External Review
                                           Draft No. 1
                                            May 1978
                     DRAFT
               DO NOT QUOTE OR CITE
HEALTH ASSESSMENT DOCUMENT
                     FOR
  POLYCYCLIC ORGANIC MATTER

                     Ui
-------

-------
                DRAFT
           NOT QUOTE OR CITE
External  Review
Draft No.  1
May 1978
      HEALTH ASSESSMENT DOCUMENT
                 FOR
       POLYCYCLIC ORGANIC  MATTER
               NOTICE

This document is a preliminary draft.
It has  not been formally  released by
EPA and should not be construed to
represent Agency policy.   It is being
circulated for comment on its technical
accuracy and policy implication.
                DRAFT
        00 NOT QUOTE OR CITE
 U.S.  Environmental Protection Agency
  Office  of Research and  Development
       Washington, D.C.    20460

-------
                                                               QUOTE OR CUE
                          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. Micheal 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
                                    ii

-------
Mr. Joseph Bumgarner
Environmental Monitoring Surveillance 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!
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 Surveillance 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
                                     ill

-------
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 Til ton
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
                                    iv

-------
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
Naylor 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

-------
                              TABLE OF CONTENTS







                                                                      Page



1.   EXECUTIVE SUMMARY                                                 1-1



2.   INTRODUCTION                                                      2-1



3.   PHYSICAL AND CHEMICAL DATA                                        3-1



4.   SAMPLING AND ANALYTICAL METHODS FOR POM's                         4-1



5.   AMBIENT LEVELS                                                    5-1



6.   HEALTH AND ECOLOGICAL EFFECTS                                     6-1
                                      vi

-------
                                                              DRAFT
                                                         NOT QUOTE OR CITE
                            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 transport,  measurement techniques, ambient levels,
toxicology, and occupational health and epidemiology.
1.1  DEFINITION AND FORMATION
     The two POM chemical groups most commonly found in ambient air are
polycyclic aromatic hydrocarbons (PAH's), such as the well known carcinogen
benzo[a]pyrene (BaP),  and the  PAH nitrogen analogs.  In addition, a small
number of oxygen-containing POM's have been detected in ambient air.  The
major environmental sources of POM's  appear to be the combustion or pyroly-
sis 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's and 32 nitrogen and oxygen analogs have been quantitatively analyzed
so far in ambient air.  The failure to detect POM's 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 suscepti-
bility varies considerably among individual compounds, the physical  form of
POM tested, and the reaction conditions used.  However, available monitor-
ing 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's 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's from various
sources and for the analysis of collected samples for quantification of
individual compounds.  POM's in ambient air are usually collected as par-
ticulate matter on filters.  Some of the more volatile POM's may not be
efficiently collected, but the collection efficiency for POM's (including
BaP) has not been determined.  Because of the large number of individual
compounds in the POM family, careful fractionation and separation proce-
dures are necessary to assure 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
only monitoring 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
                                                       3                    3
for BaP:  the median urban value declined from  3.2 ng/m  in  1966 to 0.5 ng/m
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's 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's 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 has 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's 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's are human carcinogens, several of these com-
pounds are among the more potent animal carcinogens known to exist.
     POM's gain ready access to the body's circulation either by inhalation,
ingestion, or contact with the skin.  In their parent form, however, POM's
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 non-carcinogenic 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 ROM's.  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 ROM's 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 ROM's 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 ROM's.   However,  in
real-life situations most scientists agree that the effects of  carcinogenic
ROM's will most likely be a continuous function of dose, no matter how
small.
     Whereas exposure to POM's 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's in city air.  Moreover, it has  been shown
                                     1-6

-------
that cities having the highest levels of POM-containing pollutants 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 necessary to
result in an increased cancer risk.
     Since all carcinogens are mutagens as well,  it is not surprising that
numerous POM's produce mutations in certain animal and bacterial cells in
culture.  However, POM's have not been conclusively shown to produce muta-
tions in either animals or man which can be passed from one generation to
the next.  Furthermore, there is no evidence to indicate that POM's in the
air may be producing mutations to lower organisms in the environment, or
that POM's are presently disturbing the ecological balance in any way.
POM's are apparently not bioaccumulated in any species, nor are they trans-
ferred 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's is real.  The magnitude of that risk to the general
population, however, cannot be accurately determined based on our present
knowledge.  It is almost certainly much less than the lung cancer risk
associated with cigarette smoking, since POM exposure from smoking is far
greater than that from inhaling polluted air.  Nevertheless, in conjunction
                                     1-7

-------
with smoking and the other foreign chemicals (carcinogenic and non-carcino-
genic) to which man is exposed, it is likely that the absolute risk of POM
exposure at any concentration is magnified in real-life situations.  Thus,
in adopting the most conservative approach for evaluating the human health
hazard of ROM's, most scientists conclude that although the risk of cancer
by exposure to low doses of ROM's is probably very small, it will not be
possible to completely eliminate the risk unless all  exposures are pre-
vented.
                                      1-8

-------
                                                    .00 NOT QUOTE Oft CITE
                              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

Section 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 ROM's:  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 ROM's 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's), including benzo[a]pyrene (BaP), and the aza arenes - were
focused upon because of their association with carcinogenesis.
                                      2-2

-------
                                 References
1.   Participate Polycyclic Organic Matter.   National  Academy of Sciences.
     Washington, D.C.  1972, 361  pp.

2.   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.

3.   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

-------
                                                                DRAFT
                                                       DO NOT QUOTE OR CITE
                       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)
               Imino 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's),  such as benzo[a]pyrene, and their nitrogen

analogs, aza and inn'no arenes,  which are  formed during organic combustion
          p c f
processes. ' '   Because of their common  sources, their existence in urban

air, and the considerable experimental data on their carcinogenic effects,

the PAH's and the  nitrogen analogs  have received the most attention, and

therefore this report has focused on these POM's.  However, some of the

other arenes (for  example, 7H-Benz(de)anthracen-7-one) have been detected in

urban atmospheres.   The latter compounds will be considered in this report

even though a limited amount of data are  available.
                                     3-1

-------
     ROM's 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-
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
                                       c
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
                                                                   2
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
                                                      p
to the postulated route to ROM's, Badger and Spotswood  pyrolyzed toluene,
ethylbenzene, propylbenzene, and butylbenzene and obtained the highest
                                                                 Q
yields of benzo[a]pyrene with butylbenzene.  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

-------
    I
   Co
  BaP
Figure 3-1.   Mechanism of benzo[a]pyrene formation.4
                            3-3

-------
suggest that the C2 species react to form C., Cg, and C» species, and that
reactions involving styrene and phenylacetylene are probably important in
the formation of polycyclic aromatic hydrocarbons.  Also, a C-IQH-IQ 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-
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
ROM's.
     By far the most significant source of atmospheric ROM's is pyrolitic or
combustion processes.  However, 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's 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's 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
       14
worked.
     The currently accepted nomenclature is that adopted by the International
                                                                              2
Union of Pure and Applied Chemistry (IUPAC) and by Chemical Abstracts Service.
Of major importance are the following rules that determine the orientation from
which the numbering is assigned:2'14'15'16'17'18
     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; and
     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 ROM's 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 nor are they 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 are divided into polycyclic
aromatic hydrocarbons (PAH), aza and imino arenes,  and miscellaneous  other
ROM's that have been detected in air.   The structures  are  presented in the
                                                                           p
same way as those in the NAS report, Particulate  Polycyclic Organic Matter,
in that aromatic rings are shown as plain hexagons  and a methylene group  as
C^.  The indications of carcinogenicity, taken from the NAS report,  were
obtained from several Public Health Service (PHS) surveys  of compounds tested
                    19 20 21 22  23
for carcinogenicity.  '  '  *  'In some instances, indications of carcin-
ogenicity were obtained from individual references; in these cases the refer-
ences 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
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

-------
                                  Table 3-1.  POLYCYCLIC AROMATIC HYDROCARBONS
CO
00
Compound
Structure
Formula
Carclnogenlclty
Acronym
Physical properties
Melting point, °C
Boiling point, °C
Vapor pressure, Torr
Xraax , nm
Log e
Solvent
Anthracene Benz[a]anthracene 7, 12-Dime thy Ibenz [a] anthracene
(9,10-dlmethyl-l,2-benzanthracena)
5 »o 1 «"j
(not an air pollutant)
C14H10 C18H12 C20H16
A BaA DMBA
216 159.5-160.5 122-123
340 400
1.95 x 10~* 1.10 x 10"'
308 323 338 355 375 314 327 341 359 376 386 296 345 364 384
3.15 3.47 3.75 3.86 3.87 3.67 3.81 3.87 3.72 2.73 3.86 4.90 3.83 3.94 3.83
EtOH-MeOH EtOH EtOH

-------
                                   Table 3-1 (Cont'd).   POLYCYCLIC  AROMATIC  HYDROCARBONS
         Compound
                           Dibcnz[a,j]anthracene
                         (1,2-6,7-dibenzanthracene)
   Dibenz[a,h]anthracene
(1,2-5,6-dibenzanthracene)
   Dibenz[a,c]anthracene
(1,2-3,4-dibenzanthracene)
         Structure
co
vo
Formula

Carclnogenlclty

Acronym

Physical properties
     Melting point, °C
     Boiling point, °C

Vapor pressure, Torr

Xmax, nm
     Log e
     Solvent
                                    C22H14
   C22H14
                                                                   DBahA

                                                                   266-266. S
   C12H14
                                                                    299   322   335   350   374   384
                                                                    5.20  4.28  4.23  4.16  3.00  2.61
                                                                    dioxane

-------
                                     Table  3-1  (Cont'd).   POLYCYCLIC AROMATIC  HYDROCARBONS
Compound
Structure
	 ' 	 	 1
Phenanthrene Ben2o[c]phenanthrene Fluorene
(3,4-benzphenarthrene)
/a i
t * »
CO
_J
o
Formula

CarcinogenIclty

Acronym

Physical properties
    Melting point, "C
    Boiling point, °C

Vapor pressure, Torr

Xmax, nm
         C14H10
         6.8 x 10 *

309   314   323   330   337   345
2.40  2.48  2.54  2.52  3.40  3.46
         EtOH-MeOH
                                                                          C18H12
                                         BcP


                                         68
                                                                          296   3f.3   313   323   353   372
                                                                          4.13  4.07  4.06  3.67  2.63  2.38
                                                                          EtOH
                                                                                      C13H10
                                                                                                                        116
                                                                                                                        293

-------
                       Table 3-1  (Cont'd).   POLYCYCLIC AROMATIC HYDROCARBONS
Compound
 Benzo[a]'luorene
(1,2-benzfluorene)
 Benzo[b]fluorene     Dibenzo[a,hifluorene
(2,3-benzfluorene)   (1,2-6,7-dlbenzfluorene)
 Dibenzo[a,g]fluorene
(1,2-5,6-dIbenzofluorene)
Structure
                     ofco
Formula

Carcinogenicity

Acronym

Physical properties
    Melting point, °C
    Boiling point, °C

Vapor pressure, Torr

Xmax, nm
    Log e
    Solvent
 C17H12
 189-190
 C17H12
                      208-209

-------
                                 Table  3-1  (Cont'd).  POLYCYCLIC  AROMATIC HYDROCARBONS
        Compound
        Structure
                                   Benzo[c]fluorene        Dibenzo[a,c]fluorene
                                   (3,4-benzfluorene)     (1,2-3,4-dlbenzfluorene)
                                                              **•
                                               Benzo[b]fluorant hene
                                              (2,3-benzfluoranthene)
                                               c&o
CO

ro
Formula

Carclnogeniclty

Acronym

Physical properties
    Melting point, °C
    Boiling point, °C

Vapor pressure, Torr

Xmax, nm
    Log c.
    Solvent
                                   C17H12
C21H14
C16H10
                        Flour
                        110
                        393
C20H12
                                                BbF
                                                                                                         167

-------
                                       Table  3-1  (Cont'd).    POLYCYCLIC AROMATIC  HYDROCARBONS
            Compound
 Benzo[J)fluoranthene      Benzo[k)fluoranthene
(7,8-benzfluoranthene)     (8,9-benzfluoranthene)
                                                                                            Benzo[ghl)fluoranthene
                                                                                                                          Aceanthrylene
            Structure

C*>
Formula

Carcinogenicity

Acronym

Physical properties
    Melting point. *C
    Boiling point. *C

Vapor pressure, Torr

Xmax, nm
    Log c
    Solvent
C20H12



 BJF


165-167
                                                                   C20H12
                                                                   BkF
                                                                   215-216
                                                                   9.59 x 10
                                                                           "11
                                                  C20H10
C16H12
                                                                                                                            113

-------
                                Table  3-1  (Cont'd).   POLYCYCLIC AROMATIC HYDROCARBONS
        Compound
        Structure
Benzo[j]aceanthrylene
(cholanthrene)
                                                               3-Methylcholanthrene
                                                                (not an air pollutant)
                          Naphthacene
                           (benzo[b]anthracene;
CA>
        Formula

        Carcinogenicity

        Acronym

        Physical properties
             Melting point, °C
             Boiling point, °C

        Vapor pressure, Torr

        Xmax, nra
             Log e
             Solvent
C20H14
+ +
C21H16
                                  MCA
                                  179
                                  no absorption above
                                  300 run
C18H12
                          341
                          sublimes

-------
                               Table 3-1  (Cont'd).    POLYCYCLIC  AROMATIC HYDROCARBONS
Compound
Napht ho [ 2 , 1 , 8-qra ) naphthacene Pyrene
(naphtho(2,3-a)pyrenc)
(2 , 3-naphtho-l, 2-pyrene)
1-Hethylpyrene
5"3
Denzo[a]pyrene !
(1,2-benzpyrene) 1
(3,4-benzpyrene*) I
Structure
                                                                                                            T  t   *
Formula

Carclnogenlclty

Acronym

Physical properties
    Melting point,  °C
    Boiling point,  °C

Vapor pressure, Torr

\rnax, no
    Log c
    Solvent
C24H14
C16H10
                                     152.
                                     >360

                                     6.8S x 10
                                                         C17H12
        ,-7
C20H12
                                                                                BaP
                                                                                178.8-179.3
                                           5.49 x l
-------
                            Table  3-1  (Cont'd).   POLYCYCLIC  AROMATIC HYDROCARBONS
Compound
Structure
 Benzo|e]pyrene
(4,5-benzpyrene)
(1,2-benzpyrene*)
               Dibenzo[a,1]pyrene
              (2,3-4,5-dibenzpyrene)
             (Dlbenzo [def, p ] clirysene)
              (1,2-3,4-dlbenzpyrene*)
  Dlbenzo[a,h]pyrene
 (1,2-6,7-dibenzpyrene)
(Dlbenzo[b,def ]chrysene)
 (3,4-8,9-dibenzpyrene*)
                                                                            oS9°
      Dlbenzo[a,1]pyrene
     (2,3-6,7-dibenzpyrene)
     (4,5-8,9-dibenzpyrene*)
Formula

Carcinogeniclty

Acronym

Physical properties
    Melting point, °C
    Boiling point, °C

Vapor pressure, Torr

Amax, no
    Log e
    Solvent
 C20H12
 BeP


 178.9


 5.54 x 10
               C24H14
.-9
  C22H14

  •f + +

  DBahP
                                              320-320.5
                                                290   301   313   379
                                                4.54  4.94  5.23  3.60
                                                       benzene
      C22H14
                                                                                      DBalP
                                                                             281-282.5
                                                            401
                                                            4.03
                      424
                      4.40
451
4.59

-------
                             Table  3-1  (Cont'd).   POLYCYCLIC AROMATIC HYDROCARBONS
I
~J
Compound Dibenzo[a,e]pyrene
Structure IT J. J
Formula C22Hli
Carcinogeniclty + +
Acronym
Physical properties
Melting point, "C
Boiling point, °C
Vapor pressure, Torr
A max, nm
Log c
Solvent
Dibenzo[cd,jk]pyrene Indeno[l,2,3-cd]pyrene Chrysene
(anthanthrene) (o-phenylenepyrene)
(dibenzo[def ,mno] chrysene)
U '
r'V'V^i r^r^ "r^T^**!*
rVrV rrVrS .^fWv^0^3
kAAx1 kA-kAx1 KJL/ly *
7 t
C22H12 C22H12 C18H12
+ +52
Ch
257 254
A48
344 351 360
2.88 2.62 3.00
EtOH

-------
                                   Table 3-1  (Cont'd).   POLYCYCLIC  AROMATIC HYDROCARBONS
         Compound
Ferylene
Benzo[ghi]perylene
                                                                  Coronene
         Structure
         Formula

oj       Carcinogenicity

00       Acronym

         Physical properties
              Melting point,  °C
              Boiling point,  °C

         Vapor pressure, Torr

         Xmax, nm
              Log e
              Solvent
C20H12
Per
273
ca 500
C22H12
BghiP


273


1.01 x 10
                                     ,-10
C24H12
                                                                   Cor
438
525

1.47 x 10
                                                                           -12

-------
                                  Table 3-2.   AZA AND IMINO ARENES
Compound
                                  Acridine
                           Benz[a]acrJdine
                          (1,2-benzacridine)
  Benz[c]acridine
(3,4-benzacridine)
Structure
                         OCX?
Formula

Carcinogenicity

Acronym

Physical properties
     Melting point, °C
     Boiling point, °C

Vapor pressure, Torr
111
>360
                           C17H11N
  C17H11N
Xmax ,  nm
    Log c
    Solvent

-------
                                 Table 3-2 (Cont'd).   AZA AND IMINO ARENES
Compound
   Dibenz[a,J ]acridine
(1,2-7,8-dibenzacridine)
   Dibenz[a,h]acridine
(1,2-5,6-dibenzacridine)
                                                                                                       Dibenz[c,h]acridlne
                                                                                                     (3,4-5,6-dibenzacridine)
Structure
F°rmula
Carclnogenicity

Acronym

Physical properties
     Melting point, °C
     Boiling point, "C

Vapor pressure, Torr

Xmax, nm
     Log E
     Solvent
   C21H13N
   + +
                               220
   C21H13N
                                                                228
                                                                                                       C21H13N

-------
                                     Table 3-2  (Cont'd).  AZA AND IMINO ARENES
         Compound
         Structure
                           Carbazole
 Benzo[a]ca rba zole
(1,2-benzcarbazole)
                                                      oi?
                                                                                             Dibenzo[a,g]carbazole
                                                                                           (1,2-5,6-dlbenzcarbazole)
ro
Formula

Carcinogeniclty

Acronym

Physical properties
    Melting point, °C
    Boiling point, °C

Vapor pressure, Torr

Xmax, nm
    Log c
    Solvent
                                     246
                                     355
                                                               C16H11N
                               C20H13N

-------
                                          Table  3-2  (Cont'd).   AZA  AND  IMINO ARENES
        Compound
                                   Dibenzo[c,g]carbazole
                                 (3,4-5,6-dibenzcarbazole)
  Dibenzo[a,i]carbazole
(1,2-7,8-dibenzcarbdzole)
Quinoline
        Structure
co
ro
ro
Formula

Carcinogenicicy

Acronym

Physical properties
     Melting  point,  °C
     Boiling  point,  °C

Vapor pressure,  Torr

Xmax, nm
     Log e
     Solvent
                                           C20H13N
                                           DBcgC
  C20H13N
                                                                                                                       53

-------
                                      Table  3-2  (Cont'd).   AZA AND  IMINO ARENES
Compound
Benzo[f Iquinoline
                                                                Benzo[h]quinoline
                                                                                                   H-Indeno(l,2-b)quinoline
        Structure
to
ro
Formula

Carcinogenicity

Acronym

Physical properties
     Melting  point, °C
     Boiling  point, °C

Vapor pressure, Torr

Amax , nm
     Log E
     Solvent
                                                    54

-------
                                     Table 3-2  (Cont'd).   AZA  AND  IMINO ARENES
       Compound
                               Isoquinoline
Benzo[f]isoquinoline
                                                                                                     Phenanthridine
       Structure
ro
Formula

Carcinogenicity

Acronym

Physical properties
     Melting point, °C
     Roiling point, °C

Vapor pressure, Torr

Xmax, nm
     Log e
     Solvent

-------
                                Table  3-2  (Cont'd).   AZA AND IMINO ARENES
     Compound
                               4-Azafluorene
                         (5-H-indeno[l,2b]-pyridine)
     4-Azapyrene
(Benzo(lam)phenanthridine)
     1-Azafluoranthene
(Indeno-[1,2,3-ij]isoquinoline)
     Structure
                                   cxo
                                                                                    oS.
ro
tn
Formula

Carcinogenicity

Acronym

Physical properties
    Melting point, "C
    Boiling point, °C

Vapor pressure, Torr

Xmax, nm
    Log c
    Solvent

-------
                                Table 3-3.   MISCELLANEOUS  POLYCYCLIC ORGANIC  MATTER
          Compound
                                       Phenalene-1-one
                                                            Anthanthrone
                                                                      Xanthene-9-one
7-H-Benz(de)anthracene-7-one
     (benzanthrone)
          Structure
                                                                      oio
I
l\>
CTl
Formula

Carcinogeniclty

Acronym

Physical properties
    Melting point, °C
    Boiling point, °C

Vapor pressure, Torr

Xmax, nm
    Log c
    Solvent

-------
3.2.2  Physical Properties
     Most of the POM compounds are high melting/high  boiling  point  solids  that
appear to be extremely insoluble in water.   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. '
     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's.
Depending upon the ring size and molecular weight of  the compounds, the  vapor
pressure of the pure compound varies from  6.8 x 10    Torr  for phenanthrene
                                    -12
(3 rings and 14 carbons) to 1.5 x 10"   Torr for  coronene  (7  rings  and 24
         24 ?*5 ?fi
carbons). ^»"«to  This property has considerable impact on the amount of
PAH's that remain adsorbed on particulate  matter  in the atmosphere  and retain-
ed 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
                                      1 27
matter at 1.2 cubic meters per minute: *    (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

-------
£
               200
  300                400

WAVELENGTH, nm
     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
     (3.76);  B, 2965 (4.76), 2843 (4.66), 2740  (4.50); Bf, 2655 (4.66),
     2540  (4.60); g", 2250 (4.44).
                                 3-28

-------
velocity of the air with respect to the filter,  and the adsorption character-
istics of the individual ROM's.
     The ultraviolet absorption  spectra are also available for many compounds
(Table 3-1).   Most of the PAH's  absorbed light strongly at wavelengths  longer
            28
than 300 nm.     The absorption spectra for benzo[a]pyrene (Figure 3-3)  is
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's in organic solvents, while PAH's in the environment are  usually found
associated with particulate matter and adsorbed PAH's 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's in the atmosphere.  Because of the high melting  points and low
vapor pressures of most POM's, 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's 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 10-15 ym) 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
                                     3-29

-------
Individual air path, the pattern and depth of respiration,  and the  character-
                        pa
istics of the particles.
     The particle size also affects the residence time of participate  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-
ticles 1 to 10 ym in diameter.
     Tebbens et al.    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.
                    32
     DeMaio and Corn   collected particles with a two-stage elutriator.  The
first stage collected 50 percent of particles with a diameter of 4.6 ym  and
100 percent  of 6.5 ym particles.   The second stage, which  consisted of  a
glass fiber filter paper collected 100 percent of all particles greater  than
0.3 ym.  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
                                             33
in stage one.  Kerte"sz-Saringer and coworkers   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.
                                                              32
Their results (Figure 3-4) agree with those of DeMaio and Corn   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 ym.
                                    3-30

-------
         99



         98

         97

         96

         95
-J  90

i-
z
LU

oc  80
UJ
a.


?  70



   60



   50


   40
         30
          0.1
                                   1.0


                              R ADI US (r). tun
                                                      • 9/11 to 9/24    _

                                                      O 10/15 to 11/14

                                                      A 11/21 to 12/16

                                                      D AVERAGE
10
Figure 3-4.   Size distribution of particles  containing benzo[a]pyrene.
                                                                          33
                                    3-31

-------
                    34 35
     Katz and Pierce  '   have conducted a detailed study of the relationship
of PAH's 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's.  These results
demonstrate that BaP and PAH's are associated primarily with small, respirable
particles.
     In elevated temperature combustion systems, substantial amounts  of ROM's
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.
             OC
Thomas et al.   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 al.   also determined the amount of BaP from the combustion
                                    3-32

-------
   100


    50


i-   30
 E
 s
 o
 C   10



 I   5
 I-
 Ul
 u
 a:
 §  50

 |,  30


 %
 CO

 £  10
 UJ
 U

 5  5
 a.
I      I      I      I
               YORK-S
        BATHURST-S
                                                    MacDONALD

                                                     CARTIER
                                            .COLLEGE-S
                                          COLLEGE-W
                                        STAGE NO.
           Figure 3-5.   Percent mass  of benzo[a]pyrene in relation to
           stage number  for five sampling sites in  Toronto, Canada.35
                                        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.
                      37
     Krstulovic et al.   have placed polyurethane  plugs behind a standard
glass fiber filter (1.0 pro retention with 98 percent efficiency) to collect
particles that pass through.  The authors refer to these  particles as 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
                                3                                   3
in Providence was 3709 ng/1000 m  on the foam plug but 806 ng/1000 m  on  the
                                                                   34 35
filter.  These results are consistent with those of Pierce and Katz  '   which
indicated that sizable amounts of PAH's will be associated with particles less
than 1 ym in diameter.
                        38
     Natusch and Tomkins   have developed a mathematical  description of the
process of adsorption of PAH's vapors on fly ash particles.  Their calcula-
tions indicated that PAH's would almost certainly  exist as vapors at tempera-
f
tures (>150°C) encountered in the stack systems of fossil-fueled power plants.
However at ambient temperatures (-10 to 30°C), "essentially quantitative
                                                38
adsorption [to particulates] would be expected."    Their theoretical study
also suggests that since "the extent of adsorption of PAH's is proportional to
the frequency of collision with the available surface area, then adsorption
will result in small respirable particles being preferentially enriched (per
                                              38
unit mass) in potentially carcinogenic PAH's."
                                     3-34

-------
3.3  ATMOSPHERIC CHEMISTRY
     The POM compounds emitted to the atmosphere have been shown to have con-
siderably different stability.  Unknown, however, is whether the degradation
                                                                             39
products are more or less toxic.  Preliminary evidence by Pitts and coworkers
suggests that in some instances (e.g., nitrous oxide reaction with benz[a]-
pyrene to form 6-nitrobenzo[a]pyrene and a mixture of 1-nitro and 3-nitro-
benzo[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 ROM's in the atmosphere are mostly
found in the atmosphere adsorbed to particulate matter, such factors as
particle size (affects 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 ROM's 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
                                                                          40
adsorbed on particul ate matter is probably remote.  Kuratsune and Hirohata
photolyzed 13 PAH's 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's
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's in the environment are usually found adsorbed to partic-
ulate matter, many investigators have studied the photolysis of POM's natur-
ally occurring on particulates from combustion, POM's coated on some solid
particles, and POM's coated in thin films on a smooth surface.  Typical  of
                                                                         42
photolysis studies of pure POM's adsorbed on solids is the work of Inscoe
wh'o examined photodecomposition of PAH's during thin layer chromatography.
She exposed 15 PAH's 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  eleven compounds
that were photosensitive were anthracene, naphthacene, benz[a]anthracene,
dibenz[a,c]anthracene, dibenz[a,h]anthracene pyrene, benzo[a]pyrene, benzo[e]-
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

-------
                       43
     Andelman and Suess   attempted to simulate photodecomposition  of PAH's  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
                                                                      44
constituents on the photochemistry of adsorbed POM.  Jager and Rakovic
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 S02
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 S02 is
                                                                         39
much higher than is normally present in ambient air. Pitts  and coworkers
have also indicated that POM's may react with nitrous oxide  to form nitro
derivatives.
              45
     Geacintov   studied the photolysis of 20 PAH's 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, Op and NO.  However, the PAH's were efficient photocatalysts for  the
formation of singlet oxygen, a reactive oxidizing agent which may react  with
the PAH's.
                                    3-37

-------
     Perhaps the earliest study of POM photochemistry was conducted by Falk
              46
and coworkers.    They noted that the amount of BaP in ambient air compared
with other PAH's 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.
           47
Falk et al.   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.   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

-------
GO
1O
                       Table  3-4.   PERCENT  DESTRUCTION  OF POLYCYCLIC AROMATIC  HYDROCARBONS
                                          UNDER ATMOSPHERIC CONDITIONS^
Pure Unadsorbed Compounds

Anthanthrene
Phenanthrene
Pyrene
Fluoranthene
Benzo[a]pyrene
Benzo[e]pyrene
Benzo[ ghi]perylene
Coronene
Chrysene
Air
24 hr

44
39
24
17
0
—
0
0
0
in dark
48 hr

49
61
43
20
0
--
0
0
0
Air in
24 hr

42
34
20
16
21
--
0
0
11
light
48 hr

44
60
42
24
22
—
0
0
0
Smog in light
1 hr

—
--
83
59
50
—
27
5
15
Adsorbed on
Air in light
48 hr

5
-
1
4
10
7
0
—
_-
soot
Smog
1 hr

55
—
58
59
18
51
67
—
__

-------
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.
                  oc
     Thomas et al.   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
                        •tc
minutes).  Thomas et al.   also examined the effect of particle  size by
coating glass beads (1 g of either 220 urn or 28 pm 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.
                  48
     Lane and Katz   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 Quartzline  lamp.  They
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

-------
        Table 3-5.   THE HALF LIVES OF  BaP,  BbF, and  BkF UNDER VARIOUS
                              REACTION CONDITIONS48
Reaction

Irradiation
None


Quartzline
Q500T/CL
lamp

conditions
Ozone
concentration (ppm)
0.19
0.70
2.28
0.0
0.19
0.70
2.28


BaP
0.62
0.4
0.3
5.3
0.58
0.2
0.08
Half lives,

BbF
52.7
10.8
2.9
8.7
4.2
3.6
1.9
hra

BkF
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.
                                      3-41

-------
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 non-photochemical  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's coated in a thin film  on glass.  Whether these
are the same results that would be obtained with PAH's adsorbed on  soot is
                     45
unknown.  Falk et al.   have shown that soot adsorption can have a  stabilizing
effect on BaP.
     Relatively few studies have attempted to identify the photooxidative
                                                     41          42
products.  As indicated earlier, Masuda and Kuratsune   and Inscoe    have
identified quinones of BaP and pyrene.  This is consistent with the monitoring
                           49
results of Pierce and Katz,   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
     As indicated in some studies in the previous  section, light is not
necessarily required for the atmospheric decomposition of POM's. Falk et
al.   found that for many PAH's (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's adsorbed  on soot was very high in  a
synthetic smog that had a high oxidant concentration.  In addition, Lane and
    48
Katz   found that BaP reacted rapidly in reasonably low ozone concentrations
                                    3-42

-------
(0.19-0.70 ppm) without irradiation and that the addition of irradiation had
little effect on the rate.  However, the other two PAH's  (BbF and BkF)  were
not nearly as susceptible to ozone oxidation.   Thus,  it appears  that ROM's
will be susceptible to air oxidation without light irradiation and that the
reactivity is likely to vary for different structures.
            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
weight PAH's 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 ROM's  oxidize or  photodegrade
at a significant rate under atmospheric conditions.  However, no satisfactory
quantitative rate data are available.  In contrast to the results reported
above, Lunde and Bj0rseth   published suggestive evidence that many ROM's 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's 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,
                                    3-43

-------
co
                            Table 3-6.  COMPARISON OF PAH CONTENT OF AIR SOLIDS BEFORE
                                                 AND AFTER STORAGE27

PAH
Fluoranthrene
Pyrene
Benzo[a]pyrene
Benzo[e]pyrene
Anthanthrene
Benzo[ghi]perylene
Coronene
Vapor pressure of pure
compound25,26

6.83xlO"7
5.49xlO"9
5. 54x1 O"9

l.OlxlO'10
1.47xlO"12

Analyzed immediately
after collection
225
328
111
71
70
252
142
Concentration of PAH, yg/5
Analyzed 1 yr after
collection
18
38
76
55
55
226
140
smoke
Loss
%
92
88
32
23
21
10
1

-------
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,  benzo[ghi]-
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's), 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.
     The major environmental sources of POM's appears 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 results in the formation  of hundreds  of POM's.
     Most of the POM's are high melting/high boiling point solids  that are
very insoluble in water.  The available vapor pressures of the pure  compounds
                                          A                      TO
vary for the individual compounds from 10   Torr  (3 rings) to  10    Torr  (7
rings).  The PAH's are strong adsorbers of ultraviolet light of wavelengths
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 partially found 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
                                    3-45

-------
aerosols appear to have a residence time without precipitation of 4 to 40 days
for particles less than 1 \m in diameter and 0.4 to 4 days for particles 1  to
10 ym in diameter.  Thus the atmospheric lifetime of a POM particle will be
closely related to its particle size.
     Experimental chemical and photochemical reactivities of ROM's 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-
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's 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's adsorbed on
particles are relatively reactive.  However, this may mean that there is an
inner layer of POM's that would be relatively stable and perhaps available  for
                                    3-46

-------
elution by biological fluids following inhalation.   The relatively long
stability of several PAH's 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's.
                                    3-47

-------
                                 References


 1.   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.

 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.  l^(l):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 of 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.  l_3_:553-557, 1963.

 8.   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 butyl benzene.   J. Chem.  Soc.  4420-4427, 1960.

 9.   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.

10.   Crittenden,  B. D., and R. Long.  The mechanisms  of  formation of polynuclear
     aromatic compounds in combustion systems.   In_:   Carcinogenesis.  Vol.  1.
     Polynuclear 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, J88_: 53-55, 1975.


                                     3-48

-------
13.  Hites, R. A., R. E. LaFlamme,  and J.  W.  Farrington.   Sedimentary  polycyclic
     aromatic hydrocarbons:   The historical  record.   Science,  198:829-831,
     1977.

14.  Dipple, A.  Polynuclear aromatic carcinogens.   In;  Chemical Carcinogens,
     C. E. Searle (ed.), ACS Monograph 173.   American  Chemical Society, Washing-
     ton, D. C.   1976.   pp. 245-314.

15.  Patterson, A. M., L.  T. Capell,  and  D.  F. Walker.  The  ring  index.  A
     list of ring systems used in organic  chemistry  (2nd ed.), American Chemical
     Society, Washington, D.C. 1960.   1425 pp.

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

17.  Capell, L. T. and D.  F. Walker.   Supplement  II  (1960-1961) to  the ring
     index.  American Chemical Society, Washington,  D.C.   1964.

18.  Capell, L. T. and D.  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.).  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
     for carcinogenic activity.   Supplement 1.  Public Health  Service  Publication
     149-1.  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
     for carcinogenic activity.   Supplement 2.  Public Health  Service  Publication
     149-2.  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.

23.  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.  1638 pp.

24.  Radding, S. B.,  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 hydrocarbons.   Stanford Research Institute,  Menlo Park,  Calif.,
     Environmental Protection Agency, Washington, D.  C.  Contract No.  EPA-68-
     01-2681.  PB-250 948.   1976.

25.  Murray, J. J., R. F.  Pottie, and C.  Pupp.  The  vapor  pressures and enthalpies
     of sublimation of five polycyclic aromatic hydrocarbons.  Can. J. Chem.,
     5£(4):557-563, 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  trioxide
     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-Gottingen-Heidelberg, 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., j>: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., 16:67-
     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., J>:429-431, 1971.

34.  Katz, M., and R. C. Pierce.  Quantitative distribution of  polynuclear
     aromatic hydrocarbons in relation to particle size of urban  particulates.
     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 Katz, M.  Dependency of polynuclear aromatic  hydrocarbon
     content on size distribution of atmospheric aerosols. Environ.  Sci.
     Technol., .9(4):347-353, 1975.

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

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

38.  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.  Anonymous.  Nitrogenous air pollutants:   Bigger hazard.   Chem.  Engr.  News,
     5£(13):22, 1978.

40.  Kuratsune, M. , and T.  Hirohata.   Decomposition of polycyclic aromatic
     hydrocarbons under laboratory illuminations.   Natl.  Cancer Inst.  Monograph
     No. 9, 117-125, 1962.

41.  Masuda, Y., and M. Kuratsune.  Photochemical  oxidation of benzo[a]pyrene.
     Air Wat. Pollut. Int.  J., lp_:805-8H,  1966.

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

43.  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.

44.  Jager, J., and M. Rakovic.  Sulphur-dioxide-induced qualitative changes
     in polycyclic aromatic hydrocarbons adsorbed  on solid carriers.   J. Hyg.
     Epidemiology,  Microbiology, and Immunology, Jj3(2): 137-143, 1974.

45.  Geacintov, N.  E.  Reactivity of Polynuclear Aromatic Hydrocarbons With
     02 and NO in the Presence of Light.  Prepared under Grant No. R801393.  U.S.
     Environmental  Protection Agency, Washington,  D.C., Publication  No.  EPA-
     650/1-74-010.   1973.

46.  Falk, H. L. , I. Markul , and P.  Kotin.   Aromatic hydrocarbons.   IV.  Their
     fate following emission into atmosphere  and experimental  exposure to
     washed air and synthetic smog.   A.M. A. Arch.  Ind. Hlth.,  ]3_:113-17, 1956.

47.  Falk, H. L. , P. Kotin, and A. Miller.   Aromatic polycyclic hydrocarbons  in
     polluted air as indicators of carcinogenic hazards.   Intl. J. Air Poll.,
     2.:201-209, 1960.

48.  Lane, D. A., and M. Katz.  The photomodification  of benzo[a]pyrene,
     benzo[b]fluoranthene,  and benzo[k]fluoranthene under simulated  atmospheric
     conditions.  In:  Fate of Pollutants in  the Air and Water Environments
     Part 2.  I. A. Suffet (ed.), New York, Wiley-Interscience.  1977.

49.  Pierce, R. C. , and M.  Katz.  Chromatographic  isolation and spectral analysis
     of polycyclic quinones application  to air pollution analysis.   Environ.
     Sci. Technol., J£(l): 45-51, 1976.
50.  Lunde, G. and Bjorseth, A.  Polycyclic aromatic hydrocarbons in long-range
     transported aerosols.  Nature, 268_:518-519, 1977.

51.  Hoffman, D., and E. L. Wynder.  On the carcinogenic activity of dibenzo-
     pyrenes.  Z. Krebsforsch. , 68:137-149, 1966.
                                    3-51

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

53.   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.

54.   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.
                                    3-52

-------
                                                                 DRAFf
                                                       JPO NOT QUOTE OR CITE
                4.   SAMPLING  AND ANALYTICAL METHODS FOR POM'S

4.1  INTRODUCTION
     In vivo biassays 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's) 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 ROM's 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 ROM's, 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 implementation of appropriate sampling strategy play
important roles in the acquisition of relevant data desired from ambient air
sampling.
     ROM's can be present in the air samples in three forms:   (1) adsorbed on
foreign particulate matter, (2) condensed ROM's 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's 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's, particularly the volatile components.
     The desorption of the collected compounds is no less problematic.   The
particulate adsorbed part of the POM's 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's obtained from this step.
     The lack of clean up and separation of POM's 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 quantitation 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 ROM's.   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, C  -benzo[a]-
pyrene, C  -benzo[a]anthracene or some other nonisotopic compounds not present
in the original sample.  A number of compounds including £-terphenyl, bi-
benzyl, and benzo[b]chrysene have been used for the  latter purpose.
     The instability of the ROM's 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 ROM's in the air is from vehicular
exhaust.  The sampling methods for ROM's from this source is 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-
                  1       2
strated by Spindt.   Gross  has shown the effect of accumulation of deposits
in the combustion chamber of the engine on the POM emission.  A number of
             3-7
investigators    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's 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.
                      o
     In one procedure,  the participate matter and condensable vapors from
auto exhausts are collected by a total condensation trap.  In another proce-
     g
dure,  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 ROM's 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
ROM's from auto exhausts.  In this procedure, the hot exhaust aerosol is
precooled to 75-170°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
                                                             q
collected from automobile exhausts.  While some  investigators  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 investigator
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 in 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 process   and steps involving incorporation
of additives, such as, flavoring materials.  In addition,  substantial  amounts
of ROM's 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.    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.    In some methods, a cold-trap system replaces the eleclro-
                                                            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 incinerators, 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.    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 Implnger, 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 ROM's from spiked
samples by the above adsorbent sampler method is presented in Table  4-1.
         Table 4-1.   RECOVERY OF POM'S FROM ADSORBENT SAMPLER METHOD14

POM
Pyrene
Chrysene
Perylene
Benzo[ghi]perylene
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 yg/m ,
respectively, with adsorbent sampler and 4.2 and 56.5 yg/m  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's 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.    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 scfm.  The sample from the probe
enters three cyclones with a backup filter to give four sizing intervals of
MO ym, 10-3 ym, 3-1 ym, and less than 1 ym.  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 study   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.    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
           18
and Mitzner   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's at elevated stack temperatures.
                                             19
     A recent study conducted by Adams et al.    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's.  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's,  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's.  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—all these
factors arising from vehicular traffic 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-
                                                                        on
asphalt and one of concrete, both subjected to an identical traffic load.
                                                                                20
The POM emissions due to road abrasion are also dependent on seasonal  variation.
                                     4-10

-------
     For collecting road abrasion samples according to particle size,  an
                                                21
Andersen sampler has been used by Ciaccio et al.     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 ym 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 stages  are
weighed and placed again 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
                                                                        21
filter holder containing a flash-fired glass fiber filter has been used.
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

-------
                                                              22  23
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, 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-
      24
lings.    Occurence of seasonal  variation in specific surface areas  and
                                                             25
densities of suspended particulate matters has been observed.    The effects
of wind on the collection efficiencies of particulate matter  has  been demon-
                          26
strated by Ogden and Wood.    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 ROM's 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
                                                                         2
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
 3
m /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
                       27
given by Sholtes et al.
     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
organic 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.
     ROM's are present in polluted air sorbed onto airborne particulate
matter, usually characterized as primary and secondary.  Primary particulate
                                                29
matter is found in sizes between 1 urn and 20 \im.    Secondary particulate
matter ranges in size from molecular clusters of the order of 5 nm to particles
                                              29
with diameter as large as several micrometers.    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,   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
                                              31
similar method has been used by Fox and Staley   for collecting both particu-
late and gas phase components from atmospheric samples.
                     32
     Schuetzle et al.    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 pm in
diameter.  A properly conditioned Chromosorb-102  support is  used to  trap
gaseous air pollutants.
                                                                  33
     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 contained 4 ml of conditioned Tenax  GC held in  position by plugs
of silanized glass wool.
     The usefulness of the adsorbent in collecting ROM's not otherwise retained
on the glass-fiber filter is shown by Krstulovic  et al.    Table 4-2 shows that
the amount of ROM's collected on the adsorbent depends both  on the nature of
ROM's 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 ROM's are proven carcinogens.
                                    4-14

-------
                Table 4-2.  AMOUNTS OF POM'S COLLECTED ON  GLASS-FIBER  FILTER AND  POLYURETHANE  FOAM
                                                                                                   30
I
O1

Compound
Naphthalene
Biphenyl
Phenanthrene
Fluoranthene
Benzo[a]pyrene
1 ,2,3,4-dibenzanthracene
Location
Amt. on
filter 3
ng/1000 m
248.0
95.7
337.7
1249.8
29.7
806.4
A
Amt. on
adsorbent-,
ng/1000 m
100.7
8.5
5.6
281.3
N.D.
3709.2
Location
Amt. on
filter 3
ng/1000 m
31.1
11.3
46.6
N.D.
N.D.
102.5
B
Amt. on
adsorbent.,
ng/1000 m
27.9
N.D.a
4.9
165.4
3.5
N.D.
Location
Amt. on
filter 3
ng/1000 m
31.8
N.D.
4.9
159.4
4.2
N.D.
C
Amt. on
adsorbent.,
ng/1000 mj
49.1
N.D.
6.4
N.D.
N.D.
29.7
     'N.D.:  Not detected.

-------
There is considerable disagreement regarding the aero dynamics sizes which can
cause deposition of participate matter in the three compartments of the lung
(see Section 6).
                  34
     Whitby et al.   have presented evidence that the mass distribution of
atmospheric aerosols is usually bimodal, with one mode occurring below 1.0 ym
and the other mode occurring in the 5 to 15 pm range.  A dichotomous sampler
for particulates has been designed to collect and fractionate samples into two
            35
size ranges.    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
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's 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's 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) mechanical disruption,  and (3) HF acid  dissolution.
4.3.1.1  Solvent extraction—The principle behind solvent extraction  is  to
dissolve the collected ROM's 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
                                                        44.47          fi  28 47-
have been used for this purpose.  These include acetone,      benzene, '  '
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.  '    Potthast and  Eigner  used propylene  carbonate
as a solvent for extraction of ROM's from meat products.  They claim  that  the
ROM's 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
                                              47
reproducibility of extraction.  Liberti et al.   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.
                               49
     Cautreels and Cauwenberghe   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'S
                           WITH DIFFERENT SOLVENTS^

% extraction yield
Compound
Phenanthrene
Fluoranthene
Pyrene
Benzo[b]fluorene
Benz[a]anthracene
Benzo[a]pyrene
Di benzLa ,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
28
__a
a Below the detection limit.
     Although the yields of extraction of POM's were found to be 10 to 20
percent higher with methanol than with benzene, the latter solvent showed
more selectivity towards POM's 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's.  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
                                     4-18

-------
for BaP from glass-fiber filter samples.50
4.3.1.2  Mechanical disruption—The widely used Soxhlet  extraction  is  time
consuming, decomposes some of the pollutants,  and  lacks  good precision.     In
order to avoid these errors, a mechanical  disruptive  technique which makes
                                                                      fi?
use of a specially designed blender has been used  by  Bove  and Kukreja.    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
                      52
by Golden and Sawicki.    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
                                                  C"i
entire method has been evaluated by Sawicki et al.
     The recovery of ROM's 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

weight 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 ROM's 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.     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 trie 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, such as, gas chromatography.  This results in overall  reduced re-
covery and sensitivity.  Gas-phase extraction, however, is more efficient.
According to this method the ROM's 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.   used this method for the analysis of ROM's from air-
borne particulate matters.  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

-------
                                                     00
     A very similar method employed by Bertsch et al.    consisted of heat
desorption of organic compounds from a porous polymer (Tenax GC) phase instead
of a glass-fiber filter-trapped phase.  The desorbed  compounds were trapped on
Emulphor ON-870 cooled at dry ice or liquid N« 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
                                      18
column by way of a gas sampling valve.    This method is simpler than the
other two stripping methods since it eliminated the special  accessory equip-
ment 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 instantaneous, 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 b,/ Japanese researchers.  A more recent apparatus  for extracting
seven participate samples by one operation is given by Matsushita et al.
                                     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 participate 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.  ROM's
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 evaluated
with spiked samples and for three ROM's 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 air borne particulate samples.    However, studies
                           cq
conducted by Monkman et al.   contradicted these findings.
     The vacuum sublimation technique is faster than Soxhlet extraction and
is routinely used by Japanese researchers  '   for extraction of ROM's from
                                                                           CO
Hi-Vol glass-fiber filters.  It has also been recommended by Schultz et al.
as a routine method for a limited POM analysis.
4.4  CLEAN-UP AND SEPARATION
     The next step following the desorption of ROM's 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's  as  a
group from the bulk of impurities.  According to this method,  the POM's in a
certain solvent are allowed to distribute in another immiscible solvent which
has either higher or lower distribution coefficient for the POM's than  the
original solvent.  In the former instance, the POM's 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,   the POM's in cyclohexane solution are allowed to
partition in methanol-water solution.  The polar impurities in cyclohexane
solution preferentially distribute themselves in more polar methanol-water
phase, leaving the cyclohexane phase containing the POM's 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 authors  '   as a preliminary step for getting rid of the polar com-
ponents from the POM fraction.
4.4.1.2  Tetramethyl uric acid in methano!--A solvent system incorporating
tetramethyl-uric acid in methanol as a selective complexing agent has been
used to separate several ROM's.  Distribution coefficients for a number of
                                                           74
ROM's in this solvent system have been given by Mold et al.    However, this
solvent system lacks wide applicability.
                                    75
4.4.1.3  Acetonitri1e--Haenni et al.   determined the distribution of POM's
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's from heptane phase.    Consequently, this solvent has
                          9 73
been used by other authors '   for extracting POM's from cyclohexane phase as
well.
4.4.1.5  Nitromethane--The partition coefficients of POM's between nitro-
                                                             12
methane and cyclohexane was determined by Hoffmann and Wynder   and varied
between 4.4 and 1.65 for a number of POM's tested.  The same partition coeffi-
cients determined between nitromethane and other aliphatic solvents  '
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
                              53 55 71 77
of automobile exhaust samples.  •  '  '
4.4.1.6  Dimethyl sulfoxide--The first evaluation of this solvent as a parti-
tioning medium for POM's was made by Haenni et al.    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.    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 PROCEDURE78
Compound           % Recovery	          Compound       % Recovery

                 Nitromethane  DMSO                      Nitromethane  DMSO
Fl uoranthene 48-83
Pyrene 38-42
Benz[a]anthracene 50-64
& chrysene
Benzo[k]fl uoranthene 54-59
Benzo[ghi]perylene 15.8-45
92-96
94-96
90-92

91-94
90-91
Benzo[a]pyrene
& Benzo[e]pyrene
Perylene
Indeno[l,2,3-cd]-
pyrene
Coronene
33-36

< 10-24
41-49

10.0-44
95-99

84-90
97-100

82-93
                                             78
     From their investigations Acheson et al.   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


                                               55 77
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

-------
                                                       49
IN sodium hydroxide or HC1, Cautreels and Cauwenberghe,   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 ROM's.  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:  (!) 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

-------
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, such as, 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
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,   cyclohexane-ether,   cyclohexane-
        flfi                     ft?
benzene,   and benzene-methanol   are some of the  solvent systems used by
                                        46
various researchers.  Some investigators   have successfully separated PAH's
on very long alumina column.  However, such separation procedures need auto-
matic fraction collectors to collect a number of eluting fractions.
     The main disadvantage with 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
                                            28 71
found suitable for chromatographic purposes.  '    The sample:adsorbent ratio
varies between 1:50 and 1:500.  Some investigators used silica gel with 3  to
                           47 57                                       51  71
5 percent moisture content,  '   while others used the dried silica gel  *
                                    4-28

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

done by benzene  '  *   or hexane-benzene.

     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
                                             47
gel chromatography was used by Liberti et al.    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                                      68.8
Pyrene                                            82.0
1-2 Benzanthracene                                77.6
Chrysene                                          82.1
3-4 Benzofluoranthene                             82.4
Benzo[a]pyrene                                    80.5
Benzo[e]pyrene                                    80.5
Indeno[l,2,3-cd]pyrene                            90.6
Benzo[g,h,i]perylene                              90.3
     Average                                      77.2


     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  Flori si I—Column chromatography using Florisil is not as widespread

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

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

-------
           79 80
impurities.  '    The eluting solvent for this sorbent is  usually hexane-


       87                  80
benzene   or benzene alone.    The advantage with Florisil  column is  that



PAH's 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--A1though cellulose acetate is  a promising sorbent



for separation of ROM's, its use as a column chromatographic material is very


                       79
limited.  Griest et al.   have reported using this technique for the  purifica-



tion of PAH's.


                                    88
4.4.2.5  Gel filtration—Milk et al.   were the first to use this technique for



the separation of ROM's.  Others have since found the system to be valuable for


                                           14 89                        55 90
separation of PAH from automobile exhausts,  '   air particulate matter,  '


               73                        90 91
cigarette smoke   and biological samples.  '    Separation of neutral aza-


                                                92
arenes has also been achieved by this technique.    The separation is best



achieved isothermally on Sephadex LH-20 (ratio, about 1:1000) columns with 2-


                                                         89
propanol as solvent at a flow rate of 6-7 ml/min.  Gladen    determined the



recovery efficiencies of 8 PAH's by this technique which is shown in  Table 4-7.



                                                                      on

        Table 4-7.  RECOVERY OF PAH'S BY SEPHADEX LH-20 CHROMATOGRAPHYoy
Compound                                % Recovery
Anthracene                                 101

Phenanthrene                                98

Pyrene                                      94

Fluoranthene                               102

Benz[a]anthracene                          100

Benzo[a]pyrene                              97

Perylene                                    98

Benzo[g,h,i]perylene                        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 cellulose  *   and the Chromatography is carried out by ascending tech-
nique.  A slight modification of the paper by impregnating with 10 percent
                                                                  82
liquid paraffin has also been used for chromatographic separation.    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
        83
Sawicki.    However, this method has restrictive use for reasons similar to
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 ROM's 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
                                    102
including silinazation of silica gel     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
                                          95
been discussed in detail by Woidich et al.    Although alumina-cellulose
                                             98
(2:1) has been used as a composite adsorbent,   the best resolution is
obtained when two-dimensional TLC on a composite plate of aluminum oxide-
40 percent acetylated cellulose (2:1) is used.   '     The percent recovery
of a few isomeric arenes separated by this method    is given in Table 4-8.
    Table 4-8.  RECOVERY OF VARIOUS PAH'S DURING THIN-LAYER CHROMATOGRAPHY105

Compound
BeP
BaP
BbF
BkF
Perylene
Dibenzo[def,
mno] chrysene
% Recovery
86.
85.
92.
91.
90.
90.
0+5.
2+5.
1+3.
2+3.
4+4.
9+3.
3
1
5
7
6
9
Compound
B[ghi]p
Naphtha[l,2,3,4-def]
Chrysene
Benzo[rst]pentaphene
Di benzo[b ,def ]chrysene
Naptho[2,18-gra]naphtha-
cene
Dibenzo[def,b]chrysene
% Recovery
88.
87.
91.
89.
90.
84.
7+4.
9+5.
8+4.
8+3.
7+5.
1+6.
0
5
2
9
1
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.
                                                                   98
     Two-dimensional TLC separation has also been successfully used   during
the quantification of acridine, benz[c]acridine, 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
                      Concentration                         Concentration
                        in ma/                                in ma/
Compound              1000 m3 air       Compound            1000 m3 air

Acridine                 0.870          Indeno[l,2,3-ij]-      0.030
Benzo[f]quinoline        0.420            isoquinoline
Benzo[h]quinoline        0.260          11 H-Indeno[l,2-b]-    0.190
Phenanthridine           0.020            quinoline
Benz[a]acridine          0.200          Dibenz[a,h]acridine    0.010
Benz[c]acridine          0.120          Dibenz[a,j]acridine    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's, 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 structures among some of the ROM's 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, such as, 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-
metaphenoxylene,93 apiezon L,98'108 apiezon W,79 SE-30,86*108'109 SE-52,57'110
Dexsil 300 Gc....l. 113,114 Dexsil  40Q Gc>106
nematic liquid crystals   '   '    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-
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.      Dong et al.     have used
Dexsi 1-300 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'-bis(p-substituted
benzyl i dene) -a ,a'-bi-p-tolui dine have been  shown to resolve hard to separate
PAH isomers.  Three p-substituted compounds, methoxy (BMBT), hexyloxy (BHxBT),
and phenyl (BPhBT) have been investigated.    *   '      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-
                                         1 14
thermally at 130°C has been found useful.
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
                                                                             119
theoretical plates obtained from such columns  may vary between 20,000-70,000.
Versamid 900,102 SE-30,119 XE-60,119 SE-52,11 .53,58,72,119,120.123 ,.^125
                                  191            1 A                 OO
m-bis-m-(phenoxylphenoxy)-benzene,    Dexsil 300,   Emulphor ON  870,
carbowax-20 polymer,122 OV-1,124 OV-17,9 OV-101 ,9'73'91 apiezon  L,47'76 and
SP-2100    all have been used as packing materials for capillary columns.   The
                                                                         119
most widely used column for separation of ROM's is SE-52.   Cantuti  et al.
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.
                                 125
Recent investigations by Bjtfrseth    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
                                     125
separated with this packing material.     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

-------
                                 127
     Investigations by Lao et al.    with Dexsil-300 coating liquid has shown
that the separation of ROM's 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 ROM's carried out by wall-coated capillary columns coated
with SE-52 of liquid phase thickness of 0.27 y showed excellent reproducibil-
ity, high sensitivity and high resolution between individual  isomers in
                                                                       1  to
                                                                           128
                                     128
comparison to regular packed columns.     WCOT columns have been claimed to
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

-------
               Table  4-10.   HPLC  CONDITIONS  AND SEPARATION MECHANISMS  USED  FOR  PNA ANALYSIS
OJ
Nature of Sample
Standard PAH
solution
Carbon Black
Sediment &
Sea water
Automobile
exhaust
Standard PAH
solution
Standard PAH
metabolite &
solution
Standard PAH
solution
Standard PAH
solution
Automobile exhause
condensates
Standard PAH
solution
Standard PAH
solution
Stationary Phase
120/170 mesh Alumina
ODS-Sil-X-I
Coupled Bondapak C-\g-v Bondapak
C18
Porasil T
ODS-Permaphase
ODS-Permaphase
2,4,7-trinltrofluorenone
impregnated Coras 11 I
3(2,4,5.7)tetranitrofluore-
iminopropyldiethyl siloxane
ODS-Permaphase
Spheres il XOA-400
Coras il-C18
Mobile Phase
Mode of Elutlon
n-pentane
isocratic at 211°C
30% HZ0:70% HeOH
Isocratic at SO°C
Gradient elution
30% HeOH: 70% H-0 to
100% HeOH i
n-heptane
isocratic at 51.5°C
65% MeOH: 35% H.O
isocratic at room temp.
30% HeOH: 70% H?0 to
70% HeOH: 30% H20
gradient elution
n-heptane
isocratic at 25°C
Acetonitrile-Isooctane
Isocratic
Linear gradient
20% MeOH: 80% H.O to
100% HeOH i
Isooctane
isocratic
80% HeOH: 20% H-0
Isocratic
Separation
Mechanism
Adsorption
Reversed
phase
Reversed
phase
Adsorption
Reversed
phase
Reversed
phase
Charge-
transfer
Charge-
transfer
Reversed
phase
Adsorption
Reversed
phase
Detection
Mode
Disc conveyor
FID
U.V.
U.V.. Fluor.
U.V.
U.V.
U.V.
U.V.
U.V.
U.V.
Fluor.
Fluor.
Reference
129
130
131
132
133
134
l
135
136
137
1
138
139

-------
Table 4-10.  HPLC CONDITIONS AND SEPARATION MECHANISMS USED FOR PNA ANALYSIS (CONT'D)
Nature of Sample
Standard PAH
solution
Atmospheric parti-
cipate matters
Atmospheric parti-
culate matters
Standard PAH
solution
Standard PAH
•^ solution
10
00 Standard PAH
solution
Atmospheric
participate matters
Marine sediment
ft oil samples
Coal liquefaction
products
Stationary Phase
Octadecyl trichlorosilanized
silica gel
Zorbax-ODS
Zorbax-OOS
Equal portions of 8. IX dimethyl -
formamide and carbowax-350 on
gas chromatograph
30X Cellulose Monacetate
100/115 mesh Woe 1m Basic Alumina
u-Bondapak/Cjo S p-porasil
u-Bondapak/C18 i u-Bondapak NH?
Cross-linked polyvinyl-
pyrrolidone
Mobile Phase
Mode of Elution
90S MeOH.lOX H20
isocratic
70% MeOH:30X H.O
Isocratic
65% MeOH:35X H,0
to IX MeOH/minT gradient
Isooctane
isocratic
Ethanol/MeCl- (2:1)
Isocratic
Pressure programmed
95* n-pentane:5X MeOH
20-80% MeCN in water at
2X/min and IX propanol-2
in hexane
50-70X Acetomtrile in
water and hexane
Propanol-2
Separation
Mechanism
Reversed
phase
Reversed
phase
Reversed
phase
partition
Reversed
phase
Adsorption
Reversed
adsorption
Reversed
phase
Adsorption
Detection
Mode
Fluor.
Fluor.
Fluor.
U.V.
U.V.
U.V.
Fluor.
Fluor.
Fluor.
Reference
140
31
97
141
142
143
144
145
146

-------
     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 y-Bondapak C1P has been
                                      97                144
applied for the separation of both PAH   and aza arenes.
18
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's  the
usual method of detection is either flame ionization detection (FID)  '  '  '
Q1 TOR                                     77 Q3 1 flQ
  '    or electron capture detection (ECD).//»*J'IUO  One advantage with the
EC detector is its greater selectivity towards BaP compared with BeP.
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 flame ionization detector is  its wide dynamic range of linear
response for most POM's.
                                    4-39

-------
4.5.2  U.V. Absorption
                                                                     Q TO 9fi
     This technique is the most widely used method for POM detection. '*'
  '  *  "  '    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
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  Luminiscence 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.  '
fift 71 79 fl/L Ql QQ
oo,/i,/d,oH,»i,»s  Several groups have found it to be at least an order of
                                                      138 149
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.
     Since luminescence spectroscopy is nondestructive, individual  fractions
can easily be collected and subjected to further analysis.   Furthermore, this
method provides additional selectivity.  While many interferring 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-
                                                                          152
pounds with enhanced sensitivity has been utilized by Sawicki and Johnson.
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
rigid matrix-like backing such as filter paper, thereby quenching non-radia-
                               153
tive vibrational de-excitation,    and by the addition of a heavy atom per-
turber which increases the spin-orbit coupling between the singlet and the
                       154
triplet excited states.
     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.     By means of a selective excitation
and/or oxygen quenching method, POM compounds have been quantitatively analyzed
without chromatographic separation.     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.    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
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  ShpoTskii effect--Detection of PAH compounds by this technique was
                                158
first demonstrated by ShpoTskii    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
DeLima    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 ShpoTskii 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 of these cause error in the analysis.  The use  of combined internal
standard - standard additions technique has been demonstrated  to yield accept-
                             159
able results in such samples.     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 (XEOL)—This  method is  a slight
                                                            o
variation of ShpoTskii 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.    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 sens.itizer 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's 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
                                            162
termed synchronous luminescence spectroscopy    where the excitation and
emission wavelengths are simultaneously varied with a fixed wavelength differ-
                                              1 go
ence of 3 nm between the two has been proposed    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.31'33'59'73'112'
   51     Generally, this has been accomplished by low voltage ionization
methods which produce spectra containing mainly singly-changed, intact molecu-
lar ions.
     In POM analysis, the most widely used methods for sample introduction to
                                                3? 11? 1 *>1  1 fi^
the mass spectrometer are direct insertion probe  '   '   '    and direct
interfacing GC-MS method.106'115'121*127  The sample ionization is commonly
done by electron impaction and the separation of ion beams  is accomplished by
either low resolution quadrupole instrument or high resolution magnetic scan-
                            113
ning equipment.  The claimed    advantage of quadrupole 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
amplitude versus the spectrum number (ion abundance chromatogram).
                                     4-45

-------
Identification of these chromatographic peak is accomplished by plotting the
mass spectrum of a specified peak or by specific ion monitoring (mass frag-
mentography).
     One problem with electron impact ionization is that isomeric ROM's may
give identical mass spectra making unambiguous identification sometimes
                            165
impossible.  Hites and Dubay    found that charge-exchange-chemical  ionization
mass spectrometry can be used to distinguish many isomeric ROM's.  They have
established that spectra of ROM's have characteristic ratios of the  protonated
molecular ion to the molecular ion when 5% to 10% 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
                                       CC TO
     Fourier-transform NMR spectroscopy  '   has been used for the elucidation
of the position of substitution in the benzene ring positions of the ROM's.
Other techniques of detection utilizing matrix isolation Fourier-transform
IR,    and Piezoelectric crystal    need considerable improvement to become
viable methods for POM analysis.
4.6  ANALYTICAL METHOD USED BY EPA FOR NASN MONITORED SAMPLES:
     The following method    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 y 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 nn, 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-
                      174
ing modified procedure    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 yl  of the con-
centrate was spotted on a channelled 20% acetylated cellulose plate with a
multispotter.  Plates 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 a specific form of POM's, for example, a selected group of  PAH  has  been
decided to be monitored, the reported values are subject to substantial errors.
Such errors and uncertainties may arise during defining the monitoring stra-
tegy, 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 inertial characteristics of suspended particulates bring about losses of
larger particles both by drift from the sampled air stream and  by impact!on 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 efficiency—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.
4.7.1.3  Evaporative losses and reactive conversion—Collection of ROM's on
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
participate 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
              37                    24               38            39
Thomas et al.,   DeWeist and Rondia,   Murray et al.,   and Rondia.   Pupp et
   40
al.   determined the equilibrium vapor concentration  for a number  of POM's  and
concluded that considerable losses occur during  ambient sampling of compounds
                                                  3  3
having equilibrium vapor concentration of 0.5 mg/10 m  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's 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

-------
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 Katz   using simulated atmospheric conditions have demonstrated that ROM's
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 ROM's exist on the surface and in the interstices of
soot particles in a multi-layered form, the surface exposed ROM's on the
filter may be oxidized rapidly by the prevailing oxidants in the atmosphere
followed by a slower penetration reaction of the subsurface ROM's and ROMs
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 a layer of oxidized material may protect
them from further oxidation.  Quantitative data regarding this oxidative con-
version of ROM's during collection are not available.  There is also conflict-
ing evidence regarding the importance of this process.  For example, passing
clean air at a rate of 100 liters/min through filters containing airborne
                              41
particulates Matsushita et al.   found negligible losses of PAH's when air
treatment was less than 7 days.  But by passing air for 3 weeks through
                                                                168
particulate collected filters at a rate of 3 liters/min, Commins    reported
76 percent and 87 percent losses of fluoranthene and pyrene, respectively.  No
significant losses were observed for high molecular weight compounds including
BaP.  It should be mentioned that Sawicki pointed out that he failed to observe
                                     4-50

-------
                                                                      168
significant changes in pyrene concentrations under similar conditions.
Whether the lower temperature of the filter and the passed air was responsible
for the lack of loss was not made clear in his comments made in the above
reference.  It is also not clear from these experiments whether the losses  of
PAH are due to oxidative conversion, purely evaporative process or both.
4.7.2  Losses During Transportation and Storage
     Another source of error that needs to be properly addressed is the loss
of ROM's 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 ROM's 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.     Hermann   has noted considerable error  due
to degradation or reaction on storing the sample at room temperature and
exposing to air.  Similar results have been observed by other investigators
                          168
and 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 in access to air, the losses may be due to oxidation, sublimation, or
                                                                         42
both.  By storing the sample in glass bottles not exposed to air, Hermann
reported no significant loss of POM's.  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 TIME158
                                   % losses under different storage  times3
Compound
Fl uoranthene
Pyrene
Benzo[a]pyrene
Benzo[e]pyrene
Anthranthrene
Benzo[ghi ]peryl ene
Coronene
3 wks. at
room temp.
37
34
5k
+22b
+gb
9
""
1 yr. at
room temp.
92
88
32
23
21
10
1
? The samples were stored in sealed envelopes.
  These are the % gain (instead of loss)  observed.
                                    4-52

-------
4.7.3  Errors In Analytical Procedures
4.7.3.1  Losses during desorptlon process—Several sources of losses are
possible during analytical treatment of samples.   The first of these takes
place during desorption of ROM's 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's 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's either
                                                85
due to irreversible adsorption or photoreaction.    ,The photoreactivity of a
number of POM's on 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's in evapor-
ative concentration steps is another most frequently encountered source of
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's 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
alky] substituted products of the respective group of compounds.  Most packed
columns including the widely used Dexsil 300 GC will  not separate these com-
pounds.     Although the separation is better with Dexsil 400 GC columns, it
is by no means complete.     Columns packed with liquid nemaiic crystals have
been claimed to accomplish much better resolution.   '     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 Bj/6rseth failed to separate BbF
from BkF.     Lee et al.   used a 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, y-Bondapak C,g, provides a good separation
of alkyl substituted components.  In complex samples containing alkyl substi-
tuted as well as unsubstituted parent PAH's, 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
mobile phase even though these compounds consist of two, three, and four
                    145
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's 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's which will
measure the carcinogenic potential  of atmospheric POM's 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 ROM's in the event of
a process leak or other increase in 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
methods.  The reader is referred to a report by Hawthorne et al.    for a
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
determination, 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 tech-
niques is shown in Table 4-12.  All the data in this table are estimates with
                                            31 172
some values taken from available literature.  *
                                     4-56

-------
               Table 4-12.   COMPARATIVE  COST,  THROUGHPUT,  AND OTHER RELEVANT  DATA
                           FOR FOUR COMMONLY USED ANALYTICAL METHODS.


Applicability
Operational cost/sample
Sample turnaround
Samples/day
js. Sensitivity, analytical
en
"^ Capital and peripheral

TLC/fluor.
2 compounds
$11.25
1 day
8
1-2 ng
-412,800
Methods
Column chrom/U.V.
^12 compounds
-,$600
3 days
1/3
"x/1 yg
-v-20,000


GC-MS-Computer HPLC-fluor.
80-125 compounds
$800
80 min.
6
>10 pg
$150,350
10-15 compoi
<$13
65 min.
7
25-50 pg
-416,000
investments,  initial

-------
     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 ROM's 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 ROM's 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 ROM's 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 ym particles).
Some investigators have used additional back-up filters consisting of adsorp-
tive medium to collect ROM's present in the vapor phase and to prevent evapor-
ative losses of POM's.  The high-volume air samplers do not provide informa-
tion 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 ROM's from the widely used Soxhlet extraction for desorp-
tion of ROM's from the filtering/adsorbing medium is poor for some compounds
and is dependent on the solvent(s) used.   A method utilizing dissolution by
sonification yields better results.  Techniques which use sublimation and a
thermal method for desorption of ROM's 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 ROM's,
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 ROM's (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 ROM's usually collected from environmental
samples.
     Separation procedures involving solvent partitioning is generally used
for the isolation of ROM's 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 ROM's but also
for the resolution of the individual ROM's.  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 ROM's, 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 ROM's.  Gas chromatography offers a good method for
the separation of ROM's.  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, however, do not separate
isomeric compounds.  Even with nematic crystals, column bleed at higher temp-
eratures 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 ROM's,
although the separation of isomeric ROM's 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 ROM's 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.
     The widely accepted practice of monitoring benzo[a]pyrene alone, as an
indicator for other ROM's, is questionable.  More research effort should be
directed towards defining a few representative parameters  which would be
indicative of carcinogenic potential of ROM's present in the  environment.
                                     4-62

-------
                                 References


 1.  Spindt, R.  A.   Polynuclear Aromatic Content of Heavy Duty Diesel  Engine
     Exhaust Gases, NTIS PB-267774.   1977.

 2.  Gross, G.  P.   The effect of fuel  and vehicle variables on polynuclear
     aromatic hydrocarbons and phenol  emission.   Paper No.  720210.   Presented
     at the SAE  Automotive Engineering Congress  and Exposition,  Detroit,  Mi.
     1972, pp.  20.

 3.  Begeman, C. R., and J.  M. Colucci.   Polynuclear aromatic hydrocarbon
     emissions  from automotive engines.   SAE Trans.  7£:1682-1698,  1970.

 4.  Griffing,  M.E., A. R. Maler, J.  E.  Borland, and R.  R.  Decker.   Applying
     a new method  for measuring benzo[a]pyrene in vehicle exhaust  to the  study
     of fuel factors.  Presented at the Symposium on Current Approaches to
     Automotive  Emissions Control.   American Chem.  Soc., Los Angeles,  Calif.
     1971. pp.  13.

 5.  Candeli, A.,  V. Mastrandrea, G.  Morozzi, and S. Toccaceli.  Carcinogenic
     air pollutants in the exhaust from a European car operating on various
     fuels.  Atmos. Environ., 8_:693-705, 1974.

 6.  Rinehart,  W.  E., S. A.  Gendermalik, and L.  F.  Gilbert.  Fuel  factors in
     automotive  tailpipe emissions.   Presented at American Industrial  Hygiene
     Conference, Detroit, Michigan,  Paper No. 127,  1970. pp. 15.

 7.  Stichting  Concawe.  Effect of gasoline aromatic content on  polynuclear
     aromatic exhaust emissions.  Report 6-74.  Sept. 1974.  60  Van Hogen-
     honckleaan, The Hogue,  2018.

 8.  Begeman, C. R., and J.  M. Colucci.   Apparatus for determing the contribu-
     tion of the automobile to the benzene-soluble organic matter  in air.
     Natl. Cancer Instit.  Monograph 9:17-57, 1962.

 9.  Grimmer, V. G., A. Hildebrandt,  and H. Bohnke.   Sampling and  analysis of
     polycyclic  aromatic hydrocarbons in automobile exhaust gas.  Erdoel  Und
     Kohle, Erdgas, Petrochemie Vereinigt mit Brennstoff-Chemie, 25:442-447,
     531-536, 1972.                                             ~~

10.  Bentley, H. R., and J.  G. Burgan.  Polynuclear hydrocarbons in tobacco and
     tobacco smoke.  Part II.  The origin of 3:4-benzopyrene found in tobacco
     and tobacco smoke.  Analyst, 85_:723-727, 1960.

11.  Carugno, N.,  and S. Rossi.  Evaluation of polynuclear hydrocarbons  in
     cigarette smoke by glass capillary columns.  J. Gas Chromatogr.,  5:103-106,
     1967.


                                    4-63

-------
12.  Hoffman, D., and E. L.  Wynder.   Short-term determination  of  carcinogenic
     aromatic hydrocarbons.   Anal.  Chem.,  3^: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 of polycyclic organic compounds.  Environ. Sci. Technol.,
     1£(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.  Technol., 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. Technol., 8(10):935-942,  1974.

22.  Guidelines:  Air Quality Surviellance 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.,  10(6):
     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.
     Technol., 5_:155-158, 1971.

                                    4-64

-------
26.  Ogden, T. L., and J.  D.  Wood.   Effects of wind on  the  dust and benzene-
     soluble matter captured  by a small  sampler.   Ann.  Occup.  Hyg., 17:187-195,
     1975.

27.  Sholtes, R. S., R. B.  Engdahl,  R. A.  Herrick, C. Phillips, E.  Stein,
     J. Wagman, and P. R.  Wool rich.   Tentative method of analysis of suspended
     particulate matter in the atmosphere  (high volume  method).  Health  Lab. Sci.,
     1:279-286, 1970.

28.  Sawicki, E., R. C. Corey,  A. E.  Dooley, J. B. Gisclard, J.  L.  Monkman,
     R. E. Neligan, and L.  A. Ripperton.   Tentative method  of  analysis  for
     polynuclear aromatic  hydrocarbon content of atmospheric particulate matter.
     Hlth. Lab. Sci.,  7.(suppl. ):31-44,  1970.

29.  Fennelly, P. F.  Primary and secondary particulates as pollutants.  J. Air
     Pollut. Control Assoc.,  25_(7):697-704, 1975.

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

31.  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. Anal.  Chem., 48(7):992-998,
     1976.

32.  Schuetzle, D., A. L.  Crittenden, and  R. J. Charlson.  Application  of  compu-
     ter controlled high resolution  mass spectrometry to the analysis of air
     pollutants.  J. Air Pollut. Control  Assoc.,  23.(8):704-709, 1973.

33.  Bertsch, W., R. C. Chang,  and A. Zlatkis.  The determination of organic
     volatiles in air pollution studies:   Characterization  of  profiles.
     J. Chromatogr. Sci.,  ]_2(4): 175-182, 1974.

34.  Whitby, K. T., R. E.  Charlson,  W.  E.  Wilson,  and R. K. Stevens. The
     size of suspended particulate matter in air.   Science, 183:1098-1100, 1974.

35.  Dzubay, T. G., and R.  K. Stevens.   Ambient air analysis with dichotomous
     sampler and X-ray fluorescence  spectrometer.   Environ. Sci. Technol., 9_(7):
     663-668, 1975.

36.  Katz, M., and R.  C. Pierce.  Quantitative distribution of polynuclear
     aromatic hydrocarbons in relation to  particle size of  urban particulates.
     In;  Carcinogenesis,  Vol.  1. Polynuclear Aromatic Hydrocarbons:  Chemistry,
     Metabolism, and Carcinogenesis.   R.  I. Freudenthal and P. W. Jones  (eds.),
     New York, Raven Press.  1976, pp.  413-429.

37.  Thomas, J. F., M. Mukai, and B.  D.  Tebbens.   Fate  of airborne  benzo[a]pyrene.
     Environ. Sci. Technol.,  2:33-39, 1968.
                                    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.,
     521:557-563, 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,  AS406  and Se02
     and the collection efficiencies of these air  pollutants.  Atmos. Environ.,
     8:915-925, 1974.

41.  Matsushita, H., K. Arashidani, and T.  Handa.   Investigation on  a long
     term collection of polynuclear aromatic hydrocarbons  in environmental
     air.  Bunseki Kagaku, 2£(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.

<«3.  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
     participates 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.  Kertfisz-SaYinger, M., and Z. Mori in.  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.  Sawicki, E., R. C.  Corey,  A.  E.  Dooley,  J.  B.  Gisclard,  J.  L.  Monkman,
     R. E. Neligan, and  L.  A.  Ripperton.   Tentative method  of routine  analysis
     for polynuclear aromatic  hydrocarbon content of atmospheric participate
     matter.  Hlth. Lab. Sci.,  .7:45-55,  1970.

52.  Golden, C., and E.  Sawicki.   Ultrasonic  extraction  of  total  participate
     hydrocarbons from airborne particles at  room temperature.   Intl.  J.  Environ.
     Anal. Chem., 4:9-23, 1975.

53.  Bartle, K.  D., M. L. Lee,  and M.  Novotny.   High-resolution  GLC profiles  of
     urban air pollutant polynuclear  aromatic hydrocarbons.   Intl.  J.  Environ.
     Anal. Chem., 3_:349-356,  1974.

54.  Krstulovic, A., D.  M.  Rosie,  and P.  R. Brown.   Selective monitoring  of poly-
     nuclear aromatic hydrocarbons by high pressure liquid  chromatography with a
     variable wavelength detector. Anal. Chem., ]8:1383-1386, 1976.

55.  Novotny, M., M. L.  Lee,  and K. D.  Bartle.   The methods for  fractionation,
     analytical  separation  and  identification of polynuclear  aromatic  hydro-
     carbons in complex  mixture.   J.  Chromatogr. Sci., 1_2_:606-612,  1974.

56.  Lee, M. L., M. Novotny,  and K. D.  Bartle.   Gas chromatography/mass spectro-
     metric and nuclear  magnetic resonance determination of polynuclear aromatic
     hydrocarbons in airborne  particulates.   Anal.  Chem., 48_: 1566-1572, 1976.

57.  Chakraborty, B. B., and  R. Long.   Gas chromatographic  analysis of poly-
     cyclic aromatic hydrocarbons  in  soot samples.   Environ.  Sci. Technol.,
     1:828-834,  1967.

58.  Colucci, J. M., and C. R.  Begeman.   Polynuclear Aromatic Hydrocarbons and
     Other Pollutants in Los Angeles  Air.  Proc. Int. Clean Air  Congr., 2nd,
     1970 ed. by Englund, H.  M., Academic, N.Y., N.Y., 1971,  pp.  28-35.

59.  Cautreels,  W., and  K.  Van  Cauwenberghe.   Fast quantitative  analysis  of
     organic compounds in airborne particulate matter by gas  chromatography
     with selective mass spectrometric detection.  J. Chromatogr.,  131:253-264,
     1977.

60.  Potthast, K., and G. Eigner.   A  new method  for the  rapid isolation of
     polycyclic aromatic hydrocarbons from smoked meat products.  J. Chromatogr.,
     1_03:173-176, 1975.

61.  Sawicki, E.  Analysis  of  aerotoxicants,  CRC Critical Reviews in Analytical
     Chemistry,  Nov. 1970.

62.  Bove, J. L., and V. P. KuKreja.   Determination of free carbon  collected  on
     high-volume glass-fiber  filter.   Environ. Letters,  ]£(2):89-96, 1975.
                                    4-67

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

64.  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 ]J_(8&9):517-524,  1976-

65.  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 participates  by gas-liquid chromatography. J.
     Chromatogr., 99_:697-708, 1974.

66.  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.

67.  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.

68.  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.

69.  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.

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

71.  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, 14:158-160, 1970.

72.  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.

73.  Harke, H. P., D. Schliller,  H.  J. Klimisch, and K.  Meisner.  Investigations
     of polycyclic aromatic hydrocarbons in  cigarette smoke.   Z. Lebensm.
     Unters.-Forsch., 16^:291-297,  1976.

74.  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.  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.

76.  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,
     1975.

77.  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.

78.  Acheson, M. A., R. M. Harrison,  R.  Perry,  and R.  A.  Wei lings.   Factors
     affecting the extraction and analysis of polynuclear aromatic  hydrocarbons
     in water.  Water Research, 10:207-212,  1976.

79.  Griest, W. H., H. Kubota, and M. R.  Guerin.   Resolution of polynuclear
     aromatic hydrocarbons by packed  column  GLC.   Anal. Lett.,  8_(12):949-957,
     1975.

80.  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. Technol. (in press).

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

82.  Hluchan E., and M. Jenik.  Determination of airbtfrne polycyclic  hydro-
     carbons by paper chromatography.  J. Chromatogr., 91_:531-538,  1974.

83.  Sawicki, E., M. Guyer, and C. R. Engel. Paper and thin-layer  electro-
     phoretic separation of polynuclear aza-heterocyclic compounds.  J. Chromatogr.,
     30:522-527, 1967.

84.  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
     and polynuclear aromatic hydrocarbons.   Int.  J.  Air and Water  Pollut.,
     9:291-298, 1965.
                                    4-69

-------
86.  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
     method of analysis for polynuclear aromatic  hydrocarbons in automobile
     exhaust.  Hlth. Lab. Sci., Vh228-239, 1974.

87.  Brass, H. J., W. C. Elbert, M. A. Feige,  E.  M. Gliek, and A.  W. Lington.
     United States Steel, 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.

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

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

90.  Tomingas, R., G. Voltmer, and R.  Bednarik.   Direct fluorometric analysis
     of aromatic polycyclic hydrocarbons on thin-layer chromatograms.   The Sci.
     of the Total Environ., _7:261-267, 1977.

91.  Panalaks, T.  Determination and identification of polycyclic aromatic
     hydrocarbons in smoked and charcoal-broiled  food products by high  pressure
     liquid chromatography.  J. Environ. Sci.  Hlth.,  B_M_(4):299-315,  1976.

92.  Hoffmann, D., G. Rathkamp, and S. Nesnow.  Quantitative determination of
     9-methyl carbazoles in cigarette smoke.   Anal. Chem., 41_: 1256-1259,  1969.

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

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

95.  Woidich, H., W. Pfannhauser, G. Blaicher, and K. Tiefenbacher.  Thin-layer
     chromatographic separation and in situ detection of polycyclic  aromatic
     hydrocarbon:  Studies using acetylated cellulose. Chromatographia,  10:
     140-146, 1977.

96.  Hunter, L.  Quantitation of environmental hydrocarbons  by thin-layer
     chromatography:  Gravimetry/densitometry  comparison.  Environ.  Sci.  Technol.,
     9_:241-246, 1975.

97.  Dong, M., D. C. Locke, and E.  Ferrand.  High pressure liquid chromatographic
     method for routine analysis of major  parent  polycyclic  aromatic hydrocarbons
     in suspended participate matter.   Anal.  Chem., 48:368-372,  1976.
                                    4-70

-------
98.  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, 1J;619-629, 1966.

 99. Sawicki, E., R. C.  Corey,  A. E.  Dooley,  J.  B. Gisclard, J.  L.  Monkman,
     R. E. Neligan, and  L.  A.  Ripperton.   Tentative method  of  microanalysis  for
     benzo[a]pyrene in airborne particulates  and source effluents.   Hlth.  Lab.
     Sci., ^: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.

101. Bories, G.  Use of  polyamide for the separation  of different polycyclic
     aromatic hydrocarbons  by  thin-layer chromatography.  J. Chromatogr.,  130:
     387-389, 1977.

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., 112:681-700. 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.
     Gas 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-
     ultrayiolet 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-
     genesis, Vol. 1.  Polynuclear aromatic hydrocarbons:  chemistry, metabolism,
     and carcinogensis,   R. I. Freudenthal and P.  VI. 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,
     Technologic 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., 4^: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. U., D. C.  Locke, and D. Hoffmann.   Characterization of aza arenes
     in basic organic portion of suspended particulate  matter.   Environ. Sci.
     Technol., Vl_: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., J7^60-65, 1965.

120. Hites, R. A.  Presented at the Division of Petroleum Chemistry, Inc.,
     American Chemical  Society, Chicago,  Illinois, August 24-29, 1975,  pp. 824-828.

121. Oro, J., and J. Han.  Application  of combined chromatography-mass  spectro-
     metry to the analysis  of aromatic  hydrocarbons formed by pyrolysis of methane.
     J.  Gas Chromatogr., j>: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, 13(10):304-307,  1976.

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

125. Bjjirseth, A.  Analysis of  polycyclic  aromatic  hydrocarbons  in environmental
     samples by glass capillary gas  chromatography.   In:  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., I. 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.,
     45:908-915, 1973.

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. W. E., 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., Jj2:715-721,  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, ]84:169-171, 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 ir-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.,
     i:645-651, 1971.

138. Strubert, W.  Separation of fused ring aromatics  on silica  gel:  detection
     of nanogram or ppb amounts.  Chromatographia, £: 205-206, 1973.

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,  ]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.
     J. 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., ]j[: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., 129;
     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.  0. 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.,£:85-101,  1964.

153. Wei Ions, S.  L.,  R.  A.  Paynter, and  J. D. Winefordner.  Room temperature
     phosphorimetry of biologically important compounds  absorbed on filter
     paper.  Spectrochima Acta,  30A:2133-2140,  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 polycyclic  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. Dokl.  Akad. Nauk. S.S.S.R.,  87:935-938,
     1952 (in Russian).

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.,  i]_:55-60,  1974.

                                     4-75

-------
160. D'Silva, A.  P., G.  J.  Oestreich,  and V.  A.  Fassel.   X-ray  excited optical
     luminescence of polynuclear aromatic hydrocarbon.  Anal. Chem., 48:915-917,
     1976.                                                           ~~

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. Kites, 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, 1^: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. £, 1962, pp. 225-233.

169. Butler, J. D.  Air pollution,  smoking, and lung  cancer. Chem. Br., 11(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

-------
172. Lao, R. C., R.  S.  Thomas,  and J.  L.  Monkman.   In abstracts of  the first
     ORNL 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

-------
                                                                DRAFT
                                                          NOT QUOTE OR CITE
                             5.   AMBIENT  LEVELS

5.1  AMBIENT ATMOSPHERIC LEVELS
     A variety of polycyclic aromatic  hydrocarbons  (PAH's), aza arenes, and
carbonyl arenes 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's.  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's.  Also, there are several other  social or occupational sources of POM's
which may result in larger exposures to humans than those from ambient air.
For example, Bridbord et al.  has tabulated human exposure to PAH in terms of
BaP from a number of occupational and  other sources (Table 5-1).  This treat-
ment suggests that smoking one pack of cigarettes a day results in an appro-
priate exposure of BaP twenty times greater than that of a person exposed to
ambient air BaP levels (assumed 2 ng/m BaP).  The  exact amount of BaP intake
is difficult to determine because of the  different modes of inhalation during
                                            2                 1
smoking, working, or exposure to  ambient  air.   Bridbord et al.  recognize
"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
Employee in restaurant
BaP Exposure
(yg/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 data3
1962 report
NIOSH data*
NIOSH data3
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/m3
  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
                                                                O A
the winter months (Figure 5-1), probably due to heating sources.      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.   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.
     During 1966 through 1970, the benzene-soluble  organics (BSO) of quarterly
composite samples of suspended particulate matter collected on glass fiber
filters (99 percent efficient for particles of 0.3  ym) 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.   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.  Sawicki  found that benzo[ghi]perlene  (BghiP) concentra-
tions during the summers of 1958 to 1960 were higher than  BaP levels in the
                                    5-3

-------
       I   I  I  I   I   I   I  I
CO



 00
                                     I   I   II  I   I   I  I
                                         COMPOSITE AVERAGE
                                           URBAN STATIONS
LU
O
O

°-  1
«  I
CO
           COMPOSITE AVERAGE
           NONURBAN STATIONS
       12341234123412341234
           1966        1967        1968        1969      '1970
                        TIME, year and quarter
      Figure 5-1.   Composite quarterly averages for BaP at
                  32 urban and  19 nonurban NASN stations.
                            5-4

-------
Table 5-2.  ANNUAL AVERAGE AMBIENT BENZO[a]2YRENE CONCENTRATIONS
                     AT NASN URBAN STATIONS5*
                             (ng/m3)
Station
Alabama
Birmingham
Gad sen
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

18.5
3.5

6.5


2.3

1.7
0.6

1.2
1.1

2.5


2.1
2.7

1.8



1.7
1.1

2.3

2.3
3.5
1967



3.1

2.3

1.9

2.5
0.7

0.9
2.2


1.0
2.1
1.3
1.7





1.6
1.5

2.4

2.1
1.9
1968


2.4
2.7
4.2
2.9

1.7

2.1
0.7

0.9
2.2


1.6
2.1
1.8
1.6
0.9
2.3
1.3
1.4
1.0
1.2
1.8

2.3

1.4
1.4
1969


1.8
1.8
2.6
2.0

1.3

2.2
0.5

1.1
2.4

2.9
1.6
2.3
1.9
1.6
0.6

0.8
1.8
0.9
1.4
1.2

2.5

2.0
2.1
1970


2.5
1.6

1.3

0.8


0.4

0.7
0.6

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 BEN70[a]PYRENE CONCENTRATIONS
                         AT NASN URBAN.STATIONS5
                                 (ng/m3)
Station
Delaware
Newark
Wilmington
District of Columbia
Florida
Jacksonville
Tampa
Georgia
Atlanta
Hawaii
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

-------
Table 5-2 (Cont'd).   ANNUAL AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS
                         AT NASN URBAN STATIONS5
                                 (ng/m3)
Station
Kentucky
Ashland
Coving ton
Lexington
Louisville
Louisiana
New Orleans
Maine
Portland
Maryland
Baltimore
Massachusetts
Worchester
Michigan
Detroit
Flint
Grand Rapids
Trenton
Minnesota
Duluth
Minneapolis
Moorhead
St. Paul
Missouri
Kansas City
St. Louis
Montana
Helena
Nebraska
Omaha
1966

10.5
3.1

2.5

2.3



2.8



4.7




2.2
1.6
0.7
1.8






2.7
1967


1.9
1.8
2.1

1.8



3.8



5.4
1.4
2.8



1.3

2.3


2.3

0.8

1.3
1968

9.3
3.6
3.0
2.7

1.6

2.3

2.3

1.7

5.1
0.8
3.4
1.4

2.7
1.1
0.9
1.8

1.8


0.9

1.9
1969

10.9
4.1

1.9

1.5



2.8

1.5

3.9
1.7
1.7
1.6

2.1
1.4
1.0
1.8

1.6
3.3

0.5

1.6
1970

6.7
4.4
1.6


1.1

1.1

2.1

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 NASN  URBAN STATIONS5
                                   (ng/m3)
    Station                   1966       1967      1968      1969      1970
Nevada
 Las Vegas                     1.3       1.1       1.4
 Reno                                   4.6       3.1

New Hampshire
 Concord                       0.6       1.5       1.0       0.7       0.6

New Jersey
 Camden                        3.0                 1.6       2.4       1.9
 Glassboro                     0.7       0.8       1.2       1.1       1.2
 Jersey City                   4.2       3.5       2.3       2.7       4.7
 Marlton                       1.2       1.6       1.3                 1.4
 Newark                        2.1       3.3       2.1       1.8       1.5
 Patterson                              1.9       2.0       1.2       1.2
 Perth Amboy                   2.1       2.1       1.2       1.2       1.0
 Trenton                       2.2                 1.0       1.5       1.1

New Mexico
 Albuquerque                   2.0       1.9       1.8       1.1       1.1

New York
 New York City                 4.1       3.9                 3.6       3.0

North Carolina
 Charlotte                     5.7       6.3       5.6       4.9       1.9
 Durham                                           8.0       3.4       3.9

North Dakota
 Bismarck                                         0.9       1.0       0.4

Ohio
 Akron                         4.1       3.7       3.0
 Cincinnati                    3.6       1.9       1.8       2.9       2.6
 Cleveland                     3.1       2.9       3.0       3.8       2.8
 Columbus                      2.9       1.7       2.2       2.7       1.6
 Dayton                        2.7       3.7       2.4       1.9       1.5
 Toledo                        1.8       1.9       1.8       1.5       1.4
 Youngstown                    7.3       8.2       5.6       9.9       7.1
                                     5-8

-------
Table 5-2 (Cont'd).  ANNUAL AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS
                         AT NASN URBAN STATIONS5
                                 (ng/m3)
Station
Oklahoma
Oklahoma City
Tulsa
Oregon
Eugene
Medford
Portland
Pennsylvania
All en town
Al toona
Bethlehem
Harrisburg
Lancaster
Philadelphia
Pittsburgh
Reading
Scranton
Warminster
West Chester
Mi Ikes 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

1.8
29.5
2.9


5.9
7.0
2.9
5.2
2.2
1.1

1.8

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
18.0
2.1
1.3

2.9
6.3
2.4
6.1
0.9
1.0
1.6
1.9

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

1.9
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 BENZO[a]PYRENE  CONCENTRATIONS
                         AT NASN URBAN STATIONS5
                                 (ng/m3)
  Station                   1966      1967      1968      1969       1970
Texas
Dallas
Houston
San Antonio
Utah
Ogden
Salt Lake City
Vermont
Burlington
Virginia
Danville
Hampton
Lynchburg
Norfolk
Portsmouth
Richmond
Roanoke
Washington
Seattle
West Virginia
Charleston
Wisconsin
Kenosha
Madison
Milwaukee
Superior
Wyomi ng
Casper
Cheyenne

1.4
0.9
0.6

0.5
1.2

0.8

3.2


2.8




2.7

3.4



4.1



0.5



1.4


0.7




2.2
9.2
3.5
7.7
5.2
7.5

1.8













0.9

0.8
1.0

0.7

2.5
1.5
8.7
4.9
10.2

7.7

2.0

4.6

1.4
1.3
4.7
3.3

0.9
0.6

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

1.9
1.2
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 AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS
                    AT NASN NONURBAN STATIONS5
                             (ng/m3)

Station
Arizona
Grand Canyon
Maricopa County
Arkansas
Montgomery County
California
Humboldt Coutny
Idaho
Butte County
Indiana
Monroe County
Parke County
Maine
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
0.3 0.2 0.2
0.2 0.5
0.3 0.1 0.2
0.4 0.4 0.3
0.2
0.5 0.5
0.9 0.4
0.2 0.3
0.2 0.2
0.3 0.4
0.2 0.2
0.1 0.1
0.2 0.2 0.2
0.2 0.2
1969
0.2
0.3
0.2
0.5
0.1
0.3
0.3
0.1
0.2
0.4
0.1
0.1
0.1
0.3
1970
0.1
0.3
0.1
0.1
0.1
0.2
0.4
0.2
0.2

0.1
0.1
0.1
0.2
                              5-11

-------
Table 5-3 (cont'd).  ANNUAL AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS
                       AT NASN NONURBAN STATIONS5
                                (ng/tn3)

Station
North Carolina
Cape Hatteras
Oklahoma
Cherokee County
Oregon
Curry County
Pennsylvania
Clarion County
Texas
Matagorda County
Vermont
Orange County
Virginia
Shenandoah National Park
1966 1967
0.2
0.2 0.2
0.1 1.1
1.5 2.1
0.3 0.1
0.9
0.9 0.3
1968
0.2
0.2
0.1
1.0
0.2
0.3
0.3
1969
0.1
0.2
0.1
1.2
0.1
0.3
0.3
1970
0.2
0.2
0.1
1.2
0.3
0.2
0.2
                                 5-12

-------
Table 5-4.  ANNUAL AVERAGE AMBIENT BENZO[a]PYRENE CONCENTRATIONS AT NASN
                        URBAN STATIONS IN ng/m3.6


Birmingham, Ala.
Gadsden, Ala.a»b
Montgomery, Ala.a
Grand Canyon Nat. Park, Ariz.
Jacksonville, Fla.a
Honolulu, Hawaiia
Chicago, Ill.a»b
Gary, Ind.a»b
Hammond, Ind.a»b
Indianapolis, Ind.a»b
Terre Haute, Ind.a»b
Ashland, Ky.a>b
New Orleans, La.a
Baton Rouge, La.a
Arcadia Nat. Park, Maine
Baltimore, Md.a»b
Detroit, Mich.a»b
Dearborn, Mich.a»b
Trenton, Mich.a»b
Duluth, Minn.
St. Paul, Minn.a»b
St. Louis, Missouria>b
New York City, N.Y.a
Buffalo, N.Y.a»b
Cleveland, Ohipa»b
Toledo, Ohioa»b
Youngstown, Ohioa»b
Bethlehem, Pa.a>b
Erie, Pa.a>b
Philadelphia, Pa.a»b
Pittsburgh, Pa.a»b
1966
18.5
3.5
—
0.3
--
0.2
3.3
—
3.9
10.4
__
10.5
2.3
--
0.2
2.8
4.7
__
--
2.2
1.8

4.1
__
3.1
1.8
7.3
--
__
3.8
4.9
1967
..
—
2.3
0.2
--
0.5
3.0
—
2.5
5.7
3.7
—
1.8
—
__
3.8
5.4

--
__
2.3
2.3
3.9
__
2.9
1.9
8.2
2.9
__
5.9
7.0
1968
..
2.4
2.9
0.2
2.9
0.6
3.1
--
2.1
4.1
__
9.3
1.6
--
0.3
2.3
5.1

1.4
2.7
1.8
_ _
--
__
3.0
1.8
5.6
2.1
__
2.9
6.3
1969
_.
1.8
2.0
0.2
2.3
0.6
3.9
—
3.3
5.2
4.0
10.9
1.5
--
0.1
2.8
3.9

1.6
2.1
1.8
3.3
3.6
__
3.8
1.5
9.9
2.0
__
4.0
13.8
1970
„
2.5
1.3
0.1
1.4
0.2
2.0
--
1.7
2.3
2.8
6.7
1.1
--
0.2
2.1
2.6

0.8
1.1
1.0
__
3.0
__
2.8
1.4
7.1
2.7
__
2.4
5.9
1971
4.0
1.2
0.5
--
2.2
0.2
2.5
1.6
3.8
0.9
__
9.0
0.9
0.4
__
2.8
1.4

0.7
4.8
0.5
0.8
2.3
__
__
0.8
3.7
0.9
1.5
2.3
6.1
1972
2.3
1.2
0.5
--
0.4
0.1
1.3
1.2
1.4
4.9
1.1
8.5
0.4
0.2
0.3
1.3
1.9
0.6
0.5
19.1
0.5
0.6
1.8
1.5
1.3
0.4
3.2
0.8
2.4
0.9
10.6
1973
1.5
0.8
0.3
< 0.1
0.2
< 0. 1
0^4
0.3
0.2
0.4
__
2.9
0.3
0.1
—
0.4
1.0
1.0
0.1
0.3
0.1
0.2
0.7
0.6
__
0.4
1.1
0.5
0.7
0.6
—
1974
2.5
0.5
0.4
< 0.1
--
< 0 .1
--
0.5
0.4

0.3

0.3
0.1
0.1
0.5

1.7

0.2
0.5
0.3
0.9
0.8

0.2
1.9
0.1
0.6
0.8
1.3
1975
0.9
0.6
0.3
—
0.4
— —
1.0
--
0.7

0.6
4.7
0.2
0.1
0.1
0.6
1.0

--
_ —
— —
0.3
0.9
0.5

0.4
2.1
—
0.4
0.6
2.1
1976
1.6
0.6
0.3
< 0.1
0.4
< 0. 1
—
0.7
0.5
0.6
0.6
2.9
0.2
0.3
0.1
0.5
1.1

0.3
0.2
0.5
0.3
0.5
0.3
0.6
0.5
1.5
0.3
0.2
1.0
2.0

-------
                     Table  5-4 (Cont'd).   ANNUAL AVERAGE  AMBIENT BENZO[a]PYRENE CONCENTRATIONS AT
                                           NASN  IIRRAN  STATIONS  TN nn/m3  6
NASN URBAN STATIONS IN ng/m


Chattanooga, Tenn. '
Houston, Texasa»b
Newport News, Va.
Norfolk, Va.a
Shennandoah Nat. Park, Va.
Seattle, Wash.a
Spokane, Wash.a»b
Charleston, W.V.a
Milwaukee, Wisc.a»b
1966
8.4
0.9
—
2.8
0.9
2.7
--
3.4
4.1
1967
22.9
—
—
3.5
0.3
1.8

__
--
1968
7.4
—
--
4.9
0.3
2.0

4.6
4.7
1969
4.2
—
--
3.9
0.3
1.6

2.6
4.0
1970
5.5
1.2
--
1.8
0.2
1.5

2.1
2.5
1971
• «
0.5
0.4
1.2
__
0.5
1.7
0.9
1.8
1972
9.9
0.4
0.3
0.6
0.1
0.5
1.5
0.7
3.6
1973
— —
0.4
0.2
0.4
0.1
0.3
0.4
0.2
0.6
1974
— —
0.2
—
0.2
--
__
-_
0.5
--
1975
0.8
0.2
—
0.2
--
_ _
—
0.5
1.1
1976
0.7
0.1
—
0.5
--
__
—
0.4
0.4
    ? Urban site used  for  Figure 5-2.
Y1     Urban site where a coke oven is  located.

-------
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.   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's except coronene was found, and 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's in  Birmingham, Ala.,  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
                                                                           o
does not always occur.  For example, a study by Kertesz-Saringer and Mori in
found little or no relationship between BaP and other PAH's.  They analyzed
11 PAH's 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 CITIES5
                               (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.

9
0
33.0
18.
17.
7.
7.

5.
8.
9.
2.
3.
4.
2.
0
0
3
5

1
3
5
3
4
6
6
BaP

7.
25.
31.
5.
25.
4.
2.

1.
6.
6.
0.
1.
2.
0.

4
0
0
3
0
1
3

6
4
0
5
4
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

6.0
13.0
20.0
5.7
15.0
3.9
1.7

1.3
4.6
4.9
0.5
1.0
1.8
0.2
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

-------
                     Table 5-6.  ANNUAL AVERAGE CONCENTRATION OF PAH COMPOUNDSCIN THE
                           AIR OVER GREATER BIRMINGHAM, ALABAMA, 1964 and 19655

                                                  (ng/m3)

City
Bessemer
Birmingham
Fairfield
Irondale
Mt. Brook
Tarrant
Vestavia
Site
1
3
4
5
7
1
1
1
1
1
Fluor
7.0
4.9
11.2
10.8
2.6
10.0
3.4
1.0
3.4
1.0
P
7.6
4.6
10.8
9.1
2.5
8.1
2.8
1.0
3.6
1.0
BaA
7.8
5.3
21.2
14.5
3.4
13.3
4.2
1.0
3.9
1.0
Ch
13.1
8.1
27.9
14.2
4.4
11.3
5.7
2.2
7.6
2.0
BeP
10.5
7.6
26.1
15.0
5.6
13.3
6.3
3.0
7.6
2.9
BaP
13.5
9.0
35.8
20.5
6.0
18.2
7.6
2.6
7.4
2.4
Per
1.3
0.9
4.1
2.0
0.4
1.4
0.6
0.2
0.8
0.2
BghiP
14.1
9.5
22.4
15.3
7.9
11.8
7.0
3.7
8.2
3.5
A
1.2
0.7
2.2
1.2
0.3
1.1
0.4
0.1
0.2
0.1
Cor
2.6
2.7
3.8
3.5
2.7
2.1
1.9
1.4
2.2
1.2
Average                5.5        5.1       7.6      9.6     9.8     12.3      1.2      10.3      0.8     2.4

-------
"the concentration of 3,4-benzpyrene[BaP] is a poor indicator for other com-
pounds."
                     g
     Gordon and Bryan  found similar results while monitoring PAH's 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's 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. '  '    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
                    3                           3
from 3.2 nanograms/m  in 1966 to 2.1 nanograms/m  in 1970, approximately a
30 percent decrease.    A recent analysis   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
   8.0
            !i
                                                   T
                                              T
               DASHED LINE • PERCENTILES OF QUARTERLY
                          MEASUREMENTS

                         - 4 QUARTER MOVING AVERAGE
                          OF PERCENTILE VALUES
   6.0
ee
   4.0
   2.0
                                          QUARTERLY MEASUREMENTS
       ~~     50™ PERCEWTILb
        OF QUARTERLY MEASUREMENTS
              (LOWER CURVES)
            66
67
68
69
70
71
72
73
74
75
                                     YEAR
          Figure 5-2  Benzo[a]pyrene -  seasonality and trends
                      (1966 to 1975) in  the  50th and 90th
                      percent!les for 34 NASN urban sites9
                                   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 parellel.   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-
                                                   3
ual annual average concentration of 3.2 nanograms/m  in 1966  to 0.5  nano-
       •3
grams/m  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-
ings in and around the central business districts of  urban areas.   Also
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 pm)  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.
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
                                                    12 13 14
BaP is primarily associated with particulate matter,  '  '   with a sizeable
percentage (up to 60 percent) in the less than one pm range.  '    How much
BaP might pass through the 0.3 ym glass fiber filter is unknown although the
efficient of collection should increase as the filter is loaded with particles,
Also, DeWiest and Rondia   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

-------
                                                               1 Q
     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.
            1 p
     Commins   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,
                              18
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's have also been noted in Great Britian.
      20
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
                      21                       22
effort.  Brocco et al.   and Dong and coworkers   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
                                   22
York City monitoring of Dong et al.   are compatible  with previous  data.
                                         21
However, the results of the Brocco et al.   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
                            22                  21
were used.  Both Dong et al.   and Brocco et al.   collected their samples  on
                                                   21                    22
glass fiber filters and then extracted with benzene   or benzene/methanol
                     21
(4:1).  Brocco et al.   cleaned up their sample by TLC and analyzed by gas
                            22
chromatography.  Dong et al.   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's which might result in POM contamination of areas that are down wind from
                                    5-23

-------
                                           25
large emission sources.  Lunde and Bj0rseth   monitored air samples  in Norway
which had different trajectories.  They identified 20 different PAH's  (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.
     Table 5-8 tabulates the various compounds for which monitoring  concentra-
tions have been reported.  As many as one hundred compounds have been  found
                     3 23
in urban atmospheres, '   but they are not listed in Table 5-8 because quanti-
tative information could not be found.
     Table 5-8 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 5-8 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 /
Methyl -phenanthrene/-
anthracene
Fluoranthene
Di hydrobenzo[a&b]f 1 uor-
enes
Pyrene
Benzo[a]fluorene
Benzo[b]fluorene
1 -Methyl pyrene
Benzo[c]phenanthrene
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
Total Identified PAH 32.099
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
28.252
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
--
1.108
4
Feb. 1
1976

Stationary
air
Southern
Norway
0.146

0.052
0.324

0.032
0.286
0.026
0.148
0.009
0.108
0.073
0.194
0.464
0.135
0.098
0.011
0.144
0.140
0.022
0.020
2.435
                              5-25

-------
                                 Table 5-8.  POLYCYCLIC ORGANIC MATTER  IN  AMBIENT AIR
ro
CT>

Name Concentration,
ug/1000 m3
(unless other-
wise Indicated)
Naphthalene 0.052-.350
Anthracene >1.9

0.1-2.2

0.0-0.4

0.068-0.278

6.15
Benz[a]anthracene 0.4-21.6

0.1-16.0

^4

0.74-0.041

11-14
2

1.2-8.0
4.6
102
8.2
0.5-2.8
0.18
0.011 ug/g parti -
culate
0.022 yg/g parti -
1.1-3.1
0.1-13.1
Location Publication
Date
Providence area
Near industry handling
coal tar pitch
Winter - seven cities
monthly composite
Summer - seven cities
monthly composite
Norway - air from
N. England
Budapest
3 sites in Detroit
(20 samples)
4 sites in New York
(47 samples)
Average U.S. urban
atmosphere
Norway - air from
N. England
Average urban air-Detroit
Average suburban air -
Detroit
Rome
College Park, Md.
Baltimore Harbor Tunnel
Budapest
Los Angeles
Los Angeles
Summer - Toronto

Winter - Toronto
5 samples - Rome, Italy
14 samples - Rome, Italy
1977
1962

1962

1962

1977

1975
1965

1966

1965

1977

1965
1965

1975
1976
1976
1975
1971
1976
1975

1975
1966
1972
Reference
32
4

5

4

26

8
24

24

24

26

27
27

28
29
29
8
30
31
16

16
25
25

-------
                          Table 5-8 (Cont'd).  POLYCYCLIC  ORGANIC  MATTER  IN AMBIENT AIR
ro

Name
Di benz[a ,c]anthracene
Methyl phenanthrene/-anthracene

Phenanthrene






Benzo[c]phenanthrene


Benzo[a]fluorene

Benzo[b]fluorene


Di hydrobenzo[a&b]f 1 uorenes

Fluoranthene





Concentration,
yg/1000 m3
(unless other-
wise indicated)
4.5-0.029
0.216-0.661

>7.5

0.036-1.2

2.9-25
1.0
0.011-0.340
1.021-0.038

0.9
0.815-0.021

0.571-0.117

1.1
0.874-0.032

0.19-15.0

*4

6.6-0.17

Location
Providence area
Norway - air from
N. England
Near industry handling
coal tar pitch
Norway - air from
N. England
Rome
Budapest
Providence area
Norway - air from
N. England
Rome
Norway - air from
N. England
Norway - air from
N. England
Rome
Norway - air from
N. England
3 sites in Detroit
(6 samples)
Average U.S. urban
atmosphere
Norway - air from
N. England
Publication
Date
1977
1977

1962

1977

1975
1975
1977
1977

1975
1977

1977

1975
1977

1965

1963

1977

Reference
33
26

4

26

28
8
33
26

28
26

26

28
26

24

24

26


-------
Table 5-8 (Cont'd).  POLYCYCLIC ORGANIC MATTER IN AMBIENT AIR

Name Concentration, Location Publication
yg/1000 m3 Date
(unless other-
wise indicated)
Fluoranthene (Cont'd) 3.1-11.0
4.1
93
1.04
0.31
1.5-0.16
0.3-10.6

0.6-41

0.06-2.6

V1 1.0-18.0
fv» ? 1_4 R
00 *" ' H- J
0.1-3.4
Benzo[b]fl uoranthene 2.3-7.4

0.54
0.1-1.6
Benzo[j]fl uoranthene 0.8-4.4

0.17
0.01-0.8
Benzo[k]fl uoranthene 1.1-15.7

1.7-20

0.2-4.9

0.2
0.12-0.96
Rome
College Park, Md.
Baltimore Harbor Tunnel
Budapest
Los Angeles
Providence area
20 sites - Ontario
(5 samples each location)
20 sites - Ontario
(1962 inversion period)
1 site - Sidney, Australia
(12 samples)
14 samples - Rome
5 samples - Rome
4 sites - Los Angeles
3 sites in Detroit
(6 samples)
Los Angeles
4 sites - Los Angeles
3 sites in Detroit
(6 samples)
Los Angeles
4 sites - Los Angeles
2 sites in Detroit
(6 samples)
7 cities, winter, monthly
composites
7 cities, summer, monthly
composites
Los Angeles
Toronto
1975
1976
1976
1975
1976
1977
1966

1966

1965

1972
1966
1973
1965

1976
1973
1965

1976
1973
1965

1962

1962

1976
1975
Reference
28
29
29
8
31
33
25

25

25

25
25
25
24

31
9
24

31
9
24

5

5

31
16

-------
                         Table 5-8 (Cont'd).   POLYCYCLIC ORGANIC  MATTER  IN AMBIENT AIR

Name Concentration, Location
pg/1000 m3
(unless other-
wise Indicated)
Benzo[k]fluoranthene 1.3-14.0
(Cont'd) 0.03-1.3
20 sites in Ontario
4 sites in Los Angeles
(4 samples)
Publication
Date
1966
1973
Refarer.ce
25
9
   Benzo[b&k]f1uoranthene


   Benzo[ghi]fluoranthene

   Pyrene
en
i
ro
   1-Methylpyrene
4.312-0.083


0.9-9.1

1.3-19.3

Trace-35
10

4.86-0.135

2.2-6.8
5.2
120
2.06
0.45
0.18-3.8
0.08-4.0
0.4-12.0

0.4-17.0
2.4-5.1

0.147-.009
Norway - air from           1977
  N. England

Rome                        1975

3 sites in Detroit          1965
  (6 samples)
12 U.S. cities              1962
Near industry handling      1962
  coal tar pitch
Norway - air from           1977
  N. England
Rome                        1975
College Park, Md.           1976
Baltimore Harbor Tunnel      1976
Budapest                    1975
Los Angeles                 1976
4 sites in Los Angeles      1973
1 site in Sidney, Australia 1965
20 sites (5 samples at each 1966
  site) in Ontario
14 samples - Rome           1972
5 samples - Rome            1966
Norway - air from
  N. England
1977
                     22
28

24

24
 4

26

28
29
29
 8
31
 9
25
25

25
25

26

-------
                           Table 5-8 (Cont'd).  POLYCYCLIC ORGANIC MATTER IN AMBIENT AIR
en
CO
O

Name Concentration,
yg/1000 m3
(unless other-
wise indicated)
Benzo[a]pyrene 0.01-75
0.73

2.3-31

0.3-6.4

6
2.053-0.059

6-7
1

2-37
1.0-11.0
3.2
66
2.68
0.31-2.1
0.5-1.0
0.46
0.11-0.85
0.6-8.2
20-39
Benzo[e]pyrene 1-25

5.0

12-26
2.635-0.066

2-25
Location Pub
Urban and non-urban areas
Near industry handling
coal tar pitch
Winter, 7 cities, monthly
composite
Summer, 7 cities, monthly
compos i te
"Average" urban air
Norway - air from
N. England
Average urban air-Detroit
Average suburban air -
Detroit
4 cities in Europe
Rome
College Park, Md.
Baltimore Harbor Tunnel
Budapest
Los Angeles
London
Los Angeles
Toronto
1 site - Sidney, Australia
London street
12 U.S. cities (est. from
data)
Average U.S. urban
atmosphere
London street
Norway - air from
N. England
4 cities in Europe
lication
Date
1962
1962

1962

1962

1967
1977

1965
1965

1967
1975
1976
1976
1975
1971
1976
1976
1975
1965
1965
1962

1963

1965
1977

1967
Reference
4
4

5

5

23
26

27
27

32
28
29
29
8
30
20
31
16
25
25
24

24

24
26

32

-------
                          Table 5-8 (Cont'd).  POLYCYCLIC ORGANIC MATTER IN AMBIENT AIR
in

Name Concentration,
pg/1000 nH
(unless other-
wise indicated)
Berizo[e]pyrene (Cont'd) 2.6-15.0
4.6
69
1.52
0.52-1.2
0.90
0.3-11

4.3-42

8

Dibenzo[cd,jk]pyrene 1.1
(anthanthrene) Trace - 3

0.26

2-6
0.11

0.423-0.007

0.1-1.3
0.23
1.8 ug/g parti-
Location Publication
Date
Rome
College Park, Md.
Baltimore Harbor Tunnel
Budapest
London
Los Angeles
20 sites - Ontario
(5 samples each location)
20 sites - Ontario
(during inversion)
Copenhagen (mean annual
value)
Detroit
12 U.S. cities (est. from
data)
Average U.S. urban
atmosphere
London street
Near industry handling
coal tar pitch
Norway - air from
N. England
Rome
Los Angeles
Summer - Toronto
1975
1976
1976
1975
1976
1976
1966

1966

1966

1965
1962

1965

1965
1962

1977

1975
1976
1975
Reference
28
29
29
8
20
31
25

25

25

24
24

24

24
4

26

28
31
16
    Dibenzo[a,e]pyrene
   culate
2.6 ug/g parti -
   culate

100-730 yg/g
  benzene soluble
  fraction
                                                      Winter - Toronto
                            1975
Birmingham, Ala. composite  1968
16
34

-------
                          Table 5-8 (Cont'd).  POLYCYCLIC ORGANIC MATTER IN AMBIENT AIR
OJ
PO

Name Concentration,
M9/1000 m3
(unless other-
wise Indicated)
Indeno[l ,2,3-cd]pyrene 1.5-8.2

1.9-0.062

3.8-12.8
1.34
0.03-1.2
Chrysene 1.3-11.6

Present
3.4-15.0
4.8
106
7.15
0.60
11.6 pg/g parti -
culate
17.5 ug/g parti -
culate
0.2-6.65

0.4-39.0
1.3-4.1
Perylene Trace - 5.0

0.7

0.15

0.191 - trace
Location Publication
Date
3 sites in Detroit
(6 samples)
Norway - air from
N. England
Rome
Los Angeles
4 sites - Los Angeles
3 sites in Detroit
(6 samples)
6 U.S. cities
Rome
College Park, Md.
Baltimore Harbor Tunnel
Budapest
Los Angeles
Summer - Toronto

Winter - Toronto

1 site - Sidney, Australia
(12 samples)
14 samples - Rome, Italy
5 samples - Rome, Italy
12 U.S. cities (est. from
data)
Average U.S. urban
atmosphere
Near industry handling
coal tar pitch
Norway - air from
1965

1977

1975
1976
1973
1965

1965
1975
1976
1976
1975
1976
1975

1975

1965

1972
1966
1962

1965

1962

1977
Reference
24

26

28
31
9
24

24
28
29
29
8
31
16

16

25

25
25
24

24

4

26
                                                        N. England

-------
                          Table 5-8 (Cont'd).   POLYCYCLIC ORGANIC MATTER IN AMBIENT  AIR

Name



Perylene (Cont'd)




Benzo[ghi]perylene


Concentration,
yg/1000 m3
(unless other-
wise Indicated)
0.2-1.9
1.1
0.10
0.034-0.306
0.3-0.6
0.01-1.2
2.3-11.5

2-35
Location



Rome
Budapest
Los Angeles
Toronto
4 samples - Rome
4 sites - Los Angeles
3 sites in Detroit
(6 samples)
12 U.S. cities (est.
Publication
Date


1975
1975
1976
1975
1966
1973
1965

from 1962
Reference



28
8
31
16
25
9
24

24
00
CO
8

12-46
0.55

20 yg/g parti-
   culate
30 yg/g parti-
   culate
1.971-0.064
                                  3-26
                                  3.5-21
                                  3.9
                                  85
                                  0.86
                                  0.7-3.2
                                  3.27
                                  1.0-31

                                  1000
       1
  data)
Average U.S. urban          1965
  atmosphere
London street               1965
Near industry handling      1962
  coal tar pitch
Summer - Toronto            1975

Winter - Toronto            1975

Norway - air from           1977
  N. England
4 cities in Europe          1967
Rome                        1975
College Park, Md.           1976
Baltimore Harbor Tunnel     1976
Budapest                    1975
London                      1976
Los Angeles                 1976
2 sites - London            1967
  (1954-1964)
1 site - London             1967
  (inversion - 1957)
24

24
 4

16

16

26

32
28
29
29
 8
20
31
25

25

-------
                            Table 5-8 (Cont'd).   POLYCYCLIC  ORGANIC MATTER  IN  AMBIENT AIR
CO

Name
Benzo[ghi]pery1ene (Cont'd)





Coronene



















Acrid ine
Methyl acri dines
Concentration,
yg/1000 m3
(unless other-
wise indicated)
6.3-46
45

0.9-9.7
1.4-3.6
0.2-9.2
2.3-12.2
Trace - 8

2

4-20
0.05

0.212-0

1-6
4.4-18.3
1.05
0.3-1.9
2.13
12.3
8.2
3-48
0.2-6.4
0.7-1.0
0.040-0.041
ND - 0.007
Location Publication
Date
20 sites - Ontario
(inversion - 1962)
Liverpool, England mean
value
1 site - Sidney, Australia
4 samples - Rome
4 sites - Los Angeles
2 sites in Detroit
12 U.S. cities (est. from
data)
Average U.S. urban
atmosphere
London street
Near industry handling
coal tar pitch
Norway - air from
N. England
4 cities in Europe
Rome
Budapest
London
Los Angeles
Winter - Toronto
Summer - Toronto
20 sites - Ontario
4 sites - Los Angeles
4 samples - Rome
New York City air
New York City air
1966
1966

1965
1966
1973
1965
1962

1965

1965
1962

1977

1967
1975
1975
1976
1976
1975
1975
1966
1973
1966
1977
1977
Reference
25
25

25
25
9
24
24

24

24
4

26

32
28
8
20
31
16
16
25
9
25
22
22

-------
                          Table  5-8  (Cont'd).   POLYCYCLIC ORGANIC MATTER IN AMBIENT AIR
in
i
10
en

Name
Benz[a]acridine
Benz[a]acridine
Benz[c]acridine
Dibenz[a,j]acridine
Dibenz[a,h]acridine
Carbazole
Quinoline
Methylquinoline
2,6-Dimethylquinoline
Di methyl qui no! ines
Ethyl quinolines
3C-quinolines
Concentration,
pg/1000 m3
(unless other-
wise indicated)
0.2
3-80 (ug/g benzene
soluble fraction)
1
0.6
1-30 ug/g BSF
0.1-1.5
0.04
0-8 yg/g BSF
0.1
0.08
1.9
1.1
0.022-0.069
0.6
0.033-0.035
0.3
0.044-0.088
0.014-0.022
ND-0.010
Location Publication
Date
Urban air
6 cities urban air
"Average" urban air
Urban air
Urban air 6 cities
50 samples from different
U.S. cities
Urban air
Urban air
"Average" urban air
Urban air
Near industry handling
coal tar pitch
Rome
New York City air
Rome
New York City air
Rome
New York City air
New York City air
New York City air
1977
1967
1965
1968
1965
1967
1965
1962
1973
1977
1973
1977
1973
1977
1977
1977
Reference
24
5
23
24
5
36
24
5
23
24
4
21
22
21
22
21
22
22
22

-------
                           Table  5-8 (Cont'd).  POLYCYCLIC ORGANIC MATTER  IN AMBIENT  AIR
at

Name
Benzo[f]quinoline
Benzo[h]qu1no11ne
n-Indeno[l ,2b]quinoline
Phenanthridine
Isoquinoline
5 or 8 Methyl isoqu incline
Other Methyl isoqu inclines
Di methyl i soqu i nol i nes
Ethyl isoquinolines
3C-isoquinolines
Benzo[f ] i soqui nol i ne
4-Azafluorene
4-Azapyrene (other isomers)
Concentration,
pg/1000 m3
(unless other-
wise indicated)
0.2
0-200 (ug/g benzene
soluble fraction)
0.010-0.011
0.3
1-30 (Mg/g BSF)
0.010-0.013
0.1
4-30 (pg/g BSF)
Trace
0.018-0.022
0.140-0.180
0.170-0.310
0.070-0.076
ND-0.062
0.068-0.160
ND-0.028
0.034-0.110
0.005
0.021
13.1
Location
Urban air
Urban air (6 cities)
New York City air
Urban air
Urban air (6 cities)
New York City air
Urban air
Urban air (6 cities)
New York City air
New York City air
New York City air
New York City air
New York City air
New York City air
New York City air
New York City air
New York City air
New York City air
New York City air
Rome
Publication
Date
1965
1977
1965
1977
1965
1977
1977
1977
1977
1977
1977
1977
1977
1977
1977
1977
1973
Reference
24
5
22
24
5
22
24
5
22
22
22
22
22
22
22
22
22
22
22
21

-------
                          Table 5-8 (Cont'd).  POLYCYCLIC ORGANIC MATTER IN AMBIENT AIR
GO

Name
1-Azafluoranthane

Benzo[c]cinnoline
2-Methylindole
Phenalen-1-one

7-H-benz[d ,e]anthracen-7-one
(benzanthrone)







Perinaphthanone


Concentration,
yg/1000 m3
(unless other-
wise indicated)
0.005 (other isomers
3.0
1.0
2.0
2
0.3-17
0-1.01
0.57-15.52
30.5 yg/g parti -
culate
39.1 gg/g parti -
culate
2-48
0.9-13

26.5 ug/g parti -
culate
15.9 yg/g parti -
Location Publication
Date
) New York City air
Rome
Rome
Rome
"Average" urban air
Urban site
Non-urban air
Urban air
Summer - Toronto

Winter - Toronto

Urban air
50 samples from different
U.S. cities
Summer - Toronto

Winter - Toronto
1977
1973
1973
1973
1967
1969
1969
1969
1975

1975

1969
1968

1975

1975
Reference
22
21
21
21
23
35
13
13
16

16

35
36

16

16
                                     culate

-------
5.2  SUMMARY AND CONCLUSIONS
     Numerous POM compounds have been detected in ambient air (Table  5-8)  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
                                               3                   3
annual median value for urban sites of 3.2 ng/m  in 1966 to 0.5 ng/m   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  were 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 pm (could pass through the glass fiber filter) is
unknown.  Oxidative loss of BaP during collection is another unknown  and BaP
                                    5-38

-------
degradation may occur between the time when the sample is collected and
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's while others have reported little or no relationship.
Correlations between BaP and other POM's probably vary in different areas and
are dependent upon the local POM emission sources which will provide different
ratios of individual POM's.  Because the ratio of many of the individual POM's
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's.
     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/m  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
                                    5-39

-------
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's  associ-
ated 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 ROM's 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.
                                    5-40

-------
                                 REFERENCES


1.   Bridbord, K., J. F. Finklea, J.  K.  Wagoner,  0.  B.  Moran,  and P.  Cap!an.
     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.  Carolina.   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.  p.  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. Asscc  J. 23_(2):137-143.  1962.

8.   Kertesz-Saringer, M., and Z. Mori in.  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., 1(1):33-39, 1968.
                                     5-41

-------
13.  National Academy of Sciences,  Committee on  Biological  Effects of Atmos-
     pheric Pollutants.  Participate Polycyclic  Organic  Matter.  Washington,
     D.C., 1972, pp. 361.

14.  Natusch, D. F. S., and B.  A.  Totnkins.   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.

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

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

17.  DeWiest, F., and D. Rondia.   On the validity  of determinations  of  benzo[a]
     pyrene in airborne particles  in the summer  months.  Atmos. Environ., 6:
     487-489, 1976.

18.  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.

19.  Lane, D. A., and M. Katz.   The photomodification of benzo[a]pyrene, benzo[b]-
     fluoranthene, and benzo[k]fluoranthene under  simulated atmospheric conditions.
     In:  Fate of Pollutants in the Air and Water  Environments Part  2.
     I. A. Suffet (ed.), New York,  Wiley-Interscience.   1977.

20.  Leahey, D. M.  Changing pattern in concentrations of polycyclic aromatic
     hydrocarbons in the air of central London.  Atmos.  Environ., 10:561-562,
     1976.                                                       ~~

21.  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.

22.  Dong, M. W., D. C. Locke,  and D. Hoffmann.  Characterization of aza-arenes
     in basic organic portion of suspended particulate matter.  Environ. Sci.
     Technol., jj_:612-618, 1977.

23.  Sawicki, E.  Airborne carcinogens  and allied  compounds.  Arch Environ.
     Health, 14:46-53, 1967.
                                     5-42

-------
24.  Hoffman, D., and E.  L.  Wynder.   Chemical  analysis  and  carcinogenic  bio-
     assays of organic participate pollutants.   l]n:   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 carcinogenicity.   In:   Air  Pollution, Vol.  II,
     3rd Ed., A. C. Stern (ed.), Academic Press,  New York,  1977,  op.  361-455.

26.  Lunde, G., and A. Bjprseth.  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,  615:573-580,  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. Technol., Kh370-373, 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. Technol., 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.
     Technol., 3:1198-1200,  1969.
                                     5-43

-------
36.  Stanley, T. W., M. J. Morgan,  and E.  M.  Grisby.   Application of a  rapid
     thin-layer chromatographic procedure  to  the  determination of benzo[a]-
     pyrene, benz[c]acridines,  and  7H-benz[de]anthracen-7-one in airborne
     participates from many American cities.   Environ.  Sci. Technol., 2:
     699-702, 1968.
                                     5-44

-------
                                                                 DRAFT
                                                        DO NOT QUOTE OR CITE
                      6.   HEALTH AND  ECOLOGICAL EFFECTS

6.1  ABSORPTION, DISTRIBUTION,  AND  EXCRETION
     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.   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 concentrations 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 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 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
                                     6-1

-------
tract and, unfortunately few investigators bother to measure systemic
levels of POM.  Nevertheless, it is known that BaP when administered intra-
tracheally to rats appears in the tissues with the same pattern of distribu-
                                2                      3
tion as when given parenterally.   Vainio and coworkers  recently showed
that unchanged BaP quickly appears in the perfusion fluid of isolated
perfused rat lungs following 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  monooxygenases.  The significance of
tissue-binding of POM metabolites is discussed in Section 6.2.2 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
                                                                        4
bearing on its biological effect.  In this regard, Creasia and coworkers
showed that when BaP is adsorbed to large carbon particles (15-30 ym) and
instilled into the lungs, 50 percent of both the BaP and the carrier parti-
cles 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 careinogenicity) was low.  With smaller
carbon particles (0.5-1.0 ym), however, 50 percent particle clearance was
not achieved until seven days after administration.  In this case, 15 percent
                                      6-2

-------
of the adsorbed BaP was eluted from the particles and left free to react
with the respiratory tissues.  In the complete absence of carrier parti-
cles, BaP was cleared from the lung at 20 times the rate of adsorbed BaP.
This observation may explain the difficulty in producing experimental
pulmonary tumors with BaP without the use of carrier particles.   Other
investigators  confirmed that carbon particle size affects BaP retention in
the lung, but also demonstrated that BaP retention was not affected by
particle size when adsorbed on ferric oxide or aluminum oxide.
     Regardless of its route of administration, POM, once absorbed, becomes
localized in a wide variety of body tissues.  The distribution of radio-
                      14
activity derived from   C-BaP in the rat and mouse was determined following
                                                            2
subcutaneous, intravenous, and intratracheal administration.    The pattern
of 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 recovered in the bile and feces.  Levels of radioactivity in fat,  skin,
and muscle were not determined, nor was the amount of unchanged BaP measured
                                     6-3

-------
Table 6-1.   DISTRIBUTION AND RATE OF  ELIMINATION OF  RADIOACTIVITY  AFTER INTRAVENOUS INJECTION  OF

                                    14C-BENZO[a]PYRENE INTO RATS3 2
                                      Time after injection of aqueous
                                             suspension, hr.
Time after injection of human
      plasma solution, hr.

Site
Feces
Stomach
Intestine
Liver
Urine
Kidney
Lung
Spleen
Other organs (testes and myocardium)
Percent recovered
4

0.35
0.69
23.5
3.9
4.3
1.05
0.5
1.0
0.8
36.09
24
% of injected
8.2
3.7
43.8
1.3
12.9
0.9
0.27
0.06
0.07
71.20
4
dose
2.8
0.5
47.0
4.2
4.5
1.9
0.3
0.68
0.74
62.62
10

37.4
0.31
25.5
2.5
8.3
0.99
0.69
0.15
0.16
76.00
 a Dose:   30 wg   C-benzo[a]pyrene  in 0.9 ml vehicle

-------
       Table  6-2.   DISTRIBUTION AND RATE OF ELIMINATION OF  RADIOACTIVITY AFTER INTRATRACHEAL INSTILLATION

                                          OF 14C-BENZO[a]PYRENE  INTO RATS3 2
I
01
                                                   Time after  instillation of          Time after  instillation of
                                                     aqueous suspension,  hr.         triethylene glycol solution, hr.
                                                        1             24
                         Site                                            % of administered dose
Feces
Stomach
Intestine
Liver
Urine
Kidney
Lung
Spl een
Other organs (testes and myocardium)
Percent recovered
0
0
37.0
1.3
0
0
43.2
0
0.08
81.58
28.0
0
11.1
4.3
1.2
2.4
38.6
0
0.43
86.03
17.2
0
44.2
5.8
8.8
2.6
17.7
0
0.29
96.59
            Dose:  25 pg  C-benzo[a]pyrene in 0.3 ml vehicle.

-------
in any tissue.  Bock and Dao  later showed that relative to other tissues,
unmetabolized BaP was extensively localized in the mammary gland and general
body fat after a single feeding of the carcinogen (10-30 mg).   This accumula-
tion of BaP was greater than that resulting from 3-methylcholanthrene
(MCA), 7,12-dimethylbenz[a]anthracene (DMBA),  or phenanthrene  administra-
tion (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 carcinogen trap, which slowly releases the unchanged
hydrocarbon to the glandular cells.  In the rat, the glandular tissue is
the main 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 produces mammary cancers.  A quantitative analysis of MCA in rat
tissues 24 hours after its oral administration demonstrated that the hydro-
carbon 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, sig-
nificant amounts of MCA appeared in the milk at 24 hours after administra-
                                                     Q
tion (Table 6-5).  On the other hand, West and Horton  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
                                    g
produced similar results as for MCA.   Twenty-four hours after receiving a
                                     6-6

-------
Table 6-3.  EFFECT OF MOLECULAR STRUCTURE ON HYDROCARBON LOCALIZATION IN MAMMARY
                                 GLAND AND FAT* 6
                                                  Hydrocarbon concentration,
                                                          ug/gm +_ S.E.b
Hydrocarbon
3-Methyl chol anthrene
Benzo[a]pyrene
7 , 1 2-Dimethyl benz[a]anthracene
Phenanthrene
I1UIIIL/CI
of rats
30
10
10
5
Mammary fat
18.0 + 1.6
29.1 + 4.0
23.4 + 1.7
1.8 + 0.1
Fat
39.2 + 3.9
55.9 + 5.3
39.3 + 4.4
2.8 + 0.3
  Levels of hydrocarbon were determined 24 hours after a single intragastric
  feeding of 30 mg test compound in 1 ml of sesame oil.  Quantitative assays
  were conducted by fluorescence measurements of the crude benzene extract
  of tissue containing hydrocarbon.

  + Standard deviations.
                                     6-7

-------
Table 6-4.  LEVELS OF ORALLY ADMINISTERED 3-METHYLCHOLANTHRENE IN TISSUES OF
                       50-DAY-OLD VIRGIN FEMALE RATS?
                                        Level  of 3-methylcholanthrene,
      Tissue                                     ug/g tissue^


Breast, rat 1                                        15
Fat, rat 1                                           70

Breast, rat 2                                        12
Fat, rat 2                                           17

Breast, rat 3                                        24
Fat, rat 3                                           38

Brain, pooled                                         3.5
Lung, pooled                                          2.1
Kidney, pooled                                        1.1
Thymus, pooled                                        1.4
Liver, pooled                                         0.3
Muscle, pooled                                        0.04
Heart, pooled                                        NFC
Spleen, pooled                                       NF
Uterus, pooled                                       NF
Ovaries, pooled                             •         NF
Adrenals, pooled                                     NF
Pituitaries, pooled                                  NF


a 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.

c Not found.
                                     6-8

-------
vo
       Table  6-5.  LEVELS OF ORALLY ADMINISTERED 3-METHYLCHOLANTHRENE  IN BREAST AND FAT OF VIRGIN RATS
                              AND IN MILK, BREAST, AND  FAT OF LACTATING RATS* 7
Rat No.
Virgin rats 50 days old 1
2
Lactating rat
(7 days postpartum) 1

Tissues
Breast
Fat
Breast
Fat
Breast
Fat
Milk (fetus 1)
(fetus 2)
Weight,
9
2.6
0.9
2.1
0.9
11.5
1.7

3-Methylcholanthrene,
pg/g tissueb
6.5
7.1
6.4
10.9
17
11
23
13
        One single feeding of 3-methylcholanthrene of 10 mg in 1 ml of sesame oil was given and animals
        were sacrificed 24 hours later.

        Crude benzene extracts of tissue were analyzed in a spectrophotofluorometer using either 295 my
        or 355 my excitation light.

-------
single oral dose of 30 mg  H-DMBA at two levels of radioactivity (26 pCi
and 2600 yCi), the greatest amount of activity recovered from rat tissues
was in the peri renal 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 predominant-
ly 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.
                                3
     The tissue distribtuion of  H-dibenzo[g,c]carbazole (DBcgC) following
intratracheal instillation was qualitatively and quantitatively similar to
BaP.     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).
     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.    With DMBA, the concentra-
tion of the carcinogen in fetal tissues reached a maximum (1.53 to 1.6 ug/g)
2 to 3 hours after maternal administration.  At 2 hours, the ratio of DMBA
concentrations 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
                                     6-10

-------
Table 6-6.  DISTRIBUTION AND ELIMINATION OF RADIOACTIVITY AFTER INSTALLATION
             OF 3H-DIBENZO[c,g]CARBAZOLE IN SYRIAN HAMSTERSa 10
Percent instilled dose recoverd at:
Site
Trachea
Lungs
Liver
Kidneys
Brain
Intestines (large and small)
Fat
Urine
Feces
Total % recovered

11
63
1.4
0.0
0.0
11
0.0
0.0
0.0
86.4
1 hr.
+ 1.2
+ 2.2
+ 0


+ 1.4



+ 5.44
3 hr.
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
83.7 +4.65
6 hr.
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
82.2 +4.23
a Animals were treated once with 3 mg  H-dibenzo[c,g]carbazole suspended with
  FeO- particles in 0.2 ml saline.
                                     6-11

-------
Table 6-7.   CONCENTRATION (yg/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
 Carcinogen
administered
                           Fetuses
                                         Placentas
                                                          Liver
Mean and limits
 of variations
Number of
analyses
                                                        Mean and limits
                                                         of variations
Number  of
analyses
Mean and  limits
 of variations
Number of
analyses
I

tv>
    DMBA         1.53 (0.63-3.50)       4

    BaP          2.77 (1.67-5.00)       5

    MCA        0.0013 (traces-0.002)    6
                               1.28  (0.86-1.60)        4

                               3.94  (2.24-6.15)        5

                               0.009 (traces-0.03)      6
                                              8.80  (5.50-11.80)       4

                                             23.50  (11.10-50.20)      4

                                              0.10  (0.03-0.22)        6

-------
dose of DMBA produced similar fetal tissue levels as the intragastric
administration of 200 mg/kg DMBA.
     As long ago as 1936, it was recognized that various polycyclic aromatic
hydrocarbons were primarily excreted through the hepatobiliary system and
          12 13
the feces.  '    With the advent of radiotracer methodologies, it became
                                                             14 2
possible to study more quantitatively the elimination of POM.   '   Patterns
   14
of   C-BaP excretion were determined in both intact rats and in rats with
                 2
biliary fistulas.   When BaP was given intratracheally, or by subcutaneous
and intravenous injection, maximum excretion of radioactivity always occurred
through 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 wg BaP
(intravenous dose).  Radioactivity was detectable in the bile for 24 hours
following treatment, and a total of 1 percent of the administered BaP was
recovered unchanged.  The total  recovery of radioactivity in BaP-treated
rats having biliary fistulas 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 radioactivity 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% of the dose is excreted
in the feces within 24 hours.    However, when injected intraperitoneally,
                                     6-13

-------
Table 6-8.  DISTRIBUTION OF RADIOACTIVITY IN RATS WITH BILIARY FISTULAS AFTER INTRAVENOUS INJECTION
                             of 14C-BNEZO[a]PYRENE IN PLASMA SOLUTION2




CT>
1
-£»



Dose,
ug
19
26

45
134
400

Bile flow rate,
ul /minute
7.2
18.9

18.0
12.6
15.3
Duration of
experiment,
hours
4
12

12
14
24
Recovery in bile,
% of injected dose


1 hr.
17
21

27
26
4


2 hr.
9
19

18
30
3


3 hr.
9
9

7
13
3


Total
39
78

75
96
32

-------
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 coworkers ), 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 ROM's.  On the other hand, ROM'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 most likely to occur by inhalation, it cannot be assumed that excretion
of ROM's 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
          14
the feces.    However, the rate of disappearance from the site of applica-
tion 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.
     The rate of disappearance of BaP from the various tissues can be
altered by prior treatment with any one of several POM.  Schlede and
                                     6-15

-------
         20
coworkers   reported that inducers of microsomal  enzymes (BaP,  pheno-
barbital) can enhance the rate of biliary excretion of metabolites of BaP.
In addition, pretreatment of rats with BaP, DMBA, or MCA stimulated the
rate of disappearance of an intravenous dose of BaP, and decreased the
                                                              21
concentration of BaP in various tissues (lung, fat, adrenals).     Taken
together, these results indicated that BaP can stimulate its own  metabolism.
Furthermore, in the pregnant rat, pretreatment with BaP, MCA, chrysene,
benz[a]anthracene, or DBahA had a marked stimulatory effect on the in vitro
metabolism of a subsequent dose of BaP in maternal liver, placenta, and
fetal liver.22
     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. '    However, recent studies
have also indicated that intratracheal instillation of BaP to hamsters in
combination with ferric oxide did not increase 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
                    18
tract tumors by BaP.    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
carcinogens for tumor development is the concommitant exposure to agents
                               23
which inhibit ciliary activity.    The tracheobronchial epithelium is
normally protected by a mucous blanket and ciliated cells.  These cells
                                     6-16

-------
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 6.5).
     A thorough examination has been conducted on the clearance of intratra-
                  2
cheally instilled  H-BaP (5 mg total dose) from various tissues of the ham-
     19
ster.    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 respira-
tory tract.  In the second (slow) phase of elimination, a significant
retention of radioactivity became evident only among animals given  H-BaP
together with an adsorbent.  This accumulation of  H-BaP-derived radio-
activity in the lung, although small, was considered toxicologically signifi-
cant.  The levels of radioactivity in other organs and in blood, urine, and
feces were not apparently influenced by co-administered adsorbents, but did
                                     6-17

-------
3
I
            100
         •O
         £
         a
         I

         g  10
         •a

         1
         a

         S?
         V)
         n

         "S
         a
         x
         OJ
         C>
         c
         _3


         .£  0.1
         o
         (B
         O

         1
         cc
           0.01
      [3 H]  3.4- Benzopyrene alone

      [3H]  3,4— Benzopyrene + abestos

9	^ [3H]  3,4— Benzopyrene + carbon black
                                             I
           _L
                                            14             21

                                 Time in days from single mtratracheal injection
                          28
35
              Figure 6-1.   Clearance of  radioactivity  derived  from
               o                                   17
              [ H]3,4-benzopyrene  in the lung.
                                          6-18

-------
appear to follow a biphasic pattern of elimination (Figures 6-2 - 6-6).
However, blood and fecal levels of radioactivity dropped off more rapidly
                       o
in those animals given  H-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 identify 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.    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.
                                     6-19

-------
100
                                  [3 H ]  3,4- Benzopyrene alone
                                  [3H]  3,4- Benzopy rene + asbestos
                                  [3H]  3,4— Benzopy rene + carbon black
                     6                14                21
                    Time in days form single intratracheal injection
   Figure 6-2.   Clearance of radioactivity derived from
    3                               17
   [ H]3,4-benzopyrene  in liver.
                               6-20

-------
100 r-
  10
0.01
•   • [3H]  3,4-Benzopyrene alone
o—o [3H]  3,4- Benzopyrene + abestos
•---• [3H]  3,4- Benzopyrene + carbon black
                   I
        I
                   3              7             14
                       Time in days from single intratracheal injection
                                        36
     Figure  6-3.   Excretion of radioactivity  derived from
     [ H]3,4-benzopyrene  in feces.
                                6-21

-------
       10
Q.

2

•o
CO

CC
      0.1
0.01
                                       [3H] 3,4- Benzopyrene alone

                                       [3H] 3,4-Benzopyrene + asbestos


                                  •—•• [3H] 3,4- Benzopyrene + carbon black
                                                I
                                      7             13


                            Time in days from single intratracheal injection
                                                                               36
         Figure 6-4.   Excretion  of radioactivity derived from

          3                                17
         [ H]3,4-benzopyrene in  urine.
                                     6-22

-------
  10
 0.1
0.01 J
                                     •——• [3 H]  3.4 - Benzopyrene alone
                                     o—o [3H]  3,4- Benzopyrene + abestos
                                     •— -• [3 H]  3,4- Benzopyrene + carbon black
I
                                            I
                   7             14             21             28
                       Time in days from single intratracheal injection
                                                           35
      Figure  6-5.  Clearance of radioactivity derived  from
      [  H]3,4-benzopyrene  in kidneys.
                                  6-23

-------
  10 r
                                          •	• 13 H ]  3,4- Benzopyrene alone
                                          o—o [3 H ]  3,4- Benzopyrene + abestos
                                          *—• [3 H]  3,4- Benzopyrene + carbon black
 0.1
0.01
                       7                 14                 21
                     Time in days from single intratracheal injection
28
      Figure 6-6.  Blood level  of radioactivity derived from
      [ H]3,4-benzopyrene.
                                   6-24

-------
6.2  METABOLISM AND ACTIVATION TO REACTIVE ELECTROPHILES
6.2.1  Carcinogenic Polycyclic Hydrocarbons
     That carcinogenic chemicals are human hazards became apparent with the
advent of the industrial revolution.  During the last century, the logarithmic
growth of chemical technology has resulted in the development of vast amounts
of synthetic compounds for industry, agriculture, and medical and consumer
products, such as cosmetics, food preservatives, and colorants.  Most of these
compounds are not natural to the environment and some are toxic by virtue of
their chemical structure.  However, others have low reactivity when released
into the air, water, or soil, but when taken into the organism are activated
by the biochemical pathways of the body to form detoxified and carcinogenic
species of the parent molecule.
     Several major types of carcinogenic molecules have evolved as prototype
structures for the study of chemical carcinogenesis, such as aromatic amines,
nitrosamines, aflatoxins, and polycyclic hydrocarbons.  Large numbers of
derivatives of these basic structures have been synthesized and tested for
carcinogenic activity in many animals in addition to mammals, including
birds, amphibians, reptiles, and fish.  Research into the biochemical mecha-
nisms of action of chemical carcinogens is fairly recent and is primarily the
result of innovative methods of organic synthesis, several new sensitive
bioa'ssays, and new analytical techniques for accurate analysis of highly
reactive transient intermediates formed during the metabolic processing of the
carcinogen (see Section 6.5).  At present, no complete mechanism of action is
known for any carcinogenic molecule nor is the target site definitely under-
stood, although the current literature suggests nuclear macromolecules as the
                                    6-25

-------
most probable targets (see Section 6.2.2).  However, the divergent molecular
structure of chemical carcinogens and the wide variation in tissue and species
susceptibility have only recently enabled investigators to begin comparing
structurally different chemical carcinogens with regard to their relative
carcinogenic potency, target tissues, and species specificity.   The ultimate
aim will be to draw this large body of data into formulation of a mechanism
describing the cause of cancer by chemicals.
     Early attempts to explain the careinogenicity of'polycyclic hydrocarbons
                                       24
utilized physico-chemical calculations.    These early hypotheses were based on
the assumption that those regions of the molecule most likely to react in
chemical substitution or addition reactions would be most likely to react with
cellular target sites, causing an irreversible disruption of cell homeostasis.
This concept has not been successful for polycyclic hydrocarbons.
     It is generally accepted that the ultimate forms of carcinogens are
electron deficient (see Section 6.5).  Alkylation of critical sites by these
electrophiles represents the basis of tumor initiation.  The significance of
potentiation of the transformation by tumor promoters such as phorbol esters
or environmental modifiers such as catechol is particularly relevant to man.
Since POM contains a mixture of chemical components and human exposures are
almost always in conjunction with several components in air pollution, sub-
threshold doses could be potentiated by other environmental factors.
     In animal models the initiation-promotion concept has been widely studied
(see Section 6.5).  In general, a low dose of a carcinogenic POM is applied
once as a tumor initiator to an epithelial tissue.  This dose alone does not
                                     6-26

-------
induce tumors; however, when the tissue is subsequently treated with a non-
carcinogenic tumor promoter, such as the phorbol  ester, tumors will  appear in
20 to 40 weeks.  The concurrent application of very low doses of carcinogenic
POM with a cocarcinogen such as catechol or pyrene will also induce tumors in
epithelial tissues.
     Since human exposure to carcinogenic chemicals is, in most cases, a
continuous assault of very low doses (e.g., cigarette smoke, auto exhaust,
etc.), other noncarcinogenic vectors in our environment probably enhance the
effect of the relatively low doses of initiating  carcinogen.
6.2.1.1  Historical perspectives—The human population is being exposed to
polycyclic aromatic hydrocarbon (PAH) carcinogens at an increasing rate as a
consequence of industrial growth.  Pott's observation in 1775 that scrotal
cancer found in chimney sweeps was the result of  some agent in soot was the
first suggestion that cancer could be caused by environmental contaminants.
However, it was not until more than a century later that skin tumors were pro-
                                                                        25
duced experimentally in rabbits by painting them  with coal tar extracts.
This marked the beginning of the search for the active carcinogenic substance
in soot, culminating in the isolation and identification of BaP by British
                                ?fi 77
investigators two decades later.  "    Synthetic  dibenz[a,h]anthracene (DBahA)
                                 28
had been shown to be a carcinogen   and the structural similarities between the
two molecules implicated polycyclic aromatic hydrocarbons as potential car-
cinogens in man.  Many other PAH's were subsequently synthesized, with most
                               29
proving to be non-carcinogenic.    Nevertheless,  several PAH's were discovered
that are routinely used as tumorigens in laboratory animals, such as MCA and
                                     6-27

-------
several of the methylated benz[a]anthracenes.  Elucidation of the metabolic
mechanisms by which these compounds transform normal cells into malignant
cells necessarily rests in a complete description of their physical  and chemi-
cal similarities and the molecular conformation formed between the carcinogen
and target site(s) inside the cell.  Although a general  mechanism by which PAH
carcinogens interact with the target site or sites where transformation to
malignancy occurs is now unknown, it is hoped that complete assembly of the
metabolic pathways will reveal whether there is an overall mechanism of acti-
vation for transformation.
     Polycyclic hydrocarbons are metabolized by the microsomal mixed function
oxidase system (MFO), often termed aryl hydrocarbon hydroxylase (AHH), which
is most abundant in liver.  This enzyme system has been studied extensively
and is the subject of several reviews.  '    While it is known that this
enzyme complex is involved with detoxification of xenobiotics in conjunction
with various P-450 type cytochromes, it is apparent that this system is also
directly involved with the metabolism of polycyclic hydrocarbons to their
active species.  A second microsomal enzyme, epoxide hydrase (EH), converts
epoxides into vicinal glycols.  Since some epoxides are more active carcino-
gens than the parent hydrocarbon, this enzyme would likely affect both car-
cinogenesis and detoxification.  Information on epoxide hydrase has been
                   32 33
recently summarized  '   and its importance in the formation of the three
                                                         34 35
known dihydrodiols of BaP has recently been demonstrated.  '    Figure 6.7
presents a schematic representation of the various enzymes involved in acti-
vation and detoxification pathways for BaP, that is also representative of
the known mechanisms of POM metabolism in general.
                                     6-28

-------
i
m
                          (ENDOPLASMIC
                            RETICULUM)
                      GLUTATHIONE
      CYTOCHROME P-450
      MIXED-FUNCTION OXIDASE (MFO)
                                                  MFO
         BaP-O-SG
      (DETOXIFICATION   TRANSFERASE
        PRODUCTS)      (CYTOSOL)
BaP OXIDES
                                         EPOXIDE
                                         HYDRASE
                                         (ENDOPLASMIC
                                         RETICULUM)
•»•  BaP PHENOLS
                               MFO
                                                           BaP QUINONES
                            MFO
                 BaP DIOL EPOXIDES
                (PROPOSED ULTIMATE
                   CARCINOGENS)
                                   BaP DIHYDRODIOLS (PROPOSED PROXIMATE CARCINOGENS)
             UDP-GLUCURONOSYL TRANSFERASE
                 (ENDOPLASMIC RETICULUM)
               H2O -SOLUBLE CONJUGATES
               (DETOXIFICATION PRODUCTS)
      Figure 6-7.   Enzymatic pathways  involved  in  the activation
      and detoxification of  BaP.
                                     6-29

-------
     Other intermediates currently under investigation in several laboratories,
primarily using BaP as the starting compound, are transient free radicals.
This type of intermediate is known to be generated easily both chemically and
enzymically.  '    Because it is too unstable to isolate and study, it is not
yet possible to determine if, or how, this labile species is involved in the
carcinogenic activity of BaP.
     Several reports on the metabolic production of a 6-oxy-BaP radical suggest
that this activated BaP derivative is involved in DNA binding, and thus may
              38-41                   38
induce tumors.  "    Kodama and Nagata   incubated 6-oxy-BaP radical with the
synthetic nucleotide poly G and obtained a strong ESR signal in the resulting
bound complex.  This signal was comparable to that for free 6-oxy-BaP radical
and indicated covalent binding to guanosine residues.  However, when 6-oxy-BaP
was incubated with calf thymus DNA, only a small ESR signal could be observed.
Moreover, the 6-oxy-BaP radical does not possess significant carcinogenic
activity.
     The one-electron oxidation of POM to produce radical cations which bind
to cellular constituents has been proposed as a general pathway of metabolic
                                    42                         42
activation leading to tumorigenesis.    Cavalieri and coworkers   have con-
ducted studies to show that correlations exist between the ease of radical
cation formation and carcinogenic activity for certain polycyclic hydrocar-
bons.  In particular, binding to nucleophiles by the trapping of radical
cations occurs as a result of preferential positive charge localization at
position 6 of BaP and at position Cl of MCA.  Evidence which supports this
mechanism is the observed binding of BaP to mouse skin DNA and RNA at the
6-position and the apparent lack of carcinogenic activity for MCA derivatives
                                     6-30

-------
substituted at the 1-position.  Nevertheless, numerous exceptions to the
radical cation hypothesis, and the well-demonstrated potent careinogenicity of
diol epoxides derived from POM, argue against the critical  involvement of
radical cations as either proximate or ultimate carcinogens for all  POM.
     It is suggested that hydroxymethyl derivatives of alkyl-substituted POM's
may be critical carcinogenic intermediates, even though these metabolites
generally show little tumorigenic activity.  '     In this regard, data exist
to support the view that for certain methyl-substituted POM (e.g., 7-methyl-
benz[a]anthracene), the hydroxymethyl derivative is a proximate carcinogen
                                                                42
giving rise to an ester which serves as the ultimate carcinogen.    A com-
parison of two less active intermediates showed that the carcinogenicity of
6-hydroxymethyl BaP is far less than for its sulfate ester.
     It seems clear that specific mechanisms and transient intermediates are
important steps in carcinogen activation, detoxification, and the physico-
chemical binding to transformation receptor(s)  in the cell.  It is essential
that critical methodology be employed to determine the ultimate carcinogenic
form(s) of metabolically activated POM.  Advances in systematic identification
of proximate forms or efficient and relevant trapping of the reactive ultimate
carcinogen are required to further the understanding of the mode of metabolic
activation of POM.
6.2.1.2  Chemical reactivity and care inogeni city—For many years, investiga-
tors have sought a common molecular feature among POM carcinogens which would
serve to explain their biological activity.  The "electronic theory of car-
cinogenesis" has relied upon an analysis of the influence of electron density
                                     6-31

-------
at specific molecular regions to explain unique reactivity with cellular
constituents.  A basic assumption arising from the work of the Pullmans and
      9/1
others   was that a meso-phenanthrenic region ("K-region") of high ir-electron
density and with a propensity for addition reactions was a critical  structural
feature for polycyclic carcinogens.  In expanding this hypothesis, further
biologic significance was attributed to the concommitant presence of a rather
unreactive meso-anthracenic region ("L-region") for high carcinogenicity.   In
addition, a region of comparatively low reactivity which characteristically
undergoes metabolic perhydroxylation (corresponding to the 3,4-positions  of
benz[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:                            _ _ .
                                               M-region of metabolic
                                               perhydroxylation
                  L-region  «^^          ^^  K-region
The electronic K-L theory of carcinogenic reactivity  has  encountered  numerous
inconsistencies, primarily because these relationships were  derived from
properties of the parent hydrocarbon and gave  no  consideration  to  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

                                      6-32

-------
trioi carbom'um ions.  Under the assumption that diol epoxides which are more
readily converted to carbonium ions will be better alkylating agents to pro-
duce carcinogenesis and mutagenesis, the "bay region" theory has been pro-
      43 44
posed.  '    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:
           Bay region                                   Bay region
          Benzo[a]pyrene                          Benz[a]anthracene
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 with the oxirane oxygen not situa-
ted in a "bay region."  Experimentally, the "bay region" diol epoxides of
benz[a]anthracene, BaP, and chrysene were more mutagenic in vitro and/or
tumorigenic than other diol epoxide metabolites, their precursor dihydrodiols,
the parent hydrocarbons, or other oxidative metabolites (see Sections 6.4
and 6.5).  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.
6.2.1.3.  Metabolism of PAH--Polycyclic aromatic hydrocarbon metabolites have
been arbitrarily divided into two groups on the basis of solubility.  In the
first group are those metabolites that can be extracted from an aqueous incu-
bation mixture by an organic solvent, e.g., ethyl acetate.  This group consists
                                     6-33

-------
of ring-hydroxylated products such as phenols and dihydrodiols,   '    and
hydroxymethyl derivatives of those polyaromatics having aliphatic side chains,
such as DMBA   and MCA.    In addition to the hydroxylated metabolites are
quinones that are produced enzymically by microsomes and non-enzymically from
air oxidation of phenols.  Labile metabolic intermediates such as epoxides can
                               AQ AQ Cn
also be found in this fraction. "'^'^
     In the second group are the water soluble products that remain after
extraction with an organic solvent.  While it is believed that many of these
derivatives are formed by conjugation of the hydroxylated products  to gluta-
      51                   52             53
thione   or glucuronic acid   and sulfate,   rendering the compound more
hydrophilic and presumably less toxic, this group of derivatives has not been
rigorously studied.
     The metabolite profile of BaP which has been largely worked out using
high pressure liquid chromatography is seen in Figure 6-8.  It consists of
three groups of positional isomers.  There are three dihydrodiols,  three
quinones, and two phenols.  The major BaP metabolite found in microsomal
incubations is 3-hydroxybenzo[a]pyrene, with 9-hydroxybenzo[a]pyrene present
in lesser amounts.  The BaP-4,5-epoxide has been isolated and identified as a
precursor of the BaP-4,5-diol.  Other studies indicate that epoxides are the
precursors of the 7,8-diol and 9,10-diol as well.  There have been no inter-
mediates isolated as phenol precursors, although recent evidence using deu-
terium labeling suggests that at least a portion of 3-OH-BaP is derived from
                            54
an intermediate 2,3-epoxide.    Evidence also exists for epoxide involvement
                                            qc
in the formation of 9-hydroxybenzo[a]pyrene.    The metabolite profile has
                                     6-34

-------
                                                       [9,IO-diol-7,8-ep«]

                                                       [7,8,9,10-tetroi]
   6-OH
 CONJUGATES
    7 8-epox
                                            7.8-diol
"- » c , Jt-W
    4,5-diok
           9,10-epox



— [7,8-diol-9,IO-ep(«]

    [7,8,9,10-tetroi]
                7-OH
BOUND MACROMOLECULES
       ONA
       RNA
       PROTEIN
Figure 6-8.   Metabolites of  benzo[a]pyrene.
                       6-35

-------
been extended to include two additional phenols (7-OH and 1-OH) by the use of
recycle chromatography, which allows multiple passages through the column and
                             55
will resolve all BaP phenols.    In addition, solvent regimen changes have
resolved some of the tetrols that were too polar for separation in the stand-
ard water-methanol gradients.  There is little further knowledge of the
chronological order and kinetics of metabolite formation.  The effect of
various metabolites on microsomal enzyme activity has not been thoroughly
studied.  However, it has been demonstrated that there are different types of
hepatic microsomal oxygenase induction  '  »b8»59 that appear to be dependent
upon the chemical nature of the inducer.  This may significantly influence
the amount of active intermediate species of carcinogen formed and its proba-
bility of reaching a target site.  Removal of reactive intermediates and
toxic metabolites is certain to affect the probability of cell survival, both
normal and transformed.
     Considerable progress has been made in the isolation and characterization
                         fiO 61 62                                        ?R ?6
of the AHH enzyme complex  '  '   and similar work is in progress for EH.  '
     Polycyclic aromatic hydrocarbons are relatively inert chemically and
quite hydrophobic, and cell lipids are probably needed for these compounds to
be dispersed in the cytoplasm.  Autoradiographic studies in vitro indicate
that PAH's possess access to all parts of the cell, although at present there
                                    C O C A
is no evidence for active transport.  '    It was already apparent in 1950
that for metabolism to occur, some form of biochemical action was necessary to
raise the molecule to a higher reactive state because of the stability of the
polyaromatic molecule.  Since several dihydrodiols and phenols had been isola-
ted from tissue incubations, it was suggested that reactive epoxides were
                                     6-36

-------
likely precursors that could then be hydrated to dihydrodiols or opened non-
enzymically to form phenols.    Epoxide intermediates for a number of poly-
cyclic hydrocarbons, both carcinogenic and noncarcinogenic, have now been
isolated and suggest that the arene oxide intermediate is a general biochemi-
cal scheme for biotransformation of these xenobiotics. °»49>50»^5
     Since the resonance level of the polyaromatic system makes ring openings
difficult, the enzymatic attack is designed to open double bonds and add an
oxygen moiety, such as a hydroxyl group, to give it more polarity and,
therefore, more solubility in aqueous media such as body fluid; and to
facilitate removal by the kidney.  The effect on removal is plainly seen
when extremely high concentrations of labeled polyaromatics given to test
animals are almost completely excreted within 48 hours.  Only residual
radioactivity remains, and it is below the resolution of most standard
assays to confirm that the remaining radioactivity is on the polyaromatic
molecule and has not been exchanged onto another substance.
     In the formation of a reactive metabolic intermediate, stable compounds
are converted to unstable metabolites.  In the case of polyaromatics, this
step is the formation of an arene oxide ring that is highly strained and
favors ring opening.  Thus, nucleophilic attack of this reactive intermedi-
ate either directly or through the formation of a transient carbonium ion
would be greatly enchanced.  Alkylations of this type are common to many
classes of carcinogenic chemicals.  So, paradoxically, an active carcinogen
is created in the process of detoxification. Early calculations relied upon
the participation of the most chemically reactive region of the molecule,
                                     6-37

-------
the so-called "K-region" (corresponding to the meso-phenanthrenic double
bond), for carcinogenicity predictions.  This theory was formulated at a
time before the identification of metabolic products and advanced in vitro
and in vivo assays for determining their biological  activity.
     Taking epoxides to be the intermediates and the transient ring-opened
carbonium ion to be the alkylating species, the "bay region" concept has been
developed,   using new knowledge concerning the role of electrophilic inter-
mediates.  The concept was derived from the prediction that the portion of
the molecule most likely to form an epoxide intermediate (that once opened,
would be long-lived enough to have a relatively high probability of hitting
the critical target site) would be the carcinogenic determinant of this
molecule.
     While a general mechanism that will undoubtedly be complicated by
species variation and structural and chemical differences (e.g., solubility)
between different polycyclic molecules, this concept has worked quite well
for BaP.68'69
     Since BaP has three known metabolic epoxide intermediates, the use of
this concept to predict relative carcinogenicity and mutagenicity has been
tested.  While the K-region epoxide (4,5 epoxide) is an effective mutagen in
bacterial and mammalian V79 cells, it is a very weak carcinogen in mice (see
Sections 6.4 and 6.5).  It is far less effective as a carcinogen than the non
K-region 7,8-epoxide.70'71'72'73
     Moreover, evidence for remetabolism of dihydrodiols has simply established
a higher level of certainty for this "bay region" concept.  The finding that
BAP 7,8-dihydrodiol was metabolized to BaP 7,8-diol-9,10-epoxides showed that
                                     6-38

-------
the microsomal oxygenases were capable of recycling the BaP molecule through
an additional epoxldatlon.  This dlol-epoxlde was also hydrated to form a
       7d
tetrol.
     The stereochemical nature of the diol-epoxlde intermediates allows the
possibility of two optically active enantiomers from (-)-BaP 7,8-dihydrodiol
and two optically active enantiomers from (+)-BaP 7,8-dihydrodiol.  In two of
the diol-epoxides, the 7-hydroxyl would be in the same plane (syn) as the
oxide ring, and in the other two, the 7-hydroxyl would be on the opposite
plane (anti-) to the epoxide ring.
     In vivo experiments with tumor formation in mouse skin and in vitro
assays using mammalian cells and bacteria have implicated the anti-config-
uration as the biologically active BaP 7,8-diol-9,10-epoxide isomer.  '    In
addition, studies with the (-) trans-7,8-diol isomer as substrate in the
microsomal system indicate the (+)-anti-configuration of the epoxide [r_-7,;t-
8-dihydroxy-t-9,10-oxy-7,8,9,10-tetrahydrobenzo[a]pyrene] is that which is
also biologically active.  The isolation and characterization of the DNA-
bound product in which guanosine predominates support these observations.
To date, the diol-epoxide is too unstable for direct isolation from bio-
logical systems.  While researchers conducting molecular and chemical studies
have made strides in discovering the putative reactive species of BaP and its
probable mode of macromolecular interaction, the basic problem still exists
of how the transformation event occurs and why there are different suscepti-
bilities between species.
     The advances in the "bay region" theory of PAH carcinogenesis and in
determining the identity of proximate and possibly ultimate carcinogenic
                                     6-39

-------
forms of PAH have been achieved using homogenates of liver.  In combination
with bacterial mutagenicity assays, similar advances can be expected for
carcinogens of particular environmental concern.  The need for further test-
ing of the data obtained from these in vitro experiments in suitable in vivo
animals models, however, is essential.  If such in vitro experiments can
permit the identification of proximate or ultimate carcinogenic intermediates,
the influence of other environmental agents on modifying enzyme induction or
metabolic activation to specific compounds could be studied in greater detail.
     The metabolic fate of most PAH's 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,
                                                   78
alcohols, ketones, aldehydes, and carboxylic acids.    More commonly, meta-
bolic attack on one or more of the aromatic double bonds (K-region and non-K-
region) leads to the formation of isomeric dihydrodiols by the intermediate
                               79 78
formation of reactive epoxides.  '    Dihydrodiols'are further metabolized by
the microsomal monoxygenases 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 tetrahydrotetrols, is a key step in protecting
                                                             80 81
the organism from toxic interaction with cell macromolecules.  '
     With unsubstituted POM, such as benz[a]anthracene and DBahA, oxidation
                                                                      80 82 83 84
occurs with rat liver homogenates to produce phenols and dihydrodiols.  '''
Both K-region and non K-region oxidative metabolites are formed, which may
                                     6-40

-------
subsequently be conjugated with glutathione and excreted in the urine as
mercapturic acids.  In addition, hydration of the epoxide intermediates
formed from these compounds may yield dihydrodiols.  Figure 6-9 presents
some of the significant routes of metabolism for DBahA.   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 detec-
    78
ted.   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, ROM's bearing alkyl substitutes
(e.g., MCA, DMBA) are primarily hydroxylated at the alkyl side chain.  The
                                                                          CO QC QC
metabolism of DMBA has been carefully studied using rat liver homogenates. °>oa'00
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 8,9-dihydrodiols (non K-region).  In contrast, a
K-region 5,6-epoxide is also formed, but instead of being hydrated to the
                                               QC
dihydrodiol, it is conjugated with glutathione.    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 DMA occurs through the generation of a diol epoxide
                    87
in the 1,2,3,4-ring.    Ring hydroxylation at the 3- and 4-positions of DMBA
                                    6-41

-------
                    12
            R« CH2 —CH —CO — NH— CH2—C02H


                  NH — CO — CH2 — CH2 — CH (IMH2) — CO2H
                                                               SR
Figure 6-9.   Metabolism of  dibenz[a,h]anthracene by
rat-liver  homogenate.82
                         6-42

-------
                                              u w
(M-region) had previously been shown to occur.    Thus, evidence for the
"bay region" theory being of general applicability is increasing.
     Metabolism of the carcinogens 7- and 12-methylbenz[a]anthracene is
                              88
qualitatively similar to DMBA.    The products formed by rat liver homogen-
ates 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 6-10 presents a probable scheme for the
metabolism of 7-methylbenz[a]anthracene which would also apply to 12-methyl-
benz[a]anthracene.
     The metabolites of MCA formed by rat liver homogenates, liver microsomes,
or a highly purified cytochrome P-448-dependent monooxygenase system with
                                89 90
epoxide hydrase are all similar.  '    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.
                                       90
These results led Thakker and coworkers   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).  Moreover, it is
vital to determine whether metabolism in rodents mimics metabolism in man.
                                     6-43

-------
"OH |               HO'
   CH, —CH —CO —Gly
        I
        NH—Glu
                                                              CH2 — OH
Figure 6-10.  Probable metabolic  pathways of 7-methylbenz[a]anthracene.
                                   6-44

-------
In this regard, it has been shown that the metabolism of benz[a]anthracene,



7-methy1benz[a]anthracene, and BaP was qualitatively similar in rat liver and


     91 92
lung.  '    Both K-region and non K-region dihydrodiols were detected,



which almost certainly arose from epoxide precursors.  Earlier work estab-



lished that both rat and human lung preparations converted benz[a]anthracene


                                       80
to the corresponding K-region expoxide.    Furthermore, tracheal slices of



Wistar rat and Syrian hamster, and human bronchial tissue in culture all



produced qualitatively similar products from benz[a]anthracene, 7-methyl-


                           80
benz[a]anthracene, and BaP.    Metabolite profiles with these same hydro-


                                                                      Rfl
carbons in mouse, rat, and human skin were also qualitatively similar.



Patterns of BaP metabolism in hepatic microsomes from primates, rats, and


                                           93
miniature swine were qualitatively similar,   as were the metabolite pro-



files (by thin-layer analysis) of BaP in microsomes from mouse liver, spleen,


                 94
kidney, and lung.    However, among different strains of mice with varying



responsiveness to the microsomal enzyme-inducing effects of BaP pretreatment,


                                                    94
metabolite profiles could be quantitatively altered.    Among several tis-



sues of the human adult and fetus (liver, adrenal, placenta), the same types



of BaP metabolites are obtained, although microsomal enzyme activity was


                   95
lower in the fetus.    Moreover, no significant differences were found in



the pattern of BaP metabolism between laboratory animals and man.  However,



recent studies indicate that rat liver and lung preparations differ with


                                                                    96
respect to the nature of free radical metabolites produced from BaP.    In


                             93
addition, Leber and coworkers   reported that unidentified oxidative metabo-



lites of BaP are obtained with rhesus monkey lung microsomes (analysis by
                                     6-45

-------
high-pressure liquid chromatography) which are not obtained with liver micro-
somes from the same species.  Differences are also seen in the BaP metabolites
formed by human lymphocytes and liver microsomes when analyzed by high-
                               97
pressure liquid chromatography.    Thus, it seems that interspecies compari-
sons of POM metabolism in a single tissue may be a valid practice, whereas
differences in tissue-specific metabolism may prevent the extrapolation of
data from hepatic to extrahepatic (i.e., target organ) tissues.
     An additional factor having crucial importance to the metabolic activation
of POM in various tissues and species is the level of drug metabolizing
                98
enzymes present.    As indicated in Section 6.3.4, tissue susceptibility to
POM-mediated carcinogenesis may be obligatorily coupled to the presence and
amount of microsomal mixed-function oxidase enzymes and epoxide hydrase.
Protection from POM-induced tumorigenesis, on the other hand, may be largely
determined by the.capacity for target tissues to enzymatically detoxify
(e.g., by glutathione or glucuronic acid conjugation) and remove activated
metabolites.
6.2.2  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 POM's become cova-
                                    99
lently bound to tissue constituents.    The molecular basis for carcinogenic
and cytotoxic consequences of exposure to POM was attributed to the binding
of the compounds to nuclear DNA,   '    although binding occurs to RNA and
protein as well.   '     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.     In contrast, benz[a]anthracene, a weak carcinogen, inhibited
                                     6-46

-------
DMA 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
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 non-replicating
    105
DNA.     Moreover, DNA-bound DMBA moieties could still function as templates
for further DNA synthesis.     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.
     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-
                                                      259
tionships for BaP tumorigenesis.  Albert and coworkers    found that the
formation of BaP adducts was a linear function of dose between 4 and
100 nmoles/mouse.  Since BaP carcinogenesis in mouse skin was a linear
fuction between 16 and 500 nmoles/mouse, binding to DNA may allow for ex-
trapolation 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
                                                  108
exposures to a particular POM.  Lutz and coworkers    noted that over a do:
range of 40 yg/kg to 4 mg/kg (approximately 50-5000 nmoles/animals, given
                                     6-47

-------
i.p.) the binding of BaP to rat liver DNA was linear with doses up to about
1 mg/kg.  A non-linear 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
chemically, it was believed that a reactive metabolite was responsible for
the DNA-binding phenomenon.  Gelboin   and Grover and Sims    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
    31
DNA.    Subsequent studies conducted in vitro with hamster tracheal epi-
thelial cells incubated in the presence of 7,8-benzoflavone (an inhibitor of
microsomal drug-metabolizing enzymes) confirmed that BaP binding to DNA was
dependent on an intact drug-metabolizing system. '0''''  in a related study,
the binding of BaP and MCA to rat liver DNA in vitro was enhanced in the
                                         112
presence of an epoxide hydrase inhibitor.     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.     These data
implied that active epoxide intermediates are involved in the binding of
polycyclic carcinogens to DNA.  Others have argued, however, that a 6-oxy-
benzo[a]pyrene free radical may be involved in the binding of BaP to DNA and
                          38 39 40 114
synthetic polynucleotides.  '  '  '     It was suggested that metabolic
                                     6-48

-------
hydroxylation of BaP at position -6 leading to the formation of the 6-oxy BP
                                                     41
radical may be only a minor route for binding to DNA.
     The role of K-region arene oxides as reactive intermediates in the
metabolism of polycyclic hydrocarbons has been extensively reviewed,   '   '
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.     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.   '     Similar studies using DMBA-epoxides revealed that covalent
complexes formed primarily with poly G, indicating that guanosine residues in
                                                                 119
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-
       ion
workers    examined the in vitro binding of both K- and non K-region arene
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 hydro-
carbon and position of the epoxide moiety.  Significant binding to poly G
occurred with K-region arene oxides of DMBA, BaP, and MCA (all strong carcino-
gens) and benz[a]anthracene (a weak carcinogen).  K-region arene oxides of
the non-carcinogens pyrene and phenanthrene produced no changes in the absorp-
tion spectra for poly G.  However, the non K-region BaP 9,10-, BaP 7,8-, and
                                     6-49

-------
 phenanthrene  3,4-oxides also produced alterations in poly G as measured by
 changes  in fluorescence spectra.  Further investigations revealed that RNA
 adducts  formed when cultured bronchial mucosa is exposed to BaP   or when BaP
                        121
 is painted on mouse skin    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 (jO-7B,8a-dihydroxy-
'9a,10a-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (diol epoxide 2), Moore and
         122
 coworkers     found that both diol epoxide 2 and (+)-7B,8a-dihydroxy-9B,10B-
 epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (diol epoxide 1) reacted with RNA in
 vivo by  cis and trans addition.  Diol epoxide 1 arises mainly from the (+)-
 enantiomer of BaP 7,8-dihydrodiol while the (-)-enantiomer produces primarily
 diol epoxide  2 (see Section 6.2.1).  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
 monoxygenases.  Thus the K-region BaP 4,5-oxide does not appear to be re-
 sponsible for most of the observed covalent binding of BaP to RNA.  In addi-
 tion, examination of DNA adducts formed in vitro with primary cultures of
                            123 124                        113
 Syrian hamster embryo cells,   '    cultured human bronchi,    or BHK 21/C13
                125
 cells in culture    exposed to BaP 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 demonstrated
 that BaP 7,8-dihydrodiol is highly carcinogenic, and both diastereomeric BaP
                                     6-50

-------
7,8-diol-9,10-epoxides are potent mutagens (see Sections 6.4 and
6.5).66'75'72'76'126'127'128  A diol-epoxide of DMBA was also shown to be
                                                                87
involved in the binding reaction with DMA of mouse embryo cells.
     Not all investigators agree that polycyclic hydrocarbon binding to DNA
represents the essential interaction leading to mutagenesis/carcinogenesis.
                    129
Sarrif and coworkers    were able to purify an "h-protein" from C3H mouse
liver and skin which could covalently bind to metabolites of MCA.   The pro-
tein was postulated as a primary target in hydrocarbon-induced carcinogenesis,
playing a similar role to hormone receptors in transporting carcinogens 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.      This subfraction contained 15 percent of the nuclear DNA
and represented that fraction of the nuclear chromatin which is transcrip-
tionally active.  The specific macromolecule responsible for the binding
could have been, in addition to DNA, either RNA or one of the 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 investigators    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
DNA in the nucleosome.
                                     6-51

-------
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 careinogenicity 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
     132
mice.     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 mecha-
nsim 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
                           133
phase of the mitotic cycle.     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.
                                     6-52

-------
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.
         134
Yashuhira    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.     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 coworkers    provided strong support for the argument that
DMBA-induced cytotoxicity is mediated via an interaction with DNA.  Female
                                     6-53

-------
Sprague-Dawley rats receiving 300 mg/kg B.W.  DMBA orally and male rats
receiving an intravenous injection of 50 mg/kg B.W.  DMBA displayed injury to
the intestinal epithelium, extreme atrophy of the hematopoietic elements,
shrinkage 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,
                 14
incorporation of   C-labeled thymidine into DMA of small and large intestine,
spleen, bone marrow, cervical lymph nodes, thymus, and testis was signifi-
cantly 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 particu-
larly susceptible to DMBA attack.  This contention probably holds true for
other carcinogenic POM as well.
     Similarities between the effects of DMBA and ionizing radiation on the
bone marrow suggest a common cytotoxic action.  The depressive effect of
DMBA injection (5 to 60 yg) or ionizing radiation (200 to 600 rads) on the
bone marrow of newborn DFW/D mice could be shown within a few hours of
          I og
treatment.     Either of these treatments would be sufficient to induce
thymic lymphomas in nearly all exposed mice, evidence which strongly sug-
gests a relationship between mechanisms of cytotoxicity and tumor induction
which probably involve an attack on DNA.
     Another lesion which is characteristic of that produced by X-rays is
the severe testicular damage induced by DMBA in rats.     Single intravenous
                                     6-54

-------
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 spermatogonia 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 was produced by single feedings of BaP (100 mg), MCA
(105 mg), or 2-acetoaminophenanthrene (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
                                                         130
glands, hyperplasia, hyperkeratosis, and even ulceration.     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-
                            139 452
titis, hyperkeratoses, etc.,   '    and the possible significance of these
skin disorders to human cancer is not known.
     In female animals, ovotoxicity has been reported to result from
administration of POM.  DMBA was shown to cause the destruction of small
                                     6-55

-------
oocytes and reduced the numbers of growing and large oocytes after oral
                       140
administration 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
                     141
hydroxylase activity.     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
     142
rats.     Adrenal apoplexy, increased adrenal gland weight, and increased
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-
                                  143
cutaneous injections for 40 weeks.     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
                                     6-56

-------
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[a]anthracene or anthracene.
     A more detailed study of the subchronic effects of DBahA on lymph nodes
                                  144
of male rats was reported in 1944.     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
extravascular red blood cells in the lymph spaces and the presence of large
pigmented 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-
carcinogen, anthracene, on the other hand, did not produce as dramatic a
change in the lymph nodes of rats.
     Since the most 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 POM-induced neoplasms of the
respiratory tract (see Section 6.5), an understanding of early pathological
alterations and pre-neoplastic lesions in this tissue has particular signifi-
cance.
     Repeated intratracheal applications of DMBA or BaP (100 pg 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
       145
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
                                     6-57

-------
masses filling one or more air spaces.  As this proliferating alveolar
epithelium began to destroy the adjacent reticulin framework, the lesions
were considered to be malignant.  Damage to the trachea occurred in prac-
tically all animals and consisted of mucosal and submucosal chronic exuda-
tive inflammation, ulceration, hyperplasia of reserve cells, squamous meta-
                                    /
plasia, and squamous cell carcinoma (in the DMBA group only).
     In a subsequent study conducted by Reznik-Schiiller and Mohr,    BaP-
induced damage to the bronchial epithelium of Syrian golden hamsters was
examined in detail using semi thin (1 ym) 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
month of treatment, and semi thin 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
                                     6-58

-------
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
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
in hamsters.10'149'146'150'148
6.3.3  Effects on the Immune Systems
     Numerous investigators have demonstrated that carcinogenic POM can
produce an immunosuppressive effect.  This action was first observed by
Malmgren and coworkers    in 1952 using high doses of MCA and DB[a,h]A in
mice.  Subsequent studies established that single carcinogenic doses of MCA,
DMBA, and BaP caused a prolonged depression of the immune response to sheep
                1 S7 1S^
red blood cells.   '     Noncarcinogenic hydrocarbons such as benzo[e]pyrene
and anthracene reportedly had no immunosuppressive activity.  In a recent
review on immunosuppression and chemical carcinogenesis, substantial evidence
was presented to indicate that the degree of immunosuppression was correlated
with carcinogenic potency for I
reactions are affected by POM.
                                  154
with carcinogenic potency for POM.     Both cell-mediated and humoral immune
                                     6-59

-------
     The cellular mechanism of immunosuppression by POM has been investigated
                                                                   155
by several groups of researchers in recent years.   Rowland and Hurd    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 coworkers    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
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
suppression of cell-mediated or humoral immunity.      These investigators
         T
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-
                                                          154
genesis induced by viruses than that induced by chemicals.     Moreover, a
                                     6-60

-------
dissociation between the carcinogenic and inmiunosuppresive effects of MCA
                                   l *ift
was shown at low doses in C3H mice/     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.
6.3.4  Effects on Enzyme Activity
     The metabolism of most lipophilic xenobiotics is catalyzed  by the
microsomal mixed-function oxidase system.   '    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 multiple monooxygenase systems that metabolize BaP  to fluores-
cent phenols.  The activity of AHH can be induced (i.e.,  enhanced) by sub-
strate and non-substrate xenobiotics, either in vivo or in cultured mam-
             59
malian cells.    This enzyme can be found in 90 percent of the tissues
                                              159
examined in rats, mice, hamsters, and monkeys.      These  include liver, lung,
testis, muscle, bone marrow, skin, brain, intestine, placenta, and cultured
mammary gland, lymphocytes, leukocytes, and monocytes.   "     The ultimate
cytotoxicity and/or careinogenicity 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 cell lines which were
highly resistant to the toxic effects of BaP were also lacking in AHH activi-
ty.159
                                     6-61

-------
     For more than 20 years it has been known that ROM's such as BaP,  DBahA,
                                                                 59
MCA, and chrysene are potent inducers of AHH in liver microsomes.    Thus it
is not surprising that cigarette smoke has a marked stimulatory effect on
AHH in lung, liver, intestine, and human placenta.  Intragastric administra-
tion of MCA increases AHH activity of rat liver, kidney, and intestinal
       15
mucosa,   and intraperitoneal injection increases AHH activity in rabbit
bone marrow.     In cultured hamster fetus cells, Nebert and Gelboin
established that AHH activity was readily induced by carcinogenic and  non-
carcinogenic POM.  Maximum enzyme induction was generally produced with a POM
concentration of 10 yM in the culture medium (Table 6-9).  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 subcuta-
                     l fift
neously injected MCA.
     KeHerman and coworkers    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 suscepti-
bility to chemical carcinogenesis.  These investigators observed a virtual
absence of lung cancer among patients having little capacity for AHH induc-
tion.  Therefore, it may be postulated that genetic heterogeneity with
regard to microsomal enzyme induction might account for the apparently
inherited susceptibility (i.e., familial clustering) for bronchogenic  car-
cinoma.  However, difficulty in reproducibility of the assay for AHH induci-
bility in lymphocytes and monocytes, as well as large temporal and inter-
individual variability have prevented an evaluation of its genetic control
                                     6-62

-------
     Table 6-9.   ARYL HYDROCARBON HYDROXYLASE  INDUCTION  IN  CELL  CULTURE
                                                                       167
           Compound0
Amount of compound
       medium
in    AHH activity,
    units^/mg protein
Control
Benz[a]anthracene
8-Methyl benz[a]anthracene
1 0-Methyl benz[a]anthracene
7 , 1 2-Dimethyl benz[a]anthracene
3-Methyl chol anthrene
Di benz[a ,c]anthracene
Di benz[a , h]anthracene
Chrysene
Pyrene
none
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
100 yM
10 yM
1 yM
10
118
188
56
34
8
8
65
80
33
78
84
25
3
38
29
2
67
10
4
21
57
2
59
26
27
57
22
3 Compounds tested as inducers for hamster fetus cell  aryl  hydrocarbon
  hydroxylase system.  The cells were exposed to various concentrations of
.  each compound for 16 hours.
  One unit of enzyme activity is defined as that catalyzing the production,
  during the 30 minute incubation period, of an amount of fluorescence
  equivalent to that of 1  nmole of 3-hydroxybenzo[a]pyrene.
                                    6-63

-------
in humans.   '   '   '     At the present time, attempts to confirm several
aspects of Kellerman's studies have been unsuccessful.
     Recent studies with other human tissues (liver and placenta)  have
provided important new data concerning the carcinogen-metabolizing capacity
                                                                            173
of man and its implications for cancer susceptibility.   Conney and coworkers
examined individual differences in the metabolism of drugs and carcinogens
in human tissues, and have identified drugs which may serve as model  sub-
strates 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 regula-
tory control.     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 metabo-
lism of BaP and certain drugs in many subjects indicated the presence of
several monooxygenases in human liver which catalyze the oxidative metabo-
lism of these compounds.  Furthermore, large inter-individual differences
exist in the capacity of humans to metabolize foreign chemicals both ij^
vitro and in 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.
                                     6-64

-------
These results suggested the presence of multiple monooxygenases or cytochrome
P-450's in the different 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 sub-
strates.  Multiple forms of cytochrome P-450 have been shown in the livers
of rats, rabbits, and mice, but not thus far in humans.      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.
     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.    '
Because the hydration of arene oxides may lead to the  formation of dihydro-
diol 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.
                      178
     Thomson 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
environmental chemicals.  These results may be interpreted to indicate that
a chemical carcinogen may not necessarily induce its own bioactivation, but
                                     6-65

-------
instead can be transformed into a reactive intermediate by virtue of increased
AHH activity stimulated by other non-carcinogenic 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
                                                                     1 yg
involving cell proliferation is also involved in skin carcinogenesis.
Therefore, although certain aromatic hydrocarbons are effective enzyme
inducers, 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-
                                                        180
ment of this opinion has been provided by Shulte-Hermann    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 rather than strain-dependent in animals has important implications
                                                     181
for evaluating susceptibility to POM-induced cancers.     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.
                                     6-66

-------
                                                          182
     Enzyme induction by POM is not limited to AHH.   Owens    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-tetrachloro-
dibenzo-£-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 activi-
   183
ty.     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 exacer-
bated 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 glucuronida-
tion.  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.
In summary, there is a need to further explore the relative effects of
                                     6-67

-------
enzyme induction on the metabolic activation of chemicals to toxic products
versus metabolism of chemicals via detoxification pathways.
                                     6-68

-------
6.4  MUTAGENESIS
     It is presently believed that fundamental  similarities exist between
the molecular mechanisms of mutagenesis and carcinogenesis induced by
    184
POM.     These assumptions are based on the theory that highly electrophilic
metabolites (e.g., epoxides) are generated from POM by the action of micro-
                                     49
somal mixed-function oxidase enzymes.    These reactive intermediates are
capable of binding to DMA and/or other critical cellular macromolecules
                                                                  185
presumably 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 mutation) implies that a relation-
ship exists between mutagenesis and carcinogenesis.     Therefore, it is
believed that an investigation of the mutagenicity of POM:  (1) may be
predictive of carcinogenic potential, (2) may help elucidate the mechanism
for malignant transformation in certain systems and (3) may indicate the
presence of a potential threat to human health posed by a particular POM.
6.4.1  In Vitro Studies
6.4.1.1  Microbial systems—Early attempts to demonstrate mutagenicity by
POM failed because of the inability of microorganisms to metabolize these
                                  99
compounds to reactive derivatives.    The addition of subcellular prepara-
tions and the development of sensitive tester strains of Salmonella typhi-
murium have resulted in a mutagenesis test system (Ames assay) now receiving
wide application as a powerful screening tool for evaluating environmental
       186
agents.     The most commonly employed tester strains of S. typhimurium
used when working with POM are those designated TA98, TA1537, and TA1538.
                                     6-69

-------
All of these strains are histidine-dependent and which can be reverted to
histidine independence by mutation.  When a large series of polycyclic
aromatics was tested in this system, a high correlation was found between
carcinogenicity in mammals and mutagenicity to S. typhimurium in the presence
of microsomes.  When a limited number of noncarcinogens were tested, the
number of revertant colonies produced was typically less than 0.01  revertants
per nmol of test substance.  Mutagenicity in the Salmonella/microsome
system displayed a high correlation with carcinogenicity.  Teranishi and
         187
coworkers    also found that a quantitative correlation existed between
carcinogenicity and mutagenicity of seven polycyclic hydrocarbons when
strain TA1538 was tested using rat liver enzymes.
     In discussing the significance of the Salmonella/microsome assay,
                    188
McCann and coworkers    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 DNA 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
                                      189
in vitro with carcinogenicity in vivo.     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
mutagenicity and carcinogenicity was obtained.  Studies conducted in the
United States on 102 compounds tested in vitro using S.  typhimurium generally
                                                                 190
confirm the high capability of this system to detect carcinogens.      In
addition, a further improvement in the detection of carcinogens could be
                                     6-70

-------
obtained by combining the results from Salmonella with those from polymerase
A-deficient E. coli.  On the other hand, the use of S. typhimurium in vivo
(host-mediated assay) was considerably less reliable for the detection of
                                              190
carcinogenic hydrocarbons and aromatic amines.
     In recent studies, the Ames Salmonella test system has been employed
to assess the mutagenicity of organic extracts of airborne particulate
           204
pollutants.     Although the total extract displayed marked mutagenicity,
assays of various subfractions of the extract indicated that much of the
activity is due to compounds other than polycyclic hydrocarbons and which
do not require metabolic activation.   These results, which are supported by
data from in vivo studies (see Section 6.5.1), indicate that caution must
be exercised in attributing the mutagenic/carcinogenic effects of air
pollutants solely to the presence of polycyclic hydrocarbons.  The recent
demonstration of the mutagenicity of quinoline and its 5-hydroxy- and 8-
hydroxy-derivatives in the Ames assay emphasizes the potential role of
                                                      205
nitrogen-containing ROM's as 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 BaP metabolites that are responsible
for its mutagenicity and carcinogenicity.  It had been previously demonstra-
                                                              191
ted that epoxides of carcinogenic POM are frameshift mutagens.     In 1975,
                                                  127
it was established that the K-region BaP 4,5-oxide    and non K-region BaP
        192
7,8-diol    (with metabolic activation) were more mutagenic than the parent
                                     6-71

-------
compound in S. typhimurium strains TA TOO and TA1538.  Subsequently, nearly
all of the possible primary oxidative metabolites of BaP were synthesized
and tested for mutagenicity.70'72'193'126'71   The results of these tests
have been recently reviewed  '  '    and are summarized in Table 6-10.   It
was emphasized that diol epoxide-1 ( (±)-78,8a-dihydroxy-9B,10e-epoxy-7,8,
9,10-tetrahydro BaP) is a very potent mutagen in S. typhimurium,   and  that
the BaP 7,8-diol-9,10-epoxides are likely candidates as the ultimate bio-
logically reactive metabolites of BaP.    It was subsequently demonstrated
by these researchers that the mutagenicity of the optical enantiomers of
the diastereomeric BaP 7,8-diol-9,10-epoxides is highly stereospecific.
These results are of considerable significance in light of evidence indica-
ting that the in vitro metabolism of BaP is stereoselective, depending  upon
                                                              i pc 9(")1
the source of the microsomes and conditions of the incubation.   '
     Further evidence implicating the obligatory role of epoxide inter-
                                                                        195
mediates in the mutagenicity of BaP was reported by Oesch and coworkers.
They stated in support of this contention that:  (1) epoxide hydrase inhibi-
tors 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 coworkers    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
                                      6-72

-------
     Table 6-10.  COMPARISON OF INHERENT MUTAGENIC ACTIVITY OF THIRTY BaP DERIVATIVES IN Salmonella
C*J
typhimurium TA98 AND

Compound
Diol epoxide-1
Diol epoxide-2
H4 9,10-epoxide
H« 7,8-epoxide
BaP 4,5-oxide
BaP 7,8-oxide
BaP 9,10-oxide
BaP 11,12-oxide
6-HOBaP
12-HOBaP
1-HOBaP
3-HOBaP
2-, 4-, 5-, 7-, 8-, 9-, 10-, 11-HOBaP
BaP 1,6-, 3,6-, 6,12-, 4,5-, 11 ,12-quinone
BaP 4,5-, 7,8-, 9,10-, 11 ,12-dihydrodiol
BaP
IN CHINESE HAMSTER V79 CELLSa 69
Relative °i
Strain TA98
100
35
95
10
20
1
1
0.5
5
1.5
0.5
0.5
<0.1
<0.1
<0.1
<0.1

i activity
V79
40
100
40
0.2
1
<0.1
<0.1
1
0.3
<0.1
0.1
<0.
<0.
<0.
<0.
<0.
       The relative  percent mutagenic  activities are approximations  since the data were compiled from
       several  separate  studies  conducted at different  times.   In  some experiments, BaP 7,8-dihydrodiol
       was 0.1-0.4%  as active  as diol  epoxide-2 in V79  cells.

       Abbreviations used:  BaP,  benzo[a]pyrene; 1-HOBaP,  l-hydroxybenzo[a]pyrene; 2- to 12-HOBaP,
       other BaP phenols;  BaP  1,6-quinone,  benzo[a]pyrene  1,6-quinone; BaP 3,6-quinone, BaP 4,5-
       quinone,  BaP  6,12-quinone,  and  BaP 11,  12-quinone,  other BaP  quinones; BaP 4,5-dihydrodiol,
       trans-4,5-dihydroxy-4,5-dihydrobenzo[a]pyrene; BaP  7,8-,  9,10- and 11, 12-dihydrodiol, other
       dihydrodiols  of BaP; BaP  4,5-oxide,  benzo[a]pyrene  4,5-oxide; BaP 7,8-, 9,10- and 11,12-
       oxide, other  BaP  oxides;  diol epoxide-1  (+)-7B,8a-dihydroxy-98,106-epoxy-7,8,9,10-tetrahydro BaP;
       diol  epoxide-2, (jO-7B,8a-dihydroxy-9a,10a-epoxy-7,8,9,10-tetrahydro BaP; H4 9,10-epoxide, 9,10-
       epoxy-7,8,9,10-tetrahydro BaP;  H. 7,8-epoxide, 7,9-epoxy-7,8,9,10-tetrahydro BaP.

-------
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.  Further-
more, these researchers established that epoxide hydrase converts BaP 7,8-
oxide to BaP 7,8-dihydrodiol, which is further metabolized by the monoxy-
genase 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.
6.4.1.2  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-aza-
guanine, 6-thioguanine, ouabain).  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 do not have the capacity for
metabolic activation of POM.197
                                I go
     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
potentcies.  Large numbers of 8-azaguanine-resistant mutants were produced
by treatment with as little as 0.1 yg/ml  (Table 6-11).  The non-carcinogen
benz[a]anthracene produced no mutations in this system.

                                      6-74

-------
Table 6-11.   MUTAGENICITY OF CARCINOGENIC HYDROCARBONS TO CHINESE HAMSTER  V79  CELLS  CO-CULTIVATED
                              WITH SYRIAN GOLDEN HAMSTER EMBRYO CELLSa 198
Con
Treatment
Solvent
7 , 1 2-Dimethyl benz[a ]anthracene
-lj 3-Methylcholanthrene
in
Benzo[a]pyrene
Benzo[a]pyrene
a The cul tures were treated for 2 days
and seeded to determine the cloning
resistant mutants: co-cultivation wa
centration
0
0.1
0.1
0.1
1.0
with carcinogenic
efficiency of the
s with 2 x 10-6 ir
Number of V79
x 10'6
3.0
1.3
2.3
2.7
1.9
hydrocarbons.
V79 cells and
radiated qolde
Cloning efficiency
• of the V79 cells,
99
7
78
89
15
Number of
8-azaguanine-
resistant
mutants/105
survivors
7
270
110
70
320
The cells were then trypsinized
the frequency of 8-azaguanine-
n hamster cells oer Petri dish.

-------
                       199
     Huberman and Sachs    subsequently conducted expanded studies on the
mutagem'city 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
temperature-sensitive cells.  Tables 6-12 and 6-13 summarize their results
and indicate that DMBA was consistently the most potent mutagen tested,
having significant activity at concentrations as low as 0.01 yg/ml.
     The V79 cell system without the addition of metabolizing cells  has
been used to identify structures which may be the ultimate mutagenic/carcino-
genic metabolites of BaP.70>66'72>71>69>67>75  Numerous oxidative metabolites
of BaP were surveyed, and the stereoisomeric BaP 7,8-diol-9,10-epoxides
(diol expoxides 1 and 2) were found to be highly mutagenic (Table 6-10).
The very short half-life (30 seconds) for diol epoxide 1 in tissue culture
medium made comparisons of activity between the two diol epoxides very
difficult, however.  Nevertheless, the high mutagenicity of diol epoxides 1
and 2, coupled with the high carcinogenicity 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 coworkers    combined baby Syrian hamster kidney cells (BHK21)
                                     6-76

-------
         Table  6-12.   INDUCTION  OF  OUABAIN-  AND 8-AZAQUANINE-RESISTANT MUTANTS BY  DIFFERENT CHEMICAL
crt
i
Treatment
Solvent
Pyrene
Phenanthrene
Chrysene
Benz[a]anthracene
Dibenz[a,c]anthracene
Dibenz[a,h]anthracene
7-Methyl benz[a]anthracene
3-Methyl chol anthrene



Benzo[a]pyrene

7 , 1 2-Dimethyl benz[a]anthracene

Cloning
efficiency,
%
92*
94*
79*
85*
92*
95*
79*
61*
49**
41 1
38*
47**
27t
21*
50 §
2**
Ouabain-resistant
mutants
per 10^ survivors
1
1
1
2
2
3
4
24
14
38
108
18
45
121
22
66
8-Azaguanine-
resistant
mutants per
10$ survivors
6
5
8
9
9
22
17
75
72
152
366
54
128
425
41
281
      3  The  cells  were  treated  for  2  days  with the  polycyclic hydrocarbons with *1  ug/ml,  tO.3
        **0.1  pg/ml,  or §0.01 ug/ml.   After  treatment there were with 0.01-0.1  pg/ml  of DMBA,  0.3-1  ug/ml
        of BaP,  and 0.3-1  pg/ml  of  MCA,  1.0  to 2.5  x 106 V79 cells per petri  dish.   In all  other cases,
        there  were 2.7  to  3.6 x  106 V79  cells  per petri  dish.

-------
Table 6-13.  INDUCTION OF TEMPERATURE-RESISTANT MUTANTS IN CHINESE HAMSTER
               OVARY TMPERATURE-SENSITIVE CELLS BY DIFFERENT CHEMICAL
                                 CARCINOGENS 199
Concen-
tration,
Treatment yg/ml
Solvent
Pyrene
Phenanthrene
Benz[a]anthracene
Benz[a]pyrene


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
Cloning
efficiency,
93
88
88
78
67
45
9
78
32
10
65
3
Temperature-
resistant
mutants per
10^ survivors
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
 the frequency of mutations, there were after treatment with 0.01-0.1  yg/ml  of
 DMBA, 0.3-1 yg/ml of BaP or MCA, 0.5 to 1.9 x 106 ts cells per petri  dish.
 In all other cases,  there were 2.1  to 2.3 x 10& ts cells per petri dish.
                                    6-78

-------
for metabolism of POM with V79 cells for detection of ouabain- and 8-
azaguanine-resistant mutants.  This system produced binding reactions
between DNA and BaP or 7-methylbenz[a]anthracene which accompanied mutagene-
sis 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 inter-
action.  These results strengthened the argument which supports the role of
diol epoxides in BaP-mediated mutagenesis and carcinogenesis.   Furthermore,
the cell-mediated 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
S. typhimurium in providing a rank-order correlation with carcinogenic
potency in mammals (e.g., Iball  index).  However, additional screening of
polycyclic hydrocarbons and 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
                                  202
studies of Huberman and coworkers.     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
                                     6-79

-------
                                    203
containing microsomes and cofactors.     Mutants of V79 cells were detected
by their resistance to the cytotoxic effects of 6-thioguanine.  The muta-
genic activity of BaP, MCA, DMBA, and benz[a]anthracene in this system
showed a limited correlation with their respective carcinogenic potencies.
With DBahA and DBacA, an inverse relationship between mutagenicity and
carcinogenicity was obtained.  Such a relationship has also been shown for
                                                                    186
the dibenzanthracenes when tested in the Salmonella/microsome assay.
Although mammalian cell-mediated mutagenesis is hindered because it is
technically more difficult, 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.
     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,
                                             206
MCA, and DMBA in the rat, mouse, and hamster.     However, in cultured
human leukocytes exposed to DMBA, chromosome damage was not the same as
that produced in hamster cells.  Although it is argued that chromosome
                                             207 208       209 107
changes in POM-induced tumors are non-random,   '    others   '    claim
                                     6-80

-------
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 rate of SCE's could be produced by BaP in cultured
                 210 211
human lymphocytes   '    but this was not correlated with different  rates
                  210
of BaP metabolism;    a surprising result in light of the known importance
of metabolic activation for BaP mutagenicity.   BaP-induced SCE's rates did
not differ between lymphocytes taken from normal humans and those from
                          211
patients with lung cancer.     In recent studies with cultured Chinese
hamster cells exposed to DMBA, BaP, and MCA, none of the chemicals produced
                                                                212
chromosome breaks and only DMBA could successfully induce SCE's.
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 vivo
induction of sister chromatid exchanges, the latter test proved to be the
                                                         213
most sensitive with carcinogenic polycyclic hydrocarbons.     However,
positive results were also obtained with phenanthrene, and thus limits the
usefulness of sister chromatid exchange as a screening technique for carcino-
gen detection.  BaP was positive in the sister chromatid exchange test, but
                                     6-81

-------
weakly active in the chromosome aberration test, and negative in the micro-
nucleus test.  On the other hand, DMBA was clearly positive in all  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
                                         214
also be detected by examining DMA repair.     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
                                     3
Indeed, unscheduled incorporation of  H-thymidine into nuclear DNA of
normal human cells exposed to epoxides of benz[a]anthracene and MCA has
              214
been observed.     However, since a metabolic activation system was not
present in this system, the parent hydrocarbons showed no activity.  More
recent studies confirmed that K-region epoxides of BaP, DMBA, and DBahA
caused DNA damage in human skin fibroblasts which was repaired with the
                                                                        215
same system used for repairing lesions induced by ultraviolet radiation.
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 mutagenic effect by inducing DNA lesions
which require any kind of repair synthesis at all.
                                     6-82

-------
                                                             21 fi ?1 7 ?1 ft
     Nevertheless, DNA repair synthesis in human fibroblasts,   '   '
                219                              220
rat liver cells,    and Chinese hamster V79 cells    has been successfully
used for the detection of chemical carcinogens, including numerous ROM's.
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.
6.4.2  In Vivo Studies
6.4.2.1  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 differ-
                                                207- 221
ent from sarcomas induced by Rous sarcoma virus.   '     The chromosome
patterns of DMBA- induced sarcomas were found to be identical with those
observed in primary rat leukemias    and primary carcinomas of the auricular

                                                                          223
    o no
skin    induced by DMBA.  Consistent chromosome abnormalities in DMBA-
induced rat 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
                                                                       224
showed 41 chromosomes in the modal cell lines instead of the normal  40.
        ppc
Sugiyama    obtained an apparent relationship between sarcomatogenic potency
and the incidence of chromosome breakage in rat bone marrow cells produced
                                     6-83

-------
by a single intravenous injection (50 mg/kg B.W.) of various benz[a]-
anthracene 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 metaphase 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
                        ope 997
karyotypic compositions.   '     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
             pop ppq
constitution.    '     Mice treated with 30 yg DMBA,  a dose sufficient to
produce a 100% incidence of thymic lymphomas, did not reveal an excess of
                                                  230
chromosome abnormalities in bone marrow or thymus.     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 ng) or DMBA (0.1 mg),  and cultured cell
populations derived from these tumors, failed to reveal common karyotypic
        231
changes.     Tumor cells had subdiploid, diploid, and hypotetraploid chromo-
some constitutions; further karyotype rearrangements occurred with subse-
quent growth jjrc v i tro.
                                     6-84

-------
     In humans, the presence of the "Philadelphia" chromosome in myeloid
leukemia appears to be the only example of a human chromosome abnormality
                        232
which is tumor-specific.     In ROM-induced experimental  tumors, lymphatic
leukemia in mice produced by DMBA also displays consistent chromosome
              226
abnormalities.     Beyond this common feature, convincing data have not
been presented to indicate that somatic cells exposed to  POM may suffer
characteristic 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, infec-
                                 226
tion, anoxia, lack of nutrition).
     Evidence has not been encountered in the published literature concern-
ing the likelihood of POM-induced somatic mutation in the absence of neo-
plastic transformation.
6.4.2.2  Effects in germinal tissues—The fruit fly,  Drosophila melanogaster,
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
                                                     233
mutagenic or carcinogenic in their administered form.     The use of Drosophila
has been suggested as a valuable bridge between microbial and mammalian
assays.238
     A series of studies was undertaken by Fahmy and Fahmy which demonstra-
ted conclusively that carcinogenic polycyclic hydrocarbons were selectively
                                     6-85

-------
mutagenic to the male germ cell line of Drosophila.234'235'236'237  Muta-
genicity was restricted to the t- and rRNA loci, yielding Minute (M) and
bobbed (bb) mutants, respectively.  The activity of benz[a]anthracene, 7-
methylbenz[a]anthracene, MCA, DMBA, their epoxides, and some other deriva-
tives was determined in tests of both general mutagenicity on the whole
                                                              237
genome and specific locus mutability on the t- and rRNA genes.      The
overall yield of sex-linked recessive lethals and visibles (general muta-
genicity 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 (Table 6-14).  In the specific locus
test, DMBA, MCA, and 7-methylbenz[a]anthracene produced significant increases
in the yields of phenotypic bb plus M and transmitted bb's, whereas benz[a]-
anthracene did not induce significant numbers of mutations on the RNA genes
(Table 6-15).  Thus, mutagenicity was correlated with carcinogenicity,
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 K-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
                                     6-86

-------
                           Table  6-14.   MUTAGENIC  RESPONSE  OF  THE X CHROMOSOMES IN Drosophila
                                                                                                                   a  237
cr>
oo
Compound
Controls

Benz[a]anthracene


7-Methylbenz[a]anthracene

7,12-Dimethylbenz[a]anthracene

3-Hethylcholanthrene

5,6-Epoxy-5,6-dihydrobenz[a]anthracene

8,9-Epoxy-8,9-dihydrobenz[a]anthracene
5.6-Epoxy-5,6-dihydro-7-methylbenz[a]-
anthracene

5,6-Epoxy-5,6-dihydro-7,12-dimethyl-
benz[a]anthracene

ll,12-Epoxy-ll,12-dihydro-3-methylchol-
anthrene

Benz[a]anthracene-cis-5,6-dihydro-5,6-diol
7-Bromomethylbenz[a]anthracene

Dose,
mM


1.0
5.0
10.0
5.0
10.0
1.0
5.0
10.0

1.0
5.0
1.0

5.0
10.0
1.0
5.0

5.0
10.0
5.0
1.0
5.0
Gametes
tested
3707
A
3790
3473
1773
1703
1709
1833
1753
2399
B
1639
1606
1588

2623
1709
1725
1710

1779
1695
1891
1890
3260
Lethals
Number
6(1 )C
Parent hydrocarbons
9(2)
12
7(2)
5
4(2)
7(3)
5(1)
5(3)
Reactive derivatives
10)
12(2)
3

12
14(2)
3
11(1)

12(1)
12(1)
4
5
15(1)
Visibles
Per 103
1 6

2.4
3.5
3.9
2.9
2.3
3.8
2.9
2.1

0.6
7.5
1.9

4.6
8 2
1.7
6.4

6 7
7.1
2.1
2.6
4.6
Number
2

2
4(2)
3
0
0
0
0
2(1)

5(2)
4(1)
0

6(1)
7
KD
0

3
3
0
1
5
Per 103
0 5

0.5
1.2
1.7
0 0
0.0
0.0
0.0
0.8

3.1
2.5
0.0

2.3
4.1
0.6
0.0

1.7
1.8
0.0
0.5
1.5
Total X ,
mutations/10
2.2 + 0.8

2.9 + 0 9
4.6 +" 1 1
5.6 + 1.8
2.9 +" 1.3
2.3 + 1.2
3.8 7 1.4
2.9 + 1.3
2.9+1 1

3.7 + 1.5
10.0 + 2.5
1.9 + 1.1

6.9 + 1.6
12.3 +27
2.3 + 1.2
6.4 +_ 1.9

8.4 + 2.2
88 + 23
2.1 + 1 1
3.2 + 1 3
6.1 + 1 4
        a General  mutagenicity of various benz[a]anthracenes and their reactive derivatives  as indicated by the yield of sex-linked
          recessive lethals and visible;
        b Response with administration vehicle only; 21 dimethylformamide in arachis oil
        e Mosaic mutants are entered in parentheses.

-------
                                Table  6-15.    MUTAGENIC  RESPONSE OF THE  BB  LOCI  IN Drosophila
                                                                                                                 a  237
en
i
00
00
Compound
Controls
Dose,
mM
Gametes
Phenotypic bb + M
Transmitted bb
tested Number Per 103 Number Per 103
124.724 156(7)c 1.3 + 0.1 45(1) 0.4 +.0.1
A. Parent hydrocarbons
Benz[a]anthracene


7 -Methyl benz[a]anthracene

7 , 1 2-Dimethyl benz[a ]anthracene

3-Methylcholanthrene
1.0
5.0
10.0
5.0
10.0
1.0
5.0
10.0
44,314 49(13) 1.1+0.2 61
36,854 73(10) 2.0 + 0.2 24 4
9,986 14(1) 1.4+0.4 9 !
7,798 34
10,804 52 5
29,204 259 7
16,887 134 8
18.263 86 5
4.4 +0.7 93
4.8 + 0.7 14(1
8.8 + 0.5 182(4
7.9 + 0.7 97(£
) 0.1 + 0.1
) 0.7 +0.1
) 0.9+0.3
1.2 + 0.4
1.3 + 0.3
6.2 + 0.5
5.7 + 0.6
4.7 + 0.5 62(5) 3.4 + 0.4
B. Reactive derivatives
5,6-Epoxy-5,6-d1hydrobenz[a]anthracene

8,9-Epoxy-8,9-dihydrobenz[a]anthracene
5,6-Epoxy-5,6-dihydro-7-nethylbenz[a]-
anthracene

5.6-Epoxy-5,6-dihydro-7,12-dimethyl-
benz[a]anthracene

ll,12-Epoxy-ll,12-dihydro-3-methylchol-
anthrene

Benz[a]anthracene-cis-5,6-dihydro-5,6-diol
7-Bromomethylbenz[a]anthracene
1.0
5.0
1.0

5.0
10.0

1.0
5.0

5.0
10.0
5.0
5.0
14,426 27(4) 1.9+0.4 1(1) 0.1+0.1
4,626 7(2
16,078 23(4

1.5 + 0.6 1
0.2 + 0.2
1.4 TO. 3 3(2) 0.2+0.1


32,468 56(7) 1.7 + 0.2 18(3) 0.6 + 0.1
7,828 28

3.6 + 0.7 12(2) 1.5 + 0.4

23,660 44(2) 1.9+0.3 18(1
6,356 38(2) 6.0+1.0 29(4

19,326 25(1
30.995 63(4
19,870 26(5

1.3 + 0.3 18
2.0 + 0.3 28 1
1.3 +0.3 52

0.8 + 0.2
4.6 TO. 8

0.9 + 0.2
0.9 + 0.2
0.3 + 0 1
21,233 26(3) 1.2 + 0.2 4(2) 0.2 + 0.1
          The yield of transmissible bb's among the phenotypically scored bristle mutations with various benz[a]anthracenes and their reactive derivatives

          Response with administration vehicle only, 2S dimethylformamide in arachis oil.

        c Mosaic mutants are entered in parentheses.

-------
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 correla-
tion 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
                                        239                       240
this test is not particularly sensitive.      Kennedy and coworkers
administered 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
subsequently mated to virgin females.  The females were sacrificed after
one week 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.
                                                        241
     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 6-16).  The authors noted a lack of con-
sistency in the results obtained, both in terms of dose (for MCA) and in
magnitude of response and parameters affected (for BaP).  Nevertheless, in
control animals, the mean weekly pregnancy rate was 66 percent, the mean
number of total implants per pregnancy was 11.5 to 11.9, and mean early
                                     6-89

-------
        Table  6-16.   AGENTS PRODUCING EARLY FETAL DEATHS AND/OR PREIMPLANTATION LOSSES SIGNIFICANTLY BEYOND
                        CONTROL LIMITS BUT REQUIRING  FURTHER  VERIFICATION BECAUSE OF LACK OF INTERNAL
                                                 CONSISTENCY241
(ft

o




Benzo[a]pyrene








20-Me thyl chol anthrene


Route Total
and frequency weeks
of of
administration mating
ip 3




ip 3

ip 8

1p 8




Doses,
mg/kg
500



1000
750
1000
750
1000
100

200
No. Of
males
(No. of
deaths)
5(1)



5
14
14
7(4)
9(2)
7(2)

9(1)
Parameters selected in screen (mating weeks)

%
Pregnancy
25(2)



13(3)
—
29(3)
<30(4-8)
<30(4-8)
19(1)

—

Implants/ % Pregnancy
pregnancy ED /pregnancy with ED
__. -_ _ ___



7.5(3) 1.00(1), 1.00(2) 60(1);67(2)
... — —
53(3)
7.6(3);6.3(4)
4.0(6)
	 — —

8.2(7)


Comments
Lack of
consistency
between
experiments





Dose reversal
in other weeks

       a ED, early deaths.

-------
fetal deaths per pregnancy were almost always less than 0.95.  Thus, con-
siderable differences are evident between ROM-treated groups and untreated
controls.
     Additional evidence which may indicate a potential for POM-induced
                                                                  242
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)F,  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.
                                     6-91

-------
6.5  CARCINOGENESIS
6.5.1  l*i Vivo
     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
microgram quantities of POM.  Latency periods can be short (4-8 weeks) and
the tumors produced may resemble human carcinomas.  Carcinogenesis studies
involving 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.
     Concern over potential human cancer risk posed by POM's present in the
atmosphere stems from studies demonstrating that crude extracts of airborne
                                                243-246
particulate matter were carcinogenic to animals.         Fractions soluble
in benzene or benzene-methanol produced tumors in mice by skin painting or
subcutaneous injection.  Both the aromatic and oxygenated neutral sub-
fractions were active as complete carcinogens, and indicated the presence
of numerous carcinogenic materials, including non-POM's.  The further
demonstration that the basic fraction of organic pollutants was carcino-
genic to newborn mice (see Section 6.5.1.4) indicated the presence of aza-
heterocyclic hydrocarbons.  Since the carcinogenicity of the total organic
particulates and aromatic neutral subfractions could only partly be ex-
plained by the presence of BaP, its usefulness as a measure of carcinogenic
risk may, therefore, be limited.
                                     6-92

-------
     From investigations in which polycyclic carcinogens were painted on
the skin of mice has emerged the two-stage theory of skin carcinogene-
    247 248
sis.   '     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]anthra-
cene, 1-methylchrysene, 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
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, thus complicating the inter-
pretation of experimental data.
     The carcinogenic effects of POM when applied to the skin of animals
                                            249
have been known for decades.  In 1939, Iball    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 6-17, 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
on agents tested for carcinogenicity has more more recently been published
                                     6-93

-------
                      Table 6-17.   CARCINOGENIC COMPOUNDS IN DESCENDING  ORDER OF POTENCY
                                                                                            249
er>
i
10
-p.

1.
2.
3
4

5.
6
7.

8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18
19.
20.
21.
22
23.
24.
Compound
7 ,1 2-Dimethyl benz[a ]anthracene
3-Metnylcholanthrene (a)
3-Methylcholanthrene (b)
3-lethylcholanthrene
(a ana b added together)
BenzoLajpyrene (from pitch)
Benzo[a]pyrene (synthetic)
BenzoLajpyrene
(5 and 6 added together)
Cholanthrene
5,6-cycloPenteno-benz[a]anthracene
2 -Methyl -benzo[c]phenanthrene
10-Methyl-benz[a]anthracene
5 ,6-Dimethyl -benz[a]anthracene
6-isoPropyl-benz|.a]anthracene
Dibenzo[c,g]carbazole
Dibenzo[a ,h]pyrene
5 -Methyl -benz[a]anthracene
5-Ethyl-benzj.a]anthracene
DibenzLa,h]anthracene
Benzole ]phenanthrene
Dibenzo[a,g]carbazole
5-n-Propyl-benz[a]anthracene
Dibenz[c,h]acridine
3 -Methyl -dibenz[a ,h]anthracene
Dibenz[a,h]acridine
Number of
nice alive
when first
tumor appears
20
18
8

26
10
9

19
49
14
16
18
19
IS
19
17
8
9
65
18
9
20
28
25
25
Number of
tumors
13
18
5

23
10
7

17
28
13
12
12
16
11
9
10
7
7
41
12
4
6
11
7
6
Percentage of
tumors (A)
65
100
62.5

88.5
100
78

89.5
57
93
75
66.5
84
73.5
47.5
59
87.5
77.5
63
67
44.5
30
39.3
28
24
Papilloma
6
1
0

1
2
2

4
5
1
5
2
0
1
4
0
2
2
8
5
1
3
2
1
2
Epithelioma
7
17
5

22
8
5

13
23
12
7
10
16
10
5
10
5
5
33
7
3
3
9
6
4
Average latent
period (B)
43
99
151

109
127
109

119
112
194
155
147
220
204
143
205
317
285
239
387
263
192
357
325
350
Index
(A/B x 100)
151
101
41

80
79
72

75
51
48
48
45
38
36
33
29
28
27
26
17
17
16
11
9
7
                                  Totals
                                                  305
                                                                       60
                                                                                 245

-------
                                 29
by the U.S. Public Health Service   which lists the results of tests on
hundreds of polycyclic 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.
6.5.1.1  Structure-activity relationships—Following the identification of
the first carcinogenic hydrocarbon from soot, BaP, an intensive effort was
                                                                      250
mounted to isolate the various active components of carcinogenic tars.
From the earliest studies conducted, the realization emerged that carcino-
genic POM are structurally derived from the simple angular phenanthrene
        251
nucleus.     However, unsubstituted POM with less than four condensed rings
that have been tested have not shown tumorigenic activity.  Furthermore, of
the six possible arrangements with four benzene rings, only two of these
compounds are active:  benzo[c]phenanthrene and benz[a]anthracene.  The
                                    6-95

-------
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, dibenzo[a,l]pyrene, dibenzo[a,e]pyrene, benzo[b]fluoran-
thene, and benzo[j]fluoranthene.  Somewhat less potent as carcinogens are
the dibenzanthracenes and dibenzophenanthrenes.  Only a few heptacyclic
hydrocarbons show carcinogenic activity.  These include phenanthro-
[2',3':3,4']pyrene peropyrene and dibenzo[h,rst]pentaphene.  Beyond seven
unsubstituted aromatic rings, there are very few known carcinogenic hydro-
carbons.  However, many physical-chemical and enzymatic parameters must be
dealt with when speaking of carcinogenic POM.  Factors such as solubility
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, acenaphthan-
thracene, 1,2,5,6-tetrahydrobenzo[j]cyclopent[f,g]aceanthrylene, and
"angular" steranthrene.  It should be noted that 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
                                     6-96

-------
                       251
regard, Arcos and Argus    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.  Carcinogenic trimethyl- and tetramethyl-
benz[a]anthracenes are known, and their relative potencies are comparable
to the parent 7,12-DMBA.   In general, free radical synthesis of polycyclic
hydrocarbons by pyrolysis does not favor alkyl side chain formation.
     Alkyl substitution of partially aromatic condensed ring systems may
also add considerable carcinogenic activity.  The best examples of this
        icti
         252
                                                       251
type of activation are MCA, a highly potent carcinogen,    and 5-methyl-
chrysene.
     With alkyl 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
                                     6-97

-------
indicative of a once widely-held view for most polycyclics that high
reactivity of the meso-phenanthrenic region (now called the "K-region") was
                                           251
a critical determinant for carcinogenicity.     Current studies show that
the K-region is not involved in critical binding to DNA (see Section 6.2.2).
Alkyl groups notwithstanding, the substitution of highly polar groups
(e.g., -OH, -COOH) in the 7-position of benz[a]anthracene abolishes tumori-
genic activity whereas a wide variety of less-polar substituents can enhance
activity in position 7 (e.g., -CH2OH, -CH2CH2OH, -CH2COONa, -CH2COOCH3,
-CH2OOCCH3, -CN, -CH2CN, -CHO, -NHg, -SH, -COCC13> -OCHg).
     Recent studies have indicated that methylation of the angular "bay
region" (see Section 6.2.1.2) benzene ring not only in benz[a]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
frequently increases, carcinogenicity.  For example, 7- and 8-methyl-BaP
are inactive, whereas 2-,3-,4-,5-,6-,11-, 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
                                     6-98

-------
                                     251
comparison to the parent hydrocarbon,    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 the polycyclic hydrocar-
     251
bons.     Nevertheless, several generalizations can be derived from availa-
                          251
ble data.  Arcos and Argus    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]acri-
dine or position 9 for benz[a]acridine).  Lengthening of the alkyl  sub-
stituent 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 careinogenicity.
     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 di-
benzo[c,g]carbazole.  However, dibenzo[a,g]carbazole also shows some car-
cinogenic activity.  The activity of dibenzo[a,g]carbazole, however, is
                                     6-99

-------
considerably less than for the isosteric benzopyridocarbazoles.   Further-
more, although dibenzo[a,i]carbazole shows only very weak activity, intro-
duction 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 impor-
tance as potentially hazardous environmental  pollutants is likely to
increase in the future.  Greater emphasis on  alternatives to solid fuel
combustion (e.g., coal liquefaction) is expected to result in greater
emissions of these nitrogen-containing compounds.
6.5.1.2  Skin carcinogenesis and induction of local sarcomas
6.5.1.2.1  Benzo[a]pyrene.  The ability of BaP and its various metabolic
products to initiate skin tumors in 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 princi-
ple ultimate carcinogen resulting from metabolic activation.  Various BaP-
oxides, -phenols, -dihydrodiols, and diol-epoxides were assayed for tumori-
genicity and compared to the parent compound  as a measure of bioactiva-
t.on 73,79,253-258
     The tumor response obtained when BaP alone is applied weekly to the
                                     259
skin of mice is shown in Figure 6-11.     It  is apparent from the data pre-
sented that increasing weekly doses of BaP caused a shortening of the
latency period for carcinoma formation.  Furthermore, it was determined
that the development of papillomas as a precursor lesion to carcinoma
                                     6-100

-------
                     1.4
                     1.0
                  3
                  o
                  2
                  3
                     0.1
                    0.01
                         I Slope = 8.0
                                          Slope = 7.8
                                          Slope = 10.0
                                              • 128 jug
                                              • 64/ug
                                              A 32 jug
                                              o 16jug
                                           I  I  i  i i  I
                       100
                                     Time (days)
1000
Figure 6-11.  Log-log  plot  of the cumulative incidence of skin carcinomas  per
mouse versus time.   BaP  was applied topically every week at the  indicated
doses.  Time was measured from the first BaP treatment.259
                                     6-101

-------
formation occurred only at higher BaP doses (e.g., 32 pg and 64 pg per
week).  At the lower dose levels (8 yg and 16 pg per week), carcinomas
appeared de novo without precursor papilloma formation.
     Among the arene oxides of BaP tested as complete carcinogens by
repeated topical applications to mouse skin (0.4 pmole once every two weeks
                                                                 73
for 60 weeks), BaP 7,8-oxide was shown to be highly carcinogenic.    At
lower doses (0.1 pmole), BaP 7,8-oxide was not as potent a carcinogen as
BaP.  The K-region BaP 4,5-oxide was only weakly carcinogenic at high
doses, whereas BaP 9,10-oxide was inactive.  Subsequent studies to deter-
mine if the phenolic non-enzymatic rearrangement products of BaP oxides
were carcinogenic established that none of the seven phenols of BaP tested
                                                              256
(4-,5-,6-,7-,8-,9-, and 10-hydroxybenzo[a]pyrene) were active.     Among
the remaining five possible isomeric phenols of BaP, only 2-hydroxybenzo[a]-
                                                                 257
pyrene was shown to be highly carcinogenic at the 0.4 pmole dose.     How-
ever, this compound has not been found as a metabolite of BaP in biological
systems.
     Following the demonstration of the careinogenicity of BaP 7,8-oxide to
                        255
mouse skin, Levin et al.    demonstrated that BaP 7,8-dihydrodiol, which
can form in vivo by the action of epoxide hydrase on BaP 7,8-oxide, was
                                                                254
even more potent than the parent epoxide.  In fact, it was shown    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.
                                     6-102

-------
     A summary of the available information concerning carcinogenicity of
BaP and its derivatives to mouse skin is presented in Table 6-18.   It
cannot be determined with certainty from these data,  however,  which prod-
ucts of BaP biotransformation may act as ultimate carcinogens.   The apparent
lack of carcinogenicity of the BaP 7,8-diol-9,10-epoxides to mouse skin,
despite abundant evidence concerning their mutagenicity (see Section 6.4),
binding to nucleic acids (see Section 6.2), and high  carcinogenicity to new
          yen OC"\
born mice,   '    may be explained by poor penetration of the  skin.  Since
it is a highly reactive electrophile, a BaP diol  epoxide may readily alkyl-
ate nucleophilic sites of any kind and thus have a reduced probability of
passing 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
                                                  ?fi?
related to age was examined by Peto and coworkers.     The incidence of
skin cancer in mice resulting from regular BaP application (20 pg) 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 atten-
tion.  The rationale for investigating the tumor-initiating capabilities of
BaP derivatives stems from the knowledge that hydroxylated intermediates
                                     6-103

-------
           Table  6-18.   SKIN TUMORS IN MICE TREATED WITH  BENZO[A]PYRENE AND DERIVATIVES
Treatment3
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
1-HOBaP
2-HOBaP
3-HOBaP.
4-HOBaP^
5-HOBaP^
6-HOBaP^
7-HOBaP^
8-HOBaP^
9-HOBaPcr
10-HOBaPc
11-HOBaP
12-HOBaP
Total number
of animals
25
30
26
30
27
30
30
30
30
30
30
30
25
29
29
26
26
28
30
27
26
28
28
23
Dose,
umoles
0.4
0.4
0.4
0.15
0.1
0.1
0.1
0.1
0.05
0.025
0.02
0.02
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
% of mice
with tumors
100
100
92
100
96
38
50
91
59
7
4
0
0
100
0
0
0
0
0
0
0
0
14
0
Total number .
of skin tumors
32
34
34
40
28
13
15
24
20
2
1
0
0
37
0
0
0
0
0
0
0
0
4
0
Reference
257
257
256
255
257
254
254
254
254
254
254
254
257
257
257
256
256
256
256
256
256
256
257
257
a Female C57BL/6J mice were treated with BaP or BaP derivaties (0.02-0.4 ymole) once every two  weeks for
  60 weeks  by  topical application to the shaved skin of the back.

  Skin tumors  consisted mostly of squamous  cell carcinomas; other  skin  tumors were fibrosarcomas,  papillomas,
  and keratocanthomas.

c Mice were treated once every two weeks for 56 weeks.

-------
           Table 6-18 (continued).   SKIN TUMORS IN MICE TREATED WITH  BENZO[A]PYRENE AND DERIVATIVES
O
01
Treatment3
BaP 4,5-oxide
BaP 4,5-oxide
BaP 7,8-oxide
BaP 7,8-oxide
BaP 7,8-oxide
BaP 7,8-oxide
BaP 9,10-oxide
BaP 11,12-oxide
BaP 11,12-oxide
BaP 7,8-dihydrodiol
BaP 7,8-dihydrodiol
BaP 7,8-dihydrodiol
BaP 7,8-dihydrodiol
BaP 7,8-dihydrodiol
(+)-7B,8a-Dihydroxy-
9B,10B-epoxy-7,8,9,
Total number
of animals
30-39
30-39
30-39
30
30
30-39
30-39
28
27
30
30
30
30
30

10-
Dose,
umoles
0.4
0.1
0.4
0.3
0.15
0.1
0.4
0.4
0.1
0.3
0.15
0.1
0.05
0.025


% of mice
with tumors
4
6
94
53
18
9
0
0
0
100
100
92
76
7


Total number .
of skin tumors
1
2
37
16
5
3
0
0
0
42
40
28
24
2


Reference
73
73
73
255
255
73
73
257
257
255
255
73
73
73


tetrahydrobenzo[a]pyrene
(diol epoxide 1)
diol epoxide 1
diol epoxide 1
(+)-7B,8a-Dihydroxy-
9a,10a-epoxy-7,8,9,
30
30
30

10-
0.4
0.1
0.02


0
0
0


0
0
0


73
73
73


tetrahydrobenzo[a]pyrene
(diol epoxide 2)
diol epoxide 2
diol epoxide 2
30
30
30
0.4
0.1
0.02
13
7
0
3
2
0
73
73
73
      a Female C57BL/6J mice were treated with  BaP or BaP derivatives (0.02-0.4 umoles) once  every two weeks for
        60 weeks by topical application to the  shaved skin of the back.

        Skin tumors consisted mostly of squamous cell carcinomas; other  skin tumors were fibrosarcomas, papillomas,
        and keratocanthomas.

-------
may be better substrates for activation to a proximate or ultimate carcino-
gen.  Moreover, it would not be expected that all  carcinogenic POM metab-
olites 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
    253
BaP.     In addition, the tumorigenicity of BaP 7,8-dihydrodiol  is highly
stereospecific.    At equimolar doses, (-)-BaP 7,8-dihydrodiol was more
active than BaP as a tumor initiator, whereas the  (+)-enantiomer was con-
siderably less active than BaP.  This selectivity  indicates a preferential
orientation at the active site of the enzyme complex.   Other effective
tumor-initiating metabolites of BaP are BaP 7,8-oxide and BaP 7B,8a-diol-
9B,10e-epoxide (diol epoxide 1) and BaP 4,5-, 9,10-, and 11,12-oxides (all
                                                           79
of the above compounds are racemic except where indicated).    However,
both diol epoxide 1 and diol epoxide 2 (BaP 73,8a-diol-9a,10a-epoxide) were
severalfold more active than BaP in the induction  of epidermal, hyper-
       pco
plasia.     A summary of the skin tumor initiating activities of BaP and
its metabolites is presented in Table 6-19.
     Following the subcutaneous administration of  BaP to rats and mice,
sarcomas commonly develop at the site of injection, since the carcinogen is
introduced below the basal cell layer.  A total dose of 6 to 7 ymoles (1.5
to 1.7 mg) of BaP is sufficient to produce sarcomas in nearly all treated
animals  (Table 6-20).  Although a single dose of 0.31  mg or less of BaP
                                     6-106

-------
     Table 6-19.   SUMMARY  OF THE SKIN TUMOR  INITIATING ACTIVITIES OF BENZO[A]PYRENE AND  ITS METABOLITES
o
•vl
Initiator
Bap
BaP
BaP
BaP 4,5-epoxide
BaP 7,8-epoxide
BaP 9,10-epoxide
BaP 11 ,12-epoxide
BaP 7e,8a-diol-9a,10a-epoxide
BaP 7B,8a-diol-9B,108-epoxide
BaP 7,8-dihydrodiol
(-)-BaP 7,8-dihydrodiolb
(+)-BaP 7,8-dihydrodiolD
Number
of mice
30
30
30
30
29
29
30
29
28
29
30
30
Dose,
nmoles
200
200
100
200
200
200
200
200
200
200
100
100
Weeks of
promotion
23
30
21
23
23
30
30
30
30
30
21
21
% of mice
with tumors
94
92
77
20
81
15
38
69
7
86
77
23
Papillomas/
mouse
4.8
5.3
2.6
0.2
1.9
0.15
0.45
1.5
0.07
5.0
3.8
0.43
Reference
253
79
76
253
253
79
79
79
79
79
76
76
          Female CD-I  mice were treated with a single dose of  initiator dissolved in acetone, acetoneiNfyOH (1000;!),
          or dimethyl  sulfoxide:acetone (1:3) and-followed one week later by twice-weekly applications of 10 vg of TPA.

          Promotion was  by twice-weekly applications of 16 nmoles of TPA beginning  11 days after treatment with
          initiator.

-------
                               Table 6-20.   INDUCTION OF  SARCOMA BY BENZO[A]PYRENE
o
oo
Species Number and (sex)
Rat
(Sprague-Dawley)
Mouse
Mouse
Mouse
Mouse
Mouse
Mouse
13 (female)
14 (male
16 (female)
9 (?)
10 (?)
12 (?)
15 (?)
Total dose,
ymoles
6.0a
7.1b
7.1b
15. 9C
5.0C
0.5C
.002C
Animals with
sarcoma, %
100
93
50
66.6
70
66.6
0
Average latency,
days
101 + 2.7
129
160
112
122
155
N.A.d
Reference
274
285
285
379
379
379
379
          Administered as 0.2  umole dissolved in 0.1 ml  sesame  oil  by subcutaneous injection on  alternate days for
        .  30 doses beginning at  30 days of age.
          Administered as three  injections of 2.4 pmoles each,  given at one month intervals.
         . Administered as a single injection under the skin of  the  abdomen, dissolved in 0.5 ml  of  neutral olive oil.
          Not applicable.

-------
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),
                                  250
respectively, out of groups of 50.
6.5.1.2.2  7,12-Dimethylbenz[a]anthracene.  DMBA is a potent carcinogen
used extensively in the laboratory.  It is not found, however, in ambient
air.  When applied topically, DMBA is both an effective tumor initiator and
                                                            7fiA ?fift ?7fi
a complete carcinogen, depending upon the dose administered.       '
Numerous studies have indicated that a single dose of DMBA greater than
400 nmoles is sufficient to produce tumors in mouse skin in the absence of
                     269
a promoting stimulus.     Attempts to produce skin tumors in CD-I female
mice with a single application of 200 nmoles of DMBA were unsuccessful
after 35 weeks.  DMBA is metabolized much faster than BaP and has a more
complicated metabolite profile, suggesting that more biochemical  parameters
may be modulating DMBA tumorigenesis.
                         259
     Albert and coworkers    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 6-12).   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, but they may also be formed after intravenous injection of  the car-
                                  270
cinogen.  Bennington and coworkers    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.
                                     6-109

-------
                  10.0
                o
                E
                3
                   1.0
                   0.1
                           500/jg
                      10
                                              DMBA
                                             Rat Skin
                                       100 jug
                                           20 ng
         100
Time (weeks)
Figure 6-12.   Incidence of cancer per rat after weekly topical  application of
either 20, 100, or 50 pg DMBA in 1.0 ml  acetone.259
                                     6-110

-------
     When applied to the skin of guinea pigs, high doses of DMBA can induce
                                                        271  272
malignant melanomas which resemble the melanomas of man. ''»'•"-  Four of
20 female guinea pigs developed malignant melanomas and subsequent metastases
                                                                 279
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 equal  to and possibly greater than BaP
                                269
in potency.  Slaga and coworkers    provided dose-response data in female
CD-I mice for tumor 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, (17 nmoles) was used as the
promoter 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 pro-
duced 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
                                     6-111

-------
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, how-
ever, that the effectiveness of DMBA in the two-stage skin carcinogenesis
system is influenced by species susceptibility and factors which modify the
                                          273
promotion process (diet, hormonal status).      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
2.5 mg DMBA produced sarcomas in all of a group of eight female Sprague-
Dawley rats.     In a similar study, 6 pmoles of DMBA (administered as 30
injections of 0.2 pmoles each on alternate days) produced a 100 percent
                                                      274
incidence of sarcoma in 10 female Sprague-Dawley rats.     The same treat-
ment employing BaP instead of DMBA led to essentially identical results.
Injections of principal DMBA hydroxylated 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 DMBA produced local
tumors (sarcomas and carcinomas) in four of six males and three of five
        46
females.    Monohydroxymethyl derivatives of DMBA were active as carcino-
gens, but less potent than the parent compound.
6.5.1.2.3  3-Methylcholanthrene.  As with DMBA, MCA is a potent carcinogen
which is produced synthetically for use in the laboratory.  The action of
                                     6-112

-------
MCA as a complete carcinogen in mouse skin was recently investigated by
                    275
Burki and coworkers.     A single application of 3 ymoles MCA produced
16 tumors in 18 BALB/c male mice after 57 weeks.  In a subsequent series of
experiments, repeated applications of 1.5 to 3 ymoles MCA twice weekly for
three or 17 weeks induced larger numbers of skin tumors in nearly all
animals.  Nevertheless, 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 increased by the administration of trichloropropene oxide,  an
epoxide hydrase inhibitor.  Despite this effect, which implicates the
involvement of a carcinogenic epoxide intermediate by increasing its half-
life, the K-region epoxide of MCA (MCA-ll,12-oxide) was much less effective
than the parent compound.  It is possible, however, that such 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 yg TPA
                                                               277
produced an average of 1.9 papillomas per mouse after 30 weeks.     The
tumor incidence was greater than 50 percent.  Pretreatment of mice with the
enzyme inhibitor TCPO slightly increased the yield of MCA-induced papil-
lomas per mouse, again suggesting the involvement of an epoxide intermediate.
                            278
     Cavalieri and coworkers    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
                                    6-113

-------
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
                                                       27Q
in high yields at the injection site in C57 black mice.     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.
6.5.1.2.4  Pibenz[a,h]anthracene.  A skin tumor incidence of 11.7 percent
resulted within 28 weeks after female CD-I  mice received ten daily applica-
                                           277
tions of DBahA at 1.0 mg (3.6 ymoles) each.     Other investigators have
clearly 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.281'282'283  The K-region epoxide of DBahA (DBahA-5,6-epoxide)
produced about the same tumor response as the parent compound in female
CD-I mice.277
     DBahA was shown to be as effective a tumor initiator in mouse skin as
    277
BaP.     A single application of 200 nmoles DBahA followed by twice weekly
applications of 10 yg 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 yg (0.07 nmoles) DBahA can initiate skin carcinogenesis in
n,1ce.280
                                     6-114

-------
     By subcutaneous Injection, DBahA has produced sarcomas and carcinomas
                                        46
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.
6.5.1.2.5  Dibenzo[a,h]pyrene.  Skin application of DBahP was first associated
                                             250
with the development of epitheliomas in 1939.      More recently, Hoffmann
          284
and Wynder    reported strong carcinogenic activity on mouse skin when
applied for one year, thrice weekly in 0.05% and 0.1% 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
                                              250
induce sarcomas when injected in rats or mice.     A total administration
of 1.8 mg DBahP produced 34 sarcomas among 35 male mice; only one of ten
                                         por
female mice developed a sarcoma, however.
     The reduced carcinogenic activity relative to BaP for DBahP (and other
related POM's) may be due to the very large decrease in solubility re-
sulting 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 saturation of metabolic pathways may confuse the
interpretation of results.
                                    6-115

-------
                                                   OOA
6.5.1.2.6  Pibenzo[a,i]pyrene.  Hoffmann and Wynder    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 per-
cent DBaiP solution, the results were essentially the same (13 epithelio-
mas, 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.
     Large numbers of sarcomas in mice and hamsters are produced by
                                                      250
subcutaneous and/or intramuscular injections of DBaiP.     Single sub-
cutaneous injections of only 0.5 mg DBaiP into the groin of C57 mice pro-
duced local sarcomas in 98 percent of the animals in 24 weeks.
6.5.1.2.7  Pibenzo[c,g]carbazo!e.  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
                                           250
International Agency for Research on Cancer    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
mice receiving injections.  In single doses as low as 0.2 mg, DBcgC pro-
                                                250
duced 17 sarcomas among 20 mice within 40 weeks.     Depending on vehicle
of administration (lard, olive oil, sesame oil) and strain of mouse em-
ployed (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,
                                    6-116

-------
sesame oil); 20/27 (Amice, olive oil); 5/19 (C3H mice, lard); 22/41  (C3H
mice, sesame oil); 46/67 (C3H mice, sesame oil).
6.5.1.2.8  Benzo[c]phenanthrene.  Benzo[c]phenanthrene and several  of its
monomethyl derivatives were tested for carcinogenicity by skin painting at
                                                  286
high doses and subcutaneous injection in C3H mice.     Twice weekly appli-
cation 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 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.
6.5.1.2.9  Benzo[b]f1uoranthene.  BbF is a common pyrolysis product found
                                                         250
in automobile exhaust, polluted air, and cigarette smoke.     In 1959,
                   287
Wynder and Hoffmann    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% 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%, BbF produced papillomas in 13 ani-
mals and carcinomas in 17 animals with 12 months.  At 0.01%, the lowest
dose tested, five of the ten surviving mice after 14 months developed
papillomas.  By comparison, BaP at the same 0.01% dose level caused car-
cinomas in 17 of 20 mice after 12 months of treatment.
                                     6-117

-------
                                                  288
     In later studies by Van Duuren and coworkers,    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 yg
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
                                         250 285
of 14 females (average latency 133 days).   '     The sarcomagenic potency
of BbF was considered to be not much less than for BaP, especially in
females.
6.5.1.2.10  Benzo[j]f1uoranthene.  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,
                   287
Wynder and Hoffmann    reported in 1959 that BjF produced carcinomas on the
skin of mice.  Among 20 female Swiss mice receiving thrice weekly applica-
tions of BjF  (0.5% in acetone), 19 animals developed papillomas  and car-
cinomas within seven months.  None of the mice survived beyond seven months.
At a dose level of 0.19% in acetone, BjF produced papillomas in  14 mice and
carcinomas in all 20 mice within nine months.
                                     6-118

-------
6.5.1.3  Respiratory tract carcinogenesis--Unti1 recently, the investigation
of POM-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 carcino-
genic potency of a polycyclic compound for the pulmonary tissues may not
necessarily parallel its activity either as a complete carcinogen on mouse
and hamster epidermis, as a tumor initiator on mouse skin, or as a sarcoma-
genie 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 chemi-
cals can be compared within the same bioassay system and ranked according
to their tumorigenic potency based on dose-response relationships.
     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 disper-
sions of polycyclic hydrocarbons to mice, a direct comparison of carcino-
genic potency was possible (Table 6-21).     In this test system, MCA dis-
played 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's,
since BaP is a more active carcinogen in skin and as a transforming agent
jji vitro.
                                    6-119

-------
Table 6-21.   COMPARATIVE CARCINOGENICITY OF POLYCYCLIC HYDROCARBONS AND  RELATED COMPOUNDS, MEASURED
                                 BY INDUCTION OF LUNG  TUMORS (LT)a 471
Dose,
Compound pinoles/ kg
3-Methylcholanthrene, 0.1 mg
3-Methylcholanthrene, 0.5 mg
Dibenz[a,h]anthracene
7H-Di benzo[c , g]carbazol e
Benzo[a]pyrene
Di benz[a , j ]aceanthryl ene
Dibenz[a,h]acridine
8-Methyl benzo[c]phenanthrene
7-Methylbenzo[a]pyrene
5-Methoxy-7-propylbenz[a]anthracene
Benz[a]anthracene
Untreated controls
15
74
36
38
40
33
36
42
38
33
44
--
Mice with LT/
no. of mice
15/15
6/6
10/10
12/12
10/10
9/10
11/12
6/11
5/10
1/10
2/11
4/19
Mean no.
LT/mouse
11 }
47 ;
31
5.7
3.7
2.7
2.0
0.7
0.6
0.1
0.2
0.2
gMoles/kg for
1 LT response
0.9
1.0
6.0
9.5
14
18
--
--
--
--
--
a Strain A mice, 8-12 weeks old, received single intravenous injection of 0.24 mg of methylcholanthrene
  in aqueous dispersion and were killed 20 weeks later.

-------
     Intratracheal instillation of POM to Syrian golden hamsters289'290'148
has become widely utilized for the conduct of dose-response studies for
pulmonary carcinogenesis.  Several studies are summarized in Table 6-22 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.
     Preliminary results have recently become available on a study to
establish temporal and dose-response relationships for BaP carcinogenesis
                                 259
in the hamster respiratory tract.     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
observation only, indicated that the incidence of respiratory nodule forma-
tion was dose-dependent; 0.25 mg BaP with 0.25 Fe203 given for 15 or
30 weeks produced no nodules, whereas 2.0 mg BaP with 2.0 mg Fe203 given
for 15 or 30 weeks resulted in a tumor incidence of 44% and 86%, respec-
tively.  Dose-response information is depicted in Figure 6-13.  Surpris-
ingly, the use of Fe90, as a carrier for BaP had no apparent effect on the
                    c. *5
incidence of lung nodule formation (Figure 6-14).  While tumor latency
periods decreased with increasing BaP weekly dose, no clear trend was ob-
tained in temporal aspects of tumor development between 15 and 30 week
treatments at the same dose.
                                     6-121

-------
        Table 6-22.   INDUCTION OF RESPIRATORY TRACT TUMORS IN SYRIAN GOLDEN HAMSTERS BY INTRATRACHEAL

                                             INSTILLATION OF POM
rv>
ro




Number Total dose,
Compound
BaP
BaP
BaP
BaP
BaP
BaP
BaP and Fe-0,
BaP and Fe^O,, coated
BaP and Fe^O,, ground
BaP and Fe-O.,, mixed
BaP and gelatin
BaP and Fe,0,
BaP and Fe,0,
BaP and Fe,0,
BaP and Fe,0,
BaP i J
BaP
DB[a,i]P
DB[a,i]P
DMBA and Fe-O,
DMBA and Fe^Cu
DBC and Fe-O,
DBC and Fe^O,
Respiratory tumor
incidence,
of animals mg %
30
30
30
29
28
48
48
49
49
43
46
3.25a
6.5a
13a
26*
"»?
3°h
30b r
26. 1
-------
ro
oo
               1.0 —


               0.9 —


               0.8 —
            g  0.7H
            o

            2  0.6 —
O
tr 0.4 —
2
o
£ 0.3-
               0.2 —


               0.1 —


               0.0
                                                                 2mg.\
                                                             30 WEEKS
                                                                     1 mg.
                                                                     30 WEEKS
                                                                                   2mg.
                                                                                   15WEEKS
                                       I
                                      200
                                              I
                                             400
                                          TIME (DAYS)
 I
600
 I
800
                                                                                                     259
     Figure  6-13.   Respiratory nodule incidence in hamsters  given BaP by intratracheal instillation.
     (Hamsters  were given weekly intratracheal instillations of either 1.0 or 2.0 mg BaP with equivalent
     amounts of Fe?0^ for either 15 or 30 weeks as indicated.   Proportion with tumor was calculated from
     the  life table procedure on the basis of tumors  found at death.   Time was measured from the first
     BaP  dose.)

-------
t\>
    Figure 6-14.  Respiratory  nodule  incidence in hamsters given BaP by intratracheal  instillation with
    and without Fe203."y   (Hamsters  were given weekly intratracheal instillations of  2.0 mg  BaP for
    either 15 or 3u weeks with and without 2.0 mg ferric oxide (Fe?0,).  There were 36 hamsters  per
    group and tumors found at  death were analyzed by the life table procedure to give  the proportion
    with tumor.  Time was measured from the first BaP dose.)

-------
     In addition to the hamster model system, respiratory lesions and
carcinomas 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 6-23.
     An additional approach used to evaluate the effects of POM on the
                                                291
pulmonary tissue was reported  by Flaks and Sims.     Pulmonary tissue taken
from tumor susceptible female  BALB/c mice was incubated for 30 minutes in
the 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 tracheal grafts by
                                                            292
instillation of MCA, BaP, or BaP mixed with Fe«0~ particles.     Neoplasms
were induced by 5 mg of carcinogen after 35 weeks of contact in 3 of 5, 2
of 5, and 2 of 3 tracheal grafts exposed to MCA, BaP, and BaP-Fe203,
respectively.  Subsequent studies in the same test system using DMBA in a
sustained release formulation  established that carcinomas could be induced
                                                293
in tracheal epithelium at doses as low as 40 yg.     With this procedure,
                                                                    294
the lowest dose of BaP tested  which produced a carcinoma was 300 yg.
6.5.1.4  Carcinogenesis in newborn mice—The newborn mouse is known to be
                                       295
highly susceptible to carcinogenic POM,    presumably because of its large
                                  296
numbers of rapidly dividing cells,    and is thus a useful model for car-
                                                                   297
cinogenesis studies.  In studies reported by Asahina and coworkers,
various fractions of an organic extract of atmospheric particulate pol-
lutants displayed high carcinogenic activity when injected subcutaneously
                                     6-125

-------
                             Table  6-23.    INDUCTION  OF RESPIRATORY  TRACT  TUMORS  IN  RATS  AND  MICE
CT>
CT»
Compound
DHBA and Indian ink
DHBA and Indian ink
DMBA and Indian ink
DB[a,h]A
MCA
MCA
MCA
MCA
MCA
MCA
MCA
Organism
Rat (Hi star and random-bred)
Rat (Mistar and random-bred)
Rat (Wistar and random-bred)
Mouse (DBA/ 2)
Rat (Osborne-Mendel)
Rat (Osborne-Mendel)
Rat (Osborne-Mendel)
Rat (Osborne-Mendel)
Rat (Osborne-Mendel)
Rat (Osborne-Mendel)
Rat (Osborne-Mendel)
Number of
animals
34
56
61
14 (male)
13 (female)
100
100
100
100
100
100
100
Total dose,
mg
2.5a
6b
10C
236 (male)d ,
179 (female)0
0.005f
o.osf
0.1 Of
0 20
0 30f
0.40f
0.50f
Route of
administration
Intratracheal
instillation
Intratracheal
instillation
Intratracheal
instillation
Oral
Pulmonary injection
Pulmonary injection
Pulmonary injection
Pulmonary injection
Pulmonary injection
Pulmonary injection
Pulmonary injection
Tumor
incidence, t
17.6
35.7
26.2
100 (male)6 „
77 (female)
19
139
2?9
479
409
519
459
Reference
383
383
383
384
385
385
385
385
385
385
385
        ?  Administered as  a single dose with 0.2 mg of Indian ink in 0 2 ml of a colloid protein  solution.
        °  Administered as  three 2 mg doses at monthly intervals with 0.2 mg of Indian ink in 0.2  ml of a colloid protein solution.
        .  Administered as  five 2 mg doses at monthly intervals with 0 2 mg of Indian ink in 0 2 ml of a colloid protein solution
        °  Administered as  an aqueous-olive oil emulsion of DB[a,h]A given in place of drinking water for 237  to 279 days
        ,  Tumors were alveologenic carcinomas, a 100% incidence of pulmonary adenomatosis was also observed.
          Administered as  a single MCA-containing beeswax pellet placed directly into the lower peripheral  segment of the left lung.
        s  Overt squamous cell carcinoma.

-------
to infant Swiss mice of both sexes.  Various combinations of remote tumors
(hepatomas, lymphomas, pulmonary adenomas) were produced by the basic,
neutral, and aromatic fractions of the benzene soluble organic extract.
Three oxyneutral 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
                                      298
mainly in aromatic fractions.  Walters    administered DMBA by single
injection to male and female newborn (<24 hours old) BALB/c mice at doses
ranging from 0.625 to 40 yg.  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 yg MCA or 0.003 yg DBahA were
sufficient to produce fibrosarcomas at the injection site of newborn
     299
mice.     However, on a mg/kg of body weight basis, these doses are com-
parable 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 the life span of a mouse have not produced definitive
results.300
     The K-region epoxides of DBahA and MCA were found to be negative for
carcinogenic activity in newborn mice, despite reports that the parent
                                                         301
hydrocarbons were active at the same dose levels (60 yg).     These results
suggested that K-region epoxides may not be the ultimate metabolites in-
volved in POM carcinogenesis as once was thought.  In support of the con-
tention that a dihydrodiol precusor of the "bay region" diol epoxide may be
                                                                    ?fifl
a proximate carcinogenic metabolite of BaP, Kapitulnik and coworkers ou>
                                    6-127

-------
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-7B,8a-dihydroxy-9a,10a-epoxy-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 6.2.1.2).  Since the
metabolism of BaP results in three known diols, all via epoxide intermedi-
ates, 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.
                           302
     Wislocki and coworkers    recently presented evidence supporting the
"bay region" theory with regard to the carcinogenicity of benz[a]anthra-
cene.  They demonstrated that benz[a]anthracene-3,4-dihydrodiol 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
carcinogenic 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.
                                     0-128

-------
6.5.1.5  Tumors in other tissues--Mammary cancer induced in rodents by DMBA
has become an important model system for the study of human breast can-
cer.   '   '     The demonstration of the production of hormone-responsive
mammary tumors in normal female Sprague-Dawley rats by a single intragas-
                                          305
trie dose of DMBA by Huggins and coworkers    in 1961 introduced a simple
and yet sensitive model for the experimental investigation of the biology
of mammary tumor growth.  A year later, Huggins and coworkers    demon-
strated 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, Huggins    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 consecu-
tive female Sprague-Dawley rats, age 50 days.  Tumors were detected by
palpation as early 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, Griswold303 adapted the DMBA-
induced mammary cancer technique in rats as a routine for the evaluation of
potential anti-cancer agents.  Using the Huggins 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.
                                    6-129

-------
     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-
                         308
gen to the mammary gland.     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 yg.  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
"preneoplastic" 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
                                259
formation.  Albert and coworkers    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
                        141
exposure to BaP and MCA.     In Mi star rats, a silk thread impregnated with
DMBA and inserted in the region of the ovary produced 47 ovarian solid
                         206
tumors among 121 animals.
     In addition to ovarian cancers, DMBA can also produce carcinoma by
                                                         309
local application to the pancreas of Sprague-Dawley rats.     Less than
1 mg of DMBA placed directly into the pancreas induced adenocarcinoma in
eight of ten treated animals within 180 days.
                                     6-130

-------
     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
                       250
recently been reviewed,    and some of these data are summarized in
Table 6-24.  Tumors of the stomach have also been induced by MCA    and
DMBA.     A single feeding of 3 mg DMBA induced nine stomach papillomas in
18 surviving Swiss mice after 1 year.     Ovarian, lymphoid, mammary, and
hepatic tumors were also observed.  In addition, the tumorigem'city of DMBA
                                                                312
in the rat submandibular gland has been repeatedly demonstrated.
6.5.2  In Vitro Carcinogenesis Studies
     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
                                           313
was first demonstrated by Berwald and Sachs    in 1963 that polycyclic
hydrocarbons (MCA, BaP) could cause the direct malignant transformation of
hamster embryo cells in culture.  Transformed colonies have growth charac-
teristics visually distinct from normal colonies and are readily seen above
a background of normal cells.  This 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 immunosuppressed animals.  In addi-
tion to hamster embryo cells, malignant transformation has been demonstrated
in organ cultures, liver cell cultures, fibroblastic cells derived from
mouse ventral prostate, 3T cell lines derived from mouse embryo cells, and
various types of epithelial cells from humans and other animals.   >3'
                                     6-131

-------
                 Table 6-24.   CARCINOGENICITY  OF POM BY ORAL ADMINISTRATION TO VARIOUS  MAMMALS250
Compound
BaP

Species Dose
Mouse 0.2 mg in PEG3
Mouse 50-250 ppm
(age 17-116 days)
Route of
administration
Intragastric
Dietary
(110-197 days)
Effects
14 tumors of the forestomach in
5 animals out of 11
>7Q% incidence of stomach tumors
at 50-250 ppm for 197 days; no
tumors with diets containing up
to 30 ppm for 110 days
PO
                       Mouse
Mouse
(age 18-30 days)

Rat
(Sprague-Dawley;
 age 105 days)

Hamster
                       Hamster
250 ppm




250 ppm


2.5 mg per day
                     500 ppm
                                         Dietary
Dietary
(140 days)

Oral
                                            2-5 mg bi-weekly    Intragastric
                   Dietary
                   (4 days per  week
                    for up to
                    14 months)
100% stomach tumor incidence when
 diet was  fed for 30 days; 5-7 days
 of feeding, 30-40*; 2-4 days of
 feeding,  10%; one day of feeding, 0%

Leukemias,  lung adenomas, and stomach
 tumors  produced

Papillomas  developed in the esophagus
 and forestomach in 3 out of
 40 animals

5 stomach  papillomas in 67 animals
 treated for 1-5 months; 7 papillomas
 and 2 carcinomas in 18 animals
 treated for 6-9 months; 5 papillomas
 in 8 animals treated for 10-11 months

12 tumors  (2 esophagus, 8 forestomach,
 2 intestinal) in 8 animals
            Polyethylene  glycol

-------
         Table  6-24  (continued).   CARCINOGENICITY OF POM BY  ORAL  ADMINISTRATION  TO VARIOUS  MAMMALS250
Compound
DB[a,h]A


Species
Mouse
Mouse
(A backcross)
Mouse
(DBA/ 2)
Dose
9-19 mg
(total dose)
0.4 mg per day
0.76-0.85 mg
per day
Route of
administration
Dietary
(5-7 months)
Oil emulsion
(drinking water
replacement)
Oil emulsion
(drinking water
Effects
Forestomach tumors in 7 of
22 survivors after one year;
one tumor was a carcinoma
11 papillomas of the forestomach
in 20 animals within 406 days
Pulmonary adenoma tos is in all
27 survivors at 200 days; 24 animal:
co
CO
          DB[c,g]C
                      Mouse
                      (Swiss, male)

                      Mouse
                      (BALB/c, female)
Mouse
(CBA or Strong A
 strain)
                     1.5 mg in PEGa
15 mg
 total  dose  in
 almond oil

7-23 mg
 total  dose  in
 arachis oil
                                                                   replacement)
                      Oral (single dose)
Intragastric
(twice weekly for
 15 weeks)

Oral
(twice weekly)
 had alveologenic carcinomas; 16
 had hemangio-endotheliomas; 12 of 13
 females had mammary carcinomas;
 2 pulmonary adenomatoses seen among
 35 controls

Papillomas  of  the forestomach in
 2 out of 42 animals within 30 weeks

Mammary carcinomas in one out of
 20 intact  animals and 13 out of 24
 pseudo-pregnant animals

Among 55 mice  surviving 17 to 59
 weeks, 64% had papillomas and 13%
 had squamous  carcinomas of the
 forestomach;  19 animals had benign
 hepatomas  and another 19 animals
 had malignant hepatomas; pulmonary
 adenomas observed in all Strong A
 mice and none of the CBA mice
          a Polyethylene glycol

-------
                                       317                        318
     Early reports by Berwald and Sachs    and Dipaolo and Donovan
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 addi-
tion, 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 pg/ml of BaP for 6 days.  BaP treatment at
1 pg/ml for 6 days produced 19.9 percent transformation in surviving cells.
Further data indicating the activity of several polycyclic carcinogens and
their derivatives are summarized in Table 6-25.  It is significant to note
that the K-region epoxides of DBahA and MCA are more active in the pro-
duction of malignant transformation in hamster embryo cells than the parent
                                                   320 321
hydrocarbons or the corresponding K-region phenols.   '     Although these
results confirm the view that metabolism is necessary for carcinogenic
activity, they conflict with data generated in vivo (see Section 6.5.1)
which indicate that K-region epoxides of polycyclic carcinogens are less
active than the parent compound in various species.  A possible reason for
                                     6-134

-------
  Table  6-25.
HAMSTER  EMBRYO  CELL  TRANSFORMATION  PRODUCED BY  SEVERAL POLYCYCLIC HYDROCARBONS AND
                               THEIR DERIVATIVES
Compound
DB[a,h]Aa


DB[a,h]Ab


DB[a,h]A5,6-epoxide*



Ct DB[a,h]A5,6-epoxide
i
— *
CJ
tn
MCAC


MCAd
MCA epoxide0


BaPd

Concentration,
ug/ml
2.5
5
10
2.5
5.0
10
2.5
5
7.5
10
2.5
5.0
7.5
10
2.5
5
7.5
2 5
3.5
5
7
1
5
Total number
of colonies
760
690
790
1341
1363
1365
598
601
395
350
895
866
817
707
404
370
349
664
364
245
103
1016
394
Cloning
efficiency, %
4.2
3.8
4.4
13.4
14.0
14.5
3.3
3.3
2.5
1.9
10.1
9.3
9.3
7.7
10.1
9.2
8.7
9.6
2.4
1.5
0.7
8.46
7.17
Number of
transformed
colonies
4
4
7
3
11
7
3
12
31
14
7
20
22
30
9
10
15
20
13
8
17
25
21
%
transformation
0.5
0.7
0.9
0 2
0.8
0.5
0.5
2 0
7 8
4 0
0.8
2.3
2.7
4.2
2.2
2.7
4.3
3.46
3.6
3.3
16.5
2.46
5.33
Reference
321
321
321
320
320
320
321
321
321
321
320
320
320
320
321
321
321
323
321
321
321
323
323
4 7-day treatment of cells seeded on a feeder layer.
b 7-8 day treatment of cells.
c 4-hour treatment of cells seeded in conditioned medium.
d 8-day treatment of cells.

-------
the lack of correlation is the relative instability of K-region epoxides as
compared to the parent hydrocarbon when applied to the skin.  It is likely
that in vivo 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
                                             322 323
dissociable and occur by different processes,   '    with the toxicity
being due to random alkylation of nucleophilic regions within the cell.
However, when hamster embryo cells are pretreated with weak chemical car-
cinogens which can induce microsomal enzyme activity (e.g., benz[a]anthra-
cene, methyl methanesulfonate, ethyl methanesulfonate) before the addition
of a potent carcinogen (e.g., MCA, BaP, DMBA), transformation may be con-
siderably enhanced.323'324
     As a prescreen for chemical carcinogens, cell transformation in vitro
may be one of the most sensitive techniques available.  Pienta and co-
       OOK
workers    reported that 90% (54/60) of the carcinogens they tested trans-
formed hamster embryo cells in vitro, whereas none of the non-carcinogens
tested showed any activity.  Moreover, many of the carcinogens which have
not been shown to be mutagenic toward S. typhimurium in vitro (e.g., chry-
sene) 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 E. coli, and hamster embryo cell trans-
formation is capable of detecting nearly all carcinogens tested, both POM
and non-POM types.
                                     6-136

-------
     It is known that alteration of microsomal  enzyme activity either jni
vitro or jr^ vivo can have a marked effect on the carcinogenic response to
                             326
POM.  Nesnow and Heidelberger    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
inhibitors of epoxide hydrase (e.g., cyclohexene oxide; styrene oxide;
l,2,3,4-tetrahydronaphthalene-l,2-oxide) caused an increase in transforma-
tion 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 trans-
formation in cultured cells by altering steady-state levels of oncogenic
epoxides.
                          327 328
     Chen and Heidelberger   '    developed a system using C3H mouse
ventral prostate cells to examine transformation by carcinogenic hydrocar-
bons under conditions in which no spontaneous malignant transformation
occurred.  Cells treated with MCA (1 yg/ml) for 6 days in culture produced
malignant fibrosarcomas in 100 percent of mice into which they were sub-
cutaneously 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 quantitative correlation 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.  In contrast to
                                     6-137

-------
the enhanced transforming ability of K-region epoxides relative to the
parent hydrocarbon in hamster embryo cells, the K-region epoxide 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
               329 330
prostate cells.   '     Moreover, the epoxide derived from DMBA was more
toxic than DMBA itself.  The anomalous behavior of DMBA may 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
                                                          331 332 333
extracts of airborne particulate matter has been reported.   '*
Cells preinfected with RNA tumor viruses led to accelerated transformation,
as well as transformation in cells not inducible by crude extracts alone.
                  332
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 only 1/30 as
much BaP, showed activity comparable to the benzene fraction.  Thus, a
diversity of carcinogenic material (POM or otherwise) is apparently con-
tained in extracts of particulate pollutions; a conclusion which is sup-
ported by in vivo studies (see Section 6.5.1 and 6.5.1.4).
     Attempts to transform human cells in culture with POM (e.g., BaP, MCA,
                                      O O/l                             O O C
DMBA) have generally met with failure.     However, Rhim and coworkers
reported 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 yg/ml in
                                     6-138

-------
the fifth subculture 52 to 57 days after exposure.  One of the altered cell
lines obtained from the 1 ug/ml treatment was tumorigenic in nude mice by
subcutaneous and intracerebral injection.  Interpretation of the signifi-
cance 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
             •JOC
by Lasnitzki.    Microgram quantities of MCA added to organ cultures of rat
and mouse prostate fragments caused extensive hype'rplasia and squamous
metaplasia.  However, these preneoplastic morphological effects are
generally not associated with subsequent tumor development when carcinogen-
                                                         314
treated pieces of tissue are implanted into host animals.     Limited
success has been achieved with organ cultures of rat tracheas, which showed
characteristic morphologic alterations when treated with DMBA, BaP, and
MCA.     In addition, Crocker    has exposed respiratory epithelia from the
hamster, rat, dog, and monkey to BaP at 7 to 15 yg/ml and observed occa-
sional squamous metaplasia.  More commonly, pleomorphic cells in a dys-
plastic epithelium were evident as a result of the treatment.  Using this
system, it was also possible to demonstrate a protective effect of vita-
min A against BaP-induced abnormal differentiation.  It has been suggested
that rat tracheas maintained in organ culture may be a useful system for
                                                  338
the predictive screening of potential carcinogens.
                                     6-139

-------
     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
           339
in culture.     A transplacental influence of BaP was manifested as a
proliferative stimulus in embryonic lung tissue.   Hyperplasia arising in
the bronchial epithelium led to the development of adenomas in a large
percentage of the explants.
6.5.3  Dose-Response Models in Carcinogenesis
     Published in the early 1940's, Bryan and Shimkin's    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 be-
tween 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.
                                     6-140

-------
     Since departures from parallelism in bioassays defy interpretation in
terms of simple relative potencies, the authors recommended comparison of
doses and mean latent periods at the 50 percent level  of incidence.   Other
specific problems in the design and analysis of experiments in carcinogene-
sis 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.
                                     341
     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
reference 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 relation-
ship with dose.  If this had been the result of single short-term exposures
to carcinogens, 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 6 months later.
                                    6-141

-------
     The average numbers of tumors per mouse in the latter experiment
exhibited a good fit to a straight line dose response and a poor quadratic
fit, 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 extrapolated
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.
                         342
     Mantel and coworkers    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
                                    6-142

-------
                        341
Heston and Schneider-man.     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.
     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 arti-
ficially high one if mice sensitivities are  positively associated with
their thresholds, but artificially low in the more likely case  of decreas-
ing 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  tumorigene-
sis at extrapolated low dose levels.
                                                    343
     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 extrapo-
lating 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
                                    6-143

-------
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 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:
                                       16.79
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

                                    6-144

-------
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.
              344
     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 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.
                                        344
     A regulatory rule-of-thumb is cited    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
  -6      -8
10"  to 10" .  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
                                     6-145

-------
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.
     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

                                     6-146

-------
                                                  342
complication, established clearly by Mantel et al.    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 con-
stitute 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 survived an initial period at risk.  It further seems realistic to
suppose that repeated pulses of carcinogenic exposures throughout life
would complicate the dynamics of the process.  Thus, the steady-state
solutions of this formulation may not be particularly useful in establish-
ing 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
                                     6-147

-------
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 epi-
demiologic data, coupled with the apparently conservative assumption of low
dose linearity of response,  presently appears to be the most practical
approach.
6.5.4  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
non-carcinogenic or weakly carcinogenic chemicals, or (4) the vehicle used
to deliver a carcinogen to experimental animals.  These factors further
complicate the extrapolation of experimental animal data to human situations.
     Early work on the inhibition of carcinogenesis was reviewed by Falk
              345
and coworkers.     In addition, they conducted a series of studies to
evaluate the carcinogenic action of MCA, DBahA, and BaP in the presence of
closely related POM.  Simultaneous administration of DBahA or MCA (30 yg)
with 15 times the molar equivalent of their respective dihydro- or hexa-
hydro-reduction products by subcutaneous injection to male C57 black mice
caused either a complete inhibition or dramatic reduction of the normal
carcinogenic response.  Administration of the dihydro- and hexahydro-
derivatives at intervals extending from 14 days prior to and seven days
subsequent to DBahA (60 or 135 yg) injection also caused significant
                                     6-148

-------
inhibition of tumor development.  The injection vehicle (tricaprylin or
ethyl laurate) also affected the inhibiting response.   Injections of the
non-carcinogen phenanthrene together with DBahA in molar ratios of 1:24 and
1:48 significantly reduced tumor yield.   In simulated  environmental  studies
BaP was administered to mice together with various non-carcinogenic  POM
commonly found in polluted atmospheres.   The results (Figure 6-15) demon-
strated that in all cases a marked inhibition of carcinogenisis resulted.
Neutral fractions isolated from polluted urban atmospheres produced  similar
inhibitory effects with BaP.
                                          345
     The experiments of Falk and coworkers    were modified and repeated by
         346 347
Pfeiffer,   '    who obtained different results.  Ten  non-carcinogenic
polycyclic aromatic hydrocarbons found in automobile exhaust were tested in
combination with BaP (3-100 yg) and DBahA (2-75 yg) by subcutaneous  injec-
tion in 3000 female NMRI mice.  The ten non-carcinogens tested were:
benzo[e]pyrene (2-70 yg); benz[a]anthracene (3-100 yg); phenanthrene (125-
4000 yg); anthracene (31-1000 yg); pyrene (65-2100 yg); fluoranthene (28-
900 yg); chrysene (3-100 yg); perylene (0.2-7.0 yg); benzo(ghi)perylene
(12.8-410 yg); and coronene (3-100 yg).   The tumor incidence resulting from
all 12 compounds being administered together could be  attributed to  the
presence of DBahA, with little influence from BaP or the other ten chemi-
cals.  No inhibitory effect of the ten non-carcinogens was evident;  more-
over, an increased tumor yield resulted from injection of mixtures contain-
ing increasing amounts of the components.  This effect, however, was less
                                     6-149

-------
            CARCINOGEN 400^9 BENZO(a)PYRENE
            VEHICLE: TRICAPRYLIN
            DURATION 15 MONTHS
            ANTICARCINOGENS ENCOUNTERED IN AIR POLLUTION
              BENZO(a)FLUORENE 0.1:1
                PERYLENE0.1:!
              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:!
             ANTHRACENE : PHENANTHRENE : PYRENE 10:10:10:1
                i   i   i
i   i   i  i
          0  10 20 30 40 50 60 70 80 90

                PERCENT T. B. A.
Figure 6-15.  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.345
                              6-150

-------
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 susceptibility 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
response would be obtained in cases where the tumor yield of the combina-
tion exceeds that produced by either agent alone at the same doses.   Van
                    348 349 248
Duuren and coworkers   '   '    established that a pronounced cocarcino-
genic effect could be obtained with catechol and the non-carcinogens pyrene,
BeP, and benzo[g,h,i]perylene.  Doses of 2000, 12, 15, and 21 pg, respec-
tively, were applied 3 times a week for 52 weeks to female ICR/Ha Swiss
mice.  Each animal also received 5 yg 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)
                                                                          248
also possess cocarcinogenic activity but have no tumor-promoting activity.
                                            350
Additional studies by Schmeltz and coworkers    established that most of
the naphthalenes found in cigarette smoke have an inhibitory effect on skin
                                     6-151

-------
tumorigenesis (250 yg, 3 times a week) as induced by BaP (3 pg, 3 times a
week).  On the other hand, several of the alky]naphthalenes tested (di-
methyl-, trimethyl-, tetramethyl-) enhanced 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,351'352'353 dl-a-tocopherol (Vitamin E),351'352 and
             354
ascorbic acid    in mice treated with DMBA and croton oil.  The carcino-
genic action of MCA has been reduced by tocopherol-rich diets in rats and
     ore OCC
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.   '   '     The sulfur-containing antioxidants
disulfuram, dimethyldithiocarbamate, 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.
The agricultural herbicide, maleic hydrazide, and its precursor, maleic
anhydride, can inhibit the initiating activity of DMBA in the mouse skin
                                361
two-stage carcinogenesis 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 addi-
tion of ascorbic acid (31.7 percent reduction), butylated hydroxytoluene
                                     6-152

-------
(63.8 percent), Na2Se03 (42.0 percent), or dl-a-tocopherol (63.2 per-
cent).   '     It was not apparent from this study whether anticarcinogene-
sis by antioxidants may be due to a direct protective effect against chromo-
some damage or to an interference with metabolic activation of carcinogens,
although the latter seems most likely.  This aspect was further pursued by
                       364
Rahimtula and coworkers    who examined the abilities of several antioxi-
dants 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 bac-
terial mutagenicity of BaP in the presence of rat liver microsomes and
cofactors (Table 6-26).  The authors suggested that antioxidants may exert
their protective effect in vivo by inhibiting the formation of carcinogenic
intermediates from POM.  This conclusion, however, seems to conflict with
data indicating that inducers of increased BaP hydroxylase activity can
                             365
also inhibit tumor formation.     However, flavones are also inhibitors of
BaP metabolism in vitro, thereby indicating that their specific effects
depend upon how and where they are used.  These investigators 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 profound 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
                                    6-153

-------
Table 6-26.  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 NADPH364


                                                       Revertants of strain
       Antioxidant                 Concentration          TA 98 per plate


None                                                            384

Butylated hydroxyanisole              100 uM                    193

Ethoxyquin                             25 yM                    148

Pryogallol                            100 yM                    156

Glutathione                             1 mM                     92
                                    6-154

-------
flavone, B-naphthoflavone.  Pulmonary tumor formation was also reduced
50 percent by incorporation of the naturally occurring flavone, quercetin
pentamethyl ether, into the diet.  The possibility that only specific
components of the drug metabolizing enzyme system may be induced by anti-
                                                             366
oxidants has not been fully explored.  Sullivan and coworkers    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 antioxi-
dants 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.
     In addition to flavones, other naturally-occurring compounds have
exhibited protective effects against POM-induced tumor formation.  Vita-
min A has clearly been shown to play a role in reducing carcinogen-induced
tumors.367'368'369'370  Nettesheim and Williams371 recently undertook to
determine whether inadequate vitamin A consumption may predispose indi-
viduals 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.
                                    6-155

-------
     Further studies on naturally occurring antineoplastic compounds were
                                372
recently reported by Wattenberg.     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 speculation regarding the role and importance of diet in human
susceptibility to environmental carcinogens.  In cases where dietary con-
stituents can alter the metabolism of xenobiotics  such as ROM's,  then the
anticarcinogenic effect may result from an alteration of steady state
levels of activated versus detoxified metabolites.
     Studies have shown that not only can specific substances in  the diet
affect the response to carcinogens, but protein deprivation in general may
also decrease the activation of carcinogens.     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 inactiva-
tion of the direct-acting (ultimate) carcinogen N-methyl-N'-nitro-N-nitro-
soguanidine was reduced in liver microsomes from mice receiving a protein-
deficient diet.
                                     6-156

-------
     In humans fed charcoal-broiled beef, the metabolism of the drug
phenacetin was enhanced; in pregnant rats a similar diet stimulated the
                                          59
activity of AHH in the placenta and liver.    Further studies showed that
high-protein diets enhanced the metabolism of antipyrene and theophylline
in man, while a high-carbohydrate diet depressed the rate of metabolism of
these drugs.  Additional agents in man's environment which inhibit AHH
activity include certain organophosphate pesticides, piperonyl butoxide,
carbon tetrachloride, ozone, carbon monoxide, nickel carbonyl, nickel, tin,
                         59
cobalt, and other metals.
     In a review on cocarcinogenesis Salaman and Roe    discussed the
numerous factors which can modify the host response to a carcinogenic
substance.  Dietary restriction, stress, hormone balance, and administra-
tion vehicles may all change the outcome of carcinogen administration.
Cocarcinogenesis in skin involves the action of potential tumor initiators
(e.g., ethyl carbamate, isopropyl N-phenylcarbamate, N-(3-chlorophenyl)-
carbamate) and tumor promoters (e.g., citrus oils, croton oil, anthralin,
surface-active agents).  The incidence of BaP-induced gastrointestinal
tumors in mice was also increased by ingestion of citrus oils.  In addition
to chemical agents, oncogenic viruses (Shope papilloma, Shope fibroma, Rous
sarcoma, polyoma, and Graffi leukemogenic virus) and non-oncogenic viruses
(vaccinia, poliovirus 2, myxoviruses) can increase the tumor yield result-
ing from exposure to POM.
     The effect of cyclic AMP on skin tumor formation is paradoxical in the
sense that both inhibition and enhancement can be manifested, depending on
                                    6-157

-------
the conditions of the experiment.  In the two-stage skin carcinogenesis
system, cyclic AMP inhibits tumor formation by apparently interfering with
                    375
the promotion phase.     In follow-up studies, cyclic AMP enhanced DMBA-
induced skin tumor development in strain A mice under conditions when
                                                                  T7fi
repeated doses of the carcinogen and no promoting agent were used.     In
this instance, cyclic AMP exhibited cocarcinogenic activity, although its
mechanism is not understood.  It is possible, however, that the stimulatory
effects of cyclic AMP on tumor development may be mediated via the drug
metabolizing enzyme system (MFO).  Numerous investigators have observed
that the specific events which accompany a trophic response (increased
cyclic AMP levels, induction of ornithine decarboxylase, increased RNA
synthesis) are produced by nearly all mitogenic stimuli, including carcino-
gens.   '     Although many tumors are associated with sustained increases
in cyclic nucleotide levels, the events which lead from cell proliferation
to neoplasia are not clear.
                                     6-158

-------
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 mg)  of  MCA
in olive oil  were injected in mice on the first  or seventh  day  of  pregnancy,
                                                                ooc
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 6-27).   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
                                                                       387
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 CSH(Jax) mouse
embryos.  Their results, presented in Table 6-28, 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.
                                     6-159

-------
CTi
o
                        Table  6-27.  EFFECT  OF 3-METHYLCHOLANTHRENE ON  PREGNANCY  IN MICE386
Route of
administration
Intraperitoneal



Subcutaneous






Subcutaneous

Duration of pregnancy,
days
1
1
6-7
11
1
1
6-7
6-7
6-7
11
11
7
7
Total dose of 3-methylcholanthrene,
mg
0.5
1.0
1.0
1.0
2.0
5.0
1.0
2.0
5.0
2.0
5.0
2.0
5.0
Number of animals in which
pregnancy was interrupted
Total number studied
4/5a
7/7a
2/3
0/3
3/5a
ll/ll3
3/5
3/4
6/6
0/2
0/4
2/5b
7/7°
         a A total of 50 mice, in which vaginal plugs were found, received no treatment;  39 of these animals  became pregnant.

           Mice ovariectomised and  received 1.5 mg of progresterone per day.

-------
      Table  6-28.   DATA  ON MOUSE EMBRYOS TREATED WITH 0.01 ml  OF 3-METHYLCHOLANTHRENE
                                                                                                 387
Group
0.025* Methyl chol anthrene
in benzene9
0.025% Methyl chol anthrene
In benzene"
Sham-operated control
Number
pregnant
animals
19
10
6
Number
breeders
that
delivered
10
4
6
Number
embryos
treated
122
67
43
Number
young
born
34
19
39
Prenatal
mortality.
%
72.1
71.7
9.3
Tall
Hemorrhages defects
7 5
4 2

Other
defects
2
4

Abnormalities.
Total %
14 41.2
10 52.6
0 0
* Offspring delivered normally.
b Offspring delivered by Caesarean section.

-------
     Little success has been achieved in demonstrating a marked effect of BaP
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
                 388
occur in the rat.     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
                              389
and development in Swiss mice.     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
             390
doses of BaP.     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
                                     6-162

-------
the formation of pulmonary adenomas.  As the data in Tables 6-29 and 6-30
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
        391 392
potency.   '     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-
pa thic 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
                                     6-163

-------
    Table  6-29.  EFFECT OF BENZO[A]PYRENE ADMINISTRATION TO  PREGNANT MICE  ON PULMONARY  ADENOMA FORMATION

                                                  IN  THE PROGENY390
en
-P.


group
Absolute controls
Vehicle controls
Benzo[a]pyrene


Material
administered
Nothing
Carbowax-400
Benzo[a]pyrene
in carbowax-400

No of
mice
82
89
55

No. of

adenomas9
6
10
34

% of

adenomas
7.3
11.2
61.8

No. of

per group
7
13
130

No. of

per mouse
0.09
0.15
2.36

No. of
mice with
multiple
adenomas

3-6
0
1
13'


7-9
0
0
6

            a Number of mice with pulmonary adenomas at 28 weeks of age.

-------
      Table  6-30.   EFFECT  OF  BENZO[A]PYRENE ADMINISTRATION  TO  PREGNANT MICE ON SKIN PAPILLOMA
                                    FORMATION  IN THE  PROGENY390
Experimental
group
Absolute controls
Vehicle controls
Benzo(a)pyrene
No. of
mice
82
89
55
No. of
mice with
papillomas
6
8
13
% of
mice with
papillomas
7.3
9.0
23.6
No. of
papillomas
per group
8
9
18
No. of
papillomas
per mouse
0.10
0.10
0.33
a Number of mice with skin papillomas after 24 weeks of topical treatment twice weekly with 1% croton
  oil.  Mice were 28 weeks of age.

-------
                       392
hydroxymethyl compound.     Unfortunately, the use of large doses of DMBA and



its derivatives precludes the possibility of risk assessment under realistic



environmental exposure situations.
                                     6-166

-------
6.7  HUMAN STUDIES
6.7.1  Review of Epidemiologic Evidence
     Epidemiological evidence on the health effects  of polycyclic  organic
matter (POM) comes from both occupational  studies  and  community  air  pollution
studies.  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 can cause an elevated rate  of lung  cancer as
well as cancer of other sites and chronic  non-malignant lung disease.  Com-
munity studies, which involve much lower levels of exposure  and  are  more
problematic because of the heterogenous populations  involved,  attempt to
detect an association between community morbidity  and  mortality  rates and  some
direct or indirect index of air pollution.
     In both types of studies, exposure is not limited to  POM  alone  but
includes exposure to irritant gases, such  as S0?,  NO , particulate matter, and
                                               w    A
trace metals (e.g., arsenic, nickel, chromium) which may exert their own
effects or may enhance the effects of the  POM.  Additionally,  in both occupa-
tional and non-occupational air pollution  studies, smoking is  a  powerful
                                     393 394
causative factor of similar diseases;   '     and in  community  air  pollution
studies, occupational exposure represents  another  potential  causative factor,
especially in areas where a sizeable proportion of the work  force  is employed
                                                                395
in occupations with high levels of exposure to toxic substances.
6.7.1.1  Occupational studies—The most extensive  epidemiological  studies  of
occupational groups with exposure to POM are those of  coke plant workers and
                                     6-167

-------
of gas production workers.  Other relevant studies include those of wax
pressmen and roofers.  In most of these studies, exposure is not limited to
POM, and other substances are also usually present such as gases S02 and NO ,
particulate matter, and trace elements such as arsenic, nickel, and chromium.
POM is present in the fumes from coal carbonization and coal gasification in
concentrations that are orders of magnitude higher than community levels.  It
has been hypothesized that the proportion of carcinogenic compounds released
                                                             396
into the air increases with the temperature of carbonization.
6.7.1.1.1  Historical perspective.  Epidemiological studies in different
countries have demonstrated that workers exposed to the products of the
combustion and distillation of bituminous coal experience an increased inci-
dence of cancer of several sites (lung, pancreas, kidney, bladder, skin).
These studies are discussed below, and the overall results are summarized in
Tables 6-31 through 6-33.
     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
                                                        397 398
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.399'400-401
                                      6-168

-------
              Table 6-31.    SUMMARY OF-EPIDEMIOLOGICAL  EVIDENCE OF CARCINOGENICITY
Investigator
Kuroda and Kawahata
Kawai , Amamoto
and Harada
Henry. Kennaway
and Kennaway
Kennaway and
Kennaway
i
Z? Kennaway and
10 Kennaway
Doll
Doll
Date of
study
1936
1967
1931
1936
1947
1952
1965
Type of
study
Historical
prospective
Historical
prospective
Historical
prospective
Historical
prospective
Historical
prospective
Historical
prospective
8-year pro-
spective
Site
Lung
Lung
Bladder
Lung
Lung
Lung
Lung
Results
(12 lung cancer deaths out of a total of 15
deaths for all cancers) Relative risks cannot
be calculated but there is suggestion of
26-fold excess over general population
(Lloyd. 1971)
33 times the rate for other steelworkers
1-1/2 to 4-fold Increased risk relative to
general population
3-fold excess for "gas stokers and coke-oven
chargers"
81% excess of lung cancer deaths
Those with heavy exposure showed 694 excess
relative to those with minimal exposure (by-
Characterization of exposure
Producer gas workers
Workers in producer gas works
Workers in 5 out of 10 occupations
involving exposure to coal products
British producer gas workers;
chimney sweeps and several cate-
gories of gas-works employees
Former gas retort workers
(pensioners)
Gas-worker mortality was greatest
for those with greatest exposure
                                       Bladder
Doll
1972
4-year follow- Lung
up of 8-year
prospective
products workers)

Those with heavy exposure showed 4-fold excess
relative to those with minimal exposure

Heavily exposed workers showed highly signifi-
cant excess; by-products workers showed no excess
                                                                                           (retort house workers)

-------
Table  6-31  (continued).   SUMMARY OF  EPIDEMIOLOGICAL  EVIDENCE OF  CARCINOGENICITY
Investigator
Reid and Buck





Brunsgaard







Lloyd, Lundin,
Redmond, and
Geiser



Lloyd

Date of Type of Site
study study
1956 Retrospective Lung
study of
deaths among
coke plant
workers during
1949-1954
1959 Retrospective Lung
study of
deaths among
employed and
retired gas- Bladder
workers over
a 15-year
period
1970 Prospective Lung
study of
mortality
among steel -
workers over
work- area
1971 Prospective Lung
study
Results Characterization of exposure
No excess in number of cancer deaths among Coke plant workers
coke plant workers as a whole nor of lung
cancer for oven workers



6.4-fold increase over general population All deaths occurred among gas-workers
with at least 5 years experience; most
had more than 10 years

5 deaths that appeared to be a significant
excess


Coke plant workers had 2-fold excess Coke plant workers
relative to rate among steelworkers




Coke-oven workers showed 2-1/2-fold excess Coke plant workers
over steelworker population,
                               Pancreas  Top-side workers had 5-fold risk; top-side
                                       workers with 5 years full-time
                               Intestine Top-side had 10-fold risk, nonoven workers
                                       may have excess risk of digestive cancer

-------
Table 6-31 (continued).  SUMMARY OF EPIDEMIOLOGICAL EVIDENCE OF CARCINOGENICITY
Investigator Date of Type of
study study
Redmond, Ciocco, 1972 Follow-up to
Lloyd, and Rush prospective
study


Redmond 1976 Follow-up to
previous study

en
m—l
^ Redmond et al. 1976 Historical
prospective
of by-products
workers


Site
Lung
Kidney
Prostate
Lung
Intestine
Pancreas
Kidney
Lung
Pancreas
Intestine
Buccal
Pharygeal
Results Characterization of exposure
Coke oven workers had 1.34 excess risk Coke plant workers
relative to nonoven workers
Coke oven workers had a 7.49 excess risk
over nonoven workers
Coke oven workers had 1.64 excess risk
over nonoven workers
15.72 excess for full-time top-side oven
workers
2.37 excess for nonoven workers
4.29 excess for nonoven workers
5 excess for all coke plant workers;
3.31 excess for oven workers
4.95 excess for nonoven workers
2.93 excess for nonoven workers


-------
ro
                           Table 6-32.   SUMMARY  OF  MORTALITY DATA IN GAS  WORKERS  OBSERVED  BY  DOLL

                           A.  Causes of death of pensioners from 1939  to 1948 compared  with the experience of Londoners
Cause of death Expected deaths
at England and
Wales rates
Cancer and other tumors 123.5
Cancer of stomach and 25.0
duodenum
Cancer of lungs and pleura 10.4
All causes 831.5
London
weights
1.10
0.93
1.33
1.03
No. of deaths Test of significance of difference
Exnected at Observed between observed ft* expected value
London rates
135.9 156
23.3 32
13.8 25 <0.01
856.4 840
                                            ,406
                Table 6-32 A adapted from Doirub (1952),  Table 6-32 B adapted from Doll (1956)408.

                The values for P are probabilities with which as great or greater differences  between the observed  and expected deaths  might
                occur by  chance.  Values greater than  0.05 are not recorded.


                 B.  Standardized annual death rate per 1000 men:   all  four original  boards  grouped together,  and England and Wales
Cause of death
Cancer of lung
Cancer of bladder
Cancer of skin
and scrotum
Other cancer
Bronchitis
All causes
1 Sept. 1953 to
31 Aug. 1961
Class Aa
3.39 (55)
0.28 (4)
0.07 (1)
2.11 (36)
3.53 (51)
19.68 (304)
Class C^
1.16 (5)
0.00 (0)
0.00 (0)
2.41 (9)
2 10 (9)
15.00 (61]
E&W
2.05
0.17
0.02
2.57
1.61
18.66
1 Sept. 1961 to
31 Aug. 1965
Class A
4.08 (44)
0.42 (6)
0.19 (2)
3.02 (34)
2.42 (26)
21.69 (243)
Class Cj
1.78 (6)
0.29 (1)
0.00 (0)
2.37 (8)
3.12 (9)
14.50 (46)
E&W
2.24
0.17
0.02
2.51
1.64
18.69
1 Sept. 1953 to
31 Aug. 1965
Class A
3.82 (99
0.40 (10
0.12 (3)
2.70 (70)
2.98 (77]
21.21(547]
Class C1
1.59 (11)
0.13 (2)
0.00 (0)
2.39 (17]
2.57 (18]
14.91(107]
E&W
2.13
0.17
0.02
2.55
1.63
18.67
Significance of diff-
erence between observed
[Class A) and expected
(E&W), value of P
< 0.001

0.03; 0.02

< 0.001

              Standard  population -- total number of man-years at risk  for both occupational classes and all  four  boards, 1953-65.   The one
              untraced  man  in Class A is counted as alive at the end of the study.


              a Class A = heavy exposure, i.e., coal carbonizing process workers.

                Class C. =  exposure only to by-products, i.e., process  and maintenance workers in chemical  and  by-products plant.

-------
   Table 6-33.  SUMMARY  OF OBSERVATIONS ON OBSERVED  DEATHS AND  RELATIVE RISKS OF  DEATH FROM ALL CANCERS,
                             RESPIRATORY CANCER, AND CANCER OF THE  DIGESTIVE TRACT*
cr>
co
Distribution
of workers
Length
of employment, yr
Total coke oven
Coke
Oven
Oven
Oven
oven

topside full-time
topside part-time
side
only
Nonoven
No one
coke
plant area
5+
1860
993
ISO
290
553
836
31
10+
1194
574
72
245
257
578
42
15+
790
325
29
159
137
392
73
Deaths and RR1
malignant
5+
Obs.
166
101
35
26
40
65
0

RR
1.47b
1.66b
3.70b
1.59b
1.17
1.28
d
s of death
neoplasms
10+
Obs.
136
85
22
31
32
48
3

RR
1.50
1.95
5.12
1 85
1.46
1.10
d
from
15+
Obs.
b 108
" 63
b 12
b 32
19
39
6
Deaths and RR's of death from
respiratory cancer
5+ 10+ L5+
RR Obs. RR Obs. RR Obs. RR
1.62b
2 40b 54 3 02b 44 3.42b 33 4.14b
7.63b 25 9.19b 16 11.79b 8 15.72b
2.73b 12 2.29b 16 3.07b 18 4 72b
1.51 17 1.79C 12 1.99b 7 2.00
1.13
1.34

Deaths and
digestive



All malignant neoplasms of
Large

digestive

system
Intestine
Pancreas
Other





Obs.
28
11
8
9
5+
RR
1.58
2 31
3.67
0.83
RR's of death from
system among nonoven

Obs.
c 23
c 10
b 7
6
10+
RR
1.53
2.52
3.75
0.65

cancer of
workers
15+


Obs . RR

b
b

19 1.
8 2.
6 4.
5 0.
53
37C
29b
65
        ! Redmond411 (1976).
        ° P<0.01.
        5 P<0.05.
         less than 5 deaths.

-------
     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.
           402
     Kuroda    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.
6.7.1.1.2  Recent studies.  In a study of 504 deaths among former  gas workers
                                       403
at a Japanese steel plant, Kawai et al.    found six deaths from lung cancer
in contrast with the expected number, 0.180, using other workers at the same
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
general male 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 10 to 19 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 (10 to
19 years) showed no significant excess mortality.
     Bruusgaard    studied 125 deaths among former gas works employees in
Norway, all 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
                                     6-174

-------
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 significantly
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.
                  405
     Reid and Buck    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 1950-1954 among workers in a large unspecified
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.
                                     6-175

-------
     This study was criticized by Lloyd,     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-
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
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'  mortality  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,

                                     6-176

-------
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 < p< 0.02).
                              407
     In a separate study, Doll    carried out an 8-year prospective analysis
of mortality from different causes among several occupational  groups of gas
workers and retirees covering the years 1953-1961.   The study  included
11,499 men between 40 and 65 years of age at the start of the  study with five
or more years in the gas works plant.  Observed gas worker mortality rates
were compared with those expected in populations of England and Wales and
regional metropolitan areas.
     The workers were grouped into three classes according to  their exposure:
heavy exposure; intermediate exposure; light exposure, or exposure only to
by-products.  Again, elevated mortality was attributed to respiratory system
disease, specifically, cancer of the lung and bronchitis.  The lung cancer
mortality rate was 69 percent higher for the heavy  exposure group than for
the 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
was highest among workers with greatest exposure to the products of coal
carbonization.
     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
                                     6-177

-------
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 mortality
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,
Redmond, and coworkers 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 in-
reasing cancer response rates with increasing exposure and dose.
     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, quenching, discharging, and to their
escape through improperly sealed openings.
     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
                                     6-178

-------
gas and chemical products.  Since earlier work (e.g., Doll,     Kennaway and
Kennaway   '   ) 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, Lloyd    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
workers (17 versus 9.7 expected; significant at the 5 percent level).  Cancer
of the pancreas and large intestine accounted for the greatest excess.
                    409
     Redmond et al.,    in a follow-up of earlier reports in the series,
examined the mortality records of cohorts of coke oven workers in an expanded
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 1969 to
1966 and were compared with data from 10 other plants for the same  period.
                                     6-179

-------
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 cri-
terion for inclusion in the prior Allegheny County  study  was  employment in
one of seven steel plants during 1953).
     The findings of Redmond et al.409 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
dose.
     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-
oven experience.  These data indicate a definite gradient in  response based on
both type and duration of exposure.
                                           396
     Overall, the study confirmed the Lloyd    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).
                                    6-180

-------
     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,  which Mazumdar
further substantiated by calculating cumulative exposures  of the cohort to
coal tar pitch volatiles.
                              411
     In a later paper, Redmond    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 1.79; 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 6-33).
                                     412
     A recent study by Redmond et al.    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
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
                                     6-181

-------
importance since non-oven coke plant workers are exposed to considerably lower
concentrations of coke oven emissions and,  hence, of POM than  oven  workers.
                     413
     Hendricks et al.     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 aromatic compounds.  Workers handling the finished wax did not show  a
significantly elevated risk.
     Hammond et al.    studied the lung mortality among 5,939 roofers and
waterproofers over a 10-year period.  The study population included active,
probational, and retired union members with at least 9 years membership at
the start of the study (January 1, 1960).  The total U.S. male population was
used as a standard for calculating the expected numbers of deaths.
     Workers with 9 to 19 years' experience had slightly elevated standardized
mortality ratios (SMR's) for all causes of death (102) and for all  cancers
(107); and had a marked excess for respiratory disease, not including pneu-
monia, influenza, and tuberculosis (196).  Workers with more than 20 years'
                                     6-182

-------
experience showed excess mortality from all  causes  (109),  all  cancers  (145),
and respiratory disease, not including pneumonia,  influenza, and  tuberculosis
(167).
     The rates for cancer of specific sites  included:   lung, 159  (based  on
99 deaths); upper respiratory tract (buccal, pharyngeal,  larynx,  esophagus),
195 (based on 31 deaths); stomach, 167 (based on  24 deaths); colon,  rectum,
132 (based on 37 deaths); prostate, 138 (based on  29 deaths);  bladder,  168
(based on 13 deaths); skin cancer except melanoma,  400 (based  on  3 deaths);
leukemia, 168 (based on 13 deaths).  No levels of  statistical  significance
were given.
     The SMR for lung cancer was found to increase  with increased length of
employment:  0.98 (9 to 19 years); 1.52 (20  to 29 years);  1.61  (30 to  39 years),
     Assays of some environmental conditions were obtained from breathing
samples of dust collected on masks worn by workers  engaged in  various  roofing
occupations.  The assays showed that the inhaled  BaP could potentially  range
from 53.0 yg BaP/7-hour working day to 1.4 ug BaP,  with an average over all
jobs of 16.7 pg BaP.  The authors noted the  BaP was not the only  agent  con-
tained in hot pitch fumes and that the increased  mortality of  roofers  could
be due to other factors or to the combined effect  of BaP and one  or  more
other factors.  The effects of smoking were  not controlled for in this  study.
     A 40 percent excess lung cancer rate in males  in the south-central  area
                     415                       416
of Los Angeles County    led Menck and Henderson     to examine the lung cancer
rates associated with different occupations  and industries in  Los Angeles
County.  They reviewed all lung cancer data  for white males in Los Angeles
                                     6-183

-------
County for a 5-year period.  These included 2,161  death certificates in white
males aged 20 to 64 for the period 1968-1970 and 1,777 incident cases of lung
cancer in white males of the same age group of 1972-1973.   The authors justi-
fied pooling mortality and morbidity data by the "high mortality rate of lung
cancer and the relatively high accuracy of death certification regarding lung
cancer."  The occupation and industry of each subject (obtained from death
certificates) was coded into one of 417 occupational  codes, and one of 215
industry codes, per the 1970 U.S. Census occupational  classification system.
Detailed occupational and industrial population at risk figures for white
males were calculated from sample material from the 1970 U.S.  Census for Los
Angeles County containing information on the current occupation and industry
for 31,216 white males, aged 20 to 64.
     Expected deaths and expected incidence cases were computed for each
occupation, assuming the age-specific rates of cancer in each  occupation were
the same as those for all occupations.  The standardized mortality ratio
(SMR)--the ratio of observed death plus incident cases to the  expected deaths
plus incident cases—was used to measure the risk for each occupation.  Most
occupational groups showing an excess lung cancer rate had exposure to asbestos
or polycyclic aromatic hydrocarbons (PAH) or both.  Those occupations charac-
terized by exposure to PAH and found to be at excess lung cancer risk in Los
Angeles County included:  heat treaters (SMR = 433; p<0.05), metal machine
shop workers (SMR = 137; p<0.05); plumbers (SMR = 138, p<0.05); and steel
industry workers (SMR = 279, p<0.01) (see Table 6-34).
                                     6-184

-------
           Table 6-34.  STANDARD MORTALITY RATIOS  (SMR  =  100 FOR ALL  OCCUPATIONS)  FOR  OCCUPATIONS AT

                     SIGNIFICANTLY  INCREASED RISK  TO  LUNG CANCER. WHITE  MALES  20-64  YEARS,
                                             LOS ANGELES  COUNTY4I6
cr>
i
_j
00
en
Occupation
Asbestos* insulation workers
Roofers
Heat treaters
Dental lab technicians
Decorators
Taxi drivers
Mechanics, exluding auto
Photoengravers
Pressmen
Clothing ironers
Mariners, longshoremen
Shoe repairers
Mine operatives
Electricians
Bartenders
Plasterers, dry wall workers
Cooks
Machine shop workers
Truck drivers
Restaurant, bar managers
Painters
PI umbers
Welders
Janitors
Subtotal
All occupations
Estimated
population
at risk (23)
400
2,000
550
900
1,200
3,100
8,050
1,350
5,300
1,100
3,150
1,350
1,250
12,400
7,100
3,800
11,450
52,500
34,800
8,150
19,500
9,700
15,300
27,050
231,500
1,560,800
Number of
deaths
1968-1970
3
6
1
2
6
16
19
4
10
6
11
3
5
40
16
3
23
117
58
16
45
25
21
48
504
1,738
Incident
cases
1972-1973
2
5
3
3
1
7
27
4
10
3
12
4
3
18
19
14
18
82
51
11
42
11
27
61
438
1,511
SMR
878
496
433
405
358
344
332
320
276
267
266
233
217
205
204
200
180
167
165
159
158
138
137
127
175
100

-------
     The authors noted that since smoking data were not available for the
study population, it was not possible to attribute the elevated rates among
certain occupational groups to occupational  exposure alone.
6.7.1.2  Community studies--Non-occupationa1  epidemiological  studies have,  in
general, attempted to correlate excess malignant and non-malignant disease
with differences in urbanization or with increases in air pollution, measured
directly or indirectly.  Comparison of urban and rural lung  cancer mortality
rates when adjusted for age and smoking habits shows that people who live in
cities have roughly twice the rate of lung cancer of those who live in rural
populations (Table 6-35).   »418»419  It is  apparentj however, that not only
duration of residence in a polluted area but also the degree of mobility are
factors which may determine lung cancer mortality.  While there is general
agreement on the existence of an "urban factor," there is no consensus that
the "urban factor" is air pollution.  Alternative explanations that have been
proposed to account for the urban/rural difference in lung cancer mortality
are:  (1) migration to cities for medical treatment; (2) increased bacterial
and viral infection in cities; (3) increased bronchitis in cities; and (4)  in-
creased occupational exposure.
     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
            394
lung cancer.     A second confounding factor is occupational  exposure.  This
is of much smaller magnitude for the general  population than smoking, but it
                                                                           393
is thought to be large in comparison to any effect caused by air pollution.
                                      6-186

-------
00
                Table 6-35.  STANDARDIZED MORTALITY RATIOS  FOR  URBAN  AND  RURAL AREAS  BY  CURRENT
                          RESIDENCE AND DURATION OF RESIDENCE  IN  A  SPECIFIC  COMMUNITY3


Men


Women


Current
residence
Urban
Rural
Ratio
Urban
Rural
Ratio
All
durations
113
79
1.43
108
85
1.27
Less than
1 yr
166
154
1.08
163
205
0.80
1-9 yrs
107
88
1 .22
82
117
0.70
10-39
yrs
117
83
1.41
118
79
1.49
40 & more
yrs
117
75
1.56
124
86
1.44
Lifetime
100
50
2.00
86
49
1.76
     From Higgins  (443),

-------
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.
            420
     Hammond    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 pollution,
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
agents besides POM; and these may exert their own separate effects on health
or may interact with POM.  In some studies 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,
                                     6-188

-------
etc.).    It has also been shown experimentally that BaP alone does not ade-
quately account for the careinogenicity of complex mixtures such as air
                            422 423
pollution and cigarette tar.   '
     Early studies noted the increase in lung cancer mortality with increasing
urbanization.  Males in urban areas had roughly twice the lung cancer rate of
males in highly rural areas.
                  417
     Curwen et al.    reported that SMR's for lung cancer in both sexes and
cancer of the larynx in males increased with increasing urbanization.   The
SMR's for lung cancer in males in county boroughs (towns with 100,000 popula-
tion or over), urban districts (other towns), and rural districts (the rest
of the country) for the period 1946 to 1949 were 117, 86, and 62, respectively
(see Table 6-36).
                        418
     Hoffman and Gilliam    examined the distribution of lung cancer deaths in
the U.S. from 1948 and 1949.  They adjusted state lung cancer rates for
residence (urban-rural) in addition to age, race, and sex.  Using a broad
definition of urban population (persons living in places of 2,500 or more
people), they found mortality of urban residents to be substantially higher
than that of rural residents in all race and sex groups.  Among white males
the urban rate was 22.3 per 100,000 compared to 1.3 per 100,000 for rural
white males.  Among urban non-white males the rate was 16.9 compared to 7.3
for rural non-white males.  Urban women had higher rates than rural women:
4.7 compared to 3.7 for whites and 4.2 compared to 2.1 for non-whites.  It
was not possible, however, to determine whether these differences were due to
occupational, environmental, economic, diagnostic, or other factors.  Data
from Hoffman and Gilliam are given in Tables 6-37 and 6-38.
                                     6-189

-------
     Table 6-36.   MORTALITY IN ENGLAND  AND  WALES  (1946-1949) DUE TO CANCER
                           OF THE  LUNG  AND  LARYNX417
                                            Cancer of lung    Cancer of larynx
                                            Male
          Female    Male    Female
Total registered deaths, England
 and Wales

Average death-rate (per million)
32,547

   409
7,097

   80
3,296

   41
1,115

   13
Standard mortality ratio (SMR)
England and Wales
Greater London
North
Midland
Wales
Rest
County boroughsx
Urban districts I Outside greater London
Rural districts,)

100
137
100
93
79
83
(117
\ 86
162

100
132
98
93a
69a
89
108
88
74

100
113
100
103a
83a
94
118
93
76

100a
72a
120a
84a
202?
88a
106a
96a
128a
  Based on 100-400 deaths and therefore subject to a coefficient of variation
  of 5-10 percent.  No SMR in this table is based on fewer than 100 deaths.
                                      6-190

-------
Table 6-37.   AVERAGE ANNUAL  AGE,  RACE,  AND SEX SPECIFIC MORTALITY RATES PER
               100,000,000 POPULATION FOR CANCER OF THE LUNG AMONG THE  URBAN
               AND RURAL POPULATION OF THE UNITED STATES, 1948-19493 418
Urban5
White
Age (years)
Total crude rate
Total adjusted rate
0-29
30-34
35-39
40-44
45-49
50-54
55-59
60-64
65-69
70-74
75 and over
0-44
45-64
65 and over
Male
235.2
223.4
2.6
16.8
46.0
111.7
274.8
502.8
819.5
1,053.1
1,109.5
1,104.6
867.9
20.1
622.9
1,037.5
Female
53.2
47.4
1.6
7.9
17.6
28.0
54.8
81.2
126.0
172.7
228.2
279.4
319.0
6.9
102.7
272.3
Nonwhite
Male
140.7
169.1
3.9
29.0
55.3
146.4
265.3
495.6
709.3
761.8
586.6
476.0
501.6
26.2
497.8
535.8
Female
33.7
41.6
1.2
8.2
25.6
47.0
68.4
119.4
123.7
153.8
142.7
135.9
181.6
9.1
106.6
151.5
Rural
White
Male
118.0
123.23
2.1
10.5
34.6
66.4
158.1
302.3
453.4
560.6
556.7
570.2
488.9
11.6
348.7
539.0
Female
35.2
36.8
1.3
4.7
9.1
20.1
42.4
59.6
92.0
141.4
185.1
225.3
260.1
4.0
78.7
221.2
Nonwhite
Male
53.9
73.2
2.2
19.2
26.2
53.3
108.4
212.6
317.3
311.6
270.8
158.7
255.0
8.5
220.6
236.4
Female
15.3
21.4
0.5
5.3
5.1
30.2
54.0
27.9
79.4
60.1
89.2
133.7
52.6
3.1
53.1
89.6
  Deaths according to International List Nos. 47b, c, d, and e (1948) and Nos. 162 and  163 (1949);
  population according to 1950 census.
  Urban population is generally that living in places of 2,500 and over.
  Adjusted to the age distribution of the total United States population at the 1950 census.
                                          6-191

-------
     Table  6-38.   AVERAGE  ANNUAL AGE,  RACE,  AND SEX SPECIFIC  MORTALITY RATES  PER 100,000,000  POPULATION
                              FOR CANCER OF  THE LUNG IN THE  UNITED STATES, 1948-493 418
I
•_J
PO



White
Age (years)
Total crude rate
Total adjusted rate
0-29
30-34
35-39
40-44
45-49
50-54
55-59
60-64
65-69
70-74
75 and over
0-44
45-64
65 and over
Male
185.6
183.1
2.3
14.3
41.4
93.7
229.5
427.0
679.7
861.3
881.9
874.6
698.0
16.4
517.6
824.1
Female
46.0
43.6
1.5
6.7
14.4
25.1
50.4
73.7
114.0
161.7
212.9
260.2
297.8
5.7
94.2
254.0
Total


Nonwhite
Male
103.7
130.6
3.1
25.7
45.4
114.8
211.8
397.3
558.8
578.1
439.1
320.3
372.2
18.4
392.3
389.3
Female
26.3
34.1
0.9
7.3
19.2
41.6
63.8
87.9
107.6
118.4
120.5
135.0
124.5
6.6
88.0
125.2
All persons
110.1
110.1
1.9
11.0
28.1
61.0
139.6
249.3
392.9
501.3
515.3
530.5
462.9
11.2
300.6
503.1
        a Deaths according  to  International List Nos. 47b, c, d, and e (1948) and Nos. 162 and 163  (1949);
          population according  to 1950 census.
          Urban population  is  generally that living  in places of 2,500 and over.
        c Adjusted to the age  distribution of the total United States population at the 1950 census.

-------
                     424
     Manos and Fisher    studied the correlations between the age-adjusted
mortality data (1949-1951) and different indices based on sources  of air pol-
lution, urbanization, and socioeconomic factors for 163 metropolitan areas.
The mortality ratios in this study represented the mortality experience for  a
specific population group divided by an expected number of deaths  based on
the age-specific death rates among all  the population groups in the combined
163 metropolitan areas.  A total of 10,176 linear correlation coefficients
was obtained.  Of 50 causes of death examined, cancer of the esophagus  and
stomach, cancer of the lung, arteriosclerotic heart disease, and chronic non-
rheumatic endocarditis gave the highest correlations with various  pollution
indices (based on home heating, non-specific industry, specific industry, and
motor vehicle usage).  The authors emphasize the preliminary nature of  their
findings.  They point out that simple linear correlation does not  establish  a
causal connection but, rather, indicates an area for further investigation
which would take other variables into account.
                                                            425
     A similar study was carried out by Schiffman and Landau    in which
indirect measures of air pollution were tested for correlations with mortality
from different disease categories.  The working assumption was that the 20
standard metropolitan areas (SMA's) with the highest air pollution indexes
(based on:  (1) all fuel; (2) all coal; (3) all gas; (4) coke; (5) S02
exhaust emissions; and (6) organic exhaust emissions) would have higher
mortality rates from selected diseases than the 20 SMA's with the  lowest air
pollution indices.  Comparisons showed that the 20 highest SMA's had higher
SMR's than the 20 lowest SMA's by a  margin of at least 10 percent for  chronic
                                     6-193

-------
rheumatic heart disease, arteriosclerotic heart disease including coronary
disease, and non-rheumatic endocarditis.  The differences were significant at
the 5 percent level.  Other categories of heart disease did not show a cor-
relation with the air pollution indices.  SMA's with the highest indirect
measures of industrial air polltuion showed consistently higher mortality
ratios (1.1 and above) for cancer of the esophagus and stomach (higher in
each case by a margin of at least 15 percent) and for lung cancer (higher in
each case by at least 20 percent) than the 20 lowest SMA's.  Like Manos and
Fisher, Schiffman and Landau point out the preliminary nature of their
correlations which serve to identify areas for further study rather than to
establish definite causal associations.
     Levin et al.    compared age-adjusted incidence rates for several types
of cancer in New York State, Connecticut, and Iowa and found generally simi-
lar urban-rural differences for cancer of the respiratory system, esophagus
(males), and intestine and rectum (males).  The ratios of urban and rural
age-adjusted incidence rates per 100,000 population for lung cancer were 1.41
(New York State excluding New York City), 1.57 (Connecticut), and 2.84 (Iowa).
Within New York State, the urban-rural differences persisted when data for
metropolitan and non-metropolitan counties were examined separately (see
Table 6-39).
                           419 427
     Haenszel and coworkers   '    conducted one of the few studies of lung
cancer mortality as related to residence which controlled for smoking.  They
obtained lifetime residence histories (by means of interviews with family
members) for a 10 percent sample of all white male lung cancer deaths (2,553
                                     6-194

-------
VO
cn
       Table 6-39.   AGE-ADJUSTED3  CANCER-INCIDENCE  RATES  PER  100,000  POPULATION, BY PRIMARY SITE AND SEX
       FOR URBANb.C AND RURAL  AREAS:   NEW  YORK STATE  (EXCLUSIVE OF  NEW YORK CITY), 1949-51; CONNECTICUT,
                                        1947-51; AND IOWA, 1970*26
Urban
























a
b


c
d
e

Primary site

All sites
Buccal cavity and pharynx
Lip
Digestive system
Esophagus
Stomach
Intestines
Rectum
Other digestive organs
Respiratory system
Lung and bronchus, total
Primary
Unspecified
Larynx
Hale genital organs
Prostate
Urinary organs
Kidney
Bladder and other
Skin
Lympjiatic tissues
Leukemia
Rates adjusted for age to the
For New York State and Iowa,
classified by the U.S. Bureau
76 towns with high population
New York Connecticut

264.5
15.8
6.8
87.1
6.6
23.7
23.9
18.3
14.6
37.9
27.2
16.3
10.9
5.4
32 5
28.7
18.2
4.6
13.6
34.7
9.9
8.9
total population
'urban" includes

271.8
17.9
5.4
98.3
8.7
28.5
25.6
20.1
15.4
35.2
28.0


5.8
33.2
27.8
19.4
5.6
13.8
31.0
7.9
9.9
Iowa

350.6
32.1
17.3
87.6
3.9
24.7
26.9
16.3
15.8
32.8
29.0


2.9
43.5
37.8
25.1
4.8
20.3
72.1
12.2
d
of the United States.
all Incorporated places
of the Census under special
density.
Rates for urban and rural areas have not been

adjusted to
New York
HALES
212.8
12.9
3.8
68.6
3.7
20.8
17.9
13.8
12.4
26.%
19.3
9.3
10.0
3.7
27.7
24.1
13.8
3.5
10.3
32.6
8.0
6.8
1950.
of 2.500
Rural

Ratio
'Urban/Rural
Connecticut Iowa New York Connecticut

200.3
13.5
5.6
69.1
3.7
21.3
17.5
15.5
11.1
20.5
17.8


1.4
27.0
23.3
15.0
4.1
11.0
30.2
5.1
7.5

population and

250
24
15
68
2
24
18
12
10
12
10


1
39
34
16
3
13
52
6
d


.2 1.24
.1 1.22
.7 1.79
.7
.1
.7
.6
.4
•9
.1
.2


•4
.0
•2
.5
.5
.0
.27
.78
.14
.34
.33
.18
.43
.41


.46
.17
.19
.32
.31
.32
.1 1.06
.1 1.24
1.31

over, and certain other political
rules relating to population size and density;

correct for

bias in

reporting place

for Connecticut, "urban


1.36
1.33
0.96
1.42
2.35
1 34
1.46
1.30
1.39
1.72
1.57


4.14
1.23
1.19
1.29
1.37
1.25
1.03
1.55
1.32

subdivisions
" comprises


Iowa

1.40
1.33
1.10
1.28
1.86
1.00
1.45
1.31
1.45
2.71
2.84


2.07
.12
.11
.52
.37
.56
1.38
2.00
—

so


of residence on medical records.
Comparable Iowa data not available.
In Iowa, unspecified uterus was distributed between the cervix and corpus, based on additional
histologic-type information.





data collected for a sample of cases and on





-------
                                          AGE-ADJUSTED'
10
en
iuuic u-oj v <-«-"' <• mucuy . nuc-nuuuo I tu ^Mliv*tl\- ilNUiUCINOt KM I to rtK IUU,UUU HUrULAIl
SITE AND SEX FOR URBANb.c AND RURAL AREAS: NEW YORK STATE (EXCLUSIVE OF NEW YORK
CONNECTICUT, 1947-51; AND IOWA, 1 970*26
Urban
Primary site New York Connecticut

All sites
Buccal cavity and pharynx
Lip
Digestive system
Esophagus
Stomach
Intestines
Rectum
Other digestive organs
Respiratory system
Lung and bronchus, total
Primary
Unspecified
Larynx
Breast
Female genital organs
Cervix
Corpus
Uterus, unspecified
Ovary
Urinary organs
Kidney
Bladder and other
Skin
Lymphatic tissues
Leukemia
3 Rates adjusted for age to the
b For New York State and Iowa,
classified by the U.S. Bureau
76 towns with high population

248.2
3.8
0.4
62.9
1.1
11.7
25.7
11.9
12.5
5.1
3.2
1.3
1.9
0.3
57.4
58.1
26.5
15.8
1.9
9.5
7.3
2.8
4.5
23.9
6.9
5.8
total

261.
3.
0.
73.
1.
15.
28.
14.
14.
5.
4.


0.
59.
58.
22.
12.
8.
13.
7.
3.
4.
22
5.
7.
population of
"urban" includes all
of the Census under
density.
c Rates for urban and rural areas have

1
1
2
6
5
5
4
0
2
8
7


4
8
4
1
0
9
1
7
6
0
0
6
8
Iowa New York

351.6
7.2
1.2
68.5
0.6
13.1
29.3
10.0
15.5
9.0
7.8


0.9
78.0
88.7
46.8*
21. 2e
—
15.1
10.6
3.3
7.3
45.4
7.2
d
the United States,
FEMALES
218.3
2.9
0.3
55.6
1.2
10.1
22.3
9.9
12.1
4.2
3.0
1.2
1.8
0.3
49.9
53.2
23.4
14.9
2.1
9.2
5.6
1.9
3.7
20.9
6.1
5.3
1950.
Rural
Connecticut

204.4
2.6
0.8
59.0
0.7
12.0
22.2
13.6
10.5
4.2
3.8


0.3
49.4
43.4
15.0
8.3
7.2
11.4
6.3
2.9
3.3
18.5
4.2
5.2


Iowa

261.5
4.1
0.8
65.2
1.2
14.1
23.8
11.0
15.1
6.3
5.3


0.0
62.4
59.8.
26 4*
17. 3e
—
12.3
6 1
2.8
3.3
24 9
5.5
d


Ratio:
UN, BY PRIMARY
CITY), 1949-51;
Urban/ Rural
New York Connecticut

1.
1.
1.
1.
0.
1.
1.
1
1.
1.
1.


1.
1
1.
1.
1

1.
1.
1.
1.
1
1.
1.


14
31
33
13
92
16
15
20
03
21
07


00
15
09
13
06

03
30
47
22
14
13
09


1.28
1.19
0 25
1 25
2 14
1 29
1.28
1.03
1 35
1.38
1.24


1 33
1.21
1.35
1.47
1.45

1.15
1.22
1 24
1 21
1 19
1.33
1.50

incorporated places of 2,500 population and over, and certain other political subdivisions
special rules relating to population size and density, for Connecticut, "urban" comprises
not been adjusted to
correct for
bias in
reporting place
of residence on

Iowa

1.34
1.76
1.50
1.05
0.50
0.93
1.23
0.91
1.03
1.43
1.47


—
1.25
1.48
1.77
1.23

1 23
1.74
1.18
2 21
1 82
1.31
—

so
medical records.
Comparable Iowa data not available.
e . .
__ J J _ i










         histologic-type information.

-------
deaths) in the U.S. during 1958, and for a 10 percent sample of all  white
female lung cancer deaths (852 deaths)  in the U.S.  during  1958-1959.   The
major findings of the study were:   (1)  smoking showed the  strongest  correla-
tion with lung cancer mortality in all  population subgroups; (2)  insignificant
differences in lung cancer mortality by residence history  among non-smokers;
(3) higher SMR's for urban than for non-urban population in  every residence
duration class; ranging from 123 in cities of over half a  million to 65  in
farm areas for males, and from 132 in the large cities to  69 in farm areas
for females; (4) combined effects  of residence and smoking histories on  lung
cancer rates far greater than would be  expected on the basis of additivity of
separate effects (this finding was not  confirmed by the data on female lung
                                                                      428
cancer and does not agree with some other studies—Stocks  and Campbell,
Waller   ); (5) lower risks among  lifetime residents, both urban and rural,
than those of populations with more varied residence histories; and  (6)  high
lung cancer risk among two migrant populations:  U.S. farm-born and  foreign-
born residents of large metropolitan centers (see Tables 6-40 and 6-41).
It is difficult, however, to quantify the effects observed with respect  to
air pollution because no index of pollution was used in these studies by
Haenszel and coworkers.
     An epidemiological survey of lung  cancer in Japan considered both
smoking habits and the degree of air pollution by dust, sulfer dioxide,
                                                                  430
suspended particulates, trace elements, and aromatic hydrocarbons.     The
lung cancer death rate displayed a dose-related relationship to the  total
number of cigarettes smoked, regardless of the extent of air pollution.
                                     6-197

-------
    Table 6-40.  STANDARDIZED  LUNG-CANCER MORTALITY RATIOS* - ADJUSTED FOR AGE AND SMOKING HISTORY
                 BY POPULATION SIZE OF CURRENT RESIDENCE AND DURATION OF RESIDENCE IN A SPECIFIC
                         COMMUNITY (U.S. WHITE MALES, AGE 35 AND OVER, 1965)419
vo
00
Duration (years)

Current residence

White males, total d
Metropolitan counties
500,000 and over
50,000-500,000
10,000-50,000
2,500-10,000
Rural nonfarm
Nonmetropol i tan counties
10,000-50,000
2,500-10,000
Rural nonfarm
Farm

White males, total j
Metropolitan counties
500,000 and over
50,000-500,000
10,000-50,000
2,500-10,000
Rural nonfarm
Nonmetropol i tan counties
10,000-50,000
2,500-10,000
Rural nonfarm
Farm
a SMR (adjusted for age and
b Size as of 1950.
All
nil
durations

100

123
111
164
107
91

89
84
80
65

2381

534
529
203
87
243

204
115
304
162
smoking history


Under 1

160

500
175
257
91
394

111
140
87
76

101

10
14
9
4
26

12
7
14
5
by the indirect


1-9

98

126
106
201
64
85

92
68
94
77

447

51
71
58
21
74

45
20
82
25
method)


10-39C
SMR
105

122
121
159
107
96

90
108
79
77
Observed deaths
1046

179
259
96
36
102

99
67
140
68

40 and over0

106

128
111
165
324
101

98
54
83
53

406

136
108
31
12
25

24
14
34
22
= 100 for U.S. white males, age 35



Lifetime

81

116
86
78
259
41

71
56
57
50

381

158
77
9
14
16

24
7
34
42
and over, 1958.

c Excluding lifetime residents.
Ps\iin+ i AC in c fra nHa *•<•! mA + w
irts\1 i tan a tr*a» e






-------
     Table  6-41.
STANDARDIZED  LUNG-CANCER MORTALITY RATIOS3 - ADJUSTED  FOR AGE AND SMOKING HISTORY
BY NATIVITY AND FOR  THE NATIVE-BORN BY  POPULATION SIZE  OF BIRTHPLACE AND CURRENT
         RESIDENCE  (U.S. WHITE MALES, AGE  35 AND  OVER,  1958)419
cr>
vo
10
Native-born - birthplace1*
Current residence

White males, total A
Metropolitan counties
500,000 and over
50,000-500,000
10,000-50,000
2,500-10.000
Rural nonfarm
Nonmetropol i tan counties
10,000-50,000
2.500-10,000
Rural nonfarm
Farm

White males, total d
Metropolitan counties
500,000 and over
50.000-500,000
10.000-50,000
2.500-10,000
Rural nonfarm
Nonmetropol i tan counties
10,000-50,000
2.500-10,000
Rural nonfarm
Farm
Foreign-
born

138

144
154
266
116
126

87
98
89
75

521

204
118
63
15
51

21
9
26
14
All
places

93

123
109
136
102
89

87
79
76
56

1860

330
411
140
72
192

183
106
278
148
500.000
and over

106

117
75
133
32
111

74
129
82
53

320

200
25
32
5
33

7
4
13
1
50.000-
500.000

88

116
89
63
93
88

82
184
68
28
Observed
225

25
126
10
8
28

10
7
10
1
10.000-
50,000
SMR
88

139
82
124
77
59

76
79
67
145
deaths
180

24
29
31
7
11

49
6
15
8
2.500-
10,000

78

97
109
173
158
75

33
37
60
57

130

15
30
14
22
12

7
13
12
5
Rural
nonfarm

77

125
119
103
88
64

69
56
62
84

347

35
69
16
12
50

33
18
93
21
Farm

100

159
156
247
158
129

133
109
93
50

626

25
123
36
16
55

73
57
132
109
            a SMR (adjusted for age and smoking history by the indirect method) = 100 for U.S. white males, age 35 and over,  1958.
            b Size as of 1950
            c Includes 32 descendants for whom size of birthplace was not ascertained.
            d Counties in standard metropolitan areas.

-------
Furthermore, it was shown that the age- and smoking-adjusted  lung  cancer
death rate for males increased slightly with the extent of air  pollution.
Among females and non-smokers the lung cancer death rate was  not apparently
affected by the parameters of air pollution measured.   The authors speculated
on a possible synergistic effect of air pollution and cigarettes on the
incidence of lung cancer.
     In a study of respiratory cancer in ten Polish cities, an  apparent
correlation was found between cancer death rates and degree of  pollution by
                   431
representative POM.     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 6-42).  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 log normal distribution of incubation periods  for neoplastic
disease may complicate the attempt to correlate time of exposure with car-
                 432
cinogenic effect.
                                     6-200

-------
I
ro
          Table  6-42.   RANKING  OF  THE  CITIES  INVESTIGATED BY  THE  VALUES OF THE ATMOSPHERIC AIR POLLUTION
          INDICES  AND  DEATH  RATE IN THE  POPULATION  FROM CANCER  OF THE  LUNGS AND BRONCHI  (THE NUMBERS  IN
          PARENTHESES  INDICATE  THE RANK  OR  SEQUENTIAL POSITION  OF THE  CITY BY THE GIVEN  CHARACTERISTIC)^

Average concentration
City
Gdansk
Katowice
Kratow
Lodz
Poznan
Szczecin
Warszawa
Wroclaw
Zabrze
Dust in
yg/m3
121 (1)
293 (8)
248 (7)
155 (3)
155 (4)
175 (5)
133 (2)
191 (6)
336 (9)
Tar sub-
stances-,
in ng/m
21 (1)
33 (7)
34 (8)
28 (5)
23 (3)
26 (4)
21 (2)
30 (6)
38 (9)
in the atmosphere in
BaP

74 (7)
75 (8)
63 (6)
45 (2)
49 (3)
53 (4)
29 (1)
61 (5)
117 (9)
BcP
the period
BghiPer
3
In pg/1000 m
37 (7)
39 (8)
32 (6)
22 (2)
26 (3)
27 (4)
15 (1)
29 (5)
65 (9)
79 (7)
81 (8)
69 (6)
46 (2)
53 (3)
57 (4)
32 (1)
68 (5)
117 (9)
1966-1967
P Sta
y
107 (8)
95 (7)
90 (6)
54 (2)
55 (3)
61 (4)
35 (1)
79 (5)
146 (9)
ndardized combined
eath rate for the
ears 1962 and 1963
31.9 (7)
31.3 (6)
28.4 (5)
27.1 (3)
27.7 (4)
33.1 (8)
16.8 (2)
13.8 (1)
47.3 (9)

-------
   ,  Studies on lung cancer rates in migrant populations tend to indicate that
exposures occurring early in life can lead to the development of cancers, even
though the victims have migrated to a different environment.   In 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
                                            433
their original homeland and adopted country.     Other studies of cancer
mortality among migrants support the concept that these persons are affected
                                                                        434
by their former environments and length of exposure in that environment.   '
435,436,437,438  Jt ig suggested that differences in smoking  habits are
unlikely to account for discrepancies in lung cancer rates among migrant and
resident populations.
                    439
     Hagstrom et al.    analyzed mortality from various types of cancer by
air pollution level in a population with a relatively homogeneous socio-
economic profile (Nashville Air Polltuion Study).  All cancer deaths for
the years 1949 through 1960 were classified by pollution indices (SO,,
soiling, dustfall, 24-hour S0») and socioeconomic category (upper class,
middle class, lower class) as determined by residence at time of death.  No
association was found between lung cancer and air pollution.   However, for
stomach cancer, there were significant correlations with the  level  of
dustfall (the age-adjusted rate in "high" pollution areas was twice that in
"moderate" pollution areas); and for cancer of the esophagus, prostate, and
bladder, there were positive correlations with soiling.  Bladder cancer
deaths were highest for areas of high pollution as measured by all  four
indices (see Tables 6-43 - 6-45).
                                     6-202

-------
       Table  6-43.  ADJUSTED DEATH  RATES PER 100,000 FOR  CANCER AMONG THE POPULATION EXPOSED TO MODERATE

                     POLLUTION BY FOUR AIR POLLUTANTS BY JQCJOECONOMIC CLASS, NASHVILLE SMSA, 1949 THROUGH

                                                        196C
9
ro
o
CJ


Pollutant

S03
Soiling
Dustfall
24-hr S02

so3
Soiling
Dustfall
24-hr S02

so3
Soiling
Dustfall
24-hr S02
Estimated
population
Ian 1
1955

13,633
17,972
25,009
7,237

157,862
136,171
158,117
152.660

31,367
39,142
29,669
35,516
All cancer

No

304
387
459
173

2,472
2,016
2,495
2,365

513
663
551
605

Rate

146
141.15
131.66
139.59

128.92
129.99
129.83
129.06

148.47
150.41
156.49
155.47
Stomach cancer

No.

16
20
23
11

156
122
157
151

41
49
39
43

Rate

7.51
7.14
6.56
8.27

8.09
7.9
8.14
8.23

12.01
11.19
11.06
11.19
Large
Esophagus intestine
cancer cancer
Deaths

No. Rate
Upper Class
4 1.83
5 1.86
5 1.43
2 1.61
Middle Class
47 2.47
42 2.72
50 2.62
40 2.19
Lower Class
18 5.19
23 5.16
16 4.51
19 4.86

No.

27
36
39
14

234
178
236
216

47
57
46
47

Rate

12.97
13.07
11.21
10.59

12.08
11.52
12.19
11.71

13.83
13.17
13.2
12.26
Cancer of
prostate

No.

11
13
14
4

158
126
154
145

31
40
35
38

Rate3

12.45
11.17
9.6
6.96

17.35
17.55
17.16
17.01

17.81
18.57
19.71
19.61
Cancer of
bladder

No.

11
13
13
5

63
56
61
61

9
12
11
15

Rate

5.1
4.54
3.65
3.72

3.25
3.64
3.14
3.3

2.66
2.77
3.15
3.89
      a For age-adjusted prostate cancer rates, the standard population was the total male population.

-------
       Table  6-44.  ADJUSTED DEATH RATES  PER 100,000  FOR CANCER AMONG THE  MIDDLE  CLASS EXPOSED TO AIR
                     POLLUTANTS,  BY DEGREE OF EXPOSURE,  NASHVILLE SMSA, 1949 THROUGH 1960*39
en
i
PO
o


Pollutant

S03
Soiling
Oustfall
24-hr S02
Estimated
population
Ian 1
1955

8,178
14,934
12,924
20,801
All cancer

No.

187
314
217
391

Rate

149.95
152.59
145.35
140.7
Stomach cancer

No.

17
31
29
35

Rate

13.88
14.95
19.63
12.75
Large
Esophagus Intestine
cancer cancer
Deaths

No.
High
5
8
4
12

Rate
Pollution
3.82
3.87
2.59
4 29

No.

14
26
13
35

Rate

11.03
12.46
8.89
12.42
Cancer of
prostate

No.

15
24
19
29

Rate3

22 17
20.92
23.5
19.9
Cancer of
bladder

No.

7
9
8
11

Rate

5.41
4.31
5.35
3.87
Moderate Pollution
so3
Soiling
Dustfall
24-hr S02

S03
Soiling
Dustfall
24 -hr S02
157,862
136,171
158,117
152,660

20,131
35,066
15,130
12,710
2,472
2,016
2,495
2,365

253
582
200
156
128.92
129.99
129.83
129.06

144.66
123.75
130.91
138.4
156
122
157
151

18
38
5
5
8.09
7.9
8.14
8 23

10.27
8.02
3 38
4.3
47
42
50
40
Low
4
6
2
4
2.47
2.72
2.62
2.19
Pollution
2.2
1.26
1.21
3.79
234
178
236
216

17
61
16
14
12.08
11.52
12.19
11.71

10.83
12.54
11.19
13.17
158
126
154
145

11
34
11
10
17.35
17.55
17.16
17.01

17.98
15.41
15.96
20.66
63
56
61
61

4
9
5
2
3.25
3.64
3.14
3.3

2.8
1.81
3.39
1.85
       a For age-adjusted prostate cancer rates, the standard population was the total male population.

-------
       Table 6-45.   ADJUSTED DEATH RATES PER  100,000  FOR CANCER AMONG THE MIDDLE  CLASS POPULATION EXPOSED

                       TO SUSPENDED PARTICULATES (SOILING), BY DEGREE OF POLLUTION,  BY SEX, AND BY RACE,

                                       NASHVILLE SMSA,  1949 THROUGH  1960439
PO
o
in


Sex-Race

Hale
Female
White
Nonwhite
Estimated
population
Ian 1
1955

7,123
7,811
9.716
5,218
All cancer

No.

167
147
207
107

Rate

170.51
136.79
151.97
152.05
Stomach cancer

No.

22
9
20
11

Rate

22.29
8.54
14.33
15.29
Large
Esophagus intestine
cancer cancer
Deaths

No.
High
7
1
7
1

Rate
Pollution
7.08
0.85
5.07
1.27

No.

13
13
22
4

Rate

13.13
11.6
15.69
5.29
Cancer of
prostate

No.

24
0
16
8

Rate9

20.92

19.97
24.66
Cancer of
bladder

No.

4
5
5
4

Rate

3.97
4.65
3.64
5.42
Moderate Pollution
Hale
Female
Hhlte
Nonwhite

Hale
Female
White
Nonwhite
64.761
71,410
109,873
26,298

15.830
19,236
32,651
2.415
986
1,030
1,652
364

262
320
549
33
148.51
117.29
128.98
133.42

141.24
113.49
123.46
118.94
77
45
78
44

16
22
36
2
11.70
5.09
6.09
16.51

8.65
7.68
7.96
6.87
32
10
29
13
Low
2
4
6
0
4.77
1.14
2.28
4.69
Pollution
1.13
1.24
1.33
._.
79
99
152
26

24
37
55
66
11.95
11.03
11.87
9.66

12.91
12.52
11.96
22.14
126
0
90
36

34
0
32
2
17.55
—
15.37
26.67

15.41
—
15.31
13.68
36
20
50
6

5
4
8
1
5.6
2.23
3.91
2.26

2.72
1.18
1.66
3.53
        For age-adjusted prostate cancer rates, the standard population was the total male population.

-------
                        440
     Winkelstein et al.t    using mortality data from Buffalo and Erie
County, New York, for the years 1951-1961, found a positive association
between air pollution as indexed by suspended particulates, and total
mortality and chronic respiratory disease, but none with lung cancer.
Economic status was taken into account by classifying the population into
five levels, based on a median family income assigned to each census tract.
Although low and high pollution levels were strongly correlated with high
and low economic status respectively, there was still a consistent trend of
increasing mortality with increasing air pollution within each of five
economic levels.
     Information concerning occupation, smoking, and residence history was
not available in this study.  Winkelstein and coworkers later reported
                                                                 441
associations between stomach cancer and particulate air pollution    and
                                          442
between prostate cancer and air pollution.     These two additional  findings
                                             439
confirmed similar findings of Hagstrom et al.    based on the Nashville Air
Pollution Study.
     Hammond    found no correlation between lung cancer death rates (adjusted
for age and smoking) and suspended particulate matter (in 33 cities) or
benzene soluble organic matter (in 31 cities) among men without occupa-
tional exposure.  The data are given in Table 6-46.
            443
     Higgins    found no association between BaP concentrations or particulate
matter and lung cancer death rates in some 50 Standard Metropolitan Statisti-
cal Areas (MSAS).  But there was an expected correlation between sulfate
levels and lung cancer death rates.
                                     6-206

-------
     Table 6-46.  OBSERVED AND EXPECTED NUMBER OF LUNG CANCER DEATHS BY PLACE OF RESIDENCE AND BY
                  OCCUPATIONAL EXPOSURE TO DUST, FUMES, GASES, OR X-RAYS (ADJUSTED FOR AGE AND
                  SMOKING HABITS: CONFINED TO MEN WHO HAD LIVED IN SAME NEIGHBORHOOD FOR LAST
                                               10+ YEARS)420
ro
O
Occupational ly exposed
to dust, fumes, etc.
Place of residence
Total, all male subjects
Metropolitan area, pop. 1,000,000 +
City
Town or rural
Metropolitan area, pop. <1 , 000, 000
City
Town or rural
Non-metropolitan area
Town
Rural
Los Angeles, Riverside, and Orange
Counties, California
Farmers
8 Cities: High particulates (130-180 ^g/m3) ,
11 Cities: Moderate particulates (100-129 ug/mj)
14 Cities: Low particulates (35-99 ,jg/m3)
9 Cities: High benz. sol. (8.5-15.0 ug/m3) ,
10 Cities: Moderate benz. sol. (6.5-7.9 ug/m )
12 Cities: Low benz. sol. (3.4-6.3 i.g/m3)
Observed
number
576
165
92
73
166
92
74
245
102
143

30
63
45
21
48
28
44
33
Exposed
number
530.5
134.1
69.1
65.0
145.4
83.3
62.1
251.0
104.9
146.1

21.9
77.6
32.9
18.8
37.4
21.0
32.7
29.2
Ratio
1.09
.23
.33
.12
.14
.10
.19
0.98
0.97
0.98

1.37
0.81
1.37
1.12
1.28
1.33
1.35
1.13
Not occupational ly exposed
to dust, fumes, etc.
Observed
number
934
281
168
113
271
170
101
382
200
182

38
71
66
39
110
52
65
76
Exposed
number
979.7
285.7
158.3
127.4
280.5
184.0
96.5
413.5
199.1
214.4

39.6
92.9
73.9
49.5
100.1
51.5
75.1
81.8
Ratio
0.96
0.98
1.06
0.89
0.97
0.92
1.05
0.92
1.00
0.85

0.96
0.76
0.89
0.79
1.10
1.01
0.87
0.93

-------
     On the assumption that populations exposed to PAH from the petroleum
industry could be expected to have higher rates for certain cancers  than
                                  395
unexposed populations, Blot et al.    conducted a survey of cancer mortality
from 1950 to 1969 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 lung, the  nasal
cavity and sinuses, and the skin (including malignant melanoma) compared
with the residents of counties with similar demographic features but with-
out petroleum processing plants.  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 per-
cent level); 1.10 for melanoma and other skin cancers (significant at
5 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 occupa-
                                       443
tional exposure.  Data from Blot et al.    are given in Table 6-47.
     Using death certificate data from 1968-1969, Menck et al.415 found
elevated lung cancer mortality rates in white males living in certain
heavily industrialized areas of Los Angeles County.  Populations at  risk
were determined from the 1970 Census data.  Age-adjusted lung cancer
mortality rates for Caucasian males and females were calculated for  13
                                     6-208

-------
Table 6-47.  RATIOS OF AGE-ADJUSTED MORTALITY RATES,  1950 TO 1969,  AMONG WHITE
                                 1DUSTRY COUNT I E<
                                  BY CANCER SITE^
MALES IN PETROLEUM INDUSTRY COUNTIES TO THOSE IN CONTROL COUNTIES
                                  r443
                  Cancer site                          Rate
          Buccal  cavity and pharynx                    1.04
          Esophagus                                    1.06
          Stomach                                      1.09a
          Colon                                        1.02.
          Rectum                                       1.07°
          Liver                                        1.06
          Pancreas                                     1-05a
          Nasal cavity and sinuses                     1.48
          Larynx                                       1.09
          Lung                                         1.15a
          Prostate                                     °-98h
          Testis                                       1.10°
          Kidney                                       1.05
          Bladder                                      1.02.
          Melanoma and other skin                      ^-^b
          Brain                                        0.94
          Thyroid and endocrine                        1.04
          Bone and connective tissue                   0.98
          Hodgkins's disease                           0.96
          Other lymphomas                              1.01
          Multiple myeloma                             1.05
          Leukemia                                     1.03

          All  sites combined                           1.06a
          a P<.01.

          b .01
-------
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.
     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.
                                     6-210

-------
                      444
     A follow-up study    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
                         444
County.  Henderson et al.    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 occupation did not explain  the excess male risk in  the
south-central area.  The authors note that detailed work histories might
reveal an occupational risk.  It was suggested  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.
     The authors are currently undertaking a case-control study of smoking
and occupational and residential hi sotry in south-central Los Angeles
County to determine whether air pollution contributes to the excess lung
cancer rate.
     Stocks has carried out many studies relating lung cancer mortality to
                        445
air pollution.  In 1952,    he reported that, in towns, lung cancer mortality
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
                                     6-211

-------
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.
                        428
     Stocks and Campbell    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 PAH 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 non-smokers 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
non-smokers in urban and rural areas.  The data are given in Table 6-48.
           446 447
     Stocks   '    reported significant correlations between SMR's for lung
cancer and bronchitis and air pollution (undissolved deposit and smoke)
                                     6-212

-------
     Table  5-48.   LUNG CANCER DEATH RATES  (MEAN ANNUAL PER 100,000 FROM MID-1952 TO MID-1954)  OF MEN IN
                     RURAL, MIXED, AND URBAN  AREAS CLASSIFIED ACCORDING TO  PAST SMOKING HABITS428
INS
oo
Ages 45-54
Smoking category
Non-smokers
Pipe smokers
Cigarette-
Light
Moderate
Heavy
Number of deaths
Rural
0
0

69
90
117
16
Mixed
0
0

57
83
214
26
Urban
31
104

112
138
205
124
Ages 55-64
Rural
0
34

70
205
626
25
Mixed
0
59

224
285
362
56
Urban
147
143

378
386
543
232
Ages 65-74
Rural
70
145

154
362
506
27
Mixed
0
26

259
435
412
36
Urban
336
232

592
473
588
183
S
Rural
14
41

87
183
363
68
.D.R. 54-74*
Mixed
0
25

153
132
303
118
Urban
131
143

297
287
394
539
         Standardized rate based on Liverpool population.

-------
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.
                          448
     In an expanded study,    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 correlation of four different PAH  indicated that
BaP was the "substance of prime importance" for lung cancer.  The  trace
elements beryllium and molybdenum also showed associations with lung cancer.
           449
     Stocks    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
PAH) 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 19 countries.  Smoking appeared to account for about two-thirds of the
deaths in the average country and coal consumption for about one-third.
                                     6-214

-------
     (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
social  and other factors were eliminated.  This excess  was attributed  to  air
pollution.
     The analysis of lung cancer and mortality in relation to smoking  and
                                                        450
per capita fuel consumption in 19 countries was repeated    with  more  recent
mortality data.  When lung cancer rates were adjusted for smoking differences,
the adjusted rates were generally higher in countries with high coal consump-
tion than in countries with low coal consumption.  Smoking showed a greater
effect than air pollution from solid fuel burning on the lung cancer rate
in men 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  55-64, air pollution
showed a greater effect even when smoking was held constant (the  correlation
coefficients were 0.599 for solid fuel and 0.380 for smoking).
                     451
     Carnow and Meier    carried out a regression analysis similar to  that
                   449
reported by Stocks.     They related lung cancer mortality rates  for both
sexes and for white and non-white populations during the years 1959 to 1961
to BaP concentrations in the 48 contiguous United States.  BaP was measured
quarterly at various urban and rural locations in 1967-1969, and  an urban-
rural weighting, based on the urban fraction of each state, was used.
Final estimates of BaP for each state were obtained by  a series of averaging
procedures.  Cigarette smoking was reflected in the regression analysis by
using cigarette sales per person aged 15 and over.  The regression coeffi-
cients for BaP for each age-specific group of white males indicated "that an
                                     6-215

-------
Increase in urban pollution associated with an average BaP concentration of
           3        3
1 pg/lQOO m  (1 ng/m ) corresponds to an increase of 5 percent in the death
rate."  Carnow and Meier observed that on the basis of this correlation, a
60 percent reduction in urban air pollution could be expected to reduce the
deaths of smokers and non-smokers from lung cancer by 20 percent.  This
analysis figured prominently in the National  Academy of Science's report on
Biologic Effects of Atmospheric Pollutants:  Particulate Polycyclic Organic
       452
Matter.     Data from Carnow and Meier are given in Table 6-49.
     The Carnow and Meier study has been criticized for using inadequate
cigarette consumption data and crude BaP measurements which could be
expected to have little validity for an area as large as a state.   *443
     Taking a different approach based on the risk of highly exposed workers,
           394
Pike et al.    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/m3 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/mJ 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 0.4/10$ (160/105 * 440) extra lung cancer  cases per
     year..."
     This calculation yielded an excess rate of 18/100,000 lung  cancer
cases per year for a city with 50 ng/m  BaP air pollution.  The  authors
took this calculation and Stocks's comparable findings as support for the
                                     6-216

-------
     Table 6-49.   REGRESSION ON  LUNG CANCER DEATH  RATES IN THE UNITED STATES PER 1 MILLION POPULATION
                                                                                                          a  451
i
ro
Age group
Male - white
Age-adjusted
35-44
45-44
55-64
65-74
Male - nonwhite
Age-adjusted
35-44
45-54
55-64
65-74
Female - white
Age-adjusted
35-44
45-54
55-64
65-74
Female - nonwhite
Age-adjusted
35-44
45-54
55-64
65-74
Average
death rate

867.4
101.6
497.8
1,405.8
2,064.7

844.6
132.3
606.0
1,367.6
1,722.4

129.5
32.9
86.5
163.1
255.2

156.3
34.31
85.8
184.9
381.7
Constant
(CQ)

460.7
86.5
125.2
894.4
835.4

184.2
22.8
245.9
141.6
162.6

92.1
12.7
36.4
113.5
184.7

151.5
66.0
32.4
-375.0
1,286.4
Tobacco sales
(Cj)

11.8
0.3
3.0
14.4
35.3

16.5
1.6
7.8
34.0
38.8

1.0
0.8
1.1
2.0
1.7

0.1
-1.0
-0.4
17.3
-19.2
BaP
concentration,
(C2)
ug/1 ,000 cu m

47.5
4.6
18.7
70.5
152.6

133.1
46.4
97.7
179.0
318.8

7.1
-1.1
14.2
-6.2
16.3

1.2
-1.1
46.6
44.5
-246.1
                 3 Tobacco sales  (average 28.8) expressed as dollars spent per capita by individuals
                   15 years of age or older.  Average BaP concentration was 1.38 pg/1,000 cu m.

-------
"notion of a simple proportional  relationship between Increasing BaP concen-
tration in the air and excess rate of lung cancer."   Pike et al.  further
calculated that the excess lung cancer rate associated with  smoking  one
cigarette per day (in the U.S.) was equivalent to the rate due  to exposure  to
10 ng/m  BaP in the air.
     While occupational studies have provided strong evidence that exposure to
high levels of POM causes an increased risk for cancer of several sites  and
for non-malignant respiratory disease, the results of general air pollution
studies remain inconclusive.  At present, there seems to be  no  absolute
consensus that the differences between lung cancer rates in  urban and rural
areas are due to general air pollution.  In his review of the epidemiology  of
                                443
lung cancer in the U.S., Higgins    has written that "while  an  urban effect is
undoubted, it is still not certain that it is due to carcinogenic pollutants
                                    453
in the air."  Goldsmith and Friberg,    in their review of the  literature on
air pollution and lung cancer, conclude that the evidence does  not support  the
hypothesis that "pollution per se is the urban factor," but  they add that it
is also not possible to reject the possibility.
                                     6-218

-------
6.8  ECOLOGICAL EFFECTS
     POM is ubiquitous in the natural environment, arising partly from human
activity (e.g., petroleum spillage, transport of combustion products), but also
originating from natural  sources, such as biosynthesis by plants, algae, and
         454 455
bacteria,   '    although some investigators have argued that aerobic and
                                         456
anaerobic bacteria do not synthesize POM.     Therefore, an assessment of the
ecological impact of POM in our terrestrial  and aquatic environments is com-
plicated by uncertainty over the natural  role which these compounds may have
in contributing to environmental stability.
                        457
     Payer and coworkers    have indicated that the green alga, Scenedesmus
acutus, may contain significant quantities of POM which vary depending on
the geographical location where it is grown (Table 6-50).  From the log-
normal  distribution of carcinogenic POM which was detected, these investi-
gators concluded that most of the POM present was due to bioaccumulation
from environmental (i.e., anthropogenic)  sources as opposed to endogenous
formation.  Effects of these POM on algal growth were not indicated.
     Although it was reported that phenanthroquinone, a degradation product of
phenanthrene, reduced the survival of various algal species at concentrations
                      458
of 40 to 800 yg/liter,    others demonstrated that POM at low levels can
                     459
enhance algal growth.     The growth of Chlorella vulgaris, Scenedesmus
obliquus, and Ankistiodesmus oranum'i was enhanced by exposure to 10 to
20 vig/liter fluoranthene; 1,12-benzoperylene; 3,4-benzofluoranthene; indeno-
(1,2,3,cd)pyrene; 1,2-benzanthracene; BaP; and DBahA.  Moreover, the degree of
growth enhancement appeared to correlate with carcinogenic potency.  Following
                                     6-219

-------
Table 6-50.   POLYCYCLICS CONTENT (ppb) OF  DRIED MICROALGAE  AS  A FUNCTION OF GEOGRAPHICAL LOCATION3 457





CD
1
ro
r\i
0

3,4-Benzopyrene
3,4-Benzofluoranthene
Indenopyrene
Fluoranthene
1 ,12-Benzoperylene

11 ,12-Benzofluoranthene
Dortmund,
Germany
39.5 (2.149^)
85.4 d.795^1)
50.9 (1.923^)
444 (1.505^)
52.1 (1.932-1)

36.3 d.863^1)
No. of
samples
19
19
19
18
19

19
Bangkok,
Thailand
1.39 (1.605^)
5.94 d.639^1)
2.36 (l.Sie^1)
91.2 + 38.2
3.89 + 2.36

2.40 +_ 1.02
No. of
samples
15
14
8
8
8

8
   The  scattering of values is indicated by median values and scattering factors in the case of
   lognormal distributions, or by  arithmetic means and  standard deviations  in  the case of normal
   distributions.

-------
the demonstration that Chlorella yulgaris could utilize acetate in the growth
medium to biosynthesize BaP, an hypothesis was presented that POM may act as
                                        455
an endogenous growth promoter in plants.     This contention is also supported
by the observation that growth of higher plants (rye, radish, tobacco) is
                                                           459
promoted by POM in the same fashion as the growth of algae.      However,  in
comparison to the POM accumulated in plants from anthropogenic sources,  the
contribution by endogenous synthesis is likely to be quite insignificant.   '
     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, Mass and Applegate
reported that a structure-activity correlation may exist for the effects  of
polycyclic aromatic hydrocarbons on the growth of Escherichia coli.   At  con-
centrations in the medium of 10   to 10"  molar, anthracene, phenanthrene,
chrysene, DBacA, and pentacene inhibited bacterial growth.  The more angular
configurations, 1,2-benzanthracene, DBahA, and BaP, promoted the growth  of £._
coli.  It was concluded from these limited data that growth promotion may
require the presence of POM with an angular acene conformation, whereas  inhi-
bition might occur with both linear 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.
                         462
Instead, Calder and Lader    reported that aromatic hydrocarbons inhibited the
growth of Serratia marinoruba and Vibrio parahaemolyticus in a manner which
was dose-related and a function of water solubility.  Thus,  a saturated  solu-
tion of BaP would have the same impact on bacterial growth as a saturated
                                     6-221

-------
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 en-
vironmental toxicity than the higher weight molecules because the former are
more readily lost by volatilization and degradation.
     It has also been suggested that once polycyclic aromatic hydrocarbons are
absorbed by microorganisms they can be transferred throughout the cell popula-
tion by cell surface contact.  Studies with various yeasts grown in the
presence of BaP demonstrated that greater than half of the hydrocarbon accumu-
lated by yeast cells could be transferred to recipient yeast cells which were
                ,    ..   463
grown in a normal medium.
     Sludge microorganisms were adversely affected by a variety of carcino-
          464
genie POM.     An inhibition of oxygen uptake of varying degrees was obtained
with different sludge microorganisms exposed to:  DBahA; 7-methyl-1,2-benz-
anthracene; 1,2,4,5-dibenzopyrene; MCA; 2-nitrofluorene; 2-fluoreneamine; N-2-
fluorenylacetamide; 7,9-dimethylbenz[c]acridine; 7,10-dimethylbenz[c]acridine;
dibenz[a,h]acridine; dibenz[a,j]acridine.
     The toxicity of POM has not been extensively studied in fish, amphibians,
or aquatic invertebrates.  Nevertheless, it is known that marine fish and
invertebrates can accumulate BaP from polluted waters.   '     Furthermore,
                                     6-222

-------
                     Table 6-51.   EFFECT OF AROMATIC HYDROCARBONS  IN MARINE BACTERIA'
                                                                                    462
ro
ro
co


Hydrocarbon
Naphthalene
2-Methyl naphthal ene
2 , 6-Dimethyl naphthal ene
Phenanthrene
Pyrene
Benzopyrene
2 , 3-Di hydroxynaphthal ene
a In distilled water at 25°C,
U mm — ——Jj-.-J-^ • £ 4- L« A 4> A u 4 M 4 4- n


Mol wt
128.16
142.19
156.22
178.22
202.24
252.3
160.16
1 a tin.


(A> a
Solubility3 M
2.43 x 10"J
1.73 x 10";
0.83 x 10"?
0.66 x 10"?
0.73 x 10":
~0.30 x 10 ~i
>16.5 x 10"



(B) b

0.4
2.1
12.8
4.0
170
881
.68


(C)
Toxicity index,-
(C = A x B x 10°)
97
363
106
26
124
>26
>1,120

, ^ i . • IJ.B
        to growth of controls.

-------
fresh-water and marine fish are capable of metabolizing a wide variety of
                                465             14                  3
hydrocarbons.  Lee and coworkers    showed that   C-naphthalene and  H-BaP
were readily taken up through the gills of three marine fish - mudsucker,
sculpin, and sand dab.  BaP was metabolized in the liver and, within a few
hours, metabolites were transferred to the gall bladder and excreted pre-
dominantly in the urine.  The main product of BaP metabolism was 7,8-dihydro-
7,8-dihydroxybenzopyrene, which indicated that the pattern of biotransforma-
tion was similar to that in the rat.  With naphthalene, however, a greater
tissue uptake was demonstrated than for BaP, and excretion of metabolites was
principally via the feces.  The rate of elimination of radioactivity derived
     14                                      3
from   C-naphthalene was more rapid than for  H-BaP.   Overall, it was con-
cluded that marine fish are able to dispose efficiently of certain non-halo-
genated polycyclic aromatic chemicals.
     The fact that sub-mammalian species can metabolize POM would lead one to
speculate that they may also be able to form activated intermediates from
these compounds.  Indeed it has been shown that tumors can be readily induced
in fish and amphibians by exposure to chemicals which are known to be carcino-
genic in mammals.  In addition, Ahokas and coworkers     recently demonstrated
that the oxidative metabolism of BaP by trout liver led to the formation of
reactive intermediates which became covalently bound  to 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 microsomes metabolized BaP 15 to 30 times faster when the quantity
                                     6-224

-------
of BaP metabolites produced In 15 minutes was expressed per nmole of cyto-
chrome P-450 or per unit of cytochrome c reductase.      Hepatic levels of
cytochrome P-450 are higher than in the rat, whereas cyctochrome c reductase
levels are lower.  However, the Vmax is higher than  for the rat and the Km is
lower when BaP is used as a substrate.
     Carcinogen!city 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
           ACO
Christensen    summarized much of the early work performed with the newt, a
member of the salamander family.  These studies showed that an  early prolifer-
ation 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, Xenopus laevis. that
implantation of MCA crystals-induced lymphoid tumors, a similar study  was
                                            469
undertaken to confirm this result using BaP.     The implantation of BaP
crystals (1.5 mg) in the abdominal cavity of adult Xenopus laevis laevis
produced lymphosarcomas in 11  of 13 animals within 86 to 288 days.  Advanced
tumors were found which affected the liver and kidneys and which were  trans-
pi antable to immature recipient Xenopus.
     POM's are widely distributed in the environment as evidenced by their
detection in sediments, soils, air, surface waters,  and plant and animal tis-
     470
sues.     The ecological impact of these chemicals,  however, is uncertain.
                                     6-225

-------
Numerous studies show that despite their high lipid solubility,  ROM's  show
little tendency for bioaccumulation in the fatty tissues  of animals  or man
(see Section 6.1).   This observation is not unexpected,  in  light of  convincing
evidence to show that ROM's are rapidly and extensively metabolized  (see
Sections 6.1 and 6.2).   Since only low levels of ROM's are  detected  in plants
                    470
and lower organisms,    transfer of ROM's through the food  chain does  not  seem
likely.  The direct impact of ROM's on plants, animals, or  the ecological
balance of nature is difficult to evaluate, since few data  are available which
suggest that adverse effects may occur.
                                     6-226

-------
                             SUMMARY AND CONCLUSIONS

6.1  ABSORPTION, DISTRIBUTION, AND EXCRETION
     In the environment, exposure to ROM's occurs by direct inhalation of
polluted air and tobacco smoke, and possibly in ingestion of contaminated
food and water, or by dermal contact with soot, tars, and oils.   Regardless
of the route of exposure, it can be demonstrated in laboratory animals that
ROM's are readily absorbed across all epithelia which are in contact with
the external environment.  The fact that POM's are generally highly lipid-
soluble neutral molecules greatly facilitates their passage through the
predominantly lipid-like cell membranes of animals, including man.
     Under environmental conditions, POM's reach the lungs by adsorption on
carrier particles.  Moreover, the regional deposition and retention of
inhaled POM's 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 size, 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
                                      6-227

-------
ciliary action does not necessarity remove them from the body,  since most
materials cleared from the lungs are subsequently swallowed,  thus  allowing
for absorption via the gastrointestinal  tract.
     Upon reaching the bloodstream, ROM's are rapidly distributed  to most
internal body organs.  Under experimental conditions with laboratory animals,
the route of exposure has little apparent influence on the tissue  localiza-
tion of ROM's.  Extensive localization in the fat and fatty tissues  (e.g.,
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
in 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  H-BaP from various tissues
(lung, liver, kidney, urine, feces) of the hamster following intratracheal
instillation appeared to be biphasic.  The slow phase of elimination from
the lung (involving less than 1 percent of the treatment dose)  seemed
dependent on the administration of  H-BaP together with an adsorbent.  It
is 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 animals is dependent
upon the co-administration of particulates, others maintain that carrier
particles are more important in determining localization in the respiratory
tract.
                                     6-228

-------
6.2  METABOLISM AND METABOLIC ACTIVATION
     The relative lack of chemical reactivity for tumorigenic POM's has in
the past been puzzling in light of their biological effects.  More recently,
however, it has become recognized that these molecules are enzymatically
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

-------
     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 predomi-
nantly in the liver.  In conjuction with various P-450 type cytochromes,
this enzyme complex is involved in detoxification of many xenobiotics, but
may also catalyze the formation of reactive epoxide metabolites leading to
carcinogenesis.  A second microsomal enzyme, epoxide hydrase, converts
epoxide metabolites of ROM's to vicinal glycols, a process which may also
have critical importance to carcinogenesis.
     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
most widely studied of the POM's has been BaP, one of the principal carcino-
genic products from the combustion of organic material.  The metabolites of
BaP (and all POM) can be divided into a water-soluble and an organic solvent-
soluble fraction.  Components of the latter fraction are primarily ring-
hydroxylated products, quinones, and labile epoxide intermediates.  For BaP
there are at least three dihydrodiols, three quinones, and two phenols
which can be detected as positional 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 hydrase enzyme.  A subsequent
oxidative attack by the aryl hydrocarbon hydroxylase may convert the diols
to diol epoxides, one of which (7,8-diol-9,10-epoxide) is the proposed
ultimate carcinogenic form of BaP.
                                    6-230

-------
     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.
     The general scheme of metabolism for unsubstituted ROM's 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 deriva-
tives of POM's may be preferred targets for conjugation and excretion,
whereas non K-region epoxides undergo further reductions and oxidative
attack to form toxicologically important molecules.  For POM's bearing
alkyl substituents (e.g., DMBA, MCA), the primary metabolites formed are
hydroxymethyl derivatives.  Nevertheless, epoxidation reactions at K-region
and non K-region aromatic double bonds occur which are catalyzed by aryl
hydrocarbon hydroxylase.  Removal of activated intermediates occurs by
conjugation with glutathione or glucuronic acid, or by further metabolism
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 metabolites capable of interaction with cellular constituents.
Others have indicated that a 6-oxy-BaP free radical or a hydroxymethyl
derivative of POM's may be biologically important reactive metabolites.
                                      6-231

-------
     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 DNA 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's
in the presence of rat liver 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 reactive POM metabolites also takes
place with RNA and other cellular proteins, and these processes cannot be
excluded as potentially important mediators of toxic response.  Various
epoxide derivatives of POM's 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 syn-
thetic polynucleotides (poly G) varied with the position of the epoxide
moiety; K-region and non K-region epoxides were capable of extensive co-
valent 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 diol epoxide
was also shown to be involved in DNA binding in cultured mouse embryo
cells.  These studies have helped to strengthen the concept that:
(1) 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.
                                    6-232

-------
6.3  TOXICOLOGY
     Not a great deal of attention has been paid to the non-carcinogenic
effects of exposure to ROM's.  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 ROM's has resulted
in selective destruction of hematopoietic and lymphoid elements, ovotoxicity
and anti-spermatogenic effects, adrenal necrosis, and changes in the intesti-
nal 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 malignancy predominates in evaluating POM toxicity.  For
ROM's which are not carcinogenic, very little seems to be known concerning
their involvement in toxic responses.
     Immunosuppression by exposure to ROM's 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 ROM's 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 lexicologically significant processes involved in the
response to POM absorption is the interaction with drug metabolizing enzyme
                                    6-233

-------
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 metabo-
lizing 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 activity 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 activation with POM-induced neoplasia in
animals indicates that the modulation of drug metabolizing enzymes in
humans plays a central role in carcinogenesis.
6.4  MUTAGENESIS
     The demonstration of mutagenic effect in bacterial and mammalian cells
by exposure to ROM's is generally equated with the capability to induce
tumor formation.  This assumption is based on the participation of a common
electrophilic metabolite in producing the carcinogenic/mutagem'c event, and
the common target site in the cell (i.e., DNA or other components of the
genome) for the effect to be produced.
     In recent years, considerable research effort has been directed at
determining 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
                                     6-234

-------
independence by a chemically-induced mutation, it was discoverd that epoxides
of carcinogenic ROM's displayed 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.
     Further work with cultured mammalian cells established that carcino-
genic ROM's 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 derivative is an  ultimate mutagenic agent.  Additional
investigations on the interaction  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 strengthened 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 ROM's with
the induction of chromosomal aberrations.  Although variations in chromosome
number and structure accompany POM-induced tumors in rodents, it is not
clear whether these changes are consistently observable.  No evidence in
the published literature has been  found to indicate that POM may produce
somatic mutations in the absence of neoplastic transformation.
                                     6-235

-------
     Mutations in germinal tissues induced by ROM's have been more easily
demonstrated 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.
6.5  CARCINOGENESIS
     Polycyclic aromatic hydrocarbons were the first compounds ever shown
to be associated with carcinogenesis.  To this day, carcinogenic ROM's are
still distinguished by several unique features:  (1) several  of the ROM's
are among the potent carcinogens known to exist, producing tumors by single
exposures to microgram quantities; (2) they act both at the site of applica-
tion and at organs distant to the site of absorption; and (3) their effects
have been demonstrated in nearly every tissue and species tested, regard-
less of the route of administration.  Among the more common ROM's at least
one, BaP, is ubiquitous in the environment and produces tumors in animals
which resemble human carcinomas.  The demonstration that organic extracts
of particulate air pollutants are carcinogenic to animals has raised concern
over the involvement of ROM's in human cancer formation.
     Oral administration of ROM's to rodents can result in tumors of the
forestomach, mammary gland, ovary, lung, liver, and lymphoid and hemato-
poietic tissues.  Exposure to ROM's by inhalation or intratracheal instilla-
tion 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 administration, BaP is less effective than other
ROM's (e.g., DMBA, MCA, DBcgC) in producing carcinomas.  On the other hand,
                                     6-236

-------
BaP has a remarkable potency for the induction of skin tumors in mice that
cannot be matched by any other environmental POM.  Thus, caution must be
exercised in considering the carcinogenicity of POM's as a class, or in
using BaP as a representative example in evaluating carcinogenic risk of
POM's.
     The induction of skin cancer by POM's 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's 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 non-carcinogenic 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's, several species and
modes of carcinogen administration are employed.  In studies conducted with
the Syrian golden hamster, intratracheal instillation of POM's established
that dose-related increases in tumor yield are clearly evident,  and the co-
administration of carrier particles such as Fe203 can markedly affect tumor
incidence.
     Several in vitro procedures hold promise as useful screening tools for
the detection of environmental carcinogens, including POM.  Morphological
transformation of cultured mammalian cells has proven to be a reliable
                                    6-237

-------
indicator of carcinogenicity by a chemical  in 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  transforma-
tion assays are capable of accurately distinguishing nearly all  known
carcinogens from non-carcinogens.
     An analysis of dose-response relationships for  ROM-induced tumors in
animals raises several important points having relevance to environmental
risk assessment.  A fundamental relationship which must be considered is
the deviation from linearity in dose-response curves, especially at low
doses, and whether or not 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 would  not indicate the propor-
tion of individuals still at risk of tumorigenesis at extrapolated low dose
                                    6-238

-------
levels.  Furthermore, the existence of a threshold for POM-induced careino-
genesis has not been documented in either animals or man, and, in fact, it
is more likely that in diverse populations the effect of ROM's 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  consis-
tent with a linear non-threshold pattern.  Overt tumor induction, on the
other hand, follows a dose response relationship consistent with a multi-
hit promotion process.  It is conceivable that in human populations, the
multi-hit component of carcinogenesis may be supplied by environmental
stimuli not necessarily linked or related to POM exposure.
     The we!1-documented existence of cocarcinogenic and anticarcinogenic
agents dictates that only a multifactorial analysis can provide a true
assessment of carcinogenic risk to humans for a particular POM.  Although
non-carcinogenic POM's were reported to antagonize the effects of carcino-
genic POM's in animals, others have shown that they have little influence
on tumor incidence or yield.  On the other hand, several non-carcinogenic
POM's found in cigarette smoke (pyrene, fluoranthene, BeP)  have potent
cocarcinogenic activity.  A significant decrease in carcinogenicity  can be
achieved with antioxidant food additives, certain vitamins, and other
naturally occuring 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.
                                    6-239

-------
6.6  REPRODUCTION AND TERATOLOGY
     Little information is presently available to indicate whether ROM's
present a significant hazard to reproductive success.   Furthermore,  effects
on the fetus which may be due to maternal  toxicity or  experimental conditions
(e.g., injection vehicle, stress) have not been adequately dissociated from
true embryotoxicity 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.
6.7  HUMAN STUDIES
     The presence of ROM's 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 ROM's in the absence of other potential carcinogens,
cocarcinogens, tumor initiators, or tumor  promoters.
     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 ROM-containing
substances is far greater than would occur in most communities, our under-
standing 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
                                    6-240

-------
cigarette smoking may have an overriding impact on the evaluation of the
carcinogenic threat of ROM's in occupational  or environmental  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
investigators 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 pollution exposure in  their  former residence  is still
expressed as a higher rate of lung 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, it was shown that lung cancer can be correlated with the
extent of solid fuel combustion, and particularly BaP concentration in air.
On the other hand, it has been concluded that POM in community air does
little to contribute to lung cancer incidence except insofar as  it may
interact, synergistically or additively, with the effects of cigarette
smoking.
                                    6-241

-------
     A fundamental problem in relating exposure to ROM's with the incidence
of cancer may involve the use of BaP levels as an indicator of total  exposure.
Demonstrated variation in the sources and degree of BaP emission to the
atmosphere with respect to the levels of other POM's precludes the accurate
estimation of body burdens.  In addition, animals exposed to basic and
aromatic neutral subfractions of organic air pollutants commonly develop
tumors; indicating that several highly active POM-type carcinogens are
present in polluted air.  Moreover, the carcinogenic and mutagenic activity
of organic extracts of air pollutants cannot be accounted for solely on the
basis of the amount of BaP contained in them.  Further studies which indi-
cate that BaP is a less effective carcinogen for the respiratory tissues
than other POM's (e.g., DBcgC) raise doubts concerning the validity of
risk-assessments derived from data concerning BaP.
6.8  ECOLOGICAL EFFECTS
     Although POM's are found nearly everywhere in man's environment, it is
not clear whether these agents may affect the ecological balance.  Adverse
effects on plants, microorganisms, fish, or other wildlife cannot be clearly
shown.  Moreover, there are no data to indicate that POM's may bioaccumulate
in any species, or that transfer of POM's through the food chain may occur.
Indeed, animal studies showing that POM's are rapidly metabolized and
excreted support the contention that biomagnification is an unlikely possi-
bility.
                                    6-242

-------
                                 References
 1.   Rees, E.G., P.  Mandelstam,  J.Q.  Lowry,  and  H.  Lipscomb.   A  study  of  the
     mechanism of intestinal  absorption  of benzo[a]pyrene.   Biochim. Biophys.
     Acta., 225.(1):96-107,  1971.

 2.   Kotin, P., H.L. Falk,  and R.  Busser.  Distribution,  retention, and
     elimination of  C^-s^-benzpyrene after administration  to mice and rats.
     J.  Natl.  Cancer Inst., 23_:541-555,  1959.

 3.   Vainio, H., P.  Uotila, J. Hartiala, and 0.  Pelkonen.  The fate of intra-
     tracheally installed benzo[a]pyrene in  the  isolated  perfused  rat  lung  of
     both control  and 20-methylcholanthrene  pretreated  rats.   Res. Commun.
     Chem. Path. Pharmacol.,  1_3(2): 259-271,  1976.

 4.   Creasia,  D.A.,  O.K.  Poggenburg,  Jr.,  and P.  Nettesheim.   Elution  of
     benzo[a]pyrene  from  carbon  particles  in the respiratory  tract of  mice.
     J.  Toxicol. Environ. Health,  1_( 6): 967-975,  1976.

 5.   Henry, M.C. and D.G. Kaufman.   Clearance of benzo[a]pyrene  from hamster
     lungs after administration  on  coated  particles.  J.  Natl. Cancer  Inst.
     51_(6):1961-1964, 1973.

 6.   Bock, F.G. and  T.L.  Dao.  Factors affecting the  polynuclear hydrocarbon
     level in  rat mammary glands.   Cancer Res.,  21_: 1024-1029,  1961.

 7.   Dao, T.L., F.G. Bock,  and S.  Crouch.  Level  of 3-methylcholanthrene  in
     mammary glands  of rats after  intragastric instillation  of carcinogen.
     Proc. Soc. Exptl.  Biol.  Med.,  102^635-638,  1959.

 8.   West, C.E. and  B.J.  Horton.   Transfer of polycyclic  hydrocarbons  from
     diet to milk in rats,  rabbits  and sheep.  Life Sci.,  19(10):1543-1552,
     1976.

 9.   Flesher,  J.W.   Distribution of radioactivity in  the  tissues of rats  after
     oral administration  of 7,12-dimethylbenz[a]anthracene-3H.   Biochem.
     Pharmacol., 16.: 1821-1831, 1967.

10.   Nagel, D.L., F. Stenback, D.B.  Clayson, and L. Wai leave.  Intratracheal
     instillation studies with 7H-dibenzo[c,g]carbazole in the Syrian  hamster.
     J.  Natl.  Cancer Inst., 57^(1 ):119-123, 1976.

11.   Shendrikova, I.A.  and V.A.  Aleksandrov.  Comparative characteristics of
     penetration of  polycyclic hydrocarbons  through the placenta into  the
     fetus in  rats.   Byull. Eksperiment. Biol. i  Medit.,  77(2):169-171, 1974.
                                    6-243

-------
12.  Peacock, P.R.  Evidence regarding the mechanism of elimination of 1,2-
     benzpyrene, 1,2,5,6-dibenzanthracene, and anthracene from the blood-
     stream of injected animals.   Brit.  J. Exptl.  Path., 1_7:164-172, 1936.

13.  Chalmers, J.G. and A.H.M.  Kirby.   The elimination of 3,4-benzpyrene from
     the animal body after subcutaneous injection.   I.  Unchanged benzpyrene.
     Biochem. J., 34:1191-1195, 1940.

14.  Heidelberger, C.  and S.M.  Weiss.   The distribution of radioactivity in
     mice following administration of 3,4-benzpyrene-5-Cl4 and 1,2,5,6-
     dibenzanthracene-9,10-C14.  Cancer Res., JJ_:885-898, 1951.

15.  Aitio, A.  Different elimination and effect on mixed function oxidase  of
     20-methylcholanthrene after intragastric and intraperitoneal administra-
     tion.  Res. Commun. Chem.  Path.  Pharmacol., 9_(4):701-710, 1974.

16.  Sellakumar, A., F. Stenback, and J. Rowland.   Effects of different dusts
     on respiratory carcinogenesis in hamsters induced by benzo[a]pyrene and
     diethylnitrosamine.  Europ.  J. Cancer, 1_2_:313-319, 1976.

17.  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. Abt.
     Orig. B, 163_:441-447, 1976.

18.  Kennedy, A.R. and J.B. Little.  Localization of polycyclic hydrocarbon
     carcinogens in the lung following intratracheal instillation in gelatin
     solution.  Cancer Res., J35_: 1563-1657, 1975.

19.  Pylev, L.N., F.J.C. Roe, and G.P. Warwick.   Elimination of radioactivity
     after intratracheal instillation of tritiated 3,4-benzopyrene in hamsters.
     Brit. J. Cancer,  23_:103-115, 1969.

20.  Schlede, E., R. Kuntzman,  and A.H.  Conney.   Stimulatory effect of benzo-
     [a]pyrene and phenobarbital  pretreatment on the biliary excretion of
     benzo[a]pyrene metabolites in the rat.  Cancer Res., 30_: 2898-2904, 1970.

21.  Schlede, E., R. Kuntzman,  S. Haber, and A.H.  Conney.  Effect of enzyme
     induction on the metabolism and tissue distribution of benzo[a]pyrene.
     Cancer Res., 30:2893-2897, 1970.

22.  Welch, R.M., B. Gommi, A.P.  Alvares, and A.H.  Conney.  Effect of enzyme
     induction on the metabolism of benzo[a]pyrene and 3'-methyl-4-monomethyl-
     aminoazobenzene in the pregnant and fetal rat.  Cancer Res., 32:973-978,
     1972.
                                     6-244

-------
23.  Scala, R.A.  Toxicology of participate polycyclic organic material.
     J. Occup.  Med., V7_( 12):784-788, 1975.

24.  Pullman, A. and B.  Pullman.   Electronic structure and carcinogenic
     activity of aromatic molecules - new developments.   Adv.  Cancer Res.,
     1:117-169, 1955.

25.  Yamagiwa,  K.   Ichikawa, Mitt.  Med.  Fak. Tokio,  15:295, 1915.

26.  Cook, J.W. and L.  Hewett.   J.  Chem.  Soc.,  398.  1933.

27.  Heiger, I.  J. Chem. Soc., 395, 1933.

28.  Kennaway,  E.L. and I. Heiger.   Carcinogenic substances and their fluor-
     escence spectra.   Brit. Med.  J., 1:1044, 1930.

29.  Public Health Service survey  of compounds  which have been tested for
     carcinogenic activity.  Publ.  No. 149.

30.  Conney, A.H.   Pharmacological  implications of microsomal  enzyme induction.
     Pharmacol. Rev.,  19:317, 1967.

31.  Gelboin, H.V.  A microsome-dependent binding of benzo[a]pyrene to DMA.
     Cancer Res., 29_: 1272-1276, 1969.

32.  Oesch, F.   Mammalian epoxide  hydrases:   inducible enzymes catalyzing the
     inactivation of carcinogenic  and cytotoxic metabolites derived from
     aromatic and olefinic compounds.  Xenobiotica., 3_:305, 1973.

33.  Oesch, F.  et al.,  In:  Biological Reactive Intermediates, Plenum Press,
     New York,  p.  181,  1977.

34.  Selkirk, O.K., R.G. Croy,  and  H.V.  Gelboin.  Benzo[a]pyrene metabolites:
     Efficient and rapid separation by high-pressure liquid chromatography.
     Science, 184_:109-111, 1974.

35.  Selkirk, J.K., R.G. Croy,  P.P. Roller,  and H.V. Gelboin.   High-pressure
     liquid chromatographic analysis of benzo[a]pyrene metabolism  and covalent
     binding and the mechanism of  action of 7,8-benzoflavone and 1,2-epoxy-
     3,3,3-trichloropropane.  Cancer Res.,  34_: 3474-3480,  1974.

36.  Lesko, S., W. Caspary, R.  Lorentzen, and P.O.P. Ts'o.  Enzymic formation
     of 6-oxabenzo[a]pyrene radical in rat liver homogenates from  carcinogenic
     benzo[a]pyrene.  Biochemistry, 21:3978-3984, 1975.
                                    6-245

-------
37.  Nagata, C., Y. Tagashira, and M. Kodama.  Metabolic activation of benzo-
     [a]pyrene:  Significance of the free radical.   In:   Chemical  Carcinogen-
     esis, Part A, ed. P.O.P. Ts'o and J. Dipaolo,  Marcel  Dekker,  New York.
     p. 87-111, 1974.

38.  Communication:  Binding of 6-oxygenzo[a]pyrene radical  with DNA and
     polynucleotides.  Gann, 68_:125-126, 1977.

39.  Communication:  Binding of benzo[a]pyrene-semiquinone radicals with DNA
     and polynucleotides.  Gann, 68_:253-254, 1977.

40.  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., J5^:324a, 1975.

41.  Blackburn, G.M., G.M. Milne, and L. Orgee.  Displacement of tritium from
     6-tritiobenzo[a]pyrene on covalent binding to  DNA.   J.  Chem.  Soc.,
     Chem. Comm., 10^:359-360, 1976.

42.  Cavalieri, E., R. Roth, E. Rogan, C. Grandjean, and J.  Althoff.  Mechanisms
     of tumor initiation by polycyclic aromatic hydrocarbons.  In:  Carcino-
     genesis, Vol. 3:  Polynuclear Aromatic Hydrocarbons.   P.M. Jones and
     R.I. Freudenthal (eds.).  New York, Raven  Press.  1978.  p. 273-284.

43.  Lehr, R.E., H. Yagi, D.R. Thakker, W. Levin, A.M.  Wood, A.M.  Conney, and
     D.M. Jerina.  The bay region theory of polycyclic  aromatic hydrocarbon-
     induced carcinogenicity.  In;  Carcinogenesis, Vol. 3:   Polynuclear
     Aromatic Hydrocarbons.  P.W. Jones and R.I.  Freudenthal (eds.).  New York,
     Raven Press.  1978.  p. 231-241.

44.  Wood, A.W., W. Levin, D. Ryan, P.E. Thomas,  H. Yagi,  H.D. Mah, D.R. Thakker,
     D.M. Jerina, and A.H. Conney.  High mutagenicity of metabolically activated
     chrysene 1,2 dihydrodiol:  Evidence for bay region activation of chrysene.
     Biochem. Biophys. Res. Commun., 78_(3):847-854, 1977.

45.  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.

46.  Boyland, E. and P. Sims.  The carcinogenic activities in mice of compounds
     related to benz[a]anthracene.  Intl. J. Cancer, 2^:500-504, 1967.

47.  Stoming, T.A., W. Bornstein, and E. Bresnick.   The metabolism of 3-methyl-
     cholanthrene by rat liver microsomes - A reinvestigation.  Biochem. and
     Biophys. Res. Commun., .79_(2):461 -469, 1977.

48.  Selkirk, J.K., E. Huberman, and C. Heidelberger.  An  epoxide  is an inter-
     mediate in the microsomal metabolism of the chemical  carcinogen, dibenz-
     [a,h]anthracene.  Biochem. and Biophys. Res. Commun., 43_(5):1010-1016,  1971.


                                     6-246

-------
 49. Sims, P. and P.L. Grover.  Epoxides in polycyclic aromatic hydrocarbon
     metabolism and carcinogenesis.   Adv.  Cancer Res., 20_:165-274,  1974.

50.  Selkirk, O.K., R.G.  Croy, and H.V.  Gelboin.  Isolation by high-pressure
     liquid chromatography and characterization of benzo[a]pyrene-4,5-
     epoxide as a metabolite of benzo[a]pyrene.  Arch. Biochem. and Biophys.,
     168:322-326, 1975.

51.  Nemoto, N. and H.V.  Gelboin.   Assay and properties of glutathione-S-
     benzo[a]pyrene-4,5-oxide transferase.   Arch.  Biochem. and Biophys.,
     122:739-742, 1975.

52.  Arias, I.M. and W.B.  Jakoby (eds.).  Glutathione:  Metabolism  and
     Function.  Raven Press, New York, 1976.

53.  Cohen, G.M., S.M. Haws, B.P.  Moore, and J.W.  Bridges.  Benzo[a]pyren-3-yl
     hydrogen sulphate, a major ethyl acetate-extractable metabolite of benzo-
     [a]pyrene in human,  hamster,  and rat lung  cultures.   Biochem.  Pharmacol.,
     25_: 2561-2570, 1976.

54.  Yang, S.K., P.P. Roller, P.P. Fu, R.G. Harvery,  and H.V.  Gelboin.
     Evidence for a 2,3-epoxide as an intermediate in the microsomal  metabo-
     lism of benzo[a]pyrene to 3-hydroxybenzo[a]pyrene.  Biochem. and Biophys.
     Res. Commun., 77.(4):1176-1182,  1977.

55.  Selkirk, O.K., R.G.  Croy, and H.V.  Gelboin.  High-pressure liquid  chroma-
     tographic separation of 10 Benzo[a]pyrene  phenols and the identification
     of l-phenol-7-phenol  as new metabolites.   Cancer Res., 36:922-926, 1976.

56.  Alvares, A.P., G. Schilling,  W.  Levin, and R. Kuntzman.   Studies on the
     induction of CO-binding pigments in liver  microsomes by phenobarbital  and
     3-methylcholanthrene.  Biochem.  Biophys.  Res. Commun., 2£(4):521-526,  1967.

57.  Wiebel, F.J., J.C. Leutz, and H.V.  Gelboin.  Aryl hydrocarbon  (benzo[a]-
     pyrene) hydroxylase:   Inducible in  extrahepatic  tissues of mouse strains
     not inducible in liver.  Arch.  Biochem. and Biophys., 1^:292-294, 1973.

58.  Nebert, D.W. and J.S. Felton.  Importance  of genetic factors influencing
     the metabolism of foreign compounds.   Fed. Proc. 35:1133-1141, 1976.

59.  Conney, A.H., E.J. Pantuck, K.-C.,  Hsiao,  R.  Kuntzman, A.P. Alvares,  and
     A. Kappas.  Regulation of drug  metabolism  in man by environmental  chemicals
     and diet.  Fed. Proc., 36_(5):1647-1652, 1977.

60.  Lu, A.Y.H. et al.  In:  Carcinogenesis, Vol.  1,  R.I. Freudenthal and
     P.W. Jones (eds.), Raven Press, New York.  p.  116, 1976.

61.  Lu, A.Y.H. and M.J.  Coon.  Role of  hemoprotein P-450 in fatty  acid u-hydrox-
     ylation in a soluble enzyme system  from liver microsomes, J. Biol. Chem.,
     243.: 1331-1332, 1968.

                                     6-247

-------
62.  Lu, A.Y.H., R. Kuntzman, S.  West, M.  Jacobson, and A.H.  Conney.   Recon-
     stituted liver microsomal enzyme system that hydroxylates drugs, other
     foreign compounds, and endogenous substrates.   II.  Role of the  cytochrome
     P-450 and P-448 fractions in drug and steroid  hydroxylations.  J.  Biol.
     Chem., 247.: 1727-1734, 1972.

63.  Allison, A.C. and L. Mallucci.   Uptake of hydrocarbon carcinogens  by
     lysosomes.  Nature, 203_: 1024-1027, 1964.

64.  Shires, T.K. and K.M. Richter.   The distribution of 3:4  benzo[a]pyrene
     in replicating mammalian cells  in vitro.  Exptal Cell Res., 44:617-620,
     1966.                                                       ~~~

65.  Boyland, E. and P. Sims.  Metabolism of polycyclic compounds.  The metabo-
     lism of 7,12-dimethylbenz[a]anthracene by rat-liver homogenates.  Biochem.
     J., 95^:780-787, 1965.

66.  Huberman, E., L. Sachs, S.K. Yang, and H.V.  Gelboin.   Identification of
     mutagenic metabolites of benzo[a]pyrene in mammalian cells.  Proc. Nat!.
     Acad. Sci. USA, 73:607-611,  1976.

67.  Jerina, D.M., R. Lehr, M. Schaefer-Ridder, H.  Yagi, J.M. Karle,
     O.R. Thakker, A.W. Wood, A.Y.H.  Lu, D. Ryan, S. West, W. Levin,  and
     A.H. Conney.  Bay region epoxides of dihydrodiols.  A concept which
     explains the rfluta
-------
72.  Wislocki, P.G., A.W. Wood, R.L.  Chang, W.  Levin,  H.  Yagi,  0.  Hernandez,
     P.M. Dansette, D.M.  Jerina, and  A.H.  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.

73.  Levin, W.,  A.W. Wood, H.  Yagi, P.M.  Dansette,  D.M.  Jerina, and A.H.  Conney.
     Carcinogenecity of benzo[a]pyrene 4,5-,7,8-,  and  9,10-oxides  on mouse
     skin.  Proc.  Nat!. Acad.  Sci.  USA, 73_(1 ):243-247, 1976.

74.  Yang, S.K., D.W. McCourt, P.P. Roller, and H.V. Gelboin.   Enzymatic
     conversion  of benzo[a]pyrene leading  predominantly to  the  diol-epoxide
     r-7, t-8-dehydroxy-t-9,10-oxy-7,8,9,10-tetrahydrobenzo[a]pyrene through
     a single enantiomer of r-7, t-8-dehydroxy-7,8-dihydrobenzo[a]pyrene.
     Proc. Natl. Acad.  Sci., WA., 73_: 2594-2598.

75.  Wood, A.W., R.L. Chang. W. Levin, H.  Yagi, D.R. Thakker,  D.M.  Jerina,
     and A.H. Conney.  Differences  in mutagenicity  of  the optical  enantiomers
     of the diastereomeric benzo[a]pyrene  7,8-diol-9,10-epoxides.   Biochem.
     Biophys. Res. Commun., 77.(4): 1389-1396,  1977.

76.  Levin, W.,  A.W. Wood, R.L. Chang, T.J. Slaga,  H.  Yagi, D.M. Jerina,  and
     A.H. Conney.   Marked differences in the tumor-initiating  activity of
     optically pure (+)- and (-)-trans-7,8-dihydroxy-7,8-dihydrobenzo[a]-
     pyrene on mouse skin.  Cancer Res., 37^2721-2725, 1977.

77.  Weinstein,  I.B., A.M. Jeffrey, K.W.  Jennette,  S.H.  Blobstein,  R.G. Harvey,
     C. Harris,  H. Autrup, H.  Kasai,  and K. Nakanishi.  Benzo[a]pyrene diol
     epoxides as intermediates in nucleic  acid binding in vitro and in vivo.
     Science, 193_( 4253): 592-595, 1976.

78.  Dipple. A.   Polynuclear aromatic carcinogens.   In:   Chemical  Carcinogens.
     C.E. Searle (ed.).  ACS Monograph 173, American Chemical  Society, Washington,
     D.C.  Chapter 5, 1976.

79.  Slaga, T.J.,  W.M.  Bracken, A.  Viaje,  W.  Levin. H. Yagi,  D.M.  Jerina, and
     A.H. Conney.   Comparison of the  tumor-initiating  activities of benzo[a]-
     pyrene arene oxides and diol-epoxides.  Cancer Res., 3_7_:4130-4133, 1977.

80.  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.   1976.  p.  211-224.

81.  Nemoto, N.  and H.V.  Gelboin.  Enzymatic conjugation of benzo[a]pyrene
     oxides, phenols and dihydrodiols with VDP-Glucuronic Acid, Biochem.
     Pharmacol., 25_: 1221-1226, 1976.
                                    6-249

-------
82.  Boyland, E. and P.  Sims.  The metabolism of benz[a]anthracene and dibenz-
     [a,h]anthracene and their 5,6-epoxy-5,6-dihydro derivatives by rat-liver
     homogenates.  Biochem. J., 97_:7-16, 1965.

83.  Sims, P.  Epoxy derivatives of aromatic polycyclic hydrocarbons.   The
     preparation of benz[a]anthracene 8,9-oxide and 10,ll-dihydrobenz[a]-
     anthracene 8,9-oxide and their metabolism by rat liver preparations.
     Biochem. J., 125_:159-168, 1971.

84.  Keysell, G.R., J. Booth, and P. Sims.   Glutathione conjugates as  metabo-
     lites of benz[a]anthracene.  Xenobiotica,  5^(7):439-448, 1975.

85.  Booth, J., G.R. Keysell, and P. Sims.   Formation of glutathione conjugates
     as metabolites of 7,12-dimethylbenz[a]anthracene by rat-liver homogenates.
     Biochem. Pharmacol., 22_: 1781-1791,  1973.

86.  Sims, P.  Epoxy derivatives of aromatic polycyclic hydrocarbons.   The
     preparation and metabolism of epoxides related to 7,12-dimethylbenz[a]-
     anthracene.  Biochem. J., 1_3J_:405-413, 1973.

87.  Moschel, R.C., W.M. Baird, and A.  Dipple.   Metabolic activation of the
     carcinogen 7,12-dimethylbenz[a]anthracene for DNA binding.   Biochem.
     Biophys. Res. Commun., 76_(4): 1092-1098, 1977.

88.  Sims, P.  The metabolism of 7- and 12-methylbenz[a]anthracene and their
     derivatives.  Biochem. J., N)5:591-598, 1967.

89.  Sims, P.  The metabolism of 3-methylcholanthrene and some related com-
     pounds by rat-liver homogenates.  Biochem. J., 98_:215-228,  1966.

90.  Thakker, D.R., W. Levin, T.A. Stoming, A.H. Conney, and D.M. Jerina.
     Metabolism of 3-methylcholanthrene by rat liver microsomes  and a  highly
     purified monooxygenase system with and without epoxide hydrase.  In:
     Carcinogenesis, Vol. 3:  Polynuclear Aromatic Hydrocarbons.  P.M. Jones
     and R.I. Freudenthal (eds.).  New York, Raven Press.  1978.  p. 253-264.

91.  Grover, P.L., A. Hewer, and P. Sims.   Metabolism of polycyclic hydro-
     carbons by rat-lung preparations.   Biochem. Pharmacol., 23_:323-332, 1974.

92.  Vadi, H., B. Jernstrom, and S. Orrenius.  Recent studies on benzo[a]-
     pyrene metabolism in rat liver and lung.  In:  Carcinogenesis, Vol. 1:
     Polynuclear Aromatic Hydrocarbons:   Chemistry, Metabolism,  and Carcino-
     genesis.  P.M. Jones and R.I. Freudenthal  (eds.).  New York, Raven Press,
     1976.  p. 45-61.

93.  Leber, P., G. Kerchner, and R.I. Freudenthal.  A comparison of benzo[a]-
     pyrene metabolism by primates, rats,  and miniature swine.  In: Carcino-
     genesis, Vol. 1:  Polynuclear Aromatic Hydrocarbons:  Chemistry,  Metabo-
     lism, and Carcinogenesis.  P.W. Jones and R.I. Freudenthal  (eds.).
     New York, Raven Press.  1976.  p.  35-43.


                                     6-250

-------
94.  Wang, I.Y., R.E. Rasmussen, N.L.  Petrakis,  and A-C.  Wang.   Enzyme  induc-
     tion and the difference in the metabolite patterns  of benzo[a]pyrene
     produced by various strains of mice.   In:  Carcinogenesis,  Vol.  1:
     Polynuclear Aromatic Hydrocarbons:   Chemistry, Metabolism,  and  Carcino-
     genesis.  P.M. Jones and R.I.  Freudenthal (eds.).   New York,  Raven  Press.
     1976.  p. 77-89.

95.  Pelkonen, 0.  Metabolism of benzo[a]pyrene  in  human  adult and fetal
     tissues.  In:   Carcinogenesis, Vol.  1:   Polynuclear  Aromatic  Hydro-
     carbons:  Chemistry, Metabolism,  and  Carcinogenesis.   P.W.  Jones and
     R.I. Freudenthal (eds.).  New  York,  Raven Press.  1976.  p. 9-21.

96.  Kimura, T., M. Kodama, and C.  Nagata.   Differences  in benzo[a]pyrene
     metabolism between lung and liver homogenates.  Biochem. Pharmacol.,
     26:671-674, 1977.

97.  Selkirk, J.K., S.K. Yang, and  H.V.  Gelboin.  Analysis of benzo[a]pyrene
     metabolism in  human liver and  lymphocytes and  kinetic analysis  of  benzo-
     [ajpyrene in rat liver microsomes.   In;   Carcinogenesis, Vol. 1:   Poly-
     nuclear Aromatic Hydrocarbons:  Chemistry,  Metabolism, and  Carcinogenesis.
     P.W. Jones and R.I. Freudenthal (eds.).   New York,  Raven Press.  1976.
     p. 153-169.

98.  Kouri, R.E.  Relationship between levels of aryl  hydrocarbon  hydroxylase
     activity and susceptibility to 3-methylcholanthrene  and benzo[a]pyrene-
     induced cancers in inbred strains of mice.   In:   Carcinogenesis, Vol.  1:
     Polynuclear Aromatic Hydrocarbons:   Chemistry, Metabolism,  and  Carcino-
     genesis.  P.W. Jones and R.I.  Freudenthal (eds.).   New York,  Raven  Press.
     1976.  p. 139-151.

99.  Miller, J.A.  Carcinogenesis by chemicals:   An overview, G.H.A.  Clowes
     Memorial Lecture,  Cancer Res., 3£:559-576,  1970.

100. Brookes, P. and P.O. Lawley.  Evidence for the binding of polynuclear
     aromatic hydrocarbons to nucleic acids of mouse skin:  Relation between
     carcinogenic power of hydrocarbons  and their binding to deoxyribonucleic
     acid.  Nature, 2102:781-784, 1964.

101. Marquardt, H., F.S. Philips, and A.  Bendich.  DNA binding and inhibition
     of DNA synthesis after 7,12-dimethylbenz[a]anthracene administered  during
     the early prereplicative phase in regenerating rat  liver.   Cancer  Res.,
     32^:1810-1913,  1972.

102. Kuroki, T. and C.  Heidelberger.  The  binding of polycyclic  aromatic hydro-
     carbons to the DNA, RNA, and proteins of transformable cells  in culture.
     Cancer Res., 3J_:2168-2176, 1971.

103. Corbett, T.H.  and P. Nettesheim.   Separation of multiple forms  of  protein-
     bound metabolites of the carcinogenic hydrocarbons  3-methylcholanthrene
     and 7,12-dimethylbenz[a]anthracene from several  rodent species.  Chem.
     Biol. Interactions, 8:285-296, 1974.

                                     6-251

-------
104. Slaga, T.J., G.T. Bowden, B.G. Shapas, and R.K.  Boutwell.   Macromolecular
     synthesis following a single application of polycyclic hydrocarbons used
     as initiators of mouse skin tumorigenesis.  Cancer Res.,  34_:771-774, 1974.

105. Bowden, G.T., B.G. Shapas, and R.K.  Boutwell.   The binding of 7,12-
     dimethylbenz[a]anthracene to replicating and non-replicating DMA in
     mouse skin.  Chem. Biol.  Interactions, 8_: 379-394,  1974.

106. Bowden, G.T. and R.K. Boutwell.   The replication of mouse  skin epidermal
     DNA to which is bound 7,12-dimethylbenz[a]anthracene.   Chem. Biol.  Inter-
     actions, 9_:15-23, 1974.

107. Nery, R.  Carcinogenic Mechanisms:   A critical  review  and  a suggestion
     that oncogenesis may be adaptive ontogenesis.   Chem. Biol. Interactions,
     11:145-169, 1976.

108. Lutz, W.K., A.  Viviani, and C. Schlatter.   Nonlinear dose-response
     relationship for the binding of the  carcinogen  benzo[a]pyrene to rat
     liver DNA in vivo.  Cancer Res., 38_: 575-578, 1978.

109. Grover, P.L. and P. Sims.  Enzyme-catalyzed reactions  of  polycyclic
     hydrocarbons with deoxyribonucleic  acid and proteins In Vitro.  Biochem.
     J., Vm: 159-160, 1968.

110. Kaufman, D.G.,  V.M. Genta, C.C.  Harris, J.M. Smith, M.B.  Sporn, and
     U. Saffiotti.  Binding of H-labeled  benzo[a]pyrene to  DNA  in hamster
     tracheal epithelial cells.  Cancer  Res., 33_:2837-2841, 1973.

111. Harris, C.C., D.G. Kaufman, M.B. Sporn, H. Boren,  F. Jackson, J.M.  Smith.
     J. Pauley, P. Dedick, and U. Saffiotti.  Localization  of  benzo[a]pyrene-JH
     and alterations in nuclear chromatin caused by  benzo[a]pyrene-ferric oxide
     in the hamster  respiratory epithelium.  Cancer  Res., 33_: 2842-2848,  1973.

112. Burki, K., T.A. Stoming,  and E.  Bresnick.   Effects on  an epoxide hydrase
     inhibitor on in vitro binding of polycyclic hydrocarbons  to DNA and on
     skin carcinogenesis.  J.  Nat!. Cancer Inst., 52_(3):785-788, 1974.

113. Yang, S.K., H.V. Gelboin, B.F. Trump, H. Autrup, and C.C.  Harris.   Metabo-
     lic activation  of benzo[a]pyrene and binding to  DNA in cultured human
     bronchus.  Cancer Res., 3^:1210-1215, 1977.

114. Menger, E.M., R.B. Spokane, and P.O. Sullivan.   Free radicals derived
     from benzo[a]pyrene.  Biochem. Biophys. Res. Commun.,  71_(2):610-616, 1976.

115. Grover, P.L., A. Hewer, and P. Sims.  Epoxides  as  microsomal metabolites
     of polycyclic hydrocarbons.  FEBS Letters, 18^:76-80, 1971.

116. Jerina, D.M. and J.W. Daly.  Arene  oxides:  A new  aspect of drug metabo-
     lism.  Science, 185_(4151): 573-582,  1974.


                                     6-252

-------
117. Grilli, S., P. Rocchi, and G. Prodi.  Non-enzymatic and microsome-depen-
     dent binding of polycyclic hydrocarbons to DMA and polynucleotides.
     Chem. Biol. Interactions, l]_:351-363, 1975.

118. Meehan, T.D.  Enzyme induced covalent binding of benzo[a]pyrene to poly-
     guanylic-acid.  Fed. Proc. (FEPRA), 34_(3):489, 1975.

119. Weisgras, J.M., S.H. Blobstein, I.B. Weinstein, D. Grunberger,  and
     R.G. Harvey.  Binding of dimethyl  benzathracene-5, 6-epoxide (DMBA-
     EPOXIDE) to nucleic acids in vitro.  Proc. Amer.  Assoc. Cancer  Res.,
     15:111, 1974.

120. Blobstein, S.H., I.B. Weinstein, P. Dansette, H.  Yagi,  and D.M. Jerina.
     Binding of K-  and non-K-region arene oxides  and phenols of polycyclic
     hydrocarbons to polyguanylic acid.   Cancer Res.,  36_: 1293-1298,  1976.

121. Moore, P.O., M. Koreeda, P.G. Wislocki, W. Levin, A.M.  Conney,  H.  Yagi,
     and D.M. Jerina.  In vitro reactions of the  diastereomeric 9,10-epoxides
     of (+) and (-)-trans-7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene with  poly-
     guanylic acid  and evidence for formation of  an enantiomer of each
     diastereomeric 9,10-epoxide from benzo[a]pyrene in mouse skin.   In:
     Drug Metabolism Concepts, American  Chemical  Society,  Symposium  Series
     No. 44, American Chemical Society,  Washington, D.C.,  D.M. Jerina (ed.),
     1977.

122. Moore, P.O. and M.  Koreeda.  Biochem. Biophys. Res.  Commun., 73_:459, 1976.

123. Sims, P., P.L. Grover, A. Swaisland, K. Pal, and  A.  Hewer.   Metabolic
     activation of  benzo[a]pyrene proceeds by a diol-epoxide.  Nature,  252:
     326-328, 1974.

124. Ivanovic, V.,  N.E.  Geacintov, and  I.B. Weinstein.  Cellular binding  of
     benzo|_a]pyrene to DNA characterized by low temperature  fluorescence.
     Biochem. Biophys. Res.  Commun., 70_(4):1172-1179,  1976.

125. King et al., 1976.

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

127. Wood, A.W., R.L. Goode, R.L. Chang, W. Levin, A.H. Conney, H. Yagi,
     P.M. Dansette, and D.M. Jerina. Mutagenic and cytotoxic activity  of
     benzo[a]pyrene 4,5-, 7,8-, and 9,10-oxides and the six  corresponding
     phenols.  Proc. Natl. Acad. Sci. USA, 72^(8):3176-3180,  1975.

128. Newbold and Brookes, 1976.
                                    6-253

-------
129. Sarrif, A.M., J.S. Bertram, M.  Kamarck,  and C.  Heidelberger.   The isola-
     tion and characterization of polycyclic hydrocarbon binding proteins  from
     mouse liver and skin cytosols.   Cancer Res., 35_:816-824,  1975.

130. Moses, H.L., R.A.  Webster, G.D.  Martin,  and T.C.  Spelsberg.   Binding of
     polycyclic aromatic hydrocarbons to transcriptionally active  nuclear
     subfractions of AKR mouse embryo cells.   Cancer Res., 36_: 2905-2910,  1976.

131. Jahn, C.L. and G.W. Litman.  Distribution of covalently bound benzo[a]-
     pyrene in chromatin.  Biochem.  Biophys.  Res. Commun., 76_(2): 534-540, 1977.

132. Haddow, A., C.M. Scott, and J.D. Scott.   The influence of certain carcin-
     ogenic and other hydrocarbons on body growth in the rat.   Proc.  Royal  Soc.
     B., 121:477, 1937.

133. Philips, F.S., S.S. Sternberg,  and H. Marquardt.   In vivo cytotoxicity of
     polycyclic hydrocarbons.  In:  Pharmacology and the Future of Man.  Proc.
     5th Intl. Congr. Pharmacology,  San Francisco, 1972, Vol.  2, pp.  75-88, 1973.

134. Yasuhira, K.  Damage to the thymus and other lymphoid tissues from 3-methyl-
     cholanthrene, and subsequent thymoma production,  in mice.   Cancer Res.,
     24_: 558-569, 1964.

135. Cawein, M.J. and K.L.  Sydnor.  Suppression of cellular activity  in the
     reticuloendothelial system of the rat by 7,12-dimethylbenz[a]anthracene.
     Cancer Res., 28:320-327, 1968.

136. Ball, J.K., S. Hoshino, and J.A. McCarter.  Depressive effect of 7,12-
     dimethylbenz[a]anthracene and ionizing irradiation on bone marrow colony-
     forming cells.  J. Natl. Cancer Inst., 51_(5): 1491 -1495, 1973.

137. Ford, E. and C. Muggins.  Selective destruction in testis induced by
     7,12-dimethylbenz[a]anthracene.   J. Exp. Med.,  1J8_:27-40, 1963.

138. Bock, F.G.  Early effects of hydrocarbons on mammalian skin.   Progr. Exp.
     Tumor Res., i= 126-168, 1964.

139. Hueper, W.C.  Chemically induced skin cancers in  man.  Nat. Cancer Inst.,
     Monogr. Kh 377-391, 1963.

140. Kraup, T.  Oocyte survival in the mouse ovary after treatment with 9,10-
     dimethyl-l,2-benz[a]anthracene.   J. Endocrinol.,  46_: 483-495,  1970.

141. Mattison, D.R. and S.S. Thorgeirsson.  Ovarian  metabolism of  polycyclic
     aromatic hydrocarbons and associated ovotoxicity  in the mouse.   Gynecol.
     Invest., 8(1/2):11, 1977.

142. Boyland, E., P. Sims, and C. Muggins.  Induction  of adrenal damage and
     cancer with metabolites of 7,12-dimethylbenz[a]anthracene. Nature,  207:
     816-817, 1965.

                                     6-254

-------
143. Hoch-Ligeti, C.   Studies on the changes  in the lymphoid  tissues  of mice
     treated with carcinogenic and noncarcinogenic  hydrocarbons.   Cancer Res.,
     1:484-488, 1941.

144. Lasnitzki, A.  and D.L.  Woodhouse.   The effect  of 1:2:5:6-dibenzanthracene
     on the lymph-nodes of the rat.   J.  Anat.,  78_:121-128,  1944.

145. Gross, P., E.  Tolker, M.A.  Babyak,  and M.  Kaschak.   Experimental  lung
     cancer in hamsters.  Arch.  Environ.  Health,  1_1_:59-65,  1965.

146. Reznik-Schuller, H. and U.  Mohr.   Investigations on  the  carcinogenic burden
     by air pollution in man.  IX.  Early pathological  alterations  of the
     bronchial epithelium in Syrian  golden hamsters after intratracheal  instill-
     ation of benzo[a]pyrene.  Zbl.  Bakt.  Hyg., I.  Abt. Orig.  B,  159:493-502,
     1974.

147. Henry, M.C., C.D. Port, R.R.  Bates,  and  D.G. Kaufman.  Respiratory  tract
     tumors in hamsters induced  by benzo[a]pyrene.   Cancer  Res.,  33:1585-1592,
     1973.

148. Saffiotti, U., F. Cefis, and  L.H.  Kolb.  A method  for  the experimental
     induction of bronchogenic carcinoma.   Cancer Res., 28_: 104-124, 1968.

149. Stenback, F. and A. Sellakumar.   Squamous  metaplasia and respiratory
     tumors induced by intratracheal  instillations  of 7,12-dimethylbenz[a]-
     anthracene in  Syrian golden hamsters. Eur. J. Cancer, 10:483-486,  1974.

150. Stenback, F. and A. Sellakumar.   Lung tumor  induction  by dibenzo[a,i]-
     pyrene in the  Syrian golden hamster.   Z. Krebsforsch., 82_: 175-182,  1974.

151. Malmgren, R.A.,  B.B. Bennison,  and  T.W.  McKinley.  Reduced antibody
     titres in mice treated  with carcinogenic and cancer  chemotherapeutic
     agents.  Proc. Soc. Exp. Biol.  Med.,  _79_:484, 1952.

152. Stjernsward, J.   Immunosuppression  by carcinogens.   Antibiotical  Chemother.,
     15_:213, 1969.

153. Stjernsward, J.   The effect of  non-carcinogenic  and  carcinogenic hydro-
     carbons on antibody-forming cells measured at  the  cellular level  in vitro.
     J. Natl. Cancer Inst.,  36_:1189,  1966.

154. Baldwin, R.W.   Immunological  aspects  of  chemical  carcinogenesis.   In:
     Advances in Cancer Research,  Vol.  18, G. Klein and S.  Weinhouse  (eds.).
     Academic Press,  New York, U.S.A.;  London,  England, 1973.   p.  1-75.

155. Rowland, G.F.  and C.M.  Hurd.  Target lymphoid  cell population  of carcino-
     gen induced immunodepression  in mice. Nature, 227:167-168,  1970.
                                     6-255

-------
156. Yamashita, U., Y. Matsuoka, and M.  Kltagawa.   Suppressed activity of
     t-cell in mice treated with 7,12-dimethylbenz[a]anthracene.   Gann,
     64_:317-320, 1973.

157. Dale, M.M. and J.R. Hedges.  The effect of carcinogens on immune responses
     in the guinea pig.  Brit.  J.  Exp. Path., 54_:437-448,  1973.

158. Stutman, 0.  Immunological aspects  of resistance to the oncogenic effect
     of 3-methylcholanthrene in mice.  Isr.  J.  Med.  Sci.,  3_:217-228,  1973.

159. Gelboin, H.V., N. Kinoshita,  and F.J. Wiebel.   Microsomal hydroxylases:
     studies on the mechanism of induction and  their role  in polycyclic hydro-
     carbon action.  In:  Collection of  Papers  Presented at the Annual  Sym-
     posium on Fundamental Cancer Research,  Series  24.   1972.   p.  214-240.

160. Bast, R.C., Jr., T. Okuda, E. Plotkin,  R.  Tarone,  H.J. Rapp,  and
     H.V. Gelboin.  Development of an assay  for aryl  hydrocarbon  [benzo(a)-
     pyrene] hydroxylase in human  peripheral blood  monocytes.   Cancer Res.,
     36_: 1967-1974, 1976.

161. Chuang, A.H.L., E.F. Howard,  and E. Bresnick.   Aryl hydrocarbon  hydrox-
     ylase in mouse mammary gland:  in vitro study  using mammary  cell lines.
     Chem.-Biol. Interactions,  ]7_:9-16,  1977.

162. Andrews, L.S., B.R. Sonawane, and S.J.  Yaffe.   Characterization  and induc-
     tion of aryl hydrocarbon [benzo(a)pyrene]  hydroxylase in rabbit  bone
     marrow.  Res. Commun. Chem. Pathol. Pharmacol., 1_5(2):319-339, 1976.

163. Cohn, J.A., A.P. Alvares,  and A. Kappas.  On  the occurrence  of cytochrome
     P-450 and aryl hydrocarbon hydroxylase activity in rat brain.  J.  Exp.
     Med., 145(6):1607-1611, 1977.

164. Wiebel, F.J., J.C. Leutz,  and H.V.  Gelboin.  Aryl  hydrocarbon (benzo[a]-
     pyrene) hydroxylase:  A mixed-function  oxygenase in mouse skin.   J. Invest.
     Dermatol., 64_(3): 184-189,  1975.

165. Grundin, R., S. Jakobsson, and D.L. Cinti.  Induction of microsomal aryl
     hydrocarbon (3,4-benzo[a]pyrene) hydroxylase and cytochrome  P-450|< in  rat
     cortex.  I.  Characteristics  of the hydroxylase system.  Arch. Biochem.
     Biophys., 1J8_(2): 544-555,  1973.

166. Zampaglione, N.G. and G.J. Mannering.  Properties  of  benzpyrene  hydrox-
     ylase in the liver, intestinal mucosa and  adrenal  of  untreated and 3-
     methylcholanthrene treated rats. J.  Pharmacol.  Exp.  Ther.,  185(3):676-
     685, 1973.

167. Nebert, D.W. and H.V. Gelboin.  Substrate-inducible microsomal aryl
     hydroxylase in mammalian cell culture.   J. Biol. Chem., 243:6250-6261,
     1973.


                                     6-256

-------
168. Kouri, R.E., H.  Ratrie,  and C.E.  Whltmire.   Evidence  of a  genetic  rela-
     tionship between susceptibility to 3-methylcholanthrene-induced  subcutan-
     eous tumors and  inducibility of aryl  hydrocarbon  hydroxylase.  J.  Natl.
     Cancer Inst., 51(1):197-200, 1973.

169. Kellerman, G., C.R.  Shaw,  and M.  Luyten-Kellerman.  Aryl hydroxylase
     inducibility and bronchogenic carcinoma.   New  England J. Med., 289:934,
     1973.

170. Atlas, S.A., E.S.  Vesell,  and D.W.  Nebert.   Genetic control of inter-
     individual variations in the inducibility  of aryl  hydrocarbon hydroxylase
     in cultured human lymphocytes.   Cancer Res., 36_:4619-4630,  1976.

171. Gelboin, H.V. Cancer susceptibility  and carcinogen metabolism.  N. Engl.
     J. Med., 297_(7):384-385, 1977.

172. Paigen, B., J. Minowada, H.L. Gurtoo,  K. Paigen,  N.B.  Parker, E. Ward,
     N.T. Hayner, I.D.J.  Bross, F. Bock, and R.  Vincent.   Distribution  of aryl
     hydrocarbon hydroxylase  inducibility  in cultured  human lymphocytes.  Cancer
     Res.,  37_: 1829-1837,  1977.

173. Conney, A.H., J. Kapitulnik, W.  Levin, P.  Dansette, and D.  Jerina.  Use
     of drugs in the  evaluation of carcinogen metabolism in man.   In:   Screen-
     ing Tests  in Chemical Carcinogenesis,  R. Montesano and L.  Tomatis  (eds.).
     Lyon,  France; IARC.   IARC  Publ.  No. 12.  1976.  p. 319-336.

174. Kapitulnik, J.,  W.  Levin,  P.J.  Poppers, J.E. Tomaszewski,  D.M. Jerina, and
     A.M. Conney.  Comparison of the hydroxylation  of  zoxazolamine and  benzo[a]-
     pyrene in  human  placenta:   effect of  cigarette smoking.  Clin. Pharmacol.
     and Therap., 20(5):557-564, 1976.

175. Kapitulnik, J.,  P.J.  Poppers, and A.M. Conney.  Comparative metabolism of
     benzo[a]pyrene and drugs in human liver.   Clin. Pharmacol.  and Therap.,
     21(2):166-176, 1977.

176. Kapitulnik, J.,  P.J.  Poppers, M.K.  Buening,  J.G.  Fortner,  and A.M.  Conney.
     Activation of monnoxygenases in human  liver by 7,8-benzoflavone.   Clin.
     Pharmacol. and Therap.,  £2(4):475-484, 1977.

177. Kapitulnik, J.,  W.  Levin,  A.Y.H.  Lue,  R. Morecki,  P.M.  Dansette, D.M. Jerina,
     and A.M. Conney.  Hydration of arene  and alkene oxides by  epoxide  hydrase
     in human liver microsomes.  Clin.  Pharmacol. and  Therap.,  21(2):158-165,
     1977.                                                      ~

178. Thompson,  S. and T.J. Slaga.  Mouse epidermal  aryl hydrocarbon hydroxylase.
     J. Invest. Dermatol., 66_:108-111, 1976.

179. Yuspa, S.H., H.  Hennings,  and U.  Saffioti.   Cutaneous carcinogenesis:
     past,  present, and future.  J.  Invest. Dermatol.,  67_:199-208, 1976.

                                     6-257

-------
180. Shulte-Herman, R.   Stimulation of liver growth  and  mixed-function  oxidase
     by alpha-hexachlorocyclohexane:   separation  of  inductive  pathways.  _In_:
     Microsomes and Drug Oxidations,"  V.  Ulrich  (ed.).   New  York,  Pergamon
     Press, 1977, in press.

181. Kouri, R.E., T. Rude, P.E.  Thomas, and C.E.  Whitmire.   Studies  on  pulmonary
     aryl  hydrocarbon hydroxylase activity in inbred strains of  mice.   Chem.-
     Biol. Interactions, 13:317-331,  1976.

182. Owens, I.S.  Genetic regulation  of UDP-glucuronosyltransferase  induction
     by polycyclic aromatic  compounds  in mice.  J. Biol.  Chem.,  252(9):2827-
     2833, 1977.

183. Aitio, A.  Effect of chrysene and carbon tetrachloride  administration  on
     rat hepatic microsomal  monooxygenase and upglucuronsyltransferase  activ-
     ity.   FEBS Lett.,  4i(l):46-49, 1974.

184. Assessment of the carcinogenicity and mutagenicity  of chemicals.   Geneva,
     World Health Organization,  Technical Report  Series,  No. 546,  1974.  19 p.

185. Miller, E.G. and J.A. Miller.  Biochemical mechanisms of  chemical  carcino-
     genesis.   In:  Molecular Biology of Cancer,  H.  Busch (ed.).   New York,
     Academic Press.  1974.   p.  377-402.

186. McCann, J., E. Choi, E. Yamasaki, and B.N. Ames.  Detection of  carcinogens
     as mutagens in the salmonella/microsome test:   assay of 300 chemicals.
     Proc. Natl. Acad.  Sci.  USA, 72_(12):5135-5139, 1975.

187. Teranishi, K., K.  Hamada, and H.  Watanabe.   Quantitative  relationship
     between carcinogenicity and mutagenicity of  polyaromatic  hydrocarbons
     in salmonella typhimurium mutants.  Mutation Res.,  31_:97-102, 1975.

188. McCann, J. and B.N. Ames.  Detection of carcinogens as  mutagens in the
     salmonella/microsome test:   assay of 300 chemicals:  discussion.   Proc.
     Natl. Acad. Sci. USA, 73^(3):950-954, 1976.

189. Sugimura, T., S. Sata,  M. Nagao,  T.  Yahagi,  T.  Matsushima,  Y. Seino,
     M. Takeuchi, and T. Kawachi.  Overlapping of carcinogens  and  mutagens.
     In:  Fundamentals in Cancer Prevention.  P.N. Magee (ed.),  Univ. of
     Tokyo Press, Tokyo/Univ. Park Press, Baltimore, pp.  191-215.  1976.

190. Poirier, L.A.  Personal Communication.  Nutrition and Metabolism Section,
     Carcinogen Metabolism and Toxicology Branch, National Cancer  Institute,
     National Institutes of Health, Public Health Service, U.S.  Dept. of Health,
     Education and Welfare,  Bethesda,  Md. 20014.

191. Ames, B.N., P. Sims, and P.L. Grover.  Epoxides of  carcinogenic polycyclic
     hydrocarbons are frameshift mutagens.  Science, 176:47-49.  1972.


                                     6-258

-------
192. Malaveille, C., H.  Bartsch,  P.L.  Grover,  and  P.  Sims.  Mutagenicity  of
     non-K-region diols  and diol-epoxides  of benz[a]anthracene  and  benzo[a]-
     pyrene in S. typhimurium TA  100.   Biochem.  Biophys.  Res. Conmun.,  66(2):
     693-700, 1975.

193. Wood, A.M., W.  Levin, A.Y.H.  Lu,  H. Yagi, 0.  Hernandez,  D.M. Jerina,  and
     A.M.  Conney.  Metabolism of  benzo[a]pyrene  and  benzo[a]pyrene  derivatives
     to mutagenic products by highly purified  hepatic microsomal enzymes.
     J. Biol. Chem., 2Ei]_(16):4882-4890,  1976.

194. Levin, W., A.M. Wood, A.Y.H.  Lu,  D. Ryan, S.  West, and A.H. Conney.   Role
     of purified cytochrome P-448 and  epoxide  hydrase in  the  activation and
     detoxification of benzo[a]pyrene.   In:  ACS Symposium Series,  No.  44,
     Drug Metabolism Concepts.   D.M. Jerina  (ed.).   1977.  p. 99-126.

195. Oesch, F., P. Bentley, and H.R. Glatt.  Prevention of benzo[a]pyrene-
     induced mutagenicity by homogeneous epoxide hydratase.   Int. J. Cancer,
     11:448-452, 1976.

196. 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.

197. Huberman, E., L. Aspiras,  C.  Heidelberger,  P.L.  Grover,  and P.  Sims.
     Mutagenicity to mammalian  cells of epoxides and other derivatives  of
     polycyclic hydrocarbons.  Proc. Natl. Acad. Sci.  USA, 68(12):3195-3199,
     1974.

198. Huberman, E. and L. Sachs.  Cell-mediated mutagenesis of mammalian cells
     with chemical carcinogens.  Int.  J. Cancer, 1_3:326-333,  1974.

199. Huberman, E. and L. Sachs.  Mutability  of different  genetic loci  in
     mammalian cells by metabolically  activated  carcinogenic  polycyclic
     hydrocarbons.  Proc. Natl. Acad.  Sci. USA,  73.(1): 188-192,  1976.

200. Newbold, R.F.,  C.B. Wigley,  M.H.  Thompson,  and  P. Brookes.  Cell-mediated
     mutagenesis in cultured Chinese hamster cells by carcinogenic  polycyclic
     hydrocarbons:  nature and  extent  of the associated hydrocarbon-DNA reaction,
     Mutation Res.,  43_:101-116, 1977.

201. Thakker, D.R.,  H. Yagi, H. Akagi,  M.  Koreeda, A.Y.H. Lu, W. Levin,
     A.W.  Wood, A.H. Conney, and  D.M.  Jerina.  Metabolism of  benzo[a]pyrene.
     VI.  Stereoselective metabolism of benzo[a]pyrene and benzo[a]pyrene
     7,8-dihydrodiol to diol epoxides.   Chem.-Biol.  Interactions, 16:281-300,
     1977.                                                       ~
                                     6-259

-------
202. Huberman, E., R. Mager, and L.  Sachs.   Mutagenesis and transformation of
     normal cells by chemical carcinogens.   Nature, 264(5584):360-361, 1976.

203. Krahn, D.F.  and C.  Heidelberger.   Liver homogenate-mediated mutagenesis
     in Chinese hamster V79 cells by polycyclic aromatic hydrocarbons  and
     aflatoxins.   Mutation Res., 46_: 27-44,  1977.

204. Dehnen, W.,  N.  Pitz, and R. Tomingas.   The mutagenicity of airborne
     particulate  pollutants.  Cancer Letters, 4_:5-12,  1977.

205. Hollstein, M.,  R.  Talcott, and  E.  Wei.  Quinoline:  conversion to a
     mutagen by human and rodent liver.   J. Natl.  Cancer Inst., 60(2):405-410,
     1978.

206. Kato, T., M. Yakushifi, A. Tsunawaki,  I. Ide, N.  Nirose, S. Araki, and
     M. Abe.  Studies on experimental  formation of ovarian tumors - especially,
     the discussion  of the developing  process of ovarian tumors following an
     application  of  DMBA.  Kurume Med.  J.,  221(3): 169-176, 1975.

207. Levan, G. and A. Levan.  Specific  chromosome changes in malignancy:
     studies in rat  sarcomas induced by two polycyclic  hydrocarbons.   Hereditas,
     79^:161-198,  1975.

208. Ahlstrom, U.  Chromosomes of primary carcinomas induced by 7,12-dimethyl-
     benz[a]anthracene in the rat.  Hereditas, 78_:235-244, 1974.

209. Popescu, N.C.,  C.H. Evans, and  J.A. DiPaolo.   Chromosome patterns (G and
     C bands) of  in  vitro chemical carcinogen-transformed guinea pig cells.
     Cancer Res., 36_:1404-1413, 1976.

210. Riidiger, H., F. Kohl, W. Mangels,  P. von Wichert,  C.R. Bertram, W. Wohler,
     and E. Passarge.  Benzpyrene induces sister chromatid exchanges in cultured
     human lymphocytes.   Nature, 26^(5566):290-292, 1976.

211. SchBnwald, A.D., C.R. Bartram,  and H.W. Riidiger.   Benzpyrene-induced
     sister chromatid exchanges in lymphocytes of patients with lung cancer.
     Human Genet., 36_:261-264, 1977.

212. Abe, S. and  M.  Sasaki.  Studies on chromosomal aberrations and sister
     chromatid exchanges induced by  chemicals.  Proc.  Japan Acad., 53(l):46-49,
     1977.                                                         ~~

213. Bayer, U.  In vivo induction of sister chromatid  exchanges by three poly-
     aromatic hydrocarbons.  In:  Carcinogenesis, Vol.  3:  Polynuclear Aromatic
     Hydrocarbons.  P.M. Jones and R.I.  Freudenthal (eds.).  Raven Press,
     New York.  1978.

214. Stich, H.F.  and B.A. Laishes.  DMA repair and chemical carcinogens.
     Pathobiology Annual.  Vol. 3.  H.L. loachim (ed.).  1973.   p. 341-376.


                                     6-260

-------
215. Maher, V.M., J.J. McCormick, P.L.  Grover, and P.  Sims.   Effect of DNA
     repair on the cytotoxicity and mutagenicity of polycyclic hydrocarbon
     derivatives in normal  and xeroderma pigmentosum human fibroblasts.
     Mutation Res., 43_:117-138, 1977.

216. Stich, H.F., R.H.C.  San, D.J.  Lam, L.W.  Koropatnick,  L.W. Lo,  and
     B.A. Laishes.   DNA fragmentation  and DNA repair as on i£ vitro and
     in vivo assay for chemical procarcinogens, carcinogens,  and  carcinogenic
     nitrosation products.   In:  Screening Tests in Chemical  Carcinogenesis,
     R. Montesano, H.  Bartsch, L. Tomatis (eds.).   IARC Scientific  Publ.  No.  12,
     Lyon.  1976.  pp. 15-24.

217. San, R.H.C. and H.F. Stich.  DNA  repair  synthesis of  cultured  human  cells
     as a rapid bioassay for chemical  carcinogens.  Int. J.  Cancer, 16:284-291,
     1975.

218. Stich, H.F., P. Lam, L.W. Lo,  D.J. Koropatnick, and R.H.C. San.   The search
     for relevant short term bioassays  for chemical carcinogens:  the tribula-
     tion of a modern  sisyphus.  Can.  J. Genet. Cytol., 1^:471-492, 1975.

219. Williams, G.M.  Carcinogen-induced DNA repair in  primary rat liver  cells
     cultures; a possible screen for chemical carcinogens.  Cancer  Letters,
     1:231-236, 1976.

220. Swenberg, J.A., G.L. Petzold,  and  P.R. Harbach.  In vitro DNA  damage/alka-
     line elution assay for predicting  carcinogenic potential. Bioc. Biophys.
     Res. Commun., 72J2):732-738, 1976.

221. Mitelman, F., J.  Mark, and G.  Levan.  Chromosomes of  six primary sarcomas
     induced in the Chinese hamster by  7,12-dimethylbenz[a]anthracene.  Hereditas,
     7^:311-318, 1972.

222. Mitelman, F. and  G.  Levin.  The chromosomes of primary 7,12-dimethyl[a]-
     anthracene-induced rat sarcomas.   Hereditas,  71_:325-334, 1972.

223. Kurita, Y., T. Sugiyama, and Y. Nishizuka.  Cytogenetic  studies  on  rat
     leukemia induced  by pulse doses of 7,12-dimethylbenz[a]anthracene.   Cancer
     Res., 28_: 1738-1752,  1968.

224. Stich, H.F.  Chromosomes of tumor  cells.  I.   Murine  leukemias induced  by
     one or two injections  of 7,12-dimethylbenz[a]anthracene.  J. Nat!.  Cancer
     Inst., 25:649, 1960.

225. Sugiyama, T.  Chromosomal aberrations and carcinogenesis by  various  benz-
     [a]anthracene derivatives.  Gann,  64_(6):637-639,  1973.

226. Joneja, M.G. and  D.B.  Coulson. Histopathology and cytogenetics  of  tumors
     induced by the application of 7,12-dimethylbenz[a ]anthracene  (DMBA) in
     mouse cervix.  Europ.  J. Cancer, £:367-374, 1973.

                                     6-261

-------
227. Joneja, M.G., D.B. Coulson, and E.A. MacKinnon.  Cytogenetic studies on
     two types of 7,12-dimethylbenz[a]anthracene (DMBA) induced malignant
     tumors of mice.  Anat. Rec., 969_:350, 1971.

228. Biedler, J.L., L.J. Old, and D.A.  Clarke.   Chromosome lesions associated
     with carcinogen-induced tumors in  mice.   Nature, 192_:286, 1961.

229. Hellstrom, K.E.  Chromosome studies on primary methylcholanthrene-induced
     sarcomas in the mouse.  J. Natl.  Cancer Inst., 23_:1019, 1959.

230. Ottonen, P.O. and J.K. Ball.  Lack of correlation between gross  chromo-
     some abnormalities and carcinogenesis with 7,12-dimethylbenz[a]anthracene.
     J. Natl. Cancer Inst., 50(2):497-501, 1973.

231. DiPaolo, J.A., R.L. Nelson, and P.J. Donovan.   Characteristics of primary
     tumors induced by carcinogenic polycyclic  hydrocarbons in Syrian hamsters.
     J. Natl. Cancer Inst., 46(1):171-177, 1971.

232. Nowell, P.C. and D.A. Hungerford.   Chromosome  studies in normal  and leukemic
     human leucocytes.  J. Natl. Cancer Inst.,  25_:85, 1960.

233. Vogel, E.  The relation between mutational pattern and concentration by
     chemical mutagens in drosophila.   In:  Screening Tests for Chemical
     Carcinogenesis.  R. Montesano, H.  Bartsch, L.  Tomatis (eds.).  Lyon, France;
     IARC.  IARC Publ. No. 12.  1976.   p. 117-132.

234. Fahmy, O.G. and M.J. Fahmy.  Genetic selections at specific loci by poly-
     cyclic hydrocarbons in relation to carcinogenesis.  Int. J. Cancer, 6_:250-
     260, 1970.

235. Fahmy, O.G. and M.J. Fahmy.  Mutagenie selectivity for the RNA-forming
     genes in relation to the carcinogenicity of alkylating agents and poly-
     cyclic aromatics.  Cancer Res., 32_:550-557, 1972.

236. Fahmy, O.G. and M.J. Fahmy.  Mutagenic properties of benzo[a]pyrene and
     its methylated derivatives in relation to  the  molecular mechanisms of
     hydrocarbon carcinogenesis.  Cancer Res.,  33_:302-309, 1973.

237. Fahmy, O.G. and M.J. Fahmy.  Oxidative activation of benz[a]anthracene and
     methylated derivatives in mutagenesis and  carcinogenesis.  Cancer Res.,
     33_: 2354-2361, 1973.

238. Nix, C.E. and B. Brewen.  The role of Drosophila in chemical  mutagenesis
     testing.  Biology Division, Oak Ridge National Laboratory.  1978.

239. Bridges, B.A.  Use of a three-tier protocol for evaluation of long-term
     toxic hazards particularly mutagenicity and carcinogenicity.   In;  Screen-
     ing Tests for Chemical Carcinogenesis.  R. Montesano, H. Bartsch,
     L. Tomatis (eds.).  Lyon, France;  IARC.   IARC  Publ. No. 12.  1976.  p. 549-559.


                                     6-262

-------
240. Kennedy, G.L., Jr., D.W. Arnold, and M.L. Keplinger.  Mutagenic response
     of known carcinogens.  1st International Conference, Asilomar.   The
     American Environ. Mutagen Soc., 1973, p. 224-225.

241. Epstein, S.S., E. Arnold, J.  Andrea, W.  Bass, and  Y. Bishop.   Detection
     of chemical mutagens by the dominant lethal  assay  in the mouse.  Toxicol.
     Appl. Pharmacol., 23_: 288-325, 1972.

242. Wyrobek, A.J. and W.R. Bruce.  Chemical  induction  of sperm abnormalities
     in mice.  Proc. Nat!. Acad. Sci. USA, 7201 ):4425-4429, 1975.

243. Hoffman, D. and E. Wynder.  Respiratory  carcinogenesis.  In:   Chemical
     Carcinogens.  C.E. Searle (ed.).  ACS Monograph 173, American  Chemical
     Society, Washington, D.C.  1976.

244. Wynder, E. and D. Hoffman.  Some laboratory  and epidemiological aspects
     of air pollution carcinogenesis.  J. Air Pollut. Contr. Assoc., 15:155-
     158, 1965.                                                      ~~

245. Kotin, P., H.L. Falk, P.  Mader, and M. Thomas.   Aromatic hydrocarbons.
     I. Presence in the Los Angeles atmosphere and the  carcinogenicity of
     atmospheric extracts.  Arch.  Ind.  Hyg.,  9_: 153-163, 1954.

246. Hueper, W.C., P. Kotin, E.C.  Tabor, W.W. Payne, H. Folk, and E. Sawicki.
     Carcinogenic bioassays on air pollutants.  Arch. Path., 74_:89-116, 1962.

247. Van Duuren, B.L.  Tumor-promoting  agents in  two-stage carcinogenesis.
     Prog. Exp. Tumor Res., Vh31-68, 1969-

248. Van Duuren, B.L.  Tumor-promoting  and co-carcinogenic agents in chemical
     carcinogenesis.  In:  Chemical Carcinogens.   C.E.  Searle (ed.).  ACS
     Monograph 173.  American Chemical  Society, Washington, D.C.  1976.
     p. 24-51.

249. Iball, J.  The relative potency of carcinogenic compounds.  Am. J.  Cancer,
     35:188-190, 1939.

250. International Agency for Research  on Cancer.   IARC Monographs on the
     Evaluation of Carcinogenic Risk of the Chemical  to Man.  Vol. 3.   Certain
     Polycyclic Aromatic Hydrocarbons and Heterocyclic  Compounds.  Lyon,
     France.   1973.

251. Arcos, J.C. and M.F. Argus.   Chemical  Induction of Cancer.  Vol.  IIA.
     New York, Academic Press.  1974.

252. Hecht, S.S., M. Loy, and D. Hoffmann.  On  the structure and carcinogenicity
     of the methylchrysenes.  In:   Carcinogenesis, Vol. 1:   Polynuclear Aromatic
     Hydrocarbons:  Chemistry, Metabolism, and  Carcinogenesis.   P.W. Jones and
     R.I.  Freudenthal (eds.).   New York, Raven  Press.   1976.

                                     6-263

-------
253. Slaga, T.J., A. Viaje, D.L.  Berry,  W.  Bracken,  S.G.  Butty,  and  J.D.  Scribner.
     Skin tumor initiating ability of benzo[a]pyrene 4,5-7,8-and 7,8-diol-9,10-
     epoxides and 7,8-diol.  Cancer Letter, 2(2):115-121,  1976.

254. Levin, W., A.W. Wood, P.G.  Wislocki, J.  Kapitulnik,  H.  Yagi,  D.M. Jerina,
     and A.H. Conney.   Carcinogenicity of benzo-ring derivatives of  benzo[a]-
     pyrene on mouse skin.  Cancer Res., 3^:3356-3361,  1977.

255. Levin, W., A.W. Wood, H.  Yagi, D.M. Jerina,  and A.H.  Conney.  (+)-trans-
     7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene:   a  potent skin  carcinogen when
     applied topically to mice.   Proc. Natl.  Acad. Sci.  USA,  73(11 ):3867-3871,
     1976.

256. Kapitulnik, J., W.  Levin, H.  Yagi,  D.M.  Jerina, and  A.H.  Conney.  Lack of
     carcinogenicity of 4-,5-,6-,7-,8-,9-,  and  10-hydroxybenzo[a]pyrene on
     mouse skin.  Cancer Res., 36;3625-3628,  1976.

257. Wislocki, P.G., R.L. Chang,  A.W.  Wood, W.  Levin, H.  Yagi, 0.  Hernandez,
     H.D. Mah, P.M.  Dansette,  D.M. Jerina,  and  A.H.  Conney.   High  carcino-
     genicity of 2-hydroxybenzo[a]pyrene on mouse skin.   Cancer  Res.,  37:
     2608-2611, 1977.                                                  ~~

258. Conney, A.H., A.W.  Wood,  W.  Levin,  A.Y.H.  Lu, R.L.  Chang, P.G.  Wislocki,
     R.L. Goode, G.M.  Holder,  P.M. Dansette,  H. Yagi, and D.M. Jerina.  Metabo-
     lism and biological  activity of benzo[a]pyrene  and its  metabolic  products.
     In:  Biological Reactive Intermediates.  D.J. Jallow, J.J.  Kocsis, R.
     Snyder, and H.  Vainio (eds.).  Plenum  Press.  1977.

259. Albert, R.E., F.J.  Burns, and B.  Altshuler.   Temporal Aspects of  Tumori-
     genic Response to Individual  and Mixed Carcinogens.   Comprehensive
     Progress Report.   Institute of Environ.  Med., New  York  Univ.  Med. Center,
     New York, NY 10016.   1978.

260. Kapitulnik, J., W.  Levin, A.H. Conney, H.  Yagi, and  D.M.  Jerina.  Benzo[a]-
     pyrene 7,8-dihydrodiol is more carcinogenic  than benzo[a]pyrene in newborn
     mice.   Nature, 266:378-380,  1977.

261. Kapitulnik, J., P.G. Wislocki, W. Levin, H.  Yagi,  D.J.  Jerina,  and
     A.H. Conney.  Tumorigenicity studies with  diol-epoxides  of  benzo[a]pyrene
     which indicate that (jj-trans-7&,8a-dihydroxy-9a,10a-epoxy-7,8,9,10-tetra-
     hydrobenzo[a]pyrene is an ultimate  carcinogen in newborn  mice.  Cancer Res.,
     38:354-358, 1978.

262. Peto, R., F.J.C.  Roe, P.N.  Lee, L.  Levy, and J. Clack.   Cancer  and ageing
     in mice and men.   Brit. J.  Cancer,  32^:411-426,  1975.

263. 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 benzo-
     [a]pyrene, benzo[a]pyrene diol epoxides, and other metabolites.   Cancer  Res.,
     37.: 984-990, 1977.

                                     6-264

-------
264. Turusov, V., N. Day, and L.  Andrianov.   J.  Natl.  Cancer Inst.,  47:105-111,
     1971.                                                          ~~

265. Law, L.W.  Multiple skin tumors In mice following a  single  painting  with
     9,10-dimethyl-l,2-benzanthracene.   Am.  J.  Pathol., V7:827-831,  1941.

266. Roe, F.J.  The development of malignant tumors  of mouse skin  after initia-
     ting and promoting stimuli.   1.  The effect of  a  single application  of
     9,10-dimethyl-l,2-benzanthracene (DMBA), with and without subsequent
     treatment with croton oil.  Brit.  J.  Cancer, ]0;61-69,  1956.

267. Terracini, B., P.  Shubik,  and G. DellaPorta.  A study  of skin carcino-
     genesis in the mouse with  single applications of  9,10-dimethyl-l,2-
     benzanthracene at  different dosages.  Cancer Res., 20^:1538-1542,  1960.

268. Klein, M.  Induction of skin tumors in  the  mouse  with  minute  doses of
     9,10-dimethyl-l,2-benzanthracene alone  or with  croton  oil.  Cancer Res.,
     JJ5:123-127, 1956.

269. Slaga, T.J., G.T.  Bowden,  J.D. Scribner, and R.K. Boutwell.   Dose-response
     studies on the ability of  7,12-dimethylbenz[a]anthracene and  benz[a]-
     anthracene to initiate skin tumors.  J.  Natl. Cancer Inst., 53_(5), 1974.

270. Bennington, J.L.,  E.J. Holmes, and J.W.  Combs.  Tumors  of the pilosebaceous
     unit induced in the rat by the intravenous  administration of  7,12-dimethyl-
     benz[a]anthracene.  J. Invest. Dermatol.,  66_:183-187,  1976.

271. Berenblum, I.  The carcinogenic action  of  9,10-dimethyl-l,2-benzanthra-
     cene on the skin and subcutaneous tissues  of the  mouse, rabbit,  rat, and
     guinea pig.  J. Natl. Cancer Inst., 1^:167-175, 1949.

272. Edgcomb, J.H. and  H. Mitchelich.  Melanomas of  the skin of  guinea pigs
     following the application  of a solution of  9,10-dimethyl-l,2-benzanthra-
     cene in benzene.  Acta Un. Int. Cancer, J9;706-707,  1962.

273. Boutwell, R.K.  Some biological aspects of  skin carcinogenesis.   Progr.
     Exp. Tumor Res., 4_:207-250,  1964.

274. Flesher, J.W., R.G. Harvey,  and K.L.  Sydnor.  Oncogenicity  of K-region
     epoxides of benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene.   Int. J.
     Cancer, 18:351-353, 1976.

275. Burki, K., J.E. Wheller, Y.  Akamatsu, J.E.  Scribner, G. Candelas, and
     E.  Bresnick.  Early differential effects of 3-methylcholanthrene  and its
     "K-region" epoxide on mouse skin.   Possible implication in  the  two-stage
     mechanism of tumorigenesis.   J. Natl. Cancer Inst.,  53_(4):967-976, 1974.

276. Howell, J.S.  Skin tumors  in the rat produced by  9,10-dimethyl-l,2-benz-
     anthracene and methylcholanthrene.  Brit.  J. Cancer, H[:101-109,  1962.


                                     6-265

-------
277. Berry, D.L., T.J. Slaga, A. Viaje, N.M.  Wilson, J.  DiGiovanni,  M.R.  Juchau,
     and J.K. Selkirk.  Effect of trichloropropene oxide on the ability of
     polyaromatic hydrocarbons and their "K-region" oxides to initiate skin
     tumors in mice and to bind to DNA in vitro.   J. Natl. Cancer Inst.,
     58(4):1051-1055, 1977.

278. Cavalieri, E., H. Garcia, P. Mailander,  and  K. Patil.  Isotope  effect on
     the carcinogenicity of 3-methylcholanthrene  in mouse skin by selective
     deuteration of the 1-methylene group:   biological  evidence for  a  mechanism
     of tumor initiation.  Chem.-Biol. Interactions, 1J_:179-189, 1975.

279. Sims, P.  The carcinogenic activities  in mice of compounds related to
     3-methylcholanthrene.  Int. J. Cancer, 2^:505-508,  1967.

280. Klein, M.  A comparison of the initiating and promoting  actions of 9,10-
     dimethyl-1,2-benzanthracene and 1,2,5,6-dibenzanthracene in skin  tumor-
     igenesis.  Cancer Res., 20_:1179, 1960.

281. Lijinsky, W., H. Garcia, B. Terracini, and U. Saffioti.   Tumorigenic
     activity of hydrogenated derivatives of dibenz[a,h]anthracene.  J.  Natl.
     Cancer Inst., 34_:1, 1965.

282. Van Duuren, B.L., L. Langseth, B.M.  Goldsmidt, and  L. Orris. Carcino-
     genicity of epoxides, lactones, and peroxy compounds.  VI.  Structure
     and carcinogenic activity.  J. Natl. Cancer  Inst.,  39_:1217, 1967.

283. Wynder, E.L. and D. Hoffmann.  A study of tobacco  carcinogenesis.   VII.
     The role of higher polycyclic hydrocarbons.   Cancer, 12:1079, 1959.

284. Hoffmann, D. and E.L. Wynder.  On the  carcinogenic  activity of  dibenzo-
     pyrenes.  Zeitschrift fur Krebsforschung, 68:137-149, 1966.

285. Buu-Hoi, N.P.  New developments in chemical  carcinogenesis by polycyclic
     hydrocarbons and related heterocycles:  a review.   Cancer Res., 24(9):
     1511-1523, 1964.                                               ~~

286. Stevenson, J.L. and E. Von Haam.  Carcinogenicity  of benz[a]anthracene
     and benzo[c]phenanthrene derivatives.   Amer.  Ind.  Hyg.  Assoc. J.,  Sept.-
     Oct. 1965, p. 475-478.

287. Wynder, E.L. and D. Hoffmann.  The carcinogenicity  of benzofluoranthenes.
     Cancer, 1^:1194-1199, 1959.

288. Van Duuren, B.L., A. Sivak, A. Segal,  L. Orris, and L.  Langseth.   The
     tumor-promoting agents of tobacco leaf and tobacco  smoke condensate.
     J. Natl. Cancer Inst., 37:519-526, 1966.
                                     6-266

-------
289. Saffiotti, U., R. Montesano, A.R.  Sellakumar,  and  D.G.  Kaufman.   Respira-
     tory tract carcinogenesis induced  in hamsters  by different dose  levels  of
     benzo[a]pyrene and ferric oxide.   J. Natl.  Cancer  Inst.,  49(4):1199-1204,
     1972.

290. Henry, M.C., C.D. Port, and D.G.  Kaufman.   Importance of  physical  proper-
     ties of benzo[a]pyrene-ferric oxide mixtures  in  lung tumor induction.
     Cancer Res., 35;207-217, 1975.

291. Flaks, A.  and P.  Sims.   The activities of  some polycyclic hydrocarbons
     and their "K-region" epoxides in  an in vitro-in  vivo carcinogenicity test
     system.  Brit. J. Cancer, 32^604-609, 1975.

292. Kendrick,  J., P.  Nettesheim, and A.S. Hammons.  Tumor induction  in tracheal
     grafts:  a new experimental model  for respiratory  carcinogenesis studies.
     J. Natl.  Cancer Inst.,  52^(4): 1317-1325, 1974.

293. Grisemer,  R.A., P. Nettesheim, D.H. Martin, and  J.E.  Caton, Jr.   Quanti-
     tative exposure of grafted rat tracheas to  7,12-dimethylbenz[a]anthracene.
     Cancer Res., 37.: 1266-1271, 1977.

294. Nettesheim, P., R.A. Griesemer, D.H. Martin,  and J.E. Caton, Jr.  Induc-
     tion of preneoplastic and neoplastic lesions  in  grafted rat tracheas con-
     tinuously exposed to benzo[a]pyrene.  Cancer  Res., 37^1272-1278, 1977.

295. Toth, B.   A critical review of experiments  in  chemical  carcinogenesis
     using newborn animals.   Cancer Res., 28_:727-738, 1968.

296. Roe, F.J.C., K.E.K.  Rowson, and M.H. Salaman.  Tumors of  many sites induced
     by injection of chemical carcinogens into  newborn  mice, a sensitive test
     for carcinogenesis:   the implications for  certain  immunological  theories.
     Brit. J.  Cancer,  1_5(5):5, 1961.

297. Asahina,  S., J. Andrea, A. Carmel, E. Arnold,  Y. Bishop,  S. Joshi, D. Coffin,
     and S.S.  Epstein.  Carcinogenicity of organic  fractions of particulate
     pollutants collected in New York  City and  administered  subcutaneously to
     infant mice.  Cancer Res., 32_: 2263-2268, 1972.

298. Walters,  M.A.  The induction of lung tumours  by  the injection of 9,10-
     dimethyl-1,2-benzanthracene (DMBA) into newborn  suckling  and young adult
     mice.  Brit. J. Cancer, 20^:148-160, 1966.

299. O'Gara, R.W., M.G. Kelly, and N.  Mantel.  Nature,  ]96_:1220, 1962.

300. Walters,  M.A. and F.J.C. Roe.  The time of appearance of  lung tumours in
     mice injected when newly born with 9,10-dimethyl-l,2-benzanthracene (DMBA).
     Brit. J.  Cancer,  20:161-167, 1966.
                                     6-267

-------
301. Grover, P.L., P. Sims, B.C.V.  Mitchley,  and F.J.C.  Roe.   The carcinogen-
     icity of polycyclic hydrocarbon epoxides in newborn mice.   Brit.  J.  Cancer,
     3J_:182-188, 1975.

302. Wislocki, P.G., J. Kapitulnik, W.  Levin, R. Lehr, M.  Schaefer-Ridder,
     J.M. Karle, D.M. Jerina, and A.H.  Conney.   Exceptional  carcinogenic
     activity of benz[a]anthracene 3,4-dihydrodiol  in the newborn mouse and
     the bay region theory.  Cancer Res., 38_:693-696, 1978.

303. Griswold, D.P., H.E. Skipper,  W.R.  Laster, W.S.  Wilcox,  and F.  Schabel.
     Induced mammary cancer in the female rat as a  drug  evaluation system.
     Cancer Res., 26_:2169, 1966.

304. 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.

305. Muggins, C., L.C. Grand, and P.P.  Brill antes.   Mammary cancer induced  by
     a single feeding of polynuclear hydrocarbons,  and its suppression.  Nature,
     189:204, 1962.

306. Huggins, C., R.C. Moon, and  S. Morii.  Extinction of experimental  mammary
     cancer.  I. Estradiol-17B and progesterone.  Proc.  Natl. Acad.  Sci.  USA,
     48:379, 1962.

307. Huggins, C.  Two principles  in endocrine therapy of cancers:  hormone
     deprival and hormone interference.   Cancer Res., 25^:1163,  1965.

308. Sinha, D. and T.L. Dao.  A direct mechanism of mammary carcinogenesis
     induced by 7,12-dimethylbenz[a]anthracene.  J. Natl.  Cancer Inst., 53(3):
     841-846, 1974.

309. Satake, K., K. Umeyama, K. Kami no, K.  Uchima,  K. Yamashita, and S. Yamamoto.
     Pancreatic cancer developing Sprague-Dawley rats after local application of
     7,12-dimethylbenzanthracene  (DMBA).  Osaka City Med.  J., 21_(2): 119-125,  1975.

310. Arkhipov, G.N.  Induction of cancer of the glandular stomach in rats by
     20-methylcholanthrene.  Int. J. Cancer,  2^:494-499,  1967.

311. Chouroulinkov, I., A. Gentil, and P. Sims.  Some biological effects  of
     7-hydroxymethyl-l-12-methylbenz[a]anthracene in mice.  Biomedicine,  19(10):
     438-441, 1973.                                                      ~~

312. Schmutz, J.A., A.C. Brownie, and A.P.  Chaudhry.   7,12-Dimethylbenz[a]-
     anthracene retention in the  rat submandibular  gland following intragland-
     ular injection.  Cancer Res., 34_:576-580,  1974.
                                     6-268

-------
313. Berwald, Y. and L. Sachs.   In vitro transformation with chemical  carcino-
     gens.  Nature, 200.: 1182-1184, 1963.

314. Heidelberger, C.   Chemical  oncogenesis  in  culture.   Adv.  Cancer Res.,  18:
     317-367, 1973.

315. Heidelberger, C.   Chemical  carcinogenesis.   Ann.  Rev.  Biochem., 44:79-113,
     1975.

316. Heidelberger, C.  and P.P.  Boshell.   Chemical  oncogenesis  in  cultures.
     Gann Monograph on Cancer Research,  ^7_:39-58,  1975.

317. Berwald, Y. and L. Sachs.   In vitro transformation of  normal  cells  to
     tumor cells by carcinogenic hydrocarbons.   J.  Natl.  Cancer Inst.,  35(4):
     641-657, 1965.

318. DiPaolo, J.A. and P.J.  Donovan.   Properties of Syrian  hamster cells trans-
     formed in the presence  of carcinogenic  hydrocarbons.   Experimental  Cell
     Research, 48:361-377,  1967.

319. Dao, T.L.  Carcinogenesis  of mammary gland in rat.   Prog.  Exptl.  Tumor
     Res., 5^:157-216,  1964.

320. Grover, P.L., P.  Sims,  E.  Huberman, H.  Marquardt, T.  Kuroki,  and
     C. Heidelberger.   In vitro transformation  of rodent cells  by K-region
     derivatives of polycyclic  hydrocarbons.   Proc. Natl. Acad.  Sci. USA,
     68(6):1098-1101,  1971.

321. Huberman, E., T.  Kuroki, H. Marquardt,  J.K. Selkirk,  C.  Heidelberger,
     P.L. Grover, and P.  Sims.   Transformation  of hamster embryo  cells  by
     epoxides and other derivatives of polycyclic hydrocarbons.   Cancer Res.,
     32;1391-1396, 1972.

322. Landolph, J.R., J.C. Bartholomew, and M.  Calvin.   Quantitative studies of
     the toxicity of benzo[a]pyrene to a mouse  liver epithelial  cell strain in
     culture.  Cancer Res.,  36_:4143-4151, 1976.

323. DiPaolo, J.A., P.J.  Donovan, and R.L. Nelson.   Transformation of hamster
     cells in vitro by polycyclic hydrocarbons  without cytotoxicity.  Proc.
     Natl. Acad. Sci.  USA,  6802):2958-2961,  1971.

324. DiPaolo, J.A., P.J.  Donovan, and B.C. Casto.   Enhancement by alkylating
     agents of chemical carcinogen transformation of hamster cells in culture.
     Chem.-Biol. Interactions,  9_:351-364, 1974.

325. Pienta, R.J., J.A. Poiley, and W.B. Lebherz, III.  Morphological  trans-
     formation of early passage golden Syrian hamster embryo cells derived
     from cryopreserved primary cultures as  a reliable in vitro bioassay for
     identifying diverse carcinogens.  Int.  J.  Cancer, JJ^:642-655, 1977.


                                     6-269

-------
326. Nesnow, S. and C. Heidelberger.   The effect of modifiers of microsomal
     enzymes on chemical  oncogenesis  in cultures of C3H mouse cell  lines.
     Cancer Res., 36.: 1801-1808, 1976.

327. Chen, T.T. and C. Heidelberger.   In vitro malignant transformation of
     cells derived from mouse prostate in the presence of 3-methylcholanthrene.
     J. Natl. Cancer Inst., 42_(6):915-922, 1969.

328. Chen, T.T. and C. Heidelberger.   Quantitative studies on the malignant
     transformation of mouse prostate cells by carcinogenic hydrocarbons in_
     vitro.  Int. J. Cancer, 4_: 166-178, 1969.

329. Marquardt, H., J.E.  Sodergren, P. Sims, and P.L.  Grover.  Malignant trans-
     formation in vitro of mouse fibroblasts by 7,12-dimethylbenz[a]anthracene
     and 7-hydroxymethylbenz[a]anthracene and by their K-region derivatives.
     Int. J. Cancer, 1^:304-310, 1974.

330. Marquardt, H., T. Kuroki, E. Huberman, O.K. Selkirk, C. Heidelberger,
     P.L. Grover, and P.  Sims.  Malignant transformation of cells derived  from
     mouse prostate by epoxides and other derivatives  of polycyclic hydrocarbons.
     Cancer Res., 32:716-720, 1972.

331. Gordon, R.J., R.J. Bryan, J.S. Rhim, C. Demoise,  R.G. Wolford, A.E. Freeman,
     and R.J. Huebner.  Transformation of rat and mouse embryo cells by a  new
     class of carcinogenic compounds  isolated from particles in city air.
     Int. J. Cancer, 1^:223-232, 1973.

332. Rhim, J.S., H.Y. Cho, L. Rabstein, R.J. Gordon, R.J. Bryan, M.B.  Gardner,
     and R.J. Huebner.  Transformation of mouse cells  infected with AKR leukaemia
     virus induced by smog extracts.   Nature, 239:103-107, 1972.

333. Freeman, A.E., P.J.  Price, R.J.  Bryan, R.J. Gordon, R.V. Gilden,
     G.J. Kelloff, and R.J. Heubner.   Transformation of rat and hamster
     embryo cells by extracts of city smog.  Proc. Natl. Acad. Sci. USA,
     68:445-449, 1971.

334. Leith, R.S. and L. Hayflick.  Efforts to transform cultured normal human
     cells with polycyclic aromatic hydrocarbons.  Proc. Am. Assoc. Cancer
     Res., 15:86, 1974.

335. Rhim, J.S., C.M. Kim, P. Arnstein, R.J. Nuebner,  E.K. Weisburger, and
     W.A. Nelson-Rees.  Transformation of human osteosarcoma cells by a
     chemical carcinogen.  J. Natl. Cancer Inst., 55_(6): 1291-1294, 1975.

336. Lasnitzki, I.  Growth pattern of the mouse prostate gland in organ culture
     and its response to sex hormones, vitamin A, and  3-methylcholanthrene.
     Natl. Cancer Inst. Monogr., ]2;318-403, 1963.
                                     6-270

-------
337. Crocker, T.T.  Effect of benzo[a]pyrene on  hamster,  rat,  dog,  and  monkey
     respiratory epithelia in organ culture.   In:   Proc.  of  a  Biology Division,
     Oak Ridge National  Laboratory.  Conference  held  in Gatlinburg, Tennessee,
     October 8-11, 1968.   AEC Symposium Series 18,  Oak Ridge,  Tennessee,  U.S.
     Atomic Energy Commission, Division of Technical  Information, 1970.

338. Lindsay, D.W., J.R.  Jones, W.J.  Higgins,  and  P.M. Brown.   The  bioassay  of
     carcinogenesis:   effects on the  epithelial  cell  complement of  rat  trachae
     maintained in vitro.   In:  Experimental  Lung  Cancer:  Carcinogenesis  and
     Bioassays, International Symposium.   1974.  p. 521-531.

339. Shabad, L.M., T.S.  Kolesnichenko,  T.V.  Nikonova.  Transplacental and
     direct action of benzo[a]pyrene  studied in  organ cultures of embryonic
     lung tissue.   Neoplasma, 22J2):113-122,  1974.

340. Bryan, W.R. and M.B.  Shimkin.   Quantitative analysis  of dose-response
     data obtained with  carcinogenic  hydrocarbons.  J. Natl. Cancer Inst.,
     1:807-833, 1941.

341. Heston, W.E.  and M.A. Schneiderman.   Analysis  of dose-response in  relation
     to mechanism of pulmonary tumor  induction in  mice.   Science, 117:109-111,
     1953.

342. Mantel, N., W.E. Heston, and J.M.  Gurian.   Thresholds in  linear dose-
     response models for carcinogenesis.   J.  Natl.  Cancer Inst., 27:203-215,
     1961.

343. Yanysheva, N. Ya. and Yu. G. Antomonov.   Predicting  the risk of tumor
     occurrence under the effect of small  doses  of carcinogens.  Environ.
     Health Perspect., 13:95-99, 1976.

344. Cornfield, J.  Carcinogenic risk assessment.   Science,  198:693-699,  1977.

345. Falk,  H.L., P. Kotin, and S. Thompson.   Inhibition of carcinogenesis.
     The effect of polycyclic hydrocarbons and related compounds.   Arch.
     Environ. Health, 9:169-179, 1964.

346. Pfeiffer, E.H.  Investigations on the carcinogenic burden by air pollution
     in man.  VII.  Studies on the oncogenetic interaction of  polycyclic  aro-
     matic hydrocarbons.   Zbl. Bakt.  Hyg., I.  Abt. Orig  B., l_58:69-83, 1973.

347. Pfeiffer, E.H.  Oncogenic interaction of carcinogenic and non-carcino-
     genic polycyclic aromatic hydrocarbons in mice.   In:  Air Pollution  and
     Cancer in Man.  V.  Mohr, D. Schmahl,  L.  Tomatis  (eds.).  International
     Agency for Research on Cancer. Scientific Pub!.  No.  16.  1977. p. 69-77.

348. Van Ouuren, B.L., C. Katz, and B.M.  Goldschmidt.  Brief communications:
     cocarcinogenic agents in tobacco carcinogenesis. J.  Natl. Cancer  Inst.,
     5J_(2):703-705, 1973.

                                     6-271

-------
349. Van Duuren, B.L. and B.M.  Goldschmldt.   Cocarcinogenic and tumor-promot-
     ing agents in tobacco carcinogenesis.   J.  Natl.  Cancer Inst.,  56:1237-
     1242, 1976.                                                   ~

350. Schmeltz, I., J. Tosk, J.  Hi Ifrich,  N.  Hi rota,  D.  Hoffmann,  and  E.  Wynder.
     Bioassays of naphthalene and  alkylnaphthalenes  for co-carcinogenic  activity.
     Relation to tobacco carcinogenesis.   In:   Carcinogenesis,  Vol. 3:   Poly-
     nuclear Aromatic Hydrocarbons.  P.W.  Jones and  R.I.  Freudenthal.  New York,
     Raven Press.   1978.

351. Shamberger, R.J.  Relationship of selenium to cancer.   I.  Inhibitory
     effect of selenium on carcinogenesis.   J.  Natl.  Cancer Inst.,  44:931-936,
     1970.

352. Shamberger, R.J. and 6.  Rudolph.   Protection against cocarcinogenesis by
     antioxidants.  Experientia, 22^:116,  1966.

353. Riley, J.F.  Mast cells.  Co-carcinogenesis and anti-carcinogenesis in
     the skin of mice.  Experientia, 4_:1237-1238, 1969.

354. Shamberger, R.J.  Increase of peroxidation in carcinogenesis.  J. Natl.
     Cancer Inst., 48:1491-1497, 1972.

355. Jaffe, W.  The influence of wheat germ  oil on the  production of  tumors  in
     rats by methylcholanthrene.  Exp. Med.  Surg., 4_:278-282,  1946.

356. Haber, S.L. and R.W. Wissler.  Effecto  of vitamin  E on carcinogenicity  of
     methyl cholanthrene.  Proc. Soc. Exptl.  Biol. Med.   m_: 774-775,  1962.

357. Wattenberg, L.W.  Inhibition  of carcinogenic and toxic effects of poly-
     cyclic hydrocarbons by phenolic antioxidants.  J.  Natl.  Cancer Inst.,
     48^:1425-1430, 1972.

358. Wattenberg, L.W.  Inhibition of chemical  carcinogen-induced pulmonary
     neoplasia by butylated hydroxyanisole.   J. Natl. Cancer Inst., 50:1541-
     1544, 1973.                                                   ~

359. Wattenberg, L.W., J. Speir, and A. Kotake.  Effects of antioxidants on
     metabolism of aromatic polycyclic hydrocarbons.   Adv.  Enzyme Regul.,  14:
     313-322, 1976.                                                       ~~

360. Wattenberg, L.W.  Inhibition of carcinogenic and toxic effects of poly-
     cyclic hydrocarbons by several sulfur-containing compounds.  J.  Natl.
     Cancer Inst., 5^(5):1583-1587, 1974.

361. Akin, F.J.  Anti-tumorigenic effect of  maleic hydrazide on mouse skin.
     J. Agric. Food Chem., 24(3):672-674,  1976.
                                     6-272

-------
362. Shamberger, R.J., F.F. Baughman, S.L. Kalchert, C.E. Willis, and G.C.
     Hoffman.  Carcinogen-induced chromosomal  breakage decreased by anti-
     oxidants.  Proc. Natl. Acad. Sci.  USA, 70^(5): 1461-1463, 1973.

363. Shamberger, R.J.  Antioxidants and cancer.   III.  Selenium and  other anti-
     oxidants decrease carcinogen-induced chromosome breakage.  In:   Trace
     Element Metabolism in Animals.  Proc. of Intl.  Symp., 2nd, 1973.  1974.
     p. 593-597.

364. Rahimtula, A.D., P.K. Zachariah, and P.O.  O'Brien.   The effects of anti-
     oxidants on the metabolism and mutagenicity of  benzo[a]pyrene  in vitro.
     Biochem. J., 164_:473-475, 1977.

365. Wattenberg, L.W. and J.L. Leong.  Inhibition of the carcinogenic action
     of benzo[a]pyrene by flavones.  Cancer Res., 30^:1922-1925, 1970.

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

367. 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., 1_6_:54, 1975.

368. Cone, M.V. and P. Nettesheim.  Effects of  vitamin A on 3-methylcholan-
     threne-induced squamous metaplasia and early tumors in the respiratory
     tract of rats.  J. Natl. Cancer Inst., 50_: 1599-1606, 1973.

369. 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.

370. 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.

371. Nettesheim, P. and M.L. Williams.   The influence of vitamin A  on the sus-
     ceptibility of the rat lung to 3-methylcholanthrene.  Int. J.  Cancer,
     1^:351-357, 1976.

372. 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.

373. 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.


                                     6-273

-------
374. Salaman, M.H. and F.J.C. Roe.   Cocarcinogenesis.   Brit.  Med.  Bull.,
     20(2):139-144, 1964.

375. Curtis, G.L., F.  Stenback, and W.L.  Ryan.   Inhibition of skin tumor
     formation with adenosine 3',5'-cyclic monophosphate in initiation-
     promotion carcinogenesis.   Proc.  Amer.  Assoc.  Cancer Res., lj>.:61,  1974.

376. Curtis, G.L., F.  StenbSck, and W.L.  Ryan.   Enhancement of 7,12-dimethyl-
     benzanthracene skin carcinogenesis by adenosine 3',5'-cyclic  monophos-
     phate.  Cancer Res.,  34(9):2192-2195, 1974.

377. Russell, D.H., C.V. Byus,  and C.  Manen.  Proposed model  of major sequen-
     tial biochemical  events of a trophic response.  Life Sciences, 19:1297-
     1306, 1976.                                                    ~~

378. Murray, A.M., A.K. Verma,  and M.  Froscio.   Effect of carcinogens and
     tumor promoters on epidermal cyclic adenosine  3',5'-monophosphate  metab-
     olism.  In:  Control  Mechanisms in Cancer.   W.E.  Criss,  T. Ono, and
     J.R. Sabine (eds.).  New York, Raven Press.  1976.

379. Gottschalk, R.G.   Quantitative studies on  tumor production in mice by
     benzpyrene.  Proc. Soc. Exp. Biol. Med., 5£:369-373, 1942.

380. Feron, V.J., D. De Jong, and P. Emmelot.  Dose-response correlation for
     the induction of respiratory-tract tumours  in  Syrian golden hamsters by
     intratracheal instillations of benzo[a]pyrene.  Europ. J. Cancer,  9:387-
     390, 1973.

381. Kobayashi, N. Production of respiratory tract  tumors in hamsters by
     benzo[a]pyrene.  Gann, 66_:311-315, 1975.

382. Sellakumar, A. and P. Shubik.   Carcinogenicity of 7H-dibenzo[c,g]carbazole
     in the respiratory tract of hamsters.  J.  Natl. Cancer Inst., 48(6):1641-
     1644, 1972.

383. Pylev, L.N.  Induction of lung cancer in rats  by intratracheal insuffla-
     tion of cancerogenic  hydrocarbons.  Acta Un. Int. Cancer, lj^:688-691, 1962.

384. Snell, K.C. and H.L.  Stewart.   Induction of pulmonary adenomatosis in
     DBA/2 mice by the oral administration of dibenz[a,h]anthracene.  Acta
     Un. Int. Cancer, 19_:692-694, 1962.

385. Hirano, T., M. Stanton, and M. Layard.   Measurement of epidermoid  carcinoma
     development induced in the lungs  of rats by 3-methylcholanthrene-containing
     beeswax pellets.   J.  Natl. Cancer Inst., 53_(5):1209-1219, 1974.

386. Jackson, D. and J.M.  Robson.  The effect of methyl cholanthrene on growth
     and pregnancy.  Brit. J. Exptl. Pathol., 39_: 133-138, 1958.


                                     6-274

-------
387. Svakur, L.D., O.K. Batra, B.N.  Sridharan, and L.D.  Sanghvi.   Effect of
     20-methylcholanthrene on mouse  embryos.   III. Stain C3H(Jax)-by Caesarean
     section.  J. Reprod. Fertil., 3^422-424,  1962.

388. Rigdon, R.H. and E.G. Rennels.   Effect of feeding benzpyrene on reproduc-
     tion in the rat.  Experientia,  20_:224-226, 1964.

389. Rigdon, R.H. and J.  Neal.  Effects of feeding benzo[a]pyrene on fertility,
     embryos, and young mice.  J.  Natl. Cancer Inst.,  34_(2): 297-305, 1965.

390. Bulay, O.M. and L.W. Wattenberg.   Carcinogenic  effects  of subcutaneous
     administration of benzo[a]pyrene  during pregnancy on the progeny (34993).
     Proc. Soc. Exptl. Biol.  Med., U5_:84-86,  1970.

391. Currie, A.R., C.C. Bird, A.M. Crawford, and P.  Sims.   Embryopathic effects
     of 7,12-dimethylbenz[a]anthracene and its hydroxymethyl  derivatives in the
     Sprague-Dawley rat.   Nature,  226_:911-914, 1970.

392. Bird, C.C., A.M. Crawford, A.R.  Currie, and B.F.  Stirling.   Protection
     from the embryopathic effects of 7-hydroxymethyl-12-methylbenz[a]anthra-
     cene by 2-methyl-l,2-bis-(3-pyridyl)-l-propanone(metopirone  ciba)  and
     0-diethylaminoethyldiphenyl-n-propyl  acetate (SKR 525-A). Brit. J. Cancer
     24:548-553, 1970.

393. Ayres, S.M. and M.E. Buehler.  The effects of urban air pollution  on
     health.  Clin. Pharmacol. Ther.,  Vh 337-371, 1970.

394. Pike, M.C., R.J. Gordon, B.E. Henderson,  H.R. Menck, and J.  SooHoo. Air
     pollution.  In:  Persons at High  Risk of Cancer - An Approach to Cancer
     Etiology and Control.  J.F.  Fraumeni, Jr. (ed.).   New York,  Academic Press.
     1975.

395. Blot, W.J., L.A. Bronton, and J.F. Fraumeni, Jr.   Cancer mortality in
     U.S. counties with petroleum industries.   Science,  198:51-53. 1977.

396. Lloyd, J.W. Long term mortality study of steelworkers.   V. Respiratory
     cancer in coke plant workers.  J.  Occup.  Med.,  ]3(2):53-68,  1971.

397. Sladden, A.F.  Pitch cancer.   In:   Report of International Conference  on
     Cancer.  1928.  p. 284-288.

398. Kennaway, E.L.  The anatomical  distribution of the occupational cancers.
     J. Ind. Hyg., 7_:69-93, 1925.

399. Kennaway, N.M. and E.L.  Kennaway.   A study of the incidence  of cancer  of
     the lung and larynx.  J. Hyg.,  36_:236-267, 1936.

400. Kennaway, E.L. and N.M.  Kennaway.   A further study of the incidence of
     cancer of the lung and larynx.   Brit. J.  Cancer,  l_:260-298,  1947.


                                     6-275

-------
401. Henry, S.A., N.M. Kennaway, and E.L. Kennaway.  The incidence of cancer
     of the bladder and prostate in certain occupations.  J.  Hyg., 31:125-137,
     1931.                                                         ~~

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

403. Kawai, M., H. Amanoto, and K.  Harada.  Epidemiologic study of occupational
     lung cancer.  Arch. Environ. Health, J4_:859-864, 1967.

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

405. Reid, D.D. and C. Buck.  Cancer in coking plant workers.   Brit. J.  Ind.
     Med., l_3:265-269, 1956.

406. Doll, R.  The causes of death among gas workers with special  reference to
     cancer of the lung.  Brit. J.  Ind. Med., 9_:180-187, 1952.
                                                                 •
407. Doll, 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  cancers
     of the lung and bladder, chronic bronchitis, and pneumoconiosis. Brit.  J.
     Ind. Med., 22_:1-12, 1965.

408. Doll, R., M.P. Vessey, R.W.R.  Beasley, A.R. Buckley, E.G. Fear, R.E.W.
     Fisher, E.J. Gammon, W. 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.

409. Redmond, C.K., A. Ciocco, J.W. Lloyd, and H.W. Rush.  Long term mortality
     study of steelworkers.  J.O.M., 1^(8):621-629, 1972.

410. Mazumdar, S., C. Redmond, W. Soiled to, and N. Sussman.   An epidemiological
     study of exposure to coal tar pitch volatiles among coke oven workers.
     APCA J., 25J4):382-389, 1975.

411. Redmond, C.K.  Epidemiological studies of cancer mortality in coke  plant
     workers.  For presentation at Seventh Conference on Environmental Toxi-
     cology, Dayton, Ohio, October 13-15, 1976.

412. Redmond, C.K., B.R. Strobino,  and R.H. Cypess.  Cancer experience among
     coke by-product workers.  Ann. NY Acad. Sci.  p. 102-115, 1976.

413. Hendricks, N.V., C.M. Berry, J.G. Lione, and J.J.  Thorpe.  Cancer of the
     scrotum in wax pressmen.  I. Epidemiology.  A.M.A. Arch.  Ind. Health,
     1^:524-529, 1959.

414. Hammond, E.G., I.J. Selikoff,  P.L. Lawther, and H. Seidman.  Inhalation
     of benzpyrene and cancer in man.  Ann. NY Acad. Sci.,  271:116-124,  1976.


                                     6-276

-------
415. Menck, H.R., J.T. Casagrande, and B.E.  Henderson.   Industrial  air pollu-
     tion possible effect on lung cancer.   Science,  183:210-212.  1974.

416. Menck, H.R. and B.E. Henderson.   Occupational differences  in rates of
     lung cancer.  J. Occup. Med., 1802):797-801, 1976.

417. Curwen, M.P., E.L.  Kennaway, and N.M.  Kennaway.  The incidence of cancer
     of the lung and larynx in urban  and rural  districts.   Brit.  J.  Cancer
     8:181-198, 1954.

418. Hoffman, E.F. and A.G. Gilliam.   Lung  cancer mortality,  geographic dis-
     tribution in the United States for 1948-1949.   Public Health Rep., 69:
     1033-1042, 1954.                                                   ~~

419. Haenzel, W., D.B. Loveland, and  M.G.  Sirken.  Lung-cancer  mortality as
     related to residence and smoking histories, 1,  white males.   J.  Natl.
     Cancer Inst., 28_:947-1001, 1962.

420. Hammond, E.C.  Quantitative relationship between cigarette smoking and
     death rates.  Nat.  Cancer Inst.  Monogr., 28_:3-8, 1968.

421. Sawicki, E.  Airborne carcinogens and  allied compounds.  Arch  Environ.
     Health, 14:46-53, 1967.

422. Kotin, P., H.L. Falk, P. Mader,  et al.   Aromatic hydrocarbons.   I.
     Presence in Los Angeles atmosphere and the carcinogen!city of atmospheric
     extracts.  AMA Arch. Ind. Health, 9_:153-163, 1954.

423. Wynder, E.L. and D. Hoffmann. Experimental tobacco  carcinogenesis.
     Science, Htf:862-871, 1968.

424. Manos, N.E. and G.F. Fisher.  An index of air pollution  and  its relation
     to health.  J. Air Pollution Control  Assn., 9_(1):5-11, 1959.

425. Schiffman, R. and E. Landau.  Use of indexes of air  pollution potential
     in mortality studies.  J. Air Pollut.  Control Assoc., 2^:384-386, 1961.

426. Levin, L., L. Morton, W. Haenszel, B.E.  Carroll, P.  Gerhardt,  V.H. Handy,
     and S.C. Ingraham.   Cancer incidence in urban and  rural  areas  of New York
     State.  J. Natl. Cancer Inst., 24(6):1243-1257, 1960.

427. Haenszel, W. and K.E. Traeuber.   Lung  cancer mortality as  related to
     residence and smoking histories.  II.  White females.   J. Natl.  Cancer
     Inst., 32_:803-838,  1964.

428. Stocks, P. and J.M. Campbell. Lung cancer death rates among non-smokers
     and pipe and cigarette smokers - an evaluation  in  relation to air pollu-
     tion by benzpyrene and other substances.  Brit. Med.  J., £:923-939, 1955.


                                     6-277

-------
429. Waller, R.E.  The combined effects of smoking and occupational  or urban
     factors in relation to lung cancer.   Am.  Occup.  Hyg.,  1^:67-71,  1972.

430. Hitosugi, M.  Epidemiological  study of lung cancer.   Inst.  Public Health
     Bull., 17:237-256, 1968.

431. Just, J., S. Maziarka, and M.  Dlugasiewicz.  Respiratory tract  cancer
     death rate and the atmospheric pollution  in some Polish  cities.   State
     Communal Hyg. Inst., 20(5):515-526,  1969.

432. Armenian, H.K. and A.M.  Lilienfeld.   The  distribution  of incubation
     periods of neoplastic diseases.   Amer.  J.  Epidemiology,  99_:92-100, 1974.

433. Reid, D.D., J. Cornfield, R.E. Markush, D.  Seigel, E.  Pedersen,  and
     W. Haenszel.  Studies of disease among migrants  and  native  populations
     in Great Britain, Norway, and  the United  States.   III. Prevalence of
     cardiorespiratory symptoms among migrants  and native-born in  the
     United States.  Natl. Cancer Inst. Monograph No.  19.   1966.

434. Eastcott, D.F.  The epidemiology of lung  cancer  in New Zealand.   Lancet,
     1:37-39, 1956.

435. Mancuso, T.F., E.M. MacFarlane,  and J.D.  Porterfield.  Distribution of
     cancer mortality in Ohio.  Amer. J.  Public Health, 45_:58-70,  1955.

436. Mancuso, T.F. and E.J. Coulter.   Cancer mortality among  native  white, for-
     eign-born white, and nonwhite  male residents of  Ohio:  Cancer of the lung,
     larynx, bladder, and central nervous system.  J.  Natl. Cancer Inst.,
     20_:79-105, 1958.

437. Dean, G.  Lung cancer among white South Africans. Brit. Med. J., 2:852-
     857, 1959.

438. Dean, G.  Lung cancer in South Africans and British  immigrants.   Proc. Roy.
     Soc. Med., 57:984-987, 1964.

439. Hagstrom, R.M., H.A. Sprague,  and E. Landau.  The Nashville air pollution
     study - VII:  Mortality from cancer in relation  to air pollution. Arch.
     Environ. Health, 15:237-248, 1967.

440. Winkelstein, W., Jr., S.  Kantor, E.W.  Davis, C.S. Maneri, and W.E. Mosher.
     The relationship of air pollution and economic status  to total  mortality
     and selected respiratory system mortality  in men. Arch. Environ. Health,
     1_4:162-171, 1967.

441. Winkelstein, W., Jr. and S. Kantor.   Stomach cancer,  positive association
     with suspended particulate air pollution.   Arch.  Environ. Health, 18:544-
     547, 1969.                                                        ~~


                                     6-278

-------
442. Winkelstein, W., Jr. and S. Kantor.   Prostatic cancer:   Relationship to
     suspended participate air pollution.   A.J.P.H., 52(7):1134-1136,  1969.

443. Higgins, I.T.T.  Epidemiology of lung cancer in the  United  States.   In:
     Air Pollution and Cancer in Man.  U.  Mohr,  D.  Schmahl,  and  L.  Tomatis  (eds.),
     International Agency for Research on  Cancer, Lyon, France,  1977.

444. Henderson, B.E., R.J. Gordon, H. Menck,  J.  SooHoo, S.P.  Martin, and
     M.C. Pike.  Lung cancer and air pollution in southcentral Los  Angeles
     County, California, USA.  Am. J. Epidemiol., J01_:477-488, 1975.

445. Stocks, P.  Epidemiology of cancer of the lung in  England and  Wales.   Brit.
     J. Cancer, 6:99-111, 1952.

446. Stocks, P.  Cancer and bronchitis mortality in relation  to  atmospheric
     deposit and smoke.  Brit. Med. J., 1:74-79, 1959.

447. Stocks, P.  Air pollution and cancer  mortality in  Liverpool  Hospital
     Region and North Wales.  Int. J. Air  Pollut.,  1:1-13, 1958.

448. Stocks, P.  The relations between atmospheric  pollution  in  urban  and
     rural localities and mortality from cancer, bronchitis,  and pneumonia
     with particular reference to 3:4 benzopyrene,  beryllium, molybdenum,
     vanadium, and arsenic.  Brit. J. Cancer, ]4_:398-418, 1960.

449. Stocks, P.  Recent epidemiological studies  of lung cancer mortality,
     cigarette smoking, and air pollution  with discussion of a new  hypothesis
     of causation.  Brit. J. Cancer, 20^:595-623, 1966.

450. Stocks, P.  Lung cancer and bronchitis in relation to cigarette smoking
     and fuel consumption in twenty countries.  Brit. J.  Prev. Soc. Med.,
     21:181-185, 1967.

451. Carnow, B.W. and P. Meier.  Air pollution and pulmonary cancer.   Arch.
     Environ. Health, 2^:207-218, 1973.

452. National Academy of Sciences.  Biologic  Effects of Atmospheric Pollutants:
     Particulate Polycyclic Organic Matter.  1972.

453. Goldsmith, J.R. and L.T. Friberg.  Effects  of air  pollution on health.
     In:  Air Pollution.  A.C. Stern (ed.).  New York,  Academic  Press,   (in
     press) p. 457-609.

454. Youngblood, W.W. and M. Blumer.  Polycyclic aromatic hydrocarbons in the
     environment:  Homologous series in soils and recent  marine  sediments.
     Geochim. et Cosmochim. Acta., 39:1303-1314, 1975.
                                     6-279

-------
455. Borneff, J., F. Selenka, H. Kunte, and A.  Maximos.   Experimental  studies
     on the formation of polycyclic aromatic hydrocarbons in plants.   Environ.
     Res., 2;22-29, 1968.

456. Hase, A. and R.A. Hites.  On the origin of polycyclic aromatic hydro-
     carbons in recent sediments:  Biosynthesis by anaerobic bacteria.
     Geochim. et Cosmochim.  Acta., 4_p_:1141-1143, 1976.

457. Payer, H.D., C.J. Soeder, H. Kunte, P. Karuwanna,  R. Nonhof,  and  W.  Graf.
     Accumulation of polycyclic aromatic hydrocarbons in cultivated microalgae.
     Naturwissenschaften., 62(11):536-537, 1975.

458. Fitzgerald, G.P., G.C.  Gerloff, and F. Skoog.  Chemicals with selective
     toxicity to blue-green algae.  Sewage and  Ind. Wastes, 24_:888-896,  1952.

459. Graf, W. and W. Nowak.   Growth stimulation in lower and higher plants by
     carcinogenic polycyclic aromatic compounds.  Arch.  Hyg. Bakt., 150:513-
     528, 1966.

460. Dunn, B.P. and H.F. Stich.  The use of mussels in  estimating  benzo[a]-
     pyrene contamination of the marine environment (38971).  Proc. Soc.  Exp.
     Biol. and Med., 150:49-51, 1975.

461. Hass, B.S. and H.G. Applegate.  The effects of unsubstituted  polycyclic
     aromatic hydrocarbons on the growth of escherichia coli.  Chem.-Biol.
     Interact., 10^:265-268, 1975.

462. Calder, J.A. and J.H. Lader.  Effect of dissolved  aromatic hydrocarbons
     on the growth of marine bacteria in batch  culture.   Appl. and Environ.
     Microbiol., 3£(1):95-101, 1976.

463. Meisel, M.N., G.A. Medvedeva, N.A. Pomoshchnikova, and G.E. Fedoseeva.
     Transfer of polycyclic aromatic hydrocarbons from  some yeast  cells  to
     others.  Microbiology (Eng. transl. Mikorbiologiya)., 4£(6):966-967, 1972.

464. Malaney, G.W., P.A. Lutin, J.J. Chibulka,  and L.H.  Hickerson.  Resistance
     of carcinogenic organic compounds to oxidation by  activated sludge.
     J. Water Pollut. Contr. Fed., 39:2020-2029, 1967.

465. Lee, R.F., R. Sauerheber, and G.H. Dobbs.   Uptake,  metabolism and dis-
     charge of polycyclic aromatic hydrocarbons by marine fish.  Marine  Biol.,
     1_7_:201-208, 1972.

466. Ahokas, J., R. Paakkonen, K. Ronnholm, V.  Raunio,  and 0. Pelkonen.   Oxi-
     datiye metabolism of carcinogens by trout  liver resulting in  protein
     binding and mutagenicity.  Hoppe-Seyler's  Z. Physio!. Chem. Bd.  357,
     S. 1019-1067, 1976.
                                     6-280

-------
467. 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.

468. Arffmann, E. and B.C. Christensen.   Studies on the newt test for carcino-
     genicity.  1. Benzo[a]pyrene,  dibenz[a,h]anthracene and 3-methylcholan-
     threne.  Acta Pathol. Microbiol. Scand., 52_:330-342, 1961.

469. Balls, M.  Benzpyrene-induced  tumours in the clawed toad, xenopus laevis.
     Experientia, 20(3):143-145, 1964.

470. Radding, 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
     Hydrocarbons.  Prepared by Stanford Research Institute, Menlo  Park,
     California, under Contract No. 68-01-2681.   U.S. Environmental Protection
     Agency.  Washington, D.C.  Pub!. No.  EPA-560/5-75-009.  1976.

471. Shimkin, M.B. and G.D. Stoner.  Lung  tumors in mice:  Application to
     carcinogenesis bioassay.  In:   Advances in  Cancer Research, Vol. 21,
     G. Klein and S. Weinhouse Teds.).   New York, Raven Press.  1975.  p. 1-38.
                                     6-281

-------