The mutagenicity testing of tertiary-butyl alcohol, tertiary-butyl acetate and methyl tertiary-butyl ether in Salmonella typhimurium.

Tertiary-Butyl alcohol (TBA), tertiary-butyl acetate (TBAc) and methyl tertiary-butyl ether (MTBE) are chemicals to which the general public may be exposed either directly or as a result of their metabolism. There is little evidence that they are genotoxic; however, an earlier publication reported that significant results were obtained in Salmonella typhimurium TA102 mutagenicity tests with both TBA and MTBE. We now present results of testing these chemicals and TBAc against S. typhimurium strains in two laboratories. The emphasis was placed on testing with S. typhimurium TA102 and the use of both dimethyl sulphoxide and water as vehicles. Dose levels up to 5000 microg/plate were used and incubations were conducted in both the presence and absence of liver S9 prepared from male rats treated with either Arochlor 1254 or phenobarbital-beta-naphthoflavone. The experiments were replicated, but in none of them was a significant mutagenic response observed, thus the current evidence indicates the TBA, TBAc and MTBE are not mutagenic in bacteria.

Non-mutagenicity of 3-dimethylamino-psi-saccharin to Salmonella typhimurium.

3-Dimethylamino-psi-saccharin has been confirmed as non-mutagenic to Salmonella typhimurium when evaluated in vitro in the presence of induced rat liver S9 mix.

Mechanistic relationship among mutagenicity, skin sensitization, and skin carcinogenicity.

Twenty organic Salmonella mutagens, seven of which (including benzo[a]pyrene) are established skin carcinogens, and one of which (2-chloroethanol) is a well-defined noncarcinogen to skin, have been evaluated for skin-sensitizing activity using the local lymph node assay. The relative mutagenicity of the agents to Salmonella was also established. Fourteen of the chemicals were positive in the local lymph node assay, including the seven skin carcinogens. 2-Chloroethanol was inactive as a sensitizing agent. We suggest that a variety of factors contributes to the lack of sensitizing activity of the remaining six bacterial mutagens: extremes of intrinsic chemical reactivity, high water solubility reducing dermal translocation, and inappropriate dermal metabolism. Two reference skin-sensitizing agents (an oxazolinone and fluorescein isothiocyanate) were established as in vitro clastogens after their recognition as nonmutagens to Salmonella. These data imply that mutagenicity, rather than simply activity in the Salmonella assay, is a primary stimulus for electrophilic sensitization and carcinogenic initiation in the skin. We conclude that genotoxicity data for an agent can provide indications of the agent's potential to induce skin sensitization and that genotoxins which are skin-sensitizing agents have an enhanced potential to initiate skin carcinogenesis. We suggest that common, albeit individually distinct, structure-activity relationships underpin genotoxicity, skin sensitization, and the initiation of skin carcinogenesis. These relationships should simplify the hazard evaluation of chemicals and contribute to a reduction in animal usage. Several predictions of skin carcinogenicity are made based on the data presented.

Mutagenicity to bacteria, cultured cells, and rodents of the human carcinogen chlornaphazine.

The human carcinogen and nitrogen mustard chlornaphazine (CN) has been confirmed to be mutagenic to Salmonella and, unexpectedly, the more so when evaluated in the presence of liver S9 mix. It also has been established as clastogenic to Chinese hamster lung cells exposed in vitro to dose levels greater than 2.5 micrograms/ml. Chlornaphazine subdued mice at doses of 5 g/kg, but only the occasional death occurred during the 4 days following oral administration of this dose in corn oil. Consequently, a median lethal dose level was not established. Nonetheless, dose levels of 500 mg/kg or greater gave a clear positive response in both the mouse and the rat bone marrow micronucleus assay. Although depression of erythropoeisis was observed in mice, a clastogenic response still was observed in the bone marrow 24 hr after dosing. The positive response in the rat was greater than that observed in the mouse. The present data provide a further instance of an established human carcinogen being readily detected by standard in vitro and in vivo mutagenicity assays.

Weak and unexpected mutagenicity to Salmonella of the rat hepatocarcinogen methapyrilene.

The rat liver carcinogen methapyrilene is shown to be a selective mutagen to strain TA1535 of Salmonella typhimurium when tested in the absence of S9 mix and using the standard plate-incorporation assay protocol. The activity observed was weak but was reproducible for a range of samples on many occasions of test and was not due to impurities. These data contrast with six earlier reports of the inactivity of this chemical in the Salmonella mutation assay.

Potent clastogenicity of the human carcinogen etoposide to the mouse bone marrow and mouse lymphoma L5178Y cells: comparison to Salmonella responses.

The suspect human carcinogen, etoposide, is known to be genotoxic, producing both gene and chromosomal mutations, probably by virtue of its ability to inhibit topoisomerase II activity. The present paper describes assays conducted using the Salmonella assay, the mouse lymphoma tk+/- assay (gene and chromosomal mutation analysis and molecular analysis of tk-/- mutants) and the mouse bone marrow micronucleus assay. Nonreproducible, weak, dose-related increases in mutation frequency in strain TA98 (but not TA1538 or TA1537) of Salmonella typhimurium were observed. Etoposide was highly mutagenic at the heterozygous thymidine kinase (tk+/-) locus of L5178Y mouse lymphoma cells at concentrations below 0.1 micrograms/ml. Mostly small colony mutants were induced, consistent with the potent clastogenicity also observed. Molecular analysis of mutants indicated that 83% and 92% of large and small colony mutants, respectively, had lost the entire target gene sequence. Chromosomally aberrant L5178Y cells were approximately 2 to 600-fold more prevalent than small tk-/- mutant colonies. This suggests that the viable target for etoposide-mediated clastogenesis in the selective assay is approximately one-fifth of chromosome 11b, itself being approximately one-fortieth of the mouse genome. An unusually potent response was observed for etoposide in the mouse bone marrow micronucleus assay (63.1 +/- 18 MPE/1,000 PE 24 hours after an oral dose of 1 mg/kg). The minimum detectable dose level in the assay was between 0.01 and 0.1 mg/kg. At dose levels between 1 and 15 mg/kg, an inverse dose response was observed. This reduction in assay response was not due to the small concommitant decrease in the incidence of polychromatic erythrocytes, a conclusion based on studies with N-methyl-N-nitrosourea. Animals sampled 48 hours after dosing with etoposide (10 mg/kg) had no polychromatic erythrocytes in the bone marrow. These observations for the micronucleus assay await explanation. The chemical structure of etoposide is displayed and discussed within the context of such strong mutagenic activity being associated with a nonelectrophilic agent.

Assay-specific genotoxicity of N-nitrosodibenzylamine to the rat liver in vivo.

N-Nitrosodibenzylamine (NDBzA) is mutagenic to Salmonella typhimurium and induces DNA strand breaks in isolated rat hepatocytes, yet it is reported to be non-carcinogenic to the rat. Here we report that it is inactive in both the rat and mouse bone marrow micronucleus assays and in a rat liver autoradiographic assay for unscheduled DNA synthesis. It is, however, clearly active as a micronucleus-inducing agent and mitogen in the rat liver and is capable of inducing single-strand breaks in the DNA of rat liver. The origin and implications of this curious conflict of in vivo genotoxicity data are discussed. Irrespective of that discussion, it is concluded that NDBzA is genotoxic to the rat liver in vivo.

The possible role of azoreduction in the bacterial mutagenicity of 4-dimethylaminoazobenzene (DAB) and 2 of its analogues (6BT and 5I).

The extent to which azoreductive cleavage contributes to the bacterial mutagenicity of 3 azo compounds has been investigated. The compounds studied were the rodent-liver carcinogens 4-dimethylaminoazobenzene (DAB) and 6-dimethylaminophenylazobenzthiazole (6BT), and the reported non-carcinogenic isostere 5-dimethylaminophenylazoindoline (5I). Although each of these compounds is mutagenic to Salmonella when evaluated using a pre-incubation protocol and in the presence of an induced rat-liver S9 mix, the constituent amines (cleavage products) were essentially inactive. It is therefore concluded that the mutagenic response reported for DAB, 6BT and 5I is related to metabolic activation of the intact molecules. In addition, the non-mutagenicity of 4'-phenyl-4-dimethylaminoazobenzene (4PhDAB) suggests that azoreductase activity is low in the Salmonella preincubation assay, at least as conducted in this laboratory. In the case of 4PhDAB, less than 1.4% azoreduction would yield sufficient quantities of the derived amine, 4-aminobiphenyl, for a positive mutagenic response to have been observed.

Weak mutagenicity to Salmonella of the formaldehyde-releasing anti-tumour agent hexamethylmelamine.

Hexamethylmelamine (HEMLA) is a metabolism-dependent formaldehyde-releasing agent related in structure to hexamethylphosphoramide (HMPA). Both compounds are known to be mutagenic to Drosophila. HMPA, in common with the other formaldehyde-releasing agents studied, is non-mutagenic to Salmonella. The present paper describes the mutagenicity of HEMLA to strain TA100 of Salmonella typhimurium. Activity is dependent upon both the use of a pre-incubation assay protocol and on high concentrations of Aroclor-induced rat liver in the S9 mix. HMPA was inactive under similar conditions of test.

Lack of mutagenicity to S. typhimurium of neopentyl bromide and pentaerythrityl tetrachloride: relation to chemical structure.

Neopentyl bromide and pentaerythrityl tetrachloride were shown here to be non-mutagenic to 7 strains of Salmonella typhimurium. These inactivities are reflected in the inability of either compound to produce a colour reaction in the chemical alkylation test of Epstein (4-nitrobenzylpyridine, NBP). It is suggested that the lack of biological reactivity of these two alkyl halides is due to steric crowding of the halogen group. These findings are significant within the context of the potent mutagenicity of mono-haloalkanes in general. The similarity between the odour of pentaerythrityl tetrachloride and camphor is discussed.

The role of formaldehyde and S-chloromethylglutathione in the bacterial mutagenicity of methylene chloride.

Methylene chloride was less mutagenic in Salmonella typhimurium TA100/NG-11 (glutathione-deficient) compared to TA100, indicating that glutathione is involved in the activation of methylene chloride to a mutagen in bacteria. In rodents, the pathway of methylene chloride metabolism utilizing glutathione produces formaldehyde via a postulated S-chloromethylglutathione conjugate (GSCH2Cl). Formaldehyde is known to cause DNA-protein cross-links, and GSCH2Cl may act as a monofunctional DNA alkylator by analogy with the glutathione conjugates of 1,2-dihaloalkanes. The lack of sensitivity of Salmonella TA100 towards formaldehyde (Schmid et al., Mutagenesis, 1 (1986) No. 6, 427-431) suggests that GSCH2Cl is responsible for methylene chloride mutagenicity in Salmonella. In Escherichia coli K12 (AB1157), formaldehyde was mutagenic only in the wild-type, a characteristic shared with cross-linking agents, whereas 1,2-dibromoethane (1,2-DBE) was more mutagenic in uvrA cells (AB1886). Methylene chloride, activated by S9 from mouse liver, was mutagenic only in wild-type cells, suggesting a mutagenic role for metabolically derived formaldehyde in E. coli. Mouse-liver S9 also enhanced the cell-killing effect of methylene chloride in the uvrA, and a recA/uvrA double mutant (AB2480) which is very sensitive to DNA damage. This pattern was consistent with formaldehyde damage. However, a mutagenic role in bacteria for the glutathione conjugate of methylene chloride cannot be ruled out by these E. coli experiments because S9 fractions did not increase 1,2-DBE mutagenicity, suggesting lack of cell wall penetration by this reactive species. Rat-liver S9 did not activate methylene chloride to a bacterial mutagen or enhance methylene chloride-induced cell-killing, which is consistent with the carcinogenicity difference between the species.

Activity of urethane and N,N-dimethylurethane in the mouse bone-marrow micronucleus assay: equivalence of oral and intraperitoneal routes of exposure.

Urethane is shown to be active in the mouse bone-marrow micronucleus assay when administered as a single dose by either gavage or intraperitoneal injection. The magnitude of the response using the two routes was not statistically significantly different. N,N-Dimethylurethane (DMU) is shown to be mutagenic to Salmonella and active in the bone-marrow micronucleus assay by both routes of administration. The activity of DMU in the bone marrow precludes elimination of ethanol, yielding cyanate ion, as an explanation for the micronucleus-inducing activity of urethane.

N-chloropiperidine and calcium hypochlorite: possible examples of toxicity-dependent clastogenicity in vitro.

N-Chloropiperidine (NCP) has been reported to be both toxic and mutagenic in a wide range of in vitro and in vivo genotoxicity assays, however, few experimental details or numerical data have been presented to support these claims. The latter facts, together with the lack of any clear structural precedent for the mutagenicity of this agent, led us to re-evaluate it using the Salmonella mutation assay and the mouse bone marrow micronucleus test. The absence of mutagenic activity observed in both of these systems indicates that the genotoxicity of NCP and related chloramines remains to be unequivocally established. In particular, the potent clastogenicity to CHO cells reported for NCP may be related solely to oxidative denaturation of cellular proteins induced by hypochlorous acid, a hydrolysis product of NCP. Separate reports indicate that calcium hypochlorite (bleaching powder) is also clastogenic in vitro but not in vivo. We therefore suggest that N-chloropiperidine may prove of value as a model compound for studies designed to distinguish genotoxic clastogenicity (specifically DNA-reactive) from general toxicity-mediated clastogenicity.

Genetic activity of the human carcinogen sulphur mustard towards Salmonella and the mouse bone marrow.

Sulphur mustard is clearly mutagenic to the repair-proficient (uvrB+) strain G46 of S. typhimurium. It is also a micronucleus-inducing agent to the mouse bone marrow. These data strengthen the database indicating that most human carcinogens are genotoxic.

Methyl vinyl sulphone: a new class of Michael-type genotoxin.

Methyl vinyl sulphone (MVS) is a labile, Michael-reactive chemical, similar in structure to acrylamide (AA). Given that acrylamide is a reference mammalian mutagen and a rodent carcinogen, studies were undertaken to evaluate the potential genotoxicity of MVS. In common with AA, MVS was non-mutagenic to Salmonella but active as an aneugen to cultured mammalian cells. It is concluded that vinyl sulphones should be regarded as representative of a new class of genotoxic chemical whose mode of action is probably primarily dependent upon Michael reactivity to proteins.

Chloracetamide-N-metholol: an example of an in vitro and in vivo clastogen which is non-mutagenic to Salmonella.

The industrial biocide chloracetamide-N-metholol (CAM) has been shown to be non-mutagenic to 6 strains of Salmonella using both the plate-incorporation and a pre-incubation test protocol. Its biocidal activity is unlikely to have influenced these results since Kathon 886, a more potent biocide, was concomitantly detected as mutagenic to strain TA100. In contrast, CAM was weakly clastogenic to human lymphocytes cultured in vitro and elicited a positive response in the mouse bone marrow micronucleus test when assayed using the intraperitoneal, but not the oral route of administration. A positive response was concomitantly observed for the rodent carcinogen and formaldehyde-releasing agent hexamethylphosphoramide (HMPA) in these 2 clastogenicity assays. Data are presented showing the slow hydrolysis of CAM to formaldehyde in vitro, and both [carbonyl-14C]CAM and [metholol-14C]CAM have been shown to interact covalently with calf-thymus DNA in vitro. It is concluded that CAM may be a direct-acting carcinogen to rodents, but that both the qualitative and quantitative outcome of its bioassay for carcinogenicity will be influenced critically by the bioassay protocol adopted; in particular, by the route of administration selected. These findings emphasize the need to complement the Salmonella gene-mutation assay with an in vitro assay for the induction of chromosomal aberrations if in vivo genotoxins are to be detected efficiently in vitro.

Mutagenicity to Salmonella of the mono methylamino and N-cyanoethyl analogues of 4-dimethylaminoazobenzene (DAB) and 6-dimethylaminophenylazobenzthiazole (6BT).

Replacement of one of the methyl groups of the carcinogens 4-dimethylaminoazobenzene (DAB) and 6-dimethylaminophenylazobenzthiazole (6BT) with a cyanoethyl (-CH2CH2CN) substituent dramatically increases their mutagenic potency to Salmonella (strain TA98). The corresponding monomethylamino derivatives (-NHCH3) are more mutagenic than the parent dimethylamino [-N(CH3)2] compounds, but substantially less mutagenic than the cyanoethyl derivatives. All of these mutagenic activities are liver-S-9-dependent. The very similar dose response curves observed for the two cyanoethyl compounds argues for the formation of a common electrophilic intermediate from each.

Studies on the genotoxicity of beryllium sulphate in vitro and in vivo.

There is limited evidence that beryllium is a lung carcinogen to man, and several compounds of beryllium are carcinogenic to the lungs of the rat, rabbit and monkey. One such compound is beryllium sulphate (BeSO4.4H2O). This soluble salt has been evaluated in a range of genotoxicity tests. It was non-mutagenic to Salmonella typhimurium (strains TA1535, 1537, 1538, 98 and 100) when evaluated in the plate-incorporation assay at dose levels up to 5 mg/plate (+/- induced rat-liver S9 mix). It was also non-clastogenic to Chinese hamster lung (CHL) cells cultured in vitro. When dosed to male CBA mice via oral gavage at dose levels of 80% and 50% of the medium lethal dose (2.3 and 1.4 g/kg, respectively) it failed to increase the incidence of micronucleated polychromatic erythrocytes in the bone marrow (sampled at 24, 48 and 72 h post-dosing). However, a marked depression of erythropoiesis was evident 24 h after dosing suggestive of beryllium-mediated bone-marrow toxicity. When tested in the strain A mouse lung tumour bioassay, BeSO4 induced a significant increase in the number of tumour-bearing animals but not in the number of lung tumours per animal. These findings are discussed within the contexts of other genotoxicity data published for BeSO4, and of current strategies for the detection of possible human carcinogens.

Mutagenicity to Salmonella, Drosophila and the mouse bone marrow of the human antineoplastic agent fotemustine: prediction of carcinogenic potency.

The antineoplastic agent fotemustine is shown to be a base-pair mutagen to Salmonella. Activity is more marked in the uvrB-proficient strain G46 than in the repair-deficient strain TA1535. This is consistent with its ability to cross-link DNA. Potent activity as a somatic and germ-cell mutagen to Drosophila was also observed. A potent clastogenic response was given by fotemustine in the mouse bone marrow following either oral gavage or intraperitoneal injection of a single dose of 5 mg/kg. In each of these respects it is shown to be indistinguishable from the structurally related antineoplastic agent and human carcinogen MeCCNU. It is concluded that fotemustine should be regarded as having clear potential to induce cancer in humans. Based on these data, including the preponderance of chromosome breakages over recessive lethal mutations in Drosophila, an estimated rodent carcinogenic potency (TD50) of between 15-150 mg/kg is suggested for fotemustine.

Activity of bromochlorodifluoromethane (BCF) in three mutation tests.

The halocarbon BCF was tested in 3 assays to assess its mutagenicity and clastogenicity. It produced a positive response in Salmonella typhimurium strain TA1535 but was negative in TA1537, TA1538, TA98 and TA100. In an L5178Y mouse lymphoma microwell assay (TK locus), BCF was negative. BCF was administered at 5000 and 50 000 ppm in air for 6 h to groups of C57B1/6J mice of both sexes. Animals were killed at 24, 48 and 72 h after cessation of exposure and the incidence of bone marrow micronuclei per 1000 PCEs determined. There was no significant difference in the incidences of micronuclei between untreated animals and those exposed to either concentration of BCF at any of the sampling times. These results suggest that BCF is mutagenic in vitro in only one strain of Salmonella; in mammalian cells the compound induced no gene mutation in vitro nor clastogenic activity in vivo at doses that also produced clear evidence of toxicity.