Title: Modulation of the 3′→5′-Exonuclease Activity of Human Apurinic Endonuclease (Ape1) by Its 5′-incised Abasic DNA Product
Abstract: 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 β. 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 β. 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 β 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 β. 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 β. 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 β. 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 β 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 β. The formation of apurinic/apyrimidinic (AP) 1The abbreviations used are: AP, apurinic/apyrimidinic; Ape1, AP endonuclease 1; pol β, DNA polymerase β; dRP, deoxyribose-5-phosphate; BER, base excision repair; F, tetrahydrofuran.1The abbreviations used are: AP, apurinic/apyrimidinic; Ape1, AP endonuclease 1; pol β, DNA polymerase β; dRP, deoxyribose-5-phosphate; BER, base excision repair; F, tetrahydrofuran. sites in DNA is the most common consequence of exposure of cells to DNA-damaging agents of both endogenous and environmental origin (1Demple B. DeMott M.S. Oncogene. 2002; 21: 8926-8934Crossref PubMed Scopus (86) Google Scholar). AP sites are formed as repair intermediates by DNA glycosylases, which remove certain mismatched bases or base lesions formed by reactive oxygen species, alkylating agents, or other environmental insults. AP sites can also form spontaneously via acid-catalyzed hydrolysis of the N-glycosylic bonds linking the bases to the sugar-phosphate backbone of DNA. Such spontaneous depurination forms an estimated 10,000 AP sites per day in each human cell (2Lindahl T. Nature. 1993; 362: 709-715Crossref PubMed Scopus (4325) Google Scholar), and the activity of DNA glycosylases would certainly add to this burden. Indeed, the steady-state level of AP lesions is estimated in some studies to be much higher, approaching 50,000 or more per cell depending on its age and tissue source (3Atamna H. Cheung I. Ames B.N. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 686-691Crossref PubMed Scopus (262) Google Scholar, 4Nakamura J. Swenberg J.A. Cancer Res. 1999; 59: 2522-2526PubMed Google Scholar). AP sites in mammalian cells are repaired by the DNA base excision repair (BER) pathway. The major human AP endonuclease, Ape1 (also called Apex, HAP1, or Ref-1), initiates BER by hydrolyzing the 5′-phosphodiester bond of the AP site to create a DNA repair intermediate that has a single strand break bracketed by 3′-hydroxyl and 5′-deoxyribose-5-phosphate (dRP) termini. Ape1 interacts with DNA polymerase (pol) β during BER to recruit the polymerase to the incised AP site (5Bennett R.A. Wilson III, D.M. Wong D. Demple B. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 7166-7169Crossref PubMed Scopus (326) Google Scholar), where the polymerase catalyzes both nucleotide insertion and dRP excision. The sequence of these latter two activities is still unclear, but all the combined enzymatic activities of Ape1 and pol β acting at an AP site would yield a nick that can be sealed by DNA ligase to complete the repair (6Kelley M.R. Kow Y.W. Wilson III, D.M. Cancer Res. 2003; 63: 549-554PubMed Google Scholar, 7Gros L. Saparbaev M.K. Laval J. Oncogene. 2002; 21: 8905-8925Crossref PubMed Scopus (165) Google Scholar). Ape1 is a multifunctional protein with proposed diverse roles in the cell. In addition to its DNA repair activities, Ape1 can activate transcription factors via a redox mechanism (8Evans A.R. Limp-Foster M. Kelley M.R. Mutat. Res. 2000; 461: 83-108Crossref PubMed Scopus (496) Google Scholar), form a transcriptional repressor complex for the negative calcium responsive element (9Okazaki T. Chung U. Nishishita T. Ebisu S. Usuda S. Mishiro S. Xanthoudakis S. Igarashi T. Ogata E. J. Biol. Chem. 1994; 269: 27855-27862Abstract Full Text PDF PubMed Google Scholar, 10Chung U. Igarashi T. Nishishita T. Iwanari H. Iwamatsu A. Suwa A. Mimori T. Hata K. Ebisu S. Ogata E. Fujita T. Okazaki T. J. Biol. Chem. 1996; 271: 8593-8598Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar, 11Kuninger D.T. Izumi T. Papaconstantinou J. Mitra S. Nucleic Acids Res. 2002; 30: 823-829Crossref PubMed Scopus (74) Google Scholar), and act as an important target for the granzyme A-mediated cell death pathway (12Fan Z. Beresford P.J. Zhang D. Xu Z. Novina C.D. Yoshida A. Pommier Y. Lieberman J. Nat. Immunol. 2003; 4: 145-153Crossref PubMed Scopus (214) Google Scholar). In addition to its major role as an AP endonuclease during BER, Ape1 also possesses a weak 3′-phosphatase activity and a 3′-phosphodiesterase activity against abasic residues or fragments (13Demple B. Harrison L. Annu. Rev. Biochem. 1994; 63: 915-948Crossref PubMed Scopus (1295) Google Scholar). These activities are required for the removal of 3′-blocking groups created by ionizing radiation, oxygen free radicals, radiomimetic anti-tumor drugs, and the 3′-AP lyase activities of bifunctional DNA glycosylases (1Demple B. DeMott M.S. Oncogene. 2002; 21: 8926-8934Crossref PubMed Scopus (86) Google Scholar, 14Mitra S. Izumi T. Boldogh I. Bhakat K.K. Hill J.W. Hazra T.K. Free Radic. Biol. Med. 2002; 33: 15-28Crossref PubMed Scopus (129) Google Scholar). Ape1 also has a 3′→5′-exonuclease that excises undamaged DNA nucleotides (see below). Knocking out both alleles coding for Ape1 in mice (the APEX gene) results in early embryonic lethality (15Xanthoudakis S. Smeyne R.J. Wallace J.D. Curran T. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 8919-8923Crossref PubMed Scopus (436) Google Scholar, 16Ludwig D.L. MacInnes M.A. Takiguchi Y. Purtymun P.E. Henrie M. Flannery M. Meneses J. Pedersen R.A. Chen D.J. Mutat. Res. 1998; 409: 17-29Crossref PubMed Scopus (165) Google Scholar, 17Meira L.B. Devaraj S. Kisby G.E. Burns D.K. Daniel R.L. Hammer R.E. Grundy S. Jialal I. Friedberg E.C. Cancer Res. 2001; 61: 5552-5557PubMed Google Scholar), which point to critical roles for Ape1. However, it has not yet been established which Ape1 activities are required for survival. Although Ape1 protein levels have been dramatically reduced for short periods in cells in culture using small interfering RNA (12Fan Z. Beresford P.J. Zhang D. Xu Z. Novina C.D. Yoshida A. Pommier Y. Lieberman J. Nat. Immunol. 2003; 4: 145-153Crossref PubMed Scopus (214) Google Scholar), stable cell lines lacking the protein have not been reported. The modest 3′→5′-exonuclease activity of the mouse and human Ape1 proteins has been known for some time (18Seki S. Ikeda S. Watanabe S. Hatsushika M. Tsutsui K. Akiyama K. Zhang B. Biochim. Biophys. Acta. 1991; 1079: 57-64Crossref PubMed Scopus (75) Google Scholar, 19Wilson 3rd, D.M. Takeshita M. Grollman A.P. Demple B. J. Biol. Chem. 1995; 270: 16002-16007Abstract Full Text Full Text PDF PubMed Scopus (244) Google Scholar), but the biological role of the exonuclease remained obscure. In contrast, the robust AP endonuclease activity of Ape1 is related to the demonstrated roles in BER of exonuclease III in Escherichia coli or Apn1 in yeast (1Demple B. DeMott M.S. Oncogene. 2002; 21: 8926-8934Crossref PubMed Scopus (86) Google Scholar, 13Demple B. Harrison L. Annu. Rev. Biochem. 1994; 63: 915-948Crossref PubMed Scopus (1295) Google Scholar). Recently, it was shown that Ape1 acting as an exonuclease can remove therapeutic anti-tumor nucleoside analogs incorporated at the 3′ ends of DNA oligonucleotides, which suggested that Ape1 contributes to cellular resistance to these drugs (20Chou K.M. Kukhanova M. Cheng Y.C. J. Biol. Chem. 2000; 275: 31009-31015Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar). This work was further extended to show that the exonuclease activity of Ape1 is enhanced on DNA mispairs at the 3′ termini of nicks and gaps (21Chou K.M. Cheng Y.C. Nature. 2002; 415: 655-659Crossref PubMed Scopus (199) Google Scholar). Because the primary mammalian BER polymerase, pol β, has relatively low fidelity and lacks an associated proofreading exonuclease (22Chagovetz A.M. Sweasy J.B. Preston B.D. J. Biol. Chem. 1997; 272: 27501-27504Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 23Osheroff W.P. Jung H.K. Beard W.A. Wilson S.H. Kunkel T.A. J. Biol. Chem. 1999; 274: 3642-3650Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar), the Ape1 exonuclease could provide the "missing" proofreading activity during BER (24Jiricny J. Nature. 2002; 415: 593-594Crossref PubMed Scopus (21) Google Scholar). To explore a possible role for the exonuclease activity of Ape1 during BER, we investigated the ability of the enzyme to excise nucleotides at the 3′ termini of different BER intermediates. Reagents and Enzyme—Urea was purchased from American Bioanalytical Corp. (Natick, MA). An acrylamide-bisacrylamide solution (40%, 29:1 ratio) was purchased from Bio-Rad, Inc. (Hercules, CA). E. coli exonuclease III, uracil DNA glycosylase, and T4 polynucleotide kinase were obtained from New England Biolabs (Beverly, MA). One unit of exonuclease III is defined by the manufacturer as the amount of enzyme required to release 1 nmol of nucleotides in 30 min. E. coli endonuclease IV was purified as previously described (25Levin J.D. Johnson A.W. Demple B. J. Biol. Chem. 1988; 263: 8066-8071Abstract Full Text PDF PubMed Google Scholar). Recombinant wild-type human Ape1 and the D283A/D308A mutant form were purified as previously described (26Masuda Y. Bennett R.A. Demple B. J. Biol. Chem. 1998; 273: 30352-30359Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar). The R177A mutant form of Ape1 was kindly provided by Drs. Tadahide Izumi and Sankar Mitra of the University of Texas Medical Branch (Galveston, TX). Recombinant wild-type human pol β was a generous gift from Drs. Rajendra Prasad and Sam Wilson, NIEHS, National Institutes of Health (Research Triangle Park, NC). DNA Substrates—Oligonucleotides were synthesized and high-performance liquid chromatography-purified by Operon Technologies, Inc. (Alameda, CA) or Midland Certified Reagent Co., Inc. (Midland, TX). The sequences of DNA oligonucleotides used in this study are shown in Table I. Upstream strands were 5′-end-labeled by T4 polynucleotide kinase using a molar excess of [γ-32P]ATP (3000 Ci/mmol, PerkinElmer Life Sciences, Boston, MA). Double-stranded DNA substrates were made by heating the radioactively labeled oligonucleotides with the appropriate template and downstream oligonucleotides to 90 °C and then annealing by slow cooling to room temperature. Unincorporated [γ-32P]ATP was removed using Micro Biospin P-30 columns (Bio-Rad) following the manufacturer's protocols. DNA substrate purity was monitored by autoradiography following electrophoresis on 8% non-denaturing gels and 12% denaturing polyacrylamide gels containing 7 m urea. The different substrates used in this study are listed in Table II along with the structural components of each substrate. Substrates S2, S4, S6, and S8 were incised with a catalytic amount of Ape1 just prior to use in exonuclease assays. Substrates S22 and S23 were sequentially treated with catalytic amounts of uracil DNA glycosylase and Ape1 to create nicked and gapped DNA with 5′-dRP termini. These substrates were also used immediately following pretreatment due to the labile nature of the 5′-dRP.Table IDNA oligonucleotides used for exonuclease assays The nucleotide targets for excision are indicated in boldface.OligonucleotidesDNA sequence (5′ to 3′)aAsterisks indicate the locations of the 32P label. F and U denote tetrahydrofuranyl and uridine residues, respectively. The p marks the location of a chemically synthesized 5′-phosphate groupUpstream strandsA1*GCTTGCATGCCTGCAGGTCGAA2*GCTTGCATGCCTGCAGGTCGAFTCTAGAGGATCCCCGGGTACCGAGCTCGAA3*TAGAGGATCCCCGCTAGCGGAA4*TAGAGGATCCCCGCTAGCGGAUTCTAGAGGATCCCCGGGTACCGAGCTCGADownstream strandsB1pCTCTAGAGGATCCCCGGGTACCGAGCTCGAB2CTCTAGAGGATCCCCGGGTACCGAGCTCGAB3pTCTAGAGGATCCCCGGGTACCGAGCTCGAB4TCTAGAGGATCCCCGGGTACCGAGCTCGAB5FTCTAGAGGATCCCCGGGTACCGAGCTCGAB6GTCTAGAGGATCCCCGGGTACCGAGCTCGATemplate strandsC1TCGAGCTCGGTACCCGGGGATCCTCTAGAGTCGACCTGCAGGCATGCAAGCC2TCGAGCTCGGTACCCGGGGATCCTCTAGAGGCGACCTGCAGGCATGCAAGCC3TCGAGCTCGGTACCCGGGGATCCTCTAGATCGACCTGCAGGCATGCAAGCC4TCGAGCTCGGTACCCGGGGATCCTCTAGAGCGACCTGCAGGCATGCAAGCD0TCGAGCTCGGTACCCGGGGATCCTCTAGAGCCCGCTAGCGGGGATCCTCTAD1TCGAGCTCGGTACCCGGGGATCCTCTAGACCCGCTAGCGGGGATCCTCTAa Asterisks indicate the locations of the 32P label. F and U denote tetrahydrofuranyl and uridine residues, respectively. The p marks the location of a chemically synthesized 5′-phosphate group Open table in a new tab Table IIComponents of duplex DNA substrates used for exonuclease assays Asterisks indicate substrates that were created by treatment with a catalytic amount of Apel following annealing. pF denotes a phosphorylated tetrahydrofuran residue. F denotes an unmodified tetrahydrofuran residue.SubstrateaSee Table I for oligonucleotide sequencesGap or nickNucleotides at 3′ terminiStructures at 5′ terminiOligonucleotides annealedS1GapA/G mismatchPhosphateA1:B3:C2S2*GapA/G mismatchpFA2:C2S3NickA/G mismatchPhosphateA1:B3:C4S4*NickA/G mismatchpFA2:C4S5GapA/T pairPhosphateA1:B3:C1S6*GapA/T pairpFA2:C1S7NickA/T pairPhosphateA1:B1:C1S8*NickA/T pairpFA2:C3S9GapA/T pairHydroxylA1:B4:C1S10NickA/T pairHydroxylA1:B4:C3S11GapA/T pairFA1:B5:C1S12GapA/T pairG/G mismatchA1:B6:C1S13bSubstrate was a primer-template pair lacking a downstream primerA/T pairA1:C1S14NickA/T pairFA1:B5:C3S15NickA/T pairG/G mismatchA1:B6:C3S16GapA/C mismatchHydroxylA3:B4:D0S17GapA/C mismatchPhosphateA3:B3:D0S18cTo create this substrate, oligonucleotide B5 was phosphorylated by T4 polynucleotide kinase prior to annealingGapA/C mismatchpFA3:B5:D0S19NickA/C mismatchHydroxylA3:B4:D1S20NickA/C mismatchPhosphateA3:B3:D1S21cTo create this substrate, oligonucleotide B5 was phosphorylated by T4 polynucleotide kinase prior to annealingNickA/C mismatchpFA3:B5:D1S22GapA/C mismatchdRPA4:D0S23NickA/C mismatchdRPA4:D1a See Table I for oligonucleotide sequencesb Substrate was a primer-template pair lacking a downstream primerc To create this substrate, oligonucleotide B5 was phosphorylated by T4 polynucleotide kinase prior to annealing Open table in a new tab Exonuclease Assays—Exonuclease reactions were performed in BER buffer (27Ranalli T.A. DeMott M.S. Bambara R.A. J. Biol. Chem. 2002; 277: 1719-1727Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar). Briefly, standard reactions contained 50 mm Hepes-KOH (pH 7.5), 8 mm MgCl2, 5% (v/v) glycerol, 0.5 mm dithiothreitol, 0.1 mg/ml bovine serum albumin, 10 nm DNA substrate, and enzyme concentrations as indicated in the figure legends. After incubation at 30 °C, the reactions were terminated at the indicated times by the addition of formamide loading buffer (90% formamide, 10 mm EDTA, bromphenol blue, and xylene cyanol) and heating at 100 °C for 2 min. The DNA products were then resolved by electrophoresis on acrylamide (14%) gels containing 7 m urea. After drying, the gels were analyzed using a Molecular Imager System (Model GS-525, Bio-Rad), and the results were quantified using Molecular Analyst software (Bio-Rad). The fraction of substrate degraded by the Ape1 exonuclease was calculated as previously described (28Sander M. Benhaim D. Nucleic Acids Res. 1996; 24: 3926-3933Crossref PubMed Scopus (15) Google Scholar). Briefly, the percent exonuclease activity of each reaction was calculated using Equation 1, &x0025;exonucleaseactivity=[(N-1)+2(N-2)+3(N-3)...]N+[(N-1)+2(N-2)+3(N-3)...](Eq. 1) where N represents the amount of the 21-mer DNA substrate, N - 1 is the amount of the 20-mer exonuclease product, N - 2 is the amount of the 19-mer exonuclease product, and so on. Comparison between Human Ape1 and E. coli Exonuclease III for Excision of 3′ Mismatches on BER Intermediates—Previous reports on the exonuclease activity of Ape1 focused on its activity at 3′ mismatched nucleotides at gaps and nicks in DNA (21Chou K.M. Cheng Y.C. Nature. 2002; 415: 655-659Crossref PubMed Scopus (199) Google Scholar, 29Hadi M.Z. Ginalski K. Nguyen L.H. Wilson 3rd, D.M. J. Mol. Biol. 2002; 316: 853-866Crossref PubMed Scopus (104) Google Scholar, 30Chou K.M. Cheng Y.C. J. Biol. Chem. 2003; 278: 18289-18296Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar), but the activity of Ape1 on BER intermediates with incised abasic residues was not investigated. We used DNA containing the synthetic AP analog tetrahydrofuran (F) to create a set of DNA substrates that simulate these base excision repair intermediates (Fig. 1A). Tetrahydrofuran residues are resistant to cleavage by β-elimination reactions (31Takeshita M. Chang C.N. Johnson F. Will S. Grollman A.P. J. Biol. Chem. 1987; 262: 10171-10179Abstract Full Text PDF PubMed Google Scholar) but are cleaved by Ape1 with the same catalytic efficiency as regular AP sites (19Wilson 3rd, D.M. Takeshita M. Grollman A.P. Demple B. J. Biol. Chem. 1995; 270: 16002-16007Abstract Full Text Full Text PDF PubMed Scopus (244) Google Scholar). After 5′ incision by Ape1, F residues serve as stable structural analogs of the 5′-dRP moiety that is produced following 5′ incision of natural AP sites. We found that the Ape1 exonuclease is most active on 3′ A/G mismatches adjacent to a single-nucleotide gap. However, the exonuclease was significantly less active at mismatches positioned at nicks or at gaps bearing 5′-incised F residues (Fig. 1, B and D). Exonuclease III is the E. coli homolog of Ape1 and shares many of the same DNA repair activities (13Demple B. Harrison L. Annu. Rev. Biochem. 1994; 63: 915-948Crossref PubMed Scopus (1295) Google Scholar). In addition to being an AP endonuclease, exonuclease III also possesses potent 3′-phosphodiesterase and 3′-phosphatase activities and a well characterized and robust 3′→5′-exonuclease (32Rogers S.G. Weiss B. Methods Enzymol. 1980; 65: 201-211Crossref PubMed Scopus (175) Google Scholar, 33Weiss B. Grossman L. Adv. Enzyme Rel. Areas Mol. Biol. 1987; 60: 1-34PubMed Google Scholar). Exonuclease III shares structural homology with Ape1 reflected in 28% amino acid sequence identity, including highly conserved catalytic residues (29Hadi M.Z. Ginalski K. Nguyen L.H. Wilson 3rd, D.M. J. Mol. Biol. 2002; 316: 853-866Crossref PubMed Scopus (104) Google Scholar, 34Gorman M.A. Morera S. Rothwell D.G. de La Fortelle E. Mol C.D. Tainer J.A. Hickson I.D. Freemont P.S. EMBO J. 1997; 16: 6548-6558Crossref PubMed Scopus (289) Google Scholar). To determine if exonuclease III activity was inhibited similarly to the Ape1 exonuclease by the presence of 5′-incised F residues at nicks and gaps, we assayed the ability of exonuclease III to degrade these same four DNA substrates across a range of enzyme concentrations. Exonuclease III exhibited only a small difference in activity between gaps and nicks (Fig. 1C, panels 1 and 3). The bacterial enzyme was not inhibited at all by the presence of the incised dRP analog (Fig. 1C, panels 2 and 4). Thus, the inhibition of the 3′→5′-exonuclease activity of Ape1 is a feature specific to the mammalian enzyme. Ape1 Exonuclease Activity for Correctly Matched 3′ Termini Is Inhibited by the Presence of Incised Abasic Residue—Previous experiments indicated that Ape1 is generally inefficient at excising non-mismatched nucleotides (19Wilson 3rd, D.M. Takeshita M. Grollman A.P. Demple B. J. Biol. Chem. 1995; 270: 16002-16007Abstract Full Text Full Text PDF PubMed Scopus (244) Google Scholar, 21Chou K.M. Cheng Y.C. Nature. 2002; 415: 655-659Crossref PubMed Scopus (199) Google Scholar, 35Lebedeva N.A. Khodyreva S.N. Favre A. Lavrik O.I. Biochem. Biophys. Res. Commun. 2003; 300: 182-187Crossref PubMed Scopus (40) Google Scholar) but that the enzyme may also be more active at excising some properly paired nucleotides than certain mismatched pairs (29Hadi M.Z. Ginalski K. Nguyen L.H. Wilson 3rd, D.M. J. Mol. Biol. 2002; 316: 853-866Crossref PubMed Scopus (104) Google Scholar). The difference in these reports may depend on the overall sequence context of the DNA substrate, which can affect the exonuclease activity by up to two orders of magnitude (29Hadi M.Z. Ginalski K. Nguyen L.H. Wilson 3rd, D.M. J. Mol. Biol. 2002; 316: 853-866Crossref PubMed Scopus (104) Google Scholar). Our experiments utilized a well characterized 51-mer oligonucleotide substrate that has long been used by several groups for routine BER assays (26Masuda Y. Bennett R.A. Demple B. J. Biol. Chem. 1998; 273: 30352-30359Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 36Singhal R.K. Prasad R. Wilson S.H. J. Biol. Chem. 1995; 270: 949-957Abstract Full Text Full Text PDF PubMed Scopus (294) Google Scholar). In the sequences used here (Fig. 2A), kinetic analysis showed that Ape1 was able to digest a properly matched A/T pair (Fig. 2B), albeit at a reduced efficiency compared with an A/G mismatch. On the 3′ mismatch, the same enzyme concentration resulted in digestion of most of the mismatched substrate even at the earliest time point (Fig. 2C). A comparison of estimated initial rates from exonuclease assays using a reduced amount of Ape1 indicated that the exonuclease activity is at least 3-fold more active on an A/G mismatch than on an A/T pair in this sequence context. Whereas previous studies found no difference in the exonuclease activity at nicks or gaps (21Chou K.M. Cheng Y.C. Nature. 2002; 415: 655-659Crossref PubMed Scopus (199) Google Scholar, 29Hadi M.Z. Ginalski K. Nguyen L.H. Wilson 3rd, D.M. J. Mol. Biol. 2002; 316: 853-866Crossref PubMed Scopus (104) Google Scholar), we found that Ape1 was at least 3-fold more active at gaps than at nicks in the sequence context used here, whether or not a 3′ mismatch was present (Fig. 2). Furthermore, these assays showed that Ape1 was inhibited at least 3-fold by the presence of incised F residues at both nicks and single-nucleotide gaps, consistent with the data shown in Fig. 1. Separate Inhibitory Effects of an Abasic Residue and a 5′-Phosphate on Ape1 Exonuclease Activity—We next wanted to explore in greater detail the role of different structural elements at the 5′-end of nicks and gaps that might affect the Ape1 3′→5′-exonuclease activity. Panel A and the tops of each box of panel C in Fig. 3 illustrate schematically the various DNA structures that were analyzed. The presence of a 5′-phosphate at a single-nucleotide gap did not affect the exonuclease activity (Fig. 3B). However, the presence of a 5′-phosphate at a nick inhibited the exonuclease activity 8-fold compared with a 5′-hydroxyl (Fig. 3B). We also assayed the Ape1 exonuclease activity on substrates with termini bearing 5′-unphosphorylated deoxyribose moieties or 5′-mismatched nucleotides, as well as substrates lacking a downstream strand to determine the relative inhibitory contribution made by the structural elements at the 5′ terminus of each substrate (Fig. 3C). In the sequence context of our DNA substrate, differences between single-nucleotide gaps and nicks were observed only when the 5′-phosphates were present. Consistent with data shown in Figs. 1 and 2, the Ape1 exonuclease was relatively inefficient at both nicks and single-nucleotide gaps when an abasic residue containing an adjacent phosphate group was present at the 5′ terminus (Fig. 3C, panels a and e). The Ape1 exonuclease activity increased only slightly when the 5′-phosphate was removed to yield a 5′-terminal abasic residue (Fig. 3C, panels b and f). However, substitution of the abasic residue with a guanine nucleotide to create a 5′ G/G mismatch increased the 3′→5′-exonuclease activity of Ape1 (Fig. 3C, panels c and g). DNA lacking a downstream double-stranded structure was a very poor substrate for the Ape1 exonuclease activity (Fig. 3C, panel d). This observation is consistent with structural studies, DNase footprinting analysis, and enzymatic data demonstrating that Ape1 requires at least three base pairs of duplex DNA both upstream and downstream of its active site for maximal AP endonuclease activity (37Wilson 3rd, D.M. Takeshita M. Demple B. Nucleic Acids Res. 1997; 25: 933-939Crossref PubMed Scopus (83) Google Scholar, 38Mol C.D. Izumi T. Mitra S. Tainer J.A. Nature. 2000; 403: 451-456Crossref PubMed Scopus (606) Google Scholar, 39Nguyen L.H. Barsky D. Erzberger J.P. Wilson 3rd, D.M. J. Mol. Biol. 2000; 298: 447-459Crossref PubMed Scopus (56) Google Scholar). Ape1 and E. coli Endonuclease IV Are Both Inhibited by 5′-Incised Abasic DNA at Nicks and Gaps—To verify that the inhibitory effect of 5′-dRP on the 3′→5′-exonuclease activity of Ape1 is a general effect rather than a sequence-specific feature of our DNA substrates, we assayed the Ape1 exonuclease activity on substrates whose sequences resemble some other previously characterized substrates (tops of each panel in Fig. 4A (29Hadi M.Z. Ginalski K. Nguyen L.H. Wilson 3rd, D.M. J. Mol. Biol. 2002; 316: 853-866Crossref PubMed Scopus (104) Google Scholar)). In this sequence context, which contained a 3′-A/C mismatch at nicks and gaps, Ape1 was slightly more active at the nicks than at the single-nucleotide gaps (Fig. 4A, compare panels a–c with d–f; Fig. 4B, compare panel a to b). Ape1 continued to be inhibited at nicks and gaps bearing 5′-incised F residues (Fig. 4A, panels c and f; Fig. 4B, panels a and b), although to a slightly lesser extent than previously observed (3-fold instead of 4-fold as shown in Fig. 2). Consistent with experiments shown in Fig. 3, we observed that Ape1 exonuclease was more active at nicked DNA when the 5′-phosphate was removed (Fig. 4B, panel b). However, we also observed the same increase in exonuclease activity at gapped substrates lacking the 5′-phosphate (Fig. 4B, panel a). A second AP endonuclease from E. coli, endonuclease IV, was also recently demonstrated to possess an intrinsic 3′→5′-exonuclease activity (40Kerins S.M. Collins R. McCarthy T.V. J. Biol. Chem. 2003; 278: 3048-3054Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). Although structurally unrelated to Ape1, endonuclease IV also has AP endonuclease, 3′-phosphodiesterase, and 3′-phosphatase activities. Endonuclease IV is inducible as part of the oxidative stress-activated SoxRS regulon of E. coli (41Pomposiello P.J. Bennik M.H. Demple B. J. Bacteriol. 2001; 183: 3890-3902Crossref PubMed Scopus (399) Google Scholar), and the enzyme plays a specialized role in repairing oxidatively damaged DNA (42Levin J.D. Demple B. Nucleic Acids Res. 1996; 24: 885-889Crossref PubMed Scopus (38) Google Scholar). To determine whether endonuclease IV is also inhibited by the presence of a 5′ abasic residue, which can arise from oxidatively damaged AP sites (1Demple B. DeMott M.S. Oncogene. 2002; 21: 8926-8934Crossref PubMed Scopus (86) Google Scholar), we assessed the ability of the endonuclease IV exonuclease to degrade these same substrates bearing 5′-phosphates or F residues. Endonuclease IV was nearly 2-fold more active than Ape1 as an exonuclease on gapped substrates lacking a 5′-phosphate or bearing 5′ F (Fig. 4A, compare filled wedges representing endonuclease IV