Repair and Mutagenic Potential of Oxaluric Acid, aMajor Product of Singlet Oxygen-Mediated Oxidation of

8-Oxo-7,8-dihydroguanine

Victor Duarte,† Didier Gasparutto,† Michel Jaquinod,‡ Jean-Luc Ravanat,† andJean Cadet*,†

Laboratoire des Lesions des Acides Nucleiques, Service de Chimie Inorganique etBiologique, UMR 5046, Departement de Recherche Fondamentale sur la Matiere Condensee,

CEA Grenoble, 17 Avenue des Martyrs, F-38054 Grenoble Cedex 9, France, and Laboratoire deSpectrometrie de Masse des Proteines, Institut de Biologie Structurale, 24 rue Jules Horowitz,

F-38027 Grenoble Cedex, France

Received July 31, 2000

Oxidative reactions within DNA commonly result in base modifications. Among the fourDNA bases, guanine is the most susceptible to various oxidants, and its related oxidized form,8-oxo-7,8-dihydroguanine, has been extensively studied in terms of repair and mutagenicity.However, 8-oxo-7,8-dihydroguanine is readily subjected to further oxidation, and this hasbecome a point of interest. We recently found that singlet oxygen oxidation of 8-oxo-7,8-dihydroguanine led to the predominant formation of oxaluric acid as the final product. Wereport herein on the biological features of oxaluric acid dealing with in vitro DNA synthesisand its removal from DNA by repair enzymes. Nucleotide insertion opposite oxaluric acid,catalyzed by Kf exo- and Taq indicates, that oxaluric acid induces G to T and G to Ctransversions. On the other hand, oxaluric acid represents a block when synthesis is performedwith pol â. Interestingly, DNA repair experiments carried out with formamidopyrimidine DNAN-glycosylase (Fpg) and endonuclease III (endo III) show that oxaluric acid is a substrate forboth enzymes. Values of kcat/Km for the Fpg-mediated removal of oxidative guanine lesionsrevealed that 8-oxoGua is only a slightly better substrate than oxaluric acid. Interestingly,the results obtained with endo III suggest that oxaluric acid is a much better substrate thanis 5-hydroxycytosine (5-OHC), an oxidized pyrimidine base.

Introduction

Oxidative damage to DNA is known to be mutagenicand is possibly involved in the aging process and humandiseases, including cancer (1-4). Reactive oxygen speciesare likely the most important source of spontaneousdamage to DNA. In this respect, many efforts have beendevoted to the characterization of the main oxidativeDNA lesions, with emphasis on the modified bases (5-8). Among the four DNA bases, guanine is the mostsusceptible to oxidation (6, 9). The related modifiedpurine, 8-oxo-7,8-dihydroguanine (8-oxoGua),1 has re-ceived considerable attention since it is widely used as abiomarker of oxidative stress, formed under exposure tovarious conditions, including ionizing radiation, reactiveoxygen species, chemical oxidation, and photoionization(5, 8, 10-13). It has long been recognized that 8-oxoGuais readily subjected to further oxidation since electro-

chemical detection allowed its assessment by HPLC.More recently, it was shown that 8-oxoGua may berevealed as a piperidine labile site upon selective oxida-tion by Ir(IV) and subsequent gel electrophoresis analysis(14, 15). Several studies have shown that 8-oxoGua ishighly reactive toward oxidizing reagents (13, 16-22),suggesting that the latter modified base may also besensitive to in vivo DNA oxidation and that the resultingproducts may play a role in the mutagenic processassociated with DNA damage.

Replication of oxidized DNA may induce mutations dueto the mispairing of the modified bases with an incorrectnucleobase. 8-OxoGua has been shown to pair with C andA, leading in the latter case to G to T transversions (23-25). Interestingly, a comprehensive repair system for8-oxoGua, which involves several enzymes, has beenidentified in Escherichia coli (26, 27). The Fpg protein,also known as MutM (28), removes 8-oxoGua from theduplex DNA when it is paired with a cytosine. On theother hand, the 8-oxoGua/A mismatch is repaired by theMutY protein. This enzyme removes the unmodifiedadenine when it is paired with 8-oxoGua. Another protein(MutT) is an 8-oxodTPase that prevents incorporation of8-oxodGTP into DNA (29). Although the biological fea-tures of 8-oxoGua have been extensively studied, littleis known about the mutagenic potential of secondaryoxidative products of guanine. One of the main exceptionsconcerns cyanuric acid, a singlet oxygen- or a peroxyni-

* To whom correspondence should be addressed: CEA Grenoble,DRFMC/SCIB/LAN, 17 Avenue des Martyrs, F-38054 Grenoble, France.Phone: 33 4 76 88 49 87. Fax: 33 4 76 88 50 90. E-mail: cadet@drfmc.ceng.cea.fr.

† CEA Grenoble.‡ Institut de Biologie Structurale.1 Abbreviations: ODN, oligodeoxyribonucleotide; 5-OHC, 5-hydroxy-

cytosine; 8-oxoGua, 8-oxo-7,8-dihydroguanine; Kf exo-, Klenow frag-ment exo-; pol â, DNA polymerase â; Taq, Taq DNA polymerase; Fpg,formamidopyrimidine DNA glycosylase; endo III, endonuclease III;MALDI-TOF-MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry; MB, methylene blue; TFA, trifluoroaceticacid.

46 Chem. Res. Toxicol. 2001, 14, 46-53

10.1021/tx0001629 CCC: $20.00 © 2001 American Chemical SocietyPublished on Web 12/21/2000

trite-mediated oxidation product of 8-oxoGua. Primerextension experiments using the Klenow fragment ofDNA polymerase I and a cyanuric acid-modified templatehave shown that this lesion pairs with either A or G,giving rise to G to T and G to C transversions. Interest-ingly, it was also reported that cyanuric acid was asubstrate neither for Fpg nor for endo III (30). It wasshown that oxidation of an 8-oxoGua-containing oligomerby Ir(IV), leading predominantly to guanidinohydantoin,induces misincorporation of dGMP and dAMP during invitro DNA synthesis. Moreover, when using rose bengalas a singlet oxygen generator, dGMP and dAMP werealso incorporated opposite a nonidentified 8-oxoGuaoxidation product (31).

We recently found that oxaluric acid is the majoroxidized product of 8-oxoGua in single-stranded DNA,either in the presence of methylene blue (MB, mainly atype II photooxidant) or riboflavin (predominantly a typeI photosensitizer) or by using a chemical source of singletoxygen (32). Tretyakova et al. and Burney et al. alsopointed out the formation of oxaluric acid as a secondaryoxidized product of guanine upon exposure to peroxyni-trite. Depending on the concentration of peroxynitrite,cyanuric acid was also generated in addition to oxaluricacid (33, 34). Taken together, these results suggest thatoxaluric acid is a key compound in the further oxidationof 8-oxoGua and that the latter modified base may exhibitmutagenic potential.

The determination of biological consequences of oxida-tive DNA damage has been made possible by the avail-ability of site-specific modified oligonucleotides (35-37).In the study presented here, we synthesized two oxaluricacid-containing oligomers. This was achieved by MB-mediated photooxidation of oligonucleotides having aunique 8-oxoGua residue. Emphasis was then placed onthe determination of the biological features of oxaluricacid. First, we examined the nucleotide insertion oppositethe modified base during in vitro DNA synthesis cata-lyzed by Klenow fragment exo- (Kf exo-), DNA poly-merase â (pol â), and Taq polymerase (Taq). DNA repairexperiments were performed to determine whether ox-aluric acid was a substrate for Fpg or endo III. Km andVmax values were then determined for a comparativestudy of the excision efficiencies of other modified DNAbases by the two DNA repair enzymes.

Experimental Section

Materials. T4 polynucleotide kinase, Kf exo-, [γ-32P]ATP,dNTPs, and NAP-25 Sephadex and MicroSpin G-25 columnswere obtained from Amersham Pharmacia Biotech (Uppsala,Sweden). ZipTipC18 columns were from Millipore (Milford, MA).Pol â was from Trevigen-Interchim (Montlucon, France). Taqand calf spleen 5′-phosphodiesterase were purchased fromBoehringer-Mannheim (Mannheim, Germany). Methylene bluewas from Merck (Darmstadt, Germany). Fpg and endo III werekind gifts from S. Boiteux (CEA, Fontenay-aux-Roses, France).

HPLC, MALDI-TOF-MS, and HPLC-MS/MS Analyses.Oligonucleotide mixtures resulting from methylene blue-medi-ated oxidation of 8-oxoGua-containing oligonucleotides wereanalyzed by HPLC on a Hypersil (Interchim, Montlucon, France)ODS column (5 µm, 250 mm × 4.6 mm). This was achieved usinga gradient of acetonitrile (from 0 to 10% either over 90 min forthe 15-mers or over 30 min for the 3-mer) in 25 mM TEAA (pH7) at a flow rate of 1 mL/min. The oligonucleotides were detectedat the output of the column using a UV/visible spectrophotom-eter set at 260 nm. Oligonucleotides subjected to MALDI-TOFmass spectrometry analysis were first desalted by using Zip-

TipC18 columns as described by the manufacturer. Mass spectrawere obtained with a commercial time-of-flight mass spectrom-eter (Voyager-DE; Perseptive Biosystems, Framingham, MA)equipped with a 337 nm nitrogen laser and a pulsed delay sourceextraction. The spectra were recorded from 256 laser shots withan accelerating voltage of 25 kV in the linear and positive modes.For the matrix, a mixture of 3-hydroxypicolinic acid and picolinicacid in a 4:1 (w/w) ratio was dissolved in a 50% acetonitrileaqueous solution that contained 0.1% TFA and a small amountof Dowex-50W 50X8-200 (Sigma) cation-exchange resin. Onemicroliter of a 0.1% TFA aqueous solution of the sample wasadded to 1 µL of the matrix, and the resulting solution wasstirred. The sample was subsequently placed on top of the targetplate and allowed to dry by itself. The spectra were calibratedusing synthetic oligonucleotides with known molecular masses.

Oligonucleotide Synthesis. Oligonucleotides (Figure 1)were synthesized by standard phosphoramidite chemistry usingan Applied Biosystems Inc. 392 DNA synthesizer. The 8-oxoGua-containing oligonucleotides 1 and 3 were prepared using acommercially available phosphoramidite monomer of 8-oxodGuo(Glen Research, Sterling, VA). Unmodified oligonucleotides 5-7were deprotected in a concentrated aqueous ammonia solution(32%) for 15 h at 55 °C. Oligonucleotides 1 and 3 weredeprotected in a concentrated ammonia solution of 0.25 Mâ-mercaptoethanol for 15 h at 55 °C to prevent further oxidationof 8-oxoGua during the deprotection step. Oligonucleotides 1,3, and 5-7 were purified by PAGE using a 20% polyacryl-amide/7 M urea gel and then desalted using NAP-25 Sephadexcolumns. The integrity of oligonucleotides 1, 3, and 5-7 wasassessed by MALDI-TOF mass spectrometry. Oligomers were5′-end-labeled using T4 polynucleotide kinase and [γ-32P]ATPprior to being purified on MicroSpin G-25 columns.

Synthesis of Oxaluric Acid-Containing Oligonucle-otides. Oxaluric acid-containing oligonucleotides 2 and 4 wereprepared by specific methylene blue photosensitization of oli-

Figure 1. Oligonucleotide sequences. 8OG is 8-oxo-7,8-dihy-droguanine and OXA oxaluric acid. R stands for an intrastrand2-deoxyribose unit.

Repair and Mutagenic Potential of Oxaluric Acid Chem. Res. Toxicol., Vol. 14, No. 1, 2001 47

gonucleotides 1 and 3, respectively (32). The samples wereexposed to visible light generated by a 75 W tungsten lamp ina glass vial placed 10 cm from the light source. Typically, 200µL of a 100 µM aqueous solution of the 8-oxoGua-modifiedoligonucleotide was incubated with 1 OD unit/mL of methyleneblue and irradiated for 30 min. The mixture was kept at 37 °Cfor 24 h and then purified by HPLC using the conditionsdescribed above. The presence and the integrity of oxaluric acidin the oligonucleotides, subjected or not to enzymatic digestion(vide infra), were assessed by MALDI-TOF mass spectrometry.

Enzymatic Digestion of Oligomers 2 and 4 by a 5′-3′Phosphodiesterase and Subsequent MALDI-TOF-MSAnalysis. Oxaluric acid-containing oligonucleotides 2 and 4 (0.1OD unit) were incubated at 37 °C with 10-3 unit of calf spleen5′-phosphodiesterase in 20 µL of 20 mM ammonium citrate (pH5). Aliquots (2 µL) were withdrawn at increasing periods of time,and the reactions were stopped by adding 50 µL of water priorto freezing the samples in liquid nitrogen. Lyophilized sampleswere then analyzed by MALDI-TOF-MS following the conditionsdescribed above.

Primer Extension. Reactions catalyzed by Kf exo- werecarried out in 10 µL of 50 mM Tris-HCl (pH 7.5), 10 mM MgCl2,0.05 mg/mL BSA, and 1 mM DTT. Primer extension reactionsusing pol â were performed in 10 µL solutions of 50 mM Tris-HCl (pH 8.8), 10 mM MgCl2, 10 mM KCl, 0.4 mg/mL BSA, 1mM DTT, and 1.5% glycerol. Reactions catalyzed by Taq wereconducted in 10 µL solutions of 10 mM Tris-HCl (pH 8.3) with1.5 mM MgCl2 and 50 mM KCl. The buffered solutions thatcontained the oligonucleotide template 2 and the 5′-end-labeledprimer 5′-d(GGAGTGGAGA) (5), in a template/primer nanomo-lar ratio (2/5) of 45/15, were heated to 60 °C for 5 min and thencooled to 4 °C over a period of 2 h. DNA polymerization reactionswere carried out with either 100 µM solutions of a single dNTPor a mixture of all four dNTPs. The solutions were maintainedat 37 °C for 20 min in the presence of 0.01 unit of Kf exo-. Thesamples were incubated for 1 h at 37 °C in the presence of 0.3unit of either pol â or Taq. The reactions were stopped by adding5 µL of a solution containing 95% formamide, 0.1% bromophenolblue, and 0.1% xylene cyanol (formamide dye). The samples wereheated at 70 °C for 3 min prior to being applied to a 20%polyacrylamide/7 M urea gel. Subsequently, the analysis of theradiolabeled bands was achieved by phosphorimagery (Molec-ular Dynamics Phosphorimager) using Image QuanT software.

Fpg and Endonuclease III Repair Studies: Assays forNicking Activity. DNA repair experiments were carried outwith Fpg and endo III proteins using modified double-strandedDNA fragments that contained a unique oxaluric acid residueas the substrate. Typically, 0.5 pmol of 32P-labeled modifiedoligonucleotide 4 was annealed to 0.75 pmol of the complemen-tary strand (6A, 6C, 6G, or 6T) by heating at 70 °C for 5 minand subsequent slow cooling to room temperature over a periodof 2 h. The integrity of the modified duplex was assessed by

MALDI-TOF mass spectrometry (data not shown). The enzy-matic reactions were performed in 10 µL solutions of 20 mMTris-HCl (pH 7.5), 1 mM EDTA, and 100 mM KCl at 37 °C for30 min with either 10 ng of Fpg/µL or 20 ng of endo III/µL. Forcontrol experiments that require denatured Fpg or endo III, theenzyme was first dissolved in formamide and heated at 65 °Cfor 20 min prior to incubation with the modified duplex DNA.Enzymatic reactions were stopped by adding 5 µL of formamidedye, and the samples were subjected to denaturing 20% PAGE.The gels were analyzed as previously described by phosphor-imagery.

Kinetic Studies with Determination of Vmax and Km. Theconcentration range of modified oligonucleotide 4, in the pres-ence of 1.5 equiv of the complementary strand 6C, was 0.1-2µM. Substrate concentrations were chosen so that the Michae-lis-Menten curves reach a plateau. For each reaction, in avolume of 10 µL, the amount of 32P-labeled oligonucleotide was1 pmol. The concentrations were either 0.25 ng/µL for Fpg (30.2kDa) or 0.5 ng/µL for endo III (23 kDa). The enzymatic reactionswere allowed to proceed at 37 °C for either 10 min in thepresence of Fpg or 4 min in the presence of endo III. Thereactions were stopped by adding formamide dye. The sampleswere subjected to a 20% denaturing PAGE, and the resultinggel was analyzed by phosphorimagery. Bands corresponding tocleavage products and unreacted oligonucleotides were quanti-fied using Image QuanT software. Vmax and Km constants werecalculated by nonlinear least-squares fitting of the data pointsusing Microcal Origin, on the basis of at least three separateexperiments. Reaction velocity (V) was expressed in picomolesof substrate per minute per nanogram of enzyme, while thesubstrate concentration was given as nanomolarity or micro-molarity (Figures 6A and 7A).

Results

Synthesis and Characterization of Oxaluric Acid-Containing Oligonucleotides 2 and 4. Oxaluric acid-modified oligonucleotides 2 and 4 were obtained uponphotosensitization with MB of the corresponding 8-oxo-Gua oligomers as recently reported (32). The HPLCelution profile obtained 24 h after the photooxidationreaction of 1 with MB, a predominant type II photosen-sitizer, is presented in Figure 2A. MB-mediated oxidationof 8-oxoGua leads exclusively to oxaluric acid as the finalproduct (peak 2 at 51.2 min), and peak 1 at 55.2 min isascribed to the starting 8-oxoGua oligonucleotide. Isola-tion of the oxaluric acid-modified oligomer was efficientlyperformed on a Hypersil ODS column. MALDI-TOF-MSanalysis of the related compound led to a peak at m/z4396.3 having the expected mass (caculated M + H+ )4397) (Figure 2B). Additional support for the presence

Figure 2. (A) HPLC elution profile 24 h after the irradiation of 1 with MB. (B) MALDI-TOF-MS analysis of the oxaluric acid-containing oligonucleotide 2.

48 Chem. Res. Toxicol., Vol. 14, No. 1, 2001 Duarte et al.

and the integrity of oxaluric acid in oligomer 2 was gainedfrom experiments involving enzymatic digestion of thelatter oligonucleotide by calf spleen 5′-phosphodiesteraseand subsequent MALDI-TOF-MS analysis. As shown inFigure 3, this led to the release of different fragmentsthat differ in mass by successive loss of nucleotides. Thedifference in mass between two adjacent peaks allows theidentification of the released nucleotide and the deter-mination of the overall oligomer sequence. The massspectra observed after digestion for 1 h show that themodified oligonucleotide is efficiently hydrolyzed by theenzyme, leading to the release of the oxaluric acidnucleotide with the expected mass of 310.1 (calculatedmass of 311) at the right position within the sequence.A similar HPLC profile and mass spectral data wereobserved for the photooxidation of oligomer 3 into 4.

Nucleotide Insertion Opposite Oxaluric Acid.Attempts were made to determine the mutagenic poten-tial of oxaluric acid, the major singlet oxygen oxidationproduct of 8-oxoGua in single-stranded DNA, during invitro DNA synthesis. For this purpose, a 15-mer template2 that contains the oxaluric acid residue at position 5from the 5′-end was subjected to primer extension usinga 10-mer primer 5. The primer was extended using threedifferent DNA polymerases in the presence of either asingle dNTP or a mixture of all four dNTPs (Figure 4).When the modified oligonucleotide 2 was used as thetemplate for pol â-mediated polymerization (lanes 1-5),the oxaluric acid lesion inhibited the primer extensionwith no dNTP insertion. This suggests that oxaluric acidis able to induce a change in the conformation of thetemplate which is not suitable for DNA synthesis by polâ. However, the primer extension catalyzed by Kf exo-

(lanes 6-10) induces exclusively dAMP insertion oppositeoxaluric acid (lane 6). Interestingly, on the basis of themigration of the band on the gel, the enzyme was able toextend the oxaluric acid/dA mispair and a second dAMPwas incorporated opposite dT. When all four dNTPs werepresent, complete extension of the primer was observed,

suggesting that the lesion is easily bypassed by theenzyme; however, the integrity of the complement wasnot determined. DNA synthesis catalyzed by Taq wasalso conducted on template 2 (lanes 11-15). In additionto dAMP insertion opposite the damage and extensionbeyond the oxaluric acid/dA mismatch (lane 11), signifi-cant amounts of dGMP were also incorporated. When thefour dNTPs were present, transient inhibition was ob-served opposite the lesion with only small amounts ofthe fully extended primer.

Oxaluric Acid Excision by Fpg and Endo III DNARepair Enzymes. The substrate specificity of Fpg andendo III toward the oxaluric acid-modified base wasexamined using a 15-mer DNA duplex containing thedamage opposite any of the four DNA bases (4‚6 du-plexes). Interestingly, both DNA N-glycosylases cleavedthe modified strand at the oxaluric acid site. Figure 5Ashows the cleavage efficiencies of the two enzymes. Incontrol experiments, where the 4‚6C duplex was incu-bated in the presence of denatured Fpg (lane 4) or endoIII (lane 5), no enzymatic cleavage was observed. Thesubstrate remained intact by comparison to the startingoligonucleotide (lane 3). Lanes 1, 6, 8, and 10 show thatoxaluric acid is recognized and efficiently cleaved by Fpgregardless of the base on the opposite strand. Completecleavage of the modified strand was observed for allduplexes at a concentration of 10 ng/µL. Similar resultswere observed when the damage was processed by endoIII at a concentration of 20 ng/µL (lanes 2, 7, 9, and 11).PAGE migrations of the fragments obtained upon Fpgand endo III-mediated excision of oxaluric acid andsubsequent cleavage of the strand were compared. Thisindicates a different processing of the damage by the twoproteins, providing different products. Cleavage of modi-fied duplexes by endo III (lanes 2, 7, 9, and 11) generatestwo bands, not well separated on the gel, while Fpg, gave

Figure 3. MALDI mass spectra of the products resulting fromthe digestion of 2 by calf spleen 5′-phosphodiesterase afterincubation for 1 h.

Figure 4. In vitro DNA synthesis. Primer extension reactionscatalyzed by pol â (lanes 1-5), Kf exo- (lanes 6-10), and Taq(lanes 11-15) on template 2, using a single dNTP (A, C, G, orT) or a mixture of the four dNTPs (lane N).

Figure 5. (A) PAGE analysis of oxaluric acid strand cleavageby Fpg (lanes 1, 6, 8, and 10) or endo III (lanes 2, 7, 9, and 11),using for each duplex 10 ng of Fpg/µL or 20 ng of endo III/µL.Lane 3 is the control lane without repair enzyme. D representsthe denatured enzyme. (B) MALDI-TOF-MS analysis of oxaluricacid strand cleavage by Fpg.

Repair and Mutagenic Potential of Oxaluric Acid Chem. Res. Toxicol., Vol. 14, No. 1, 2001 49

rise to a single band which migrates faster on the gel(lanes 1, 6, 8, and 10). MALDI-TOF-MS analysis of thefragments generated by Fpg from the 2‚7 duplex ispresented in Figure 5B. The peak at m/z 4732.1 corre-sponds to the complementary strand 7. The fragmentsat m/z 1204.8 and 2979.8 arise from the cleavage of 2 asfollows. The fragment at m/z 1204.8 is ascribed to the4-mer ODN released 5′ to the lesion (5′-CTCTp-3′,calculated M + H+ ) 1205.8), while the fragment at m/z2979.8 corresponds to the 10-mer ODN released 3′ to thedamage (5′-pTCTCCACTCC-3′, calculated M + H+ )2980). Taken together, these results indicate that oxalu-ric acid is excised by Fpg, followed by enzymatic cleavageof both the 3′- and 5′-phosphodiester bonds at the site ofthe damage. These findings are in agreement with a â-δelimination mechanism as previously reported for Fpg(38).

Comparative Kinetics and Determination of Vmax

and Km. To assess the relative efficiency of the excisionof oxaluric acid by Fpg and endo III, the modifiedoligonucleotide 4 was annealed to the complementarystrand 6C and subjected to Fpg and endo III cleavage.Experiments in which the concentration of the substratewas varied were carried out, and values of Vmax and Km

were determined (Figures 6A and 7A). Kinetic valuesobtained either with Fpg for 8-oxoGua and oxazolone orwith endo III for oxazolone and 5-OHC are also reportedfor a comparative study. Consideration of the kcat/Km

values, presented in Figure 6B, indicates that the re-moval of 8-oxoGua, one of the best substrates for Fpg, isonly favored by 2-fold in comparison to oxaluric acid. Onthe other hand, oxaluric acid is a much better substratefor endo III than 5-OHC as inferred from the comparisonof the kcat/Km constants listed in Figure 7B. It should beadded that the processing of oxazolone by the latter DNA

repair enzyme is favored by approximately 2-fold withrespect of that of oxaluric acid.

Discussion

In this study, the mutagenic potential of oxaluric acidduring in vitro DNA synthesis and the excision of thelatter oxidized purine by Fpg and endo III were investi-gated. The oxaluric acid-modified template 2 was usedas a substrate for primer extension catalyzed by Kf exo-,pol â, and Taq DNA polymerases. DNA synthesis oppositethe lesion and full extension of the primer past thedamage were studied using a single dNTP and a mixtureof all four dNTPs, respectively. Replication of template2 indicates that dAMP was incorporated by Kf exo- andTaq. Moreover, dGMP was also inserted by Taq oppositethe modified base (Figure 4). Primer extension reactionscatalyzed by pol â on template 2 led to different resultsin comparison with the results provided by Kf exo- andTaq. In fact, pol â was unable to insert a nucleotideopposite the lesion, leading to a stop at the site of thedamage. This result may suggest a change in thetemplate conformation in such a way that the modifiedbase is no longer a substrate for pol â. The resultspresented here would indicate that oxaluric acid forma-tion in DNA may lead to G to T and G to C transversions.Furthermore, oxaluric acid may also represent a blockduring DNA synthesis.

Repair experiments were performed to assess theremoval of oxaluric acid by Fpg and endo III. The latterenzyme has already been shown to recognize a widerange of modified pyrimidine bases (39-43), while Fpgacts mainly on altered purines (26, 44-46). Our resultsindicate that oxaluric acid is removed from DNA duplexesby both DNA repair enzymes, thus broadening the

Figure 6. (A) Michaelis-Menten kinetics. Cleavage of the oxaluric acid-containing oligonucleotide by Fpg (0.25 ng/µL). (B) Km andVmax values for the removal of oxaluric acid, oxazolone, and 8-oxoGua by Fpg. The numbers in parentheses indicate standard errors.

50 Chem. Res. Toxicol., Vol. 14, No. 1, 2001 Duarte et al.

substrate specificity of these enzymes. Interestingly,oxaluric acid is efficiently excised regardless of the baseon the opposite strand. While this work was nearingcompletion, Tretyakova et al. reported that oxaluric acid,being paired with C, was cleaved by Fpg (47). Our dataconfirm this finding and extend the analysis to thecleavage efficiencies by assessing the kinetic parametersfor the excision of oxaluric acid by Fpg and endo III. Thiswork, together with the results obtained recently in ourlaboratory for the oxazolone modified base, providedstrong evidence for the removal of oxidized purines byendo III (48). The relatively small size of oxazolone andoxaluric acid in comparison to other oxidized purineproducts might explain their incision by endo III. On theother hand, oxaluric acid cleavage by Fpg confirms thatthe latter DNA N-glycosylase can also accommodatesmall substrates in its active site. The latter feature ofthe Fpg protein has been previously reported sinceseveral oxidized pyrimidine bases, including 5-hydroxy-cytosine, 5-hydroxyuracil, uracil glycol, and thymineglycol, are cleaved by Fpg (40, 49, 50). The kineticparameters for oligomer cleavage at the oxaluric acid siteby Fpg and endo III are reported in Figures 6B and 7B.By comparison of kcat/Km values, 8-oxoGua was found tobe a slightly better substrate for Fpg than oxaluric acid.The removal of 8-oxoGua is only favored by 2-fold.Consideration of the kcat/Km values for the excision ofoxaluric acid by endo III indicates that the latter modifiedbase is a much better substrate than 5-OHC, which isan established substrate, efficiently removed by endo III(39). As previously reported, uracil glycol and oxazolonewere found to be the best substrates for endo III (39, 48).From our results, it appears that oxaluric acid is alsoefficiently recognized and cleaved by endo III since

removal of uracil glycol and oxazolone is only favored by3- and 2-fold, respectively.

As previously reported in our group for other DNAlesions (50, 51), MALDI-TOF-MS analysis of the frag-ments generated upon enzymatic cleavage provides astraightforward method for gaining insight into themechanistic pathway of repair enzymes. As shown inFigure 5B for oxaluric acid, Fpg removes the modifiedbase and cleaves the resulting abasic site, giving rise tofragments which are in agreement with a â-δ elimina-tion mechanism. In the case of endo III, it is generallyassumed that the excision of a base damage is followedby a â-elimination of the resulting AP site (52, 53). PAGEanalysis of such a R-â unsaturated fragment leads totwo distinct bands on the gel as shown on Figure 5A (54,55). However, the MALDI-TOF-MS analysis of the prod-ucts obtained upon excision of oxaluric acid by endo IIIfailed to detect the presence of a fragment bearing analkenal residue (data not shown). More detailed studies,including the identification of a chemically generatedalkenal residue by MALDI-TOF mass spectrometry, arein progress in an effort to assess the mechanism of endoIII-mediated cleavage of a modified base.

G to T and G to C transversions have been reportedduring replication of DNA treated with oxygen radicals.For example, McBride et al. (56) found that Fe2+/oxygendamage to DNA led to single-base substitutions with themost frequent being G to C transversions followed by Cto T and G to T. These authors also reported that singletoxygen-mediated DNA damage generates predominantlyG to C transversions, with G to T transversions occurring

2 J.-L. Ravanat, V. Duarte, and J. Cadet, unpublished results.

Figure 7. (A) Michaelis-Menten kinetics. Cleavage of the oxaluric acid-containing oligonucleotide by endo III (0.5 ng/µL). (B) Kmand Vmax values for the removal of oxaluric acid, oxazolone, and 5-OHC by endo III. The numbers in parentheses indicate standarderrors.

Repair and Mutagenic Potential of Oxaluric Acid Chem. Res. Toxicol., Vol. 14, No. 1, 2001 51

at a lower frequency (57). It has to be noted that G to Ctransversions induced by singlet oxygen have also beendetected in a mutation assay involving mammalian cells(58). Recently, Valentine et al. (59) studied the DNAdamage generated by peroxyl radicals. The spectrum ofmutations induced in E. coli upon transfection of DNAexposed to ROO• led mainly to G to C and G to Ttransversions. The authors reported that oxidized gua-nines were sensitive to Fpg treatment; however, effortsto detect 8-oxoGua indicated that the latter modified basewas not present, thus suggesting that G to T transver-sions were not induced by 8-oxoGua. Taken together,these results suggest that in oxygen-mediated DNAdamage, 8-oxoGua may not induce all of the G to Ttransversions and that this oxidized purine is not re-sponsible for the G to C tranversions. On the other hand,the importance of oxaluric acid as a secondary oxidizedproduct of guanine in DNA was recently reported byTretyakova and co-workers and Burney et al. Both of thelatter works showed that oxaluric acid is generated as afinal stable product when 8-oxoGua-containing oligo-nucleotides are treated with peroxynitrite (33, 34). Giventhe high susceptibility of 8-oxoGua to further oxidationand the results reported herein, oxaluric acid may be agood candidate with which to explain some transversionsin the mutation spectra observed during DNA replicationfollowing exposure to oxidizing agents.

In conclusion, it was shown that photooxidation of8-oxoGua-containing oligonucleotides, under singlet oxy-gen generation conditions, leads to the formation ofoxaluric acid as the final stable product. In vitro DNAsynthesis experiments suggested that the latter modifiedpurine may play a role in the mutagenesis associatedwith oxidative reactions to DNA since 8-oxoGua is aprimary target for oxidizing agents. However, repairexperiments showed that oxaluric acid is efficientlycleaved by, at least, Fpg and endo III proteins. It willnow be interesting to assess how this lesion is recognizedby other DNA N-glycosylases, including human DNArepair enzymes. Moreover, we recently gained evidencefor the formation of oxaluric acid in isolated double-stranded DNA.2 It will also be of interest to search forthe formation of oxaluric acid within cellular DNA uponexposure to various oxidizing agents. Such studies arecurrently in progress using a highly sensitive HPLC-tandem mass spectrometry assay.

Acknowledgment. We are grateful to Dr. SergeBoiteux for the gift of endonuclease III and Fpg proteins.We thank the Comite de Radioprotection (Electricite deFrance) for financial support.

References

(1) Ames, B. N. (1983) Dietary carcinogens and anticarcinogens.Science 221, 1256-1264.

(2) Ames, B. N., Shigenaga, M. K., and Hagen, T. M. (1993) Oxidants,antioxidants, and the degenerative diseases of aging. Proc. Natl.Acad. Sci. U.S.A. 90, 7915-7922.

(3) Ames, B. N., and Gold, L. S. (1991) Endogenous mutagens andthe cause of aging and cancer. Mutat. Res. 250, 3-16.

(4) Floyd, R. A. (1990) The role of 8-hydroxyguanine in carcinogenesis.Carcinogenesis 11, 1447-1450.

(5) Breen, A. P., and Murphy, J. A. (1995) Reactions of oxyl radicalswith DNA. Free Radical Biol. Med. 18, 1033-1077.

(6) Burrows, C. J., and Muller, J. G. (1998) Oxidative nucleobasemodifications leading to strand scission. Chem. Rev. 98, 1109-1151.

(7) Cadet, J., Berger, M., Douki, T., and Ravanat, J. L. (1997)Oxidative damage to DNA: Formation, measurement, and bio-logical significance. Rev. Physiol. Biochem. Pharmacol. 131, 1-87.

(8) von Sonntag, C. (1987) The Chemical Basis of Radiation Biology,Taylor and Francis, London.

(9) Steenken, S., and Jovanovic, S. V. (1997) How easily oxidizableis DNA? One electron reduction potentials of adenosine andguanosine radicals in aqueous solution. J. Am. Chem. Soc. 119,617-618.

(10) Kasai, H., Yamaizumi, Z., Berger, M., and Cadet, J. (1992)Photosensitized formation of 7,8-dihydro-8-oxo-2′-deoxyguanosine(8-hydroxy-2′-deoxyguanosine) in DNA by riboflavin. A non singletoxygen mediated reaction. J. Am. Chem. Soc. 114, 9692-9694.

(11) Ravanat, J. L., and Cadet, J. (1995) Reaction of singlet oxygenwith 2′-deoxyguanosine and DNA. Isolation and characterizationof the main oxidation products. Chem. Res. Toxicol. 8, 379-388.

(12) Spassky, A., and Angelov, D. (1997) Influence of the local helicalconformation on the guanine modifications generated from one-electron DNA oxidation. Biochemistry 36, 6571-6576.

(13) Adam, W., Saha-Moller, C. R., Schonberger, A., Berger, M., andCadet, J. (1995) Formation of 7,8-dihydro-8-oxoguanine in the 1,2-dioxetane-induced oxidation of calf thymus DNA: evidence forphotosensitized DNA damage by thermally generated tripletketones in the dark. Photochem. Photobiol. 62, 231-238.

(14) Shigenaga, M. K., Park, J.-W., Cundy, K. C., Gimeno, C. J., andAmes, B. N. (1990) In vivo oxidative DNA damage: measurementof 8-hydroxy-2′-deoxyguanosine in DNA and urine by high-performance liquid chromatography with electrochemical detec-tion. Methods Enzymol. 186, 521-530.

(15) Muller, J. G., Duarte, V., Hickerson, R. P., and Burrows, C. J.(1998) Gel electrophoretic detection of 7,8-dihydro-8-oxoguanineand 7,8-dihydro-8-oxoadenine via oxidation by Ir(IV). NucleicAcids Res. 26, 2247-2249.

(16) Hickerson, R. P., Prat, F., Muller, J. G., Foote, C. S., and Burrows,C. J. (1999) Sequence and stacking dependence of 8-oxoguanineoxidation: comparison of one-electron vs singlet oxygen mecha-nisms. J. Am. Chem. Soc. 121, 9423-9428.

(17) Niles, J. C., Burney, S., Singh, S. P., Wishnok, J. S., andTannenbaum, S. R. (1999) Peroxynitrite reaction products of 3′,5′-di-O-acetyl-8-oxo-7,8-dihydro-2′-deoxyguanosine. Proc. Natl. Acad.Sci. U.S.A. 96, 11729-11734.

(18) Adam, W., Saha-Moller, C. R., and Schonberger, A. (1996)Photooxidation of 8-oxo-7,8-dihydro-2′-deoxyguanosine by ther-mally generated triplet-excited ketones from 3-(hydroxymethyl)-3,4,4-trimethyl-1,2-dioxetane and comparison with type I and typeII photosensitizers. J. Am. Chem. Soc. 118, 9233-9238.

(19) Raoul, S., and Cadet, J. (1996) Photosensitized reaction of 8-oxo-7,8-dihydro-2′-deoxyguanosine: Identification of 1-(2-deoxy-â-D-erythro-pentofuranosyl)cyanuric acid as the major singlet oxygenoxidation product. J. Am. Chem. Soc. 118, 1892-1898.

(20) Goyal, R. N., Jain, N., and Garg, D. K. (1997) Electrochemicaland enzymic oxidation of guanosine and 8-hydroxyguanosine andthe effects of oxidation products in mice. Bioelectrochem. Bioenerg.43, 105-114.

(21) Sheu, C., and Foote, C. S. (1995) Photosensitized oxygenation ofa 7,8-dihydro-8-oxoguanosine derivative. Formation of dioxetaneand hydroperoxide intermediates. J. Am. Chem. Soc. 117, 474-477.

(22) Sheu, C., and Foote, C. S. (1995) Reactivity toward singlet oxygenof a 7,8-dihydro-8-oxoguanosine (“8-hydroxyguanosine”) formedby photooxidation of a guanosine derivative. J. Am. Chem. Soc.117, 6439-6442.

(23) Cheng, K. C., Cahill, D. S., Kasai, H., Nishimura, S., and Loeb,L. A. (1992) 8-Hydroxyguanine, an abundant form of oxidativeDNA damage, causes G‚T and A‚C substitutions. J. Biol. Chem.267, 166-172.

(24) Shibutani, S., Takeshita, M., and Grollman, A. P. (1991) Insertionof specific bases during DNA synthesis past the oxidation-damaged base 8-oxodG. Nature 349, 431-434.

(25) Lowe, L. G., and Guengerich, P. (1996) Steady-state and pre-steady-state kinetic analysis of dNTP insertion opposite 8-oxo-7,8-dihydroguanine by Escherichia coli polymerases I exo- andII exo-. Biochemistry 35, 9840-9849.

(26) Michaels, M. L., Tchou, J., Grollman, A. P., and Miller, J. H.(1992) A repair system for 8-oxo-7,8-dihydrodeoxyguanine. Bio-chemistry 31, 10964-10968.

(27) Michaels, M. L., Cruz, C., Grollman, A. P., and Miller, J. H. (1992)Evidence that MutY and MutM combine to prevent mutationsby an oxidative form of guanine. Proc. Natl. Acad. Sci. U.S.A.89, 7022-7025.

(28) Boiteux, S., O’Connor, T. R., Lederer, F., Gouyette, A., and Laval,J. (1990) Homogeneous Escherichia coli Fpg protein. J. Biol.Chem. 265, 3916-3922.

52 Chem. Res. Toxicol., Vol. 14, No. 1, 2001 Duarte et al.

(29) Maki, H., and Sekiguchi, M. (1992) MutT protein specificallyhydrolyses a potent mutagenic substrate for DNA synthesis.Nature 355, 273-275.

(30) Gasparutto, D., Da Cruz, S., Bourdat, A.-G., Jaquinod, M., andCadet, J. (1999) Synthesis and biochemical properties of cyanuricacid nucleoside-containing DNA oligomers. Chem. Res. Toxicol.12, 630-638.

(31) Duarte, V., Muller, J. G., and Burrows, C. J. (1999) Insertion ofdGMP and dAMP during in vitro DNA synthesis opposite anoxidized form of 7,8-dihydro-8-oxoguanine. Nucleic Acids Res. 27,496-502.

(32) Duarte, V., Gasparutto, D., Yamaguchi, L. F., Ravanat, J. L.,Martinez, G. R., Medeiros, M. H. G., Di Mascio, P., and Cadet, J.(2000) Oxaluric acid as the major product of singlet oxygen-mediated oxidation of 8-oxo-7,8-dihydroguanine in DNA. J. Am.Chem. Soc. (in press).

(33) Burney, S., Niles, J. C., Dedon, P. C., and Tannenbaum, S. R.(1999) DNA damage in deoxynucleosides and oligonucleotidestreated with peroxynitrite. Chem. Res. Toxicol. 12, 513-520.

(34) Tretyakova, N. Y., Niles, J. C., Burney, S., Wishnok, J. S., andTannenbaum, S. R. (1999) Peroxynitrite-induced reactions ofsynthetic oligonucleotides containing 8-oxoguanine. Chem. Res.Toxicol. 12, 459-466.

(35) Wang, D., Kreutzer, D. A., and Essigmann, J. M. (1998) Mutage-nicity and repair of oxidative DNA damage: insights from studiesusing defined lesions. Mutat. Res. 400, 99-115.

(36) Gasparutto, D., Bourdat, A.-G., D’Ham, C., Duarte, V., Romieu,A., and Cadet, J. (2000) Repair and replication of oxidized DNAbases using modified oligodeoxyribonucleotides. Biochimie 82, 19-24.

(37) Cadet, J., Bourdat, A.-G., D’Ham, C., Duarte, V., Gasparutto, D.,Romieu, A., and Ravanat, J.-L. (2000) Oxidative base damage toDNA: specificity of base excision repair enzymes. Mutat. Res. 462,121-128.

(38) Bhagwat, M., and Gerlt, J. A. (1996) 3′- and 5′-strand cleavagereactions catalyzed by the Fpg protein from Escherichia coli occurvia successive â- and δ-elimination mechanisms, respectively.Biochemistry 35, 659-665.

(39) Wang, D., and Essigmann, J. M. (1997) Kinetics of oxidizedcytosine repair by endonuclease III of Escherichia coli. Biochem-istry 36, 8628-8633.

(40) Hatahet, S., Kow, Y. W., Purmal, A. A., Cunningham, R. P., andWallace, S. S. (1994) New substrates for old enzymes. J. Biol.Chem. 269, 18814-18820.

(41) Demple, B., and Linn, S. (1980) DNA N-glycosylases and UVrepair. Nature 287, 203-208.

(42) Breimer, L. H., and Lindahl, T. (1984) DNA glycosylase activitiesfor thymine residues damaged by ring saturation, fragmentation,or ring contraction are functions of endonuclease III in Escheri-chia coli. J. Biol. Chem. 259, 5543-5548.

(43) Boorstein, R. J., Hilbert, T. P., Cadet, J., Cunningham, R. P., andTeebor, G. W. (1989) UV-induced pyrimidine hydrates in DNAare repaired by bacterial and mammalian DNA glycosylaseactivities. Biochemistry 28, 6164-6170.

(44) Tchou, J., Bodepudi, V., Shibutani, S., Antoshechkin, I., Miller,J., Grollman, A. P., and Johnson, F. (1994) Substrate specificityof Fpg protein. J. Biol. Chem. 269, 15318-15324.

(45) Chetsanga, C. J., and Lindahl, T. (1979) Release of 7-methylgua-nine residues whose imidazole rings have been opened fromdamaged DNA by a DNA glycosylase from Escherichia coli.Nucleic Acids Res. 6, 2673-2684.

(46) Tchou, J., Kasai, H., Shibutani, S., Chung, M. H., Laval, J.,Grollman, A. P., and Nishimura, S. (1991) 8-Oxoguanine (8-hydroxyguanine) DNA glycosylase and its substrate specificity.Proc. Natl. Acad. Sci. U.S.A. 88, 4690-4694.

(47) Tretyakova, N. Y., Wishnok, J. S., and Tannenbaum, S. R. (2000)Peroxynitrite-induced secondary oxidative lesions at guaninenucleobases: chemical stability and recognition by the Fpg DNArepair enzyme. Chem. Res. Toxicol. 13, 658-664.

(48) Duarte, V., Gasparutto, D., Jaquinod, M., and Cadet, J. (2000)In vitro DNA synthesis opposite oxazolone and repair of this DNAdamage using modified oligonucleotides. Nucleic Acids Res. 28,1555-1563.

(49) Purmal, A. A., Lampman, G. W., Bond, J. P., Hatahet, J., andWallace, S. S. (1998) Enzymatic processing of uracil glycol, amajor oxidative product of DNA cytosine. J. Biol. Chem. 273,10026-10035.

(50) D’Ham, C., Romieu, A., Jaquinod, M., Gasparutto, D., and Cadet,J. (1999) Excision of 5,6-dihydroxy-5,6-dihydrothymine, 5,6-dihydrothymine and 5-hydroxycytosine from defined sequenceoligonucleotides by Escherichia coli endonuclease III and Fpgproteins: kinetic and mechanistic aspects. Biochemistry 38,3335-3344.

(51) Bourdat, A. G., Gasparutto, D., and Cadet, J. (1999) Synthesisand enzymatic processing of oligodeoxynucleotides containingtandem base damage. Nucleic Acids Res. 27, 1015-1024.

(52) Kim, J., and Linn, S. (1988) The mechanisms of action of E. coliendonuclease III and T4 UV endonuclease (endonuclease V) atAP sites. Nucleic Acids Res. 16, 1135-1141.

(53) Kow, Y., and Wallace, S. S. (1987) Mechanism of action ofEscherichia coli endonuclease III. Biochemistry 26, 8200-8206.

(54) Bailly, V., and Verly, G. (1987) Escherichia coli endonuclease IIIis not an endonuclease but a â-elimination catalyst. Biochem. J.242, 565-572.

(55) Mazumder, A., Gerlt, J. A., Absalon, M. J., Stubbe, J., Cunning-ham, R. P., Withka, J., and Bolton, P. H. (1991) Stereochemicalstudies of the â-elimination reactions at aldehydic abasic sites inDNA: Endonuclease III from Escherichia coli, sodium hydroxydeand Lys-Trp-Lys. Biochemistry 30, 1119-1126.

(56) McBride, T. J., Preston, B. D., and Loeb, L. A. (1991) Mutagenicspectrum resulting from DNA damage by oxygen radicals. Bio-chemistry 30, 207-213.

(57) McBride, T. J., Schneider, J. E., Floyd, R. A., and Loeb, L. A.(1992) Mutations induced by methylene blue plus light in single-stranded M13mp2. Proc. Natl. Acad. Sci. U.S.A. 89, 6866-6870.

(58) Costa de Oliveira, R., Ribeiro, D. T., Nigro, R. G., Di Mascio, P.,and Menck, C. F. M. (1992) Singlet oxygen induced mutationspectrum in mammalian cells. Nucleic Acids Res. 20, 4319-4323.

(59) Valentine, M. R., Rodriquez, H., and Termini, J. (1998) Mutagen-esis by peroxy radical is dominated by transversions at deoxy-guanosine: evidence for the lack of involvement of 8-oxo-dG and/or abasic site formation. Biochemistry 37, 7030-7038.

TX0001629

Repair and Mutagenic Potential of Oxaluric Acid Chem. Res. Toxicol., Vol. 14, No. 1, 2001 53