Induction of Reversible Complexes between Eukaryotic DNA Topoisomerase I and DNA-containing Oxidative Base Damages

We recently showed that abasic sites, uracil mismatches, nicks, and gaps can trap DNA topoisomerase I (top1) when these lesions are introduced in the vicinity of a top1 cleavage site (Pourquier, P., Ueng, L.-M., Kohlhagen, G., Mazumder, A., Gupta, M., Kohn, K. W., and Pommier, Y. (1997) J. Biol. Chem. 272, 7792–7796; Pourquier, P., Pilon, A. A., Kohlhagen, G., Mazumder, A., Sharma, A., and Pommier, Y. (1997)J. Biol. Chem. 26441–26447). In this study, we investigated the effects on top1 of an abundant base damage generated by various oxidative stresses: 7,8-dihydro-8-oxoguanine (8-oxoG). Using purified eukaryotic top1 and oligonucleotides containing the 8-oxoG modification, we found a 3–7-fold increase in top1-mediated DNA cleavage when 8-oxoG was present at the +1 or +2 position relative to the cleavage site. Another oxidative lesion, 5-hydroxycytosine, also enhanced top1 cleavage by 2-fold when incorporated at the +1 position of the scissile strand. 8-oxoG at the +1 position enhanced noncovalent top1 DNA binding and had no detectable effect on DNA religation or on the incision step. top1 trapping by 8-oxoG was markedly enhanced when asparagine adjacent to the catalytic tyrosine was mutated to histidine, suggesting a direct interaction between this residue and the DNA major groove immediately downstream from the top1 cleavage site. Altogether, these results demonstrate that oxidative base lesions can increase top1 binding to DNA and induce top1 cleavage complexes.

DNA is highly susceptible to reactive oxygen species in vivo, especially hydroxyl radicals that are generated by various forms of oxidative stress such as lipid peroxidation, products of inflammation, cellular respiration, and nearly ultraviolet light (1,2). Many oxidized bases have been identified in mammalian DNA (3). In vivo measurements indicate that oxidatively modified bases range from a few hundred to 4 ϫ 10 5 /cell/day depending on the technique used (4). Among these, 7,8-dihydro-8-oxoguanine (8-oxoG) 1 represents the most abundant lesion with estimates of about 100,000 8-oxoG being formed and ex-cised on average per rat cell per day (5). A recent study comparing different techniques reported a lesser but still formidable amount of 8-oxoG produced in young rat liver with a steadystate level of 0.04 adducts/10 5 G (6). Conversion of G to 8-oxoG results in retention of three hydrogen bonds between 8-oxoG and C and hence does not severely alter DNA architecture as determined by x-ray crystallography (7) (see Fig. 1B). However, polymerases involved in DNA replication selectively incorporate A opposite to 8-oxoG (8). In this case, 8-oxoG adopts the syn conformation, a geometry favoring two hydrogen bonds with A (7, 9, 10) (see Fig. 1B). In the absence of repair, A misincorporation leads to G-T transversions (11)(12)(13)(14) that could be relevant to the carcinogenic process. Pyrimidines also undergo oxidation in vivo and can lead to potential mutagenic derivatives such as 5-hydrocycytosine (5-ohC) (15,16).
In Escherichia coli, oxidized bases are removed by specific enzymes. 5-ohC is removed by endonuclease III and formamidopyrimidine DNA N-glycosylase also known as MutM (17,18). MutM also removes 8-oxoG paired to a C. When 8-oxoG is paired with A, the MutY glycosylase excises the complementary strand A residue that has been misincorporated by the polymerase (13). It has been proposed that the abnormal DNA major groove hydrogen bonding properties of oxoG, attributable specifically to the 8-oxo carbonyl, are responsible for recognition by glutamine or asparagine residues of the glycosylases (7). A number of cellular enzymes from yeast, Drosophila, and humans are capable of recognizing 8-oxoG, including glycosylases and the nucleotide excision repair complex (4, 19 -23). The effect of 8-oxoG incorporation on other ubiquitous DNA processing enzymes is still unknown.
In this study we have investigated the effect of 8-oxoG on eukaryotic topoisomerase I (top1) using purified enzyme and oligonucleotides containing a unique top1 cleavage site (24,25). DNA topoisomerases play a critical role in DNA metabolism by changing the topological state of DNA (26). Eukaryotic top1 is an essential (27,28) and ubiquitous nuclear enzyme (26,29,30). It catalyzes a reversible transesterification reaction by forming a covalent 3Ј-phosphotyrosyl bond between a DNA strand and an enzyme tyrosine residue (Tyr 723 for human top1) while leaving a 5Ј-hydroxyl terminus downstream to the cleavage site (31). These reversible cleavage complexes are the key catalytic intermediates that permit relaxation of DNA supercoiling and recombination. top1 inhibitors such as camptothecin and its derivatives inhibit the religation of the top1 cleavage complexes (32,33). DNA modifications such as uracil mismatches, abasic sites, nicks, and vinyl chloride adducts can also trap top1 cleavage complexes and in some cases lead to the formation of irreversible cleavage complexes also referred to as "suicide complexes" (25, 34 -38).
In this report, we show for the first time that 8-oxoG can promote formation of top1 cleavage complexes by increasing top1 binding to DNA. We also demonstrate the influence of the amino acid residue immediately flanking the catalytic tyrosine by using various top1 mutants including an asparagine-toserine mutant that was previously found to induce camptothecin resistance (39). These findings further extend the broad variety of DNA lesions affecting top1 activity and provide novel details on the interactions of specific residues of eukaryotic top1 with DNA.
Oligonucleotide Labeling and Annealing Procedures-3Ј Labeling was performed using terminal deoxynucleotidyl transferase (Life Technologies, Inc.) with [␣-32 P]cordycepin as described previously (41). Labeling mixtures were subsequently centrifuged through a G25 Sephadex column to remove excess unincorporated nucleotide. Radiolabeled single-stranded DNA oligonucleotides were annealed to the same concentration of unlabeled complementary strand in 1ϫ annealing buffer (10 mM Tris-HCl, pH 7.8, 100 mM NaCl, 1 mM EDTA). Annealing mixtures were heated to 95°C and slowly chilled overnight to room temperature.
top1 Purification-Overexpression of the human recombinant top1 was performed in Sf9 insect cells using a baculovirus virus construct containing the top1 full-length cDNA (42). Nuclear extracts were prepared from infected Sf9 cells 4 days after infection as described previously (25) and kept at Ϫ80°C before used. Human top1 was purified from nuclear extracts using nickel nitrilotriacetic acid-agarose beads (Qiagen, Santa Clarita, CA). 50 l of beads were placed into microcentrifuge tubes and washed three times with 1 ml of wash buffer (50 mM Hepes, pH 7, 0.5 mM dithiothreitol, 10 mM MgCl 2 , 3 mM MnCl 2 , 50 mM KCl). 500 l of nuclear extract were incubated with beads for 30 min at 4°C and washed two times with 300 l of the wash buffer. Endogenous top1 was eluted with wash buffer containing 40 mM imidazole. Human recombinant top1 was eluted with wash buffer containing 300 mM imidazole and dialyzed 3 h with 30 mM potassium phosphate buffer, pH 7, 0.1 mM EDTA, 5 mM dithiothreitol. Bovine serum albumin (0.2 mg/ml) and glycerol (30% v/v) were added to the purified top1, and aliquots were stored at Ϫ20°C. Yeast top1 extracts were performed as described previously. Proteins have been corrected for concentration by Western blotting, and the standard catalytic activity of all mutants was checked in standard assays. A baculovirus construct expressing the yeast mutant, top1Y727F, was constructed from plasmid YCpGAL1-top1Y727F (43) using the Bac-to-Bac baculovirus expression system from Life Technologies, Inc. Purification of recombinant yeast top1Y727F protein from infected Sf9 insect cells was essentially as described (44), except for a final phosphocellulose chromatographic step in place of Mono-S. 2 Yeast wild-type top1 and mutant proteins in which Asn 726 was mutated to serine, histidine, or aspartate 3 were partially purified from galactose-induced yeast cultures as described (45,46). top1 protein concentrations were corrected in Western blots, and the catalytic activities of all the mutant proteins were assessed in a plasmid DNA relaxation assay (45,46). top1 Reactions-DNA substrates (approximately 50 fmol/reaction) were incubated with yeast or human recombinant top1 or nuclear extracts for 30 min at 25°C in reaction buffer (10 mM Tris-HCl, pH 7.5, 50 mM KCl, 5 mM MgCl 2 , 0.1 mM EDTA, 15 g/ml bovine serum albumin). Reactions were stopped by adding SDS (final concentration, 0.5%). For reversal experiments, the SDS stop was preceded by the addition of NaCl (0.5 M for 30 min at 25°C for standard reactions or 0.35 M NaCl at 4°C for reversal kinetics).
Gel Electrophoresis and Analysis of Cleavage Products-Before loading of the electrophoresis, 3.3 volumes of Maxam Gilbert loading buffer (98% formamide, 0.01 M EDTA, 1 mg/ml xylene cyanol, 1 mg/ml bromphenol blue) were added to reaction mixtures. 16% denaturing polyacrylamide gels (7 M urea) were run at 40 V/cm at 50°C for 2-3 h and dried on 3MM Whatman paper sheets. Imaging and quantitations were performed using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
DNA Binding Assay-3Ј-Labeled DNA substrates (approximately 25 fmol/reaction) were incubated in the presence of increasing concentrations of yeast top1Y727F mutant protein (final volume, 20 l) in 50 mM Tris-HCl, pH 7.5 (47), and 50 ng of double-stranded poly dI-dC DNA (Sigma) for 5 min at room temperature. 10 l of 3ϫ neutral loading buffer (10 mM Tris HCl, pH 7.5, 30% glycerol, 0.1% bromphenol blue) were added to each reaction before loading into a 6% native polyacrylamide gel in 0.25ϫ TBE. Electrophoresis was for 2 h at 100 V in 0.25ϫ TBE buffer. Imaging and quantitations were performed using a PhosphorImager.

Induction of Reversible top1 Cleavage Complexes by 8-Oxoguanine Incorporation at the ϩ1 Position of a top1 Cleavage
Site-In this study, we investigated the effects of the abundant oxidative lesion 8-oxoG on top1 activity and how CPT would affect top1-DNA interactions under such conditions. For this purpose, we used oligonucleotides derived from a Tetrahymena ribosomal DNA sequence (24) containing a single top1 cleavage site that had been previously mutated to maximize top1 sensitivity to camptothecin and its derivatives (25,48) (Fig. 1A). We first studied the effect of 8-oxoG incorporation at the ϩ1 position of the scissile strand (X in Fig. 1A) on top1-mediated cleavage using purified recombinant human enzyme. Replacement of G by 8-oxoG at the ϩ1 position enhanced cleavage 3-5-fold (Fig. 1C, compare lanes 2 and 7). The same increase (approximately 6-fold) was observed when 8-oxoG was paired with A as it would be after polymerase misincorporation ( . These results are consistent with our previous finding that DNA alterations at this ϩ1 position can trap top1 and increase DNA cleavage (36,38). Under normal conditions, top1 cleavage is reversible, leading to religation of the DNA strand. The top1 cleavage/religation equilibrium can be shifted toward religation by increasing the salt concentration (48). In the case of the 8-oxoG-containing oligonucleotides, top1 cleavage remained reversible after the addition of 0.5 M NaCl (Fig. 1C, lanes C). Similar results were obtained when 5-ohC was used as a modified base. When introduced at the ϩ1 position of the scissile strand, 5-ohC enhanced top1 cleavage approximately 2-fold (Fig. 2B, compare lanes 2 and 7). However, 5-ohC failed to affect top1 activity when it was incorporated at the ϩ1 (or the ϩ2) position of the nonscissile strand opposite to G (Fig. 2C,  compare lanes 17 and 22 to lane 12). Interestingly, both the 8oxo and the 5-hydroxy group of the oxidized ϩ1 base on the scissile strand occupy the same region of the DNA major groove (Fig. 1B, middle panel, and Fig. 2A, right panel). Together, these results indicate that presence of an additional oxygen group in the DNA major groove on the ϩ1 base of the scissile strand selectively enhances the formation of top1 cleavage complexes.
Because we previously hypothesized that CPT traps top1 cleavage complexes by interacting with the ϩ1 purine of the top1 cleavage site (49), we tested whether the 8-oxoG and 5-ohC substitutions at the ϩ1 position would affect the activity of CPT. Fig. 1C shows that the 8-oxoG modifications had no effect on top1 sensitivity to CPT when 8-oxoG was paired to C (lane 9). Cleavage was also salt-reversible (compare lanes 9 and 10). However, when 8-oxoG was paired to A, CPT was not able to further enhance top1 cleavage (compare lanes 14 and 12). Induction of top1 Cleavage Complexes by 8-Oxoguanine Incorporation at the ϩ2 or Ϫ3 Position of a top1 Cleavage Site-We next studied the influence of the position of the 8-oxoG by replacing the guanines present in the immediate vicinity of the top1 site (Fig. 3A). 8-oxoG at the ϩ2 position on the scissile strand also led to reversible top1 trapping (Fig. 3B), and top1 cleavage was increased to approximately the same extent (4 -6-fold) as for the 8-oxoG at the ϩ1 position (Fig. 1C). By contrast, incorporation of 5-ohC at the ϩ2 position of the nonscissile strand did not increase the formation of top1 cleavage complexes (Fig. 2C, compare lanes 12 and 22). When the guanine at the Ϫ3 position of the nonscissile strand (lower strand) was replaced by 8-oxoG, cleavage at the normal site remained detectable (Fig. 3B). However, a new cleavage site was observed two bases upstream (Fig. 3B, open arrowhead). This observation is consistent with our previous studies showing that the presence of uracil, abasic site, or of a nick at the Ϫ2 position of the nonscissile strand all induced this new top1 cleavage upstream to the normal cleavage site (25,36). It is interesting to note that the 8-oxoG was incorporated at position Ϫ1 relative to the new cleavage site. We also tested the effect of this 8-oxoG incorporation at the Ϫ3 position of the nonscissile strand on possible top1-induced cleavage on the lower strand and did not detect any cleavage of this strand (data not shown).
CPT remained active at the pre-existing top1 cleavage site in the oligonucleotide with 8-oxoG at positions ϩ2 and Ϫ3 (com-pare lanes D and B) and CPT-induced cleavage was reversible (compare lanes E and D). However, CPT did not increase top1mediated cleavage at the new site (Fig. 3B, compare lanes 7  and 9). We have previously described the same lack of CPT activity when an abasic site was present at the Ϫ3 position (36). These results demonstrate that 8-oxoG can enhance top1 cleavage when present at the ϩ1 or ϩ2 position relative to a preexisting top1 site and that it can induce new top1 cleavages.
Effects of 8-Oxoguanine on top1-mediated DNA Cleavage and Religation-To assess the manner by which 8-oxoG enhanced top1 cleavage, we first performed kinetic experiments with the substrates shown in Fig. 1. Cleavage was enhanced in the 8-oxoG oligonucleotide after 1 min and reached a plateau at about 3-5-fold the cleavage level measured in the control oligonucleotide (Fig. 4A). top1-mediated DNA religation was studied at 0°C after addition of 0.35 M NaCl (Fig. 4B). Under these conditions, religation is slower than at 37°C and in the presence of 0.5 M NaCl (48). As a comparison, we used the control oligonucleotide in the presence of CPT, because enhancement of top1 cleavage by CPT is known to result from inhibition of religation of top1 cleavage complexes (48). After NaCl addition, cleavage was reversed in the case of 8-oxoG after 1 min, whereas CPT-induced cleavage remained detectable for more than 1 h. When compared with the control oligonucleotide in the absence of CPT, religation rates were comparable, as 80% of the cleavage complexes reversed after 5 s (data not shown). Thus, 8-oxoG at the ϩ1 position does not appear to mimic the effect of CPT, e.g. to inhibit top1-mediated DNA religation.
To investigate the effects of 8-oxoG on the induction of DNA cleavage by top1, we used a partially single-stranded 19-mer oligonucleotide, also referred to as suicide substrate, containing the 8-oxoG at the ϩ1 position relative to the top1 cleavage site (Fig. 5). Once top1 becomes covalently attached to the 3Ј DNA terminus, it releases a 5Ј-hydroxyl hexamer (35, 50). The lower strand was phosphorylated at its 5Ј-end to prevent intramolecular religation (41,50). Thus, using this oligonucleotide, cleavage can be studied under single turnover conditions (51). Fig. 5 shows that the presence of 8-oxoG at the ϩ1 position did not increase top1 cleavage rate as compared with the control oligonucleotide. Cleavage appeared even slightly slower in the case of the 8-oxoG-containing oligonucleotide (Fig. 5C). This result demonstrates that the increase in top1 cleavage complexes observed at equilibrium (such as in Fig. 1C) was not due to an increased rate of DNA cleavage.
8-Oxoguanine Enhances the Formation of Noncovalent top1-DNA Complexes-We next investigated the effect of 8-oxoG on the formation of noncovalent top1-DNA complexes using a yeast top1 mutant where the catalytic tyrosine 727 was mutated to phenylalanine (top1Y727Fp). This catalytically inactive mutant cannot perform the incision step but is still able to bind the top1 consensus sequence (52,53). 4 Using electro-phoretic mobility shift assays, we found that increasing concentrations of top1Y727Fp produced an increase in noncovalent top1-DNA complexes (Fig. 6A). More efficient top1Y727Fp binding was observed with the oligonucleotide containing 8-oxoG rather than G at the ϩ1 position (Fig. 6A, compare lanes 2-5 between the G:C (control) and the 8-oxoG:C oligonucleotides)). Quantitation of the top1-DNA complexes indicated about a 3-fold increase with the highest concentrations of enzyme (Fig. 6A, lanes 5) and about 7-fold with the lowest concentrations (Fig. 6A, lanes 2). Incorporation of 5-ohC at the ϩ1 position of the nonscissile strand, however, did not enhance top1Y727Fp DNA binding under the same conditions (Fig. 6B). Thus, enhancement of top1 cleavage by incorporation of 8-oxoG appears related to enhanced binding of top1 to the 8-oxoGcontaining DNA. terminus region (29) (Fig. 7A). We investigated the contribution of the asparagine immediately flanking the catalytic tyrosine for the enhancement of top1 cleavage complex by 8-oxoguanine. For this purpose, we used three different yeast top1 mutants where asparagine 726 was mutated to histidine, aspartic acid, or serine. Each of the partially purified mutant proteins was tested with the control oligonucleotide and with oligonucleotides containing an 8-oxoG at the ϩ1 or the ϩ2 position (Fig. 7B). With wild-type top1 and the asparagine to serine mutant protein, cleavage complexes were almost undetectable with the control oligonucleotide (G:C at ϩ1 position). In contrast, when asparagine was mutated to aspartic acid or histidine, top1 cleavage complexes were enhanced significantly, although the overall percentage of DNA cleavage was  (lanes B and D), or first with 0.5 M NaCl (final concentration) for an additional hour at 25°C before addition of 0.5% SDS (lanes C and E). Numbers to the right indicate product size (in nucleotides).
FIG. 4. Kinetics of formation and religation of top1 DNA cleavage complexes in an oligonucleotide with 8-oxoG incorporated at the ؉1 position. 3Ј end-labeled oligonucleotides with either a G (Control) or 8-oxoG paired to C (8-oxoG:C) at the ϩ1 position (see Fig. 1, A and B) were reacted with purified human top1. A, kinetics of cleavage induction. Reactions were stopped with 0.5% SDS at the indicated times. B, kinetics of religation were performed in the absence of CPT for the 8-oxoG/Cϩ1 oligonucleotide and in the presence of 10 M CPT for the control oligonucleotide. DNA substrates (lanes a) were incubated with purified top1 for 15 min at 25°C. Reactions were either stopped with 0.5% SDS (lanes 0) or reversed at 4°C by adding 0.35 M NaCl (final concentration) before adding 0.5% SDS. Incubation times (in min) are indicated above each lane. Quantitations are plotted in the right panels. below 5% (Fig. 7C). Incorporation of 8-oxoG at the ϩ1 or ϩ2 positions led to a marked increase in top1-mediated cleavage for wild-type yeast top1, which confirms our previous findings using the human recombinant enzyme (Fig. 1C). Enhancement of cleavage was also observed when the asparagine was mutated to histidine or serine (Fig. 7, B and C). When 8-oxoG was incorporated at the ϩ2 position of the scissile strand, only the histidine mutant led to marked enhancement of top1-induced DNA cleavage (approximately 13-fold increase as compared with the control oligonucleotide). Thus, replacement of the asparagine immediately adjacent to the catalytic tyrosine of top1 by histidine enhances DNA cleavage in 8-oxoG-containing oligonucleotides. DISCUSSION This study shows for the first time that 8-oxoG, which is one of the most abundant oxidative DNA lesions in mammalian cells (5), can induce the formation of reversible top1-linked DNA single-stranded breaks. In general, an increase of top1mediated cleavage complexes results from a shift in the cleavage/religation equilibrium. Until the present study, enhanced top1 cleavage by DNA damages (base mismatches, abasic sites, or vinyl chloride adducts; Refs. 36 and 38) was found to result from inhibition of religation, as in the case of CPT (30,32,54). By contrast, in this study we found that 8-oxoG paired with C at the ϩ1 position enhanced top1 cleavage complexes by increasing the binding of top1 to DNA. We also found that oxi-dation of cytosine ϩ1 on the scissile strand (5-ohC, another oxidative damage found to be mutagenic in vivo; Refs. 15 and 55) similarly enhanced top1-mediated DNA cleavage. This effect on top1 is remarkably specific because we did not detect any significant change in cleavage or binding when 8-oxoguanine or 5-hydroxycytosine were incorporated at the ϩ1 position on the nonscissile strand. Because replacement of G by 8-oxoG in DNA does not change base pairing or overall DNA structure (7), increased top1 binding is probably due to the presence of the nucleophilic 8-oxo group in the vicinity of the cleaved strand in the DNA major groove.
As shown in Fig. 1B, oxidation of G converts the 7 position from hydrogen bond acceptor to donor and the 8 position from a weak hydrogen bond donor to a strong hydrogen acceptor, O8 (7). Hydrogen bond donor/acceptors within the DNA major groove are probably essential for recognition by DNA repair enzymes such as the DNA glycosylase, MutM (7). Furthermore, because the hydroxy group of 5-ohC and the 8-oxo group of 8-oxoG are occupying approximately the same position in the DNA major groove and because both enhance top1 cleavage complexes, it is plausible that they exert their activity by the same mechanism. We propose that the differential arrangement of hydrogen bond donors/acceptors within the major groove enhances top1 binding to 8-oxoG or 5-ohC-containing DNA. Based on the recently published crystallographic structure of a top1-DNA complex (56,57), asparagine 722 in human FIG. 5. Influence of 8-oxoG incorporation on top1-induced DNA cleavage using a suicide oligonucleotide. A, structure of the 3Ј-labeled suicide oligonucleotide. X represents the ϩ1 base of the scissile strand: either G or 8-oxoG. The lower strand was phosphorylated (p) at the 5Ј end to prevent recombination (41). B, oligonucleotides (lanes 0) were incubated at 25°C with human top1 for 5 s, 10 s, 30 s, 1 min, 3 min, 5 min, 10 min, and 30 min (lanes 1-8, respectively). Reactions were stopped with 0.5% SDS. C, quantitation of the data shown in panel B. top1 (and the corresponding Asn 726 in yeast Saccharomyces cerevisiae; Fig. 7A) is a plausible hydrogen bond donor for the O8 of 8-oxoguanine. Asparagine 722 is pointed roughly in line with the ϩ1 base on the scissile strand (57). Changing asparagine 722 to histidine would probably enhance a potential interaction with 8-oxoG because the 5-membered ring would probably bring the amino group even closer to the oxygen, which would be covalently bonded to the carbon 8 of guanine. 5 It is also possible that the asparagine to histidine mutation increases the reactivity of the adjacent catalytic tyrosine.
A frequent in vivo DNA modification is the incorporation of an A opposite 8-oxoG by DNA polymerases. This lesion is mutagenic. top1-mediated DNA cleavage was further enhanced when 8-oxoG was paired to A at the ϩ1 position. 8-oxoG:A base pair is in an syn-anti-conformation, resulting in the protrusion of the 8-oxo group in the minor grove of the DNA and altered hydrogen bonding (Ref. 7 and Fig. 1B). Moreover, 8-oxoG:A base pairing was able to inhibit top1-mediated religation in the absence of CPT, whereas 8-oxoG:C could not (data not shown). We previously reported that DNA mismatches at the ϩ1 posi-tion of a top1 cleavage site reversibly enhance top1-mediated cleavage complexes (36,58). Thus, in the case of the 8-oxoG:A, we propose that DNA structural distortions and inhibition of religation contribute to top1 trapping.
CPT has been shown to interact with the ϩ1 base immediately downstream of the top1 cleavage site (49). This interaction has been suggested to inhibit DNA religation by top1 and thus to trap the top1 cleavage complexes. We found that CPT remained inhibitory when 8-oxoG was paired with C at the ϩ1 position. However, the CPT activity was markedly decreased when 8-oxoG was paired to A at this same position. We previously reported such a decrease of CPT activity in the case of mismatches or abasic sites at this ϩ1 position, whereas wobble base pairing such as that of G:U did not reduce CPT activity (36). These results suggest the importance of base pairing at the ϩ1 position for CPT activity. These results are also consistent with the selective interaction of camptothecin at the top1-DNA interface with the ϩ1 base pair.
It is also interesting that the top1 trapping effect of 8-oxoG was observed with crude nuclear extracts from leukemia cells 5 L. Stewart, personal communication.  Fig. 1) were incubated with increasing concentrations of yeast top1Y727Fp for 5 min at 25°C. Electromobility shift assay was then performed. Lanes 1-5 correspond to 0, 1.5, 3, 7.5, and 15 ng of purified yeast top1Y727Fp, respectively. B, control (G:C) and G:5-ohC oligonucleotides were incubated with increasing amount of purified yeast top1Y727Fp as described in A. Electromobility shift assay was performed under the same conditions as in A. (data not shown). This result suggests that top1 can recognize the 8-oxoG modifications in the presence of other nuclear proteins and can efficiently compete for binding. Induction of top1 cleavage complexes by oxidized bases could potentially lead to DNA damage and cell death (as in the case of CPT) or possibly to recombination and mutagenic effects. The potential role that specific DNA repair pathways play in the processing of these lesions has yet to be established.