Modulation of proto-oncogene expression by polychlorinated biphenyls in 3T3-L1 cell line.

The effects of two substituted polychlorinated biphenyls, the 3,4,5,3',4,5' (PCB-169) and the 2,3,4,2',4',5' (PCB-138) forms, were examined on the expression of c-myc, c-jun, c-ras, and jun-b in 3T3-L1 cells. Northern blot analysis demonstrated that the two PCBs, which exhibit a coplanar and di-ortho-substituted configuration, activated these oncogenes differently. PCB-138 markedly induced overexpression of ras, jun, and myc, whereas PCB-169 led to the overexpression of jun-b. High-performance liquid chromatography analysis of the cell samples treated in medium without serum revealed a higher intracellular concentration of the 2,3,4,2',4',5'-hexachlorobiphenyl (hexaCB), whereas the 3,4,5,3',4'5'-hexaCB reached the same concentration in the sonicated samples of cells with or without serum. These results indicated that there was a relationship between PCB structure, bioavailability, and the capacity to stimulate oncogene expression.

Isolation and characterization of RAD51C, a new human member of the RAD51 family of related genes.

The yeast and human RAD51 genes encode strand-transfer proteins that are thought to be involved in both recombinational repair of DNA damage and meiotic recombination. In yeast, the Rad51 family of related proteins also includes Rad55, Rad57 and Dmc1. In mammalian cells, five genes in this family have been identified (HsRAD51, XRCC2, XRCC3, RAD51B/hREC2 and HsDMC1), and here we report the isolation of the sixth member, RAD51C. RAD51C was originally identified by a computer screen of the EST database. A full-length approximately 1.3 kb cDNA clone has been isolated that encodes a protein of 376 aa, having a 18-26% aa identity with other human Rad51 family members. RAD51C includes a previously mapped sequenced-tagged site location near the end of chromosome 17q. The RAD51C transcript is expressed in various human tissues, with highest level of expression in testis, followed by heart muscle, spleen and prostate. Yeast two-hybrid experiments indicate that the Rad51C protein binds to two other members of the Rad51 protein family (Xrcc3 and Rad51B) but not to itself. These findings suggest that Rad51C may function similarly to the yeast Rad55 or Rad57 proteins, rather than as a Rad51 functional homolog.

Suppression of apoptosis by overexpression of Bcl-2 or Bcl-xL promotes survival and mutagenesis after oxidative damage.

Apoptosis is the physiological process by which unwanted cells in an organism are killed. Bcl-2, a membrane-bound cytoplasmic protein, and its close relative Bcl-xL, are both effective inhibitors of apoptosis induced by a wide variety of stimuli in many different cell types. In a previous study, we reported that suppression of apoptosis by Bcl-2 or Bcl-xL, markedly elevates the levels of radiation-induced mutations at the specific locus thymidine kinase. We investigated the effect of the Bcl-2 or Bcl-xL overproduction on hydrogen peroxide-induced mutagenesis. Oxidative DNA damage has been implicated in biological processes such as mutagenesis, carcinogenesis and aging. Overexpression of either Bcl-2 or Bcl-xL enhances oxidative stress mutagenesis in cells with wild type p53 as well as with mutated p53 protein. These results support the hypothesis that apoptosis plays a crucial role in maintaining genomic integrity by selectively eliminating highly mutated cells from the population.

Synthesis and properties of oligonucleotides containing the mutagenic base O4-benzylthymidine.

The preparation of synthetic oligodeoxynucleotides containing O4-benzylthymidine (Tbn) is described. The use of standard and t-butylphenoxyacetyl amino protecting groups is compared. The thermal stabilities of duplexes containing Tbn paired with adenine and guanine have been measured.

Synthesis of a 25 base oligonucleotide containing a styrene oxide modification at the O6 position of 2'-deoxyguanosine at a defined site and incorporation studies of the similarly modified 2'-deoxyguanosine-5'-triphosphate.

A diastereomeric mixture of the regioisomers O6-(2-hydroxy-2-phenylethyl)-2'-deoxyguanosine (st6G, beta-isomer) and O6-(2-hydroxy-1-phenylethyl)-2'-deoxyguanosine (alpha-isomer) was site-specifically placed in a 25 base oligonucleotide template 5'-CCGCTAst6GCGGGTACCGAGCTCGAAT-3' using CED phosphoramidite chemistry. Using 32P-post-labeling we found the oligonucleotide to contain 95% of the beta-isomer and 5% of the alpha-isomer of st6G. st6G as the 3'-phosphate was found to be considerably more acid labile than O6-methyl-2'-deoxyguanosine-3'-phosphate, leading to dealkylation during oligonucleotide synthesis. The diastereomeric mixture of O6-(2-hydroxy-2-phenylethyl)-2'-deoxy-guanosine-5'-triphosphate (st6dGTP) was chemically synthesized and used as a substrate for the exonuclease-free Klenow fragment of Escherichia coli DNA polymerase I. This study demonstrated that st6dGTP could be incorporated opposite deoxycytidine and did not completely block replication.

1,N6-ethenoadenine is preferred over 3-methyladenine as substrate by a cloned human N-methylpurine-DNA glycosylase (3-methyladenine-DNA glycosylase).

A lethal DNA adduct induced by methylating agents, 3-methyladenine (m3A), is removed by both the constitutive (Tag) and inducible (AlkA) bacterial m3A-DNA glycosylases. The human 3-methyladenine-DNA glycosylase also releases m3A as well as other methylated bases. The rate of release of m3A from alkylated DNA by the purified or recombinant human m3A glycosylase is much higher than that of the other methylated bases. We now find that a partially purified recombinant human m3A-DNA glycosylase, expressed in Escherichia coli, releases at least 10-fold more 1,N6-ethenoadenine (epsilon A) than m3A from DNA. epsilon A is completely unrelated to m3A since it is a heterocyclic adduct produced by the carcinogen vinyl chloride. The rates of release of epsilon A and m3A were both dependent on protein concentration and time. The differential release of epsilon A and m3A occurs regardless of whether DNA containing each adduct is assayed separately or is assayed in a mixed substrate containing both DNAs. This result raises the question of what structural features are involved in recognition and excision by the human m3A-DNA glycosylase and what may be its primary substrate.

All four known cyclic adducts formed in DNA by the vinyl chloride metabolite chloroacetaldehyde are released by a human DNA glycosylase.

We have previously reported that human cells and tissues contain a 1,N6-ethenoadenine (epsilon A) binding protein, which, through glycosylase activity, releases both 3-methyladenine (m3A) and epsilon A from DNA treated with methylating agents or the vinyl chloride metabolite chloroacetaldehyde, respectively. We now find that both the partially purified human epsilon A-binding protein and cell-free extracts containing the cloned human m3A-DNA glycosylase release all four cyclic etheno adducts--namely epsilon A, 3,N4-ethenocytosine (epsilon C), N2,3-ethenoguanine (N2,3-epsilon G), and 1,N2-ethenoguanine (1,N2-epsilon G). Base release was both time and protein concentration dependent. Both epsilon A and epsilon C were excised at similar rates, while 1,N2-epsilon G and N2,3-epsilon G were released much more slowly under identical conditions. The cleavage of glycosyl bonds of several heterocyclic adducts as well as those of simple methylated adducts by the same human glycosylase appears unusual in enzymology. This raises the question of how such a multiple, divergent activity evolved in humans and what may be its primary substrate.

Both O4-methylthymine and O4-ethylthymine preferentially form alkyl T.G pairs that do not block in vitro replication in a defined sequence.

The mutagenic potential of O4-methylthymine (m4T) and O4-ethylthymine (e4T) was determined by a primer extension assay on a 25mer oligonucleotide containing a single site-specifically incorporated modified thymine. The e4T-containing oligonucleotide was prepared by using a new synthetic procedure suitable for large alkyl groups on thymine. The second-order rate constants, K(app)m and V(rel)max, permitted calculation of the frequency of formation and extension of modified base pairs compared to Watson-Crick pairing. With both m4T and e4T, the T.G type pairing was formed at least 10-fold more frequently than the nonmutagenic alkyl T.A pairing. However, there was a small but reproducible preference for m4T.G pairing. In both cases T-->C transitions would result. There was no evidence for formation of alkyl T.C or T.T. These data suggest that reported T-->A transversions by ethylation are not likely to result from O4-alkylthymine. In contrast to insertion, extension beyond alkylthymine under kinetic conditions did not occur with alkyl T.A. but only with the alkyl T.G termini. For this latter T.G type pairing, the larger ethyl group did not hinder extension compared to that of the methyl group, in the sequence studied. Under non-limiting conditions of dNTP concentration and time, complete replication could be demonstrated for both methyl- and ethyl-containing oligonucleotides. We conclude that the increase in size of the alkyl group from methyl to ethyl does not significantly affect the mutagenic potential and type of mutations of O4-alkylthymine in vitro.

Evidence from in vitro replication that O6-methylguanine can adopt multiple conformations.

The effect of O6-methylguanine (m6G) on replication, in a partially double-stranded defined 25-base oligonucleotide, has been studied under nonlimiting conditions of unmodified dNTPs and over an extended time period, using the Klenow fragment of Escherichia coli DNA polymerase I. The sequence surrounding m6G has flanking cytosines (C-m6G-C), and the initial steady-state kinetics have been reported. When the primer was annealed so that the first base to be replicated was m6G, replication was virtually complete in approximately 5 min, although the reaction appears biphasic. When annealed with a primer where thymine or cytosine is paired opposite template m6G, about half the molecules were replicated in the first 15 sec, and no significant further replication was seen over a 1-hr period. When m6G was dealkylated by DNA-O6-methylguanine-methyltransferase, replication was rapid with no blockage. These data suggest that there can be two (or more) conformations of m6G. In these studies the term syn refers to conformers interfering with base-pairing, whereas anti refers to those allowing such base-pairing. Previous physical studies by others indicate that syn- and anti-conformers of the methyl group relative to the N1 of guanine are possible. Here molecular modeling/computational studies are described, suggesting that syn- and anti-m6G can be of similar energy in DNA, and, therefore, these two conformers may explain the two types of species observed during in vitro replication. An alternative explanation could be the possibility that the different species may manifest differential interactions of m6G with Klenow fragment. These results may provide a rationale for why m6G lesions in vivo have been reported to be lethal as well as mutagenic.

Further studies of the mixed acetals of nucleosides.

We reported in 1988 on a new nucleoside modification reaction: the exocyclic amino groups of (d)adenosine and (d)cytidine react rapidly at ambient temperature with acetaldehyde and alcohols to give stable mixed acetals (N-ethylethoxy-acetal). NH2 + O = CH(CH3) + ROH-->NH-CH(CH3)-O-R + H2O. Here we report in detail on the occurrence of this reaction in very dilute aqueous solution (ie under biological conditions), on its mechanism and kinetics, on the mixed acetal formation with other aldehydes and other nucleic acid components, and on the question of whether these adducts are mutagenic.

Both purified human 1,N6-ethenoadenine-binding protein and purified human 3-methyladenine-DNA glycosylase act on 1,N6-ethenoadenine and 3-methyladenine.

We previously described a protein, isolated from human tissues and cells, that bound to a defined double-stranded oligonucleotide containing a single site-specifically placed 1,N6-ethenoadenine. It was further demonstrated that this protein was a glycosylase and released 1,N6-ethenoadenine. We now find that this enzyme also releases 3-methyladenine from methylated DNA and that 3-methyladenine-DNA glycosylase behaves in the same manner, binding to the ethenoadenine-containing oligonucleotide and cleaving both ethenoadenine and 3-methyladenine from DNA containing these adducts. The rate and extent of glycosylase activities toward the two adducts are similar.

Partial purification of a human DNA glycosylase acting on the cyclic carcinogen adduct 1,N6-ethenodeoxyadenosine.

We previously reported that a variety of human cells and tissues contained a Mr35,000 DNA-binding protein which selectively recognized a single 1,N6-ethenoadenine in a defined 25-base double-stranded oligonucleotide (B. Rydberg et al., Proc. Natl. Acad. Sci. USA, 88: 6839-6842, 1991). We now demonstrate that incubation of the same duplex with 50-fold partially purified binding protein from human placenta results in release of the free 1,N6-ethenoadenine base, indicative of DNA glycosylase action. This enzyme activity appears unique in that it excises a cyclic adduct resulting from a known human carcinogen.

Kinetics of extension of O6-methylguanine paired with cytosine or thymine in defined oligonucleotide sequences.

The frequency of extending m6G.C or m6G.T pairs, when the 3' and 5' flanking neighbors of m6G are either cytosines or thymines, was investigated using primed 25-base-long oligonucleotides and the Klenow fragment of Escherichia coli DNA polymerase I (Kf). The efficiency, Vmax/Km, of extension to the following normal base pair was up to 40-fold greater than for the formation of the m6G.T or m6G.C pair. The frequencies of inserting either dCMP or dTMP opposite these m6G bases did not appear to be different in the two sequences, C-m6G-C and T-m6G-T, but extension was favored in the C-m6G-C sequence. The m6G.T pair extended to a C.G pair most efficiently, indicating that it was not a strong block to continued replication past the template lesion. Thus, m6G.T flanked by cytosines replicates more readily than when flanked by thymines, increasing G----A transitions. These data lend further support to the importance of sequence context in mutagenesis.

The vinyl chloride DNA derivative N2,3-ethenoguanine produces G----A transitions in Escherichia coli.

Vinyl chloride is a known human and rodent carcinogen that forms several cyclic base derivatives in DNA. The mutagenic potential of these derivatives has been examined in vitro but not in vivo. One of these derivatives, N2,3-ethenoguanine (epsilon G), is known to base pair with both cytosine and thymine during in vitro DNA synthesis, which would result in G----A transitions. To determine the base pairing specificity of this labile guanine derivative in Escherichia coli, we have developed a genetic reversion assay for guanine derivatives. The assay utilizes DNA polymerase-mediated analogue insertion into a bacteriophage vector, M13G*1, which detects all single-base substitutions at position 141 of the lacZ alpha gene by change in plaque color. After the insertion of a single epsilon G opposite the template cytosine at position 141 by use of epsilon dGTP and DNA polymerase and further extension with all four normal dNTPs, the DNA was transfected into E. coli. Transfection of M13G*1 containing epsilon G at the target site yielded 135 mutants among 26,500 plaques, 134 of which represented G----A transitions. The uncorrected mutation frequency was 0.5%, as compared with the control value, approximately 0.02%; when corrected for epsilon G content and penetrance, the calculated mutagenic potential of epsilon G (mutations/analogue) was about 13%. We thus conclude that epsilon G specifically induces G----A transitions during DNA replication in E. coli. The M13G*1 assay may permit the testing of other labile guanine derivatives not otherwise amenable to mutagenesis studies.

Human cells contain protein specifically binding to a single 1,N6-ethenoadenine in a DNA fragment.

A human DNA binding protein has been characterized from cell-free extracts of liver, placenta, and cultured cells. This protein, apparent molecular mass approximately 35 kDa, to our knowledge, does not resemble other proteins reported to bind to carcinogen-modified DNA. The probe used for characterization was a 25-base oligonucleotide containing a single site-specifically placed 1,N6-ethenoadenine (epsilon A), a product of vinyl chloride metabolism. When annealed to form an epsilon A.T or epsilon A.C pair, a strong affinity to the protein was observed, with a binding constant of approximately 1 x 10(9) M-1. In contrast, very little binding was found with an epsilon A.A pair and none was found with an epsilon A.G pair. This suggests protein recognition of a specific structural alteration. Other defined probes with alkyl adducts did not bind. In addition, the human cell extracts and a rat liver extract were found to nick specifically at the 5' side of the epsilon A adduct, which could indicate a possible associated repair activity.

Comparative mutagenesis of O6-methylguanine and O4-methylthymine in Escherichia coli.

The qualitative and quantitative features of mutagenesis by two DNA adducts of carcinogenic alkylating agents, O6-methylguanine (m6G) and O4-methylthymine (m4T), were examined in vivo. The deoxyhexanucleotides 5'-GCTAGC-3' and 5'-GCTAGC-3' were synthesized, where the underlined bases are the positions of m4T or m6G, respectively. By use of recombinant DNA techniques, the respective hexanucleotides or an unmodified control were inserted into a six-base gap in the otherwise duplex genome of the Escherichia coli virus M13mp19-NheI. The duplex adducted genome was converted to single-stranded form and introduced into an E. coli strain that was phenotypically normal with regard to m6G/m4T repair, a strain deficient in repair by virtue of an insertion in the gene encoding the Ada-m6G/m4T DNA methyltransferase, or the same two cell lines after challenge with N-methyl-N'-nitro-N-nitrosoguanidine. Treatment with this alkylating agent chemically compromises alkyl-DNA repair functions. The mutation efficiency of m6G was low or undetectable (0-1.7%) in all cell systems tested, owing, we believe, to rapid repair. In striking contrast, the mutagenicity of m4T was high (12%) in cells fully competent to repair alkylation damage and was roughly doubled when those cells were pretreated with N-methyl-N'-nitro-N-nitrosoguanidine to suppress repair. Taken together, these data suggest that m4T is potentially more mutagenic than m6G and, if formed by a DNA methylating agent, may pose a significant threat to the genetic integrity of an organism.

Evidence for the mutagenic potential of the vinyl chloride induced adduct, N2, 3-etheno-deoxyguanosine, using a site-directed kinetic assay.

N2,3-Ethenoguanine (epsilon G) is a product of vinyl chloride reaction with DNA in vivo and of its ultimate metabolite, chloroacetaldehyde, in vitro. The synthesis of the very labile 5'-triphosphate of N2,3-etheno-deoxyguanosine (epsilon dGuo) has made it possible to study the base pairing properties of this derivative placed opposite a defined normal base in a 25-base oligonucleotide template. The kinetic parameters, Km and Vmax were determined from elongation of a [32P]5'-end labeled primer annealed one base prior to the designated template base, epsilon G.T pairs, which would be mutagenic, were formed with a frequency 2- to 4-fold greater than the analogous wobble pair, G.T. The non-mutagenic pairing, epsilon G.C, occurs with a lower frequency than G.C but neither epsilon G.T or epsilon G.C constitute a significant block to replication. The frequency of epsilon G.T formation was similar with all polymerases tested: Escherichia coli DNA polymerase I (Klenow fragment), exonuclease-free Klenow, Drosophila melanogaster polymerase alpha-primase complex and human immunodeficient virus-I reverse transcriptase (HIV-RT). It is concluded that these prokaryotic and eukaryotic replicating enzymes apparently recognize the same structural features, and on replication G----A transitions would occur, which in turn, could initiate malignant transformation. In contrast to the G.T mismatch which is known to have a specific repair system, etheno derivatives are apparently not repaired in vivo.

Relative efficiencies of the bacterial, yeast, and human DNA methyltransferases for the repair of O6-methylguanine and O4-methylthymine. Suggestive evidence for O4-methylthymine repair by eukaryotic methyltransferases.

The suicidal inactivation mechanism of DNA repair methyltransferases (MTases) was exploited to measure the relative efficiencies with which the Escherichia coli, human, and Saccharomyces cerevisiae DNA MTases repair O6-methylguanine (O6MeG) and O4-methylthymine (O4MeT), two of the DNA lesions produced by mutagenic and carcinogenic alkylating agents. Using chemically synthesized double-stranded 25-base pair oligodeoxynucleotides containing a single O6MeG or a single O4MeT, the concentration of O6MeG or O4MeT substrate that produced 50% inactivation (IC50) was determined for each of four MTases. The E. coli ogt gene product had a relatively high affinity for the O6MeG substrate (IC50 8.1 nM) but had an even higher affinity for the O4MeT substrate (IC50 3 nM). By contrast, the E. coli Ada MTase displayed a striking preference for O6MeG (IC50 1.25 nM) as compared to O4MeT (IC50 27.5 nM). Both the human and the yeast DNA MTases were efficiently inactivated upon incubation with the O6MeG-containing oligomer (IC50 values of 1.5 and 1.3 nM, respectively). Surprisingly, the human and yeast MTases were also inactivated by the O4MeT-containing oligomer albeit at IC50 values of 29.5 and 44 nM, respectively. This result suggests that O4MeT lesions can be recognized in this substrate by eukaryotic DNA MTases but the exact biochemical mechanism of methyltransferase inactivation remains to be determined.

Site-directed mutagenesis for quantitation of base-base interactions at defined sites.

Two alkylation products implicated in initiation of carcinogenesis are O6-alkylguanine (m6G) and O4-alkylthymine (m4T). We have used site-specific insertion of these derivatives into oligonucleotides and measured the kinetic constants of various pairings, using both prokaryotic and eukaryotic polymerases for replication. Preliminary data are also reported for another carcinogen product, N2,3-ethenodeoxyguanosine ( epsilon G). The immediate neighbor bases play an important role in determining the frequency of specific changed basepairing and subsequent elongation of the annealed primer. However, both m4T and m6G prefer to form a type of G.T pairing which would lead to the transitions: G.C----A.T or T.A----C.G. The enzymes were the Klenow fragment of E. coli DNA polymerase I (Kf), engineered 3'----5' exonuclease-free Kf (exo-free Kf), polymerase alpha-primase complex from Drosophila melanogaster or calf thymus, and human immunodeficient virus-I reverse transcriptase (HIV-I RT). All enzymes led to approximately the same frequency of transitions. It is postulated that the mutation frequency at a given site is primarily a function of the structure of the sequence around the target site.

Comparative efficiency of forming m4T.G versus m4T.A base pairs at a unique site by use of Escherichia coli DNA polymerase I (Klenow fragment) and Drosophila melanogaster polymerase alpha-primase complex.

Synthesis of a 25-mer oligonucleotide template containing O4-methylthymine (m4T) at a unique site is reported. The sequence used is analogous to that studied previously to determine the mutation frequency of O6-methylguanine in vitro and in vivo. The templates containing m4T or unmodified T were used in a primer-extension gel assay to determine kinetic parameters for incorporation by DNA polymerases of dGTP and dATP opposite either m4T or T. Both Escherichia coli DNA polymerase I (Klenow fragment, Kf) and Drosophila melanogaster polymerase alpha-primase complex (pol alpha) were used. On the basis of the Vmax/Km ratios, the pairing of m4T.G was preferred over that of both m4T.A and T.G by more than 10-fold. The two polymerases gave almost identical values for the frequency of formation of all pairs investigated including m4T.G pairs, suggesting that the 3'----5' exonuclease activity of the Klenow fragment does not efficiently edit such pairs. Extension beyond m4T.G was demonstrated with both Klenow and pol alpha. In similar kinetic experiments, bacteriophage T4 DNA polymerase, which has a very high 3'----5' exonuclease activity, allows stable incorporation of G opposite m4T in contrast to G opposite T. This kinetic approach allows quantitation of the mutagenic potential in the absence of alkylation repair and additionally provides qualitative data on mutagenesis that are in accord with our previous in vivo studies showing that replication of m4T causes T----C transitions.