Repair of a synthetic abasic site involves concerted reactions of DNA synthesis followed by excision and ligation.

A synthetic analog of an abasic site in DNA is efficiently repaired by a short-patch repair mechanism in soluble extracts of Xenopus laevis oocytes (Y. Matsumoto and D. F. Bogenhagen, Mol. Cell. Biol. 9:3750-3757, 1989). We present a detailed analysis of the repair mechanism, using extracts depleted of endogenous nucleotide pools. ATP was required for repair with a sharp optimal concentration of 5 mM. The initial rate of repair was increased by preincubation of the DNA in the extract in the presence of ATP. During this preincubation, the DNA was cleaved on the 5' side of the lesion by a class II apurinic-apyrimidinic endonuclease, but removal of the abasic sugar residue was not observed prior to addition of deoxynucleotides to the reaction. Immediately following DNA synthesis, excision and ligation proceeded in a coordinated manner to complete repair. DNA preincubated in the extract in the absence of deoxynucleotides remained associated with repair enzymes during gel filtration. These observations suggest that the enzymes involved in concerted repair of the abasic site form a complex on DNA.     

Translesion DNA synthesis-assisted non-homologous end-joining of complex double-strand breaks prevents loss of DNA sequences in mammalian cells

Double strand breaks (DSB) are severe DNA lesions, and if not properly repaired, may lead to cell death or cancer. While there is considerable data on the repair of simple DSB (sDSB) by non-homologous end-joining (NHEJ), little is known about the repair of complex DSBs (cDSB), namely breaks with a nearby modification, which precludes ligation without prior processing. To study the mechanism of cDSB repair we developed a plasmid-based shuttle assay for the repair of a defined site-specific cDSB in cultured mammalian cells. Using this assay we found that repair efficiency and accuracy of a cDSB with an abasic site in a 5′ overhang was reduced compared with a sDSB. Translesion DNA synthesis (TLS) across the abasic site located at the break prevented loss of DNA sequences, but was highly mutagenic also at the template base next to the abasic site. Similar to sDSB repair, cDSB repair was totally dependent on XrccIV, and altered in the absence of Ku80. In contrast, Artemis appears to be specifically involved in cDSB repair. These results may indicate that mammalian cells have a damage control strategy, whereby severe deletions are prevented at the expense of the less deleterious point mutations during NHEJ.     

Repair of clustered DNA lesions. Sequence-specific inhibition of long-patch base excision repair be 8-oxoguanine

Ionizing radiation induces clustered DNA damage where two or more lesions are located proximal to each other on the same or opposite DNA strands. It has been suggested that individual lesions within a cluster are removed sequentially and that the presence of a vicinal lesion(s) may affect the rate and fidelity of DNA repair. In this study, we addressed the question of how 8-oxoguanine located opposite to normal or reduced abasic sites would affect the repair of these sites by the base excision repair system. We have found that an 8-oxoguanine located opposite to an abasic site does not affect either the efficiency or fidelity of repair synthesis by DNA polymerase beta. In contrast, an 8-oxoguanine located one nucleotide 3'-downstream of the abasic site significantly reduces both strand displacement synthesis supported by DNA polymerase beta or delta and cleavage by flap endonuclease of the generated flap, thus inhibiting the long-patch base excision repair pathway.     

In vitro replication and repair of DNA containing a C2'-oxidized abasic site

Abasic lesions are unable to form Watson-Crick hydrogen bonds with nucleotides. Nonetheless, polymerase and repair enzymes distinguish between various oxidized abasic lesions, as well as from nonoxidized abasic sites (AP). The C2-AP lesion is produced when DNA is exposed to gamma-radiolysis. Its effects on polymerases and repair enzymes are unknown. A recently reported method for the chemical synthesis of oligonucleotides containing C2-AP at a defined site was utilized for studying the activity of Klenow exo(-) and repair enzymes on templates containing the lesion. The C2-AP lesion has a similar effect on Klenow exo(-) as do AP and C4-AP sites. Deoxyadenosine is preferentially incorporated opposite C2-AP, but extension of the primer past the lesion is strongly blocked. C2-AP is incised less efficiently by exonuclease III and endonuclease IV than are other abasic lesions. Furthermore, although a Schiff base between C2-AP and endonuclease III can be chemically trapped, the location of the 3'-phosphate alpha with respect to the aldehyde prevents beta-elimination associated with the lyase activity of type I base excision repair enzymes. The interactions of the C2'-oxidized abasic site with Klenow exo(-) and repair enzymes suggest that the lesion will be mutagenic and that it will be removed by strand displacement synthesis and flap endonuclease processing via a long patch repair mechanism.     

Translesion Synthesis of Abasic Sites by Yeast DNA Polymerase ?

Studies of replicative DNA polymerases have led to the generalization that abasic sites are strong blocks to DNA replication. Here we show that yeast replicative DNA polymerase ϵ bypasses a model abasic site with comparable efficiency to Pol η and Dpo4, two translesion polymerases. DNA polymerase ϵ also exhibited high bypass efficiency with a natural abasic site on the template. Translesion synthesis primarily resulted in deletions. In cases where only a single nucleotide was inserted, dATP was the preferred nucleotide opposite the natural abasic site. In contrast to translesion polymerases, DNA polymerase ϵ with 3′–5′ proofreading exonuclease activity bypasses only the model abasic site during processive synthesis and cannot reinitiate DNA synthesis. This characteristic may allow other pathways to rescue leading strand synthesis when stalled at an abasic site.     

Involvement of poly(ADP-ribose) polymerase in base excision repair

Poly(ADP-ribose) polymerase (PARP) is a zinc-finger DNA binding protein that detects and signals DNA strand breaks generated directly or indirectly by genotoxic agents. In response to these lesions, the immediate poly(ADP-ribosylation) of nuclear proteins converts DNA interruptions into intracellular signals that activate DNA repair or cell death programs. To elucidate the biological function of PARP in vivo, the mouse PARP gene was inactivated by homologous recombination to generate mice lacking a functional PARP gene. PARP knockout mice and the derived mouse embryonic fibroblasts (MEFs) were acutely sensitive to monofunctional alkylating agents and gamma-irradiation demonstrating that PARP is involved in recovery from DNA damage that triggers the base excision repair (BER) process. To address the issue of the role of PARP in BER, the ability of PARP-deficient mammalian cell extracts to repair a single abasic site present on a circular duplex plasmid molecule was tested in a standard in vitro repair assay. The results clearly demonstrate, for the first time, the involvement of PARP in the DNA synthesis step of the base excision repair process.     

DNA polymerase beta promotes recruitment of DNA ligase III alpha-XRCC1 to sites of base excision repair

Base excision repair is a major pathway for the removal of simple lesions in DNA including base damage and base loss (abasic site). Base excision repair requires the coordinated action of several repair and ancillary proteins, the impairment of which can lead to genetic instability. Using a protein-DNA cross-linking assay during repair in human whole cell extracts, we monitored proteins involved in the initial steps of repair of a substrate containing a site-specific abasic site to address the molecular events following incision of the abasic site by AP endonuclease. We find that after dissociation of AP endonuclease from the incised abasic site, both DNA polymerase beta (Pol beta) and the DNA ligase IIIalpha-XRCC1 heterodimer efficiently bind/cross-link to the substrate DNA. We also find that the cross-linking efficacy of the DNA ligase IIIalpha-XRCC1 heterodimer was decreased about 2-fold in the Pol beta-deficient cell extract but was rescued by addition of purified wild type but not a mutant Pol beta protein that does not interact with the DNA ligase IIIalpha-XRCC1 heterodimer. We further demonstrate that Pol beta and the DNA ligase IIIalpha-XRCC1 heterodimer are present at equimolar concentrations in whole cell extracts and that Pol beta has a 7-fold higher affinity to the incised abasic site containing substrate than DNA ligase IIIalpha. Using gel filtration of whole cell extracts prepared at physiological salt conditions (0.15 M NaCl), we find no evidence for a stable preexisting complex of DNA Pol beta with the DNA ligase IIIalpha-XRCC1 heterodimer. Taken together, these data suggest that following incision by AP endonuclease, DNA Pol beta recognizes and binds to the incised abasic site and promotes recruitment of the DNA ligase IIIalpha-XRCC1 heterodimer through its interaction with XRCC1.     

Interaction of a spin-labeled adenine-acridine conjugate with a DNA duplex containing an abasic site model

The abasic site is a common lesion in DNA that is also formed as an intermediate in the base excision repair of damaged bases. We have previously reported the adenine-acridine conjugate 1 that was designed to bind to the abasic site and interfere with the repair process. High-field NMR had shown that 1 forms specific complexes with a DNA duplex containing an apurinic abasic site model. We report here the dynamics of the interaction of the nitroxide-labeled analogue 3 of the conjugate 1 with the same apurinic oligonucleotide and with the parent unmodified duplex. Identical study of the labeled acridine subunit 5 used as a reference is also reported. In the presence of the apurinic duplex and depending on the concentrations and drug ratios, three species are observed: the radical "free in solution", the "intercalation" complex characterized by its similarity to that observed in the presence of the parent unmodified duplex, and the "abasic-site-specific" complex which is the sole species visible at low drug ratios. The experimental data reinforced by molecular modeling of the complex and theoretical calculation of correlation times suggest (i) the most immobilized form corresponds to that observed by NMR and (ii) complexation of the drug is little or not modified by the spin-label. We also show that the abasic site constitutes a binding site for the propylaminoacridine intercalator 5.     

Improved measurement of dibenzo[a,l]pyrene-induced abasic sites by the aldehyde-reactive probe assay

Dibenzo[a,l]pyrene (DB[a,l]P) induces abundant amounts of depurinating adducts that spontaneously dissociate to form abasic sites in DNA. However, several previous studies that used the aldehyde-reactive probe (ARP) assay, could not verify abasic site formation by DB[a,l]P. Therefore, we examined whether a modification of the ARP assay would allow greater quantification of abasic sites. A previous study indicated that the abasic site quantification is improved by letting ARP trap the nascent abasic sites in cells, before extracting DNA for the assay. To test whether the addition of ARP to the DB[a,l]P-DNA adduct-forming reaction would improve abasic site quantification, we treated calf thymus DNA (0.625 mg/mL) with DB[a,l]P (80 microM) and 3-methylcholanthrene-treated rat liver microsomes with or without ARP (3 mM). The inclusion of ARP in the adduct-forming reaction resulted in significantly greater detection of abasic sites (62 lesions/10(6) bp versus 3.7 lesions/10(6) bp). DB[a,l]P also induces DNA strand breaks. The strand breaks may occur at abasic sites and by other mechanisms, such as oxidative damage. ARP/O-methoxyamine-abasic site conjugates are refractory to strand breakage, however, ARP or O-methoxyamine (3-10 mM) could only partially protect DB[a,l]P-induced DNA degradation, presumably by protecting the abasic sites, but not the other strand breaks. These results suggest that if DNA strand breakages occur at the abasic sites or at bases flanking them, and the fragments are lost during DNA extraction, abasic site estimation could be compromised. To obtain an independent line of evidence for abasic site formation in DB[a,l]P-treated cells, mouse Mbeta16 fibroblasts were treated with DB[a,l]P and O-methoxyamine. O-Methoxyamine is known to potentiate cytotoxicity of abasic site-inducing chemicals by forming abasic site conjugates, which partially inhibits their repair. O-Methoxyamine was found to increase DB[a,l]P cytotoxicity in these cells, supporting the idea that DB[a,l]P formed abasic sites. In summary, the inclusion of ARP in the DB[a,l]P-DNA adduct-forming reaction traps and protects the nascent abasic sites, allowing an improved quantification of abasic sites.     

Long-patch base excision DNA repair of 2-deoxyribonolactone prevents the formation of DNA-protein cross-links with DNA polymerase beta

Oxidized abasic sites are a major form of DNA damage induced by free radical attack and deoxyribose oxidation. 2-Deoxyribonolactone (dL) is a C1'-oxidized abasic site implicated in DNA strand breakage, mutagenesis, and formation of covalent DNA-protein cross-links (DPCs) with repair enzymes such as DNA polymerase beta (polbeta). We show here that mammalian cell-free extracts incubated with Ape1-incised dL substrates under non-repair conditions give rise to DPCs, with a major species dependent on the presence of polbeta. DPC formation was much less under repair than non-repair conditions, with extracts of either polbeta-proficient or -deficient cells. Partial base excision DNA repair (BER) reconstituted with purified enzymes demonstrated that Flap endonuclease 1 (FEN1) efficiently excises a displaced oligonucleotide containing a 5'-terminal dL residue, as would be produced during long-patch (multinucleotide) BER. Simultaneous monitoring of dL repair and dL-mediated DPC formation demonstrated that removal of the dL residue through the combined action of strand-displacement DNA synthesis by polbeta and excision by FEN1 markedly diminished DPC formation with the polymerase. Analysis of the patch size distribution associated with DNA repair synthesis in cell-free extracts showed that the processing of dL residues is associated with the synthesis of >or=2 nucleotides, compared with predominantly single nucleotide replacement for regular abasic sites. Our observations reveal a cellular repair process for dL lesions that avoids formation of DPCs that would threaten the integrity of DNA and perhaps cell viability.     

Inactivation of the 3'-5' exonuclease of the replicative T4 DNA polymerase allows translesion DNA synthesis at an abasic site

Here, we have investigated the consequences of the loss of proof-reading exonuclease function on the ability of the replicative T4 DNA polymerase (gp43) to elongate past a single abasic site located on model DNA substrates. Our results show that wild-type T4 DNA polymerase stopped at the base preceding the lesion on two linear substrates having different sequences, whereas the gp43 D219A exonuclease-deficient mutant was capable of efficient bypass when replicating the same substrates. The structure of the DNA template did not influence the behavior of the exonuclease-proficient or deficient T4 DNA polymerases. In fact, when replicating a damaged "minicircle" DNA substrate constructed by circularizing one of the linear DNA, elongation by wild-type enzyme was still completely blocked by the abasic site, while the D219A mutant was capable of bypass. During DNA replication, the T4 DNA polymerase associates with accessory factors whose combined action increases the polymerase-binding capacity and processivity, and could modulate the behavior of the enzyme towards an abasic site. We thus performed experiments measuring the ability of wild-type and exonuclease-deficient T4 DNA polymerases, in conjunction with these replicative accessory proteins, to perform translesion DNA replication on linear or circular damaged DNA substrates. We found no evidence of either stimulation or inhibition of the bypass activities of the wild-type and exonuclease-deficient forms of T4 DNA polymerase following addition of the accessory factors, indicating that the presence or absence of the proof-reading activity is the major determinant in dictating translesion synthesis of an abasic site by T4 DNA polymerase.     

Caught bending the A-rule: crystal structures of translesion DNA synthesis with a non-natural nucleotide

Damage to DNA involving excision of the nucleobase at the N-glycosidic bond forms abasic sites. If a nucleotide becomes incorporated opposite an unrepaired abasic site during DNA synthesis, most B family polymerases obey the A-rule and preferentially incorporate dAMP without instruction from the template. In addition to being potentially mutagenic, abasic sites provide strong blocks to DNA synthesis. A previous crystal structure of an exonuclease deficient variant of the replicative B family DNA polymerase from bacteriophage RB69 (RB69 gp43 exo-) illustrated these properties, showing that the polymerase failed to translocate the DNA following insertion of dAMP opposite an abasic site. We examine four new structures depicting several steps of translesion DNA synthesis by RB69 gp43 exo-, employing a non-natural purine triphosphate analogue, 5-nitro-1-indolyl-2'-deoxyriboside-5'-triphosphate (5-NITP), that is incorporated more efficiently than dAMP opposite abasic sites. Our structures indicate that a dipole-induced dipole stacking interaction between the 5-nitro group and base 3' to the templating lesion explains the enhanced kinetics of 5-NITP. As with dAMP, the DNA fails to translocate following insertion of 5-NIMP, although distortions at the nascent primer terminus contribute less than previously thought in inducing the stall, given that 5-NIMP preserves relatively undistorted geometry at the insertion site following phosphoryl transfer. An open ternary configuration, novel in B family polymerases, reveals an initial template independent binding of 5-NITP adjacent to the active site of the open polymerase, suggesting that closure of the fingers domain shuttles the nucleotide to the active site while testing the substrate against the template.     

DNA polymerase beta is required for efficient DNA strand break repair induced by methyl methanesulfonate but not by hydrogen peroxide

The most frequent DNA lesions in mammalian genomes are removed by the base excision repair (BER) via multiple pathways that involve the replacement of one or more nucleotides at the lesion site. The biological consequences of a BER defect are at present largely unknown. We report here that mouse cells defective in the main BER DNA polymerase β (Pol β) display a decreased rate of DNA single-strand breaks (ssb) rejoining after methyl methanesulfonate damage when compared with wild-type cells. In contrast, Pol β seems to be dispensable for hydrogen peroxide-induced DNA ssb repair, which is equally efficient in normal and defective cells. By using an in vitro repair assay on single abasic site-containing circular duplex molecules, we show that the long-patch BER is the predominant repair route in Pol β-null cell extract. Our results strongly suggest that the Pol β-mediated single nucleotide BER is the favorite pathway for repair of N-methylpurines while oxidation-induced ssb, likely arising from oxidized abasic sites, are the substrate for long-patch BER.     

Repair of abasic sites in DNA

Repair of both normal and reduced AP sites is activated by AP endonuclease, which recognizes and cleaves a phosphodiester bond 5' to the AP site. For a short period of time an incised AP site is occupied by poly(ADP-ribose) polymerase and then DNA polymerase beta adds one nucleotide into the repair gap and simultaneously removes the 5'-sugar phosphate. Finally, the DNA ligase III/XRCC1 complex accomplishes repair by sealing disrupted DNA ends. However, long-patch BER pathway, which is involved in the removal of reduced abasic sites, requires further DNA synthesis resulting in strand displacement and the generation of a damage-containing flap that is later removed by the flap endonuclease. Strand-displacement DNA synthesis is accomplished by DNA polymerase delta/epsilon and DNA ligase I restores DNA integrity. DNA synthesis by DNA polymerase delta/epsilon is dependent on proliferating cell nuclear antigen, which also stimulates the DNA ligase I and flap endonuclease. These repair events are supported by multiple protein-protein interactions.     

Base excision repair is impaired in mammalian cells lacking Poly(ADP-ribose) polymerase-1

In mammalian cells, damaged bases in DNA are corrected by the base excision repair pathway which is divided into two distinct pathways depending on the length of the resynthesized patch, replacement of one nucleotide for short-patch repair, and resynthesis of several nucleotides for long-patch repair. The involvement of poly(ADP-ribose) polymerase-1 (PARP-1) in both pathways has been investigated by using PARP-1-deficient cell extracts to repair single abasic sites derived from uracil or 8-oxoguanine located in a double-stranded circular plasmid. For both lesions, PARP-1-deficient cell extracts were about half as efficient as wild-type cells at the polymerization step of the short-patch repair synthesis, but were highly inefficient at the long-patch repair. We provided evidence that PARP-1 constitutively interacts with DNA polymerase beta. Using cell-free extracts from mouse embryonic cells deficient in DNA polymerase beta, we demonstrated that DNA polymerase beta is involved in the repair of uracil-derived AP sites via both the short and the long-patch repair pathways. When both PARP-1 and DNA polymerase beta were absent, the two repair pathways were dramatically affected, indicating that base excision repair was highly inefficient. These results show that PARP-1 is an active player in DNA base excision repair.     

The choice of nucleotide inserted opposite abasic sites formed within chromosomal DNA reveals the polymerase activities participating in translesion DNA synthesis

Abasic sites in genomic DNA can be a significant source of mutagenesis in biological systems, including human cancers. Such mutagenesis requires translesion DNA synthesis (TLS) bypass of the abasic site by specialized DNA polymerases. The abasic site bypass specificity of TLS proteins had been studied by multiple means in vivo and in vitro, although the generality of the conclusions reached have been uncertain. Here, we introduce a set of yeast reporter strains for investigating the in vivo specificity of abasic site bypass at numerous random positions within chromosomal DNA. When shifted to 37 °C, these strains underwent telomere uncapping and resection that exposed reporter genes within a long 3′ ssDNA overhang. Human APOBEC3G cytosine deaminase was expressed to create uracils in ssDNA, which were excised by uracil-DNA N-glycosylase. During repair synthesis, error-prone TLS bypassed the resulting abasic sites. Because of APOBEC3G's strict motif specificity and the restriction of abasic site formation to only one DNA strand, this system provides complete information about the location of abasic sites that led to mutations. We recapitulated previous findings on the roles of REV1 and REV3. Further, we found that sequence context can strongly influence the relative frequency of A or C insertion. We also found that deletion of Pol32, a non-essential common subunit of Pols δ and ζ, resulted in residual low-frequency C insertion dependent on Rev1 catalysis. We summarize our results in a detailed model of the interplay between TLS components leading to error-prone bypass of abasic sites. Our results underscore the utility of this system for studying TLS bypass of many types of lesions within genomic DNA.     

Aphidicolin-resistant and -sensitive base excision repair in wild-type and DNA polymerase beta-defective mouse cells

Several DNA polymerases (Pols) can add complementary bases at the gap created during the base excision repair (BER). To characterize the BER resynthesis step, the repair of a single abasic site by wild-type and Pol beta-defective mouse cell extracts was analysed in the presence of aphidicolin, a specific inhibitor of replicative Pols. We show that there is a competition between distributive and processive Pols for the nucleotide addition at the primer terminus. In wild-type cell extracts, the initial nucleotide insertion involves mainly Pol beta but the elongation step is carried out by a replicative Pol. Conversely, in Pol beta-null cell extracts the synthesis step is carried out by a replicative Pol without any switching to an auxiliary polymerase. We present evidence that short-patch repair synthesis occurs even in the absence of both Pol beta and replicative Pols. Exogeneously added purified human Pol lambda was unable to stimulate this back-up synthesis.     

Evaluation of the role of the vaccinia virus uracil DNA glycosylase and A20 proteins as intrinsic components of the DNA polymerase holoenzyme

The vaccinia virus DNA polymerase is inherently distributive but acquires processivity by associating with a heterodimeric processivity factor comprised of the viral A20 and D4 proteins. D4 is also an enzymatically active uracil DNA glycosylase (UDG). The presence of an active repair protein as an essential component of the polymerase holoenzyme is a unique feature of the replication machinery. We have shown previously that the A20-UDG complex has a stoichiometry of ～1:1, and our data suggest that A20 serves as a bridge between polymerase and UDG. Here we show that conserved hydrophobic residues in the N' terminus of A20 are important for its binding to UDG. Our data argue against the assembly of D4 into higher order multimers, suggesting that the processivity factor does not form a toroidal ring around the DNA. Instead, we hypothesize that the intrinsic, processive DNA scanning activity of UDG tethers the holoenzyme to the DNA template. The inclusion of UDG as an essential holoenzyme component suggests that replication and base excision repair may be coupled. Here we show that the DNA polymerase can utilize dUTP as a substrate in vitro. Moreover, uracil moieties incorporated into the nascent strand during holoenzyme-mediated DNA synthesis can be excised by the viral UDG present within this holoenzyme, leaving abasic sites. Finally, we show that the polymerase stalls upon encountering an abasic site in the template strand, indicating that, like many replicative polymerases, the poxviral holoenzyme cannot perform translesion synthesis across an abasic site.     

Human apurinic/apyrimidinic endonuclease is processive

Apurinic/apyrimidinic endonuclease (AP endo) is believed to play a critical role in repair of oxidative damage of DNA and is proposed to initiate repair of most abasic sites in the base excision repair pathway. AP endo makes a single nick 5' to an abasic site in double-stranded DNA. In this study, we investigated whether AP endo locates an abasic site through a processive or a distributive mechanism. We used a linear multi-abasic site substrate (concatemer), synthesized by ligating together identical 25-nucleotide monomeric units (25-mers). We first determined that the 25-mer monomer from which the concatemers were prepared was nicked by AP endo in a fashion similar to that of the previously published 49-mer substrate with a different sequence. Steady state parameters K(m) and k(cat) and single-turnover parameters for substrate binding were comparable to previously published values. Using the multi-abasic site concatemer, we demonstrated that AP endo was capable of cleaving approximately seven to eight abasic sites, traveling at least 200 nucleotides, before dissociating from its substrate. Thus, AP endo, like uracil DNA glycosylase, behaves in a quasi processive fashion. Processivity could be separated from catalysis, since processivity was maximal at 25 mM NaCl, while the rate of cleavage was maximal at 125 mM salt. In short, nicking activity was maximized close to physiological salt molarities while processivity was midrange at physiological salt concentrations. The latter is likely to be subject to tight regulation by small changes in ionic strength.     

Modulation of the 3'-->5'-exonuclease activity of human apurinic endonuclease (Ape1) by its 5'-incised Abasic DNA product

The major abasic endonuclease of human cells, Ape1 protein, is a multifunctional enzyme with critical roles in base excision repair (BER) of DNA. In addition to its primary activity as an apurinic/apyrimidinic endonuclease in BER, Ape1 also possesses 3'-phosphodiesterase, 3'-phosphatase, and 3'-->5'-exonuclease functions specific for the 3' termini of internal nicks and gaps in DNA. The exonuclease activity is enhanced at 3' mismatches, which suggests a possible role in BER for Ape1 as a proofreading activity for the relatively inaccurate DNA polymerase beta. To elucidate this role more precisely, we investigated the ability of Ape1 to degrade DNA substrates that mimic BER intermediates. We found that the Ape1 exonuclease is active at both mismatched and correctly matched 3' termini, with preference for mismatches. In our hands, the exonuclease activity of Ape1 was more active at one-nucleotide gaps than at nicks in DNA, even though the latter should represent the product of repair synthesis by polymerase beta. However, the exonuclease activity was inhibited by the presence of nearby 5'-incised abasic residues, which result from the apurinic/apyrimidinic endonuclease activity of Ape1. The same was true for the recently described exonuclease activity of Escherichia coli endonuclease IV. Exonuclease III, the E. coli homolog of Ape1, did not discriminate among the different substrates. Removal of the 5' abasic residue by polymerase beta alleviated the inhibition of the Ape1 exonuclease activity. These results suggest roles for the Ape1 exonuclease during BER after both DNA repair synthesis and excision of the abasic deoxyribose-5-phosphate by polymerase beta.