Using polymerase arrest to detect DNA binding specificity of aristolochic acid in the mouse H-ras gene

The distribution of DNA adducts formed by the two main components, aristolochic acid I (AAI) and aristolochic acid II (AAII), of the carcinogenic plant extract aristolochic acid (AA) was examined in a plasmid containing exon 2 of the mouse c-H-ras gene by a polymerase arrest assay. AAI and AAII were reacted with plasmid DNA by reductive activation and the resulting DNA adducts were identified as the previously characterized adenine adducts (dA-AAI and dA-AAII) and guanine adducts (dG-AAI and dG-AAII) by the (32)P-post-labeling method. In addition, a structurally unknown adduct was detected in AAII-modified DNA and shown to be derived from reaction with cytosine (dC-AAII). Sites at which DNA polymerase progress along the template was blocked were assumed to be at the nucleotide 3' to the adduct. Polymerase arrest spectra showed a preference for reaction with purine bases in the mouse H-ras gene for both activated compounds, consistent with previous results that purine adducts are the principal reaction products of AAI and AAII with DNA. Despite the structural similarities among AAI-DNA and AAII-DNA adducts, however, the polymerase arrest spectra produced by the AAs were different. According to the (32)P-post-labeling analyses reductively activated AAI showed a strong preference for reacting with guanine residues in plasmid DNA, however, the polymerase arrest assay revealed arrest sites preferentially at adenine residues. In contrast, activated AAII reacted preferentially with adenine rather than guanine residues and to a lesser extent with cytosine but DNA polymerase was arrested at guanine as well as adenine and cytosine residues with nearly the same average relative intensity. Thus, the polymerase arrest spectra obtained with the AA-adducted ras sequence do not reflect the DNA adduct distribution in plasmid DNA as determined by (32)P-post-labeling. Arrest sites of DNA polymerase associated with cytosine residues confirmed the presence of a cytosine adduct in DNA modified by AAII. For both compounds adduct distribution was not random; instead, regions with adduct hot spots and cold spots were observed. Results from nearest neighbor binding analysis indicated that flanking pyrimidines displayed the greatest effect on polymerase arrest and therefore on DNA binding by AA.     

Etheno adducts formed in DNA of vinyl chloride-exposed rats are highly persistent in liver.

Preweanling rats were exposed to 600 p.p.m. (4h/day) of the human carcinogen vinyl chloride for 5 days to determine the molecular dosimetry of DNA adducts in liver, lung and kidney. 7-(2'-Oxoethyl)guanine (7OEG) was the major DNA adduct detected, representing approximately 98% of all adducts. N2,3-Ethenoguanine (epsilon G) and 3,N4-etheno-2'-deoxycytidine (epsilon dC) were present at approximately 1% of the 7OEG concentration, while 1,N6-etheno-2'-deoxyadenosine was present in even lower concentrations. Liver had 3- to 8-fold higher amounts of the DNA adducts than lung and kidney. The persistence of all four adducts was determined at 3, 7 and 14 days post-exposure. Whereas 7OEG had a t 1/2 of -62 h, all three etheno adducts were highly persistent. After accounting for dilution due to growth-related cell proliferation, epsilon G had a t 1/2 of approximately 30 days, while epsilon dC and epsilon dA were not repaired. These data suggest that these cyclic adducts are poorly recognized by liver DNA repair enzymes and have the potential for accumulation upon chronic exposure. This, coupled with the known miscoding properties of the ethenobases, provides a strong rational for examining their role in vinyl chloride-induced cancer and their utility as biomarkers of exposure.     

Formation of etheno adducts and their effects on DNA polymerases

Etheno (epsilon) and related DNA adducts are formed from the reaction of certain bifunctional electrophiles with DNA. Our interest has been focused on oxiranes substituted with leaving groups, e.g. 2-chlorooxirane, the epoxide derived from the carcinogen vinyl chloride. The chemical mechanisms of the formation of the major etheno products derived from adenine, cytosine and guanine have been elucidated by nuclear magnetic resonance analysis and 13C-labelled precursors. The amounts of all major etheno adducts have been quantified in DNA treated with 2-chlorooxirane by coupled high-performance liquid chromatography of nucleoside and base products. 1,N2-epsilon-Gua, its formally hydrated but stable hemiaminal HO-ethanoGua (5,6,7,9-tetrahydro-7-hydroxy-9-oxoimidazo[1,2-a]purine) and 1,N2-ethanoGua have all been inserted at a single site in oligonucleotides. All three of these bases block polymerases, cause misincorporations and produce some mutations in bacteria. The patterns of blockage and substitution vary among polymerases. In nucleotide excision repair-deficient Escherichia coli, 1,N2-epsilon-Gua yielded a calculated 16% mutation frequency (base-pair substitutions) when the results were corrected for strand usage. 1,N2-epsilon-Gua was also examined in Chinese hamster ovary cells with a stable integration system; the mutants are more complex than observed in bacteria and include rearrangements, deletions and base-pair substitutions other than at the adduct site.     

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.     

Mutagenic and promutagenic properties of DNA adducts formed by vinyl chloride metabolites

Published results and work from this laboratory permit the characterization of the possible promutagenic lesions induced by chloroethylene oxide (CEO) and chloroacetaldehyde (CAA), both known as bifunctional alkylating metabolites of vinyl chloride (VC). The mutagenic effectiveness of CEO and CAA in Escherichia coli, when compared to their nucleophilic selectivity, suggests that the critical target site in DNA bases is not an oxygen atom, and/or that the reaction mechanism of CEO and CAA is different from a simple alkylation. CEO-mutagenicity in E. coli is recA-independent, and CEO preferentially induces GC----AT transitions; accordingly, the mutagenicity of CEO in bacteria may result mainly from a miscoding guanosine or cytosine adduct. Two observations argue against the role of 1,N6-ethenoadenine (epsilon A) and 3,N4-ethenocytosine (epsilon C) in VC-induced mutagenesis/carcinogenesis: i) the lack of detection in double-stranded DNA in vivo and in vitro; ii) the inconsistency between mutational specificity of CEO and miscoding properties of epsilon A and epsilon C. The lack of miscoding properties of 7-(2-oxoethyl)guanine (oxet-G), the major in-vivo VC-DNA adduct, suggests a minor miscoding base adduct. Several lines of evidence point to N4-(2-chlorovinyl)cytosine as one possible putative promutagenic lesion produced by VC, but this compound has yet to be identified in DNA.     

Role of etheno DNA adducts in carcinogenesis induced by vinyl chloride in rats

Vinyl chloride, a hepatocarcinogen in humans and rodents, can form promutagenic etheno bases in DNA after metabolic activation. The formation of 1,N6-ethenoadenine (epsilon A) and 3,N4-ethenocytosine (epsilon C) was measured in adult Sprague-Dawley rats by immunoaffinity purification and 32P-postlabelling. A highly variable background was found in all tissues from untreated animals: the mean molar ratios of epsilon A:A and epsilon C:C in DNA ranged from 0.043 x 10(-8) to 31.2 x 10(-8) and from 0.062 x 10(-8) to 20.4 x 10(-8), respectively. After exposure to 500 ppm vinyl chloride by inhalation (4 h/day, 5 days/week for 8 weeks), increased levels of epsilon A were found in the liver, lung, circulating lymphocytes and testis, the mean (+/- SD) of induced levels (treated-control values) being (4.1 +/- 1.5) x 10(-8) for these tissues. No increase in the epsilon A:A ratio was observed in kidney, brain or spleen. The levels of epsilon C increased in all the tissues examined except the brain. The mean value of the induced epsilon C:C ratios was (7.8 +/- 1.2) x 10(-8) for the liver, kidney, lymphocytes and spleen, and these ratios were higher in the lung (28 x 10(-8)) and testis (19 x 10(-8)). The results suggest a variable repair capacity for epsilon A or epsilon C in different tissues. The results are discussed in relation to published studies on the accumulation and persistence of etheno bases in the liver during and after exposure to vinyl chloride and on mutation spectra in the ras and p53 genes in liver tumours induced by vinyl chloride. In addition, we show that the linear relationship established for monofunctional alkylating agents between their carcinogenic potency in rodents and their covalent binding index for promutagenic bases in hepatic DNA holds for vinyl chloride. It is concluded that etheno bases are critical lesions in hepatocarcinogenesis induced by vinyl chloride. For a better understanding of the mechanism of action of this compound, further work is needed on the role of DNA repair pathways and of endogenous lipid peroxidation products in the formation and persistence of etheno bases in vivo.     

DNA polymerase action on bulky deoxyguanosine and deoxyadenosine adducts

In order to determine how individual hydrocarbon-DNA adducts give rise to specific mutations, a single-stranded oligonucleotide, 5'-T8GT10AT8C2T4CT3CT-3', was reacted with the carcinogen 7-bromomethylbenz[a]anthracene which generates both deoxyguanosine and deoxyadenosine adducts in DNA. The products were separated by HPLC to yield unmodified oligonucleotide and oligonucleotide modified either at the single guanine, or at the single adenine, residue. Incubation of these products with 32P-5'-end-labeled primer, 5'-AGA3GA4G2-3', modified T7 DNA polymerase (Sequenase) and deoxyribonucleoside-5'-triphosphates followed by gel electrophoretic analysis indicated that unmodified oligonucleotide template allowed the primer to be rapidly extended to give species of the same length as the template (40 nucleotides) and of 41 nucleotides in length. However, primer extension for the templates containing the guanine and adenine adducts was held up initially (1 min) at the nucleotide preceding the adduct. At longer times (up to 15 min) a nucleotide was added opposite the adduct and, to a lesser extent, another nucleotide was added beyond this. Some full-length oligonucleotide was also synthesized with these carcinogen-modified templates. When synthesis was allowed to proceed only to the nucleotide preceding the adduct, and this template-extended primer complex incubated with individual nucleotide triphosphates plus Sequenase, it was found that deoxyadenosine residues were most readily incorporated opposite the adduct irrespective of whether it was a deoxyguanosine or deoxyadenosine adduct. These results, which suggest that G.C----T.A and A.T----T.A transversions would be the mutagenic consequences of formation of bulky hydrocarbon adducts at guanines and adenines respectively, are consistent with the most frequent hydrocarbon-induced mutational changes reported thus far.     

Formation and repair of DNA adducts in vinyl chloride- and vinyl fluoride-induced carcinogenesis

Vinyl chloride is a known human and animal carcinogen that induces angiosarcomas of the liver. We review here studies on the formation and repair of DNA adducts associated with vinyl chloride and vinyl fluoride in exposed and control rodents and unexposed humans. These vinyl halides induce etheno (epsilon) adducts that are identical to those formed after lipid peroxidation. Of these adducts, N2,3-ethenoguanine (epsilon G) is present in greatest amounts in tissues of exposed animals. After exposure to vinyl chloride for four weeks, epsilon G levels attain steady-state concentrations, such that the amount of newly formed adducts equals the number of adducts that are lost each day. We report the first dosimetry of epsilon G in rats exposed to 0, 10, 100 or 1100 ppm vinyl chloride for five days or four weeks. The number of adducts increased in a supralinear manner. Exposure to 10 ppm vinyl chloride for five days caused a two- to threefold increase in epsilon G over that of the controls, while four weeks' exposure resulted in a fivefold increase. This was confirmed with [13C2]vinyl chloride and by measuring exogenous and endogenous adducts in the same animals. Exposure to 100 ppm vinyl chloride for four weeks caused a 25-fold increase in epsilon G levels over that found in control rats, while exposure to 1100 ppm resulted in a 42-fold increase. The amount of endogenous epsilon G was similar in liver DNA from rats and humans. A comparable response to exposure was seen in rats and mice exposed to 0, 25, 250 or 2500 ppm vinyl fluoride for 12 months. There was a very high correlation between epsilon G levels in rat and mouse liver at 12 months and the incidence of haemangiosarcoma at two years. We were able to demonstrate that the target cell population for angiosarcoma, the nonparenchymal cells, contained more epsilon G than hepatocytes, even though nonparenchymal cells are exposed by diffusion of vinyl halide metabolites formed in hepatocytes. The expression of N-methylpurine-DNA glycosylase mRNA was induced in rat liver after exposure to either 25 or 2500 ppm vinyl fluoride. When this induction was investigated in hepatocytes and nonparenchymal cells, it was found that the latter had only 20% of the N-methylpurine-DNA glycosylase mRNA of hepatocytes, and that only the hepatocytes had induction of this expression after exposure to vinyl fluoride. Thus, the target cells for vinyl halide carcinogenesis have much lower expression of this DNA repair enzyme, which has been associated with etheno adduct repair.     

Mammalian enzymatic repair of etheno and para-benzoquinone exocyclic adducts derived from the carcinogens vinyl chloride and benzene

Two human carcinogens that have been extensively studied are vinyl chloride and benzene. The active metabolites used in this study are chloroacetaldehyde (CAA) and para-benzoquinone (pBQ). Each forms exocyclic adducts between the N1 and N6 of A, the N3 and N4 of C and the N1 and N2 of G. Only CAA has been found to form the N2,3 adduct of G. CAA and pBQ adducts differ structurally in size and in the number of added rings, pBQ adding two rings to the base, while etheno bases have a single five-membered ring. The mechanism of repair of these two types of adducts by human enzymes has been studied in our laboratory with defined oligodeoxynucleotides and a site-specific adduct. The etheno derivatives are repaired by DNA glycosylase activity; two mammalian glycosylases are responsible: alkylpurine-DNA-N-glycosylase (APNG) and mismatch-specific thymine-DNA glycosylase. The former repairs 1,N6-ethenoA (epsilon A) as rapidly as the original substrate, 3-methyladenine, while the latter repairs 3,N4-ethenoC (epsilon C) more efficiently than the G/T mismatch. Our finding that there are separate enzymes for epsilon A and epsilon C has been confirmed by the use of tissue extracts from an APNG knockout mouse. As pBQ is much less efficient than CAA in modifying bases, the biochemical studies required total synthesis of the nucleosides. Furthermore, the pBQ adduct-containing oligomers are cleaved, to various extents by a different class of enzyme: human and bacterial N-5'-alkylpurine (AP) endonucleases. The enzyme incises such oligomers 5' to the adduct and generates 3'-hydroxyl and 5'-phosphoryl termini but leaves the modified base on the 5'-terminus of the 3' cleavage fragment ('dangling base'). Using active-site mutants of the human AP endonuclease, we found that the active site for the primary substrate, abasic (AP) site, is the same as that for the bulky pBQ adducts. There appears to be no clear rationale for the widely differing recognition and repair mechanisms for these exocyclic adducts, as shown for the repair of the same types of modification on different bases (e.g. epsilon A and epsilon C) and for completely unrelated lesions (e.g. AP site and pBQ adducts). Another important variable that affects the rate and extent of repair is the effect of neighbouring bases. In the case of epsilon A, this sequence-dependent repair correlates with the extent of double-strandedness of the substrate, as demonstrated by thermal stability studies.     

DNA polymerase-mediated nucleotide incorporation adjacent to hydrocarbon-deoxyadenosine and hydrocarbon-deoxyguanosine adducts

To examine the effect of DNA adducts on nucleotide incorporation by DNA polymerase at 3' neighboring bases, synthetic oligonucleotides (16mers) containing a purine at position 13 from the 3' end and any one of the four possible bases at position 12 were prepared and reacted with 7-bromomethylbenz[a]anthracene. Using HPLC, unmodified oligonucleotide was separated from oligonucleotide containing a single adduct, at either an adenine or a guanine residue. These products were annealed with a 32P 5'-end labeled primer (11mer) and incubated with modified T7 DNA polymerase (Sequence, version 2.0) in the presence of deoxyribonucleoside 5'-triphosphates. Analysis by gel electrophoresis showed that unmodified oligonucleotide template allowed the primer to be rapidly extended to the entire length of the template. However, the presence of an adduct caused primer extension to stop at the base 3' to the adduct. While correct base pairing occurred at this termination site with most adducted templates, there was a high frequency of misincorporation of guanine opposite a thymine located 3' to an adenine adduct. This result suggest that some bulky carcinogen--DNA adducts may lead to base mismatches at neighboring bases.     

Detection of 1,N6-etheno-2'-deoxyadenosine and 3,N4-etheno-2'-deoxycytidine occurring endogenously in DNA

1,N6-Etheno-2'-deoxyadenosine (epsilon dA) and 3,N4-etheno-2'-deoxycytidine (epsilon dC) are DNA adducts formed by a number of genotoxic chemicals, including vinyl chloride. They are also formed endogenously in tissue DNA, probably from a reactive metabolite of lipid peroxidation. Both the qualitative and quantitative detection of endogenous adducts is important in order to place adduct formation by chemicals such as vinyl chloride in the context of this natural background level. Methods with sufficient sensitivity are therefore being developed to measure the natural background of epsilon dA and epsilon dC adducts. We have developed a high-performance liquid chromatography (HPLC)-32P-postlabelling method to measure epsilon dA and epsilon dC at alkylation frequencies of 1 adduct in 10(7)-10(8) nucleotides in 10-microgram samples of DNA. In HPLC-32P-postlabelling analysis of liver DNA from control Wistar rats, epsilon dA and epsilon dC were determined at levels of 1 adduct in 8.1 x 10(7) and 1 adduct in 1.8 x 10(7) nucleotides, respectively. The levels of epsilon dA and epsilon dC measured in liver DNA of animals exposed orally to five daily doses of 50 mg/kg body weight vinyl chloride were found by this method to be 1 adduct in 2.9 x 10(7) and 1 adduct in 1.4 x 10(7) nucleotides, respectively. In contrast, in a direct labelling study, radiolabelled epsilon dA and epsilon dC were not detected in liver DNA of rats exposed for 6 h by nose-only inhalation to [1,2-14C]vinyl chloride at up to 45 ppm v/v. Immunochemical procedures are also being developed for recognizing etheno adducts. Thus, a monoclonal antibody raised to protein conjugates of epsilon dC showed high selectivity in the recognition of this DNA adduct. When the antibody was immobilized on a solid support and used in an immunoenrichment procedure to purify epsilon dC from a large excess of normal nucleotides, one epsilon dC adduct from about 10(8) normal nucleotides could be resolved. Coupling the immunoaffinity enrichment procedure with capillary zone electrophoresis permitted the detection of approximately one epsilon dC adduct in 3 x 10(6) nucleotides.     

Solution conformation and mutagenic specificity of 1,N6-ethenoadenine

Site-specific studies in several laboratories established that each of the three etheno adducts, 1,N6-ethenoadenine (epsilon A), 3,N4-ethenocytosine (epsilon C) and N2,3-ethenoguanine (N2,3-epsilon G), is mutagenic. In Escherichia coli, epsilon A is only weakly mutagenic in single-stranded DNA (mutation frequency, 0.1%), and epsilon C is at least 20 times more mutagenic than epsilon A. Prior treatment of host cells with ultraviolet irradiation enhances the mutagenic frequency of epsilon C by 30-60%, even when the E. coli is recA. Likewise, enhanced mutagenicity was observed when the host cells lacked 3'-->5' exonuclease activity of DNA polymerase III. epsilon A induces all three base substitutions, but A-->G predominates. epsilon C induces epsilon C-->T and epsilon C-->A substitutions, but only the latter was enhanced after ultraviolet irradiation of host cells. In contrast to the results in bacteria, both epsilon A and epsilon C are potent mutagenic lesions in simian kidney cells, inducing 70 and 81% base substitutions, respectively. In simian kidney cells, epsilon A exclusively induces epsilon A-->G transitions, whereas epsilon C-->A transversions are the major type of mutation induced by epsilon C. Nuclear magnetic resonance (NMR) spectrometry of the four possible pairs containing epsilon C indicated that both epsilon C:G and epsilon C:T pairs are stabilized by hydrogen bonds. Even though the latter forms the most stable pair containing epsilon C, the etheno adduct is in syn alignment. DNA polymerase appears to continue DNA synthesis with a syn-orientated base only in the absence of proofreading exonuclease activity or when ultraviolet irradiation-inducible proteins are present. For epsilon A, only epsilon A:T and epsilon A:G pairs have been studied by NMR, which showed that the former has no hydrogen bond whereas the latter maintains two hydrogen bonds with the etheno base in syn orientation. Determination of the relationship between a particular conformation of epsilon A and its mutagenic activity must await further studies. In a site-specific study of epsilon A with human cell extracts, an 11-mer oligonuclotide with a single epsilon A was inserted into an M13 bacteriophage containing an SV40 origin of replication. This vector was replicated in vitro with human fibroblast cell extracts, and the replicated products were analysed. In this experiment, epsilon A induced predominantly epsilon A-->G transitions but at a mutation frequency of 0.14%.     

Adduction of human p53 gene by fecal water: an in vitro biomarker of mutagenesis in the human large bowel.

A polymerase arrest assay was designed to determine sites of adduction in the human p53 gene induced by incubation with fecal water. Significant formation of adducts was observed on p53 DNA after a 2-h incubation in fecal water from 10 of 17 samples studied. Large sample-to-sample variation was observed. The major sites of polymerase termination occurred at nucleotides 3' to guanine residues. Adduct sites coincided with colorectal cancer p53 mutation "hotspots," highlighting the potential carcinogenicity of fecal material.     

32P-Postlabelling of DNA adducts formed by allyl glycidyl ether in vitro and in vivo

32P-Postlabelling analysis of allyl glycidyl ether-treated DNA after adduct enrichment on anion-exchange cartridges revealed two major and one minor DNA adducts. The major adducts were shown to originate from alkylation at N-7-guanine and N-1-adenine, respectively, while the minor adduct was at N-3-cytosine. In addition, rearrangement products of the 1-adenine and 3-cytosine adducts to N6-adenine and 3-uracil were indicated. The relative amounts of adenine, cytosine and uracil products appeared to be dependent upon conditions (in particular pH) during sample processing and analysis. When nuclease P1 was used for adduct enrichment the adenine, cytosine and uracil adducts, but not the 7-guanine adduct, were detected. The labelling efficiency of the 7- guanine adduct standard was 40-45%. Total recovery of this adduct from allyl glycidyl ether-modified DNA was 9-12%. The labelling efficiency of the 1-adenine adduct standard was 78-82%. Total recovery of this adduct from DNA was approximately 20% when using anion-exchange chromatography for adduct enrichment and 30-34% when using nuclease P1. Preliminary analysis of DNA from mice treated with allyl glycidyl ether indicated 57 times higher level of the 7-guanine adduct, per unit dose, in skin DNA (120 per 10(8) normal nucleotides) after topical application when compared to liver DNA after i.p. administration. The 1- adenine adduct could not be quantified in liver DNA (due to an interfering background product present in untreated animals) and the level of the 3-cytosine adduct was below the detection limit of the method. After topical application the level of the 1 adenine adduct in skin DNA was approximately 30 per 10(8), using either column or nuclease P1 enrichment. The 3-cytosine adduct was detected in skin, but was not quantified.      

Dna polymerase arrest by adducted trivalent chromium

Carcinogenic chromium (Cr⁶⁺) enters cells via the sulfate transport system and undergoes intracellular reduction to trivalent chromium, which strongly adducts to DNA. In this study, the effect of adducted trivalent chromium on in vitro DNA synthesis was analyzed with a polymerase-arrest assay in which prematurely terminated replication products were separated on a DNA sequencing gel. A synthetic DNA replication template was treated with increasing concentrations of chromium(III) chloride. The two lowest chromium doses used resulted in biologically relevant adduct levels (6 and 21 adducts per 1,000 DNA nucleotides) comparable with those measured in nuclear matrix DNA from cells treated with a 50% cytotoxic dose of sodium chromate in vivo. In vitro replication of the chromium-treated template DNA using the Sequenase version 2.0 T7 DNA polymerase (United States Biochemical Corp., Cleveland, OH) resulted in dose-dependent polymerase arrest beginning at the lowest adduct levels analyzed. The pattern of polymerase arrest remained consistent as chromium adduct levels increased, with the most intense arrest sites occurring 1 base upstream of guanine residues on the template strand. Replication by the DNA polymerase I large (Klenow) fragment as well as by unmodified T7 DNA polymerase also resulted in similar chromium-induced polymerase arrest. Interstrand crosslinking between complementary strands was detected in template DNA containing 62, 111, and 223 chromium adducts per 1,000 DNA nucleotides but not in template containing 6 or 21 adducts per 1,000 DNA nucleotides, in which arrest nevertheless did occur. Low-level, dose-dependent interstrand cross-linking between primer and template DNA, however, was detectable even at the lowest chromium dose analyzed. Since only 9% of chromium adducts resulted in polymerase arrest in this system, we hypothesized that arrest occurred when the enzyme encountered chromium-mediated interstrand DNA-DNA cross-links between either the template and a separate DNA molecule or the template and its complementary strand in the same molecule. These results suggest that the obstruction of DNA replication by chromium-mediated DNA-DNA cross-links is a potential mechanism of chromium-induced genotoxicity in vivo. © 1994 Wiley-Liss, Inc.     

DNA adducts of the antitumor agent diaziquone

We have studied adduct formation of the antineoplastic agent diaziquone (AZQ; NSC 182986) with DNA and nucleotides in vitro. The aziridine moieties of AZQ can be expected to interact covalently with DNA which, in turn, presumably elicits the antitumor activity. We analyzed AZQ-DNA adducts by a modified 32P-postlabeling assay involving purification of the nuclease P1-enriched labeled adducts by high-salt C18 reversed-phase thin-layer chromatography and separation of the eluted adducts on a polyethyleneimine-cellulose layer using non-urea salt solutions. Modification of calf thymus DNA with AZQ produced two major (22% and 40%) and at least eight minor adducts. At equal concentrations of AZQ and DNA (1 micrograms/microliters each), peak binding was observed in about 2 h [1926 +/- 378 (SD) fmol/micrograms of DNA] with the binding levels remaining practically unchanged through 4 h. However, incubation for 24 h resulted in over 40% decline, indicating adduct instability. AZQ was found to be highly reactive in vitro as evidenced by its substantial binding (49 +/- 14 fmol/micrograms of DNA) even at a DNA:AZQ ratio of 100:1. When incubated with mononucleotides, AZQ reacted extensively with adenine, guanine, and cytosine but only slightly with thymine. Cochromatography of the modified DNA and nucleotides revealed that one of the major adducts and several minor adducts were guanine derived. The aziridine rings of AZQ were found to be the main reactive sites as its monoaminoalcohol derivative showed as much DNA reactivity as did the parent compound, but no activity was observed when both aziridine groups were hydrolyzed to diaminoalcohols. The improved 32P-postlabeling assay seems capable of detecting relatively polar adducts such as those formed with AZQ at a level of one adduct/10(9) nucleotides.     

AlkB homologue 2-mediated repair of ethenoadenine lesions in mammalian DNA

Endogenous formation of the mutagenic DNA adduct 1,N(6)-ethenoadenine (epsilon A) originates from lipid peroxidation. Elevated levels of epsilon A in cancer-prone tissues suggest a role for this adduct in the development of some cancers. The base excision repair pathway has been considered the principal repair system for epsilon A lesions until recently, when it was shown that the Escherichia coli AlkB dioxygenase could directly reverse the damage. We report here kinetic analysis of the recombinant human AlkB homologue 2 (hABH2), which is able to repair epsilon A lesions in DNA. Furthermore, cation exchange chromatography of nuclear extracts from wild-type and mABH2(-/-) mice indicates that mABH2 is the principal dioxygenase for epsilon A repair in vivo. This is further substantiated by experiments showing that hABH2, but not hABH3, is able to complement the E. coli alkB mutant with respect to its defective repair of etheno adducts. We conclude that ABH2 is active in the direct reversal of epsilon A lesions, and that ABH2, together with the alkyl-N-adenine-DNA glycosylase, which is the most effective enzyme for the repair of epsilon A, comprise the cellular defense against epsilon A lesions.     

Quantitation of etheno adducts by fluorescence detection

Etheno adducts can be formed by the reaction of vinyl chloride metabolites with DNA and may play a role in the carcinogenicity of this chemical. These adducts are highly fluorescent and may be quantitated by sensitive photometric methods in conjunction with high-performance liquid chromatographic (HPLC) separation. Three HPLC systems were evaluated on the basis of maximal fluorescence intensity and resolution of two etheno adducts, ethenodeoxycytidine and ethenodeoxyadenosine. Analyses were conducted with enzymatically digested DNA that had been incubated with chloroacetaldehyde, a vinyl chloride metabolite which may cause etheno adduct formation in vivo. All three known etheno adducts of DNA were tentatively identified in DNA reacted in vitro. The sensitivity of the method was highest for the ethenodeoxyadenosine adduct, with the limit of detection (1 pmol per injection in the HPLC system) being similar to that for O6-methylguanine, another promutagenic DNA adduct for which quantitation by HPLC with fluorescence detection has been reported. The method described here may be useful for the analysis of DNA from animals or humans exposed to vinyl chloride.