Excision of 8-oxoguanine from methylated CpG dinucleotides by human 8-oxoguanine DNA glycosylase

FEBS Letters 587 (2013) 3129–3134

journal homepage: www.FEBSLetters.org

Excision of 8-oxoguanine from methylated CpG dinucleotides by human 8-oxoguanine DNA glycosylase Rustem D. Kasymov a, Inga R. Grin a,b, Anton V. Endutkin a, Serge L. Smirnov c, Alexander A. Ishchenko b, Murat K. Saparbaev b, Dmitry O. Zharkov a,d,⇑ a

SB RAS Institute of Chemical Biology and Fundamental Medicine, 8 Lavrentieva Ave., Novosibirsk 630090, Russia Groupe ‘‘Réparation de l’ADN’’, CNRS UMR8200, Université Paris-Sud, Institut de Cancérologie Gustave Roussy, Villejuif, France Department of Chemistry, Western Washington University, Bellingham, WA 98225-9150, USA d Department of Molecular Biology, Faculty of Natural Sciences, Novosibirsk State University, 2 Pirogova St., Novosibirsk 630090, Russia b c

a r t i c l e

i n f o

Article history: Received 30 May 2013 Revised 5 August 2013 Accepted 6 August 2013 Available online 13 August 2013

Edited by Varda Rotter Keywords: DNA repair Epigenetic methylation 5-Methylcytosine 8-Oxoguanine OGG1

a b s t r a c t CpG dinucleotides are targets for epigenetic methylation, many of them bearing 5-methylcytosine (mCyt) in the human genome. Guanine in this context can be easily oxidized to 8-oxoguanine (oxoGua), which is repaired by 8-oxoguanine-DNA glycosylase (OGG1). We have studied how methylation affects the efficiency of oxoGua excision from damaged CpG dinucleotides. Methylation of the adjacent cytosine moderately decreased the oxoGua excision rate while methylation opposite oxoGua lowered the rate of product release. Cytosine methylation abolished stimulation of OGG1 by repair endonuclease APEX1. The OGG1 S326C polymorphic variant associated with lung cancer showed poorer base excision and lost sensitivity to the opposite-base methylation. The overall repair in the system reconstituted from purified proteins decreased for CpG with mCyt in the damaged strand. Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Methylation of cytosine at C5 in CpG dinucleotides is an epigenetic modification almost universally used by higher eukaryotes to regulate gene expression [1,2]. Two mechanisms accounting for the change in gene activity by methylation have been described: direct changes in the affinity of transcription factors for their recognition sites [3,4] and recognition of 5-methylcytosine (mCyt, Fig. 1A) by specific binding proteins that further modulate the accessibility of CpG-containing DNA for transcription factors through direct competition or various modes of chromatin remodeling [5,6]. Proper establishment or removal of methylation at CpG is critical for developmental regulation and suppression of carcinogenesis [7,8]. While cytosine in CpG is the methylation mark carrier, its neighbor, guanine, is often a target for DNA damage. Notably, guanine is easily oxidized to produce 8-oxoguanine (oxoGua, Fig. 1B) [9], a pre-mutagenic base instructing DNA polymerases

⇑ Corresponding author at: SB RAS Institute of Chemical Biology and Fundamental Medicine, 8 Lavrentieva Ave., Novosibirsk 630090, Russia. Tel.: +7 (383) 3635128; fax: +7 (383) 3635153. E-mail address: [email protected] (D.O. Zharkov).

to incorporate dAMP opposite it and thus inducing G?T transversions [10]. The adverse effects of oxoGua in humans are normally kept in check by the base excision DNA repair (BER) system: a specific 8-oxoguanine-DNA glycosylase (OGG1) excises the damaged base, and then abasic site endonuclease (APEX1), DNA polymerase b (POLB), and DNA ligase IIIa conclude the repair replacing the damaged nucleotide with a regular one [11]. Other subpathways and participating proteins can be involved in oxoGua repair but it inevitably starts with the action of OGG1 and APEX1 [12]. Conservative estimates put the level of oxoGua in human DNA at 1/106 guanines, with a potential to increase many fold under oxidative stress conditions [13]. The frequency of CpG dinucleotides in the human genome is 1% [14]; considering the size of the human diploid genome is 6  109 nucleotide pairs, it would contain 6  107 CpG dinucleotides, each with two guanines. Thus, each human diploid cell at any moment will harbor 120 CpG dinucleotides containing a spontaneously formed oxoGua base. Moreover, chromatin remodeling involving histone demethylation locally produces reactive oxygen species that oxidize guanine [15]. In mammals, 70–80% of CpG dinucleotides is methylated [16], making the appearance of oxoGua in the methylated CpG context a potentially biologically relevant event. It is well known that mCyt, being easily deaminated to thymine, carries a mutagenic

0014-5793/$36.00 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2013.08.008

3130

R.D. Kasymov et al. / FEBS Letters 587 (2013) 3129–3134

A

B

G

A

C

oG

T

C

G

A

mC

oG

T

C

C

T

G

C

A

G

C

T

G

C

A

G

non-methylated

5-methylcytosine

8-oxoguanine

cis-hemimethylated

G

A

C

oG

T

C

G

A

mC

oG

T

C

C

T

G

mC

A

G

C

T

G

mC

A

G

trans-hemimethylated

fully methylated

C

Fig. 1. (A) Structure of modified bases used in this work. (B) Schematic structure of the substrates (six central base pairs are shown). oG, 8-oxoguanine, mC, 5-methylcytosine. The strand containing 8-oxoguanine was 32P-labeled. (C) Kinetic scheme of catalysis by OGG1. DNAoG, DNA containing 8-oxoguanine; DNAAP, DNA containing AP site after oxoGua excision; DNAnick, final nicked product after b-elimination.

risk, and C?T transitions at CpG are among the most frequent point mutations in humans; however, guanine in this context is also a mutation hotspot [17]. As a part of an attempt to understand the interplay between epigenetic methylation, DNA damage, and DNA repair, in this paper we have characterized the action of BER enzymes on oxoGua in non-methylated, hemimethylated and fully methylated CpG dinucleotides. 2. Materials and methods 2.1. Oligonucleotides and enzymes Recombinant full-length wild-type human OGG1, its mutant S326C form, Escherichia coli Fpg protein, human APEX1, and human POLB were purified as described [18–22]. Bacteriophage T4 polynucleotide kinase and DNA ligase were purchased from SibEnzyme (Russia). c[32P]ATP and a[32P]dGTP were from Laboratory of Biotechnology (Institute of Chemical Biology and Fundamental Medicine). Oligonucleotides (Table 1) were synthesized from commercially available phosphoramidites (Glen Research, Sterling, VA) on an ASM-2000 DNA/RNA Synthesizer (Biosset, Russia) and purified by reverse-phase HPLC on a PRP-1 C18 column (Hamilton, Reno, NV). The oligonucleotides were 50 -labeled with 32P using c[32P]ATP and T4 polynucleotide kinase according to the manufacturer’s protocol. The labeled oxoGua-containing oligonucleotides were annealed to an appropriate complementary strand to form the non-methylated substrate (CO annealed to CG, CO//CG), the hemimethylated substrate containing mCyt in the damaged strand (cis-hemimethylated, MO//CG) or in the non-damaged strand (trans-hemimethylated, CO//MG), or a fully methylated substrate (MO//MG) (Fig. 1B). The complementary strand was taken in a twofold molar excess to ensure that all labeled strand is in the duplex form; as the affinity of all used enzymes for single-stranded DNA is low, this is not expected to affect the results.

2.2. Kinetic experiments The reaction mixture contained 50 mM Tris–HCl (pH 7.5), 100 mM NaCl, 1 mM EDTA, and 1 mM DTT. To determine the base excision rate constant (k2), the mixture (120 ll) was supplemented with 10 nM 32P-labeled DNA substrate and 200 nM OGG1, and the reaction was allowed to proceed at 13 °C for 20 s–120 min. At the required time, 10-ll aliquots were withdrawn, quenched with 1 ll of 1 M NaOH with 2-min heating at 95 °C and neutralized with the equal molar amount of HCl. An equal volume of formamide dye solution (80% formamide, 20 mM Na-EDTA, 0.1% xylene cyanol, 0.1% bromophenol blue) was added, and the products were resolved by 20% denaturing PAGE and quantified using a Molecular Imager FX (Bio-Rad, Hercules, CA). The data were fit to a single turnover equation:

½PðtÞ ¼ ½S0 ð1  ek2 t Þ where [S]0 and [P] are the initial substrate concentration and product concentration, respectively, and t is reaction time. To determine the product release rate constant (k3), the mixture (60 ll) was supplemented with 100 nM 32P-labeled DNA substrate and 5 nM OGG1, and the reaction was allowed to proceed at 37 °C for 1–7 min. The reaction progress was followed as above save for alkaline treatment. The data were fit to a burst rate equation:

k3 ¼

m ½E0

where [E]0 is the OGG1 concentration, and v is the rate of product accumulation calculated from the linear part of the reaction progress curve. In all cases, the rate constants were calculated from at least three independent experiments using SigmaPlot v9.0 (Systat Software, Chicago, IL). Statistical difference between the calculated parameters was estimated by F test [23] using R software package.

Table 1 Oligonucleotides used in this study. Oligonucleotide ID

Sequence

Modification

CO MO CG MG

50 -GAGTGAGACXTCAAGTTGGG-30 50 -GAGTGAGAYXTCAAGTTGGG-30 50 -CCCAACTTGACGTCTCACTC-30 50 -CCCAACTTGAXGTCTCACTC-30

X = oxoGua X = oxoGua, Y = mCyt X = mCyt

R.D. Kasymov et al. / FEBS Letters 587 (2013) 3129–3134

2.3. Stimulation of OGG1 by APEX1 The reaction was performed under the burst phase conditions described above except 5 mM MgCl2 substituted for EDTA. The concentration of OGG1 was 10 nM, and that of APEX1, 100 nM. The products were analyzed as described in the previous section. 2.4. Reconstitution of base excision repair The reaction mixture (20 ll) included 20 nM DNA substrate, 50 mM Tris–HCl (pH 7.8), 10 mM MgCl2, 10 mM DTT, 1 mM ATP, 0.25 mM dGTP (or 2.5 lM dGTP spiked with 0.01 MBq of a[32P]dGTP in the [32P]dGMP incorporation experiments) or each dNTP, 100 nM OGG1, 100 nM APEX1, 100 nM POLB, and 5 cohesive end units of phage T4 DNA ligase. The reactions were carried out for 30 min at 37 °C, quenched by NaOH as described above, and analyzed by PAGE. If necessary, the reaction was terminated by heating for 15 min at 95 °C and slowly cooled to the room temperature over 2 h to allow re-annealing, then Fpg was added to 100 nM and incubated for 20 min at 37 °C to cleave all remaining oxoGua-containing duplex. 3. Results and discussion 3.1. Removal of oxoGua from methylated CpG dinucleotides by OGG1 To estimate the effect of cytosine methylation on OGG1 activity, we have used four types of oxoGua-containing substrates (Fig. 1B): one non-methylated, two containing one mCyt (termed cishemimethylated if mCyt and oxoGua are in the same strand and trans-hemimethylated otherwise), and one containing mCyt in both strands. We did not address the situation with two oxoGua bases in the CpG dinucleotide because it would appear exceedingly rarely and is probably of little biological relevance. Kinetics of OGG1-catalyzed reaction deviates from the minimal Michaelis–Menten scheme due to formation of a stable reaction intermediate [20,21]. After initial binding, the enzyme hydrolyzes the N-glycosidic bond of the damaged nucleotide, leaving an abasic (AP) site, which is then slowly cleaved by b-elimination and released (Fig. 1C; k2  k3). Although KM can be formally determined for such a scheme, it is a more complicated combination of rate constants than in the minimal scheme and provides little information on the reaction mechanism. However, in this case, k2, the rate constant of damaged base release, and k3, the rate constant of product release (principally reflecting

3131

the rate-limiting step of b-elimination), can be measured separately in single turnover and burst rate experiments, respectively. In single turnover experiments, the enzyme is in a large excess over the substrate, and the time course of base excision, revealed by alkaline treatment of the reaction products, follows exponential increase to a maximum (Fig. 2A, see also Section 2.2). In burst rate experiments, the substrate is in a large excess, and the nicked product (observed in the absence of NaOH) shows a fast initial burst and then accumulates linearly, limited by the enzyme turnover (Fig. 2B, see also Section 2.2). The amplitude of the burst equals the active enzyme concentration, providing an internal check of data consistency. From these experiments, k2 and k3 were determined (Table 2). Note that these two rate constants cannot be compared directly to each other: due to fast rate or base excision, k2 could not be reliably estimated at 37 °C (>90% cleavage at the first time point; not shown), so the single turnover experiments were performed at 13 °C. Thus, the values of k2 at 37 °C will be 5- to 10-fold higher, depending on the exact temperature coefficient of the reaction. The results show that, in general, OGG1 efficiently removes oxoGua from CpG dinucleotides in all methylation contexts; however, the presence of mCyt opposite oxoGua, in both hemimethylated and fully methylated CpG, negatively affects k3, whereas the cishemimethylated substrate is processed with a lower k2. The same tendencies were observed when 5 mM MgCl2 substituted for EDTA (Table 2), although MgCl2 had an overall inhibitory effect on the reaction [24]. The effect of opposite-base methylation may be rationalized from the structure of OGG1DNA complex [25]. When bound to OGG1, damaged DNA is kinked, and the bases adjacent to the cytosine opposite oxoGua pack against C5 atom of the cytosine that would sterically strain the C5 methyl group if mCyt were present (Supplementary Fig. 1A and B). Also, a water molecule is present near C5 in all OGG1 structures along the reaction coordinate [25–27], coordinated with adjacent protein and DNA residues and possibly forming part of the strict OGG1 specificity determinant for Cyt as the opposite base; k3 is mostly affected when other bases are present against oxoGua [28]. Thus, distortion of the OGG1DNAAP complex by the presence of a methyl group at C5 could interfere with the correct positioning of reacting groups required for b-elimination. The influence of the cis-hemimethylation on k2 is harder to explain structurally, since all solved structures of OGG1DNA complexes contain adenine in this position. In any case, changes in OGG1 activity due to the presence of mCyt are within the normal context-dependent variation [29].

Fig. 2. Determination of k2 and k3 for OGG1. (A) Time course of oxoGua excision under single turnover conditions to determine k2. [DNAAP], concentration of AP sitecontaining DNA. (B) Time course of DNA nicking under burst rate conditions to determine k3. [DNAnick], concentration of the product of b-elimination. In both plots, data for the CO//CG substrate is shown (mean ± S.D., n = 3, lines are fits to the respective equations; see Section 2.2).

3132

R.D. Kasymov et al. / FEBS Letters 587 (2013) 3129–3134 Table 2 The values of k2 and k3 for oxoG excision from CpG dinucleotides by wild-type OGG1 and OGG1 S326C. Substrate

Conditions

Wild-type OGG1 1

k2 (min CO//CG (non-methylated) MO//CG (cis-hemimethylated) CO//MG(trans-hemimethylated) MO//MG(fully methylated) a b c

EDTA Mg2+ EDTA Mg2+ EDTA Mg2+ EDTA Mg2+

OGG1 S326C 1

)

k3 (min a

1.5 ± 0.2 0.091 ± 0.008 0.79 ± 0.07a,c 0.051 ± 0.004c 1.2 ± 0.2a 0.17 ± 0.03c 1.6 ± 0.2a 0.10 ± 0.02

k2 (min1)

) a

0.64 ± 0.09 0.30 ± 0.10 0.77 ± 0.08a 0.31 ± 0.05 0.33 ± 0.03a,c 0.13 ± 0.04c 0.37 ± 0.05c 0.25 ± 0.06

k3 (min1) a,b

0.17 ± 0.02 0.05 ± 0.02b 0.09 ± 0.01a,b,c 0.015 ± 0.005b,c 0.47 ± 0.06a,b,c 0.08 ± 0.04b 0.39 ± 0.08a,b,c 0.09 ± 0.05

0.70 ± 0.09 0.67 ± 0.08b 0.54 ± 0.06b 0.66 ± 0.02b 1.00 ± 0.09a,b,c 0.68 ± 0.04b 0.55 ± 0.09 0.69 ± 0.05b

Statisticaly different from Mg2+ conditions (F-test, P < 0.05). Statistically different from wild-type (F-test, P < 0.05). Statistically different from CO//CG (F-test, P < 0.05).

Fig. 3. Stimulation of OGG1 by APEX1. The substrates were: (A) CO//CG; (B) MO//CG; (C) CO//MG; (D) MO//MG. Open circles are OGG1 alone, filled circles are OGG1 with APEX1. See Section 2.3 for the reaction conditions. Mean ± S.D. are shown (n = 4). [DNAnick], concentration of the nicked product. The ordinate has the same scale in all panels. The slope is statistically different between OGG1 with and without APEX1 (F-test, P < 0.05) only for the CO//CG substrate.

3.2. Removal of oxoGua from methylated CpG dinucleotides by OGG1 S326C OGG1 S326C is a polymorphic variant of OGG1 with a significant incidence in the human population (allele frequency from <0.1 in Western Europe and Africa to 0.5 in East Asia), which is associated with a moderately elevated risk of lung cancer, especially in people exposed to environmental stress (smoking, air pollution) [30–32]. In many functional tests, purified OGG1 S326C or extracts from OGG1 326Cys/Cys cells show impaired activity [22,33–36], although the reasons for that are not entirely clear at the moment; according to crystal structure, Ser326 is unlikely to be involved in catalysis or oxoGua recognition but may rather affect protein stability, folding, dimerization or cellular localization. To assess the possible influence of S326C mutation on the interaction of OGG1 with CpG-substrates, the purified recombinant protein was subject to the same set of kinetic experiments as wild-type OGG1, and k2 and k3 were determined (Table 2). It should be noted that our reactions were performed under reducing (1 mM DTT) conditions, which support higher activity of OGG1 S326C preventing it from dimerization through disulfide bonds between Cys326 residues [34,37]. The mutant was less active in the repair of oxoGua, with the effect mostly due to a 3- to 9-fold decrease in k2. The k3 rate constant was affected to a much lesser degree or even increased up to 3-fold compared with the wild-type OGG1. Considering the adjustment for temperature differences, this may cause k2 to become comparable with k3 for some substrates of OGG1 S326C, making substrate release not rate limiting. Interestingly, the substrates with mCyt opposite oxoGua were preferable to their Cyt counterparts by OGG1 S326C, and, unlike in the wild-type enzyme, k3 was unaffected by the opposite Cyt methylation. This observation is consistent with the reported

relaxed opposite-base specificity of OGG1 S326C [34]. Therefore, the S326C polymorphism of OGG1 is likely to affect the genomic stability of CpG dinucleotides, possibly contributing to the methylation pattern disruption seen in tumors. 3.3. Stimulation of OGG1 by APEX1 Human AP endonuclease APEX1 is known to stimulate OGG1 by increasing the turnover of the glycosylase [21,38–40]. Studies on a number of OGG1 substrates have shown that APEX1 hardly affects k2 of OGG1 but increases k3, sometimes up to a point where k3 is no longer rate-limiting [21,41]. It has been suggested that APEX1 binds 50 to the OGG1 molecule retained at the nascent AP site and directly displaces the glycosylase to expose the baseless deoxyribose to the active site of APEX1 [21]. Hence, variation of DNA structure 50 to oxoGua could affect the degree of OGG1 stimulation by APEX1. Since APEX1 is a Mg2+-dependent enzyme, we have replaced EDTA in the reaction mixture with 5 mM MgCl2 and performed the reaction under the burst phase conditions. This slowed down the turnover of OGG1 (compare Fig. 3 and Fig. 2B), consistent with the literature data [24]. However, addition of APEX1 caused a 3fold increase in the enzyme turnover rate for the non-methylated substrate, evident from the increase in the slope of the linear part of the time course curve (Fig. 3A). Both hemimethylated substrates did not show the stimulation, whereas with the fully methylated substrate, a moderate increase that was observed was not statistically significant (Fig. 3B–D). Thus, methylation of CpG decreases the stimulation of OGG1 by APEX1; since the glycosylase action is rate-limiting for oxoGua repair, this may affect the overall efficiency of the BER process in methylated CpG. Although the magnitude of reaction condition effect for CO//CG was comparable with

3133

R.D. Kasymov et al. / FEBS Letters 587 (2013) 3129–3134

OGG1 APEX1 POLB T4 ligase Fpg

– – – – +

– – – – –

+ – – – –

1

2

3 4

+ + – – –

+ + + – –

+ + + + –

– – – – +

+ + + + +

– – – – +

– – – – –

+ – – – –

+ + – – –

+ + + – –

+ + + + –

– – – – +

+ + + + +

– – – – +

– – – – –

+ – – – –

+ + – – –

+ + + – –

+ + + + –

– – – – +

+ + + + +

– – – – +

– – – – –

+ – – – –

+ + – – –

+ + + – –

+ + + + –

– – – – +

+ + + + +

S

P

% repaired

9 10 11 12 13 14 15 16

17 18 19 20 21 22 23 24

25 26 27 28 29 30 31 32

CO//CG non-methylated

5 6

7 8

MO//CG cis-hemimethylated

CO//MG trans-hemimethylated

MO//MG fully methylated

52

41

60

71

Fig. 4. Reconstitution of base excision repair of oxoGua in CpG-dinucleotides in the presence of dGTP. The arrows indicate the mobility of: S, substrate and fully repaired product (lanes with T4 DNA ligase present also show non-specific ligation products of lower mobility than the substrate); P, repair intermediates (products of DNA cleavage by OGG1, OGG1/APEX1 or products of dGMP insertion by POLB). Substrates and enzymes in the reaction mixture are indicated in the figure. % repaired, the fraction of Fpginsensitive full-length product after repair.

the effect of APEX1 stimulation, we note that APEX1 stimulates OGG1 over a variety of conditions, as reported in the literature [21,38–41]. Therefore, lack of stimulation, also observed for several OGG1 substrates of low specificity [41], may be relevant for the substrate flow through the multienzyme BER pathway. 3.4. Reconstitution of oxoGua repair in CpG-dinucleotides To address the overall BER efficiency for oxoGua in methylated CpG-dinucleotides, we reconstituted BER using purified recombinant human OGG1, APEX1, POLB, and T4 DNA ligase (Fig. 4). OGG1 alone excised the damaged base, the reaction outcome after the alkaline treatment evident as a cleavage product (Fig. 4, lanes 3, 11, 19, and 27) co-migrating with the product of cleavage of the same substrate by E. coli formamidopyrimidine-DNA glycosylase (Fpg). When APEX1 was also added to the reaction mixture, the labeled product rather unexpectedly migrated faster than the alkaline elimination product, likely due to the 30 ?50 exonucleolytic activity of APEX1, which is easily observed during BER [42,43] (Fig. 4, lanes 4, 12, 20, and 28). When APEX1 cleavage is limited to AP sites, it yields a product with a 30 -terminal hydroxyl group, which migrates slightly above the elimination band, present in the case of the non-methylated substrate but not with the methylated ones. The mechanism of the strong exonuclease activity of APEX1 and its relevance to methylation is currently under investigation. POLB inserted dGMP in place of the damaged nucleotide (Fig. 4, lanes 5, 13, 21, and 29) forming a ligatable end (Fig. 4, lanes 6, 14, 22, and 30) or fully extended the nascent primer when all four dNTPs were present (not shown). When the reactions were treated with Fpg after the repair, a significant fraction (40–70%) of the products was refractory to cleavage, indicative of the replacement of oxoGua with the normal base (Fig. 4, lanes 8, 16, 24, and 32). The least efficient repair was found for the cishemimethylated substrate (Fig. 4, lane 16), while trans-hemimethylated and fully methylated substrates were repaired somewhat better than the non-methylated one. While it is tempting to correlate the efficiency of the complete repair cycle with the kinetic constants of OGG1 on the respective substrates, it would be premature for the multienzyme system in which different enzymes both compete for the substrate and stimulate each other. As an alternative assay, we have followed incorporation of a[32P]dGTP into an otherwise unlabeled oxoGua-containing duplexes

(Supplementary Fig. 2). In all cases, we have observed incorporation of labeled [32P]dGMP by DNA polymerase, with the label going into the band of the expected size of the fully repaired product upon DNA ligase addition, with little apparent difference between the four substrates. Overall, we conclude that methylation of Cyt 50 to oxoGua may interfere with efficient repair of this lesion, partially explaining why not only C, but also G in CpG-dinucleotides is a hot spot for mutations [17]. All methylation seems to negatively affect the ability of APEX1 to stimulate OGG1. Moreover, the exonuclease activity of APEX1 may remove mCyt 50 to oxoGua, providing an intriguing possibility for DNA demethylation coupled with oxidative damage repair. It would be interesting to investigate whether the in vitro results reported here would translate into methylation effects on oxoGua repair in vivo. Acknowledgments This research was supported by the Presidium of the Russian Academy of Sciences (Molecular and Cell Biology Program 6.12), Russian Foundation for Basic Research (10-04-91058-PICS_a, 1104-00807-a, 12-04-31733-mol_a, 12-04-33231-mol_a_ved), Centre National de la Recherche Scientifique (PICS N5479-Russie); Electricité de France Contrat Radioprotection (RB 2013); Fondation de France (#2012 00029161). S.L.S. acknowledges the support from the Murdock Charitable Trust. The sponsors had no role in study design, in collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the article for publication. The authors thank Jacob Brockerman (University of British Columbia) for help with molecular graphics. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2013.08. 008. References [1] Suzuki, M.M. and Bird, A. (2008) DNA methylation landscapes: provocative insights from epigenomics. Nat. Rev. Genet. 9, 465–476. [2] Iyer, L.M., Abhiman, S. and Aravind, L. (2011) Natural history of eukaryotic DNA methylation systems. Prog. Mol. Biol. Transl. Sci. 101, 25–104.

3134

R.D. Kasymov et al. / FEBS Letters 587 (2013) 3129–3134

[3] Petrovich, M. and Veprintsev, D.B. (2009) Effects of CpG methylation on recognition of DNA by the tumour suppressor p53. J. Mol. Biol. 386, 72–80. [4] Rishi, V., Bhattacharya, P., Chatterjee, R., Rozenberg, J., Zhao, J., Glass, K., Fitzgerald, P. and Vinson, C. (2010) CpG methylation of half-CRE sequences creates C/EBPa binding sites that activate some tissue-specific genes. Proc. Natl. Acad. Sci. USA 107, 20311–20316. [5] Nan, X., Ng, H.-H., Johnson, C.A., Laherty, C.D., Turner, B.M., Eisenman, R.N. and Bird, A. (1998) Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 393, 386–389. [6] Sarraf, S.A. and Stancheva, I. (2004) Methyl-CpG binding protein MBD1 couples histone H3 methylation at lysine 9 by SETDB1 to DNA replication and chromatin assembly. Mol. Cell 15, 595–605. [7] Reik, W. (2007) Stability and flexibility of epigenetic gene regulation in mammalian development. Nature 447, 425–432. [8] Jones, P.A. and Baylin, S.B. (2007) The epigenomics of cancer. Cell 128, 683– 692. [9] Burrows, C.J. and Muller, J.G. (1998) Oxidative nucleobase modifications leading to strand scission. Chem. Rev. 98, 1109–1151. [10] Grollman, A.P. and Moriya, M. (1993) Mutagenesis by 8-oxoguanine: an enemy within. Trends Genet. 9, 246–249. [11] Zharkov, D.O. (2008) Base excision DNA repair. Cell. Mol. Life Sci. 65, 1544– 1565. [12] Svilar, D., Goellner, E.M., Almeida, K.H. and Sobol, R.W. (2011) Base excision repair and lesion-dependent subpathways for repair of oxidative DNA damage. Antioxid. Redox Signal. 14, 2491–2507. [13] ESCODD (European Standards Committee on Oxidative DNA Damage) (2002) Comparative analysis of baseline 8-oxo-7,8-dihydroguanine in mammalian cell DNA, by different methods in different laboratories: an approach to consensus. Carcinogenesis 23, 2129–2133. [14] International Human Genome Sequencing Consortium (2001) Initial sequencing and analysis of the human genome. Nature 409, 860–921. [15] Perillo, B. et al. (2008) DNA oxidation as triggered by H3K9me2 demethylation drives estrogen-induced gene expression. Science 319, 202–206. [16] Jabbari, K. and Bernardi, G. (2004) Cytosine methylation and CpG, TpG (CpA) and TpA frequencies. Gene 333, 143–149. [17] Pfeifer, G.P. (2006) Mutagenesis at methylated CpG sequences. Curr. Top. Microbiol. Immunol. 301, 259–281. [18] Drachkova, I.A., Petruseva, I.O., Safronov, I.V., Zakharenko, A.L., Shishkin, G.V., Lavrik, O.I. and Khodyreva, S.N. (2001) Reagents for modification of protein– nucleic acid complexes. II: site-specific photomodification of mammalian DNA polymerase b complexes with primers extended by dCTP exo-N-substituted arylazido derivatives Russ. J. Bioorg. Chem. 27, 173–179. [19] Gilboa, R. et al. (2002) Structure of formamidopyrimidine–DNA glycosylase covalently complexed to DNA. J. Biol. Chem. 277, 19811–19816. [20] Kuznetsov, N.A., Koval, V.V., Zharkov, D.O., Nevinsky, G.A., Douglas, K.T. and Fedorova, O.S. (2005) Kinetics of substrate recognition and cleavage by human 8-oxoguanine-DNA glycosylase. Nucleic Acids Res. 33, 3919–3931. [21] Sidorenko, V.S., Nevinsky, G.A. and Zharkov, D.O. (2007) Mechanism of interaction between human 8-oxoguanine-DNA glycosylase and AP endonuclease. DNA Repair 6, 317–328. [22] Sidorenko, V.S., Grollman, A.P., Jaruga, P., Dizdaroglu, M. and Zharkov, D.O. (2009) Substrate specificity and excision kinetics of natural polymorphic variants and phosphomimetic mutants of human 8-oxoguanine-DNA glycosylase. FEBS J. 276, 5149–5162. [23] Johnson, L.A. (1992) Analysis of variance of parameter estimates: F tests and t tests. Anal. Biochem. 206, 195–201. [24] Morland, I., Luna, L., Gustad, E., Seeberg, E. and Bjørås, M. (2005) Product inhibition and magnesium modulate the dual reaction mode of hOgg1. DNA Repair 4, 381–387. [25] Bruner, S.D., Norman, D.P.G. and Verdine, G.L. (2000) Structural basis for recognition and repair of the endogenous mutagen 8-oxoguanine in DNA. Nature 403, 859–866.

[26] Fromme, J.C., Bruner, S.D., Yang, W., Karplus, M. and Verdine, G.L. (2003) Product-assisted catalysis in base-excision DNA repair. Nat. Struct. Biol. 10, 204–211. [27] Chung, S.J. and Verdine, G.L. (2004) Structures of end products resulting from lesion processing by a DNA glycosylase/lyase. Chem. Biol. 11, 1643– 1649. [28] Zharkov, D.O., Rosenquist, T.A., Gerchman, S.E. and Grollman, A.P. (2000) Substrate specificity and reaction mechanism of murine 8-oxoguanine-DNA glycosylase. J. Biol. Chem. 275, 28607–28617. [29] Sassa, A., Beard, W.A., Prasad, R. and Wilson, S.H. (2012) DNA sequence context effects on the glycosylase activity of human 8-oxoguanine DNA glycosylase. J. Biol. Chem. 287, 36702–36710. [30] Weiss, J.M., Goode, E.L., Ladiges, W.C. and Ulrich, C.M. (2005) Polymorphic variation in hOGG1 and risk of cancer: a review of the functional and epidemiologic literature. Mol. Carcinog. 42, 127–141. [31] Hung, R.J., Hall, J., Brennan, P. and Boffetta, P. (2005) Genetic polymorphisms in the base excision repair pathway and cancer risk: a HuGE review. Am. J. Epidemiol. 162, 925–942. [32] Wei, W., He, X.-F., Qin, J.-B., Su, J., Li, S.-X., Liu, Y., Zhang, Y. and Wang, W. (2012) Association between the OGG1 Ser326Cys and APEX1 Asp148Glu polymorphisms and lung cancer risk: a meta-analysis. Mol. Biol. Rep. 39, 11249–11262. [33] Dherin, C., Radicella, J.P., Dizdaroglu, M. and Boiteux, S. (1999) Excision of oxidatively damaged DNA bases by the human a-hOgg1 protein and the polymorphic a-hOgg1(Ser326Cys) protein which is frequently found in human populations. Nucleic Acids Res. 27, 4001–4007. [34] Hill, J.W. and Evans, M.K. (2006) Dimerization and opposite base-dependent catalytic impairment of polymorphic S326C OGG1 glycosylase. Nucleic Acids Res. 34, 1620–1632. [35] Kershaw, R.M. and Hodges, N.J. (2012) Repair of oxidative DNA damage is delayed in the Ser326Cys polymorphic variant of the base excision repair protein OGG1. Mutagenesis 27, 501–510. [36] Simonelli, V. et al. (2013) Genotype–phenotype analysis of S326C OGG1 polymorphism: a risk factor for oxidative pathologies. Free Radic. Biol. Med. 63, 401–409. [37] Bravard, A., Vacher, M., Moritz, E., Vaslin, L., Hall, J., Epe, B. and Radicella, J.P. (2009) Oxidation status of human OGG1-S326C polymorphic variant determines cellular DNA repair capacity. Cancer Res. 69, 3642–3649. [38] Hill, J.W., Hazra, T.K., Izumi, T. and Mitra, S. (2001) Stimulation of human 8oxoguanine-DNA glycosylase by AP-endonuclease: potential coordination of the initial steps in base excision repair. Nucleic Acids Res. 29, 430–438. [39] Vidal, A.E., Hickson, I.D., Boiteux, S. and Radicella, J.P. (2001) Mechanism of stimulation of the DNA glycosylase activity of hOGG1 by the major human AP endonuclease: bypass of the AP lyase activity step. Nucleic Acids Res. 29, 1285–1292. [40] Saitoh, T., Shinmura, K., Yamaguchi, S., Tani, M., Seki, S., Murakami, H., Nojima, Y. and Yokota, J. (2001) Enhancement of OGG1 protein AP lyase activity by increase of APEX protein. Mutat. Res. 486, 31–40. [41] Sidorenko, V.S., Nevinsky, G.A. and Zharkov, D.O. (2008) Specificity of stimulation of human 8-oxoguanine-DNA glycosylase by AP endonuclease. Biochem. Biophys. Res. Commun. 368, 175–179. [42] Wong, D., DeMott, M.S. and Demple, B. (2003) Modulation of the 30 ?50 exonuclease activity of human apurinic endonuclease (Ape1) by its 50 -incised abasic DNA product. J. Biol. Chem. 278, 36242–36249. [43] Dyrkheeva, N.S., Lomzov, A.A., Pyshnyi, D.V., Khodyreva, S.N. and Lavrik, O.I. (2006) Efficiency of exonucleolytic action of apurinic/apyrimidinic endonuclease 1 towards matched and mismatched dNMP at the 30 terminus of different oligomeric DNA structures correlates with thermal stability of DNA duplexes. Biochim. Biophys. Acta 1764, 699–706.