Document 44M5g6ka8YkrY7wJ3m9nYk9Op

I Progrtu in Mutation RtHareh, Vol 2 p 'JJ? A, Kappat, tdttor. - -------- o ISM/ Kliun/tor/Nurllt-Jltfilutul lliumvdiculi'ruu . 'ttCEfVtQ MAY 10 1982 SARR [LOW RELEVANT AUK HIGH DOSES IN MUTAGENICITY AND CARCINOGENICITY STUDIES IN ANIMALS? H. CHEIM, U. ANDUAE, W. O&GQBLMANN, S. HESSE, L.ft. SCHWARZ and K.H. SUMMER i Department of Toxicology, Getellechafl f&r Strahkn- und UmwtUfortohung, D-&042 Seuherberg (F.R.G.) Summary Toxicity, including genotoxicity of chemicals, is frequently affected by metabolic conversion of the chemicals by the host, leading to inactive deriva tives. Inactivation mechanisms can be overcome by high doses of a chemical resulting in a greatly altered dost--effect relationship and drastically increased toxicity. This is exemplified by the dose<iependent effect of chloroform and 1,1-dichloroethylene in animals, by investigations on the glutathione-dependent mutagenicity of chlorodinttrobenzene in the Ames test, and by the effects resulting from inhibited metabolic inactivation of naphthalene in isolated hepatocytes. At tow doses genotoxic effects of chemicals may be prevented by meta bolic inactivation, e.g. chloroprene is mutugonic in vitro but not carcinogenic, possibly due to glutathione conjugation. Moreover, disputable data on muta genicity or carcinogenicity of styrene must be judged in that its mutagenic metabolite styrene oxide is inactivated by efficient cellular epoxide hydrolase activity in vivo. Hepatic monoxygenase activity creates mutagenic hydrogen peroxide which is trapped in vivo by catalase and glutathione peroxidase. Thus, information on the mechanism and extent of inactivation is essential for extrapolation from high dosee of chemicals used in toxicity studies in animals to low doses that are relevant to human exposure. Introduction For risk estimeiion of man's exposure to carcinogenic chemicals it is cus tomary to extrape ate from high doses at which tumor induction is observed in animals to low doses in the human environment. However, there is increasing evidence that several carcinogens induce other biological responses at low doses than those at high doses. Furthermore, species differences in the metabolic acti vation and inactivation processes may also question a linear dose--effect relationship. j apoooio632 130 An evaluation of these facts is essential because one has to be aware that man is continuously exposed to small amounts of mutagens and carcinogens. Commoner et al. (1978) reported the presence of mutagens in broiled ham* burgers; the presence of mutagenic activity in charred fish and beef was pre viously reported by Nsgao et al. (1977) and Sugimuraet al. (1977). Such muta gens apparently stem from pyrolisation of the amino acids tryptophan and glutamic acid (Sugimura, 1978). Nitrosamines present in food (Spiegtlhalder et al., 1979; Scanlan et al., 1960) are constantly formed in the human stomach when secondary and tertiary amines axe ingested together with nitrite (Lljinsky, 1977; Tannenbaum, 1980). Chloroform has been used for many yean in tooth pastes and in cough syrup formulations and it is a drinking-water contaminant whenever chlorine ia added for hygienic reasons (Fishbein, 1979; Stockinger, 1977). Sugimura recently evaluated that by the age of 50 a penon has eaten approx. 10 tons dry weight of food (Sugimura, 1978). Even though concen trations of environmental contaminants in food arc very low, their continuous intake could be sufficient to cause genetic damage possibly resulting in the appearance of cancer. Models to extrapolate from high to low doses and between species Animal studies to investigate the potential hazards of such chemicals are usually performed at relatively high levels of exposure (Crump et al., 1976; Cornfield, 1977; Scientific Committee, 1978; Maugh, 1978; Tomatis, 1979; Carter, 1979; Interagency Regulatory Liaison Group, 1979). In general, human risk is determined by extrapolating from the high doses used in the animal experiments to the much lower levels to which man is actually exposed assuming that no threshold exists. This is bused on the concept that cancer Is induced after the reaction of a single or several molecules with DNA or other macro molecules in a cell. Once such a reaction has occurred the cell is damaged irreversibly and initiates a new disorganized cell line which finally produces the tumor. This mechanism implies that there will be a certain tumor incidence in a population no matter how long the dose of the carcinogen has been. In this case, cancer incidence at low doses will be a linear function of the high doses. The formula generally applied in extrapolation from high lo low doses is (Reitz et al., 1978): Linear P & X d where P is the fractional incidence of excess tumors, d the average daily dose in mg/kg and 0 is a dimensionless parameter to be estimated from the experimental data. Additionally, a species correction factor can be applied to ff which has been chosen as the cube root of the ratio of the body weights to calculate the dose per surface area. This always implies that man is considered to be more sensitive to a carcinogen than mice or rats. Rail (1960) established this correlation with 18 anti-neo plastic agents. They produced, however, direct toxicity, and Rail stated that these compounds are not Involved in the various drug-metabolizing systems. However, there is ample evidence that most of the environmental chemicals undergo metabolism by the host, and it becomes evident from this that the rate of metabolism must be included into the process of risk estimation. Reitz et al. (1973) suggested therefore that the relative rates of production of toxic metabolites be introduced into risk estimation. And, indeed, relative 131 carcinogenicity of chloroform, vinylidene chloride, 2-acetylaminoQuorene end trichloroethylene In rats and mice correlated well with the relative rate of metabolic activation to a reactive intermediate of each of these chemicals. The resolution of these approaches by animal experiment* teems to be impossible. To decide which of the correction factors is correct would mean that data on the response to low doses of the chemical concerned would have to be collected. But this would moan measuring low incidences of cancer in animal populations, say 0--16%, and these low incidences can only be dis tinguished with any degree of statistical certainty from the natural rate of cancer in animals by trials involving impossibly large numbers of both control and treated animals. Thus, in practice, only relatively high doses can yield satis* tically significant data. But in many cases such high doses may produce cancer simply because their very quantity overwhelms the biochemical pathways that would detoxify smaller, more realistic doses. Since no conclusive solution of the problem can be obtained from the standard carcinogenicity or mutagenicity test procedures the only approach to solve the problem should be the evaluation of mechanisms which may affect the Lumor inciUoneu at low and high doses of carcinogens. This may provide a more solid basis for the presently highly speculative end uncertain quantitative risk assessment. In general several mechanisms are considered to be involved in the inacti vation of chemical carcinogens. Important contributors to these mechanisms are membranes that may not permit potential carcinogens to come into contact with DNA and enzyme systems for repair of damage to DMA. Some evidence also suggests that the immune system can identify and destroy many kinds of operant ceils produced by chemicals that slip through the other defenses. Moreover, enzyme systems for metabolic toxification and detoxification modify the biological impact of chemicals and have to be considered in the course of risk estimation. This is explained in moru detail by discussing the following points: (1) Is there any indication of threshold doses of carcinogenic chemicals in man due to metabolism? (2) Arc the high doses of carcinogens used in animal studies generally repre sentative for the low doses to which man is exposed? (3) Does metabolism result in qualitative and quantitative differences in the susceptibility of different species to carcinogens? Indication of threshold doses due to metabolic inactivation (a) Carcinogenicity of ettrone The hormone estrone, for example, has been shown by several investigators to be carcinogenic when given to laboratory animals in large doses (IACR, 1974). However eotrone is present in very small concentrations in all humans, yet without demonstrable evidence of harm. It is present in much higher con centration* in women than in men which might bo at least part of the cause of the much greater incidence of breast cancer in women. Moreover, glucuronidation of estrogens seems to be a significant inactivating mechanism. A recent report by Fishman et al. (1979) indicates a correlation 132 between low glucuronidation capacity and the occurrence of breast cancer in women. These authors studied steroid excretion In the urine of 30 young women who by genetic analysis were at risk for familial breast cancer as com* pared to a carefully matched control group. Highly significant differences were observed in the excretion of giucuronides of estrone and estradiol with the high-risk women exhibiting lower values than the controls. Besides the finding of a potential discriminant for identifying women at high risk for breast cancer, tills observation strongly indicates that the status of inactivation mechanisms affects susceptibility to potential carcinogens. (b) Increased DNA-repalr by reactive oxygen species Recently, we demonstrated that reactive oxygen species, which are formed during monooxygentse-mediated metabolism of chemicals, induce DNA-repair activity in the human lymphoblaatoid cell line NC37 BaEV (Andiae and Greim, 1979). This was observed in the course of measuring the induction of DNArepair replication by mutagens requiring metabolic activation (Fig. 1). Mute. o to *0 M 40 M Xrattwn Wit. 1. IMpair rtpUcalton in NC37 htv Us. Cultutaa wort (Mitoatad la to* prttaaaa aX l mM 4-Quoio* eaa*yurtdifia and 10 MM Mtemodaoxyuridlna tor 1 H. Aflar addiliaa of 2.5 mM hydrosyurta tall* warn ImuImM for 3 h wtUi At NADVUi Ui 0,0 m Uvar mlBnawMiw of Ctapkan A SO pnUiitiX rats r NAUfU; C- Uvar miemomn * NADPK * IViiU b*ntUlpyNMiI)! UvwaitMMm * NADfH * 45 min mM dimathytaitroaaiatoa. and labtlad for 4.5 h with 10 mCJ (*H)thynidina (iOCi/mmola). Call* rtn Irate to 0.M MdiuadodaayUtydrotanaulfaU, diftatad with ptotatoam K (50 w(ml) and *adJ- manted in alkaUna CaO-cradtanu (Aadna and Crain. 1979). Ttia location of ['Xldabal at aamal danaliy (eo>neidttt with too ohaorbaaeo pk) w indicativa tor repair reputation, Hi gen* luch as benzo [a] pyrene end dimethylnitrosamine induce repair activity. However, in control experiment* without the mutagen a marked incorporation of repair label was observed. The incubate was comprised of washed microsome* of phenobarbital-treated mice, NADPH and hydroxyurea which is used to suppress semiconservative DNA synthesis. When this inhibitor was omitted, no repair replication occurred. Hydroxyurea-induced repair was dependant on the presence of microtomes and NADPH, and was reduced to 60% by the meta bolic inhibitor SKF 626-A. This highly indicates that hydroxyurea-induced repair is a consequence of microsomal monooaygenase-mediated metabolism. Since addition of glutathione (GSIi) and superoxide dlsmutase reduced and catalase prevented hydroxyurea-induced DNA repair, hydrogen peroxide (HjO,) or reactive oxygen species deriving from H,Oa axe likely to be the genotoxic agents. As HjOj production generally occurs during monooxygenase function (llildcbrundt and lloots, 1975; Jones ct al., 1978) wo investigated tho effect of ethylmorphine. Accordingly, ethylmorphine induced DNA repair in tho test system but the effect was prevented in the presence of catalase (Fig. 2). Other in vitro coll systoms, c.g. human A649 lung tumor cells, apparently have a sufficient HjOi-inactivation capability as these systems did not show enhanced repair activity. The precise mechanisms by which reactive oxygen species are trapped in vivo remain unknown. GSH peroxidase is likely to be involved. It is evident that efficient HiOi-inactivation must be present in vivo La prevent ruuelivu oxygen species from exceeding a no-effect concentration and thus preventing genotoxic effects. (c) Chloroprene: Mutagenicity but no carcinogenicity Further evidence for no-effect levels due to metabolic inactivation is given by toxicity studies on chloroprene (2-chloro-l,3-butadiene). Ettymorphrva (rM) Kit. 1. K4t>ii tapttaatiM In NC37 BaEV Mila Uv Ova pcaaanea p{ llvai mictoaomaa iaolstaS teem pbanobarlMUi-imMd miaa. NAUXH and aUtylmoipttna. Fat tlauila in lasaad to FI*. 1. / AP000106J6 IJ'I Chloroprene is a reactive chemical which is widely used in the manufacturing of the synthetic rubber neoprene. Chloroprene has been suggested to be respon sible for the increased incidence of skin and lung cancer in workers exposed to the chemical. However, the data are questionable and no carcinogenic effects of chloroprene have been noted in animal studies to date (Fishbein, 1979; Ponemarkov and Tomatis, 1980). Bartsch et aJL (1975, 1979) have demon* strated a slight mutagenicity of chloroprene in Salmonella typhimurium strains without metabolic activation. Mutagenicity was increased about 3*foLd in the presence of S9 fractions. These observations were considered to reflect the probable formation of an epoxide intermediate of chloroprene. Why was chloroprene shown to be mutagenic but not carcinogenic? In analogy to vinylchloride and 1,1-dichloroethyIene, a subsequent detoxification of the chloroprene by conjugation with glutathione has been suggested (Haley, 1978; Plugge and Jaeger, 1979). Since in vitro mutagenicity test procedures do not include sufficient glutathione (Summer et al., 1980), the apparent dis crepancy of mutagenicity in vitro without producing carcinogenicity in the animal may be explainable. We tested the possible involvement of glutathione in inactivation in isolated rat hepatocytea and in the whole animal (Summer and Greim, 1980). Similar to the rat Uvur, cellular glutathione decreased in isolated rat hepatocytes to about 50% of die control values within 15 min in the presence of 3 mM chloroprene (Fig. 8). l*his depletion was concentration-dependent and increased with time. In hepatocytes of animals pretreated with phenobarbital or Clophen A50, 3 mM of chloroprene almost completely depleted glutathione within 30 min. This strongly indicates that a Phase I reaction, presumably epoxide for- Fig. 3. OluUlhion* dopfetion la bobud tat tupatoertot by ohleroproae. Dote an from 1 ropr**nuUv *poiiiBnt out of 3 wttfe different eti propaationa. Conuol l*vl of wlliilw OSH imnunfd to 1.9.9 * 1.1 (uaolMfois ctu ptotoia. CIom4 symbol*; Hpatocytti (S mg eU ptoMkaJoU) of unutd iomtawd with 0.S mM (>. 1.0 mM (t) and 3.0 mM (*) ehloroprtn*. Opn tyrnfelis HipiMeCM from eithtr pkwnaburbiUI- (0) or Cloptwn A SO* (a) promoted onimab ifkcuboteil wttb 3 mM obkosoprona. m mation which U enhanced after Clophen A50- or phenobarbital-prstreatment, precedes glutathione conjugation. Glutathione-dependent detoxification in the animal was further verified by determining urinary thioether excretion comprised of GSH conjugates and mercapturic acids. Chloroprene administration to rats resulted In a dose-dependent increase in the excretion of urinary thioolhen (Fig. 4). This increase was reversible and completed within 24 h at ail dose levels administered. At dose levels of 50 and 100 mg chloroprene/kg, the additional excretion of urinary thioethers was almost 200 and 450 amoles/kg daily, respectively. It is to be noted that no linear dote--response relationship was observed in thioether excretion, Indicating that at higher doses of chloroprene the availa bility of cellular glutathione becomes rate-limiting. Thus, a dose level at which high concentrations of chloroprene overcome GSll-inacUvation is suggested. Only sufficiently high doses are expected to induce carcinogenicity. They have not been used so far in carcinogenicity studies. Effect* of high doses on metabolism (a) Overwhelmed metabolic inactivation The latter observation suggests that high doses of a chemical may have other biological effects than lower doses by overcoming metabolic inactivation mechanisms. VLnyiidene chloride (1,1-dichloroethylene) is also conjugated to glutathione. In the rat, glutathione is depleted by exposure to high levels of 1,1-dichloroethylene (Jaeger t al. 1974). Fasting of rats prior to 1,1-dichloroethylene exposure further depleted the glutathione and, at the same time, dramatically increased the hepatotoxicity of the chemical. Furthermore, McKenna et al. (1977) reported that a sudden disproportionate increase in macromoJocular binding of 1,1-dichluroethylene metabolites occurred when glutathione was depleted by more than 30%. These observations are strongly in favour of the protective function of such inactivating systems. Fit. 4. CMoropzM*e*pndnt as*r*un f thloUtr la tha urtn* of nu. Chloreprao* vu dmlaiaund in oiiv oil by itomaclt tub*. Data msmmiH nnit t J.O. from 4 animal*. 1 AP00010638 (2) Naphthaline: Protein binding and conjugation Isolated hepatocytes metabolized naphthalene to water-soluble compounds (Schwarz et el., 1980). Metabolism of the hydrocarbon was linear for 1 h reaching a plateau after 2 h. During biotransformation of naphthalene reactive intermediates were formed which became irreversibly bound to cellular protein. Binding almost paralleled the increase in the formation of metabolite*. For mation of water-soluble compounds and binding was due to metabolism since frozen-thawed cells in the presence of 1 mM SKF 52&-A showed neither formation of water-soluble compounds nor binding to cellular macromolecules. Conjugation with UDP-gIucuronic acid and sulfate is one of the main path ways of naphthalene metabolism in hepatocytes (Bock et ai., 1976). We In hibited these conjugation mechanisms by interfering with the synthesis of their respective co-factors. Addition of D-galactoaamine reduces levels of uridinediphosphoglucuronic acid by trapping uridine-triphosphate and by inhibiting UDPG dehydrogenase activity (Bauer and Reutter, 1973). Sulfation is inhibited by incubation of the cells in sulfate-free medium which decreases synthesis of 3'-phosphoadenoslne-S-phosphosuIfate (PAPS) and, thereby, sulfation (Schwarz, 1980). Incubation of hepatocytes in a sulfate-free medium in the presence of 3 mM D-galactosamine did not affect formation Of water-soluble metabolites from naphthalene. However, a drastic increase in covalently bound metabolites was demonstrated (Fig. 6). U is evident from this thul cellular formation of reactive and toxic species of chemicals is frequently counteracted by metabolic detoxification. A dis proportionate increase in the toxic effects can occur when high concentrations of chemical are present which overcome inactivation processes. In toxicology, such threshold levels on the basis of overwhelmed inactivation have long been established. For example, furosamide, a diuretic, is excreted predominantly unaltered in the urine when low therapeutic doses are given. Administration of high doses, however, overwhelm renal clearance and cause a disproportionate increase in toxic metabolites which react with macromolecules. Moreover, bromobenzene is metabolized in the liver to the reactive 3,4-bromobenzene oxide. This molecule Is detoxified by glutathione con ns. S. ItfMt or S mM I>sslattM*miAS (D*) and suifaie dtpIrUM oa Imvtttible blading ( ['*01 MptailM*. HipiiMfW won pitputi Cron 200 Hpiagiie Qawlar mu by is tUu pnluiiPi Wits eoUaieitaee. RiptMVtM (1.28 X 10* eeBa) were incubated is Weymouth US T8*sl medium in iplmn flask*. Mttaboil kb we* ttooped with tcwdold 4-ftauM.i-WaUr soluble metabolite*" of aapblbaleae were quantified by Liquid-ecinUlUtioa counting ol the water pktaae piu* nattUaUscd NaOH eatract ot the orgaaiq Phase containing the pknali.1 product*. To determine izrevetable binding ol "C radioactivity. O.S ml of the eeU luapendon waa tonified twice wltb 0.9 mi ethanol. The niheequent tclveat estraetions were performed aa previously described {Kerne et at., 1970. 138 jugation. Whan high doses of bromobenzene are given, glutathione is depleted and the reactive metabolites interact with cellular macromolecules (Gillette, 197-la, b). Other chemicals for which there is evidence that high doses cause a disproportionate increase in toxicity and possibly carcinogenicity are aspirine, salicyiamide, acetaminophen, styrene, ethylene glycol and aniline (Gehrlng et al., 1976), (b) Impaired metabolic actuation On the other hand, potential carcinogens not only overwhelm inactivating capabilities but also inhibit metabolic activation reactions. The animal bio* assay results on vinylchloride of Maltoni (1977), Lee et el, (1978) and Viola et al. (1971) suggest that inhalation beyond 150 ppm results in a smaller increase in carcinogenic response in rodents than that predicted from a linear extrapolation of the responses in the dose range of 0--150 ppm. Recent pharmacokinetic studies indicate that at high doses of vinyl chloride the mecha* nisms which metabolically activate this chemical axe saturated. Bolt et al, (1980) observed no further increase of DNA and protein-bound vinyl chloride metabolites in rats beyond a certain level of exposure. A similar effect is reported with carb-TM hy Ugazio et aL (1972). A preceding low dose of CClnignificanuy increased the LDt of rats when a high dose was given thereafter. Modification due to species differences in metabolism Moreover, species differences in the metabolic activation and inactivation systems further complicate extrapolation from high doses given in animal studies to the low doses to which man is exposed. Several well-established examples indicate qualitative and quantitative species differences in the activating and inactivating metabolism interfering with the capability of the reactive intermediates to bind to the genetic material. (a) Species differences in metabolic activation (i) X'AcelyUttni iofluoretia For example, & strong hepatocarcinogen in male rats, is without effect in guinea pigs. 2-AAF is an indirect carcinogen. It requires aromatic iV-hydroxylatlon forming the iV-hydroxy-2-AAF (Miller ut oi., 1961). Since guinea pigs do not have this enzyme this species is insensi tive to the carcinogen although they develop tumors when the N-hydroxy product is given (Miller et al., 1964). Other species differences in the carcinogenicity of 2-AAF cannot be related lo this activating mechanism. However, carcinogenicity of 2-AAF is also dependent on the formation of a sulfo-derivative. This reaction is catalysed by a .V-sulfotransferase which differs quantitatively in several animal species. DeBaun et ai. (1970a, b) determined the activity of this enzyme and were able to explain further the species differences of 2*AFF carcinogenicity (Table 1). Male rats, which axe highly sensitive to this carcinogen, possess a high sulfo- transferase activity. It is low in female rats and in mice of both sexes which are 139 TABUS l HEFATOCaXCIHOGENICITY OF 3-AC3TYLAMINGFLUOIIENC AS ACLATKO TO 1ULFO-TXANSFERASE ACTIVITIES IN SEVERAL. SPECIES Du IMvn el al,, 1 I10as b. HtiwiwlMiwtoUy iHUfitutinm Mtivity1 Rat Mowie OulnM pig Rabbit M +++ F M r* M .U _ ... . -- * itmelu/0-04 at wfeuiMt ft tO min. 71.0 4.0 04 04 04 xo .. nfW' As*mMbf. ~ _j' s. ** *'-?.* less susceptible than the male rat. The other species tested which have moderate or lower sulfotransferase activities are not known to produce hepatocarcinoma due to 2-AAF. (2) Chloroform Moreover, not only qualitative but also quantitative differences in the meta bolizing rate exist between the different animal species and man. Several inves tigators have studied the metabolism of chloroform in rats and mice and have concluded that mice metabolize chloroform more rapidly than do rats (Fohl, 11)79). Thu most -unipiulu studios are thosu of liruwn cl ui. (1974) who administered a 60 mg/kg dose of chloroform orally to rats, squirrel monkeys and 3 strains of mice (Table 2). Very little unmetabolixed chloroform was rovovuruU from the mice, in Uw rut, 3 times us much unmvlabolixed chloro form was found, while in the monkey 13 times as much unaltered chloroform was exhaled. Fry et al. (1972) administered chloroform to human volun teers and found that 17--66% of the material' was exhaled unchanged. However, the dose of chloroform used in these investigations was only 7 mg/kg. Thus, the relative amount of chloroform that would be metabolized at the higher doses used in the animal experiments may even have been overestimated. Based on these investigations, Reitz et al. (1978) suggested man to be the least sensitive of the species to the carcinogenic action of chloroform. This is also based on the data of Weiss et al. (1977) who observed that man normally metabolises TAHLSS* iNTcasvtcucs differences in chloroform metabolism as determined ky urinary excretion or the unchanged chemical Onldnaa (ag/kt) Unchanged CHO, axmwd (O of 40) Mouaa Hat Monkay Man SO 00 0 T SO fl iT-e Modlfiad teas Riiu n *1. U*tS). / APOOO10641 140 material* much more slowly than the small laboratory animals such as mice or rau. Consequently, a direct extrapolation from carcinogenicity teste on chloro form in mice or rats would overestimate the risk of man to this potential carcinogen. (3) Halogenated ethylene* In addition to chloroform, there are several other examples where the rela tive carcinogenicity of chemicals correlates well with their rate of activation to a reactive species. 1,1-Dichloroethylene is metabolized more extensively in the mouse than In the rat (McKenna et al., 1977). Correspondingly, tumors have been observed in mice exposed to 1,1-dichloroethylene (Maltoni, 1977), whereas in rats, although conflicting results appeared, no tumor formation seems to be related to this chemical (Fishbein, 1979; Maltoni, 1977; Rampey et al., 1977; Maltoni et al., 1977). Trichloroethylene is carcinogenic in B6C3F1 mice but not in Osborn-** Mendel rats (Natl. Cancer Inst., 1976). This again correlated with the greater uupaclty of mouse raicrosome* to catalyze binding of trichloroethylene metabolites to DNA (Bannezjee and van Duuren, 1978). These data indicate that the status of metabolic activation systems in the different species significantly affects susceptibility to carcinogens in the animal and explain species differences. (b) Species difference* in metabolic inactivation (1) Afiatozin B, In addition, there is also evidence that species differences of inactivating mechanisms such as binding of the reactive intermediates with glutathione modify adverse effects of chemicals. The hepatotoxic and carcinogenic effects of aflaloxin Bi are attributed to the reactions of metaboiically formed AP oxide (Croy et al,, 1978). The marked species differences in susceptibility to carcinogenicity could not be related so far to the different ability to generate the epoxide. Recently, Degen and Neumann (1978) found that AF B(-epoxide is inactivated by conjugation with glutathione. Degen (1979) has since reported that mouse-liver preparations most effectively catalysed the formation of the glutathione conjugate. These result* support the view that the lower suscepti bility of mice is not a result of leu effective activation but rather of more effective inactivation of the reactive AF Bi intermediate due to the formation of glutathione conjugates. (2) Styrene epoxide hydrolase Oesch (1980), Oesch et ai. (1978, 1974) and Jerina et al. (1977), studying the effects of microsomal epoxide hydrolase activity, further stressed the importance of considering species differences in metabolic inactivation. The hydrolase generally converts reactive epoxides into more stable dihydrodiols (Jerina et al., 1977) which are frequently further conjugated with glucuronic acid (Oesch, 1980). For example hydrolase activity is involved in the biotransformation of styrene (Leibman, 1975). It is highly suggested that hydro lase activity being present in the S9 fraction prevents styrene from being muta- 141 genic in the pUite*incorporation assay (Greim et aL, 1977) although styrene oxide, the metabolite which is formed during metabolic activation of styrene, is a potent mutagen in this test (Miiy and Garro, 1976). Epoxide hydrolase generally plays an important role in the inactivation of many aromatic and oleflnic compounds of industrial interest. Since man hu a much higher hydrolase activity than most of the investigated laboratory animals (Oesch at aj., 1974) it is to be expected that man is the least eensitive to such compounds. iv > We can oonclude the following from the experimental data priftsnted:Mt- abolism plays an important role in the carcinogenic ox mutagenic acuity of chemicals. It is species-dependent, can be influenced by the doseof t&tchemical administered, and thus has to be considered hi extrapolationfromdm high doses given to the experimental animal to the low doses to which man is exposed. Regarding metabolism, roughly 4 groups of chemicals can be differentiated: (1) Compounds which undergo neither metabolic activation nor significant inactivation, e.g. methyl methanesulfonate. (2) Compounds which only undergo metabolic activation, . dimethyl- or diethyl-nitrosamine. (3) Compounds which only undergo metabolic inactivation, e.g. tf-methyl//'-nitroW-nitrosoguanidine. (4) Compounds which undergo both metabolic activation and inactivation, e.g. vinyl chloride or chloroform. For risk estimation, it is apparent that linear extrapolation from high to low doses and from tne animal species to another is only possible for compounds of Group One which require neither metabolic activation nor inactivation. Apparent species differences in the metabolic activation and inactivation system have to be considered concerning compounds of Groups Two to Four. Extrapolation can be further complicated when high doses of the chemical inhibit or overcome metabolic activation or inactivation. Carcinogen* acting by socontlary mechanisms Another group of compounds induces tumors by secondary mechanisms, for example, by suppressing production of hormones or by inducing necrosis at high doses whioh finally results in tumor formation. Several chemioda inhibit function of the thyroid gland. These goitrogeru induce tumors by a secondary mechanism. Thiourea blocks thyroxin synthesis by specific inhibition of iodine peroxydation {Davidson et aL, 1979). As a consequence, insufficient amounts of thyroxin are produced by the thyroid gland leading to an Increased hormonal stimulation of the thyroid by the hypophysis. This results in a hypertrophy of the thyroid, and after continuous exposure, thyroid adenoma appears (1ARC, 1974). Finally, carcinomas of the thyroid are observed. At the lower doses of thiourea when no hypertrophy of the thyroid is observable, carcinomas of this organ have not been detected. Carcinomas of the thyroid have been detected in rats and mice receiving a diet of 0.26% thiourea (Purves and Griesbach, 1947). This dose is equivalent to approx. 100 mg/kg daily. When administered as a drug, repeated daily intake AP00010643 142 of approx. 10 mg/kg was goitrogenic. Presently, thiourea is used as an industrial chemical only and sufficient measures to protect man at his working place have to be provided (Fishbein, 1979). However, workers would not be endangered by thiourea exposure unless exposed to concentrations affecting thyroid functions which can cosily be detected. Nitriiotriacedc acid and chiorothalonll, both excreted almost quantitatively via the kidney, induce urinary-tract tumors which have been preceded by tubular necrosis (Natl. Cancer last., 1977; Anderson and Kanerva, 1978; WHO, 1975; FAO, 1978; Hicks et al., 1975). Due to the almost 200-fold con centration of the glomerular filtrate in the tubular system, high intratubular concentrations of both compounds occur. When high doses are given to an animal, intratubular supersaturation of the chemicals with the consequences of precipitation and cristalluria appear. This frequently is associated with hema turia. Both symptoms, cristalluria and hematuria, have been observed when high concentrations of such chemicals are given. During long-term feeding of high doses to animals, tubular necrosis preceded renal-tumor formation (Anderson and Kanerva, 1978). At lower doses which do not induce tubular necrosis, neither cristalluria nor hematuria, renal necrosis or kidney tumors have been observed. One might also refer to the tumor-inducing capability of saccharin and cycla- mate. Both compounds have been shown to induce renal tumors at high doses (Hicks tt al., 1975; Oser et aL, 1975). Both substances, when given in large quantities to animals, increase the urinary pH substantially. This gives rise, at least in the case of saccharin as well as nitrflotriacetic acid, to subepithelial microcaiculi (Anderson and Kanerva, 1978; Armstrong, 1977). It has been demonstrated with other carcinogens that alkalization of the urine acts in a cocarcinogenic fashion, thus increasing tumor incidence. There fore, the tumor inducing effect of such compounds may be a physicochemical one which is related to their alkalizing properties. Now if this is true, then this is a dose-related phenomenon. The doses humans would take or would be exposed to in the case of other similarly acting chemicals, would be virtually insignificant in that tho buffer systems in the organism and in the urine would maintain the physiological pH of the urine. Again, extremely high doses given to animals are irrelevant to human exposure in such cases. This sort of consideration requires more basic research than is presently available and, if true, we would be dealing with at least 2 classes of carcinogens. One group Includes the classical type of the electrophilic reactants which react with target nudeoptilles resulting in mutagenic or carcinogenic effects. For these mechanisms sc* far there is no clear evidence for a threshold. Another group of compounds which indirectly induce tumors by other mechanisms associated with extremely high doses are likely to have a threshold dose. These considerations have been taken into account only marginally so far. If this hypothesis can be clearly established, then it would be possible to deal with different classes of chemicals that induce tumors and one can start to deal with questions of acceptable dose levels. This would also imply that high doses of such compounds used in animal studies would be irrelevant to human exposure. 143 Effect of contaminants at high doses finally, the problem of Uui pruscncc of impurities in chemicals being sub jected to mutagenicity or carcinogenicity testing is pointed out as exempli fied by trichloroethylene. This chemical has been reported to produce high incidence of hepatocellular carcinoma in both male and female mice but not in rats offer high daily oral doses. (Memorandum, HEW, 1975). Pure trichloro ethylene has been found to be slightly mutagenic in a modified bacterial liquid incubation system (Greim et el., 1975) and noncarcinogerdc in mice and hamsters (Henschler et al., 1980). Technical-grade trichloroethylene,; however, contains several impurities. Using gas-chromatogrephie mass-spectroscopic analysis, Henschler et al. (1977) recently found 10 different compounds in trichloroethylene which amounted to 0.6695 of the sample. Two of these, namely epichlorohydrin arid epoxybutane, constitute strong alkylating agents which have been shown to be mutagenic (Henschler et al., 1977; Pishbein, 1976) and carcinogenic (IARC, 1976; van Duuren and Banerjee, 1976). It has been concluded that the carcinogenic effect of technical trichloroethylene is most probably due to these epoxide# which arc' added to several brands for stabilization. This is supported by the observations that hepatocellular carcinomas were produced in mice but not in rats (IARC, 1976). Mice show a relatively lower activity of epoxide hydrolase, the anzyre* which detoxifies epoxides such as epichlorohydrin and epoxybutane fCMffjTl 110<Ta> T" the carcinogenicity studies, the animals vere exposed to high nf tri^hiorttAthvlana with concomitant high dose of the contaminants. Thus, when only high doses of chemical become mutagenic or carcinogenic, possible contamination by impurities has to be considered as well. Conclusions Firat Interspecies extrapolation and high to low doses extrapolation of chemicals undergoing metabolic activation and inactivation is most complicated. One has to consider: (1) Specie* differences in metabolic activation; (2) Species differences in metabolic inactivation; (3) Sufficient inactivating capacity may almost completely prevent interaction of reactive intermediates with the genetic mate'ial; (4) High amounts of a chemical may overwhelm inactivating processes with the consequence of disproportionately Increased interaction with genetic material. Second Several chemicals act by secondary mechanisms which only become effective at doses much higher than those to which man is exposed by the chemical. Third When only high doses result in toxic effects including mutagenicity or carcinogenicity, contribution of impurities have to be considered as well. AP00010645 144 Acknowledgement The excellent secretarial help of Ms. Judy Byera is gratefully acknowledged. References A*dm*. *.I, ui R.L. K>&n (1171) Hrpcalthns u4 eMiluto during la|Mlts t( 4Uun ivUlttsUtodUtt, 74. Cwmtl. Tile.( 14. M> 674. Aitdrav. U., wmI II. Untm (lultl) imiueuvn ol UNA mimIi npilatiian iir nrUfutruiw In kunw Irmptte- blutatd eU medlnladby Uvermteroromeaaad N ADPH. Bieeham. ftlopbya. Cemaaun_ I?. 10-44. Armstrong (1177) 4nUm and im Marian. In: HJ(. Watt, J.D. Wataoa and 4a, WlaMi (14*,), Oiidat of Human Canaat, Cold Sprint Harbor Wb^ pp. 1701--1703, ftanartaa. t.. and IX. van Duumr (17S) Covalent binding at iba tarrinnmn Triableraaibyla-- to hepatic miataaomal protein* and to axotenooa ONA ta vUro, Canaat la.. SI, 77*--7S0. Bartoeh. H.. C. MalavaUla, ft. Montaaano and b. TomatU (1071) TMauanaadiated muteganlrity 04 vtnyUdinr chloride aad 2<hle(obutadiac tn WMarUa (yphimaWnm, Nature (lt*4M), Ml. (41-441. Banach. H.. c. MaievaiUe. A. ftaibla aad 0. Pleneba (1979) Muusaaie aad alkrUUa* mttaballtM af baleethylenoa, chlorobutadlane* aad diehlerobuuaea predueed by >ada*( ar human livor tlnn, Evidence far oxirana formation by PASO liahad mtcroeamel tnoBOeaygtWM. Arab ToxleeL, 41,144--277. Iiuh. C.. and W. ftautur (1971) inblbitlan af dipbaapbasluaaaa dahydMgaaaaa by gaiootoeambte-lpboephate and VDP-galeetcaemia*, Blochla. Biepbya. Acil. 201,11--14. leek, K.W.. (3. aa Aahaaab, T. I/oecb aad P.w. Birka (1974) Maubodam af naphthalan* to naphthalene dihydrodlo) alucoroatde In iaolattd bepatoeytoa aad In liver mtaraaamaa. lUaabam, Pharmacol. 26, 2061--23041. Belt. HU, 4.0. Fitter. ij. Laib aad K. OttenwAder (1940) Undlni blnatloa of vtaytebteridn and vtnrl bromida at *7 low doaaa. Arab. TMKaL, SuppL 9, IBS--141. Brown. D.M.. tS. lAnglay. D. Smith aad D.C. Taylor (1974) MataboLiam of chloroform -- jbo mote* bottom of ["Cl chloroform by dlffmat tpaeUa. Xanabiotkca. 4, ill--ill. Carter, I.J. (1970) llaw to hum (unr rieke, 204, ail--*10. Commoner. B.. AJ. VHhayatbU. P. Dolan. S. Naif, P. MadyaaLha aad G.C, Cuaa <1474) Formation el mutagene la baal aad baat extract duilhd coolant. Science. 201.913--914. Comflatd. 4. (1077) Caret-eeaota itak immimiat, Beteaee, 194. 001-400. Ctur. i.M. UMOM1M, V.M. Hainituld add Od. Wwn <1U7*> MmUI4Ml<M Of the principal iil uiiia B, -OKA adduet formed in vivo la rat Uvar. Proa. Natl. Aaad. Sal. (U.S.A.), 76,1746--1749. Crump, KB.. D.C. Hoat, C.H. Lanai*)' and H. Pato (1979) Fundamental aarriaagania cimimu aad thvir implication* for low doaa flak amaarraant, Cancer Ke* 34, 2979--3*79. oavidtaa. M. Soodak. H.V. sout, J.T. Nwy. C. Nakamura and F. Maloof (1979) Thiourea and evaaamide j< inhibitor* of thyroid paroxidaict Th rota of Iodide, Sndocriooloty, 104,919--924. Uclbiun, 4.It.. K.C. MiUar and J.A. MHIar (11170a) Af*llydrnsy>2aiylwmlnonuwrvnv iuU<iMuns(erva: I( ittuiubUi tubs In ureiaooaiteai* and in prouln (matlitoadt-yl) 7f1**ri ia rat Uvar. Caacar Ite*30, 677--69 6. Oaftaun, J.R.. C.Y.R. Smith, C.C, Millar and 4.A. KUlar (1970b) Jtegetivtty is vivo of the eardaagao JV-bydraxy-2ACtamlnefluoran: Irene** by tulfaia ioa, Selenec, 147.164--lit. Ovan. U4l. <)97U) MataboUc InacUvaUaa of aQatoxm U,-*p*itle aa a catue for the diffaraao* la alia* wiioauKitllbillty af tat and smm. An in vitro iavartjgation. NaunyBcbmlodebaisb Arch. Pharma- aoL. 307, SuppL JUS. Dagan, C.W. and K.C. Nrumana (197B) Tka major metabolite* of afUtoai* B, in tba rat ia a glUtaUtlona aoaiuiata, Cham.-Blol. i stared. 29.339--265. Uueka. t.U, aad G. Kiitkia (1974) Tba rote of elutathioao in ebteraferm4nduaad hapatatoairity, Kapil. Mol. PathoL. 24,13-s , KAO Float Produatlen Papar. 10 SuppL (1976) Paatteidoa icalduaa in food! 1677 evmlvatieaa. p. 47--99. FUatMla. L (1976) laduatiial mutagen* and potantial stutagana, 1. Halogaaatad alipbatlc tiarivaUvaa, Muudoa Haa., 32. 237--307. KiahbaUt. L. (1974) Potential induatrial Carcinogana and Hutatitt. Kbavter, Aautardam. n, 211--217. Kkahmaa, ),, OJC. Kukuabidta. 4.0. 0<Canao* and U.T. Lraah (1979) Uv urinary aatrogaa glucuroaldat ia woman at riak for lanuUal braaat naw, Soiaaea, 304.1069--1091. Fry. BJ. T. Taylor and D.C. Hathwar (1972) Pulmonary edmlnaiion of otalototoim gad ita mctaboUta in man. Atriulat. PhanueodynN 194.99--111. Gahring. P.J.. SJL Biau aad F.Q. Watanaba (]g76) Pharmaaeklaatia iludteg in avaluatlon af tba Mil, cnUigical olid aiivlruuiuantal ha/anl of clwatiiwk, ut: Advances ia Mwdaro Tualcwlugl1 -- Naw CilO* eapta la 8alr ^valuation, Htmlapbar*. WiUUkha*. GUlatta. 4Jl. (1974a) A panpoativa oa tha rola of ehamicaUy naetlva mataboUtet ot loraign eoapouada 145 la toxicity. I. CentUllon of <Iubim l tdwt Wading e( iHtttv* mttilitUW with ihttw Ui ilu iitedaaoe and severity of toxicity, iioebem. phunmesi. 29, a?at--17*4. Gillette. J.R. <197b) a pempeottv# an iki tele at atomically nuttn metabolite* at foreign compounds la toxicity. II. AlleraUona in Uia Unatka at eevalent binding. Btoshem. PhamtaeaL. as, 3997--Ml*, flwim. IIM <1. IIomd, X. Iiadwan, U. Kiiekon uW U. lltinaeMer (179> Mtilxaanliriiy lu vitro and iMiiouUal careiaurentalty at ehiorbiated atkylenae w a IvaMiai af naUMii onimoo farms(ton, iioohom. Ptormoaol.. 34.2011--2017. Grate* H, D. tteimi, 0. tgrrt. v- Olooalmaiui aadM. Krinar (1977) inf Hiinnranmii aberrsttoai r* in analytical teal ter in vitro detection at wiwailla sniymo mediated ol nwte W*rmllaln, Anh. Tnrleal, W, lta-101, llaiar. TJ. <1070) ghlntonceue <3hlorol.3diMtMillnnr) -- Wbu ta Um eebUnoO tea g aaSji&gsnloitr7 Clin. TeotaaL. tl. 1M-LT0. Hopple. 1TM. and X. Kohtentaahi <19T7) Induction of Oak pairtk in n-iritlr i bantam, Inpctt, g, 1001--1001. .' --Ts"' tfcT .......... * HeneeMer. D,, t, Idea, T. Keudeckar and W. Mauler atTT) Carelnogtnlaltr at MUkUadidWanatfan n ^ tnli^O AMk, T*1mU it, IU-SU. . -iy . > ^Hanff^kr, D,,"W, Raman. XJd. Uaiaaar, O. Betaken. S. Mar and X. Radarwg (1990) Carcinogenicity itudy of trldUereaUiylana by ionium Inhrlatioi la.ihnt anteal Maiaa, AKb. TooteoL. 41, 337-- 340, Hama, 3.. U. Master and T. Welft (1371) Activation si (`*Cl-okIorchi*kanrle a protetebiadlng metaboiuee by rat Uvex lukrotomrs, Ch*ra.'SioL unmet., 30.104--911. Hiokr. SJd,, J4U. Wakefield and i. Cboweniee (1071) EvahiaUen a( a saw modal to dataai bladdar earainogwie or eoeoMUtoaent: RaaulU obtaiaad vtu eetrhaiin. irdinils and eyelopkerphaaaide, Owat.*Rlol. Internal, 11, 114-191. Hidrbmndi. A.O.. and I. Reota <1374) Raduoad nieeiliudanlna diaudwiMi rksMkti* (NAWHWsMfr tlM iwnsiiM and breakdown at byUseaeo iMteside during miaad laaaUaa euoaUsa raaaUona in liver mteoaemaa, Arab. Rteeham. liepkyi., 171.314--137. 1ACR 0374) Moaoiiapki on the Evaluation at the CaielaodnU Wak et Chemicals to Humana. Vol. 4. Sax hormones, Lyon. IARC (1374) Monographs on the evaluation et Careinoganie Xiak ol Cbamiaalt to Man. Vol, 7. Some anti thyroid and mlatad ubetaneva, aUrofwsos and IndwUiol atomical*, Lyon, pp. UO--J04, IAHC <1374) Monotmpk *ana an evaluation of carcinogenic defer ox abomkaala to man. VoL 11. Lyon. p. til--134. IflUntMty Isiultlny Uaiaon Croup <1373) SeientfAe bun (or IdmittReMian ol potential ei-- and aatlmottona of risk*. i. NUI. Canoar Inal. 33.141--200. icritr. RJH R.R. CpnoUy and S.D. Murphy (1174) Utaet of 14 h taat and glutathione daplotlon on 1J.> dicMorcetkrlenodnduaed hepetotexietty and lotfealUy in rata. Etas Mol. PatboL. 30.1ST--L90. Jprina. D.M.. R.lehr, M. Schaetet.RUldf. tt. Yai. J.M. XarU, D.X. Thakkar, A.W. Wood. A.YJi. Lu. 1). Uyad. 4. Watt. W. Levin and A.H. Comity (1977) Bay tvaiun apogidnof dlhydiotUols; A eonaapt eapUtaifts tka mutagenic and euetnoeanie eativity ol kasMialPm** and beftao|)anthnsena, lit: II.II. lUatt. J.D. nrataon ami J.A, Wiuatvn <KI.). Oridna uf llumon Canaar, Cold NpriM llartMir . ... LaU.,Pt>.ttltt-4d. donas. O.P,, H. Tbor, S. Andermon and 0. Omnlua <1973) Datoxifleatloa caaottona In laolatad hepatoyws. d.BtOj. Ckanu, 143,6341--4017. La*. C.C.. J.C. RandhaH. M. Winston. WJ. Hour*. R.L. Dixan and JJ. Wooda <l*7t> OaroinocanMtr of vinyl aklnrhle ami vlpylMana slilaiWs, J. TvaiaoL Knviron. liaaitk, 4.13 20. Loibman. K.C. <1374)Mataboliam and (oxielty of atymna, Environ. HmiU Ponpook, 11,114--1)9. Uteaky. W. (1.877) RtandaMndttino tor nitrttoa and nitromminaa. ins H.K. Hiatt. J.D. Stetson and J-A. Wlnatas (kda.). Gridina of livmau Canaar. Cold Sortna llarirur Lob., pp. 1007--i 311. MoUtml. C. <1077) V iyl ablufiila aareiouantUdLy: Au oxMtbiioMal rnmloi tur aarainooMinda vimjlas. Im tl.ll. Ibait, J.l> WaUmn ami J.A. Wlnatan (Kda.) Orlaina of Itwinan Cancer. Cold kPMd lUrbor Lab., Cold dptlno ubnr, pp. 113--144. Maltoni. CH O. Cotti, L. Mote and 9. Ckloao (1377) Camloooanldty bioanmra of vtarUdano cblerida, raoearen piana and tarty rtaulta. Mod. Lavoro, 40.141--MS, Maiijk. T4t, <1371) Ckamlcal eantneomada: How danooreua are lev deaeat Manta, 90S, 17--41. MeXonn*, 9.0. Watanabo and 9J. Cahrioa <1977) Phaimsaobinatlai I vinylidane ekloride in tko cat. Bovlron. Kaoltk Nwrnt. 41. 93--104. Mamezandum. Dept, of Hoalth, BduaaUen mod WaUara, Washington. <304.1374). MiUar. CJ. Wyatt. S.C. Miller and K.A. HOltRMM (1341) The A-bydraxylaLon of 4-atTiamlnii- biphaayl by the ml and dog and tka strong aaratnegaaWty of ^rd^irH-tt^niTnlmWrrbinTl in the rat. Canaar See.. 31.1434-1473. Millar. C.C., MiUar and M. Knomoto <134) The aomparaUva aaraiaogrnUitlM et gmaetylamino* nuerane and lu AMiydroxy mataboilta in mice, kuaatam. and guinan pte, Canaar Raa^ 24, 3014-- 2024. t AP00010647 14* Mlly. P,, and AJ. 0*i*o (1974) Mutaarnia activity of nyraoa oxlda (U*MV athyl banian*), * pta- tutnadimiM ntuboUM. Muutten In,40, 19--It. Niim, M. M. Honda. V. HUN, T. YuMfl, T. Kawaahi ud T. Swinun (1977) MutataniaiUaa of ptoiakn pyrolyaataa, Mutation R4. 93. J&MU, National Canaor Inatituta <1970) KIn on tha CarciaoMMria Bloomy of Triehioroathylana, Batbatd*. MD. National Canaor laatilutt (18771 Hmtmyt of nittUotriacatia uM (NTA) and aiUUssrtacatic aotdrirl* odium MtHwtakrdnM (Na,NTAHjO) for pnaalftla udwmidti', NCI Taeh. In. tit No. 4. January, DREW PubUaatioa No. (NtM> 77--904. Otaah. F. (1971) ---lt*" tpoxida Brdnaaa: iadtMibl* tnsymaa eatalytinc tha InattivaUon of tartiao- lanta and artotoaio ntubatm darivad from aromatto and olaflnto aompouada, XonoWoliaa. 3. 10fr--349. patch. F. (1980) Spaetaa diffttancai in wtividai and tnaetfemtias anxymm rolatod to la vttso mutt* mMir madiatad by Ham* preparation* from thaaa apaclaa, Arch. ToxiceL, SuppL 3,179--194. Oman. F.. N. Mama, l.V, Daly. J.S. Glolan and D.W. Nabort (1979) Oanatla aapraaaton of tit induatioa of apoatda hydraat and aryl hydroearbon hydnxyitn aetivtuaa in tha sauaa br phanohcThllal or 3*mthyLeholaatfanna, MoL PhamaooU 9.993-494. Oaaah. T,. K. Toanan and H. raftrtdndar (1974) Spoxlda hydnuaa in human livai hiopay rpaaiBsaoa: Aaaay and propartiaa, dtoeham. fkimutL, 33,1107--1917. Oaw, 9.L- S. Canon, a.I, con. R.S. Votin and IJ. lumbarc 0971) Chrenla toniaity nudy of ayata* mala: meetiarina (10:1) in rata. Toniaoloar, 4. 919--310. Mum, H.. and IJ. Jaitn (1979) Aouto Inhalation toxirity of 2*hiofo*V,3'biitadieaa (Chlaraprana); Klttu on Uv* and Ivflf, Toxical. APPl. Pharmacol, 90, 949--978. Pohl. b, (1079) Biochemical Toaiaakany of Chloroform. In: llodaaon, hand and PIlliwl (Md*.), liavirwa in ttioahamloai Toxicotopy, Eiaavitr/NarthHoU*ad, Now York, pp. 7^107. Fonamarkov, V., and L. Tomatla (1990) Cona-tan* tiatina of vinyUdano ahlortd* aAd eUoropraiu rot carctnotealeity in rata, OMOloay, 97,199--141. Furvos. H.D,, and VI. Oriaabaah (1947) studlaa oa ecparimantal |oiw*. VIII. Thyroid lumoa in mu iftuil with thiourea, Br. J. Fathol.. 39. 44--63. Hull, u.l'. (low) wfflautUra in strapuiatina Uw mulu uf wsicity itudki* In iatwaMry animnia to man. Environ. Raa.. 3,1(94(7. Raotpay. J.F. Quota. C.O. Hamilton. M.F. Balma? and S.A. Sahwatx (1977) Intatim raauita of wt. poar tonleoloalaal ttudioa in raw of vinyddonr efclerlda inaorporaiad in tit drinktnp watar or anminl^ wt( by repeated inhalkUon. Enviren. HUh, Panpaet.. 31, 39--43. Raui. R.H.. 9J. Othrina and C.M. Park (1979) Cuclaageeil# rick cctiautloa for chloiofonn: An altar* native to CPS*a ptoorOurvs, Fd. Com*t. Toxleel.. Id, 911--914. Scanlon, K.A., J.K. Barbour, J,M. HoUhkti* and LJd. Libtoay (1090) N-NitrumilimeUiylemln* in Ixwr, Ytl. Cnvinvt. ToUCOL, 1 *. 97--iltt. Schwvtx. I, (I9M Modulation in sulfation and alucuronidalinn of Imaphtteii la Isolaiwl fat livor <olti. Afflt. Tuipi4,.44, I37 Hft. Software, L.. M. Macaar and 9. Htatr (19M) Effaot or doeraaMd ttueuronldation and tuifaUan on fovnlant blndtns of naphthalan* in iaotated rat hopatoeytm. Tonicolnsy, 90, in Ptau. SriantWa CommittM. Food Safety Council (1879) Quantitative riak tarcatmint, Fd. Coamat. ToxiccL, Id.SuppL 2.109--134. SpirarilutlUvf. It., U. Ktoanbnuid and It. yntuamann (197U) ContamlnaUun uf boar with tmev ituaiiUtiaa of V-aitroao*dimrtfcylamina, Fd. Coonav. ToaleoL. 17,1*--34. .StoakiapaT. Ml. (1977) Tonlnolocy and dtiakinc watar eoaumlnanu, i. Am. Watar Wortta Aao~ 99 309--404. Sudmuca. T. (19T4) Ut7 io miantlflo about tha problam of mutantaa In human aookad food. Mutation Haa., 49.149--194. Suatmum. Tm m. Naaao. 1. Kawaehi. M. Honda. T. Yahafi, Y. Stino. 9. Sato. N. Matankiti. T. Matau* hima, A. Skint, M. Sawamum and H. Mtiauaoto (1977) Mutatint rtralaofana'la food, with n>aoi*l rafamne* to blgb mutadonia pyroiyti* produau id brodad fooda, in: Hiatt, 3A Wataon and J.A. Wimn (tda.). OiWm of Humaa Caaaar, Cold Sprint Harbor til, PP. 1991--1977. Sum mar. K.IJ_ u( W. aopcaimaan UOtIO) Khluro>S.4^1inUrobaaana daptatm pamMont ill ml akin and la muUtania in Saimoortla typhlmurium, Muution Rat., 77,91--93. Summar. K.H.. W. Gdttatmann and H. Oraim <19M0) GluUtiUona and alutaUtioaa yAnahalatam in tha SalmonaUa maaimaUanoniaroaoma mutaianieity Last. Mutation In., 70. 999--379. Summar, K.K., and H. Omim (1990) Ootoatftehtien Of ohloconvna (>ehioo*l>butndlana) with chiU* tMona In tha rat. Rincham. Btopbya. Rat. Ctnmvn. 09. dttd--979. `raiuianitaum, S.K. (1UOO) InaalUl aula ( ntUiiaa, llw ileianee*. 30.7--9. Tomatla. L. (1979) Tha pradlativa valua of rodant earcUoianielty tatu in iha mluttioo of human riaka. Aaau. Rv. Pbamaeot Toxlaol.. 19. Sll--530.