiating
 activity,  only  BaP 7,8-dihydrodiol  is as potent a tumor  initiator as BaP.
 In  addition, the tumorigenicity  of  BaP 7,8-dihydrodiol is  highly  stereo-
 specific.223  At equimolar doses,  (-)-BaP  7,8-dihydrodiol  was  more active
 than BaP as  a tumor  initiator,  whereas the (+)-enantiomer  was  considerably
  less active than BaP.   Other effective tumor-initiating  metabolites  of  BaP
                                                                       215
  are BaP 7,8-oxide and BaP 7B,8a-diol-9p,10p-epoxide (diol  epoxide 1).
  However, both diol epoxide 1 and diol epoxide 2 (BaP 7p,8crd1ol-9a,lOcr
  epoxide) were  severalfold more active than BaP in the induction of epidermal
  hyperplasia.278  A summary of the skin tumor initiating activities of BaP and
  its metabolites  is presented in Table 6-17.
216
                                      6-90

-------
     o
     ca
     «c
    LU
    ce

    a.
   CQ


   U.

   O
   uu
ai
u
c
0)
                    a,
                    ae
                 S  O)
                 O  in
                f—  3
                •—  Q

                •§.E

                O-
                    s-
                 >- o
                o •<-
                -
                O) O
                01 S.
                3 O.
                Ol  0)
                I/I ^
                o  o
                S- 0)
                u u
                             • IT) CV  O ~ O O ~ O CO CO O
                                                   oo r«»
                                                                 O O
                                                                 O
                                                                 O Ol
                                                                 O
                                                                  ^'

                                                                    °
                                             ai  c

                                             §•2
                                             4-1 4-1
                                             a> >a
                                             u u
                                               O.
                                               10
                                            ai i—
                                            U -*
                                            0; ..-
                                            > i
                                                                 O      *9~
                                                     c

                                                     i
                                                                         v.
                                                                         OJ
                                                                          QJ   r~

                        a A   •£
                        o o   ^


                     „  7^° I
                  QJ ^3  &\ (f\ *Q  ^^

            -g-g^-x^.^. £5

             x'x g S..2.2-S.T
             O  O O-ttJ'O'O.CCO
             Q.  C. 01 I   i   i .7:  .
               i  csj  3 a
             I  I O ^- 00 CO I
            m CO i— •  •  «co Q.
              •  «  .^-ocia! .^
                                                                 tn >,
                                            O  10
                                           4W ^»
                                            ID
                                           -^- ^
                                           *-»   ^
                                           •— o
                                                                         en
                                                                         03
                                                                        o

                                                                        vt

                                                                       Si


                                                                        c

                                                                       10


                                                                       It-
                                                                       O

                                                                       in
                                                                       c
                                                                       O
                                                   >o
                                                   u
                                                                       Q.
                                                                       Q.
                                                                       IO

                                                                      ^,

                                                                      J£
                                                                       QJ
                                                               Ol C

                                                               4-1 O
                                                               (O 4J

                                                               Ol Ol

                                                               V u
                                                               4J (O
                                                               Ix
                                                              C (/I    (/)


                                                              T x   -  .

                                                              O J=    c s-
                                                              (_) 4J    00
                                                                 a;    •r-4-i
                                                              Ol E    4-1 m
                                                              
-------
     Following the subcutaneous  administration  of  BaP to  rats and mice,
sarcomas commonly develop at the site of injection since  the  carcinogen  is
introduced be!ow the basal cell  layer.  A total dose of 6 to  7  umol.es (1.5  to
1/7 mg) of BaP is sufficient to produce sarcomas in nearly all  treated
animals (Table 6-18).  Although a single dose of 0.31 mg or less of BaP
produced  no tumors  in C3H mice, others  reported that 0.0004, 0.004, and
0.04 mg produced sarcomas in one, five,  and 23 mice (strain unknown), respec-
                        * cn 208
tively, out  of groups of bu.
6432  Benzlalanthracene--The weakly carcinogenic  activity of BA has  been
 known for many years.   Recent studies have attempted to  apply  the  bay region
 theory (see Section 6.4.2) to explain the carcinogenicity of BA using mouse
 skin as the experimental bioassay system.  As predicted by the bay region
 theory, the 3,4-dihydrodiol of BA and BA 3,4-diol-l,2-epoxide showed high
 tumorigenic activity on mouse  skin,  and act as proximate and ultimate
 carcinogens,  respectively.534'536'537   Further studies  involving  the production
 of pulmonary tumors in newborn mice with the  diastereomeric BA 3,4-diol-l,2-
  epoxides and the «  and (-)  enantiomers of  BA 3,4-dihydrodio! have confirmed
                              . .  531
  the results shown in mouse skin.
  6433  yo-n^hv^nzralanthracene-DHBA is a potent carcinogen used
  extensively  in the laboratory.  It  is  not found, however, in the environment.
  When applied topically, DMBA  is both  an effective tumor initiator and a
  complete carcinogen,  depending upon the dose administered.227'     Numerous
  studies have indicated that a single  dose of DMBA  greater than 400  nmoles i.
   sufficient to produce tumors  in mouse skin  in the  absence  of a promoting
   stimu!us.233  Attempts to produce skin tumors in CD-I  female mice with a
                                       6-92

-------
















UJ
•z.
UJ
C£
Q.
1— l
« •
O
r*^*
UJ
CO
CO

y
o
£jj
<:
u_
0
u
sz
o

^.
u
0
z=




CO
^~

S£)
 o CM CM ui <
o CM 10 f— CM ir> •





10 vo

8 CO O ID O IO O
Ot UV *£> f— l£>








(J
CM
lO JO vQ U O O ^*%
o .— »— a> o to o
IO f^ ^> If) If) O
F—








«~^ x^^
Ol Ol
IO Ol IO
= « — e ^ ^





»-»
aT
•r
IO
o
i
Ol

ai
10 Ol Ol Ol Ol Ol Ol
L. in in in t/i in (/i
• •—
IO O
"*>
01 >a
4-> U
ie -u
u u
Ol =
r— U-
10 O
e »-
o =

c • ir>
O v •

4->  -
Ol t^ *^»
"^ Ol
C *^ "^
— C Ol
in ^
§J= 0
"c in
s is
4-1
3 Ol •
u e c
JO O Ol
3 E
VI 4J O
IO -D
J3 C 10
i— > Ol
o cr>4->
ai •it-
is J= 0
IO U
vi 10 c
01 tt>-«-
i/i -zt
in vi
t— O)

= 1 j=
o s-
«J- Ol
c • -o
•^ • CM C
Ol 3
TD CJtl^
Ol IO O C
> 0
*0 0 C *J
in O u
1/1 in i- ai
^ IO O C
•a 01 —
Ol *•">
>— C3 C Ol
o m •*- <—
a 4-> 01 c
IO. Ol •"-
CM S- VI
C3 C *•> 10
•1—
10 c 10 10 01
^7 O>^ ^ J3
Ol Ol Ol Ol 10
i_ J2 S_ i- o
Ol Ol OJ -i—
^^ VI 4^ 4~f ^^
in oi vi vi i^
•f- VI •>-••- O.
C O = C 10
•^ T3 *»~ *r~
•5 o -a -5 o
cC m «£ «< Z
IO J3 O "O
6-93

-------
single application-of 200 nmoles of DMBA were unsuccessful  after 35 weeks.
DMBA is metabolized much faster than BaP and has a more complicated metab-
olite profile suggesting that more biochemical parameters may be modulating
DMBA tumorigenesis.
     Albert and coworkers142 found that DMBA readily induces tumors on the
skin of rats as well as mice.  In general, the temporal and dose-response
features of the action of DMBA on rat skin were very similar to that shown in
mouse  skin treated with BaP  (Figure 6-13).  The data suggested  that a similar
mechanism  of action  may be operating for  these  compounds in both species.
      Not only  are skin tumors  induced by  single topical  applications of  DMBA,
they may also  be  formed  after  injection of  the  carcinogen.   Bennington and
coworkers234  reported that  six of 70  female Sprague-Dawley rats developed  seven
pilosebaceous  tumors within one year  after  receiving 3 mg  of DMBA  by  intravenous
 injection.   Slaga and coworkers233 reported that a single  intraperitoneal
 injection (0.6-4.8 nmoles)  of DMBA to mice, followed by repeated application of
 phorbol ester to the skin,  caused the appearance of multiple papillomas.
      When applied to the skin of guinea pigs, high doses of DMBA can induce
                                                         rtOC O^fa
 malignant melanomas which resemble the melanomas of man.   '     Four of
 20 female guinea pigs developed malignant melanomas and subsequent metastases
                                                                  ooc
 after receiving  50  to 100 applications of 0.5 mg DMBA in benzene.      Twenty-
 six  male  guinea  pigs developed only alopecia and hyperpigmentation at the
 site of DMBA  application; one male developed a non-metastasizing  subcutaneous
 liposarcoma.
       As a tumor  initiator,  DMBA is greater than  BaP in  potency.   Slaga  and
  coworkers233  provided dose-response  data in female CD-I mice for  tumor
                                     6-94

-------

                    10.0
                            500 ng
                 o

                 I
                     1.0
                    0.1
                                               DMBA
                                               Rat Skin
                      10
                                        100/ug
                                            20 ag
                                                          '-
J	1—'   ' »  »
                                             100
                                    Time (weeks)
Figure 6-13.   Incidence of cancer  per  rat  after  weekly  topical  application of
either 20, 100, or 500 |jg DMBA in 1.0 ml acetone.^-
                                     6-95

-------
production by single dose DMBA application followed by twice weekly  treatment
with 0.25 percent croton oil.   A single dose of either 0.5,  5,  10, 20,  100,
or 200 nmoles DMBA produced 0.4, 2.3, 6.0, 5.5, 6.2, and 16  papillomas  per
mouse, respectively, at 24 weeks.  At the lowest DMBA dosage (0.5 nmoles),
31 percent of treated mice bore tumors after 34 weeks.  With higher  doses of
DMBA the tumor incidence ranged from greater than 60 percent (5 nmoles) to
100 percent (200 nmoles).  When TPA, the active phorbol ester ingredient in
croton oil, was used as the promoter (17 nmoles) instead of croton oil, both
tumor incidence and tumor yield were increased.  In this study, the effect of
higher DMBA doses  (i.e., 10,  20, or  100 nmoles) was primarily a shortening of
the tumor  latency  period since tumor incidences and the average number of
tumors per mouse were similar after  24 weeks.  However, at  the 200 nmole  dose
level DMBA produced more than twice  the number of  tumors as  the 100 nmole
treatment.  The  authors  suggested  that by increasing  the dose of DMBA  to
200  nmoles a  promoting  stimulus  is superimposed upon  initiation while  at  the
same time increasing  the number of effectively initiated sites.  Therefore,
 it was  implied that the magnitude  of an  initiating stimulus can  be  evaluated
 in terms of the number of tumors eventually produced and their  latency period,
 rather than by the final tumor incidence.   It must be kept  in  mind,  however,
 that the effectiveness of DMBA in the two-stage skin carcinogenesis system is
 influenced by species susceptibility and factors  which modify  the promotion
 process (diet, hormonal status).237  These principles probably apply as well
 to most polycyclic carcinogens.
      Sarcomas developing at the site of injection are frequently observed
 with DMBA, much the same as with  BaP injection.   A single injection of
                                    6-96

-------
   2.5 mg DMBA produced sarcomas  in all of , group of e1ght fema)e Sprague.
   Oawley rats.  In a similar study, 6 urcoles Of DMBA (administered as 30
   injections of 0.2 umoles each on alternate days) produced a 100 percent
   1-cidence of sarcoma in 10 female Sprague-Dawley rats.  The sa»e treatment
  employing BaP instead of DMBA led to essentially identical  results.   In-
  jections  of principaT  DMBA hydroxylatad metabolites  or its  K-region  epoxide
  on the other hand,  were much  less  effective  in  producing sarcomas.   In  C57 '
  black  mice,  10 weekly  injections of  one mg OMBA  produced local  tumors
  (sarcomas  and carcinomas)  in  four of six males and three of five females
  Monohydroxymethyl derivatives of DMBA were active as carcinogens, but less
'  potent than the parent  compound.
  6-4.3.4  3-Methylch0lanthrPnP-As with DMBA,  MCA is a potent carcinogen which
  is produced synthetically for use in the laboratory.   The action of MCA as a
 complete carcinogen in mouse skin was recently investigated  by Burki  and
 coworkers.  3*  A  Slngle appllcation of  3 Mmoles  MCA ^^ ]g ^^  ^
 18 BALB/c  male  mice  after 57 weeks.   In  a  subsequent  series  of experiments,
 repeated applications of 1.5 to  3 umoles MCA  twice weekly for  three or
 17 weeks induced  larger  numbers  of skin  tumors in nearly  all animals.  Never-
 theless, MCA  is not as potent  a  skin carcinogen as either BaP or DMBA.
 However, the  incidence and number of tumors produced by MCA could be in-
 creased by  the administration of trichloropropene oxide,  an epoxide hydrase
 inhibitor and scavenger of glutathione.   Despite this  effect, which implicates
the involvement of a carcinogenic epoxide intermediate by increasing its
half-life,  the K-region epoxide of MCA (MCA-11,12-oxide) was  much less
effective than the parent compound.   It  is  possible, however,  that  such
                                  6-97

-------
epoxides are too unstable to pass through the cornified layer in the  skin  to
reach the dividing cells in the epidermis.
     The skin tumor initiating activity of MCA is comparable to that of BaP
when TPA is used as a promoting agent.  In CD-I female mice, a 50 nmole
application of MCA followed by twice weekly applications of 10 ug TPA pro-
duced  an average of 1.9  papillomas per mouse after 30 weeks.239  The tumor
incidence was greater than 50 percent.   Pretreatment of mice with TCPO
slightly increased the  yield  of  MCA-induced  papillomas per  mouse, again
 suggesting  the  involvement of an epoxide intermediate.
      Cavalieri  and coworkers240  have demonstrated an  isotope effect for skin
 tumor induction by MCA.  Deuteration at positions C-l  and C-5 diminished the
 tumorigenic activity of MCA (0.2 umoles twice weekly for 20 weeks)  on the
 skin  of female Swiss mice.  It was suggested that metabolic activation and
 binding of MCA may involve C-l hydroxylation followed by binding to critical
 intracellular  sites.   In  support of this hypothesis, it was noted that 1-
 hydroxy-MCA and MCA-1-one are sarcomagenic  for  mice by subcutaneous injection.
       Subcutaneous injection  of  relatively high  doses of MCA produced tumors
  in high yields at the  injection site in C57 black mice.241  Three weekly
  injections of 1  mg  MCA produced sarcomas in all of a  group of 10 male  mice.
  Ten of 11  female mice developed local tumors after  receiving 10 injections of
  1 mg MCA each.
  6.4.3.5  MhenZfa.h1anthracene--DBahA was the first pure chemical  shown to
  produce tumors in animals.  A skin tumor incidence of 11.7 percent resulted
  within 28 weeks after  female CD-I mice  received ten daily applications of
  DBahA at  1.0  mg (3.6  umoles) each.239   Other investigators have clearly
                                      6-98

-------
  established the ability of DBahA to act as a complete carcinogen in mouse
  skin; tumor production by DBahA is dose-related and similar in potency to
  BaP.  The K-region epoxide (see Section 6.4.2) of DBahA (DBahA-5,6-epoxide)
  produced about the same tumor response as the parent compound in female CD-I
       OQQ
  mice.
       239
       DBahA was shown to be as  effective a tumor initiator in mouse  skin  as  "
      239
  BaP.      A single  application  of 200  nmoles  DBahA  followed by twice weekly
  applications  of 10 ug TPA  for  30 weeks  yielded  an  average of four papillomas
  per  animal  in about 90 percent of a group  of 30  CD-I female  mice.   As little
  as 0.02.ug  (0.07 nmoles) DBahA can initiate skin carcinogenesis in  mice.242
      By subcutaneous  injection,  DBahA has  produced sarcomas  and carcinomas at
  the injection  site  in  C57 black mice.   Ten weekly injections of 1 mg DBahA
  produced tumors in all of a group of 20 male mice and in 17 of 19 female
 mice.  The K-region epoxide of DBahA produced tumors in mice of both sexes,
 but was not as potent as the parent compound.
      Beuning and coworkers532 recently demonstrated that DBahA 3,4-dihydrodiol
 is a  proximate carcinogen when  applied to mouse  skin and when injected  in
 newborn mice,  as would be predicted by the bay region theory (see Section
 6.4.2).  The authors established  that  of the  three  possible dihydrodiol
 metabolites  of DBahA (1,2-,  5,6-,  and  3,4-dihydrodiol) only DBahA 3,4-
 dihydrodiol  showed  significant  tumorigenic  activity.  In  addition, 42 percent
 of male  newborn mice  injected with DBahA 3,4-dihydrodiol  (420 nmole  total)
 developed  liver tumors  whereas  all mice treated with DBahA, DBahA 1,2-
 and 5,6-dihydrodiol showed normal   liver histology.
 6-4.3.6  DibenzoKhlpyrene-Skin  application of  DBahP was first associated
with the development of epitheliomas in 1933™   More recently, Hoffmann  and
                                   6-99

-------
Wynder243 reported strong carcinogenic activity on mouse skin when applied
for one year, thrice weekly, in 0.05 percent and 0.1 percent concentrations.
DBahP was also a strong tumor initiator.  Both the carcinogenic and tumor-
initiating activity of DBahP, however, were less than that of BaP.
     Several  investigators  have demonstrated that DBahP will consistently
induce  sarcomas when  injected in  rats or mice.208   A total administration of
1.8 mg  DBahP produced 34 sarcomas among 35 male mice; only one of ten female
                                   225
mice developed a  sarcoma, however.
      The reduced  carcinogenic  activity  relative to  BaP  for  DBahP  (and other
 related POM) may  be due to the very large decrease  in  solubility  resulting
 from the addition of another benzene ring.   A decreased solubility would
 imply a reduced ability to metabolize (and presumably activate)  these  compounds,
 although confirming biochemical  studies are not available.   Moreover,  it is
 difficult to compare compounds such as the dibenzopyrenes to the more potent
 carcinogens  (e.g., BaP,  DMBA), since high doses over long periods are used to
 assess  tumorigenicity.   In such  cases, purity of the test compound and  satur-
 ation  of metabolic pathways may  confuse  the  interpretation  of results.
 6.4.3.7  Dibenzora,i1pyrene-Hoffmann  and Wynder243 have established the
  carcinogenicity  of DBaiP in mouse skin.   Among 20  female Swiss Albino mice
  painted three times a week for one year with a 0.1 percent DBaiP solution,
  16 animals developed 15 epitheliomas and 29 papillomas.  With a 0.05  percent
  DBaiP solution,  the results were essentially the same (13  epitheliomas,
  28 papillomas).   The tumor latency period for DBaiP-induced skin tumors was
  longer than the latency period  with BaP.  DBaiP was also an active tumor
  initiator,  but  less effective than either DBahP or BaP.
                                      6-100

-------
       Large numbers of sarcomas in mice and hamsters are produced by subcu-
  taneous and/or intramuscular injections of DBa!^208  Single subcutaneous
  injections of only 0.5 mg DBaiP into the groin of C57 mice produced local   •
  sarcomas in 98 percent of the animals in 24 weeks.
  6-4.3.8  Dibenz0fc,q]carba7olP-Skin tumors have been produced in  mice  by
  painting with DBcgC,  although a high mortality associated  with the treatment
  hampers the  interpretation  of results.   Studies  summarized by  the  Inter-
  national  Agency for Research  on Cancer20*  reported  that repeated applications
  of DBcgC  in solutions of 0.1  to 0.5  percent induced papillomas and carcinomas
  in high percentages.
      Local sarcomas have been induced by DBcgC in as many as 100 percent of
 .ice receiving injections.   In single doses as low as 0.2 mg, DBcgC produced
 17 sarcomas among 20 mice within 40 weeks.208  Depending on vehicle of admin-
 istration (lard,  olive oil,  sesame oil)  and strain of mouse employed (C3H, A,
 C),  the  incidence  of DBcgC-induced sarcomas by a single 0.2 mg  injection was'
 as  follows:   17/93 (C3H mice,  lard);  72/133 (A mice,  sesame oil); 20/27  (A
 mice, olive oil);  5/19  (C3H  mice,  lard);  22/41  (C3H  mice, sesame oil); 46/67
 (C3H mice, sesame  oil).
 6-4.3.9  Ben2ofc1phenanthrpnP-Benzo[c]phenanthrene.and several of  its mono-
 methyl derivatives were tested for carcinogenicity by skin painting at high
 doses and subcutaneous injection in C3H mice.244  Twice weekly application of
 a 0.5 percent solution in acetone to the shaved backs of 20 mice produced
three carcinomas and two sarcomas over the 638 day experimental  period.
Single subcutaneous injections of 5 mg benzo[c]phenanthrene  to 20 mice
resulted  in three  sarcomas.   Markedly enhanced  tumorigenic potency was evident
                                  6-101

-------
for the 4-methyl and 5-methyl derivatives of benzo[c]phenanthrene when
applied to the skin surface, and for the 5-methyl  derivative when injected
subcutaneously.
6.4.3.10  REnzofblfluoranthene-BbF is a common pyrolysis product found in
automobile exhaust, polluted air, and cigarette smoke.208  In 1959, Wynder
and Hoffmann245 reported that BbF had relatively high carcinogenic activity
when  applied  repeatedly to  the  skin of female Swiss mice.  Among 20 mice
receiving thrice  weekly applications of  BbF  (0.5 percent  in  acetone), all of
the  animals  developed papillomas within  five months,  and  19  of the 20 mice
 developed carcinomas  after eight months.   None  of  the mice  survived beyond
 eight months.  At a dose  of 0.1 percent, BbF produced papillomas in 13  ani-
 mals and carcinomas in 17 animals  with 12 months.   At 0.01-percent, the
 lowest dose tested, five of the ten surviving  mice after 14 months developed
 papillomas.   By comparison, BaP at the same 0.01  percent dose level  caused
 carcinomas in 17 of 20 mice after 12 months of treatment.
       In  later studies by Van Duuren and coworkers,246 BbF was shown to be a
 potent tumor initiator.  A single application of  1 mg BbF in acetone to
 20  female KCR/Ha Swiss mice, followed by  thrice weekly applications of 25 ug
 croton  resin,  produced 18  mice with papillomas and five  mice with carcinomas
 within  63 weeks.   The same dose of  BbF,  followed  with  no promoting agent,
 yielded no  tumors.  By comparison,  a  150 ug dose  of  BaP  followed by the  same
  croton resin treatment produced 20  mice with  papillomas  and six mice with
  carcinomas over the same time period.   The tumor initiating potency of BbF  in
  this test system was slightly greater than for chrysene.
       Three  subcutaneous injections of 0.6 mg BbF given over two months
  produced local  sarcomas in 8  of 16 males (average latency 130 days)  and 10 of
                                      6-102

-------
   14  females  (average  latency  133 days).208'225  The  sarcomagenic potency Qf
   BbF was considered to be not much less than for BaP, especially in.females.
   6.4.3.11  Benzomfluoranthene-Little biological information is available
   concerning BjF, despite the fact that it is a pyrolysis product found in
  automobile exhaust, polluted air,  and cigarette smoke.   Nevertheless, Wynder
  and Hoffmann245 reported in 1959 that BjF produced carcinomas on the  skin  of
  mice.   Among 20 female Swiss  mice  receiving  thrice weekly applications of  BJF
  (0.5 percent in acetone),  !9  animals  developed  papillomas and carcinomas
  within  seven months.   None  of the  mice survived beyond  seven  months.   At a
  dose  level of 0.!9  percent  in acetone, BJF produced  papillomas in 14  mice  and
  carcinomas in all 20  mice within nine months.
  6.4.4   Respiratory Tract Carcinogenesis
      Until recently, the investigation of ROM-induced carcinogenesis  in
 tissues of the respiratory tract has been hindered by the lack of a reliable
 animal model  system.  Nevertheless, many compounds of the POM class, depending
 upon the assay system, display marked carcinogenic activity in the  respiratory
 tissues.  However,  the carcinogenic potency of a polycyclic compound for the
 pulmonary tissues may  not necessarily  parallel its  activity either as  a com-
 plete carcinogen on  mouse and  hamster  epidermis, as a tumor initiator  on
 mouse  skin, or as a  sarcomagenic agent.  Nevertheless, bioassay data can
 serve as guidelines  for the  prevention of  neoplastic  diseases  in man in at
 least two ways:   (1) the  carcinogenic potential of  a  single chemical can be
 assessed by examining  its effects in multiple animal  systems with comparison
 of target tissues, and by various routes of administration; or (2) a series
 of related chemicals can be compared within the same bioassay system and
ranked according to  their tumorigenic potency  based on dose-response  relation-
ships.
                                   6-103

-------
     An examination of comparative carcinogenicities  within  the  same  tumor
system can provide valuable insight concerning relative risks of various  POM.
By single intravenous injection of about 0.25 mg of aqueous  dispersions of
polycyclic hydrocarbons to strain A mice, a direct comparison of carcinogenic
potency was possible (Table 6-19).247  In this test system,  MCA displayed the
greatest lung tumor-forming capability; DBahA followed closely in activity,
with DBcgC and BaP being considerably less potent.  These results demonstrate
the variability in tissue-specific response to POM, since BaP is a more
active carcinogen  in  skin  and as  a transforming agent In vitro.
                                                     _,   u    .    188,248,249
      Intratracheal  instillation of POM to Syrian  golden hamsters
has  become widely  utilized for the conduct of  dose-response studies  for
pulmonary carcinogenesis.   Several studies are summarized in Table 6-20  and
 indicate that:   (1) dose-response relationships  are  clearly evident,  and (2)
 the co-administration of carrier particles  such as Fe203  (i.e.,  with BaP)  can
 markedly increase tumor incidence,  depending upon the physical  characteristics
 of the particle.   Since environmental  exposures to BaP occur in conjunction
 with particulate material in air, this effect may be particularly relevant to
 human situations.  In addition, increasing doses of carcinogen generally
 decreased the tumor latency period.
       Recently, preliminary results have become available on a study to estab-
 lish  temporal and dose-response  relationships for BaP carcinogenesis  in the
 hamster  respiratory  tract.142  Weekly intratracheal instillations of  BaP
 (2.0,  1.0,  0.5, or 0.25 mg) with or without an equivalent  amount of  Fe203
 were conducted for 15  or  30 weeks.  Initial results,  based on  gross  observa-
 tion only,  indicated that the  incidence of  respiratory nodule  formation was
                                     6-104

-------
     oo
     a
                       i. Ol
                       O wi

                          o
                       Cft ^^
                          OJ
    UW
    CCCVJ

    a LU
   CO Z
   Q£ i— i
   Q
   >-
      oo
  - o:
  —J o
  o s:
 o u.
 CJ3 O
 O i— i
 UJ
    CQ
ce o:
«c =3
ex. oo
                       in
                       O)
  • tt)
 O to
                      c s
                         HI
                         U
0)  •
u o
                        Ol
                    91  O)
                                     o     •—
                                                                CO
                                     r*;  O  r>.   to'  *-   CM  CM

                                     CM  CM  O   O   O   O  O
                                                                                                    a>
                                                                               10

                                                                              *0

                                                                               u

                                                                              *=»!
                     10   oo
                     co   ro
                                                             Jg
                                                             CO
  o
  o.

  o
 t_J
BS
Ol
C
Ol
                               u
                                     cu

                                     O)
                T3
               r—
                O
               4=
                u
               »—
                >>
     Ol        4)

g   S       J

O)  ja        ^"
u   $_       .C
ID   ID       4-J
b   O        C
                                                   c
                                                   a>
              en
               «
              u
             i — i   .
              o   >»
              M   S
              C  I — i
               c   -a
               ID   ••—
               Ol   S_
               U   O
                                         C  •.-
                                         Ol  Q
                        ID    ID
                       —I  I—I
                        M    N
                             C
                             Ol
                                             01

                                             Ol
                                             i.
 10
 C
 O)
-C
 0.
i— i
 O
l_l
 o
 IM
 C
 Ol
.a
                                                                       co   PO
                                                                            O)

                                                                            a>
                                                                            u
                                                                            £
      C
      ID

      ia

 01   IM
 C   C
 oi   ai
 ^  ^2

 CL  "~
I—1
 ID
                                                                           CL
                                                                           O
                             o
                             IM
                             C
                             Ol
                            .a
      CL
      i
                                                                                01
                                                                 Si   >,
                              O    _
                              IM    Ol
                              =   J=   _
                              Ol   -i-   —
                             CO   O   O
                                                                 Ol
                                                                          =f
                                                                          m
                                      C
                                      ID
                                     r—i
                                      ID

                                      IM
                                      C
                                      Ol
                                     CO
                                                                                     o
                                                                                     o
                                                                                     u
                                                                                     01
                                                                                                   o
                                                                                                   c
                                                                                                  o
                                                                              u
                                                                              O)
                                                                             •*•>
                                                                  o

                                                                  Ol

                                                                  ID iJ
                                                                  i. 01
                                                                             0)  V)
                                                                            r- -ii
                                                                             cn 01
                                                                               o
                                                                            •O CM
                                                                            Ol

                                                                            •^ Ol
                                                                            01 •—
                                                                            O i —
                                                                            Ol "-
                                                                            L. J*
                                                                           •o  v.
                                                                           •—  01
                                                                            o  s
                                                                                                Ol  ID
                                                                                                O>
   I   S-
  OO  01
      a.

  01 •.-
  
-------
co
on
tu

OO
1
(S
C£

CO O
   o.

gu.
   o
CO
oi z:
o o
«C I—
Q£ CO
t— z:
 cc
 H^


 UD
 cc.

 u.
 o


 o


 o
 to
  a>
 J3
  (O
Oi
u
c

i.
a)
          £J
          H""
           in
           OJ
              O)
              c/>
              o
                c



                o
                 g.
                             o o «o \aa\ o\ o\   o\ oo coco ao r— <—
                        o o o o «o \a
                        iennutmi— r—
                      CM CM CSI CM CM
                                                                        aooOLOir)
                                       CM CM CM   CM CM CM CM CM CM CM r— r— •— r— CM CM
                                                                  o o
                                                                  oScri
                                                                                                    at 1/1    c
                        ca co m a CD CD co co oo ca
                                                  6-106

-------
  dose-dependent;  0.25 mg BaP  with 0.25  Fe^ given for 15 or 30 weeks  produced
  no  nodules,  whereas  2.0 mg BaP  with  2.0  mg Fe^  given  for  15  or  30 weeks
  resulted  in  a  tumor  incidence of 44  percent and 86  percent,  respectively.
  Dose-response  information is depicted  in Figure 6-14.   Surprisingly,  the use
  of  Fe203  as  a  carrier for BaP had no apparent effect  on  the  incidence of lung
  nodule formation (Figure 6-15)  in this recent study.  While  tumor latency
 periods decreased with  increasing BaP weekly dose, no clear  trend was obtained
 in temporal aspects of tumor development between 15 and 30 week treatments at
 the same dose.
          540
      Mohr    recently compared several  methods of carcinogen administration
 for the production of respiratory tract tumors in  the-Syrian hamster.   Three
 different methods were employed  for  the administration of BaP,  instillation,
 implantation, and inhalation.  Whereas  both instillation and direct  implantation
 of BaP were very  effective  in the induction of respiratory tumors, inhalation
 exposure yielded  no tumors.   Hamsters were exposed to  BaP aerosols of  40-50 mg/m3
 of air for 4.5  hours  per day  over a period of three  to four  months and
 observed for  one year.   These negative  results underscore  the need to  care-
 fully  examine carcinogenic activity of  POM under environmentally realistic
 conditions  of exposure if reliable dose-response data  are  desired.
     In addition to the  hamster  model system, respiratory  lesions and carcin-
 omas have  been  readily induced by  POM in  rats and mice by various routes of
 administration.   The results of  several  representative studies are summarized
 in Table 6-21.
     An additional approach used to evaluate the  effects of POM on the
pulmonary tissue was reported  by Flaks and Sims.253  Pulmonary tissue taken
from tumor susceptible female  BALB/c  mice was incubated for 30 minutes  in the
                                   6-107

-------
                                                                                                ~  g£~
                                                                                                i—  ,— ev«  m
                                                                                                  . to     J-
                                                                                                 c > -o ••-
                                                                                                 o •<-  ai u-
                                                                                                •»- 3 +•»
                                                                                                +•> o" 

                                                                                                ,£ ""55

                                                                                                *"" "**  u  =
                                                                                                •M -f-  

                                                                                                •— (S  3   i-  3

                                                                                                 cj     E  *o
                                                                                                 (O O  3  CJ
                                                                                                 L.   ••<->£
                                                                                                 4-> CM
                                                                                                 ,i— or»-
                                                                                                 Q. OJ  t.
                                                                                                  IO f  O

                                                                                                 "^o
                                                                                                  C 01  J-
                                                                                                  0>    CL
                                                                                                  I_ O -t-3 -0

                                                                                                  4-» -4J  O  3
                                                                                                  en   O
                                                                                                  T— c   01
                                                                                                  •t- 4.)     -=
                                                                                                  •O — — •!->

                                                                                                  C  >, S_ C
                                                                                                     •— O) O
                                                                                                    )-   -
                                                                                                   J_  O) *->  OJ
                                                                                                   O    3 O)  3
                                                                                                   (O        T3
                                                                                                   S_  C L.   «<-  o
                                                                                                   ul  -i-     S-
                                                                                                   a>  01  me.
                                                                                                  a:     O
                                                                                                      ai   C^MI
                                                                                                      S. OJ r—
                                   HUM NOIlUOdOUd
                                                                                                                a)
n-uttn
 i   01 o
vo  $_
    ttJ en •*—  <"
 4)  jJ 4J •*-  O
 t.  wi C i—  ~

 O 5 O  O)  O.
f-  n S J=  f
U_^ 5 4J  CD
                                               6-108

-------
o     o      o     o      d

saowni HUM Nouaodoad
                                                                         35-    c
                                                                            O i. O
                                                                         C <4- Ol •!-
                                                                         O     Q-4->
                                                                        •r- Q.    S_
                                                                        4J  re in o
                                                                         re CQ £_ Q.'
                                                                               O) O
                                                                                  i.

                                                                        ^J  "™ (••*

                                                                        oo o  re 
                                                                        at     a> -r—
                                                                        J= 00  S-  CD
                                                                        o c  ai
                                                                        re o  3  o


                                                                        re re i- at
                                                                        i- i— OJ S_
                                                                        •u i— J= 3
  •r- +•>
      t/5
                                                                                 d)
                                                                                 (J
                                                                               .  O
                                                                              -s 1_
                                                                               coca.
                                                                       aj re    +J
                                                                       > J_  0)
                                                                       •i— 4-J -53 ,  ll +J "°
                                                                       E .^  S_    Q_

                                                                      •c ai «*— .Q ca
                                                                          cu
                                                                       s: 3 OTO -»->
                                                                      ••-    E a> oo
                                                                          c:    N s_
                                                                       vr-
                                                                       O  >  • r- M-
                                                                       c -i- cvj re
                                                                         CD    CO)
                                                                      "o    +j re jc
                                                                      •>-  d)  3   -(->
                                                                       u  s_ o a)
                                                                       =  at ^= s_ E
                                                                      ••- 3 -t-> at o

                                                                      at to 3    4_
                                                                      r— S-    _C

                                                                      •o -tJ c  re a>
                                                                      o oo re aj i-
                                                                         re
                                                                                  00
                                                                               -M re
                                                                               .re ai
       oo c oo
       -^ 3 re

       S 4? S
 oo  co s    a>
 0) O     oo E
Cc:  CvKD 5- -r=
    at  co o i—
                                                                     LO -U  O -P   •
                                                                     i—3        £_
                                                                      I  O LT5 T3  O
0) -^
S- 3
3
                                                                               re  3
                                                                           OJ
                                                                              Q.
                                                                           4= 3 J=
                                                                           4-J O -t-J
                                                                        C -r- i- -i-
                                                                        re cu en 3
                  6-109

-------
co
i
CO
 CO
 UJ
 (X
 o
 CO
 OJ
                                    e-no

-------
 presence  of  4 ug/ml  of DMBA,  DBahA,  BaP,  or one  of their related  K-region
 epoxides,  and subsequently  implanted into isologous mice.   Under  these
 conditions,  only DMBA  showed  marked  carcinogenic  activity.
     Squamous  cell carcinomas have been induced  in  rat trachea! grafts by
 instillation  of MCA, BaP, or  BaP mixed with  Fe^ particles.254  Neoplasms
were induced by 5 mg of carcinogen after  35 weeks of contact in 3 of 5, 2 of
5, and 2 of 3 trachea! grafts exposed to MCA, BaP, and BaP-Fe^,  respectively.
Subsequent studies in the same test system using DMBA in a sustained release
formulation established that carcinomas could be induced in trachea! epithe-
                              255
                                   With this procedure,  the lowest dose of
                                                 256
 lium at doses as low as 40 |jg
 BaP tested which produced a carcinoma was 300 pg.
 6.4.5  Carcinogenesis in Newborn Mice
      The newborn mouse is known  to  be highly susceptible  to  carcinogenic
     257
 POM,     presumably  because of its large  numbers  of  rapidly dividing  cells,258
 and is  thus  a useful  model  for carcinogenesis  studies.  In studies reported
 by  Asahina and coworkers,259  various  fractions of an organic extract of
 atmospheric  particulate  pollutants  diSp!ayed high carcinogenic activity when
 injected  subcutaneously  to  infant Swiss mice of  both sexes.   Various combina-
 tions of  remote  tumors (hepatomas,  lymphomas, pulmonary adenomas) were pro-
 duced by the basic, neutral, and aromatic fractions of the benzene soluble
 organic extract.   Three oxyneutra! subfractions also produced multiple tumors.
These results indicated that a variety of different chemical  carcinogens are
present in organic extracts of air pollutants in addition  to  BaP, which is
found mainly in aromatic fractions.   Walters263 administered  DMBA by  single
injection to male and female newborn (<24 hours old) BALB/c mice  at doses
                                  6-111

-------
ranging from 0.625 to 40 ug.   At the lowest dosage,  the incidence of lung

tumors was"greater than 50 percent in females and greater than 75 percent in

males.  Subcutaneous injections of 0.005 ug MCA or 0.003 ug DBahA were suffi-
                                                                     264
cient to produce fibrosarcomas at the injection site of newborn mice.

However, on a mg/kg of body weight basis these doses are comparable to

carcinogenic doses in adult animals.  Attempts to determine whether the

appearance of DMBA-induced lung tumors is more likely at a definite point in
                                                              pCC
the life span of a mouse  have  not produced conclusive results.

      The K-region epoxides of  DBahA  and MCA were found to be  negative for

carcinogenic activity  in  newborn mice, despite reports that the  parent
                                                         peg
hydrocarbons were active  at the  same dose  levels (60 ug).     On the  other

hand, the 3,4-dihydrodiol of  DBahA  is highly  tumorigenie in the  newborn

mouse,532 as would  be  predicted by  the "bay  region" theory  (see  Section

 6.4.3.5).   These results  suggested  that K-region epoxides  may not be the

 ultimate metabolites involved in POM carcinogenesis as once thought.   In

 support of the contention that a dihydrodiol  precusor  of the  "bay region"
                                                  ••4
 diol  epoxide may be a proximate carcinogenic metabolite of BaP,  Kapitulmk

 and coworkers66'68 recently showed that BaP 7,8-dihydrodiol is more carcinogenic

 than BaP in newborn mice.  Moreover, one of the stereoisomeric BaP 7,8-diol-

 9,10-epoxides, (+)-trans-7p,8crdihydroxy-9a,1Ocrepoxy-7,8,9,10-tetrahydro-

 BaP, demonstrated extremely high carcinogenic activity in newborn mice, as

 would be predicted by the "bay region" theory (see Section 6.4.2).  Since the

 metabolism of BaP results in  three  known diols, all via epoxide  intermediates,

 it thus appears that the major pathway for carcinogenic activation may  be:

 BaP  	> BaP 7,8-oxide 	» BaP 7,8-dihydrodiol —> BaP 7,8-diol-9,10-epoxide


 which  results in formation of the  diol epoxide  as  an  ultimate carcinogen.
                                     6-112

-------
       Wislocki and coworkers267 recently presented evidence supporting the
  "bay region"  theory with regard to the carcinogenicity of benz[a]anthracene
  (see Section  6.4.3.2).,  They demonstrated that benz[a]anthracene-3,4-dihydrodi0l .
  possessed  about  35  times the activity  of the  parent  hydrocarbon  and  more  than
  300  times  the activity of the other  four isomeric  dihydrodiols in producing
  lung adenomas in  the newborn mouse.  The "bay region"  theory predicts that
  benz[a]anthracene-3,4-diol-l,2-epoxide would  be the  reactive ultimate car-
  cinogenic metabolite of  benz[a]anthracene.  Since the  3,4-dihydrodiol is the
  metabolic precursor of the 3,4-diol-l,2-epoxide, it acts as a proximate carcinogen
  according to  "bay region" theory predictions.
  6.4.6  Tumors in Other Tissues
      Mammary cancer induced in rodents  by DMBA has become an important model
 system for the study of human breast cancer.268'2^,270  The deBonstrat1on Qf
 the production of hormone-responsive mammary tumors in normal  female  Sprague-
 Dawley rats by a  single  intragastric dose of DMBA  by  Muggins and  coworkers271
 in  1961  introduced a simple,  yet sensitive,  model  for the experimental
 investigation  of  the biology  of mammary tumor  growth.   A year later,  Muggins
             O"7O
 and coworkers     demonstrated that  a  single  intragastric feeding  of 20 mg DMBA
 resulted in  the production  of mammary cancer or fibroadenoma in 100 percent  of
 the treated  animals.
      In 1965,  Muggins273  reported that  intragastric feeding  of DMBA could be
 substituted with a single intravenous injection of 5 mg DMBA in lipid emulsion
 for the purpose of producing mammary cancer  in  rats.  By this technique,
Huggins successfully induced mammary carcinoma  in 1500 consecutive female
Sprague-Dawley rats,  age 50 days.  Tumors were detected by palpation  as early
                                     6-113

-------
as 20 days after DMBA Injection.  In a study with 90 rats, mammary cancer was
detected in 28 to 92 days, with a mean time of appearance of 42.8 + 11 days.
In 1966, Griswold268 adapted the DMBA-induced mammary cancer technique in
rats as a routine for the evaluation of potential anti-cancer agents.  Using
the Muggins system, Griswold found that tumor regression was produced in
DMBA-treated  tumor-bearing  rats by hypophysectomy or ovariectomy  at  90 days
after  feeding of DMBA and by injection  of  testosterone  propionate beginning
120  days  after DMBA feeding.
      It is  now known that mammary cancer  can be induced by DMBA in rats  not
 only by systemic administration but also  by local  application of the carcino-
 gen to the mammary gland.274  These tumors were primarily adenocarcinomas
 of ductal origin and were microscopically evident as early as 30 days after
 local application of 1 mg DMBA.  The minimal dose of DMBA to produce a
 mammary tumor was 300 Mg.  It  is important.to note that in this  study
 hyperplastic alveolar nodules  did not form  in the mammary glands after local
 DMBA  application, thereby  disputing earlier contentions that these  "preneo-
 plastic" lesions were a  prerequisite to mammary carcinogenesis.  Thus,  it was
  demonstrated that  transformation to a  neoplastic state may  occur directly and
  require  no intermediate  steps.
       In  addition to DMBA,  ingestion  of BaP may also result in mammary tumor
  formation.  Albert and coworkers142 "reported the  development of multiple
  mammary nodules in Lewis rats following a single intragastric feeding  of
  50 mg BaP.  Numerous nodules were present within 50 days of the treatment.
       POM-induced cancers of the endocrine  target organs are not limited to
  the  mammary gland.  Ovarian granulosa cell tumors in  mice result from expo-
                                        6-114

-------
  sure to BaP and MCA.275  In Wistar rats,  a silk thread impregnated with DMBA
  and inserted in the region of the  ovary produced 47 ovarian solid tumors
  among 121  animals.276
       In addition to ovarian cancers,  DMBA  can  also  produce  carcinoma  by local
  application  to  the  pancreas of Sprague-Dawley  rats.277   Less than  1 mg  of
  DMBA  placed  directly into  the pancreas  induced  adenocarcinoma in eight  of ten
  treated  animals within  180  days.
  6.4.6.1  Oral exposure to  POM-Oral administration of POM frequently results
  in papillomas and carcinomas in the forestomach of rodents.   Numerous studies
 conducted with BaP, DBahA, and DBcgC given orally to rats, mice, and hamsters
 have recently been reviewed,208 and some of these data are summarized in
 Table 6-22.  Tumors of the stomach  have also been induced by MCA279 and
      280
 DMBA.      A single feeding of 3 mg  DMBA induced nine stomach papillomas  in  18
 surviving Swiss  mice after 1 year.280   Ovarian,  lymphoid, mammary,  and hepatic
 tumors were also observed.   In addition, the tumorigenicity  of DMBA in the
 rat  submandibular gland  has been  repeatedly demonstrated.281
      Very little work has  been  done to characterize  dose-response  relation-
 ships  for POM-induced tumors by oral administration.  The most extensive
 studies  in  this  area have been conducted by  Rigdon and Neal.282'284  Their
 studies  involved  gastric tumors (papillomas  and  squamous  cell carcinomas),
 pulmonary adenomas,  and  leukemia in CFW mice fed BaP.  Small  groups of mice
were fed BaP at various concentrations in their chow, and for different
periods of time,  up to a maximum of 197 days.  In some groups, an observation
period of from 104 to 114 days followed the treatment period.  The results of
Rigdon and Neal283  are summarized in Table 6-23.  The authors noted that
                                     6-115

-------





3
"3
E
CO
H- 4 |
ee.
o
1
^•^

i
t—
in
^•4


^H
|
§
0
CO
1 .
Uw
o
1—
t— 1
z:
o
»— i
o
cc
<:
o


OJ
OJ
0)

ja
«
H*











i


4J
U
Ol
<*-
l*-
UJ


c
o

*•*

O J;

IO



O>
en
o
a


-

in
ai
•*~
y
a.
on



•O
3
|

O
tj

C
u
1
in
Ol
JF
•g'S
4-> 4J
3
U- 0
I. "re
4-> re
^






u
^
tn
a
en
R>
L.
£

Ul
J^
C
i1
CM
O







01
in
i




a.
CO



in
3
en
IO

3"
in
Ol O
4-1 4-1
3
C O
iplllomas developed 1
and forestomach in 3
40 animals
o.








,
o ^
01 in
30
U 0>
4-> o. en
IO  IO








in
in in JS
f4-*
s

•J= 'o.^ S.^
c re re re i
re o. E cxo
r~. r> "c in *~
10 re ^ t
c m co m "*-
stomach papillomas 1
treated for 1-5 nionti
and 2 carcinomas in 1
treated for 6-9 mont!
1n 8 animals treated
in






w
5
10
en
re
u
c •

01
i1
IT)
1
CM






1-
01
in
re








.

w
i

in
Ol
o

i- vn .
§OI
3-S-
4-1 t-
CM CVJ
i^


Ol

*

fc- ^^^
ai O m
>, m a. c
C >, 3 g
ai Of
SS. *~


H
a.
o
o
in






01
in
re
re













LT
re
<^ aj 2
r- a; c
coo
Forestomach tumors
22 survivors after
one tumor was a ca





in
r-

>,|
IO
4J P~
Ol 1
5' — •



in
O
•o
ID
er> 4-1
7.S







OJ
i/»
I


<£
- j-

. ,
CO
o



u
re in
= >.
o re
4J-D
in
Ol VO
S.O
O T
<4>
C
11 •»-
4-1 43
•s*
(/I
VI F—
II
^- c
p— re
"0.0
re CM

t.
01
4J —
c re 4->
0 X c
.^. Ol
in en E
•— C Ol
§•£ » re
Pulmonary adenoma to
27 survivors at 20
had alveologenic c
had hemangio-endot
fanales had inammar
2 pulmonary adenoni
35 controls

u
Ol

c re 4->
0 I C
^- Ol
in en E

»— u 01
•*- T! I-
O' — •


1"
in
CO >,
o-S
r^ 01
O






CM
Ol -~.
in «c
3 a
o a



































^_
O
"en
01
c
Ol
*>,
.c
Ol
^1
o
O-

TO
6-116

-------
CO
o
CM
CO

"-JC
*^?

3
SE
oo
2
o
C£.
4
>
g
z
0
•— 1
1—
g
CO
>— 1
JT

•
CQ

SET
o
Q.
U_
0
1—
o
^4
LJJ
CD
1

O
 ~c
ocg
VI
§(*•
0

J~
Ol

o
•c
Ol
*0>
c
VI
"*
"iS
u
o


"l3
LU

C
I"

^



a>

E
O) VI
3 '5
5~









^"
«M
O. %.
0 0


2 3
0 0

01 m
c i— vi
o •—

C C £
•— 10 -t-
c

§ 1 'c
"p 10 o
1. Ol
IO *•> U
u u a.

lo 1
w oj a.
i.
o
It-

O-
il ^J—~
4^ Q^ vl
io ai
01 ai ai
IO O X
*•> x tn
C i * .
^« ^ ^


c
Ol i—

1°
i*2 i
0 •—
in *j 
 i- C
•*• o.-—
L. a. u
3 10
v> -a o
Ol -C VI
O 3
•—wo
E «• =
VO fO
uo - o-
VI VI
01^:
C 01 -0



-^
^
ai

s
Ol
o
*->
lid"1




r— i
Ol
O
s*


C >t
01 U CC
'5 — C 01
01 10 o c
-Q E E o a)
— •— u o
T3 C 3 *J •—
2 10 a. on =

VI ^ VI ^ CO

= aj 5 c aj
C ^ ro 4-1
C O) O) «-
Ol 10 — > O
TJ 4J Ol Ol
• • e e in e
-= us re .a o
o c o c
10 VI OJ
S~ 10 ••- VI -O
= •— n c
4-> O IO = 10
in 4-1 £ S
OJ 10 C Ol
1- C.-O Ol O
O Ol (0 -0 •—
i*- ^ i 10 6












































t









i





























"o
Ol
Ol
c
Ol
•rr
^
01
o
0.
(O
6-117

-------
CO
CO
CM
 LU
  LU
  co

  CO
  CM
   0)
   ro
             <3 in
            Et
               I
                           o o o o   ooooo  oo
                                                  o
                                        jr-g>f~— ae —
                  eoooeeooooooo*"- o>
               .1
                   —.r-^-^—~-r—OOO
             0. O
              at- I •—
                §-
                                      OOOOOOOtA
                         is. o *~ CM ^" O
                       — »- CM —
                                            e-no

-------
  among 289 mice fed a control  ration,  none developed gastric tumors.   As would
  be  expected,  the  frequency of BaP-induced gastric tumors  was dependent both
  on  the  concentration of  BaP in the  food,  and  the  number of  days  it was fed.
       The  role  of  genetic factors  in determining tumor  response to orally
  administered BaP  in  mice was  subsequently explored  by  Rigdon  and Neal.284
  Since it  -is known that the  development of pulmonary adenomas  in certain
  strains of mice is genetically  influenced, a comparison was made between the
  incidence of pulmonary adenomas and gastric tumors and/or leukemia in CFW
 Swiss mice.  A ration containing 0.25 mg of BaP per gram of food was fed to a
 total of 108 mice for 80 to 140 days.   Their results are summarized in
 Table 6-24.  A large number of mice developed leukemia, even though the rate
 of spontaneous occurrence in this  strain  of mouse  was considered  to be very
 low.   Among leukemic mice a statistically significant increase in pulmonary
 tumor formation was observed,  whereas  a negative statistical  correlation with
 gastric  tumors  was seen  in this group.  Furthermore,  there was no relationship
 between  the occurrence of pulmonary  adenomas and gastric tumors in mice with-
 out  leukemia.
 6.4.7  Carcinogenicity of POM  Mixtures
     In  the environment,  man is  unlikely to come in contact with only a
 single POM, regardless of the  route  of exposure.   Instead, POM occur as
 complex mixtures in all environmental media.  Despite this generally accepted
 fact, very few studies have  been conducted on the carcinogenicity of defined
 POM mixtures.  The importance of studying the effects of POM in mixtures is
that it allows for the expression of possible cocarcinogenic, anticarcinogenic,
and tumor-promoting effects.
                                     6-119

-------
   Table 6-24.   RELATIONSHIP  OF  LEUKEMIA TO  LUNG AND STOMACH TUMORS
                IN 108 MICE FED  BENZO[a]PYRENE.284
                                     Number of mice with
Total no.
of mice
9
23
5
3
7
17
44
Leukemia
9
23
5
3
0
0
0
Lung
tumors
0
23
5
0
7
17
0
15 or more
lung tumors
0
12
4
0
3
1
0
Stomach
tumors
0
0
5
3
0
.-47
44
108
40
52
20
                                                                69
                                6-120

-------
       Among the most relevant studies conducted on the effects of POM mixtures

  were those concerned with the carcinogenic components of automotive engine

  exhaust.   Pfe1ffer285,286 treated groups  of 100 female NMRI mice with single

  subcutaneous  injections  of a mixture containing 10 noncarcinogenic  POM,  in

  addition  to.BaP  and/or DBahA.   The treatment  combinations  and dosages are

  summarized in  Table  6-25.   As  the  results  depicted in Table 6-26 indicate,

  increases  in tumor incidence could be attributed  to the  presence of  increased

  amounts of BaP and of DBahA.   It is  noteworthy  that,  at  the  lower dosages,

  DBahA was  more effective  in producing tumors at the injection site than was BaP.

 Moreover,  no effect of the  10 noncarcinogens on tumorigenic response was evi-

 dent.  Probit analysis of tumor incidence data indicated that the tumorigenic

 response from application of all 12 POM was attributable solely to DBahA.

      Similar studies intended to reveal  carcinogenic interactions among POM

 found in automobile  exhaust were conducted by Schmahl  and coworkers.287

 Eleven POM were selected  for their experiments,  and various combinations were

 applied to the  skin  of NMRI mice in a proportion based on their respective

 weights in automobile exhaust (Table 6-27).  Animals received twice weekly

 treatments  for  life (or until a carcinoma  developed).   Their results  (Table 6-

 28)  indicated that a  mixture of  carcinogenic POM was more effective than BaP

 alone,  and  that the whole  mixture (carcinogenic  plus noncarcinogenic  POM) was

 not  significantly more effective than the carcinogenic  POM group  alone.

 Thus, the carcinogenic effects observed were solely attributable  to the

 carcinogenic components of the mixture.

     Early  studies have focused on the carcinogenic activity of simple com-

binations of two POM,  each having different carcinogenic potency when admini-
             poo OQQ
stered alone.    '      Compounds were administered by subcutaneous injections to
                                     6-121

-------
Table 6-25.  Classification of Test Groups
                                          285
A



dose substance
(£g) '
A, 3.12 benzo(a)pyrene
A; 6.25
A, 12.5
AT 25.0
Ac 50.0
Ag 100.0
C
Substance C1
dose
*
benzo (e)pyrene
benzo (a) anthracene
phenanthrene
anthracene
pyrene
fluoranthene
chrysene
perylene
benzo (ghi) perylene
coronene
D
D, AT + BL
D, A, + B,
D-5 An + Bo

DT A? + Bl
DC Ac + B|

D6 A6 + B6
yag)
2.
3.
125.
31.
65.
28.
3.
0.
12.
3.










15
125
0
25
1
1
125
2
8
125









B


dose
%
B!
S3
C2
dose
(pg
4
6
250
62
131
56
6
0
25
6










.3
.25
.0
.5
.2
.25
.25
.4
.6
.25
E
u.l
.2
E?

4
ES

^6
2.35
4.7
9.3
18.7
37.5
75.0
C3
dose
(>jg
8
12
500
125
262
112
12
0
51
12









)
.75
.5
.0
.0
.5.
.5
.5
.87
.25
.5

?1
*"2
C-a

4
CS

C6





substance
dibenz (
dose
(^g)"
17.
25.
1000.
250.
525.
225.
25.
1.
102.
25.

+ DL

D3

* D4
* D?

+ D6

5
0
0
0
0
0
0
75
5
0









a, h) anthracene
C5 C6
dose dose
gag)
35
50
2000
500
1050
450
50
3
205
50










.5
.0
.0
.0
.0
.0
.0
.5
.0
.0










70.0
100.0
4000.0
1000.0
2100.0
900.0
100.0
7.0
410.0
100.0









                      6-122

-------
    (U
*r  o
t—  E
i—  ID

O>  M
f X)


£"
o +•>
 IV O
 .C
 •M C
    10

 «;^

1- Q.
 CU  3
ce  o
    O)
 u  c
 U  as

 CU  IQ
•o  o
                o

               o
                        <*- V)
                         0 S-
                        O 3
   00
vo
vo
  r- CM CO CM
                                       •er in 143 rs. ^o co
                                      \o oo vo
                                                  «n in
                                      00 *!^ r— 00 O"l
                                      ^r *s- vo  o>
                                                vo vo
                                                                    A3
                                                                    U

                                                                    g
                                                                   •o
                                     in
                                     co r*»
                                              t>. in o
                                     CM
                                             oo f>*in
                                             r- CO !-»
                                     in in ^- r-» r*. co
                                        co in in r*. oo
             CM in
             i— CM in o o o

             co vo CM in o o
                  •p- CM in o
                                                                   s-


                                                                   u
                                                                    u
                                                   6-123

-------
      Table 6-27.  Doses (pg) Applied in Dermal Administration Experiments,
                  in Relation to Benzo(a)pyrenez87
Controls

Acetone
Benzo(a)pyrene

C PAHa

Benzo(a)pyrene

Dibenz(arh)anthracene
Benzo(a)anthracene
       as
          solvent
j_j^.*itj»* \ *»/—•--	              A a         .1  •>
Benzo (b) f luoranthene        0^9.         £i=
        1.0
        1.0

        0.7
        1.4
        0.9
1.7
 1.7

 1.2
 2.4
 1.5
                  total
         4.0
                                          6.8
                               3.0
 3.0

 2.1
 4.2
 2.7

12.0
 NC PAH

 (Benza(a)pyrene

 Phenanthrene
 Anthracene
 Fluoranthene
 Pyrene
 Chrysene
 Benzo(e)pyrene
 Benzo(ghi)perylene
         1.0
 3.0
                                9.0
                        27.0)
total   65.0
                                        195.0
243.0
76.5
97.2
124.2
10.8
5.4
27.9
729.0
229.5
291.6
372.6
32.4
16.2
83.7
                               585.0
                                                                1755.0
 C  PAH + NC PAH
                                                      3.0)
  (Benzo (a) pyrene      	j-«	^	__1	
  Total C PAH
  Total NC PAH

  Total C PAH +  NC PAH
          4.0
         65.0

         69.0
  6.8
110.5

117.3
207.0
  Relation
   aPAH, polycyclic aromatic hydrocarbon
                                      6-124

-------














CO
CM

3
£

O
4-*
=
°-
o
o
•f"
4->
to
0
•f"»
Q.
Q.
,,_
• o
01

co
0)
£
ro
to
c
•c
c
•r-

CO
C\J
1
VD
O)
£






Cl
0
fQ f0
U E
•H o
fHIO •
QJu CQ
QJ«J J2
(0 JCQ RJ
MI
Q
V (0
J-> E e»P
<0 C
a>(o •
J= U W
4-> flj .Q
(^ fQ
(0

•""• (0 e»P
CQ g
olo
§fH
(OfH •
•fH Ol W
•o «j ja
di (0
n>
o
*^H
(7> 4)
O > <*>
fH H
0 *J
^W j IQ •
CO JD^ CQ
sic "S

0)
>
•H OJ
0 O *«
kU ^
sgi


•H CO

fH O 3
to t:


C
o
JJ
Applica


'H'''lH'-'ll|-HfH ^4




""* ' ' 1 1   VO CM
fH

.
SS2SSSSSSSS5 S S






oop»o3oocooocftSS00vO0^ fin
co oo oo co en en



i 1 1 1 1 1 I 1 1 1 1 1 i s
• •-if-ii— 1,— t iH »H iH fH fH



or>ri°4 ^ °*
WfflfflfflUOUZZZZOZ U2 CJ2





fll
\i/
.0
CQ

n
fH
(R
*u
4J
0
C
•H
fH
fH

CQ
0)
3
fH
>

&

O
E
CQ
CO
£
JJ

^ B
2 §
of
4} u
0) °
^
4J f

«W JJ
u-i re
0 §
•D ^
TJ o
C ^
o >,
w u
cu °"

^2 n
-T-
(y ^
« °-
CQ c«P
4J O
C 0
"o
a o
. *^
<0 -W
E C
"0 fH
CU (Q
•D >
(U 3
J= CT
§H 01
(C
6-125

-------
C57 black mice, and the resulting incidence of sarcomas used as the index of
carcinogenic activity.  The combination of benz[a]anthracene and chrysene
produced a tumor response which was more than additive; the combination of
benz[a]anthracene and DBahA produced less than additive effects; the combination
of DBahA and MCA produced a summation of response.  When a moderately active
carcinogen  (benz[a]anthracene) was  combined with  a powerful carcinogen, the
results varied from summation  effects  (with MCA)  to  no-less-than-additive
response  (with BaP  or DBahA).  The  combination  of a  strong  carcinogen  (DBahA)
with a weak carcinogen (chrysene) produced no summation,  and  the combination  of
 strong carcinogen (DBahA) with a noncarcinogen (anthracene, phenanthrene)  had
 no effect on tumor yield.
 6.4.8  .In Vitro Carcinogenesis Studies
      Bioassay systems to study the carcinogenic effects of chemicals in vitro
 are desirable because of their relatively low cost and rapid generation of
 results in comparison to in yjvo bioassays.  Cell culture systems for the study
 of  chemical carcinogenesis are still evolving, and  a  highly  reproducible In
 vitro assay system is  not yet available.
       The  published literature regarding  chemical carcinogenesis in  cultures  is
  vast, despite the  fact that  systematic studies were not  begun until the early
  1960's due to the lack of a  reproducible transformation  assay.   It  was  first
  demonstrated by Berwald and  Sachs290 in 1963 that polycyclic hydrocarbons (MCA,
  BaP) could cause the direct morphological transformation of hamster embryo
  cells in culture with a low rate of spontaneous transformation in untreated
  cells.  Transformed colonies have growth characteristics visually distinct from
  normal colonies and are readily seen  above  a  background of  normal  cells.  This
                                        6-126

-------
  assay can therefore be easily  used as a  screen to compare carcinogenic activity
  of suspect compounds.  A common feature  of these, and nearly all, transformed
  cells is that they give rise to fibrosarcomas upon inoculation into immuno-
  suppressed animals.  In addition to hamster embryo cells, malignant trans-
  formation has been demonstrated in organ cultures,  liver cell  cultures,  fibro-
  blastic  cells derived from mouse ventral  prostate,  3T3 cell  lines derived from
  mouse embryo  cells, and various types  of epithelial  cells from  animals.291'297
  In  general, these  test  systems  are effective  with much lower quantities of test
  material  than in bacterial  mutagenicity assays.   However, the very  low numbers
  of  transformed colonies which are  usually obtained, even with potent carcinogens,
  require a careful  evaluation of negative  control  data  for the occurrence of
  spontaneous transformation.  In addition, the possibility of contamination of
  test compounds must always be considered when positive results are obtained.
      Early reports by Berwald and Sachs298 and Dipaolo and Donovan299 described
 alterations in hamster embryo cells induced by BaP,  DMBA,  and MCA which could
 be used as indicators of a change from  a normal  to neoplastic state.   The
 compounds  were applied to  cells  in  culture either  dissolved  in paraffin and
 impregnated on filter disks  or as a colloidal  suspension  in growth medium.
 Following  marked cytotoxicity, foci  of  transformed cells developed which
 displayed  continuous proliferation  in vitro, chromosomal abnormalities, and  the
 ability to grow indefinitely  in  culture.   In addition,  these transformed mass
 cultures, when transplanted to 4- to 6-week old hamsters, continued to grow and
 form tumors.  A good correlation was obtained between in vitro carcinogenicity
of a polycyclic hydrocarbon and the number of transformed clones  they produced
The maximum rate of cell  transformation  in these studies was  25.6 percent  in
surviving cells, obtained by treatment with 10  ug/ml  of BaP for 6 days.  BaP
                                     6-127

-------
treatment at 1 ug/ml for 6 days produced 19.9 percent transformation in sur-
viving cells.  Further data indicating the activity of several polycyclic
carcinogens and their derivatives are summarized in Table 6-29.  It is sig-
nificant to note that the K-region epoxides of DBahA and MCA are more active in
the production of malignant transformation in hamster embryo cells than the
parent hydrocarbons or the corresponding  K-region phenols.300'30   Although
these results confirm the view that metabolism is necessary for carcinogenic
activity,  they conflict  with  data generated  in vivo  which  indicate that  K-
region  epoxides  of polycyclic carcinogens are less active  than the parent
compound in various species.   A possible reason  for  the lack  of  correlation  is
the relative instability of K-region epoxides as compared  to  the parent hydro-
 carbon when applied to the skin.  It is likely that in yjvo far less of the
 reactive K-region epoxide can survive passage through the skin to reach the
 basal cell layer.  Several investigators have also made it evident that the
 toxicity and transforming activity of POM are dissociable and occur by dif-
 ferent processes,302'303 with the toxicity being due to random alkylation of
 nucleophilic regions within  the cell.   However, when hamster  embryo cells are
 pretreated with weak chemical carcinogens which can induce microsomal enzyme
 activity (e.g., benz[a]anthracene,  methyl methanesulfonate,  ethyl methane-
  sulfonate) before the  addition of a potent  carcinogen  (e.g.,  MCA,  BaP,  DMBA),
                                             . 303,304
  transformation  may be considerably enhanced.
       As a prescreen for chemical carcinogens, cell  transformation in yrtro  may
                                                                          305
  be one of the most sensitive techniques available.   Pienta and coworkers
  reported that 90 percent (54/60) of the carcinogens they tested transformed
  hamster embryo cells in vvtro, whereas  none of the noncarcinogens tested
                                        6-128

-------
                             III   gggg  S8SS  555  g  555  gg |
  o
  g
  UJ
  LU
  oo
  CO

  o
 o

 §
 a.


 O LU
 OO LU
 a? o
LU
O O

ol

Q£ 00
CQ Z

5g
LU O
1—0
OO Qi
z: o
-
CM
 I
IO


0)


JQ
               »< ,0  I  o o o'  o' o o  o cv r^ V  ocve*^
               01 X

              •S =

               o —
              — o
              30.

                                                           la

                                                           0t
                    §


                                                                           II
                                                                           ,  s
                                                                            0*  *J  i. O»
                                            6-129

-------
showed any activity.  Moreover, many of the carcinogens which have not been
shown to be mutagenic toward S^ typhimurium in vitro (e.g., chrysene) were
capable of transforming the hamster cells.  Calculations have been made which
show that a battery of tests using S._ typhimurium (Ames assay), polymerase A-
deficient £._ coli, and hamster embryo cell transformation are capable of
detecting nearly all carcinogens tested, both POM and non-POM types.
     It is known that alteration of microsomal enzyme activity either jj}
vitro or ijn vivo can have a marked effect on the carcinogenic response to
POM.  Nesnow and Heidelberger306 reported that in 10T1/2CL8 cells,  a line of
contact-sensitive C3H mouse embryo fibroblasts, transformation in culture was
altered by chemical modifiers  of microsomal enzymes.   Pretreatment  of  10T1/
2C18 cells with benz[a]anthracene, a microsomal enzyme  inducer, caused a
doubling in MCA-mediated transformation.  Similarly, treatment with inhibi-
tors of epoxide h'ydrase (e.g.,  cyclohexene  oxide; styrene oxide;  1,2,3,4-
tetrahydronaphthalene-l,2-oxide) caused an  increase in transformation  over
that obtained  with  MCA treatment alone.   Thus,  it seemed  that  treatments
which  can  induce  epoxide-forming enzymes and/or  lower  the activity  of  epoxide-
degrading  enzymes will enhance the degree of  transformation  in cultured  cells
by altering steady-state  levels of oncogenic  epoxides.
      Chen  and  Heidelberger307'308  developed a system (which  is no longer used)
 using C3H  mouse ventral  prostate cells to examine transformation  by carcinogenic
 hydrocarbons under conditions in which no spontaneous  malignant transformation
 occurred.   Cells treated with MCA (1  Mg/ml) for 6 days in culture produced
 malignant fibrosarcomas in 100 percent of mice into which they were subcutan-
 eously injected.   When treated for only one day with MCA at the single cell
 stage, 100 percent of the clones were transformed to malignancy.   A good
                                      6-130

-------
   quantitative  correction was  obtained  between  the  in  vivo  oncogenic  activity
   of eight hydrocarbons  (including BaP,  MCA, DMBA, and  DBahA) and the  number of
   transformed colonies produced in this  system.  !„  contrast to the enhanced
   transforming ability of K-region epoxides relative to the parent hydrocarbon in
  hamster embryo cells, the K-region epoxjde derived from DMBA was less active and
  the K-region epoxides from MCA,  DBahA,  and benz[a]anthracene were more active
  than  the parent compound in mouse prostate cells.309-310  Moreover,  the
  epoxide derived from DMBA  was  more  toxic than DMBA  itself.   The  anomalous
  behavior of DMBA «y have  been due,  however,  to a decreased intracellular
  half-life of the epoxide because  of  its greater chemical  reactivity.
      Transformation  of  mouse,  rat, and  hamster  embryo  cells by organic
  extracts of airborne particulate matter has been reported.311'313  Cells
  preinfected with  RNA tumor viruses led  to accelerated transformation, as well
  as transformation in cells not inducible by crude extracts alone.   Rhim and
  coworKers    reported that the benzene extract had 100 to 1000 times the cell
 transformation  activity attributable to  the pure BaP contained in it.   In
 addition, the methanol  extract, although containing  on!y 1/30 as  much BaP
 showed  activity comparable  to the  benzene fraction.   Thus,  a diversity of
 carcinogenic material  (POM  or otherwise, is apparently  contained  in extracts
 of particulate  portions, a  conclusion  which  is supported by in vivo  studies
 (see Section 6.4).

     Attempts to  transform human cells in culture with POM (e.g., BaP  MCA
 DMBA) have generally met with failure.31"  However, „,,. and coworker,315 '
 ^ported that a human osteosarcoma clonal cell  line could be further trans-
formed in vitro  with DMBA.   Morphologic alterations and abnormal  growth
patterns became  evident in cells treated  with  DMBA  at 2.5 and 1.0  ug/.l ,„
                                     6-131

-------
the fifth subculture 52 to 57 days after exposure.   One of the altered cell
lines obtained from the 1 ug/ml treatment was tumorigenie in nude mice by
subcutaneous and intracerebral injection.  Interpretation of the significance
of these results is made difficult by the fact that an aneuploid sarcomatous
cell line had to be employed in order to demonstrate successful transformation.
     The use of organ cultures for the assessment of chemical carcinogenicity
suffers from the lack of reliable biochemical and morphological parameters
for measuring early neoplastic changes.  Nevertheless, pioneering work in the
application of organ culture to chemical carcinogenesis was performed by
Lasnitzki.316  Microgram quantities  of MCA added to organ cultures of rat and
mouse  prostate fragments caused extensive hyperplasia  and squamous metaplasia.
However, these preneoplastic morphological effects  are generally  not  asso-
ciated with  subsequent  tumor  development when carcinogen-treated  pieces  of
tissue are implanted into  host animals.291   Limited success  has been  achieved
with organ cultures of rat tracheas, which  showed  characteristic  morphologic
 alterations  when treated with DMBA,  BaP, and MCA.291   In addition,  Crocker
 has exposed respiratory epithelia from the  hamster, rat, dog, and monkey to
 BaP at 7 to 15 ug/ml and observed occasional squamous metaplasia.  More
 commonly,  pleomorphic cells in a dysplastic epithelium were evident as a
 result of the treatment.  Using this system, it was also possible to demon-
 strate a protective effect of vitamin A against BaP-induced abnormal
 differentiation.  It has been suggested that rat tracheas maintained in organ
 culture may be a useful system for  the predictive  screening of potential
 carcinogens.
             318
                                       6-132

-------
      A unique organ culture technique has recently been reported in which BaP
 (4 or 12 mg) was administered to pregnant mice (strain A and C57 Bl), and
 lung tissue of their 19 to 20-day-old embryos was subsequently explanted in
         319
 culture.      A transplacental influence of BaP was manifested as a proli-
 ferative stimulus in embryonic lung tissue.   Hyperplasia arising in the
 bronchial epithelium led to the development of adenomas in a large percentage
 of the explants.
 6.4.9  Dose-Response Models in Carcinogenesis
      Published in the early 1940's,  Bryan and Shimkin's320 paper is a good
 summary of the "state of the art"  in the  design and  analysis of experiments
 with  carcinogenic agents.   The development of tumors  at local  sites of
 injection with varying  doses of carcinogenic  hydrocarbons  (MCA,  BaP,  DBahA)
 into  mice or  rats was characterized  in terms  of:   (1)  latent periods  between
 injection and  appearance  of tumor, and (2) incidence  (percent)  of tumor
 bearing animals.
      Assuming  that the  logarithms of minimal  effective  doses  for individual
 animals are normally distributed, an S-shaped cumulative normal incidence
 curve is  expected as a function of log dose.   Consequently, the inverse
 transformation to probits is expected to fit a straight line.  The authors
 presented data which fit this well, at least when doses were not extremely
 high.   Average latent periods, on the other hand, appeared as a linearly
decreasing function of log dose until a flat minimum was reached at high
doses of carcinogen.
     Since departures from parallelism in bioassays defy interpretation in
terms  of simple relative potencies,  the authors recommended comparison of
                                     6-133

-------
doses and mean latent periods at the 50 percent level  of incidence.   Other


specific problems in the design and analysis of experiments in carcinogenesis


were discussed:  species and animal variability, planning of sample sizes,


duration of observation, and competing risks of death.

     The principal focus of this paper was on demonstrating the usefulness of


formal statistical methods.  No concern was expressed about the low-dose


validity of the incidence and latency models, and no attempt was made to


extrapolate to human populations.  In the context of the times when this


article was written, however, it is quite understandable that the authors had


to set their  sights on  more modest objectives.
                                     op!
      In 1953,  Heston and Schneiderman    referred briefly  to the initiation-


promotion  theory  but subsequently  focused on  the question  of whether one  or


more somatic  gene mutations are  involved in carcinogenesis.  Specific refer-


ence was made to  earlier experiments by Charles and Luce-Clausen who produced


lung papillomas  in  mice painted  repeatedly with BaP;  the square roots of


numbers of tumors per  mouse were found to fit a linear  relationship with


dose.   If  this had  been the  result of  single  short-term exposures  to carcin-


ogens,  it  would suggest a  two-step recessive  mutation mechanism.   However,


the confounding of  a long-term dose schedule  with  latency  made  interpretation


 unclear.   Heston and Schneiderman therefore designed  a  new experiment  using


 2-month-old strain  A mice  given a single  injection of 0,  0.1,  0.2,  0.3,  0.4,


 or 0.5 mg DBahA and sacrificed six months  later.

      The average numbers  of tumors per mouse in the latter experiment  exhi-


 bited a good fit to a straight line dose  response and a poor quadratic fit,
                                      6-134

-------
  thus  providing  greater  support  for  a  one-step  dominant  than  for  a  two-step

  recessive gene  mutation mechanism.


       The mean number of tumors  at zero dose exceeded the  linearly  extrapo-

  lated value.  This discrepancy  from linearity was attributed by  the authors

  to a  combination of natural carcinogenic processes active at zero  dose and

  deactivation or other pharmacological factors operating to discount nominal

  doses to lower effective levels.  The authors mentioned a "threshold" dose of

 0.03 mg under the experimental circumstances.   However,  they quickly rejected

 this idea in favor of speculation that the dose-response curve may have

 convex curvature in the 0 to 0.1 mg range of DBahA,  presumably because of

 variability  among the thresholds of individual  mice.


      Mantel  and  coworkers322 considered the consequences of assuming a

 linear rise  in expected  numbers  of tumors elicited by  injecting a carcinogen

 into  a mouse at  doses  above some threshold level.  The authors  interpreted

 the  slope of the line  as indicative  of the animal's  sensitivity to  the

 carcinogen,  once the dose exceeds  the  threshold  level.   It was  shown that,  in

 a population of  mice having variable dose thresholds and variable sensi-

 tivities, the average number of  tumors per  mouse remains a linear function of

 dose at  levels above the maximum value of the mouse thresholds.   It was shown

 that at  doses lower than this the relationship is convex.


     A weighted  least squares fit to a replicate experiment with  DBahA gives

 additional support to the linearity result  reported previously by Heston and
             321
 Schneiderman.     While  the authors also demonstrated a linear range of dose

 response with intraperitoneal urethane, deceleration at very high doses

 occurs, presumably as the result of competition between,  or coalescence of,

multiple tumors.
                                     6-135

-------
     More important from the standpoint of safety regulations,  it was  shown
that extrapolation of the straight portion of the dose response for groups of
mice could give a biased estimate of the average threshold—an  artificially
high one if mice sensitivities are positively associated with their thresholds,
but artificially low in the more likely case of decreasing sensitivity with
an increasing threshold.  In the latter case, a linear extrapolation would
provide a conservatively low estimate of the average threshold.  However, one
may criticize the use of an "average threshold" as a parameter for safety.
It is not ideal for such a purpose, since it does not directly indicate the
proportion of individuals at risk of tumorigenesis at extrapolated low dose
levels.
                                                    323
     More recently, in  1976, Yanysheva  and Antomonov    emphasized the
importance of,  but really provided  no principles  for, choosing animal species
and  schedules of  exposure which  might be  appropriate  for  extrapolating to
permissible  levels in man.  They,  like  others,  were forced  to  use  highly
 informal  judgments in choosing their experimental conditions.
      From 16 to 40 random-bred white rats at each of  several doses received
 intratracheal  BaP at monthly  intervals  for up to  10 months.  Total  adminis-
 tered doses  ranged from 0  to  25  mg. Presumably (though this is  not entirely
 clear) each rat was  observed  throughout its entire  life in  order to determine
 the age at onset, the location,  and the histological  type of the first
 malignant or benign  tumor.   The authors do not say  how they determined the
 age at onset, and it is difficult to imagine that they could be very  accurate
 in that endeavor unless special  attention were given to the problem.
      The reported lifetime incidences  of tumor-bearing rats vary from 9.5 percent
 (non-lung sites) in controls and 0 percent in those at low dose levels of
                                      6-136

-------
 benzpyrene, to over 80 percent reported after 25 mg.  Responses appeared as a


 concave increasing function of untransformed doses or as a convex function of

 log doses.


      The reported "time to occurrence" of first tumor appeared to be the


 average of onset times of first tumors for affected rats at a given dose


 level, but the authors were not explicit about this.   Generally, these times


 decreased with increasing doses in the authors'  experience.


      By a crude empirical fit, the expected occurrence time of the first

 observed tumor in an animal  was given as:
                            T-L02U.T6.79
 where T = average first appearance time and d = total  dose (mg).   Extrapo-


 lating from this, the  authors  estimated that a dose  of 0.02 mg would  cor-


 respond to first  tumors at  an .expected  time of 67.9  months.   Since this


 exceeds the lifespan limit  of  such rats,  it became the authors' recommended


 dose.   One fault  with  this  approach  is  that,  even when correctly  indicated,


 the expected time of first  tumor onset  is merely an  average;  many  tumors


 could  first appear within the  lifetimes of  members of  the  species,  even if


 the extrapolated  average exceeded  the lifespan  limit.  Another major diffi-


 culty  is  lack of  clarity as to whether, or  how, the  experimental experience

 should  be  extrapolated  to man.

              324
     Cornfield     provided a valuable review of approaches to the regulation


 of involuntary exposure to harmful  substances.  Particular attention was


 given to carcinogens in food.   In addition,  he examined a formal kinetic


model  which suggests the existence  of a threshold,  or near-threshold,  in the
                                     6-137

-------
dose-response curve.   Cornfield recognized the governmental  regulatory
principle of balancing risk against benefits, but the subsequent discussion
focused on problems relating to risks as expressed by lifetime cancer
incidences at low doses.
     A regulatory rule-of-thumb is cited324  by which the no-observed-effect
level (NOEL) is  divided by  100 to set an acceptable daily intake (ADI)
tolerance  specification for humans.  Another technique  is to  extrapolate
downward from doses  that  elicit observable lifetime  incidences  so  as  to reach
doses expected  to produce arbitrarily  low incidences  of the order  of  10
to 10"8.   The  difficulty  cited here is  in the diversity of. results,  depending
 on the  choice  of models which perform comparably well in ranges of observable
 responses.
      The theoretical case for low dose linearity of response (down to 0
 response at 0 dose) is favored by:  (1) its conservativeness compared with
 probit model extrapolation; (2) an approximate low dose linearity in the one-
 hit Poisson-like model of  carcinogenesis; and (3) by results of all too
 few epidemiological studies of large groups of people  exposed  to  low doses
 (references by the  author are to H.A.  Kahn, "National  Cancer Institute
 Monograph No.  19,"  1966, and  J.  Cornfield et al.,  "Proceedings of the First
  International  Toxicology Congress,"  in press).
       At least when using formal  models, Cornfield argues  against  the practice
  of mingling informal  conservatism with extrapolated incidence  expectations,
  since this may overly weight the risks of  harm against offsetting benefits.
  A similar argument also would apply to using pessimism or optimism factors
  when estimating benefits.
                                        6-138

-------
       Cornfield presented a kinetic model  which yields steady state equations
  with interesting implications for dose response curves.   The steady state
  solutions  indicate:
       1.    No  toxic  reaction  in  a  host  as  long  as  the  quantity (moles)  of
  toxic  substance  is  less  than  the  quantity (moles)  of  host deactivator,
  provided that  the host deactivation process  is  irreversible.
      2.    No threshold,  but an  approximately linear dose-response  at low  dose
  levels, provided that the host  deactivation process is reversible.
      3.   That any irreversible component in a vertical chain of deactivation
 steps results in a threshold.
      4.   That any incremental step in a chain of reversible deactivation
 components  lowers the linear low dose slope..
      One assumption  in the underlying kinetic model which may not prove to be
 correct is  that the  lifetime probability of a toxic effect is said to be
 proportional  to the  steady state quantity of activated carcinogen.   Another
 complication,  established clearly  by  Mantel  et  al.322  and explicitly recog-
 nized by Cornfield,  is that  individual  thresholds  (or  near-thresholds)  do  not
 preclude convexity in  the dose-response for  groups  of  animals  because of
 variability in  individual  thresholds.
      Finally, and perhaps most important,  the formal results apply  only  to
 steady-state conditions.   Even in  irreversible deactivation processes, there
 may be extended periods during which activated carcinogens can constitute a
 risk of  harmful effects prior to the host achieving a steady state.  A zero
 risk achieved in the steady state would apply only after the host had sur-
vived an initial period at risk.   It further seems realistic to suppose  that
repeated pulses of carcinogenic exposures throughout life  would complicate
                                     6-139

-------
the dynamics of the process.   Thus, the steady-state solutions of this
formulation may not be particularly useful in establishing new rules for
acceptably low doses of harmful substances in the human population.  Perhaps
a far greater potential for this kind of model may be in stimulating new
investigations of variability  in low dose thresholds in very  large groups of
different  species and  strains  of experimental animals, possibly  leading to
the identification  of  endogenous deactivation mechanisms  for  carcinogenic
substances and,  consequently,  the  raising of thresholds  in  those species
having increased capacity for carcinogen deactivation.
      Unfortunately, the application of animal-derived dose-response relation-
 ships to human situations is not likely to  provide reliable estimates of
 cancer risk.  Assuming that the incidence of carcinoma is proportional to
 steady state levels of activated carcinogen, the dynamics of this process
 would be  dependent on the species employed  and the target tissue examined.
 Wide variation  in  both species susceptibility and metabolite kinetics are
 known to  exist, and would require the  compilation  of numerous data  sets to  _
 provide meaningful results.   For  human risk assessment,  reliance on epidemi-
 ologic  data,  coupled  with the apparently conservative assumption of low dose
  linearity of response,  presently  appears to be  the most practical  approach.
  6.4.10  Synergistic and Antagonistic Interactive Effects
       It is well-known that the development of polycyclic hydrocarbon-induced
  tumors can be altered by:   (1) components   in the diet,  (2) agents which
  affect the activity of certain enzymes, (3) other co-administered noncarcino-
  genic or weakly carcinogenic chemicals, or (4) the vehicle  used to deliver a
  carcinogen to  experimental animals.   These influences can have  an  important
  effect on  possible threshold levels for tumorigenic response,  although the
                                        6-140

-------
  .echanism of such interactive effects is unknown.  These factors further compli-
  cate the extrapolation of experimental animal data to human situations.
       Early work on the inhibition of carcinogenesis was reviewed by Falk and
  coworkers.32*  In ada,tion,  tney conducted a series Qf ^.^ ^ ^^ ^

  carcinogenic action  of MCA,  DBahA,  and BaP in the presence  of closely related
  POM.   Simultaneous administration of DBahA or MCA (30 ug) with 15 times the
  .olar equivalent  of  their respective dihydro- or  hexahydro-reduction  products
  by  subcutaneous injection to male C57 black mice  caused either a  complete
  inhibition or dramatic  reduction  of  the normal carcinogenic response.   Admin-
  istration of the dihydro-  and hexahydro-derivatives at intervals extending
  from  14 days prior to and  seven days  subsequent to DBahA (60 or 135 Mg)
  injection also caused significant inhibition of tumor development.  The in-
 jection vehicle (tricaprylin or ethyl laurate) also affected the inhibiting
 response.   Injections of the noncarcinogen phenanthrene together with DBahA
 in molar ratios  of i:24 and 1:48 significantly reduced tumor yield.   In
 simulated  environmental  studies  BaP was administered to mice together  with
 various  noncarcinogenic  POM commonly  found in polluted atmospheres.  The
 results  (Figure 6-16)  demonstrated that in all  cases a marked  inhibition of
 carcinogenesis resulted.   Neutral  fractions  isolated from polluted urban
 atmospheres produced similar inhibitory effects with BaP.
     The experiments of  Falk and coworkers325 were modified and  repeated by
 Pfeiffer who obtained different results (see also Section 6.4.7).  Ten non-
 carcinogenic polycyclic aromatic hydrocarbons found in automobile exhaust
were tested in combination with BaP (3-100 Mg) and DBahA (2-75  Mg) by sub-
cutaneous injection in 3000 female NMRI mice.  The tumor incidence resulting
from all 12 compounds  being administered together could be  attributed to the
                                     6-141

-------
  o
  *T
  to


  §
CARCINOGEN 400MgBENZO(a)PYRENE


VEHICLE: TRICAPRYLIN

DURATION 15 MONTHS                     t.~.~*i

ANTICARCINOGENS ENCOUNTERED IN AIR POLLUTION
             BENZO(a)FLUORENE 0.1:1
               PERYLENE 0.1:1
              PERI NAPTHOXANTHENE 0.1:1
              BENZ (a) CARBAZOLE0.15:1
              CHRYSENE 0.15:1
                BENZO(k)FLUORANTHENE 1:1
              BENZ(m, n, o) FLUORANTHENE 1:1
               2-NAPHTHOL5:1
             ANTHRACENE : PHENANTHRENE : PYRENE 10:10:10:1




             ..il	1	L—J	1	1
          0  10 20 30 40 50 60 70 80 90


                PERCENT T. B. A.
Figure 6-16.  The inhibiting effect of hydrocarbons found in polluted urban

air and cigarette smoke in concentrations approximating those occurring

naturally when injected in combination with BaP.
                               6-142

-------
  presence of DBahA, with little influence from BaP or the other ten chemicals.
  No inhibitory effect of the ten noncarcinogens was evident; moreover, an
  increased tumor yield resulted from injection of mixtures containing in-
  creasing amounts of the components.   This effect, however,  was less dramatic
  than if BaP were administered alone,  and paralleled the dose-response curve
  of DBahA acting singly.   It is not known how the  relative tumor  suscepti-
  bility of NMRI  mice,  as  compared  to C57  mice used by Falk,  may have affected
  the  results.

      Many studies on  cocarcinogenesis have been concerned with the  identifi-
  cation of tumor-accelerating substances present in cigarette smoke.  These
  compounds are generally tested for cocarcinogenic activity by repeated
 application to mouse skin together with low doses of BaP.  A positive res-
 ponse would be obtained in cases where the tumor yield of the combination
 exceeds that produced by either agent  alone at the same doses.   Van Duuren
 and coworkers206>326,327 establ1shed ^ g pronounced C0carcinogen1c effect

 could be obtained with catechol and the  noncarcinogens pyrene,  BeP."and
 benzo[g,h,i]perylene.   Doses of 2000,  12,  15,  and  21  Mg,  respectively,  were
 applied 3 times  a week for  52 weeks to female ICR/Ha  Swiss mice.   Each  animal
 also  received 5  Mg of  BaP in 0.1 ml acetone with each  dose of test substance.
 Although  phenol  has been regarded as a tumor-promoter  in  the,two-stage
 carcinogenesis system, this  compound had a slight  inhibitory effect on BaP
 carcinogenesis when administered in combination.  These results therefore
 indicated that tumor-promoters and cocarcinogens may not have the same mode
 of action, and that the two terms should not be used interchangeably.  Other
 POM's (e.g.,  fluoranthene, pyrene,  pyrogallol) also possess cocarcinogenic
activity but have no tumor-promoting activity.  Additional studies by Schmeltz
                                     6-143

-------
and coworkers328 established that most of the naphthalenes found in cigarette
smoke have an inhibitory effect on skin tumorigenesis (250 M9, 3 times a
week) as induced by BaP (3 ug, 3 times a week).   On the other hand, several
of the alky!naphthalenes tested (dimethyl-, trimethyl-, tetramethyl-) en-
hanced the carcinogenic activity of BaP on mouse skin.
     Numerous investigators have shown that antioxidants are effective
inhibitors of POM-induced tumor development.   This action has been demon-
strated with selenium,329'331 dl-crtocopherol (Vitamin E),    '    and
ascorbic acid332  in mice treated with DMBA and croton oil.  The carcinogenic
                                                                         333,334
 action  of MCA has  been  reduced  by  tocopherol-rich  diets  in  rats and mice.
 The antioxidant food additives  butylated  hydroxytoluene  (BHT), ethoxyquin,
 and butylated hydroxyanisole (BHA) have inhibited  lung,  breast, and gastric
 tumor formation induced in rats and mice  by various  carcinogens in the
 diet.335"337  The sulfur-containing antioxidants disulfuram,  dimethyl-
 dithiocarbamate, and benzyl thiocyanate inhibited  DMBA-induced mammary  cancer
 in rats when they were added to the diet; in the mouse,  disulfuram prevented
 the formation of forestomach tumors induced by BaP in the diet, but  had no
 effect on BaP-induced pulmonary adenoma.338  The agricultural herbicide,
 maleic hydrazide, and its precursor, maleic anhydride, can inhibit the
 initiating activity of DMBA in the mouse skin two-stage carcinogenesis
        339
 system.
      Investigators have attempted to explain the mechanism for the anti-
 carcinogenic  action of antioxidant chemicals.  Chromosomal breakage in human
 leukocyte cultures  induced  by  DMBA was reduced by simultaneous addition of
 ascorbic acid (31.7 percent reduction),  butylated hydroxytoluene
                                       6-144

-------
                                       340,341
  (63.8 percent),  Na2Se03 (42.0 percent), or dl-a-tocopherol (63.2 percent).



  It was not apparent from this study whether anticarcinogenesis by antioxidants



  may be due to a  direct protective effect against chromosome damage or to an



  interference  with  metabolic  activation  of carcinogens,  although the latter



  seems  most likely.   This  aspect was  further pursued  by  Rahimtula and coworkers



  who  examined  the abilities of several antioxidants to affect BaP hydroxylation



  by rat  liver  microsomal mixed-function  oxidases.  Their results  indicated



  that antioxidants can markedly inhibit  BaP  hydroxylation by an apparently



  direct action on microsomal oxidation mechanisms.  Furthermore,  all of the



  antioxidants tested reduced the bacterial mutagenicity of BaP in the presence



 of rat liver microsomes and cofactors (Table 6-30).   The authors suggested



 that antioxidants may exert their protective effect in v|vo by inhibiting the



 formation of carcinogenic intermediates  from POM.  This  conclusion, however,



 seems to conflict with data indicating that inducers  of  increased BaP hydrox-



 ylase activity can  also inhibit tumor formation.343   However,  flavones are



 also  inhibitors of  BaP metabolism  in  vrtro,  thereby indicating  that their



 specific effects  depend  upon  how and  where they  are used.   These  investiga-



 tors  found  that several  synthetic  and naturally  occurring flavones,  when



 incorporated in the  diet  (3 to 5 mg/g) or  applied to  the skin, caused  a pro-



 found increase in BaP  hydroxylase  activity  in the  small intestine and  skin,



 respectively.  In addition, pulmonary adenoma formation resulting from oral



 administration of BaP was totally prevented and skin tumors initiated by BaP



 application to mice were significantly reduced (>50 percent) by treatment



with the synthetic flavone, p-naphthoflavone.  Pulmonary tumor formation was



also reduced 50 percent by incorporation  of the.naturally occurring flavone,
                                           342
6-145

-------
     Table  6-30.   EFFECT OF  ANTIOXIDANTS ON BENZO[a]PYRENE METABOLISM
                   AS  ASSAYED BY  Salmonella typhimurium  (STRAIN TA98)
                   MUTAGENICITY iFTHE^ENCElTW^LIVER MICROSOMAL
                   FRACTION AND NADPH34^
       Antioxidant
     •HI*
None
Butylated hydroxyanisole
Ethoxyquin
Pryogallol
Glutathione
Concentration
   «••—••»

   100 uM
    25 uM
   100 yM
      1 mM
                                                       Revertants of strain
                                                          TA 98 per plate
384
193
148
156
 92
                                       6-146

-------
  quercetin pentamethyl ether, Into the diet.  The possibility that only specific



  components of the drug-metabolizing enzyme system may be induced by anti-



  oxidants has not been fully explored.   Sullivan and coworkers344 recently



  demonstrated that BHA,  BHT, phenothiazine,  phenothiazine methosulfate,  and



  ethoxyquin can  all  reduce  the  quantitative  yield of BaP metabolites  in



  incubations  with rat  liver microsomes.   On  the  other hand,  these  antioxidants



  did not  react with  6-oxy-BaP free  radical,  although some  reduction in the



  enzymatic  in vitro  formation of 6-oxy-BaP free  radical  could be obtained.



  The authors concluded that  the antioxidant  effect on BaP tumorigenesis is



 probably not due to a direct reaction with  BaP  free radicals.




      Benson and coworkers345 recently provided evidence suggesting that the



 protective effects of antioxidants are mediated, at least in part, by an



 elevation of glutathione S-transferase activity.  The incorporation of BHA



 or ethoxyquin in the diet of mice  greatly reduced the conversion  of BaP  to



 •utagenlc metabolites  that  were subsequently excreted in the urine.   In  both



 rats and  mice, BHA in  the diet  increased  hepatic glutathione S-transferase



 specific  activities.   The mutagenic activity of  BaP  metabolites in the urine



 could be  reduced  by  the  addition of glutathione  together with either  purified



 glutathione S-transferases  or liver cytosols from rats or mice fed on BHA,



 containing diets.  The authors cited several  studies indicating that various



 epoxide metabolites of POM are substrates for conjugation with glutathione.



These studies are considered as important supporting evidence for the ro!e



of conjugation with glutathione in  the detoxification of carcinogenic  POM.
                                     6-147

-------
     In addition to flavones,  other naturally occurring compounds  have
exhibited protective effects against ROM-induced tumor formation.   Vitamin A
has clearly been shown to play a role in reducing carcinogen-induced
tumors.345"349  Nettesheim and Williams350 recently undertook to determine
whether inadequate vitamin A consumption may predispose individuals to
carcinogenesis, or whether increased vitamin A intake exerts a protective
effect against  neoplasia.  They found that a diet deficient  in vitamin A
increased  the formation  of MCA-induced  metaplastic lung nodules in  female
Fisher 344 rats,  even though  adequate amounts of the  vitamin were  stored  in
the liver.  On  the other hand,  moderate amounts  of vitamin A added to  the
diet markedly reduced the development of MCA-induced  lesions of the lung.
 High doses of the vitamin given intragastrically provided no additional
 protection, however.                     ,
       Further -studies on naturally occurring antineoplastic compounds were
 recently  reported by Wattenberg.351  Benzyl isothiocyanate and phenethyl
 isothiocyanate,  both  found in cruciferous plants such as cabbage, brussel
 sprouts,  cauliflower, etc.,  inhibited  DMBA-induced mammary  cancer  in Sprague-
 Dawley  rats.   When  added to  the diet together with DMBA, these compounds
 inhibited the  development  of forestomach tumors and  pulmonary  adenomas in
 female  ICR/Ha  mice.   Similar anticarcinogenic  actions were  obtained when BaP
 was incorporated into the  diet.   These results lead  to interesting specula-
  tion regarding the role and importance of  diet in human susceptibility to
  environmental  carcinogens.   In cases  where dietary constituents  can alter the
  metabolism of xenobiotics such as ROM's, the anticarcinogenic effect may
  result from an  alteration of steady state levels of activated versus detoxi-
  fied metabolites.
                                        6-148

-------
      Studies have shown that not only can specific substances in the diet
 affect the response to carcinogens,  but protein deprivation in genera!  may
 also  decrease the activation of  carcinogens.352  The  feeding of protein-
 deficient  diets  to male mice caused  decreased  liver weights and reduced
 cytochrome P-450 content in  the  total  liver.   Diets deficient  in both protein
 and choline  produced even further reductions in  liver weight and cytochrome
 P-450 content.   Liver microsomes isolated from these animals displayed  a
 decreased  ability to activate dimethylnitrosamine to a mutagen  (in the Ames
 Salmonella test system), which paralleled the reduction in cytochrome P-450
content produced by the diet.  Conversely, the inactivation of the direct-
acting (ultimate) carcinogen N-methyl-N'-nitro-N-nitrosoguanidine was reduced
in liver microsomes from mice receiving a protein-deficient diet.
                                    6-149

-------
6.5  MUTAGENESIS

     It is presently believed that fundamental similarities exist between the

molecular mechanisms of mutagenesis and carcinogenesis induced by POM.  These

assumptions are based on the theory that highly electrophilic metabolites

(e.g., epoxides) are generated  from POM by the action of the MFO system  (see

Section  6.2).  These reactive intermediates  are capable of  binding to DNA

and/or other  critical cellular  macromolecules (see Section  6.2.6) presumably
                                                        o c o
to initiate a mutagenic or carcinogenic  transformation.      The  concept  that

 carcinogenesis is an expression of an alteration  in the genetic  material of a

 treated cell  (i.e., somatic mutat-ion) implies that a relationship exists

 between mutagenesis and carcinogenesis.133  Therefore, it is believed that an

 investigation of the mutagenicity of POM:   (1) may be predictive of carcino-

 genic potential, (2) may .help  elucidate the mechanism for malignant trans-

 formation  in  certain systems,  and (3) may indicate the presence of a potential

 threat  to  human  health posed by a particular POM.

 6.5.1   In  Vitro  Studies

 6.5.1.1 Microbial systems-Early attempts  to demonstrate  mutagenicity  by POM

  failed because of the  inability of microorganisms to metabolize these com-

  pounds to reactive derivatives.  The addition of subcellular preparations and

  the development of sensitive tester strains of Salmonella typhimurium have

  resulted in a mutagenesis test  system (Ames assay) now receiving wide appli-
                                                                          354
  cation as a powerful  screening  tool for evaluating environmental agents.

  All  of these strains  are histidine-dependent and can be reverted to histidine
                                        6-150

-------
   independence by mutation.  Mutagenicity in the Sajmonena/microsome system
   displays a high correction with carcinogenicity for many chemicals, although
   exceptions are known to occur with POM (see Section 6.5.3).   Teranishl and
   coworkers355 found that a quantitative correction existed between carclno-
  genicity and mutagenicity of seven polycyclic hydrocarbons when strain TA1538
  was tested using rat liver enzyn.es.   In discussing the  significance of the
  Salmonella/microsome assay,  McCann  and Ames356 believed that their results
  supported  the somatic mutation  theory  of cancer, and that  a  common link between
  mutagenesis and carcinogenesis  by chemicals is the  ability to produce DMA  damage.
      In Japan, a data base has  been developed on 60 chemicals (POM and  non-
  POM) which correlates mutagenic activity in various strains of S,  typhimurium
  in vitro with carcinogenicity in vivo.3"  With the use of an activating
  system from liver microsomes and the introduction of the R-factor plasmid into
 S, typhimurium for increased sensitivity,  a strong correlation between muta-
 genicity and carcinogenicity was obtained.   Studies conducted in the United
 States  on 102  compounds  tested in vitro using  S, typhimurium  generally confirm
 the high capability of this  system  to detect carcinogens.358   !„ addition,  a
 further Improvement in the detection of carcinogens  could be  obtained by
 combining the results  from Salmonella with  those from polymerase  A-deficient
 !. coll.  On the other hand, the use of s,  typhimurium In vivo (host-mediated
 assay) was considerably less reliable for the detection of carcinogenic  hydro-
 carbons and aromatic amines.358
     In recent studies, the Ames Salmonella test system has been employed to
assess the mutagenicity of organic extracts of airborne particulate pollutants.359
Although the total  extract displayed marked mutagenicity, assays  of various
                                     6-151

-------
subtractions of the extract indicated that much of the activity is due to
compounds other than polycyclic hydrocarbons and do not require meta-
bolic activation.  These results, which are supported by data from in yjvo
studies (see Section 6.4), indicate that caution must be exercised in attri-
buting the mutagenic/carcinogenic effects of air pollutants solely to the
presence  of polycyclic  hydrocarbons.  The recent demonstration of the muta-
genicity  of quinoline and its  5-hydroxy-  and 8-hydroxy-derivatives in the
Ames assay  emphasizes the potential  role  of nitrogen-containing  ROM's as
                          ocn
 hazardous air pollutants.
      Several  years ago  a joint program was undertaken by scientists  at the
 National  Institute of Arthritis, Metabolism, and Digestive Diseases, and  at
 the Department of Biochemistry and Drug Metabolism of Hoffmann-LaRoche, Inc.,
 to determine the structures of POM metabolites that are responsible for their
 mutagenicity and carcinogenicity.  It had  been previously demonstrated that
 epoxides of carcinogenic  POM  are frameshift mutagens.361  In  1975,  it was
 established that the K-region BaP 4,5-oxide362 and  non-K-region BaP 7,8-
 diol363  (with metabolic activation)  were more mutagenic than  the parent
  compound in  S, tvphimurium strains  TA100 and  TA1538.   Subsequently, nearly
  all of the possible primary oxidative metabolites of BaP were synthesized and
  tested for mutagenicity.364-368  The results  of these tests have been recently
  reviewed48'64>65-and are summarized in Table 6-31.   It was emphasized that diol
  epoxide-1 ( (±)-7p,8a-dihydroxy-9p,10-pePoxy-7,8,9,10-tetrahydro BaP) is a very
  potent  mutagen in S,  tvohimurium, and that the BaP 7,8-diol-9,10-epoxides are^
  likely  candidates as  the ultimate biologically reactive metabolites of BaP.
  It was  subsequently demonstrated by these researchers  that the mutagenicity  of
                                        6-152

-------
  OO
  UJ
  > 10
  >-" 00
  CC _l
   o
Q.

car*.
 t- CC
 C£ UJ
 •-• t—
 ?= CO
    oo
              U
              re
               O>
 UJ «C
 0
 «C CO
 a:
 UJ
                     en
                     CO
                     en
                     2
                    4->
                    OO
                             O O O
                             ^3- O «3-
                                         °   •"
— ° <=>
       V
                                        o o
                                         VVVVV
                               \r> ITS
                               co en
             O  Or-,
             I—  CM
      in in in i— ,— ,— ,—

     > •— o o d d d d
                V  V  V  V
OO i—
•-H  O)


II
o  re
o oo
CO

vo

o>
 re
                JO
                 TO



                  O
                  O.
                                                                        Ol

                                                                        O i—
                                                                        c  o
                                                                       •r- .*-

                                                                    Q.  3 TO
                                                                    re  CT O
                                                                    CO  I   S.
                                                                    o CM TJ
                                                 I
                                           •  "CM
                                          I  I i—
                                         O IT)  •>
                                                                     I
                                                                   CD
                                                                            I
                                                                           o
                                                                      I    »  •
                                                                     co 10 en
       O)
^- CM TO
 i  i t-
 O)  QJ  X
TO TO  O
•i- -f-  Q.
 X  X  Ol
 O  O  I
 CL  Q.O
 O>  O) i —
                                      oi
                                     TO
                                     f-
                                      X
                                      o
                                      CL
                                      01
                                      i
                                     CO
                                            o o

                                           m co
                                              "
                            <— f— d    r-
                            o o
                                          re  re
                                          ca OQ
                                                    O)
                                                     CM

                                                      .

                                                     CQ U3 f— r— f) CSJ CO O3 CO
 o TO
 ££

 •o'S
 (U -C
 ^— »f—

 •5.7
 S CO
                                                            O) 0_
                                                            s-  re
 re «/>
+•> •!->
 re c
•o oi

 a,*
-e s.
4J Ol
    Q.
 CO X
 U Oi

T- CO
 v> E

«/> 10

O  C
                                -^
                                i-  in  «/»
                                X  Ol r—
                                O  E •—
                                j. •;=  co
                                C.*->  U
                                O.
                                «s +j en
                                    c r-^
                                «,  «,>

                                re  oi  c
                                    re •!-
                               ••-     x
                               •w -o o
                                O Ol O.
                                re *J co
                                   u
                                o 3 i—
                               •I- TO O
                                C C -i-
                                CO O TO
                                en u
                                re    
 Ol 01 O
 u    re
 s- co
 oi +•> in
 o. re re

 co re s*
 > Q.T3-
T- 01  •
4J in o
 re    i

 oi re  •
 s- i- O
    eg
 CO > 10
j= oi re
                                                                                                              Ol
                                                                                                                   o.  i
                                                                                                                   re o
                                                                                                                   CO .—

                                                                                                                   o en

                                                                                                                   TO  -
                                                                                                                   >> o>
               03  I  O
              oo to M-
              O  «T3    INI  rtJ -f-
              Z ^ O  « r-  1.  X
               I     S. r—  «+J  O
              'CsJ Q. T3 O i—  (U  Q.
              f— (O >i-^- ^— *J  0)
                 CO J= -O     II
               q   -i- o -a o o   •
              +J  ••oi.c^-r— a.
                  CD I  T3 (a  »   •  in
              J. S *" >>    cn eft ea
              CM o  «jz  i   «
                  C «3- f- O CO  *T O
               •••i—    T3 r—  «>Z  S_
               ftt 3 Q. I   «r>.    -o
               c cr re CM cr>
               eu i ca_        .
               ^ >  * •"  «  I  O CQ  S-
               c.co i/> •— . co a.    •<->
              f"™i    CU !•—  « CU O  CJ
              ,o
              N    t- ns  > o a_ F—  aj   i  co co
                 c 10 --o
                                                                                                    o-»- CQ en f-
                                                                                                                    X  I
                                                                                                          i  i
                                                                                                           co
                                                                                                                        » x
                                                                                                                       n o
             >> i
            j=vo

            i-I-i-^  O        ^     ~~

              . co   -a. oi  i  r»Tcn
            Cv.  c  co  re c  8  i
             re  co  c CQ co co
            CQ  I.  O     i.  - X
            O  >,  C  •• >, OQ o  «
            31  Q.*r-  oi Q.r^» Q.  Ol
             II—I  3  Ci—it  CO TO
            r—  its  cr oi re *—«  i  ••-
               i—i  i  i.i—i -t-  o  x
             •-OCM  >, O — O  O
             CO  M r—  Q. Nl    r—  Q.
             C  C    r—i c r—   •  Ol
             oi  oi   «  re co  i  a  i
             v. -Q •— i—i js oi  en co
             >>  ^-  O   TO  I
             ex •     M   .1-
           i—i  01 o.  c  01 x  x
            jo  c re  oi TO o  o  •&
           l__l  o CD -O -I— g>- <- -r-
            O  C     O  X Ol TO

            S  =1-15 ?r-££
            oi  cr re  >^ LO  o *^~  re
           JD  I     .^   .-r- TO CQ
               IO  • •»- «3- TO  I
             -  • co TO        BO
           O- •— C  I  O. ««CO  S.
            re    o m re m   • TO
           CO O- C  «CQCOC£>s
               re ••- ^3-    TO r^ .e
              ca 3  i  •••^-  i   re
                  cr >>o_ x-^ si
           •o ••  i  x re o -t- <-M
            oi in CM o CD   -^» o>
            inr- r- S_    0.    4-J
            3 o   «TO «*- re  •  i
                                                                                                  ••-   03
                                              Q.
                                               re
            Ol

           J3
           J3
                                                  OI
                                                  C
                                               S-  O  1/1
                                               O>  C  C
                                              jz -i-  re
                                              •»->  3  S-
                     I  O J= -r-

                      •TO O O
                     a-  o    a.
                     i  s-  « QJ
                                                                                                                        co
           i
 -  -     »i
 >>TO r— X
 a -r-  o o
 1-  X -r- a.
TO  O TO O)
                                                         6-153

-------
the optical enantiomers of the diastereomeric BaP 7,8-diol-9,10-epoxides is
highly stereospecific,72 but does not correlate with carcinogenic activity for
these same isomers.  These results are of considerable significance in light
of evidence indicating that the in vitro metabolism of BaP is stereoselective,
                                                                             47
depending  upon the source of the microsomes and conditions of the incubation,
and suggests that the Ames assay may  not be capable of distinguishing between
certain  stereoisomers.  On the other  hand, certain mammalian cells in culture
(e.g., hamster V79 cells) are capable of detecting the stereoisomer of  BaP  diol-
epoxide  which  is active in vivo.
      Further  evidence  implicating the obligatory role of epoxide intermed-
                                                                      369
 iates in the  mutagenicity of BaP was  reported by Oesch and coworkers.
 They stated in support of this  contention  that:   (1)  epox'ide hydrase  inhi-
 bitors potentiated BaP mutagenicity,  even  though epoxide hydrase activity is
 necessary for the conversion of BaP-oxides to BaP-dihydrodiols prior  to
 formation of the diol-epoxide;  and (2) homogeneous epoxide hydrase prevented
 mutagenicity even though it is required for the formation of diol epoxides
 from BaP-oxides.
      In this same regard, Levin and  coworkers366'370 reported on extensive
 studies into the role of epoxide  hydrase  in the metabolic activation of BaP
 to mutagenic products.  The addition of purified epoxide  hydrase decreased
 mutation  frequency  in S.. typhimurium by 30 to 50 percent.   However, muta-
 genicity  could  not  be completely abolished by high levels of  epoxide hydrase.
 Evidence  showing  that mutations  induced by BaP  4,5-oxide  could  be reduced
 90  percent by the addition  of  very small  quantities  (units) of  epoxide
  hydrase argued against a central role for the K-region  epoxide.  Furthermore,
                                       6-154

-------
  these researchers established that epoxide hydrase converts BaP 7,8-oxide to
  BaP 7,8-dihydrodiol,  which is further metabolized by the monoxygenase system
  to  the highly  mutagenic  BaP diol  epoxides.   Moreover,  these proposed ultimate
  mutagens were  not substrates for  epoxide  hydrase,  and  therefore were not
  affected by addition  of  the purified  enyzme
      A marked  effect  on  BaP mutagenicity  in  the Ames assay  has  been  achieved
  by  the addition of a  purified rat  liver dihydrodiol dehydrogenase enzyme to
  the incubation system.371   When BaP was activated by liver microsomes from
 MCA-pretreated rats, the addition of dihydrodiol dehydrogenase caused up to
 a 60 percent reduction in mutagenicity.  From these data it was concluded
 that diol  epoxide metabolites play a significant role in BaP-induced muta-
 genesis, and that the discovery of the action of dihydrodiol dehydrogenase
 in reducing this  effect may have important implications for the control  of
 metabolic  pathways involved in carcinogenic bioactivation.   The fact that
 dihydrodiol  dehydrogenase did not  completely abolish mutagenic  activity
 indicated  that  BaP metabolites other than  diol  epoxides may  also be  mutagenic.
 This possibility  is  supported by the observation that certain simple  epoxides,
 Phenols, and phenol  derivatives  of  BaP  are  mutagenic  in the  Ames  assay.372
     Support for the theory  that diol epoxide metabolites of  POM  are  ultimate
mutagens in the Ames assay has been  obtained from studies with MCA, DMBA, BA,
7-methylbenz[a]anthracene, 5-methylchrysene, chrysene, and DB(ah)A.373>374>535>539
In particular,  it was shown  that the immediate dihydrodiol precursor of the
bay-region (see Section 6.4) diol epoxide for each of these compounds was
consistently the most highly mutagenic derivative.
                                     6-155

-------
6.5.1.2  Somatic cells In culture—Current methods of identifying chemically



induced mutants from cultured mammalian cells rely upon the mass selection of



variants resistant to the cytotoxicity of certain drugs (e.g., 8-azaguanine,



6-thioguanine, ouahain).  Among the most suitable cells for this type of



study are Chinese hamster cell lines, particularly V79 cells derived from



male lung tissue.  However, V79 cells apparently do not have the capacity for


                            375
metabolic activation of POM.

                                O~7C
     In 1974, Huberman and Sachs    reported a successful system for cell-



mediated mutagenesis which combines V79 cells for the  identification of



mutants and  lethally irradiated rodent embryo "feeder" cells  for the meta-



bolic activation of test chemicals.  Mutagenicity was  obtained with DMBA,



MCA, and BaP at  rates which correlated with  their respective  carcinogenic



potencies.   Large  numbers  of  8-azaguanine-resistant  mutants were produced  by



treatment with  as  little as 0.1 ug/ml  (Table 6-32).  The  noncarcinogen



benz[a]anthracene  produced no mutations  in this  system.


      Huberman and  Sachs377 subsequently  conducted expanded studies  on the



mutagenicity of ten polycyclic hydrocarbons  at  three different genetic  loci



 in hamster cells.   V79 cells  (together with  metabolizing cells) were used for



 identification of ouabain- and 8-azaguanine-resistant mutants induced by



 these polycyclics.  The induction of temperature-resistant mutants  was  also



 examined by using Chinese hamster ovary cells derived from a clone  of temper-



 ature-sensitive cells.  Tables 6-33 and 6-34 summarize their results and



 indicate that DMBA was consistently the most potent mutagen tested, having



 significant activity at concentrations as low as 0.01 ug/ml.


      The V79 cell system without the addition of metabolizing cells has been



 used to identify structures which may be the ultimate mutagenic/carcinogenic
                                       6-156

-------
oo

~*J

LU
*m^


Oi

^^

Q£
LU
^^
CO ID

3=10
LU _J
^O *•!
LU LU
z o
HH
3= O

O CO
i— s:
LU

Z Q£
O LU
CQ I—
ce. oo
0 Z
o z:
>- LU
3= a
^•j
£* CD
z
LU Z
O I-H
z a:
i— i >»
O «/>
a:
5 HJ
1 t • »
o3
a
>- LU
tn "~
0 >
i— i t— i

^i ^^
UJ wvj

«C c_?
1— i
=> O
2: u

.
«VJ
PO
1


eu


_Q
n3
*


i
• CU ID
l»- C-W 0 l/>
O i- C r— S-
c « ^ o
!_ us *J in >
CU 3 in *J ••-
JD CTi-^ C >
PS «g t/1  3
Z (O ^ 3 (/)
1 S
CO


U l/l
at •—
t- CU
u u
C cn
M^ P**» %4
cu s»
cn cu
•S5
o >«-
CJ

^

r— in c
^ >•» cu
o a t
CU Wl -4J
-Or- X
E f— i.
3 CU CU
Z U4j
A3




C C
o cu
•4-> Oi—
 f— CM CVJ f—









^- r— ~- O
O C3 O O ^





CU
• C
cu
o
a

ja
*^
w
r— ^ CU
I_l CU
N S-
C f
CU 4->
-Q c a) ai
•— > i— CU CU
j= 0 s- i.
+J j: >, >,
cu o o. a.
E *— i — i r— i

c o j: i__i u—i
CU 1 4J o O
> CM CU N N
— •— Z c c
o • i cu cu
V) f>» c*l CD CD




•g
M
•F- 1 .£
c cu in
•i- C -^
Wl 1--O
a. c
>) 1O -^
S. 3 S.
•M cn-u
10 cu
c no.

CUlt-
i- O u>
CU r—
X >»—
o cu
V) C O
— eu
"3 crcu
U CU 4->
^ (/)
5"~ §
H- CU.C
.c
• Qi

c =.—
o >*0l BH
jc r^ fo
=» i.
0 !-
i- eu-»-
cu j-**f^
cn i
C Oi—
u >>x
J. 0
CO CCM
O CU
•^ *~
-C O -t->
i-i^"?
CU u>
in IQ
nj e:
<= o
CM O-^
l-lj 10
o >
11. CU-^-

•o •>-> ^~
CU 3
+J cu o
IQ C 1
eu •" o
^ ^ ^
cu •-
Of 4-» tn
L. CU •*->
cu -a c
2 T3
0 4-»
M •<-> 3
CU E
3 CU 4J
•w -a c

3 CU *J
o m in

CU T3 in
JZ = CU
P— It! S-


-------
CO

CO
 CO
 (—4
 CO
 LU
 UJ
 =>!•«.
 R CO
  QJ o
•r-  C O. >
 c:  re   ••-
 10 4J t/J >
 3  in -4-> s-
 O>i- C 3
 nj  1/1 re «/i
 M  o» -t->
             CO       «—
    re     o
    4-»     >
    «n    •*-
    •t-  in  >
    «n +J  i.
    ai  c  3
    j_  re  «>
                 re    r-
                       OJ
                 en u
                 O U
                    (U
             I/I
                                                                                                                   • a. a>
                                                                                                                 I—    (S>
                                                                                                                  £•- «3
                                                                                                                    .
                                                                                                                     O  OJ
                                                                                                                  O <  O
                                                                                                                  •t- ea
                                                                                                      i—     re
                                                                                                      E t-
                                                                                                      •^  O  C
                                                                                                      01   •—
                                                                                                                   *J ^- -O
                                                          COO)
                                                                                     «3-   CM   csj
                                                                                              f—  O "CM
                                                                                                       e-s
                                                                                                                          -
                                                                                                                    O  i-
                                                                                                                    >> 
                                                                                                                   +J «3 CM
                                                                                                                       OJ
                                                                                                           s_ o|
                                                                                                           a;
aj   a>
c   c
O)   O)
                                                     0)   re
                                                  u
                                  a>
                                 r—   S_
                          0)

                          (U
                                           0)

                                           'OJ
                                            0)
                                                     u
                                                     re
                                                       a>

                                                       at
                                                       u
                                                       re
                                                      i—i
                                                       re
      
                                                          re   re   N   i—
                                         c    -•-ao
                                    o)   re   re    re   ^^  •"**
                                    C  r—it—ii—i   >»  >>
>,   N   
                                                                    a>
                                                                                        a>
                                                                                        u
                                                                                        re
                                                                                        re
                                           ai
                                  (U        J3
                                                                                           re       r:
                                  N       CO
                                  C       i—
                                      o.  a.   o   co   o  o
                                                                        co
                                                                                                                    CM
                                                                                                                    «4-  CD
                                                                    a> »—
                                                                    4-> O
                                                                    (O  •
                                                                    O) O
                                                                    S_ un i
                                                                    •)->

                                                                    0)

                                                                    O)
                                                                                                        i—  en
                                                    ai o
                                                   £1 *
                                                   H- *
                                                                                                                    >a
                                                          6-158

-------
         Table 6-34.   INDUCTION  OF TEMPERATURE-RESISTANT MUTANTS IN
                      CHINESE  HAMSTER OVARY TEMPERATURE-SENSITIVE
                      CELLS  BY DIFFERENT CHEMICAL CARCINOGENS
Concen-
. tration,
Treatment pg/ml
Solvent
Pyrene
Phenanthrene
Benz[a]anthracene
Benz[a]pyrene

f
3-Methyl chol anthrene


7 , 1 2-Dimethyl benz[a]anthracene
— — — - • . __
0
1.0
1.0
1.0
0.1
0.3
1.0
0.1
0.3
1.0
0.01
0.1
Temperature-
Cloning resistant
efficiency, mutants per
% 1Q6 survivors
93
88
88
78
67
45
9
78
32
10
65
3
0.6
0.9
0.7
0.7
21
56
170
20
46
125
51
494
Three days after treatment with the polycyclic hydrocarbons at 34°, the cells
were seeded to determine the cloning efficiency of the ts cells and the fre-
quency of temperature-resistant mutants.  For selection of temperature-resis-
tant mutants, the ts cells were shifted from 34° to 39°, 3 days after cell
seeding.  At the time when the cells were seeded for cloning efficiency and
JMC«frequency of mutations, there were after treatment with 0.01-0.1  ug/ml  of
In 111  ;?h  W9/   °fJaP °r MCA; ?'5 to ]'9 x 1°  ts cells Pfir Petri  dish-
in all  other cases,  there were 2.1 to 2.3 x 10^ ts cells per petri  dish.
                                    6-159

-------
metabolites of BaP.64,65,72,364,367,378  Numerous oxidative metabolites
of BaP were surveyed, and the stereoisomeric BaP 7,8-diol-9,10-epoxides were
found to be highly mutagenic (Table 6-31).  The very short half-life (30 seconds)
for BaP 7p,8ot-diol-9p,10p-epoxide in tissue culture medium made comparisons of
activity between the two diol epoxides very difficult, however.  Nevertheless,
the high mutagenicity of both diol  epoxides, coupled with the high carcinogeni-
city of their precursors, BaP 7,8-dihydrodiol and BaP 7,8-oxide, suggests a
strong formal relationship  between  mutagenicity  and carcinogenicity insofar as
a  common pathway of  metabolic activation  is concerned.
     Modifications of the V79 cell-mediated mutagenesis  system  have recently
been developed  which involve new means of metabolizing test compounds.
Newbold  and coworkers379 combined baby Syrian  hamster kidney cells  (BHK21)
for metabolism  of POM with  V79  cells for detection  of ouabain-  and  8-aza-
 guanine-resistant mutants.   This system produced binding reactions  between
 DMA and BaP or 7-methylbenz[a]anthracene which accompanied mutagenesis in  the
 same cells and were typical of those which occur in vivo or in primary cell
 cultures.   Chromatographic analyses of the BaP-DNA reaction indicated that
 BaP 7,8-diol-9,10-epoxide was involved in the binding interaction.   These
 results strengthened the argument which supports the role of diol epoxides in
 BaP-mediated mutagenesis and carcinogenesis.  Furthermore, the cell-mediated
                                                               oyq
 mutagenesis system  as modified  and validated by Newbold et al.    is proposed
 as a convenient and representative mammalian cell screening system which may
 prove to  be superior to mutagenicity  assays with $._  typhimurium in providing
 a rank-order correlation with  carcinogenic potency in mammals  (e.g.,  Iball
 index).   However, additional screening of polycyclic hydrocarbons  and
                                       6-160

-------
   aza-arenes  Is  required before a definitive conclusion may be drawn for these
   classes of  carcinogens.
       The current belief that neoplastic transformation may arise from a
  chemically  induced somatic mutation was made more convincing by the recent
  studies of Hubeman and coworkers.38"  They demonstrated for the first time
  that BaP and BaP 7,8-dihydrodiol  can induce both neoplastic transformation
  and mutagenesis (ouabain  resistance) in the same normal  diploid hamster
  embryo  cells. A ratio  between transformation  and mutagenesis  for ouabain
  resistance was  only about  20:1.
      A  further  adaptation  of the V79 cell-mediated mutagenesis  system has
  been employed which provides metabolic  activation by rat liver  homogenates
  containing microsomes and cofactors.3«,381  ^^ Qf v?g ^ ^ ^
  by their resistance to the cytotoxic effects of 6-thioguanine.  The mutagenic
  activity of BaP, MCA,  DMBA, and benz[a]anthracene in this system showed a
  United correlation with their respective carcinogenic potencies,  with OBahA
 and DBacA,  an inverse  relationship  between mutagenicity and  carcinogenicity
 was obtained. Such  a  relationship  has also been  shown  for the dibenzanthra-
 cenes when tested .in the SalmonelWmicrosome  assay.355  Although mammalian
 cell-mediated mutagenesis is hindered because  it  is technically  more diffi-
 cult, time-consuming, and thus permits fewer compounds to be screened, there
 are advantages over the microbial system:   (1) bacteria indicate  reverse
mutations and V79 cells indicate forward mutations, the latter type of
mutation being more relevant to the process of neoplastic transformation;  and
(2) cells can be employed which are the same kind as found in the actual
targets  for mutagenesis and carcinogenesis induced by chemicals.
                                    6-161

-------
     The analysis of chromosomal  aberrations and sister chromatid exchanges

(SCE's) is often recommended as a screening technique for potential  mutagens

and carcinogens.  Several investigators have examined the effects of POM on

the chromosomes of mammalian cells.  Early studies indicated that variations

in chromosome number and structure may accompany tumors induced by BaP, MCA,

and DMBA  in the  rat, mouse, and  hamster.382  However,  in cultured human

leukocytes exposed  to DMBA, chromosome damage was  not  the same as that pro-

duced  in  hamster cells.  Although  it is  argued  that  chromosome changes  in

POM-induced tumors  are  nonrandom,382 others141'385 claim that detectable

chromosome changes  are  not specific for  the carcinogenic agent  nor  are they  a

prerequisite  for neoplastic growth.  Moreover,  it was shown that an increased
                                                                  *   386,387  hll+
 rate of SCE's could be produced by BaP in cultured human lymphocytes        but
                                                                387
 this was not correlated with different rates of BaP metabolism;    a sur-

 prising  result  in  light of the  known importance of metabolic activation for

 BaP mutagenicity.  BaP-induced  SCE's rates did not differ between lymphocytes
                                                                   OQ"7
 taken from normal  humans  and  those  from patients  with lung cancer.       In

 recent studies  with cultured  Chinese  hamster cells  exposed to DMBA, BaP, and

 MCA,  none of the chemicals produced chromosome breaks and  only  DMBA could

  successfully induce SCE's.388  Although it cannot be denied  that POM  causes

  chromosome damage, it is  not clear whether this effect may represent  an

  epigenetic phenomenon which is merely secondary to  mutagenesis and neoplastic

  transformation.   Furthermore,  in cases where a chemically-induced  mutation  is

  "silent" (i.e., neutral amino  acid substitutions),  there is no reason to

  believe that detectable  chromosome damage should occur.

        In recent comparisons of  three cytogenetic  tests,  (1)  induction of

  chromosome  aberrations,'(2)  induction  of micronuclei,  and (3) in yjvo
                                        6-162

-------
   induction  of  sister chronatid exchanges,  the  latter test proved to be the
   most sensitive with carcinogenic polycyclic hydrocarbons.389  However,
   positive results were also obtained with  phenanthrene and thus limits'the
   usefulness of sister chromatid exchange as a screening technics for carcino-
  gen detection.  BaP was positive in the sister chromatid exchange test,
  weakly active in the chromosome aberration test,  and negative in the  micro-
  nucleus test,   on the  other hand,  OHM was clearly ,„.,„„. ,„ ,„  three
  tests.   It  was concluded that cytological  tests do not provide reliable
  correlations with all  carcinogens  tested and thus  cannot be  used alone in
  mutagenicity/carcinogenicity  evaluations.
      Damage-to the genome resulting from chemical  insult can theoretically
  also be detected  by examining DNA repair.390  The  suggestion that DNA repair
  is applicable as  a screening procedure for evaluating potential chemical
 mutagens is based on the assumption that the level  of DNA repair synthesis in
 a cell  reflects the extent of DNA damage produced  by a chemical.   Indeed
 unscheduled incorporation of 3H-thymidine  into  nuclear DNA  of normal  human
 cells exposed to epoxides of benz[a]anthracene  and  MCA has  been observed  39°
 However,  since  a  metabolic activation  system was not present ,„ this system
 the parent hydrocarbons  showed no activity.  More recent studies confined
 that  K-region epoxides of BaP, DMBA, and DBahA  caused DNA damage in hunan
 skin  fibroblasts which was repaired with the same system used for repairing
 lesions induced by ultraviolet radiation.391  As would be expected, the
parent hydrocarbons exerted no effect.   More importantly, results were
obtained which indicated that the DNA repair process itself does not induce
mutations, but rather that mutagenesis  occurs before the DNA lesion can be
excised.   In  fact,  there is no a  priori reason why  POM should exert  a
                                     6-163

-------
mutagenic effect by inducing DNA lesions which require any kind of repair syn-
thesis at all.
                                                             392~394
     Nevertheless, DNA repair synthesis in human fibroblasts,        rat liver
cells,395 and Chinese hamster V79 cells396 has been successfully used for the
detection of chemical carcinogens, including numerous POM.  However, the
percentage of carcinogens giving positive results for DNA repair is considerably
less than in the  cell transformation or microbial mutagenesis assays.  For
example, benz[a]anthracene  is negative  in the DNA repair assay utilizing human
or rat liver  cells,  and  BaP is  only positive  in the DNA damage assay utilizing
hamster V79 cells.
6.5.2   In Vivo  Studies
6.5.2.1  Effects  in somatic tissues—Tumors  induced in vivo  by  POM are
commonly associated with chromosome  abnormalities  in  the  neoplastic cells.
 In particular,  sarcomas  induced by DMBA,  MCA, and  BaP in  the rat display
 karyotype variations which were reportedly nonrandom and  distinctly different
                                             OQO "30*7
 from sarcomas induced by Rous sarcoma virus/0-5'-537   The  chromosome patterns
 of DMBA-induced sarcomas were found to be identical  with  those observed in
                                                                      384
 primary rat leukemias399 and primary carcinomas of the auricular skin
 induced by DMBA.  Consistent chromosome abnormalities in DMBA-induced rat
                                                               399
 leukemias were first reported  in 1967 by Kurita and coworkers.     These
 changes were characterized by  trisomy  in the C-l and/or A-6 chromosomes.
 Earlier researchers reported that DMBA-induced leukemias in mice  consistently
 showed 41 chromosomes in the modal cell lines  instead of the normal 40.
 Sugiyama401  obtained an apparent  relationship  between sarcomatogenic potency
 and the incidence of chromosome breakage  in  rat bone marrow cells produced
 by a  single  intravenous injection (50  mg/kg  B.W.) of various  benz[a]anthracene
                                       6-164

-------
 derivatives.   It was apparent that benz[a]anthracene and several non- or
 weakly-carcinogenic derivatives (7,12-diethyl-; 7-ethyl-; 12-ethyl-) produced
 no elevation of chromosome aberrations, whereas the more active carcinogens
 (7,12-dimethyl-; 6,8,12-trimethyl; 7,8,12-trimethyl) produced aberrant meta-
 phase cells.
      Considerable evidence is also available to indicate that chromosome
 alterations in POM-induced tumors in vivo are not consistent either in
 frequency or in pattern.   DMBA-induced tumors (fibrosarcoma, squamous carcinoma,
 lymphosarcoma) of the  uterine cervix in ICR mice revealed various  karyotypic
 compositions.402'403  These tumors  displayed diploid,  aneuploid, tetraploid,
 and octaploid chromosome  constitutions.   Tumors induced in mice with MCA and
 dibenzo[a,i]pyrene  also showed a wide variation in  chromosome constitution  404>
 405
      Mice treated with  30 ug  DMBA,  a  dose  sufficient to produce a 100 percent
 incidence of thymic lymphomas, did  not  reveal an excess of chromosome abnor-
 malities  in  bone marrow or thymus.406  Even  at  higher doses  (60 ug DMBA), the
 incidence of abnormal chromosomes did not  significantly differ from  controls.
 Subcutaneous  tumors  in  Syrian hamsters  induced  by single injections  of BaP
 (0.1 ug)  or  DMBA (0.1 mg), and cultured cell populations derived from these
 tumors, failed to reveal  common karyotypic changes.407  Tumor cells  had
 subdiploid, diploid, and  hypotetraploid chromosome constitutions;  further
 karyotype  rearrangements occurred with subsequent growth |n vitro.
     In humans, the presence of the "Philadelphia" chromosome in myelotd
 leukemia appears to be  the only example of a human chromosome abnormality
which is tumor-specific.408  In POM-induced experimental tumors,  lymphatic
 leukemia in mice produced  by DMBA also displays  consistent chromosome ab-
            402
normalities.      Beyond this common  feature,  convincing data  have not been
                                     6-165

-------
presented to indicate that somatic cells exposed to POM may suffer charac-
teristic or reproducible damage to the genome.  Instead, it is proposed that
random karyotypic mutants of transformed cells are selected in response to
growth pressures in the host environment (e.g., tissue necrosis, infection,
                           402
anoxia, lack of nutrition).
      Evidence  has not been encountered  in the published  literature concerning
the  likelihood of POM-induced  somatic mutation  in  the  absence of  neoplastic
transformation.
6.5.2.2  Effects  in germinal  tissues—The  fruit fly,  Drosophila melanoqaster,
 is commonly used as a whole animal submammalian genetic test system.   Assays
 can be conducted for gene mutations at specific loci  and for small  deletions
 resulting in recessive lethal and visible mutations.   Moreover, Drosophila is
 apparently capable of activating compounds that are not mutagenic or carcino-
                                  410
The use of Drosophila has been suggested
                      410
 genie in their administered form
 as a valuable bridge between microbial and mammalian assays.
      A series of studies was undertaken by Fahmy and Fahmy        which
 demonstrated conclusively that  carcinogenic polycyclic hydrocarbons were selec-
 tively mutagenic to the male germ  cell line of  Drosophila.  Mutagenicity was
 restricted to the  tRNA and  rRNA loci, yielding  Minute  (M) and  bobbed  (bb) mutantj
 respectively.   The activity of  benz[a]anthracene,  7-methylbenz[a]anthracene,
 MCA,  DMBA, their epoxides,  and  some other derivatives  was determined  in  tests
  of both general mutagenicity on the whole genome and specific  locus mutability
  on the tRNA and rRNA genes.414  The overall  yield of sex-linked recessive  lethal]
  and visibles (general mutagenicity test) was not significantly enhanced over
  control levels by the parent hydrocarbons.  However, several  of the K-region^:
  epoxides  of these compounds were appreciably mutagenic (Tabla 6-35),   Ip the
                                        6-166

-------
   o

   a
  to
  co
  o
  0
 CO

 O
 a.
to
oo

vo

tno<   
-------
specific locus test, DMBA, MCA, and 7-methylbenz[a]anthracene produced signifi-
cant increases in the yields of phenotypic bb plus N and transmitted bb's,
whereas benz[a]anthracene did not induce significant numbers of mutations on
the RNA genes (Table 6-36).  Thus, mutagenicity was correlated with carcinogen-
icity, although mutation  frequencies did not vary with the dose employed.  On
the other hand, the mutagenicity of the hydrocarbon epoxides on the RNA genes
was apparently dose-related.
      It was  significant to note that the  It-region epoxides  of  all  derivatives
tested displayed appreciable nonspecific  mutagenicity  as  evidenced by the
high  number of X chromosome mutations  (lethals and  visibles) in  comparison  to
controls  and the parent hydrocarbons.   An explanation  was offered based  upon
 the high  reactivity of electrophilic carbonium ions derived from the epoxides
 which may react nonspecifically with nucleophilic centers in DNA, especially
 the N-7 of guanine.  The fact that the parent hydrocarbons displayed greater
 mutagenic specificity than their K-region epoxides argued in favor of the
 involvement  of a different reactive metabolite (i.e., non-K-region epoxides,
 dihydrodiols, or diol-epoxides).  Moreover, the  correlation of mutagenic
 activity to  RNA genes with  carcinogenicity  in rodent  skin  (DMBA>MCA>benz[a]-
 anthracene)  implied a common intracellular  target  and/or ultimate reactive
 metabolite.
       For the detection of mutagenicity in the whole mammal, the male dominant
  lethal assay in mice is most commonly employed.   However, it is felt that
  this test is not particularly sensitive.415  Kennedy and coworkers416 admin-
  istered BaP (50 and 100 mg/kg B.W.) and benz[a]anthracene (15 and 30 mg/kg
  B.W.) by intraperitoneal injection to male albino mice which were subse-
  quently mated to  virgin females.  The females were sacrificed after  one week
                                        6-168

-------












^
(0
ro
^
Q.
o
V)

0
M
1— 1
_J
CO
CQ
UJ
H-
U.
O
UJ
to
•p»
o
Q_
co
UJ
o£
<_>
2
UJ
o
t—



vo
CO
1
VO
a;

3

i—










r=
:
a>
-a a.
A !
1
g
! =
. !*
i f
i 0
•a
3



^
O
X a!

«
U
1
.c
1
z




S"S
K £
3-




ai












c
1
§








_ -_^^«u,«,»


o o o o •— *- *o *n <•*






C- C-S.^.-C-C.^.00 "^
**"> * 2SS2




eeoooooo
^vnomo^ino







g' '
o
u
a> 2

u c
£ r^, 4,
J= « C
4^ l_^ Q)
C ^S C £
o r? 5 •£

.= = 5" "5
J= c 5 jj "5
£ S . £ ° 5
c S ^ ~ £
0 «l I . ,
«D r« f^ <*)
-CM^
O O O
•— CM CM
O OO






^ c^
^r^ . c*r



«,
•B  OJ
C C i
Of Oi r*.
A ^D i
£ S£
"x "&•§.
£ £:S
"? T T

i/C as urT
i it
X XX
a o o
CL 
— «• "CM CO
OO OO
*in co«
o*- o v






»-%CM ^"^
CD CM CO CTi
^ — *— CM


CM r^ mo
O O O —
^* io en o
^ <*» ^ u>




c sr^:
tft CM ^ m




ScS Slo
^r as to <->
<-» CM




OO OO
U) O ^ */>



1
- 1

«C ^i
** -C
1 i

CM i
*- o
• 1_
rx ^
E I"
">. o> "V
.C C CM
•*- QJ ^»


i*C *
C I C X
OJ >t T3 O
W X f— ^ Q.
T> Q — < i
5 *Y c S—
S • Q !••
*n *-
CM CM ^ P—
O O O O
Ch'eh'm'rt
O OO O






•— CM CM

»— CM


«"i m *»j CM
O o o o

^ CM P- *->




^^?~
CM IA CM CM




CM SS S
rncncocM
^— m *— CM




o o O o


*o

I
*n
0
t.
1" =

i  H* t.
M S C°
CO r^. •

III
1
«

k.
01
3
1 ig
' £
J:
ai
^
*
i
•B
U
1 JC
c
*5
*^J*
1
3
"w
I
"5
I .

IB O
+*
3 VI
• i
— —
h 5
"S •§
I 1
2 S.
«B >.
*Su a;
>» J
2 "^
S v«
« as w»
£ *e vi
O Of
i1 « 5
1 u E
•B — fc.
VI 0) Q>
" >
•° § ~

JO s 0 —
£0» *rt «»

• AW
6-169

-------
and the mutation'rate determined by counting the numbers of fetal resorptions
versus the total number of implantations.  These investigators found no
evidence of genetic damage which might have been manifested as a dominant
lethal mutation.
      In contrast to these results, Epstein and coworkers    observed an
excess incidence of dominant  lethal mutations in ICR/Ha Swiss mice using high
doses of BaP  and MCA  (Table 6-37).  The  authors noted  a lack  of  consistency  in
the results obtained,' both  in terms of dose  (for MCA)  and in  magnitude of
response and  parameters affected (for BaP).   Nevertheless,  in control  ani-
mals the mean weekly pregnancy rate was 66 percent, the mean  number of total
 implants  per pregnancy was 11.5 to 11.9, and the mean early fetal  deaths per
 pregnancy were almost always less than 0.95.  Thus, considerable differences
 are evident between POM-treated groups and untreated controls.
      Additional evidence which may indicate a potential for POM-induced
                                                                   4-18
 mutagenesis  in germinal tissues was provided by Wyrobek and Bruce.     They
 found that five daily  intraperitoneal injections of BaP or MCA  at doses up  to
 100  mg per day caused  abnormally  shaped  sperm in (C57BL x C3H)F1 mice.  These
 effects were most pronounced at four and ten weeks after treatment, thereby
 indicating damage to the  primary spermatocytes and spermatogonia.   Since
 abnormalities in  sperm morphology were found,  these results  may be  indicative
  of mutagenesis in the male germ cell line.
  6.5.3  Correlation of Mutagenicity with .Carcinogenicity
       The recent widespread use of in vitro mutagenicity bioassay  systems
  (e.g., Ames" assay) to predict the carcinogenic activity of chemical pollutants
  and complex environmental mixtures has  led to considerable controversy over
  the reliability  of obtained results.  In particular, the practice of
                                        6-170

-------



J

2>

U O
1 1 SX
1-4 -
- t—
«E CO •— i
.
co
i
vo

(U
i->
•o




(A
C

§









t/t
Oj
*

1
e
k
u
c
•g

w
•^
^f
E
I
)Q
5



u
CO
T) UJ
e
of
£•-
a. x
••



1
1
k
"fe

>s-
** C
c ra
« c
•» cn
E £
•— O.

fc?

X C
1

<*-*-*
<*- O w»
0 «i J=
01 • <-»
• ^ o «o

Of JH
s e"

^- M C*
o^*5?
^— * E

e
i? 5
Of to
II*. -s

1 i











.*
^ Ol
X VI (Q (U
u *• vi X
C C l»
*J C E > Of

O •*-  i a* o

w o ai x vi e


	 	 ^
CM
t f»> I 1 1 I i
i 
S. 3


^^,
CM
1 * 1 1 1 1 t |
t *- 1 1 1 1 I i
I •* 1 1 1 1 i i

8

£

l II IJ9 I
I II.* t
^•T ~»>r*~Le-i *™^
*« tO O CM
•"• »*»'t» W

— . ^ ^°F? -^

*M • •— CM m cn ^ ,
V V


*— ^ CM CM ^»

*** '"S^'^'0* *** °*

O O iA O in O O O
tn o ^- o i*- o ~- CM
— *- —


c** m co eo





^ Q. a. a.


u
E
£
•_*
at S
£ S
S: a
1~~l X
^2, 5
O tt;
c "T
* o
03 M
6-171

-------
extrapolating from data regarding in vitro mutagenic potency to predict in


vivo carcinogenic potency has often been criticized.  It now appears that


these criticisms are at least partially justified in the case of POM.

                                                         419
     In a recent paper published by Andrews and coworkers    24 POM were


tested for mutagenicity using the Ames Salmonella assay in the presence of


Aroclor 1254-stimulated rat  liver microsome S9 mix.  Tester strains TA1535,


1537, 1538,  98, and  100 were employed using the plate incorporation method


and liquid culture assay.  Among the 24 compounds tested,  mutagenicity was


correlated with carcinogen!city for 14  (58 percent), while there was  no


correlation  for  10 (41  percent) of the  compounds  (Table 6-38).  Of the compounds


showing a negative correlation, eight were false  positives and two were


 false negatives.   Difficulty in the interpretation of the false positive


 results arises from the fact that information concerning carcinogenicity


 may be incorrect  (e.g., BeP is considered as noncarcinogenic by the authors,


 but as a weak carcinogen by others).  Interpretation of the false negative


 results, on the other hand, is  considerably more difficult.


      One of the most likely explanations  for the lack  of  qualitative agree-


 ment between Ames assay and carcinogenicity bioassay results  is the type of


 metabolic activation system employed for mutagenesis testing.  Variables


 such as  the choice  of  organ from  which the S9 mix  is prepared, the choice


 of species, the  choice of enzyme-inducing agent,  and the volume of S9 mix


 employed are all  believed to  influence the outcome of  a particular experi-


  ment.420'421   For example,  Tang and Friedman422  demonstrated that while


  human liver enzymes were effective in activating aflatoxin B-, to  a mutagen


  in the Ames assay, they were highly variable in activating aromatic amines,
                                       6-172

-------
     Table 6-38.  COMPARISON OF MUTAGENICITY IN AMES ASSAY
                  WITH CARCINOGENICITY OF POLYNUCLEAR
                  HYDROCARBONS 419        ™LTNULLEAR
  dibenz[a,j]anthracene
  5,6-dihydrodibenz[a,j]anthracene
  l,2,3,4,8,9-hexahydrodibenzra,j]-
       anthracene

  dibenz[a,h]anthracene
  1 «2,7,8-tetrahydrodibenz[a,h]-
       anthracene
  5,6-dihydrodibenz[a,h;ianthracene
  7,14-djhydrodibenz[a,h]anthracene
  I »2,3,4,12,13-hexahydrodibenz[a,h"l-
       anthracene

  dibenz[a,c]anthracene
                      .
  iu,ll,l2,]3-tetrahydrodibenzra,c]-
      anthracene

 7 , 1 2-dimethyl benz[a]anthracene
 5 , 6-di hydrodimethyl benz[a]-
      anthracene
 8,9,10,11-tetrahydrodimethylbenz-
      Ujanthracene

 anthanthrene
 4 , 5-di hydroanthanthrene
 1,2,3,7,8,9-hexahydroanthanthrene

 3-methyl chol anthrene
 1 1 , 1 2-di hydromethyl chol anthrene
 6,7,8,9,10,1 2b-hexahydromethy 1 -
      chol anthrene

 benzo[ghi]perylene
 5,6,7,8,9,10-hexahydrobenzoCghil-
      perylene

 9 , 1 0-dimethyl anthracene

 benzo[a]pyrene
 benzo[e]pyrene
_**
_**
  Only in liquid test
**
  Not mutagenic in plate or liquid test
                               6-173

-------
and generally ineffective with BaP or MCA.  .In a comparison of BaP metabolism
in liver microsomes (rat and hamster), hamster embryo cell microsomes, and
intact hamster embryo cells, significant differences were noted in the forma-
tion of metabolic products.423  In particular, the author showed how cell
disruption during preparation of microsome-containing mixtures leads to
altered metabolite ratios which could have a  significant  impact on the out-
come of mutagenesis assays  that employ  such mixtures.  The author concluded
that the  use of  intact  cell-feeder  layers to  provide metabolic activation
for in vitro bioassays  should lead  to more  reliable results  for in vivo
comparisons.
                                                           424'
      A comprehensive investigation  by LaVoie and coworkers    has recently
provided a uniform basis for the comparison of mutagenicity with  carcinogenic
 activity for various POM.  They examined both qualitative and quantitative
 differences in complete carcinogenicity, tumor-initiating activity,  and
 mutagenic response in the Ames assay for compounds tested under identical
 experimental conditions. .  In all, 40 compounds were tested which contained
 four  or  more fused ring  systems.  These included:  chrysene and its methyl-
 and fluoro-derivatives;  fluoranthene and its methylated  derivatives; benzo-
 fluoranthenes;  benzopyrenes; perylene;  dibenzopyrenes; naphtho[2,3-e]pyrene;
 indeno[l,2,3-cd]pyrene;  benzo[ghi]perylene;  and pyrene.   The Ames Salmonella/
 microsome assay was  96 percent  (26 of  27)  effective  in detecting positive
  carcinogens, and 36 percent (4  of  11)  effective in detecting negative  carcino-
  gens.  In addition,  one false negative result (6-methylchrysene) was obtained.
  Data from the tumor initiation bioassay indicated that it is a better pre-
  dictor of carcinogenic potential than the Ames Salmonella/microsome assay
                                       6-174

-------
  system.   Furthermore, quantitation of the mutagenic response among structur-

  ally related POM in Salmonella did not accurately reflect either tumor-

  initiating activity or complete carcinogenicity.


       These results have important implications for the use of the Salmonella/

  microsome  mutagenicity assay system in evaluating carcinogenic hazards.   Most

  important,  the  authors concluded  that  the  mutagenicity of POM mixtures  should

  not  be equated  with  carcinogenic  hazard  in the absence of analysis of rela-

  tive composition,  and  quantitative  comparisons  of mutagenic potency for POM

  do not provide  a reliable indication of  relative carcinogenic potential.

      As an  illustration of the misleading  results which can be obtained with

 bacterial mutagenicity assays of complex'environmental mixtures, Andrews and
      i    419
 coworkers    considered the situation with petroleum and coal  tar products.

 These substances have a high content of benzo[ghi]perylene relative to BaP

 content.   However,  since both compounds show nearly equal  mutagenic activity

 in Salmonella,  an incorrect assumption  concerning  carcinogenic potential  of

 the mixture could result.


      The  use of  mammalian  mutagenicity  bioassay systems may also lead  to

 erroneous predictions regarding  carcinogenic  activity  of POM,  although large

 numbers of  compounds  have  not yet  been  tested in any one system. -  Among the

 cytogenetic  assays  which are  currently  being  used  for  screening  of environ-

 mental chemicals, the sister-chromatid  exchange  (SCE)  test  is  regarded as being

 more sensitive than the test  for induction  of chromosomal aberrations.   A

 comparison has been made of the ability of  eight POM to  induce SCE in Chinese

 hamsters bone marrow cells by intraperitoneal injection.425  Their results,

which are summarized in Table 6-39, revealed a poor correlation between muta-

genicity and carcinogenicity for several of the compounds tested.
                                     6-175

-------
                               CM  CM  •—
                 in
                 0)
                               S  S  o  o  o  o   o
                               o   p
                                            o  p  p
                                    0.  0.  Q.
                lO O
               *> */>
                
                in
                 (O
                o   o
                O   Q
8
 UJ
 I—UJ
 istyy
 l—I LU
 ooz
in

15 <
                         CO  00  00
                                      00  CO  CO   00  00  CO
                                   0)
                                                    
  co
  vo

  o>
   (O
    CT)
                                                OJ   (O   (U   «
               Q)
               C

               S
                                   N
                   us
                                        »  s-
                                                0)
                                   -
                                            •.
                                    S
                                                       6-176

-------
Phenantnrene (a noncarcinogen) and benzoUJpyrene (inactive to weakly-active
carcinogen) both induced more SCE's than DBahA (a strong  carcinogen).
                                  6-177

-------
6.6  REPRODUCTION AND TERATOLOGY
     The interference with reproductive success by exposure to POM has only
been documented for a few compounds.  When large doses (0.5 to 5.0 rag) of MCA
in olive oil were injected in mice on the first or seventh day of pregnancy,
                                                                 AOC
fetal resorption usually occurred in most of the treated animals.     However,
sham-treated controls were not included in the study, thereby making inter-
pretation of the results quite difficult.  Treatment with MCA at mid-term
(day  11) was ineffective in  interrupting pregnancy (Table 6-40).  Among those
animals going  to full term and giving  birth, the  offspring displayed  reduced
body  weights in comparison to controls but  no  congenital  abnormalities were
evident at  birth or during a 3-month observation  period.   Although  similar
treatment of non-pregnant mice with MCA had no apparent effect,  it  is diffi-
 cult to determine  whether pregnancy was interrupted  by a direct action on the
 embryo.   This  is especially difficult in light of evidence which indicates
 that MCA does not readily cross the placenta (see Section 6.1).
      Subsequent studies did establish, however, that if MCA directly contacts
                                                                        427
 mid-term mouse embryos, fetal  resorptions and malformations will occur.      In
 these experiments, 0.01 ml  of a 0.025 percent solution of MCA in benzene was
 injected directly  into the embryonic  fluids of 10-day-old C3H(Jax) mouse
 embryos.   Their results, presented  in Table 6-41, indicate that fetal resorp-
 tions and  malformations were very  common.  From  the data presented,  it is  not
 possible to determine  the contribution of  the  benzene  injection vehicle  in
 producing  the embryopathic  effects  observed.   Moreover,  a 9.3 percent inci-
 dence  of fetal resorptions  in  sham-operated  (incisions made  but no MCA  or
 vehicle  injected) control mice  was observed,  which  could not be attributed to
  a specific cause.
                                       6-178

-------





to
CM
•a-
LU
y
2-
i— i
O
<
z
LU
cei
o.

o

LU
LU
ce.
t—
z
• '
o
x
^i
>-
LU
1
CO

u_
0

»
u
1 , 1
UJ
u.
u_
LU


o
1
10
V
jQ

1—






i

«*
"§ 5" "S
c - -5
— 01 3
tf) C VI
C Uli 0)
^ >J c
°li •£
.2 °1 °
1 ^

.
01
c
0)
i_
c
>e
I -"
^
j:
Ji1

•
o
V
1 "
1

^



X
c .
c.
Ol

Q. (/)
o -5
c
o
•us
^



c
o
O TJ

4-> (/)
11

•5
1


•

«
rf^ f*» (*l fo u? ^* 1.0 ^ \o cj ^ if> r^
«• r»M O PO •— ntrivooo' ^i^


i

/


'





u>ooe ooooooo oo
O •—•—»- cj irt »— t\j in pj ir> cvjui

















i TT^
t— ^VO,— — ^- «» «> VO — — («. ^







«o
1 i I

" « S
o- *j 2
-b ^ S
^ •§ -°

C
C
o>
Ol
u
Q.
m
1
U1
1
c
ai
01
f
«.
o
o\
m
.„
c
E?
s
i;

•
O> 13
•? I-
01 at
U K-
01
W 01
-0 §
§ 2
0 u>

-------
CM


 LU
 o
  O



  1=



  O


  o
  o
  LU

  5
  LU
  to
  o
  LU
  00
   O

   eC
   OJ




   10
                 2
 c e


ill






ill     8    S  3
III
   I
rene
                         £gj  Of ttJ
                         5  £ y

                                 6-180

-------
       Little  success  has been achieved  in demonstrating a marked effect of B.aP
 on reproduction, despite its ease of absorption and extensive tissue distri-
 bution (including transplacental).  Early studies involved the feeding of BaP
 to dogs, rats, mice, chickens, ducks, and cockroaches and indicated that a
 period of infertility may occur in the chicken, and fetal resorptions may
 occur in the rat.428  Rats fed 1.0 mg BaP per gram of food displayed no abnor-
 malities in the ovarian cycle,  ovulation, fertilization,  and implantation;  a
 few resorptions were observed,  however.   A more detailed  investigation was
 subsequently pursued to establish if lower doses of BaP might affect fertility
 and development in  Swiss .mice.429  At levels of 0.25 mg or 0.50 mg BaP per
 gram  of food given  either  before,  during,  or after mating,  no adverse  effects
 could be  shown on fertility  or  the developing embryo.  Even when BaP concen-
 trations  were increased to 1.0  mg  per gram  of food,  mice  continued to  repro-
 duce  normally.   Neither males nor  females were affected by this ration.  Mice
 fed BaP in their  diet at 1.0 mg per gram of  food since the time of weaning
 appeared  normal and gained weight  readily.
     Although the effects of BaP on reproductive function are apparently not
 severe, a transplacental carcinogenic action  has been shown with very large
 doses of BaP.430  Mice of the Ha/ICR strain were administered BaP  (4 mg) in
 carbowax-400 subcutaneously on days 11, 13, and 15 of pregnancy.   Offspring
were delivered by Caesarean section and nursed on untreated foster mothers to
ensure that no BaP exposure could occur through the milk.   After weaning at
4 weeks of age, mice received twice weekly applications of 1  percent croton
oil in acetone dropped on the back to determine if BaP-initiated  tumors could
be promoted.   Mice were  sacrificed at age 28 weeks  and examinations made for
                                     6-181

-------
the formation of'pulmonary adenomas.  As the data in Tables 6-42 and 6-43
indicate, prenatal treatment with BaP considerably enhanced tumor development
in both lung and skin during the postnatal period.
     In contrast to the questionable teratogenic effects produced by MCA or
BaP, DMBA and its hydroxymethyl derivatives possess considerable teratogenic
potency.431'432  Treatment of Sprague-Dawley rats with a single intravenous
dose of 7-hydroxymethyl-12-methylbenz[a]anthracene  (about  5.5 to 7.0 mg) in
olive  oil on day 8 of pregnancy produced  100 percent fetal  resorptions.  A
similar dose of  DMBA on day  8 of pregnancy  produced a  smaller incidence  of
fetal  resorptions.  Resorptions  in  control  rats  given  olive oil with or  with-
out saline  averaged 7 percent  (maximum 9  percent).  Treatment with  7-hydroxy-
methyl -12-methylbenz[a]anthracene  on day  12 to day 14  of pregnancy  resulted in
malformations  in every  surviving fetus.   Characteristic pathology was  found in
 all littermates:  stunting,  lordosis of cervical and upper thoracic parts  of
 the vertebral  column,  and an encephalocele and a spina bifida.   No  skeletal
 deformities were seen among 2000 controls.   Teratogenic effects in  the absence
 of an increased resorption rate were also obtained with DMBA given  on  day 13
 of pregnancy.   However, these effects were not as pronounced as those  produced
 by the hydroxymethyl  derivative.  Since 7-hydroxymethyl-12-methylbenz[a]-
 anthracene is a principal metabolite of DMBA and also exceeds the parent
 compound in teratogenic potency, it was postulated that metabolic activation
 precedes fetal  damage.  Furthermore,  it was proposed that the ultimate  embryo-
 pathic substance derived from DMBA may be  a metabolite of 7-hydroxymethyl-12-
 methylbenz [a] anthracene, since teratogenesis  could be inhibited by pretreat-
 ment  of mice with microsomal enzyme inhibitors  before administration of the
                                       6-182

-------
Z i £ 01 VI ri
° °'*!s- i
<£
g

3.
O_
O
UJ
o



H-
«C
mm
CCf

^M
0 Z
1— UJ
ts
Z 0
O fv
>-i Q.
1—
< UJ

^ S
to
i— i 3£

t_i
y
o
o i
0>

«*. m 3
°|i
o'  VI
o *5 §
• 01 C

¥ «


o 

*— ^






>c o «r





CM 91 m
00 00 K)


o
O Of X
S il
m i A5
? S ^.fe
S J S u
— k C =
o m aj •—
•C U CO

VI
~ VI
S 0 ^
•*-» k c
c *j S
0 C k
" S S:
4* f"-I
•^* 0) «
*O -^ 2
A QJ ,JJ
< > ffl























ai
Ot
«
<*-
o
VI
^*
K
x
CO
CM
(A
1

-------
   in
   
   E ">
<4- O 3
 Or- O       r-
                                           CO
                                                   U


                                                   **
                 O O- S_
                 zr ra oi
                    Q. Q.
o_
to
                        -.
                     to  o.
                                    o>
                                    o
UJ
                                CO
                                            CO
                  O O.S.
                                                    
                                                    QJ
 UI
 0-fo
                                                     (O
    UJ
 2g
   t- O


    
                       C C3.
                       
-------
hydroxymethyl compound.432  Unfortunately, the use of large doses of DMBA and
its derivatives precludes the possibility of risk assessment under realistic
environmental exposure situations.
                                   6-185

-------
6.7  HUMAN STUDIES
     Epidemiological evidence on the health effects of POM comes from both
occupational and community air pollution studies.   The effects of ingested
POM on humans have been largely ignored, despite the fact that this is poten-
tially the most significant route of exposure in non-tobacco smokers.
The absence of health effects data on POM ingestion is the most important
information gap in the present epidemiology data base.
     Occupational studies, especially among gas production workers and coke
plant workers, have shown that long-term exposure to the products of the
destructive distillation of coal is associated with an elevated rate of lung
cancer as well as cancer of the digestive tract and other sites (e.g., pancreas,
kidney).  Community studies, which  involve much lower levels  of inhalation ex-
posure and  are more problematic because of the heterogeneous  populations  in.volv-|
ed,  attempt to detect an association between  community morbidity and mortality
rates and some direct or indirect  index of air pollution.  These studies  have
shown, in certain cases, an excess  of  respiratory  cancers and malignancies of
other sites (e.g.,  stomach, prostate)  associated with elevated POM  levels in
urban settings.
      In  both  types  of studies,  inhalation  of  pollutants  is  not limited to POM
 alone but includes  exposure  to  irritant gases such as S02 and NOX,'particulate
matter,  and trace metals  (e.g.,  arsenic,  nickel,  chromium)  which may exert
 their own effects or may enhance the effects  of the POM.   Additionally,  in
 both occupational  and nonoccupational  air pollution studies,  smoking is  a
                                               433 434
 powerful causative factor of similar diseases.    '     In community air
 pollution studies,  occupational exposure represents another potential  causative
                                      6-186

-------
                                      435
  employed in occupations with high levels of exposure to toxic substances.'
       It is tempting to speculate that cancers of the non-respiratory tissues
  associated with airborne POM are not in fact the result of POM absorption
  via the respiratory tract.   It is known that POM are adsorbed to particulate
  matter in  air,  and that POM-containing particulates may be rapidly cleared
  from the lungs  by mucociliary action and subsequently swallowed.   Thus,  it
  is  conceivable  that significant  ingestion of POM occurs secondary to  the
  inhalation  of POM adsorbed to particulates.   However,  the  anticipated inges-
  tion of  POM via particulates  in  air  would constitute  only  a minor increment
  relative to POM ingestion from foods.  Therefore, the more relevant factor
  to consider in evaluating the role of POM  in non-respiratory cancers  is the
  dietary contribution to total POM exposure.  Unfortunately there  has been no
 attempt in  either occupational or community studies to separate dietary fr6m
 airborne sources of POM.
 6.7.1  Occupational Studies
      In occupational situations the effects of long-term exposure to high
 levels  of a toxicant are more easily quantified than in community settings,
 thus  allowing for  extrapolation back to  the effect of small doses  received'
 via  the ambient  atmosphere.   The  most extensive epidemiological studies of
 occupational groups  with exposure to  POM  are  those of  coke  plant workers  and
 gas production workers.   Other relevant studies  include  those of wax pressmen
 and roofers.
     Epidemic-logical studies in different countries have demonstrated that
workers exposed to the products of the combustion and distiliation of bituminous
6-187

-------
coal experience an increased incidence of cancer of several  sites (lung,
pancreas, kidney, bladder, skin).  These studies are discussed below, and the
overall results are summarized in Tables 6-44 and 6-45.
     The earliest association of skin cancer with occupations involving
exposure to coal-combustion products was that of Percival Pott, who in 1775
observed a high incidence of scrotal cancer among chimney sweeps exposed to
soot.  His observation has now become a classic reference of occupational
medicine for cancer and for discussions of coal tar products.  In the early
20th century, several studies established the association between the handling
of  coal tar products and pitch products and skin cancer.    '     Kennaway and
Kennaway, in a  later series of reports, found an increased  rate of bladder and
lung cancer in  occupations involving exposure to coal  gas,  tar, pitch, and
soot.438'439'440  ,
     In  these studies, the number  of workers in any one  occupational  group was
small, and  it was not possible to  calculate relative  risks  for exposed groups
with certainty  or to obtain evidence  for  a dose-response relationship..
     Kuroda441  demonstrated a high incidence of  lung  cancer among  Japanese gas
generator workers.  Although  lung  cancer  was a  relatively rare form  of cancer
 in Japan during the 1930's  (accounting for only  3.1 percent of all cancer),
this  study showed that  lung cancer accounted for 80 percent of all cancer
 among  the gas  generator work force who were  exposed to extremely high quanti-
'ties of coal-tar-pitch  volatiles.
      In a more recent study of 504 deaths among former gas  workers at a
 Japanese steel  plant,  Kawai  et a!.442 found six deaths from lung cancer  in
 contrast with the expected number, 0.180, using other workers at the same
                                      6-188

-------
  o
  UJ
  CD
  O
  o
  a:
EVIDENCE
OGICA
 o
 a.
 UJ
 u.
 o
 C£

 21
 00
VO
 O)
02
 .
as-
a

0
tfl
1





M
4)
§
S»
1.
41
U
°-
o
2I
I/I C
•-.* e o
*J W» O —
o"kZ %
•«-» V^r—
41 4) 3
* > cr> a.
O <0 irt
°^ 4* £
4> c a>
•o « ** >
t» k «B xSl
w.S'S*-
§«.S2,*
— .c « o >>
•*-> u <*- o
CM ro , ,Z
^ O» CU vO _J

en
4>
o a.
t/t o
•«— k
= a. .

(0
2
•M
.c
«B
C
TS
1
3


^
I
M
producer gi
e
xl
M
fc.
41
^f
U
S


U
teelworki
vt
1
01
£
vt
I

I
a»
*— >
 cn-o J*
*j 3 k « e
3 M a* 2
O 0 U o T
c 4) 0 Oj 0»
"^ OJl & 1A <••
II |s|
o = u^o
a»- CD u a

2 I
s i
5 S
I 1
•** M
^ c
k ^
I 5
V»
§ S,
Ij 1
^- *B
**- ^ in
* ^t U
^ o u
a. x s
S- *c
T^S *? ™
1

4* OJ ai
— > — > _ >
3= 3= 3=
It 1* 11
— £ *" ? •" °
£ o. So. z a.

^_ ' \
S5 5
V
«
IfL L
.-.2*1 * i-
^*ac JB «O to t)
£1 II 11
„
££!
*^ 3
k C.
*• (/)
i i-SS
*r «w » o
S £.c *
^"c t *g
•O ttJ 4> 4J ?
.O k O *>
«> *^ S ^ ^
£l 3£i

si-
ti*"*
x at
41 k
«2
SS
£ |s
1 II
||
vi JT
S.^
- «:_
X *^ wi
?0 k
S£5
5S?
i -i|5
0>^M 3
« S^l
5 Ifife.

1 1
-Si
10— C
II if

Z Q. CD vi

UT> to
fc


O *Q










0
aJ
u
•n ai
ed 4-foli
exposun
II
Bl
Is
S*
rl
u *-*
*3
**
*!
41 •«-*
i-
Is;
k
1
v
OD




















1 u
C 0
en c
II
O» (A
S M
k
|5
1^
Vt
4A **
t- U
41 3
fi
Sk
•gi
«/t
a. t/i
X vt
4> at
^g
> — •
Si
z u

en
c
3
_J
•— 
^ >«~


w s a.

2



^
                                                                   S
                                         6-189

-------
UI
UJ
C3
UJ


U_
CO
 a>
 O
 vo

 cu
              g.
                      II
fc£
                     ^-s
                                  S-s
                                  *"i
                                  F-«-"13
                                         S.
                                                  -2
                                                                  3*5
                                                               &
                              'S   SM-
                                      i:
                                                        •52
                     to k.    •—   >.o>
                        «    **!*-** *J
 i   * a. u •—
 a X-c   
 LI *^ « O O C*
 I WO «J *^-
                                      x-SCuio   o.>,«o.o>.  o.
                                      w —— j^ui—   ui'a.jcj^j*  ^
                                                                 .--
                                                               i c-o w   at
                                                               ~--
                                           So k. f^ u

                                           Ji"SS
                                          e QI   >
                                          V Q; «l   01 Cl
                                          > w-0 jrt-0 >
                                          o «/*•»• L <^ «
                                          i   wt o» ft .c:
                                          4i w t jc i
                                         •^ a* a. w a. >»
                                          o > o p o a
                                         *j o*- Si- e
                                                               §   i
                                                               S.:
                                            6-190

-------
t—
G
a;
LU
0
5
s
0
u_
0
LU
0
LU
O

LU
_J
»— 1
CD
F EPIDEMIOLO
0

>-
<
s


^.
"O
a;
c
1"^
c
o
^**^
,3.
i

~ V» T3

u. a. IA


o«

o .c
U tn
O tt

c
• IQ
"H
I'
^ o










« g
"E ^ o
VI t. a;
S 2
s s T;
X O I
«1 S 0.
X o
^ s g
« *^ *rf
T3 T9 «L
« W>  o > ui o
Ol
4^
I 1 J

o|
(rt
&
1!
•r «;
if k


10






•o
i













if
Ol
*t
III E £
i i 1 1 1 1
£ s ^ § g g
g 9 •£ e o S
I 1 S I 1 1
1 £ f I | |
v» VI O VI 1/1 tn
g s *- s s i
2 s « " x S


^ ft;  ^ ew
81 «* *tO
•S « « C *
** a> .x £ Z _ ?
SbSt«fcSS§
2 1 S J 1 « S5
•"•a-ac— JO. —  "^
*l — O
*J •*-» U
•»- 0 Q. vi
O 0. X0>
vt O w
•^ U >*. p
X 0.0 J


r*.


^
^J

-------
CO
cc
CJ
z
 LU
h- 1
»-1 LU
•3* S
_i a
LU
OC LU
^^
O f—
 a.

LU O.
CO LO
ca LU
o ca
u_
CD
tt

j,^8;
§
CO
LO
1
*sO
0)
3
re



e

I + •
£ 15-
!|1
IJ
a —
IS
irt
i

in



+
«••

|
*•
C *A
** ™
S  m *o e u* o
o m CM v to

glT> Ol O» »^» «VI ^*
CM CM »O PO Jl
^ „ _ "" ;

v «- CM m »*» » «^£
2 5 •*• S CM u» *
^»

o <^ o o «^ *° z:
\o cw i*> o* i« **i ^«
eo ot *— CM «n co


; II
*j *j ai
• ' J. 1:
I i I - i
> £ a a. "5 5
v > -0 XI 0 O.
5 ° e 1 * «, e>
1 (U 5 0. 0. -0 •"
; -S 5 S S -5 8
• *"' tJ e e e o*> «*
r» "^ *§ > > > o o
g o " ° ° ° i i








































































K
•*- *
b m
. 01 -— •
I L. A
; o o
e
i >

i*. g
\if
'i «•
> -C
e o
M at
il «
« > x
i- •*•
n #^fn
5 aj •
I? I



































o» ^
in R CM
_' 

-------
  Plant with no gas generator work experience as a control.  Age-standardized
  mortality from lung cancer in the control group was close to that of the
  genera! .ale population.  The large excess of lung cancer deaths among the gas
  workers could not be attributed to smoking.   The authors noted that the
  excess of lung cancer mortality occurred only in the age group of 45 to
  54 years.   Data for those in this  group with ,0  to  !9 years  of gas generator
  work experience showed a marked increase in  lung cancer  risk,  whereas  data for
  those under 45 years  of age  having the  same  work experience  (70  to 19 years)
  showed no significant excess mortality.
       Bruusgaard443 studied 125 deaths among  former gas works employees in
  Norway, al, of whom had at least five years'  work experience and most of whom
  had more than ten years.  The.number of respiratory cancers was higher than
 expected (12,  or 1.6 percent of the total number of deaths, against 1.5 per-
 cent in males  for the country as a  whole). The proportion of lung cancers to
 cancer of all  sites among the gas workers (29.8 percent)  was  also signifi-
 cantly higher  than that in the general  population, 9.2 percent.   In addition
 there were five  deaths from cancer of the bladder-,2 percent of  all  cancers.
 Although Bruusgaard gives  no  exposure data, and occupational  histories  for most
 cases  are incomplete,  he  notes that workers with  a history  of employment in
 the  retort houses had  an especially high  incidence of respiratory cancer.
     Reid and Buck444 conducted a mortality study in 1956 among 800 coke plant
workers randomly selected from a total  of 8000 employed over the years 1949-
1954, inclusive.   The study did not show an elevated cancer risk when death
rates for all  causes and for cancer  were compared  with age-specific rates
prevailing in the period 7950-7954 among workers in a large unspecified
                                    6-193

-------
industrial organization.   The cause of death was ascertained either by refer-
ence to the union's funeral fund records, which were required to be supported
by a copy of the death certificate, or by a special search at the General
Register Office.  The coke plant workers were categorized by occupation:  coke
oven workers, those handling by-products, and maintenance workers  (further
grouped as  laborers, workers, and  foremen).  No total excess in the number of
cancer deaths was  found  among the  coke plant workers  as  a whole, and  there was
a "complete lack"  of  any excess of respiratory  cancer for men  working on the
ovens. When occupational history  was taken into  account, no excessive cancer
 risk was  found for by-product workers and only a small  excess was  found for
 men who had at some time worked at the oven.
      This study was criticized by Lloyd,445 who pointed out that Reid and Buck
 may have underestimated  the number of lung cancer deaths, since the records
  included only men dying  while  still  "on the books" during the period  1949-
  1954.  Lloyd also states that  "the population at  risk and the distribution by
  age and  area of prior employment  were based on an estimate  of figures which
  excluded retirees and those who had  left employment."
       In  an effort to further  quantify the  Kennaway  and  Kennaway data suggest-
                                                                           446
  ing a correlation between occupational  exposure  and cancer mortality, Doll
  studied the mortality among male  pensioners (over age  60)  of a large London
  gas works company for a 10-year period (1939-1948) and compared the data with
  mortality data for the population of Greater London.  Men who retired early
  were included in the study on  reaching 60 so as  not to bias the  investigation
  by the  exclusion of a particularly  unhealthy group who retired early because
  of health  reasons.  Age-standardized mortality ratios were calculated  by use
                                        6-194

-------
  of mortality rates for England and Wales, which were weighted to approximate
  higher rates in Greater London.  The causes of death recorded by the company
  had been copied from death certificates.  The pensioners' mortamy from all
  causes was close to the expected (840 deaths against 856 expected), but the
  mortality from cancer was  in excess of the expected (156 against 123.5; p<0.01);
  cancer of the lung accounted for the greatest excess (25 against 10.4;  p<0.001)'
  which constitutes  a significant increase in  mortality.
       To assess  differences  in  risk  among different  jobs  within the  gas  works,
  Doll  categorized the  pensioners  as  those employed outside the works and those
  involved directly  in  the production of gas or  in handling of the waste  products,
  representing a  low- and a high-exposure  group, respectively.  Excess lung
  cancer among the high-exposure group was  significant (17  observed versus 8.6
 expected,  0.01
-------
low exposure group.   A fourfold higher rate of bladder cancer observed In the
heavy exposure group as compared with the light exposure group verged on
significance (p = 0.06) according to Doll.  He concluded that the mortality of
gas workers varied significantly with the type of work and that mortality »as
highest among workers with greatest exposure to the products of coal carbon-
ization.                                                              448
      A report  on  an  additional  four years of  observation  of  the cohort
provided  follow-up  information on  2,449 coal-carbonizing  process workers and
 579 maintenance workers for mortality rates gathered at annual  intervals from
 1961 to  1965,   Additional employees of four other gas boards were  also fol-
 lowed over periods of 7 to 8 years.   Cause of death in the original studies
 was obtained from death certificates.  The new data showed a pattern similar
 to that of data for the first eight years.
       Heavily  exposed workers  experienced a highly significant elevated  mortal-
  ity from  lung cancer  (p<0.001)  and bronchitis  (p<0.001).  Data on  by-product
  workers  showed no  excessive mortality and over the  12-year  period  provided no
  substantial  evidence of increased occupational risk for  this  group.
       The additional four years of data in this study support the  earlier
  association between exposure to the products of  coal carbonization and in-
  creased lung cancer and also a risk of bladder cancer (p = 0.06).
       The long-term study of mortality among steel workers conducted by Lloyd
  and  Redmond  and coworkers449 has confirmed and extended the well-established
  findings that workers  exposed  to the  coal-carbonization process experience  a
  markedly increased cancer  risk.  The  successive phases  of  this study also
   showed  increasing cancer response  rates with  increasing exposure  and dose.
,445
                                        6-196

-------
       The coke plant workers  studied by Lloyd were  employed in  by-product  coke
  plants.   Exposure  to effluents  from by-product  coke  ovens  is due  to  the
  escape of volatiles during charging, discharging,  escape through  improperly
  sealed openings, and quenching.
       In  these  studies, the workers were classified by work area within the
  plant in  terms of  function and exposure to effluents.  The by-product coke
  Plant was analyzed  in terms of three distinct areas:   (1) the coal-handling
  area, (2) the coke oven area, and (3) the by-products plants for recovery of
  gas and chemical  products.   Since earlier work (e.g., Doll,448 Kennaway and
 Kennaway438'439)  had shown no apparent increased cancer risk for men involved
 in work similar to that performed in areas (1) and  (3),  some of the initial
 study groups included only those workers employed in  area (2).
      In the first phase of this  long-term study,  Lloyd445 undertook a 9-year
 prospective analysis of 2,552 coke plant workers  employed in  1953.   He examined
 the mortality records of  the  workers  in relation  to length  of employment and
 work area within  the coke plant and  compared  the  cause-specific mortality  of
 coke plant workers  as  a whole with the  mortality  of the total steelworker
 population of 58,828 workers.  Coke plant workers were categorized as oven
 workers and non-oven workers.  The excess lung cancer mortality among coke
 plant workers indicated that risk was elevated nearly threefold among coke
 oven workers (20 observed deaths versus 7.5 expected).  Men working on the
 tops of the coke ovens had nearly a fivefold increased risk, and men employed
 five or more years at full-time topside jobs had a tenfold risk (15 observed
deaths versus 1.5  expected).   Furthermore, a significant excess  of cancers  of
the digestive system was observed in  both long- and  short-term non-oven
                                     6-197

-------
workers (17 versus 9.7 expected, significant at the 5 percent level).   Cancer
of the pancreas and large intestine accounted for the greatest excess.
     Redmond et a!.,449 in a follow-up of earlier reports in the series,
examined the mortality records of cohorts of coke oven workers in an expanded
                                                                       445
study at 12 steel plants.  In addition, the data from the earlier study    of
two Allegheny  County  steel plants with coke plants were updated from 1965 to
1966 and were  compared with data from 10 other plants for the same period.  The
cohorts at the 10 additional plants  included all men who had worked at  or on
the oven at any time  in  the 5-year period,  1951  through 1955.   (The criterion
for inclusion  in the  prior Allegheny County study  was employment  in one of
 seven  steel plants  during 1953).
     The  findings of  Redmond  et al.449  further indicate that both the  level
 and  duration  of exposure to  coke oven emissions  are correlated  with mortality
 from various  types of cancer,  further demonstrating the  additivity of  time  and
 concentration.
      Overall  mortality of white coke oven workers is somewhat less than
 expected.   When mortality is categorized by cause, the rates for coke  oven
 workers are significantly elevated for malignant neoplasms (relative risk,
 RR 1.34;  p<0.01), for malignant neoplasms associated primarily with respira-
 tory cancer (RR 2.85; p<0.01), for kidney cancer (RR 7.49; p<0.01), and for
 prostate  cancer  (RR  1.64; not significant).
      The  study showed that men employed at full-time topside jobs for  five
 years or  more have a relative risk  of lung cancer  of 6.87  (p<0.01), compared
 to risks  of 3.22 (p<0.01) for men with five years  of mixed topside and side-
 oven  experience and  2.10 (p<0.05) for men  with  more than five years of side-ovenl
                                       6-198

-------
  experience.   These data indicate a definite gradient in response based on
  both type and duration of exposure.
       Overall,  the  study confirmed the  Lloyd445 findings of a more than twofold
  excess  of mortality due to  respiratory cancers in  all  coke oven  workers.  A
  new  finding of Redmond's  was a  significant  excess  of kidney cancer among  coke
  oven workers  (RR 7.49;  p<0.01).
      Redmond concludes  that the 6.87 (p<0.01)  relative  risk for malignant
  neoplasms of the respiratory system for men employed full-time topside, and the
  1.70 relative risk (not significant) for men employed less than five years,
 constituted further evidence of a dose-response relationship.
      In a later paper, Redmond450 summarized and analyzed additional  data
 (from 1967 to  1970) for the cohort that demonstrated a  consistent increase in
 the level  of risk of malignant  neoplasms of various sites with increased
 exposure for each of the coke oven exposure groups  studied.   Further,  the  risk
 for side-oven  workers,  which had not been statistically significant in the
 earlier  studies,  reached significance (RR l.79;  p<0.05).   Although no  dose-
 response relationship was  apparent, the relative risk for cancer  of the
 pancreas and the  relative  risks  for respiratory diseases  other than cancer
 increased  markedly with  length of  exposure (see  Table 6-45).
     A recent study by Redmond et  a!.451 again confirmed  elevated  risks for
 coke oven workers for lung cancer  (44 deaths versus 24.5  expected;  RR 2.01;
P<0.01), and genitourinary cancer  (RR 1.82; p<0.05), due primarily to a
fivefold increase in kidney cancer.  Data on non-oven, coke plant workers
continue to demonstrate excess  kidney cancer, and the most recent studies in
this series show that incidences of buccal and pharyngeal malignancies are
highly significant.   The overall  conclusion of the  paper is that "these
                                     6-199

-------
observations indicate the need to consider non-oven as well  as oven workers
when evaluating cancer hazards in the coke plant."  This finding is of great
importance since non-oven coke plant workers are exposed to considerably lower
concentrations of coke oven emissions and, hence, of POM than oven workers.
     Hendricks et al.452 examined the incidence of cancer of the scrotum among
82 wax pressmen over a 20-year period.  The refinery population as a whole had
a cancer  incidence  close to that expected on the  basis of the age-adjusted
rate for  the  general white male population.  The  82 wax pressmen, with 10 or
more years  of employment, represented 3 percent  of the  refinery population but
accounted for about 11 percent of  the cancer cases.   The cancer rate  among wax
pressmen  was  4 times that of  refinery workers  as a whole, which was similar to
 that of the general male population.  Of  the  19  cancer cases among pressmen
 with 10 years or more experience,  11  were scrota! cancers.   The greatly
 increased risk for scrotal  cancer of the  pressmen can be accounted for by
 their skin contact with the crude wax,  which contained large amounts  of aro-
 matic compounds.  Workers handling the finished wax did not show a signifi-
 cantly elevated risk.
      Hammond et al.453 conducted a prospective study of mortality from 1960
 through  1971 among 5,939 male hot pitch roofers  and waterproofers.  Motivation
 for the  study was  provided by the knowledge that most of these workers were
 frequently exposed to extremely high concentrations  of BaP  in the air they
 breathed.  The  principal criterion for inclusion of  an  individual  in the
 cohort was that he must have been a  member of the United Slate, Tile and
 Composition  Roofers,  Damp  and Waterproof Workers' Association  for at least
  nine years prior to the onset of follow-up on January 1,  I960.  The  inclusion
                                       6-200

-------
  of workers  irrespective of whether active, probational or  in  retired  status
  tended to reduce the heaHhy worker bias which would apply to the follow-up
  only of active workers.  The quality of follow-up for mortality was excellent-
  only 151 men were unsuccessfully traced with respect to survival status during
  the 12 year follow-up period; 1,798 deaths were recorded in all.
       The cohort person-years  of experience were classified according to years
  of past membership  in the  union.   Observed deaths in major cause categories
  were  compared with  expected numbers.   The  latter  were obtained by applying age
  and calendar year specific rates  for the entire male population  of the United
  States  to the corresponding person-years of exposure in the cohort classes.
       In all  age classes except 50-59 years, the ratios of observed to  expected
  deaths slightly exceeded 1.  Thus, if there was a healthy worker bias,  it was
 more than exceeded by generally higher rates of death in the worker cohort
 than among the general population of United States males.   The overall mortal-
 ity ratio for workers with  only 9-,9 attained years since joining the union
 was 1.02,  compared  to 1.09  for those with  20 or more  attained  years since
 joining.   The respective cancer mortality  ratios of 1.07  and 1.45 showed the
 levels of  these ratios were not generally greater  than for  certain other
 causes  of  death such  as  accidents  (1.59 and  1.41,  respectively) and deaths
 from other respiratory causes  (1.96 and 1.67,  respectively).  The  mortality
 ratios for all  other  causes than specified above, however, were less than 1.
 For cancer of  the lung, the mortality ratio rose from 0.92 for 9-19 years of
membership to 2.47 for 40 or more years.  Taken in itself, the latter gradient
might be dismissed as merely the result of an attenuation in the healthy
worker bias with the passage of time.   However, similar gradients  were not
                                     6-201

-------
seen for the mortality ratios in the other major cause of death categories.
Thus, there was support for the concept that there was an increased hazard of
lung cancer with extended duration of exposure.
     As to whether excess mortality can be attributed to BaP exposures, the
question is moot.  This is because, even though there were extremely large BaP
dose gradients (from averages near 14,000 ng/lOOOm  in the roof tarring areas
to 6,000,000 ng/lOOOm3 near kettles), internal dose-response comparisons were
not presented in the article.   Presumably the  data were not available for
different categories in the cohort of workers.
6.7.2   Community Studies
     The observation of a  link  between  an occupational chemical exposure and
irreversible disease inevitably prompts a similar investigation in the  commun-
ity  setting.  With  POM, investigators  have  long  sought a measure  of  the excess
risk of lung cancer presumably  attributable to ambient air pollution by POM.
None of these  attempts  have  provided unequivocal  results and thus a  concensus
has  not yet been reached  regarding a causal association  between cancer  and  POM
 in air.  Nevertheless,  it is commonly assumed that the  incidence  of  disease
 from exposure  to a carcinogen is directly proportional  to  the  concentration
 to which the population is exposed,  regardless of how small  the  dose may  be.
 The difficulty in evaluating the results of community studies  involving low
 level  carcinogen exposures is that the excess risk ratio relative to "background1
 cancers in the population is usually extremely low.  However,  a ratio of
 1.05 or less would still  represent a large number of cancer deaths per year.
 Several major attempts have recently been made to quantitate the carcinogenic
 risks of POM present in ambient air.
                                     179,454,458
                                       6-202

-------
       A great problem in the analysis of community air pollution studies is the

  confounding effect of cigarette smoking which is the major etiologic agent in
  i            434
  lung cancer.      A second confounding factor is occupational  exposure.   This

  is of much smaller magnitude for the general  population than  smoking,  but  it

  is thought to  be  large  in comparison to any  effect  caused  by  air  pollution.433

  Thus,  an  increase  in  lung cancer caused by air  pollution will be  difficult to

  attribute  to a  particular etiologic  agent.  Few of  the  community  studies of

  air pollution have  controlled for these  two factors.


      Hammond459 has pointed out other problems  that arise in air pollution

  studies.  First, the kinds of information on which estimates of increased risk

  in smokers and certain occupational  groups are based are not obtainable for

 the subjects in general  air pollution studies.  These include  the  interval

 since first exposure and a rough estimate of the degree of  exposure.   Second,

 the amount and  type of air pollution  varies within the same metropolitan area

 from day-to-day and from year-to-year.   Also,  the levels of air  pollution in

 most cities have changed substantially over the  past 30  years  so that recent

 measurements do  not give an accurate  picture of  exposure at  an earlier time-

 Even current sampling procedures  may  not  adequately  reflect  differences  in

 exposure in different neighborhoods of the same  metropolitan area.   The  high

 degree  of mobility among Americans further complicates the problem of deter-

mining  the type and  degree  of exposure.


     Even if excess  risk for certain diseases is attributable to air pollu-

tion, there is the problem of identifying the specific etiologic agent(s) in

air pollution that may be responsible  for an  increase in certain diseases.   As

in the case of occupational exposure,  community air pollution contains other
                                    6-203

-------
agents besides POM and these may exert their own separate effects on health
or may interact with POM.  In some studios of community air pollution, BaP (a
known animal carcinogen) is used as an indicator for the presence of other
POM.  However, BaP is an imperfect indicator because the proportion of BaP to
other POM varies with the type of pollution (automobile exhaust, coal burning,
etc.).460 It has also been shown experimentally that BaP alone does not ade-
quately account for the  carcinogenicity of complex mixtures such as air
pollution and cigarette  tar.    *
     Several approaches  have been taken in the conduct of large scale epidemi-
ologic studies of the association between general levels of air pollution and
cancer mortality.   In general,  four methodological categories can be  desig-
nated:   (1)  urban-rural  comparisons;  (2) migrant  studies; (3) regression
analyses  of lung cancer death  rates for residents of countries  or states with
respect  to  gross indicators  of cigarette  consumption,  industrialization,  and
air pollution;  and  (4)  retrospective  and  prospective analyses of data for
sampled  persons with individually identified exposure  and background char-
acteristics.
6.7.2.1   Urban-rural comparisons—The urban-rural studies  (in which pollutant
 levels are not measured) were conducted by:   Mancuso    '    in  Ohio; Griswald
 in Connecticut; Haenszel and coworkers467'468 in Iowa; in the  United States by
 Hoffman and Gilliam,469 Manos and Fisher,470 and Prindle471;  by Stocks472 in
 the United Kingdom; and by Curwen and coworkers    in England and Wales.   With-
 out exception, a high to low urban-rural  gradient in overall  age-adjusted death
 rates was observed.  For lung cancer in particular, the ratio of age-adjusted
 mortality rates for males in urban versus rural  locations was consistently in
                                      6-204

-------
  the  neighborhood of 2:1  (Tables  6-46  and 6-47).   While there is  general  agree-
  ment on  the  existence  of an  "urban  factor"  for  cancer  mortality,  there  is  no
  consensus  that  the  "urban factor" is  air pollution.  Alternative  explanations
  proposed to  account for  the  urban-rural  differences  in lung  cancer mortality
  include:   (1) migration  to cities for medical treatment; (2)  increased bacterial
  and  viral  infection  in cities; (3)  increased bronchitis in cities; (4) increased
  occupational exposure to  carcinogens; and (5) differences in smoking habits.
      However, studies of  local variations by Winklestein and coworkers474 in
 Buffalo and Hagstrom and coworkers475 in Nashville showed no consistent rela-
 tionship between air pollution and lung cancer incidence within the urban
 environment.   These studies tend to  support the  contention  that not all  of
 the urban-rural  gradient should be attributed to the effects  of air pollution.
 An additional finding in both studies,  which could not  be explained,  was  an
 association between  stomach cancer and particulate air  pollution.   The possi-
 bility that particulates  were cleared  from the lungs  by mucociliary action  and
 subsequently  swallowed  seems  obvious.   However,  the possible  contribution of
 dietary factors  was  not considered in these  or any other studies of POM   '
 pollution and human  cancer.   In addition,  fundamental differences  in  urban-
 rural dietary habits may  be confounding any  interpretation of cancer mortality
 data with regard to  air pollution in general and POM pollution in particular.
 The quantitative importance of dietary sources of  POM relative to that which
 is inhaled necessitates a more careful consideration of this parameter in future
epidemiclogic research.
     Therefore,  after adjusting for age and cigarette smoking habits,  there
seems to be an unwarranted tendency to equate the residual  urban-rural gradient
                                     6-205

-------
                                                    f- CM co in in in •
                                                CM IO r—


                                                I— O CO
                                                ^-00
                                                   co in
                                                                                       cri
                                                                                                   Ol
8?


D£ cn
UJ «3-
O.  I
   CO
CO •«•
uj CT»
                                                                         a>
        i r.   o> co CM vo oo a> vo «• 10 o in    vo o evj
                                    co
                                   i— esicoevi
                                                   i oo i
                                                    ooi
   CO
                                                                                        CO
                                                                                        «r
                                                                                        O)
•C CO
    UJ
                         0)
                                 p.. to    i— i*. <• oo eo co oo i— f— co CM
                                                                              • CO PO
                                 CO O
                                               <«r p- r- cr. u?
                                                                co cj r^    ^-
                                                                               CO I
                                                                     in   a>
                                                                      •>  4->
                                                                     CM   (O
                                                                                                ai
                                                                                                o
                                                                                                10
UJ CS
CO 2=
                                                  i— ** r^ o
                                                                 en CM oo
                                  to co    ^- i
                                                           •— i— CM
    UJ
UJ
o u-
< o

   0£
  ••UJ
UJ O
CD -Z.
 =3 O
 UJ

 CD
                    a>
                          (U
                                  tot—   coco«rr^inor^.cocnvoo
                                  00 CO      •—'
                                    u
                                                          • to co ao co v
                                                                               to r—
                                                                               in oo
                                                                                         f— !/)  IO
                                                                                         re c  f
                                                               4J O   •—  =
                                                               ce in   F-  -O

                                                               S2   £  T
                                                               O)     Ol  4->
                                                               *j o   e  m
                                                               c 4J   O>  ••-
                                                               •—     0>  T3
                                                                  o>
                                                               O C   
 •£)
 VD

  0)
- r~       «a- to
                                                                                          00
                                                                                                r—   O
                                                                                                 Q.  01
                                                                                          10  O-
                                                                                          Ol  O
                                                                   Q.  =>  «*
                                                        6-206

-------
X •>
LU 1—
00 =3
S P
< <-}
LU
til ^^
kU -*^
M O
oo c_j
en ^
eC LO
t«-i cS
or «a-
Q» O^
>- *""
co .
sTi—
*-^ (_j
i—
^ ^
=> 0
Q. >-
O
O- 3
1 1 1

f^ bO
0 Lug
§LU °
I— " i— •
O£ 00
LU =) -
o 	 1 «c
OO X O
LU LU i— i

rv t i i ^^

LU c£
ej i — .«•
Z OO r—
LU in
*•"•» NX |
*™* cc? r**»
f ^ <*•**) ^^
j_4
1 -^
rv* r i t
LU Z
O
z
0 00

(C LU
o cc
LU «C
h—
OO i
•"3 OC.
"T c=
LU Q
CD 2:
r>» ^
o:
(T3 O
I— U_

1


1
1

"•B
5;
£
| ="
=
t
•^
fl
JC









*fl
cc












c:
_g
3














1



^
3
U

U
c
c
o



1
X


1


4J
3
U
i
5
n
o
X
&


n
0



^
CJ
OJ
c
c
o


^
o
1


S
k
Q.



O mo co *o o m f— m
W t*l r— CMCOOV<*^ ci 5;



CM *" *Vd v-~v-




n axo ~»~ntn«,^


V»
uj eo en co *o P*» co o*» co «•
CM *" CM^»— ^«



« ^ *n tf> en p«k en  en <•! vo
S S" S*!C522

X
C- i/l
£• S
_g „
O. o
= g >
- 2, ^ S
'i « if = -
I |o. Illlli
S ^ 5


-s s






CM f*. «9-




5? S



CM O —




moo "T
ss -

>• ff^ n o (**
to o> oC Ofn
CM ^ fv



coo en
SS ~




C\)C3 00
(X co tri
o
0
El
4J (J
C ^
•» o at
.0 v-
; = ^ a
•*• «rt C
en u c >>
CO. = k
3 




.*«. en








on
ss


CM CM



m co
5m




CM CD
m r^.
m CM




mrs.
CM CO
m CM


•o
CP
b.
o
UJ
*-* O)
12
en it
a, S
•— a.
£


fM f*. tO CO O



4


Sf^ in PO in CM
^*> CM c3 m tn




CM *- CM JP «r "^—
"


vo r^ P^ CM *n v
*— *— in




O^o CM ^- in
*n •»• •— o in r»»

co *n m »o o co
mmo CM « «




S'S K ~ '




v rn ^ r^ ch
— — m




CM Id 
2^2 S " °°


1 g
c -^ ^
«J = ^~
cn «a *J
0 k U
x a» -3 ZI ^2
U C TD q C=
it) *O rg ^, 5
I2S 1 t 1
^3 •/> — ( »J

1
!

O
0
; O 0)
«•• trt
"> CL
1 3
i r =
W k

~*
II"
O) 4^
^ U
-~ k»
k O
(J
.*!
* c
k U
$'
o-o
1-

*^ c
S *•*
I-
fi. ZJ
a.
o' ~"S
S 'Si'

IU **
*rt U »2
** ^ k
** Wf
•s le
** k
"c §.*"
r? k -2
o u
***
*o Z^ at
 w» 3

** TD *n
*0 , 3 3
tu c
"^ " >»
i— c x: -^
« |Q 4^ Wl
*J -0 C
O k «*. O)
01 *^ 3 C
£ *«r 3-2
0 S^%
0. -0 .S.
S1 s^S.
k u x:
° Slf
3 O 13 *
^D •»• in
«» X *^ C
i ^l«











VI

O
0)
k
1

§
a>
ai
ai
t_
»*-
o -
€)
U
"a.
C'
!
c
^
2

k
iS
^
£
k
o
U
o
**

2.
3
•c?
e'
e;

o
c
> ' •
«a 4)
JC ^~
ns ^
01 .^-
" I
1 =
1 5
1 -
o 2
*• z
irt qj
^ 5
QC VJ
j Y«








"TJ
Wl
01
«
<*_
O
ft)
E
k
O

"S
U


5

§*

'S
c
o
VI
J

I
^
o
U
1

4)


-------
>- r-
oa to
a: en
O-r-
>-   *
          ! ^
  .
03
Q- UJ
  i U_
O LU
    I/Ta-
 ta: =>
 LU _J O
 Q. O f-»
    x cn
 CO UJf—
 1 1 1 -_ -
 (—     «
 c£UJ-  •«

£3 5
 I UJ  I
a: == r-^

CJ    cn
Z  •• r—
et CO

   S H^

O«C O
in    t—«
f- —>r—
CO ^
                       s  £g  ssssss  25
                       ,J  r-^  — OO — O<—  "• —
                                                          ffl   5RS  5  ZS.X   «   «

                                            S5  SS    S  R   S5S  2  «SR  2   R

                                                       8  £  S2S  S   85?~  *  R

                                                            ^  oj^-<*> * *^  »—com
                                                            «e   utcvj ^
                        «  
                                              4rt  «* W   •*-
                                           tft  Q*    £   ^
                                         ; oi-s s  c o>t
                                         S ••-• u f  -^ c o
                                                                  - ^ = 3 >,
                                                              +1   0> > O. U U
                                                  M .j     -j   a*
                                                                                                 S    S"5 •£ 3  S2"
                                                                                                  in  u  —  3
                                 5  J «- c:  ^  gl  5 _c

                                 wt  3C 
-------
   in  cancer death  rates  with  the  direct  effects  of  ,nna,ed  air  po,,utants    The
   studies by Winklestein et ,,.«74  and ^^  et  ^m ^  ^
   that tendency,.   The major question which remains  unanswered is  what part of
   the urban-rural  gradient can be attributed to the air portion gradient to
  which it corresponds.
  6.7.2.2  Miarant^udies-studies on lung cancer rates ,„ migrant populations
  (in the absence of portion monitoring data) tend to indicate that exposures
  occurring  ear,y in ,„. can  lead to the ^^ Qf cancers ^ ^
  the  victims have  Migrated to a different environment.   !n  this regard   it
  can  be  seen that  the age-standardized lung cancer  death rates  among British
  and  Norwegian  migrants  to the United States are  intermediate between those
  in their original  homeland and adopted  country.««  other  studies of cancer
  mortality among migrants support the concept that these persons  are affected
  by^their former environments and length of exposure in that environment "",477-
      It is  suggested that differences in smoking habits are unlikely to
 account for discrepancies  in  lung cancer rates among migrant and resident
 populations.
 6-7.2.3   Regression an.lys.s-On  the assumption  that populations  exposed to
 POM from  the petroleum industry could be expected to  have higher  rates  for
 certain cancers  than unexposed populations, Blot  et .,.«5  conducted
 of cancer mortality from ,950 to  ,969 in 39 counties where  the  petroleum
 industry is most heavily concentrated.  They found that white male residents
 of these counties had significantly higher rates for cancer of the ,ung, the
 nasal  cavity and sinuses, and the  stomach compared with the residents of
counties with similar demographic  features but  without petro,eum-Process,ng
                                    6-209

-------
plants.  However, these correlations do not necessarily establish cause-and-
effect relationships.  Ratios of age-adjusted mortality rates among white
males in the petroleum-industry counties to the rates in control counties
were:  1.15 for  lung cancer  (significant at 1 percent level); 1.48 for cancer of
nasal cavity and sinuses  (significant  at 1 percent level); 1.09 for stomach
cancer (significant  at 1  percent  level).  The ratio for all  sites combined was
1.06 (significant at the  1 percent level).  White females  in petroleum-industry
counties  also  had significantly elevated  rates  for lung cancer  (though not  for
cancer of the  nasal  cavity and skin,  nor  for  all  cancers  combined),  suggesting
the possibility that community exposure might be involved as well  as occupational]
 exposure.   Data from Blot et al.435 are given in Table 6-48.
                                                                  481
      Using death certificate data from 1968 to 1969,  Menck et al.     found
 elevated lung cancer mortality rates in white males  living in certain heavily
 industrialized  areas of  Los Angeles County.  Populations at risk were deter-
 mined from the  1970 Census  data.  Age-adjusted lung cancer mortality rates for
 Caucasian males and females were  calculated for 13 areas from Los Angeles
 County death  certificates.  A  direct  method of age-adjustment using the U.S.
 1970 Census population as a standard  was carried out, and socioeconomic class
 index based on  average income  and education  level was assigned to each census
 tract.
       The age-adjusted lung  cancer rates  in 1968 and  1969  for male Caucasians
  in the  13 study areas ranged  from 43 to  75 per 100,000.   In three contiguous
  areas of south-central  Los  Angeles County, however,  the  mortality  rate  was 70
  per 100,000 or greater.   The excess of male lung cancers for these  areas  was
  40 percent above the rate for the rest of Los Angeles County.
                                       6-210

-------
TablS 6-*3-  SSU2^;™U5IE? «9?TALITY  RATES,
                                                   TO  KKQ
                 Cancer site
                 	•	_

         Buccal cavity and pharynx
         Esophagus
         Stomach
         Colon
         Rectum
         Liver
         Pancreas
         Nasal  cavity and sinuses
         Larynx
         Lung
         Prostate
        Testis
        Kidney
        Bladder
        Melanoma and other skin
        Brain
        Thyroid and  endocrine
        Bone and connective tissue
        Hodgkins's disease
        Other  lymphomas
        Multiple myeloma
        Leukemia
                                               Rate
                                               1.04
                                               1.06
                                               1.09a
                                               1.02,
                                               1.07b
                                               1.06

                                              1 .'48a
                                              1.09
                                              1.15a
                                              0.98.
                                              1.10b
                                              1.05
                                              1.02.
                                              1.10b
                                              0.94D
                                              1.04
                                              0.98
                                              0.96
                                              1.01
                                             1.05
                                             1.03
All sites combined
                     6-211

-------
     The Inhabitants of the three areas of south-central  Los Angeles  County
are mainly lower and middle class, and the lung cancer rates for Los  Angeles
County were 1.73 times higher in the lower than in the upper socioeconomic
classes.  However, the excess for the three south-central areas remained when
deaths for the 13 study areas were examined for the lower socioeconomic
groupings.
     Measurements of BaP and other PAH in the air and soil (made at four
sampling stations with and adjoining the three south-central areas) showed the
highest pollution levels at the center of the three study areas with increased
lung cancer mortality.
     Neither smoking nor occupational  history data were  available in this
study, 'but the authors argue that neither smoking nor occupational exposure
could account for the excess.  They  suggest that synergistic action between
smoking and neighborhood air pollution, primarily of  industrial origin,
provides  the best explanation  of  the elevated  lung cancer rates in south-
central Los Angeles County.
     A  follow-up study482  including  additional  mortality data  from 1970  and
morbidity data  from 1972 confirmed  the findings of an elevated lung  cancer
rate for  south-central  Los Angeles  County.  Age-adjusted lung  cancer rates per
100,000 were  70.9,  70.2, and 69.2 for the three south-central  areas  compared
to an average rate  of 55.8 for all  14 study areas  in  Los Angeles  County.
Henderson et al.482 reported that the increased risk  was present  in  different
 social  classes  as well  as  in six of eight non-factory occupational categories.
 Females did not have an elevated risk.  Although  current occupation  was an
 important factor in the lung cancer risk of Los Angeles County men,  current
                                      6-212

-------
  occupation did not explain the excess male risk .in the south-central area.
  The authors noted that detailed work histories might reveal an occupational
  risk that could account for the observed increased cancer risk.   It was sug-
  gested that the excess lung cancer rate was  not due to differences in smoking,
  since cancers  of other sites normally associated with  smoking showed no excess
  in  the three south-central  areas.
       In a study of  respiratory  cancer in ten Polish cities, an apparent cor-
  relation was found  between  cancer  death rates  and degree  of pollution by rep-
  resentative POM.483  Data on ambient  levels of dust, tars, BaP, BeP, BghiPer,
  and pyrene for  the period 1966 to  1967 were compared with the age- and  sex-
  adjusted  respiratory cancer death  rates for the period 1962 to 1963 in the
  same cities (Table 6-49).  For the most part,  it was shown that cities with
 the highest death rates had the highest levels  of POM pollution (correlation
 coefficient, R  = 0.66).  No apparent relationship was seen between respiratory
 cancer death rate and air pollution as measured by other indices  (e.g.,  dust,
 tar  substances).  Caution must be exercised,  however,  in the interpretation  '
 of cancer incidence  data in  the  absence of  historical  information  on  ambient
 levels of POM.   The  long latency period for tumor appearance (10 to 30 years)
 often  makes  it  impossible to establish levels of  exposure  at the time of the
 initial  carcinogenic  event.   Moreover,  it cannot  be  stated with certainty
 whether carcinogenesis  in humans  is determined  by  intensity or duration  of
 exposure,  or a combination of both.  In addition, a presumed lognormal distri-
 bution  of  incubation periods for neoplastic disease may complicate the attempt
 to correlate time of exposure with carcinogenic effect.484
     Stocks has  carried out many studies relating lung cancer mortality to air
pollution.  In 1952,485 he reported that, in towns, lung cancer mortality
                                     6-213

-------



JC AIR POLLUTIC
(THE NUMBERS U
CHARACTERISTIC]
a~z
ft r ^ ^>
oo z *-*
O O C3
JE CC.
p. ga 1 1 1
Q t—
UJ Z
f£ **" en
U- C3 >-
°§s
OO —J O
UJ
^ u. u.
LLJ CD
>- 0 •-«
oagt:
Ijj 1^3
So0"
ca cc _j
»-i U, «C
1— "— '
CO Z J-
LU CD Z
> •— < UJ
z i— r>
_J UJ
1 • 1 Q_
»-< OQT
1— Q. O
i— «
O UJ ^£
in 2S
UI t— — i
• Q 00
en z LU
*f *x.ts)
1 UJ
vo oo ac
UJ J—
cu o z
r— •— • UJ
x> a c£.

c
c
o
e concentrat
CO
S_
O)
>

 VD
(M rO CT»
?i^
CO -I-3 S-
"O CO *O
C U CO
«o -o >>
4-1

Q.
J_
CO
Q_
S ME
« 8
o
o>
3.
O. r-
S «
Q_
(0
ca
1 CO^
i/i (J en
c a.
S- 10
eo +•> C
1 — in -i-
C
•«-> »*»
I/I Cl
3 S.
O
•r—


en co «a-
• • •
r- f— OO
CO CO CVJ
eo t>. vo
i-. in o
e~\ en en
r^. eo vo
en i-" en
r- co vo
r-. co vo
t>. en cvj
CO CO CO
t^ CO VO
ia- m co
i*- r^ vo

i— co *i-
CVJ CO CO
i— CO l>~
i— CO CO'
cvj en «a-
Gdansk 1
Katowice 2
Kratow 2

CO «9" CO
r- f- —
r^ r«> co
CVJ CVJ CO
CVJ CO ^1-
•a- in >—
in in vo
cvj co *r
^jD ^O ^^
^r in in
CVJ CO «*
cvj vo t--
CVJ CVJ CVJ
cvj co «er
in en co
*3- «3- m
in co «3-
oo co vb
CVJ CVJ CVJ
co «a- m
tr> in in
in in r~
c
•r-
C 0
(0 0
N C N
-0 N 0
O O N
_J Q. 00

CVJ •— CT>
CO ^Q CO
vo co r-.
<— r— ^"
r— in en
co fv. «r

cvj , cd K
co vo f—
r- in en
in en in
r— CVJ VO

en i— 't>
CVJ VO •—
cvj vo en
r— O CO
evj co co

CO ^~ VO
CO O> CO
1 1 '.
IM i— N
I/) {—? ^—
S- O -O
ro i- *O
3 3 r^i
6-214

-------
  increased in proportion to the number of inhabited dwellings (from a compara-
  tive mortality ratio of 89 in towns with less than 20,000 occupied dwellings
  to 162 in towns with over 200,000 occupied dwellings):  Differences in smoking
  habits could not account for the lung cancer excess in towns compared with
  rural  areas.   Stocks put forward the hypothesis  that  the effects of tobacco
  and  atmospheric pollution  are additive.
      Stocks  and Campbell472  compared the  lung  cancer  rates among men  with
  different  smoking habits in  a rural  area, a mixed urban-rural area, and a
  highly urbanized area (Liverpool County borough) using data  from  a study of
  environmental histories of persons with and without cancer carried out by the
 British Cancer Campaign.  The death rates were related to levels of BaP and
 other POM and sulfur dioxide in the air in each area.
      The rural death rate increased with increased number of cigarettes
.smoked  per week.  Liverpool rates.were higher than the rural  rates in every
 smoking category,  but the urban-rural ratio  decreased  progressively from about
 9 to  1  for nonsmokers to near unity for heavy cigarette  smokers.
      There  was an  "absolute urban excess" in  each  smoking group,  suggesting
 that  an urban  factor was  added to the effects of smoking.  Stocks  and  Campbell
 estimated that about half the  male  lung cancer  deaths  in  Liverpool were due to
 smoking and about three-eighths were  due to "a  factor which is only slightly
present in the rural  areas."
     The level of air pollution increased with increasing urbanization.  The
concentration of BaP  in Liverpool was from 8 to 11  times that in rural areas,
and this ratio corresponded to the estimated mortality  ratio among nonsmokers
in urban and rural areas.
                                     6-215

-------
     Stocks486'487 reported significant correlations between SMR's for lung
cancer and bronchitis and air pollution (undissolved deposit and smoke)
levels in 58 county boroughs (towns) in England and Wales when population
density was held constant.  Stomach cancer was also positively correlated with
air pollution.  Lung cancer gave a correlation of 0.500 with the amount of
deposit when the population density was held constant (p<.002).  Lung cancer
gave a significant partial coefficient with smoke of 0.510 (p<.01).  Bronchitis
gave a partial coefficient with amount of deposit of 0.579 in males and 0.511
in females.
     In an expanded  study,488 a high correlation was reported between  smoke
density and lung cancer  mortality  (r = 0.873).  Differences in mortality were
only partially explainable by social differences (housing indices  based on
numbers of persons per room  and the proportion of employed  and  retired males
in unskilled  jobs).   Bronchitis and pneumonia in males  (r = 0.869  and  r =  0.666,
respectively) and  bronchitis in females  (r =  0.751,  significant at the 5 percent
level) were also  strongly related  to  smoke.   Analysis  for the  relative correla-
tion  of  four  different  POM indicated  that BaP was  the  "substance of  prime
importance" for  lung cancer.  The  trace  elements beryllium  and molybdenum  also
showed associations  with lung cancer.
      Stocks489 reported the results of three  separate  analyses:
      (1)  Lung cancer mortality showed "substantial and independent  correla-
 tions" with smoking and air pollution (measured by smoke,  BaP, and three other
 POM)  in six European cities and two areas of Wales.
      (2)  Lung cancer mortality showed positive correlations with the per
 capita consumption of cigarettes and solid fuel  (but not with liquid fuel) in
                                      6-216

-------
   19  countries.   Smoking appeared  to  account  for ^ ^^^  of
   deaths  in the average  country and coal consumption for about one-third.
       (3)  There was a  large urban excess of lung cancer in English metropoli-
   tan areas relative to surrounding regions after the effects of differences in
   soc,a,  and other factors were eliminated.  This excess was attributed to air
  pollution.
       The analysis of ,ung cancer and mortality in relation to smoking and per
  capita  fuel  consumption in 19  countries was  repeated490 with  more recant
  mortality data.   When  lung cancer rates were adjusted for  smoking differences
  the  adjusted  rates were generally higher  in  countries with high coal
  tlon than in  countries  with low coa, consumption.  Smoking showed  a greater
  effect than air pollution  from solid fuel burning on the lung cancer rate in
 -n aged 35 to 44 (the correlation coefficients with lung cancer were 0 634
 for smoking and 0.470 for solid fuel), but in men aged 55 to 64,  air pollution
 showed a greater effect even when snoMng was held constant Cthe  correlation
 coefficients  were 0.599 for solid fuel  and 0.380  for smoking).
      Carnow and  Meier491 reviewed  the multiple  regression analysis performed
 by Stocks and  Campbell4*9 on the age  and sex-specific and adjusted lung cancer
 death rates for  ,9 countries.  After  taking account of the  influence of the
 covariate  defined as per capita cigarette  consumption, the  analysis indicated
 a  substantial  residual association between the age-sex adjusted lung cancer
 death rate and per capita fuel consumption as an index of air pollution; the
 average age-sex adjusted lung cancer death rates were estimated to increase
by a factor of about 20 percent per metric ton of solid fuel consumed per
capita per year.   However, since fuel  consumption  may be  a  surrogate for
                                    6-217

-------
numerous industrial, climatic, and other national  characteristics,  it would
be presumptuous to attribute all of this relationship to the effects of air
pollution.
     In a similar regression analysis of age-sex-race specific and adjusted
lung cancer death rates for persons 35 years of age and older in the 48
contiguous United States, cigarette sales per person and average BaP con-
centrations in  the  ambient air  were employed as the  regressors.  Population-
weighted  averages of BaP  concentrations  in  samples taken from  1967  to  1969  in
 urban and non-urban strata  throughout the country were  used by the  authors  as
 a convenient representation of general  air pollution levels.   The  specific
 BaP measure was used as an "alias" or surrogate variable to represent all
 forms of general air pollution in the states.   Thus, a unit of "air pollution-
 was defined as a one-year average of 1 ug BaP/lOOOm3.   It was found that the
 lung cancer death  rates for white males increased by a factor of 5 percent for
 each such unit of  pollution; the  increase was 15 percent for non-white males,
 6  percent for  white females, and  1 percent for non-white females.
      The authors felt that  the foregoing effects could have been overestimates,
  because  1967-69 levels of  BaP  were only about 40 percent of their  general
  levels in  the more typical  period of the early  1950's.   It is questionable,
  however, whether secular changes in  BaP adequately  reflect long-term changes
  in overall  air pollution levels.   Then, if BaP is  an inadequate surrogate  for
  !ong term changes, perhaps downward adjustment by a factor of 40 percent would
  be an over-correction.  This  analysis by Carnow and Meier491  was included in
  the National  Academy  of Science's report on Biologic Effects of Atmospheric
   Pollutants:   Particulate  Polvcyclic Organic Matter.
                                                     179
                                        6-218

-------
                     ive and prospective analy.es-Thls category of community

  studies on POM includes both prospective and retrospective large scale

  sampling designs.  Prospective studies of lung cancer death rates were

  carried out by Buell  and Dunn492 in male veteran residents of California

  and by Hammond and Horn493 in United States white male veterans.   Retrospec-

  tive studies  were performed by Dean494 in Northern Ireland,  Haenszel  et al>7>
  T"OO •
      m the  United States,  and by Hitosugi495 near Osaka.   The  designs of

  these  studies permitted  finer adjustment  for the  individually  determined

  background  characteristics  of age,  sex, and  smoking  habits.  In some  of  the

  designs, individual and family  histories  permitted the evaluation of  the

  duration of urban or rural  residence and  other factors in conjunction with

  the urban-rural exposures.  The,characteristic high-to-low gradient in

  urban versus rural residence was consistently confirmed in all  of these

  studies.  When expressed per unit.of BaP as an index of pollution, male lung

 cancer death rates increased by a factor of 4 to 5 percent per  Mg/1000m3.

 These agreed well  with the gradients estimated by the cruder population

 and regression studies  previously cited.

      Carnow  and Meier's491  proposal.of a  1 Mg BaP/lOOOm3  "unit  of  pollution-

 may have been  more for  analytical  convenience than as an  index  of  widespread

 practical use.   In addition  to BaP concentrations,  Hitosugi felt it necessary

 to  characterize  air pollution  levels  in terms  of measured rates of dust fall,

 rates of sulfur  dioxide accumulation, and  suspended particulate concentrations

 Carnow and Meier491 have also pointed out  that long term trends in predominant

 types of fuel might vitiate the validity of BaP as an index.  Specifically,

 data from Finland and New Orleans did not conform well with the  BaP index/

primarily,  it was believed, because of atypical fuel usage in those areas.'
                                     6-219

-------
Even where the BaP index has seemed to perform well  in past studies,  the
habitual use of a single component as an index and its possible incorporation
into the language of regulations would seem to invite misinterpretation and
improper manipulation.  A more prudent approach would involve a multiple
component  index based on such measures as obtained in the study by
Hitosugi.495   It  seems  that Carnow and Meier would also prefer using a com-
posite index,  since one of  their strongest  recommendations  was to obtain more
precise measures  of POM, particulates,  and  irritant  gases  in ambient air
 sampling.   They also strongly recommended better documentation of cigarette
 smoking habits and obtaining a greater variety of samples  from occupational
 and residential strata.
      Taking a different approach  based on the risk of highly exposed workers,
 Pike  et al.434 calculated  a small  but non-negligible risk  associated with air
 pollution.   Pike et al.  used the  level of  exposure to BaP  among British gas
 workers  showing  an excess  lung cancer  rate in order  to extrapolate  to the
  lung cancer rate attributable  to BaP in  general  urban air pollution.  They
  explain the algorithm used in  the extrapolation as  follows:              ^
      "   The carbonization workers were exposed to an estimated 2,000  ng/m
       BaP for about 22 percent of the year (assuming a 40-hour working week,
       2 weeks paid  leave,  1 week  sick leave); very roughly, the men were
       exposed to the equivalent of 440 (2000 x 0.22) ng/m   BaP general air
       pollution.  This  exposure caused an  extra  160/105 lung cancer cases,
        so  that we may estimate,  assuming  a  proportional effect, that each
        ng/m3  BaP  causes 0.4/105 (160/105  *  440)  extra lung  cancer cases per
        year..."
        This calcuUtion yielded an excess rate of 18/100.000 lung cancer cases
   per year for a city with 50 ng/m3 BaP air portion.   The authors  took this
   calculation and Stocks comparable findings as support for the  "notion  of  a
   simple proportional relationship between increasing BaP  concentration in the

                                        6-220

-------
Qf
             r

   «r an, ^ rate of
   excess ,ung cancer rate

   U.S.)  was  eo.uiva,ent  to the  ,ate due to exposure to  ,0 noV Bap ,„

       WM1.  occupationa, stud,es have p.ov,'^ stTOna evince that exposure
   to  high  1TOJ. of POM causes an jncreased Mi

   and fo, non^io.nant respl>ato,y d1.M.a.  the
   t-n studies remain 1nconclu.1v..   flt
  consensus that th. dWmnCM  between  ,ung
  -as  a,e due to  8ene,a, .1r  po,lutl.on.  ,„

  1-n. cance, in  the  UnUe, states, „,„,„.<*  has wr,tten that
  ef ect  ,. undoubted, n  „ ..„„ not certal.n  ^ n  ^ ^ ^ ^     ^^
  Ponutants  in the .,r. .  Goldsraitn and ^^^ ^ ^^ ^^ ^ ^

  »t.«tur. on a,> ponton and lung cancer,  conc,ude ^ ^ .^
  not suppon the hypothesis  that .p.llut1on  ^ ^ ,.  ^ urban factor „
 they add that it is a,so not „...,„,.  to reject the possibi,ity

      Other reviews have conceded  that *h,,e air p0llution My be a  signif-
 icant facto, in  the  etiology of  ,ung cancer,

 of r,.k  vary  t,emendously.   « the Znternationa, Symposium on Genera, A1r
 Ponution and Human HeaHh With Specia, Reference to Uong-Te™ Effects held in
 Stockholm, March 8-n, ,977 severa,  concisions Were reached concernin, carcin-
ogenic effects of PCM.  participants  at the  Symposium  agreed on the
conclusions:  w

     1;    Lung cancer is  more common  in urban versus rural areas.
    2-    Cigarette smoking is  the major  cause  of  lung  cancer, and
          contributes  to  urban-rural differences.
                                    6-221

-------
    3.    Differences in lifestyle (e.g.,  alcohol  consumption,
         nutrition) and occupation are not important causes of
         the urban-rural difference.
    4.    Combustion products of fossil fuels, probably acting
         together with cigarette smoke, can account for cases
         of lung cancer in large urban areas.  The number of
         such cancers  is in the order of 5 to 10 cases per 100,000
         males per year.
    5.   Although well-documented  urban-rural differences  have
         been described for cancers  in other organs,  it  is, not
         possible  to evaluate the  role of air pollution  in deter-
         mining such differences.
     When evaluating the carcinogenic threat of  POM to man  no one has  yet
considered the relevance of dietary sources or its impact on  gradients of  can-
cer for sites other than the respiratory tract.   This may be especially impor-
tant in light of the fact that the absolute rate of digestive tract cancer in
both rural and urban populations can be considerably higher than the rate of
respiratory tract cancer.465'474'475  In addition, the oral administration of
POM to experimental  mammals commonly produces lymphomas, leukemia, and tumors of
the lungs  and endocrine tissues; thus the possibility  is raised  that  the car-
cinogenic  impact of POM in humans  may be  underestimated  by considering only
inhalation as the  primary  route  of exposure,  and lung  cancer  as  the.endpoint
of greatest concern.
                                      6-222

-------
  6.8  ECOLOGICAL EFFECTS
       POM ,. ubiguitous ,„ the natural  environment, ar^ng
           (e.g., petrol spmaae,  transport of combustion products)  but

      ;ls; °:19r;9r ™
      bactena,    -499 aUhoggh SQme
  anaerobe .acterta do  not syntn^ze POM.=00
  •col^lc.,  impact of POM ,„ our terrestM3l  and
  P> Seated by uncertainty OV6r th
  6.8.1  Effects on Microorganisms  and  Algae
      Pav^r and cow^ers^  have  indicated tHat tne green a,ga> Scanedes»S
 -tus,  ,ay contain  significant quantfties of POM wnicn vary       ~
 geograpMca,  Ration where  it tt grown (Table 6-50).   From the
 distribution of  carcinogenic POM wMcn was datected_  these 1nwi1g(ttori
 duo-ed that ..t of the POM present was due  to Moaccu^ation  fro™ envi
 -ntal Ci.e., anthropogenic) sources as opposed to endogenous Nation
 Effects of these POM on algal growth were  not indicated.
     Although it was  reported that phenanthroquinone, a degradation product
of Phenantnrene,  reduced the  surviva,  of various a,ga, species « concentra-
tions of 40  to  BOO Mg/,iter,^ others demonstrated that POM at  low ,eve,s
                                        Qf
can enhance a,ga, growth. =«3  The
        , and AnJdstiodesHs oranunii  was  enhanced by exposure to ,0 to
20 Mg/liter f,uoranthene;  ,,,2-benzopery,ene; 3,4-benzofluoranthene; indeno-
C,2,3,cd)pyrene;  1 ,2-benzanthracene; BaP; and DBahA.  Moreover,  the degree
of growth enhancement appeared to correlate with carcinogenic potency
                                  6-223

-------
r—
O
in
CO
z.
o
»-i
^
s
_J

c_)
HH
C3C
a.
**C
^i*
CD
O
UJ
CD
f f
I_J i
0
•z.
o
I-H
•z. ' '
u_
«£
to

1 —
UJ
O
o

to
o
1— t
_1

o
o
a.
o
in
CO
CO
H—








(f- CU
O i—
a.
• E
O CO
•z. to








O cO
Ol'r-
C. CO
CO J=
CQ 1—








CO
H— CU
O r—
Q.
O CO
•z. to





•o >>
c e
CO g
5g
J- CU
0 CD
a



















m •*
r— P—











o co
to to
r~™ r"™
co en
r-^ Lfj









en en






/-> /-N
+ 1 +1
en m
*d- en

m •*
CO 00







cu
c
cu
•p
eu co
c s-
-Benzopyre
-Benzofl uo
•3- «*•
CO CO







CO











,-^
Ti
to
in
C-
to
CO
CM









en






/->
CM
cn
s^
cn
0









enopyrene
-a
t— i







00 00












CM tD
CO.
00
CO CM
+ 1 +1
CM cn
00
en co









CO cn






/-; /—s
o co
in cn
^ v^
•3- I—
«^- CSJ
in







cu
cu
oranthene
2-Benzoper
3 i—
LI_ i —







CO












CM
O
+ 1
3
CM









cn






'~s
+ 1
CO
UD
CO
<— '
co
co





cu
c
cu
e
CO
t-
o
12-Benzofl
**
•—

"cO
E
O •
c w
0) C
O 0
1- 3
0 J3
•r™
o> s-
co -P
CO (A

CU
x: i—
•P CO
e
C. i-
•r- 0
c.
U)
0 0
u cu
CO CO
<4- CO
u
c cu
•r- x:
i- -P
cu
-P C
tO
U (/>
LO C
o
TJ «i-
e -p
CO CO
•r-
CO >
CU 0)
3 T3
'io -a
> s-
(0
c -a
•5 -P
CU 10
E ^3
•O IO
CU C
-P (0
CO CU
u e
T3 U
•i- '-P
CU
LO E
•r- -C
•p
10 'r-
0) t-
3 CO

CO >>
^> f^
4- S-
0 0
O5 •>
c to
^ 	 ^—
le scatter!
istributior
i— "5
CO
6-224

-------
   Blowing.the demonstration that Chlorella vul^aris  cou.d utiHze  acetate
   -  the growth medium to  biosynthesize  BaP,  an  hypothesis  was  presented  that
   POM may act as  an endogenous  growth promoter ,„ plants.^  ^ Contant1on
   is  also  supported by the observation that growth of  higher pUnts  (rye
   radish^tobacco) is  promoted  by POM in the  same fashion as the growth of
   algae.     However,  in comparison to the POM accumulated in pUnts from
  anthropogenic sources, the contribution by endogenous synthesis is ,ike,y
  to be quite insignificant.501'504

       In various microorganisms,  POM may either  promote or  inhibit  growth
  Toxicity may be expressed more as  a function of increasing concentration
  rather than specific  chemical  structure.  On the other hand, Hass and
  ^legate   reported that a structure-activity correlation may exist for
  the  effects of polycyclic aromatic hydrocarbons on the growth of Escherichia
  coll.  At concentrations in  the medium of lo'7 to lo'5 mo!ar, anthracene
 Phenanthrene,  chrysene, DBacA, and pentacene inhibited bacterial growth
 The more angular configurations,  1,2-benzanthracene,  DBahA, and BaP  pro-
 moted the growth of E, co,,.   lt was concluded from these limited data  that
 growth  promotion may  require  the  presence  of POM with  an angular acene  con-
 formation, whereas  inhibition might  occur  with both  Hnear  and  angular acene
 molecules.

      Recent studies conducted with marine  bacteria do  not support the hypo-
 thesis  that the effect of POM on microorganisms might  be structure-specific
 Instead, Ca,der and Lader505  reported that aromatic hydrocarbons inhibited
 the growth of Serratia marinoruba and Vibrio parahaemolvti^ ,„ .  manner
which was dose-related and a function of water so,ub,my.   Thusi a  saturated
                                   6-225

-------
solution of BaP would have the same impact on bacterial  growth as a saturated
solution of naphthalene, even though their respective solubilities differ by
several orders of magnitude..  Consequently, when evaluating the effects of
POM on aquatic organisms it may be necessary to balance inherent toxicity on
a molar basis against water solubility.  The significance of considering both
variables  in the ranking of POM for toxicity to marine bacteria  is depicted
in Table 6-51.  Furthermore,  one must  recognize that  under  normal  conditions
the  low-to-medium  molecular weight POM may contribute less  to  overall  environ-
mental  toxicity than the  higher weight molecules  because  the  former  are  more
 readily lost by volatilization and degradation.
      Poglazova and Meisel507 reported that a wide variety of  bacterial types
 can accumulate BaP without metabolizing it.  Localization'of  BaP occurs in
 Hp1d granules and lipoprotein membranes, as would be expected based upon the
 lipophilicity of most POM.  Other investigators508 have reported that cultures
 of Pseudomonas aeruginosa and E, coJl absorbed 90 percent of the BaP added to
 the culture medium, 10-26 percent of which was metabolized.  Whereas these
                                                                        509
 authors reported  a  growth-stimulating  effect of  BaP  on bacteria,  others
 observed  a BaP-induced disorganization of colony structure.   The  localization
 of  POM in bacterial  membranes is  the  most likely explanation  for  observations
 that  BaP  depressed  the metabolism of  lipids and  inhibited  the formation of the
  electron  transport  system in Staphvlococcus aureus.510'511   Whereas BaP was
  inhibitory to membrane formation and function,  benzo[e]pyrene was not.
       Several studies conducted with cultures of yeast have confirmed that
  BaP readily'penetrates the cell, is partially metabolized, and becomes
                                      6-226

-------
 l-D
 O
 LO
      en
v-'  >>
   -P X

   U <=C


   i"
   t— o
    -Q
      O)
     r—

  ^^ ^5
  03  E
                                    r-» co CD co <• 10 o
                                    cn ID o csj CM csj evj
                                   "*• •— oo o

                                   O C\J CM «^ o i
                                                     '00
  to
  z
  o
  CO
  oe.
  

 O)

-Q
 (O
                                                  QD
                                               r— 00
                 **• •=*• ID LO to fx. ^>

                  ooo o'o'o'o


                   x  x  x x.  x;  x.  x

                  CO CO CO l£> 
  •r—
  4->
  rb

  "cii


  ns
  •r—
  S_
  cu
  *j
  u
  CO
  .a
                                                      s-
                                                      ai
                                                     ^3

                                                      rj
                                                      c

                                                      c
 01

 u
 0)
 T3

 •P

 ai
 u
 i.
 ai
 a.

 in
 re

•a
 , c
c: ro .c 
CO Q.T-
C O i_j
'- C CO
>i CO ~
X 03 CM
.
-P
03
1 —

•*
o
o
LD
CM
_p
(0
S-
O)
-p
m
T3
CO
-p
.J2
-a
c
l^-l
co
£
a.
S-
ai
a.

^.
1 ^
., —
u
x .
O u)
-P r—
O
CO S-
•P C
^s
o
t-
co o
"O
3 -C
•p +J
C O
CD S-
ro en
xi
                                                       6-227

-------
                                                                            512-51'
localized in I1p1d components of the cell,  including Uproprotein membranes.
As a result, qualitative and quantitative alterations in the formation of
11pids, especially phospholipids, have been observed when Saccharoses
cerevisiae was exposed to BaP or DBahA.  On the other hand, weak- or non-
carcinogenic POM  (e.g., pyrene, dibenz[a,c]anthracene) were reportedly without

effect.
      It has also  been suggested that once  polycyclic aromatic  hydrocarbons
are absorbed by microorganisms  they can  be transferred  throughout the cell
population.by cell  surface contact.  Studies  with various yeasts grown in the
 presence of BaP demonstrated that greater than half of  the hydrocarbon accum-
 ulated by yeast cells could be transferred to recipient yeast cells .which
                               515
 were grown in a normal medium.
      Sludge microorganisms were adversely affected  by a variety  of carcino-
  genic  POM.516  An  inhibition of oxygen  uptake of  varying  degrees was obtained
  with different sludge microorganisms exposed to:   DBahA;  7-methyl-l,2-benz-
  anthracene;'l,2,4,5-dibenzopyrene; MCA; 2-nttrofluorene;  2-fluoreneamine;
  N-2-fluorenylacetamide;  7,9-dimethylbenz[c]acridine; 7,10-dimethylbenz[c]-
  acridine; dibenz[a,h]acridine; dibenz[a,j]acridine.
  6.8.2  Effects on Aquatic Organisms and Amphibians
       The toxicity of POM has not  been  extensively  studied in fish, amphibians,
  or  aquatic invertebrates.  Nevertheless, it is known that marine fish and
                                               J    .    504,517-520  Furtner-
  invertebrates can accumulate  BaP  from  polluted waters.             Further
  more,  freshwater  and marine fish  are capable of  metabolizing a  wide variety
  of hydrocarbons.
                                      6-228

-------
        The induction of cytochrome P450-dependent MFO activity ,„ „.„ and
   invertebrates is no* receiving considerab]e ^^ ^ ^ ^ ^

   activation of carcinogenic POM.™  Investigators  postdate that the capa.
   city of aquatic organisms to  absorb  and  metaboHze POM f.  important  ^
   in deter^in, carcinogem.c r,sk  for ^ speci.es_  ^ ^  a]so  ^

        fo, the  t™,.f.r of POH and  its activate, metabo,ites throu8h the food
        .

        Ue ana c^or^" shoKed that ' VnaphthaUne an, 3H.Bap were
        up throuflh the „„,. Qf three ^^  ^^..^^^^  ^^   ^

          .  BaP Was metabo,iZed in  the^ive,  and, within a few hours>  metaho-
  1-t.. we.  t™sfer,ed to the  „„  „«„.,. and  excreted predomjnant,y  )n u-

    ,ne.  The  main product  of BaP metabol1sn, was  7>a-  flhokas  and  coworkers52, ^^^^ demnstrstea

that the oxidative metabo,ism of BaP by trout  ,iver ,ed to  the formation  of
                                  6-229

-------
reactive intermediates which became covalently bound by protein.   Furthermore,
trout liver microsomal enzymes activated BaP, 2-aminofluorene, and 2-acetyl-
aminofluorene to compounds which were potent frameshift mutagens in the Ames
Salmonella assay system.  In comparison to the Sprague-Dawley rat microsomes,
trout liver .1crosoa.es metabolized BaP 15 to 30 times faster when the quantity
of BaP  metabolites  produced  in  15 minutes was  expressed per nmole of cytochrome
P-450 or per  unit  of cytochrome c  reductase.522   Hepatic  levels  of  cytochrome
P-450 are higher than'in the rat,  whereas  cytochrome c reductase levels  are
 1oWer.   However, the Vmax is higher than for the rat and  the  Km is  lower when
 BaP is  used as a substrate.
      Carcinogenicity testing of POM in lower organisms has been pursued for
 many years.  For the most part, compounds which are carcinogenic in mammals
 also produce hyperplastic reactions or tumors in lower animals.  Arffman and
 Christensen523  summarized much of the early work performed with the newt, a
 salamander.  These studies  showed that  an early proliferation of the epidermis
 accompanies  the subcutaneous  injection  of tar,  BaP,  and  MCA  in  the tail  region
  of the newt.   They confirmed these results  by showing that an  epithelial
  proliferation commonly occurred with the injection of BaP, MCA, or DBahA.   The
  highest incidence of epithelial reaction was obtained with DBahA.
        Following the demonstration in the clawed toad, Xeno£us laevjs,  that
   implantation of MCA crystals  induced lymphoid tumors, a similar study was
   undertaken to  confirm  this result using BaP.524  The implantation of BaP
   crystals  (1.5  mg) in the abdominal cavity  of adult  Xenoeus  laeyis laevls
   produced  lymphosarcomas in 11 of  13  animals  within  86 to 288  days.  Advanced
                                      6-230

-------
  tumors were found which affected the ,.ver
  plantable to Immature recipient Xenopus.
      POM are wide,y distributed ,„ the environment as evidanced by their
 detection^ sediments,  soi!s, air, surface waters,  and p,ant  and  ani»a,
 t-sues.      The eco108ica,  i^act  of these chemicals,  however,,is  unc.rtain
 Numerous  studies show that despite  their high Hpid  soiubiHty, POM show
 Uttl. tendency for bioaccunu,ation ,„ the  fatty tissues of anira,s or „
 This observation is not unexpected  in Hght of convincing evidence to show
 that POM are rap,d,y and extensively metabo,ized.   since on,y ,„« ,eve,s of
 P0« are detected in p,ants and ,ower organ,s.s,«5 transfer of  pflM  ^
the food chain does not see* me,y.  The direct impact of  POM  on p,ants
ani.a,s,  or the eco,ogica, ba!ance of nature is  difficuH to eva,uate, since
    data  are avai^e which  suggest that adverse effects «y occur
                                 6-231

-------
6.9  SUMWRY AND CONCLUSIONS
6.9.1  Absorption, Distribution, and Excretion
     In the environ«nt, exposure to POM occurs by direct inhalation of
ponuted air and  -obacco smoke, by ingestion of contaminated food and water,
and by demal  contact with  soot, tars, and oils.  Regardless of the route of
exposure,  it can  be demonstrated in  laboratory animals that POM are readily
absorbed across all epithelia which  are  in contact with  the external environ-
 ment.   The fact that POM are generally highly lipid-soluble neutral molecules
 greatly facilitates their passage through the predominantly  lipid-like cell
 Kenbranes of animals, including man.
       Under environmental conditions, POM reach the lungs by adsorption on
 carrier particles.   Moreover,  the regional deposition and retention of
  inhaled POM in the respiratory tract will be primarily  determined by the
  physical  size of carrier particles  and, to. a lesser  extent, by their composi-
  tion.  Once deposited in the lungs, two processes begin to act on  the
  particulate material.  First, depending on their si«,  particles are  cleared
  from the respiratory tract by upward flow in the mucociliary tree or by
  phagocytic action of pulmonary macrophages.  Second, adsorbed POM is eluted
  from the carrier particles  and left free to react with the respiratory
  tissues  or traverse the  epithelium to  reach the systemic circulation.  A
   balance apparently exists between  the.rate  of particle clearance and the
   degree of POM elution which ultimately establishes  the degree of  toxicant
   exposure via the lungs.   However,  clearance of particles by mucus and
   ciliary action does not necessarily remove them from the  body,  since most
   trials cleared  from the lungs are subsequently swallowed, thus allowing
   for absorption  via the gastrointestinal tract.
                                       6-232

-------
        Upon reaching the bloodstream, POM are rapidly distributed to most
   internal  body organs.   Under experiments  conditions with Moratory animals
   the route of exposure  has  mtl. apparent  influence on  the tissue  localiza-  '
   tion of POM.   Extensive  Ioc.11z.t1.,,  in  the  fat and fatty tissues  (e  ,
   breast, is observed, and suggests that these tissues may  act as a  chemical
   trap, creating a situation for sustained release of the unchanged  substance
   In pregnant rats, it is apparent that BaP and DMBA, but probably not MCA
  are capable of transplacental passage and localization in the fetus
       Excretion of POM is rapid and occurs mainly via the feces;  elimination
  m the bile may account for a significant percentage of  administered doses
  The influence of route  of administration  on  patterns of  POM excretion  is
  not entirely  clear.   However,  elimination of  W  from „„,„„ ^^
  dung, liver,  Mdney, urine,  feces)  of the hamster following intratracheal
  insolation appeared to be biphasic.  The slow phase of elimination from
 the lung (involving less than l percent of the treatment dose) seemed
 dependent on the administration of 3H-BaP together with an adsorbent   it
 1. not known to what extent  bioexchange of the tritium label may  have
 accounted  for retention  of radioactivity.  Although  some  investigators  find
 that retention of BaP  in the lungs of experimental anima!s  is dependent
 upon the co-administration of  particulates, others maintain that carrier
 particles are more important in determining localization in  the respiratory
 tract.
 6.9.2  Metabolism and Metabolic Activation
     The relative lack of chemical  reactivity for tumorigenic POM  has in
the past been puzzling in light of their biological  effects. . More recently
however, it has become recognized that these  molecules are enzymatically
                                  6-233

-------
activated by oxidative mechanisms to form reactive  electrophiles.  These
activated metabolites are capable of covalent interaction with cellular
constituents (RNA, DNA, proteins), and it is suggested that one  of these
metabolites is the true ultimate carcinogenic form.
     Metabolic reactions on the POM skeleton may take place at  nearly any
position, although chemical theory predicts that certain locations will  be
more reactive.  The  earliest theories designated a K-region having  particu-
lar relevance to  the biological activity of the molecule subsequent to
oxidative attack.  More  recently, a "bay region" hypothesis was formulated
which  takes into  account the ease of  benzylic carbonium ion formation such
as would be formed from  the epoxides  of  diol-epoxides on tetrahydrobenzo
 rings  in the "angular" region  of a  POM.  These  diol  epoxides are postulated
 as ultimate carcinogenic metabolites  of  POM,  a  contention  which is supported
 by circumstantial experimental data showing high mutagenicity/carcino-
 genicity for such structures  and specific  alkylation products of genetic
 material identical to that obtained with the parent compound under metabolic
 activation conditions.  Examples of a K-region  and a bay  region on the  BaP
 molecule can be depicted as follows:
                   Bay region *--v\     .
                                  rlo-
                                               K-region
       POM is  metabolized by the microsomal mixed-function oxidase system,
  often designated aryl  hydrocarbon  hydroxylase.  This enzyme system is
  readily inducible and  is  found in  most mammalian tissues, although predominantly
                                     6-234

-------
   m the liver.   !„ conjunction with various P-450 type cytochro.es,  this enzyme
   complex is involved ,-n detoxification „, ^ ^^  ^ ^
   the formation  of reactive  epoxide  metabolites fading to  carcinogenesis   A
   second  microsoma,  enzyme,  epoxide  hydrase,  converts epoxide metaboMtes of POM
   to  vicina, g,ycols, a  process which may a,so  have critical instance to carcin
   ogenesis.
       Because of the importance of metabolic activation for the expression
  of carcinogenic effects by POM,  the chemical fate of many representative
  compounds in mammalian  cells has been  extensively explored.   By far  the
  ».t widely studied of  the  POM has  been BaP, one  of  the principal carcino-
  genic products  from the combustion  of  organic  material.  The metabolites of
  BaP  (and a!l  POM)  can be divided  into  a water-soluble  and  an organic solvent-
  soluble  fraction.   Components of the latter  fraction are primary ring-
  hydroxyuted products, quinones, and ,*„. epoxide intermediates   For BaP
 there are at !east three dihydrodiols,  three quinones,  and two phenols
 which can be detected as positiona!  isomers.   The  K-region (4,5-) and non-
 K-region (7,8-;  9,10-) epoxides are  precursors of  the corresponding diols
 which are formed by the  action of the epoxide nydrase enzyme.   A subsequent
 oxidative attack by the  ary,  hydrocarbon hydroxylase  may convert the  diols
 to dio, epoxides,  one of which  (7,8-diol-9,10-epoxide)  is an uHimate carcino-
 genie  form of  BaP.
      In the water-soluble fraction containing BaP metabolites are mainly
conjugates of hydroxylated products with glutathione,  glucuronic acid, and
sulfate.  This group of metabolites is tenatively regarded  to be composed
of non-toxic excretion products.
                                  6-235

-------
     The general scheme of metabolism for unsubstituted POM closely parallels
that for BaP, although several other major environmental PAH and aza-arenes
have not been studied.  It is also evident that K-region derivatives of POM
may be preferred targets for conjugation and excretion, whereas non-K-region
epoxides undergo further reductions  and oxidative attack to form theologically
important molecules.   For  POM bearing alkyl substituents (e.g., DMBA, MCA), the
primary metabolites  formed are  hydroxymethyl derivatives.   Nevertheless,  epoxi-
dation reactions  at  K-region and non-K-region  aromatic double bonds occur which
are catalyzed by aryl hydrocarbon hydroxylase.   Removal of activated intermediatej
 occurs by conjugation with glutathione or glucuronic acid, or by further metabo-
 lism to tetrahydrotetrols.
      Alternative explanations for the generation of reactive POM metabolites
 exist, although supporting evidence for their toxicologic significance is not
 as strong  as for the diol  epoxide  (bay region) theory.  One  argument supports
 the  role of reactive radical cation intermediates  of  POM  as  the critical  meta-
 bolites  capable  for interaction with cellular constituents.   Others have indicate
 that a 6-oxy-BaP free radical  or a hydroxymethyl  derivative  of POM may be bio-
  logically important reactive metabolites.
       The exact intracellular event by which a reactive POM metabolite initiates
  a toxic or carcinogenic response is not known.   However,  it is a widely held
  view that covalent  binding of metabolites to DMA forms the molecular basis of
  the carcinogenic and/or mutagenic  consequence of exposure to certain POM.  In
  this regard it  has  been  shown  that carcinogenic POM  in the  presence of  rat liverj
  microsomes  become  bound  to  DNA and synthetic polynucleotides, and moreover that
  the extent of binding is correlated with  the known carcinogenic  potency of the
  compound  and  levels of microsomal enzyme  activity.   Covalent binding  of reactivel
                                      6-236

-------
  POM metabolites also takes place with RNA and other cellular proteins, and these
  processes cannot be excluded as potentially inportant Gators of toxic response
  Various epoxide derivatives of POM have been tested for their ability to bind to
  nucleic acids,  and it appears that guanosine residues may be the preferred
  targets for reactive arene oxides.   Reactivity with nucleic acids and synthetic
  polynucleotides-Cpoly G)  varied with the position of the epoxide moiety;  K-
  region  and  non-K-region epoxides were capable  of  extensive covalent  binding,   it
  was further shown  that, for BaP, the binding to RNA  occurring after  in vivo
  exposures may be due  to the reaction  of  BaP  7,8-diol-9,10-epoxides with the 2-
  amino group  of guanosine.   A DMBA dlol epoxide was also  shown to be involved in
  DNA binding  in cultured mouse embryo  cells.  These studies have helped to strength-
  en the concept that:  (!)  reaction with nucleic acids may be the molecular basis
 by Which a POM exerts its  oncologic effects,  and (2) non-K-region diol epoxide
 metabolites  may be the ultimate reactive chemical  forms of POM.
 6.9.3   Toxicology
     Not a great  deal  of attention  has been paid to the noncarcinogenic effects
 of  exposure  to POM.   Nevertheless,  it is  known  that tissues of the  rapidly
 proliferating type  (e.g.,  intestinal  epithelium, bone marrow, lymphoid organs,
 testis)  seem to be  preferred targets  for  ROM-induced  cytotoxicity.  This action
 is probably  due to  a specific attack  on DNA of cells  in the S phase of the
mitotic cycle.
     Acute and chronic exposure to various carcinogenic POM has resulted in  '
selective destruction of hematopoietic and lymphoid elements, ovotoxicity and
anti-spermatogenic effects, adrenal  necrosis,  and changes in the intestinal and
respiratory epithelia.   For the most part, however, tissue damage occurs  at dose
levels  that would also be expected to induce carcinomas, and thus the  threat of
                                   6-237

-------
malignancy predominates in evaluating POM toxicity.   For POM which are not
carcinogenic, very little seems to be known concerning their involvement in
toxic responses.
     lwino.uppnui.1on by exposure to POM  has been reported, although it is not
clear whether this effect may be  involved  in carcinogenesis.  However,  it appears
that the degree of immunosuppression (cell-mediated  and humoral)  by  POM is
 correlated with carcinogenic potency.   On the  other  hand,  a dissociation between
 carcinogenic and immunosuppresive effects can  be shown, whereby doses which are
 insufficient to affect immunity can still induce tumors.
      One of the most toxicologically significant processes, involved in the
 response to  POM absorption  is the interaction with  drug-metabolizing enzyme
 systems.   The  induction of  this  enzyme activity in  various body  tissues by
 substrate  and  non-substrate xenobiotics may be  a critical  determinant  in the
 generation of  reactive  POM metabolites at the target site for  tumor induction.
  Recent emphasis has been placed on determining  the  drug-metabolizing capacity of
  humans as a measure of their potential to form activated POM metabolites.
  Although  it has not thus far been possible to definitely correlate enzyme activ-
  ity in humans with susceptibility to  carcinogenesis,  it is known that wide
  variations  occur in human  carcinogen-metabolizing  capacity.  Moreover, tissue-
  specific  enzyme inducibility may affect  the response  of  different  organs to
  carcinogen action.   Nevertheless,  the obligatory coupling of  metabolic  activ-
  ation with POM-induced neoplasia in animals  indicates that  the  modulation  of
   drug-metabolizing enzymes in humans plays a  central role in carcinogenesis.
   6.9.4  Mutagenesis
        The demonstration of mutagenic  effect in bacterial and mammalian cells by
   exposure to  POM is generally equated with the capability to induce tumor formatil
                                       6-238

-------
  This assumption is based on the participation of a common electrophilic metabo-
  lite in producing the carcinogenic/mutagenic event, and the common target site
  in the cell (i.e., DNA or other components of the genome) for ^ ^^ ^ ^
  produced.
       In recent years,  considerable research effort has been directed at deter-
  mining the mutagenicity of various POM derivatives as  a means  of identifying
  structural  features associated  with the biological  effect  produced.   Working
  with  bacterial  mutants  which can be reverted  to histidine  independence  by a
  chemically-induced mutation, epoxides  of carcinogenic  POM were shown  to possess
  significant mutagenicity.   In particular, it was found that a non-K-region 7,8-
  diol-9,10-epoxide of BaP possesses the highest mutagenic activity of all its
 possible oxidative metabolites.V
      Further work with cultured mammalian cells established that carcinogenic
 POM can produce forward mutations when a drug-metabolizing enzyme system is
 available.   Once again,  a 7,8-diol-9,10-epoxide of BaP  displayed the highest
 mutagenicity among  its  various metabolites.   This  effect was seen in the absence
 of  drug-metabolizing enzymes, strongly suggesting  that  the  diol  epoxide  deriva-
 tive is  an  ultimate mutagenic agent.   Additional Investigation* on the inter-
 action of BaP derivatives with constituents  of the  same mammalian cells  in which
 mutations are induced revealed that  a  BaP 7,8-diol-9,10-epoxide was  involved in
 binding to DNA.  The link between carcinogenicity and mutagenicity was strength-
 ened by the demonstration that both neoplastic transformation and mutagenesis
 could be induced by BaP or BaP 7,8-dihydrodiol (precursor of the 7,8-diol-9,10-
epoxide) in the same normal  diploid hamster embryo cells.
     Numerous attempts  have  been  made to correlate exposure to POM with the
induction of chromosomal  aberrations.  Although variations in chromosome  number
                                   6-239

-------
and structure accompany POM-induced tumors in rodents,  it is not clear whether
these changes are consistently observable.  No evidence in the published liter-
ature has been found to indicate that POM may produce somatic mutations in the
absence of neoplastic transformation.
     Mutations in germinal tissues  induced by POM have been more easily demonstratj
 in Drosophila than  in mammals.  The male  dominant lethal  assay  in mice  has
 produced conflicting results,  although  it is known  that  BaP and MCA  can induce
 sperm abnormalities.
 6.9.5  Carcinogenesis
      Polycyclic aromatic hydrocarbons were the first compounds ever shown to be
 associated with carcinogenesis.  To this day, carcinogenic POM are still distin-
 guished  by severa! unique features:  (1) several of the  POM are among the potent
 carcinogens  known-to exist, producing  tumors by single  exposures to microgram
 quantities;  (2)  they act both at the site of.application and at organs distant
 to the site of absorption;  and (3) their effects have been demonstrated in
  nearly every tissue and species tested,  regardless of the route of administrati on |
  Mong the more co-on POM at least one,  BaP, is ubiquitous in the environment
  and produces tumors in animals which resemble human carcinomas.  The demonstrat,o|
  that organic extracts of particulate air pollutants are carcinogenic to animals
   has raised  concern over the  involvement of POM in human cancer formation.
       Ora! administration of  POM to rodents can result  in  tumors of the forestomac
   mammary gland,  ovary,  lung,  liver, and  lymphoid and  hematopoietic tissues.
   Exposure to POM by inhalation or intratracheal  instillation can  also  be an
   effective means of producing tumors  of  the respiratory tract  using very small
   doses of chemical.  However, for both oral and intratracheal  routes of admin,-
   stration, BaP is less effective than other POM (..„..  DMBA, MCA, DBcgC) in
                                       6-240

-------
   producing  carcinomas.  On the  other  hand,  BaP  has a  remarkable  potency  for  the
   induction  of skin tutors in mice that cannot be matched by any  other environ-
   mental POM.  Thus, caution must be exercised in considering the carcinogenicity
   of POM as a class, or in using BaP as a representative example in evaluating
   carcinogenic risk of POM.
       The induction of skin cancer by POM is regarded to be a two-stage process
  involving an irreversible  initiation step and requiring the subsequent presence
  Of a promoting  agent  for tumors to  develop.  Certain  POM may act only as initiating
  agents  (e.g., DBacA,  chrysene,  benz[a]anthracene), or may  supply both initiating
  and  promoting stimuli  (e.g., BaP, DMBA,  MCA, DBahA, DBcgC, DBahP, DBaiP)  The
  administration  of a single subcarcinogenic  dose of a  POM (e.g.,  100 nmoles of
  BaP)  followed by repeated application of a  noncarcinogenic promoting agent
  (e.g., croton oil) can cause the appearance of  numerous skin tumors in mice.
 The two-stage mechanism of carcinogenesis may also apply to nonepidermal tissues.
      For the induction of respiratory tumors by POM,  several  species and modes
 of carcinogen administration  are employed.   In studies conducted with  the Syrian
 golden hamster,  intratracheal  instillation of POM established that dose-related
 increases  in tumor yield are clearly evident,  and  the  co-administration  of
 carrier  particles such as Fe^  can  markedly increase  tumor incidence.
      Several  u,  vitro  procedures  hold  promise as useful  screening tools for the
 detection  of environmental carcinogens, including POM.   Morphological transform-
 ation  of cultured mammalian cells has proven to  be a reliable indicator of
 carcinogenicity  by a chemical 1» vivo.  Examination of DNA repair synthesis in
 cultured human and animal cells exposed to a carcinogen  is also predictive of in
 vivo carcinogenicity for certain compounds.  When combined with the results of
microbial mutagenicity tests,  cell transformation assays are capable of accurately
distinguishing nearly all  known carcinogens from noncarcinogens.
                                   6-241

-------
     An analysis'of dose-response relationships for POM-induced tumors  in
animals raises several important points having relevance to environmental  risk
assessment.  A fundamental relationship which must be considered is the devia-
tion from linearity in dose-response curves, especially at low doses, and
whether this  indicates the existence of a threshold level.  It can be argued
that individual variability  in  thresholds for  tumor induction  resulting from
saturation of detoxification mechanisms may account for a  convex curvature in
the dose-response curve  in  the  low dose range.   Once  the dose  exceeds  threshold
 levels, tumor yield should  remain a linear  function of dose, with  the  slope  of
 the dose-response curve being indicative  of the animal's sensitivity to the
 carcinogen.  Thus, the extrapolation of the straight portion of the dose-response
 curve for groups of  animals  is often regarded to provide a conservatively low
 estimate  of  the  average  threshold  for tumor induction,  assuming the most likely
 case  of decreasing sensitivity with an increasing threshold.  However, the use
 of an average threshold as  a parameter of  safety wouTd not indicate the  propor-
 tion  of individuals  still  at risk of  tumorigenesis  at extrapolated low dose
  levels.   Furthermore,  the  existence of a threshold for POM-induced carcino-
  genesis has not been documented in either  animals or man, and, in fact,  it  is
  more likely that in diverse populations  the effect of POM will be a continuous
  function of dose.   In support of  this contention are  data which indicate that,
  in the two-stage model, tumor initiation  with  BaP is  consistent with a  linear
  non-threshold pattern.  Overt tumor  induction,  on the other  hand, follows a dos<
  response relationship  consistent with a multi-hit promotion  process.  It is
  conceivable'that in human populations,  the multi-hit component of carc1nog.n..1.
   may be supplied by environmental stimuli  not necessarily linked  or related to
   POM exposure.
                                      6-242

-------
       The well-documented existence of cocarcinogenic and anticarcinogenic
  agents dictates that only a muHifactorial  analysis can provide a true assess-
  ment of carcinogenic risk to humans for a particular POM.   Although noncarcin-
  ogenic POM were reported to antagonize the  effects  of carcinogenic POM's  in
  animals,  others have shown that  they have little  influence  on  tumor incidence or
  Yield.  On  the  other hand,  several  noncarcinogenic  POM found in  cigarette smoke
  (pyrene,  fluoranthene, BeP)  have potent  cocarcinogenic acitivity.   A significant
  decrease  in carcinogenicity  can be  achieved with antioxidant food  additives,  .
  certain vitamins, and other  naturally occurring components in the  diet in the
  experimental setting.  The molecular basis of anticarcinogenesis cannot be fully
 explained, but may be due, at least in part, to an effect on the production of
 activated POM metabolites.
 6.9.6  Reproduction and Teratology                              .         ;
      Little information is presently available to  indicate whether POM  present a
 significant hazard to reproductive  success.   Furthermore,  effects on the  fetus
 which may be due to maternal  toxicity or experimental  conditions  (e.g.,  injec-
 tion  vehicle, stress) have not been adequately dissociated form true embryo-
 toxicity or  teratogenesis.   In cases where teratogenic effects are  clearly
 evident (e.g., with  DMBA),  the required  doses  are far  in excess of  realistic
 environmental exposures.
 6.9.7   Human Studies
     The presence of  POM  in the air, or as components of soot, tars, and oils,
 have long been associated with an excess incidence of cancer in human populations.
However, it has never been possible to study a population having exposure to
POM in the absence of other potential carcinogens,  cocarcinogens, tumor  initi-
ators, or tumor promoters.
                                   6-243

-------
     Convincing evidence indicates an excess in lung cancer mortality among
workers exposed to large amounts of coal gas, tars, and coke oven emissions.   In
such cases, cancer mortality can be correlated with both the type and duration
of exposure.  Although occupational exposure to POM-containing substances is far
greater than would occur  in most communities,  our  understanding of chemical
carcinogenesis would lead to the  conclusion that the number of cancers produced
is  directly proportional  to the dose received.  One must  assume,  therefore, that
the smaller amounts  of'POM in  ambient air contribute  in some  degree  to the
 observed incidence of lung cancer in most populations.   It should be recognized,
 however, that the influence of cigarette smoking may have an overriding  impact
 on the evaluation of the carcinogenic threat of POM in occupational  or environ-
 mental settings.
       Nevertheless,  an approximately  two-fold  excess of lung cancer occurs In
  urban settings  as compared to rural  environments, which  generally cannot be
  accounted for  by differences  in  cigarette smoking, age,  or sex.  Several invest-
  igators have shown  that the  incidence of lung cancer  is  highest in  cities where
  POM pollutants are the most concentrated.  Moreover,  it  is clear that among
  groups which have migrated from one country to another,  the influence of pollu-
  tion exposure in their former residence  is still expressed as a higher  rate of
  ,ung cancer mortality while  in their adopted country.  This observation cannot
  be explained on the basis of differences in  cigarette smoking habits.
       Sampling  studies  from different  communities have attempted to  separate the
  various factors  which  may be contributing to lung cancer excesses  in urban
   settings.  These studies have not provided the  kind  of  definitive  results which
   have been achieved with homogeneous worker populations  exposed to  pollutants
   which are well-defined both quantitatively and qualitatively.   On  the  one  hand,
                                      6-244

-------

      been conceded that POM  in co.unit  ai
                                           air does
             -pec     t                                        *"   "
                                          .. ,
unde.sti.tion of Hsks derived from data concerning BaP.
6.9.8  Ecological Effects
             POH are found
             these agents
                        ,  f,,h>
                          data
                                                        may thus
                                             ^       ^
                                                                  »
                                                            cc
•"               inS that POH are  rapidly ^^ ^               '
                                                          , hiaher
                                 6-245

-------
6.10  REFERENCES
2
 3.
 4.
 5.
8.

     1962.
HIGWIIfc*!! !»*••• » •  ..- — — -
Acta., 225(1):96-107, 1971.

                                 3CI.  Distribution, retention, and
                                  after administration to mice and rats.

j"NatT!"cancer"lnst!, 23:541-555, 1959.





 Chem. Path.  Pharmacol.,  13(2):259-271,  1976.
       . Toxicol. Environ. Health, 1_
       5J.(6): 1961 -1964,  1973.
       Environ.  Health,  3:1003-1010,  1977.
 Bock,  F. G., and T.  L. Dao.   Factors affecting the
 level  in rat mammary glands.   Cancer Res., 2V.1024 1029,
                                                                      hydrocarbon

                   F  G  Bock  and S. Crouch.  Level of 3-methylcholanthrene in
                                                       1-1" - —
        1976.
        Pharmacol.,  16:1821-1831,  1967.

        ..-.HI    P  stPnhSck  D  B   Clayson,  and L.  Wall cave.   Intratracheal
         nX llati n'studferwithV^dibenzolc.gicarbazole in the Syrian  hamster.
        J.  Natl.  Cancer Inst., 5Z(1):119-123,  1976.
                                           6-246

-------
             ,„ rats.  Byull.
   14.
  15.
  16.
  17.
                                                            •££•
 Science, J99:3o5-307.
      Of  injected -
                                  ns stu"'ed by fluorescence  spectroscopy.
                                                "'"  "' •"•""««. of ,,2-
  18.
                     after
 benzpyrene.   Blochem. J. , wl
                                                              3,4-benzpyrene
                                             mo
                              a
      stratlon.   Res. Co«n. Cbem.
      pyrene metabolites ,„
                                                                    of benzofa]-
 19.  Heidelberger,  C.,  and S. M. Weiss.  The distribution  f   „••
      mice following administration of 3 4-benz   ributlol?4of radioactivity  in

                                                *     v> JO j | zyZj I B
 20.
 21.
 22.
23.
24.
               eeaosn  t
Cancer Res. ,  30- zS" 19?ob
                                               H" C°nney-  Effect "^ enzyme
                                           dlstri'bu"<>" of benzo[a]pyrene
                                      6-247

-------
25.
26.
27.
     Pelfrene, A.  F.   Experimental evaluation of the clearance of 3,4-benzo[a]-
     pyrene in association with talc from hamster lungs.  Am. Ind. Hyg. Assoc. J. ,
     37:706-710.

     Feron  V. J., D. De Jong, and M. A. H. Rijk.  Clearance of benzo[a]Pyrene
     from the respiratory tract of hamsters following its intratracheal
     instillation with or without ferric oxide.  Zbl. Bakt. Hyg., i. MDT;.
     Orig. B, J63:t41-447, 1976.

     Kennedy, A. R., and J.B.  Little.   Localization  of  polycyclic hydrocarbon
     carcinogens in the lung  following  intratracheal Instillation in gelatin
     solution.  Cancer Res.,  35:1563-1657,  1975.

28   Scala,  R.  A.  Toxicology of  particulate  polycyclic organic material.
     J. Occup.  Med., 17(12):784-788,  1975.

29   Pylev  L.  N., F. J.  C.  Roe,  and G.  P.  Warwick.   Elimination  of radioactivity
     after-intratfacheal  instillation of tritiated 3,4-benzopyrene  in  hamsters.
     Brit. J.  Cancer, 23:103-115, 1969.

30.  Pullman,' A.  and B.  Pullman.   Electronic structure and carcinogenic activity
     of aromatic  molecules - new developments.  Adv. Cancer Res., 3:117 169,  1955.

31    Lehr  R  E   H.  Yagi, D. R.  Thakker, W.  Levin, A.  W. Wood,  A.H  Conney,  and
      DM  Jerina.   The bay region theory of polycyclic aromatic hydrocarbon-
      induced carcinogenicity.  In:  Carcinogenesis, Vol  3:  Polynuclear
      aromatic hydrocarbons.  R. I. Freudenthal and P.W. Jones (eds.), p. 231
      241, Raven Press,  New York, 1978.

 32.  Conney, A. H.  Pharmacological implications of microsomal enzyme induction.
.   "  Pharmacol. Rev., 19:317, 1967.
 33.
  34.
  35.
  36.
      Marquardt, H.  Microsomal metabolism of chenncal carcinogens  in  animals
      and man.  In:  Screening Tests  in Chemical  Carcinogenesis.  R. Montesano
      and L  Tomatis (eds.),  Intl. Agency for Research on  Cancer, Lyon, .France,
      IARC Publication No.  12, p.  309-328, 1976.

      Sims,  P.  The metabolism of polycyclic hydrocarbons  to dihydrodiols and
      diol epoxides by human  and  animal tissues.   In:  Screening Tests in
      Chemical  Carcinogenesis.  R. Montesano H.  Bartsch ,  and L.  Tomatis (eds.).
      Lyon,  France; IARC.   IARC Publication  No.  12,  p. 211-224, 1976.
       Gelboin,  H.  V. ,  N.  Kinoshita, and F.  J..W1ebel.   Microsomal
       studies" on the mechanism of induction .and their role jn polyjycl c hydro
       carbon action.   In:   Collection of Papers Presented at the Annual
       Symposium on Fundamental Cancer Research, Series 24.  p. 214 240, 19/^.

       Ra«?t  R  C   Jr   T  Okuda, E. Poltkin, R. Tarone, H. J. Rapp, and H. V.
       Gelboin:   Development of an assay for aryl hydrocarbon  [benzo(a)Ryrene]
       hydroxylase in human peripheral blood monocytes.  Cancer Res., 36.19b/
       1974, 1976.
                                           6-248

-------
                                  n
            .  Interactions, !7:9-T6,^§77



                          hy'dro^rbo
                Res.  Commun.  Chem.
   37.
   38.
  39.
       pyrene) hydroxylase.'

       Dermatol., 64(3) : 184-1 89




                  '
 42.





 43.




 44.





45.
                                                                ^ocarbon hydroxylase
                                                       mammary cell lines.  Chem.-
                                                                            and
                                                                              bone
                                                                       i.
                                               oxygenase-in-mouse"skinr"ibe^ear
                            oapyrene
       cortex.   I.   Characteristics of

       Biophys.,
                                   Induction of microsomal  aryl
                                   and cytochrome P-450.,  in rat

                                   " system.  Arch. BiSchem.




r^ntCest?n1lGmucoS^^^^
" + v>,->-»-i.«,j   j.        anu aurenai  or  unt.rpatoH ar.^4 o	,_j.i...-i
  treated rats.  J.
                .   Biochem. and B
                                         ^




                                                             eP-ide is an  inter-

                                neVhu^        ?• '  and H'  V'  Gelbo^-  In vitro
      Cancer Res. ,. 35:3651-3655, 1975.             nncrospmes and lymphocytesT -
46.  Pezzuto, J. M. , C. S.

     Metabolism of benzo(>

                   liver nuclei
47.
                                  C
                                                                      '  Gelb<""-

                                        ox
    detoxification  of benzo[a]pyrene   ?n
         Metabolisn, Concepts.   oT Jel"na
                                                              actl'»ation and
                                                                               Role
                                        6-249

-------
 «   ^ -,,- •   i 'v   P  r  rrnv  P   P   Roller  and H. V. Gelboin.   High-pressure
49'   ? Sfch^.atographi/an^y'sis'of'b^zoiajpyren, ..tabolls™ and covjlant
     binding and the mechanism of action  " J^"^"01" a"d 1>Z ePOXy 3>3>3
     trichloropropane.  Cancer Res., 34:3474-3480,  19/4.
50   Sims  P    Qualitative and quantitative  studies on the metabolism of a
     series of aromatic hydrocarbons by rat-liver preparations.  Biochem.
     Pharmacol., 19:795-818, 1970.
51.  Boyland, E., and P. Sims.  The carcinogenic; activities in mice of compounds
     related to  benz[a]anthracene.   Intl. J.  Cancer, 2.500 bU4,  iy&/.
K»  c+nnrfnn  TAW   Bornstein,  and E. Bresnick.  The  metabolism of 3-methyl-
52'  Solanthrene by  rat liver microsomes - A reinvestigation.   Biochem.  and
     Biophys.  Res.  Commun., 79(2):461-469, 1977.
 "•  ^^Zil^^^^X^^ L"a ffi
      tiSZtf^^^i^^
      and P.  W.  Jones (eds.).   Raven Press, New York, p. 253 264, 19/8.
 KA   v=,nn  *  K   P  P  Roller  and H. V. Gelboin.  Benzo'[a]pyrene metabolism:
 54.  Yang, S.  K., JjJ'J^'  af epoxides,  phenols, dihydrodiols  and the
                   i-epoxides.   In:   Carcinogenesis^Vol. ^ .^^J^l^p^uSnall
       1978.
       Npmoto   N    and S  Takayama.  Modification of benzo(a)pyrene metabolismVitt
       m?crosomes'by  addition of uridine 5'-diphosphoglucuronic acid.   Cancer Res.
       37:4125-4129,  1977.
             J  R  Z Ben-Zvi, J. Van Anda, P. M. Dansette, and D. M.  Jerina.
56.
57.
       Chemistry, Metabolism,  and Carcinogenesis, Vpl
       P? W. Jones (eds.),  pp.  63-75,  Raven  Press, New York, 1976.
  58   Jerina, D. M.,  and J.  W.  Daly.   Arene oxides:  A new aspect of drug metabol
       Science, 185(4151):573-582, 1974.
  59   Wiebel- F  J   J  C. Leutz, and H.  V. Gelboin.  Aryl hydrocarbon (benzo[a]-|
    '  pyrene) hydroxylase:  Inducible in extrahepatic tissues jyJJ^e strains   |
       not  inducible  in liver.  Arch.  Biochem.  and  Biophys., 154:292 294,
                                          6-250

-------
   60.
  61.
  62.
                                     abo
   and diet.   Fed.  Proc., 36(5): 1547-1652} "977
                                                               .*-  >•  «»»«.  and
                                                            env1r°™enta]  chemicaTs
                           rern,
   and  epoxide  hydrase.   InP  cc nogene is
   Hydrocarbons,  R.  I.  Freudenthal 2nd  P   w
   Press, New York,  1978.
                                                                               Enzy»o
                                                                        system  ^
                                                       ^^ P°'y"cul^r- Aromatic
                                                       (eds-). P- 243-252, Raven
                                                                          and P. „.
                                              L           '   '          .   •   • .
          Biomedical  Press.  1976."
                                       p.
                                                                      Elsevier/North-

 66.
67.
68.
 Carcl
                                ^^
                                     W
                                                      1.  D.  R  Thaler,  H. Akagl
                                                         differences 'in the
      indicate that

    obenzo[a]pyrene
    , 38:354-358, 1978b.
                                                                           . H.
                                                        of benzo[a3pyrene

                                                        ^
                                                      newborn mice.   Cancer

skin.
            genic^^                                           and A.  H.  Conney.

            Proc.  Natl.  Acad.  Scf SsA, ^.243-2^7,       ™**5 O"
                                       6-251

-------
70.
71.
    Gelboin, H. V., J. K. Selkirk, S.  K.  Yank,  F.  J. Wiebel, and N. Nemoto

    SfeMJ^^S^
    In  Glutathione:  Metabolism and Function,  I.  M. Anas  and W.  B. Jakoby,
          ,  Raven Press, New York, 1976.. pp.  339-356.

                               ^
     binding to DNA.   Proc.  Natl. Acad. Sci., USA, 7:
72.
     ofthe diasereomeric benzo[a]pyrene
     Biophys. Res. Commun. , 77 (4): 1389- 1396,  1977.
                                                                  Biochem.
73'  Hor^' W^^\^.' Tu^^en^ilyTth/o^c^
     enaSers of the diastereomeric benzo[a]pyrene 7,8-d1ol-9VlO-epoxides in
     newSorn m?ce:  Exceptional activity of (^-^^a-dihydroxy-g^lOa-epoxy
     7,8,9,10-tetrahydrobenzo[a]pyrene.  PNAS,  75:5358-b^bi,  iy/o.

74.  Meehan, T., and K. Straub   Double-stranded DNA stereoselectively binds
     benzo(a)pyrene diol epoxides.  Nature, 277:410-412,  1979.

75  Smith   B   R   R  M. Philpot, and J. R. Bend.  Metabolism of benzo[a]pyrene
     by the isolated perfusedPrabbit lung.  Drug Metabolism and Dispositnon,
     6:425-431.

 76  VahSkanaas  K  K Nevasaari, 0.  Pelkonen, and N.  T. KSrki.  The disposition
 ?6-   of beSI)pyrene in  isolated perfused-rat lung.  Symp  Microsomes  Drug
      Oxidations,  3rd Berlin,  1976, Microsomes and Drug Oxndations, ed. Volker,
      Oil rich Pergamon  Press,  1977. ,

 77   vshSkanaas  K   K  Nevasaari, 0.  Pelkonen,  and N.  T. Karki.  The metabolism
 11'   of beSa)pyrene'in  isolated perfused lungs from variously-treated rats.
      Acta.  Pharmacol.  etToxicol., 41_:129-140,  1977.

 7ft   Simbera  N   and P  Uotila.   Stimulatory  effect of  cigarette smoke on the
 78'  mefabolism'and covalent binding of benzo(a)pyrene in the  trachea of the
      rat.  Int. J. Cancer, 22:28-31, 1978.

 -/a   M«r^o  R  P   and G  M  Cohen   Metabolism of  benzo(a)pyrene and its major
      metabolites to ethyl  acetate-soluble and  water-soluble  metabolites by
      cultured  rodent trachea.  Cancer Res. 38:3066-3075..           ^

 80.
       B     Trump   'Metabolsm of 'benzo[a]Pyrene and 7,1
       in cultured  human  bronchus and pancreatic duct.  Cancer Res.  38:3349  3355,
       1977.
                                          6-252

-------
  81.
  82.  Grafstrom,'R. ,  s.  J.
  83.
  84.
  85.
 86.
 87.
 88.
       1976,
  Fang, W.  F.,  and H.  W.

  rat colon microsomes.
                                                                             Benzo(a)
                                                                     mUtagens
              of 'cL
      Brief communication.  j.
                                     n
 Adv. Enzyme Regulation, J5:339-62, 1977
                                                              ites to DNA.
 pyrene metabolites to DNA      yl

 -spons.ve and nonresponsive miL.   eocne
                                                                      °f
          ,1977.
                                          ssues.   Drug Metabolism and Disposition,
89.
                             mes               ^               f  benzofaj-

     Vol. 1:  Polynuclear    matic' Hydrocarbons   S^-  f1"^   In:   C^inogenesis,
     Carcinogenesis,  R. I  FreudenthaV«nH p  u  (?hemistry>  Metabolism, and
     Press, New York, 1976!  Fr6Udenthal  and P-  W-  Jones  (eds.),  p.  35-43, Raven
90.
                                                  obr       Enzyme 1
        various strains of mice    In-   Carcfnnnlnf °  benzo[ a] pyrene produced
     Aromatic Hydrocarbons:   Chemistry          c esis» vol. 1:  Polynuclear
     Freudentnal  and P.  W.      •
91.
CJ-Ut*  Meta^s^andC^c^e UW
(eds.).   Raven Press,  New York, p. 9-21, 1976.
                                                  han aduit
                                                                  and P.  W.  Jones
                                       6-253

-------
92.
93.
                               ^
     26:671-674,  1977.
 94.
  98.
                                  W  Levin,  P.  Dansette, and D.  Jerina.  Use
      Lyon, France;
                             Publ.
                                        12,
      ^ V V/lljl*^*II***-J'~ 	

 "
      1978.

 "
      York, 1978.
97.   Booth,
     cyclic
     1974.
                                     JS WEU!"
                                              K. Selkirk,  P. W. Schaefer, and
     B  F  Trump  Metabolism or L HioenzoL^pyrene by cultured  human bronchus
     and cultured human pulmonary alveolar macrophages.  Lab.  Inv., 38^).
     224, 1978.

 99   Prough  R.  A.,  Z.  Spial, and S. W.  Jakobsson.  Metabolism of benzo[a]pyrene
     by human lung microsomal fractions.  Life Sciences, 21:1629-1636,  1977.

 100  DiDDle  A   Polynuclear aromatic carcinogens.  In:  Chemical Carcinogens.
     C  E  Searle (ed.).   ACS Monograph 173, American  Chemical Society,
     Washington, D.C.  Chapter 5,  1976.
            «                   «
      pyrene arene oxides and diol-epo
                                              Cancer
                                                                 4133,
       25:1221-1226, 1976.
                                         6-254

-------
105-
'°6'
Biochem. Pharmacol. ,
                                        -:    ««»., «"*.tM.n. conjugates
                                     l" 2,[|j|nthl-a«"« . "V rat-liver horaogenaies.

                                                                   and
                                         oc
                                       Biochem.
                                                           , 23:323-332, 1974


                                                                            '

   Sy.posl-um on Fun^eni
                                      6-255

-------
116.  Kellerman, G., C. R. Shaw, and M.  Luyten-Kellerman
     ducibility and bronchogemc carcinoma.   New England J. Mea. , ^os.

117  Kellerman, G., M. Luyten-Kellerman, and C.  R.  Shaw.   Genetic variation of
     aryl  hydrocarbon hydroxylase in human lymphocytes.  Am. J. Hum. Genet.,
     25:327-331,  1973b.

118  Atlas  S  A    E   S.  Vesell, and D. W. Nebert.   Genetic control of inter-
118' ?nd?vidual variations  in the inducibility of ^^Jg;^^"^11188
     in cultured  human lymphocytes.  Cancer Res., 36:4619  46du,  iy/b.

119  Gelboin, H.  V.   Cancer susceptibility and carcinogen  metabolism.  N.  Engl
     J. Med., 297(7):384-385,  1977.
     aryl'hydroSfbon hydroxylase inducibility in cultured human lymphocytes.
     Cancer Res., 37:1829-1837, 1977.

     Paiaen' B   H  L  Gurtoo, E. Ward,  J.  Minowada, L. Houten, R. Vincent,  N  B.
            ' and J. VaugSt.  Human aryl  hydrocarbon hydroxylase and cancer risk.
      ?Tes?r?n S^"cSi^UnrR:~BSti»i and L. Tonatls-Teds.). Lyon,
      France:  IARC.   IARC Publ. No. 12, 1976.  p.  319-336.



      f£S£-&'Ji&Ssa:&l£*££'
      Therap., 20(5):557-564,  1976.

      Kanitulnik  J   P  J.  Poppers,"  and A.  H. Conney.  Comparative metabolism of
      bePnzo[l]pyreneJand drugs'in human liver.  Cl1n. Pharmacol.  and Therap.,
      21(2):166-176, 1977b.

      Kanitulnik  J   P  J.  Poppers,  M. K.  Buening,  J. G. Fotner, and AH  Conneyi
      Xctivai?onJof'monooxygenaseS in human liver  by 7,8-benzoflavone.  Clin.
      Pharmacol. and Therap., 22(4):475-484, 1977c.
       158-165,  1977d.

  127  Thomas, S.,  and T.  J.  Slaga.  Mouse epidermal aryl  hydrocarbon hydroxylase.
       J. Invest.  Dermatol.,  66:108-111, 1976.
                                          6-256

-------
  128'
  129-
 2833  1977
SS
 135'
32l80-TI9l3   92
"'
     Biol. Interactions,
                                      - ulHch
                                    °f U?p-9l"curonosyltransf erase induction
                                    in m7Ce.  J. Biol.  Chen.-, 252(9):2827-
                                Carbon
                                                        administraion on
                                           UDp-9l"CuronsyHransf erase activity.
                                                              carcinogenesis.
                                               ^
                                                rat liver.  Cancer Res. ,
139"
     sn
     sKin.
                 .  Shapas, and R.  K. Boutwell.  The binding of 7 12-
      cta:-Bio-i:"irS^MoS?1gflw-n§9«?diS7rrap1fc"t1n9 om''"mouse
                                    6-257

-------
140



141



142




143
 '


144


 145
    "
 147
Bowden, G. T. , and R. K. Boutwell.  The replication of mouse skin epidermal
DNA to which is bound 7512-dimethylbenz[a]anthracene.  Chem. Biol. Inter-
actions, 9:15-23, 1974.

Nery, R.  Carcinogenic Mechanisms:  A critical review and a suggestion that
oncogenesis may be adaptive ontogenesis.  Chem. Biol. Interactions, 12:
145-169,  1976.

Albert, R. E. , F. J. Burns, and  B. Altshuler.  Temporal  aspects  of tumori-
genie  respons4 to individual  and mixed  carcinogens   Comprehensive
Progress  Report.  Institute of Environ. Med. ,  New York Univ. Med. Center,
New York, NY 10016, 1978.

Lutz   W.  K.,  A. .Viviani,  and  C.  Schlatter.   Nonlinear dose-response  relation-
ship  for  the binding of the carcinogen  benzo[a]Pyrene to rat  liver DNA
In vivo.  Cancer Res.,  38:575-578,  1978.

Gelboin  H.  V.   A microsome-dependent binding of benzo[a]pyrene to  DNA.
Cancer Res., 28:1272-1276, 1969.

Grover, P.  L. ,  and P.  Sims.   Enzyme- catalyzed reactions of polycyclic hydro-
 carbons with deoxyribonucleic acid and proteins in vitro.  Biochem.
J.,  VIO: 159-1 60, 1968.

 Kaufman  n  G   V  M  Genta,  C.  C.  Harris, J. M. Smith, M. B.  Sporn, and
 teufjan  k G. .^-^ ^abe,ed benzo[a]pyrene to DNA in hamster
 trachea! epithelial cells.   Cancer Res., 33:2837-2841,  1973.

 Harris, C.  C. , D. G. Kaufman, M. B. Sporn, H. Boren, F. Jackson, J.  M. Smith,
 J  Pau ley,  P. Dedick, and U.  Saffiotti.  Localization of benzo[a]pyrene- H
 and alterations in  nuclear chromatin caused by benzo[a]pyrene-f erne  oxide
 in the hamster  respiratory epithelium.   Cancer Res., 33:2842-2848, 1973.
 148. BUrki,  K., T. A.  Stoming,  and  E.  Bresnick.   Effects  on  an  epoxlde  hydrase
  149



  150.


  151


  152
    r,   .,   .   .         ,       .          .
  inhibitor  on in vitro  binding of  polycyclic hydrocarbons  to  DNA  and on
  skin  carcinogiHesis.   J.  Natl.  Cancer Inst. , 52(3) -.785-788,  1974,

  Yang, S.  K. , H. V.  Gelboin,  F.  B.  Trump,  H.  Autrup,  and C. C.  Harris.
  Metabolic activation of benzo[a]pyrene and binding to DNA in cultured
  human bronchus.  Cancer Res. , 37:1210-1215, 1977.

  Communication:   Binding of 6-oxygenzo[a]pyrene radical with  DNA and poly-
  nucleotides.  Gann, 68:125-126, 1977.

  Communication:   Binding of benzo[a]pyrene-semiquinone radicals with DNA
  .and polynucleotides.  Gann, 68:253-254, 1977.

  Lesko  S   W.  Caspary, R. Lorentzen, and P. Ts'o.  Enzymic formation of
  6-oxo-benzo [a] pyrene radical from carcinogenic benzo[ a] pyrene.  Bio-
  physical J., Tj5:324a, 1975.
                                      6-258

-------
                            V:
 '"'
                    ••

                                        =
'"



                           6-259

-------
    Ivanovic  V   N  E  Geracintov, H.  Yamasaki , and I.E.  Weinstein   DNA and
    RNA adducts formed in hamster 'embryo cell  cultures exposed  to benZo[a]-
    pyrene.  Biochem. , 17(9): 1597- 1603,  1978.

          H W S   MR  Osborne, F.A. Beland,  R.G. Harvey, and P. Brookes.
                               -7,8,9,10-tetrahydrobenzo[a]Pyrene  is an

                                                         •"•"-
"
     229-235,  1978.




     mouse liver and  skin cytoso.ls.   Cancer  Res., 35.816 824,  19/b.


     MOCP^  HI    R A Webster, G.D.  Martin,  and T.C. Spelsberg.   Binding of
      B. ,  122:477-507, 1937.
      5th Intl.  Congr-
San
                                                , 1972,  Vol. 2, pp. /a »»,
      24:558-569, 1964.


      »r-h^^^
      Cancer Res., 28:320-327, 1968.

  176. Ford, E., and C. Huggins.  Selective destruction in testis induced  by
      7,12-dimethylbenz[a]anthracene.  J. Exp.  Med. ,  118.27 40, 1963.

  177 Bock F G   Early effects  of  hydrocarbons on mammalian skin.  Progr. Exp.
     ' Tumor Res., 4:126-168, 1964.
                                         6-260

-------
 816-817, 1965.


 £S^^ c
 1:484-488,  1941
184-
the bronchial epithelium in
    anthracene ,„ Syan
         -n  ^he  t^n
                                                      » «•   ^t.  Cancer !nst. ,
                                                                   1c Pollutants:
                                                    ]antliracene.   Nature,  207:
                                              "
                                                   hydrocarbons.   Cancer  Res. ,
                                                              alterations of

                                      6-261

-------
192.  Stjernsward, J.   I-unosuppression by carcinogens.   Antibiotical Chemother. ,
     r5:213, 1969.

«•
     J. Mat!. Cancer Inst. , 36:1189S  1966.
194.  Baldwin, R.W.  I.unological aspects  of
     ssxss
                                   sr
                                                   carcinogenesis. J:
                                                            P.  i-75.
      320, 1973.

      Society,  Washington, D.C.   1976.
       158, 1965.
                    i   Falk  P  Mader,  and M. Thomas.   Aromatic hydrocarbons
                                                                      °f
  -•
       densate.   Cancer, 40:203-210, 1977.
                           SsJal" area"
                                         6-262

-------
 '  car^nes^'
   Honograph ,73.
                                                               **  1n
  2°7'
                                                                          „.
                               potency of carcinogenic compounds.  Am. J. Cancer,
                                                                     on the
                                              Induct1«n
                                                                       .  IIA.
                                   cnoa
  Hydrocarbons:   Chemistry.-MetaboMsm  and r,ir-
  R.I.  Freudenthal  (eds.)   New ?ork! Rave" Prass"°?9?I
                                                             .polV""<:lear  Aromatic

                                                                            a"d

                                               ""  ^'
 aromatic hydrocrbons   Cancer R
                                                                 '" P"'ycyc,ic
2'3-
                       2n65.an
            .  Natl.  Acad.  Scf USA,  73ci):143-l«
                                                                     A.H. Conney.

                                                                     m°USe
215'
     pyrena arene oxides and «.,
                 naabmobenzo
     epoxides  and 7,8-diol.   Cancer Le?ter!
pyrene on mouse sk,n.
                                                                .  O.N.  .erina,
                                                                of »enzo[a]-'
                                       6-263

-------
     1976.
219.  Kap1tu,n1K, J.. W. Uv In. H  Vagi   D.M.  JeHna  an< ^H-^

                                          «8.  ,976.
220. W1.1.CH. P.G.
              oyo.    nceRes. , 37:2608-26,,
     1977.





                                                  -,sr   —



      pyrene on mouse skin.  Cancer Res., 37:2721  2725,  is//.
      Cancer, 18:351-353, 1976.
       1511-1523, 1964.
                                                                    1ce  by
  227. Turusov, V.,  N.  Day, and L. Andrianov.   J.  Nat,. Cancer Inst.  47: ™5-
       m, is?,.

                                      "
                                  of maHgnants tumors of  ouse.sMn
   229.  Roe, F.J.  The
                                                                   —
                                          6-264

-------
 230.
 231.
               ,n  ?he

      benzanthracene  at differe
                                           - *
                                                                             ponss
     benz[a]anthracena.
            pig.  J. Nat). Cancer
236.
      in benzene.   Acta Un* int.'canclr^Tg:

     II ix     •  a      k-»*i i jf  uiilCicflLlo|
      K-regTon  epoxide on mouse  skin

     mechanism of tumorigenesis.  J.


239. Berry, D. L. , T.J.
                                                   !  1962   y'2"benzanthracene
                                                                      Progr.
                                         .  «.  SchHcner, G. CandeU,, and

                                                   "'hh-        Its
    ability of
               in tumors i
                                                             i"  oxides  to
                                                     In vitro,  j.  Natl.
     the carcinogenn-city of ^
     deuteratfon of the 1-
     of tumor initiation.
                                                  '
                                                 •
                                                   m°use=k1n "V  selective
                                      6-265

-------
?42  Klein  M   A comparison of the initiati ng -and promoting actions of 9, 10
     dime?hyl-l,2-benzanthracene and 1 ,2,5,6-dibenzanthracene in skin tumor-
     igenesis.  Cancer Res., 20:1179, 1960.
     Oct.  1965, p.  475-478.
 245  Wynder,  E.L..  and D.  Hoffmann.   The card nogeni city  of  benzofluoranthenes.
     Cancer,  12:1194-1199, 1959.
                  ?^
     J.  Natl. Cancer Inst. , 37:519-526, 1966.
 247  Shimkin  M B.  , and G. D. Stoner.  Lung tumors in mice:   Application to
                                                                -
      of benzo[a]pyrene and ferric oxide.  J. Natl. Cancer
      1204, 1972.
      Honvw  MC    CD  Port   and D.G.  Kaufman.   Importance  of physical properties!
      of Senzoiaipyrene-ferric oxide mixtures in  lung  tumor  induction.  Cancer
      Res., 35:207-217, 1975.
    intratracheal  instillations of
    390,  1973.
                                                    .   Europ.  J.  Cancer,  9.
  251. Kobayashi, N.  Production of respiratory tract tumors in hamsters by
       benzo[a]pyrene.  Gann, 66:311-315, 1975.
     1644, 1972.
J»«M  Flaks  A   and P  Sims.  The activities of some polycyclic hydrocarbons
     and tne?r'"K-region'' epoxides in an in vitro-in vjvo carcinogenicity te
     system,  Brit. J. Cancer, 32:604-609, 1975.
                                                                            test
        J  Natl.  Cancer Inst.,  52(4):1317-1325,  1974.
                                           6-266

-------
                 of
 Cancer Res.,  37:1266-1271,
  256.
  257.
                 exposed to ben2o[a]pyrene
                                                        d-
                                                        d«>ethylbenz[a]anthracene
                                                                     •*•
  258.
 259.
tteorlel CaR^"t°91ner1S:  the ^P^atfons
tneones.  Brit. J. Cancer, !5(5):5, 1961.
                                                      TUm°hrS °f »a"V sites in-
                                                       newborn mice, a sensitive

                                                      ""'*<"" i™™logical
      cutaneous,y to infant n,1ce.  Cancer  Res^
 260.
 261.
 262.  Hi
263.
264.
265.
                                              M-«urement of epidermoid
                                          ~

  bay region tneory.

                                                                  mouse and
                                       6-267

-------
268



269




270


271



979



 273.


 274
   Griswold, D.P.,  H.E.  Skipper, W.R.  Laster, W.S. Wilcox   and  F.  Schabel.
   Induced mammal  cancer  in  the female  rat  as  a  drug  evaluation  system.
   Cancer Res.,  26:2169, 1966.

   Pearson,  O.H.  Animal models of endocrine- related cancer.   In:   Host
   Environment Interactions in the Etiology  of  Cancer in Man.   Proc.  of a
   meeting.   Primosten,  Yugoslavia.   August  27  -  September 2,  1972.   Lyon,
   France;  IARC.   IARC Publ.  No.  7.   1973.   p.  137-141.

   Dao, T.L.  Carcinogenesis of mammary gland in rat.   Prog.  Exptl.  Tumor
    Res., 5:157-216, 1964.

    Huqgins.  C. , L.C. Grand, and F.P. Brillantes.   Mammary cancer induced by
    a single feeding of polynuclear hydrocarbons, and its suppression.
    Nature, 189:204, 1962-
             C   R.C. Moon, and S. Mori i.  Extinction of experimental mammary
             I/Estradiol-17B  and  progesterone.   Proc. Natl. Acad. Sci.
    USA, 48:379, 1962.

    Huggins, C.  Two principles  in endocrine therapy of  cance rs:   hormone
    deprival and hormone interference.   Cancer  Res., 25:1163,  1965.
Sinha  D , and T.L. Dao.  A direct mehcanism of mammary
induced by 7"l2-dimethylbenz[a]anthracene.  J. Natl. Cancer Inst. , 53(3):
841-846, 1974.
Mattison  D R   and S.S. Thorgeirsson.  Ovarian metabolism of polycyclic
a?omat?c' hydrocarbons and associated ovotoxicity in the mouse.  Gynecol.
Invest. 8(1/2): 11, 1977.
Kato  T   M  Yakushifi, A. Tsunawaki ,  I.  Ide,  N. Nirose,  S. Araki , and
         Studies  on experimental  formation  of ovarian  tumors  -especially

                                                           followins an
97G
                             K  Kamino, K. Uchima, K. Yamashita, and S. Yamamoto.
     Pancreatic cancer developing Sprague-Dawley rats after  Ya\???v 119-125   '
     of 7,12-dimethylbenzanthracene  (DMBA).  Osaka City Med. J., 21(2).119  i*>,

278  Bresnick  E., T. F. McDonald, H.  Yagi,  D.M. Jerina, W.  Levin   A.W. Wood,
     and A H   Conney.  Epidermal hyperplasia after topical application of
     benzoiaipyrene!  benzo[a]Pyrene  diol  epoxides, and  other metabolites.
     Cancer Res.,  37:984-990,  1977.

279  Arkhipov, G.N.   Induction of  cancer  of  the glandular  stomach  in  rats  by
     20-methylcholanthrene.   Int.  J. Cancer, 2:494-499,  1967.
                                         6-268

-------
   280. Chouroulinkov  I   A

        injection.   Cancer Res. ,    :576-5M

   282-
                                                       ,
                                                      9     fon«i"3 fntraglandular
                                                                              fn

  285'
  28S-
                   Cancer.
                                            ^
                            Scientific PubT  No.
                hydrocarbons.  Ab,.
               K   fn
      and Cancer in Man.  V. Mohr  D

             for Research on
                                                        '  12-  Air Pollution
                      ,  and
                                                                      of weak and
 289'

291 '
292.  Heideiberoer, C.
293-
                                          transfo™at'°" »1th chemical
                                                                        carcino-
                                »«»-'«'• 1" culture.  Mv. Cancer Res., J8:



                       CheMca, carcino9enesis.  Ann. Rev. Biochem. , 44:79-113.-
                                      . 6-269

-------
                -
    from newborn mice.  Cancer Res., 38:2246-2252,  1978.
    Si I GC t5 U I  |J U • j ^*J      *              ** A *^ oyirt  TQ"7Q
    I.  Benzo[a]pyrene.  Cancer Letters, 5:343-348, 19/B.   .


    M         u   s  Baker  B  Tierney  P L  Grover,  and P. Smis.  Induction

    ofrmaliqnant transformation and mutagenesis by dihydrodiols derived from
    7;i2-dimethylbenz[a]anthracene.  Biochem. Biophys.  Res. Commun., 85.357 362,

    1978.
    Cancer,  12:463-473,  1973.


298. Berwald, Y. , and L.  Sachs.  In vitro transformation ?                .
    tumor cells by carcinogenic hydrocarbons   J.  Natl. Cancer Inst. ,  m«}.

    641-657, 1965.
«•  SSfoi*-,; S y
     Cell Research, 48:361-377,  1967.
           .
     d^rivativefof poly^yZTi^hydrocarbons.  Proc.  Natl. Acad.  Sci.
     68(6): 1098-1101,  1971.
            overd       .a
     epoxides  and o?her derivatives of polycyclic hydrocarbons.  Cancer Res. ,

     32:1391-1396, 1972.
      in culture.  Cancer Res., 36:4143-4151, 1976.
 303  DiPaolo  J A , P.J. Donovan, and R.L. Nelson.  Transformation of hamster

                                                            t-   proc-
      Chem.-Biol. Interactions,  9:351-364,
      Pienta  RJ   JA  Poi ley, and W.B.  Lebherz, III.  Morphological trans-

      SSS:ss.!ss5 as-^-
      identifying diverse carcinogens.  Int.  J. Cancer,  19:642-655,  1977.
                                      6-270

-------
Cancer Res.,  36:1801-7808,  1976.



      den-^lro; ElS'JSEi
 .  Natl.  cancer  Inst. , 42(6^9l1-922,
                                                          mouse cell lines.
                                   TH
                                                  "11|IBmt
                                                                           of
                                                         of ^thylcholanthrene
  309-
                           sae c
   fo-   Int.  J. Cancer, 4: 166-178^1969
                                                           ic nylrocarbons  In
anthracene and 7-hpd73Jyn,ethylbenz
denvatives.  int. J. cancer, ]l"
                                                           * thei> K"re3lon
mouse prostate by epoxdes  and
Cancer Res.,  32:716-?20,  1972
                                                         .  •     eeroer
                                           *ra"Sfrraatfon of cells derived from
                                           """vates of polycyclic hydrocarbo°s.
 312
314'
              carcinogen.   J.
                                                                     . and
316.
                Natl.  CacerjntMonogr
                                       6-271

-------

                   Effect of benzofalpyrene on hamster,  rat, dog, and monkey
317.
    October 8-11,
    Atomic Energy



"-
     Bioassays, International Symposium.   1974.  pp. 521  531.

319
     direct action of benzoiajpyrene buuuieu
     lung tissue.  Neoplasma, 22(2):113-122, 1974.

320  Bryan, W.R., and M.B.  Shimkin.  Quantitative     -
     data obtained with carcinogenic hydrocarbons.  J. Natl.
     1:807^833,  1941.
321  Heston  WE  and M.A. Schneiderman.  Analysis of dose:response  in relation
321' £ mechanism'of pulmonary tumor induction in mice.  Sconce, 117.109 111,
     1953.

 322.
                                                                 Inst.
      1961.


 ~ 2±2 ^^^^^«^<^^