|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 273
CIRCULATORY SYSTEM
Juchau et al .76~77 summarized a body of literature bearing on the
hypothesis that PAHs may play an important role in the pathogenesis of
arteriosclerotic lesions. The validity of this hypothesis apart, these
investigators clearly demonstrated that aortic tissues from a number of
species, including man, have detectable, albeit low, monooxygenase
activities using BaP and 7,12-DMBA as substrates. Enzyme activities
were comparable with those characterizing mouse skin. Cytochrome P-450
could be detected in primate aortas, and epoxide hydratase activity for
BaP 4,5-oxide was identified in homogenates of the arterial walls of
chickens and rabbits. The characteristics of the aortic monooxygenase
for BaP resembled those of the enzyme system found in other tissues.
It could be markedly induced, for example, by 3-MC, polychlorinated
biphenyls (PCBs), and 5,6-benzoflavone; and, surprisingly, aortic
homogenates produced higher than expected quantities (by as much as a
factor of 28) of alkali-extractable metabolites when hematin was added
to the reaction mixtures. Interestingly, hematin has been shown in
other studies to degrade, in vitro, components of the monooxygenase
system.lll The primary BaP metabolites formed in rabbit aortic
homogenates were the 3-OH and 9-OH derivatives, phenolic compounds
known to be cytotoxic. The authors cited unpublished data to show that
the aortic metabolites of BaP form covalent bonds with such macro-
molecules as calf-thymus DNA. Treatment of chickens with the inducer
3-MC markedly increased the amount of the PAM-DNA adducts, whereas
addition of 7,8-benzoflavone in vitro inhibited binding. Aortic
enzymes also have been shown to catalyze the formation of mutagenic
metabolites from 7,12-DMBA. Thus, both cytotoxic and mutagenic
metabolites of PAHs can be generated in vascular tissues. The possible
relation of the formation of these compounds to the initial vascular
injury that may presage the local development of an atherosclerotic
plaque is of considerable interest.
The interaction of benz~aJanthracene and BaP with crystalline human
serym albumin in solution has been studied fluorimetrically by Ma et
al. 10 Equilibrium studies indicated that both PAHs hind to the pro-
tein to the same extent. Evidence of energy transfer from the trypto-
phan residue of the protein (increase in the weak B region--395-420
nm--fluorescence of the PAHs) permitted an assessment of the mean dis-
tance between the tryptophan and the bound ligand, thus identifying two
different binding sites in the same general area. The authors sug-
gested that structural differences among hydrocarbons, which may
greatly affect their orientations on the protein molecule, influence
mainly the selection of the binding site, rather than the binding
equilibrium.
In vivo BaP associates very little with serum albumin in the
presence of lipoproteins. The kinetics of BaP transfer between human
plasma lipoproteins have been examined by Smith and Doodyl63 with
high-density lipoproteins (HOL), low-density lipoproteins (LDL), and
6-ll
OCR for page 274
very-low-density lipoproteins (VLDL) prepared from fresh unfrozen human
plasma by ultracentrifugal flotation. BaP-lipoprotein interactions
were analyzed fluorimetrically, and kinetic measurements were
determined by stopped-flow techniques. The half-times of BaP transfer
between HDLs, between LDLs, and between VLDLs were 40, 180, and 390 ms,
respectively. The transfer of these PAHs among lipoproteins of the
same density class was about one-twentieth that of pyrene under the
same conditions. The rate of BaP transfer between lipoproteins also
decreased with increasing size of lipoprotein; at equilibrium in vitro,
VLDLs contain about 10 times more of the BaP than LDLs, and LDLs
contain 20-50 times more than HDLs. The distribution between plasma
and erythrocytes is different for 7,12-DMBA, BaP, benzanthracene, and
anthracene, the mass of the PAH being associated with red cells (50,
70, 93, and 100t, respectively). Plasma lipid concentrations and the
dynamics of lipid and lipoprotein metabolism clearly may have an impact
on PAH distribution in blood and into specific tissues. For example,
transfer of BaP is quite rapid, compared with the half-time for either
hydrolysis of chylomicron triglyceride (about 2-5 min in humans) or
clearance of the most abundant lipoproteins from the circulation (3-5 d
in humans). The data of Smith and Doodyl63 concerning the role of
plasma lipoprotins in the transport of PAHs corroborated and extended
the findings of other invle~t~ai2ors wh6O examined the interaction of
PAHs and plasma proteins. , , ,57,1 2
The specific process of BaP uptake from human LDLs into cultured
human cells was examined by Remsen and Shireman.149 The cell lines
used were WI-38, a human embryonic lung-fibroblast line, and GM 1915, a
skin-fibroblast line derived from a patient with homozygous familial
hypercholesterolemia; the former cells are LDL-receptor-positive, and
the latter LDL-receptor-negative. Thus, in these studies, it was
possible to explore the role of LDL receptors in the cellular uptake of
PAHs that enter the bloodstream transported by chylomicrons and plasma
lipoproteins. The results indicated that cellular uptake of the
tritiated PAH by both cell lines from delipidated or serum-free medium
varied linearly with concentration, whereas incorporation of PAH bound
to LDLs was much less and, at higher lipoprotein concentrations, varied
nonlinearly. The prese2ce of the PAH in the LDL preparation did not
affect the binding of 1 5I-labe]ed lipoprotein to receptor-positive
cells. The study provided several findings of special importance
relative to the biologic impact of PAHs--or at least BaP as a model
compound--on tissues in vivo. Clearly, although LDLs carry substantial
amounts of PAH, the presence of LDL receptors on cel Is is not necessary
for tissue uptake. The fact that PAH bound to LDL was incorporated into
cells more slowly than PAH in a delipidated serum or serum-free medium
raises questions about the biologic significance of experimental models
in which increased incorporation of BaP from particles into lipid
vesicles has been demonstrated. The data £rom these experiments also
indicate that cells that may be directly exposed to a PAH (i.e.,
tracheobronchial, intestinal, and cutaneous cells) before the compound
reaches the bloodstream may accumulate PAH in much higher
concentrations than cells exposed to the PAH bound to lipoproteins,
inasmuch as the latter significantly slowed as well as limited the
6-12
. -
OCR for page 275
cellular uptake of BaP. Finally, the report indicated that BaP
previously incorporated into WI-38 cells could be substantially removed
(by 55-797) in a 120-min posttreatment study period by 10% delipida ted
serum or LDL-containing medium. This finding implies a potential for
considerable PAN redistribution and a requirement for a not
insignificant period for progression of the hydrocarbon from the plasma
membrane to the endoplasmic reticulum, where metabolism takes place.
The ability of human monocytes to oxidize BaP and the induction of
this enzyme activity by bt~z9agth:gcene have been demonstrated by
several investigators. 6, , ,1 Lake and colleagues96
re-examined this problem with the goal of developing a practical assay
for measuring whole-cell metabolism of BaP under highly standardized
conditions, eliminating--among other problems--the need for a large
volume of blood (50 ml) in the fluorometric assay developed earlier for
AHH activitity in this cell type. By measuring whole-cell generation
if water-soluble BaP metabolites over a 3-d culture period, using
H-labeled substrate and closely controlling other character-
istics, they provided a useful alternative cell system to that using
mitogen-stimulated lymphocytes for characterizing BaP oxidation
activity in humans.
Because of the advantage gained by the much greater inducibility of
AHH activity (-up to 40-fold) in cultured monocytes, compared with
mitogen-stimulated 1 ~hocytes (about 5-fold), the monocyte system was
used by Okuda et al. to study the contribution of genetic factors
to the control of individual variation in AHH inducibility. Ten sets
of monozygotic tissues were assayed two to four times and 17 sets of
dizygotic tissues one to three times for basal and induced monocyte AHH
activity. The results indicated that 55-70% of the individual
variation in AHH inducibility of monocytes was genetically determined.
Variation in AHH inducibility within subjects in repeat assays was wide
and approached the magnitude of the variation between subjects. Thus,
a single AHH assay is an imprecise biochemical characterization of a
subject. Alternatively expressed, the method then available (late
1977) made it impractical to characterize a population with genetically
distinct differences in AHH inducibility. The large intrasubject
variation in AHH inducibility of monocytes also indicated that, in
addition to the clear genetic influences on this process, unknown
environmental or technical factors expressed themselves in the test
procedure.
An abundant literature exists related to the monooxygenase activity
of lymphocytes; the inducibility of this activity by mitogens, which
have the property of stimulating lymphocyte transformation, during
which a number of metabolic activities are concurrently greatly
increased; and the use of mitogen-stimulated lymphocytes to study the
genetic control of AHH in man and its relation to the occurrence of
Some human cancers--notably those of the l~lg,2 Kouri and colleagues
have1 revi2ewed key aspects of this subject; ' McLemore et
al. 19-12 have also provided a detailed analysis of the genetics of
AHH and its purported relation to human cancer. Only a brief summary
of these findings can be included here.
6-13
OCR for page 276
The identification of AHH activity in lymphocytes in 197229,192
and its increase during lymphocyte blastogenesis led quickly to
clinical studies, the earliest being that of Kellermann et al. ,87 in
which this induced enzyme activity was measured in cultured lymphocytes
of normal controls, non-lung-tumor controls, and lung-cancer patients.
In a preceding study in the same year, this grouped had examined the
genetic variation in AHH activity in lymphocytes of 353 normal subjects
and had categorized the population into three groups--low, inter-
mediate, and high responders with respect to AHH inducibility; the
population frequencies were about 50t, /~0X, and 10%, respectively. The
conclusion was reached that the enzyme activity was controlled by two
alleles at a single gene locus and that the high and low responders
were homozygous and the intermediate group ~7eterozygous for those
alleles. In the initial lung-cancer study, there was a virtual
absence of cases in the low-inducibility population, and all but two
cases were in the intermediate- and high-inducibility categories. All
the lung-cancer cases were in heavy smokers; of the 50 subjects, 48 had
an average consumption of two packs of cigarettes per day. When the
two control groups (normal subjects and a non-lung-cancer tumor group)
and the lung-cancer group were compared for risk of lung cancer, those
with intermediate and high inducibility (48 of the 50 lung-cancer
cases) had risks for lung cancer 16 and 36 times, respectively, the
risk in the low-inducibility group. This study prompted considerable
controversy over the next few years, during which the findings of
Kellermann and associates were cast in doubt.
85 strong correlation (r = 0. 923) was also found by Kellermann et
al. between the plasma elimination rate of antipyrine and the rate
of BaP metabolism in human lymphocytes from a "carefully selected
homogeneous" population, compared with the much lower correlation (r =
0.425) found in a "heterogeneous" population. The authors interpreted
their findings as supporting the existence of common oxidative systems
or common genetic control of the systems for antipyrine and BaP
oxidation. Atlas et al. confirmed that plasma antipyrine half-life
is correlated to some extent with AHH inducibility (r = 0.84), although
no intrasubject correlations were found between AHH inducibility and
the oxidation of other drug substrates, such as phenylbutezone and
bishydroxycoumarin. Most importantly, this group, 7 while affirming a
significant heritable determinant of AHH inducibility in human
lymphocytes, failed to confirm the monogenic model and trimodal distri-
bution of AHH indu86bility in the general population, proposed by
Kellermann _ al.; rather, the population distributions for AHH
inducibility (and for plasma antipyrine half-life) were consistent with
polygenic control of both traits in man. In other studies in which the
relation of AHH indu-cibility t~ th~ occurrence of lung cancer was
re-examined by Paigen et al., 331 4 low AHH activity was found in
half the tumor patients studied, in contrast with the earlier findings
of Kellermann et al.,87 and no characteristic alterations in this
enzyme activity were found in the progeny of these patients. A con-
siderable number of technical problems related to the lymphocyte-AHH
assay may confound the results obtained in studies of this enzyme
6-14
. . .
OCR for page 277
activity and its relation to human cancer, as noted by Kouri et
al.91 However, recent methodologic advances made by this group,
particularly the use of cryopreserved lymphocytes and close control of
a number of assay variables, have added an important degree of
precision to the assay.
Chrysene, one of several PAR derivatives (benzanthracene is
another), has been shown by Snodgrass et al.164 to induce AHH
activity in cultured human lymphocytes (from normal subjects) with BaP
as substrate. The individual variation in the monooxygenase activity
observed with other inducers was also seen with chrysene.
The comparative metabolism of BaP in human lymphocytes and human
liver microsomes has been studied by Selkirk et al.,l6u who examined
the nature of the metabolites formed by each cellular system. The
patterns of metabolizes formed in both cell systems had characteristics
quite similar to each other, with some exceptions--for example, among
the derivatives formed in a 30-min incubation, all three dihydrodiols
produced by liver were absent in the lymphocyte incubation mixture. In
a~24-h incubation of lymphocytes, however, all three dihydrodiols
formed by liver microsomes were also formed by the blood cells, and new
metabolite peaks were observed, presumably reflecting more extensive
biotransformation of already formed metabolizes in the reaction
mixture. The authors concluded that, although the ratios of some
metabolites may differ and although lymphocytes form several more
derivatives than does liver, many identical metabolizes are produced in
these two human cell types.
Schonwald et al.l58 studied the effect of BaP on sister chromatic
exchange in mitogen-stimulated lymphocytes of 11 normal subjects and 18
patients with lung cancer. Patients and controls differed neither with
respect to the spontaneous rate of sister chromatic exchange nor in
their responses to the hydrocarbon, although it did double the number
of exchanges in both population groups.
Barfknecht et al.l5 studied the ability of dichloromethane
extracts of automobile diesel soot at high concentrations (100 mg/m3)
to induce trifluorothymidine-resistant mutants in human lymphocytes
incubated in the presence of rat-liver postmitochondrial supernatant.
A significant induction of such mutants was observed. Anthracene,
phenanthrene, and their alkylated derivatives accounted for one-fourth
of the observed biologic activity. Among eight related compounds,
there was general agreement between responses in lymphoblasts and in
bacterial test systems. Phenanthrene was an exception, in that it was
positive in the human-lymphoblast test system, but negative in bacteria
at a concentration 60 times higher. The data in this report indicate
that methyl substitution at some sites of anthracene and phenanthrene
greatly increases their mutagenicity in both S. typhimurium and human
lymphoblasts. A similar effect for chrysene has been observed.
Methylations at the 1 and 3 positions of phenanthrene and the 2 and 9
positions of anthracene result in PAHs that are particularly mutagenic
6-15
At;
OCR for page 278
in the human and bacterial test systems used. Methylations at other
positions had the capability of eliminating the mutagenic activity of
the PAH derivative. No correlation between the results of the
mutagenesis studies with the soot-derived PAHs and the reported
capacity of the compounds studied to elicit neoplastic or carcinogenic
responses in test animals could be made.
REPRODUCTION
The title of this section refers collectively to studies related to
the ability of some genital tissues (including the placenta) to
metabolize or otherwise respond biochemically to Part. There is an
abundant and detailed literature on transplacental and peri-
natall8 carcinogenesis. These and related topics in reproduction
were reviewed in a 1981 special issue of the Journal of Environmental
Pathology_and Toxicology and are not summarized here. It is
perhaps appropriate, however, to refer to the report by Sir Percival
Pott in 1775, 42 in which there was first described an increased
incidence of scrotal cancer in chimney sweeps exposed to soot and to
note that almost 150 yr elapsed before Yamagiwa and Ichikawal§4
demonstrated that the repetitive application of crude coal tar to the
rabbit ear produced skin cancer and that the identification of specific
carcinogenic coal-tar constituents, such as BaP, required the passage
of additional decades.20~43' 8 Over this period, the question of why
only scrotal cancers, and not other genital cancers or even other
cancers in general, were found in-excess in chimney sweeps appears to -
have remained unanswered.
Grover _ al.66 investigated the metabolism--including the
Specific identification of biotransformation products--of three
H-labeled PAHs by nonneoplastic human mammary epithelial-cell
aggregates maintained in culture. The lobuloalveolar units from which
these aggregates are derived are thought to be the site of origin of
many human mammary carcinomas; two of the PAHs studied, 7,12-DMBA and
BaP, are known to be relatively potent mammary carcinogens in rats,
whereas benz~aJanthracene is not a mammary carcinogen in rats. Tissues
from eight patients were studied. The extent of metabolism of the PAHs
is summarized in Table 6-4. There was considerable individual
variation in PAH metabolism among the subjects studied, but the
formation of water-soluble metabolites by the tissue samples accounted,
in each instance, for a major portion of the total of each PAH
metabolized. The extent of binding of each PAH to cellular DNA and
proteins also varied considerably. Interestingly, the extent to which
H-labeled metabolites of benz~aJanthracene--a noncarcinogen for
mammary tissue in the rat--were bound seemed, from the limited data
obtained, to be consistently lower than the binding displayed by the
other two PAHs. The results of chromatographic characterization of
PAM-DNA adducts formed suggested that, with Hap, the hydrocarbon was
activated by the cultured cells through the formation of anti-BaP
7,8-diol-9, 10-oxide, a bay-region diol-epoxide that appears to be
responsible for most of the nucleic acid adducts formed in several
6-16
OCR for page 279
other biologic systems. The situation was less clear with 7,12-DMBA,
although a portion of the adducts formed with this PAH
cochromatographed with adducts present in a DNA hydrolysate that had
been treated with anti-7,12-DMBA 3,4-diol-1,2-oxide--a derivative that
is also classified as ~ bay-region diol-epoxide. The authors
interpreted their data with caution, considering all the factors known
to bear on the development of mammary cancer; but the possibility of
partial causal relationships among the PAHs, their metabolic
transformations, and tumor stimulation is implicit in this work.
Stampfer and colleagues168 did similar studies with BaP and
cultured mammary epithelial cells and fibroblasts. They showed that
the breast epithelial cells were 50-100 times more sensitive (growth
inhibition) to BaP than the fibroblasts; that the epithelial cells
formed adducts as early as 6 h after addition of the PAH to the
cultures; and that the adducts between the 7R anti stereoisomer of BaP
diol-epoxide and deoxyguanosine predominated at all times and, with two
minor adducts that were consistently present, persisted in the
epithelial cells for at least 72 h in a BaP-free medium. No adducts
were detected in fibroblasts until 96 h after exposure to the PAH, at
which time the type and extent of adduct formation were similar to
those observed with epithelial cells. As with the report of Grover et
al.,66 caution concerning the direct relation of these findings to
the role of PAHs in mammary carcinogenesis is necessary. On this
matter, Stampfer and co-workers168 stated, however, that "chemical
carcinogens, particularly BaP, should not be minimized as possible
factors in the initiation of breast cancer."
Mass et al.113 studied 26 specimens of normal human endometrium
_ _ ~
to determine the patterns of metabolism of [~H]BaP in short-term
explant cultures. Three of the tissue samples were from postmenopausal
women; of the remaining 23, it was possible to approximate the stage of
the menstrual cycle at which the tissue was removed during surgery.
Eight of the latter subjects were smokers. In summary, it was clear
that normal human endometrium could enzymatically convert BaP to a wide
variety of oxygenated derivatives that cochromatographed with
dihydrodiols, quinones, and monohydroxy products of the PAH; sulfation
was also identified. HPLC analysis of metabolites revealed marked
individual variation in metabolite formation among the subjects
studied; smoking did not account for this difference, but some evidence
of hormonal influences on the patterns of PAH metabolism was adduced.
In a study by Dorman et al.,52 BaP binding to DNA in human
endometrial tissue was studied in samples obtained from 41 subjects
and, again, a striking (70-fold) range in the observed specific
activities of carcinogen binding to UNA was identified (see Figure
6-1~. Tissues obtained late in the proliferative phase or early in the
secretory phase of the menstrual cycle had the highest mean specific
activity of PAM-DNA binding (Table 6-5~. Binding was significantly
reduced when tissue specimens from low-estrogen periods of the
menstrual cycle were studied. The reason for this apparent association
between estrogen content (actually, the estimated phase of the cycle)
6-17
~.. . ..
OCR for page 280
and PAM-DNA binding is obscure, but clearly merits further study. Such
study would have to deal with the important confounding factor of the
broad range of individual variation in binding, which may mask
systematic but small changes that can occur during a menstrual cycle,
but which cannot now be detected.
Namkung and Juchaul28 studied the oxidative biotransformation of
BaP in preparations of human placental microsomes with HPLC. The
investigations revealed that the use of substrate concentrations high
enough to ensure zero-order reaction kinetics markedly inhibited the
formation of dihydrodiols in the reaction mixtures. The relative
quantities of dihydrodiols generated increased with decreasing
substrate concentrations between 200 and 2.7 ~M. Addition of manganese
or ferric ions to reaction mixtures altered the ratios of generated
phenols to dihydrodiols. Identical results were obtained with C-
and 3H-labeled BaP as substrate. The data suggested that
considerable amounts of 7,8-dibydroxy-7,8-dibydro-BaP, a proximate
mutagen-carcinogen, may be generated in vivo by placental tissues of
women who smoke.
The formation of PAH metabolite-nucleoside adducts when human tumor
placental microsomes were incubated with i3H]BaP and salmon sperm DNA
has been studied by Pelkonen and Saarni.L 9 There were significant
differences between the PAH metabolite patterns and the nucleoside-
metabolite complexes formed, compared with rat liver, for example.
Specifically, in the human placenta microsomes, the absence of the
nucleoside complex of 9-'nydroxy-4,5-oxide implied the inability of this
tissue to form 4,5-oxides of BaP. Indirect evidence of epoxide
hydratase activity in placental tissue was obtained. The extent of
PAM-DNA binding in this tissue correlated significantly with both
7,8-diol metabolite formation and fluorometrically determined AHH
activity. The question of whether the 7,8-dioi-9,10-epoxide of BaP is
formed by the human placenta in vivo could not be answered unequi-
vocally, but the authors' inferential conclusion is that it is probably
formed in the human host. The interplay of possible genetic influences
and clearly established regulatory influences of environmental factors
on Herman placental AHH has been incisively discussed by the same
group.l38
Cigarette-smoking has been shown by Conney and
associates42~189~190 to be one of the most potent and consistent
inducers of human placental AHH activity yet identified. In the
initial report of the group, 189 the enzymatic hydroxylation of BaP
could not be detected in nonsmokers in homogenates of placentas frozen
immediately after birth and studied within 48 h. In contrast, the
enzyme activity was present in all 11 placentas from women who smoked
during gestation, although enzyme activity in this small group did not
correlate with the number of cigarettes smoked. BaP administration to
pregnant rats also was shown to induce AHH activity in the placenta.
The effect was related to PAH dose. This study constituted the first
demonstration that compounds in cigarette smoke could induce a
carcinogen-metabolizing enzyme in human tissues. These studies were
6-18
OCR for page 281
extendedl90 to related enzymatic reactions in human placentas and to
other types of pyrenes as probes for AHH-inducing activity in rat
placenta (see Table 6-6~. Extremely active inducers included chrysene,
1,2-benzanthracene, pyrene, 3,4-benzofluorene, and a number of related
compounds. 90 The wide variability in the induction of AHH activity
in human placentas is exemplified by the data in Table 6-7--a range in
activity of the enzyme in smokers approaching 1,000-fold (a nearly
2,000-fold range if smokers are compared with nonsmokers). The basis
for this extreme range of responses to a chemical exposure (15-20
cigarettes/d for each subject) is not known. However, data presented
by Harris et al.70a suggests that pulmonary alveolar macrophages can
metabolize BaP to proximate and ultimate mutagens released into extra-
cellular space.
LUNG
-
The respiratory tract comprises an extremely disparate and complex
set of tissues containing some 40 different cell types.166 As
Devereux et al.47 have noted, whereas pulmonary cytochrome P-450 and
the metabolism of xenobiotics have been studied with various
preparations of lung tissue (microsomes, isolated perfused lung, cells
obtained by pulmonary Savage, direct instillation of xenobiotics in
various portions of the respiratory tract, etc.), little is known about
the localization of the cytochrome P-450 monooxygenase components in
the pulmonary system. This section deals exclusively with the
metabolic properties of human respiratory tissues with respect to PAH
metabolism, but the lack-of information just cited needs to be kept in
mind. There are facets of the investigation of Devereux et al.47 in
rabbits that probably bear significantly on problems of human pulmonary
tissue biotransfo'=ations that depend on cytochrome P-450; these
aspects include the observation that the alveolar macrophage that
accumulates PAH has little or no measurable cytochrome P-450 or
monooxygenase activity58~71~148 and that there is selective cellular
distribution of cytochrome P-450 species.
The ability of human bronchial epithelial cells to bind and
presumably to activate such PAHs as 7,12-DMBA, 3-MC, BaP, and
dibenz~ahianthracene was described by Harris and colleagues in
1974.7 Four tissue samples were studied (one control and three lung
cancer) in explant cultures, and radiolabeled PAHs were used;
radioactivity from all four compounds tested was found in both
cytoplasm and nuclei and in all tissue samples studied (see Table
6-8~. The number of tissues examined precluded comparisons between
normal and tumorous lung PAH metabolism
adduct identification were carried out, ~
activity from the labeled PAHs was found tightly
by CsC1 gradient. A more detailed study by this
tissues obtained from an additional four patients
pulmonary malignancy.
and no studies of PAM-DNA
althou~h, as noted, radio-
bound to DNA isolated
groupl95 used
, three of whom had
Explants of human bronchi also metabolized BaP and released
deriv,iives that are mitogenic in the Chinese hamster V-79 cell
line. The 7,8-diol of BaP was approximately 5 times more potent as
a promutagen than the parent PAH; binding of the diol to DNA was 5-20
6-19
OCR for page 282
times greater than that found with BaP. When 13 samples of bronchial
cells were studied with cloned Chinese hamster V-79-4A cells, a
positive correlation between ONA-PAH binding (in the cultured bronchial
cells) and induction of Or (ouabain-resistant) mutants was found, but
no correlation between this mutation frequency and AHH activity was
identified. This may be attributable, as the authors noted, to the
difficulty in correlating AHH activity with the consequences of the
multistep pathway of metabolic activation for BaP. The individual
variation in mutation frequency was 9-fold, and the variation in
binding of PAH to UNA 5-fold. This important investigation pointed the
way toward study of the metabolic activation of chemical carcinogens
into promutagens and mutagens directly in differentiated epithelial
cells derived from human tissues; and the human tissue-mediated mutagen
assay opened the possibility of testing the hypothesis that people
differ in mutagenic and oncogenic susceptibility to environmental
chemicals, depending on individual capacity to activate and deactivate
chemical procarcinogens. Autrup et al.12 compared the metabolism of
BaP by cultured tracheobronchial tissues from humans and four other
species (mice, hamsters, rats, and cows). They provided evidence that
the metabolism of BaP is qualitatively similar in tracheobronchial
tissues from humans and from animal species in which PAHs have been
shown experimentally to be carcinogenic.
A similar study limited to a comparison of human lung microsomal
fractions and rat microsomes was carried out by Prough et al.144 The
results indicated that human microsomes form a higher percentage of
dihydrodiol products from BaP than do rat microsomes. The wide
variation of PAH metabolize profiles formed by the 15 samples of human
lung studied may be due in part to differences in clinical diagnosis
when the samples were obtained. Bronchial tissues cultured in a
chemically defined medium were exposed to radiolabeled BaP or its
metabolizes, and their binding to TUNA was measured. Radiolabeled
metabolizes were prepared by incubating the parent PAH with rat liver
microsomes and then purifying and identifying with silica gel and
HPLC. The binding data showed that (-~-trans-7,8-diol bound to
bronchial mucosal DNA to a considerably greater degree (S- to 23-fold)
than did BaP; binding was also much greater (25- to 80-fold) than with
the (-~-trans-9,10-diol. The trans-7,8-diol constituted 3-6% of the
total identified metabolizes when human bronchi were exposed to BaP.
Diol-epoxides were formed from (-~-trans-7,8-diol in two of the
bronchial explants, and strong evidence was provided that the major
tumor bronchial mucosal DNA-binding BaP metabolize is in fact derived
from (-~-trans-7,8-diol.195 The specific adducts formed between DNA
and the metabolic intermediates of BaP were not isolated, but the
author concluded that the predominant bound metabolite is a single
enantiomer of diol-epoxide I derived as indicated above.
In an extension of their earlier work, Harris and colleagues69
examined the metabolism of BaP and 7,12-DMBA in explants of human
bronchus and made a metabolic comparison with human pancreatic duct
explants. As in the prior study, both normal and malignant human
bronchi (37 subjects) metabolized BaP actively and in generally similar
6-20
hi . .
OCR for page 283
fashion, except for a higher percentage of organic-solvent-extractable
metabolites formed by bronchi from noncancer patients. In addition,
prior exposure of the bronchial explants to benz~aJanthracene altered
the qualitative features of the metabolite profile of BaP, as analyzed
by HPLC. Benz~aJanthracene specifically increased the binding of BaP
to cellular DNA and the activity of AHH. Among a group of 28 of the
patients' tissues studied, 7,12-DMBA was bound to DNA more often (26 of
28) than BaP. In the comparison with pancreatic duct explants,
7,12-DMBA-ONA binding was consistently lower in the latter tissue than
in the bronchial explants.
Cohen et al.34 showed, with cultured human bronchial epithelium,
that BaP was converted promptly to metabolites that cochromatographed
with 9,10-dihydro-9,10-dihydroxy-BaP and 7,8-dihydro-7,8-dihydroxy-
BaP. Similar results were obtained with human lung cultures, except
that a major metabolite, benzota~pyrene-3-yi hydrogen sulfate, was
identified. The biologic activity of this sulfate ester of 3-hydroxy-
BaP is of interest, because, owing to its physicochemical properties,
it could be extremely persistent in man.
- Covalent adducts between DNA and BaP in treated cultured explants
of peripheral human lung tissue and in the continuous human alveolar
tumor cell line were identified by Shinohara and Cerutti.161 From
the chromatographic analysis of digests of ONA extracted from these
tissues, it was concluded that both the lung specimens and the human
alveolar tumor (A549) cells metabolized BaP to diastereomeric
7,8-dihydroxy-9,10-epoxytetrahydro-BaP intermediates that mostly
reacted with the exocyclic amino groups of deoxyguanosine to form
N'-~10-t 76 , 8a , 9a- and 96-trihydroxy-7,8,9,10-tetrahydro-
benzotaipyrene~yl~deoxyguanosine (dGua-BaP I and II). Although
comparable amounts of dGua-BaP I and II were formed in A549 cells,
dGua-BaP I was the predominant adduct in the DNA of lung specimens from
six different donors.
The wide range of metabolic capacities for PAHs exhibited by other
buman tissues studied also extends to lung tissue, as shown by Cohen et
_ .35 They observed a 44-fold variation in the ability of short-term
organ cultures of peripheral lung tissues from human cancer patients to
metabolize BaP to organic-solvent-soluble derivatives. The total
amounts metabolized ranged from 1: to 96.2t in a 24-h culture period.
The authors concluded that, although caution must be exercised in
measuring metabolic activities of human tissues derived from diseased
patients, the use of short-term organ explant cultures mimics the in
vivo metabolic disposition of PAH better than the use of lymphocyte AHH
activity would. A solution to the practical problem of obtaining lung
tissue from large populations to study the validity of this conclusion
is not apparent.
Kahng et al.78 concluded from a study of 11 immediately autopsied
subjects that bronchial tissue exposed to benz~aJanthracene produced
induction responses of AHH that correlated with induced AHH activity in
monocytes from the same subjects. A reconfirmation of the wide range
6-21
~;
~... .
OCR for page 338
194. Yamagiwa, K., and K. Ichikawa
195.
196.
197.
. Experimental study of the
pathogenesis of carcinoma. J. Cancer Res. 3:1-29, 1918.
Yang, S. K., H. V. Gelboin, B. F. Trump, H. Autrup, and C. C.
Harris. Metabolic activation of benzo~a~pyrene and binding to
ONA in cultured human bronchus. Cancer Res. 37:1210-1215,
1977.
Yoshida, D., H. Nishigata, and T. Matsumoto. Pyrolytic yields
of 2-amino-9H-pyredot2,3-b~indole and 3-amino-1-methyl-5H
pyridot4,3-b] indole as mutagens from proteins. Agr. Biol.
Chem. 43:1769-1770, 1979.
Yoshinaga, T., S. Sassa, and A. Kappas. A comparative
heme degradation by NADPH-cytochrome c reductase alone
study of
_ and by
the complete heme oxygenase system: Distinctive aspects of heme
degradation by NADPH-cytochrome c reductase. J. Biol. Chem.
257:7794-7802, 1982.
198. Yoshinaga, T., S. Sassa, and A. Kappas. Purification and
properties of bovine spleen heme oxygenate. Amino acid composi-
tion and sites of action of inhibitors of heme oxidation. J.
Biol. Chem. 257:7778-7785, 1982.
199. Yoshinaga, T., S. Sassa, and A. Kappas. The occurrence of
molecular interactions among NADPH-cytochrome c reductase, heme
oxygenate, and biliverdin reductase in heme degradation. J.
Bio1. Chem. 257:7786-7793, 1982.
200. Yoshinaga, T., S. Sassa, and A. Kappas. The oxidative degrada-
tion of heme c by the microsomal heme oxygenase system. J.
Biol. Chem. 257:7803-7807, 1982. -
lung in railroad workers. J.~.M.A. 171:2039-2043, 1959.
201. Zedeck, M. S. Polycyclic aromatic hydrocarbons : A review.
J. Environ. Pathol. Toxicol. 3: 537-567, 1980.
6-76
OCR for page 339
7
SOME FACTORS THAT AFFECT SUSCEPTIBILITY OF HUMANS TO
POLYCYCLIC AROMATIC HYDROCARBONS
The interaction of chemical pollutants, including the PAHs, and
mammalian cells can result in a variety of problems, including toxicity,
mutagenesis, carcinogenesis, and teratogenesis. This interaction of
chemicals with somatic cells probably results in such end points as
cancer, and the interaction of chemicals with germ cells probably results
in a variety of hereditary disorders. Many genetic disorders result in a
predilection to the development of cancer. The cancer burden in the male
population in the United States, although speculative, is distributed
approximately as follows: 40% from tobacco-smoking, 10-20% from all
diet-related causes, 5t from occupational exposures, it from single-gene
inheritance, and 35% from other causes, which may include unknown genetic
predisposition and environmental effects.59 The birth-defects burden in
the United States is distributed approximately as follows: 5-10% from
known teratogens, such as viruses, chemicals, and radiation; 25% from
genetic anomalies; and 60-65% from unknown mixtures of genetic predisposi-
tion and environmental effects.59 Although monogenic disorders (includ-
ing dominants), X-linked recessive disorders, and chromosomal abnor-
malities account for only about 5% of the human disease burden, the impact
of heterozygous recessively inherited abnormalities similar to the mono-
genic disorders is very ill-identified, but could outweigh all other
contributions.115
The heterozygous recessively inherited disorders may be the major
reason why cancer incidences are not uniformly distributed.95397 In
fact, of the millions of people exposed to such environmental chemicals as
diethylstilbestrol, estrogen oral contraceptives, vinyl chloride, and
cigarette smoke, only a very small proportion develop or express the
cancer thought to be associated with these exposures. It is likely that
genetic variability within the human population accounts in part for the
distribution pattern. As depicted in Figure 7-1, cancer sensitivity can
be viewed as a function of inborn susceptibility. Where this inborn or
genetic susceptibility is low, cancer expression is low. Where this
susceptibility is high (e.g., in single-gene defects), cancer expression
is high. The major question is whether the combination of chemical
exposure and genetic susceptibility can change significantly the numbers
of persons who develop cancer.
PAHs are ubiquitous chemicals capable of producing a broad spectrum
of biologic responses. Some can cause cancer in a variety of tissues,
including lung, liver, kidney, colon, skin, and bladder. In humans,
epidemiologic evidence has demonstrated that the incidences of cancers of
stomach, nasal cavity and sinuses, lung, and to a lesser extent rectum,
testis, skin (e.g., melanoma), brain, liver, pancreas, and hemopoietic
7-1
OCR for page 340
tissue i.e., leukemia) are correlated with area" captaining high con-
centrations of industrial pollutants. 10 ,46' 102,1 2, For many
persons, the amount of these agents in the environment may be the rate-
deterTnining factor for cancer susceptibility. Thus, the primary need
would be to identify and measure the amount o f exposure to the environ-
mental pollutants. The advent of a variety of in vitro and in vivo
bioassays promises the development of me thods for identi Eying chemicals
that are potential carcinogens.
In animal-model systems, susceptibility to chemically induced cancers
is usually dose-related. However, route, duration, and frequency of
administration and such genetic factors as species, sex, and strain all
tend to modify the relationship. In humans, mixtures containing PAHs can
certainly cause cancer, hut inadequacies in the information on age and
trauma but especially on duration, frequency, and intensity of exposure
and on the size and characteristics of the exposed population make
quantitative estimation of dose-response relationships and the concept of
thresholds difficult to interpret.
EFFECT OF GENETIC D IFFERENCE
The hypothetical stages in carcinogenesis are depicted in Figure
7-2. PAHs probably can show biologic effects at any of these stages.
Thus, answers are needed to the following questions: Which stages can
PAHs modify in humans? Are there naturally occurring variations in the
expression of some of these steps in humans? Can a genetic basis be
identified for the regulation of these naturally occurring differences?
If so, can the differences result from the action of a single gene
system? Can a relationship be shown between the express ion in the gene
locus and PAH-mediated effects?
PAH-induced effects in humans could depend on exposure, uptake, and
distribution of the chemicals; their metabolic activation and inactiva-
tion; DNA-repair capacity; "promoters"; and the extent of immunocompe-
tence. Each of these is discussed below.
UPTAKE AND DISTRIBUTION OF PAHs IN TISSUES
The distribution of PAHs in tissues or cells depends on the route of
exposure. According to the results of Rees et al.,1 the distribution
of benzota~pyrene (BaP) in tissue other than at the site of absorption
(i.e., intestine) depends on two phases: accumulation of the BaP on the
tissue and passive diffusion through the tissue. These two phases
underlie these authors' views about the apparent exponential nature of the
accumulation of BaP as a function of dose. The exponential increase could
be very important, but it must be pointed out that humans are rarely
exposed to BaP at concentrations greater than 200 AM (i.e., 50 legal)
under "normal" circumstances. Concentrations of a variety of PAHs (e.g.,
pyrene, anthracene, and BaP) in human tissues average about 1,100 parts
per trillion (ppt) in fat tissue and 380 ppt in liver.ll9 BaP can vary
7-2
OCR for page 341
from 0.3 No 15,000 parts per billion (ppb) in bronchial-carcinoma
tissue.18 Most of the subjects in the study of Tomingas et al.184
were cigarette-smokers, but no obvious correlation between BaP
concentration and extent of smoking was seen. The PAHs observed in
addition to BaP included fluoranthene, benzotbifluoranthene, and perylene.
Reasons for differences in tissue distribution are not known, but,
inasmuch as most of these chemicals are inducers and substrates for
microsomal enzymes, tissue variation in cytosol and nuclear receptors
could be important. In rodents, the induction of the microsomal mono-
oxygenase system by some PAHs depends on the presence of particular
cytosol receptor proteins.56314231 3 These receptor proteins are not
evenly distributed in all tissues, but are highest in thymus and lung,
lower in liver and kidney, lower yet in testes, brain, and skeletal
muscle, and not detectable in pancreas, adrenal, or prostate.l5 Most
importantly, receptor proteins are found in high concentrations in strains
of animals or cultured cells in which PAHs induce the enzyme aryl
hydrocarbon hydroxylase (AHH) and are nondetectable in those in which AHH
is nonrespons ive . 56 , 142 This correlation also extends to humans, in
whom the concentration of a BaP-binding plasma component is correlated
with the capacity of lymphocytes to be induced for AHH activity in
culture.104 A cytoplasmic receptor for BaP, which did not cross-react
with 7,12-dimethy31benz~ajanthracene, has also been reported for human
cells in culture. 1 The presence of some of these receptors is under
specific genetic control in animal models,l22~143, so uptake and
distribution, at least in particular persons, could be under a form of
genetic control.
METABOLISM
PAHs are metabolized in a variety of ways, with the microsomal mono-
oxygenases (e.g., AHH) probably most important. Steady-state activities
of these enzymes vary Animals and are linked to susceptibility to some
PAH-mediated cancers.7 ~ 1 In humans, the data are much less clear.
Table 7-1 summarizes the studies that suggest a correlation between high
AHH inducibility (and usually high induced-AHH activity) and cancer
susceptibility, and Table 7-2 summarizes the studies that suggest the
converse.
Reasons for the contradictory results probably lie in methodologic
variations, such as the use of different cell types, different assays,
and different assay conditions. The most easily accessible and therefore
commonly used Lyman tissue is She peripheral blood lymphocytes. Nutri-
tional state,1 drug intake,8 age,35 and disease state74
influence the capacity of the lymphocytes to respond to mitogen. These
influences have not been assessed in determining their relationship to the
AHH activity observed in cultured lymphocytes. Variations in AHH activ-
ity in lymphocytes have been observed to occur seasonally in some geo-
hi 1 ations 128$129,154 but whether they result from in vivo or in
vitro factors is not known.
7-3
~ . it. . ..
OCR for page 342
A variety of in vitro conditions are known to influence AHH
activity. The initial concentration of lymphocytes affects the time
course and amount of control and induced AHH activity.7 The type and
lot of serum Supplement infuence the control and induced AHH
activity.5 ,7 In fact, some lots of feta)-galf serum are capable of
causing mitogen activation of lymphocytes. O The numbers of cultured T
cells may affect the AHH activity observed.67 In studies using cultured
human tissue, two important aspects are the question of the variable
degree of AHH activity in different cell types202 and the question of
large variations between and within individuals in both AHH activity and
microsome-mediated BaP-DNA adduct formation. ' 3 Blood monocytes and
pulmonary alveolar macrophages are examples of other human cell types
whose AHH activity is Correlated with that in lymphocytes or
cultured human tissue, but there are problems of accessibility with
each of these cell types.
If the cell samples are cultured and assayed on the same days,
the v5r~2t~30 seems to be ac3eptable.5°379 Culturing lympho-
cytes ~ ~ or monocytesl2 from fraternal or identical twins at the
same time has shown that AHH activity is under a degree of genetic
control, and the numbers of genes in question are probably small. Thus,
the genetic component most likely results in a unimodal frequency
distribution that is skewed in the populations of individuals toward those
with higher AHH activity,7831ll rather than the trimodal distribution
originally reported.72
To circumvent many of the in vitro problems, the use of cryopre-
served tissue may be an alternative, in that lymphocytes can be cryo-
preserved before mitogen activation and still have ted capacity to be
mitogen-activated and then assayed for AHH activity. The relative AHH
activities among the lymphocyte samples from different individuals are
similar, whether the assays are conducted on freshly cultured lymphocytes
or after cryopreservation.82 Cryopreservation allows the culture and
assay at the same time of cells from different organisms collected in
diverse geographic locations and over extended periods.
The use of cryopreserved lymphocytes 3 control of some basic culture
variables--such as initial lymphocyte concentration (1.0 x 106 cells/ml)
and lot and type of serum supplement (e.g., human AB serum)--and assaying
AHH activity at two times to ensure detection of peak activity can yield
the data presented in Figure 7-3. Data were taken on a group of 51 per-
sons who were on hospital diets for at least 2 d before phlebotomy, who
were not on any medication, and who were eventually followed for complete
clinical diagnosis. Viability of cells was measured by assay for the
NADH-dependent cytochrome by reductase (using cytochrome c as a sub-
strate) activity (Cyt c). Carcinogen-metabo~i~gg activity is presented
in terms of units of AHH per unit of Cyt c.7 ~ The degree of mitogen
activation was also measured. Data analyses showed that:
~ Cryopreserved lymphocytes from over 957 of the normal and cancer
patients were mitogen-activated.
7-4
OCR for page 343
· Lymphocytes from lung-cancer patients were mitogen-activated as
efficiently as lymphocytes from noncancer patients (actually better;
= 0.001).
O The 14 highest AHH activities were found in patients with lung
cancer, with the mean in the 21 lung-cancer patients (0.89 unit AHH/unit
Cyt _) being significantly higher than that in the 30 non-lung-cancer
patients (0.47 unit AHH/unit Cyt c).
The higher AHH activities were not directly related to higher degrees
of blastogenesis and were not related to cigarette-smoking history, tumor
type, tumor location, or family history of cancer. Whether high AHH
activity is the cause or the result of lung cancer cannot yet be answered.
In animal-model systems, some PAHs cause tumors of the lymphoreticu-
lar system, and a genetic association for this activity at the Ah locus
has been suggested.29~1l3 Although this is only presumptive, there may
be a similar relationship in human leukemia patients who were recently
shown to express lower AHH activity (as in animal-model systems);ll in
other studies, the first-degree relatives of leukemia patients expressed
normal AHH activity.90
The results of these studies are interesting and certainly need to be
confirmed and extended. The extended studies should be multifaceted; that
is, they should simultaneously measure more than one enzymatic end point.
Perhaps an appropriate group of assays would include an assay for AHH, as
described in the literature; an assay for all B3aP ~§abolites via HPLC; an
assay for particulate P-450s via immunoassays;1 1, and an assay for
mRNA expression of the P-450 genes with cloned UNA fragments containing
the P-450 genes.116 Human tissues should be used where possible. There
is probably a degree of genetic control of AHH activity in the human
population, and this enzyme may play a role in determining susceptibility
to PAH-mediated cancer and other diseases.
DNA BINDING, DAMAGE, AND REPAIR
Many PAHs are converted by the microsomal monooxygenases to forms
that bind covalently to a variety of cellular macromolecules, including
nucleic acids (see Chapter 5 and Phillips and Simsl40~. Evidence of the
importance of DNA binding is exemplified by the observation that varia-
tions in DNA-repair capacity seem to play a major role in determining the
toxic, mutagenic, and3 t :00s f49ming ac t ivit ie s o f many chemical care i 0
In animal-model systems, the amount of PAH metabolism is determined
by the activity of the microsomal monooxygenases, and variations in these
enzymes result in concomitant changes in the binding of chemicals to
DNA.112 In cultured human tissue, hydrocarbon-DNA binding also occurs
as the result of microsomal monooxygenase-mediated metabolism,6~55 and
variations in metabolic activity are asso5:iated with concomitant varia-
tions in binding of hydrocarbons to DNA. The major DNA adduct often
results from the interaction of specific metabolites of PAHs (diol-
epoxides) and the N7 of deoxyguanosine.l24,l35 Other products a
7-5
~;
~` ..
OCR for page 344
found, including interactions with the N4 deoxyadenosine,177 the back-
bone phospt~tl,70f ONA, 1 and the exocyclic amino group of deoxy-
adenosine. ~ The latter may be important, because its formation
from various PAH-like chemicals closely parallels their carcinogenic
potencies on mouse skin.27 No apprec~,b{~ syggificity of binding with
respect to base sequence is apparent, ~ 8, but binding may be
influenced by chromatic structure, with a greater extent of binding
associated wi th internuc leosomal regions.6 ,76 '
A potentially important anomaly is that, although in vitro metabolism
of BaP to forms that bind to DNA parallels the AHH activity of the micro-
somal preparations and the Genetic background of mice used to generate
these microsomal samples,13 the in viva results from strains of mice
that differ widely in ASH activity so that there is very little
strain variation in BaP-DNA binding. ~
Probably more crucial to carcinogenicity is the geometry of the
binding in relation to later excision repair by endonucleases.48 The
binding of different residues and different chemical groups within
residues dramatically affects excisibility. These chemical-DNA adducts
are either repaired, not repaired, or misrepaired (see Figures 7-2 and
7-4). The fate of these adducts determines whether a cell remains normal,
mutates, or dies.
Repair capacity can be separated into two major types--excision
repair and postreplication repair.l49 Excision repair is the in situ
removal and replacement of chemically modified ONA so that the original
DNA sequence is re-established. For a variety of reasons, excision-
repair systems usually do not remove all the modified bases; so the ONA
very often replicates, even though some unexcised damage may be present.
This replicated UNA usually has gaps in the newly synthesized strand
opposite the DNA adduct. The gaps are fillip in by postreplication
repair--also termed "recombination repair." O Figure 7-3 depicts how
these two processes of repair contribute to the cells' survival of the
damaging effects of chemicals like PAHs. A combination of both methods is
involved in the repair of hydrocarbon-bound DNA.1OO
A large number of both constitutive and inducible enzymes are
involved in this ONA-repair process.47 The exact role of these enzymes
is not known, but it seems that rather small changes in any of the enzyme
activities can have great effects on the repair process and eventual bio-
logic expression of the DNA adducts. Moreover, it has been recently shown
in prokaryotes that the DNA adduct itself is not likely to be mutagenic,
but rather that the mutagenic event is induced by the action of the
DNA-repair enzymes themselves.]
Natural variations in DNA-repair capacity occur in humans. These
variations are exemplified by the existence of genetic diseases that are
associated with defects in DNA repair. Table 7-3 presents a list of such
diseases, their modes of inheritance, the specific tumors associated with
them, and their proposed DNA-repair defect. These genetic diseases are
7-6
Am .. .. ..
OCR for page 345
associated with a high incidence of malignancy, compared with the
incidence in the general population, and often a specific malignancy is
involved (see Table 7-3~. The incidences for persons who are genetically
homozygous for xeroderma pigmentosum, ataxia telangiectasia, and Fanconi's
anemia are about 10-5, and for those who are heterozygous, about
10-2. Those who are heterozygous for ataxia telangiectasia and are less
than 45 yr old have a fivefold increase in the risk of cancer,180 and
those heterozygous for Fanconi's anemia may account for 5% of all leukemia
deaths (approximately a fivefold increase in susceptibility).179
Because these people are deficient in the ability to repair radiation-
induced DNA damage and chemical-induced DNA damage, 169 it has been
suggested that alteration in DNA-re~air capacity may put them at greater
risk of chemically induced cancers. 68 It must be pointed out that many
of these diseases, especially ataxia telangiectasia, are also associated
with abnormalities of the immune system. Thus, genetic disease may result
in higher risks of cancer via deficiencies in DNA-repair capacity or
immunocompetence. Among the normal population of humans, there are
probably subtle variations in DNA-repair capacity, but whether these
variations are genetically controlled or are related to cancer risk
remains to be determined.
PROMOTION AND COCARCINOGENESIS
Many s tudies have shown that a number of modifying fac tars can
increase the effect of low-dose or low-potency carcinogens that by
themselves would be insufficient to induce malignancies.l09~200 Many
PAHs are complete carcinogens; that is, they have both initiating and
promoting activities. Others--such as pyrene, benzote~pyrene,
fluoranthene, and benzotghi~perylene--are weak complete carcinogens and
weak cocarcinogens.l9l'192 It is difficult to determine what role PAHs
might have in tumor promotion in humans, because there are no good methods
for measuring this activity in the human population. Such end points as
induction of ornithine decarboxylase activity,13 phosDholipid
synthesis,l59~178 inflammation,~73 protease activit ,~87 cellular
proliferation,57 decrease in differentiated states,;6~201 and
formation of "dark cells''l48~172 are manifestations of man romoters,
and many PAHs can induce at least some of these changes.l95~500 But no
single end point correlates with the promoting activity of all the
different chemicals that have promoting activity.
In animal systems, there seems to be a genetic basis for promota-
bility, in that different strains of mice express different suscepti-
bility to promotion during the standard two-stage carcinogenesis assay.
Such strains as CD-1 and BALB/c are relatively resistant, whereas the
specifically derived SENCAR strain (i.e., sensitive to carcinogenesis) is
very sensitive to promotion of skin cancer~78~58 The molecular basis of
this difference has not been defined, but recent informal ion suggests that
the skin itself has the sensitivity, inasmuch as skin from SENCAR mice
remains sensitive to promotion even after grafting to BALB/c mice.203
7-7
OCR for page 346
No genetic variation in promotability in humans has been described.
However, the fact that pyrenes may have promoting and cocarcinogenic
activity, the possibility that such activity plays a major role in cancer
formation in humans, and the absence of effective end points in the human
population all suggest that much more work is necessary before the role of
PAHs in promotion can be understood.
I~4UNOCOMPETENCE
_
Substantial interest has centered on the role of the immmune system
in preventing the expression of malignancy by recognition and destruc-
tion of newly formed malignant cells. The concept of "immunosurveil-
lance," however, has not been well supported, and, in fact, "stimula-
tion" of malignant cells may even occur.75
Immunodeficient persons do have a greatly increased risk of develop-
ing a malignancy of the lymphoreticular system.17~60~93~126 The exact
mechanism responsible for the increase, however, is not clear.
A number of genetic disorders in humans are associated with immuno-
deficiencies. These disorders include ataxia telangiectasia, Wiskott-
Aldrich syndrome, Bloom's syndrome, common variable immunodeficiency,
selective IgA deficiency, Bruton's agammaglobulinemia, severe combined
immunodeficiency, selective IgM deficiency, and immunodeficiency with
normal or increased immunoglobulins.73
These immunodeficient genetic disorders are usually heterogeneously
linked with a variety of other distinct underlying defects. For example,
persons with ataxia telangiectasia and Bloom's syndrome have severely
impaired DNA-repair capacities,169~195 and those with severe combined
immunodeficiency also have adenosine deaminase deficiency.73 Therefore,
it is difficult to determine the reasons for the increased cancer
susceptibility of these persons. Epidemiologic evidence fails to support
the idea that immunosurveillance mechanisms are generally involved in
carcinogenesis 5 but does provide clues to immunologic processes that may
predispose to particular neoplasms.38
In animal-model systems, PAHs can cause tumors of the lympho-
reticular system, and association with the Ah locus has been
suggested.29~113 In humans, exposure to some hydrocarbons, such as
benzene, has been repeatedly associated with leukemia. Whether variations
@ ~
Ln 1mmunocompetence occur naturally in the normal population and whether
PAHs, as a group of environmental contaminants, pose a special risk to
persons with such variations are not known.
STAGE OF DEVELOPMENT
.
Some cell types undergo periods of heightened sensitivity to
chemicals during their normal growth cycles. For example, in animal-
model systems there are striking differences between geru-cell stages in
OCR for page 347
the chemical induction of dominant lethals, translocations, and specific-
locus mutations. l4, 160 Moreover, the fetus is at greater risk than
the mother, owing to high doses of environmental chemicals; the
permeability of the blood-brain barrier is greater, and liver-enzyme
conjugating function is poorer.61 The greater the lipid solubility of
a chemical, the greater its placental transfer; and the placenta is
readily permeable to chemicals with molecular weights less than 600.
Most PAHs fit into these categories, and in animal-model systems such
PAHs as BaP, 3-methylcholanthrene, and 7,12-dimethylbenz~aJanthracene
cause oocyte and follicle destruction and embryo lethality and
resorption and have a greater incidence of malformation and even cancer
in surviving embryos.88~103~171~190
In humans, gross congenital abnormalities occur in some 2t of all
infants and are the cause of about 15t of the deaths of infants less
than a year old. Exposure to such agents as viruses, mercury, DUT, CO,
and polybrominated biphenyls probably accounts for 5-10t of the birth
defects; genetic abnormalities cause 25%; and the causes of the
remainder are largely unknown.61 Interactions in the intrauterine
environment between genetic predisposition and chemical and biologic
factors are probably responsible for these birth defects. Although
occupational exposure of human males52 and both paren-ts204 to PAHs
was not associated with increased cancer incidences in the offspring,
recent work has suggested that a combination of chemical exposures of
both parents (especially the mother) resulted in higher incidences of
brain tumors in the offspring. 137 Maternal cigarette-smoking is
associated with decreased birthweight, increased perinatal morbidity and
mortality, and other harmful effects on the newborn.l41 The PAHs in
cigarette smoke may account for some of its biologic activity, inasmuch
as a relationship has been shown between cigarette-smoking, induction of
AHH activity in human placental tissue,ll4'l98 and a decrease in
placental size;135 PAHs are the major class of AHH inducers found in
cigarette smoke, 83 and thus it is important to note that BaP, which is
in cigarette smoke, can cross the placental barrier.92
Because PAHs must be metabolized before they produce a biologic
effect, the impact of PAHs on maternal and fetal tissues can be quite
complex. Some examples of these complexities are differences in
developmental patterns of specific enzymes, the relative importance of
maternal and fetal metabolism, the role of metabolism in placental
tissue, the relative importance of hepatic and extrahepatic metabolism,
and sex differences in developmental patterns. The induced and control
forms of AHH and acetanilide 4-hydroxylase are temporally regulated both
before and after the birth of animals.68 The deactivation of
conjugating enzymes (e.g., UBP-glucuronyltransferase, sulfotrans-
ferase, and N-acetyltransferase) is also temporally regulated both
before and after birth, but this regulation can be quite different from
that of AHH.94 The relation between activation and inactivation can
be influenced by the sex of animals.70 Shum et al.171 showed that
both the fetal and maternal enzymes play an active role in determining
the ultimate fetal toxicity of BaP. Using specific crosses between
AHH-responsive and AHH-nonresponsive strains, these authors could show
7-9
OCR for page 348
that when the mother was nonresponsive the enzyme capacity of the fetal
tissue determined the toxicity of BaP, but that when the mother was
AHH-responsive there was no difference in fetal toxicity between
nonresponsive and responsive fetuses. Mice seem to have AHH activity as
early as about 7.5-8.5 d of gestation.40 This activity slightly
increases before birth, but increases greatly in the first few days
after birth94 and then slowly decreases as the mouse ages.68 It
should be pointed out that in vivo exposure to BaP, in addition to
inducing higher AHH activity in mouse fetal tissue, can suppress bumoral
immunity in animals that survive and can cause about a 10-fold increase
in the incidence of various tumors in surviving animals.l90 It seems
likely that, in rodents (and perhaps in bumans), PAHs can be taken up
and distributed through the placenta intact or in the form of
metabolites, that the metabolites themselves can cause fetal toxicity or
the delayed effects of immune suppression or cancer, and that intact
PAHs can cause fetal enzyme induction, metabolism, and the sequelae
mentioned earlier.
MODIFYING FACTORS
-
A variety of environmental factors can mitigate or exacerbate the
inherent sensitivity of mammalian tissues to PAHs. These factors are
probably at least as important as some of the genetically controlled
differences discussed earlier and tend to make genetic differences less
distinct. Two factors known to modify PAH carcinogenesis, at least in
animal-model systems, are the physical state of the PAH and the
nutritional state of the exposed organism.
PHYSICAL STATE OF PAH
.
The sources and the formation of PAHs in the environment are dis-
cussed in Chapters 1-3. Most of them are found as mixtures and many are
found in association with particles, such as cigarette-smoke
particles, i74 fossil-fuel combustion products, 9 coal flyash, 77
and asbestos fibers.86~105 This association can be important, because
PAHs in the presence of or adsorbed on particles are transported through
membranes more e f f ic iently,75 are cleared from tissue more
slowly, 25 and have a different tissue distribution--that determined
by the particle size, rather than by the size of the free PAH.147 The
increased uptake results in more efficient induction of AHH activity at
low PAH concentrations.l°5 Those exposed to particles containing PAHs
are probably at greater risk of various cancers.166 Uptake,
distribution, and metabolism of PAHs can be so altered by particles that
those who normally would be unaffected by the PAHs may be adversely
affected.
NUTRITIONAL STATE OF HOST
Nutritional status can substantially modify the toxicity of some
environmental pollutants. For example, specific dietary
7-10
~... . .
Representative terms from entire chapter:
aromatic hydrocarbons
