Document NeRa5Z2rMxamv2ML5nQw84d6D

WORLD I-IEALTH ORGANIZATION INTERNATIONAL AGENCY FOR RESEARCH ON CANCER IARC MONOGRAPHS ON THE EVALUATION OF THE CARCINOGENIC RISK OF CHEMICALS TO HUMANS The Rubber Industry VOLUME 28 This publication represents the views and expert opinions c of an IARC Working Group on the Evaluation of the Carcinogenic Risk of Chemicals to Humans which met'in Lyon, 16-23 February 1981 April 1982 w C/I o GO O Q H* 'i INTERNATIONAL AGENCY FOR RESEARCH ON CANCER * VI. 310LOGICAL EFFECTS 1 51 -r This section includes information on chemicals for which rs some data on carcinogenic, mutagenic or teratogenic effects were f available to the Working Group. The selection of chemicals included in this section is not the result of an exhaustive screening of the literature on all chemicals used or formed in " the rubber industry. Some of the chemicals have already been * considered in previous IARC Monographs, and, for these, reference to the pertinent Monograph is given (see also Appendices 2 and 3). For a number of chemicals, information on carcinogenicity, mutagenicity or teratogenicity was derived from the literature and was not critically analysed or evaluated in the same way as that for chemicals previously considered in IARC Monographs. This section should therefore be read more as an introduction than as an exhaustive and complete review. 1. Carcinogenic, mutagenic and teratogenic effects j 1.1 Accelerators, antioxidant and curing agents I j Of the**organic peroxides used as curing agents, dicumyl j peroxide, bis(2,4-dichlorobenzoyl) peroxide and q.g'-bisf tert- b u tyl per ox y iso propy I) be n z e ne were reported to be skin irritants, while tert-butylperoxide was not. Their acute toxicity in animals has been summarized by Holmberg and Sjostrom (1977), but virtu ally no evaluable data are available on their carcinogenic, muta genic or teratogenic effects. Hydrogen peroxide was reported to be carcinogenic to mouse duodenum when administered at a concen tration of 0.496 in drinking water for 108 weeks (Ito et al.. 1981). It was also shown to induce DNA repair in Escherichia coli (Rosenkranz, 1973) and mutations in S ac c h aro m vc es cerevisiae (Thacker, 1975). The accelerators and curing agents commonly used are thiuram compounds, including tetram ethylthiuram disulphide (thiram, TMTD), tetraethylthiuram disulphide (disulfiram, TETO) and tetram ethylthiuram monosulphide (TMTM). (See also Appendix 1). 1 52 IARC MONOGRAPHS VOLUME 28 TMTD and TETD were evaluated for carcinogenicity in an IARC Monograph (IARC, 1976b). The two carcinogenicity studies in experimental animals and the one case report of human thyroid cancer were considered an insufficient basis for evaluating the carcinogenicity of TMTD; the data for TETD were also too limited to evaluate carcinogenicity. In one study of oral administration to mice, the incidence of liver-cell tumours was increased in males of two strains, and the number of lung tumours was in creased in males of one strain. Both TMTD and TETD react with nitrite to form carcinogenic nitrosodialkylam ines. TETD is meta bolized in rats and man to diethyldithiocarbam ate, glucuronide and sulphate conjugates, diethylamine and carbon disulphide. These compounds were investigated in a series of reactivity studies in vitro (Hemminki et al.. 1980). TMTM, TMTD, TETD, zinc and copper dim ethyld ithiocarb am ate s, zinc, cadmium and tellurium d ie thyld ith ioc arb am a t es, and zinc d ib ut y I d it h ioc a rb a m at e were incubated with synthetic nucleophiles, 4-(para-nitrobenzyl)~ pyridine and deoxyguanosine. The thiuram compounds were not reactive; copper dim ethyldithiocarbam ate, cadmium and tellurium d ie th y 1 d ith io c arb a m a t e and zinc d ib ut y Id it h io c arb a m at e reacted slowly with both deoxyguanosine and 4-(para-nitrobenzyl)pyridine. There seemed to be no correlation between mutagenicity in E schcrichia coli WP2 and chemical reactivity. In a series of mutagenicity tests on rubber accelerators (Hedenstedt et al., 1979), TMTD and TMTM were mutagenic but TETD was non--m utagenic to Salmonella typhimurium. TMTD was the most active of the thiurams. Five out of nine tested dithiocarbam ates were also mutagenic, ziram being the most active. 4,4'-Dithiomorpholine, zinc ethylphenyldithiocarbamate, N, N ' -- d ic y clohexyl-para-phen ylenediamine, N.fa'-dipheny I- para-- phenylenediamine, 4,4'-diam in odiphenyl methane, 6-ethoxy-1^-di ll y dr o-2,2,4^-trim ethylquinoline and N-m ethyl-N,4-dinitrosoaniline were all direct or indirect mutagens for S. typhimurium in another study (Hedenstedt, 1982). TMTD was also found to be muta genic in another study on S. typhimurium TA100 (Shirasu et al.. 1977). TMTD, TMTM and TETD (in order of decreasing toxicity) pro duced deaths and malformations in chicken embryos (Korhonen et al., 1 982a). The dithiocarb am ates investigated were cadmium and zinc d ie th y Id ith ioc arb a m a t e , zinc ethylphenyldithiocarbamate. zinc tellur thioc erabr (I AR (Khe embr tetr; c arb comp cone urea ethy 1.2 i nd u: ethar lizer reaci th eir c arc i genit rats and chlor chlor liver st at repo chIo rode *3 sligh cn and T Al o c hi oi o has o an IA R C tid ics in i yroid it the o . i it e d nis ^ration eased in was in:act with is m eta il c ur on id e reactiv ity TD, zinc tellurium te were b enzyl)rere not tell urium reacted ) pyrid ine. lie ity in elerators but TETD the most m ates I b am ate, yl-paraf-1,2-di-- scan il ine r iu m in 'e mutai et al.. ty) protonen _et turn and b am ate. BIOLOGICAL EFFEuiS 153 -rr j. zinc dibutyldithiocarbamate, copper dimethyldithiocarbamate, -* tellurium diethyidithiocarbam ate and piperidine pentam ethylenedi-- i'S thiocarbamate. The cadmium and zinc salts were particularly nr embryotoxic (Korhonen et al., 1982b). * Ethyl enethiourea has been judged to be carcinogenic in rats (IARC, 1974). It was also teratogenic in two studies in rats ~ (Khera, 1973; Ruddick & Khera, 1975). When tested on chicken embryos, together with 1,3-dibutylthiourea, 1,3-diphe'nylthiourea, t e tr a m ethy Ithiour ea, tri m ethy11hiour e a, 1,3-die thy11hiourea and carbon disulphide, e th y I e n e thio ur e a was the least embryotoxic com pound in the series. Carbon disulphide was inactive at the concentrations tested. Tetram ethylthiourea and 1,3-diphenylthiourea were the most active teratogens in this test system; ethylenethiourea was a weak teratogen (Korhonen et al., 1982c). 1.2 Solvents Solvents that have been used or are used in the rubber industry include benzene, trichloroethylene, 1,1,1-trie hlor oethane and methylene chloride; 1,4-dioxane is used as a stabi lizer in solvents. Generally, these compounds are-.not chemically reactive and do not appear to react with macromolecules, but their metabolites may do so. Benzene is a known clastogen and has been considered to be carcinogenic in humans and rodents (IARC, 1982). It was terato genic in rats ('Kuna & Kapp, 1981). 1,4-Dioxane is carcinogenic in rats and guinea-pigs (IARC, 1976b). The evidence for the carcinogenicity of trichloroethylene and 1,1,1-trichloroethane is limited; and the data on methylene chloride were inadequate for evaluation (IARC, 1979c). 1,1,1--Tri-- chloroethane induced mutations *in E. coli in the presence of a liver microsomal activation system; the increase was small but statistically significant (Norpoth et al., 1980). The data reported in the literature regarding the mutagenicity of tri chloroethylene are conflicting. However, in the presence of a rodent liver activation system, a technical-grade product slightly increased mutagenicity in E. colt (Greim et al., 1975); and a product 99.5% pure increased mutagenicity in S. typhimurium TA100 (Bartsch et al., 1979). Irreversible binding of **C-tri til chloroethylene to mouse liver constituents, in vitro and has been described (Uehleke & Poplawski--Tabarelli, 1977). in vivo, o G3 O o 154 IARC MONOGRAPHS VOLUME 28 D i c hi or o m eth me is weakly mutagenic in S. typhimurium T A98 and TA100, both in the presence and absence of a liver activation system (Jongen et al., 1 978). Covalent binding and metabolism to carbon monoxide, both in vivo and in vitro, have been described (Ahmed et al.. 1 980). Studies on the teratogenicity of trichloroethylene, t r i c h I or o e t h a n e, methylene chloride and 1,4-dioxane were inconclusive or failed to demonstrate skeletal or visceral formations (IARC, 1976b, 1979c). 1,1,1either mal Di m ethylform amide, also used in the rubber industry, is metabolized in rats to N-m ethylform am ide (Barnes & Ranta, 1972), which is teratogenic and embryotoxic to rats (Stula St Krauss, 1 977). Ethylene dichloride, which has been used as a solvent in the rubber industry, is carcinogenic to mice and rats and is mutagenic to S. typhimurium with or without metabolic activation (IARC, 1979c). In human whole blood cultures, low concentrations (0.0125-- 0.05%) of a rubber solvent refined from a straight-run petroleum distillate of paraffin-based crude oil (consisting of a mixture of paraffins, monocycloparaffins, monoolefins, benzene and alkyl benzenes) caused an increased frequency of chromatid gaps and breaks; high concentrations (>0.05%) caused increases in the frequency of chromosome breaks. Concentrations up to a toxic level failed to produce sister chromatid exchanges (Altenburg _et^ aL, 1979). 1.3 Monomers Small amounts of monomers, e.g., acrylonitrile, butadiene, chloroprene, ethylene, propylene, styrene, vinyl acetate, vinyl chloride and vinylidene chloride, may remain in solid rubber and could be released into the work environment. Most of these mono mers do nort bind directly with cellular macromolecules, but their metabolites may do so. Metabolic activation of acrylonitrile, chloroprene, styrene and vinyl chloride leads to reaction pro ducts which are mutagenic in S. typhimurium and in other bacte rial test systems. In the case of acrylonitrile, chloroprene and vinyl chloride, a low mutagenic response was obtained in S. typhimurium and E. coli without metabolic activation (IARC, 1979b). Chloroprene, styrene and vinyl chloride also produce clastogenic effects. a cr y 1 limit were is c a aery L terat that hum ai react: genic al., cinog (IA R < 1.4 It hi tial encoi being talc may hum a to cases 1.5 cult, i many che m 3 of p babI y N O 03 O T A98 iv ation is to sc. .oed 1.1,1either m al- try, is 1972), Krauss, 'ent in 2nd is tivation 0,0125Croleum i ixtur e I alkyl ps and in the toxic urg et adiene, , vinyl er and m on ot their nitrile, n pr o- bactene and in S. (I ARC, troduce f BIOLOGICAL EFFECT. 155 There is sufficient evidence for the carcinogenicity of acrylonitrile and vinyl chloride in experimental animals, and limited evidence for the carcinogenicity of styrene. The data '/ere inadequate for an evaluation of chloroprene. Vinyl chloride is carcinogenic to humans, and there is limited evidence that - acrylonitrile is too (1ARC, 1979b,c). (See also Appendix 3.) Acrylonitrile, chloroprene, styrene and vinyl chloride are teratogenic to experimental animals, and there is a suspicion that styrene and vinyl chloride have teratogenic effects in humans (IARC, 1979a,b) Epichlorohydrin is used in the manufacture of elastomers. It reacts directly with macromolecules and it is mutagenic, clasto- genic and carcinogenic in mice and rats (IARC, 1976a; Laskin et al., 1980). No adequate epidemiologicalobservations on the car cinogenicity of epichlorohydrin to humans have been published (IARC, 1976a, 1979b). 1.4 Talc French chalk (talc) is widely used as an antitacking agent. It has been used in many processes and is still used in substan tial quantities, wit is a major component of environmental dusts encountered in the rubber industry. Its composition is variable, being dependent upon the source of supply; for the most part, talc particles are 'platey1 in character. Some talcs, however, may contain asbestos fibres, which are carcinogenic in both humans and animals (IARC, 1977). Prolonged and excessive exposure to talc has been shown to give, rise to a number of documented cases of respiratory disease. (See section VI, .2.2). 1.5 Carbon blacks Assessment of possible hazards from carbon black is diffi cult, because different preparations contain variable amounts of many compounds, some of which have still not been identified. The chemical composition and physical properties differ with methods of preparation and the nature of the starting material and, pro bably, from day to day. BFG39154 1 56 IARC MONOGRAPHS VOLUME 28 The most probable carcinogenic hazard is associated with the 'benzene e x t r a c t ab I e s1. The material, extracted from furnace black, consists mainly of aromatic hydrocarbons and sulphur com pounds. Some compounds have been identified by computerized gaschromatographic mass spectrometry and high-resolution mass spectrometry (Lee & Hites, 1 976). Among them, benz[^]anthracene, ben:o[_a]pyrene and indeno(1,2,3-cd)pyrene are known carcinogens (IARC, 1973); chrysene has been described as an initiating agent, and pyrene and fluoranthene as cocarcinogens for mouse skin (Van Duuren, 1976). Oxygen derivatives (Gold, 1975) and nitro deriva tives (Fitch & Smith, 1 979) of polycyclic compounds have also been found in extracts of commercial carbon blacks. Benzoylpyrene was found in benzene extracts of all types of carbon black examined (Troitskaya et al., 1975). The amounts of some poly cyclic hydrocarbons found in the benzene extracts of furnace carbon blacks are shown in Table 13 (from Locati et al.. 1979). Existing data indicate that the known carcinogens present in carbon blacks are strongly adsorbed but can be eluted by bio logical fluids. Although Neal et al. (1962) reported that poly cyclic hydrocarbons, including b e n z o y) p y r e n e, in furnace and channel blacks are not eluted by human blood plasma or gastric juice, Kutscher et al. (1967) found that bovine serum elutes 10- 20% of benzoy] pyrene from various types of carbon blacks within periods ranging from 4-60 hours depending upon the particle size. When a commercial carbon black (500-nm particle size) was incu bated with sterile human plasma, several polycyclic aromatic hydrocarbons, including benzo[ a]pyrene, were extracted within 16-192 hours. Elution of benzo[ji] pyrene from fine-particle carbon black (80 nm) was more difficult, but 65% was extracted within eight days (Falk et al., 1958). The rates of elution by benzene and the amounts of material extractable from carbon black vary with the type of carbon. Thus extraction "Of two carbon blacks which contained over 0.1% of extractable material, was almost complete in 150 .tours (Locati et al.. 1979). In a study using three organic solvents for extrac tion of adsorbates from five rubber-grade oil-furnace blacks, toluene was found to be a better extractant than benzene and both to be superior to cyclohexane (Taylor et al., 1980). l' i t Table 13. Values expr black type" PAH Phenanchre anthrac Fluorancr.e Benzo(d,, thiopoe acenapr Pyrene Benzofc.h fluorai Cycloper.r* pyrene Benzofluo (coral Benzooyre Dimemyic and/or benzol Indenopy: Benzo f c An than tn; - 0 C/1 19 O PS Benzoacr deriv vith the f ace ir - o m -- :e-1 gas- hracene, cinogens 5 agent. in (Van deriv a-- ve also n z 0 fa 1n black e polyfurnace ) sent in >y biot polyc e and gastric tes 10- with in le size. _ 0 atic carbon .hin 1aterial . Thus .1% of cati et extrac-- blacks, id both BIOLOGICAL EFFECTS - 157 Table 13. Polycyclic aromatic hydrocarbons (PAH) identified in five types of carbon black. Values expressed as percent of benzene extract; mean value and range of 5 samples for each black type3 PAH Sample % AB C D e Phenanthrene and/or anthracene Fluoranthene Benzo [d ,e_, f| d Ibenzothiopnene + benzo(a|acenaphthylene Pyrene Benzo(g,h,ilfluoranthene Cyclopentafcdlpyrene Benzofluoranthenes (total) %* Benzopyrenes (total) Dimethylcyclopentapyrene and/or dimethylbenzofluoranthene Indenopyrene Benaoig ,h, Ilpervlene _ Anthanthrene Benzoacridine derivative 0.07 absent 0.05-0.1) 4.1 (3.3-4.3) 3.7 (3.4-3.9) <0.05 <0.3 <0.35 3.5 (2.9-4.3) <0.3 absent 4.1 (3.4-6. 4) absent 0.2 (0.1-0.5) 4.3 (3.9-5.2) <0.05 22.2 23.2 16.5 17.2 17.2 (21.3-24.9) (21.6-24.0) (13.4-18.7) (14.3-18.1) (14.3-20.1) 7.2 (5.9-7.6) 6.3 (5.6-7.2) 6.9 (6.5-7.3) 4.8 (4.2--5)- 7.8 (7.5-8.1) 10.2 <0.3 (5.6-14.7) 7.0 (6.8-7.3) 4.4 (4.2-5) 6.5 (5.2-7.8J 0.7 (0.4-1.1) absent <0.3 1.2 (0.9-1.4) 0.6 (0.4-0.9) 1.4 (1.1-1.9) 0.5 0.3-0.7) 1.1 (0.4-2.1) 2.6 (1.5-3.2) 2.7 (2.0-3.9) 2.5 (2.0-3.5) 0.5 0.9 0.3-0.7) (0.3-1.3) 1.9 (1.7-2.3) 1.4 (1.1-1.7) 1.7 (1.5-2.2) 0.5 (<0.3-0.7) 11.7 6.2 " (10.9-13.8) (5.5-8.5) 3.2 (2.8-3.5) - 0.05 absent 0.1 (abs.-0.3) 8.7 (8.2-9.5) 0.8 (<0.3-1.2) absent 2.3 (2.1-2.5) 13.5 (12-16.5) 2.3 (1.9-3.3) absent 2.9 (2.7-3.1) 13.6 (12.8-17) 3.5 (2.8-5.5) 0.2 (abs.-O.S) 5 M O O O (Qfi 158 Table 13 (contd) IARC MONOGRAPHS VOLUME 28 PAH Sample Isoner of coronene Coronene 6.3 (5.1-7.2) 7.2 (4.1-10) 1.1 (abs.-2.2) 1.2 (<0.3-3.2) 0.1 (abs.-0.3) 3.8 (3.5-4.4) 0.5 (abs.-l) 8.9 (8.5-9.8) 5.1 (4.6-5.7) 11.7 (9.5-13.8) TOTAL 78.5 43.8 50.05 63.5 77.5 a From Locati et al. (1979) Methylene chloride extracts of soots prepared by burning either kerosene or equal parts of pyridine, decaline and ortho- xylene or thiophene, were assayed in S. typhimurium in the pre sence of liver supernatant from Ar o c lo r-ind u c e d rats for the induction of 8-azaguanine-resistant mutants. Mutation frequencies decreased in the following order: kerosene soot > furnace blacks > nitrogen-containing soot > sul ph ur-c o n t ai n i n g soot (Kaden et al., 1 979). Several nitrated polycyclic aromatic hydrocarbons were mutagenic in S. typhimurium and induced unscheduled DNA repair in cultured human (HeLa) cells (Campbell et al., 1981). Samples of furnace blacks used as photocopy toners were found to be mutagenic in S. typhimurium, partly because of the presence in them of nitroarom atics (Rosenlcranz et al., 1980; Lofroth et al., 1980). Studies on the carcinogenicity of carbon blacks, by oral, skin, s.c. or inhalation exposure (Nau et al.. 1 958a,b, 1 960, 1962, 1976), provided no evidence of carcinogenic effects; but the studies were inadequate in terms of number of animals and duration. Furnace black (but not channel black) when suspended in tricaprylin induced sarcomas upon injection into mice. As it was not carcinogenic when implanted alone, it is probable that it became so when polycyclic aromatic hydrocarbon carcinogens were eluted by the tricaprylin. Injection of benzene extracts of furnace black also induced sarcomas (Steiner, 1954). w blacks cation 1958a,t s.c. in been s its c a freduce | given i I m anuf i neopl a: sm all 1.6 A 1 form ec m aero Miller (1 979) JO (A JO o G O a amine: activ a largely susce; arom a hydrox com po conv er interm cation nitrog icai c ive e cul ar a cor maligi and < . rr~5.7) .7 .5-13.3) .5 burning ortho-- he prefor the quencies i blacks ad en e t carbons :d D N A 1981). Found to sence in t al.. ra, I 960, cts; but tals and I in triwas not becam e e eluted fur n a c e BIOLOGICAL EFFECTS 159 When benzene-extractable materials from channel and furnace blacks were administered to mice by s.c. injection, skin appli cation or feeding, malignant tumours were produced (Nau ct al., 1 958a,b, 1960). Malignant tumours were also produced following s.c. injections of cotton-seed oil in which the carbon blacks had been suspended for 1-6 months (Nau et al., 1960). When 3-m ethylcholanthrene was adsorbed onto carbon black, its carcinogenic effect following s.c. administration to rats was reduced in comparison to that produced when the same dose was given alone (von Haam et al., 1 958). In a retrospective cohort study of workers involved in the manufacture of carbon black, no excess mortality from malignant neoplasms was noted; but the number of deaths considered was small (Robertson A Ingalls, 1980) 1.6 A ro m atic a mines The metabolism and the nature of reactive intermediates formed from aromatic amines and their interaction with cellular macromolecules have been reviewed by Boyland (1 95"), Miller and Miller (1969), Clayson and Garner (1976), Miller (1978), Irving (1 979) and Kriefc and Westra (1 979). The carcinogenic or genotoxic activity of all aromatic amines that have been studied in detail is dependent on metabolic activation in vivo. Differences in these metabolic pathways largely account for the differences seen in tissue-- and speciessusceptibilities to cancer induction. The carcinogenicity of aromatic amines or amides is dependent on their oxidation to N -- hydroxy derivatives, while the carcinogenicity of aromatic nitro compounds is linked to their reduction to hydroxylam ines. Further conversion of the N-hydroxyl am ine or N--hydroxyam ide to reactive intermediates can occur in several ways, including (1) esterifi cation of the N-hydroxy group, (2) non-enzymic protonation of the nitrogen of the hydroxylam ine, and (3) oxidation to a free rad ical of arylhydroxam ic acids. Following generation of such react ive electrophilic intermediates in tissues or cells, macromolecular binding to nucleic aids and proteins has been observed. As a consequence, mutations (and other genotoxic effects, including malignant transformation) have been induced in a variety of cells and organisms when the reactive intermediates or the parent C/I ^ ^ Ho BFG39158 160 IARC MONOGRAPHS VOLUME 28 amines were assayed in the presence of a mammalian drug metabo lizing system. In many cases, arylamidated and arylaminated products were formed with nucleic acid bases. A great number of aromatic amines have now been studied for carcinogenicity, and considerable data have been reported in the first 27 volumes of the 1ARC Monographs. However, many of the aromatic amines considered in that series could not be evaluated for carcinogenicity, due to the inadequacy of experiments in animals and/or the lack of data concerning man. (See also Appendix 2.) Naphchylam ine-acetaldehyde condensates, of which the most important was known by the trade name 'Nonox S', were introduced around 1928 as antioxidants in the rubber product manufacturing industry. They were used in the UK and other countries (but apparently not in the US) for the production of tyres, tubes, air bags, electric cables and miscellaneous rubber goods, and were present in rubber compounds in concentrations of up to 196. (It was the opinion of the members of the Working Group that concen trations of up to 4% may have been present in some products,' particularly in air bags.) Their use was discontinued in the UK in 1949. Although the manufacturing process has changed several times, early products contained up to 2.5% 1-naphthylamine and up to 0.25% 2-naphthylam ine. Exposure to Nonox 5 in rubber factories has been associated with an increase in the incidence of bladder cancer in workers (Veys, 1969).' 4--A m inobiphenyl, benzidine and 2 --naphthyl am ine are known to be carcinogenic for humans (IARC, 1979b). There is sufficient evidence of the carcinogenicity of 2,4-diam inotoluene and 4,4'methylenebis(2-chloroaniline) in experimental animals. For 1 -naphthylam ine, 2,5-diam inotoluene, 4,4'-diam inodiphenyl m ethane and N-phenyl--2--naphthylam ine, the available data were inadequate to assess--carcinogenicity in experimental animals or in humans. (See Appendix 3.) The experimental animal species most frequently used to investigate the carcinogenicity of aromatic amines are the dog and the hamster, in which the urinary bladder is the main target organ, as in humans. Aromatic amines have been shown to be carci nogenic in other rodent species, but the urinary bladder was not necessarily the target organ. M. im porta 1.7 M Ju M ducts i try as in . tyre or ev e ties tc m inera residue and ci cyclic m ay c m ay al A depend all ind carcinc PAHs ( B gated; an th e n diben z consid e IV found test (I 1.8 N >0 C/I h w dim e tl o nitroso anim a O co m pc m utag %\ metiborr .ted uc I for I in the f of the ev aluated n ents in See also :he most ntroduc ed jfacturing ries (but tubes, air and were 1*. (It C concen- produc ts, the UK d several te and up factories >f bladder known to s*"fficient iL 4.4'" ials. For nyl m ethane inadequate n humans. y used to : the dog tain target > be carcir was not BIOLOGICAL EFFECTS ...... . . . . 161 Methods of analysis for aromatic amines of major industrial importance have been summarized (Egan et al., 1981). 1 .7 Mineral oils, tar products and polycyclic aromatic hydroc arbons Mineral oils (e.g., coal-tar oils, petrolatum) and tar pro ducts (e.g., bitumen, pitch) are widely used in the rubber indus try as extenders. The amount of mineral oils (rich in aromatics), in tyres, for example, increased over past years up to about 2096 or even more, since they are cheap and provide desirable proper ties to the finished rubber. Extenders such as the high-boiling mineral oil distillates (so-called aromatic oils), obtained from residues of solvent-refining and the manufacture of lubricating and cutting oils, contain relatively large quantities of poly cyclic aromatic hydrocarbons (PAH), and 3096 and more of the oil may consist of 4-6-ring PAHs. (See also section VI, 1.5.) PAHs may also be formed when tars and mineral oils are heated. Although mineral oils and tar products vary in composition depending on their origin, the methods of production, etc., they all induce carcinogenic effects in mammals, including man. Their carcinogenicity may be dependent on the presence of carcinogenic PAHs (IARC, 1979b). Benzo[ a,] pyrene is the carcinogenic PAH most widely investi gated; b en z[_a ] anthr ac e ne, b en z o [J>] fl uor an the ne, b en z o (jj fl uoranthene, chrysene, dibenz(a,h|anthracene, dibenzof a.hlpyrene, dibenzof a.ilpyrene and ind e n o (1,2,3-^_d) py r en e) have also been considered by an IARC Working Group (IARC, 1973). Mineral oils containing found to be mutagenic in test (Hermann et al.. 1982). different amounts of PAHs have been the Salmonella typhimurium/microsom e 1.8 Nitroso compounds N--Nitrosodi--ji-butylam ine, N-nitrosodiethylamine, N-nitrosodim ethylam ine, N-nitrosom orpholine, N-nitrosopiperidine and Nnitrosopyrrolidine have been shown to be carcinogenic in various animal species, after administration by various routes. These compounds are metabolized to reactive metabolites, which are mutagenic in various experimental systems (IARC, 1978). There is ^ q o H 162 I ARC MGNOGRAPHS VOLUME 28 limited evidence for the carcinogenicity of N-nitrosodiphenyl- amine in rats; but it was not mutagenic in prokaryote or euka ryote cells (IARC, 1 982), although it is a tr ansnitrosa ting agent in vivo (O h s h i m a e t a I., 1982). No carcinogenic effects were observed when dini tr oso pe n t a m e thyl en e t e tr a m in e was tested in rats by oral administration and i.p. injection (IARC, 1 976a); and in a cell-transformation assay, negative results were also obtained (Styles, 1978). The data on the carcinogenicity of N-m ethyl,N-4- dinitrosoaniline were considered to be inadequate for evaluation (IARC, 1 972). para-Nitroso-N.N-dimethylanilinc has produced oesophageal tumours in rats (Goodall et al., 1968). 1.9 Phthalate and adipate esters Among the phthalic acid esters used in the rubber industry, di(2-ethylhexyl) phthalate has been most extensively investigated for its toxicological properties and is the most widely used. It inhibits mitochondrial function and causes peroxisome prolifera tion, hepatomegaly and testicular atrophy in rodents. Various mutagenicity tests in bacteria and mammalian cells showed nega tive or contradictory results. Long-term administration of di(2-e th yl h ex y I) phthalate to rats and mice resulted in hepato--, cellular tumours. Di (2-ethyl hexyl) adipate produced hepato cellular tumours in mice; but. di(_n-butyl) phthalate was apparent ly negative when tested for' carcinogenicity in mice and rats (IARC, 1982). 1.10 Curing fumes and other emissions The compounds present in the air during vulcanization (curing fumes) are mainly produced by the volatilization of rubber chemicals and their impurities and of the products of chemical reactions occurring during curing. (See also Sections IV and V.) Air -samples from industrial curing of acrylonitrilebutadiene styrene copolymer, ethylene-propylene rubber and chlo-- roprene rubber were mutagenic in S. typhimurium (Hedenstedt, 1982). In laboratory model systems, the curing of ethylene pro pylene rubber and chloroprene rubber produced mutagenic gases and vapours. On the basis of these and other series of tests of con densates from cured rubber material, it was concluded that its mutagenicity was dependent to a large extent on the composition of the polymer (Hedenstedt, 1982). iph e nyI-- >r 'ukag gent t $ were rats m in a obtained :hyl,N-4valuation iroduc ed industry, estig ated used. It r ol if e r a- Vario us id negai of di- hepatoh e pa totpparen tand rats tni zation a I t of ducts of :tions IV lonitril end chloJenstedt, ene pro gases and of con-- that its m position 3I0L0GICAL EFFECTS _ .... 163 In other model systems, in which mixing and curing condi tions were simulated (Hedenstedt et al., 1981), air samples from the curing of styrene-butadiene rubber and from the mixing and curing of chloroprene rubber and ethylene-propylene rubber were mutagenic to S. tyohimurium. No experimental data are available concerning the long-term toxicity of curing fumes. 2. Effects in humans (other than cancer) In this part, some effects in humans are described which are indicative of exposures that occurred within the working or general environment. Some of the data may be useful in assessing that exposure levels were sufficiently high to produce the ad verse effect. However, such effects may have no relation per se to the existence of a carcinogenic risk. 2.1 Dermatological effects Dermatological effects resulting from exposure to rubber and rubber chemicals have been known for years in rubber goods manu facture (Cronin, 1 980) and have been described in consumers. Non-specific dermatological reactions to rubber gloves and to footwear were reported many years ago (Downing, 1933; Marcussen, 1 943). 'Accelerator agents', especially m ere ap tob en zo thia z ol e, were described as the specific cause (Bonnevie & Marcussen, 1945). Shortly thereafter, seven cases of dermatitis of the eye lids were reported, due to the use of eyelash curlers and to the presence in them of the antioxidant, N-phen y l-2-naph th yl am i n e (Curtis, 1945). In a case report of 42 individuals with dermati tis due to rubber gloves, m ercaptobenzothiazole, TMTD and dipentame thyl e n e thiur am disulphide were identified as the causative agents (Wilson, 1960). Of 100 cases of 'shoe dermatitis', 94% were attributed to exposure to m ercaptobenzothiazole or TMTD or both (Cronin, 1966). Another report indicated that TMTD was res ponsible for a number of cases of rubber-glove dermatitis (van Ketel, 1968). Another series of 104 cases were attributed to exposure to thiuram accelerators or mercaptobenzothiazole (Wilson, 1969). N-lsopropyl-N'-phenyl-para-phenylenediamine has been reported to be responsible for 23 cases of dermatitis following exposure to rubberized fabrics (Batschvarov & Minkov, J j ( ( f \ 164 I ARC MONOGRAPHS VOLUME 28 1968) and for dermatitis due to contact with automobile tyres (Jordan, 1971). It has also been reported to produce crosssensitivity to the hair dye, para-aminodiphcnylamine (Sch^nning & Hjorth, 1969). Alkyl phenols and hydroquinone have been reported to cause leucodermas in rubber workers (Calnan, 1973). Phenylenediam ines in rubber products are common causes of dermatitis (te Lintum & Nater, 1974; Nater, 1975), and a cross-reactivity occurs between different phenylenediam ines (Rudzki et al.. 1976). 4,4'-Dithiodi- morpholine, another rubber chemical, has been reported to be a skin sensitizer (Heydenreich & 0lholm-Larsen, 1976). Sensitivity to N-iso pro py I- N '-phenyl-par a- phenylencdiaminc. N-phenyl-N-cycIo- hex yl- par a-phenyl ened iam ine and N-dim ethyl-1,3-butyl-N'-phenyl- para-phenylenediam ine has been reported in workers handling tyres. All subjects sensitive to the first were also sensitive to the second, and 3796 of them to para-phenylenediamine. The last- mentioned was the most powerfully allergenic compound (Herve- Bazin et al., 1977). Disulfiram (TETD) and thiram (TMTD), like most dithio-- carbamates, alter ethanol metabolism (van Logten, 1972); there has been a report that in three patients with eczema from rubber gloves the condition was aggravated after consumption of ethanol (van Ketel, 1968). Superficial keratitis has been observed in rubber workers exposed to ethyl isothiocyanate (Groves & Smail, 19,69). L The International Contact Dermititis Research Group has proposed the following patch-test allergens in relation to exposure to rubber chemicals: * N-< hyd 1.3 phe N4.4 2-r he> tet bis me tet phi dip dib mc tei dip d ibis 2.2 R_e In factorie found a rubber had bro In taicosis ex po s ui foot) * that tir tvres < sning & I cause iam ines turn & etween thiodibe a sitivity cyclohenylndl in g ive to lastferve- ithiother e rubber thanol .r rs J n BIOLOGICAL EFFECTS N-cyciohexviben zothiazylsulphenamide hydroquinone monobenzylether I ,3 -- diphenylguanidine phenyl-3-naphthyl am in e N-phcny l-N '-isopropyl-- para-phenylenediam ine 4,4'-dihydroxydiphenyl 2--m ercaptobenzimidazole hexam ethylenetetramine tetr am ethyl thiuram monosulphide bis(diethyldithiocarbamato)zinc mercaptobenzothiazole tetram ethyl thiuram disulphide phcnylcyclohexy I-par a-phenyl enediam ine diphenyl-para-phenylenediam ine dibenzothiazyldisuiphide morpholinylmercaptobenzothiazole tetraethylthiuram disulphide dipentam ethyl enethiuram disulphide di-a-naph thy I- par a-phenyl enediam ine bis( dibutyldithiocarbam ato) zinc (Malten et al.. 1976).- 165 2.2 Respiratory Vffects In a study of 239 Egyptian rubber workers In three factories, chronic, bronchitis and chronic bronchial asthma were found among mixing and compounding workers exposed to dust and rubber additives, while workers exposed only to by-product gases had bronchial asthma (Noweir et al.. 1972; Osman et al.. 1972). In a Swedish report (Fristedt et al., 1968), five cases of talcosis were- described in one rubber factory. The mean average exposure to respirable dust (37-66 million particles per cubic foot) was twice to three times above the occupational standard at that time, which was 20 million particles per cubic foot. The substitution for talc of powder with a fine particle size and containing a high content of crystalline silica in a rubber processing plant resulted in the development of silicosis in a number of exposed workers. The average time of exposure was 6.8 years (Gaubatz & G aubatz-Trott, 1973). 91030215?: BFG39164 166 IARC MONOGRAPHS VOLUME 28 Workers exposed to curing fumes had a higher prevalence of chronic bronchitis than controls. Respiratory morbidity was related to both intensity and length of exposure to fumes. In addition, forced vital capacity was significantly decreased in curing workers; and those workers re-examined oge year later suffered an excessive loss in pulmonary function compared with those not exposed (Fine & Peters, 1976a,b). Sixty-five men exposed to dust in the processing area of three rubber factories had a higher prevalence of chronic pro ductive coughs and a decrease in the ratio of forced expiratory volum e tforced vital capacity when compared with controls. The effects on pulmonary function were related to length of exposure. The authors concluded that exposures to processing dusts at the level measured produced chronic respiratory disease (Fine & Peters, 1976c). Eighty talc workers were also studied in_ the three rubber manufacturing plants. Although exposure to talc was below the current threshold limit value (20 million particles per cubic foot), workers exposed for 10 years had statistically signific antly decreased forced expiratory volume and ah increase in respiratory morbidity, despite the absence of changes in chest X-rays (Fine et al., 1976). Rubber workers exposed to the hexamethylenetetramineresorcinol adhesive system had an excess of acute respiratory symptoms and statistically insignificant reductions in forced expiratory volume and forced vital capacity. However, statis tically significant decrease's in pulmonary function were seen over a six-hour period of a work shift (Gamble et al., 1976). A questionnaire on respiratory symptoms was answered by 1820 of 2856 white male production workers. The data were analysed according to where the individuals worked in rubber production. The prevalence of respiratory symptoms was high in those workers currently involved in milling, tube building, tube curing and tube inspection (McMichael et al.. 1976). Another study of chronic respiratory disease in therubber industry was based on a cohort on 4302 male rubber workers, of whom 73 terminated their employment with pulmonary disability. These 73 workers had speni significantly longer in curing preparation, curing and finishinj and inspection work areas, each of which implied exposure tc particulate material and/or to solvents. The authors concludec 25208017 BIOLOGICAL EFFECTS 167 that a significant risk of developing a pulmonary disability was associated with smoking and exposure to dust and fumes, and in particular to talc and carbon black (Lednar et al., 1977). Two further studies (Weeks et al., 1981a,b) revealed that employment in curing departments was associated with shortness of breath, chest tightness, wheeze, reduced forced expiratory volume and forced vital rapacity and a decline in the ratio of forced expiratory v ol u m e : f o r c e d vital capacity. In addition, tingling and numbness occurred in the extremities. All of these associa tions became stronger with increase in duration of employment in the curing department. Exposure to respirable particulates causes not only increased symptoms of respiratory disease but also nausea and abdominal pain. Workers exposed to emissions from heated uncured rubber reported chest tightness on return to work. At a US tyre factory, 210 workers (172 men and 38 women) developed a syndrome related to exposure to the volatile products released from a synthetic rubber formulation, first used in 1960. The factory employed approximately 2200 workers at the time of the outbreak, about 600 of whom were repeatedly exposed to the new formulation in their daily work. Symptoms occurred most fre-' quently in workers in the B a n b ur y- c al e n d er and tyr e-buil din g areas, where 3296 and 2296, \ respectiv ely, of the workers were afflicted. The symptoms included severe respiratory irritation, weakness, weight loss and general malaise. Fever and radiographic and pulmonary function abnormalities were also noted in many of those affected. More than 1796 of the affected employees were permanently disabled. A simitar syndrome was reported in at least four other factories in which the adhesive system had been intro duced. After the system was modified, in late 1962, the preva lence of symptoms remained low, and no new cases appeared (doPico et al.. 1975). Chronic lung diseases (pneumoconiosis, pulmonary fibrosis, bronchitis and emphysema) occur in workers exposed to carbon black (National Institute for Occupational Safety & Health, 1978); and a higher incidence of pneumoconiosis has been reported in those exposed to channel black than in those exposed to thermal or furnace black (Troitskaya et al., 1975). 10 CvT 168 IARC MONOGRAPHS VOLUME 28 2.3 Effects on reproduction Data on hospitalized patients who had had miscarriages (spontaneous abortions) were analysed in conjunction w'lth member ship files of the Union of Rubber and Leather Workers (about 10 000 women) and records of the personnel of a rubber factory (about 1600 women) in Finland. Spontaneous abortions were calcu lated in two ways for each population analysed: rate (no. of spontaneous abortions x 100/no. of pregnancies) and ratio (no. of spontaneous abortions x 100/no. of births). The two frequencies were slightly increased for all Union members, as compared with all Finnish women. Moreover, age-standardized frequency was higher when the pregnancy occurred during Union membership than when it occurred before or after joining the Union. The frequency of spontaneous miscarriages among Union members employed only in the rubber industry was higher among those emplpyed for 3 to 35 months than for those employed for longer periods of time. The employees ^f the rubber factory had slightly fewer spontaneous abortions on average than did the community population. However, women employed for 3 to 23 months had markedly higher frequencies of spontaneous abortion than those employed for longer times (Hemminki et al.. 1982). (Because the data in these two studies present a conflicting picture with respect to the relation between rubber work and spontaneous abortions, no conclusions can be drawn.] 3. Summary The number and diversity of chemicals used or formed and the multiplicity of exposures experienced within the rubber industry make it difficult to attribute a particular toxic effect to a given exposure. Carbon blacks and mineral oils are major constituents in such products as rubber tyres. Data on the carcinogenicity of carbon blacks in experimental animals are inconclusive. Carbon blacks contain varying amounts of polycyclic aromatic compounds, some of which are carcinogenic and which may become bioavailable through elution. Solvent extracts of carbon blacks have been found to be mutagenic and carcinogenic in experimental systems. The mutagenicity and carcinogenicity of mineral oils are also 25208C19 BFG39167