Drinking Water and Health,: Volume 6 (1986)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 Mechanisms of Carcinogenesis The public health effects resulting from reductions in exposures to various drinking water contaminants can be predicted with greater accuracy as the mechanisms underlying those effects become better understood. Among the possible chronic effects of concern to regulators faced with the task of estimating risk of such exposures, cancer ranks foremost. Carcinogenesis is a complex, multistep process that has been extensively reviewed by Becker (1981), Farber (1982), Farber and Cameron (1980), Slaga et al. (1980b), and Weinstein et al. (19841. This chapter provides an overview of principles that should be considered when assessing risk of exposure to drinking water. It is not intended to be comprehensive. THE MULTISTAGE THEORY OF CARCINOGENESIS Cancer is the product of a process involving complex interactions be- tween environmental and endogenous factors. It is usually manifested by the uncontrolled proliferation of cells that have sustained heritable alter- ations. The discovery that many carcinogens interact with DNA and thus alter the genotype, i.e., specific DNA sequencing of encoded information, is important to the development of the current theories of carcinogenesis. It has also been learned that the inheritance of a single mutation (i.e., a gene with altered DNA) may not be sufficient to produce cancer (Farber, 1982; Weinstein et al., 19841. In the human body, many millions of cells are at risk, and many of them can be shown to have DNA lesions; however, few cells give rise to malignant tumors. When DNA is damaged, the body responds with the cellular mechanisms of repair or eliminates the aberrant 139

]40 DRINKING WATER AND HEALTH cell through immune-surveillance mechanisms. This process provides pro- tection against both exogenous and endogenous mutagens and carcinogens. Cellular mutation is often an early stage in a multistage process. Carcinogenesis has been experimentally demonstrated to be a multistage process in the cells of certain animal tissues, including skin, lung, liver, and bladder. This process is believed to occur in much of human tumor- igenesis as well. According to current theories, at least three such stages (initiation, promotion, and progression) are evident in many experimen- tally induced cancers (Farber, 1984a,b; Slaga, 1983; Weinstein et al., 19841. These stages are phenomenological, and their mechanisms of action are not well understood. The distinction between the stages has been defined experimentally. Each stage appears to be influenced by several exogenous and endogenous factors, such as age, sex, diet, metabolic activity, and the dose and type of xenobiotic substance to which the . . OrgaIllSm IS exposer . The early work of Rous and Kidd (1941), Berenblum (1941), and Mottram (1944) demonstrated that cancer could be induced in experimental animals in two steps: initiation and promotion. Subsequent studies in animals showed that sequential induction of cancer can occur in a number of tissues or organs. Epidemiological evidence on epithelial cancers col- lected by Armitage and Doll (1954) also suggested a multistage process in carcinogenesis. This was later supported by the toxicological studies of Boutwell (1964), Slaga et al. (l98Ob), and Van Duuren et al. (19731. Consequently, the initiation-promotion model has been generally accepted as being representative of tumor induction. initiation Initiators are mutagens that act either directly or indirectly by forming electrophilic species that interact with and modify DNA structure, or otherwise damage the DNA sequence, but do not by themselves induce tumor formation. Initiation is believed to cause a lesion that persists over a long period, as demonstrated by Van Duuren et al. (1975), who showed that mouse skin initiated more than 1 year before treatment with phorbol esters is still very susceptible to tumor induction. Thus, the initiation step is considered to be irreversible. In addition to demonstrating this, Boutwell (1964) showed that repeated doses of an initiator were additive in the number of tumors produced. Promotion A promoter is a substance that usually does not induce a carcinogenic response by itself but that results in a carcinogenic response when applied

Mechanisms of Carcinogenesis 141 in multiple doses following a single, subcarcinogenic dose of an initiator. This temporal sequence of administration is only demonstrable in the laboratory; such distinctions are difficult to demonstrate in humans, who receive simultaneous environmental exposures to many types of chemicals. Some promoters can also show weak initiating activity at high doses. Unlike initiators, promoters do not form electrophilic species that interact with DNA. Some evidence indicates that promotion itself involves several stages and that it may be possible to characterize a promoter as a complete or a first- or second-stage promoter (Furstenberger et al., 1983; Slaga et al., 1980a). The effects of a first-stage promoter are thought to be re- versible; i.e., if promoter administration is terminated, a carcinogenic response is not produced. Administration of a second-stage promoter pro- duces irreversible effects. A compound capable of acting as both an initiator and a promoter in the same tissue is defined as a complete (whole) carcinogen. Most chem- icals that seem to have performed as initiators appear to be complete carcinogens. Progression The period during which ill-defined stages lead from benign tumor to malignant tumor is called progression. The transformation of neoplastic cells to a malignant tumor during this stage may involve several steps, such as oncogene activation (Weinberg, 1985), chromosome aberration (Weinstein et al., 1984), interaction between tumor cells and host defenses (Kripke and Morison, 1985), and various selection processes (OSTP, 1985~. Progression can be considered a dynamic process, since tumors may continue to increase in their degree of malignancy and heterogeneity (Weinstein et al., 19841. Cocarcinogenesis Cocarcinogenesis is the process by which two or more compounds, when administered concurrently, increase the risk of tumor development. In some cases, a cocarcinogen is not carcinogenic by itself, but enhances the carcinogenic potency of an initiator. In other cases, both compounds are carcinogenic by themselves, but together elicit a response that is greater than that expected on the basis of simple additivity. Cocarcinogens differ from promoters by definition because promoters are administered after initiators and are not usually carcinogenic alone. Some compounds may be both cocarcinogens and promoters. However, not all tumor promoters are cocarcinogens and not all cocarcinogens are tumor promoters, sug-

|42 DRINKING WATER AND H"LTH gesting that promotion and cocarcinogenesis proceed by different mech- anisms. ONCOGENE ACTIVITY Oncogenes are naturally occurring genes that code for factors that reg- ulate, among other things, cellular growth. They have been identified in several human tumors as well as in spontaneous and xenobiotically induced tumors in animals. The most recent studies on the mechanisms of cell interactions in cancer induction have attempted to elucidate the role of oncogenes in carcinogenesis (Weinberg, 19851. In the last 4 years, such research has led to several key discoveries that strongly support the mul- tistage theory of carcinogenesis (Land et al., 1983; Slamon et al., 19841. Approximately 40 oncogenes have been discovered (Weinberg, 19851. Most of these have been operationally classified as immortalizing genes (myc type) and transforming genes (ras type). Recent studies have shown that rag-type oncogenes can be activated by chemical carcinogens and by ultraviolet light (Sukumar et al., 1983~. Immortalization of cells in culture has been carried out using chemical carcinogens, and these cells have been transformed and promoted by oncogene products and xenobiotic substances. Studies have shown that the DNA from chemically induced tumors contains active oncogenes, and recent research has demonstrated the potential importance of chromosome translocations in oncogene ac- tivation (Bishop, 1982; Land et al., 1983; Leder et al., 1983~. Immor- talization (presence of a myc-type oncogene) and transformation (presence of a rag-type oncogene) may be regarded as the biological counterparts of initiation and promotion. Recent findings have identified a third type of cancer gene (not yet classified) that may cause a cancer cell to metas- tasize (Bernstein and Weinberg, 19851. This finding of a unique oncogene in a metastatic tumor strongly supports the multistage model of carcino- genesis, since it indicates that different types of oncogenes may correspond to different stages of carcinogenesis. The literature on the specificity of oncogenes and their activation, their location in spontaneous and induced tumors, and the ability of transformed cells in culture (activated ras gene) to induce tumors in viva further in- dicates that tumor induction is indeed a multistage event. To reflect these experimental findings, risk modeling must consider at least three, if not more, stages in order to be consistent with the experimental and human evidence. In particular, the observation that certain promoting agents can induce chromosome aberrations (including translocations) and aneuploidy is of interest in view of the fact that both translocations and aneuploidy have been observed in certain animal and human neoplasms, such as Burkitt's

Mechanisms of Carcinogenesis 143 lymphoma, retinoblastoma, and Wilms' tumor (Barrett et al., 1983; Cairns, 1981; Leder et al., 1983; Tsutsui et al., 1983; Yunis, 1983~. Increased risk of tumor development has also been observed in patients with con- genital aneuploidy, such as Down's syndrome (Windham et al., 19851. Recent research has illustrated the potential importance of chromosome translocations in oncogene activation (Bishop, 1982; Land et al., 1983; Leder et al., 19831. MODEL SYSTEMS Mouse Skin The mouse has served as an especially useful test animal, since tumor induction in its dorsal skin is relatively rapid and can easily be observed and quantitated without sacrificing the animal. For 30 years, up to the late 1960s, the mouse skin model was used almost exclusively for initi- ation-promotion studies in chemical carcinogenesis. For nearly 40 years, croton oil (an oil obtained from the seeds of Croton tiglium L.) was used as a promoting substance in mouse skin studies. The active component was identified in the late 1960s as phorbol myristate acetate (PMA) (Hecker, 1971; Van Duuren, 19691. Since that time, several PMA analogs have been synthesized. Although these PMA analogs have less promoting activity, they are important in studies designed to determine the mechanism of action of phorbol ester promoters (Boutwell, 19741. In addition to the phorbol ester tumor promoters, there are other classes of compounds with promoting activity in the mouse skin. Dihydroteleo- cidin B. a derivative of a natural product isolated from streptomyces, has strong promoting activity in this model (Fujiki et al., 1981), as do certain natural products other than the phorbol esters extracted from plants (Muf- son et al., 19791. Phenol and certain phenol derivatives were shown by Boutwell and Bosch (1959) to be weak promoters in skin, as are certain fatty acids and fatty acid methyl esters and long, straight-chain alkanes such as decane (C-10) and tetradecane (C-14) (Arffmann and Glavind, 1971; Van Duuren and Goldschmidt, 19761. Anthralin (1,8-dihydroxy-9- anthrone) was shown by Segal et al. (1971) to be a strong promoter in mouse skin, whereas detergents such as sodium lauryl sulfate and Tween 60 are weak promoters (Boutwell, 1964; Setala, 19601. A class of tumor promoters discovered recently are peroxides such as benzoyl peroxide, which is a moderately strong promoter in mouse skin (Slaga et al., 19811. Direct-acting alkylating agents such as iodoacetic acid and 1-fluoro-2,4- dinitrobenzene are weak to moderate promoters (Bock et al., 1969;.Gwynn and Salamon, 19531. Benzofe~pyrene, which is not a whole carcinogen in skin, is a moderate promoter in that tissue (Slaga et al., 1979), whereas

]44 DRINKING WATER AND HEALTH 7-bromomethyl benz~ajanthracene is a complete carcinogen and also a strong promoter (Scribner and Scribner, 19801. Suganuma et al. (1984) have recently identified a new class of tumor promoters, aplasiatoxins, which are structurally unrelated to PMA. A number of chemicals that are tumor promoters in mouse skin are also cocarcinogens in mouse skin. The existence of cocarcinogenic substances was inferred from studies of carcinogenesis resulting from exposures to cigarette smoke condensate (CSC). Since the concentration of benzota~pyrene [B(a)P] in CSC and the dose delivered in CSC skin-painting experiments could not account for the observed tumor yield, it was surmised that other substances present in CSC enhanced the response to B(a)P (Hecht et al., 19811. Van Duuren and Goldschmidt (1976) showed that catechol (1,2- dihydroxybenzene), which is an abundant phenol in CSC, is a potent cocarcinogen in B(a)P-induced skin carcinogenesis. Thus when low car- cinogenic doses of B(a)P were applied to the skin repeatedly over a long period in the presence of catechol, the number of skin tumors increased significantly compared with the number induced by treatment with B(a)P without catechol. Treatment with catechol in the absence of B(a)P did not cause tumor induction. Van Duuren and Goldschmidt (1976) also showed that PMA is a cocarcinogen as well as a promoter. In addition to serving as a model system for identifying whole chemical carcinogens, initiators, and promoters, the mouse skin has been extremely useful for studying the mechanism of action of initiators and promoters since the early 1970s. At the tissue level, phorbol ester promoters cause hyperplasia and inflammation (Boutwell, 19641. Cellular responses ob- served within 24 hours after exposure include increased synthesis of DNA, RNA, and protein (Baird et al., 1971), increased phospholipid turnover (Rohrschneider and Boutwell, 1973), changes in cyclic nucleotide me- tabolism (Mufson et al., 1979), increases in protease activity (Troll et al., 1978), induction of ornithine decarboxylase—a key enzyme in polyamine metabolism (O'Brien et al., 1975), and decreases in the activities of the antioxidant defense enzymes superoxide dismutase and catalase (Solanl~i et al., 1981~. Largely as a result of these studies, early investigations on the inhibition of promotion focused on substances that might reverse the tissue and biochemical responses caused by promoters. Thus antiinflam- matory compounds such as cortisol, dexamethasone, and fluocinolone acetonide were found to be effective promotion inhibitors (Berman and Troll, 1972; Schwartz et al., 1977), as were protease inhibitors (Troll et al., 19781. Vitamin A derivatives (Verma et al., 1979) and free-radical scavengers such as dimethyl sulfoxide (DMSO) (Loewengart and Van Duuren, 1977) were also effective promotion inhibitors. In the mouse skin, the two-stage initiation-promotion protocol results in the formation of a large number of papillomas, many of which progress

Mechanisms of Carcinogenesis 145 to squamous cell carcinomas. Burns et al. (1976) identified two classes of papillomas one class that regressed after treatment with PMA was discontinued and another class that showed a decreased tendency to re- gress, i.e., a class of papillomas with autonomous growth. The greatest tendency to undergo malignant transformation occurred in the autonomous group. Klein-Szanto et al. (1983) have shown that the acquisition of ~y- glutamyltranspeptidase (GOT) and the loss of high-molecular-weight ker- atins appear to be good indicators of progression from benign to malignant tumor in mouse skin. Slaga et al. (1980a) demonstrated that the promotion phase in mouse skin can be subdivided into two stages. In the first stage, the initiated dorsal skin of mice is treated with a first-stage promoter such as PMA for 2 weeks. This treatment is followed by repeated treatments (18 weeks or longer) with a second-stage promoter (e.g., mezerein) (Slaga et al., 1980a,b). Treatment of initiated mouse skin with a first-stage promoter for 2 weeks without subsequent treatment with a second-stage promoter does not result in tumor induction, nor does treatment for a long time with a second- stage promoter without prior treatment with a first-stage promoter (Slaga et al., 1980a). In the two-stage promotion model, the complete promoter PMA is also a potent first-stage promoter. The weak promoter mezerein is a strong second-stage promoter, but the PMA analog 4-O-methyl-PMA, which is not a promoter in the two-stage carcinogenesis model, is a weak first-stage promoter (Slaga et al., 1980b). Furstenberger et al. (1983) have shown that the first stage of promotion was not reversible for at least 2 months when PMA was given as a first-stage promoter and was followed by 12-retinylphorbol-13-acetate (RPA). They also demonstrated that par- tial inversion can occur between initiation and the first stage of promotion. In their model, initiation with a subthreshold dose of dimethylbenzan- thracene (DMBA) could occur successfully up to 6 weeks after treatment with PMA. Rat and Mouse Liver This model system has in the past involved an invasive procedure (partial hepatectomy), followed by administration of the initiator [usually 2-ace- tylaminofluorene (2-AAF), diethylnitrosamine (DEN), or DMBA] and then by phenobarbital as the promoter. The interaction of phenobarbital with 2-AAF was first reported by Peraino et al. (1971), who established the promotional effect of phenobarbital in the liver. Recent results indicate that the rat and mouse liver models of carcinogenesis do not require use of the invasive technique and that dietary restriction of choline for less than 2 months after pretreatment with a cytochrome P450 modifier is sufficient to induce altered foci (Mylecraine, 19841.

]46 DRINKING WATER AND HEALTH Recent evidence suggests that intraperitoneal pretreatment of Fischer 344 adult male rats with Q-naphthoflavone, a 2,3,7,8-tetrachlorodibenzo- p-dioxin (TCDD)-type of P450 cytochrome inducer, enhances the pro- duction of altered foci by DEN (Mylecraine, 19841. A small percentage of these altered foci have been shown to progress to cancer (Williams and Weisburger, 19831. Mylecraine (1984) has shown that altered foci in rodent livers are enhanced when phenobarbital is injected intraperitoneally to the animals before initiators are administered. Thus, apart from in- creasing the production of reactive derivatives, some aspect of the pro- liferation induced by compounds known to induce one or more forms of cytochrome P450 may have an effect on the ability of the liver to respond to a carcinogen. Induction of uncontrolled cell replication is believed to be an important contributing factor in the process of cocarcinogenesis, but the role of proliferation in promotion is not entirely clear. Investigations of several organ systems indicate that cells in the process of normal replication are more susceptible to mutagenesis and carcinogenesis than are cells that are relatively dormant and replicate more slowly. For example, synchronized rat liver epithelial cells were shown to be most sensitive to mutation at the phosphoribosyltransferase locus by methyl methane sulfonate or by the gastric carcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MUNG) during DNA synthesis (Tong et al., 19801. Following partial hepatectomy in the B6C3F~ mouse, Newberne et al. (1982) observed a small increase in spontaneous liver tumors. Rats given a single injection of dimethylnitro- samine 24 hours after undergoing partial hepatectomy (the peak time of DNA synthesis in the regenerating liver) developed hepatocellular carci- nomas, whereas nonhepatectomized rats did not (Craddock, 19711. How- ever, the lack of proportionality between the rate of cell division and induction of tumors in carcinogen-exposed rat livers indicates that there are modulating factors beyond the rate of cell division (Becker, 19794. J. M. Ward et al. (1984) have demonstrated that continuous, long-term exposure is not necessary for tumor promotion in the mouse liver. For example, di(2-ethylhexyl) phthalate administered after initiation with DEN resulted in the same significant increase in altered hepatic foci after 24 or 84 days of exposure. See next section, entitled Other Animal Systems, for further examples. Studies in which 2-AAF is administered serve as excellent examples of the complexity of multistage (and multiorgan) carcinogenicity. This com- pound behaves as a promoter in the mouse bladder but as an initiator in the liver of the same animal (Hughes et al., 19831. Further evidence for tissue specificity has been provided by studies demonstrating that liver tumors in rodents are not promoted by PMA or by other phorbol esters but, rather, by polyaromatic hydrocarbons (PAHs), polychlorinated bi-

Mechanisms of Carcinogenesis 147 phenyls (PCBs), TCDD, and phenobarbital (Peraino et al., 1980; Pitot et al., 1980~. This finding is supported by evidence in humans for tissue- specific promotion by estrogens, cigarette smoking, and asbestos, the latter serving as a promoter of lung cancer development and as an initiator of mesothelioma (NRC, 1984, pp. 165-1991. Many inducers of hepatic cytochrome P450, e.g., phenobarbital, PCBs, dichlorodiphenyltrichloroethane (DDT), and TCDD, have been implicated as promoters in the rodent liver model. These agents may act as promoters through the generation of active oxygen molecules such as superoxide, hydroxyl radical, or peroxide, which may cause genetic effects, direct effects on the cell membrane through alteration of transport mechanisms, and immunotoxicity (see Kensler and Trush, 1984, for a review). Indeed, Dean et al. (1983a) have shown that 12-O-tetradecanoyl-phorbol-13- acetate (TPA) depresses cell-mediated immunity, T-cell function, and in viva tumor resistance. These responses are consistent with findings using other suspected promoters, e.g., diethylstilbestrol (DES), TCDD, and dimethylvinylchloride (DMVC) (Dean et al., 1983a). Other Animal Systems The discovery of experimental tumor promotion in other tissues, such as the bladder (Hicks, 1983; Verma et al., 1983), breast (Rogers, 1983; Wotiz et al., 1984), colon (Ready and Maeura, 1984; Rogers, 1983), lung (Witschi, 1983; Witschi et al., 1977), pancreas (Ohyama, 1985), and respiratory tract (Mossman et al., 1985), is relatively recent. Research in these areas has produced a great deal of information on possible mecha- nisms of carcinogenicity, including tissue specificity and a memory effect of promoters. Multistage carcinogenesis in the liver and skin has been more clearly defined (J. M. Ward et al., 1984) than in the lung, colon, brain, kidney, and thyroid. Estrogen-sensitive tissues (i.e., tissues with high levels of estrogen receptors) such as the endometrium, prostate, and breast appear to respond to estrogens and nonhormonal compounds that exhibit estrogen- like activity in a manner similar to the initiator-promoter (multistage) model (Baxter and Funder, 1979; Lippman and Allegra, 1978; McGuire et al., 1978; Wein and Murphy, 19731. Some promoters are active at only one site. However, others, such as 2-AAF, can act as an initiator in one tissue and a promoter in another (Hughes et al., 19831. Some agents can act as an initiator at one site, a promoter at another, and an inhibitor at a third. TCDD is an example of this phenomenon: in the mouse, it is a weak initiator in skin (DiGiovanni et al., 1977) and possibly in liver (Kociba et al., 19781; a promoter in liver (Pitot et al., 19801; and an inhibitor in skin (DiGiovanni et al., 1983)

|48 DRINKING WATER AND H"LTH and possibly in the pituitary, uterus, mammary glands, pancreas, and adrenal gland (Kociba et al., 1978~. The effect of a promoter may vary, depending on the sequence of administration with the initiator. For ex- ample, when given to mice before an injection of urethan, the antioxidant butylated hydroxytoluene (BHT) will decrease the numbers of lung ad- enomas induced. When administered after urethan, BHT increases tumor yield (Malkinson and Beer, 19841. The sensitivities of different species and strains to promoter activity vary in many cases; mice are the most sensitive, forming multiple skin papillomas in response to an initiation- promotion regimen, as compared to less frequent basal cell carcinomas in rats and melanomas in hamsters (Slaga and Fischer, 1983~. In light of these observations, it may be desirable to test potential promoters and inhibitors of promotion in viva in several species and strains and in both sexes, to examine several organs for the response, and to administer the compounds in different sequences. Studies in Humans Evidence for multiple stages in human tumor promotion results from analysis of epidemiological data on several cancers. For example, epi- demiological data on cigarette smoking and lung cancer have been used to develop a simple multistage model in which the incidence rate is pro- portional to timer, where k is the slope of the log-log age-incidence curve and is believed to approximate the number of stages necessary for tumor development (Peso, 19771. Cessation of cigarette smoking reduces the risk of lung cancer, although not to the level of risk for nonsmokers (Doll, 1978), and is thus believed to affect the penultimate stage of lung cancer development (Peso, 19771. Cigarette smoke contains compounds capable of both initiating and promoting lung carcinogenesis. In another example, liver cancer in Africa appears to be linked to both aflatoxin Be exposure and hepatitis B virus infection (Linsell and Peers, 1977), each of which may affect different stages in liver cancer development. The risk of lung cancer for asbestos workers who smoke is 50 times greater than that of nonsmokers (NRC, 19841. Asbestos appears to be acting both as a pro- moter in the lung and as an initiator of mesothelioma (NRC, 1984, pp. 165-1991. Tumors of the endocrine organs are linked to hormone avail- ability; the hormones apparently exert a promoting effect (Day and Brown, 1980; Sivak, 19791. GENETIC TOXICITY Several contaminants in drinking water react chemically with DNA. A few of these are reviewed in Chapter 9. This genetic toxicity is of great

Mechanisms of Carcinogenesis 149 concern since the information that controls the structure, function, and reproduction of cells is encoded in the DNA. Moreover, both theoretical considerations and experimental evidence indicate that alteration of the DNA is involved in at least one step in the complex process of carcino- genesis. Genotoxicity and, preferably, genetic toxicity are the general terms used to describe the niode of activity of an agent that itself or by way of its metabolites can interact with genetic material to induce heritable alterations in DNA sequence or chromosome number or structure in either somatic or germ cells. Mutagens are agents that can alter the primary base sequence of the DNA. Clastogenicity and aneuploidy refer to chromosome breakage or change in chromosome number, respectively. A distinction is sometimes made between genetic toxicants that act in different ways. In primary genetic toxicity, either the parent compound or its metabolite directly alters or binds to genetic material. For example, both methyl methane sulfonate (MMS) and B(a)P are primary genetic toxicants. MMS, however, is called a direct-acting agent because it interacts directly with DNA without the need for metabolic activation. In contrast, B(a)P is an example of a primary genetic toxicant that requires metabolic activation to exhibit genetic tox- icity. Secondary genetic toxicity refers to activity in which genotoxicity is a secondary result of the primary action of the agent. Examples of this type of toxicity are effects on the DNA polymerase, inhibition of DNA repair, induction of a physiological state that results in genetic toxicity, or forced cell proliferation, resulting in an increase in the frequency of spontaneous mutations. Some carcinogens are believed to act without affecting DNA. One example is DES, which can lead to a rare form of vaginal cancer in adolescent women who had been exposed transplacentally (Herbst et al., 1977~. DES may affect the differentiation of endocrine organs, leading to cancer later in life due to either hormonal imbalances or altered hormone receptor response (Weisburger and Williams, 1982~. Another example is asbestos, whose fibers do not seem to damage DNA directly (Fornace, 1982) or act as mutagens (Chamberlain and Tarmy, 1977) but which may transport PAHs (known initiators) into target cells because they adhere to asbestos fibers (Eastman et al., 19831. THE ROLE OF THE IMMUNE SYSTEM IN CARCINOGENESIS The immune system provides a major defense against invading micro- organisms and altered cells. It functions in overall host resistance to in- fections, in the maintenance of homeostasis, and in surveillance. against uncontrolled cell proliferation (Fidler, 1985~. The immune system pro- vides a natural defense against cancer. This conclusion is supported by

|50 DRINKING WATER AND HEALTH reports of increased incidences of some rare types of cancer among people with an impaired immune system, such as patients with acquired immu- nodeficiency syndrome (AIDS) (Bigger et al., 1985; Cairns, 19851. Fur- thermore, malignancies occur with enhanced frequency in patients on immunosuppressive therapy (McKhann, 1969; Penn, 19781. Although a variety of lymphoid cell types and macrophages are involved in immune response, evidence to date has indicated that the most effective host re- sponses leading to tumor cell elimination are those of the T cells (Schatten et al., 1984; Zoller, 19851. Neoplasms are frequently associated with the presence of a mononuclear cell infiltrate containing numerous macrophages that, in collaboration with B and T lymphocytes, can destroy neoplastic cells. Adams and Snyderman (1979) have suggested that tissue macrophages may recognize small col- onies of newly formed tumor cells, cluster about them, and destroy the nascent neoplasm. Recently, Fidler and Kleinerman (1984) have shown that human blood monocytes (phagocytic blood cells that become mac- rophages in tissue) destroy tumor cells but not normal cells when they are cultured together. This effect is related to immunosurveillance. The possibility that the immune system developed as a useful and effective mechanism for the early sensing and elimination of neoplastic cells was originally proposed by Thomas in 1959 and was later given the name immunosurveillance by Burnet (19701. Burnet's concept of im- munosurveillance was based on two assumptions: (1) that cancer cells might be arising in small clones, or specialized subpopulations, in one organ or another, all the time and (2) that something foreign might always be displayed at their surfaces indicating their alien nature. Both T cells and macrophages have been shown to participate in host immunosur- veillance. Studies in laboratory animals have demonstrated that the immune system is particularly sensitive to the toxic effects of xenobiotic substances. The cellular components of the immune system are derived from bone marrow stem cells, a population of very rapidly proliferating cells that constantly replenish the immunocytes in the peripheral blood and tissues. Rapidly dividing cells are believed to be more sensitive to mutations than those that divide more slowly or not at all. Both the cell-mediated and the humoral immune systems are affected by toxic chemicals. Suppression of immune responsiveness is typically manifested by a decreased resistance to viruses, parasites, bacteria, and tumor cell grafts (for a review, see Dean et al., 1984; NRC, 19771. Immune suppression has also been observed in people accidentally exposed to polybrominated biphenyls and PCBs (Bekesi et al., 1979; Chang et al., 1980; Shigematsu et al., 19781. Xenobiotic compounds may also induce

Mechanisms of Carcinogenesis 151 immunologic enhancement, resulting in autoimmune-type disorders and possibly hypersensitivity reactions. A variety of carcinogens and cocarcinogens have been shown to have profound effects on the immune system. Immune suppression has been reported following exposure of animals to B(a)P, ultraviolet irradiation, and DMBA (Dean et al., 1983b; Kripke and Morison, 1985; E. C. Ward et al., 1984~. These effects were manifested by decreased resistance to infections and tumor cell grafts and by decreased spleen weight and mar- row cellularity. Tumor-promoting agents such as TPA, DES, TCDD, DMVC, and urethan have been shown to decrease both the cell-mediated immune response and overall resistance to infections (Dean et al., 1983a; Luster et al., 1980a,b, 19821. However, phorbol ester tumor promoters have been reported to activate macrophages and to stimulate reactive oxygen release from macrophages (Laskin and Pilaro, 1985; Laskin et al., 1981; Pilaro and Laskin, 19841. In this case, stimulation of the immune response by tumor promoters may also lead to enhanced development of neoplasia. Thus both suppression and enhancement of the immune system by xenobiotic compounds and carcinogens may lead to increased cancer in- cidence. These effects may be manifested at any stage during carcino- genesis. CARCINOGEN EVALUATION Animal Bioassays Positive results in properly conducted animal bioassays are considered to be predictors of qualitative response in humans (IARC, 1980; NRC, 1977, 1983; NTP, 1984; OSTP, 1985; OTA, 19811. The scientific ratio- nale for this approach is simply that animals are the closest models to the human for cancer studies. In addition, many carcinogens produce cancer in several species, and all known human carcinogens have been shown to produce tumors in at least one animal model (NTP, 19841. Benzene and arsenic trioxide, the two former holdouts from this general rule, have now been shown to be carcinogenic in animals (Goldstein et al., 1982; Maltoni and Scarnato, 1979; Pershagen et al., 19841. For some chemicals (e.g., aflatoxin B~, DES, vinyl chloride, mustard gas, melphalan, and 4- aminobiphenyl), the positive results in experimental animals preceded the epidemiological evidence. The overall patterns of chemical metabolism are generally similar in humans and laboratory animals (Rail, 1979), although the rates of metabolism and the type and site of cancer may

|52 DRINKING WATER AND H"LTH differ (IRLG, 1979; OTA, 19811. For example, the metabolism of B(a)P is qualitatively the same in all species and systems studied (Sims, 19761. At present, the most widely used data for carcinogen evaluation and regulation are those derived from animal bioassays. Alternative approaches to identifying carcinogens, such as short-term tests and the study of struc- ture-activity relationships, are generally viewed as supplemental to but not as substitutes for long-term bioassays (NTP, 1984; OSTP, 19851. These two approaches are discussed in later sections of this chapter. The standards for experimental design and conduct as well as for the analysis and reporting of data from animal bioassays have been well described (IARC, 1980; see OSTP, 1985, for a review). Subjects of extensive discussion have been the necessity of using high doses or the maximum tolerated dose (MTD) to maximize the sensitivity of the bioassay and the validity of this approach (Food Safety Council, 1980; NRC, 1977; NTP, 1984; OSTP, 1985; OTA, 19811. The MTD has been operationally defined as the dose that will neither reduce longevity nor inhibit the maximum growth or weight gain of the experimental animals by more than 10%. For some materials that produce important deleterious effects without causing serious weight loss, however, the MTD must be related to the other toxic responses and normal longevity of the animal (NTP, 19841. Recently, two expert panels have reaffirmed the use of MTD as a valid and necessary procedure (NTP, 1984; OSTP, 1985~. There are several important limitations to animal bioassays. For ex- ample, the test animals are usually homogenous, healthy animals exposed to a single agent, whereas the human population is genetically heterog- enous with widely varying susceptibilities to disease and multiple expo- sures to carcinogenic substances. Moreover, the assays do not necessarily predict the target site of action in humans. For example, 2-naphthylamine induces bladder cancer in humans, monkeys, dogs, and hamsters, but hepatic cancer in rats (IARC, 19771. These problems are discussed further in Chapter 8 in the context of quantitative risk assessment based on animal bioassay data. In addition, animal bioassays are time-consuming, difficult to conduct, and expensive. For example, a well-conducted inhalation study usually requires 3 to 4 years from the planning stage to completion of the final report and can cost in excess of $500,000. Despite their limitations, however, bioassays are currently the most acceptable way to predict effects in humans when, as is usually the case, there are no adequate-epide- miological data on humans (OSTP, 19851. Short-Term Tests Short-term tests can be completed in only a fraction of the time it takes to complete traditional 2-year bioassays in rodents. They can be used to

Mechanisms of Carcinogenesis ~ 53 determine the mutagenicity of chemicals in in vitro systems of bacteria and mutagenicity or transformation in mammalian cells in culture. They can also be used to study very specific end points, such as DNA binding or sister chromatic exchange, in in viva systems following limited exposure to a chemical. Short-term assays often can provide information regarding the putative carcinogenicity of a chemical; that is, a positive response in a short-term test can indicate the likelihood of a positive response in a · . · . c Come carclnogenesls bioassay. Numerous assays can be conducted to obtain a measure of the genetic toxicity and potential carcinogenicity of a chemical (Bridges et al., 1982; Butterworth, 1979; Hollstein et al., 19791. Various combinations of these assays are currently used in batteries or tiers by industry and government agencies to characterize chemicals regarding potential carcinogenicity and to assist in regulatory decision making (EEC, 1979; EPA, 1984; OECD, 1981; OTA, 19811. The theories supporting the use of such short-term tests have been reviewed in detail (Ashby, 1983; de Serres and Ashby, 1981; IARC, 1980, 1982; OSTP, 19851. In the widely accepted multistage theory of carcinogenesis, genetic toxicity occurs during the early or initiating phase of neoplasia. Such end points are of considerable value in detecting initiating and complete car- cinogens, which are believed to exert their effect, at least in part, by interacting with genetic material. However, these assays are limited in their ability to detect events such as promotion, which may also be of considerable importance in the process of carcinogenesis. As more is learned about the molecular biology of cancer, new assays that measure other events in the process of carcinogenesis are becoming available. For example, cell transformation assays measure the ability of chemicals to convert a normal cell to a preneoplastic state or to convert a preneoplastic cell to a tumorigenic state (Barrett et al., 1984; Heidelberger et al., 19831. Chemicals that participate in carcinogenesis by means other than altering genetic material have been termed nongenetic, or epigenetic, carcinogens. These chemicals can include promoters, which accelerate the multistage process of carcinogenesis. Some cell culture assays, such as the cell transformation assay, may be useful in detecting promoting activity (Aber- nethy et al., 1984; Frazelle et al., 1983), but there is still a need for short- term tests for promotion (IARC, 1983; NTP, 1984; OSTP, 1985~. Short-term tests are highly complex, sometimes technically demanding procedures for measuring the molecular effects of chemicals on cells. Skilled personnel are required to perform them properly and to interpret the results (Butterworth, 19811. For example, cytotoxicity can have a profound effect on the end points measured. Thus, experiments must be designed so that results are not artifacts of excessive cytotoxicity. In many cases, a large amount of data is generated, especially for chemicals of

}54 DRINKING WATER AND HEALTH interest to the general public (see, for example, EMIC, 1986~. Occasion- ally, there is inconsistency between the results of different tests. The reasons for varying responses must be examined and reconciled. When testing a known carcinogen, investigators may look for and give weight to any positive response at the expense of a more balanced view of the data. Because chemicals may act by different mechanisms, a battery of tests is required for an appropriate evaluation of the genetic toxicity of a chem- ical. Assays vary in complexity from tests on bacteria that are exquisitely sensitive to mutagens (McCann et al., 1975) to measurements of genetic toxicity in the treated animal or cells from exposed individuals (Bridges et al., 1982~. Results from bacterial assays alone are rarely sufficient to classify a chemical as genetically toxic or not and should generally be confirmed in mammalian cells. For example, vitamin C is positive in the Salmonella mutation assay under certain conditions (Norkus et al., 1983), but TCDD, a potent carcinogen, is not (Geiger and Neal, 19811. Thus, knowledge of the principal mechanism of action of a chemical can be critical in determining which short-term tests will have meaningful pre- dictive value. This consideration could be important for assessing the safety of a substance, identifying noncarcinogenic analogs of a compound for use in new products, or choosing alternatives to currently used car- cinogens. For example, data from promotion assays would be far more valuable than mutagenicity data in searching for a noncarcinogenic analog of TCDD. In reality, however, the mechanism of action of chemicals is generally not known. Both false-negative and false-positive results may occur when using short-term tests, because they do not reflect the complexity of interactions in the whole animal. In the absence of an appropriate metabolic activating system, false negatives can occur. Measurements in the whole animal are particularly useful because the results are influenced by such inherent factors as metabolism, distribution, excretion, and repair (Ashby, 19831. One must be concerned that the events measured in vitro truly reflect the critical events that occur in the animal. The importance of this concern is illustrated by the effects of nitroaromatic carcinogens. The potent he- patocarcinogen technical-grade dinitrotoluene (DNT), as well as the in- dividual DNT isomers, are weakly mutagenic in the Salmonella mutation assay without the need for an added metabolic activation system (Couch et al., 1981~. No activity is seen, however, in cell culture assays, including those of metabolically competent primary hepatocytes (Bermudez et al., 19791. The weak activity in bacteria bears no resemblance to the complex pattern of activation in the whole animal, which involves the enterohepatic circulation and sequential steps of metabolism by the liver, gut flora, and the liver once again (Rickert et al., 19841. Only when genetic toxicity,

Mechanisms of Carcinogenesis 155 forced cell proliferation, and promotion are measured in the whole animal do the results correlate with the striking differences in carcinogenic potency of the various isomers and the sex-specific susceptibility to the carcino- genic action of DNT (Rickert et al., 19841. When conducting short-term tests to determine the potential carcinogenicity of nitroaromatic com- pounds, it is mandatory to examine the effects in the whole animal, because the presence of gut flora is obligatory for the metabolic activation of this class of chemicals (Doolittle etal., 1983; Mirsalis et al., 19821. GERM CELL MUTAGENESIS Genetic toxicants must be regarded as potential germ cell mutagens and, thus, as threatening irreversible damage to the human gene pool. Although there is as yet no documented case of chemically induced her- itable genetic disease in humans (Mohrenweiser and Neel, 1982), epi- demiological and animal studies suggest that such an association may exist (Strobino et al., 19781. This potential association is of particular concern when chemical exposure occurs through contaminated drinking water, since exposure may be chronic throughout the reproductive years. Systems for evaluating human germ cell mutagens are evolving. Cig- arette smoke is a potent human carcinogen and is the most important factor in more than 120,000 lung cancer deaths annually in the United States (Cairns, 1975; Mommsen and Aagaard, 1983; Silverberg, 19841. It con- tains so many mutagens that smokers have mutagenic urine (DeMarini, 1983; Yamasaki and Ames, 19771. There is also an increase in mor- phologically abnormal sperm in smokers, and smoking during pregnancy increases the incidence of spontaneous abortion (Evans, 1982; Kline et al., 19771. Thus, there is reason to suspect that cigarette smoke may be a germ cell mutagen. Nonetheless, there is as yet no evidence directly linking the induction of heritable mutations in humans to cigarette smoking (Bridges et al., 19791. Whether these negative results reflect the inade- quacy of current germ cell mutagenicity assays or the relative mutagenic potency of cigarette smoke in germ cells is not clear. One of the few tests that can be conducted in humans is the sperm morphology assay, which at least provides an indication that the parent compound or its metabolites reach the testes (Wyrobek et al., 19821. The value of this assay is limited because induction of abnormal sperm mor- phology may not be related to genetic damage in the exposed male. Biochemical assays for human gene mutation (Mohrenweiser and Neel, 1982) show promise as mutational screening techniques, but they also require precise assessment of the background germinal mutation rate. Although these biochemical assays are of little use in measuring mutational events in germ cells of exposed individuals, they do provide information

|56 DRINKING WATER AND H"LTH on mutation rates in exposed populations. There is no effective assay for measuring chemically induced mutations or even chemically induced DNA damage in germ cells of human females. STRUCTU RE-ACTIVITY RELATIONSH I PS An important consideration in the evaluation of a chemical for potential carcinogenicity is that of structure-activity relationships with other chem- icals in the same class. Carcinogenic potency and target organ specificity can sometimes be predicted from such relationships. A number of methods have been developed to relate structure with activity; one of the best known is the Hansch-Taft relationship. Quanti- tative Hansch-Taft relationships have been established for several classes of chemicals, based on water-hexane partition coefficients, electronic fac- tors (Taft axe values), which are believed to provide an indication of the electron-donating or -withdrawing properties of substituents present on a methylene group bound to a reactive center (Gould, 1959; Hansch, 1969), and dose-response data for chemicals within the class for which such data exist. These relationships can be used to predict the carcinogenicity of other chemicals within the class. For example, a Hansch-Taft relationship has been established for the carcinogenicity of nitrosamine (Wishnok et al., 19781. This was based on data for 60 nitrosamines that had been tested in rats (Druckrey et al., 1967~. The relationship for nitrosamines was extended to include target organ specificity, revealing that specificity is directly associated with molecular structure and a complex interplay between the parent molecule and its metabolites with the exposed organism (Edelman et al., 19801. Structure-activity relationships have also been established for groups of promoters such as anthralin derivatives (Van Duuren et al., 1978) and phenolic compounds (Boutwell and Bosch, 19591. RISK ASSESSMENT Risk assessments for carcinogenesis reflect assumptions about the un- derlying mechanisms of the disease. Considerable data from studies in animals and humans support the multistage concept of chemical carci- nogenesis. Classic noncancer toxicology assumes that there is a threshold dose for toxic effects, and that below the threshold dose no ill effects will or can occur. Sound biological arguments support this assumption. How- ever, cancer is a disease originating at the molecular level, involving unrepaired or ill-repaired damage to cellular genetic material (DNA) (Hat- tis and Ashford, 19824. Consequently, the insult by an initiating xenobiotic compound is believed to add to an already on-going natural process, so

Mechanisms of Carcinogenesis 157 that a threshold for cancer induction by a xenobiotic compound does not exist. Most laboratory evidence indicates that promoters need to be admin- istered continuously, or in large doses, implying that promoters should have discoverable thresholds (Hoer et al. 19831. A difficulty arises when the same chemical appears to behave both as an initiator with respect to one tissue and a promoter with respect to another tissue, raising questions of the importance not of the chemical but of the chemical-tissue interaction. Examples of this chemical-tissue interaction are asbestos, which is prob- ably a promoter with respect to lung cancer and an initiator with respect to mesothelioma in humans (NRC, 1984), and 2-AAF, which has pro- moting activity in both bladder and liver carcinogenesis in laboratory animals (Hughes et al., 1983~. In assessing cancer risk, projections from high animal doses to low- dose exposures of humans are driven more by the choice of the extrap- olation model than by the available data (Whittemore, 19801. Thus, further research, development, and validation of more sophisticated models are required. For example, models are needed to distinguish between re- sponses to lifetime exposures at relatively uniform increments of dose and those resulting from more erratic accumulation of doses that range from high to no dose for varying periods of time at irregular intervals. Studies of cohorts exposed to radiation or asbestos indicate that dose rate needs to be included in future dose-response models. When assessing the risk of exposure to carcinogens in drinking water, one should consider information on metabolism and pharmacokinetics. This information may be used to ascertain the validity of a linear dose- response assumption, which can misstate the risk because of inadequate information on metabolism. Determination of the biologically effective dose delivered to the target molecule (such as DNA) will allow a more realistic evaluation of risk. Because of the multistage nature of carcinogenesis, the concept of threshold dose, or no-observed-effect level, is not applicable as it is for other toxic effects. Instead, the multistage, no-threshold, low-dose ex- trapolation model described in Chapter 8 can be used to predict a finite excess risk of cancer at any low carcinogen dose. However, no model can unambiguously predict low-dose effects from data obtained at high doses. A major reason for this ambiguity is that at high doses, the behavior of an enzyme system responsible for the activation or detoxication of a carcinogen may often be governed by a set of parameters different from those that apply at low doses. Chapter ~ contains a more detailed discussion of risk assessment for carcinogens.

|58 DRINKING WATER AND HEALTH REFERENCES Abernethy, D. J., J. C. Huband, W. F. Greenlee, and C. J. Boreiko. 1984. The effect of TCDD upon the transformation, initiation, and promotion of C3H/10T~/: cells. (Abstract Gb-37.) Environ. Mutagen. 6:461. Adams, D. O., and R. Snyderman. 1979. Do macrophages destroy nascent tumors? J. Natl. Cancer Inst. 62: 1341-1345. Arffmann, E., and J. Glavind. 1971. Tumour-promoting activity of fatty acid methyl esters in mice. Experientia 27:1465-1466. Armitage, P., and R. Doll. 1954. The age distribution of cancer and a multistage theory of carcinogenesis. Br. J. Cancer 8:1-12. Ashby, J. 1983. The unique role of rodents in the detection of possible human carcinogens and mutagens. International Commission for Protection Against Environmental Mutagens and Carcinogens (ICPEMC Working Paper 1/1). Mutat. Res. 115:177-213. Baird, W. M., J. A. Sedgwick, and R. K. Boutwell. 1971. Effects of phorbol and four diesters of phorbol on the incorporation of tritiated precursors into DNA, RNA, and protein in mouse epidermis. Cancer Res. 31:1434-1439. Barrett, J. C., D. G. Thomassen, and T. W. Hesterberg. 1983. Role of gene and chro- mosomal mutations in cell transformation. Ann. N.Y. Acad. Sci. 407:291-300. Barrett, J. C., T. W. Hesterberg, and D. G. Thomassen. 1984. Use of cell transformation systems for carcinogenicity testing and mechanistic studies of carcinogenesis. Pharmacol. Rev. 36(Suppl.):53S-70S. Baxter, J. D., and J. W. Funder. 1979. Hormone receptors. N. Engl. J. Med. 301:1149- 1161. Becker, F. F. 1979. Lack of proportionality between rate of cell division and induction of tumors in carcinogen-exposed regenerating livers. Cancer Res. 39:5177-5178. Becker, F. F. 1981. Recent concepts of initiation and promotion in carcinogenesis. Am. J. Pathol. 105:3-9. Bekesi, J. G., H. A. Anderson, J. P. Roboz, J. Roboz, A. Fischbein, I. J. Selikoff, and J. F. Holland. 1979. Immunologic dysfunction among PBB-exposed Michigan dairy farmers. Ann. N.Y. Acad. Sci. 320:717-728. Belman, S., and W. Troll. 1972. The inhibition of croton oil-promoted mouse skin tu- morigenesis by steroid hormones. Cancer Res. 32:450-454. Berenblum, I. 1941. The cocarcinogenic action of croton resin. Cancer Res. 1:44-48. Bermudez, E., D. Tillery, and B. E. BuKerworth. 1979. The effect of 2,4-diaminotoluene and isomers of dinitrotoluene on unscheduled DNA synthesis in primary rat hepatocytes. Environ. Mutagen. 1:391-398. Bernstein, S. C., and R. A. Weinberg. 1985. Expression of the metastatic phenotype in cells transfected with human metastatic tumor DNA. Proc. Natl. Acad. Sci. USA 82:1726- 1730. Biggar, R. J., J. Horm, J. H. Lubin, J. J. Goedert, M. H. Greene, and J. F. Fraumeni, Jr. 1985. Cancer trends in a population at risk of acquired immunodef~ciency syndrome. J. Natl. Cancer Inst. 74:793-797. Bishop, J. M. 1982. Oncogenes. Sci. Am. 246(3):80-92. Bock, F. G., A. Fjelde, H. W. Fox, and E. Klein. 1969. Tumor promotion by 1-fluoro- 2,4-dinitrobenzene, a potent skin sensitizer. Cancer Res. 29:179-182: Boutwell, R. 1964. Some biological aspects of skin carcinogenesis. Prog. Exp. Tumor Res. 4:207-250

Mechanisms of Carcinogenesis ~ 59 Boutwell, R. K. 1974. The function and mechanism of promoters of carcinogenesis. CRC Crit. Rev. Toxicol. 2:419-443. Boutwell, R. K., and D. K. Bosch. 1959. The tumor-promoting action of phenol and related compounds for mouse skin. Cancer Res. 19:413-424. Bridges, B. A., J. Clemmesen, and T. Sugimura. 1979. Cigarette smoking Does it carry a genetic risk? Mutat. Res. 65:71-81. Bridges, B. A., B. E. BuKerworth, and I. B. Weinstein, eds. 1982. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 580 pp. Burnet, M. 1970. Immunological Surveillance. Pergamon, New York. 280 pp. Burns, F. J., M. Vanderlaan, A. Sivak, and R. E. Albert. 1976. Regression kinetics of mouse skin papillomas. Cancer Res. 36:1422-1427. Butterworth, B. E. 1979. Strategies for Short-Term Testing for Mutagens/Carcinogens. CRC Press, West Palm Beach, Fla. 134 pp. Butterworth, B. E. 1981. Predictive assays for mutagens and carcinogens. Pp. 157-177 in E. Gralla, ed. Scientific Considerations in Monitoring and Evaluating Toxicological Research. A CIIT Series. Hemisphere Publishing, New York. Cairns, J. 1975. The cancer problem. Sci. Am. 233(5):64-78. Cairns, J. 1981. The origin of human cancers. Nature 289:353-357. Cairns, J. 1985. The treatment of diseases and the war against cancer. Sci. Am. 253(5):51- 59. Chamberlain, M., and E. M. Tarmy. 1977. Asbestos and glass fibres in bacterial mutation tests. Mutat. Res. 43:159-164. Chang, K., F. Lu, T. Tung, and T. Lee. 1980. Studies on patients with polychlorinated biphenyl poisoning. Res. Commun. Chem. Pathol. Pharmacol. 30:547-554. Couch, D. B., P. F. Allen, and D. J. Abernethy. 1981. The mutagenicity of dinitrotoluenes in Salmonella typhimurium. Mutat. Res. 90:373-383. Craddock, V. M. 1971. Liver carcinoma induced in rats by single administration of di- methylnitrosamine after partial hepatectomy. J. Natl. Cancer Inst. 47:899-905. Day, N. E., and C. C. Brown. 1980. Multistage models and primary prevention of cancer. J. Natl. Cancer Inst. 64:977-989. Dean, J. H., M. I. Luster, G. A. Boorman, L. D. Lauer, and E. C. Ward. 1983a. Immunotoxicity of tumor promoting environmental chemicals and phorbol diesters. Adv. Immunopharmacol. 2:23-31. Dean, J. H., M. I. Luster, G. A. Boorman, L. D. Lauer, R. W. Leubke, and L. Lawson. 1983b. Selective immunosuppression resulting from exposure to the carcinogenic con- gener of benzopyrene in B6C3F1 mice. Clin. Exp. Immunol. 52:199-206. Dean, J. H., L. D. Lauer, R. V. House, M. J. Murray, W. S. Stillman, R. D. Irons, W. H. Steinhagen, M. C. Phelps, and D. O. Adams. 1984. Studies of immune function and host resistance in B6C3F1 mice exposed to formaldehyde. Toxicol. Appl. Pharmacol. 72:519-529. DeMarini, D.M. 1983. Genotoxicity of tobacco smoke and tobacco smoke condensate. Mutat. Res. 114:59-89. de Serres, F. J., and J. Ashby, eds. 1981. Evaluation of Short-Term Tests for Carcino- genesis: Report of the International Collaborative Program. Progress in Mutation Re- search. Vol. 1. Elsevier/North-Holland, New York. 827 pp. DiGiovanni, J., A. Viaje, D. L. Berry, T. J. Slaga, and M. R. Juchau. 1977. Tumor- initiating ability of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and Arochlor 1254 in

|60 DRINKING WATER AND HEALTH the two-stage system of mouse skin carcinogenesis. Bull. Environ. Contam. Toxicol. 18:552-557. DiGiovanni, J., P. C. Decina, and L. Diamond. 1983. Tumor initiating activity of 9- and 10-fluoro-7,12-dimethylbenz[a]anthracene (DMBA) and the effect of 2,3,7,8-tetrachlo- rodibenzo-p-dioxin on tumor initiation by monofluoro derivatives of DMBA in SENCAR mice. Carcinogenesis 4: 1045-1049. Doll, R. 1978. An epidemiological perspective of the biology of cancer. Cancer Res. 38:3573-3583. Doolittle, D. J., J. M. Sherrill, and B. E. Butterworth. 1983. The influence of intestinal bacteria, sex of the animal, and position of the nitro group on the hepatic genotoxicity of nitrotoluene isomers in viva. Cancer Res. 43:2836-2842. Druckrey, H., R. Preussmann, S. Ivankovic, and D. Schmahl. 1967. Organotropic car- cinogenic effects of 65 different N-nitroso-compounds on BD-rats. Z. Krebsforsch. 69: 103- 201. (In German; English summary) Eastman, A., B. T. Mossman, and E. Bresnick. 1983. Influence of asbestos on the uptake of benzo(a)pyrene and DNA alkylation in hamster tracheal epithelial cells. Cancer Res. 43:1251-1255. Edelman, A. S., P. L. Kraft, W. M. Rand, and J. S. Wishnok. 1980. Nitrosamine car- cinogenicity: A quantitative relationship between molecular structure and organ selectivity for a series of acyclic N-nitroso compounds. Chem.-Biol. Interact. 31:81-92. EEC (European Economic Community). 1979. Council Directive of 18 September 1979 amending for the sixth time Directive 67/548/EEC on the approximation of the laws, regulations and administrative provisions relating to the classification, packaging and labelling of dangerous substances (79/831/EEC). Off. J. Eur. Comm. 22(L 259):10-28. EMIC (Environmental Mutagen Information Center). 1986. Automated data base file from Oak Ridge National Laboratory library systems. TOXLINE information system, National Library of Medicine, Bethesda, Md. EPA (Environmental Protection Agency). 1984. Proposed guidelines for carcinogen risk assessment. Fed. Regist. 49:46294-46301. Evans, H. J. 1982. Sperm morphology in cigarette smokers. Pp. 543-554 in B. A. Bridges, B. E. Butterworth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Farber, E. 1982. Sequential events in chemical carcinogenesis. Pp. 485-506 in F. F. Becker, ed. Cancer: A Comprehensive Treatise, 2nd ed. Vol. 1. Etiology: Chemical and Physical Carcinogenesis. Plenum, New York. Farber, E. 1984a. Cellular biochemistry of the stepwise development of cancer with chem- icals: G. H. A. Clowes Memorial Lecture. Cancer Res. 44:5463-5474. Farber, E. 1984b. The multistep nature of cancer development. Cancer Res. 44:4217-4223. Farber, E., and R. Cameron. 1980. The sequential analysis of cancer development. Adv. Cancer Res. 31: 125-226. Fidler, I. J. 1985. Macrophages and metastasis A biological approach to cancer therapy: Presidential address. Cancer Res. 45:4714-4726. Fidler, I. J., and E. S. Kleinerman. 1984. Lymphokine-activated human blood monocytes destroy tumor cells but not normal cells under cocultivation conditions. J. Clin. Oncol. 2:937-943. Food Safety Council. 1980. Proposed System for Food Safety Assessment. Final Report of the Scientific Committee of the Food Safety Council, June 1980. Food Safety Council, Washington, D.C. 160 pp. Fornace, A. J., Jr. 1982. Detection of DNA single-strand breaks produced during the repair of damage of DNA by DNA-protein cross-linking agents. Cancer Res. 42:145-149.

Mechanisms of Carcinogenesis ~ 6 ~ Frazelle, J. H., D. J. Abernethy, and C. J. Boreiko. 1983. Factors influencing the promotion of transformation in chemically-initiated C3H/lOT1/2 C1 8 mouse embryo fibroblasts. Carcinogenesis 4:709-715. Fujiki, H., M. Mori, M. Nakayasu, M. Terada, T. Sugimura, and R. E. Moore. 1981. Indole alkaloids: Dihydroteleocidin B. teleocidin, and lyngbyatoxin A as members of a new class of tumor promoter. Proc. Natl. Acad. Sci. USA 78:3872-3876. Furstenberger, G., B. Sorg, and F. Marks. 1983. Tumor promotion by phorbol esters in skin: Evidence for a memory effect. Science 220:89-91. Geiger, L. E., and R. A. Neal. 1981. Mutagenicity testing of 2,3,7,8-tetrachlorodibenzo- p-dioxin in histidine auxotrophs of Salmonella typhimurium. Toxicol. Appl. Pharmacol. 59:125-129. Goldstein, B. D., C. A. Snyder, S. Laskin, I. Bromberg, R. E. Albert, and N. Nelson. 1982. Myelogenous leukemia in rodents inhaling benzene. Toxicol. Lett. 13:169-173. Gould, E. S. 1959. Polar effects in aliphatic compounds: The Taft treatment. Pp. 227-230 in Mechanism and Structure in Organic Chemistry. Holt, Rinehart & Winston, New York. Gwynn, R. H., and N. H. Salamon. 1953. Studies on co-carcinogenesis. SH-reactors and other substances tested for co-carcinogenic action in mouse skin. Br. J. Cancer 7:482- 489. Hansch, C. 1969. A quantitative approach to biochemical structure-activity relationships. Acc. Chem. Res. 2:232-239. Hattis, D. B., and N. A. Ashford. 1982. Defining regulatory alternatives and baselines for comparison. Pp. 2-1 through 2-16 in Analyzing the Benefits of Health, Safety, and Environmental Regulations. Report CPA-82-16. Center for Policy Alternatives, Mas- sachusetts Institute of Technology, Cambridge, Mass. Hecht, S. S., S. Carmella, H. Mori, and D. Hoffmann. 1981. A study of tobacco carci- nogenesis. XX. Role of catechol as a major cocarcinogen in the weakly acidic fraction of smoke condensate. J. Natl. Cancer Inst. 66:163-169. Hecker, E. 1971. Isolation and characterization of the cocarcinogenic principles from croton oil. Methods Cancer Res. 6:439-484. Heidelberger, C., A. E. Freeman, R. J. Pienta, A. Sivak, J. S. Bertram, B. C. Casto, V. C. Dunkel, M. W. Francis, T. Kakunaga, J. B. Little, and L. M. Schechtman. lg83. Cell transformation by chemical agents A review and analysis of the literature. A report of the U.S. Environmental Protection Agency Gene-Tox Program. Mutat. Res. 114:283- 385. Herbst, A. L., R. E. Scully, S. J. Robboy, W. R. Welch, and P. Cole. 1977. Abnormal development of the human genital tract following prenatal exposure to diethylstilbestrol. Pp. 399-412 in H. H. Hiatt, J. D. Watson, and J. A. Winsten, eds. Origins of Human Cancer. Book A. Incidence of Cancer in Humans. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Hicks, R. M. 1983. Effect of promoters on incidence of bladder cancer in experimental animal models. Environ. Health Perspect. 50:37-49. Hoel, D. G., N. L. Kaplan, and M. W. Anderson. 1983. Implications of nonlinear kinetics on risk estimation in carcinogenesis. Science 219:1032-1037. Hollstein, M., J. McCann, F. A. Angelosanto, and W. W. Nichols. 1979. Short-term tests for carcinogens and mutagens. Mutat. Res. 65:133-226. Hughes, D. H., R. D. Bruce, R. W. Hart, L. Fishbein, D. W. Gaylor, J. M. Smith, and W. W. Carlton. 1983. A Report on the Workshop on Biological and Statistical Impli- cations of the EDo~ Study and Related Data Bases. Fund. Appl. Toxicol. 3:129-136.

|62 DRINKING WATER AND H"LTH IARC (International Agency for Research on Cancer). 1977. General principles for the evaluation. Pp. 11-14 in IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Man. Vol. 13. Some Miscellaneous Pharmaceutical Substances. Interna- tional Agency for Research on Cancer, Lyon, France. IARC (International Agency for Research on Cancer). 1980. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. Supplement 2. Long- Term and Short-Term Screening Assays for Carcinogens: A Critical Appraisal. Inter- national Agency for Research on Cancer, Lyon, France. 426 pp. IARC (International Agency for Research on Cancer). 1982. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. Supplement 4. Chemicals, Industrial Processes and Industries Associated with Cancer in Humans. International Agency for Research on Cancer, Lyon, France. 292 pp. IARC (International Agency for Research on Cancer). 1983. Approaches to Classifying Chemical Carcinogens According to Mechanism of Action. Report of a meeting held at the International Agency for Research on Cancer, Lyon, April 11-15, 1983. IARC Internal Technical Report No. 83/001. International Agency for Research on Cancer, Lyon, France. 69 pp. IRLG (Interagency Regulatory Liaison Group). 1979. Scientific bases for identification of potential carcinogens and estimation of risks. Report of the Interagency Regulatory Liaison Group, Work Group on Risk Assessment. J. Natl. Cancer Inst. 63:242-268. Kensler, T. W., and M. A. Trush. 1984. Role of oxygen radicals in tumor promotion. Environ. Mutagen. 6:593-616. Klein-Szanto, A. J. P., K. G. Nelson, Y. Shah, and T. J. Slaga. 1983. Simultaneous appearance of keratin modifications and y-glutamyltransferase activity as indicators of tumor progression in mouse skin papillomas. J. Natl. Cancer Inst. 70:161-168. Kline, J., Z. A. Stein, M. Susser, and D. Warburton. 1977. Smoking: A risk factor for spontaneous abortion. N. Engl. J. Med. 297:793-796. Kociba, R. J., D. G. Keyes, J. E. Beyer, R. M. Carreon, C. E. Wade, D. A. Dittenber, R. P. Kalnins, L. E. Frauson, C. N. Park, S. D. Barnard, R. A. Hummel, and C. G. Humiston. 1978. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8- tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl. Pharmacol. 46:279-303. Kripke, M. L., and W. L. Morison. 1985. Modulation of immune function by W radiation. J. Invest. Dermatol. 85(1 Suppl.):62s-66s. Land, H., L. F. Parada, and R. A. Weinberg. 1983. Cellular oncogenes and multistep carcinogenesis. Science 222:771-778. Laskin, D. L., and A.M. Pilaro. 1985. Activation of rat liver macrophages (Kupffer cells) by the tumor promoters phenobarbital (PB) and 12-O-tetradecanoylphorbol-13-acetate (TPA). (Abstract 572.) Proc. Am. Assoc. Cancer Res. 26:144. Laskin, D. L., J. D. Laskin, F. K. Kessler, I. B. Weinstein, and R. A. Carchman. 1981. Enhancement of macrophage-induced cytotoxicity by phorbol ester tumor promoters. Cancer Res. 41 :4523-4528. Leder, P., J. Bauey, G. Lenoir, C. Moulding, W. Murphy, H. Potter, T. Stewart, and R. Taub. 1983. Translocations among antibody genes in human cancer. Science 222:765- 771. Linsell, C. A., and F. G. Peers. 1977. Field studies on liver cell cancer. Pp. 549-556 in H. H. Hiatt, J. D. Watson, and J. A. Winsten, eds. Origins of Human Cancer. Book A. Incidence of Cancer in Humans. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Lippman, M. E., and J. C. Allegra. 1978. Receptors in breast cancer. N. Engl. J. Med. 299:930-933.

Mechanisms of Carcinogenesis 163 Loewengart, G., and B. L. Van Duuren. 1977. Evaluation of chemical flame retardants for carcinogenic potential. J. Toxicol. Environ. Health 2:539-546. Luster, M. I., G. A. Boorman, J. H. Dean, R. W. Luebke, and L. D. Lawson. 1980a. The effect of adult exposure to diethylstilbestrol in the mouse: Alterations in immuno- logical functions. J. Reticuloendothel. Soc. 28:561-569. Luster, M. I., G. A. Boorman, J. H. Dean, M. W. Harris, R. W. Luebke, M. L. Padarathsingh, and J. A. Moore. 1980b. Examination of bone marrow, immunologic parameters and host susceptibility following pre- and postnatal exposure to 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD). Int. J. Immunopharmacol. 2:301-310. Luster, M. I., J. H. Dean, G. A. Boorman, M. P. Dieter, and H. T. Hayes. 1982. Immune functions in methyl and ethyl carbamate treated mice. Clin. Exp. Immunol. 50:223-230. Malkinson, A. M., and D. S. Beer. 1984. Pharmacologic and' genetic studies on the modulatory effects of butylated hydroxytoluene on mouse lung adenoma formation. J. Natl. Cancer Inst. 73:925-933. Maltoni, C., and C. Scarnato. 1979. First experimental demonstration of the carcinogenic effects of benzene. Long-term bioassays on Sprague-Dawley rats by oral administration. Med. Lav. 70:352-357. McCann, J., E. Choi, E. Yamasaki, and B. N. Ames. 1975. Detection of carcinogens as mutagens in the Salmonella/microsome test: Assay of 300 chemicals. Proc. Natl. Acad. Sci. USA 72:5135-5139. McGuire, W. L., K. B. Horwitz, D. T. Zava, R. E. Garola, and G. C. Chamness. 1978. Hormones in breast cancer: Update 1978. Metabolism 27:487-501. McKhann, C. F. 1969. Primary malignancy in patients undergoing immunosuppression for renal transplantation. Transplantation 8:209-212. Mirsalis, J. C., T. E. Hamm, Jr., J. M. Sherrill, and B. E. BuKerworth. 1982. The role of gut flora in the genotoxicity of dinitrotoluene. Nature 295:322-323. Mohrenweiser, H. W., and J. V. Neel. 1982. Models to man: Establishment of reference points for estimating genetic risk in man. Pp. 471-486 in B. A. Bridges, B. E. Butter- worth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Mommsen, S., and J. Aagaard. 1983. Tobacco as a risk factor in bladder cancer. Carci- nogenesis 4:335-338. Mossman, B. T., G. S. Cameron, and L. P. Yotti. 1985. Cocarcinogenic and tumor promoting properties of asbestos and other minerals in tracheobronchial epithelium. Carcinog. Compr. Surv. 8:217-238. Mottram, J. C. 1944. A developing factor in experimental blastogenesis. J. Pathol. Bac- teriol. 56:181-187. Mufson, R. A., S. M. Fischer, A. K. Verma, G. L. Gleason, T. J. Slaga, and R. K. Boutwell. 1979. Effects of 12-O-tetradecanoylphorbol-13-acetate and mezerein on epi- dermal ornithine decarboxylase activity, isoproterenol-stimulated levels of cyclic aden- osine 3':5'-monophosphate, and induction of mouse skin tumors in vivo. Cancer Res. 39:4791-4795. Mylecraine, L. 1984. Effect of Pretreatment with Modifiers of Mixed Function Oxidases on the Development of Hepatic Foci of Cellular Alteration in Rats After a Single Dose of a Genotoxic Carcinogen (Liver, Diethylnitrosamine, Pre-neoplasia). Ph.D. disserta- tion. Rutgers, The State University of New Jersey, Newark. 185 pp. [Abstract in Diss. Abstr. Int. B 45(06):1635.] Newberne, P. M., J. L. V. de Camargo, and A. J. Clark. 1982. Choline deficiency, partial hepatectomy, and liver tumors in rats and mice. Toxicol. Pathol. 10:95-109.

|64 DRINKING WATER AND H"LTH Norkus, E. P., W. Kuenzis, and A. H. Conney. 1983. Studies on the mutagenic activity of ascorbic acid in vitro and in viva. Mutat. Res. 117:183-191. NRC (National Research Council). 1977. Drinking Water and Health. National Academy of Sciences, Washington, D.C. 939 pp. NRC (National Research Council). 1983. Risk Assessment in the Federal Government: Managing the Process. National Academy Press, Washington, D.C. 50 pp. NRC (National Research Council). 1984. Asbestiform Fibers: Nonoccupational Health Risks. National Academy Press, Washington, D.C. 334 pp. NTP (National Toxicology Program). 1984. Report of the NTP Ad Hoc Panel on Chemical Carcinogenesis Testing and Evaluation. U. S. Department of Health and Human Services, Washington, D.C. 280 pp. O'Brien, T. J., R. C. Simsiman, and R. K. Boutwell. 1975. Induction of the polyamine- biosynthetic enzymes in mouse epidermis and their specificity for tumor promotion. Cancer Res. 35:2426-2433. OECD (Organisation for Economic Co-operation and Development). 1981. OECD Guide- lines for Testing of Chemicals. Organisation for Economic Co-operation and Develop- ment, Paris, France. (400+ pages, unnumbered) Ohyama, K. 1985. Effects of anti-tumor drugs and gastrointestinal hormones on the growth of pancreatic duct cell adenocarcinoma in the homologous transplanted animals models. Hokkaido Igaku Zasshi 60:206-216. (In Japanese; English summary) OSTP (Office of Science and Technology Policy). 1985. Chemical Carcinogens; A Review of the Science and Its Associated Principles, February 1985. Fed. Regist. 50:10371- 10442. OTA (Office of Technology Assessment). 1981. Assessment of Technologies for Deter- mining Cancer Risks from the Environment. Congress of the United States, Office of Technology Assessment, Washington, D.C. 240 pp. Penn, I. 1978. Malignancies associated with immunosuppressive or cytotoxic therapy. Surgery 83:492-502. Peraino, C., R. J. M. Fry, and E. Staffeldt. 1971. Reduction and enhancement by pheno- barbital of hepatocarcinogenesis induced in the rat by 2-acetylaminofluorene. Cancer Res. 31:1506-1512. Peraino, C., E. F. Staffeldt, D. A. Haugen, L. S. Lombard, F. J. Stevens, and R. J. M. Fry. 1980. Effects of varying the dietary concentration of phenobarbital on its enhance- ment of 2-acetylaminofluorene-induced hepatic tumorigenesis. Cancer Res. 40:3268- 3273. Pershagen, G., G. Nordberg, and N. Bjorklund.1984. Carcinomas of the respiratory tract in hamsters given arsenic trioxide and/or benzo[a]pyrene by the pulmonary route. En- viron. Res. 34:227-241. Peto, R. 1977. Epidemiology, multistage models, and short-term mutagenicity tests. Pp. 1403-1428 in H. H. Hiatt, J. D. Watson, and J. A. Winsten, eds. Origins of Human Cancer. Book C. Human Risk Assessment. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Pilaro, A. M., a: ~ D. L. Laskin. 1984. Characterization of resident and inflammatory Kupffer cells; activation by the tumor promoter, 12-O-tetradecanoylphorbol-13-acetate (TPA). (Abstract 1056.) Proc. Am. Assoc. Cancer Res. 25:266. Pitot, H. C., T. Goldsworthy, H. A. Campbell, and A. Poland. 1980. Quantitative eval- uation of the promotion by 2,3,7,8-tetrachlorodibenzo-p-dioxin of hepatocarcinogenesis from diethylnitrosamine . Cancer Res . 40:3616-3620. Rall, D. P. 1979. Validity of extrapolation of results of animal studies to man. Ann. N.Y. Acad. Sci. 329:85-91.

Mechanisms of Carcinogenesis 165 Reddy, B. S., and Y. Maeura. 1984. Tumor promotion by dietary fat in azoxymethane- induced colon Carcinogenesis in female F344 rats: Influence of amount and source of dietary fat. J. Natl. Cancer Inst. 72:745-750. Rickert, D. E., B. E. Butterworth, and J. A. Popp. 1984. Dinitrotoluene: Acute toxicity, oncogenicity, genotoxicity, and metabolism. CRC Crit. Rev. Toxicol. 13:217-234. Rogers, A. E. 1983. Influence of dietary content of lipids and lipotropic nutrients on chemical Carcinogenesis in rats. Cancer Res. 43(Suppl.):2477s-2484s. Rohrschneider, L. R., and R. K. Boutwell. 1973. The early stimulation of phospholipid metabolism by 12-O-tetradecanoyl-phorbol-13-acetate and its specificity for tumor pro- motion. Cancer Res. 33: 1945-1952. Rous, P., and J. G. Kidd. 1941. Conditional neoplasms and subthreshold neoplastic states: A study of the tar tumors of rabbits. J. Exp. Med. 73:365-390. Schatten, S., R. D. Granstein, J. A. Drebin, and M. I. Greene. 1984. Suppressor T cells and the immune response to tumors. CRC Crit. Rev. Immunol. 4:335-379. Schwartz, J. A., A. Viaje, T. J. Slaga, S. H. Yuspa, H. Hennings, and U. Lichti. 1977. Fluocinolone acetonide: A potent inhibitor of mouse skin tumor promotion and epidermal DNA synthesis. Chem.-Biol. Interact. 17:331-347. Scribner, N. K., and J. D. Scribner. 1980. Separation of initiating and promoting effects of the skin carcinogen 7-bromomethylbenz(a)anthracene. Carcinogenesis 1:97-100. Segal, A., C. Katz, and B. L. Van Duuren. 1971. Structure and tumor-promoting activity of anthralin (1,8-dihydroxy-9-anthrone) and related compounds. J. Med. Chem. 14: 1152-1154. Setala, K. 1960. Progress in carcinogenesis, tumor-enhancing factors. A big-assay of skin tumor formation. Prog. Exp. Tumor Res. 1:225-278. Shigematsu, N., S. Ishimaru, R. Saito, T. Ikeda, K. Matsuba, K. Sugiyama, and Y. Masuda. 1978. Respiratory involvement in polychlorinated biphenyls poisoning. Environ. Res. 16:92-100. Silverberg, E. 1984. Cancer statistics, 1984. Ca Cancer J. Clin. 34:7-23. Sims, P. 1976. The metabolism of polycyclic hydrocarbons to dihydrodiols and diol- epoxides by human and animal tissues. Pp. 211-224 in R. Montesano, H. Bartsch, and L. Tomatis, eds. Screening.Tests in Chemical Carcinogenesis. IARC Scientific Publi- cations No. 12. International Agency for Research on Cancer, Lyon, France. Sivak, A. 1979. Cocarcinogenesis. Biochim. Biophys. Acta 560:67-89. Slaga, T. J. 1983. Mechanisms involved in multistage chemical carcinogenesis in mouse skin. Pp. 577-585 in J. Rydstrom, J. Montelius, and M. Bengtsson, eds. Extrahepatic Drug Metabolism and Chemical Carcinogenesis. Elsevier, New York. Slaga, T. J., and S. M. Fischer. 1983. Strain differences and solvent effects in mouse skin carcinogenesis experiments using carcinogens, tumor initiators and promoters. Prog. Exp. Tumor Res. 26:85-109. Slaga, T. J., L. Jecker, W. M. Bracken, and C. E. Weeks. 1979. The effects of weak or non-carcinogenic polycyclic hydrocarbons on 7,12-dimethylbenz[a]anthracene and benzo[a]pyrene skin tumor-initiation. Cancer Lett. 7:51-59. Slaga, T. J., A. J. P. Klein-Szanto, S. M. Fischer, C. E. Weeks, K. Nelson, and S. Major. 1980a. Studies on mechanism of action of anti-tumor-promoting agents: Their specificity in two-stage promotion. Proc. Natl. Acad. Sci. USA 77:2251-2254. Slaga, T. J., S. M. Fischer, K. Nelson, and G. L. Gleason. 1980b. Studies on the mechanism of skin tumor promotion: Evidence for several stages in promotion. Proc. Natl. Acad. Sci. USA 77:3659-3663. Slaga, T. J., A. J. P. Klein-Szanto, L. L. Triplett, and L. P. Yotti. 1981. Skin tumor promoting activity of benzoyl peroxide, a widely used free radical-generating compound. Science 213:1023-1025.

|66 DRINKING WATER AND H"LTH Slamon, D. J., J. B. deKernion, I. M. Verma, and M. J. Cline. 1984. Expression of cellular oncogenes in human malignancies. Science 224:256-262. Solanki, V., R. S. Rana, and T. J. Slaga. 1981. Diminution of mouse epidermal superoxide dismutase and catalase activities by tumor promoters. Carcinogenesis 2:1141-1146. Strobino, B. R., J. Kline, and Z. Stein. 1978. Chemical and physical exposures of parents: Effects on human reproduction and offspring. Early Hum. Dev. 1:371-399. Suganuma, M., H. Fujiki, T. Tahira, C. Cheuk, R. E. Moore, and T. Sugimura. 1984. Estimation of tumor promoting activity and structure-function relationships of aplysia- toxins. Carcinogenesis 5:315-318. Sukumar, S., V. Notario, D. Martin-Zanca, and M. Barbacid. 1983. Induction of mammary carcinomas in rats by nitroso-methylurea involves malignant activation of H-ras-1 locus by single point mutations. Nature 306:658-661. Thomas, L. 1959. Discussion. Pp. 529-532 in H. S. Lawrence, ed. Cellular and Humoral Aspects of the Hypersensitive States. Hoeber-Harper, New York. Tong, C., M. Fazio, and G. M. Williams. 1980. Cell cycle-specific mutagenesis at the hypoxanthine phosphoribosyltransferase locus in adult rat liver epithelial cells. Proc. Natl. Acad. Sci. USA 77:7377-7379. Troll, W., M. S. Meyn, and T. G. Rossman. 1978. Mechanisms of protease action in carcinogenesis. Carcinog. Compr. Surv. 2:301-312. Tsutsui, T., H. Maizumi, J. A. McLachlan, and J. C. Barrett. 1983. Aneuploidy induction and cell transformation by diethylstilbestrol: A possible chromosomal mechanism in carcinogenesis. Cancer Res. 43:3814-3821. Van Duuren, B. L. 1969. Tumor-promoting agents in two-stage carcinogenesis. Prog. Exp. Tumor Res. 11:31-68. Van Duuren, B. L., and B. M. Goldschmidt. 1976. Carcinogenic and tumor-promoting agents in tobacco carcinogenesis. J. Natl. Cancer Inst. 56:1237-1242. Van Duuren, B. L., A. Sivak, A. Segal, I. Seidman, and C. Katz. 1973. Dose-response studies with a pure tumor-promoting agent, phorbol myristate acetate. Cancer Res. 33:2166-2172. Van Duuren, B. L., A. Sivak, C. Katz, I. Seidman, and S. Melchionne. 1975. The effect of aging and interval between primary and secondary treatment in two-stage carcino- genesis on mouse skin. Cancer Res. 35:502-505. Van Duuren, B. L., G. Witz, and B. M. Goldschmidt. 1978. Structure-activity relationships of tumor promoters and cocarcinogens and interaction of phorbol myristate acetate and related esters with plasma membranes. Carcinog. Compr. Surv. 2:491-507. Verma, A. K., B. G. Shapas, H. M. Rice, and R. K. Boutwell. 1979. Correlation of the inhibition by retinoids of tumor promoter-induced mouse epidermal ornithine decarbox- ylase activity and of skin tumor promotion. Cancer Res. 39:419-425. Verma, A. K., E. Erturk, and G. T. Bryan. 1983. Specific binding, stimulation of rodent urinary bladder epithelial ornithine decarboxylase, and induction of transitional cell hyperplasia by the skin tumor promoter 12-O-tetradecanoylphorbol-13-acetate. Cancer Res. 43:5964-5971. Ward, E. C., M. J. Murray, L. D. Lauer, R. V. House, R. Irons, and J. H. Dean. 1984. Immunosuppression following 7,12-dimethylbenz[a]anthracene exposure in B6C3F1 mice. I. Effects on humoral immunity and host resistance. Toxicol. Appl. Pharmacol. 75:299- 308. Ward, J. M., M. Ohshima, P. Lynch, and C. Riggs. 1984. Di(2-ethylhexyl) phthalate but not phenobarbital promotes N-nitrosodiethylamine-initiated hepatocellular proliferative lesions after short-term exposure in male B6C3F1 mice. Cancer Lett. 24:49-55.

Mechanisms of Carcinogenesis 167 Wein, A. J., and J. J. Murphy. 1973. Experience in the treatment of prostatic carcinoma with cyppoterone acetate. J. Urol. 109:68-70. Weinberg, R. A. 1985. The action of oncogenes in the cytoplasm and nucleus. Science 230:770-776. Weinstein, I. B., S. Gattoni-Celli, P. Kirschmeier, M. Lambert, W. Hsiao, J. Backer, and A. Jeffrey. 1984. Multistage carcinogenesis involves multiple genes and multiple mech- anisms. Pp. 229-237 in A. J. Levine, G. F. Vande Woude, W. C. Topp, and J. D. Watson, eds. Cancer Cells. 1. The Transformed Phenotype. Cold Spring Harbor Lab- oratory, Cold Spring Harbor, N.Y. Weisburger, J. H., and G. M. Williams. 1982. Chemical Carcinogenesis. Pp. 42-95 in J. F. Holland and E. Fret, m, eds. Cancer Medicine, 2nd ed. Lea and Febiger, Philadelphia. Whittemore, A. S. 1980. Mathematical models of cancer and their use in risk assessment. J. Environ. Pathol. Toxicol. 3:353-362. Williams, G. M., and J. H. Weisburger. 1983. New approaches to carcinogen bioassay. Pp. 200-209 in F. Homburger, ed. Safety Evaluation and Regulation of Chemicals. First International Conference, Boston, Mass., February 24-26, 1982. Karger, New York. Windham, G. C., T. Bjerkedal, and F. Langmark. 1985. A population-based study of cancer incidence in twins and in children with congenital malformations or low birth weight, Norway, 1967-1980. Am. J. Epidemiol. 121:49-56. Wishnok, J. S., M. C. Archer, A. S. Edelman, and W. M. Rand. 1978. Nitrosamine carcinogenicity: A quantitative Hansch-Taft structure-activity relationship. Chem.-Biol. Interact. 20:43-54. Witschi, H., D. Williamson, and S. Lock. 1977. Enhancement of urethan tumorigenesis in mouse lung by butylated hydroxytoluene. J. Natl. Cancer Inst. 58:301-305. Witschi, H. P. 1983. Promotion of lung tumors in mice. Environ. Health Perspect. 50:267- 273. Wotiz, H. H., D. R. Beebe, and E. Muller. 1984. Effect of estrogens on DMBA induced breast tumors. J. Steroid Biochem. 20:1067-1075. Wyrobek, A. J., L. A. Gordon, G. Watchmaker, and D. H. Moore, II. 1982. Human sperm morphology testing: Description of a reliable method and its statistical power. Pp. 527-541 in B. A. Bridges, B. E. BuKerworth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Yamasaki, E., and B. N. Ames. 1977. Concentration of mutagens from urine by adsorption with the nonpolar resin XAD-2: CigareKe smokers have mutagenic urine. Proc. Natl. Acad. Sci. USA 74:3555-3559. Yunis, J. J. 1983. The chromosomal basis of human neoplasia. Science 221:227-236. Zoller, M. 1985. Evaluation of in vivo and in vitro effectivity of immune defense against a spontaneously arising, nonlymphoid rat tumor. II. T cell response after induction of immunogenicity. Cancer Immunol. Immunother. 19:189-197.