Connexin 37 mutations in rat hepatic angiosarcomas induced by vinyl chloride.

Connexin genes have been shown to restore normal cell growth when transfected into certain tumorigenic cells and thus are considered to form a family of tumor suppressor genes. In this study, we have analyzed mutations of the connexin 37 (Cx37) gene in rat hepatic angiosarcomas induced by vinyl chloride. A total of 25 rat liver tumors (22 hepatic angiosarcomas and 3 hepatocellular carcinomas) were analyzed by PCR-single-strand conformation polymorphism analysis and DNA sequencing. Four mutations were detected in three tumors: (a) one GGG(Gly) to GAG(Glu) mutation at codon 168; and (b) three silent mutations, CGA(Arg) to CGC(Arg), at codon 166. In addition, we found that codon 88 is polymorphic (GAG(Glu) to GAA(Glu)). Cx37 proteins are detectable in endothelial cells of normal liver by immunohistochemical analysis, but none of the angiosarcomas showed Cx37-positive spots. These results suggest that Cx37-mediated gap junctional intercellular communication may be disturbed in most of these angiosarcomas, but mutation of the Cx37 gene is rare.

Lack of miscoding properties of 7-(2-oxoethyl)guanine, the major vinyl chloride-DNA adduct.

Chloroethylene oxide, an ultimate carcinogenic metabolite of vinyl chloride, was reacted with poly(deoxyguanylate-deoxycytidylate); the nucleic acid base adducts, 7-(2-oxoethyl)guanine and 3,N4-ethenocytosine, were analyzed by reverse-phase high-performance liquid chromatography. Chloroethylene oxide-modified poly(deoxyguanylate-deoxycytidylate) was assayed as template in a replication fidelity assay with Escherichia coli DNA polymerase I, and the newly synthesized product was subjected to nearest-neighbor analysis. Misincorporation rates of deoxyadenosine monophosphate and thymidine monophosphate were found to increase with the level of template modification. About 80% of the mispairing events were located opposite minor cytosine lesions. 7-(2-Oxoethyl)guanine, the major adduct identified (greater than 98% of the adducts), did not miscode for either thymine or adenine, failing to support an earlier hypothesis that the cyclic hemiacetal form, O6,7-(1'-hydroxyethano)guanine, could, by analogy with O6-methyl- and O6-ethylguanine, simulate adenine. Our results indicate that direct miscoding of 7-(2-oxoethyl)-guanine may contribute only slightly to the induction of mutations by chloroethylene oxide or vinyl chloride.

Copper-dependent formation of miscoding etheno-DNA adducts in the liver of Long Evans cinnamon (LEC) rats developing hereditary hepatitis and hepatocellular carcinoma.

Formation of etheno-DNA adducts in the liver was investigated in Long Evans cinnamon (LEC) rats, a Long Evans strain with hereditary abnormal copper metabolism, which develop spontaneous hepatitis and later hepatocellular carcinoma. Using an ultrasensitive immunoaffinity/32P-postlabeling assay (J. Nair et al., Carcinogenesis, 16: 613-617, 1995), the etheno adducts 1,N6-ethenodeoxyadenosine (epsilon dA) and 3,N4-ethenodeoxycytidine (epsilon dC) were measured in the liver of 7-, 18-, 30-, and 87-week-old LEC rats. Levels were highest in the liver of 18-week old rats 85 +/- 17 (epsilon dA) and 85 +/- 30 (epsilon dC) adducts per 10(9) parent nucleotides, and the increase in the levels of etheno adducts was age dependent. Age-matched Long Evans agouti rats, a tumor-free sibling line of LEC rats, had much lower levels of both etheno adducts. Etheno adduct levels in LEC rats were well correlated with the hepatic copper levels, and peak adduct levels coincided with the age of commencement of fulminant hepatitis. Our results demonstrate for the first time a copper- and age-dependent formation of highly miscoding etheno-DNA adducts in the liver of LEC rats. These adducts are formed from lipid peroxidation products (F. El-Ghissassi et al., Chem. Res. Toxicol., 8: 273-283, 1995) and thus could arise in the liver of LEC rats from oxygen radicals generated by copper-catalyzed Fenton-type reactions. Etheno-DNA adducts along with other oxidative DNA base damages may thus be involved in liver carcinogenesis in LEC rats.

Carcinogenicity of chloroethylene oxide, an ultimate reactive metabolite of vinyl chloride, and bis(chloromethyl)ether after subcutaneous administration and in initiation-promotion experiments in mice.

Repeated s.c. administration of chloroethylene oxide, a reactive metabolite of the carcinogen vinyl chloride, induced local tumors in mice, with an incidence comparable to that of bis(chloromethyl)ether, a structurally related human and animal carcinogen, when both compounds were applied at maximum tolerated chronically toxic doses; no tumors distant from the injection site were produced. Bis(chloromethyl)ether, chloroethylene oxide, and its rearrangement product chloroacetaldehyde, a highly toxic compound, were further tested in an initiation-promotion experiment. Application to the skin of a single dose of either bis(chloromethyl)ether or chloroethylene oxide, followed by 3-times-weekly applications of 12-O-n-tetradecanoylphorbol-13-acetate for 42 weeks, produced skin tumors in mice; chloroacetaldehyde under comparable conditions produced no increase in benign or malignant tumors. A good correlation between the chemical reactivity, on the basis of hydrolysis constants in aqueous media, and the carcinogenicity of the three compounds was noted. Our results support the hypothesis that epoxidation of the thylenic double bond in vinyl chloride yields an ultimate carcinogenic metabolite, chloroethylene oxide, a highly reactive compound which appears also to be largely responsible for the known genetic changes caused by the parent compound.

Mutagenesis of ras proto-oncogenes in rat liver tumors induced by vinyl chloride.

Vinyl chloride is a DNA-damaging carcinogen which induces liver angiosarcomas in humans and animals. Activation of the Ki-ras 2 gene by a GC-->AT transition at the second base of codon 13 in human liver angiosarcomas associated with occupational exposure to vinyl chloride has been reported recently. In order to compare the molecular pathways of carcinogenesis in humans and animals, Sprague-Dawley rats were exposed to vinyl chloride and hepatic tumors, including two hepatocellular carcinomas and five liver angiosarcomas, were investigated for mutations at codons 12, 13 and 61 of the Ha-ras, Ki-ras and N-ras genes. High molecular weight DNA was amplified by the polymerase chain reaction and point mutations were analyzed by allele specific oligonucleotide hybridization, direct sequencing of polymerase chain reaction products and sequencing after cloning. None of the tumors exhibited a mutation in codons 12, 13 and 61 of the Ki-ras gene, nor in codons 12 of the Ha-ras gene or 61 of the N-ras gene. However, an activating AT-->TA transversion at base 2 of codon 61 of the Ha-ras gene was detected in the two hepatocellular carcinomas. Mutations involving codon 13 (GGC-->GAC) and codon 36 (ATA-->CTA) of the N-ras A gene were detected in two liver angiosarcomas, suggesting that the nature of the ras gene affected by a given carcinogen depends on host factors specific to cell types. Several additional base pair substitutions were found in exon 1 of the N-ras B and C sequences. NIH 3T3 transfection assays and Southern blot analysis of DNA from transformed NIH 3T3 cells confirmed the presence of a dominant activated N-ras gene. These results emphasize the differences in the molecular pathways leading to tumors in humans and rats and within a given species between different cell types.

p53 gene mutation pattern in rat liver tumors induced by vinyl chloride.

Vinyl chloride (VC) induces angiosarcomas of the liver (ASL) and hepatocellular carcinomas (HCCs) in humans and rodents. We examined the presence of p53 gene mutations in ASL and HCC induced by VC in Sprague Dawley rats; 25 ASL and eight HCCs were analyzed for point mutations in exons 5-8, using PCR amplification, single-strand conformation polymorphism analysis, and direct DNA sequencing. Mutations were found in 11 (44%) of the ASL and in 1 HCC. A 12-base pair deletion was found in one tumor; all others were base pair substitutions. Nine of the point mutations were observed at A:T base pairs (5 A:T --> T:A; 2 A:T --> G:C, and 2 A:T --> C:G), and of three G:C --> A:T transitions, only one was at a CpG site. In ASL, four mutations were found in exon 5, two in exon 6, and six in exon 7; the base pair substitution found in one HCC was in exon 8. One ASL exhibited two point mutations, including a silent one. Two ASL exhibited the same mutation in codon 203 and two other samples in codon 253. Codon 235 was found to be mutated in three ASL. These data show that p53 is often mutated in ASL induced by VC in rats and, as observed in ASL in humans exposed to VC, the majority of the missense mutations involved A:T base pairs. The characteristic patterns of mutations found suggest that a common mechanism operates in VC-induced p53 mutagenesis in both species, and these mutations are consistent with the formation of DNA etheno adducts by VC in the liver. The A:T --> T:A transversion observed in the first nucleotide of codon 253 in two rat ASL is equivalent to the A:T --> T:A transversion characterized previously in codon 255 in one human ASL associated with VC exposure.

[Phospholipid composition in thermosensitive mutants of Escherichia coli (author's transl)]. / Composition phospholipidique de mutants thermosensibles d'Escherichia coli k 12

Quantitative phospholipid composition was investigated on various thermo-sensitive mutants of Escherichia coli K 12. Phosphatidylethanolamine was the major component. Three mutants (T1, T46 and T5) grown at 40 degrees C, had more anionic phospholipids (phosphatidylglycerol and diphosphatidylglycerol) and less bipolar phospholipids (phosphatidylethanolamine and lysophosphatidylethanolamine) than the wild type. Two mutants strains T46 and T83 which have the same Dna A mutation and identical membrane proteins alterations, have different phospholipid composition. Other mutants, such as T46, exhibit a blocking o? THE INITIATION OF DNA synthesis or morphological alterations under restrictive growth conditions (40 degrees C); however, the phospholipid composition of these mutants is normal and differs from the phospholipid composition of T46.T1 and T5 mutants have the same alterations in the phospholipid composition as T46 although they do not possess Dna A and Div A mutations of the T46 strain. Mutations leading to the alteration of the phospholipid composition are unrelated to other mutations leading to changes in the morphology and cellular division.

Conformational analysis of a 139 base-pair DNA fragment containing a single-stranded break and its interaction with human poly(ADP-ribose) polymerase.

The conformational changes induced by the introduction of a central and unique single-stranded break in a 139 base-pair DNA duplex have been analysed by means of polyacrylamide gel electrophoresis, HPLC and dark-field electron microscopy. Compared to the control DNA, the disruption of the covalent sugar-phosphate backbone induces a retardation detected both by gel electrophoresis and anion exchange based HPLC. Electron microscopic visualization of the DNA molecules reveals that most of them present a central fracture at the position of the nick. Measures of the angle at the apex were very well fitted by a simple model of isotropic flexible junction assuming spatial Hooke's law and simple basic Boltzmann statistics. This amounts to using a folded Gaussian distribution. The fit yields an angle equilibrium value phi 0 = 122 degrees for the nicked fragment. The angle distribution could also result from an equilibrium between two forms of the molecule with isotropic flexibility at the nicked site: a stacked and a very flexible unstacked form. The majority of bound poly(ADP-ribose) polymerase, a zinc-finger enzyme involved in DNA break detection, was localized at the apex of the V-shaped DNA duplex, with an accentuation of its general V-shaped conformation (phi 0 = 102 degrees).

Effect of split doses of N-methyl-N-nitrosourea on DNA repair synthesis in cultured mammalian cells.

DNA repair was measured in human fibroblasts, mouse C3H 10 T 1/2 fibroblsts and rat hepatocytes by the non-semi-conservative incorporation of [3H]-TdR during DNA repair synthesis using liquid scintillation techniques. Confluent monolayers of these cells grown on cover slips were exposed to split doses (125 or 250 microgram/ml) of the mutagenic and carcinogenic alkylating agent MNU and DNA repair synthesis compared with that produced by a single dose (500 microgram/ml). No significant difference in DNA repair capacity was detected in the three cell lines treated with a single dose or split doses of MNU.

Investigations on the relationship between DNA ethenobase adduct levels in several organs of vinyl chloride-exposed rats and cancer susceptibility.

The levels of 1,N6-ethenodeoxyadenosine (epsilon dAdo) and 3,N4-ethenodeoxycytidine (epsilon dCyd) were measured in DNA of several target organs of vinyl chloride (VC)-exposed rats. Seven-day-old (group I) and 13-week-old (group II) BD VI rats were exposed during 2 weeks to 500 ppm VC in air (7 hr per day and 7 days per week). epsilon dAdo and epsilon dCyd were measured by a combination of prepurification of DNA hydrolysates by HPLC and competitive radioimmunoassay using specific murine monoclonal antibodies. Both ethenodeoxynucleosides were detected in liver, lungs and brain (levels ranging from 0.6 x 10(-7) to 1.3 x 10(-7) for epsilon dAdo/2'-deoxyadenosine and from 1.95 x 10(-7) to 4.92 x 10(-7) for epsilon dCyd/2'-deoxycytidine) but not in kidneys of group I rats. In group II rats, only liver DNA was analysed and the levels of each adduct were six times lower than in young (group II) rats. These findings are in good agreement with the organotropism and the age-related sensitivity of VC-induced carcinogenesis in rodents.

Metabolic activation of vinyl chloride by rat liver microsomes: low-dose kinetics and involvement of cytochrome P450 2E1.

The metabolism and pharmacokinetics of vinyl chloride (VC) have been extensively studied in rodents and humans, but the maximum velocity (Vmax) and Michaelis constant (K(m)) for the activation of VC by microsomal monooxygenases in vitro have not yet been determined. Using a new sensitive assay, the epoxidation of VC by rat liver microsomes (adult Sprague-Dawley) at concentrations from 1 ppm to 10(6) ppm in the gas phase was measured. In the assay, the reactive VC metabolites chloroethylene oxide and 2-chloroacetaldehyde were trapped with excess cAMP, yielding, 1,N6-etheno-cAMP (epsilon cAMP) which was quantitated by HPLC fluorimetry. The trapping efficiency of electrophilic VC metabolites by cAMP was close to 10%. The specificity of the method was confirmed by purification of epsilon cAMP on an immunogel. The VC concentration in the gas phase was measured by GC/flame ionization detection, while in the aqueous phase it was calculated from the partition coefficient between air and the microsomal suspension. Activation of VC by rat liver microsomes followed Michaelis-Menten kinetics with K(m) = 7.42 +/- 0.37 (+/- SD) microM and Vmax = 4674 +/- 46 pmol.mg protein-1.min-1. Inhibitor studies and immunoinhibition assays showed that VC was activated by cytochrome P450 (CYP) 2E1 down to 1 ppm in the air phase. Based on the metabolic parameters determined, the uptake of VC by rats in vivo can be accurately predicted.

Mutagenicity and metabolism of vinyl chloride and related compounds.

The various adverse biological effects of vinyl chloride appear to be dependent upon the metabolic conversion of this compound into chemically reactive metabolites. The metabolism of vinyl chloride in mammals and in man, including the formation of monochloroacetic acid and some identified sulfur conjugates is reviewed. Hepatic microsomal mixed function oxidases from rats, mice, and humans were equally effective in transforming vinyl chloride into alkylating agents in vitro. Two of the enzyme reaction products, i.e., chloroethylene oxide and 2-chloroacetaldehyde, showed potent genetic activity in microorganisms and Chinese hamster V79 cells. The role of liver microsomal enzymes in the generation of electrophilic mutagenic vinyl chloride metabolites is discussed.

Formation of DNA etheno adducts in rodents and humans and their role in carcinogenesis.

Ethenobases are exocyclic adducts formed with DNA by some environmental carcinogens such as vinyl chloride or urethane. In the last few years, they have received a renewed interest due to the development of sensitive techniques of analysis that made it possible to measure their formation in vivo. This minireview summarizes the information gained recently from the work of several laboratories, including ours. Increased levels of DNA etheno adducts have been measured in target tissues from rodents exposed to vinyl chloride or urethane. Hepatic tumours caused by exposure to vinyl chloride in humans and in rats and lung tumours induced by urethane in mice exhibit base pair substitution mutations in the ras and p53 genes which seem to be exposure-specific and consistent with the promutagenic properties of ethenobases. Background levels of etheno adducts have been detected in DNA from non-exposed humans or animals, pointing to an alternative, endogenous pathway of formation. This background may be affected by dietary factors. It could arise from the reaction of trans-4-hydroxy-2-nonenal (or its epoxide 2,3-epoxy-4-hydroxynonanal), a lipid peroxidation product, with nucleic acid bases. Elevated levels of etheno adducts are found in hepatic DNA from humans and rodents with genetic predisposition to oxidative stress and lipid peroxidation in the liver, and with an associated increased risk of liver cancer. These data suggest that DNA ethenobases could serve as new biomarkers of oxidative stress/lipid peroxidation.

Nucleophilic selectivity and reaction kinetics of chloroethylene oxide assessed by the 4-(p-nitrobenzyl)pyridine assay and proton nuclear magnetic resonance spectroscopy.

The nucleophilic selectivity (Swain-Scott's constant s) of chloroethylene oxide (CEO), an ultimate carcinogenic metabolite of vinyl chloride, was determined to be 0.71 using the 4-(p-nitrobenzyl)pyridine (NBP) assay (Spears method). The molar extinction coefficient of the adduct formed between NBP and CEO was measured; and the second-order rate constants for the reactions of CEO with NBP and with thiosulfate were estimated at three temperatures. The disappearance of CEO and the formation of chloroacetaldehyde (CAA) and glycolaldehyde (GCA) were followed in D2O or a mixture of D2O/hexadeuterated acetone (acetone-d6), using Fourier transform proton nuclear magnetic resonance spectroscopy (1H-FTNMR). Evidence was obtained that CEO reacts with chloride ions to yield CAA at a rate constant of about 17 M-1 h-1 in D2O/acetone-d6 (1 : 1, v/v) at 280 K. Under the same conditions, the first-order rate constant kr for the thermal rearrangement of CEO into CAA was estimated to be approximately 0.41 h-1. These data suggest that the isomerization of CEO may be a minor reaction in physiological saline. These chemical properties of CEO are discussed in relation to the mechanism of vinyl chloride-induced carcinogenesis.

New adducts of chloroethylene oxide and chloroacetaldehyde with pyrimidine nucleosides.

Pyrimidine nucleosides were treated with chloroethylene oxide (CEO) and 2-chloroacetaldehyde (CAA) in methanol and, following trimethylsilylation, the products were analysed by combined gas chromatography-mass spectrometry (GC-MS). Reaction of CEO with 2'-deoxycytidine gave 3,N4-etheno-2'-deoxycytidine and diadduct isomers in which a 1-hydroxy-2-chloroethyl group was substituted for hydrogen on either deoxyribose hydroxyl group. When the N-3-position of 2'-deoxycytidine was blocked by a methyl group, CEO or CAA added a 2-chlorovinyl group at the exocyclic N4 amino nitrogen, as evidenced by a pair of cis/trans isomers. Reaction of 3-methylcytidine and CEO also gave the cis/trans 2-chlorovinyl base adducts, as well as six isomers with a 1-hydroxy-2-chloroethyl group attached to ribose and nine isomeric diadducts, which are possibly positional and optical isomers. Although CEO and CAA were less reactive towards uracil in 3-methyluridine than to cytosine in 3-methyl(deoxy)-cytidine, both electrophiles were able to alkylate 3-methyluridine on ribose, yielding 1-hydroxy-2-chloroethyl derivatives. These data suggest that CEO and CAA may also yield non-cyclic adducts with cytosine in double-stranded DNA where the N-3 position is of low accessibility. Such adducts are of interest in view of their potential promutagenic properties. The data also imply a new mechanism of reaction of CEO with nucleophiles.

Etheno-adduct-forming chemicals: from mutagenicity testing to tumor mutation spectra.

During the past 25 years, ethenobases have emerged as a new class of DNA lesions with promutagenic potential. Ethenobases were first investigated as DNA reaction products of vinyl chloride, an occupational carcinogen causing angiosarcoma of the liver (ASL). They were subsequently shown to be formed by several carcinogenic agents, including urethane (ethyl carbamate), and more recently, to occur in various tissues of unexposed humans and rodents. The endogenous source of ethenobases in DNA is thought to be a lipid peroxidation (LPO) product. Initial studies on metabolic activation, mutagenicity and carcinogenicity moved to the analyses of the formation of ethenobases in vivo and to the determination of their promutagenic properties. Quantification of etheno adducts in vivo became possible with the development of ultrasensitive techniques of analysis. To study the miscoding properties of ethenobases, the initial assays on the fidelity of replication or of transcription were replaced by site-directed mutagenesis assays in vivo. Ethenobases generate mainly base pair substitution mutations. With the advent of new techniques of molecular biology, mutations were investigated in the ras and p53 genes of tumors induced by vinyl chloride and urethane. In liver tumors induced by vinyl chloride, specific mutational patterns were found in the Ki-ras gene in human ASL, in the Ha-ras gene in hepatocellular carcinoma (HCC) in rats, and in the p53 gene in human and rat ASL. In tumors induced by urethane in mice, codon 61 of the Ha-ras gene (liver, skin) and of the Ki-ras gene (lung) seems to be a characteristic target. These tumor mutation spectra are compatible with the promutagenic properties of etheno adducts and with their formation in target tissues, suggesting that ethenobases can be initiating lesions in carcinogenesis. Another recent focus has been given to the repair of etheno adducts, and DNA glycosylases able to excise these adducts in vitro have been identified. The last two decades have brought ethenobases to light as potentially important DNA lesions in carcinogenesis. More research is needed to better understand the environmental and genetic factors that affect the formation and persistence of ethenobases in vivo.

Nucleophilic selectivity as a determinant of carcinogenic potency (TD50) in rodents: a comparison of mono- and bi-functional alkylating agents and vinyl chloride metabolites.

Using published data, the carcinogenic potency (TD50) in rodents of a series of monofunctional alkylating agents, bifunctional antitumor drugs and the vinyl chloride (VC) metabolites chloroethylene oxide (CEO) and chloroacetaldehyde (CAA) was compared to their nucleophilic selectivity (Swain and Scott's constant s or initial ratio of 7-/O6-alkylguanine in DNA). A positive correlation between the log of TD50 estimates and the s values for a series of 14, mostly monofunctional, alkylating agents was observed. This linear relationship also included 2 bifunctional chloroethylnitrosoureas, although their carcinogenic potency was compared to their initial 7-/O6-alkylguanine ratio rather than their s values (n = 16, r = 0.91, p less than 0.005). In addition, the carcinogenic potency of 2 alkyl sulfates, which is not yet known accurately, may correlate with their nucleophilic selectivity through the same relationship. By contrast, 2 methyl halides and 5 bifunctional antitumor drugs (nitrogen mustards and azyridinyl derivatives) did not follow this linear relationship: at similar nucleophilic selectivity, they were more potent carcinogens than the above 18 alkylating agents; this may hold true for CEO and CAA too, although further carcinogenicity experiments are needed to calculate their precise TD50 values. The possible molecular mechanisms involved in tumor induction by these agents are discussed on the basis of these findings. Comparison of the estimated TD50 for CEO, CAA and VC in rodents confirms that CEO is the ultimate carcinogenic metabolite of VC and suggests that only a very small proportion of metabolically generated CEO is available for DNA alkylation in vivo.

Etheno DNA-base adducts from endogenous reactive species.

Promutagenic etheno (epsilon) adducts in DNA are generated through reactions of DNA bases with LPO products derived from endogenous sources or from exposure to several xenobiotics. The availability of sensitive methods has made it possible to detect three epsilon-adducts in vivo, namely epsilon dA, epsilon dC and N2,3-epsilon dG. One probable endogenous source for the formation of these adducts arises from LPO products such as trans-4-hydroxy-2-nonenal (HNE), resulting in highly variable background epsilon-adduct levels in tissues from unexposed humans and rodents. The range of background levels of epsilon dAx10-8dA detected inhuman tissues was <0.05 to 25 and in rodent tissues 0.02 to 10; the corresponding values for epsilon dCx10-8dC were 0.01 to 11 and 0.03 to 24, respectively. Part of this variability may be associated with different dietary intake of antioxidants and/or omega-6 PUFAs which oxidize readily to form 4-hydroxyalkenals, as epsilon dA and epsilon dC levels in WBC-DNA of female volunteers on a high omega-6 PUFA diet were drastically elevated. Increased levels of etheno adducts were also found in the liver of cancer-prone patients suffering from hereditary metal storage diseases, i.e., Wilson's disease (WD) and primary hemochromatosis (PH) as well as in Long-Evans Cinnamon rats, an animal model for WD. Increased metal-induced oxidative stress and LPO-derive epsilon-adducts, along with other oxidative damage, may trigger this hereditary liver cancer. Epsilon-Adducts could hence be explored as biomarkers (i) to ascertain the role of LPO mediated DNA damage in human cancers associated with oxidative stress imposed by certain lifestyle patterns, chronic infections and inflammations, and (ii) to verify the reduction of these epsilon-adducts by cancer chemopreventive agents. This article summarizes recent results on the formation, occurrence and possible role of epsilon-DNA adducts in carcinogenesis and mutagenesis.

recA-independent mutagenicity induced by chloroethylene oxide in E. coli.

The mechanism of mutagenicity of chloroethylene oxide (CEO), an ultimate carcinogenic metabolite of vinyl chloride, was investigated in 3 Escherichia coli strains (E. coli "multitest"). In this system, the mutagenicity of CEO was found to be mainly SOS-independent. CEO did not induce recombinational events at a detection level of about 10(-2) recombinants/survivor. Our results indicate that CEO- (or vinyl chloride-) induced bacterial mutagenesis arises mainly from miscoding DNA adducts.

Induction of specific base-pair substitutions in E. coli trpA mutants by chloroethylene oxide, a carcinogenic vinyl chloride metabolite.

Chloroethylene oxide (CEO), an ultimate carcinogenic metabolite of vinyl chloride, induces base-pair substitution mutations but not frameshift mutations in bacteria. The mutational specificity of CEO was investigated in Escherichia coli, using the trpA mutants developed by Yanofsky. Reversion frequencies to tryptophan prototrophy were analysed, and CEO was found to induce more GC----AT transitions than AT----TA transversions, in addition to a low frequency of other types of substitution. This specificity indicates that CEO is mutagenic through a miscoding DNA adduct. The results are discussed in relation to the various CEO-DNA adducts formed and to their reported or expected mispairing properties.