Effects of DNA repair deficiency on the mutational specificity in the lacZ gene of Escherichia coli

Chloroethylating anticancer drug-induced mutagenesis and its repair in Escherichia coli

Background

Chloroethylnitrosourea (CENU) derivatives, such as nimustine (ACNU) and carmustine (BCNU), are employed in brain tumor chemotherapy due to their ability to cross the blood-brain barrier. They are thought to suppress tumor development through DNA chloroethylation, followed by the formation of interstrand cross-links (ICLs) that efficiently block replication and transcription. However, the alkylation of DNA and ICLs may trigger genotoxicity, leading to tumor formation as a side effect of the chemotherapeutic treatment. Although the involvement of O⁶-alkylguanine-DNA alkyltransferase (AGT) in repairing chloroethylated guanine (O⁶-chloroethylguanine) has been reported, the exact lesion responsible for the genotoxicity and the pathway responsible for repairing it remains unclear.

Results

We examined the mutations induced by ACNU and BCNU using a series of Escherichia coli strains, CC101 to CC111, in which reverse mutations due to each episome from F’101 to F’106 and frameshift mutations due to each episome from F’107 to F’111 could be detected. The mutant frequency increased in E. coli CC102, which can detect a GC to AT mutation. To determine the pathway responsible for repairing the CENU-induced lesions, we compared the frequency of mutations induced by CENU in the wild-type strain to those in the ada, ogt (AGT-deficient) strain, uvrA (nucleotide excision repair (NER)-deficient) strain, mismatch repair (MMR)-deficient strains, and recA (recombination deficient) strain of E. coli CC102. The frequencies of mutations induced by ACNU and BCNU increased in the ada, ogt strain, demonstrating that O⁶-chloroethylguanines were formed, and that a portion was repaired by AGT.

Mutation induced by ACNU in NER-deficient strain showed a similar profile to that in AGT-deficient strain, suggesting that an NER and AGT play at the similar efficacy to protect E. coli from mutation induced by ACNU. O⁶-Chloroethylguanine is reported to form ICLs if it is not repaired. We examined the survival rates and the frequencies of mutations induced by ACNU and BCNU in the uvrA strain, the recA strain, as well as a double-deficient strain of CC102. The mutation profile of the double-deficient strain was similar to that of the NER-deficient strain, suggesting that an NER protects E. coli from mutations but not recombination. In addition, cell death was more pronounced in the uvrA, recA double-deficient strain than in the single-deficient strains.

Conclusion

These results suggest that the toxic lesions induced by CENU were repaired additively or synergistically by NER and recombination. In other words, lesions, such as ICLs, appear to be repaired by NER and recombination independently.

Carcinogenicity of the chlorination disinfection by-product MX

3-Chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone, better known by its historical name 'mutagen X' or MX, is a chlorination disinfection byproduct that forms from the reaction of chlorine and humic acids in raw water. MX has been measured in drinking water samples in several countries at levels that ranged from non-detectable to 310 ng/L. Although the concentration of MX in drinking water is typically 100- to 1000-fold lower than other common chlorinated by-products of concern (e.g., trihalomethanes), some have hypothesized that MX might play a role in the increased cancer risks that have been associated with the consumption of chlorinated water. This hypothesis is based on observations that MX, in some test systems, is extremely potent relative to trihalomethanes in inducing DNA damage and altering pathways involved in cell growth, and that in some epidemiological studies increased cancer rates are associated with the bacterial mutagenicity of disinfected water of which MX contributes a significant portion. MX also appears to be more potent than other chlorination by-products in causing cancer in animals. This article reviews the available evidence on the carcinogenicity of MX. MX induced cancer at multiple sites in male and female rats, acted as a tumor initiator and promoter, enhanced tumor yields in genetically modified rodents, induced a myriad of genotoxic effects in numerous in vitro and in vivo test systems, and was a potent inhibitor of gap junction intercellular communication. Although the precise mechanism of MX-induced DNA damage is not known, MX is able to cause DNA damage through an unusual mechanism of ionizing DNA bases due to its extremely high reductive potential. MX may also cause mutations through DNA adduction. This article develops a mean cancer potency estimate for MX of 2.3 (mg/kg-d)(-1) and an upper 95% percentile estimate of 4.5 (mg/kg-d)(-1), and examines the potential health risks posed by this chlorination contaminant in drinking water. A discussion of additional data that would be desirable to better characterize the risks posed by MX and other halogenated hydroxyfuranones follows.

Frameshift mutations induced by three classes of acridines in the lacZ reversion assay in Escherichia coli: potency of responses and relationship to slipped mispairing models

The frameshift mutagenicity of 9-aminoacridine (9AA) was compared with that of quinacrine, the acridine mustards ICR-191 and quinacrine mustard (QM), and the nitroacridine Entozon in the lacZ reversion assay in Escherichia coli. As intercalating agents, 9AA and quinacrine cause mutations through noncovalent associations with DNA. Mustards and nitroacridines form covalent adducts in DNA and give rise to different spectra of mutations. Quinacrine and 9AA most effectively induced -1 frameshifts in a run of guanine residues, with 9AA being the more potent mutagen. They also induced +G frameshifts. The acridine mustard ICR-191 was a stronger mutagen than 9AA, owing largely to its potent induction of +G frameshifts. QM induced +G frameshifts more strongly than did its nonreactive counterpart quinacrine. The nitroacridine Entozon differed from the other acridines in being a potent inducer of -2 frameshifts, but it was less effective in inducing +/-1 frameshifts. Quinacrine, although a simple intercalator, induced all five kinds of frameshift mutations detected in the assay, as did the acridine mustards. Although +A and -A frameshifts were induced, adenine runs were less susceptible to acridine mutagenesis than guanine runs. The patterns of frameshift mutagenicity in the lacZ assay are similar to those in an assay based on the reversion of mutations in the tetracycline-resistance gene of the plasmid pBR322. The similarity suggests that the responses reflect the inherent bacterial mutagenicity of the compounds in the local sequence context and are not highly dependent on the broader sequence context. The results are interpreted with respect to slipped mispairing models of frameshift mutagenesis.

Induction of -2 frameshift mutations by 2-nitrofluorene, N-hydroxyacetylaminofluorene, and N-2-acetylaminofluorene in reversion assays in Escherichia coli strains differing in permeability and acetyltransferase activity

The mutagenicity of 2-nitrofluorene (NF), N-hydroxyacetylaminofluorene (N-OH-AAF), and N-2-acetylaminofluorene (AAF) was measured in strains of Escherichia coli that contain a lacZ allele that reverts by -2 frameshift mutations from CG(5) to CG(4). Mutagenesis was compared in a strain having wild-type permeability and metabolism, a strain with increased permeability caused by a lipopolysaccharide-defective (LPS(d)) mutation, a strain with N- and O-acetyltransferase (NAT/OAT) activity conferred by the Salmonella nat gene on plasmid pYG219, and a strain carrying both an LPS(d) mutation and pYG219. The LPS(d) mutation facilitated the measurement of mutagenicity but was not absolutely required, in that lower levels of mutagenicity were detected in LPS(+) strains. The NAT/OAT activity conferred by pYG219 strongly potentiated the mutagenicity of NF and N-OH-AAF. Surprisingly, AAF was mutagenic in the NAT/OAT LPS(d) strain without an exogenous P450 metabolic activation system. Its activity may be ascribable to the detection of a directly mutagenic impurity by the highly sensitive strain or to a low level of metabolic activation by the bacteria under the assay conditions. The findings add to the evidence that the lacZ allele derived from E. coli strain CC109 is an effective indicator of -2 frameshift mutagenesis and that strains expressing high levels of NAT/OAT activity are highly sensitive in monitoring the mutagenicity of nitroarenes and aromatic amides.

Characterization of the mutational specificity of DNA cross-linking mutagens by the Lac+ reversion assay with Escherichia coli

The mutational specificities of DNA cross-linking compounds such as cisplatin, transplatin, carboplatin, mitomycin C, psoralen, and 8-methoxypsoralen were investigated in lacZ reversion assay systems of Escherichia coli. Tester strains were constructed by introducing the six kinds of F' plasmids (lacI-, lacZ461, and proAB+), each of which carries a different base-substitution mutation within the lacZ gene. Each of the six possible base-substitution mutations was assayed by Lac+ reversion. Cisplatin induced G.C-->A.T transitions and G.C-->T.A transversions, with the former predominating. Transplatin induced A.T-->G.C transitions in addition to G.C-->A.T transitions and G.C-->T.A. Carboplatin weakly induced G.C-->A.T transitions. On the other hand, mitomycin C induced only G.C-->T.A transversions, while psoralen and 8-methoxypsoralen reactivated with near-UV irradiation induced A.T-->G.C transitions preferentially. The Lac(+) reversion system was very convenient for rapidly determining mutational spectra.

Comparable DNA and chromosome damage in Chinese hamster ovary cells by chlorohydroxyfuranones

Chlorinated drinking water contains several chlorohydroxyfuranone (CHF) by-products whose contribution to cancer risk is not presently known. 3,4-Dichloro-5-hydroxy-2(5H)-furanone (MCA), 3-chloro-4-(chloromethyl)-5-hydroxy-2(5H)-furanone (CMCF), and 3- chloro-4-methyl-5-hydroxy-2(5H)-furanone (MCF) were studied for the induction of DNA damage, using the alkaline single-cell gel (SCG)/comet assay, and for chromosome damage, using sister-chromatid exchange (SCE) and chromosome aberration (CA) tests, in Chinese hamster ovary (CHO) cells. 3-Chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX), the known genotoxic chlorination by-product and a rat carcinogen, was used as a reference chemical. The SCG analyses were done using concentrations that caused little or no cytotoxicity compared to that of the concurrent control cultures. In the cytogenetic tests, the CHFs were tested up to maximum cytotoxicity. MX and MCA were the most cytotoxic of the compounds in CHO cells followed by CMCF and MCF. All of the CHFs induced DNA damage, SCEs and CAs (mainly chromatid-type breaks and exchanges) in a concentration-related manner, with the exception that MCA was a weak inducer of SCEs. There were no significant differences between the lowest concentration of MX, MCA, and CMCF to cause DNA damage (SCG assay). Based on comparisons of the slopes of regression lines, MX was somewhat more potent than either MCA or CMCF, and MCF was clearly less potent than the other three compounds in the assay. The order of potency was MX > CMCF > MCA > MCF in inducing SCEs and MX > MCA > CMCF > MCF in inducing CAs. The data show that there are differences in the potency of genotoxicity among the CHFs tested. In many cases, however, the extent of maximum effect observed was comparable between the compounds. The results suggest that besides MX other CHFs should be considered in the evaluation of genotoxic risks associated with the consumption of chlorinated drinking water.

Identification of ICR170-induced XPD mutations in UV-sensitive CHO cells

We highlight selected contributions of Dr. Richard Setlow that contributed to our earlier understanding of excision repair processes and set the stage for dissecting nucleotide excision repair (NER) in mammalian cells through molecular genetics. More than 20 years ago, large-scale screens for UV-sensitive mutants of hamster CHO cells isolated approximately 200 mutants, many of which were assigned to the XPD/ERCC2 complementation group, but the nature of the mutations was not determined. The XPD protein performs not only an essential viability function as a structural component of transcription initiation factor TFIIH, but also an NER function as a 5' to 3' DNA helicase within TFIIH that unwinds DNA on the 3' side of bulky lesions. Alterations in these XPD functions are responsible for three UV-sensitivity genetic disorders that have distinguishable clinical features. In this study, we sequenced six UV-sensitive ICR170-induced Chinese hamster ovary (CHO) cell mutants that previously were assigned to the XPD complementation group to determine whether they carry frameshift mutations. All six mutants show 3- to 5-fold increased hypersensitivity to UV irradiation, similar to the XPD mutant prototype UV5. Even though ICR170 is a strong frameshift mutagen, all six cell lines contain base substitution mutations, five of which are unique among all mutations identified so far in human and rodent cells. The sixth mutation was identical to the R75W mutation previously found in CHO UVL-1. The results presented here contribute to a mutation database that should prove useful in structure-function studies of this unique DNA-structure-specific helicase and its complex mutant phenotypes.

The Escherichia coli lacZ reversion mutagenicity assay

The Escherichia coli lacZ reversion assay, based on the set of episomal lacZ alleles engineered by Miller et al., provides an attractive system for studies of mutagenesis and mutational specificity. Each strain in the lacZ set reverts by a specific base substitution or frameshift event. Revertants are selected by growth on lactose minimal medium. In this review, I describe the development of the assay and its subsequent modifications and improvements. Examples of its application are presented and detailed protocols for the implementation of the assay are given.

Mutation spectra in Salmonella of analogues of MX: implications of chemical structure for mutational mechanisms

We determined the mutation spectra in Salmonella of four chlorinated butenoic acid analogues (BA-1 through BA-4) of the drinking water mutagen 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX) and compared the results with those generated previously by us for MX and a related compound, MCF. We then considered relationships between the properties of mutagenic potency and mutational specificity for these six chlorinated butenoic acid analogues. In TA98, the three most potent mutagens, BA-3, BA-4, MX, and the organic extract, all induced large percentages of complex frameshifts (33-67%), which distinguish these agents from any other class of compound studied previously. In TA100, which has only GC sites for mutation recovery, >71% of the mutations induced by all of the agents were GC-->TA transversions. The availability of both GC and TA sites for mutation in TA104 resulted in greater distinctions in mutational specificity than in TA100. MX targeted GC sites almost exclusively (98%); the structurally similar BA-4 and BA-2 produced mutations at similar frequencies at both GC and AT sites; and the structurally similar BA-3 and BA-1 induced most mutations at AT sites (69%). Thus, large variations in structural properties influencing relative mutagenic potency appeared to be distinct from the more localized similar structural features influencing mutagenic specificity in TA104. Among a set of physicochemical properties examined for the six butenoic acids, a significant correlation was found between pK(a) and mutagenic potency in TA100, even when the unionized fraction of the activity dose was considered. In addition, a correlation in CLOGP for BA-1 to BA-4 suggested a role for bioavailability in determining mutagenic potency. These results illustrate the potential value of structural analyses for exploring the relationship between chemical structure and mutational mechanisms. To our knowledge, this is the first study in which such analyses have been applied to structural analogues for which both mutagenic potency and mutation spectra date were available.

No consistent pattern of mutations in p53 and ras genes in liver tumors of rat treated with the drinking water mutagen 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX)

The frequency of point mutations in p53 (exons 4-7) and in Ki-ras, Ha-ras, and N-ras (exons 1 and 2) and the expression of p53 protein were evaluated in the liver tumors of Wistar rats of a 104-week carcinogenicity study on 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX), a chlorine disinfection by-product in drinking water. Mutations were analyzed in 16 hepatocellular adenomas, 7 hepatocellular carcinomas, 23 cholangiomas, and 2 cholangiocarcinomas of the MX-treated animals and one hepatocellular carcinoma and cholangiocarcinoma in control animals using PCR-SSCP (polymerase chain reaction-single-strand conformation polymorphism) or PCR-TGGE (temperature gradient gel electrophoresis) and direct sequencing. The expression of the p53 protein (wild-type and mutated protein) was detected by immunohistochemistry (CM5 antibody). The expression of p53 and that of the proliferating cell nuclear antigen (PCNA, 19 A2) were also evaluated in livers of female animals exposed to MX for 1 week, 3 weeks, or 18 weeks. Altogether, four mutations were found in p53 in three tumors, in two hepatocellular adenomas, and one cholangiocarcinoma, all in females receiving the highest MX dose (6. 6 mg/kg/day) of the study. Three of the mutations were G:C --> A:T transitions and one was an A:T --> T:A transversion. The mutations were scattered at different codons and positions of the codon. One hepatocellular adenoma contained two p53 mutations. All cholangiomas and cholangiocarcinomas, but no hepatocellular adenomas and carcinomas, overexpressed the p53 protein. MX treatment did not induce p53 expression at any age in the liver or alter the expression of the PCNA in the liver of younger animals. The p53 protein was overexpressed in hyperplastic bile ducts in aged rats but not in bile ducts of younger rats (up to 24 weeks). No mutations were observed in either Ki-ras, Ha-ras, or N-ras of the liver tumors. These data suggest that point mutations in p53, Ki-ras, Ha-ras, and N-ras are not involved in the MX-induced liver carcinogenesis in rats.

Mutation spectra of chemical mutagens determined by Lac+ reversion assay with Escherichia coli WP3101P-WP3106P tester strains

We previously reported the development of mutation-specific Escherichia coli B tester strains WP3101 to WP3106 from strain WP2uvrA. In this study we constructed their pKM101-containing derivatives WP3101P to WP3106P, and further isolated their rfa derivatives WP4101-WP4106 and WP4101P-WP4106P. The six kinds of F' plasmids (lacI-, lacZ-, proAB+), each of which carries a different lacZ allele, contained in the above strains were originally derived from E. coli K-12 strains CC101-CC106. All the tester strains show Lac- and Trp- phenotype. Assays for transitions and transversions are based upon Lac+ reversion of a specific mutation located within the lacZ gene on an F' plasmid. The trpE65(ochre) allele in the same strains enables them to be used for Trp+ reversion assays as well. In the present paper, we evaluated the sensitivity, specificity, and usefulness of the newly developed tester strains. Strains WP3101P-WP3106P were highly sensitive to determine mutational profile of heterocyclic amines with S9 mix-mediated metabolic activation and most of the oxidative mutagens and free radical generators tested. Every type of base-pair substitutions induced by 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) or 5-diazouracil were detected in strains WP3101P-WP3106P, while A:T-->C:G and G:C-->A:T mutations induced by MeIQ, and A:T-->C:G, G:C-->A:T, and G:C-->C:G by 5-diazouracil were not detected in pKM101-free tester strains. In pKM101-carrying strains, cumene hydroperoxide induced all types of base substitutions, while formaldehyde preferentially induced G:C-->T:A transversions. Phenazine methosulfate induced predominantly G:C-->A:T transitions and G:C-->T:A transversions, while H2O2 induced predominantly G:C-->T:A and A:T-->T:A transversions. Introduction of the rfa mutation considerably enhanced sensitivity to bulky mutagens such as polycyclic aromatic compounds. All six possible base substitutions induced by 9, 10-dimethyl-1,2-benzanthracene (DMBA) were detected in tester strains WP4101P-WP4106P. In conclusion, our tester strains WP3101P-WP3106P and WP4101P-WP4106P permitted rapid and simple detection of specific mutations induced by variety of mutagens.

Development of new tester strains derived from E. coli WP2uvrA for the determination of mutational specificity

We have developed a set of multipurpose tester strains (WP3101 to WP3106) derived from E. coli WP2uvrA for the detection and classification of mutagens. Six kinds of F' plasmid (lacI, lacZ, proAB+) in strains CC101-CC106, each of which carried a different lacZ allele, were transferred to a delta(lac-pro) derivative of WP2uvrA. Assays for transitions and transversions are based upon Lac+ reversion of a specific mutation located within the lacZ gene on an F' plasmid in strains WP3101-WP3106. In addition, the trpE65(ochre) allele in the same strains is available for Trp+ reversion assays. Using the new tester strains, we investigated the mutational specificities of various chemical mutagens. Base analog mutagens and alkylating mutagens induced specific types of base substitutions. G:C-->A:T transitions and G:C-->T:A transversions predominated in mutagenesis induced by 4-nitroquinoline 1-oxide. Only a slight increase in G:C-->T:A transversions was observed in cells treated with 2-(2-furyl)-3-(5-nitro-2-furyl)acrylamide (AF-2), although the potent mutagenicity of AF-2 was detected in a concurrent Trp+ reversion assay in the same strain. Sodium azide, on the other hand, was negative in the Trp+ reversion assay but specifically induced G:C-->A:T transitions. Present finding suggested that target sites for AF-2- and azide-induced lesions may largely depend on sequence context.

Food-derived heterocyclic amines potentiate the mutagenicity of a drinking water mutagen 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX)

We investigated the enhancing effect of heterocyclic amines on base-substitution mutations with 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 2-amino-3,4-dimethyl-imidazo[4,5-f]quinoline (MeIQ). We compared the mutagenicity of 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX) in the presence and absence of the heterocyclic amines in E. coli WP2 (trpE) and in excision repair-deficient strains WP2s (uvrA, trpE) and ZA500 (uvrA, rfa, trpE). Since the assay was performed without microsomal metabolic activation, Trp-P-1 and MeIQ alone were not mutagenic. In WP2, trp+ reversions induced by MX were greatly potentiated by Trp-P-1 and slightly potentiated by MeIQ. Mutation enhancement was not observed in strains WP2s and ZA500, suggesting that a functional DNA excision repair system is necessary for the combined action of MX and heterocyclic amines. Our finding implies that the combined effect of mutagens as well as the effect of individual mutagens, should be considered in risk evaluation.

Mutation spectra in salmonella of chlorinated, chloraminated, or ozonated drinking water extracts: comparison to MX

Drinking water samples were prepared in a pilot-scale treatment plant by chlorination (Cl2), chloramination (NH2Cl), ozonation (O3), or O3 followed by Cl2 or NH2Cl; and the nonvolatile acidic organics of the raw and treated waters were extracted by XAD/ethyl acetate and evaluated for mutagenicity in Salmonella (-S9). The extracts were 2-8 times more mutagenic in TA100 than in TA98, and the mutagenic potencies of the water extracts ranked similarly in both strains: Cl2 > O3 + Cl2 > NH2Cl > O3 + NH2Cl > O3 > raw. 3-Chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX), which was estimated to account for approximately 20% of the mutagenic activity of the extracts, was shown to be the most potent compound tested thus far in a prophage-induction assay in Escherichia coli and a forward-mutation assay in Salmonella TM677. The mutations in approximately 2,000 revertants of TA98 and TA100 induced by MX and the water extracts were analyzed by colony probe hybridization and polymerase chain reaction/DNA sequence analysis. The water extracts and MX produced similar mutation spectra, which consisted in TA100 of predominantly of GC-->TA transversions in the second position of the CCC (or GGG) target of the hisG46 allele. This spectrum resembles that produced by large aromatic compounds and is distinct from that produced by alkylating agents and the semivolatile drinking water mutagen dichloroacetic acid. In TA98, MX and those water extracts resulting from the introduction of the chlorine atom produced 50-70% hotspot 2-base deletions and 30-50% complex frameshifts (frameshifts with an adjacent base substitution--mostly GC-->TA transversions as found in TA100). No other compound or mixture is known to induce such high frequencies of complex frameshifts. These results suggest that MX and "MX-like" compounds (possibly halogenated aromatics, such as halogenated polycyclic aromatic hydrocarbons) account for much of the mutagenic activity and specificity of the nonvolatile organics in drinking water and that these halogenated organics are especially capable of promoting misincorporation by the DNA replication complex. This study provides further evidence that the mutation spectrum of a complex mixture reflects the dominance of one or a few classes of chemical mutagens within the mixture.