Environmental Mutagenesis 6593-616 (1984)
Role of Oxygen Radicals in Tumor
Promotion
Thomas W. Kensler and Michael A. Trush
Division of Experimental Pathology and Toxicology, Department of Enviromental Health
Sciences, Johns Hopkins University School of Hygiene and Public Health, Baltimore
Tumor promoters provoke the elaboration of oxygen radicals by direct chemical
generation and through the indirect activation or alteration of cellular sources
including membrane oxidases, peroxisomes, and electron transport chains in
mitochondria and endoplasmic reticulum. Although direct measurement of amplified oxygen radical production in response to tumor promoters in target tissues
remains problematic, studies with scavengers of reactive oxygen species demonstrate inhibition of biochemical and biological sequelae of tumor promoter exposure and provide strong presumptive evidence for oxygen radical involvement in
this late stage of carcinogenesis. The critical macromolecular targets for these
oxygen radicals remain undefined; however, they may include lipids, DNA, DNA
repair systems, and other enzymes.
Key words: multistage carcinogenesis, tumor promotion, phorbol esters, reactive oxygen, free
radicals, superoxide dismutase, antioxidants, DNA damage, lipid peroxidation
INTRODUCTION
Respiring cells produce free radicals from molecular oxygen through enzymatic
and nonenzymatic reactions. The univalent pathway of oxygen reduction generates,
in turn, the superoxide radical (OF), hydrogen peroxide (H202),the hydroxyl radical
('OH), and finally water. The intermediates of oxygen reduction to water are highly
reactive and present a challenge to the morphological and functional integrity of cells.
As either an oxidant or reductant 02 can, directly or indirectly as a precursor for
more potent radicals, modify a variety of biologically important molecules. For
example, fluxes of 02 generated enzymatically, photochemically, or radiochemically
have been shown to peroxidize lipids, depolymerize polysaccharides, alter enzyme
activity, cleave DNA, and lull cells [Bors et al, 1980; Fridovich, 19811. While it has
long been recognized that oxygen has a primary role in radiation toxicity and that
radicals and peroxides derived from molecular oxygen participate in radiation damage
Received February 22, 1984; revised and accepted April 26, 1984.
Address reprint requests to Dr. Thomas W. Kensler, Division of Experimental Pathology and Toxicology,
Department of Environmental Health Sciences, Johns Hopkins University School of Hygiene and Public
Health, Baltimore, MD 21205.
O 1984 Alan R. Liss, Inc.
594
Kensler and Trush
to cells, it has only recently been appreciated that these reactive oxygen species are
also important mediators of toxicity for a wide spectrum of chemical agents [Bus and
Gibson, 1979; Holtzman, 1981; Trush et al, 1982; Hochstein, 19831. The cumulative
effect of oxygen radical-initiated reactions may be cell death, possibly resulting in
tissue necrosis or fibrosis, or may be more subtle and delayed as evidenced by the
development of neoplasms.
Epidemiological and experimental evidence has established that the induction of
cancer by chemicals is represented by stages of cellular evolution from normal,
through preneoplastic and premalignant cells, to highly malignant neoplasia. In the
best-studied model system, mouse skin, three distinct stages can be defined: initiation,
promotion, and progression. The initiation stage requires only a single application of
a direct or an indirect carcinogen at a subthreshold dose and is essentially irreversible
whereas the second stage, promotion, which follows initiation, requires repetitive
treatments and, although initially reversible, later becomes irreversible. Progression
entails a second round of exposure to initiating agents and allows for the selection of
autonomous, malignant cells. The concept of multistage carcinogenesis has recently
been extended to other species and tissues. For example, multistep carcinogenesis
systems have been shown in many epithelial tissues including liver, lung, bladder,
mammary gland, pancreas, esophagus, stomach, and colon, as well as cells in culture.
Such findings provide strong presumptive evidence for the role of multistage processes in the etiology of human cancer.
Tumor promotion is a phenomenon of gene activation in which the latent altered
genotype of the initiated cell becomes expressed through selection and clonal expansion. Thus, although tumor promoters are considered to be epigenetic in action, their
effects are ultimately on the genome. Tumor promoters are not carcinogenic by
themselves, but must work in sequence with initiators. The work of Rous and Kidd
[1941], Berenblum [1941], and Mottram [I9441 provided the early evidence for
promotion. These studies were based primarily on studies on mouse skin using croton
oil or its later identified active principal, 12-0-tetradecanoylphorbol-13-acetate
(TPA),
as the model promoter. Boutwell [I9641 and later Slaga et a1 [1980a] have shown that
promotion itself could be divided into two steps, conversion (I) and propagation (11).
After initiation, the conversion step is accomplished by a limited number of TPA
treatments, which with no further treatment produce only a few tumors. The propagation step is accomplished by repeated treatment with turpentine or mezerein,
compounds that are ineffective as complete promoters. Selective inhibition of these
two stages of promotion with various antipromoters suggests that stage I reflects a
dedifferentiation to the embryonic state as manifested by the induction of dark basal
keratinocytes while stage I1 reflects specific gene activation and cell proliferation
[Slaga et al, 1980b, 19821.
The molecular mechanisms of action of tumor promoters are not fully understood and the bulk of the available information pertains to studies of the tumor
promoting activity of phorbol esters in mouse skin. In general though, most tumor
promoters stimulate cell proliferation. Several recent excellent reviews have dealt
with this subject [Diamond et al, 1980; Blumberg, 1981; Hicks, 19831. Phorbol esters
act early and directly on cell membranes, presumably through their interaction with
high-affinity, saturable receptors. This receptor appears to be a calcium and lipid
binding protein kinase C that is activated upon binding by TPA. Phosphorylation of
proteins by this kinase may in turn modulate many cell regulatory functions including
.
Oxygen Radicals in Tumor Promotion
595
gene activation. Within minutes following exposure of cells to TPA increases in
calcium ion flux, cyclic nucleotide and phospholipid metabolism, amino acid and
glucose transport, and activation of the arachidonate cascade are observed. Sequential
activation of RNA. protein, and DNA synthesis occurs later. The importance of any
of these specific biochemical events in the tumor promotion process is not clearly
defined.
Although reactive oxygen species appear to have a role in multiple steps in
chemical carcinogenesis, a rapidly expanding literature presents a particularly compelling argument for the involvement of oxy radicals in tumor promotion and is the
subject of this review. The reader is also refered to other commentaries on free
radicals in carcinogenesis [Ts'o et al. 1977; Floyd, 1982; Pryor, 1982; McBrien and
Slater, 1982: Ames, 1983; Copeland. 1983; Kensler and Trush, 19841. Figure 1
provides a general scheme for the elaboration of oxy radicals by tunlor promoters and
the subsequent cascade of aggressive oxygen species that can interact with tjiomolecules and contribute to the development of a neoplastic state. As is the case with
mechanistic studies on tumor promotion in general, most experimental observations
linking oxygen-free radicals to the tumor promotion process are derived from studies
on the action of agents active in the skin, particularly the phorbol esters. However.
additional presumptive evidence suggests that these oxygen species might also be
involved in tumor promotion induced at nonepidermal organ sites. Table I summarizes
the classes of tumor promoters that may evoke, by direct or indirect mechanisms, the
elaboration of oxy radicals and hence a cellular pro-oxidant state.
STUDIES WlTH EPIDERMAL TUMOR PROMOTERS
Tumor Promotion by Free Radical Generating Compounds
A number of free-radical generating peroxides used in the chemical and pharmaceutical industries are skin irritants and tumor promoters. Benzoyl peroxide, lauryl
TUMOR PROMOTERS
Chemlcal
Ir
Act!vat~onof
REACTIVE OXYGEN SPECIES
t
INTERACTIONS WlTH BIOMOLECULES
LIOIP
~ erox~dal~on
DNA
DamageIRepa~r
Enzyme A r t ~ v a t o n
lnact\vat~on
Fig. I . General scheme for the elaboration of reactive oxygen by tumor promoters and the accompanying enzymatic and non-enzymatic pathways leading to more aggressive oxygen species that in turn
may m o d i ~critical molecular targets and participate in the development of a neoplastic state.
596
Kensler and Trush
TABLE I. Classes nf Tumor Promoters That May Act Through
an Oxidant hlechanism
U V radiation
Reactivc oxygen-pencrating con~pounds: Peroxides, anthralin.
gossypol. complete carcinogens that autooxidire
Membrane active agents: phorbol c<ters. indolc alkaloids
Peroxisome proliferators: hypolipiderrlie drugs. phthalate esters
Mixed function oxidarc inducers: TCDD. DDT, phenobarbital. polychlorinated hiphenyls, a-hexachlorocyclohcxane
Hyperoxia
peroxide. decanoyl peroxide, cumene hydroperoxide, perbenzoic acid, m-chloroperbenzoic acid. and p-nitroperoxybenzoic acid are very active in the mouse epidermis
system. while. tert-butyl hydroperoxide and methylethylketone peroxide are weak
promoters and H 2 0 2 is only marginally active (Bock et al, 1975; Slaga et al,
1981,1983; Klein-Szanto and Slaga, 19821. Interestingly, however, when H202 is
examined in the multistage promotion scheme, it is found to be an effective stage I
promoter when followed by phorbol ester during stage I1 promotion. Both benzoyl
peroxide and lauroyl peroxide induce epidermal hyperplasia and the production of
dark basal keratinocytes, a phenotypic marker of stage I promotion, in approximately
10% of the basal cell population during the first week after single topical application.
H202-induced epidermal hyperplasias also exhibited numerous dark cells, but their
presence was less sustained (Klein-Szanto and Slaga, 19821. Although active as tumor
promoters. these peroxides are largely inactive as either initiators or complete carcinogens [Kotin and Falk, 1963: Van Duuren et al, 1963; Slaga et al, 19811.
The antipsoriatic agent anthralin is another mouse skin tumor promoter [Bock
and Burns. 16631 that can directly produce free radicals. Anthralin may be readily
oxidized by air and light to form intermediate hydroperoxyl and anthralin radicals
[Ashton et al. 19831. The hydroperoxyl radical can then react with basic groups to
form 0 2 . The anthralin free radical undergoes further oxidation to anthraquinone
and anthralin dimer. Neither of the end products of anthralin oxidation, anthraquinone
nor dimer. have tumor promoting activity [Segal et al. 19711. Interestingly, anthralin
can also inhibit tumor promotion when applied before application of TPA [Boutwell
et al, 19811.
Gossypol, the toxic principle of cottonseed oil, is a tumor promoter in the skin
of mice initiated with 7.12-dimethylbenzlalanthracene[Haroz and Thomasson, 19801.
Gossypol occurs in three tautomeric forms, hemiacetal. phenol quinoid, and aldehyde,
which are reactive with oxygen (Berardi and Goldblatt, 19691. This compound is
presently being evaluated for antifertility activity [Zatuchni and Osborn. 19811 and
Troll and coworkers [Coburn et al. 19801 have suggested that the spermicidal action
of gossypol results from the production of oxygen metabolites inasmuch as SOD and
catalase inhibit this activity. De Peyster et al [I9841 have observed that gossypol
enhances the formation of oxygen radicals wher, incubated with rat liver microsomes
and human sperm. Thus, it is possible that the promoting activity of gossypol relates
to its oxygen radical generating action.
Several polycyclic aromatic hydrocarbons including benzo(a)pyrene are carcinogenic to the skin [Slaga et al, 19781. Benzo(e)pyrene is a skin promoter [Slaga et al,
Oxygen Radicals in Tumor Promotion
597
19791. A major pathway of activation of benzo(a)pyrene entails sequential metabolism
to 7,s-diol-9,lO-epoxy benzo(a)pyrene. the putative ultimate metabolite. However.
this metabolite lacks complete carcinogen activity in mouse skin although it is an
initiator in the multistage model [Slaga and Fischer. 19831. Other metabolic pathways
also generate reactive benzo(a)pyrene intermediates and these may contribute to the
"promoter" action of benzo(a)pyrene. One electron oxidation at the C-6 position
results in the formation of a radical cation which can undergo nucleophilic attack by
water to yield the labile 6-hydroxy benzo(a)pyrene derivative, which in turn rearranges to the 6-oxy radical [Ts'o et al, 19771. Interaction of molecular oxygen with
this radical yields 1,6-, 3,6-, and 6,12-quinone metabolites of benzo(a)pyrene [Lesko
et al, 1975; Lorentzen et al. 19751. In the presence of molecular oxygen the quinones
undergo oxidation-reduction cycling that is accompanied initially by the generation of
02 and subsequently by H 2 0 2 and 'OH [Lesko et al, 1975; Lorentzen and Ts'o.
19771. Accordingly, Lorentzen et al 119791 have shown that the toxicity of
benzo(a)pyrene quinones in Syrian hamster embryo fibroblasts is oxygen dependent.
In further support of this concept. we find that a biomimetic superoxide dismutase
(SOD) inhibits the oxidative metabolism of 7.8-diol-benzo(a)pyrene [Trush et al,
19841. Biomimetic SOD also inhibits both the initiating [Solanki et al, 19841 and
complete carcinogenic action [Egner PA and Kensler TW, unpublished observations)
of 7.12-dimethylbenz[a]anthracene in mouse skin. Other carcinogens with diverse
target specificities, such as benzene, anthracyclines. diethylstilbestrol, bleomycin.
and aromatic amines, can also produce oxy radicals that may be important to their
respective "promoter" activities [Greenlee et al, 1981; Doroshow. 1983; Metzler,
1981: Borek and Troll, 1983; Nakayama et al, 1983; Stier et al, 1980; Kensler and
Trush, 19841.
Phorbol Ester-Mediated Oxy Radical Generation
Specific interaction of either soluble or particulate stimuli with membranes of
inflammatory cells such as leukocytes and macrophages causes these cells to consume
and metabolize oxygen to reactive intermediates. Concentrations of TPA as low as 5
nglml provoke rapid and marked changes in oxidative metabolism of polymorphonuclear leukocytes (PMNs) including increased oxygen consunlption [Repine et al,
19741, increased oxidation of glucose via the hexose monophosphate shunt pathway
[Briggs et al. 19751; increased generation of 02 and H 2 0 2 and chemiluminescence
[Goldstein et al, 1975; DeChatelet et al, 1976; Kensler and Trush. 19811. These
processes appear to be mediated through the interaction of the phorbol ester and a
specific membrane receptor (Goodwin and Weinberg, 19821 and the subsequent
activation of a pyridine nucleotide-dependent oxidase system localized to the plasma
membrane [Briggs et al, 19751. Structure-activity studies comparing phorbol ester
analogs of varying tumor promoter activity with their ability to act as stimulators of
active oxygen metabolism. as monitored by 0 2 production and chemiluminescence,
have shown strong concordance between the two processes [Goldstein et al, 1981;
Kensler and Trush. 19811. Phorbol and phorbol triacetate, which are inactive as tumor
promoters are also inactive as stimulators of active oxygen production. Phorbol
diacetate. phorbolol, phorbol dibutyrate, phorbol dibenzoate, and phorbol didacanoate. which range from weak to potent tumor promoters, show a comparable range
of activities in the PMN system. TPA is the most active phorbol ester examined.
Several nonphorbol-promoting agents, iodoacetic acid, Tween 60, and anthralin. and
598
Kensler and Trush
the intlammatory but nonpromoting agent. ethylphenylpropiolate, were without stimulator~activity. Chemiluminescence is also stimulated by the calcium ionophore
A23187 [Wilson et al, 19781, a stage-I promoter [Slaga et al. 19831. Indole alkaloids
such as dihydroteleocidein B and teleocidin, which are chemically distinct from the
phorbol esters, represent another class of epidermal tumor promoters that activate
oxygen metabolism in PMNs [Troll et al. 19821. Mezerein, a diterpene related to the
phorbol esters, is only a weak tumor promoter [Mufson et al, 19791, but a potent
stimulator of the generation of 02 production and chemiluminescence [Kensler and
Trush. 198 1 ; Troll et al, 19821. Mezerein is however a potent stage-I1 promoter [Slaga
et al. 1980a], suggesting that elaboration of oxygen radicals may be particularly
important in the propagation stage of promotion.
Utilizing lymphocytes from bovine lymph nodes, Mueller and coworkers [Mueller et al. 1980; Wertz and Mueller, 1980; Mueller and Wertz, 1982: Wrighton and
Mueller, 19821 have presented evidence that TPA activates certain preexisting enzymatic processes and induces gene expression for other enzymes through an oxygenrnediated mechanism. Early membrane events that are oxygen independent involve
the facilitation of glucose transport and the capping reaction observed using fluorescein-labeled concanavalin A. The oxygen-dependent group of lymphocyte responses
to TPA include the stimulation of choline phospholipid synthesis, amino acid transport. cytidyl transferase activity. and the alkylation of protein by arachidonic acid. all
of which occur after a 20-30 min lag period. The possibility that the requirement for
molecular oxygen in these later events reflects the involvement of reactive oxygen
species remains conjectural at present. However, in contrast to the oxygen-independent events. the oxygen-dependent events are sensitive to inhibition by antipromoting
retinoids and archidonate cascade inhibitors that also impede oxygen radical elaboration in phorbol ester-activated PMNs.
Other than the inferential nature of inhibition of tumor promotion by freeradical scavengers discussed in a subsequent section. direct evidence for the elaboration of oxygen radicals in response to phorbol esters in target tissue, ie, mouse
epidermis. is very limited. This limitation reflects in large part the difficulty in
measuring these species in intact biological systems. However, Fischer and Adams
119841 have observed a dose-dependent generation of chemilumineseence in mouse
epidermal cells following TPA treatment. Additionally, Goldstein et a1 119831 have
demonstrated the TPA-mediated production of H202 in mouse skin in vivo by
measuring the ability of 3-amino-1.2,4-triazole to inhibit catalase, a process dependent on H202. Treatment with increasing concentrations of 3-amino-1,2,4-triazole
produces a dose-dependent decrease in catalase activity in mouse s b n . This inhibitory
effect is markedly enhanced when 2.5 pg TPA is also applied, although this dose of
TPA alone has no effect on catalase activity. The cellular source of the H202remains
undefined, although it is hypothesized by these investigators that it is produced by
infiltrating intlammatory cells activated by the phorbol ester. The findings of Fischer
and Adams 119841 suggest this need not necessarily be so, since epidermal cells
themselves can be shown to produce radicals. Discussions of the overall role of
intlammation in carcinogenesis can be found in several recent reviews [Demopoulos
et al, 1980, 1983: Roman-Franco, 19821.
Modulation of Oxidant Defenses by Tumor Promoters
Multiple defense mechanisms are present in cells for coping with oxidant stress.
Protection of cell constituents from damage by oxygen and its metabolites can be
Oxygen Radicals in Tumor Promotion
599
accomplished through enzymatic and nonenzymatic means. SOD, catalase, and peroxidases (particularly gluthathione peroxidase) catalyze reactions to remove 0; or
H202. Enzymes that catalyze the removal of 'OH or lo2have not been described so
that these species must be removed by quenching molecules such as glutathione and
vitamins A, C. and E. Levels or activities of these defenses are known to be
modulated in response to oxidant stress.
Solanki et a1 [I9811 have investigated the effects of tumor promoters on epidermal SOD and catalase activites. Treatment of mouse s h n with TPA results in a rapid
and sustained decrease in SOD and catalase activities which is diminished 75% and
60%. respectively. The decline in SOD activity occurs within 3 h of TPA application
and is maximal at 16 h. The alterations in both enzymes occur against a background
of enhanced protein synthesis suggesting that the effect of phorbol ester is selective
for SOD and catalase. Nonphorbol tumor promoters such as anthralin and mezerein
(a potent stage-I1 promoter) are also active. In a separate study [Logani et al, 19821,
it was observed that these tumor promoters have no effect on glutathione peroxidase
activity. The underlying mechanism for the reduction in SOD and catalase activities
is unclear. Although 01 and H202 inhibit catalase and SOD, respectively [Bray et al.
1974: Kono and Fridovich. 19821, the requirement for high oxidant concentrations
and the lack of protective effect of either 2(3)-tert-butyl-4-hydroxyanisole (BHA) or
3,5-tert-butyl-4-hydroxyanisole(BHT) may mitigate against this possibility. Kinsella
et a1 119831 report that TPA provokes a lowering of SOD activity in both human
lymphocytes and fibroblasts. although maximal reduction is not observed until 48 h.
The role of lowered SOD levels and cancer has been recently reviewed [Oberley,
19821.
Inhibition of Tumor Promotion by Free Radical Scavengers
Antioxidants have been shown to protect against tumor induction by a spectrum
of chemical carcinogens in several rodent tissues [Wattenberg and Lam, 1981 1.
Current evidence suggests that the protective effects may arise from enhanced carcinogen inactivation through selective induction of enzymatic detoxification pathways
[DeLong et al, 19831. Alternatively. antioxidants may directly or indirectly through
elevation of free thiol pools serve as scavengers of electrophilic reactants. thus
protecting critical biomolecules from attack. Such a mode of action may be more
pertinent to the antipromoting activity of phenolic antioxidants such as BHA and
BHT. Wattenberg and Lam [I9831 have observed inhibition of carcinogenesis of the
large bowel when 3-BHA (the ma.jor isomer in commercial preparations) was given
subsequent to dimethylhydrazine. Slaga et a1 [ 19831 have reported that BHA and BHT
inhibit skin tumor promotion after 7,12-dimethylbenz[a]anthraceneinitiation by both
benzoyl peroxide and TPA. In addition to the phenolic antioxidants, other antioxidants, namely, a-tocopherol and disulfiram, are also effective inhibitors. Many
inhibitors of tumor promotion, such as retinoids, protease inhibitors, antioxidants,
and arachidonate cascade inhibitors, block oxy radical production by activated PMNs
(Goldstein et al, 1979; Witz et al. 1980; Kensler and Trush, 1981, 1983; Yavelow et
al, 1982; Hoffman and Autor, 19821 and epidermal cells [Fischer and Adams, 19841.
O'Brien et a1 [I9751 have reported an excellent correlation between the tumor
promoting activity of various compounds and their ability to induce ornithine decarboxylase (ODC). the rate limiting step in polyamine biosyntheses. Kozumbo et a1
[1983] have utilized epidermal induction of ODC by TPA to examine the immediate
600
Kensler and Trush
effects of antioxidants as inhibitors of tumor promoter action. Structure-activity
relationships among BHA analogs show that both free-radical scavenging potential
and lipophilicity are essential properties. Phenolic antioxidants are particularly reactive with oxygen-centered radicals such as peroxy and hydroxyl radicals. but show
little reactivity with 0; [Simic and Hunter: 19831. However, recent observations by
Kensler et a1 [I9831 demonstrate the probable role of 0- as well in tumor promotion.
Here a low molecular weight, lipophilic copper coordination complex with SODmimetic activity inhibits phorbol ester-mediated tumor promotion in the skin of CDI mice. This complex, copper(IIj(3,5-diisopr~pylsalicylate)~
(CuDIPSj, catalyzes the
disproportionation of 0; and displays a rate constant comparable to native Cu-Zn
SOD [Lengfelder and Weser, 19811. Compounds devoid of SOD-mimetic activity
such as the ligand alone (DIPS) and the corresponding zinc complex (ZnDIPS) are
without antipromoter activity. CuDIPS also blocks the induction of ODC by TPA,
anthralin, and mezerein. Friedman and Cerutti [I9831 also find that antioxidants with
specificity for 0 2 , H202. and 'OH, namely, SOD, catalase, and mannitol, partially
inhibit the induction of ODC by TPA in mouse mammary tumor cells.
A number of systems have been proposed as models of in vitro promotion of which
the mouse embryo fibroblast C3HlOT112 cells are the most thoroughly studied.
Exposure of these cells at low density to carcinogenic agents such as polycyclic
hydrocarbons, ultraviolet, or X-ray irradiation followed by sustained exposure to
phorbol esters results in augmented formation of transformed foci [Mondal and
Heidelberger. 1976; Mondal et al, 1976; Kennedy et al, 19801. In general, pron~oting
agents produce similar effects for in vivo carcinogenesis and in vitro transformation.
Utilizing the C3HlOT112 system, Little et a1 [I9831 have observed that the enhancement of X-ray-induced transformation by TPA can be suppressed by concommitant
incubation with either SOD or catalase. A similar finding has been made by Borek
and Troll [I9831 who demonstrated that SOD and to a lesser extent catalase inhibit
the TPA enhancement of radiogenic transformation of freshly explanted hamster
embryo cells. Zimmerman and Cerutti [I9841 have recently reported that reactive
oxygen acts directly as a promoter of transformation in C3HlOT112 cells. Cells
initiated with either X-rays or benzo(a)pyrene-diol-epoxide I at doses that produce
minimal transformation show a strong enhancement in formation of malignant foci
following daily treatment for three weeks with xanthine oxidaselxanthine. Addition
of this extracellular 02 generating system produces an effect comparable to that
engendered by TPA addition, and like the TPA response, is inhibited by simultaneous
addition of SOD. Phenolic antioxidants, SOD, and CuDIPS also inhibit promotion of
neoplastic transformation by TPA in JB-6 mouse epidermal cells [Nakamura et al.
19841, implying that oxygen radicals may be elaborated directly without the involvement of inflammatory cells.
STUDIES WITH NONEPIDERMAL TUMOR PROMOTERS
Prolonged inhalation of asbestos produces pulmonary fibrosis and two forms of
cancer. mesothelioma and bronchogenic carcinoma [Nettescheim et al, 1981; Mossman et al, 19831. The action of asbestos in bronchogenic carcinoma appears to be
primarily cocarcinogenic and promoting. Topping and Nettescheim [1980] have shown
that asbestos has a promoting effect in rodent tracheal grafts previously exposed to
subcarcinogenic amounts of 7,12-dimethylbenz(a)anthracene. Phenotypic markers of
Oxygen Radicals in Tunlor Promotion
601
promotion. such as induction of ODC, increased DNA synthesis, hyperplasia, and
metaplasia are induced by asbestos in tracheal organ cultures and tracheal epithelial
cells [Mossman et al, 19831. Exposure and subsequent phagocytosis of asbestos by
tracheal epithelial cells in culture produces cell damage that can be greatly diminished
by SOD [Mossman and Landesman. 19831, suggesting that 0 2 may be an important
mediator of asbestos toxicity. Catalase or singlet oxygen scavengers are ineffective.
Prolonged exposure (3-4 days) to asbestos produces marked increases in endogenous
SOD activities, perhaps reflecting an adaptive response. Concordantly, Gulumian et
al [I9831 observed a dose-dependent increase in lipid peroxidation when asbestos is
incubated with either rat liver or lung microsomes. Using spin trapping techniques,
Weitzman and Graceffa [I9841 have demonstrated that asbestos catalyzes the generation of hydroxyl and superoxide radicals from H 2 0 2 . Interestingly, in the tracheal
epithelial system OF is produced in the absence of inflammatory cells. In fact, the
role of an inflammatory process in the development of bronchogenic carcinoma is
uncertain [Netteschein et al, 198 1 1, although asbestos engenders pulmonary inflammation. Endotracheal lavage of mice 1 y after exposure for 75 days showed populations of lyniphocytes and PMNs elevated 3- and 9-fold, respectively, compared to
unexposed mice [Bozelka et al. 19831. Additionally. human PMNs respond to coincubation with asbestos with phagocytosis and generation of cherniluminescence [Gaunier et al, 19791. The chemiluminescent response to different forms of asbestos
correlates with in vivo fibrogenic and carcinogenic potentials. Asbestos also elicits a
vigorous chemiluminescent response in guinea pig alveolar macrophages [Hatch et
al, 19801. However, analogous to the situation with TPA in mouse skin, the importance of oxidant production by inflammatory cells to the promoting action of asbestos
is essentially unexplored.
There are a number of enzymes localized to mitochondria, peroxisomes, endoplasmic reticulum. and cytosol, which catalyze the univalent or divalent reduction of
molecular oxygen [Chance et al, 19791. In several instances, tumor promoters are
known to modify the biochemical action of these organelles, such that output of 0;
and H 2 0 2 are enhanced. Whether these effects directly relate to promotion is
unresolved.
Hyperoxia induces an oxidant stress that is manifested primarily in the lung.
Low-level chemiluminescence is increased in perfused lung as well as in perfused and
in situ liver under conditions of hyperbaric oxygenation [Chance et al, 19791. High
oxygen tension increases 0; forination by lung submitochondrial particles [Turrens
et al, 1982aI and H102 release by intact lung mitochondria and microsomes [Turrens
et a]. 1982b] by 2-10-fold. Hyperoxia also appears to have a promoting effect. Heston
and Pratt [I9561 and DiPaulo [I9591 have shown an enhancement in the number of
pulmonary tumors by placing mice under increased oxygen tension following treatment with dibenz[a.h]anthracene or urethan, respectively. Dettnier et a1 119681 have
reported that exposure of rats to hyperoxia for 6 wk following 400 rads whole-body
irradiation diminished the latency period and enhanced the incidence of mammary
tumors.
Hepatocarcinogenesis proceeds in qualitatively distinct stages similar to the
initiation-promotion process of skin carcinogenesis. Peraino et al [I9711 originally
demonstrated the enhancing effects of dietary phenobarbital exposure on hepatic
tumorigenesis in rats previously fed 2-acetylaminofluorene and an expanding array of
hepatic tuinor promoters have since been enumerated [Pitot et al. 1980: Periera. 1982;
602
Iiensler and Trush
Jensen et al, 19821. A common feature of many, but not all, of these agents (ie.
phenobarbital, DD'T. TCDD, polychlorinated, and polybrorninated biphenyls, and
dii2-ethylhexy1)phthalate) is their pronounced inductive effect on hepatic mixed function oxidase activity. The mixed function oxidase system, in the presence of pyridine
cofactors. generates reactive oxygen metabolites [White and Coon. 1980; Trager,
1982; Ingelman-Sundberg and Hagbjork, 19821. Microsomes induced with phenobarbital or pregnenolone-16a-carbonitrile generate H z 0 2 at a 5-fold greater rate than
noninduced microsomes [Hildebrandt et al, 19731. However, considerations that the
elaboration of reactive oxygen species in response to enzyme induction is an important
component of the action of these chemicals as tumor promoters must be tempered by
the observations that not all inducers of hepatic mixed function oxidase activity are
promoters (ie. arnobarbital) [Peraino et al , 19751. Additionally. Leonard et a] [ 19821
and Jensen et al [I9831 have recently reported that induction of hepatic mixed function
oxidase enzymes is neither a prerequisite for nor an indicator of tumor promotion
ability in a study utilizing brominated biphenyls.
A number of structurally divergent xenobiotics. including hypolipidemic drugs
and phthalate plastisizers. have been described as peroxisome proliferaters in rodent
liver. Peroxisomes are single membrane-limited cytoplasmic organelles that have
been implicated functionally in gluconeogenesis, lipid metabolism, and the detoxification of H202. Sustained peroxison~eproliferation has been associated with an
increased occurrence of hepatocellular carcinoma and it appears that peroxisome
proliferaters as a class are carcinogenic [Reddy and Azamoff, 1980; Warren et al.
19821. Included in this grouping is the ubiquitous environmental contaminant, di(2ethylhexy1)phthalate (DEHP). which produces hepatocellular carcinomas in F344 rats
and B6C3F 1 mice following long-term dietary exposure [Kluwe et al. 19821. Ward et
al [I9831 have recently reported that DEHP exhibits promoting activity, and an
apparent absence of initiating activity, in a B6C3F I mouse liver assay using diethylnitrosamine as an initiating agent and that DEHP is a stage I1 promoter in mouse skin
[Diwan et al. 19841. Similarly. Reddy and Rao [I9781 and Mochizuki et a1 119821
have reported that the hypolipidemic drugs WY-14,643 and clofibrate promote the
appearance of hepatocellular carcinomas following initiation of F344 rats by diethylnitrosamine. The antioxidants ethoxyquin and BHA inhibit the hepatic tumorigenic
effect of another peroxisome proliferator, ciprofibrate [Rao et al, 19841, although
peroxisome proliferation itself is unaffected [Lalwani et al, 19831. The underlying
mechanism of the promoting activities of these compounds may relate to the production of active oxygen species. Peroxisomes of liver and other organs contain a number
of H202-generating enzymes, including some flavoproteins, D-amino acid oxidase,
L-a-hydroxyacid oxidase, fatty acyl CoA-oxidase and urate oxidase [Chance et al,
19791. Clofibrate and other hypolipidemic drugs cause an 11-18-fold increase in the
capacity of rat liver to oxidize long-chain fatty acids [Lazarow, 1977; Osunii and
Hashimoto, 19781. Increased H z 0 2 generation has been observed in the livers of rats
administered peroxisome proliferaters [Lalwani et al, 19811. Catalase activities are
also increased, but in a manner disproportionately small compared to the increase in
peroxisome volume [Moody and Reddy, 19761 and H202-generating fatty acid 0oxidation [Lazarow and DeDuve, 19761. Prolonged exposure to peroxisome proliferaters results in the excessive accumulation of autofluorescent lipofuscin in the liver
[Reddy et al, 19821 providing indirect evidence for the increased production of
biologically damaging free radicals. Although the exact nature of the fluorescent
Oxygen Radicals in Tumor Proniotion
603
chromophores in lipofuscin is poorly defined, it is ascribed to products of lipid
peroxidation.
In addition to peroxisomal effects, phthalic acid esters also inhibit electron and
energy transport in rat liver mitochondria [Takahashi, 19771. Inhibition of mitochondrial respiration can result in enhanced H 2 0 2production [Boveris, 19771. An inhibitor
of mitochondrial electron transport. rotenone, is a hepatocarcinogen [Gosalvez. 19831.
Interestingly, Backer et a1 [1982] have shown that TPA inhibits mitochondrial respiration in C3HlOT112 mouse fibroblasts at nanomolar concentrations. Although the
nonpromoter 4-0-methyl TPA does not effect oxygen consumption, the promoters
phorbol-12,13-dibutyrateand teleocidin are also inhibitory. Tangeras and Malviya
119831 find that micromolar concentrations of TPA have no effect on respiration or
H 2 0 2production by isolated rat liver mitochondria. On balance, although mitochondria may be the major source of intracellular 0 2 and H202,the importance of altered
mitochondrial function in carcinogenesis remains undefined.
OXYGEN RADICAL-BIOMOLECLILE INTERACTIONS AND CARCINOGENESIS
A variety of chemicals, including tumor promoters. can create a cellular prooxidant state either directly by elaborating oxy radicals or indirectly by disrupting the
homeostasis that normally exists between the rate of cellular radical generation and
the rate of radical dissipation. Although cells have various defense and repair mechanisms, this pertubation often leads to cellular injury and death [Halliwell. 1978; Bus
and Gibson, 1979: Holtzman 1981; Trush et al, 1982; Hochstein, 1983; Sies and
Cadenas, 1983; Thaw et al. 19831. While it may be that similar reactive oxygentarget interactions are involved in both chemical-induced acute toxicity and carcinogenesis, the alterations resulting in the latter are likely to be more subtle and selective.
Membrane lipids. free amino acids, nucleic acids, polysaccharides, enzymes, receptors, and structural and transport proteins have all been shown to be altered by oxy
radicals [Demopoulos et al. 1980, 1983: Fridovich. 19811. Many of these investigations have been conducted in well-defined chemical and/or isolated biochemical
systems. However. the critical macromolecular target(s) and the cellular processes
involved in reactive oxygen toxicity and carcinogenicity remain unresolved. The
following paragraphs examine some of these interactions and how they may be related
to carcinogenesis.
The interaction of reactive oxygen metabolites with DNA is well characterized
[Brawn and Fridovich, 1981; Floyd, 1982: Schmidt and Borg, 1976; Schulte-Frolinde, 1983; Van Hemmen and Mueling, 1975; Von Sonntag et al, 19811 and may
result in chromosome aberrations, mutagenesis and modulation of gene expression.
Tumor promoter-stimulated PMNs produce a series of DNA damaging oxidant species that are derived in part from oxy radicals [Birnboim, 1982a,b, 1983; Birnboim
and Biggar, 1982; Emerit and Cerutti, 1981, 1982. 1983; Cerutti et al, 1983; Emerit
et al, 19831. DNA strand breaks are detectable in PMNs 5 minutes following TPA
exposure and are maximal by 45 min. Nonphorbol epidermal tumor promoters such
as dihydroteleocidin B, anthralin, benzoyl peroxide, H202, iodoacetic acid, and
canthardin are also effective inducers of DNA strand breaks in PMNs [Birnboim.
19831. An extensive evaluation by Birnboin [l982b] of inhibitors of TPA-mediated
DNA damage suggests a mechanism involving 0; and H202, but the nature of the
ultimate damaging radical remains unclear. SOD and catalase block the response;
604
Kensler and Trush
however, 'OH scavengers such as DMSO and glycerol are not very effective and
scavengers produce no decrease. 02 generated from a xanthine oxidaselxanthine
system also produces DNA strand breaks in unstimulated PMNs [Birnboim. 19831.
Emerit and Cerutti [1981. 1982, 19831 have shown that exposure of PMNs,
monocytes. or phytohemagglutinin-stimulated lymphocytes in the presence of moncytes or platelets to TPA. but not its weakly promoting derivatives, induces chromosomal aberrations such as breaks, gaps, fragments. dots, double minutes. interchanges.
and pulverization. This clastogenic action of TPA, which is inhibited by SOD, is
indirect and is mediated by secondary stable products which are formed by the cell in
response to the interaction with the tumor promoter. In addition to active oxygen
species. intermediates of arachidonate metabolism are involved in the formation and
action of clastogenic factor because phospholipase A2 inhibitors as well as inhibitors
of the cyclooxygenase and lipoxygenase pathways are anticlastogenic [Cerutti et al,
1983; Emerit et al, 19831. Preliminary chromatographic evidence suggests that the
clastogenic factor is composed of lipid hydroperoxides. aldehydic break-down products. and free arachidonic acid. Cerutti and coworkers have proposed a model for
.'membrane mediated chromosomal damage" in which membrane-active agents such
as some tumor promoters and complete carcinogens (eg, aflatoxin B,) elicit an
oxidative burst and stimulate the archidonate cascade to perturb membrane integrity
so that phospholipids are more susceptible to autooxidation and form a DNA-damaging clastogenic factor [Cerutti et al, 19831. In addition to the clastogenic factors.
Weiss et a1 [I9831 have found that TPA-activated PMNs produce a stable oxidant that
has sufficient potential to modify sulfhydryl and thioether containing molecules. The
elaboration of these phagocyte-derived N-chloramines is inhibited by catalase. but not
SOD. The differential susceptibilities to inhibitors suggests that these three DNA
damaging and oxidant factors are probably distinct.
The elaboration of oxygen radicals by PMNs also mediates genotoxicity in other
cells. PMNs stimulated by phagocytosis increase reversion frequency in Salmonella
hphimurium strain TAlOO [Weitzman and Stossel, 19811 and induce dark mutants of
the luminous bacteria Photobacterirun jischeri to revert to heredity stable luminescent
forms [Barak et al, 19831. Phorbol ester-stimulated PMNs produce cytogenetic changes
in mammalian cells as evidenced by a dose-dependent increase in sister chromatid
exchange in CHO [Weitberg et al. 19831 and V-79 (Popescu NC, Trush MA and
Kensler TW. unpublished observations) fibroblasts. Mutagenesis can also be induced
in CHO cells by potassium superoxide and is inhibited by SOD [Cunningham
and Lokesh, 19831. Birnboim [I9831 reports that when TPA-stimulated PMNs are
cocultured with erythroleukemia cells appreciable DNA strand breakage is observed
in the secondary target cells, but in this instance SOD is not protective, although
catalase remains so. Collectively, these studies demonstrate that diffusable species act
as mediators of inflammatory cell-induced genotoxicity. Addition of TPA to primary
cultures of epidermal cells results in substantial increases in chromosomal aberrations
[Fusenig et al. 19831. However, Gensler and Bowden [I9831 have shown that DNA
strand scission is unrelated to irreversible late stage promotion by TPA in JB-6 mouse
epidermal cells although TPA does induce DNA single strand breaks in primary
mouse epidermal cells cocultured with macrophages [Dutton and Bowden, 19841.
Thus the importance of DNA damage in promotion certainly requires further clarification. Birnboim [I9831 has presented several hypotheses suggesting how reactive
*
Oxygen Radicals in Tumor Promotion
605
oxygen-mediated DNA damage might be interlinked with oncogene activation and
expression of a malignant phenotype. For example, studies by Sher and Friend [I9781
and Terada et a1 [I9781 suggest that DNA strand breaks induced by a variety of
stimuli may trigger derepression of genes required for hemoglobin synthesis in
erythroid leukemia cells. Aft and Mueller [I9831 have recently proposed that hemin
may implement changes in gene expression associated with erythroid differentiation
by the generation of localized reactive oxygen species that produce DNA nicks.
DNA damage by chemical carcinogens, radiation, or ultraviolet light is subject
to enzymatic repair [Hanawalt et al, 1979; Kimball, 19791. While these systems have
been extensively characterized in bacterial systems, their existence in mammalian
cells are implicated as a result of defective DNA repair in cell lines established from
individuals with ataxia telangiectasia and xeroderma pigmentosum [Patterson et al,
19831. DNA repair-deficient bacteria strains are hypersensitive to H 2 0 2 [Demple et
al. 19831 and it has been demonstrated that the rec+ gene product may be more
important than SOD and catalase in protecting E coli against H z 0 2 toxicity [Carlson
and Carpenter. 19801. A newly described component of DNA repair processes is
poly(ADP-ribose) polymerase. This enzyme catalyzes the formation of poly(ADPribose) from NAD+ in a reaction that requires histones and damaged DNA [Ydmada
et al, 19711. Although the exact cellular function of poly(ADP-ribose) is not known.
there appears to be a relationship between its synthesis and DNA damage and repair
in mammalian cells. Inhibition of poly(ADP-ribose) polymerase enhances DNA
strand breakage and sister chromatid exchange induced by alkylating agents [Park et
al, 19831 and potentiates cell killing by ultraviolet light [Durrant and Boyle, 19821.
Sugimura and Miwa [I9831 have recently summarized the relationship of poly(ADPribose) to various aspects of cancer. Studies with alloxan suggest a relationship
between poly(ADP-ribose) synthesis. NAD+ levels and oxidant-mediated toxicity.
Alloxan undergoes both enzyme-catalyzed and enzyme-independent redox cycling
resulting in reactive oxygen generation [Malaisse, 19821 and SOD and catalase protect
against alloxan-induced DNA strand breaks and inhibition of proinsulin synthesis in
alloxan-exposed pancreatic islets [Yamamoto et al, 1981; Uchigata et al, 19821. This
latter effect is attributed to the depletion of NAD' resulting from its utilization by
nuclear poly(ADP-ribose) polymerase. Increased oxygen tension also reduces cellular
nicotinamide adenine dinucleotide levels through oxygen-mediated inactivation of the
enzymes involved in NAD+ biosynthesis [Brown et al, 19791. Thus, depletion of
NAD+ by this mechanism would also influence the subsequent repair of DNA damage
by the poly(ADP-ribose) system.
In his now classic review on cellular mechanisms of oxygen toxicity, Haugaard
[I9681 emphasized the importance of enzyme inactivation in this process. More
recently, Fucci et a1 119831 have put forth the concept that oxidant-mediated inactivation of key metabolic enzymes may serve a function in regulating cellular protein
turnover. Enzymes shown to be inactivated in this study included various kinases,
synthetases and NAD(P)H-dependent dehydrogenases. The inclusion of catalase in
the admixture prevented these oxidant-mediated processes, no matter whether the
oxidant was generated by NAPH oxidase, the NADPH cytochrome P-450 reductasel
P-450 system, or the ferrous iron-H202 system. Addition of H 2 0 2 to the reduced
ferrous form of P-450 destroys this hemoprotein [Guengerich, 19781 and inactivates
SOD [Bray et al, 19741. Willson and co-workers [Packer et al, 1978; Searle and
Willson. 1980; Willson, 1977, 19821 have discussed the sensitivity of various amino
606
Kensler and Trush
acids as targets in the inactivation of enzymes such as lysozyme, ribonuclease, and
glutathione peroxidase by 'OH and the trichlorornethyl peroxy and thiocyanate
radicals.
The relevance of enzyme inactivation at the cellular level has been demonstrated
by the studies of Brown et a1 [1979]. Exposure of E coli to hyperbaric oxygen results
in the inactivation of one of the key enzymes in NAD+ biosynthesis, quinolate
phosphoribosyl transferase. The inhibition of this enzyme is accompanied by significant decline in total cellular pyridine nucleotide content. Niacin, which enters the
NAD' biosynthetic pathway beyond this enzyme, prevents oxygen-dependent toxicity
in E coli and the alterations in pyridine nucleotide content. Niacin also prevents
hyperoxia-induced toxicity to alveolar macrophages and paraquat-mediated toxicity
in rats [Brown et al, 1981: Pearl and Ruffin, 19831. As discussed previously, NAD+
is important, not only as a coenzyme in a large number of cellular reactions, but in
the poly(ADP-ribose) DNA repair system as well.
In contrast to the inhibiting effect of a pro-oxidant state on enzymes, guanylate
cyclase, a prominent enzyme involved in regulating cell growth and transformation.
is activated in the presence of oxidant-generating systems. DeRubertis and Creaven
[I9821 have recently reviewed the various aspects of guanylate cyclase activity in
relationship to the actions of a number of carcinogens. White et al [I9761 demonstrated that either the direct addition of H202, or its generation by glucose oxidasel
glucose, activates this enzyme in lung supernatant. The tumor promoter benzoyl
peroxide also activates guanylate cyclase as does 0; resulting from the redox cycling
of paraquat [Goldberg et al, 1978; Vesely et al, 1979). Oberley et a1 119811 have
discussed how the ratio of cGMP to CAMPmay be involved in regulating normal cell
division and how an increase in the cGMP1cAMP ratio, along with the attainment of
cellular immortality, may be integral to the malignant transformation of cells.
Whereas guanylate cyclase represents an example of direct enzyme activation
by an oxidant exvironment. exposure of cells to a pro-oxidant state also results in the
induction of enzyme synthesis. Examples of this process, include induction of manganese-containing SOD in E coli, benzo(a)pyrene hydroxylase activity in rat liver
cells, and a DNA repair system in E coli [Hasson and Fridovich, 1979; Paine and
Francis, 1980; Demple and Halbrook, 19831. In summary then, oxygen radicals may
mediate enzymic denaturations and inactivations, activations, and inductions. which
could result in pertubations in control of cellular growth and differentiation.
A number of carcinogens that are metabolized to radical intermediates result in
increased peroxidation of various cell organelles. Reactive oxygen-dependent lipid
peroxidation has also been implicated in the mechanism of radiation-induced carcinogenesis [Petkau, 19801, ultraviolet light-induced skin carcinogenesis [Logani and
Davies, 19801, and the action of epidermal tumor promoters [Logani et al, 1982;
Stohs et al, 19831. It appears that the hydroxyl radical and iron chelates, which
function like a hydroxyl radical. are of particular importance in lipid peroxidation
[Lai and Piette, 1977: Tien et al. 19821.
Both nonspecific membrane peroxidation and enzymatically controlled stereospecific lipid peroxidation associated with prostaglandin synthesis can mediate the
formation of highly reactive and diffusable aldehyde and carbonyl products and
clastogenic factors which can amplify these forms of peroxidation and enhance
damage. For example, 4-hydroxynoenal, formed during iron-dependent microsomal
peroxidation, has been shown to alter mitochondria1 function, polyamine synthesis,
Oxygen Radicals in Tumor Promotion
607
tubulin function. adenylate cyclase activity, and protein and DNA synthesis [Dianzani.
19821. The effects and manifestations of membrane peroxidation are diverse. Peroxidation of plasma membrane can result in decreased ligand binding to receptor and
inhibition of sodium-potassium ATPase or adenylate cyclase activity [Demopoulos et
al. 1980; Fourcans. 1974; Maridonneau. 1983; Muakkassah-Kelly et al, 19821. Such
alterations can be attributed to both direct effects on enzymes and indirect effects
mediated through changes in the lipid environment. Lipid peroxidation in mitochondria induces swelling and alteration of respiratory function [Aono et al, 19811.
Peroxidation products of the nuclear membrane have been implicated as mediators of
DNA damage [Ames et al. 19821. Nuclear-cytoplasmic communication, such as RNA
transport [Yannarelli and Awad, 19821, may also be affected by peroxidation at this
site. Endoplasmic reticulum subjected to a peroxidation reaction is functionally
modified as characterized by inhibition of microsomal ATP-dependent calcium sequestration and mixed-function oxidase activity [Bidlack and Tapel, 1973; Jones et al,
19831. Thus. all cell membranes can be subject to radical-mediated modification.
While the initial consequences of peroxidation are on membrane structure and function and integrity of cell compartmentalization. the ultimate manifestations. short of
cell death, could be reflected in altered cellular metabolic and regulatory activities.
SUMMARY
The concept that oxygen radicals are involved in the tumor promotion process
is presently based on three groups of experimental evidence. First. it has been
dernonstrated that different classes of tumor promoters can evoke the elaboration of
oxygen radicals by direct chemical generation or through the indirect activation or
alteration of cellular metabolic sources. Although the bulk of information concerning
the indirect action of promoters deals with effects in non-target cell populations, ie,
inflammatory cells. these cells may be coordinately involved in some aspects of
promotion or may serve as a model to presage similar effects in target cells, as
suggested by an incipient literature [Fischer and Adams, 1984; Nakamura et al.
19841. Second. tumor promoters can be shown to modulate cellular antioxidant
defenses. For example, epidermal SOD and catalase activities are diminished following promoter exposure. perhaps allowing for amplified oxygen radical-biomolecule
interactions. Third, reactive oxygen scavengers and detoxifiers, such as SOD. SODmimetics, catalase, and phenolic antioxidants, inhibit biochemical and biological
actions of tumor promoters in vitro and in vivo, thus providing strong presumptive
evidence for oxygen radical involvement in the later stages of neoplastic development.
The critical macromolecular targets for these oxygen radicals remain undefined:
however. they may include lipids. DNA. DNA repair systems, and other enzymes,
and their effects on these biomolecules and cellular processes are compatible with
current theories of carcinogenesis.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the financial support of the American
Cancer Society. Maryland Division, cancer prevention grant SIG-3, and the National
Institutes of Health. grants ES00454. CA36380. and BRS.
608
Kensler and Trush
REFERENCES
Aft RL. Mucller GC (1983): Hemin-mediated DNA strand scission. J Biol Chern 258: 12069-12072.
Anies BN. Hollstein MC, Cathcart R (1982): Lipid peroxidation and oxidative damage to DNA. In Yagi
K (ed): "Lipid Peroxides in Biology and Medicine." New York: Academic Press. pp 339-351.
Ames BN (1983): Dietary carcinogens and anticareinogens: Oxygen radicals and degenerative diseases.
Science 22 1: 12561264.
Aono K, Shiraishi N. Arita T. Inouye B, Nakazawa T, Utsumi K (1981): Changes in mitochondria1
function by lipid peroxidation and their inhibition by biococlaurin alkaloid. Physiol Chern Phys
13: 137-144.
Ashton RE, Andre P. Lowe N. Whitefield M (1983): Anthralin: Historical and current perspectives.
Scm Dermatol 2:287-303.
Backer J M , Boersig MR, Weinstein IB (1982): Inhibition of respiration by a phorbol ester tumor
promoter in murine cultured cells. Biochem Biophys Res Common 105:855-860.
Barak M, Ulitzur S. Merzbach D (1983): Phagocytosis-induced mutagenesis in bacteria. Mutat Res
121:7- 16.
Berardi LC. Goldblatt LA ( 1969): Gossypol. In Liener IE (ed): "Toxic Constituents of Plant Foodstuffs."
New York: Academic Press. pp 21 1-266.
Berenblum I(1941): The cocarc~nogenicaction of croton resin. Cancer Res 1:44-48.
Bidlack WR. Tapcl AL (1973): Damage to rnicrosomal membrane by lipid peroxidation. Lipids 8: 177182.
Birnboim HC (1982a): DNA strand breakage in human leukocytes exposed to a tumor promoter. phorbol
myristate acetate. Science 215: 12471249,
Birnboim HC (1982b): Factors which effect DNA strand breakage in human leukocytes exposed to a
turnor promoter. phorbol myristatc acetate. Can J Physiol Pharmacol 60: 1359-1366.
Birnboim HC. Biggar WD (1982): Failure of phorbol myristate acctate to damage DNA in leukocyte:.
from patients with chronic granulomatous disease. Infect Immunol 38: 1299- 1300.
Birnboim HC (1983): Importance of DNA strand-break in tunlor promotion. In Nygaard OF. Simic MG
(eds): "Radioprotectors and Anticarcinogens." New York: Academic Press, pp 539-556.
Blumberg PM (1981): 111 citr-o studies on the mode of action of the phorbol esters, potent tunlor
pronloters. CRC Crit Rev Tox 8: 199-234.
Bock FG. Burns R (1963): Tumour-promoting properties of anthralin (1.8.9-anthratriol). J Natl Cancer
Inst 30:393-397.
Bock FG, Myers HK, Fox HW (1975): Cocarcinvgenic activity of peroxy compounds. J Natl Cancer
Inst 55: 1359-1361.
Borek C , Troll W (1983): Modifiers of free radicals inhibit in r,irro the oncogenic actions of X-rays.
bleomycin. and the tumor promoter 12-0-tetradecanoylphorbol-13-acetate. Proc Natl Acad Sci
USA 80:1304-1307.
Bors W. Saran M. Czapski G (1980): The nature of intermediates during biological oxygen activation.
In Bannister WH. Bannister J V (eds): "Biological and Clinical Aspects of Superoxide and
Superoxide Dismutase." New York: ElsevierINvrth Holland. pp 1-31.
Boutwell RK (1964): Some biological aspects of skin carcinogenesis. Prog Exp Tumor Res 4:207-250.
Boutwell RK. Hoe1 MJ. Dig~ovanniJ (1981): The role of anthralin in mouse skin tunlour promotion. Br
J Dermatol (Suppl) 20:68-69.
Boveris A (1977): Mitochondria1 production of superoxide radical and hydrogen perouide. Adv Exp
Med Biol 20:67-82.
Bozelka BE, Sestini P. Hammond Y. Lenzini L , Gaumer HR. Salvaggio JE (1983): A murine model of
asbestosis. Cheat 83S:YS-IOS.
Brawn K, Fridovich I (1981): DNA strand scission by enzymatically generated oxygen radicals. Arch
Biochem Biophys 206:414-419.
Bray RC. Cockle SA. Fielden EM. Roberts PB, Rotillo G, Calabrese L (1974): Reduction and
inactivation of superoxide dismutase by hydrogen peroxide. Biochem J 139:43-48.
Briggh RT. Drath DB, Karnovsky ML. Karnovsky MJ (1975): Localization of NADH oxidase of the
surface of human polymorphonuclear leukocytes by a new cytochemical method. J Cell Biol
67:566-585.
Brown OR. Yein F. Boehme D, Foudin L, Song C-S (1979): Oxygen poisoning of NAD biosynthesis: A
proposed site of cellular oxygen toxicity. Biochem Biophys Res Commun 9 1 :982-990.
Oxygen Radicals in Tumor Promotion
609
Brown OR. Heitkamp M. Song C-S (1981): Niacin reduces paraquat toxicity in rats. Science 212: 15101512.
Bus JS. Gibson JE (1979): Lipid peroxidation and its role in toxicology. Rev Biochem Toxicol I: 125149.
Carlaon J. Carpenter VS ( 1980): The reci A gene product is more important than catalase and superoxide
dismutase in protecting Etchrrichicr coli against hydrogen peroxide toxicity. J Bacteriol 142:319321.
Cerutti PA. Amstad P. Emerit I (1983): Tumor prornotcr phorbol-myristate-acetate induces membranemediated chromosomal damage. In Nygaard OF, Simic MB (eds): "Radioprotectors and Anticarcinogcns." Ncw York: Academic Press, pp 527-538.
Chance B. Sies H . Boveris A (1979): Hydroperoxide metabolism in mammalian organs. Physiol Rcv
59:527-605.
Coburn M. S~nshrimerP, Segal S, Burgos M. Troll W (1980): Oxygen free radical generation by
,oo\sypol: 4 possible mechanism of antifertility action in sea urchin sperm. Biol Bull 359:468.
Copcland ES (ed) (1983): A National Institutes of Health workshop report: Free radicals in promotiona chem~calpathology study section workshop. Canccr Rcs 43:5631-5637.
Cunningham ML, Lokesh BR (1983): Superoxide anion generated by potassium superoxide is cytotoxic
and mutagenic to Chinese hamster ovary cells. Mutation Res 111 :299-304.
DcChatclct LR. Shirley PS, Johnson RB Jr (1976): Effcct of phorbol rnyristate acetate on the oxidative
metabolism of human polyrnorphonuclear leukocytcs. Blood 47:545-554.
DeLong MJ. Prochaska HJ, Talalay P (1983): Substituted phenols as inducer\ of enzymes that inactivate
electrophilic compounds. In McBricn DCH. Slater TF (eds): "Protective Agents in Human and
Experimental Cancer." London: Academic Press, pp 175196.
Demple B. Halbrook J. L ~ n nS (1983): E,tch~richincoli .rrh mutants are hypersensitive to hydrogen
peroxide. J Bacteriol 153: 1079-1082.
Dcmplc B. Halbrook J (1983): Inducible repair of oxidative DNA damage in Escherichia coli. Naturc
304:366-368.
Dernopoulo:, H . Pietronigro D. Seligman M. Flarnrn E (1980): The possible role of free radical reactions
in carcinogenesis. J Environ Path Toxicol 3:273-304.
Demopoulos HB. Pietronigro DD. Seligrnann ML (1983): The development of secondary pathology
with free radical reactions as a threshold mechanism. J Am Coll Toxicol 2:173-183.
De Peystcr A. Quintanilha A. Packer L, Smith ML (1984): Oxdygen radical formation induced by
gosaypol in rat liver microsomes and human sperm. Biochem Biophys Res Comm 118:573-579.
DeRubcrtis FR. Craven P,4 (1982): Activation of guanylate cyclase by N-nitro carcinogens and related
agonists: Evidence for a free radical mechanism. In Floyd RA (ed): "Free Radicals in Cancer."
New York: hldrcel Dekkcr. pp 201-244.
Dcttmer Chl. Krarner S. Gottlieb SF. Aponte GE (1968): Carcinogenic properties of increased oxygen
tensions. I. Effects on radiation-induced mammary tumors. J Natl Cancer Inst 41:751-765.
Diamond L. O'Brien TG. Baird WM (1980): Tumor promoters and the mechanism of tumor promotion.
4dv Canccr Rcs 32: 1-74.
Dianzani M U (1982): Biochemical effects of saturated and unsaturated aldehydes. In McBricn DCH,
Slatcr TE (eds): "Free Radicals. Lipid Peroxidation and Canccr." New York: Academic Press,
pp 129-158.
DiPaulo JA (1959): Influence of altered atmospheric oxygen on urethan-induced pulmonary tumora in
mice. J Natl Cancer lnst 23:535-540.
Diwan BA. Ward JM. Rice JM. Colburn NH. Spanglcr EF (1984): Tumor-promoting effects of di(2ethylhexy1)phthalate in JB6 mouse epithelial cells and mouse skin. Cancer Res in press.
Doroshou, JH (1983): Anthracycline antibiotic-atimulated superoxide, hydrogen peroxide and hydroxyl
radical production by NADH dehydrogcnase. Cancer Res 43:4543-4551.
Durrant LG. Boylc JM (1982): Potentiation of cell killing by inhibitors of poly(ADP-ribose) polymerase
in thur rodent cell lines exposed to N-methyl-N-nitrosourea or UV light. Chem Biol Inter 38:325338.
Dutton D. Bowden GT (1984): Clastogenic cffccts of tumor promoters in primary mouse epidermal cells
coincubated with macrophages. Proc Am Asaoc Cancer Res 25: 151.
Ernerit I. Cerutti PA (1981): Tumor promoter phorbol-12-myristate-13-acetate induces chromosomal
damage via indirect action. Naturc 293: 144-146.
Emerit I. Cen~ttiPA (1982): Tumor promoter phorbol 12-myristate 13-acetate induces a clastogenic
factor in human lymphocytes. Proc Natl Acad Sci USA 79:7509-7513.
610
Kensler and Trush
Emerit I. Cerutti PA (1983): Clastogenic action of tumor promotcr phorbol-12-myristate-13-acetatein
mixed human leukocyte cultures. Carcinogenesis 4: 13 13-1316.
Emerit I, Levy A, Cemtti P ( 1983): Suppression of tumor promoter phorbol myristate acetate-induced
chromosome breakagc by antioxidants and inhibitors of arachidonate metabolis~n.Mutat Res
110:327-335.
Fischer SM, Adams LM (1984): Inhibition of tumor promotcr stimulated chemiluminescence in mouse
epidermal cells by inhibitors of arachidonic acid metabolism. Proc Am Assoc Cancer Res 25:82.
Floyd RA ( 1981): DNA-ferrous iron catalyzed hydroxyl free radical formation from hydrogen peroxide.
Biochem Biophys Res Commun 99: 1209-1215.
Floyd RA (ed) (1982): "Frce Radicals and Cancer." New York: Marcel Dekker.
Fourcans B (1974): Role of phospholipids in transport and enzyme reactions. Adv Lipid Rcs 12: 147226.
Fridovich 1 (1981): Superoxide radical and superoxide dismutases. In Gilhert DL (ed): "Oxygen and
Living Processes." New York: Springer-Vcrlag. pp 250-271.
Friedman J. Cemtti P (1983): The induction of ornithine decarhoxylase by phorbol 12-myristate 13acetate or by aerurn is inhibited hy antioxidants. Carcinogenesis 4: 14251327.
Fucci L. Oliver CN. Coon MJ. Stadtrnan ER (1983): Inactivation of key metabolic enzymes by rnixedfunction oxidation reactions:possible implication in protcin turnover and aging. Proc Natl Acad
Sci USA 80: 15211525.
Fusenlg NE. Dzarlieva RT (1982): Phenotypic and chromosomal alterations in cell culturca as indicators
of tumor-promoting activity. In Hecker E, Fusenig NE. Kunz W. Marks F, Thielmann HW (eds):
"Carcinogenesis--A Comprehensive Survey." Vol 7, New York: Raven Press. pp 201-216.
Gaumer HR. Cairo J, Sal~aggioJE (1979): Chemiluminescent response to asbestos fibcra. Clin Res
27:37A.
Gensler HL. Bowden GT (1983): Evidence suggesting a dissociation of DNA strand scissions and latestage promotion of tumor cell phenotype. Carcinogenesis 4: 1507-151 1.
Goldberg ND. Graff G. Haddox MD, Stcphenson JH. Glass DB. Moser ME (1978): Redox modulation
of splenic cell soluble guanylate cyclase activity: Activation by hydrophilic and hydrophobic
oxidants represented by ascorbic dehydroascorbic acids. fatty acid hydroperoxides, and prostaglandin endoperoxidex. Adv Cyclic Nucleotide Res 9 : 101-130.
Goldstein BD. Witz G. Amomso M. Troll W (1979): Protease inhibitors antagonize the activation of
polymorphonuclear leukocyte oxygen consumption. Biochem Biophys Res Commun 88:854-860.
Goldstein BD. Witz G, Amomso M. Stone DS. Troll W (1981): Stimulation of human polymorphonuclear leukocyte superoxide anion radical by tumor promoters. Cancer Lett 11:257-262.
Goldstein B, Witz G , Zimmerman J. Gee C (1983): Frce radicals and reactive oxygen specie, in tumor
promotion. In Greenwald RA. Cohen G (edx): "Oxy Radicals and their Scavenger Systems.
Volume 11: Cellular and Medical Aspects." New 'Yhrk: Elsevier. pp 321-325.
Goldstein 1M. Roos D, Kaplan HB. Weissman G (1975): Complement and immunoglobulins stimulate
superox~deproduction hy human lcukocq tes independently of phagocytosis. J Clln Invest 56: 11551163.
Goodwin BJ, Weinberg JB (1982): Receptor-mediated modulation of human monocyte. neutrophil.
lymphocyte and platelet function by phorbol diesters. J Clin Invest 70:699-706.
Gosalvcz M ( 1983): Carcinogencsis with the insecticide rotenone. Life Sci 32:809~-816.
Greenlee WF. Sun JD. Bus JS (1981j: A proposed mechanism of benzene toxicity: Formation of reactive
intermediates from polyphenol metabolites. Toxicol Appl Pharmacol 59: 187-195.
Guengcrich Y (1978): Destruction of heme and hemoproteins mediated by liver microsomal reduced
nicotinamide adenine dinucleotide phosphate-cytochrome P-450 reductase. Biochem 17:36333639.
Gulumian M, Sardianos F. Kilroe-Smith T, Ockerse G (1983): Lipid peroxidation in microsorncs induced
by crocidolite fibres. Chcm Biol Int 43: 111-118.
Halliwell B (1978): Biochemical mechanisms accounting for the toxic action of oxygen on living
organisms: The key role of superoside dismutase. Cell Biol Int Rep 2: 113128.
Hanawalt PC. Cooper PK, Ganesan AK, Smith CA (1979): DNA repair in bacteria and mammalian
cells. Annu Rev Biochem 48:783-836.
Haroz RK, Thomasson J (1980): Tumor initiating and promoting activity of gossypol. Toxicol Lett
(Suppl) 6:72.
Hasson HM. Fridovich I (1979): Intracellular production of superoxide radical and of hydrogen peroxide
by redox active compounds. Arch Biochcm Biophys 196:385-395.
Oxygen Radicals in Tumor Promotion
611
Hatch GE. Gardncr DE. Menzel DB (1980): Stimulation of oxidant production in alveolar macrophages
by pollutant and latex particles. Environ Res 23: 121136.
Haugaard N (1968): Cellular mechanisms of oxygen toxicity. Physiol Rev 48:311-313.
Hcston WE. Pratt AW (1956): Increase in induced pulmonary tumors in mice associated with exposure
to high concentrations of oxygen. Proc Soc Exp Biol Med 92:451454.
Hicks RM (1983): Pathological and biochemical aspects of tumour promotion. Carcinogenesis 4 : 12091214.
Hildehrandt AG. Speck M , Roots I (1973): Possible control of hydrogen peroxide production and
dejiradation in microsomes during mixed function oxidation reaction. Biochem Biophys Re.;
Comrnun 54:968-975.
Hochstein P (1983): Futile redox cycling: Implications for oxygen radical toxicity. Fund Appl Touicol
3:215-2 17.
Hoffman M. Autor AP (1982): Effect of cyclooxygenase inhibitors and protease inhibitors on phorbolinduced stimulation of oxygen consumption and superoxide production by rat pulmonary macrophages. Biochem Pharmacol 31 :775-780.
Holtzman JL (1981): Role of reactive ox)yen and metabolite binding in drug toxicity. Life Sci 30: 1-9.
Ingelman-Sundberg M , Hagbjork AL (1982): On the significance of thc cytochrome P-450-dependent
hydroxyl radical-mediated oxygenation mechanism. Senobiotica 1?:673-686.
Jensen RK. Sleight SD. Goodman J1. Aust DS. Trosko JE (1982): Polybrominated biphenyls as
promoters in experimental hepatocarcinoyenesis in rats. Carcinogenesis 3: 1 1 8 3 1 186.
Jensen RK. Sleight SD, Aust SD. Goodnian JI. Trosho JE (198.3): Hepatic tumor-promoting ability of
3.3'.4.4'5.5'-hexabromohiphenyl: The interrelationship between toxicity, induction of hepatic
microsoma1 drug metabolizing enzymes, and tumor-promoting ability. Toxicol Appl Pharrnacol
71: 163-176.
Jones DP, Thor H. Smith MT, Jewcll SA. Orrenius S (1983): Inhibition of ATP-dependent inicrosomal
c a Z t sequestration during oxidative stress and its prevention by glutathione. J Biol Chem
258:6390-6393,
Kennedy AR. Murphy G. Little JB (1980): Effect of time and duration of exposurc to 12-0-tetradecanoylphorbol- 13-acetate on X-ray transtor~nationof C3H 10 TI12 cells. Cancer Res 40: 19151920.
Kensler TW. Trush MA (1981): Inhibition of phorhol ester-stimulated chemiluminescencc in human
polyn~orphonucIearleukocytes hy retinoic acid and 5.6-epouyretinoic acid. Cancer Res 41:216222.
Kensler TW. Trush MA (1983): Inhibition of ouygcn radical metabolism in phorbol ester-activated
poly~norphonuclear leukocytes hy an antitunlor promotins copper complex with superoxide
dismutase-minetic activity. Biochem Pharmacol 32:3485-3187.
Kensler TW. Bush DM, Kozumbo WJ (1983): Inhibition of tumor promotion by a biomimetic superoxide
dismutasr. Science 221 :75-77.
Kensler TU'. Tru\h MA (1984): Oxygen free radicals in chemical carcinogenesis. In Oberley LW (ed):
"Superoxide Dismutase. Volume 111: Pathological States." Boca Raton, FL: CRC Press, in press.
Kinihall RF (1979): DNA repair and its relationship to mutagenesis. carcinogcnesis and cell death. In
Prescott R. Goldstein L (eds): "Cell Biology." New York: Academic Press, pp 439-528.
Kinsella AR. St.C.Gaincr H. Butler J (1983): Investigation of a possible role for superoxide anion
production in tuniour promotion. Carcinogenesis 4:717-719.
Klcin-Szanto AJP. Slaga TJ (1982): Effects of peroxides on rodent skin: epidermal h>perplasia and
tumor promotion. J Invest Dermatol 79:30-34.
Kluwc WM. Haseman JK. Douglas JF. Huff J E (1982): The carcinogenicity of dietar) di(2-ethylhexy1)phthalate (DEHP) in Fiseher 344 rats and B6C3F1 mice. J Toxical Environ Health 10:797815.
Kono Y. Fridovich I (1982) Superoxide radical inhibits catalase. J Biol Chem 257-57515754.
Kotin P. Falk HL (1963): Organic Peroxides. hydrogen peroxide. epoxides. and neoplasia. Radiation
Res 3: 193-2 11.
Kozumbo WJ. Seed JL, Kensler TW (1983): Inhibition by 2(3)-ten-butyl-4-hydroxyanisoleand other
anrioxidants of epidermal ornithine decarboxylase activity induced by 12-0-tetradecanyolphorbol13-acetate. Cancer Res 43:2555-2559.
Lai C . Pictte LH (1977): Hydroxyl radical production involved in lipid peroxidation of rat liver
nllcro.rolne5. Biochem Biophys Res Commun 78:51-59.
Lalwani ND. Rcddy MK. Qureshi SA. Reddy JK (1981): Development of hepatocellular carcinornrl
and increased peroxisomal fatty acid P-oxidation in rats fed (4-chloro-6-(2,3-xylidino)-2-pyri1~iidiny1thio)acetic acid (WY-14,643) in the semi-purified diet. Carcinogenesis 2:645-650.
612
Kensler and Trush
Lalwani ND. Reddy MK, Qureshi SA. Moehle CM, Hayashi H. Reddy JK (1983): Nuninhibitory effect
of antioxidants cthoxyquin. 2(3)-tert-butyl-1-hydroxyanisclle and 3.5-di-tcrt-butyl-4-hydroxytoluene on hepatic perovisorne proliferation and peroxisornal fatty and fi-oxidation induced by a
hypolipidemic agent in rats. Cancer Res 43: 16801687.
Lazarow PB. DcDuve C (1976): A htty acyl-CoA oxidizing system in rat liver peroxisomcs: Enhanccment by clotibrate, a hypolipidernic drug. Proc Natl Acad Sci USA 73:2013-2046.
Lazarow PB (1977): Three hypolipidemic drugs increase hepatic palmitoyl-coenzymc A in the rat.
Science 197:580-581.
Lcngfelder E. Weser U (1981): Superoxide dismutation by low niolecular weight Cu-complexes. Bull
Eur Phyhiopath Resp 17:73-80.
Leonard TB. Dent JG, Graichen E, Lyght 0. Popp JA (1982): Comparison of hepatic carcinogen
initiation-promotion syatems. Carcinogenesis 3:851-856.
1-esko S , Caspary W. Lorentzen R. Ts'o POP (1975): Enzymic formation of 6-oxobenzo(a)pyrene
rad~calin rat liver homogenates from carcinogenic benzo(a)pyrene. Biochem 14:3978-3984.
Little JB. Kcnnedy AR. h'agasawa H (1983): Involvement of free radical intermediates in oncogenic
transformation and tumor promotion in vitro. In Nygaard OF. Simic MG (eds): "Radioprotectors
and Anticarcinogens." New York: Academic Press. pp 487-493.
Logani MK. Davies RE (19801: Lipid oxidation: Biologic effects and antioxidants--A review. Lipids
15:485-495.
Logani MK. Solanki V. Slaga TJ (1982): Effect of tumor promoters on lipid peroxidation in mouse skin.
Carcinogencsis 3: 13011306.
Lorentren RJ. Caapary WJ. Lesko SA. Ts'o POP (1975): The autooxidation of 6-hydroxybenzo(a)pyrene
ant1 6-oxobenzo(a)pyrene radical. reactlve metabolities of benzo(a)pyrene. Biochem~stry 14:39703977.
Lorentzcn RJ. Ts'o POP (1977): Benzo(a)pyrenedioncIhenzo(a)pyrenediol oxidation-reduction couples
and the yeneration of reactive reduced rnuleeular oxygen. Bio~.hemistry 16: 1467-1473.
Lorcntzen RJ. Lcsko SA. McDonald K . Ts'o POP (1979): Toxicity of metabolic benzo(a)pyrenediones
to cultured cells and the dependence upon molecular oxygen. Cancer Res 39:3194-3198.
Malaisse WJ (1982): Alloxan toxicity to the pancreatic 0-cell. Biochem Pharmacol 31 :3527-3534.
b1:iridonncau I. Braquet P. Garay PB (1983): Nat and K+ transport darnuge induced by oxygen free
radicals In human red cell membranes. J Biol Chern 258:3107-3113.
McBricn DCH. Slatcr T F (eda) (1982): "Free Radicals. Lipid Peroxidation and Cancer." London:
Academic Press.
Metzler M (1981): The metabolism of dicthylstilbestrol. CRC Crit Rev BiochemlO: 171-213.
Mochizuk~Y, Furukawa K. Sawacla N (1982): Effects of various concentrations of ethyl-a-p-chlorophcnoxyisobutyrate (clofibratc) on diethylnitrosamine-indu~dhepatie tumorigenesis in the rat.
Carcinogencsis 3: 1027-1029.
Mondal S. Brankow DW. Heidclberper C (1976): T\s,o-stage ehemical carcinogenesis in cultures of 10
T l I? cells. Cancer Res 36:2251-2260.
Mondal S. Heidelbcrgcr C (1976): Transformation of 10TlI2 cl 8 mouse embryo fibroblasts by
ultraviolet irradiativn and a phorbol cster. Nature 260:710-711.
Moody DE. Reddy JK (1976): Morphometric analysis of the ultrastructural ehanges in rat liver induced
by the peroxisome proliferator SaH 42-348. J Cell Biol 71:768-780.
Mossman B. Light W. Wei E (1983): Asbestos: mechanisms of toxicity and carcinogenicity in the
respiratory tract. Annu Rev Pharniacol Toxicol 2?:595-615.
Mnssman BT. Landesman JM (1987): Importance of oxygen free radicals in asbestos-induced injury to
airway epithelial cells. Chest 83S:50S-51S.
Mottrarn JC (1944): A developing factor in e~pcrimentalblastogenesis. J Pathol Bacterial 56: 181-187.
Muakkasah-Kelly SF. Andresen JW, Shih JC. Hochstein P (1982): Decreased and serotonin and
I-'Hlsp~peronebinding consequent to lipid peroxidation in rat cortical membranes. Biochem
Biophys Res Commun 104: 1003-1010.
Muellrr GC. Wertz PW. Kwong C H , Anderson K. Wrighton SA (1980): Dissection of the early
rnvlecular events in the activation of lymphocytes by 12-0-tetradeeanoylphorbol-13-acetate.
In
Pullman B. Ts'o POP. Gelboin H (eds): "Carcinogcnesis: Fundamental Mechanisms and Environmcntal Effects." Dordrecht. Holland: D. Reidel, pp 319-333.
Mucller GC. Wcrtz PW (1982): A possible role of protein alkylation in phorboI ester action. In Hecker
E. Fusenig NE. Kunz W, Marks F. Thiclrnann HW (eds): "Carcinogenesis - A Comprehensive
Survey." Vol. 7. New York: Ravcn Press. pp 499-5 11.
Oxygen Radicals in Tumor Promotion
613
Mufsor~RA, Fischer SM, Verma AK. Gleason GL, Slaga TJ (1979): Effects of 12-0-tetradecanoylphorbol-13-acetate and mererein on epidermal ornithinc decarboxylasc activity. isoproterenol-stiniuluted lcvela of cyclic adenosine 3:5-monophosphate. and induction of mouse skin tumors. Cancer
Rrs 39:479 1-4795.
Nakarnura Y. Gindhrin T D , Colburn NH (1984): Antioxidants. superoxide disrnutase and Cu(I1)DIPS
inhibit promotion of neoplastic transformation by TPA in JB6 cells: Role of reactive oxygen in
tumor promotion. Proc Am .4ssoc Cancer Res 25: 139.
Nakayalna T. Kirnura T, Kodama M. Nagata C (1983): Generation of hydrogen peroxide and superoxide
anion from active metabolites of napthylamines and arninoa~odyes: Its possible role in carcinogenesis. Carcinogenesis 4:765-769.
Nettescheim P. Topping DC. Jarnasbi R (1981): Host and environmental f'actors enhancing carcinogencsis in thc respiratory tract. Annu Rev Pharmacol Toxicol 21:133~-163.
Oberley LM'. Oherley TD. Buettner GR (1981): Cell division in normal and transformed cells; The
possible role of superoxide and hydrogen peroxide. Med Hypoth 7:21-42.
O b e r l e LM' (1982): Superoxide disrnutase and cancer. In Oberley LW (ed.): "Superoxide Dismutase."
Vol. 11. Boca Raton, FL: CRC Preas. pp 127.165.
O'Brien TG Simsiman KC. Boutwell RK 1975): Induction of thc polyamine biosynthetic enzymes in
mouse epidermis and their specificity for tumor promotion. Cancer Res 35:2426-2433.
O s u m ~T. Hashimoto T (1978): Enhancement of fatty acyl-CoA oxidizing activity in rat liver peroxisomes
by dl(?-ethylhexy1)phthaIatc. J Biochem 83: 1361-1365.
Packer JE. Slatcr TF. Willson RL (1978): Reactions of the carbon tetrachloride-relatcd pcroxy free
radical (CCI,O-) ) with amino acids: pulse radiolysis evidence. Life Sci 23:2617~-2620.
Paine AJ. Francis JE (1980): Thc induction of benzy(a)pyrenc-3-mono-oxygenase by singlet oxygen in
l ~ v e cell
r culture is mediated by oxidation products of histidine. Chern Biol Inter 30:343-353.
Park SD. Kim CG. Kim MG (1983): Inhibitors of poly(ADP-ribose) polymerase enhance DNA strand
hreaks. excision repair. and sister chromatid exchanges induced by alkylating agents. Environ
Mutagen 5:5 15-525.
Paterson MC. Bech-Hanscn NT. Blattner \VA. Fraumeni JF, Jr (1983): Survey of human hereditary and
familial disorders for ?-ray response in vitro: Occurence of both cellular radiosensitivity and radioresistance in cancer-prone families. In Nygaard OF. Si~nicMG (eda): "Radioprotectors and
Ant~carcinogcns."New York: Academic Press, pp 6 1 5 6 3 8 .
Pearl RG. Ruffin TA (1983): Niacin reduces oxygen toxicity in mouse alveolar macrophages. Pharmacol
27:219-222.
Peraino C. Fry RJM. Staffeldt E (1971): Reduction and enhancement by phenobarbital of hepatocarcinogencsis induccd in the rat by 2-acetylaminofluorcne. Cancer Res 3 I : 1506- 15 12.
Peraino C. Fry RJM. Staffeldt E. Christopher JP (1975): Comparitive enhancing effects of phenobarbital. arnobarbital. diphenyl hydantoin. and dichlorodiphenyltrichloroethane on 2-acetylaminotlurene-induced hepatic tumorigenesis in the rat. Cancer Rcs 35:2884-2890.
Pereira MA (1982): Rat liver foci assay. J Am Coll Toxicol 1: 101-117.
Petkau A (1980): Radiation carcinogenesis from a membrane perspective. Acta Physiol Scand Suppl
492:81-90.
Pitot HC. Goldsworthy T, Can~pbellHA. Poland A (1980): Quantitative evaluation of the promotion by
2.3.7.8-tetrachlorodibenzop-dioxinof hepatocarcinogenesis from diethylnitrosaminc. Cancer Res
40:3616-3620.
P n o r WA ( 1982): Free radical biology: Xenobiotics, cancer, and aging. Ann NY Acad Sci 393: 1-22.
Rao MS. LaIwani ND. Watanabe TK. Reddy JK (1984): Inhibitory cffect of antioxidants ethoxyquin and
Z~i)-t~.rr-butyl-4-hydroxyanisole
on hepatic tumorigcnesis in rats fed ciprofibrate, a peroxisomr
proliferator. Cancer Res 44: 10721076,
Rcddy JK. Rao MS (1978): Enhancement by W y l 4 , 6 4 3 , a hepatic peroxisome proliferator. of dicthylnitrosamine-initiated hepatic tutnorigenesis in the rat. Br J Cancer 38:537-543.
Reddj JK. Azarnoff D L (1980): Hypolipidernic hepatic peroxisome proliferators form a novel class of
chemical carcinogens. Nature 283:397-398.
Reddy JK. Laluani ND. Reddy MK. Qureshi SA (1982): Excessive accurnulation of autofluorescent
lipofuscin in the liver during hepatocarcinogenesis by methyl clofenapate and other hypolipidemic
peroxisome proliferators. Cancer Res 42:259-266.
Repine JE, White JG, Clawson CC. Holmes BM (1974): The influence of phorbol myristate acetate on
oxygen consumption by polymorphonuclear leukocytes. J Clin Lab Med 83:911-920.
614
Kensler and Trush
Ronian-Franco AA (1982): Non-enzymatic extramicrosomal hioactivation of chemical carcinogens by
phagocytes: A proposed new pathway. J Theor Biol97:543-555.
Rous P, Kidd JG (1941): Conditional neoplasms and subthreshold neoplastic states: A study of tar tumors
in rabb~ts.J Exp Med 73:365-390.
Schmidt J, Bore DC (1976): Frcc radicals from p ~ ~ r i nnucleosides
e
after hydroxyl radical attack. Rad
Res 65:220-237.
Schultc-Frohlinde D (1983): Kinetics and mechanism of polynucleotide and DNA strand break formation. In Nygaard OF, Simic MG (eds): "Radioprotectors and Anticarcinogens." New York:
Academic Press, pp 53-7 1.
Searle AJ. Willson RL (1980): Glutathione peroxidasc: effect of superoxide. hydroxyl and bromine free
radicals on enzyme activity. Int J Rad Biol 37:213-217.
Segal A. Katz C . Van Duuren BL (1971): Structure and tumor pro~notingactivity of anthralin (1.8dihydroxy-9-anthrone) and related compounds. J Med Cheni 14: 1152-1154.
Shcr W. Friend C (1978): Breakagc of DNA and alterations in folded genomes by inducers of differencclls. Cancer Rea 38:841-849.
tiation in Friend erythrole~~kemia
Sies H. Cadcnas E (1983): Biological basis of detoxification of oxygen free radicals. In Jakoby W (ed):
"Biological Basis of Detoxification. New York: Academic Press. pp 181-2 1 1 .
Simic MG. Hunter EPL (1983): Interaction of free radicals and antioxidants. In Nygaard OF, Simic MG
reds): "Radioprotectors and Anticarcinogens." New York: Academic Press, pp 449-460.
S l a y TJ. Bracken WM. Viaje A, Berry DL. Fischer SM. Miller DR, Leven W. Conney AH, Yagi H,
Jcrina DM (1978): Tumor initiating and promoting activities of various bcnzo(a)pyrene metabolites in mouse <kin. In Jones PW, Freudenthal RI (eds): "Carcinogcnesis-A Comprehensive
Survey." Val. 3. New York: Raven Press. pp 371-382.
Slaga TJ, Jcckcr L. Bracken WM. Weeks CE (1979): The effects of weak or noncarcinogenic polycyclic
hydrocarbons on 7.12-dimethylbenz(a)anthracene and benzo(a)pyrene skin tumor initiation. Cancer Lett 7:51--59.
Slaga TJ. Fischer SM. Nelson K, Glrason GL (1980a): Studies on mechanism of action of anti-tumorpromoting agents: Evidence for several stages in promotion. Proc Natl Acad Sci USA 77:3659-3663.
Slaga TJ. Klein-Szanto AJP, Fischer SM. Weeks CE, Nelson K . Major S (1980b): Studies on mechanism
of action of anti-tumor-promoting apents: Their specificity in two-stage promotion. Proc Natl
I 77:2251-2254.
Acad S ~USA
Slaga TJ. Klein-Szanto AJP. Triplett LL, Yotti LP. Trosko JE (1981): Skin tumor promoting activity of
benzoyl peroxide. a widely used free radical generating compound. Science 213: 1023-1025.
Slaga TJ. Fischer SM. Weeks C E , Nelson K , Mamrack M , KIein-Szanto AJP (1982): Specificity and
mcchanism(a) of promoter inhibitors in multistage promotion. In Hecker E, Fusenig W. Kunr W.
Marks F. Thielmann H (eds): "Carcinogenesis-A Comprehensive Survey." Vol. 7, New York:
Ravcn Press. pp 19-74.
Slaga TJ. Solanki V. Logani M (1983): Studies on the mechanism of action of antitumor promoting
apents: Suggestive evidence for the involvement of free radicals in pronlotion. In Nygaard OF,
Simic MG (eds): "Radioprotectors and Anticarcinogena." New York: Academic Press, pp 471485.
Slaga TJ. Fischer SM (1983): Strain differences and solvent effects in mouse skin carcinogenesis
experiments using carcinogens, tumor initiators and promotera. Prog Exp Tumor Res 26:85109.
Solanki V. Rana KS. Slaga TJ (1981 ): Diminution of mouse epidermal superoxide dismutase and catalase
activities by tumor promoters. Carcinogcnesis 2: 1141-1146.
Solanki V. Yotti L. Logani MK. Slaga TJ (1984): The reduction of tumor initiating activity and cell
mediated mutagenicity of dimethylbmz[a)anthracene by a copper coordination compound. Carcinopcnesis 5: 129- 13 l .
Stier A. Clauss R. Luckc A, Reitz 1 (1980): Redox cycle of stable mixed nitroxides formed from
carcinogenic aromatic arnines. Xenobiotica 10:661-673.
Stocha SJ. Hassan MQ. Murray WI (1983): Lipid peroxidation aa a possible cause of TCDD toxicity.
Biochem Biophys Res Comm 11 1 :854-859.
Sugimura T. Miwa M (1983): Poly(ADP-ribose) and cancer research. Carclnogenesis 4: 1503-1506.
Takahashi T ( 1977): Biochemical studies on phthalic esters. 11. Effects of phthalic esters on mitochondrial
respiration of rat liver. Biochem Pharrnacol 26: 19-24.
Tangeras A. Malviya AN (1983): Oxygen uptake, ATPase activity. and superoxide dismutase activity in
isolated rat liver mitochondria are not influenced by the tumor promoter 12-0-tetradecanoylphorhol-13-acetate. Carc~nogenesis4:509-512.
"
Oxygen Radicals in Tumor Promotion
615
Terada M. Nadel Ll. Fibach E. Ritkind RA. Marks PA (1978): Changes in DNA asmciated with
induction of crythroid differentiation by dimethyl s~~lfoxide
In rnurlne erythroleukernia cells.
Cancer Res 38:835-840.
Thaw HH. Hamberg H . Brunk UT (1983): Acute and irreversible injury following exposure of cultured
cells to reactive oxygen. Eur J Cell Biol 31:46-54.
Ticn M. Svingen BA. Aust SD (1982): Superoxide dependent lipid peroxidation. Fed Proc 40: 179182.
Topping DC. Ncttescheini P (1980): Two-stage carcinogeneGs atudies with asbestos. J Natl Cancer Inst
65:627-630.
Trager W F (1982): The postenzymatic chemistry of activated oxygen. Drug Metab Rev 13:51-69.
Troll W. Witz G. Gnldstein B. Stone D. Sigirnura T (1982): The role of free oxygcn radicals in tumor
promotion and carcinogenesis. In Hccker E. Fusening NE. Kunz N!. Marks F. Thiclrnann HN!
(eds): "Carcinogenesis-A Comprehensive Survey." Vol 7. New York: Raven Prcss. pp 593597.
Trush MA. Mimnaugh EG. Gram TE (1982): Activation of pharmacologic agents to radical intermcdiare<: Implications for the role of free radicals in drug action and toxicity. Biocheln Pharmacol
3 1 :33?5-3346.
Trush MA. Sccd JL. Kcnsler TW (1984): Activation of benzo[a]pyrene-7.8-dihydrodiol by reactive
ox!gen penerated by phorbol ester-stimulated human polyniorphonuclear leukocytes (PMNs).
Proc 9th Intern Cong Pharmacol. in press.
Turrens JF. Freeman BA. Lcvitt JG. Crapo JD (1982n): Thc effect of hypcroxia on superoxide
production by lung submitochondrial particles. Arch Biochem Biophya 217:401-410.
Turrens JF. Freeman B.4. Crapo JD (1982b): Hyperoxia increaxs H 2 0 , release by lung mitochondria
and microscxnes. Arch Biochcrn Biophqs 217:411-421.
Ts'o POP. Caspary WJ, Lorentzen RJ (1977): The invol~ernentof free radicals in chemical earcinogenesis. In Pryor WA (ed): "Free Radicals in Biology." Volume 111. New York: Academic Press. pp
251-303.
Uchigata Y. Yamamoto H. Kawamura A, Okamoto H (1982): Protection by superox~de dismutase
catalase. and poly (ADP-ribose) synthetase inhibitors against alioxan and streptozotocin-induced
islet DNA \trand breaks and against the inh~bitionof proinwlin synthesis. J Biol Chem 267:60846088.
Van Duuren BL. Nclson N. Orris L. Palrnes ED. Schmitt FL (1963): Carcinogenicity of epoxidcs.
lactones. and peroxy compounds. J Natl Cancer Inst 31:41-55.
Van Herllrnen JJ. hleuling WJA (1975): Inacti~ationof hiologically active DNA y-ray-induced superoxide ri!dicai, and their dismutation products, singlet molecular oxygcn and hqdrogen peroxide.
Biochim Biophks Acla 402: 133-141.
Vesley DL. Watson B. Levey GS (1979): Activation of liver guanylate cyclase by paraquat: Possiblc
role of wperoxide anion. J Pharmacol Exp Ther 209: 162-164.
Von Sonntiig C. Hagen U. Schon-Bopp A. Schulte-Frohlinde D (1981): Radiation-induced atrand brcaks
in DNA: Chemical and enzymatic analysis of end groups ancl mechanistic aspects. Adv Rad Biol.
9: 109- 142.
Ward JM. Rice JM. Creasla D. Lynch P. Riggs C (1983): Dissimilar patterns of prornorion by di(2ethylhexy1)phthalate and phenobarbital of hepatocellular neoplasia initiated by diethylnitrosanline
in B6C3F1 nlice. Carcinogenesis 4: 10211029.
Warren JR. Lalwani ND. Rcddy JK (1982): Phthalate esters as pcroxisome proliferator carcinogen\.
Environ Health Perspect 45:35-40.
Wattenberg LW. Lain LKT (1981): Inhibition of chemical careinogencsis by phenol\. coumarins.
aromatic isothiocyanates. flavones. and indolcs. In Zedcck MS. Lipkin M (eds): "Inhibition of
Tumor Induction and Development." New York: Plenum Prcss. pp 1-22.
Wattenberp LW. Lam LKT (1983): Phenolic antioxidants as protective agents in chemical carcinogene\ls. In Nggaard OF. Simic MG (cds): "Radioprotectors and .-2nticarcinogens." New York:
Academic Press. pp 461-496.
Weiss SJ. Lampert hlB. Tcst ST (1983): Long-lived oxidants generated by human neutrophiles: Character~rationand bioactivity. Science 222:625-628.
N!eitberg AB. Weitzman SA. Destrernpes M. Latt SA. Stossel T P (1983): Stimulated human phagocytes
produce cytogenetic changes in cultured mammalian cells. N Engl J Med 308:26-30.
Weitzrnan SA. Stohael T P (1981): Mutation caused by human phagocytes. Science 212:546-547.
Weitzman SA. Graceffa P (1984): Asbestos catalyzes hydroxyl and superoxidc radical generation from
hydrogen peroxide. Arch Biochem Biophys 228:373-376.
616
Kensler and Trush
Wertz PW. Mueller G C (1980): Inhibition of 12-0-tetradecanoylphorbol13-acetate-accelerated phospholipid ~nctabolisniby 5.8.11.14-eicosatetraynoic acid. Cancer Res 40:776-781.
White AA. Crawfbrd KM, Patt CS Lad PJ (1976): Activat~onof solublc guanylate cyclase from rat lung
incubation or by hydrogen peroxide. J Biol Chem 25 1 :7304-73 12.
White. R.E. and Coon. M.J.. Oxygen activation by cytochrome P-450. Ann Rev Biochem 49:315-356.
1980.
Willson RL (1977): Chemically reactive species in biology - Free"! radicals and electron transfer in
biology and mcdicinc. Chem Indust 1 8 3 1 9 3 .
Willson RL (1982): Iron and hydroxyl frcc radicals in cnzymc inactivation and cancer. In MacBrien
DCH. Slatcr T F (eds): "Free Radicals, Lipid Peroxidation and Cancer." New York: Academic
Prcss. pp 275-303.
Wilson M. Trush MA. Van Dyke K (1978): Induction of chemiluminescence in human neutrophila by
the calciun~ionophore A23187. FEBS Lctt 94:387-390.
Wit7 G. Goldstein BD. Ainoruso M. Stone DS. Troll W (.1980): Rctinoid inhibition of superoxide anion
radical production by human polymorphonuclear leukocytes stimulated with tumor promoters.
Biochern Biophys Res Commun 97:883-888.
Wr~ghtonSA. Mueller G C (1982): Rapid acceleration of deoxyglucose transport by phorbol esters in
bovine lymphocytcs. Carcinogenesis 3: 14 15- 1418.
Yamada M. Miwa M. Sugimura T (1971): Studies on poly(adenosinc diphosphate-ribose): X . Properties
of a partially purified poly(adenosine diphosphate-ribose) polymerase. Arch Biochcm Biophys
146:579-586.
Yamanloto H. Uchigata Y. Okamoto H (1981): Streptozotocin and alloxan induce strand breaks and poly
(ADP-ribose) synthetase in pancreatic islets. Nature 294:284-286.
Yannarelli A. Awad AB (1982): Thc effect of alteration of nuclear lipids on messenger RNA transport
from isolated nuclei. Bioche~nBiophys Rcs Cornrnun 108: 1056-1060.
Yavelow J. Gidlund M. Troll W (1982): Protease inhibitors froni proccsscd legumes effectively inhibit
superoxide gencratlon in response to TPA. Carcinogenesis 3: 1 3 5 1 3 8 .
Zatuchni GI. Osborn CK (1981): Gossypol: A possible male anrifertility agent, report of a workshop.
Rcs Front Fertil Reg 1:l-15.
Zinimernian R. Cerutti P (1984): Active oxygen acts as a promoter of transformation in mouse embryo
C3H IOT1/2/C18 libroblasts. Proc Natl Acad Sci USA. 81:2085-2087.