Document jgJwzamxvdGJRad0pxJRd6ZOZ

"Literature Update on Potential Health Effects From Exposures to Polychlorinated Biphenyls and Related Chlorinated Hetarocycles" Prepared by Victor A. Drill, M.D., Ph.D. Seymour L. Fness, Ph.D Harry W. Hays, Ph.D. Tad A. Loomis, M.D., Ph.D. December 20, 1983 HONS 212*^6 TABLE OF CONTENTS Executive Summery........................................................ I. Introduction................................................................................... II. Yusho Episode................................................................................ III. Body Burdens, Metabolism andKinetics........................... IV. General Toxicity.................. V. Dermal Effects............................................................................... VI Effects of PCBs on Liver...................................................... VII. Gastric Lesions............................................................................. VIII. Carcinogenesisi Experimental..................................... IX. Reproductive Effects................................................................. X. Mutagenesis...................................................................................... XI. Other Health Effects................................................................. XII. Epidemiology.............. ........................................... XIII. Recent Findings on Health Effects of Polychlorinated Dibensofurans......................................... XIV. References........................................................................................ i 8 11 n 26 31 33 35 38 44 45 47 53 60 72 MOWS 212*67 Executive Summary This Addition to a previous study of the available bio medical literature on potential health effects of the poly chlorinated biphenyls (PCBs) and related impurities is essen tially an update from literature and other data retrieved during late 1981 and up throu9h November 15, 1982. As in the first study, the review and conclusions were formulated by a Category I team from DFHL&S, Inc. and subjected to a second level review and critique by an independent Category II team. The comments and suggestions from the Category II review have been incorpo rated into the final revised text of this annex, for which DFHL&S, Inc. takes full responsibility. All new conclusions for each specific health effect in test animal systems and human populations are discussed in the context of both the new studies reviewed and conclusions from the previous report material based on the earlier literature. A summary of the major new findings and conclusions with respect to specific health effects resulting from PCB exposure is given below. Also included is a summary of observations on health effects attributable to exposure to polychlorinated dibenzofurans (PCDFs) in test animal models and humans, for this same literature retrieval period. 1. Recent findings on the health status of individuals previously exposed to the mixtures of PCBs, PCDFs and PCQs HONS 212868 2 (polychlorinated quaterphenylsI In the contaminated rice bran cooking oil* consumed during the Yuaho episodes in Japan (1960's) and Taiwan (1970's) have served to emphasize and re* inforce two points made in the 1982 report, first, it appears that elevated PCD? levels in the cooking oils My have been a principal cause of the enhanced toxicity of those oils over that of the parent commercial PCB mixtures originally charged Into the heat exchangers. Second, because of the mixed PCBi PCDFi PCQ content of those oils, the Yusho episodes are not appropriate models for prediction of human health effects from exposure to commercial PCB mixtures. 2. Hew findings in test animal models and exposed human populations have served to emphasize distinct species specificity In biological processes of tissue metabolism, distribution and excretion. The residual concentrations of PCB congeners in fatty and other tissues are seen to be par ticularly dependent on ease of metabolism and subsequent excretion of polar metabolites, free or conjugated, from specific congeners in commercial PCB mixtures. 2. In the domain of acute toxicity of PCBs, recent attention has been focused ons (a) the substantial toxicity of 2,3,6-hexa CB (chlorinated biphenyl) in inhibition of growth of a mouse embryo cell line; and (b) the marked hepatotoxicity of the 3,4,5-hexa CB congener in young mice. The latter was a rare congener In commercial PCB mixtures. MONS 212BB9 3 These findings do not alter significantly the generalizations on structure vs. toxicity In the 1982 report. 4. Dermal effects In humans attributable to exposure to PCBs and related Impurities are of continued interest, even though most of the rare cases of chloracne found In PCB-exposed industrial workers are not severe and not long-lasting after exposure has stopped. Such signs were particularly severe and long-lasting in certain of the Yusho poisoning patients. The most recent evidence reviewed by Kashlmoto et al. (1981) for chloracne in Yusho victims points to the enhanced levels of contaminants such as the PCDPs in the rice oils as primary factors in severe chloracne. 5. Recent results on the potential impact of PCB ex posure on liver structure or function have derived from an experimental study in the rat and an epidemiological study of capacitor workers. It appears that Aroclor 1254 does not in duce benign liver lesions (hyperplastic foci, nodular hyper plasia) in the male Sprague-Dawley rat, and that extended ex posure of people in the electrical industry setting to various Aroclors has not resulted in findings of significant liver dysfunction or disease. 6. In two studies, gastric lesions were observed in rats and monkeys receiving Aroclor 1254 orally; such lesions have not been seen in occupationally exposed human populations. HONS 212870 4 In th nt, at ait** near tha glandular stomach that were high In alkallna phosphatase, an Increased frequency of occurrence of cystic intestinal metaplasia was observed. In the monkey, the principal finding in the stomach was hypertro phic and hyperplastic mucous gastrophy. These are noncancerous lesions, 7. Studies in the rat using Aroclor 1254 and focused on potential carcinogenesis in chronically exposed rats at target organs such as liver, bladder, stomach and intestinal tract have proven essentially negative, in our opinion. The recent reevaluation of rat stomach tissues from an NCI bio assay (1978) by Morgan et al. (1981) did not result in a sta tistically significant, dose-related increase in the numbers of stomach adenocarcinomas observed. However, the finding by Preston et al. (1981) of a possible tumor-promoting effect of Aroclor 1254 in the Sprague-Dawley rat is judged as biolog ically significant and requiring further investigation. 8. A new study has evaluated the effect of Aroclor 1254 on reproduction, the compound being administered in concen trations of 25-900 ppm in the diet of female rats during days 6-15 of gestation. Noted wares (a) decrease in placental protein and glycogen; and (b) decreases in fetal weight and survival rate. Since decreases in food intake were also noted, resulting in reduced nutritional status and a decrease in weight gain, attribution of the placental and fetal effects HONS 212*71 -5- to either direct PCB action or to altered nutritional state cannot clearly be made. 9. Recent test results on mutagenic potential of PCBs in Ames tests with Salmonella systems .in vitro, with or without metabolic activation, have been negative. 10. The ability of Aroclors to induce a variety of micro* somal ensyme systems in test animal tissues has been widely documented. A new and unexpected finding in this area of bio logical activity has been negative) Aroclor 1254 leads to a decrease in microsomal ensyme content of P-450j in rabbit lung, rather than an increase, with no alteration in P-450jx content. On the positive side, Aroclor 1254 causes the induction of glucuronyl transferases and inhibits the release of polymorphonuclear glucuronidase from leucocytes. These biological actions have not been related to human disease states or dysfunctions. 11. Recent experimental studies have shown that immuno logical effects in test animal systems exposed to PCBs and in Yusho victims exposed to mixtures of PCBs, PCQs and PCDfs can indeed occur. Humans exposed to Yusho oils and monkeys exposed to synthetic mixtures of PCBs, PCQs and PCDFs show signs of immuno-suppression or alterations in immunoglobulins and T-cells. immunogenic alteration or immunotoxicity have also been observed in rabbits and mice dosed with PCBs. How- MONS 212872 ver, the evidence to dat* dot* not point to a direct, do*rtlattd effect of PCBs alona In exposed human population*, txprtsalblt a* a change In immunological competence. 12. The new epidemiological study of Lawton at al. (1981) on electrical industry workers, taken in conjunction with other recent studies on workers occupationally exposed to PCBs, leads to additional insights in the areas of skin problems, lipid metabolism, pulmonary function, blood gluco** levels, blood uric scid levels, and general complaints and symptoms. The Lawton study found no signs of chloracne, past or present, and no dermatological findings specifically re lated to PCB exposure. Associations were found between blood levels of PCBs and serum triglyceride levels in exposed per sonnel. The Lawton study extends previous conclusions that PCB exposure under occupational conditions does not lead to signs or symptoms of clinical disease, despite the presence of normal incidence rates for symptoms associated with inter current disease and/or general stress. Further, the Lawtorr study did not reveal any abnormal findings with respect to pulmonary function in PCB exposed workers, or any changes in either fasting blood glucose levels or serum uric acid levels vs. those levels in control populations. 13. In the domain of potential behavioral effect* attri butable to PCB exposure, several recent studies are of inter est. In offspring of rats given prenatal treatment with Xanechlor 300 and 500, and of monkeys dosed with Aroclor 1248 MONS 212873 -7 in the diet during pregnancy, signs of decreased learning ability were observed. Offspring of mice after prenatal and postnatal exposure to Aroclor 1254 showed a latency period In making responses, following training in an avoidance response pattern. These behavioral manifestations of PCB exposure in animal models have not been related to any poten tial behavioral changes in exposed human populations. 14. The PCDFs have been studied with increasing interest in recent years as realization grows that they are toxic derivatives of the PCBs and may be important contributors to the toxicity of mixtures in which they are present if their concentrations are at high enough levels. In recent animal studies they have been shown to be powerful aryl hydrocarbon hydroxylase (AHH) inducers, especially when the lateral 2,3,7,8-positions are fully substituted by chlorine atoms. They produce many toxic signs in test animal systems at doses in the ug/kg/day range, including severe skin changes, swollen eyelids, enlargement of meibomian glands, squamous metaplasia, hyperplasia of the gastric mucosa, thymic involution, hypo plasia of the bone marrow, and hepatotoxicity. Aside from continued analysis of residual health effects in Yusho pa tients, in which the opinion is growing that the PCDFs are principal contributors to the persistent toxic syndrome ob served, no human PCDF exposure vs. effect information has been documented in the 1981-1982 interval. -8 I. INTRODUCTION A study on the potential health effects documented in the literature throu9h 1981 lor human populations, from exposure to commercial polychlorinated biphenyl (PCB) mixtures under occupational or other environmental conditions, was completed by this firm in 1982. Entitled "Potential Health Effects in the Human from Exposure to Polychlorinated Biphenyls (PCBs) and Related Impurities", the study (hereafter labeled as the dfhLS report), was sponsored by the Edison Electric Institute (EEI) and the National Electrical Manufacturers Association (NEMA). The literature coverage in that study included important toxi cological and epidemiological papers generally available through the middle of 1981. It was issued under a date of February 12, 1982. Since many important PCB studies have appeared in publica tions or presentations since mid 1981, an update of the DFHLS study in the form of an annex was deemed advisable by both spon sors. The present document constitutes that updated annex to the 1982 DFHLS report. It covers literature retrieved from library and other sources up to a cutoff search date of November 15, 1962. As in the original DFHLS study, the present annex was constructed in draft by a Category I team composed of members of this firm (Drs. Drill, Friess, Hays and Loomis), reviewed HONS 212875 I 9 for content and with suggestion* for revision by two external peer reviewers (Drs. Stendeert and Alvaresl, and finally revised by the Category I team which takes full responsibil ity for its content and conclusions. The annex is cast in a format which largely tracks the original report* section by section, in the sequence of health effects. For each effect, the results of now toxicological and epidemiological studies are considered, and then reviewed in the framework of results cited in the 1982 report to reach judgment on the possibility of occurrence of that effect in exposed human populations. Also included in the annex are sections dealing with recent publications on behavioral effects attri buted to PCB exposure, and with new findings on potential health effects from exposures to polychlorinated dlbenzofurans (PCDFs). It is of considerable interest that modern receptor theory has been brought to bear on interpretation of health effects of of PCBs and related compounds. It is possible that manifes tations of toxicity may result from a, common initial path way of interactions of certain PCBs, PCDFs or PCDDs with tissue elements in mammalian organisms. In particular, recent interpretations of experimental findings in test animal systems have centered on the possibility that on* or more receptor pro teins may be involved in initial binding of a suitably substi tuted compound leading to ultimate expression of a biological NONS 212876 -10- activity such aa chloracne or thymic involution. In this picture of events, under a theory proposed by Poland and Knudsen (1982) and Nebert et al. (1981), the effects stemming from exposure to a given polychlorinated molecule would be 'tra ceable to initial binding to an inherent receptor surface whose structure and location in tissues are genetically controlled. Such binding could occur broadly across mammalian species and strains, providing a unifying basis for correlating phenomena of many types in different species. The PCBs as a group are considered to be the least potent in this binding - mobili sation process, the PCDFs orders of magnitude more potent, and the PCDDs the most potent of all. The receptor substance itself is thought to be regulated genetically, and to be cytoplasmic in location. If the receptor in induced by a given stimulus and triggered by binding of the agonist, the mobilisation of the endogenous transmitter substance could then account under this theory for hyperplasia at a variety of cutaneous and hepatic tissues, involution of the thymus, and metabolic derrangements leading to lesions of chloracne. The theory is presently quite useful, in the material that follows, in efforts to under stand phenomena that extend across many species and are triggered by specific molecular structures in the PCB and PCDP families of compounds. HONS 212877 -11- II. YUSHO EPISODE Sine* the occurrence of the 1968 Yusho episode in Japan Involving consumption of Kanemi rice bran cooking oil contain ing PCB, PCQ, snd PCDF impurities which were accidentally leaked into the oil from heat exchanger coils originally charged with Kanechlor 400, the medical follow-up and history of recovery of these patients has been persistent and well documented, Yoshihara end Yoshimura (1981) have recently reviewed residual health effects from this intoxication episode, s seen in medical examinations made over the past decade. In general, most of the clinicsl symptoms from this bran oil intoxication have diminished gradually, accom panied by a decrease in blood PCB levels to a mean residual of 7 ppb (vs. a mean level of 3 ppb in control subjects). Since the estimated average total intakes of PCBs, PCQs and PCDFs in these patients during 1968-1969 amounted to 633, 596 and 3.4 mg respectively (Yoshihara and Yoshimura, 1981) the persistence of traces of markers of the original intox icant load in the circulatory systems of these individuals after a 15 year period is striking. Perhaps stemming from residual body burdens in storage tissues, this occurrence of traces of marker substances in blood has been accompanied by the persistence of certain symptoms, e.g. subcutan eous cyst formation, pigmentation of the skin, eyelids and gingiva, eye discharge, and abnormal changes of the tarsal gland in a relatively small number of severely affected Japanese patients. HONS 212878 -12- Conslderable variance has bean recorded in estimates of PCB, PCO and PCDF Intakes In patients and In oil samples from the Taiwanese episode. In a very recent analytical study, Masuda e_t al,. (1982) have reported on analyses for PCB. PCQ, and PCDP congeners In the 1979 Taiwan Yusho oil, and in the blood of Taiwanese patients sampled several months after the onset of symptoms from that episode. The levels in oil and blood were compared with those from the Japanese Yusho epi sode of 1968, except that blood samples from Japanese patients were taken 5 years after that poisoning episode. It was found thatt (1) the concentrations of PCBs in the Taiwan oil (60-100 ppm) were 10 percent or less of that in the Japanese oil (900 ppm), but the concentrations in the blood of Taiwanese patients were about ten or more times those of Japanese patients. These results follow from the ingestion of higher amounts of oil by the Taiwanese (about 12,3 kg per person vs. about 800 ml for Japanese consum ers), the net ingestion of about equal amounts of PCBs by the two populations, the longer time interval for metabolism/ elimination by the Japanese, and the fact that the Taiwanese oil had a higher proportion of more highly chlorinated congen ers. (2) The Taiwan rice oil contained a much lower concen tration (90-180 ppm) of PCQs than did the Japanese Yusho oil (800 ppm). (3) Both the Taiwanese, and Japanese oils con tained PCDFs, chiefly 2,3,4,7,8 - penta CDF and 2,3,4,6,7 penta CDF, which were also detected in the blood of Taiwanese HOHS 21279 -13- patients. In view of previous findings by these Investiga tors that 2,3,4,7,8 - penta CDF is one of the most accumula tive congeners In the livers of humans, monkeys and rats, and Is very potent In Its biological actions (P-448 Induc tion, thymic atrophy or liver hypertrophy), Kasuda et al. (1982) conjecture that this congener may well be one of the most Important etlologlc agents for the Yusho poisonings in both countries. Yoshlhara and Yoshimura (1981) also take note of the In creasing body of evidence Implicating the PCDF component In the Japanese Yusho oil as an Important contributor to the total toxic syndrome. Further, they cite supporting experi mental data in the rat with a variety of PCDF congeners, given at single doses in the range 1-10 mg/kg, which Included significant lowering of thymus weight, increase in liver weight, and retention in liver of large fractions (20-97%) of the administered PCDF congeners that contain 5 or 8 Chlor ine atoms per molecule. Nagayama at al. (1981) have also contributed to the data base on possible contributions of the PCDF component to the Japanese Yusho syndrome, from experiments on controlled heat ing of Kanechlor 400 which parallel the cumulative heating cycles characteristic of the use of this fluid in a metal heat exchanger. In the presence of metallic iron, nickel or stainlees steel and a small amount of water contaminant as MOMS 212880 -14- n oxygen source, heating at 300*C for seven weeks leads to an inert as* o PCDF content in the Kanechlor to levels of 240-560 ppm; further heating at 350*C for four more weeks led to enhancement of PCDF content to 610-740 ppm. The possible role of PCDFs as a principal agent in Yusho poisoning has been studied in detail by Kashimoto et al. (1991). This work was based on comparative health effects observations on a group of 56 patients in the Japanese epi sode, 15 Yusho patients in the Taiwanese event that occurred about ten years later, control populations, and groups of workers occupationally exposed to PCBs in Japan and Taiwan. Kashimoto et al. have presented analytical data on relative PCBt PCQt PCDF residual levels in liver in Yusho patients from the Japanese episode, showing relatively high retention of PCDFs In that tissue (PCBst PCOst PCDFs 100i76;20>, and on concentrations of these components in several samples of the toxic Yusho oils obtained during the two separate but parallel intoxication episodes. From their data analyses, Kashimoto and colleagues concluded that the PCDFs in the Yusho oils were the major pathogenic compounds contributing to the development of the Yusho syndrome in the Taiwanese patients, despite their low levels in blood compared to those of the PCBs and PCQs (PCBst PCQs; PCDFs * 500t 160:1). This conclusion was based principally on the following points. 1, The clinical manifestations and duration of Yusho symptoms were disproportionately more severe and MONS 212861 -15- perslstent In consideration of the blood PCB levels observed In Taiwanese patients than in PCS - exposed populations of industrial workers with considerably highar blood PCB levels who had much more benign or mild clinical syndromes. 2. PCDTs show a marked tendency to accumulate in the liver, which could well explain the frequent initial finding of jaundice and other serious abdominal symptoms in Yusho patients and the absence of such changes in people occupationally exposed to PCBs. 3. Prom biological dose vs. response data for a series of toxic effects in many test animals, including the chick, monkey, mouse, and rat, it is apparent that the toxicities of PCBs and the derived PCQs are of the same order of magnitude, but that the toxic potencies of PCDFs are greater by factors ranging from 100 - 10,000. finally, Chen et al. (1961) have analyzed the toxic con taminants in a Taiwanese Yusho cooking oil that was ingested by approximately 2,000 people. These investigators studied six samples of the oil, both for total PCB and PCDP content and for the distribution of congeners in each family of com pounds. They found that the distribution in the PCB fraction favored the more highly chlorinated structures HONS 212882 -16- (penta-CBt tetrat hexai heptai tri* 47t30tl6i6il), and that tha PCDF conganara ware mostly of tha panta - CDF and tatra CDF variety with only minor amounts of haxa - CDF and trl - CDF. This distribution of components differs from that in Japanese Kanachlor 400. and would be In accord with more lasting Yusho symptoms by persistence in tissue storage of more highly chlorinated molecular species that are less active metabolically and less likely to be excreted as polar metabolites or their conjugates. Discussion and Conclusions These recent findings, especially as they have focused increasingly on the potential role of elevated levels of PCDFs in contaminated cooking oils as the principal contri butor to the Yusho syndrome in the Japanese and Taiwanese intoxication episodes. serve to emphasize and reinforce points made in the DFHLS analysis of earlier studies. First, heat exchanger action on PCB fluids leading to elevated PCDF levels contributed . to the enhanced toxicity of the Yusho cooking oils over that of the parent commercial PCB mixture. Second, because of the unique PCBtPCQtPCDF distribution of the Yusho oils, it is not correct to use these intoxication episodes as models of acute or subdhronic PCB poisoning in humans, or for prediction of human health effects from expo sure to commercial PCB mixtures. HONS 212883 -17- III. BODY BURDENS, WTABOLISM AND KINETICS A. Animal Experlmenti In continuation of studies on the biological ftt of individual PCB congeners in aniiaal models following adminiatration of a ainglt doaa of a radiolabaltd compound, Sipta e al. (1982) hava raportad on tha diatribution, metabolism and excretion of 2,4,5, 2*,6,,5`- haxa-CB i n tha baagla dog and tha cynoaiolgus monkey by following tha lovala of 1<C labeled matarial in excreta, blood and tiaauaa for up to IS daya in tha dog and 90 days in tha monkey. Thasa workara obaarvad tha rataa of elimination of this paraistant haxa-CB from thasa taat animals, Elimination from blood occurrad aa a biphasic kinatic procaaa. Aftar tha first rapid phasa, tha alowar tarminal elimination from blood was characttritad by rata constants that wars thraa tiotas as graat in tha dog as in tha monkey. Tha kinatics of axeration of tha PCB in tha two spacias also d if far markadlyi by 19 days tha dog had axcratad 86 parcant of tha dosa, mainly in tha facas, wharaas tha monkay in a 90 day intarval had axcratad only 18 parcant of tha dosa. Tha ramaindar of tha dosa in both spacias was found largaly as tha parant compound in adiposa tissue and skin as tha major storaga dapot. Further study on biliary axeration of tha haxa-CB in tha two spacias again showed that axeration processes in tha dog are faster than in tha monkay, fourfold in this case during two hours, with 0,8 percent of the label appearing in the bile of anesthetixed dogs during this period. However, tha monkey axcratad a MONS 212884 -18- greater percentage of the dose as parent compound into the biliary flow than did the dog, reflecting a lower degree of metabolism of this hexa-CB in the liver of the primate. Prom these and other observations, Sipes et al. reached a number of significant conclusions. First, the dog ia clearly able to eliminate this hexa-CB from the body substan tially faster than the monkey, indeed, the dog is the only test species that has been shown to reduce its body burden of this persistent PCB substantially, probably because of its ability to metabolise the molecule despite the lack of vicinal unsubstituted ring position pairs. The monkey appears to be similar to other species in its pattern of elimination of this hexa-CB. The processes of tissue distri bution in the dog are relatively rapid, beginning with early distribution into skin and adipose tissue. In the dog liver, metabolism is relatively facile in that approximately 60 percent of the radiolabel present in that tissue at 24 hours was in the fora of metabolites. In contrast, for both dog and monkey, the major molecular species present in muscle, skin and adipose tissue was in the form of unmetabolised parent compound. For both animal species, the major route of elimination of the hexa-CB and metabolites was via the feces. These results of Sipes et al, with 2,4,5 ,2 ,4',5hexa-CB can be integrated with findings from previous studies by Matthews and his colleagues (e.g. Matthews and Anderson, 1975) HONS 212885 -19- 1975) to Afford a brief survey of th pharmacokinetic beha vior of three PCBs In three apeciea, doga, monkeys and rats. For 4,4* dl-CB, data from the rat afford reasonable predic tive power for distribution, metabolism and elimination in the dog, but not in the monkey. For 2,3,$- hexaCB, the dog and the rat eliminate the compound at similar rates which were, however, faster than the observed rate in the monkey. Finally, in the present study by Sipes at al., the 2,4,5 hexaCB data from the monkey are quite similar to previous findings in the rat, but the dog is decidedly different in that its metabolism and elimination rates are much faster. Clearly, species specificity operates markedly and not always predictably in the pharmacokinetics of PCBs in test animal models. The reason behind the ability of the dog to metabo lize the persistent 2,4,5 - hexaCB, presumably by direct biochemical action at unsubstituted ring meta positions, is presently unknown. A point bearing on the direction of PCB metabolism In liver by monooxygenases induced by Aroclor 1254 or by 3-methycholanthrene (3-MC) was demonstrated in a recent paper by Halpaap-Wood et 1. (1981). These workers studied the actions of enzymes contained in 9000g fractions derived from the PCB induced Sprague-Dawley rat and the Ha/ICR mouse on unsubstituted biphenyls. When a single polar group is already in place, as in the monohydroxylated biphenyls, it appears that induction of the 3-MC type (P-448 requiring HONS 212886 -20 enzymes) leads to further monohydroxylet Ion in the ring that already bears the first hydroxyl group. The diols produced in this metabolic step result from ortho-para direction by the initial hydroxyl group, leading to 2,5 - and 2,4 -diols. These workers also demonstrated a species difference in the effects of PCBs on the hydroxylatlon of biphenyl in mice. They showed that Aroclor 1254 induction of biphenyl metabo lism differed in the rat and in the mouse. In the rat, PCBs induced hydroxylatlon both in the 2- and 4- positions. In the mouse, Aroclor 1254 resulted in an Increase in 2- OH biphenyl formation, but not in the 4- hydroxylation. Work on metabolic hydroxylation of commercial Aroclor mixtures in test animal systems has recently been extended to the terphenyl series. Voss (1981) has observed that the mixture of polychlorinated terphenyls (PCTs) in Aroclor 5460 and a representative hexaCT congener (2,2)2" 5,5',5'' -hexa-CT) undergo metabolism in the Sprague-Dawley rat to yield monohydroxylated metabolites, which as yet have not been fully characterised. The rates of excretion of these hexa-CT congeners and their metabolites in feces and urine were quite lowt within 7-14 days only about 10 percent of the radiolabel in the administered dose was recovered in excreta, largely in the feces. HONS 212887 -21- B. Human Observations Previous studies by Jensen and Sundstrom (1974), who determined the fate of 40 PCB congeners in human adipose had shown that PCBs with no chlorines at the 3,4- position of one ring (e.g. 2,5- diCl-> constituted about 201 of PCBs in human adipose tissue. Two compounds, 2,4,5,2',4',5'- and 2,3,4,2',4',5'- hexachlorobiphenyl accounted for 36% of the PCBs is tissue. Similar findings were reported by Kuroki and Pasuda ( 1977), In an effort to assess exposure among workers in capacitor manufacture, Wolff e a^. (1982) determined PCB concentrations in plasma (290) and adipose tissue (61). The highly chlorin ated congeners appeared to be present mostly in tissues of workers who had been exposed to the higher chlorine-containing Aroclors in the past. The less highly chlorinated PCBs, di-, trt-, and tetrachlorobiphenyls, were found in workers who were recently employed and had thus been exposed to the lower chlorine-containing PCBs mixtures, such as Aroclors 1016 and 1221 which were used from 1971 to 1976. further they found that higher exposure occurred among persons with direct contact with PCBs, in jobs such as capacitor filling. Adipose tissue concentrations were proportional to those in plasma, Wolff eal, (1982) have reported recently on epidemio logical findings in a limited cohort of 26 workers employed HONS 212888 -22- over a period of years in tha manufacture of electrical capaci tors. They presented analytical data on the content of 37 different PCB congeners in paired samples of adipose tissue and plasma in these workers. It was noted in general that the concentrations of PCBs in tissues were related to the duration and relative intensity of exposures to these chemi cals in the workplace, with a median range of 0,1-9 ug/g for individual congeners in adipose tissue and 0.3 - 27 ng/ml in plasma. As a rule, the PCB concentration in adipose tissue was proportional to that observed in plasma, with partition ing between adipose and plasma in an approximate ratio 190t1. Congeners with chlorine atoms in 4 and 4'positions, such as 2,4,4'-tri. 2.4.2',4'-, 2,4,5,4'- and 2,4, 3',4' -tetra-CB, were major components in plasma and adipose tissue. These workers had been exposed primarily to Aroclors 1016, 1242, and 1254 (20-54% chlorine). Congeners with unsubstituted 3,4 positions (e.g. 2,5-dichloro substitu ted) were observed at lower tissue concentrations and lower adiposetplasma ratios than other congeners, probably reflec ting metabolism to polar derivatives and excretion by virtue of vicinal positions unsubstituted in either ring. Conver sely, compounds with 2,4 and/ or 3,4 positions substituted by Cl on both rings were present in much higher proportions in plasma or adipose tissue than in the PCB mixtures used to fill capacitors, probably because of diminished capability for hydroxylation and elimination, and were characterized by much higher adiposetplasma partition ratios than those for HONS 212889 -23- congeners with unsubstituted 3,4-vicinal ring positions. Also, ss expected on ths basis of increased PCB tissue storage when Metabolic possibilities are limited, those congeners with 2,4-substitution on both rings had higher adiposeiplasma partition ratios than congeners with 3,4-sub stitution on at least one of the rings. Finally, metabolic oxidation and excretion of polar metabolites in human sub jects of those PCB congeners with two vicinal unsubstituted carbons correlate rather well with lower adiposeiplasma partition ratios previously observed by Matthews and Tuey (1980). Bickal and Muehlebach (1980) seam to agree with Matthews that by blocking metabolically vulnerable groups (such as 2 unsubstituted carbons) a persistent compound may become an unmetabolisable one. They cite the "key work* of Matthews and Anderson (1975) that the hexa isomer, 2,4,5,2 *,4 *,5*CB has no unsubstituted pair and was practi cally unmetabolixed. They also state that in addition to degree of chlorination, the para positions (4,4 *} are most metabolically vulnerable positions. Thus 2 ,4,2',4'- tetra-CB is more persistent than the 2,5,2',5*- isomer. Finally, they speculate that the general structure-persistence relationships apply to other polyhalogenated hydrocarbons, such as PBBs, and polychlorinated naphthalenes. Bickal and Muehlebach noted, for correlation purposes be tween congener distributions in tissues and in commercial HONS 212890 24- Aroclor mixtures, that the distribution of PCB congeners in human adipose tissues sampled in their survey centered about a mean degree of chlorination near that of ho^a-CB and therefore was closer to the composition of Aroclor 1254 (mean degree of chlorination, 5.1) than to Aroclor 1242 (mean, 3.1} or Aroclor 1221 (mean, 1.2). It is clear, however, that the distribution of congeners in human tissues after extended Aroclor exposure intervals, and a description of their mean degree of chlorination, do not necessarily constitute quanti tative indices of identity or relationship to any single Aroclor source. Failure to establish this firm linkage is due, at least in part, to the shifts in residual retention in adipose tissue that can accompany structure-specific metabolism and excretion of suitably substituted congeners from a complex exposure mixture. Discussion and Summary The current findings on tissue levels and pharmacokin etic handling of PCBs in test animal models and exposed human populations have not revealed any significant new insights on the mechanisms employed by mammalian systems in coping with the presence of these molecules in tissues. Nearly all studies reviewed suggest strongly that in most species, including humans, the higher chlorine containing isomers are more biopersistent and resistant to metabolism to polar metabolites. The total data base on PCB handling HONS 212891 I -25mechanisms including tissue distribution, metabolism excre tion, points to distinct species specificity in these pro cesses, with rate and tissue distribution constants that are also linked to molecular structure of the individual pcb congeners. The residual concentrations of PCB congeners in fstty and other tissues are seen to be particularly dependent on the ease of metabolism and subsequent excretion of polar metabolites of specific components of the cosunercial PCB mix tures to which humans and other animal species are exposed. MONS 212892 26 IV. GENERAL TOXICITY in a i*cant study by Norback at al, (1981) a call culture from a mouse embryo call line was used to examine the toxicity of two hexachlorobiphenyls. this method involves measurement of growth rate and plating efficiencies of the cell line. Evaluation of two hexachlorobiphenyls (2,4 ,5.2'.4',5* and 2,3,6,2*,3' .6 ') demonstrates that the 2,3,6,2*,3*,6'-isoer was the most toxic end the order of toxicity paralleled the long-term whole animal studies. Not all cases showed a linear dose (concentration)-responee related effect on the cell numbers after 10 days exposure. Biocca et al. (1981) fed hexachlorobiphenyl isomers in the diet of five week old mice. Hexachlorobiphenyl residues were determined in adipose tissue and liver. The 3,4,5,31,4',5' -isomer (not commonly encountered in commercial mixtures) was the most toxic of those studied and was the most concen trated in the fat and liver. It also produced the greatest mortality as well as hepatocellular markings, hepatocellular swelling and necrosis, subcutaneous edema and changes in body and organ weights at doses ranging from 0.3 to 300 ppm in the feed. In a review article by McConnell and Moore (1979) on the general toxicities of halogenated aromatics, including the dioxins, biphenyls, dibenxofurans and naphthalenes, they noted that the structure-toxicity relationships are similar HONS 212893 -27- Cor *11 th compound* in non-hum*n primtt* but the absolute doses to produce toxic effects *re quite different. The affects include loss of body weight, Alopecia, chloracne, blepharitis, waxy ear canals, dry-scaly-skln and nails and progressive weakness. Fetal deaths often occur and offspring may die early. The signs are less conspicuous in rodents than in non-human primates. Rabbits specifically show chloracne. There are other effects, such as, testicular degeneration, hyperplasia of the epithelial tissue in the urinary tract, leucopenia, pancytopenia and altered serum enzyme systems which may be reflective of organ damage. Behavioral Effects The use of behavioral effects in the assessment of the toxicity of xenobiotlcs, particularly those affecting the ner vous systems, is a new field of toxicology which may event ually provide sensitive biological endpoints. There are some who believe that behavioral changes in offspring as well as reduced growth rates should be considered as teratological effects, but until more is known about behavioral lesions and the underlying mechanisms we prefer to adhere to the original concept that teratology refers to abnormal structural changes during embroyonlc development. Several animal species have been used to detect subtle behavioral effects of PCBs in offspring of mothers exposed to MONS 212894 -28- the environmental pollutant during various stages of gestation. Shiota (1976) using the Japanese brand of PCBs, Kanechlors 300 and 500. adsiinistered by gavage on 8-14 and 15-21 days of gestation to rats* ahowed that the offspring had lower birth weights than controls and learning impairments were observed in aase testing. However, other field activities and swimming ability of the treated and control offspring were similar. Storm et al. (1961) administered Aroclor 1254 to female mice prior to mating and during gestation. There was no effect on the ability of the mice to learn an avoidance response; but they took longer to learn the response. The data obtained with primates appear to be more defin itive. Bowman et al. (1978) observed that feeding of Rhesus monkeys at 2.5 ppm of Aroclor 1248 in the diet resulted in disruption of discrimination learning. A dose-effect rela tionship between PCB body burdens early in life and later behavioral effects were observed. Surviving monkeys were hyperactive in locomotor tests and deficient in learning various types of discrimination problems at 6 and 12 months of age. These same monkeys became significantly hypoactive. when compared to controls, at 44 months of age (Bowman and Heironimus (1981). These data may indicate that primates may have a greater sensitivity to _in utero PCB exposure than do mice. In other studies. Bowman et al. (1981) also ob served hyperactivity in the early stages of development of monkeys whose mothers had received the PCBs. However, in MONS 212895 -29- these studies tha hyperactivity did not correlate with PCB dosage, leading to the desirability of further study. More extensive studies are required to. determine which particular period of exposure (gestation or lactation) contri butes to the observed behavioral effects. It would also be important to correlate these effects with physiological or biochemical tissue markers, in particular at the central nervous system (CDS) level. Discussion and Conclusions Recent activity in the study of tha acute toxicity of commercial Aroclors and synthetic or purified congeners has tended to focus on evaluation of specific changes induced in target organs and cell lines. Findings of interest that complement previous literature includei (1) substantial tox icity of 2,3,6-HCB in inhibition of growth of a mouse embryo cell line; (2) marked hepatocellular toxicity of the 3,4,5-HCB congener in young mice; and (3) an explanation for the inability of 4-chloromethylbiphenyl to induce ijn vitro genotoxic effects in terms of its ready metabolic conversion to the corresponding hydroxy compound. These findings do not alter the generalizations on * PCB toxicity, acute or chronic, pointed out in the 1982 report. The relevance of the behavioral studies in animals to man can only be a matter of speculation, for little or nothing is HONS 212896 -30- known about any structural or matabollc changes that might hava taken placa during tha pariod of gastatlon and tha lngastlon of PCSs In tha animal tests. In animal studies, it is wall known that housing conditions and malnutrition may altar tha davalopmant of offspring, laading to soma abnormal behav ioral pattern, It is our vlaw that ona cannot draw any ganaral conclusion from thasa studies in ralation to behavloral changas in man and that much mora work will ba ragulrad to undarstand tha machanisms involvad in altarad bahavioral activity. Further, tha currant findings do not cast add itional light on tha problam of whathar or not laslons laad ing to bahavioral abarrations ara tha pracursors of patho logic changas in tissuas in animal modals or humans. MOMS 212897 31- V. DERMkL EFFECTS in the previous report on potential human health effects from exposure to PCBs, It was pointed out that one of the clinical manifestations of Yusho disease was the appearance of chloracne In about 62% of the patients who had consumed contaminated rice oil containing PCBs, PCDFs and PCOs. Fischbein, e al_. (1979) reported on the clinical findings among PCB-exposed capacitor manufacturing workers, and noted a high Incidence (38-41%) of dermatologic problems In his study population which Included about 5% of cases with acneform eruptions. Other investigators (inoue et al., 1975; Ouw et al. , 1976; Chase et al., 1981; Maroni et al., 1981; Baker et al., 1980; Smith et al., 1981) have found low to zero incidence rates for chloracne in populations occupation ally exposed to commercial PCBs and derived products. In a later paper (1982), Pishbein et al. made a more de tailed study of dermatological findings among capacitor work ers and noted a high prevalence of dermatological effects such as erythema, swelling, dryness, thickening of the skin, rash, hyperpigmtntation, comedones and acne. However, acne was far less prevalent in these industrial workers than among patients with Yusho disease, .despite high concentra tions of PCBs found in the blood of some of the subjects in the Fischbein survey. HONS 212898 32- Dlscusslon and Conclusions It seems clear that prolonged exposure to certain Aroclor or Kanechlor type PCS mixtures In the occupational setting, sufficient to establish high blood levels of PCBs, could lead to the development of signs of chloracne In sensitive Individuals, but this effect has been seen only rarely. It has generally been observed that an Industrial outbreak of dermatitis. Including chloracne cases, coincided with the production or use of a poor or oxidized grade of PCB containing enhanced levels of contaminants, presumably the PCDFs, and PCQs. Such contamin ants, which have been Identified as major components In samples of the rice oils from the Yusho evgnts In Japan and Taiwan, In our opinion, have led to the high Incidence of chloracne In Yusho patients. The recent findings of Kashlmoto et al. (1981) also point to this conclusion. Compat ible with these observations Is the finding that most of the signs of chloracne found In Industrial workers are not severe and not long-lasting after exposure has stopped, and are not at all comparable In severity or duration to the symptoms seen In some Yusho disease patients. HONS 212899 -33- VI, EFFECTS OF PCBi ON LIVER A. Hyperplastic Foci and Nodular Hyperplasia in the Liver in a recent study, Preston et al. (1961) reported that the oral administration of Aroclor 1254 alone or with PCDF impuri ties did not induce foci of cellular alteration, hyperplas tic nodules or cholangiomas in the livers of male SpragueOawley rats. Further, pretreatment of the animals with diethylnitrosamine (DENA) followed by administration of Aroclor 1254 did not significantly affect the incidence of foci of cellular alteration or other benign hepatic lesions. These results are summarized in Table I. The tumor-promoting activity of Aroclor 1254 recorded in the Table is discussed in Section vii of this annex. These results are of particu lar interest from the standpoint that the studies of Calandra (1975), the National Cancer Institute bioassay (1978) and Preston et al. (1981) with Aroclor 1254 all fail to confirm the increase in benign hepatic lesions reported by Kimbrough et al. (1975) for Aroclor 1260 in the rat. B. Liver Function and Liver Disease in PCB-Exposed Workers In an unpublished, comprehensive epidemiologic study of General Electric Co, employees, Lawton et al. (1961) studied liver function in 172 capacitor workers previously exposed to PCBs. Serum PCB levels for the workers averaged 276 ppb for MON5 212900 34 Aroclor 1242, 58 ppm Cor Aroclor 1254, and 34 ppb Cor Aroclor 1260. Tha hepatic atudlta showed that tha naan serum valuas + 2 standard daviations (SD) wara within tha normal ranga Cor SCOT, SGPT, alkaline phosphatasa, total bilirubin and diract reacting bilirubin analyses. Tha mean sarun SGPT valua oC 34.5 iu was within tha normal ranga (1-55 IU), but tha uppar level oC *2 SD exceeded the normal ranga. As a part oC this study, Lawton at al. conducted an evaluation oC tha history oC 194 PCB-exposed capacitor workers. There was no relation ship between the exposure and episodes oC jaundice or hepati tis, most oC which occurred prior to employment. Summary and Conclusions These recent studies add important Cindings on the poten tial impact oC PCB exposure on liver structure or function in tha rat and in human populations. Clearly, Aroclor 1254 does not lead to benign liver lesions in tha male SpragueDawley rat, and extended exposure to Aroclors in the indus trial setting has not resulted in Cindings oC significant liver dysfunction or disease in capacitor workers surveyed in the Lawton study. HONS 212901 TABLE r Incidence of Liver Lesions In pale Sprague-Davley Rat3 Treated with Aroclor 1254 (Data from Preston et al., 1981) Number of: Treatment* Ar 1Control Ar 1254 DEN A + DEN A + AR PCDF DEN A 1254 AR 12 PCDF Animals necropsied Foci of cellular alteration Neoplastic nodules Hepatocellular carcinoma Cholangioma Choiangiocarcinoma 72 0 0 0 0 0 34 34 32 33 32 0 0 10(31) 0 1(3) 0 0 6(19) 10(30) 5(16) 0 0 5(16) 21(64) 27(04 ) 0 0 1(3) 1(3) 2(6) 00 0 1(3) 1(3) Numbers in parentheses are % incidence MOWS 212902 -35 VII. GASTRIC LESIONS Morgan et al. (1981) examined histologically the stomachs of 144 rats treated with Aroclor 1254. (These preserved tis sues were from the NCI, 1978 bioassay study). All stomachs were examined (a) by routine section through the junction of the pyloric region and the duodenum and (b) by section of any areas that gave a positive staining reaction for alkaline phosphatase. Although the phosphatase stain may aid in detecting areas with cellular change, the authors state that most of the stomachs showing positive alkaline phosphatase activity were histologically normal. The authors reported an increased frequency of cystic in testinal metaplasia in the treated animals* comparative inci dences were 0/47 in control rats, 3/46 in low-dose, 3/48 in medium-dose, and 13/48 in the high-dose treated animals. Treatment did not change the frequency of "crypt type" intes tinal metaplasia or diffuse intestinal metaplasia. The total number of rats with all types of intestinal metaplasia was 3/47 in control rats, 4/43 in low-dose, 5/43 in mediumdose, and 15/43 in the high-dose treated group. Thus, the high-dose group, but not the low or medium-dose groups, showed a significant increase in intestinal metaplasia. Metaplasia is characterized by an adaptive substitution by one type of adult or fully differentiated cell for another MONS 212903 -36- type of adult cell. Although intestinal metaplasia and stomach cancer may co-exist in the pyloric region of the rat stomach, it does not mean that the changes are related or that metaplasia is a premallgnant change. Findings in this study with respect to stomach or intestinal tract cancer are discussed elsewhere in this review. An unusual form of gastric lesion has been observed tCJeistCeld e_t a_l., 1932) in monkeys housed m new Cncllitit.-a which resembled the morphological changes that have been described for animals receiving PCBs in their diet. Forty- five of 259 male nonkeys died of chronic progressive disease. The principal necropsy findings were hypertrophic and hyper plastic mucous qastrophy. The gastric lesion was characterized by hyperplasia of the gastric epithelium, hypertrophy of mucus- socreting glands, cyst formation and o?not rat ion of glands into the submucosa. PCBs were implicated in tit is finding because of the presence of Aroclor 1254 In tbs adipose tissue of the animals and in samples of concrete from pens in which the monkeys were housed. The source of the PCDs appeared to be the concrete sealer used in the construction of the pens. Re moval of successive layers of concrete greatly reduced the level of PCD exposure, with a corresponding reduction in animal deaths. Quantitative information on levels of PCBs in the concrete and on dose transfer to the monkeys was not obtained. HONS 212904 -37- Summary and Conclusions It seems clear that relatively high doses of commercial Aroclorn delivered chronically to the stomachs of rats or mon keys can produce pathologic changes in stomach or nearby intestinal tissues. In the rat the ma^or change is seen m an increase in the frequency of cystic Intestinal metaplasia in tissue loci otherwise characterized by elevated alkaline phosphatase levels, tn the monkey the principal pathologic findings in the stomach are hypertrophic and hyperplastic mucous changes. These recent findings amplify and extend indications of target action on the stomach seen previously in the monkey and rat. HONS 212905 38 VIII, CARCINOGENESISi BXPERIMNTAL A. Hepatocellular Carcinoma Preston st al^ (1981) reported that administration jo( Aroclor 1254 alone or with added PCDF content for 23 weeks In mala Sprague-Dawley rats did not Indues hapatocallular carcinoma or cholanglocarcinona In tha traatad animals (saa Table I, Sactlon VI), Thay found, however that the number of hepatocellular carcinomas Induced by dlethylnltrosamlne (DENA) was In creased when the DENA-treated rats also received Aroclor 1254, either with or without PCDF, indicating a tumor promo ting effect of the Aroclor (Table I). S. Bladder Cancer [Madder tumors were not observed in rats given Aroclor 1254 for 23 weeks (Preston t al,, 1981). C, Stomach, Intestinal Tract Cancer Using fixed tissues from an earlier bioassay, Morgan et al, (1911) conducted a reexamination of tha gross morph ology and histology of tha stomachs of 191 animals preserved from the 1971 NCI bioassay of Aroclor 1254 in F344 rats. The stomachs of these animals were stained for alkaline phosphatase activity, and tha alkaline phosphatase positive areas were used to select areas of possible pathological significance. By this means, thay detected histological changes in an additional three animals which thay characterised as early adenocarci noma! one lesion was in the medium-dose group and two in the HONS 212906 -39- high-dose group. Sex of the animals was not stated. They also noted two regions of severe dysplasia in the high-dose group. The lesions which were observed in the glandular stomach and at the 3unctlon between the pyloric region of the stomach and the duodenum by Morgan e_t a_l, included intes tinal metaplasia, cystic lesions, and three new adenocarcin omas not seen in the original examination. The total inci dence of adenocarcinomas observed in a total dosed population of 144 animals was therefore one cancer at the low-dose, three at the medium-dose and two at the high-dose of Aroclor 1254, for a total of six adenocarcinomas. The authors con cluded that since the incidence of stomach adenocarcinomas was less than two percent (no tumors in 4? animals) in simultaneous controls, the observation of six cancers in 144 treated rats constituted a finding in which the likelihood of occurrence at random is less than 0.05. Further, they referred to this cancer incidence rate relative to the incidence of stomach adenocarcinomas in historical controls (cited to be one case in 3541 animals), and concluded anew that the occur rence of six adenocarcinomas in 144 Aroclor-treated rats is highly significant (p < 0.001). These conclusions as to the statistical significance of the adenocarcinoma findings in the stomach tissues reexamined by Morgan e al. should be subjected' to scientific scrutiny and challenged on several grounds, detailed below, 1. First, Morgan et al. do not specify what statistical MONS 212907 40 methodology they employed to retch their conclusions about the significance of the cancer Incidence in Aroclor-treated animals vs. either the concurrent controls or the historical controls. The comparisons are further confounded In that the cancer Incidence is not separated by both sex and dose for test animals, and proper comparisons are not made between cancer incidences for each sex at each dose level and the corresponding control group values. 2. Since the rate of occurrence of cancers by sex group was not recorded, the only valid test that one can apply to these experimental results is the simultaneous set of com parisons for three dose groups and controls. When this is dona, for example, by chi-square analysis with three degrees of free dom, the chi-square statistic is 3.39 for the Morgan data. A value of 7.B15 would be required for significance at a level of 0.05. Therefore, the rates of occurence of stomach adenocarcinomas among the dosed and concurrent control groups as presented in the paper could represent Just * random draw ing from the same populations. 3. Attribution of possible biological significance to the adenocarcinoma incidence rates In the throe treated groups is further compromised by the fact that there is no clear Aroclor dosage vs. response relationship among the results. Indeed, fewer cancers were observed at the high-dose than at the the medium-dose. The authors do not discuss this point. HONS 212908 -41- 4. Finally, the reference by Morgan e_t al. to cancer rates in historical controls is inappropriate and possibly mislead ing, since the tissues of historical controls were not exam ined in the same painstaking manner and by tha same personnel who performed tha Aroclor animal reexamination. Bearing ;n mind that Morgan at al. picked out cancers in their tissue reexamination that had not been recorded by the pathologists in the 1978 bioassay procedures, it could well be that cancers in the large number of historical controls went unrecorded. These points constitute serious objections to any possible conclusion that the 197B NCI bioassay provided consistent evidence for the occurrence of stomach adenocar cinomas in F344 rats treated with Aroclor 1254. Another aspect of this general problem of potential stom ach tumors in test animals dosed with Aroclor mixtures is pro vided by the findings from studies in monkeys discussed in our previous health effects document. Workers in the laboratory of Dr. J. R, Allen at the University of Wisconsin had performed experiments with Aroclor 1249 administered in the diet of the Rhesus monkey, leading to findings of gastric lesions in this primate species. These findings included hypertrophy and hy perplasia of the gastric mucosa, edema of the stomach submu cosa, and mucous conversion of the gastric epithelium at doses as low as 100 ppm in the feed. No such gastric changes were documented in monkeys fed at 2.5 or 5 ppm in the diet. Additionally, Aroclor 1248 in the diet as low as 2.5 ppm HONS 212909 also produced reproductive problems in monkeys end some fetotoxioity m their offspring m the experiments conducted hv Dr. Allen and his colleagues during the period 1974-1980, The work oC Allen et al, (1973) on gastric lesions In monkeys received some confirmation In the work of Becker et al. (1979) who produced gastric lesions In Rhesus monkeys receiving dotages of Aroclor 1242 In the mg/kg range. How ever. latropoulos et al. (1977) working with the PCU mixture Clophen A-30 and the monkey aa test animal did not obsarve the production of gastric lesions. The question of whether low doses of PCB mixtures of a sufficient degree of chlorin ation can indeed produce gastric lesions in a primate model, possibly as the precursors of the carcinogenesis which Porgan et al. claim to have aeen in Aroclor 1254-treated rats, is an important health effects Issue that requires further experimental study. Finally, in the recent study of Preston et al. (1981) it was found that treatment of male Sprague-Dawley rats with Aroclor 1254 (100 ppm in the diet) for 23 weeks did not induce gastrointestinal carcinomas. Summary and Conclusion Tht recent evidence on carcinogenic potential of Aroclor 1254 in the rat, as cited above in two experimental studies HONS 212910 -43focused on the liver, bladder, stomach and intestinal tract as target organs, most be viewed as essentially negative in our opinion. However, the finding by Preston et al, (1981) of a possible tumor promoting effect of Aroclor 1154 in the Sprague-Dawley rat must be judged as biologically significant. Further studies to probe possible promoter activity of PCB mixtures in mammalian systems seem worthy of early attention. HONS 212911 -44- IX. REPRODUCTIVE EFFECTS Spencer (19B2) has made an assessment of the fetotoxic effects of Aroclor 1254 in rats by adding the compound to the diet of pregnant and pseudopregnant Sprague-Dawley rats; doses ranged from 25 to 900 ppm administered on days 6 through 15. There was a decrease in ovarian protein in pseudopregnant rats and a decrease in placental protein and glycogen in pregnant rats. Embryonic resorption did not occur up to ex posure levels of 900 ppm, but there was a decrease in fetal survival rate and fetal weight at concentrations of 300 ppm and 100 ppm respectively. There was also a significant decrease in food intake and maternal body weight at doses of 300 ppm and above, and nutritional factors alone may account for the reduced fetal survival and fetal weights observed at these high concentrations. Summary and Conclusion This new study on the impact of Aroclor 1254 on reproduc tion in female rats fed 25 to 900 ppm in the diet at a critical period during gestation confirms previous reports that high doses of the PCBs may be fetotoxic. Although it is Known that high doses of PCB may affect food intalce and weight gain, the degree to which the fetotoxic effects are directly caused by PCBs or are a secondary consequence of altered nutritional state in the dosed animals is presently unknown. - HONS 212912 -45- x. mutagenesis Although tht Annex dealt principally with polychlorin ated biphenyls, it ia of interest to note results from a study by Ashby e_t _al. (1981) on mutagenic properties of substructural elements of the PCBs in the hi vitro Ames test systems. Starting with single aryl rings, these workers found that aniline and benzyl chloride give a negative re sponse in the Ames test for mutagenicity, using a number of strains of S.tvohimurium. The addition, however, of a benzene ring to form 4-aminobiphenyl (4AB) and 4-chloromethylbiphenyl (4CfS) greatly enhances the biological activ ity above that of the parent compounds. Positive responses for 4AB were observed in strains TA 1537, TA 1538, TA 100 and TA 98. A negative response was observed in TA 1535 . Positive responses for 4C>*B were obtained in strains TA 90, TA 1538 and TA 100 but weak and non-reproducible positive effects in strains TA 1535 and TA 1537. On the basis of these observations one might expect that certain PCBa would also be mutagemc in the same Ames assay systems. Schoeny (1982) tested 2,4, 2',4'-tetrachlorobiphenyl, 3,4,3',4J-tetrachlorobiphenyl, 4-chlorobipheny1, 2,4,6, 2 *,4',6'hexachlorobiphenyl, and several chlorinated d ibenzof urans. All the compounds were found to be non* mutagenic for strains TA 9B and TA 100 when tested over a 3-log dose range. They were also non-mutagenic when tested with either induced or non-induced microsomal extracts as HONS 212913 -46- activators. These results are reminiscent of the recent finding on the lack of mutagenic potency, with or without so activation, of the important PCDD congener 2,J,7,(1 dibenzop-dioxin in the Ames systems. Summary and Conclusion These recent test results on mutagenic potential of PCBa and their analogs in the Ames test with Salmonella test strains add to a body of previously available data pointing to a lack of potency with regard to this important biological effect. Although the biphenyl nucleus substituted with e single polar or reactive group INH2 or CHjCl) yields posi tive results in Ames systems, single or multiple substitution of Cl groups on the biphenyl rings generally does not yield mutagenic activity, with or without mstabolic activation by S9 fractions. It it of further interest that this leek of nutagenic activity alto extends to several polychlorinated dibenzo-p-dioxins. MOWS 212914 -47- XI. OTHER HEALTH ErrECTS A. Effects on Enzyme Svtmi Many reports have appeared in the past two or three years which dealt mainly with effects of the PCBs and related halogen- ated compounds on liver microsomal enzyme systems. Most of theee reports simply confirm what is known of the effects of PCBs on the liver and its enzymes. Gartotf at al. (1901) showed that single or repeated doses of various Aroclors in rats produced decreased hcxabarlutal sleepmg-time and in creased liver weight. Shull et al. (1912) studied the induc tive capacLtyof certain Aroclors in mink and ferrets and found that at 20 ppm in the diet, Aroclor 1016 and 1242 produced only weak induction of mixed function oxidases (MFO). Related poly chlorinated compounds such as the polychlorinated naphthalenes induce mixed hepatLC microsomal enzymes in immature male rats, as shown in a study by Cockerline et al. (1981), They Induced both the phenobarbital and methylcholanthrene types of enzymes. Pereira ej a^. (1912) compared single doses of Aroclor 1254 to single doses of diethylnitrosamine (DENA) as inducers of foci in the liver which produce gamma glutamyl transpeptidase and found that Aroclor was a much weaker initiator of those foci than DENA. ' The rate of progression of the effect of single doses of Aroclor 1254 on the different liver forms of cytochrome P-450s (a,b,c,) and epoxide hydrolase (EH) in rat liver mcrosomes was reported by Parkinson et al. (1982) to vary MONS 212915 48 In regard to rata of induction and duration of affact, Ryan et al. (1982) isolated a new htmoprotein from rat liver called P-450* and found that it could be induced by Aroclor 1254 and phenobarbltal, but not by 3methylcholanthrene. Nilsan and Tuftgard (1981) reported that various fonu of cytochrome P-450 were Inducible in rat livers by paraffins. Ryan and Levin (1981) have purified and characterised the spectral peaks of three different cytochrome P-450 ensymes (a.b.c) from rats given Aroclor 1254 Lilienblum at si. (1912) showed that Aroclor 1254 stimu lated glucuronidation systems, that is. the glucuronyl tranferases. GTi and GT2- It was also found that GT^ is induced by methylcholanthrene. whereas GT2 is like the phenobarbltal induced ensyme. In connection with the effects of PCBs on glucuronidase release when polymorphonuclear leucocytes are treated with certain releasers, such as human blood serum plus zymogen, Lee (1981) showed that Aroclor inhibited release of polymor phonuclear glucuronidase. The possible significance of this finding is not clear. A clearly protective mechanism has been shown to result when the microsomal system is induced by the PCBs in Aroclor 1254 since this induction partially protects mice from benseneinduced leucopenia (Wierda et al.. al.. 1981) and Greenlee et al., 1979). In chicks, Hansen et al. (1981) found that Aroclor 1254 was less potent as an HONS 212916 -49- enzyme inducer than was the same amount of Aroclor 1254 in animal Cat obtained from swine which was subsequently fed to the chicks. High-dose levels (i.e.j 9g/kg) of the Aroclor, were required, but the chicks did not show any other toxic effects from the PCBs. The effects of the pesticides mirex, photomirex and kepone were found by Chu et al. (1980) to be additive, but they did not potentiate the induction of microsomal enzymes produced by polybrominated and polychlor inated biphenyls in rats. Ueng and Alvares (1981), Alvares e al. (1982), Alvares (undated workshop report, 1981) and Serab]it-Singh et al. (1982) have shown that in the rabbit lung, microsomal enzyme content as well as cytochrome P-450 were decreased, rather than induced, by administration of Aroclor 1254. The de crease was due to a decrease in P-450j with no alteration in P-450jj. At the doses of the Aroclor used by Alvares et al. (1982) the PCBs were poor inducers of arylhydrocarbon hydroxylase activity in the livers of mice as well as in the kidney Of rabbits, but resulted in suppression of this enzyme in rabbit lung. The Serabjit-Singh et al. study (1982) showed that the effect on the rabbit lung enzyme system was dose related and that 200 mg/kg of Aroclor 1260 in one or two doses produced no change in lung cytochrome P-450. The effect of topically applied Aroclor 1254 on the same enzyme systems in the skin has been investigated by Bickers et al. (1982) who showed induction in whole skin, epidermis and MOHS 212917 -50 dermis of arylhydrocarbon hydroxylase, 0-deethylase and epox ide hydrolase. They emphasized the importance of the skin as an organ for metabolizing xenobiotics. Summary and Conclusions The ability of Aroclors to induce a variety of micro somal enzyme systems in tissues of test animals has been widely documented. The recent results add to the body of information on the specificity of the induction process in terms of the typos of enzyme systems induced and of the specific tissues in which the process occurs. Notable new findings include) (1) the surprising observation that Aroclor 1254 leads to a decrease in microsomal enzyme content of rabbit lung rather than an increase, due to a decrease in P-450 j with no alteration in P-450j j content; (2) the finding of a new induced hemoprotein in rat liver designated as P-450#; (3) the induction of glucuronyl transferases by Aroclor 1254; (4) Aroclor 1254 inhibition of release of polymorphonuclear glucuronidase from leucocytes; and (5) the demonstration that PCB induction in skin leads to an impor tant contribution to metabolism of xenobiotics. HONS 212918 -51 B. Immunologic Effects Chang et al. (1901) followed 30 patients who had "YuCheng" disease, due to lngastlon of rice bran oil contamin ated with PCBs, PCDFs and PCQs, in 1979, and found differen ces in these subjects compared to controls in regard to the serum immunoglobulins. They found that serum IgA and Igl* but not igG globulins were lower in the patients than m controls. Also total T-cells, active T-cells and Tu-c#lls were decreased m the patients but the percentages of acells and T-cells were not different in the patients. Hori at al. (1981) succeeded in simulating the Yusho disease in monkeys by feeding them daily doses of a mixture of 5.0 mg of Kanechlor 400, a polychlorinated dibenzofuran (5 ug) plus a polychlorinated quaterphenyl (5.0 mg). The animals exhi bited the usual signs and symptoms of Yusho, such as edema, liver necrosis. and immune suppression. These results strongly suggest that PCDFs and/or PCQs play a major role in the etiology of Yusho. Rabbits which were fed Aroclor 1243 for 4 weeks prior to mating, during gestation, and nursing of ths young, wert tes ted for immunogenic alterstions (Thomas and Hinsdill, 1900). Mo genaral toxicity was scan in ths mothers. Only the off spring from the mothers fad 250 ppm (ths highsst dose in the study) showed some impaired contact sensitivity but these offspring also showed impaired body weight gains. It was qusstionsd whether the effect observed on contact ssnsitlv- HQNS 212919 52- ity was a direct sffsct rathsr than an indirect affect du to the general action on body weight gain. Sillcworth and Grabstein (1982) showed that PCB-produced lmmunotoxicity was mediated through the Ah (aryl hydrocarbon) gene locus in mice and that only the "planar" structure of the PCBs (planar being 2,3,3',4*- versus non-planar 2,5,2*,5' -isomers of tetrachlorobiphenyl) was active. Summary and Conclusions Continuing work on immunological effects in husians ex posed to PCBs mixed with other chlorinated substances in contaminated rice bran oil (PCBs, PCOPs, PCQs), and in mon keys exposed to synthetic mixtures of these three components, has shown signs of immunosuppression or alteration in immun oglobulins and T-cells. Causality with respect to PCBs alone has yet to be established in these species. Additionally, signs of immunogenic alteration or lmmunotoxicity in rabbits and mice dosed with PCBs have been observed. Despite the growing numbers of studies on this health effect, the body of available evidence does not suggest that PCBs alone can produce a dose related effect on immunological competence in the human, in human PCB workers there is no reported effect on food intake and body weight, an effect seen only at relatively high-dose levels in animals. HONS 212920 -53- XII. Epidemiology K. Skin Problems Lawton at al. (1981) made a ravlaw of mad leal records of workers at several General Electric plants; the record* from a five year period (1970-1978) ware reviewed at which time the employee census varied between 1,800 and 1,800, Skin Irritations and dermatitis were reported among these employees. Medical records for a 15 year period of individuals occupationally exposed to PCBs were reviewed. During this period, 49 individuals presented with "allergic contact derma titis" to PCBs or other constituents of the compounded dielec tric fluid; the dermatitis cleared rapidly, usually without medication, upon removal from exposura, No cases of chloracne were observed, Lawton at al. (1981) also axammed the dermatological history of 194 PCB-exposad capacitor workers; 75 employees had 100 dermatological findings of various types in thair madical history, Many of the tabulated cases had dermatolog ical findings that would not be considered work-related, Inca employees recognised the contact nature of most rashes (fiberglass, dirty glovas, watch straps, etc.). Two cases of rash ware related to epoxy exposure and not to dielectric fluid. Physics! examination of these employees was unramark- MQNS 212921 -54- able and no signs of chloracns past or prasant wera discov erad. Pigmentation, thickening of tha skin or discoloration of tha fingarnails wara not notad. Many of tha dermetological apisodas wara noted to antedate employment at General Electric. Most employees had multiple industrial exposures arising from job mobility (maintenance man, use of other chemicals such as solvents, etc.). Tha authors conclude that it was difficult to relate any dermatological findings specifically to PCB exposure. B. Lipid Metabolism The recent epidemiologic study of Lawton at al. (1981) also focused on workers with elevated serum PCBs, The average serum levels weret Aroclor 1242, 278 ppbj Aroclor 1254, 58 ppbj Aroclor 1260, 34 ppb. The mean serum triglyceride level for 172 PCB-exposed workers was 144.5 g/dl, which is within the normal range of 50*200 mg/dl, but the upper level of 2 SD was above the normal range. Detailed statistical analysis demonstrated! 1. A significant positive association between log serum triglycerides and log serum PCB levels, either as Aroclor 1242, Aroclor 1260 or total PCB; 2. When PCB and DOB levels were considered as MOWS 212922 55- distrlbuted In a serum lipid fraction (sum of triglycerides end cholesterol), the log serum triglyceride level was not signifLcantly dependent on log serum PCBs plus DDE) 3, When PCBs and DDE were considered as body burden, the log serum triglyceride level was significantly dependent on burden only when the percent body fat was removed as an indepen dent variable in the case of Aroclor 1242r 4. That the log serum triglycerides were positively related to percent body fat but negatively related to sex. Following discontinuetion of the use of PCBs at the '.eneral Electric plants on 6/30/77, Lawton at al. (1901) found that the serum cholesterol level for each sex of PCBtxposed workers showed a statistically significant decline :hrough 1979. The 1979 determinations showed that the range 2SD) in the exposed group was above the upper limit for .he standard range. The serum cholesterol had a significant >ositive association with log serum PCBs but not when stated is PCB in adipose tissue or PCB body burdens. The serum tholesterol level was service and/or age dependent. C. General Complaints t Symptomatology it should be recognised that any survey data on sympto- MONS 212923 -56- matology are subjective in nature and that the "normal" background incidences of various symptoms can vary widely in different studies, depending in part on how questions are asked and on the type of questions used. Thus, symptomatol ogy reported by workers Is the least reliable of the data available for analysis. Fischbtln et al. (1979), studying volunteers from Gen eral Electric plants, did not find a significant correlation between reported symptoms and plasma PCS level or between symptoms and duration of employment. Lawton at al, (1981), studying symptoms reported by 194 PCS exposed capacitor workers at General Electric plants, found reported symptoms (headache, dizziness, depression, memory loss, fatigue, ner vousness, sleeplessness, somnolence) to be less frequent than reported by Fischbein at al. (1979). The analysis of Lawton at al, (1981) suggests that the symptoms reported can be reasonably associated with intercurrent disease and/or the general situations of life stress. Smith at al. (1981), summarizing data from combined work sites, reported that coughing at work and eye Irritation were related to both L-PCB and H-PCB levels in serum whereas the symptoms of loss of appetite and peripheral "tingling" were associated only with L-PCB. Many of these differences were not observed when individual work site data were anal yzed separately; the lack of a difference in each site may HONS 212924 -57- result because jobs were Included as a confounder variable. Most studies have not reported any major symptoms associated with exposure to PCBs (Smith et al., 1981; Kreiss et u. 1981). Except for Yusho disease, which Is a special situation (see Section II), and an occasional report of chloracne (section XII), there has been no reported clinical disease in PCB-exposed populations (Xarppanen and Kolho, 1972; Humph rey, 1980; smith et al. 1981; Chase et aK 1981;' Kreiss et al. 1981; Balter et al., 1980; Lawton et a^., 1981). The common but not consistent findings of abnormalities of lipid metabolism, especially blood triglyceride levels, and minor abnormalities of liver enzysies, especially SGOT and GGTP, were not associated with symptoms or clinical disease. 0, Pulmonary Function The work of Marshaw t al, (1979) of pulmonary function in workers in a capacitor plant, has bssn reviewed previously, in contrast, the study of Lawton et al, (1981) did not demonstrate that FCB exposure, as measured by serum PCB levels, is correlated with spirometnc abnormalities. Their data are calculated on the basis of Knudson et al. (1976 ), using a normal range to >70% for both vital capacity and PEVj^/FVC as recommended by the American Thoracic society. Abnormal findings occurred primarily in smokers and ex-smokers, HONS 212925 58 as ahown in Table II. E. fasting Blood Glucose Lavala In the epidemiologic study of capacitor workers (Lawton et l., 1981), the mean fasting blood glucose level and range for PCS exposed workers were within normal limits. Compari son of 1976 and 1979 values showed an apparent decline in 1979 that was statistically significant. It was also obser ved that fasting blood glucose levels increased with age, which is in agreement with the literature. Glucosuria was not observed. P. Uric Acid Levels Normal serum uric acid levels were obtained in PCBexposed workers (Lawton et al, 1981), Summary and Conclusions The epidemiological review of medical records of General Electric workers by Lawton et al, (1961), focused on employees with records of PCB exposure in electrical capacitor manufacture provides additional clinical data to assess the possible hasards of PCBs. For workers occupationally exposed to PCBs over many years, it is important to note that dermato logical findings were relatively minor, physical examinations were unremarkable, and no signs of chloracne past or present MONS 212926 -59- were discovered. Soiee relationships indicative of altered triglycerides in the circulating serum, were observed; they were not associated with symptoms of clinical disease. Other reported symptoms, e.g. headache, dissiness and depres sion are reasonably associated with general life stress, not PCB exposure. Complaints of coughing and eye irritation have been reported by Smith et al. (1981) as related to serum PCB levels, but this has not been reported by other authors. The study by Lawton demonstrates that extensive exposure to PCBs in the occupational setting, sufficient to have established blood levels in excess of 200 ppb, has not led to the production of clinical disease states in General Elec tric capacitor writers. HONS 212927 TABLE II S9* Prevalence of spirometric abnormalities In 107 PCB-exposed capacitor workers. Criteria for abnormality of FEVi/FVC given as percent of predicted using standards of Rnudson et al. (<70% compared to the older standard (<75%) used by Warshav et jl.l. rvc* donsmoke rs 0/41 MALE Current i Ex-Smokers A) 1 Non Smokers 5/105 (4,8%) 5/146 (3.4%) 1/15 (7.0%) FE HALF. Current t Ex-Smokers 4/26 (15.4%) All ' 5/41 ( 12.2%) FEV| * 0/41 rEVj/ FVC* 0/41 11/105 (10.5%) 11/146 (7.5%) 0/15 21/105 (20.0%) 21/146 (14.4%) 1/15 (7.0%) 8/26 (30.8%) 8/41 (19.5% 11/26 (42.3%) 12/41 (29.3%) 12/41 (29,3%) 36/105 (34.3%) 48/146 (32.9%) 4/15 (26.7%) 20/26 (76.9%) 24/41 (58.5%) < 79.5% + < 70,0% + < 75,0% HONS 212928 -60- XIII. Recent Findings on Health Effects of Polychlorinated Dlbenzofurn~ 1. Introduction with increasing realization that the relatively po tent polychlorinated dlbenzofurans {PCDFs) occur as trace constituents in commercial PCB mixtures, and are also formed in small yields at temperatures above 270 *c (Choudhry and Hutzinger, 1982) by oxidatlon/cyclUation of suitably substituted PCB congeners, attention has turned to evaluation of the contributions that these tricyclic compounds may make to the toxicity syndrome elicited by exposure to PCB/PCDF mixtures or to individ ual PCB congeners. A workshop on chlorinated aromatic heterocyclic com pounds was held in Rome in October, 1980. The papers presented at that workshop represent summaries of impor tant studies performed to that date and form a useful framework for integration with work on the PCDFs that has appeared in the past two years. Specifically, major workshop papers included! Cl) a presentation (Jansson and Sundstrom, 1980) which described the formation of PCDFs during a fire in electrical capacitors containing PCBs; and (2) a paper by Me Hulty et al. (1980) which presented results on the chronic toxicity of 2,3,7,8-tetrachlorodibensofuran in Rhesus macaques. HONS 212929 -61 Since the biological potencies of PCDFs are of the order of 102 - 10* greater than the toxicities of corresponding substi tu ted PCBs with respect to produc tion of such health effects as chloracne and P-448 oxi dase induction it is clearly important to sort out the contributions to total toxicity patterns made by PCDFs even when they are minor constituents (at a few ppm) in PCB mixtures. Indeed) Kashimoto et al^. (1981) have already high-lighted this point in their analysis of the progress of the Yusho disease syndrome in Japanese patients. These patients were exposed to contaminated rice oil in which the PCDF content was only 5 ppm (on a rice oil basis) but may have been almost entirely respon sible for the persistence of chloracne andskin pigmenta tion over many years in some of the more severely affected patients. Therefore, the following sections are devoted to a brief review of some recent advances toward understanding of the structure vs. activity rela tionships for this important class of halogenated poly cyclic compounds. 2. Animal Studies Recently. Hori et al. ( 1982) reported on an inter esting study in which the PCB and PCDF fractions Isola ted from a batch of the original Japanese Yusho rice oil were administered orally as the PCB fraction alone (15.4 ug/mouse/day) or mixed PCB PCDF fractions (15.4 ug + HONS 212930 -62- 1.0 ug/mouse/day) for seven days to mala ICR itrain mica, followed by analysis of tha ratantion of individual PCB congeners in livar and adiposa tissua for tha naxt sO days. Overall, rasults wars assantially tha same for tha two experimental groups in tha following raspactsi (1) by day 50 in both livar and adiposa tissua tha relativa contant of hexa-and hapta-CBs ratainad was naarly 90 parcant of tha PCBs in thosa tissues, probably bacausa of matabolic allmination of lowar chlorinatad congenersj and (2) rasidual amounts of PCBs in tissua with tha mats and para positions unsubstitutad by Cl atoms (i.e. tha 3.4-, 4,5-, 3',4'-, or 4*, 5'-posltlons) wara small, prasumably bacausa of low original concantrations and raady matabolism via vicinal unsubstitutad ring posi tions. Tha principal affact of tha PCDPs on PCB ratan tion in tissuas was to lowar tha half-livas of thraa marker PCB conganars in aach tissua by a factor of about two. This affact was prasumad to originata from salactiva matabolic activation/alimination of thasa conganars by P-44B oxidasas inducad by tha PCDPs. Yoahihara at al, (1981) hava raportad on an impor tant study of tha acuta actions of 13 diffarent PCDF conganars in mala wistar rats, with singla lntraperitonaal injaetions of conganars in corn oil vahlcls at lavals naar 1 mg/kg leading to onsyme induction of tha 3-methylcholanthrana (P-448) typa and to variable accumulations HONS 212931 63- ln liver tissue, A major finding of the study concerned the controlling influence of ring substitution by Cl atoms on the course of 3-MC type enzyme induction in liver tissue; all 9 PCDF congeners having at least 3 cl atoms in the critical lateral positions (2,3,7 ,| on the PCDF rings) exhibited typical 3-MC type induction. The most potent congeners for this effect, active at a single dose near 1 mg/kg, were those with the lateral positions completely filled, i.e. both 2,3,7,8 - tetra CDF and 2,3,4,7,8 - penta CDF. Congeners with two or less Cl atoms in the lateral positions did not show any enzyme inductive effect at a single dose as high as 5-10 mg/kg. in this study, the PCDF congeners active as 3-MC type inducers also produced marked atrophy of the thymus and hypertrophy of the liver in rats; congeners inactive as inducers failed to elicit these effects on thymus and liver. The ability of congeners to concentrate in liver and produce acute hepatotoxicity turned out to be quite variable and to bear no clear relationship to PCDF struc ture. The comparative acute and subchronic toxicities of two of these highly substituted PCDFs with lateral posi tions completely filled, the 2,3,7,8 - tetra CDF and 2,3,4,7,8 penta CDF congeners, were studied earlier by MOHS 212932 -64- Moor* *t *1. (1979) in single strains of mice and guinea pig* end also in the Rhesus monkey, Ml animal* ware dosed by gastric intubation in corn oil, and showed toxic signs ranging through the sett body weight loss, decrease in body weight gain, thymic atrophy, liver hypertrophy, anemia, edema of face and lips, loss of facial and other hair, loss of fat, hyperkeratosis, squamous metaplasia, etc. Curiously, both the guinea pig and th* monkey were quit* senaitiv* to the acute toxicities of these PCDFs 5-10 ug/kg (12-15 days) and 1000 ug/kg, respectively, but th* (57Bl/6fh(J67) strain of mouse was relatively insensitive to acute toxic actions of these PCDFs, showing no clinical signs of toxicity ov#r 30 days at doses up to 6600 ug/kg. These acute lethality data could also be compared with acute data on th* correspondingly substituted dibenzo-p-dioxin, th* 2,3,7,8 - tetra CDD congener, to give an estimate of th* relative lethal potencies of fully laterally substi tuted PCDFs vs. PCDDs. In the guinea pig th* PCDFs were only 2-4 fold weaker than th* tetra CDD; in th* monkey th* factor by which the PCDFs was weaker was about 20 fold; and in th* resistant mouse the factor difference in 30 day LD50s was approximately 2*3 fold. These differ ences amount to one to two orders of magnitude in lethal strength of fully laterally substituted PCDDs over PCDFs. Th* PCDFs are not unconuson contaminants in th* eco- MONS 212933 65- sphere. They arc formed in the course of synthesis of polychlorinated phenols. Rappe et al. (1979) have noted that they are found in fly ash and flue gases, resulting from pyrolysis of PCB residues by at least three cyclization reactions involving either loss of two ortho Cl atoms, loss of ortho H and Cl atoms, or migration of cl atoms on rings followed by cyclixation. Cavallaro at al. (1980) have also measured the amounts of PCDF con geners in effluents from incinerated solid urban wastes. Rappe et al. (1961) have found many PCDF congeners stored in the fat of animal species as widely separated as the snapping turtle (U.S.) and the grey seal (Sweden), using mass fragmentograms as the detection/identification tool. PCDF congeners have been studied in recent pharmaco kinetic and toxicologic experiments with a variety of animal models. Decad et al. (1981) studied the distribu tion and excretion of the fully laterally substituted 2,3,7,8 - tetra CDF in two strains of mice, with the findings that the liver was the major site of initial accumulation, redistribution to adipose tissue then occurred, 55-80 percent was excreted as polar metabolites in the feces within 10 days, and the whole body half life for total elimination of parent compound and metabo lites was of the order of 2-4 days. Veerkamp et, al. (1981) also studied PCDF metabolism, using the Wistar rat. These workers found that monohydroxy and dihydroxy HONS 212934 -66- darivativas ware in various tissuas and axcrata (fat, livar, urina, facas), involving, at laast in part, the occurranca of cl atom shifts plus hydroxylation, Lowar dagraas of PCDF chlorination lad to a variaty of hydroxylatad darivativas, plus soma containing tha -SH sulfhydryl group isolatad as tha mathylthloathar, but tha highar PCDFs (a.g. tha fully subatitutad octa CDF) rasistad mataboliss. Much unchangad PCDf was axcratad in tha facas, prasumably bafora intastlnal raabsorption had occurrad. Me Nulty at si. (1981) axtandad tha study of tha actions of tha potant conganar 2,3,7,8 - tatra CDF to Rhasus monkays. Dosing in tha food at 50 ug/kg/day for 2 months or 5 ug/kg/day for $ months yialdad savara signs of toxicity (and aarly daaths) in thasa primatas, including raduction in physical activity, savara skin changas, swollan ayalids and anlargamant of tha maibomian glands, hyparkaratonic nail bad, atrophy or squamous mataplasla of sabacaous glands, mucous mataplasia and hyparplasia of tha gastric mucosa, involution of tha thymus and hypoplasia of the bona marrow. In animals that survived, racovary was apparantly complata aftar 3 months on control diat. Bacausa of tha biological potancias of suitably subs* titutad PCDFs, tha ability to datact and quantitata small amounts of thasa compounds in tissuas and othar samplas bacomas vary important. Hutsingar at a^. (1981) HONS 212935 7- heve reported on a bioanalytical approach to this problam which makes usa of an initial stap involving reversibla binding to a spscific racaptor protain of tha hepa tic cytosol of rodants or in cultured mammalian calls. Essentially, this technique measures the ability of a compound in a sample to displace radiolabeled 2,3,7,8TCDD from its binding to the receptor system, and quanti tates the molar amount of that compound in reference to that concentration of unlabeled TCDD which can effect half-displacement of the label. Prom extracts of air samples containing PCDPs or PCDDs in the ng/m3 concentra tion range, this method is able to afford estimates of the amounts of tetra-octa chloro-substituted congeners in either family of compounds. k recent report (Silkworth et al., 1982) on a 1981 fire in the State Office Building in Binghamton, New York involved fire products from a transformer containing as dielectric fluid 8S percent of a PCB mixture and 35 percent of chlorinated benxenes. The fire resulted in release of large quantities of soot contaminated with PCBs, PCDDs and PCDPs. Congeners present included the fully laterally substituted 2,3,7,8-TCDD (two samplesi 2.8, 2.9 ppm) and 2,3,7,8-TCDP (272, 124 ppm). Both the soot and equivalent sample of its benxene extract were tested for acute oral toxicity and lethality in guinea pigs and dermal toxicity in rabbits. In guinea pigs, MOttS 212936 -68- the spectrum of toxic signs expected for PCDD-PCDF mixtures (weight decreases* effect on thymus and kidney* hyperpla sia and metaplasia* triglyceride elevation, etc.) was about equivalent for either carrier mode* soot vs. extract, as were the respective LD50 values (410 mg soot/kg * 327 mg soot equivalent/kg), in rabbits* dermal application of the soot or the extract resulted in liver pathology, but local akin serous inflammation occurred only with the benzene extract* Indicating that the soot matrix prevented local inflammatory reaction from particlebound PCDOs and PCDFs. Finally* in a comprehensive recent review, Parkin son and Safe (1981) have attempted correlations of PCDF toxicities with aryl hydrocarbon hydroxylase (AHK) induc tion, as a key biological action of these chemicals. The animal toxicities sort into the following major health categoriesi (1) wasting syndrome* involving pro gressive weight loss that is not simply related to de creased food consumption; (2) a variety of akin disor ders including chloracne, alopecia, etc.; (3) lymphoid involution* with thymic and splenic atrophy and immunosup pression; porphyria* involving an overproduction of uroporphyrinogen; (5) endocrine and reproduction dysfunc tion; (6) teratogenesia; (7) carcinogenesis. These auth ors tested correlations between specific health effects MOMS 212937 -69- in these categories and the ability to induce AHH, reaaoning from the baaia that corresponding atructure vs. activity relationships in each chemical class should Influence or govern toxic potency for a given effect, receptor binding affinity to the specific cytosol recep tor, and the ability to induce P-448 requiring AHH. Taking account of age dependent, aex-dependent and organ-specific differences in susceptibility seen in different teat species and atrains, and effects dependent on the presence or absence of a genetic Ah locua control ling AHH synthesis, some positive correlations were in deed noted. Good correspondence was found between swlecular structure lending to AHH induction and that producing significant binding affinity to the cytosol receptor. However, it was noted that AHH induction stay correlate with one toxic effect but not another aa atruc ture is varied, and that in some cases the relationship between AHH induction and toxicity is not established unequivocally. No compelling evidence was found that AHH per e . mediates the lethal effects of these com pounds, even though AHH induction correlates with toxi city for certain spacific health effects. Parkinson and Safe also pointed out that other factors impinge on the ability to sort out clear correlations between AHH induc tion and specific toxic affects. For example, these polyhalogenated compounds alter the vitamin status of HONS 212936 -70- animal nodal* (vitamin A, ate.), which may in itatlf hava great impact on biological raaponaa patterns, and iron daficiancy in a taat animal modal may b* abla to nagata or perturb the ability of a compound to initiate porphyrl** 3. Human Observations No new information bearing on PCpr-inducad health affect* in exposed human population* ha* appeared in the past two year*, aside from the increasing emphasis placed on potential causation of persistent Yueho disease symp toms in Japanese and Taiwanese patients by these chemi cals. It is possible that FCDF effects such as ehloracne and ensyme induction might also occur in other sufficient ly exposed human populations, but detection would depend on being able to sort out the specific contributions of these agents alone from the total spectrum of chemical mixtures (e.g. FCB* and chlorinatad phenols> in whleh they are generally encountered aa trar* ontnin*nt*. To date, the -identification of unique FCDF effects in humans has not baen documented. Summary and Concluaiona Tha FCDFa hava bean atudiad with incraaaing intareat in recent yeara as raalisation grows that they are impor tant contributors to tha toxicity of mixtures in which they may be present at only tiny mole fraction!. In HONS 212939 P P recent animal studies they have been shown to be powerful AHH Inducers, especially when the lateral 2,3,7,8-posi tions are fully substituted by chlorine atoms. They produce many toxic signs in test animal systems including severe skin changes, swollen eyelids, enlargement of meibomian glands, squamous metaplasia, hyperplasia of the gastric mucosa, thymic involution, hypoplasia of the bone marrow, and hepatotoxicity. They undergo metabolic hydroxylation and excretion when the pattern of chlorine substitution is suitable, with major excretion of polar metabolites in the feces. Aside from continued analysis of residual health effects in Yusho patients, in which the opinion is growing that PCDFs are major contributors to the persistent toxic syndrome observed, little in the way of human PCOF exposure vs. effect information has been documented. HONS ZIZ9W -72- RE F REN CSS ALLEN, J.R., ABRAHAICCN, L.J., andNORBACK, D.H. (1973), Biological Affects of polychlorinated biphenyls end triphenyls on the subhuman primate. Environ. Res. 6*344-354. ALVARES, A.P., UENG, T.H. and EISEMAN, J.L. (1982), Polychlorinated Biphenyls (PCBs) Inducible Mono-oxygenases in Rabbits and nice: Species and Organ Specificities. Life Sciences. 30*747-751, ASHBY, J., TRUEMAN, R.W., STYLES, J., at al. (1981). 4-Chloromethylbiphenyl (4-CfB): a novel mutagen an3 potential carcinogen. Cardnoqenesla 2(l)i33-36. BAKER. E.L., LAM ORIGAN, P.J., GLUECK, C.J., ZACK, M.M., et al. (1980). Metabolic consequences of exposure to polychlorinated biphenyls (PCB) in sewage sludge. Am. J. Eoldem. 112*553-563. BECKER, G.H. at al. (1979). Polychlorinated biphenyl-induced morphologic chengea in the gaatrlc mucosa of the rhesus monkey. Laboratory Investigation 40i373-383. BICKEL, M.H. and MUEHLEBACH, S. (1980), Pharmacokinetics and ecodisposition of polyhaloganated hydcocacbonsi aspects and concapta. Drug Metabolism Reviews 11(2)il49-190. BICKERS, D.R., DUTTA-CHOUDKURY, T., MUKHTAR, H. ( 1982). Studies on Cytochroms P-450 contsnt and mixed-function oxidase and epoxida hydrolase activity, molecular Pharm. 21j239-247, BIOCCA, M. , GUPTA, B.M. and CHAE, K. , at al. (1981). Toxicity of selected symmetrical hexachlorobiphenyT-isomers in the mouse. Toxicol. Applied Pharmacol. 58*461-474. BOWMAN, R.E., HEIROMIMUS, M.P., and ALLEN, J.R. (1978). Correlation of PCB body burden with behavioral toxicology in monkeys. Pharmacol. Btochem. Behavior 9*49-56. BOWMAN, R.B., and HEIROMIMUS, M.P. (1981). Locomotor hyperactivity In PCB-expoaed rhesus monkeys. Neurotoxlcol. 2(2)*251-268. BOWMAN, R.E., and HEIRONIHJS, M.P. (1981). Hypoactlvlty in adoles cent monkeys perinatally exposed to PCBa -and hyperactive aa juveniles. Neurobthavloral Tox. Teratol. 3*15-18. CALAN ORA, J.C., ( 1975). Summary of toxicological studies on commercial PCBs. Proc. of the National Confer, on Polychlorinated Bjphenvla. Nov. 19-21, Chicago, IL. CAVALLARO, A., BAN 01, G, , INVESNIIZI, G., e el. (1980). Sampling occurrence and evaluation of PCDDa and PCDFs Trom incinerated solid urban waste. Chsmoaphere 9^(611-621. MONS 212941 -73- CHANG, K., HSIEH. K., LEE. T., fit 1. (1980). Immunologic evaluation of patients with PCB poiaoningi Determination of lymphocyte sub populations. Toxicol Applied Pharmacol. 61(58-63. PHASE, K.H., W0NG,0., THOMAS, D. , BEFNEY, B.H., and SIWN, R.K. (1981;. Clinical and metabolic abnormalities in occupational exposure to polychlorinated biphenyls. Presented at the American Occupational Health Conference, Atlanta, Georgia. CHEN, P.H., CHANG, K.T., and LU, Y.o. (1981). Polychlorinated biphenyls and polychlorinated dibenzofurans in the toxic rice-bran oil that caused PCB poisoning in Taichung. Bull. Environ. Contam. Toxicol. 26i489-495 . CHOUDHRY, G.G., and HUTZINGER, 0. (1982). Mechanistic aspects of the thermal formation of halogenated organic compounds including polychlorinated dibenzo-dioxins. Part IIi Thermo-chemical generation and destruction of dibenzofurans and (Jioenzo-p-dioxins. Toxicol. Environ. Chem. 5t67-93. CHU, I., VILLENEUVE, D.C., BECKING, G.C., IVERSON, F., and RITTER, L. (1980). Short-term study of the combined effects of mirex, photomi rex, and kepone with halogenated biphenyls in rats. Jour. Tox. Environ. Health 6*421-432. COCKERLINE, R., SHILLING, M., and SAFE, S. (1981). Polychlorinated naphthalenes as hepatic microsomal enzyme inducers in the immature male rat. Gen. Pharm. 12*83-87. DECAD, G.M., BIFNBAUM, L.S., and MATTHEWS, H.B. (1981). Distribution and excretion of 2,3,7,8-tetrachlorodibenzofuran in C57BL/6J and DBA/2J mice. Toxicol. Applied. Pharmacol. 59*564-573 . FISCKBEIN, A., WOLFF, K.S., LILIS, R. , THOFNTCN, J. and SELIKOFF, I. (1979). Clinical findings among PCB-exposed capacitor manufacturing workers. Ann. N.Y. Acad. Sciences 320*703-715. FISCHBEIN, A. WOLFF, M.S., BE FN STEIN, J., SELIKOFF, I. and THOFNTON, J. (1982). Dermatological findings in capacitor manufacturing workers exposed to dielectric fluids containing poly chlorinated biphenyls (PCBs). Arch, of Environ. Health. Vol.37 (2)*69-74. GARTHOFF, L.H., CERRA, F.E. and MARKS, E.P. (1981). Blood chemistry alterations in rats after single and multiple gavage administration of polychlorinated biphenyl. Toxicol. Applied Pharm. 60*33-44. GEISTFELD, J.G., BOND, M.G., BULLOCK, B.C. and VARIAN, M.C. (1982). Mucinous gastric hyperplasia in a colony of rhesus monkeys (Macaca mulatta) induced by polychlorinated biphenyl (Aroclor 1254). Laboratory Animal Science 32(1)*83-86. MONS 212942 -74- GREENLEE , W.F., DENT, J.G. and IRONS, R.D. (1979). Effact of pratraatmant with Aroclor 1254 and aalactad polychlorinatad biphanyl isomars on banzana-inducad myalotoxicity In tha rat. Acadany procaadlnqa and Abatracta, John gllaha Hltchall Scl. soc. 95(2)7119. ----- HALPAAP-WOOD, K., HOMING, E.C. and HOMING, M.G. (1981). Tha affact of 3-mathylcholanthrana, Aroclor 1254, and phanobarbltal Induction on tha matsbolism of biphanyl by rat and sauaa 9000g auparnatant llvar fractions. Drug Watab. and felapoalt. 9(2)1103-107. HANSEN, L.G. and STRIK, J.J., at al. (1981), Biological activity of tachnlcal Atoclor 1254 comparaato Aroclor 1254 rtsiduast Swlna fat rasiduaa fad to broilar cockaraia. Toxicol. 21_t203-212. HAY, A.W.M. (1980). Expoiura to TCDDi Tha Haalth Risks. In Chlorlnatad Dioxins and Ralatad Compounds. Ed. by 0. Hutxingar, a t. a 11, vSt 5fli-locT, 19807 N . Y . pargamon Prass, 1982. HEIRONIKUS, H.P., LAUGHLIN, N.R., snd BOWMAN, R.E. (1981). Effacts of aarly axpoaura to FCBa on laarnad irralavancy of cuaa in rhasua monkaysi An lncidantal laarning paradigm. TaratolooV 24(1)t55A. HORI, S., OBANA, H., and KASHIMOTO, T., sj al. (1981). Biological affacts of PCBa and ralatad polychlorinatad compounds on Croo monkaya. ` J. Pharmacobla Dyn. 14iS-65. HORI, S., KASHItDTO, T., and KUNITA, N. (1982). Effact of poly chlorinatad dibanzo-furan on tha ratantlon of polychlorinatad biphanyl isoisara in tha livar and adipoaa tiaaua of mica. J. Food Kvc. Soc. Japan 23(2)il67-l75. , HU ((PH REY, H. (1980). Evaluation of humana axpoaad to hsloganatad biphanyla. Aiaar. Chaw. Soc. Div. of Environ. Cham. 201 (2). HUTZINGER, 0., OLIE, K., and LUSTENHOUWER, J.W.A. (1981). Poly chlorinatad dibanzo-p-dtoxina and dlbanzofuranat a bio-analytical approach. Chamoaphara 10>19-25. IATROPOCL08, M.J., FELT, R., at Si. (1977). Hlatopathology of low chlorlnatad biphanyla and haxachlorobanzana in famala rhasua raonkaya. Toxicol. APPllad Pharmacol. 41tl73,l00. INOUE, Y., at al. (1975), Dlacovary of PCB pollution in a taxtlia factory. I. PcB laval in blood aarum of laborara and raaults of physical axamination, Jen, j. Public Haalth 22i461-463, (As raportad in NXOSH Critaria Documant). JANSSON, B. and SUNDSTROM, G. (1980). Formation of polychlorinatad dibanzofurana (PCDF) during a fira accldant in capacitors containing polychlorinatad biphanyla (PCB). In Chlorlnatad Dloxlna and Rtlata^ Compounda. Bd. by o. Hutzlnaar. at al.. v5i20l-20B", l8u. n,Y.Pargamon Pcaaa, 1982. HONS 2129** 75- JENStN S. and SUNDSTROM, G. (1974 ). Structures and levels of moat chlorobiphenyls In two technical PCS products and in human adipose tissue. Amblo 3 * ( 2) 70-74. KMtPPAHES , E. and KOLHO, I>. (1972). The Concentration of PCB In human blood and adipose tissue In three different research groups. PCB conference II, Stockholm. KASHIMDTO, T., at al. (1981). Role of polychlorinated dibenxofuran In YushO (PCB poisoning), Arch. Environ. Health. 3i(6)*321-364. KIMBROUGH, R.D., et al. (1975), Induction of liver tumors in Sherman strain female rats by polychlorinated biphenyl (Aroclor 1260). J. Natl. Cancer Inst. 55*1453-1459. KNUDSON, R.J., SLATIN, R.C., LEBOWITZ, M.D., and BURROWO, B. (1976). The maximum expiratory flow volume curve* normal standards, variabi lity, and effects of age. Am. Rev. Reep. Pis. 113*587. KREISS, K., et el. (1981). Association of blood prsssure and poly chlorinated biphenyl levels. JAMA. 24S(24)*2505-2509. KUROKI, H., and MASUDA, Y. (1977), Structures end concentrations of the main components of polychlorinatad biphenyls retained in patients with yusho, Chemosohere, 8*469-474. LAWTON, R.W., SACK, B.T., ROSS, M.R. snd FEINGOLD, J. (1981). Stud ies of employets occupationally exposad to PCBs. General Electric Procress Report. LEE, Tee-Ping, (1981). The inhibition of PHI B-glucuronidase release by Aroclor 1254. Environ. Res. 25*386-390. LILXENBLUM, W., WALLI, A.K., AMD BOCK, K.W., (1982). Dlfftrsntial induction of rst liver microsomal UDP-glucuronosyltransferase activltiaa by various inducing, aganta. Blochtm. Pharm.31(6)*907-193. KARON I, M., 4$, i,. (1981). Occupational axpoaure to polychlorinated biphenyls in slectricel workers. I., Environmental end blood polychlorinated biphenyls concentrations. British of Indust. Mad. 28*49-54. MARONt, M., at, al. (1981). Occupational exposure to polychlorinated biphenyls in electrical workers. II., Health Effects. British Indus. Med. 38*35-60. flASUDA, Y., KUROKI. H., YAMARYO, T., HARAGUCHI, K., et el. (1982). Comparison of csussl sgents in Tsiwen and Fukuoks PCB poisonings. Chemosphere. 11.(2)* 199-206. MATTHEWS, 11.B. and ANDERSON, M.W. (1975). Effect of ehlorination on the distribution snd exerstion of polychlorinated biphenyls. Drug Me tab. Plan. 3,1 * 5 ^ MOWS 212944 -76 MATTHEWS, H. B. and TUEY, D.B. (1980). The effect of chlorine position on the distribution end excretion of four hexachlorobipheny1 isomers. Toxicol. Appl. Pharma. 52:377-388. MCCONNELL, E.F., and MDORE, J.A., (1979). Toxicopathology charac teristics of the halogenated aromatics. Annals. N.Y. Acad. Scien., 320i138-150 . ------ -------- MCNULTY, W.P., POWRANTZ, 1.8. and FARRELL, T.J. (1981). Chronic toxicity of 2,3,7,8-tetrachlorodibensofuran for rhesus macaques. MCNULTY, W.P., POME RAN TZ, I.M. and TARRELL, T.J. (1980). Chronic toxicity of 2,3,7,8-tetrachlorodibenxofuran for rhesus macaques. In Chlorinated Dioxins and Related Compounds. Ed. by 0. Hutzinger, et.ai., 511 ll-4~nn n7y. Pergamon Press, 1982. MX) RE, J.A., MCCONNELL, E.F.,, DALGARD, D.W. and HARRIS, M.W. (1979). Comparative toxicity of three halogenated dibenzofurans in guinea pigs, mice and rhesus monkeys. Annals N.Y. Acad. Scl. 320>151-163. fORCAN, R.W., WARD, J.M. and HARTMAN, P.E, (1901), Aroclor 1254- induced intestinal metaplasia and adenocarcinoma in the glandular stomach of F-344 rats. Cancer Res. 41;5052-5059. NAGAYAMA, J., KURATSUNE, M. and MASUDA, Y. (1901). Formation of polychlorinated dibenzofurans by heating polychlorinated biphenyls. Fukuoka Acta Med. 66i$93-599. NATIONAL CANCER INSTITUTE, (1978). Biosassay of AROCLOR 1254 for possible carcinogenicity. Technical Report Series. No.38. NEBERTS, DANIEL, w., (1981). Genetic mechanisms controlling the induction of polysubstrate mono-oxgenase (P-450) activities. Ann. Rev. Pharm. Tox.. 21)431-462. NILSEN, O.G. and TOFTGARD, R. (1981). Effects of polychlorinated terphenyls and paraffins on rat liver microsomal cytochrome P-450 and in vitro sietabolic activities. Arch. Toxicol. 4711-11. NORBACK, D.H., CIHLA, H.P. and REDDY, G. (1981). In vitro measur ement of PCI toxicity. Toxicol. Halogenated Hydrocarbons) Hlth. Ecol. Effect (Pap. Svisp. > 1^3-186. OUW, H.K., SItVSCN, G.R. and SIYALI, D.S. (1976). Use and health effects of Aroclor 1242, a polychlorinated biphenyl, in an electr ical industry. Archives Environ. Health 31)189-194. PARKINSON, A. and SAFE, S. (1981). Aryl hydrocarbon hydroxylase induction and its relationship to the toxicity of halogenated aryl hydrocarbons. Toxicol. Environ, chem. Rev. 4:1-45. PARKINSON, A., THOMAS, P.E., and RYAN, D.E., et al. (1982). Time course of induction by Aroclor 1254 of three forms of cytochrome P-450 and epoxide hydrolase (EH0 in ret liver microtomes. Fed. Proc. 41 (4)t1299. HONS 2129*5 -77 PEREIRA, M.A., HERRIN, S.L., BRITT, A.L. and KHOURY, M. M. (1982). Promotion by polychlorinated biphenyls of enzyme-altered foci in ret liver. Cancer Letters 151185-190 . PTAND, ALAN and KNUTSON, JOYCE, C., (1982). 2,3,7 ,8-tetrachlorodibento-p-dioxln and related halogenated aromatic hydrocarbons. Ann. Rev. Pharm. Ton. 22*517-554. PRESTON, B.D., VAN MILLER, J.P., MOORE, R.H. and ALLEN, J.R. (1981), Promoting effects of polychlorinated biphenyls (Aroclor 1254) and polychlorinated dibenzofuran-free Aroclor 1254 on diethylnitrosamine-induced tumorigeneais in the rat. J. Mat. Cancer Inst. 66(3)i509-515. RAPPE, C. , BUSER, H.R. 6 BOSSHARDT, H-P. (1979), Dioxins, dibenzo furans and other polyhalogenated aromatics* production, use, for mation and destruction. Annals N.Y. Acad. Scl. 320il-18. RAPPE, C., NYC REM, M., BUSER, H.R. and KAUPPINEN, T. (1981). Occupational exposure to polychlorinated dioxins and dibenzofurans. In Chlorinated Dioxins and Related compounds. Ed. by O. Hutzinger, et a~l7 v51495-51X1 mTy. Pergamon Press, 1982, RYAN, D.E. and LEVIN, W. (1981). Comparisons between a major and minor form of liver microsomal cytochrome P-450 from rats treated with Aroclor 1254. Fed. Proc. 40(61*1640, Abet.*582. RYAN, D.E., THOMAS, P.E. and LEVIN, W. (1982). Purification and characterization of a minor form of hepatic microsomal cytochrome P-450 from rats treated with PCBs. Archives Blochem. Biophysics 216(1)*272-288. SERABJIT-SIHGH, C.J., ALBAO, P.H. and PHILPOT, R.M. (1982), The effects of Aroclor 1260 on rabbit pulmonary cytochrome P-450 isozymes, red. Proc. 41(5)*1479. SHIOTA, X. (1976). Postnatal behavioral effects of prenatal treatment with PCls in rats. Okalimaa Fol. Anat. Jap. 53*105-114. SCHOIMY, R, (1982). Mutagenicity testing of chlorinate3~biphenyls and chlorinated dibenzofurans. Mutation Res. 101*45-56. SHULL, L.R., BLEAVINS, M.R., OLSON, B.A. and AULERICH, R.J. (1982). Polychlorinated biphenyls (Aroclora 1016 and 1242)* Effect on hepatic microsomal mixed function oxidases in mink and ferrets. Environ. Contam. Toxicol. 11*313-321. * SILXHORTH, J.B., and GRABSTEXN, E.M. (1982). PCB immunotoxicity* Dependsncs on isomer planarity and the Ah gene complex. Toxicol. Applied Pharmacol. 65*109-115. SIPES, I.G., SLOCUW, M.L., PERRY, D.F. and CARTER, D.E. (1982), 2,4,5,21,4',5'-hexachlorobiphenyl* Distribution, metabolism, and excretion in the dog and the monkey. Toxicol. Applied PJiarmacol. 65*264-272. HONS 212946 70- SMITH, A. B., et *1. (1981;. Metabolic and health consequences of occupational-exposure to polychlorinated biphenyls (PCBs). British J. of. Ind. Med. 39;361-369. SPENCER, F. (1982). An assessment of the reproductive toxic poten tial of Aroclor 1254 in female Spragut-Dawley rets. Bull. Environ. Contain. Toxicol. 28 1290-297 . STORM, J.E, HART, J.L. end Sf3TH, R.E. (1981). Behavior of mice after pre- and postnatal exposure to Aroclor 1254. Weurobehav. Toxicol, and TeratoJ. JUl)5-9. THOMAS, P.T. and HIHSDILL, R.D. (1980). Perinatal PCS exposure and its effect on the immune system of young rabbits. Drug Chem. Toxicol. 3(2)i173-184. DENG, T.H. and ALVARES, A.P. (1981). Selective loss of pulmonary cytochrome P-450 in rabbits pre-treated with polychlorinated biphenyls. J. Biol. Chem. 256(14):7536-7542. VEERKAKP, W. , WEVER, J. and HUTZINGER, 0. (1981). The metabolism of some chlorinated diben?ofurans toy rats. Chemosohere 10(4)i397-403. VOSS, K.A. (19611. In vivo metabolism and disposition of Aroclor 5460 and 2,2,2'*, 5,575 *' hexachloroterphenyls (HCTa) in the rat. Dissertation Abstract Int. 42i No. 3. WARSHAW, R. et al. (1979). Decrease in vital capacity in PCBexposed workers-In a capacitor manufacturing facility. Annals N.Y. Acad. Set. 320i277-203. WIERDA, D., IRONS, R.D. and GREENLEE, W.F. (1981). Immunotoxicity in C57BL/6 mice exposed to benzene and Aroclor 1254. Toxicol. Applied Pharmacol. 60i410-417. WOLFF, K.S., et al. (1982). Disposition of polychlorinated biphenyl congener* in occupationally exposed persons. Toxicol. Applied Pharmacol. 62(1)<294-306. YOSHIKARA, S., NAOATA, K., YOSHIMURA, H., KUROKI, H., and MASUDA, Y. (1981), Inductive effect on hepatic enzymes and acute toxicity of individual polychlorinated dlbenzofuran congeners in rats. Toxicol. Applied Pharmacol. S9iSOO-S88. YOSHIHARA, S. and YOSHIMURA, H. (1961). Study for "Yusho"- the progress in this decade. Elsel Kacaku 27(3)144--155. HONS 2X2942