Document MMjJJzoG2jBgbOxpOoMk0w6GV

Ann. N.Y. Acad. Sci., 246, 258-267 (1975) I ON THE POSSIBLE MECHANISM OF CARCINOGENIC er ACTION OF VINYL CHLORIDE Benjamin L. Van Duuren ^^ Laboratory of Organic Chemistry and Carcinogenesis Institute of Environmental Medicine New York University Medical Center-- (2.1 T-) "5 1 6 New York, New York 10016 2 rt (1 Introduction Although we have to date not studied the carcinogenicity or metabolism of vinyl chloride, some suggested modes and sites of action are presented in this paper. This presentation is based on our past experience with structure-activity relationships of direct-acting alkylating carcinogens and also on various im portant considerations on mechanisms of action of chemical carcinogens in gen eral. Direct-Acting and Indirect-Acting Alkylating Carcinogens Studies on the mechanism of the carcinogenic action of vinyl chloride involve the consideration of whether the chloroolefin is the active carcinogenic agent or whether activated carcinogenic metabolites are involved. Therefore, it is instruc tive to consider the same question in earlier studies with a wide variety of chemical carcinogens. No attempt is made here to give a review of the literature on this subject, but only an indication of what appears to us to be the current thinking in this area. There is a considerable body of evidence which supports the view that many well-known carcinogenic agents such as nitrosamines, aromatic amines, azo dyes and aromatic hydrocarbons have to be metabolized in situ to their activated carcinogenic intermediates21 Although not firmly established for all of these carcinogens, much of the evidence suggests that the intermediate agents arc alkylatitmiiinaturej i.e.. electrophilic aeents. .Such^l^rojihjles undergo a variety of reacticmsTmhnucleochiles in vivo and it is reasonable to assume that one or more of these nucleophilic reactions is or are responsible for the carcinogenicity of the parent compounds. These carcinogens can, therefore, be referred to as indirect-acting alkylating carcinoens. Many of thtfsc fflmpmmili arc also capable oT physical binding with 1tissue-constituents. This noncovalent bigdige qf aromatic tiy(jpv--arhnr|<j in pnclcic acids, for example, has been extensively studied in many laboratories.1 Mo<jt rpsparrhpr* in r.hemicaI carcinogenesis now believe that such physical hinHino k not related to the critjq^l interactions involved in r-areinogene sis"by aromatic carcinogens. However, these studies and conclusions derived from them do not exclude the possibility that physiciHbindjnigvvTthtissuc^constitLietits olays an important role for some chemicalraremnppns ^ ~ The direct-acting alkylating carcinogens include such classes as the epoxides, O /?-lactones, and halo ethers. These agents do not have to be metabolized to exert their carcinogenic effect and for this reason they are usually carcinogenic only at the site of administration in laboratory experiments. There are some excep tions, e.g., propane sultone, which is chemically less reactive than other direct- 238 1 [ ] 0496 CJj|K>yu Van Duuren: Carcinogenic Action of iwi/g acting alkylating carcinogens in this group,' hence giving rise to tumors . site of administration and at distant sites.* During the past decade we have determined the carcinogenicity u; a scr:.-. o: approximately 100 direct-acting alkylating agents including the following chemi cal structural types: epoxides, lactones, their sulfur and nitrogen iscsters, and halo ether compounds. The carcinogenicity assays were conducted by use of mice and rats by various routes of administration.1'5 In parallel with the bioussays we examined the chemical reactivity of these compounds with a variety of nucleo philes by in vitro experiments. Nucleophilic reactions included rate of hydrolysis studies, rates of reaction with --S,Or and --N,~ ions,"-* reactions with :'iid/r: isolation of products of some of these agents with purines,11 DNA,'"'1 and whole mouse skin chromatin." In addition to these in vitro and in vivo studies we have studied the stereochemistry, molecular flexibility, and functionality (i.e.. mono- functional, bifunctional or polyfunctiona!) of these molecules since these facto: have an important bearing on cross-linking abilities in nucleic acids, react on; ... phosphate sites in nucleic acids, cross-linking in the chromatin complex' and re actions with such body chemicals as cysteine and glutathione." These studies have made a contribution to a basic understanding of the mode of action of direct- acting alkylating carcinogens. Furthermore, these studies have made it possible to predict carcinogenicity within this group of compounds, i.e., direct-acting alkylating carcinogens. Examples of these are: i. chloromethyl methyl ether (CME)`* " ii. bis(chloromethyl) ether (BCME)1*" iii. propane sultone' iv. N, N-dimethylcarbamyl chloride" These compounds are cited because they were, and in some cases still are, being used extensively in industry and in the cases of compounds i and ii our findings . have had far-reaching implications. As a result of our work on the carcinogenicity of CME and BCME, these two compounds were included in the recent list of human carcinogens by the U.S. Department of Labor.Several recent epidemio logic studies in this country and abroad have implicated these two compounds as being responsible for lung cancer in workers exposed to either or both of these materials.1'"* Activated Carcinogenic Intermediates of Vinyl Chloride Vinyl chloride is a small molecule which is relatively unreactive compared to its atiyl analogs and to such potent and chemically reactive direct-acting alkylat ing carcinogens as the a-halo ethers mentioned above. It may bind to microsomal membranes or other constituents such as serum albumin prior to metabolism. It is nonetheless likely that it is metabolized to activated carcinogenic intermediates in the liver and other sites. We have considered the possibility that hydroperoxide, peroxide, or free radical intermediates may be involved. While such intermediates should be kept in mind, we feel that epoxide intermediates are likely candidates as activated carcinogenic intermediates, particularly in the liver. Many studies have shown that aromatic hydrocarbons are metabolized to epoxides in vitro in rat liver." The liver is known to be particularly rich in such enzymes. These en zymes occur in other organs and tissues but at lower levels and the proposed .epoxide intermediate can therefore also account for primary tumors induced at EC- 0497 2(j0 Annals New York Academy of Sciences other sites in the rat inhalation experiments reported by Viola et al.!< and Maltoni." Furthermore, our own studies on structure-activity relationships of epox ides, lactones, and halo ethers referred to above point very strongly to an epoxide, monochloroethylene oxide, as the activated carcinogenic intermediate of vinyl chloride. This is illustrated in Figure 1. This compound is not only an epoxide as arc the carcinogens cpichlorohydrin and glycidaldehydc; it is also an n-chioro ether, like chloromethyl methyl ether and bis(chloromethyl) ether, i.e., the carbon atom bearing the cyclic ether (epoxide) oxygen also carries a chlorine atom. Such a-halo ethers are known to be highly reactive alkylating agents* BCME is a par ticularly potent direct-acting aikylating carcinogen resulting in tumors at the site of application, e.g., mouse skin, subcutaneous tissue in mice and rats, and rat lung. The metabolic studies on trichloroethylene described below had resulted in the suggestion of an epoxide intermediate in the metabolism of trichloroethylene. Earlier attempts to synthesize an epoxide of vinyl chloride by pcracid oxida tion have failed for obvious reasons. The compound, monochloroethylene oxide, has been reported" but it has only been partially characterized. Its synthesis and spontaneous rearrangement to monochloroacetaldehyde are shown in Figure 2. Monochloroacetaldehyde is a highly toxic agent2" which is intensely irritating to eyes, skin, mucous membranes, and respiratory tract. Evidence has been presented which indicates that chloroacetaldehyde is an intermediate in the metabolism of 2-chloroethanol." In vivo and in vitro experiments in rat liver have shown that chloroacetaldehyde reacts with glutathione to give S-carboxymethyl Cl CHj-O-CHj CICHj-O-CHjCl CME BCME A H,C----- C-Cl H Chloroethylene oxide 0, HjC-CH-CHjCl HjC-CH-CIln Epichlorohydrin Glycidaldehyde Figure I. Alkylating carcinogens related to chioroethylene oxide. h2ochci Peracid /0\ h2-c-c-h Cl _ Monochloroethylene oxide Ya 0 H-C' \h2ci Monochloroace taldehyde *Spontaneous rearrangement at room temperature Monochloroethylene carbonate Figure 2. Synthesis and rearrangement of chloroelfaylenc oxide. Van Duurcn; Carcinogenic Action of V' ci-ch2-cw2-oh NAD css-ch2-coon 4 NADII 2 ci-ch2ciio GSH 1' GS-ciij-cin S-Carboxymethyl glutathione S-Formylmethylglutathione Figure 3. Metabolism of 2-chloroethanol in rat liver. glutathione as a final metabolic product. These experiments are summarized in Figure 3. The role of reactions of activated carcinogenic intermediates of comoo'inds such as nitrosamincs or of direct-acting alkylating carcinogens such as glyciduicehyde and other agents with DNA or its constituent bases has been extensively studied. Therefore, it is of interest to examine this type of reaction as shown in Figure 4. Glycidaldehydc (1) was shown in our earlier studies to react with guanine derivatives (II) to give the triheterocyclic ring compound III.1" The spontaneous rearrangement product of monochloroethylene oxide is monochloro acetaldehyde as shown in Figure 2. Adenine and its derivatives (V, Figure 4) react with monochloroacetaldehyde (IV, Figure 4) to give a similar triheterocyclic productTM1" (VI, Figure 4). Some of the possible metabolic products of VC and the mechanisms of their formation are depicted in Figures 5 and 6. Analogous known in vivo and fn v/fro reaction schemes lend support to those given in Figures 5 and 6. It is ciear irom the schemes presented in Figures 5 and 6 that a variety of metabolic routes are possible and only laboratory experiments will enable vs to determine the exact in vivo metabolic pathways. It is also expected that in the liver the NAD* NADH + H* and alcohol dehydrogenase pathways play an important role and that rKjctioiKwith^uEhydrjdcompoundssuch^a^cystejjj^jjjd^gJjj^yjjgjj^are involved, as describeaSefowTonF^liToroetllyleneT"*""1^^^^^^^^^^^^^^* Trichloroethylene (TCE) This compound is a close structural analog of vinyl chloride and hence it is worth summarizing here our present knowledge concerning this compound. TCE is a liquid, b.p. 87C;M m.p. --88'C." It is prepared by the addition of chlorine to acetylene followed by treatment of the resultant tetrachloroethane with calcium hydroxide. TCE is extensively used in: a) the dry-cleaning industry; b) as a de greasing agent for metal surfaces, leather, and textiles, and c) as a solvent for the extraction of residual oils from vegetable oil cakes, etc. TCE has found wide use for these purposes because of its noninflammability. Also as a result of its non inflammability, it is frequently used in open-vat operations in poorly ventilated work areas. TCE was used since 1941 as an anes thetic but its use for this purpose has decreased sharply in recent years since it has been replaced by fluoro ethers and related anesthetics. A number of reports have appeared on the acute toxic effects of TCE in ani mals and man.TM"TM Most of these studies were by inhalation exposure for short periods of time and few effects were observed. Mice given TCE by intraperitoneal injection showed mild hepatic dysfunction." Mice exposed to TCE by inhalation 262 Annals New York Academy of Sciences In structures II and III R - ribose or deoxyribose. In structures V and .VI R -CHj, ribose or deoxyribose. Figure 4. Reactions of glycidaldebyde and chioroacetaldehyde with purine de rivatives. 0 A In vivo ROM ' Cl 1 H-C--CM Z1 1 +0R O' H K,C-CH0 Z1 OR r 7` ! H,C -- OC1 * J 21 OTI HO H \/ C-C /\ H Cl Oil I H2,C-C! H2, Cl EC- 0498 H-C-COOH 2I Cl H,C-CH0 Zt Cl Figure 5. Possible reactions of vinyl chloride. Van Duuren: Carcinogenic Action of VC r -I * H-C-CH 51 L Cl J ROH H 1 H.C -C --PR Cl H on, t` H,C - Oil JI Cl j II H.C-C-UR JI Cl OH I H.C-C-I! 3I Cl HjC-CHO Olj-COOH Figure 6. Possible reactions of vinyl chloride. for up to 8 weeks showed fatty degeneration of the liver.* Only ver;. >ircii_c .nic ies appear to have been carried out on workers exposed to TCE." Tp our knowh edgejiQ_carcinogenic effects have been eia5Xaed-ffiitlLJC-in--EMiSIimfltaLani mals. "***" Several reports have appeared describing the in animals and man. Trichloroacetic acid and trichloroethaaol or its glucuronide have been found in the urine of humans or dogs exposed to TCE by inhalation." K Powell," as early as 1945, suggested a metabolic scheme shown in Figure 7 in which an epoxide was proposed ac an intermediate in a subsequent study* rats were fed "Cl-labeled TCE and the metabolites were isolated from the urine. Both tri chloroacetic acid and trichlniwihanni wp... found. The specific activities ot tEe two metabolites isolated were approximately equal to that of the administered trichloroethylene. This result suggested that there is a pathway via trichloro- acetaldehyde, as suggested by Powell* (Figure 7) and that no exchange of chloride with the body chloride pool occurred. Daniel* also found that trichloror ethanol is not (he precursor of trichloroacetic acid, i.e,, these two metabolites are Tp^mcri via in^f-nepfj|-p| ' --i--------l--------------- ~-t fJial IHCfll0PHJIhyJene is converted to trichloroacetaldehyde (chloral) in liver microsomes of rats, rabbits Cl n V/ c.7 ci Cl /Vv\/ Cl H , j;------------------------------------------ CljC-CHO ------------------------- * CljC-COOH Figure 7. Proposed epoxide intermediate in the metabolism of trichloroethylene. 264 Annals New York Academy of Sciences Glucuronide A -on Figure 8. Other proposed metabolic intermediates of trichloroethylene. and dogs in a reaction requiring NADPH -- H* and oxygen. Leibman1" also ex amined the effects of activators and inhibitors of the enzyme system. Work by Byington and Leibman'1 presented similar evidence concerning the metabolism of TCE in rat liver. Other possible intermediates have been proposed,11 e.g., a chloronium ion or carbonium ion intermediate as shown in Figure 8. effect of disuifiram fbisfdiethylthiocarbamoyi)disulfidel on the metaboljsm of TCE has been examined in humans exposed to TCE by inha|^finrU 7as iound that the elimination oi trichloroethanol in the urine was decreased hit~5t) percent and ot trichloroacetic acid bv ~80 percent when disuifiram was ad- 'l was ^oun4 that unqhaneed TCE excreted from tile lungs increased <^-irni compared to the con trols. ft was conciiitkd that disultlram inhibits the oxidation of TCE. McKinney et at." reportedontheautoxidationToPT^S1 and "suggested that an epoxide intermediate may be formed. However, the isolation and identification of this compound has not yet been accomplished. In summary, a number of metabolites of TCE have been isolated; however, the exact metabolic route, the nature of the intermediates involved, and the effects on liver function remain to be clarified. Induction of Angiosarcomas of the Liver in Humans and Experimental Animals by Other Chemical Agents Angiosarcoma_of the liver, which has been much discussed at this conference in connection with vinyl chloride and polyvinyl chloride carcinogenesis, is a rare disease in humans. It is, therefore, worthwhile to mention some other chemicals which cause the same type of liver tumor in animals and man. A number of reports have appeared indicating that vineyard workers in France and Germany exposed to arsenical insecticides had an unusually high inci dence of these tumors,``~" Regelson et at." reported on one case of human hemangioendothelial sarcoma of the liver in a patient treated for psoriasis with sodium arsenite (Fowler's solu tion). This drug was still in use for this purpose at lhe time of theTeport of Regel son et al." Toth and co-workers" ** reported on the experimental induction of angiosar coma of the liver in mice by treatment with dime)hY|ni(rosamine" and 1,1-di methyl hydrazine.1* Both compounds were administered orally in these ex EC- 0499 Van Duuren; Carcinogenic Action of VC periments. Dicthvlnitrosamine administered to rats also resulted in . .:r angiosarcoma.TM All three of these compounds arc currently bciicsed to be e- li veried to alkylating agents in vivo. Rats fed a diet_flnlaifliaj^iimfant bcnecio tongiiobus (wlpch contains nvrrolizi3me alkaloids! resulted in ihcimlSjraL ^owriijiJgtu^inh^tu^firs7!r~--""* Summary and Conclusions An attempt has been made in this report to bring together relevant informa tion bearing on the possible mechanism of action of the carcinogen vinyl chloride. The relationship of indirect-acting alkylating carcinogens and their direct-acting counterparts is described. It is suggested that an -chloro ether or related -oilium ion intermediate is involved as an activated carcinogenic intermediate in vivo. Based on earlier studies it is also suggested that noncovaicnt hindinr in -rr'i' . albumin, microsomal membranes, or otner large! sites may precede activatinn ` > a CarcinOlieniC intPrqyniiHp ^rnm *nW<utlinffMlllliynHnv hr* rid.-il.-tl m fh^ sir- of origin ot acroostcolysis, a disease associated with exposure to vinyl ch.oricic. Trichloroethylene is a widely used industrial chemical. Based on its structural similarity to vinyl chloride and known information concerning its metabolism in animals and man, an a-chloro ether or -onium ion is suggested as an important intermediate in its metabolism, it is likely to be carcinogenic, particularly to the liver but this remains to be established. Other chemical agents that have been shown to be carcinogenic to animals and man are briefly mentioned. Acknowledgments The author is indebted to Drs. G. Witz and B. M. Goldschmidt tor their col laboration. Our previous work on direct-acting alkylating agent carcinogens was supported by U.S. Public Health Service Contract N01 CP 4-3221 from the Na tional Cancer Institute and U.S. Public Health Service Grant ES-00260 from the National Institute of Environmental Health Sciences. Refefrences L Bergmann, E. D. & B. Pullman, Eds. 1969. Physico-chemical mechanisms of car cinogenesis. In Proc. Intern. Symp. Quantum Chem. Biochem., Jerusalem, 1968. Israel Academy of Sciences and Humanities, Jerusalem. 2. Miller, S. A. 1970. Carcinogenesis by chemicals: an overview. G. H. A. Clowes Memorial Lecture. Cancer Res. 30: 559-576. 3. Van Duuren, B. L., S. Melchionne, R. Blair, B. M. Goldschmidt & C. Katz. 1971. Carcinogenicity of isosters of epoxides and lactones: aziridine ethanol, propane sultone, and related compounds. J. Natl. Cancer Inst. 46: 143-149. 4. Druckrey, H., H. Kruse, R. Preussmann, S. Ivankovic, Ch. Landschutz & S. Gimmy. 1970. Canccrogene alkylierende Substanzen IV. 1,3-Propansulton und I,4-Butansutton. Z. Krebsforsch. 75: 69-84. 5. Van Duuren, B. L., N. Nelson, L. Orris, E. D. Palmes & F. L. Schmitt. 1963. Carcinogenicity of epoxides, lactones, and peroxy compounds. J. Natl. Cancer Inst. 31: 41-55. 6. Van Duuren, B. L. 1969. Carcinogenic epoxides, lactones, and halo-ethers and their mode of action. Ann. N.Y. Acad. Sci. 163: 633-651. (See references cited therein.) 7. Van Duuren, B. L., B. M. Goldschmidt, C. Katz, I. Seidman & J. Paul. 1974. The carcinogenic activity of alkylating agents. J. Natl. Cancer Inst., in press. 8. Van Duuren, B. L. & B. M. Goldschmidt. 1966. Carcinogenicity of epoxides, * lactones, and peroxy compounds. HI. Biological activity and chemical reactivity. J. Med. Chem. 9: 77-79. 9. Van Duuren, B. L., C. Katz, B. M. Goldschmidt, K. Frenkel & A. Sivak. 1972. 266 Annals New York Academy of Sciences Carcinogenicity of halo-ethers. II. Structure-activity relationships of analogs of bisfchloromethyltether. J. Nat!. Cancer Inst. 48: 1431--1439. 10. Golsdchmidt, B. M., T. P. Blazej & B. L. Van Duuren. 1968. The reaction of guanosine and deoxyguanosine with glycidatdehyde. Tetrahedron Letters 13: 1583-1586. 11. Goldschmidt, B. M., K. Frenkel & B. L. Van Duuren. The reaction of propane sultone with guanosine, adenosine, and related compounds. J. Heterocycl. Chem., in press. 12. Segal, A., M. Schroeder, P. Barnett & B. L. Van Duuren. 1974. Studies of the effects in vitro of p-propiolactone and 0["C]one on whole mouse skin chromatin. Biochem. Pharmacol. 23: 937-946 13. Van Duuren, B. L. & B. M. Goldschmidt, Unpublished data. 14. Van Duuren, B. L., B. M. Goldschmidt, C. Katz, L. Langseth, C. Mercado & A. Sivak. 1968. Alpha-haloethers: a new type of alkylating carcinogen. Arch. Environ. Health 16: 472-476. 15. Van Duuren, B. L., A. Sivak, B. M. Goldschmidt, C. Katz & S. Melchionne. 1969. Carcinogenicity of halo-ethers. 1. Natl. Cancer Inst. 43: 481-486. 16. Van Duuren, B. L., B. M. Goldschmidt, C. Katz, & I. Seidman. 1972. Dimethylcarbamyl chloride, a multipotential carcinogen. J. Natl. Cancer Inst. 48: 15391541. 17. Federal Register. 1973. Title 29-Labor. Chap. XVII. Occupational Safety and Health Administration, Department of Labor. Part 1910. Occupational Safety and Health Standards. Emergency Temporary Standard on Certain Carcinogens. 38: 10929-10930. 18. Federal Register. 1974. Title 29-Labor. Chap. XVII. Occupational Safety and Health Administration, Department of Labor. Part 1910. Occupational Health and Safety Standards. Carcinogens. 39: 3756-3797. 19. Figueroa, W. G., R. Raszkowski & W. Weiss. 1973. Lung cancer in chloromethyl methyl ether workers. New Engl. I. Med. 288: 1096-1097. 20. Brown, S. M. & S. Selvin. 1973. Lung cancer in chloromethyl methyl ether work ers. New Engl. J. Med. 289: 693-694. 21. Thiess, A. M., W. Hey & H. Zeller. 1973. Zur Toxikologie von Diehlordimethyl- ather--Verdacht auf kanzerogene Wirkungauch beim Menschen. Zenlr. Arbeitsmed. 23:97-102. 22. Sakabe, H. 1973. Lung cancer due to exposure to bis(chloromethyl)ether. Ind. Health. 11: 145-148. 23. Conney, A. H. 1967. Pharmacological implications of microsomal enzyme induc tion. Pharmacol. Rev. 19: 317-366. 24. Viola, P. L., A. Bigotti & A. Caputo. 1971. Oncogenic response of rat skin, lungs, and bones to vinyl chloride. Cancer Res. 31: 516-522. 25. Maltoni, C. 1974. Unpublished data, presented at: Informal Fact-finding Hearing on Possible Hazards of Vinyl Chloride Manufacture and Use, U.S. Department of Labor, OSHA, Washington, D.C., February 15. Carcinogenicity Bioassays of Vinyl Chloride: Current Results. This monograph. 26. Powell, I. F. 1945. Trichloroethylene: absorption, elimination and metabolism. Brit. J. Ind. Med. 2: 142-145. 27. Zief, M. & C. H. Schramm. 1964. Chloroethylene oxide. Chem. Ind. April 18: 600-661. 28. Lawrence, W. H., E. O. Dillingham, J. E. Turner & J. Autian. 1972. Toxicity profile of chloroacetaldehyde. J. Pharm. Sci. 61: 19-25. 29. Johnson, M. K. 1967. Metabolism of chloroelhanol in the rat. Biochem. Phar macol. 16: 185-199. 30. Kochetkov, N. K., V. N. Shibaev & A. A. Kost. 1971. New reaction of adenine and cytosine derivatives, potentially useful for nucleic acids modification. Tetra hedron Letters 22: 1993-1996. 31. Secrist, J. A., Ill, J. R. Barrio, N. J. Leonard & G. Weber. 1972. Fluorescent modification of adenosine-containing coenzymes. Biological activities and spec troscopic properties. Biochemistry 11: 3499-3506. 32. Chemical Rubber Co. 1968-1969. Handbook of Chemistry and Physics. Cleve land, Ohio, 49th edit. 33. Defalque, R. J. 1961. Pharmacology and toxicology of trichloroethylene. Clin. Pharmacol. Therap. 2: 665-688. 34. Klaassen, C. Cki& G. L. Plaa. 1966. Relative effects of various chlorinated hydro- EC- 0 5 0 0 Van Duuren: Carcinogenic Action of VC ' c1a51rb. ons on liver and kidney function in mice. Toxicol. Appl. Pharmacol. 1- 35. Kylin, B,, H. Reichard, 1. Sumegi & S. Yllner. 1963. Hepatotoxicity of inh.il.-j trichloroethylene, tctrachloroethylene and chloroform. Single exposure. Acta Pharmacol. Toxicol. 20: 16-26. 36. Kylin, B,, Sumegi & S. Yllner. 1965. Hepatotoxicity of inhaled tricr.ioroethykne and tctrachloroethylene. Long-term exposure. Acta Pharmacol, ic.icol. 22: 379-385. 37. Soucf.k, B., & D. Vlachova. I960. Excretion of trichloroethylene metabolites in human urine. Brit. J. Ind. Med. 17: 60-64. 38. Daniel, J. W. 1963. The metabolism of **Cl-tabelled trichloroethylene and tetra- chloroethylene in the rat. Biochem. Pharmacol. 12: 795-802. 39. Leirman, K. C. 1965, Metabolism of trichloroethylene in liver microsomc--. I. Characteristics of the reaction. Mol. Pharmacol. 1: 239-246. 40. Byington, K. H. & K. C. Leibman. 1965. Metabolism of trichloroethylene in !:-er microsomes. 11. Identification of the reaction product as chloral hydrate, '.foi Pharmacol. 1: 247-254. 41. Van Dyke, R. A. & M. B. Chenoweth, 1965. Metabolism of volatile a nexlh Anesthesiology 26: 348-357. 42. Bartonicek, V. & J. Teisinger. 1962. Effect of tetraethyl thiuram disvi I - ' - c (disulftram) on metabolism of trichloroethylene in man. Brit. J. Ind. Men. 19: 216--221. 43. McKinney, L. L., E. H. Uiiing, J. L. White & J. C. Picken, Jr. 1955. Auto oxidation products of trichloroethylene. J. Agr. Food Chem. 3: 413-419. 44. Roth, F. 1958. Ueber den Bronchialkrebs arsengeschadigter Winzer. Virchows Arch. Pathol. Anat. Physiol. Klin. Med. 33J: 119-137. 45. Roth, F. 1956. Ueber die chronische Arsenvergiftung der Moselwinzer unter be- sonderer Beriicksichtigung des Arsenkrebses. Z. Krebsforsch. 61: 287-319. 46. Roth, F. 1957. Arsen-Lebcr-Tumoren (Hamangioendotheliom). Z. Krebsforsch. 61: 468-503. 47. Regelson, W., U. Kim, J. Ospina, el al. 1968. Hemangioendotheliat sorro-'r the liver from chronic arsenic intoxication by Fowlet's solution. Cancer 21: ' i-522. 48. Toth, B., P. N. Magee & P. Shubik. 1964. Carcinogenesis study with dimethyl- nitrosamine administered orally to adult and subcutaneously to newborn BALB/c mice. Cancer Res. 24: 1712-1719. 49. Toth, B. 1973. 1, l-Dimethylhydrazine (unsymmetrica!) carcinogenesis in mice. Light microscopic and ultrastructural studies on neoplastic blood vessels. J. Natl. Cancer Inst. 50: 181-194. 50. Hadjiolov, D. 1972. Hemangioendothelial sarcomas of the liver in rats induced by diethylnitrosamine. Neoplasma 19: 111-114. 51. Harris, P. N. & K. K. Chen. 1970. Development of hepatic tumors in rats fol lowing ingestion of Senecio longiiobus. Cancer Res. 30: 2881--2886. F