Mammalian DNA Polymerase Beta Research Articles

Structures of Minimal Catalytic Fragments of Topoisomerase V Reveals Conformational Changes Relevant for DNA Binding

Topoisomerase V is an archaeal type I topoisomerase that is unique among topoisomerases due to presence of both topoisomerase and DNA repair activities in the same protein. It is organized as an N-terminal topoisomerase domain followed by 24 tandem helix-hairpin-helix (HhH) motifs. Structural studies have shown that the active site is buried by the (HhH) motifs. Here we show that the N-terminal domain can relax DNA in the absence of any HhH motifs and that the HhH motifs are required for stable protein-DNA complex formation. Crystal structures of various topoisomerase V fragments show changes in the relative orientation of the domains mediated by a long bent linker helix, and these movements are essential for the DNA to enter the active site. Phosphate ions bound to the protein near the active site helped model DNA in the topoisomerase domain and show how topoisomerase V may interact with DNA.

The inhibitory action of kohamaic acid A derivatives on mammalian DNA polymerase beta.

We previously isolated a novel natural product, designated kohamaic acid A (KA-A, compound 1), as an inhibitor of the first cleavage of fertilized sea urchin eggs, and found that this compound could selectively inhibit the activities of mammalian DNA polymerases (pols). In this paper, we investigated the structure and bioactivity of KA-A and its chemically synthesized 11 derivatives (i.e., compounds 2–12), including KA-A - fatty acid conjugates. The pol inhibitory activity of compound 11 [(1S*,4aS*,8aS*)-17-(1,4,4a,5,6,7,8,8a-octahydro-2,5,5,8a-tetramethyl-naphthalen-1-yl)heptadecanoic acid] was the strongest among the synthesized compounds, and the range of IC50 values for mammalian pols was 3.22 to 8.76 μM; therefore, the length of the fatty acid side chain group of KA-A is important for pol inhibition. KA-A derivatives could prevent human cancer cell (promyelocytic leukemia cell line, HL-60) growth with the same tendency as the inhibition of mammalian pols. Since pol β is the smallest molecule, we used it to analyze the biochemical relationship with KA-A derivatives. From computer modeling analysis (i.e., docking simulation analysis), these compounds bound selectively to four amino acid residues (Leu11, Lys35, His51 and Thr79) of the N-terminal 8-kDa domain of pol β, and the binding energy between compound 11 and pol β was largest in the synthesized compounds. The relationship between the three-dimensional molecular structures of KA-A-related compounds and these inhibitory activities is discussed.

Structure and function of 2:1 DNA polymerase·DNA complexes

DNA polymerases are required for DNA replication and DNA repair in all of the living organisms. Different DNA polymerases are responsible different stages of DNA metabolism, and many of them are multifunctional enzymes. It was generally assumed that the different reactions are catalyzed by the same enzyme molecule. In addition to 1:1 DNA polymerase.DNA complex reported by crystallization studies, 2:1 and higher order DNA polymerase.DNA complexes have been identified in solution studies by various biochemical and biophysical approaches. Further, abundant evidences for the DNA polymerase-DNA interactions in several DNA polymerases suggested that the 2:1 complex represents the more active form. This review describes the current status of this emerging subject and explores their potential in vitro and in vivo functional significance, particularly for the 2:1 complexes of mammalian DNA polymerase beta (Pol beta), the Klenow fragment of E. coli DNA polymerase I (KF), and T4 DNA polymerase.

Mismatched dNTP incorporation by DNA polymerase β does not proceed via globally different conformational pathways†

Understanding how DNA polymerases control fidelity requires elucidation of the mechanisms of matched and mismatched dNTP incorporations. Little is known about the latter because mismatched complexes do not crystallize readily. In this report, we employed small-angle X-ray scattering (SAXS) and structural modeling to probe the conformations of different intermediate states of mammalian DNA polymerase β (Pol β) in its wild-type and an error-prone variant, I260Q. Our structural results indicate that the mismatched ternary complex lies in-between the open and the closed forms, but more closely resembles the open form for WT and the closed form for I260Q. On the basis of molecular modeling, this over-stabilization of mismatched ternary complex of I260Q is likely caused by formation of a hydrogen bonding network between the side chains of Gln260, Tyr296, Glu295 and Arg258, freeing up Asp192 to coordinate MgdNTP. These results argue against recent reports suggesting that mismatched dNTP incorporations follow a conformational path distinctly different from that of matched dNTP incorporation, or that its conformational closing is a major contributor to fidelity.

DNA Polymerase β Catalysis: Are Different Mechanisms Possible?

DNA polymerases are crucial constituents of the complex cellular machinery for replicating and repairing DNA. Discerning mechanistic pathways of DNA polymerase on the atomic level is important for revealing the origin of fidelity discrimination. Mammalian DNA polymerase beta (pol beta), a small (39 kDa) member of the X-family, represents an excellent model system to investigate polymerase mechanisms. Here, we explore several feasible low-energy pathways of the nucleotide transfer reaction of pol beta for correct (according to Watson-Crick hydrogen bonding) G:C basepairing versus the incorrect G:G case within a consistent theoretical framework. We use mixed quantum mechanics/molecular mechanics (QM/MM) techniques in a constrained energy minimization protocol to effectively model not only the reactive core but also the influence of the rest of the enzymatic environment and explicit solvent on the reaction. The postulated pathways involve initial proton abstraction from the terminal DNA primer O3'H group, nucleophilic attack that extends the DNA primer chain, and elimination of pyrophosphate. In particular, we analyze several possible routes for the initial deprotonation step: (i) direct transfer to a phosphate oxygen O(Palpha) of the incoming nucleotide, (ii) direct transfer to an active site Asp group, and (iii) transfer to explicit water molecules. We find that the most probable initial step corresponds to step (iii), involving initial deprotonation to water, which is followed by proton migration to active site Asp residues, and finally to the leaving pyrophosphate group, with an activation energy of about 15 kcal/mol. We argue that initial deprotonation steps (i) and (ii) are less likely as they are at least 7 and 11 kcal/mol, respectively, higher in energy. Overall, the rate-determining step for both the correct and the incorrect nucleotide cases is the initial deprotonation in concert with nucleophilic attack at the phosphate center; however, the activation energy we obtain for the mismatched G:G case is 5 kcal/mol higher than that of the matched G:C complex, due to active site structural distortions. Taken together, our results support other reported mechanisms and help define a framework for interpreting nucleotide specificity differences across polymerase families, in terms of the concept of active site preorganization or the so-called "pre-chemistry avenue".

A Unified Kinetic Mechanism Applicable to Multiple DNA Polymerases,

After extensive studies spanning over half a century, there is little consensus on the kinetic mechanism of DNA polymerases. Using stopped-flow fluorescence assays for mammalian DNA polymerase beta (Pol beta), we have previously identified a fast fluorescence transition corresponding to conformational closing, and a slow fluorescence transition matching the rate of single-nucleotide incorporation. Here, by varying pH and buffer viscosity, we have decoupled the rate of single-nucleotide incorporation from the rate of the slow fluorescence transition, thus confirming our previous hypothesis that this transition represents a conformational event after chemistry, likely subdomain reopening. Analysis of an R258A mutant indicates that rotation of the Arg258 side chain is not rate-limiting in the overall kinetic pathway of Pol beta, yet is kinetically significant in subdomain reopening. We have extended our kinetic analyses to a high-fidelity polymerase, Klenow fragment (KF), and a low-fidelity polymerase, African swine fever virus DNA polymerase X (Pol X), and showed that they follow the same kinetic mechanism as Pol beta, while differing in relative rates of single-nucleotide incorporation and the putative conformational reopening. Our data suggest that the kinetic mechanism of Pol beta is not an exception among polymerases, and furthermore, its delineated kinetic mechanism lends itself as a platform for comparison of the kinetic properties of different DNA polymerases and their mutants.

A dideoxynucleotide‐sensitive DNA polymerase activity characterized from endoreduplicating cells of mungbean (Vigna radiata L.) during ontogeny of cotyledons

Within this work we describe the purification and biochemical characterization of a ddNTP-sensitive DNA polymerase purified from mungbean (Vigna radiata cv B1, L.) seeds at 18 days after fertilization, when > 70% of the nuclei are reported to be in the endoreduplicated state. The purified enzyme is a single polypeptide of 62 kDa and many of its physicochemical properties are similar to those of mammalian DNA polymerase beta. Similar to the other X-family DNA polymerases, it lacks 3'-5' exonuclease activity and has short gap-filling and strand-displacement activity. The enzyme shows moderately processive DNA synthesis on a single-strand template. The determined N-terminal heptapeptide sequence of the enzyme showed clear homology with helix 1 of the N-terminal single strand DNA-binding domain (residues 32-41) of rat and human DNA polymerase beta. These results represent the first evidence for the identification and characterization of a ddNTP-sensitive DNA polymerase expressed during the endoreduplication cycle that shares biochemical and immunological similarity with mammalian DNA polymerase beta.

Affinity Capture of a Mammalian DNA Polymerase β by Inhibitors Immobilized to Resins Used in Solid-Phase Organic Synthesis

The application of resins normally used in solid-phase organic synthesis to the affinity capture of a mammalian DNA polymerase beta (pol beta) is reported. Lithocholic acid (LCA), an inhibitor of pol beta, was immobilized on various solid supports, and the batch affinity purification of pol beta from a mixture of proteins using these LCA-immobilized resins was examined. Of the resins tested, TentaGel was the most effective at purifying pol beta and at resisting nonspecific absorption of proteins. The immobilized LCA recognized pol beta specifically, which resulted in pol beta binding to the resin. Using the LCA-immobilized resin, it was possible to purify pol beta from a mixture of proteins. Furthermore, it was possible to concentrate pol beta from a crude nuclear extract of human T lymphoma Molt4 cells. To facilitate the immobilization of compounds on TentaGel resins, we also designed and prepared photoaffinity beads containing a photoreactive group at the free termini of the TentaGel resin. The pol beta inhibitors LCA, C18-beta-SQDG, and epolactaene were immobilized on the photoaffinity beads by photoreaction. The batch affinity purification of pol beta from a protein mixture could be also achieved with these beads.

A New Highly Efficient Photoreactive Analogue of dCTP. Synthesis, Characterization, and Application in Photoaffinity Modification of DNA Binding Proteins

A new base-substituted analogue of dCTP, exo-N-{2-[N-(4-azido-2,5-difluoro-3-chloropyridine-6-yl)-3-aminopropionyl]aminoethyl}-2'-deoxycytidine-5'-triphosphate (FAP-dCTP) has been synthesized and characterized. FAP-dCTP is an efficient substrate of mammalian DNA polymerase beta in the reaction of primer elongation displaying substrate properties as an analogue of dCTP and dTTP. FAP-dCTP was used for the photoaffinity modification of mammalian DNA polymerase beta. Two approaches to photoaffinity labeling were utilized. In one approach, photoreactive FAP-dCTP was first incorporated into radiolabeled primer-template, and photoreactive DNA was UV-irradiated in the presence of DNA polymerase beta, which resulted in the polymerase labeling by photoreactive primer. In an alternate approach, FAP-dCTP was first UV-cross-linked to the enzyme; subsequently, radiolabeled primer-template was added, and the enzyme-linked FAP-dCTP was incorporated into the 3'-end of radioactive primer. This "catalytic" modification pathway was shown to be less specific in recognition of FAP-dCTP as an analogue of dCTP than dTTP. FAP-dCTP was used as substrate of endogenous DNA polymerases of HeLa cell extract to synthesize photoreactive DNAs for photoaffinity modification of cell proteins. UV irradiation results in modification of DNA binding proteins of cell extract. The level of photoaffinity labeling of protein targets in the cell extract was strongly dependent on the efficiency of synthesis of photoreactive DNA.

Structural relationship of lithocholic acid derivatives binding to the N-terminal 8-kDa domain of DNA polymerase beta.

We reported previously that lithocholic acid (LCA, 3-alpha-hydroxy-5-beta-cholan-24-oic acid), one of the major compounds in the secondary bile acids, selectively inhibited the activity of mammalian DNA polymerase beta (pol beta) [Mizushina, Y., Ohkubo, T., Sugawara, F., and Sakaguchi, K. (2000) Biochemistry 39, 12606-12613]. The purpose of this study was to investigate the molecular structural relationship of LCA and its 10 chemically synthesized derivatives. The inhibitory activities of pol beta by some derivative compounds were stronger than that by LCA, and these compounds bound tightly to the 8-kDa domain fragment but not to the 31-kDa domain fragment of pol beta. Biacore analysis demonstrated that the 8-kDa domain bound selectively to compound 9 (3-alpha-O-lauroyl-5-beta-cholan-24-oic acid), which was the strongest pol beta inhibitor tested, as a 1:1 complex with a dissociation constant (K(d)) of 1.73 nM. From computer modeling analysis (i.e., molecular dynamics analysis), the 8-kDa domain had two inhibitor binding areas. Three amino acid residues (Lys60, Leu77, and Thr79) of the 8-kDa domain bound to LCA and compound 2 (3-alpha-methoxy-5-beta-cholan-24-oic acid), and four amino acid residues (Leu11, Lys35, His51, and Thr79) of the 8-kDa domain bound to compound 9. From these results, the structure-function relationship among pol beta and its selective inhibitors was discussed.

Three-dimensional structural model analysis of the binding site of an inhibitor, nervonic acid, of both DNA polymerase beta and HIV-1 reverse transcriptase.

Previously, we reported the three-dimensional molecular interactions of nervonic acid (NA) with mammalian DNA polymerase beta (pol beta) [Mizushina et al. (1998) J. Biol. Chem. 274, 25599-25607]. By three-dimensional structural model analysis and comparison with the spatial positioning of specific amino acids binding to NA on pol beta (Leu11, Lys35, His51, and Thr79), we obtained supplementary information that allowed us to build a structural model of human immunodeficiency virus type-1 reverse transcriptase (HIV-1 RT). In HIV-1 RT, Leu100, Lys65, His235, and Thr386 corresponded to these four amino acid residues. These results suggested that the NA binding domains of pol beta and HIV-1 RT are three-dimensionally very similar. The effects of NA on HIV-1 RT are thought to be same as those on pol beta in binding to the rhombus of the four amino acid residues. NA dose-dependently inhibited the HIV-1 RT activity. For binding to pol beta, the kinetics were competitive when the rhombus was present on the DNA binding site. However, as the rhombus in HIV-1 RT was not present in the DNA binding site, the three-dimensional structure of the DNA binding site must be distorted, and subsequently the enzyme is inhibited non-competitively.

A Plant Phytotoxin, Solanapyrone A, Is an Inhibitor of DNA Polymerase β and λ

Solanapyrone A, a phytotoxin and enzyme inhibitor isolated from a fungus (SUT 01B1-2) selectively inhibits the activities of mammalian DNA polymerase beta and lambda (pol beta and lambda) in vitro. The IC50 values of the compound were 30 microm for pol beta and 37 microm for pol lambda. Because pol beta and lambda are in a family and their three-dimensional structures are thought to be highly similar to each other, we used pol beta to analyze the biochemical relationship with solanapyrone A. On pol beta, solanapyrone A antagonistically competed with both the DNA template and the nucleotide substrate. BIAcore analysis demonstrated that solanapyrone A bound selectively to the N-terminal 8-kDa domain of pol beta. This domain is known to bind single-stranded DNA, provide 5'-phosphate recognition of gapped DNA, and cleave the sugar-phosphate bond 3' to an intact apurinic/apyrimidinic (AP) site (i.e. AP lyase activity) including 5'-deoxyribose phosphate lyase activity. Solanapyrone A inhibited the single-stranded DNA-binding activity but did not influence the activities of the 5'-phosphate recognition in gapped DNA structures and the AP lyase. Based on these results, the inhibitory mechanism of solanapyrone A is discussed.

Three-dimensional structural model analysis of the binding site of lithocholic acid, an inhibitor of DNA polymerase beta and DNA topoisomerase II.

The molecular action of lithocholic acid (LCA), a selective inhibitor of mammalian DNA polymerase beta (pol beta), was investigated. We found that LCA could also strongly inhibit the activity of human DNA topoisomerase II (topo II). No other DNA metabolic enzymes tested were affected by LCA. Therefore, LCA should be classified as an inhibitor of both pol beta and topo II. Here, we report the molecular interaction of LCA with pol beta and topo II. By three-dimensional structural model analysis and by comparison with the spatial positioning of specific amino acids binding to LCA on pol beta (Lys60, Leu77, and Thr79), we obtained supplementary information that allowed us to build a structural model of topo II. Modeling analysis revealed that the LCA-interaction interface in both enzymes has a pocket comprised of three amino acids in common, which binds to the LCA molecule. In topo II, the three amino acid residues were Lys720, Leu760, and Thr791. These results suggested that the LCA binding domains of pol beta and topo II are three-dimensionally very similar.

Reagents for modification of protein-nucleic complexes. III. Site-specific photomodification of elongation complex of DNA polymerase beta with arylazide derivatives of primers sensitized with fluorescent ATP gamma-amide

ATP gamma-amides containing in gamma-N-position 1-methylpyrene, 9-methylanthracene, 10-chloro-9-methylanthracene, and 3-methylperylene residues were synthesized and characterized. These compounds were used as sensitizers of site-specific photomodification of the reconstituted elongating complex of the mammalian DNA polymerase beta. The photomodification was carried out with the use of photoaffine reagents, which were synthesized in situ by the 5'-(32)P-labeled primers extension with photoactive analogues of dCTP containing in the exo-N-position of cytosine various perfluoroarylazide groups. The effect of structures of the sensitizers and photoactive reagents on the efficiency and selectivity of photolinking of primers to the enzyme and template, as well as formation of a number of other photomodification products was studied. It was shown that the sensitizers containing 10-chloro-9-methylanthracene and 3-methylperylene residues allow preparation of photolinks in such irradiation conditions when photomodification in their absence is not essentially observed.

Identification of the active site of phosphoribosyl-dephospho-coenzyme A transferase and relationship of the enzyme to an ancient class of nucleotidyltransferases.

Malonate decarboxylase from Klebsiella pneumoniae contains an acyl carrier protein (MdcC) to which a 2'-(5' '-phosphoribosyl)-3'-dephospho-CoA prosthetic group is attached via phosphodiester linkage to serine 25. We have shown in the preceding paper in this issue that the formation of this phosphodiester bond is catalyzed by a phosphoribosyl-dephospho-coenzyme A transferase MdcG with the substrate 2'-(5' '-triphosphoribosyl)-3'-dephospho-CoA that is synthesized from ATP and dephospho-coenzyme A by the triphosphoribosyl transferase MdcB. The reaction catalyzed by MdcG is related to nucleotidyltransfer reactions, and the enzyme indeed catalyzes unphysiological nucleotidyltransfer, e.g., adenylyltransfer from ATP to apo acyl carrier protein (ACP). These unspecific side reactions are favored at high Mg(2+) concentrations. A sequence motif including D134 and D136 of MdcG is a signature of all nucleotidyltransferases. It is known from the well-characterized mammalian DNA polymerase beta that this motif is at the active site of the enzyme. Site-directed mutagenesis of D134 and/or D136 of MdcG to alanine abolished the transfer of the prosthetic group to apo ACP, but the binding of triphosphoribosyl-dephospho-CoA to MdcG was not affected. Evidence is presented that similar to MdcG, MadK encoded by the malonate decarboxylase operon of Malonomonas rubra and CitX from the operon encoding citrate lyase in Escherichia coli are phosphoribosyl-dephospho-CoA transferases catalyzing the attachment of the phosphoribosyl-dephospho-CoA prosthetic group to their specific apo ACPs.

Hydrophobic interactions in the hinge domain of DNA polymerase beta are important but not sufficient for maintaining fidelity of DNA synthesis.

We previously described a general mutator form of mammalian DNA polymerase beta containing a cysteine substitution for tyrosine 265. Residue 265 localizes to a hydrophobic hinge region predicted to mediate a polymerase conformational change that may aid in nucleotide selectivity. In this study we tested the hypothesis that van der Waals and hydrophobic contacts between Y265 and neighboring residues are important for DNA synthesis fidelity and catalysis, by altering interactions in the hinge domain via substitution at position 265. Consistent with the importance of hydrophobic interactions, we found that phenylalanine, leucine, and tryptophan substitutions did not alter significantly the steady-state catalytic efficiency of DNA synthesis, relative to wild type, while the polar serine substitution decreased catalytic efficiency 6-fold. However, we found that all substitutions other than phenylalanine increased the error frequency, relative to wild type, in the order serine > tryptophan = leucine. Therefore, maintenance of the hydrophobicity of residue 265 was not sufficient for maintaining fidelity of DNA synthesis. We conclude that while hydrophobic interactions in the hinge domain are important for fidelity, additional factors such as electrostatic and van der Waals interactions contributed by the tyrosine 265 aromatic ring are required to retain wild-type fidelity.

Translesion replication by DNA polymerase delta depends on processivity accessory proteins and differs in specificity from DNA polymerase beta.

Mutations caused by DNA damage lead to the development of cancer. The critical step in the formation of these mutations is the replication of unrepaired lesions in DNA by DNA polymerases, a process termed translesion replication. Using a newly developed method for preparation of gapped plasmids, containing a site-specific synthetic abasic site, we analyzed translesion replication with purified mammalian DNA polymerases delta and beta. DNA polymerase delta was found to be unable to replicate through the abasic site. Addition of the sliding DNA clamp PCNA, the clamp loader RFC, and ATP caused a drastic 30-fold increase in translesion replication. Thus, similar to Escherichia coli DNA polymerase III, the processivity accessory proteins enable DNA polymerase delta to bypass blocking lesions. Under comparable conditions, DNA polymerase beta was unable to bypass the abasic site, unless its concentration was greatly increased. Analysis of translesion replication products revealed a marked difference in the specificity of bypass: whereas 90% of bypass events by DNA polymerase delta holoenzyme involved insertion of a dAMP residue opposite the abasic site, DNA polymerase beta tended to skip over the abasic site, producing mainly minus frameshifts (73%). The significance of these results for in vivo translesion replication is discussed.

Mode Analysis of a Fatty Acid Molecule Binding to the N-terminal 8-kDa Domain of DNA Polymerase β: A 1:1 COMPLEX AND BINDING SURFACE

We reported previously that long-chain fatty acids are potent inhibitors of mammalian DNA polymerase beta. At present, based on information available from the NMR structure of the N-terminal 8-kDa domain, we examined the structural interaction with the 8-kDa domain using two species, C(18)-linoleic acid (LA) or C(24)-nervonic acid (NA). In the 8-kDa domain with LA or NA, the structure that forms the interaction interface included helix-1, helix-2, helix-4, the three turns (residues 1-13, 48-51, and 79-87) and residues adjacent to an Omega-type loop connecting helix-1 and helix-2 of the same face. No significant shifts were observed for any of the residues on the opposite side of the 8-kDa domain. The NA interaction interface on the amino acid residues of the 8-kDa domain fragment was mostly the same as that of LA, except that the shifted cross-peaks of Leu-11 and Thr-79 were significantly changed between LA and NA. The 8-kDa domain bound to LA or NA as a 1:1 complex with a dissociation constant (K(D)) of 1.02 or 2.64 mM, respectively.

A human DNA editing enzyme homologous to the Escherichia coli DnaQ/MutD protein.

Mammalian DNA polymerases alpha and beta lack 3' exonuclease activity and are unable to edit errors after DNA synthesis. However, editing exonucleases can be functions of separate polypeptides. We isolated a widely distributed DNA-specific 3' exonuclease from rabbit liver nuclei, sequenced tryptic peptides by mass spectrometry, and identified the corresponding human open reading frame. The protein expressed from the cloned human sequence exhibits 3' exonuclease activity. The human clone shares sequence homology with the editing function of the Escherichia coli DNA polymerase III holoenzyme, i.e., the DnaQ/MutD protein, and weakly with the editing 3' exonuclease domain of eukaryotic DNA polymerase epsilon. The gene maps to human chromosome 3p21.2-21.3. In a reconstituted human DNA repair system containing DNA polymerase beta and DNA ligase III-XRCC1, accurate rejoining of a 3' mismatched base residue at a single-strand break is dependent on addition of the exonuclease.

Translesional synthesis on DNA templates containing an estrogen quinone-derived adduct: N2-(2-hydroxyestron-6-yl)-2'-deoxyguanosine and N6-(2-hydroxyestron-6-yl)-2'-deoxyadenosine.

Miscoding properties induced by estrogen quinone-derived DNA adducts were analyzed using an in vitro experimental system to quantify base substitutions and deletions. Site-specifically modified oligodeoxynucleotides containing a single N2-(2-hydroxyestron-6-yl)-2'-deoxyguanosine (2-OHE1-N2-dG) or N6-(2-hydroxyestron-6-yl)-2'-deoxyadenosine (2-OHE1-N6-dA) were prepared postsynthetically and used as templates in primer extension reactions catalyzed by mammalian DNA polymerases (pol) alpha, beta, and delta. The 2-OHE1-N2-dG adduct blocked primer extension reactions more strongly than 2-OHE1-N6-dA. Using pol alpha and delta, 2-OHE1-N2-dG promoted incorporation of dCMP (6.3 and 3.1%, respectively), the correct base, opposite the lesion: when pol delta was used, misincorporation of dTMP (0.52%) was detected. 2-OHE1-N6-dA also promoted incorporation of dTMP, the correct base, opposite the lesion, accompanied by misincorporation of dCTP (0.54% for pol alpha and 3.2% for pol delta) and one-base deletion (0.3-0.5%). Using pol beta, no miscoding was detected. The miscoding occurred only when replicative DNA polymerases were used. Kinetic data were consistent with those obtained from the analysis of fully extended products formed by pol alpha or pol beta. These results indicate that endogenous estrogen quinone-derived DNA adducts have miscoding potential: G --> A and A --> G transitions and deletions are predicted in mammalian cells.