Document QgVZXzOj2Kg330QZ24vmqDn3R

V A REVIEW AND EVALUATION OF CARCINOGENICITY STUDIES IN MICE AND RATS AND MUTAGENICITY STUDIES WITH POLYCHLORINATED BIPHENYLS ' George J. Levinskas October 14, 1981 ADM 0059 Lb WATER PCB-00050093 INTRODUCTION* This is a review and evaluation of studies which deal with the potential carcinogenicity and mutagenicity of polychlorinated biphenyls (?C3s). It is subdivided into 4 sections: Chronic Rodent Studies, Metabolism Studies, Co-Carcinogenesis Studies and Mutagenicity Studies. A brief summary of Epidemiology Studies is added to complete coverage of the issue of carcinogenicity. This review does not discuss the effects of impurities or contaminants, particularly polychlorinated dibenzofurans (PCDF), which are reported to be present in some PCB mixtures (Brinkman and deKok, 1980). The presence and amounts of such impurities have not been specified in the materials used in many studies. Thus, attempts to apportion the observed biological effects between impurities and PCBs would only add further conjecture to a subject which currently is rife with speculation. - By contrast, specific chemical and trade names have been used to identify materials that were studied instead of the all encompassing term PCBs. This was deemed necessary because there are too many one-sided generali zations in the literature. Every adverse finding is, by implication at least, extrapolated to the entire class of materials designated as PCBs. Conversely, every report which has failed to find an adverse effect is careful to stipulate that it relates only to the substance studied. Even more difficult to contend with are misleading references to adverse findings which are not substantiated by the actual publications referred to. An example of this occurred in a discussion of in vitro metabolism studies with liver microsomal enzymes which stated that the formation of PCBmacroroolecular adducts had been demonstrated. Among the references ' 1 ADM 005919 WATER PCB-00050094 cited was a paper entitled "The in vitro binding of 2,2',5,5'-tetrachiorobiphenyl metabolites to rat liver microsomal proteins''. The latter paper clearly states "There is no clear evidence in the data obtained from the work reported here to substantiate covalent binding; however, it is also not possible to exclude the nondialyzable radioactivity as being covalently bound". (Hargraves and Allen, 1979). To emphasize the point that dose is important in evaluating safety, test exposure conditions have been described so that the reader can compare them to ambient exposure levels encountered in the literature and else where. Hoopingarner, et al. (1972) in their studies with Aroclor 1254 on cultured human lymphocytes are among the few authors who acknowledged that "The toxic dose of the chemical was several times greater than is usually found biologically". They used concentrations of 100 ppm of Aroclor 1254 in their studies discussed under Mutagenicity. ' Chronic Rodent Studies ' Difficulties in comparing and assessing different studies are compounded not only by problems of histopathologic diagnosis but also by variations in experimental design and animal strain differences. Animal feeding studies cited in the literature as evidence for the carcinogenicity of PCBs and related studies of similar duration have been tabulated for mice in Table 1 r-K ` and for rats in Table 2. Studies have been grouped by the approximate weight percent of chlorine in the materials tested to facilitate comparisons between products from different manufacturers with the same chlorine content. \ ADM 005920 WATER PCB-00050095 Mouse studies m Table 1 were conducted with high dietary levels of FC3s. Kimbrough and Linder (1974) reported a 52% mortality for mice fed 300 ppm of Aroclor 1254 for 6 months and then held an additional 5 months. During that interval, control mice had a 32% mortality. This appears to be quite high for control mice about 1 year of age. In a report discussed under Co-carcinogenesis, Roller (1977) studied mice injected with Aroclor products and Moloney leukemia virus. Some data from his study, although not shown in Table 1, are quite similar to those of Kimbrough and Linder (1974). Roller (1977) fed diets containing Aroclor 1221, 1242, or 1254 to groups of 25 Balb/c male mice for 6 months. After that time, some mice were sacrificed for examination and others were returned to a control diet for another 3 months and then sacrificed. Dietary levels of each Aroclor were 375, 37.5, or 3.75 ppm. The only feeding regimen that was toxic was 375 ppm of Aroclor 1254. Only 8 mice of that group survived for 6 months, giving a mortality of 68%. Other authors cited in Table 1 -(Ito, ' et al., 1973a,b and Nagasaki, et al., 1972, 1974, 1975) did not indicate how many of the mice in their studies died when fed 500 ppm of Kanechlor 500, which has a chlorine content similar to Aroclor 1254. Roller (1977) reported that "PCBs did not produce hepatic neoplasia". There was marked liver injury m mice fed 375 ppm of Aroclor 1254, mild liver injury which persisted through the 3-month recovery period at 37.5 ppm, and no hepatic lesions at 3.75 ppm. Moderate liver injury was produced by 375 ppm of Aroclor 1242, but this regressed and no hepatic lesions were seen 3 months after mice were returned to a control diet. Other dietary levels of Aroclor 1242 and all levels of 1221 did not produce liver changes. Highly significant mean liver weight increases after 6 ' \. months of feeding occurred in' all Aroclor 1254 groups and in the group fed 375 ppm of Aroclor 1242. Three months after mice were placed on a -3- A UM 00592 1 WATER PCB-00050096 control diet, mean Liver weights of each group had decreased. The decreases were highly significant for the 37.5 ppm Aroclor 1254 and the 375 ppm Aroclor 1242 groups. These observations are generally consistent with those of other investigators which show regression of lesions when PCB dosing is ended, the degree of regression depending upon the duration of the recovery period. Data for mice reported earlier by Nagasaki, et al. (1972) appear to have been included in the later publications of I to, et al. (1973a,b). It is interesting to note that the tumors were called hepatomas in the former publication and were labeled well-differentiated hepatocellular carcinomas in. the latter. Kimbrough and Linder (1974) apparently also felt that these same data on mice were reported in these 2 papers as they reference the production of hepatomas in male dd mice to the later publication by Ito, et al. (1973b). Similar terminology troubles beset the 1974 and- 1975 Nagasaki publications. Data from the later publication list 9/17 males and 4/17 females fed Kanechlor 500 as having liver tumors. Table 1 shows data from the earlier publication; among males, 9/17 had nodular hyperplasia and 7/17 had hepatocellular carcinoma while the female incidence of 4/17 was described as nodular hyperplasia. The hepatomas described by Kimbrough and Linder (1974) were later referred to as "neo plastic nodules (hematomas [sic], hyperplastic nodules)" by Kimbrough, `r. ' et al. (1978). The use of the term hepatoma has created confusion. With respect to terminology of tumors in the mouse liver, "Hepatoma is a collective term used to describe the progressive stages of tumour development from the lesiom called "hyperplastic nodule" or "simple hyperplastic growth" to the morphologically and biologically malignant neoplasms" (Turusov and Takayama, 1979). Nevertheless, whatever 0059/2 WATER PCB-00050097 terminology is used to describe them, tumors were attributed only to PC3s containing 52-54 percent chlorine. Lower doses of the S2-54 percent chlorine materials, as well as lower chlorinated materials, were not described as tumorigenic. Table 2 summaries results of rat studies with PCBs. As in the mouse studies, only some studies with materials having a chlorine content of 52-54 percent, or higher, were reported to produce carcinomas in the livers of rats, and even for those materials this was not a consistent, reproducible observation. Odashima (1976) reported the results from a series of different tests used to evaluate potential carcinogenicity of several compounds, including PCBs. Two tests are of sufficient interest to' mention here. One is the transplacental method in which rats were treated for 3 days during pregnancy. These were days 15, 17, and 19 or 14, 16, and IS, depending on the strain used. The total dose was approximately the maximum one that did not cause abortion or early death of the weanlings. In the other, the newborn method, pups were dosed subcutaneously on days 1, 8, 1-5, and 22 after birth. The maximum dose was one that did not cause early death of over 20% of the animals. The subsequent observation period in each test was limited to one year after birth. For both Kanechlor 300 and Kanechlor 500, the transplacental test was scored' negative and the newborn one as equivocal. Both of these tests gave positive results with some known carcinogens such as 4-aminobiphenyl, beazo[a]pyrene, butyl-nitrosourea and N-methylrN-nitrosourea. - 5 0059 Z3 *0* WATER PCB-00050098 Objective appraisal of chronic rodent feeding- studies is difficult. Some of the rodent studies Ln Table 1 and Table 2 are of the type used to detect potent carcinogens. They use relatively few animals, high doses of the test material, and have a relatively short duration. Studies of that nature have been included in Tables 1 and 2 to illustrate the various diagnoses of the rodent hepatic lesions observed after dosing with PCSs and to aid in the stepwise analysis of the results of all the reported carcinogenicity studies. , In any short-term test with a few animals, the chance occurrence of some tumors is a possibility which must be considered. The probability that an event is a chance occurrence is decreased if it is repeatable. Many of the studies in Table 1 did not produce tumors. Those which were reported to produce hepatocellular carcinoma apparently were also reported in other publications as producing nodular hyperplasia, or tumors, or hepatomas. Since terminology of tumors and the use of those terms have subjective, i.e., judgemental, elements it may not be valid to make direct comparisons between different studies on the basis of terminology alone. The rodent studies which were of less than lifetime duration raise a question as to whether or not cancers would have occurred had the feeding periods been extended. That question is speculative, and it f^ ' cannot be answered from the present data. Nevertheless, negative studies even if they are of relatively short duration are part of the overall evidence which has to be considered. As a first step in the analysis of the data, the evidence shows that at a minimum, PCBs are not potent carcinogens to rodents because they do not produce cancers when tested in studies designed to detect potent carcinogens. -6- ADM 005924 WATER PCB-00050099 With respect to further analysis of the mouse studies, greater significance can be attached to the data of Kimbrough and Linder (1974) because their study had a larger number of animals, a longer duration," and a relatively ~ high dietary level of PCB. In those respects, it more closely resembles conventional cancer studies. They did not observe any hepatocellular carcinomas. Thus, it can reasonably be concluded that PCBs are not carcinogenic to mice. Review of the rat data in Table 2 shows that there are 2 studies on PCBs with an average chlorine content of 40% (Levinskas, 1981 and Weltman and Norback, 1979) and 3 on PCBs with an average chlorine content of 52-54% (Levinskas, 1981; NCI, 1978; Wassenaan, et al., 1978). Results of those studies are consistent with respect to the absence of hepatocellular carcinomas. This consistency reasonably suggests a conclusion that PCBs with those chlorine contents are not carcinogenic. - Of the 3 studies with PCBs having an average chlorine content of 60%, one reported hepatocellular carcinomas (Kimbrough> et al. 1975) and 2 did not (Levinskas, 1981 and Weltman and Norback, 1978) i Since Kimbrough, et al. (1975) and Levinskas (1981) both used Lot No. AK-3 of Aroclor 1260, the different conclusions they reached are not related to differences in the test material. In addition to the use of a different strain of rat, Kimbrough, et al. (1975) used a different histologic diagnostic criteria. Kimbrough, et al. (1975) used the criteria for classification of specific hepatocellular lesions in rats developed at a National Cancer Institute Workshop (Squire and Levitt, 1975). That workshop recommended that the term "neoplastic nodules" replace so-called "hyperplastic nodules" because "Such nodules are proliferative lesions and - 7- A0'1 0059,? 5 WATER PCB-00050100 are known to be induced by carcinogens and, at the least, they indicate an increased probabiity for the development of hepatocellular carcinoma" (Squire and Levitt, 1975). However, Pitot, et al. (1973) in a discussion of stages in the progression of hepatocarcinogenesis in rat liver have noted that "The demonstration of carcinoma cells that appeared to arise within the nodules was also reported (13)1, suggesting the nodule was a precursor to the malignant neoplasm. On the other hand, as was shown by Farber and others, the vast majority of the regenerating or hyper plastic nodules disappeared on removing the animals from the carcinogenic diet. Thus, despite the more recent suggestion that these nodules be termed "neoplastic nodules" (14)2, it is difficult to understand a precursor relationship of the nodule to carcinomas if the existence of the putative precursor is so transient." Further, the use of those criteria for classifying experimental hepatic lesions was considered and rejected by Kimura, et al. (1976) m their studies on the co-carcinogenesis of Kanechlor 400 and 3'-methyl-4-dimethylaminoazobenzene. There is concern that a presently benign lesion might at some future date or under other circumstances be transformed into a malignant lesion. Kimbrough (1979) apparently had that concern when she stated "Although it has been shown many times that the neoplastic nodules are part of the carcinogenic response they are not always included in the statistical f^ ' evaluation of bioassays, which may lead to erroneously interpreted results, particularly when they are classified as hyperplastic nodules or "nodular hyperplasia" and when the number of animals studied was small (Carcinogenesis Testing Program, 1977)".3 A similar concern appears to have been behind the statement in a recent review (Anon. , 1981) that "hexachlorobiphenyl administered to groups of 50 male and 50 female -8 Aum 00 5^6 WATER PCB-00050101 Sprag-ue Dawley rats at dietary levels of 0 and ICO ppm for 105 weeks was carcinogenic in female rats, producing an increased incidence of liver hepatocellular carcinomas among the dosed animals (Norback and Weitman 1980. Personal communication)". Contact with Dr. Norback (Ribelin, 1S81) revealed that her data were available only as an abstract (Weitman and Norback, 1979), and that the abstract and her oral presentation of the data referred to the lesions as neoplastic nodules, i.e., not distinctly tumorous. These 2 examples illustrate the difficulty in establishing that cancer is not present, and they strongly suggest that the different findings reported by various researchers are a reflection of their orientation, training, and philosophical perspective. The latter is defined as the difficulty in separating what is actually being observed under the microscope, from a concern over what it might have become if the animals had lived longer. * Since only an abstract has been published, details on the studies by Weitman and Norback (1979) are limited . Their s til dies are of particular interest because they observed the sequential development of liver changes over a 2-year period. They noted that hexachlorobiphenyl was more toxic than tetrachlorobiphenyl, and that the more toxic, more highly chlorinated hexachlorobiphenyl produced neoplastic nodules only. Again, the weight of the evidence leads to a reasonable conclusion that the carcinogenicity of biphenyls with an average chlorine content of ^60% has not been established. \ ADM 005927 WATER PCB-00050102 Overall, one can surmise that the inability to resolve the crucial issue of whether or not a lesion is neoplastic in character was one of the factors which led to the following statement from a recent Surgeon General's report: "Science and society have not yet arrived at a final consensus on the definition of a carcinogen either in the human population or in experimental animals" (DHHS, 1980). Metabolism Studies Several reviews (Goldstein, 1980; IARC, 1978; Matthews and Kato, 1979; Roberts, et al., 1978; and Safe, 1980) discuss the metabolism of specific isomers as well as mixtures of PCBs. While there are some exceptions, the following general conclusions can be drawn. Both the degree of chlorination and the positions of the chlorine substituents determine the ease with which PCBs are metabolized. In general, the lower chlorinated ones are metabolized and excreted more readily while the more highly chlorinated materials are stored in fat. The process of metabolism converts the fat soluble PCB into a water soluble hydroxylated derivative which can be excreted in the urine. This metabolism and excretion appears to occur via arena oxide intermediates, and the carcinogenicity of some compounds has been attributed to formation of arene oxide interme diates and their binding to subcellular macromolecules. Chlorination at the 4,4' position blocks the metabolism and excretion of PCBs as illustrated below. ' \ - 10 ADM 0059ZB WATER PCB-00050103 Studies in rats (Hansell, et al. , 1977) and in mice (Morales and Matthews 1978, 1979) are of interest because they report the metabolism cf isomers which were fed to rats for 2 years by Weltman and Norback (1979). Hansell, et al. (1977) gave groups of male rats single intraperitoneal injections of 0.2 nmoles/kg of 2,2',5,5'-tetra- or 2,2' ,4,4' ,5,5'hexachlorobiphenyl (TCB and HCB, respectively). They measured changes in hepatic mixed function oxidases and the persistence of the PCS in the livers of animals killed at selected intervals for 35 days after dosing. TCB produced a transient, but significant increase in O-deaethylase activity only at day 3 while HCB produced significant increases in O-demethylase and aniline hydroxylase activities within 24-48 hours. Induced enzyme activities by HCB peaked at 4-6 times control activity during days 7-14 and were about 3 times control activity at the end of 35 days. Even though equimolar amounts of each isomer were given, liver residues of HCB one day after dosing were about 7 times higher- than those of TCB. HCB residues decreased relatively slowly with time while TCB residues were more rapidly eliminated and had almost returned to control values by 35 days. Livers from HCB treated rats had centrolobular necrosis at day 35. They also were significantly increased in size, showed increased amounts of smooth endoplasmic reticulum (SER), and appeared to contain increased numbers of microbodies and reduced amounts of rough endoplasmic reticulum throughout the 35 day interval. By contrast, TCB treated livers were similar to corftrol ones except for occasionally increased aggregates of SER on days 3 and 4. Hansell, et al. (1977) stated "It would be interesting to speculate that.. .2,4,5,2'.4',5'hexachlorobiphenyl undergoes^ direct hydroxylation whereas 2,5,2' tetrachlorobiphenyl undergoes biotrans formation via the arene oxide intermediate, thereby accounting for the different slope for the ` -11- AOM 005929 WATER PCB-00050104 elimination curve of the latter compound." Thus, the metabolism (Hanseli, et al. 197") and feeding1 (Weltman and Norback, 1979) study results lead to somewhat different conclusions. The more potent enzyme inducing-, less readily excreted HCB which appears to resist hydroxylation produces neoplastic nodules in rat livers. By contrast, the more readily excreted TCB, which apparently is excreted via an arene oxide intermediate, does not produce tumors when fed to rats for 2 years. Covalent binding to cellular macromolecules also appears to be greater for the more readily metabolized PCB isomers. Morales and Matthews (1978, 1979) compared the covalent binding of 2 hexachlorobiphenyls; the more readily metabolized 2,2',3,3',6,6'-hexachlorobiphenyl (2,3,6-isomer) and the more slowly metabolized 2,2',4,4',5,5'-hexachlorobiphenyl (2,4,5-isomer). Each PCB, with a radiocarbon label, was given orally to groups of mice at a dosage of 7.28 mg/kg on each of five successive days. Animals' were killed 1, 5, and 8 days later. The concentration of each PCB was determined in liver, muscle, and kidney. AH tissues had consistently higher concentrations of the less readily metabolized 2,4,5-isomer. The more readily metabolized 2,3,6-isomer showed a consistently greater binding to purified macromolecules. The binding was at least one order of magnitude greater than that seen with the 2,4,5-isomer. Results of animal feeding studies with the 2,3,6-isomer would be of ^particular interest to determine the degree of correlation, if any, between the carcinogenicity of this isomer and its covalent binding to cellular macromolecules. Taken as a whole, metabolism studies suggest that if PCBs are carcinogenic, it should be those PCBs which are more readily metabolized and excreted, i.e., the lower chlorinated materials. This is in sharp - 12 005930 WATER PCB-00050105 contrast to results on animal studies in which questions of carcinogenicity arise regarding' higher chlorinated materials only. On balance, metabolism studies on the formation of arene oxide intermediates would lead cr.e to expect lower chlorinated materials to show a more pronounced carcinogenic response in animals. Conclusions drawn from metabolism studies are not supported by animal feeding studies. Co-Carcinogenesis Studies The position and degree of chlorination of PCBs also are important in inducing microsomal mixed function monooxygenases (Goldstein, 1980 and Yoshimura, et ah, 1979). Such induction of microsomal monooxygenases could alter the metabolism of exogenous and endogenous substances in the body, as reported in a variety of studies which have been conducted to determine whether PCBs might be cocarcinogens. These are summarized in Table 3 and discussed in greater detail below. Uchiyama and Chiba (1974) inserted 20-methylcholanthrene impregnated threads into the uteri of virgin mice and fed them diets containing up to 100 ppm of Kanechlor 400 or DDT. Animals were killed at various times and the cervical epithelium was examined for cancerous changes. Kanechlor 400 dosed mice showed no significant changes as compared to the controls, but those receiving 100 ppm of DDT "showed a remarkable tendency towards the induction of cancer." *v ' Diets- containing benzene hexachloride (BHC) isomers and Kanechlor 500 separately and in various combinations were fed to male mice by I to, et al. (1973a,b). Concentrations of the BHC isomers ranged from 250 ppm to - 13 - >0M 005,3, WATER PCB-00050106 50 ppm. The amount of Kanechlor 500 was 250 ppm or 100 ppm. Test groups consisted of 20 to 30 animals. When fed alone, only a-BHC at 250 .ppm- produced a high incidence of nodular hyperplasia and a moderate incidence of hepatocellular carcinoma. A diet containing 250 ppm each of a-BHC and Kanechlor 500 increased the number of hepatocellular carcinomas. In comparison, combinations of 100 or 50 ppm of a-BHC and 250 ppm of Kanechlor 500 yielded a moderate incidence of nodular hyperplasia and produced only a few hepatocellular carcinomas. There were a few nodular hyperplasias in mice fed 100 ppm each of a-BHC and Kanechlor 500. A mixture of 50 ppm a-BHC and 100 ppm of Kanechlor 500 was without effect. Diets containing 250 ppm or 100 ppm of g-BHC and 250 ppm of Kanechlor 500 produced both nodular hyperplasia and hepatocellular carcinomas at a lower incidence than that seen with comparable diets of a-BHC. No liver nodules were seen with 50-ppm of p-BHC and 250 ppm of Kanechlor 500 or with 100 ppm of each. y-BHC and Kanechlor 500 did not produce liver nodules either singly or in combination. Nagasaki, et al. (1974, 1975) reported a similar series of experiments except that Kanechlor 400 was included. The same concentrations of the PCBs, 250 and 100 ppm, were used. Their test groups consisted of 20 to 38 male mice. For a-BHC and Kanechlor 500, they presented essentially the same data as I to, et al. (1973a,b). While Kanechlor 400 did not produce liver nodules when fed alone at 250 ppm, ' 'V ` . there was an increase in the incidence of hepatocellular carcinomas when it was fed in combination with 250 ppm of a-BHC. The increase was similar to that reported by I to, et al. (1973a,b) for a-BHC and Kanechlor 500. A combination of 100 ppm vcf a-3HC and 250 ppm of Kanechlor 400 produced only a few nodular hyperplasias and 100 ppm of each in the diet did not induce any liver nodules. - 14 - ADM 005932 WATER PCB-00050107 The effects of PC3s on tumor induction by diethylnitrosamir.e (DEN) have been studied extensively. Male rats were given 25 ppm of DEN in their drinking1 water and 500 ppm of Kanechlor 500 in their diet concurrently for 20 weeks, returned to a control diet for 4 weeks, and then killed (Makiura, et al. 1974). Neither liver tumors nor nodular hyperplasia were seen, even though rats receiving the same DEN treatment alone had a 92% incidence of liver cancer. The livers of rats treated with Kanechlor 500 only also had no tumors. When administered after DEN, Kanechlor 500 markedly enhanced liver tumor production as reported in a series of papers by Nishizumi (1976, 1979a, 1979b). He described studies in which male rats were 'given 50 ppm DEN in drinking water for 2 to 10 weeks, after which they were intubated twice weekly with a corn oil solution of Kanechlor 500 for 6 to 12 weeks. Kanechlor 500 doseswere either 0.2 ml of a 5% or 0.1 ml of a 10% solution. Since experiments were terminated at 20 or 52 weeks after dosing, some animals received untreated diets prior to sacrifice. Only one paper (Nishizumi, 1976) contained results for animals dosed only with Kanechlor 500 alone. That treatment produced enlargement of the livers of rats but no neoplastic lesions. After pre treatment with DEN, however, hepatocarcinogenesis was enhanced as evidenced by the earlier appearance and a significant increase in the number of tumors. A similar dosing pattern was used by Preston, et al. (1981). Male rats were given drinking water containing 66 Mg DEN/ml (66 ppm) for 5 weeks after which they were fed for 18 weeks on a control diet or one containing 100 ppm of either of 2 Aroclor 1254 diets. One was Aroclor 1254 as received. The other was an Aroclor 1254 which `the authors claimed to have purified by removing polychlorinated diber.zofuran (PCDF) impurities by adsorption onto and subsequent elution from activated rlorisil. No analysis was made of Aroclor 1254 as received for . - 15 - AOM 005933 WATER PCB-00050108 PCDF. The authors reported recovery of PCDF at a level comparable to 30 pgr of tetrachlorodibenzofuran. (TCDF) per 10 g of A roc lor 1254. TCDF was used as a positive control and a standard for quantification for the purification process. Other animals received one of the 2 Aroclor 1254 diets only during1 the latter 18-week interval. No evidence of hepatic tumor formation was seen in control animals or those receiving ICO ppm of either Aroclor 1254 diet. However, using the diagnostic criteria of Squire and Levitt (1975), they reported that both Aroclor 1254 diets resulted in significantly greater incidences of hepatocellular carcinomas in rats pretreated with DEN as compared to those dosed with DEN alone. Thus, they concluded that the hepatic tumor-promoting ability appears to reside in Aroclor 1254 itself. As the authors of this paper pointed out, it cannot be concluded that PCDFs are not promoters of hepatocarcinogenesis since appropriate studies have not been done on PCDFs alone. Similarly, their findings do not invalidate the hypothesis that some of the other effects reported for some PCBs may have been due to PCDF impurities. The dependence of the inhibition or promotion effect of PCBs on DEN-induced tumors on the dosing sequence was illustrated in a different manner by Nishizumi (1980). Pregnant female rats were given oral doses of 200 mg/kg or 50 mg/kg Kanechlor 500 on days 5, 10, and 15 of gestation and were allowed to deliver their pups. At^ 28 (days. of age, pups were given 50 ppm of DEN in their drinking water for 5 weeks. Groups of these pups were sacrificed at 16, 20, and 24 weeks after the start of DEN dosing and their livers were examined. The number of liver tumors in the offspring from Kanechlor 500vtreated dams was significantly decreased as compared to the controls. The decreases were more pronounced in males. This decrease in DEN-induced tumors apparently resulted from induction of - 16 A0H OO5934 WATER PCB-00050109 microsomal enzymes in the livers of the pups. Those livers contained Kanechlor 500 residues and electron microscopy showed an increase o the smooth endoplasmic reticulum of hepatic cells. Rainbow trout were fed 6 ppb of aflatoxin B: (AFBj), or 100 ppm of Aroclor 1254, or a combination of both for a year (Hendricks, et al., 1977). At intervals during the year, some fish were killed. The remainder were killed at the end of that time. All livers were examined grossly and microscopically. There were no tumors in those fed a control diet or Aroclor 1254 alone, and growth was not affected by 100 ppm of Aroclor 1254. The number of tumor-bearing fish in the group fed AFB:, plus Aroclor 1254 was significantly reduced (to less than one-half) when compared to those fed AFBi, alone. There also were fewer tumors per liver and the tumors were smaller in those fed the mixture. In the studies cited earlier, Makiura, et al. (1974) also fed combinations of 500 ppm Kanechlor 500 with 300 ppm of 3,-methyl-4-dimethylaminoazobenzene (3'-Me-DAB) or 150 ppm of N-2-fluorenylacetamide (2-FAA). Results similar to those when PCBs were fed concurrently with DEN were obtained. Kanechlor 500 reduced the incidence of liver tumors to zero from 65% for those treated with 3'-Me-DAB and from 54% for those treated with 2-FAA. However, the effect of PCBs on the tumorigenicity of 3'-Me-DAB appears to depend on the dosing sequence as shown for DEN. Kimura, et al. (1976) used different sequences to feed groups of rats diets containing 400 ppm of Kanechlor 400 for six months and 600 ppm of 3'-Me-DAB for 2 months. Spme received Kanechlor 400 alone and some only 3'-Me-DAB. Others received one diet, then the other, after a 2-month interval on control diet. A final group received Kanechlor 400 for 005935 WATER PCB-00050110 4 months and both materials for another 2 months. No hepatocarcinomas were produced by Kanechlor 400 alone or when it was given before or during' the overlap with 3'-Me-DAB. No hepatocarcinomas occurred in control animals. The incidence of liver cancer rose from 13% Ln those receiving 3'-Me-DAB alone to 64% in the group which received Kanechlor 400 after 3'-Me-DAB. The inhibitory effect of -Kanechlor 500 on rat liver tumors was evident when rats were dosed with two carcinogens (Makiura, et al.. 1974). Combinations of 3'-Me-DAB and DEN or 2-FAA and DEN were administered with and without Kanechlor 500 at the concentrations stated earlier. The liver cancer incidence fell from 92% to 8% when Kanechlor 500 was given to rats dosed with 3'-Me-DAB and DEN. It went to zero from 82% for those receiving 2-FAA and DEN. .' Nishizumi (1979a,b) studied the effects of various combinations of DDT and sodium phenobarbital (SPB), in the absence and in the presence of Kanechlor 500, on the induction of rat liver. hepatocellular carcinoma by DEN. Pretreatment of rats with DEN followed by DDT, by SPB, or by a combination of both produced low incidences of liver cancer while DEN pretreatment alone did not induce cancer. When Kanechlor 500 was included with each of the preceding dosing regimens, there was a marked f^ * ' increase in liver cancers. Increases resulting from joint administration of compounds after pre treatment with DEN were lower than those observed for DEN followed by Kanechlor 500 alone discussed earlier. I to, et ah (1978) fed diets containing 200 ppm of 2-FAA to male rats for two weeks and then a diet containing 1000 ppm of an unspecified Kanechlor - 18 - AOM 0059 WATER PCB-00050111 mixture for 8 weeks. Partial hepatectomies (PH) were performed on seme rats during the third week of the study. Feeding of 2-FAA was without effect, but that diet plus PH produced a few hyperplastic nodules in the liver. Treatment with both diets caused an even greater incidence of hyperplastic nodules and both diets combined with PH produced a marked rise in the number of liver nodules. DiGiovanni, et al. (1977) and Berry, et al. (1978, 1979) studied Aroclor 1254 in a two-stage mouse skin carcinogenesis assay to see if it was a tumor initiator or promoter. The shaved skin of female mice was dosed with Aroclor 1254 at a level of 100 pg or 625 pg per mouse. This was applied alone as an initiator or from 5 minutes to 72 hours before initiation with 7,12-dimethylbenzla]anthracene (DMBA). One week later, mice received twice weekly applications of 5 pg of the phorbol diester promoter 12-0-tetradecanoylphorbol-13-acetate (TPA) for 32 weeks. Animals were observed for both papillomas and carcinomas. Aroclor 1254 alone produced a few papillomas. The authors of these reports apparently faced a dilemma common to scientists, i.e., how much significance should be attached to the observation of a low incidence finding which may or may not have a causal relationship to treatment. On the basis of these few papillomas, Aroclor 1254 was called a "weak tumor initiator" (DiGiovanni, 1977). Later, it was described as possessing "little or no tumor-initiating properties" (Berry, et al., 1979). While Aroclor 1254 had a negligible effect on tumor induction when given 5 minutes before an initiating dose of DMBA, it markedly inhibited tumor induction when given 18 to 72 hours before DMBA. In other studies, an initiating dose of 200 nmole of DMBA was applied to the shaved backs of mice. After one week, 100 pg of Aroclor 1254 was applied twice weekly for 30 weeks. Aroclor 1254 failed to 00593 WATER PCB-00050112 promote tumors while 0.2 pg doses of TPA applied m a similar manner induced an average of 8 papillomas per mouse. The authors concluded that Aroclor 1254 possessed little or no tumor-initiating- or tumor-promoting properties. The transplantability and growth of Walker 256 carcinosarcoma in rats was inhibited by Aroclor1254 (Kerkvliet and Kimeldorf, 1977a,b). Diets containing up to 800ppm of Aroclor1254 were fed to rats of both sexes for 30 days, after which they were given intramuscular injections of Walker tumor cells. Nine days later, during which time they continued to receive Aroclor 1254in their diets, animals were killed and the tumors were dissected out and weighed. Mean tumor weights of all Aroclor 1254 fed groups were significantly reduced as compared to their sex-matched controls, and the reductions were dose-related to the dietary level of Aroclor 1254. Body weight gain was depressed in males receiving 400 ppm or more of Aroclor 1254. Females showed reduced body weight at 100 ppm, the lowest level fed to that sex. Intraperitoneal injections of 50 to. 200 mg/kg every other day for 14 days after injection of a small inoculum of Walker 256 cells (103) inhibited both the development and growth of the Walker tumor. The number of tumor takes and the size of the tumors were reduced and the tumor latency period was increased. Body weight gain was reduced only at the 200 mg/kg dosage. Alternate day intraperitoneal injections of 100 mg/kg and 200 mg/kg of Aroclor 1254 after a large inoculum of Walker 256 cells (107) were continued for 60 days. Control animals receiving tumor cells only had 100% mortality by day 20 with a mean survival time of 13.5 days. Mean survival times were significantly increased to 17.7 days and 18.9 days for animals dosed with 100 mg/kg and 200 mg/kg of Aroclor 1254. A few animals survived to 60 - 20 ADM 00593 6 WATER PCB-00050113 days, and four receiving' the higher dosage of Aroclor 1254 showed total regression of their bilateral tumors. Tumor growth rates in Aroclor-treated animals were significantly reduced. Again, body weight gain was depressed only at the 200 mg/kg dosage. An experiment was undertaken to determine if there was an optimum time period for the antitumor effect of Aroclor 1254. Intraperitoneai injections of 100 mg/kg of Aroclor 1254 were given on 5 consecutive days before tumor inoculation, cn 5 days after inoculation, and daily from 5 days before until 10 days after inoculation. This study was ended 14 days after the initiating tumor inoculum was given. While all dosing schedules reduced tumor growth, there were variations in the responses. The greatest inhibition of tumor growth resulted from dosing animals for 15 days. Almost equally effective were pretreatment and treatment starting with inoculation of the Walker ? cells. The delayed treatment starting after 5 days was least effective as the tumor had become established. While early treatment reduced tumor growth, it was less effective in preventing metastases and death of the animals. Kerkvliet and Koller (1980) offered groups of male' mice diets containing 10, 100, or 500 ppm of Aroclor 1254 for 15 weeks prior to injecting them with MSB, an established tissue culture cell line derived from Moloney sarcoma virus. They measured tumor growth and cell-mediated toxicity over the next 19 days. A dietary level of 500 ppm of Aroclor 1254 resulted in marked weight loss and deaths. This was reduced to 250 ppm at 11 weeks, and a few weeks later surviving animals were placed on a control diet. They reported that Aroclor 1254 pretreatment enhanced the growth rate of the MSB tumor and inhibited or delayed -the development of cellular immunity against the tumor. - 21 - AOM 005V3 WATER PCB-00050114 KoUer (197?) fed groups of mice diets containing 3.75, 37.5, or 375 ppn cf Aroclor 1221, Aroclor 1242, or Aroclor 1254 for 6 months. About 2 weeks after being placed on test diets, some mice in each group were Inoculated intraperitoneaUy with Moloney leukemia virus. None of the dietary levels of any Aroclor affected, i.e., did not promote or induce, the oncogenesis of Moloney leukemia virus. Since PCBs' are enzyme inducers, their biological effects resemble those of other enzyme inducers. Peraino, et al. (1978) noted that phenobarbetal had a specific tumorigenic enhancing effect that was restricted to the liver. With the exception of the mouse skin painting studies and those in which ttimer cells or a virus were injected, all of the studies in Table 3 were concerned with liver tumors. Peraino, et al. (1978) also pointed out resemblances between phenobarbetal and PCBs. Both substances "enhanced hepatic tumorigenesis" when given after DEN, and both "exerted a protec tive effect against hepatic tumorigenesis" when administered concurrently with AAF or DEN. The comparison between PCBs and phenobarbital is extended by the observation of Berry, et al. (1978, 1979) that PCBs did not promote skin tumors in mice pretreated with DMBA. Peraino, et al. (1978) refer to a study4 in which skin tumorigenesis was not enhanced in mice fed phenobarbital after skin painting with DM3 A. Lichti, et al. (1978) described the induction of ornithine decarboxylase (ODC) in mouse epidermal cell cultures by TPA. Aroclor 1254, apparently at much higher concentrations than TPA, induced a low but reproducible \' stimulation of ODC in the same 'system. They also noted a report5 that a high intraperitoneal dose of Aroclor 1254 induced ODC in the liver of rats. After commenting that PCBs "warrant further investigation into their WATER PCB-00050115 possible action as tumor promoters," they added that "These compounds are being' tested on carcinogen-initiated mouse skin (T.J. Slaga, personal communication)". The series of publications by Berry, et al. (1978, 1979) and DiGiovanni, et al. (1977) show that PCBs are not promoters on mouse skin. , Authors of some of the mutagenicity studies to be discussed (Danz and Urban, 1980; Norback, et al., 1981; and Stadnicki, et al. , 1979) have suggested that PCBs may act as promoters of carcinogenicity. In general, those comments appear to have been offered as hypotheses to aid in understanding the observations which had been made. This also appears to be the case for the NCI study on Aroclor 1254 which was reviewed by the Data Evaluation/Risk Assessment Subgroup of the Clearinghouse on Environmental Carcinogens. That group, charged with the responsibility of providing a peer review of NCI bioassay reports on chemicals studied for carcinogenicity, made and accepted a motion to add the following to the report summary: "Based on the liver proliferative lesions in the treated rats and published reports, it is suggested that Aroclor 1254 may be a tumor promoter" (NCI, 1978). -` . With respect to tumor promotion, Weisburger and Williams (1980) state that "certain inducers of liver metabolic enzyme systems, such as phenobarbital, DDT and BHT, when administered after minimal doses of primary hepatocarcinogens exerted a powerful promoting effect". As mentioned earlier, PCBs also induce liver metabolic enzyme systems. However, since the carcinogen alone produced tumors in animals in the co-carcinogenesis studies summarized in Table 3, it appears that the doses were greater than minimal. In addition, even though the co-carcinogenesis studies were of - 23 - q 1,0 *0* WATER PCB-00050116 shorter duration, several used much higher dietary levels of PCBs than those fed to rats for 2 years. The latter studies are the ones which gave rise to questions of carcinogenicity. Thus, while the evidence Indicates that PCBs are not carcinogenic, their reported effects in rodent livers following prolonged exposure in conjunction with an initiator may be promotion or inhibition. When administered to animals together with a biological agent, PCBs show a promoting or inhibitory effect similar to that seen with chemical agents. Pretreatment with PCBs inhibited the growth of Walker 256 carcinosarcomas in rats and increased their survival time (Kerkvliet and Kimmeldorf, 1977a,b), enhanced the growth of MSB tumor in mice (Kerkvliet and Roller, 1978) and neither promoted nor induced the oncogenesis of Moloney leukemia virus in mice (Roller, 1977). Mutagenicity Studies Mutagenicity testing of PCBs has ranged from bacterial systems to intact animal studies. A series of studies with eleven bacterial test strains (Heddle and Bruce, 1977; Hsia, et al. 1978; McMahon, et al. 1979; Odashima,, 1976; Probst, et al., 1981; Sugimura, et al. , 1976; and Wyndham, et al., 1976) are summarized in Table 4. Hsia, et al. (1978) used both phenobarbital and Aroclor 1254 to induce jliver enzymes in rats for preparation of S-9 activation systems. They also tested 4-hydroxy2,2',5,5'-tetrachlorobiphenyl and 2,2',5,5'- tetrachlorobiphenyl-3,4-oxide, a known and a presumed metabolite of 2,2',5,5'-tetrachlorobiphenyl, \ respectively. All three materials gave negative responses with each activation system. Odashima (1976) presented data in tabular form from a series of screening tests. In addition to the results shown in Table 9, - 24 - . ADM 005^ 1 WATER PCB-00050117 bacterial strains WP2, TA100, TA98, H-17, >1-45, W3I10, and TA1978 are shown in groups in their Table 5 with a text notation that not aU chemicals were tested with each strain. For Kanechlor 300 and 500, such of the preceding: strains as were tested showed a negative response. The group consisting: of H-17 & M-45, WP2 try" (her'' & her') shows a plus sign for Kanechlor 300. Presumably, one or more of those bacterial strains gave a positive response. Kanechlor 500 was listed as giving' a negative response with the latter group. With the exception of Wyndham, et al., (1976), all of the studies were negative with and without the addition of a microsomal enzyme activation system. It is somewhat difficult to reconcile the text and the graphs in the publication by the latter authors. They stated that their "results clearly showed that as the degree of chlorination decreased the mutagenicity increased, a concentration of 100 pg of 4-chlorobiphenyl in the test medium gave over 2000 revertant colonies per plate. The higher chlorinated biphenyls show very little activity as mutagens". While Figure 3 in their article shows their marked increase in revertants for 4-chlorobiphenyl, the same figure also shows results for Aroclor 1221, which averages 1.15 chlorine units per molecule. The mutagenicity of Aroclor 1221 is not discussed in the text, but Figure 3 shows that at a concentration of 100 pg per plate, it produced about one-tenth of the revertant colonies that 4-chlorobiphenyl did. Thus, a 15% increase in chlorine content produced a 10-fold decrease in the reported mutagenic activity. In light of that sharp reduction of mutagenic activity with a slight chlorine increase, it is difficult to understand the statement in their text that Aroclor 1254 "was only weakly mutagenic." Results from Aroclor 1254 (average of 4.96 chlorine per molecule) are not shown in their - 25 - ADM 005942 WATER PCB-00050118 Table 3, but 2,2',5,5'-tetrachlorobiphenyl (average of 4 chlorine per molecule) was presented. At 100 pg per plate, 2,2,,5,5'-tetrachlorobiphenyl was virtually devoid of mutagenic activity. Overall, primarily because McMahon, et al. (1979) reported that 4-chlorobiphenyl was not mutagenic, it can only be concluded that the findings of Wyndham, et al. (1976) are aberrant. A similar conclusion that the findings reported by Wyndham, et al. (1976) are unfounded because attempts to repeat them have been unsuccessful was reached by the State of California (Anon. 1981). In a variant of this test, Stott and Sinnhuber (1978) used Aroclor 1221, 1242, 1254, and 1260 to induce the mixed function oxidase system of the liver in trout. The submitochondrial fraction of those livers was used to activate the metabolism of ATBi which was assayed for mutagenicity using strain TA 1538 of S. typhimurium The mutagenic response was decreased compared to the concurrent control using untreated trout liver. A general pattern of decreasing response with increasing degree of chlorination was observed, except for Aroclor 1260. Induction of mutagen detoxifying enzyme systems was suggested as a possible explanation for the apparent conflict between the reported induction of trout mixed function oxidases and the decrease in mutagenic activity. These results are consistent with those of Hendricks, et al. (1977) discussed earlier who reported that Aroclor 1254 reduced hepatic tumors in trout fed AFBi. On the basis of slower sedimentation rates in alkaline sucrose gradients, Stadnicki, et al. (1979) reported that 2,2' ,5,5'-tetrachiorobiphenyl (TC3), a mixture of the 3-hydroxy and 4-hydro.xy derivatives of TC3 and the 3,4-epoxide of TC3 induced single strand breaks in DNA of L-329 cells. - 26 ADM 005943 WATER PCB-00050119 At 100 pg/ml, each of the three materials caused all of the Dh'A to come out in fractions at the top of the gradient. The epoxide caused some breakage down "to I jjg/ml. " The mixture of hydroxy derivatives caused significant breakage at 20 pg/ml and only slight breakage at 1 and 10 pg/ml. TCB, the least potent, caused lesser breakage at 20 pg/ml and was without effect at 1 and 10 pg/ml. ' Nilsson and Hamel (1974) conducted genetic tests on adults and Larvae of Drosophila melanogaster fed Clophen 30 and Clophen 50. These PCB mixtures were without effect on the loss of sex chromosomes used to measure chromosome breaking action and nondisjunction of the sex chromosomes. Tazima (1980) has described a specific locus test using the silkworm (Bombyx mori). Neither Kanechlor 300 nor Kanechlor 500 showed mutagenic activity in that system. Hoopingarner, et al. (1972) induced mitosis in cultured human lymphocytes with phytohemagglutinin and treated them with 100 ppm of Aroclor 1254. There was no effect on the mitotic index, satellite association, chromatid gaps or chromatid breaks in comparison to control human lymphocytes. Odashima (1976) also studied the incidence of chromosomal aberrations. Their in vitro test used Yoshida ascites sarcoma cells cultured for 6 to 72 hours in the presence of PCBs at concentrations producing a minimal or 50% growth inhibition. Kanechlor 300 gave a positive response and Kanechlor 500 a negative one. For an in vivo system, they examined bone marrow cells 6 to 48 hours after adult or newborn animals were given an approximately lethal dose of PCBs. This test gave the opposite result; Kanechlor 500 was positive and Kanechlor 300 was negative. They noted . - 27 - AOM 00594-* WATER PCB-00050120 that chromosomal aberrations frequently . occurred in controls and that chemicals were called positive when they induced more than twice the aberrations in controls. In a Syrian' hamster cell transformation system (Pienta, 1980), Aroclor 1254 was one of several chemicals tested double-blind. It gave a negative result. Mouse embryo fibroblasts also have been exposed to different PCBs. Nesnow, et al. (1981) studied the cocarcinogenic action of agents which increase microsomal mixed-function oxidase activity in the C3H10T5sCL8 transformation assay. After a 48-hour pretreatment with Aroclor 1254, cells were then treated with benzo(a)pyrene [B(a)P] and the agent for an additional 24 hours. Aroclor 1254 did not increase B(a)P-mediated transformation and no Type II or Type III foci were observed. Norback, ' et al. (1979, 1980, 1981) exposed C3H10T% cells continuously for 6 weeks to 10 pg Aroclor 1254 per ml of medium. Treated cells developed Type III foci. Cells exposed to the same concentration of Aroclor 1254 for 24 hours or to 1 pg of Aroclor 1254 per ml of medium did not develop Type III foci. Continuous exposure of cells to Aroclor 1260 and 2,4,5,2',4',5'-hexachlorobiphenyl also caused formation of Type III foci. A clone from a focus transformed by Aroclor 1254 induced sar coma formation when inoculated in irradiated mice. Foci from Aroclor 1260 and 2,4,5,2',4',5'-hexachlorobiphenyl had not been characterized further. Since transformation occurred only after continuous^ exposure, Norback, et al. (1981) suggested that the effects of PCBs in culture include promotion. When fed to rats, however, 2,4,5,2',4\5'-hexachlorobiphenyl produced neoplastic liver nodules, not hepatocarcinomas (Weltman and Norback, 1979). - 28 - A DM 00 59 *t 5 WATER PCB-00050121 Wong1, et al. (1979) claimed that 4-chlorobiphenyl induced an increase in unscheduled DNA synthesis in Chinese hamster ovary cell cultures in the presence of hydroxyurea, a chemical agent which suppresses normal replicative DNA synthesis. Few details were presented regarding the validation and reliability of their test system. By contrast, Aroclor 1254 gave a negative response in an unscheduled DNA synthesis m primary cultures of adult rat hepatocytes (Probst, et al. , 1981). The latter authors presented results of an extended series of compounds tested in their system. Danz and Urban' (1980) have proposed using the mitogenic response of the rat adrenal cortex after dosing the animals with a test substance as a short-term test for evaluating the promoting action of chemical compounds. Clophen A60, Delor 103, Delor 106s, and Phenoclor DP6 all gave positive responses in their test system. - At 3-6 days of incubation, embryos from ring doves (Streptopelia risoria) fed a control diet or one containing 10 ppm of Aroclor 1254 were examined for cytogenetic changes (Peakall, et al., 1972). The relative frequencies of chromosome aberrations in the 8 largest chromosome pairs in metaphase cells of allantoic sac and limb bud origin were scored. The 6 control embryos had a mean aberration rate of 0.8% (0-2.0). The aberration rate in 17 embryos from Aroclor 1254 treated birds was 1.8% (0-9.4). In the latter group, there was one chromosome rearrangement, 13 embryos with aberration rates exceeding the mean control rate, and 4 embryos vyhich exeeded the highest control rate. Aroclor 1242 was injected into fertile White Leghorn eggs to give estimated final concentrations of 10 or 20 ppm (Blazak and Marcun, 1975). After 4 or 5 days of incubation, eggs were - 29 ADM 005946 WATER PCB-00050122 injected with colcemide, incubated an additional 45-60 minutes, and embryos were harvested for examination. There was a high degree of early embryonic death, and a few live embryos showed drastically retarded development without malformation. The first five pairs of chromosomes, constituting over 50% of the chromatin material per cell, were examined for detection of clastogenesis. There were no chromosomal aberrations. Heddle and Bruce (1977) injected mice with Aroclor 1254 for five consecu tive days and then examined bone marrow preparations for chromosomal breakage and sperm cell preparations for sperm with abnormally shaped heads. While nonspecific factors can induce sperm abnormalities, the authors believe the latter also can result from point mutations or small deletions. Aroclor 1254 was judged to be neither carcinogenic or mutagenic. Dikshith, et al. (1975) gave male rats oral dosages of 50 mg/kg of Aroclor 1254 on each of 7 consecutive days. Animals were killed at intervals over the next 3 days. Those scheduled for cytogenetic analysis were injected with colchicine 2 hours before killing, after which time the seminiferous tubules were prepared for such study. At autopsy of the other animals, testis, epididymis, and liver weight were recorded and sections were prepared for microscopic study and histochem.ical determi nations . Livers were markedly enlarged and showed significant increase in weight compared to controls. There were no differences in body weight or weight and appearance of testis, epididymis, or vas deferens between control and treated rats. Arqelor 1254 treated rats showed a few meta phase figures with abnormal chromosomes whch appeared to be sporadic and not specific to any one type. Histological examination showed testis - 30 - AOM 005997 WATER PCB-00050123 and epididymis from Aroclor 1254 treated rats were comparable to controls although interstitial cells were increased in Aroclor 1254 dosed rats. These cells showed an increase in acid phosphatase activity. The authors concluded that they "found no evidence to suggest that Aroclor 12S4 causes significant chromosome damage or histopathological changes in Lie rat testis." Similar studies were conducted by Green, et al. (1973, 1975a) who gave male rats single oral dosages of 5000 mg/kg, 2500mg/kg or 1250 mg/kg or four successive daily dosages of 500 mg/kg of Aroclor 1242. Other rats received five consecutive daily dosages of 300 mg/kg, 150 mg/kg or 75 mg/kg of Aroclor 1254. Animals were injected with colcemide 3 or 4 hours before sacrifice which occurred about 24 hours after the single dose or the series of doses. There were some body weight losses and some deaths occurred. Bone marrow from rats dosed with both PCB mixtures and spermatogonia! preparations from Aroclor 1242 treated --its were prepared for cytogenetic study. Repeated dosages of 150 mg/kg and 300 mg/kg of Aroclor 1254 produced decreases in the number of mitoses in bone marrow cells. Repeated dosages of 75 -mg/kg of Aroclor 1254 and all dosages of Aroclor 1242 were without effect. Neither PCB mixture at any dosage produced a significant number of chromosomal abnormalities in bone marrow cells. At the lower dosages, Aroclor 1242 did not affect mitoses of spermatogonial cells, but 5000 mg/kg or 4 dosages of 500 mg/kg significantly reduced the rate of cell division. No dosage of Aroclor 1242 produced cytogenetic abnormalities in spermatogonial cells. The authors concluded from their studies that Aroclor 1242 and Aroclor 1254 did not possess mutagenic potential. Garthoff, et al. (1977) fed male rats diets containing 5, 50, or 500 ppm of Aroclor 1254 for 5 weeks, after which they 005948 WATER PCB-00050124 examined bone marrow and testis samples. There was no significant difference between , test and control animals with respect to the incidence of chromosomal abnormalities and the number of cells in mitosis from bene marrow and spennatogonial cells. In a dominant lethal study (Keplinger, et al., 1972 and Calandra, 1976), albino mice were given single intraperitoneal dosages of 500 or 1000 mg/kg of Aroclor 1242, Aroclor 1254, or Aroclor 1260 and mated on successive weeks to virgin females. There was no evidence of mutagenic effects. Although details on their studies are limited, their results are in general agreement with those from another dominant lethal study in rats conducted by Green, et al. (1975b) with Aroclor 1242 and Aroclor 1254. The former was administered orally at a single dosage of 625, 1250, or 2500 mg/kg or in five daily dosages of 125 or 250 mg/kg. Aroclor 1254 was given orally in five daily dosages of 75, 150, or 300 mg/kg. After dosing, they' were mated with untreated females for 10-11 weeks. Another group of rats was given 150 mg/kg of Aroclor 1254 for five- successive days and starved overnight before admittance to females. Other male rats were offered diets containing 25 or 100 ppm of Aroclor 1254 for 70 days, then mated with untreated females for one week. TEM (triethylenemelamine) was used as a positive control. There was a body weight loss and some deaths occurred in a few groups. Neither the oral dosing nor the dietary feeding of Aroclor 1242 or Aroclor 1254 had any effect on the number of implantations or the number of dead implantations per pregnant female while TEM produced significant postimplantation losses in weeks 2, 3, and 4. These authors noted that the reduction in the number of dividing spennatogonial cells reported earlier for Aroclor 1242 (Green, et al. . 1973, 1975a) did not impair the reproductive performance of male animals. - 32 - A O/M 5949 WATER PCB-00050125 Review of the various mutagenicity tests and their results prompts three comments. Chlorinated hydrocarbons do not generally give positive results in Salmonella strains. Therefore, the mutagenic responses with TA153S (Wyndham, et al. 1976) are either aberrant or else the response is, indeed, limited to essentially monochlorobiphenyl. The degree o'f correlation between in vitro mutagenicity tests and carcinogenicity depends upon the initial classification of the test materials. Pienta (1980) classifies Aroclor 1254 as a non-carcinogen. Rinkus and Legator (1980) regard Aroelor 1254 and Kanechlor 500 as having known or suspected carcinogenic activity. Odashima (1976) and Sugimura, et al. (1976, 1977) consider Kanechlor 500 to be carcinogenic and Kanechlor 300 to be non-carcinogenic. Apart from the difficulties they present for correlation, these different opinions about the carcinogenicity of PCBs are a reflection ? of what data these individuals considered and how they evaluated those data. Finally, in light of the large number of tests conducted to evaluate various mutagenic parameters, it is not surprising that an occasional suspicious or positive finding resulted, simply on a statistical basis. Those occurred in in vitro systems. In vivo tests gave negative results and provide a basis for concluding that ambient levels of PCBs do not present a mutagenic risk. Epidemiology Studies ^ With, respect to epidemiologic studies, it should be noted that there are frequent references in the literature to an epidemiologic study of workers exposed to Aroclor 1254 (Bahn, et al., 1976, 1977). Those articles are cited in two recent reviews, tone states "The epidemiological data provide suggestive evidence of a relationship between exposure to polychlorinated biphenyls and the development of malignant melanoma. . . .for practical - 33 40m 0q5 950 WATER PCB-00050126 purposes, polychlorinated biphenyls should be regarded if they were carcinogenic to humans" (IARC, 1978). The other states "No conclusive evidence has thus far been reported which demonstrates that occupational exposure to PCBs has caused an increased incidence of cancer" (Kimbrough, 1980). The latter author then proceeds to discuss the study by Bahn, et al. (1976). Among the references for these 2 renews are NIOSH (1978) which has the following comment on the Bahn study, "PC3 exposure histories were based on recollections of two company employees. Exposures to other chemicals could not be ascertained.----- To correct these deficiencies in the preliminary study, a more intensive investigation is being conducted (B.N. Kightlinger, written communication, November 1976). A substantial change has occurred in the cohort since release of the preliminary report by Bahn and her coworkers, and it seems likely that the findings on this new cohort will differ significantly from those of the preliminary study. The final report is not yet available." - Recently, Gaffey (1981) reviewed and evaluated existing reports concern ing health effects and exposure to PCBs. The reports he discussed dealt with diverse potential health effects. With respect to liver effects , he concluded that "Alterations of liver function and fat metabolism associated with PCB exposure have been observed in several studies, but are characterized by investigators as mild and of no clinical significance. " '> ' He also concluded that "Mortality studies concerned primarily with cancer present problems of interpretation due to the small sample size of some of the studies, and to the confounding effect of other exposures. However, they do exhibit a pattern, which is that none of the studies agree on the cancer sites at which an excess mortality was found, and the excesses that . ' - 34 - .' ' AOA 0059b1 WATER PCB-00050127 were found are in general not statistically, significant. One must conclude that the findings of the mortality studies reflect a sporadic pattern cf excess mortality at different sites which is not consistent with a carcinogenic effect of PCBs. In addition, where an examination of duration and latency of exposure was possible, no association with these variables was found [32]7. "Taken as a whole, the epidemiologic studies find that high occupational exposures to PCBs may cause dermatitis of various kinds, but that there are no other clinically observable effects, including the occurrence of cancer." Summary Some remarks by Cole and Merletti (1980), although written in ' a more general vein, appear to be particularly suited for ending a discussion on the potential carcinogenicity of PCBs. "In view of the difficulty of deciding whether a "suspect" carcinogen is in fact a weak carcinogen or in fact harmless, we point out one aspect of this scientific judgement which, though well known, is often lost sight of; namely, that in prin ciple, it is more likely that a non-carcinogen will appear to be a weak carcinogen than the reverse. This asymmetry should be fully appreciated by legislators and regulators. It is summed-up succinctly, if not very accurately, in the oft-heard phrase "you can prove a positive but not a negative". This is clearly illustrated in the case of PCBs. Review of the results of a lar^e number and wide range of chronic feeding and metabolism studies in animals and of mutagenicity studies in various systems has failed to establish that PCBs are carcinogenic. However. 005952 WATER PCB-00050128 presently available rodent data raise some suspicions and lead to disputes about the carcinogenicity of PCBs. Attempts to compare and evaluate results of rodent studies reported in the literature are complicated by different uses of similar terminology and variations in the criteria employed for diagnosing hepatic tumors in rodents. Co-carcinogenesis studies have reported both promotion and inhibition of tumors in rodent livers following prolonged administration of PCBs in conjunction with an initiator. Thus, it is not likely that animal studies and other laboratory procedures will lead to a universal consensus on the carci nogenicity or non-carcinogenicity of PCBs. While epidemiology studies of humans exposed to PCBs present problems of interpretation due to their small sample size and the confounding effect of other exposures, they present the best available evidence for assessing the carcinogenic potential of PCBs to humans. Epidemiology studies, including recent studies involving large cohorts with lengthy exposures to PCBs, demonstrate a pattern that is not consistent with a carcinogenic effect of PCBs. \ - 36 AOM 005953 WATER PCB-00050129 Footnotes I-. Farber, E. (1973). Hyperplastic Liver nodules. In:Methods in Cancer Research, Vol. 7, H. Busch, ed. Academic Press, N.Y. pp. 345-375. 2. See reference: Squire, R.A. and Levitt, M.H. (1975). 3. See reference: NCI (1978). 4. Grube, D.D., Peraino, C., and Fry, R.J.M. (1975): The effects of dietary phenobarbital on the induction of skin tumors in hairless mice with 7,12- dimethylbenz(a)anthracene. J. Invest. Dermatol., 64, 2S8-262. " 5. Costa, M., Costa, E.R., Manen, C.A., Sipes. I.G., and Russell, D.H. (1976): Adenosine cyclic 3' ,5'-monophosphate-dependent protein kinase and ornithine decarboxylase involvement in the induction of cytochrome P-450 and hepatic hypertrophy. Mol. Pharmacol., 12, 871-878. 6. Delor is a tradename for PCBs made by Chemko in Czechoslavakia. 7. Brown, D.P. and Jones, M. (1981). Mortality and industrial hygiene study of workers exposed to polychlorinated biphenyls. Arch. Environ. Health 36, 120-129. \ - 37 00^54 *0* WATER PCB-00050130 References 1. Anon. (1981). The Toxicology of PCB's. An overview with emphasis on human health effects and occupational exposures. Hazard Evaluation System. Epidemiological Studies Section. State of California Berkeley. 2. Bahn, A.K., Grover, P., Rosenwaike, I., O'Leary, K., SteUman, J. (1977) PCB? and melanoma. New Engl. J. Med. 296, 108. 3. Bahn, A.K., Rosenwaike, I., Herrmann, N., Grover, P. , Stellman, J. and O'Leary, K. (1976). Melanoma after exposure to PCB's. New Engl. J. Med. 295, 450. 4. Berry, D.L., DiGiovanni, J., Juchau, M.R., Bracken, W.M., Gleason, G.L., and Slaga, T.J. (1978). Lack of tumor-promoting' ability of certain environmental chemicals in a two-step mouse skin tumorigenesis assay. Res. Commun. Chem. Pathol. Pharmacol. 20, 101-108. ' 5. Berry, D.L'. Slaga, T.J., DiGiovanni, J., and Juchau, M.R. (1979). Studies with chlorinated dibenzo-p-dioxins, polybrominated biphenyls and polychlorinated biphenyls in a two-stage system of mouse skin tumorigenesis: potent anti-carcinogenic effects. Ann. N.Y. Acad. Sci. 320, 405-414. 6. Blazak, W.F. and Marcun , J.B . (1975). Attempts to introduce chromosomal breakage in chicken embryos with Aroclor- 1242. Poultry Sci. 54, 310-312. - 7. Brinkman, V. A. Th. and deKok, A. (1980). Production, properties and usage. In: Halogenated biphenyls, terphenyls, naphthalenes , dibenzodioxins and related products. R.D. Kimbrough, ed. Elsevier, North-HoLland, New York. pp. 1-40. 8. Calandra, J.C. (1976). Summary of toxicological studies on commercial PCB's. National Conference on Polychlorinated Biphenyls, November 19-21, 1975. Chicago, Illinois. EPA Report No. 560/6-75-004. March. . NTIS PB-253 248. pp. 35-42. 9. Cole, P. and Merletti, F. (1980). Chemical Agents and Occupational Cancer. J. Environ. Pathol. Toxicol. 3, 399-417. 10. Danz, M. and Urban, H. (1980). Elevated mitotic number' in the adrenal certex as a reflection of early events in' growth induction by . carcinogens and promoters - a possible short-term assay. Further studies in the assessment of toxic actions. Arch. Toxicol. Suppl. 4, 19-21. 11. DHHS. Public Health Service (1980). Health Effects of Toxic Pollution: A report from the Surgeon General. Serial No. 96-15. August. U.S. Government Printing Office. - 38 ADM 005955 WATER PCB-00050131 12. DiGiovanni, J. , Viaje, A., Berry, D.L., Slag^a, T.J., and Juchau M.R. (1977). Tumor-initiating ability of 2 ,3,7,8-tetrachlorodibenzop-dioxin (TCDD) and Aroclor 1254 in the two-stage system cf mouse ' skin carcinogenesis. 3ull.' Environ. Contam. Toxicol., l_3, 552-557. " 13. Dikshith, T.S.S., Reckwoed, W. , Abraham, R., Coulston, F. (1973) Effects of a polychlorinated biphenyl (Aroclor 1254) on rat testis Exp. Mol. Pathol. 22, 376-385. ' 14. Gaffey, W.R. (1981). The epidemiology of PCBs. Presented at the Annual Meeting of the American Chemical Society. 15. Garthoff, L.H., Friedman , L., Farber, T.M., Locke, K.K., Sobotka, T.J., Green, S., Hurley, N.E., Peters, E.L., Story, G.E., ' Moreland, F.M., Graham, C.H., Keys, J.E., Taylor, M.J., Scalera, J.V., Rothlein, J.E., Marks, E.M., Cerra, F.E., Rodi, S.3., Sporn, E.M. (1977). Biochemical and cytogenetic effects in rats caused by short-term ingestion of Aroclor 1254 or Firemaster BP6. J. Toxicol. Environ. Health 3, 769-796. 16. Goldstein, J.A. (1980). Structure-activity relationships for the biochemical effects and the relationship to toxicity. In: Halogenated biphenyls, terphenyls, naphthalenes, diben2odioxins and related pro ducts . R.D. Kimbrough, ed. Elsevier/North-Holland New York, pp. 151-190. 17. Green, S., Carr, J.V., Palmer, K.A. and Oswald, E.J. (1975a). Lack of cytogenetic effects in bone marrow and spermatogonial cells in rats treated with polychlorinated biphenyls (Aroclors 1242 and 1254). BuU. Environ. Contain. Toxicol. 13, 14-22. 18. Green, S., Palmer, K.A. and Oswald, E.J. (1973). Cytogenetic effects of the polychlorinated biphenyls (Aroclor 1242) on rat bone marrow and spermatogonial cells. Toxicol. Appl. Pharmacol. 25, 482. 19. Green, S. , Sauro, F.M. and Friedman, L. (1975b). Lack of dominant lethality in rats treated with polychlorinated, biphenyls (Aroclors 1242 and 1254). Food Cosmet. Toxicol. 13, 507-510. 20. Hansel!, M.M., Ecobichon, D.J., Comeau, A.M. and Cameron, P.H. (1977).' The relationship between retention of pure chlorobiphenyl cogeners and hepatic function in the rat. Exp. Mol. Pathol. 26, 75-84. '? ` 21. Hargraves , W. A. and Allen, J.R. (1979). The in vitro binding of ' 2,2',5,5'-etrachlorobiphenyl metabolites to rat liver microsomal pro teins. Res. Comm. Chem. Pathol. Pharmacol. 25, 33-52. 22. Heddle, J.A. and Bruce, W.R . (1977). Comparison of tests for mutagenicity or carcinogenicity using assays for sperm abnormalities . formation of micronuclei, and mutations m Salmonella. In: Origins cf Human Cancer. Cold Spring Harbor Conferences on Cell Proliferation. H.H. Hiatt, J.D. Watson and J.A. Wins ten, eds. Cold Spring Harbor Laboratory. Vel. 4 Book C, pp. 1549-1557. - 39 - ADM 005956 WATER PCB-00050132 23. Hendricks, J.D., Putnam. T.P., Bills, D.3., and Sinnhuber, R.O. (1977) Inhibitory effect of. a polychlorinated biphenyl (Arocior 1254) . on Aflatoxin B1 carcinogenesis in rainbow trout (Sajoo gaircr.eri). J. Natl. Cancer Inst. 59, 1545-1551. " : '. 24. Hoopingamer, R., Samuel, A., and Krause, D. (1972). Polychlorinated biphenyl interactions with tissue culture cells. Environ. Health Perspect. 1, 155-158. 25. Hsia, M.T.S. , Lin, F.S.D. and Allen, J.R, (1978). Comparative mutagenicity and toxic effects of 2,5,2'5'-tetrachiorobiphenyl and its metabolites in bacterial and mammalian test systems. Res. Comm. Chem. Pathol. Pharmacol. 2JL, 485-496. 26. IARC (1978). Polychlorinated biphenyls and polybroninated biphenyls IARC Monographs on the evaluation of the carcinogenic risk of " chemicals to humans. Volume 18, 43-103. IARC. Lyon, France. 27. Ito, N., Nagasaki, H., and Arai, M. (1973a). Interactions of liver tumorigenesis in mice treated with technical polychlorinated biphenyls (PCBs) and benzene hexachloride (BHC). In: New methods in environmental chemistry and toxicology. F. Coulston, F. Korte and M. Goto, eds. Int. Acad. Print. Co., Ltd. Tokyo. pp. 141-147. 28. Ito, N., Nagasaki, H., Arai, M., Makiura, S., Sugihara, S. and Hirao, K. (1973b). Histopathologic studies on liver tumorigenesis induced in mice by technical polychlorinated biphenyls and its promoting effect on liver tumors induced by benzene hexachloride. J. Natl. Cancer Inst. 51, 1637-1646. - 29. Ito, N., Nagasaki, H., Makiura, S. and Arai, M. (1974). Histopathological studies on liver tumorigenesis in rats treated with poly chlorinated biphenyls. Gann 65, 545-549. 30. Ito, N., Tatematsu, M., Hirsoe, M. Nakanishi, K. and Murasaki, G. (1978). Enchancing effects of chemicals on production of hyperplastic liver nodules induced by N-2-fluorenylacetamide in hepatectomized rats. Gann 69, 143-144. 31. Keplinger, M.L., Fancher, O.E., Calandra J.C. et al. (1972). ' Toxicological studies with polychlorinated biphenyls. Read at PCB conference, Quail Roost Conference Center, Rougemont, North Carolina, 20-21 December 1971. Cited in Polychlorinated Biphenyls Environmental Impact. A Review by the Panel on Hazardous Trace Substances. March, 1972. Environ. Res. 5, 249-o82. 32. Kerkvliet, N.I. and Kimeldorf, D.J. (1977a). Inhibition of tumor growth in rats by feeding a polychlorinated biphenyl, Arocior 1254. Bull. Environ. Contam. Toxicol. IS 243-246. * 33. Kerkvliet, N.I. and Kimeldorf, D.J. (1977b). Antitumor activity of a polychlorinated biphenyl mixture, Arocior 1254, in rats inoculated with Walker 256 carcinosarcoma ceils. J. Mail. Cancer Inst. 59, 951-955. - 40 - A tJM OG595 7 WATER PCB-00050133 34. KerkvLiet, I. and Holier, L.D. (1980). .Effect of cadmium. arsenic and PCS's on tumor-directed cell-mediated cytotoxic immune reactions m * . mice injected with MSB sarcoma cells. In: Inadvertent Modification of the Immune Response. Proceedings of the Fourth FDA Science'* Symposium. U.S. Naval Academy, August 28-30, 1973. HKS Dubi FDA-80-1074, pp. 275-279. *' 35. Kimbrough, R.D. (1979). The carcinogenic and other chronic effects of persistent halogenated organic compounds. Ann. N.Y. Acad Sci 320, 415-418. '' 36. Kimbrough, R.D. (1980). Occupational exposure. In: Halogenated biphenyls, terphenyls, naphthalenes, dibenzodicxins and related products. R.D. Kimbrough, ed. Elsevjer/North-Holland, New York pp. 373-397. 37. Kimbrough, R. , Buckley, J., Fishbein, L., Flamm. G., Kasza, L. , Marcus, W., Shibko, S. and Teske, R. (1978). Animal Toxicology. Environ. Health Perspect. 24, 173-184. 38. Kimbrough,- R.D. and Linder, R.E. (1974). Induction of adenofibrosis and hepatomas of the liver of BALB/cJ mice by polychlorinated bi phenyls (Arocior 1254). J. Nad. Cancer Inst. 53, 547-552. 39. Kimbrough, R.D., Squire, R.A., Linder, R.E.. Strandberg, J.D., Mon tali, R.J., and Burse, V.W. (1975). Induction of liver tumprs in Sherman Strain female rats by polychlorinated biphenyl Arocior 1260. J. Nad. Cancer Inst. 55, 1453-1459. 40. Kimura, N. and Baba, T. (1973). Neoplastic changes in the rat liver induced by polychlorinated biphenyl. Gann 64, 105-108. 41. Kimura, N.T., Kanematsu, T., and Baba, T. (1976). Polychlorinated biphenyl(s) as a promoter in experimental hepatocarcinogenesis in rats. Z. Krebsfersch. Klin. Onkol. 87, 257-266. 42. Roller, L.D. (1977). Enhanced polychlorinated biphenyl lesions in Moloney leukemia virus-infected mice. Clin. Toxicol. U,, 107-116. 43. Levinskas, G.J. (1981). Toxicity of Arocior products 1242, 1254 and 1260 to the liver of albino rats. . 44. Lichti, U. , Yuspa, S.H. and Hennings , H. (1973). Ornithine and S-adenosylmethionine decarboxylases in mouse epidermal cell cultures . treated with tumor promoters. In: Carcinogenesis, Vol. 2 Mechanisms of tumor promotion and cocarcinogenesis. T.J. Slaga. A. Sivak, and R.K. Boutwell, eds. Raven Press, N.Y. pp. 221-232. . 45. Makuira, S., Aoe, H., Sugihara, S., Hirao, K., Arai, M. and Ito, N. (1974). Inhibitory effect of polychlorinated biphenyls on Liver tumorigenesis in rats treated with 3'-methyl-4-dimethyianinoazobenzer.e, N-2-fluorenvlacetamide, and diethylnitrosamine. J. Natl. Cancer Inst. , 53, 1253-1257. - 41 - AuM 005958 WATER PCB-00050134 46. Matthews, H.B. and Kato, S. (1979), The metabolism and disposition of halogenated aromatics. Ann. N.Y. Acad. Sci. 320, 131-137, * 47. McMahon, R.E-., Cline, J.C. and Thompson, C.2. (1979). Assay of 855 test chemicals in ten tester strains using a new modification of ' the Ames test for bacterial mutagens. Cancer Res. 39, 682-69*3. 48. Morales , N.M. and Matthews, H.B . (1978). In vivo binding of PC3s to hepatic micromolecules in mice. Toxicol. Appl. Pharmacol. 45, 334. 49. Morales, N.M. and Matthews, H.B. (1979). In vivo binding of 2,3,6,2,\3',6'-hexachlorobiphenyl and 2,4,5,2',4' ,5'-hexachloro- biphenyl to mouse liver macromolecules. Chem-Biol. Interact 27 99-110. ~' 50. Nagasaki, H., Tomii, S., Mega, T., Maru garni, M. and Ito, N. (1972). Hepatocarcinogenicity of polychlorinated biphenyls in mice. Gann 63, 805. 51. Nagasaki, H., Tomii, S., Mega, T. Sugihara, S., Miyata, Y. and Ito, N. (1974). Analysis of various factors on liver carcinogenesis in mice induced by benzene hexachloride (BHC) and technical polychlorinated biphenyls (PCB's). J. Nara. Med. Assoc. 25: 635-648. 52. Nagasaki, H. , Tomii, S., and Tsumashika, Y. (1975). On a main cause for the formation of hepatic tumors of mice by BHC and PCBs. Nippon Eiseigaku Zasshi 30, 134. (Translation by Ralph McElroy Co., Custom Division, Texas). \ . 53. NCI (1978). Bioassav of Aroclor 1254 for possible carcinogenicity. NCI-CG-TR-38, DHEW. Publication No. (NIH) 78-838, Washington, D.C., U.S. Department of Health, Education and Welfare. 54. Nesnow, S., Leavitt, S., Garland, H. , Vaughan, T.O., Hyatt, 3., Montgomery, L., and Cudak, C. (1981). Identification of cocarcinogens and their potential mechanisms of action using C3H10T4CL8 mouse embryo fibroblasts. Cancer Res . 41, 3071-3076. 55. Nilsson, B. and Ram el, C. (1974). Genetic tests on drosophila melanogaster with polychlorinated biphenyls (PC3). Hereditas 77, 319-322. 56. NIOSH (1977). Criteria for a recommended standard. Occupational exposure to polychlorinated biphenyls (PCBs). DHEW (NIOSH) Publication No. 77-255, Washington, D.C., U.S. government . Printing Office. 57. Nishizumi, M. (1976). Enhancement of diethylnitrosamine hepatocarcinogenesis in rats by exposure to polychlorinated biphenyls or phenobarbital. Cancer Lett. , 2, 11-16. 58. Nishizumi, M. (1979a). Effect of DDT and phenobarbital on the accelerating action of PCBs in liver cancer generation due to diethylnitrosamine. Nippon Eiseigaku Zasshi (Jpn. J. Hyg.) 34, 197. (Translation for EPA by Literature Research Company). - 42 - AL>M OO5959 WATER PCB-00050135 59. Nishizumi, M. (1979b). Effect of phenobarbital, dichlorodiphenyltrichlo roe thane, and polychlorinated biphenyls on diethylnitrosamine- . induced hepatocarcinogenesis. Gann 70, S35-337. 60. Nishizumi, M. (1980). Reduction of diethylnitrosamine-induced hepatoma in rats exposed to polychlorinated biphenyls through their dams. Gann jT, 910-912. 61. Norback, D.H., Reddy, G. and Cihla, H.P. (1980). Oncogenic, , cytotoxic and enzymatic effects of PCBs on C3H10T's cells. ACS, Ann. Mtg., San Francisco, Abstract no. 153 in section on Environment. 62. Norback, D.H. and Swede, G.J. (1979). Transformation in vitro induced by Aroclor 1254: Ultrastructure of transformed ceils. Toxicol. Appl. Pharmacol. 48, A182. 63. Norback, D.H. , Weltman. R.H., and Cihla, H.P. (1981). Malignant transformation of C3HlQT<s cells by polychlorinated biphenyls. Fed Proc. 40, 757. . 64. Odashima, S. (1976). The cooperative development in Japan of methods for screening chemicals for carcinogenicity. In: Screening tests in Chemical Carcinogenesis, International Agency for Research on Cancer. R. Motesano, M. Bartsch and L. Tomacis, eds. IARC Scientific Publ. No. 12, Lyons, France, pp. 61-75. 65. Peakall, D.B., Lincer, J.L. and Bloom, S.E. (1972). Embryonic mortality and chromosomal alterations caused by Aroclor 1254 in ring doves. Environ. Health Perspect. _1, 103-104. - 66. Peraino, C., Fry, R.J.M., and Grube, D.D. (1978). Drug-induced enhancement of hepatic tumori gene sis. In: Carcinogenesis , Vol. 2. Mechanisms of Tumor promotion and Cocarcinogenesis. T.J. Slaga, A. Sivak, and R.K. Boutwell, eds. Raven Press, N.Y. pp. 421-432. 67. Pienta, R.J. (1980). Transformation of Syrian hamster embryo cells by diverse chemicals and correlation with their reported carcinogenic and mutagenic activities. In: Chemical Mutagens: Principles and Methods for Their Detection Vol. 6. F.J. deSerres and A. Hollaender, eds. Plenum Press, New York, pp. 175-202. 68. Pitot, H.C., Barsness, L. and Kitagawa. T. (1978). Stages in the process of hepatocarcinogenesis in rat liver. In: Carcinogenesis, Vol. 2. Mechanisms of Tumor Promotion and Cocarcinogenesis. T.J. Slaga, A. Sivak, and R.K. Boutwell, eds. Raven Press'/1 N.Y. pp. 433-442. 69. Preston, 3.D., Van Miller, J.P., Moore, R.W. and Allen, J.R. (1981). Promoting effects of polychlorinated biphenyls (Aroclor 1254) and polychlorinated dibenzofuran-free Aroclor 1254 on diethylnitrosamineinduced tumorigenesis in the rat. J. Natl. Cancer Inst. 66, 509-515. 70. Probst, G.S., McMahon, ''R.E. , Hill. L.E.. Thompson, C.Z., Epp. J.K. and Neal, S.B . (19S1). Chemically-induced unscheduled DNA synthesis in primary rat hepatocy te cultu res . Environ. Mu tag. 3 , 11-32. * ~ - 43 - ADM 005960 WATER PCB-00050136 71. Ribelin, W.E. (1981). Personal communication. 72. Rinkus, S.J. and Legator, M.S. (1980). The need for both in vitro and in vivo systems in mutagenicity screening. In: Chemical Mutagens: Principles and Methods for their detection. Vol. 6. F.J. deSerres an d' A. Hollaender, eds. Plenum Press, New York, pp. 365-473. 73. Roberts, J.R., Rodgers, D.W., Bailey, J.R. and Rorke, M.A. (1978). Polychlorinated biphenyls: Biological criteria for assessment of their effects on environmental quality. National Research council of Canada, Publication no. NRCC 16077. 74. Safe, S. (1980). Metabolism, uptake, storage and bioaccumulation. In: Halogenated biphenyls, terphenyls, naphthalenes, dibenzodioxins and related products. R.D. Kimbrough, ed., Elsevier/North-Holland New York. pp. 81-107. 75. Squire, R.A. and Levitt, M.H. (1975). Report of a workshop on classification of specific hepatocellular lesions in rats. Cancer Res. 35, 3214-3223. 76. Stadnicki, S., Lin, F.S.D. and Allen, J.R. (1979). DNA single strand breaks caused by 2,2',5', 5-tetrachlorobiphenyl and its ' metabolites. Res. Comm. Chem. Pathol. Pharmacol. 24, 313-327. 77. Stott, W.T. and Sinnhuber, R.O. (1978). Trout hepatic enzyme activation of aflatoxin Bt in a mutagen assay system and the inhibitory effects of PCB. Bull. Environ. Con tarn. Tox-col. 19, 35-41. 78. Sugimura, T., Kawachi, T., Matsushima, T., Nagao, M., Sato, S. and Yahagi, T. (1977). A critical review of submammalian systems for mutagen detection. In: Progress in genetic toxicology. D. Scott, A.B. Bridges, and F.H.. Sobels, eds. Elsevier/North-Holland Biomedical Press, pp. 125-140. 79. Sugimura, T., Sato, S., Nagao, M. , Yahagi, T. , Matsushima, T., Seino, Y. , Takeuchi, M., and Kawachi. T. (1976). Overlapping of carcinogens and mutagens. In: -Fundamentals in Cancer Prevention. P.N. Magee, S. Takayama, T. Sugimura and T. Matsushima, eds. Univ. of Tokyo Press, Tokyo/Univ. Park Press, Baltimore, pp. 191-215. 80. Tazima, Y. (1980). Chemical mutagenesis in the silkworm. In: Chemical Mutagens: Principles and Methods for their Detection. Vol. 6, F.J. deSerres and A. Hollaender, eds. Plenum Press, N.Y. pp. 203-238. 81. Turusov, V.S. and Takayama, S. (1979). Tumours of the liver. In: Pathology of Tumours in Laboratory Annuals . International Agency for Research on Cancer. World Health Organization. Lyon France. V.S. Turusov, Editor-in-Chief. Vol. II-Tumours of the mouse, pp. 193-233. 82. Uchiyama, M. and Chiba, T. (1974). Co-carcinogenic effect of DDT and PCB feedings on me thylchoianth re ne-induced chemical carcinogenesis . Bull. Environ . Contam. Toxicol. 1_2, 687-693 . - 44 - 005 v WATER PCB-00050137 83. Wasserman, D. , Miller, H.J., and Wasserman, M. (1978). Polychlorinated biphenyl-induced rat liver adenomas. Toxicol. Eur. Res. li, 159-172. 84. Weisburger, J.H. and Williams, G.M. (1980). Chemical Carcinogens. In: Casarett and Doull's Toxicology, J. Doull, C.D. Klaasen, and M.O. Amdur, eds., Macmillan Publishing' Co., Inc., New York. pp. 84-138. 85. Weltman, R.H. and Norback, D.H. (1979). Hepatic ultrastructural changes of rats exposed to polychlorinated biphenyl isomers: Hexachlorobiphenyl-induced neoplastic nodules. Toxicol. Appl. Pharmacol. 48, A181. 86. Wong, \., Basrur, P.K. and Safe, S. (1979). The metabolically mediated DNA damage and subsequent repair by 4-chlorobiphenyl in Chinese hamster ovary cells. Res. Commun. Chem. Pathol. Pharmacol. 24, 543-550. 87. Wyndham, C. Devenish, J., Safe, S. (1976). The in vitro metabolism, macromolecular binding and bacterial mutagenicity of 4-chlorobiphenyl, a model PCB substrate. Res. Comm. Chem. Pathol. Pharmacol. 15, 563-570. 88. Yoshimura, H., Yoshihara, S., Ozawa, N. and Miki, M. (1979). Possible correlation between induction modes of hepatic enzymes by PCBs and their toxicity in rats. Ann. N.Y. Acad. Sci. 320, 179-192. - 45 Aqh 596c, WATER PCB-00050138 TABLE 1 PCBs : Tabulation of Liver Modules Reported in Mice1i Approx. wt. I Cl2 Product Strain Dosing Duration PCB in diet (ppm) Mo. Animals Nodular Hyperplasia Hepatoma Hepato cellular Carcinoma 52-54 Aroclor 1254 BALB/cJ 6 nos3 11 nos Kanecblor 500 dd 32 weeks Kanecblor 500 dd 32 weeks Kanecblor 500 dd7 32 weeks 0 300 0 300 0 100 250 500 0 100 250 500 0 100 250 500 34 24 24 22 6 12 12 12 20 18 20 17 12 19 20 17 * 0 0 0 7 - 9 - 4 0 1 0 94 - (-)s - - - - - - - 0 0 0 5s 0 0 0 7 0 0 0 o6 Reference Kimbrough + Linder (1974) Nagasaki, et al. (1972) 1 to , et a 1 . ( 1973a ,1 Nagasaki , et al. (1974, 1975) Nagasaki, et al. (1974, 1975) 48 Kanecblor 400 dil 32 weeks 0 6 100 12 250 12 500 12 0. 0 0 0 Kanecblor 400 dd 32 weeks 0 100 250 500 0 17 19 20 ' * " - Kanecblor 400 dd7 32 weeks - 0 100 250 500 12 20 20 . 17 -- - 0 Nagasaki, et al. - 0 (1972) - 0 I to, el a 1. ( 1973a, 1 -0 - 0 Nagasak i, et al. - 0 ( 1974, 1975) -0 -0 - 0 Nagasaki, el a 1. - 0 (1974, 1975) -0 --0 WATER PCB-00050139 005963 TABLE 1-continued PCBs: Tabulation of Liver Nodules Reported in Mice* Approx. Dosing wt. X Cl12 3 P* 5 *roductStrain Duration PCB in diet (ppm) No. Animals Nodular Hypcrplasia_____ Hepatoma Ilepatocellular Carcinoma Kancchlor 300 dd 32 weeks 0 ' 100 250 500 6 12 12 12 0 0 0. 0 - - 0 0 0 0 Kancchlor 300 dd 32 weeks 0 100 250 500 20 19 19 20 - '- -0 -0 -0 -0 Kanechlor 300 dd7 32 weeks 0 100 250 500 12 19 20 20 - - - - _0 -0 -0 -0 Reference Nagasaki , et al. (1972) Ito, et al. (1973a, Nagasaki, et al. (1974, 1975) Nagasaki , et al. (1974, 1975) 1 Hales, except as noted. ' 2 From Brinkman and dcKok (1980). 3 Held an additional 5 months before sacrifice. * No. of anim ils. One hail 2 hepatomas for a tumor total of 10. 5 Apparent Iy same data reported as hepatoma and hepatocellular carcinoma. See references' and text. Apparently same data reported as nodular hyperplasi. and hepacellular carcinoma or tumor. See references text. 7 Females. 0059 6^ jo> s WATER PCB-00050140 TABLE 2 PCBs: Tabul aiIon of Liver Modules Reported in Rata Approx Product 60 Aroclor 1260 PCI Dosing in diet Ho. Adenoastous Nodular Strain Sea Duration (PP) Animals Nodules Hyperplasia Slieras F 21 os.* 0 100 m 184 _ * '" Aroclor 1260 Charlea H. 24 aos. River F 0 1 10 IM 23 26 25 25 - | 0 ? 6 2,2-,4,4*,S,5'- - lleaathlorobiplienyl1 - 24 aos. 100 - - Heoplaatic Modules AUenoaa UepaLoaa Hepato- cellular Carcinona Cholanglollepatona Reference 0 i _ Klabrough. 144 - * 26 * et al. (19/5 - - 0 _ 0 Leviuskaa - -0 - 0 (1981) - -1 - 0 * -I- 4 _ _ Urllaaii 6 Noihack (19/ Aroclor 1254 Charles H, 24 aoa. River F 0 I 10 100 23 30 26 26 - 1 0 3 14 0 -' - 0 - - -0_ * -4 - 0 Uviiulus . 0 (1981) 0 2 Ai odor 1254 Ki seller H 344 |0S weeks 0 25 so 100 24 24 24 24 --- '- -- - 0_ 0 NCI (1978) - 0- 0 _ - 0- 1 _ - 1-2 - Aroclor 1254 ludier F 344 SOS weeks 0, 25 50 100 23 24 22 24 -. - _0 _ - 0- - NCI (1978) - 1-_ - 2-- - Aroclor 1254 Local F 28 aos.* 0 4 - - - 0_ _ 200 6 * * - - - et al. (1978) KaitrcKilor Vi alar H 28-52 vlis 0 18 - 0 -- _ t 500 O 2 100 soo 1000 25 16 13 *' - 3 5 5 - -- - --_ _ (1974) " _ "- -- 005 -J6 5 WATER PCB-00050141 < Appro*, wt. t Cl* Product 4( Kanechlor 400 Kaneclilor TABU 2-coatInued PCIi: Tabulation of Liver Module* Reported In Rata PC Doaiag la diet No. Adenoaatoua Modular Strain Sex OuratiM Cm) Aniaala Modulea Hyperplaaia Donryu M 23-62 aka 3R.5616 r 5 10 5 10 0* 0 '* 06- Ulatar H 21-52 wka 0 100 5M IM0 18 16 8 10 - - 0 2 0 3 Meoplaatlc Modulea Adenoaa Mepatoaa Hepatocellular Carcfnoaa CbolangloHepatoaa Reference _ Kiaura * Ba> - --- (19/3) - --- - * *" * - -- - - llo, et a I. - - - - - (1974) - --- - ** - " ** 40-42 Aroclor Cbarlea M. 24 aoa. 0 23 1 0_ 0 Levinxkea 1242 River F | 32 - 1 - - 0 - 0 11981) 10 29 - 1 - -0- 0 100 19 " 11 Kaneclilor Uialar N 28-52 wka 0 ia - 0 - --- Ilo, el a 1 . 100 22 - 1 * - - - - (19/4) 500 19 - 0 - - - - - 1000 15 0 '- - 2,2',5,5"--let re- - - 24 aoa. 100 - - - - - - - - Weltoajn & chI orobiphenyl* Mot Iiai k ( H/'i * Fhm Br InkaAtt Anti detufe ( I9B0) . * lltrltl An Atltlii ionA I 2 aunt be before ncrl flee. 1 Abstract. Del AI I* I iaitetl. ' * Presrnre lepurleil. * Ethanol solution in drinking water. ADM GQ59o6 I WATER PCB-00050142 Product Mode Time Other Treatment1 TABLE 3 suttwty or co-cabcinocehesis studies with pcb Hode Speciea Ohdervatioaa Kanechlur 400 Diet With Kaneclilor 500 Diet Kanci It lor 500 ^ Diet With With Kancrhlor 400 SCaneclilor 500 Kanechlor 500 Aroclor 1254 Kaneclilor 500* lliet With Diet With Cavsge After Diet Alter Cavage Before A roc1ur 1254 Kanerhlar 500 Diet Diet With With Kanci lil or 500 Diet Wi tit Kanrt li 1 or 500 Diet Before Kaiiciiilor 500 Diet After 20-HeCh -BIIC p-BIIC Uterine implant Diet Diet a-BMC Diet DEN Water DEN Water OEM Water DEN* Water AFB, Diet 2-EAA Diet 3*-He-DAB Diet 3-He-DAB Diet 3"-Hc-IMB Diet dd Muie Hepatocellular carcinoma With 20-HeCh. None with combination. dd Mouse Combination increased hepatocellular carcinoma over o-BIIC alone. dd souse Cuabiiulloa produced hepstocellulsr carcinoma. Each alone negative. dd aouae Combination increased hepatocellular carcinoma over a-BIIC alone. SpragueDawley rat Nepatocellular carcinoma with DEN. None with combination. Uiatar rat Combination increased hepstocelluisr carcinoma over DEN aloue. SpragueDawley rat Combination increased hepatocellular carcinoma over DEN alone. Wiatar rat Combination decreased hepatocellular carcinoma in offspring. Ha i nbow t rout Hepatocellular carcinoma with AEB(. None with combination. SpragueDawley rat Hepatocellular carcinoma with 2-FAA. None with combination. Spraguc- Hepatocellular carcinoma with 3'-He-DAB. Dauley rot .. None with combination. SpragueDawley rat Hepatocellular carcinoma with 3*-Me-DAB. None with combination. SpragutDawley rat Coadiinalion Increased hepatocellular carcinoma over 1`-He-DAB alone. Ref. Uchiyaaa, and Chiba, 1974 Ito, el al. 1973a,b Naaaaki.~el al 1974. 1975 llo, et al^ 1973a,b Nagasaki, et _a1. 1974, 1975 Nagaaaki, e^ aK 1974, 1975 Hakiura, et a 1 1974 Niahizumi, 1976, 1979a, b Preston, el al. I9BI Niahizumi, 1980 Hendricks, et a 1. 1977 Hakiura, el al. 1974 Hakiura, el al. 1974 Kimura, el il: 1976 1Cimora , et a 1 . 1976 Kimura, el_ a 1. 1976 AOM 0 0 5 9 6 7 WATER PCB-00050143 PsodkICt Node Other Treataest1 TABLE 3-continued BIBtfAHT Of CQ-CAKCmoCEBESIS STUDIES MTM PCBn Moile Species . Observations Kancchlor 500 Diet With Kinechlor 500 Diet With Konetlilor 500 Gevage Kanechlor 500 . Gevi|e Kanechlor 500 Cavage Unspec i (ted Kanechlor Diet After fc With After L Utb After A Vi tk After A Vi th Aroclor 1254 Aroclor 1254 Skin Skin After Before 3*-He-DAB DEN Diet Sprague* Coabiaation decreased hepatocellular Dawley rat ` carclnoara over 3*-He-DAB * DEN. 2-FAA DEM Diet SpragueDawley rat Coahinatlon decreased hepatocellular carcinoaa over 2-FAA DEN. KN WT Water Csvsge Wiatar rat Coablnalton Increased hepatoce!lular carcinoaa over DEM DDT. DEM SI'B Water Water Uitlar cal Combinat ioa increased liepatoce i lular carcinoaa aver DM SPft. DEM SPB, DOT Water Meter, Cavage Viitar rat Combination Increased hepatocellular arcinna over DC*, SPfi, 00T. 2-FAA Partial hepalecloay Diet ' Flecker rat Harked increase in hyperplastic liver nodules. ?, 12-DHBA Skin CDl BMIC ` Not a proaoter of skin tuaora. TPA Skin C01 aouae Little or no akin tuaor ini 1iation. Aroclor 1254 Aroclor 1254 Aroclor 1254 Aroclot 1254 Skin Before Diet Before Biel Before Diet . Cefore 7,12-DHBA TPA Skin CD! aouae Vainer 254 lutrstuoor cell* auscutar HSB cells lotsnuicuUr Moloney leukeaU virus Inlr*" peritoneal SpragueDawiey rat C571L/6 BOttCS ' Salb/c aouae No difference or decrease in skin tuaor induction over 7,12-DHBA k TPA, depending on pretreataienl interval. Tuaor growth inhibited. Survival tlae increased. Enhanced growth of HSB tuaor. Did not affect oncogenesis of Holoney leukcaia viroe. Net. Haklura, et al. 1974 Haklura, et al. 1974 Niahlzuat, 1979a,b Nlahizual, 1979a,b Nishizuai, 1979a,b 1 to, et al. I97B Berry, et al. 1918. 1979 Berry, el al. 1979 DiCiovanni, et a 1. 1977 Berry, el al. 1979 DiCiovanni, et al. 1977 lerkvl id And JCiwieidurf, I9fl*. b Kerkvliel and toller, 1978 Koller, 19/7 ADM 0 0 5 9 oo f1 1 WATER PCB-00050144 Product Holt Tlaa TAHE 3-contlnued satwii or co-cmcinocehesis stupies mitw rea Otbar Treataeat 1 Mode Speclea Obaervatlooa Aroclor 1242 let Aroclor 1221 let S elart Before Moloney leukeala virua Moloney leukeala virua latraperltoneal Bslb/c mouse . Id not affect oncogeaeala of Moloney leukeala virua. lot ro per lloaeal sib/c mouse Id not affect oncogeneaia of Moloney leukeala virus. 'Cheat cals used: 20-fleCli = 20-aelbylcholanlhrene a-BHC = u-l>euzeae heaocbloride p-ilic = jl-benzene hcaachlorfdc DEM = ilielliylnitrosaaiiie AKBj = af laloain B, 2-fAA = N-2-fluurenylacetaaide I'-ric-DAH = J ` -Be t liy 1-4-dlaetby Iaal aoazobenzeae ,s DOT = illi lilorotli|iheiiyl I r ithloroe^hane Sm = toilnui |ilieiiok<ikilal - 7 ,12-DBA = 7 , 2-tlielliy Ibeiu: |a|anlfcr*ceoe , TfA = 12-0-let rtlccnoyl pliorbo) - 13-Acetate 2Pre|naat 4mi dosed with Ksnechlor 400. Offspring dosed vllh DEN. sEslsbIished (issue culture cell line derived from Moloney ssrcorns virus~Induced tumors in 6 mice. ef. toiler. 1977 toller. 1977 AOM 0 0 5 9 6 9 }i WATER PCB-00050145 l Teller Strain Product AKOCI.OK 1268 AHOCI.ON 1254 AKOCIOH 1254 AHIHII.OH 1254 TABU 4 PCBa: Tabulation ol Microbial Mutagenicity Teif S. liphlaiitliia C3#?i M52 CM TAM TAI00 TAI000 TAIS3S TAI536 TAIS37- TAIS3B "eg. fc. # a Meg. 0 "eg. fcg. o "eg. N . a "eg. "*- a Meg. Meg. "eg"eg. POB1 . ICjiici Itlor 5Q0* Kannblnr 500 m Neg. "eg. Meg. Meg. Meg. "eg' 2,2*,5,5-tet ra dii ocobigiltcity 1 2,2*5,5*-trlracMorobi|luity 1 K*itei h loi 30OJ Kafliiechlor 300 AROCI.UN 1221 4 - < li 1 rob i |>heuy 1 v- # "eg* Met# o Neg* * Meg. # "eg* Net- e "eg. Meg* "eg. "eg "eg- Poa* Poe* 4-ihlurobiphenyl "eg. Neg. Meg. "eg. eg. Meg. I. Al 1 trsls done with aad without ai.croaoaal ctivatioa except a noted. 2. Otlier aliaina tealed. Sec text and reference. - Hot trrted. Meg. = Mega I ivc. Mc - On 1 y tested without, act Ivil ion. Meg4 = Only trsted with activation, P*s* = I'osiIive only with activat ion. Poa^ = Uncertain. See teat. "eg. "eg- r* i s E. call WWP22 WP2 uvrA' *ef. Meg. o Meg. Wyndbaa, el al. 1916 Probat, cl al, 1981 Neddie and Bruce 1911 Wyndbaa. el l . 1916 Odashiaa, 1916 Sugiaura , el a I 1916 1 Uyodliea, cl a I 1916 "eg "eg- Hata, el al . 1918 Odaaliiau, 1916 Sugiaura, el al. 1916 Wyndbaa, el al. 1916 Wyndbaa, el a I . 1916 HcMaliUu, el al. 1919 WATER PCB-00050146 AOM 0 0 5 9 7 0