Chk2-dependent phosphorylation of XRCC1 in the DNA damage response promotes base excision repair
2008; Springer Nature; Volume: 27; Issue: 23 Linguagem: Inglês
10.1038/emboj.2008.229
ISSN1460-2075
AutoresWen‐Cheng Chou, Hui‐Chun Wang, Fen-Hwa Wong, Shian-ling Ding, Pei-Ei Wu, Sheau-Yann Shieh, Chen‐Yang Shen,
Tópico(s)Cancer-related Molecular Pathways
ResumoArticle30 October 2008free access Chk2-dependent phosphorylation of XRCC1 in the DNA damage response promotes base excision repair Wen-Cheng Chou Wen-Cheng Chou Institute of Public Health, National Yang-Ming University, Taipei, Taiwan Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Search for more papers by this author Hui-Chun Wang Hui-Chun Wang Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Graduate Institute of Natural Products, School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan Search for more papers by this author Fen-Hwa Wong Fen-Hwa Wong Institute of Public Health, National Yang-Ming University, Taipei, Taiwan Search for more papers by this author Shian-ling Ding Shian-ling Ding Department of Nursing, Kang-Ning Junior College of Medical Care and Management, Taipei, Taiwan Search for more papers by this author Pei-Ei Wu Pei-Ei Wu Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Section of Medical Genetics, Taiwan Biobank, Academia Sinica, Taipei, Taiwan Search for more papers by this author Sheau-Yann Shieh Corresponding Author Sheau-Yann Shieh Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Search for more papers by this author Chen-Yang Shen Corresponding Author Chen-Yang Shen Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Section of Medical Genetics, Taiwan Biobank, Academia Sinica, Taipei, Taiwan Life Science Library, Academia Sinica, Taipei, Taiwan Graduate Institute of Environmental Science, China Medical University, Taichung, Taiwan Search for more papers by this author Wen-Cheng Chou Wen-Cheng Chou Institute of Public Health, National Yang-Ming University, Taipei, Taiwan Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Search for more papers by this author Hui-Chun Wang Hui-Chun Wang Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Graduate Institute of Natural Products, School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan Search for more papers by this author Fen-Hwa Wong Fen-Hwa Wong Institute of Public Health, National Yang-Ming University, Taipei, Taiwan Search for more papers by this author Shian-ling Ding Shian-ling Ding Department of Nursing, Kang-Ning Junior College of Medical Care and Management, Taipei, Taiwan Search for more papers by this author Pei-Ei Wu Pei-Ei Wu Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Section of Medical Genetics, Taiwan Biobank, Academia Sinica, Taipei, Taiwan Search for more papers by this author Sheau-Yann Shieh Corresponding Author Sheau-Yann Shieh Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Search for more papers by this author Chen-Yang Shen Corresponding Author Chen-Yang Shen Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan Section of Medical Genetics, Taiwan Biobank, Academia Sinica, Taipei, Taiwan Life Science Library, Academia Sinica, Taipei, Taiwan Graduate Institute of Environmental Science, China Medical University, Taichung, Taiwan Search for more papers by this author Author Information Wen-Cheng Chou1,2, Hui-Chun Wang2,3, Fen-Hwa Wong1, Shian-ling Ding4, Pei-Ei Wu2,5, Sheau-Yann Shieh 2 and Chen-Yang Shen 2,5,6,7 1Institute of Public Health, National Yang-Ming University, Taipei, Taiwan 2Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan 3Graduate Institute of Natural Products, School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan 4Department of Nursing, Kang-Ning Junior College of Medical Care and Management, Taipei, Taiwan 5Section of Medical Genetics, Taiwan Biobank, Academia Sinica, Taipei, Taiwan 6Life Science Library, Academia Sinica, Taipei, Taiwan 7Graduate Institute of Environmental Science, China Medical University, Taichung, Taiwan *Corresponding authors: Institute of Biomedical Sciences, Academia Sinica, 128 Section 2, Academia Road, Taipei 11529, Taiwan. Tel.: +886 227 899 036; Fax: +886 227 823 047; E-mail: [email protected] or E-mail: [email protected] The EMBO Journal (2008)27:3140-3150https://doi.org/10.1038/emboj.2008.229 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The DNA damage response (DDR) has an essential function in maintaining genomic stability. Ataxia telangiectasia-mutated (ATM)-checkpoint kinase 2 (Chk2) and ATM- and Rad3-related (ATR)-Chk1, triggered, respectively, by DNA double-strand breaks and blocked replication forks, are two major DDRs processing structurally complicated DNA damage. In contrast, damage repaired by base excision repair (BER) is structurally simple, but whether, and how, the DDR is involved in repairing this damage is unclear. Here, we demonstrated that ATM-Chk2 was activated in the early response to oxidative and alkylation damage, known to be repaired by BER. Furthermore, Chk2 formed a complex with XRCC1, the BER scaffold protein, and phosphorylated XRCC1 in vivo and in vitro at Thr284. A mutated XRCC1 lacking Thr284 phosphorylation was linked to increased accumulation of unrepaired BER intermediate, reduced DNA repair capacity, and higher sensitivity to alkylation damage. In addition, a phosphorylation-mimic form of XRCC1 showed increased interaction with glycosylases, but not other BER proteins. Our results are consistent with the phosphorylation of XRCC1 by ATM-Chk2 facilitating recruitment of downstream BER proteins to the initial damage recognition/excision step to promote BER. Introduction The DNA damage response (DDR), that is, the response of the cell to genetic injury, is essential for maintaining the integrity of the genome (Hoeijmakers, 2001; Friedberg et al, 2006). Failure of the DDR results in genomic instability and a predisposition to malignancy (Hoeijmakers, 2001). The DDR is carried out by DNA damage signalling mechanisms that are triggered and precisely regulated by checkpoint proteins (Kastan and Bartek, 2004; Sancar et al, 2004). DDR targeting of repair proteins to lesions for proper DNA repair is equally critical. Structurally complicated DNA damage, including double-strand breaks (DSBs), stalled replication forks, and bulky lesions, activates the checkpoint mechanisms regulated by p53 and the kinases ataxia telangiectasia-mutated (ATM) and ATM- and Rad3-related (ATR) (Kennedy and D'Andrea, 2005; Stiff et al, 2006). Furthermore, there is mounting evidence that the proteins involved in repairing DNA DSBs and interstrand cross-link damage are regulated and targeted to damaged sites (Petrini, 2007; Wang, 2007). In contrast, the damage repaired by the base excision repair (BER) pathway is structurally simple and relatively small (Friedberg et al, 2006; Caldecott, 2007; Fortini and Dogliotti, 2007), and the nature of the DDR that senses this damage and leads to BER activation is still unclear. The BER pathway consists of a number of coordinated sequential steps that detect and process the damage; the six core steps are (i) base removal, (ii) strand incision, (iii) incised strand processing to enable DNA synthesis, (iv) DNA synthesis to fill the gap, (v) flap removal, and (vi) ligation. In contrast to the other DNA repair pathways, BER depends on specific glycosylases that recognize and process different forms of DNA damage in the first step, initiating BER by releasing the damaged base (Sancar et al, 2004; Almeida and Sobol, 2007). Some glycosylases (the bifunctional glycosylases) have an associated apurinic/apyrimidinic (AP) lyase activity and further catalyse the cleavage of the sugar-phosphate chain and the excision of the abasic residue, leaving a single nucleotide gap. This gap is filled by DNA polymerase β (polβ) and the nick is sealed by the DNA ligase III/X-ray repair cross-complementing group 1 (XRCC1) complex. Other glycosylases (the monofunctional DNA glycosylases) have no associated lyase activity. When such enzymes initiate repair, the phosphodiester bond on the 5′ side of the intact AP site is incised by AP endonuclease (APE1/APEX1). DNA polβ, DNA ligase III, and XRCC1 then complete the repair process, and the net result is the replacement of a single nucleotide (short-patch repair). In contrast, long-patch repair, which is involved in the removal of reduced abasic sites, requires further DNA synthesis, resulting in strand displacement and the generation of a damage-containing flap that is later removed by flap endonuclease. Strand displacement DNA synthesis is performed by DNA polδ/ε, and DNA ligase I restores DNA integrity (Hung et al, 2005). The BER pathway is distinguished from other DNA repair pathways by the relatively short excision patch generated in double-stranded DNA after removal of the base lesion. To explore whether, and what, DDR signalling mechanisms are required to activate BER, we investigated possible involvement of established DDR pathways, and demonstrated that checkpoint kinase 2 (Chk2), which is usually activated by ATM once ATM senses DNA DSBs (Melchionna et al, 2000; Ismail et al, 2005), may have a function in regulating BER by phosphorylating XRCC1, the scaffold protein of BER (Caldecott, 2003). To explore possible functions of phosphorylated XRCC1, we examined whether site-specific phosphorylation of XRCC1 affects BER. Because BER is the predominant repair pathway responsible for the processing of a broad spectrum of ubiquitous small base lesions caused by alkylation and oxidative damage (Hoeijmakers, 2001; Slupphaug et al, 2003; Friedberg et al, 2006), the elucidation of the mechanisms underlying BER might result in a more comprehensive understanding of DDR. Results ATM-Chk2 activation is associated with base damage To obtain initial information about the role of Chk2, we analysed checkpoint activation in human cells of different tissue origin (HeLa cells, MCF-7 cells, U2OS cells, and 293T cells) in response to different forms of insult to DNA, because Chk2 is central in transducing DDR signalling and is located downstream of ATM (Chaturvedi et al, 1999; Melchionna et al, 2000). Interestingly, as shown in Figure 1A, phosphorylation of Chk2 Thr68, a site known to be phosphorylated in cells after ionizing radiation (IR)-induced DSB formation (Ahn et al, 2000; Melchionna et al, 2000), was induced not only by DNA DSB, but also by base damage, including DNA alkylation (treatment with methyl methanesulphonate; MMS) and oxidative damage (H2O2 treatment), which are known to be repaired by BER (Sancar et al, 2004). This finding of Chk2 activation was confirmed by immunofluorescence studies using a Chk2 Thr68 phosphospecific antibody after treatment of U2OS cells with either MMS or H2O2 (Figure 1B). Chk2 activation might be linked to activation of ATM, the upstream kinase of Chk2, as a time-dependent increase in autophosphorylation of Ser1981 of ATM, a widely used biological marker to identify the active form of ATM (Bakkenist and Kastan, 2003), was also detected after MMS treatment, accompanied by Chk2 Thr68 phosphorylation (Figure 1C). This finding lends support to the idea of a role of the activated ATM-Chk2 pathway in the response to DNA insults caused by the formation of small base lesions, such as DNA alkylation by MMS. Confirming this, ATM-targeting siRNA (siATM) caused a significant decrease in MMS-induced Chk2 Thr68 phosphorylation, whereas ATR-targeting siRNA (siATR) had no effect (Figure 1D). We also examined whether TTK/hMps1 and DNA-PKcs, which have been implicated in IR- and UV-induced Chk2 Thr68 phosphorylation (Li and Stern, 2005; Wei et al, 2005), caused MMS-induced Chk2 Thr68 phosphorylation and found that was not the case (Figure 1D; Supplementary Figure S1). Figure 1.Induction of ATM and Chk2 checkpoint activation by alkylation- or oxidation-induced damage. (A) Immunoblots (IB) showing phosphorylation of ATM Ser1981 (ATM-pS1981) and Chk2 Thr68 (Chk2-pT68) following exposure of HeLa, MCF-7, U2OS, or 293T cells to methyl methanesulphonate (MMS) (alkylation damage), H2O2 (oxidative damage), or ionizing radiation (IR). (B) Phosphorylation of endogenous Chk2 Thr68 in U2OS cells after treatment with either MMS or H2O2, detected by immunofluorescence using a phosphospecific antibody (green in left panels). The right panels show DAPI-stained nuclei. (C) IB analysis showing time-dependent activation of ATM and Chk2 checkpoint, as evidenced by phosphorylation of ATM Ser1981 (ATM-pS1981) and Chk2 Thr68 (Chk2-pT68), following exposure of HeLa cells (left panel) or MCF-7 cells (right panel) to MMS. (D) IB analysis showing that siRNAs targeting ATM or Chk2, but not ATR or TTK, block MMS-induced Chk2 activation in 293T cells, as evidenced by the decreased fraction of phosphorylated Chk2 Thr68 in the total Chk2. (E) Decreased expression of the glycosylase MPG abrogates MMS-induced phosphorylation of ATM Ser1981 and Chk2 Thr68. 293T cells were transfected with siRNA targeting either MPG or XRCC1 prior to MMS treatment. Whole-cell lysates were then analysed by immunoblotting (IB) using indicated antibodies. Download figure Download PowerPoint The relationship between ATM-Chk2 and BER was further revealed by the knockdown of BER proteins. Although knockdown of XRCC1 did not significantly affect ATM-Chk2 activation by MMS, knockdown of MPG, a gene coding for a glycosylase that recognizes DNA alkylation caused by MMS and initiates BER, almost abolished the phosphorylation of ATM Ser1981 and Chk2 Thr68 (Figure 1E; Supplementary Figure S2A and B). This suggests that initial processing of base damage by glycosylases may be required for ATM-Chk2 activation. It is possible that some less common alkylation and oxidative damage, such as N3-alkyladenine, or glycosylase-processed lesions (for example, single-strand break (SSB)) may block DNA replication in S phase, resulting in the collapse of the replication fork and the production of DSBs (Brem et al, 2008), and thus activate ATM-Chk2. To exclude this possibility, we first synchronized cells in the G1 phase by nocodazole arrest and then released them (Supplementary Figure S3A). Under this condition, activation of ATM-Chk2 by MMS was still observed as that seen in IR treatment (Supplementary Figure S3B), suggesting that the MMS-induced DDR we observed could occur without DNA replication. Chk2 interacts with XRCC1 in vivo and in vitro As Chk2 regulates DSB repair through its kinase activity (Zhang et al, 2004; Wang et al, 2006; Zhuang et al, 2006), it was of interest to know whether it regulated BER by the same mechanism and which BER proteins were Chk2 substrates. We hypothesized that XRCC1 might be the BER component interacting with Chk2, the rationale being: (a) that, as Chk2 was activated by DNA alkylation and oxidative damage (Figure 1), each of which is recognized by different glycosylases in BER (Slupphaug et al, 2003), it seems unlikely that Chk2 would phosphorylate a range of glycosylases to efficiently regulate BER, and (b) XRCC1, the major BER scaffold protein, has both upstream and downstream functions (Caldecott, 2003; Marsin et al, 2003), suggesting that it may be a suitable protein to interact with the checkpoint protein. To investigate whether Chk2 kinase regulated BER function by interacting with XRCC1, we performed co-immunoprecipitation experiments with whole-cell extracts of 293T cells transfected with Myc-tagged Chk2 and His-tagged XRCC1. Using immunoprecipitation with anti-His antibodies followed by immunoblotting with anti-Myc antibodies, we found that XRCC1 was associated with Chk2 (Figure 2A). More importantly, an interaction was also detected between endogenous Chk2 and XRCC1 (Figure 2B and C; Supplementary Figure S4A). As a control, XRCC1 was not immunoprecipitated by Chk2 antibody from cells in which Chk2 expression was knocked down (Supplementary Figure S4B). The interaction was damage-inducible, as the amount of endogenous XRCC1 associated with Chk2 increased as the concentration of MMS used to treat the cells increased (Figure 2B, upper panel). This interaction between endogenous Chk2 and XRCC1 was also observed after H2O2 treatment (Figure 2C), further supporting the idea that Chk2 is associated with XRCC1 during BER. Interestingly, an IR-induced interaction between Chk2 and XRCC1 was also detected (Figure 2C). This is probably due to the fact that, in addition to DSB formation, IR induces oxidative base modification and SSBs (Friedberg et al, 2006; Jeggo and Lobrich, 2006), both of which are repaired by BER, thus providing additional support for the role of Chk2 in regulating XRCC1 in BER. Figure 2.Interaction between Chk2 and XRCC1. (A) In vivo association of Chk2 with XRCC1. 293T cells were co-transfected with Myc-tagged Chk2 and/or His-tagged XRCC1, then whole-cell extracts were subjected to immunoprecipitation (IP) with anti-His antibody and the immunoprecipitated proteins analysed by immunoblotting (IB) with anti-Myc or anti-His antibodies (upper panel). The lower panel shows IB of the whole-cell extract. (B) Association of Chk2 and XRCC1 is damage-inducible. 293T cells were either untreated or treated for 1 h with increasing concentrations of methyl methanesulphonate (MMS), then whole-cell lysates were subjected to IP with anti-Chk2 antibody and the immunoprecipitated proteins were analysed on IB using anti-XRCC1 or anti-Chk2 antibodies (upper panel). The lower panel shows IB of the whole-cell extract for Chk2, XRCC1, and phosphorylated p53 (p53-pS15), an indicator of DNA damage. (C) In vivo association of endogenous Chk2 with endogenous XRCC1 is induced by alkylation (MMS), oxidative damage (H2O2), or ionizing radiation (IR) (upper panel). Whole-cell lysates were analysed in the lower panel as in (B). (D) Constructs of GST-fused XRCC1 segments used in the in vitro binding assay. (E) Upper panel, purified Flag-tagged Chk2 produced in bacteria was incubated with purified GST–XRCC1 fragments, and the proteins pulled down by glutathione beads were analysed by IB with anti-Flag antibody. Lower panel, pull down of GST–XRCC1 fusion proteins was confirmed in a separate IB probed with anti-GST antibody. (F) The His–Chk2 constructs containing different segments used in the in vitro binding assay to detect direct interaction between Chk2 and XRCC1. (G) Upper panel, purified GST–BID produced in bacteria was incubated with the different purified His–Chk2 fragments and the GST pull-down proteins were analysed by IB using anti-His antibody. Lower panel, pull down of GST fusion proteins were confirmed in a separate IB probed with anti-GST antibody. Download figure Download PowerPoint To determine whether the co-immunoprecipitation experiments reflected a direct interaction between Chk2 and XRCC1, in vitro binding assays were performed with purified recombinant GST–XRCC1 and His–Chk2 proteins, either full length or truncated (Figure 2D and F). The results were consistent with a direct physical interaction between Chk2 and XRCC1, with the B1D segment of XRCC1 (Figure 2E) interacting with the Chk2 N-terminal domain, including the SCD motif (an N-terminal domain rich in serine or threonine residues followed by the glutamine) and the FHA (forkhead-associated) domain (Figure 2G). This finding is mechanistically interesting, given that a BRCT (BRCA1 carboxyl-terminal) domain is located within B1D of XRCC1 and has been shown to mediate many protein–protein interactions involved in DNA damage and repair (Manke et al, 2003; Yu et al, 2003). XRCC1 is a substrate of Chk2 To determine whether XRCC1 could be phosphorylated by Chk2, we performed in vitro kinase assays with His–Chk2 using GST-fused full-length or truncated XRCC1 as a substrate. Fusion proteins containing the NTL or B1D segment of XRCC1 were found to be phosphorylated by Chk2 (Figure 3A and B), suggesting that phosphorylation sites lie between amino acids 173 and 413 of XRCC1. The XRCC1 segment containing amino acids 1–196 (i.e. the NTD segment) was also phosphorylated to a lesser degree (Figure 3A and B). To precisely identify the residue(s) of XRCC1 phosphorylated by Chk2, we examined the XRCC1 sequence within the region that was most probably phosphorylated, but found no evidence of putative Chk2 phosphorylation sites (O'Neill et al, 2002). We therefore decided to map the site(s) using mass spectrometry, and, as a result, several residues in the NTL segment, including Ser184, Ser210, and Thr284, were identified (Figure 3C). Ideally, mass spectrometry of XRCC1 immunoprecipitated from extracts before and after damage would also provide convincing evidence about the site that is phosphorylated in a damage-dependent manner, in cells. Using mass spectrometry, however, we were not able to unambiguously determine the phosphorylation sites in B1D. To further localize the major sites in the NTL segment, we performed an in vitro kinase assay using various GST–NTL mutants, each lacking a different putative phosphorylation site (S184A, S210A, or T284A), as substrates for Chk2. Thr284 was identified as the major site, as the Thr284 → Ala (T284A) mutation almost abolished phosphorylation (Supplementary Figure S5). On the basis of this finding, we generated GST-fused full-length XRCC1T284A as a substrate for Chk2 in an in vitro kinase assay, and, as expected, XRCC1 Thr284 was required for XRCC1 phosphorylation by Chk2 (Figure 3D). Furthermore, to confirm the finding of XRCC1 Thr284 phosphorylation by Chk2, we generated a phosphospecific antibody against this residue, and found that XRCC1 Thr284 was phosphorylated by wild-type Chk2 (Chk2WT), but not by a catalytically inactive (i.e. kinase dead) Chk2 mutant (Chk2KD) (Figure 3E). The T284A mutant was not recognized by this phosphospecific antibody, demonstrating its specificity (Figure 3E). Figure 3.Localization of sites on XRCC1 phosphorylated by Chk2. (A) In vitro kinase assay using the different GST–XRCC1 segments and His-tagged wild-type Chk2. Top panel, in vitro Chk2-mediated phosphorylation of the different XRCC1 segments detected by autoradiography. Centre and bottom panels, input His–Chk2 or GST–XRCC1 segments were analysed by immunoblotting (IB) with anti-His or anti-GST antibodies, respectively. (B) Quantification of phosphorylation shown in the top panel of (A) after normalization to the amounts of proteins. (C) Identification of Thr284 of XRCC1 as the main site of phosphorylation by mass spectrometry. In vitro Chk2-phosphorylated GST–NTL was trypsinized and analysed by LC/MS/MS spectrometry. Three phospho-tryptic fragments of XRCC1 were identified to be derived from the tryptic peptide 281TPAT*APVPAR290 (T* represents the site of phosphorylation). The y6, y7, b3, and b4 product ions define the site of phosphorylation as Thr284. (D) In vitro kinase assay using Chk2 and either wild-type XRCC1 or the XRCC1 mutant with the Thr284 → Ala (T284A) mutation. (E) Top and third panels, in vitro kinase assay using wild-type Chk2 (Chk2WT) or a Chk2 mutant lacking kinase activity (Chk2KD) and either wild-type XRCC1 or XRCC1 T284A. Thr284 phosphorylation of XRCC1 (top panel) and autophosphorylation of Chk2 Thr68 (third panel) were detected using phosphospecific antibodies against Thr284 or Thr68. Second and fourth panels, input XRCC1 and Chk2 used in the in vitro kinase assay. Download figure Download PowerPoint XRCC1 Thr284 phosphorylation is induced by base damage To confirm the finding of XRCC1 Thr284 phosphorylation by Chk2 in vivo and, more importantly, to examine whether this is of biological relevance, we used this phosphospecific antibody to demonstrate increased phosphorylation of endogenous XRCC1 when U2OS cells were treated with H2O2 (Figure 4A). Furthermore, this H2O2-induced site-specific phosphorylation of XRCC1 became undetectable when cells were pretreated with a Chk2-specific inhibitor (Figure 4A). Consistent with these findings, Thr284 phosphorylation was detected in vivo when XRCC1-defective EM9 cells stably transfected with wild-type XRCC1 (XRCC1WT), but not the XRCC1 T284A mutant (XRCC1T284A), were subjected to MMS or H2O2 treatment (Figure 4B; Supplementary Figure S6). These results indicate that XRCC1 Thr284 phosphorylation by Chk2 is induced by either MMS or H2O2 in cells. In addition, we also performed immunoprecipitation using the anti-phospho-Thr284 XRCC1 antibody to demonstrate that XRCC1 was immunoprecipitated when cells were treated with MMS and that the amount of XRCC1 immunoprecipitated was reduced when Chk2 expression was downregulated by siRNA (Figure 4C). Figure 4.XRCC1 Thr284 phosphorylation in vivo in response to MMS or H2O2 treatment. (A) Phosphorylation of endogenous XRCC1 at Thr284 after 10 min treatment of U2OS cells (pretreated with either DMSO- or a Chk2-specific inhibitor, Chk2 inhibitor II) with 2 mM H2O2, detected by immunofluorescence (IF) using phosphospecific antibody against Thr284. (B) Phosphorylation of XRCC1 Thr284 in XRCC1-defective EM9 cells transfected with vector or wild-type XRCC1 after treatment with MMS (0.3 mg/ml) or H2O2 (2 mM), detected by IF using antibody against XRCC1 (red) or phosphospecific antibody against XRCC1 Thr284 (green). Nuclei were stained with DAPI (blue). (C) Phosphorylation of XRCC1 at Thr284 was induced by MMS in a Chk2-dependent manner. Whole-cell lysates prepared from 293T cells were subjected to IP with anti-phospho-Thr284 XRCC1 antibody and the immunoprecipitated proteins were analysed by immunoblotting (IB) using anti-XRCC1 antibody. The amount of XRCC1 immunoprecipitated was reduced when Chk2 expression was ablated by siRNA. Download figure Download PowerPoint Interaction of XRCC1 with glycosylases is enhanced by Thr284 phospho-mimicking mutation The requirement for glycosylases to activate the ATM-Chk2 checkpoint (Figure 1E) leading to XRCC1 phosphorylation (Figures 1, 2, 3 and 4) raises the question of how the glycosylases and phosphorylated XRCC1 interact with each other to mediate BER. The evidence that XRCC1 is a scaffold protein and can interact with each BER component (Caldecott, 2003; Campalans et al, 2005) suggests that phosphorylation of XRCC1 may facilitate its interaction either with glycosylases or with other BER downstream proteins and thus promote BER. In co-immunoprecipitation assays, the XRCC1 mutant mimicking phosphorylation by Chk2 (T284D) showed a higher affinity for the glycosylases MPG and UNG2 (a glycosylase that removes uracil near replication forks and in non-replicating DNA) than either wild-type XRCC1 or the XRCC1 mutant lacking the phosphorylation site (T284A) (Figure 5A and B). In contrast, the downstream BER proteins polβ and PARP1 interacted equally with wild-type XRCC1 and the T284D and T284A mutants (Figure 5C and D). Figure 5.Interaction between phosphorylated XRCC1 and glycosylase promotes the binding of XRCC1 to DNA. (A–D) Increased interaction of phosphorylation-mimicking XRCC1 mutant (T284D) with glycosylases, but not with downstream proteins of BER. In vivo association of glycosylases (MPG in (A) and UNG2 in (B)) or downstream BER proteins (polβ in (C) and PARP1 in (D)) with wild-type XRCC1 (XRCC1WT), a phosphorylation-mimicking form (XRCC1T284D), or a phosphorylation-deficient form (XRCC1T284A). The 293T cells were co-transfected with His-tagged wild-type or mutant XRCC1 and Myc-tagged MPG, Myc-tagged UNG2, Myc-tagged polβ, or HA-tagged PARP1, then whole-cell lysates were subjected to immunoprecipitation (IP) with anti-His antibody and the immunoprecipitated proteins were analysed by immunoblotting (IB) with anti-His, anti-HA, or anti-Myc antibodies. (E) Chromatin fractionation analysis showing that decreased expression of MPG and Chk2 by siRNA blocks MMS-induced binding of XRCC1 to chromatin. 293T cells were transfected with siRNA specific for MPG and Chk2, then, after treatment with 0.3 mg/ml MMS for 1 h, whole-cell lysates were prepared, separated into soluble and chromatin-enriched fractions as described in Materials and methods, and analysed by IB using the indicated antibodies. Tubulin and histone H4 were protein controls present in the soluble and chromatin fractions, respectively. (F) In vivo association of chromatin with XRCC1WT, XRCC1T284D, or XRCC1T284A showing that the XRCC1 mutant lacking the phosphorylation site (T284A) displays lower affinity for DNA after exposure to MMS. 293T cells were transfected with His-tagged wild-type (XRCC1WT) or mutant XRCC1 (XRCC1T284D or XRCC1T284A), then whole-cell lysates were separated into soluble or chromatin-enriched fractions as in (E), and analysed by IB using the indicated antibodies. Download figure Download PowerPoint Because glycosylases have a function in DNA lesion recognition/excision to initiate BER, the finding that phosphorylation of XRCC1 Thr284 enhances the interaction between XRCC1 and glycosylases suggests a mechanism in which XRCC1 phosphorylation promotes the recruitment of XRCC1 to the initial DNA lesion recognized/excised by glycosylases. In line with this notion, an MMS-induced increase in chromatin-bound endogenous XRCC1 was significantly decreased when either MPG or Chk2 expression was knocked down by siRNA (Figure 5E). Moreover, more wild-type XRCC1 and the T284D mutant were associated with chromatin than the T284A mutant after MMS treatment, thus lending further support to the model (Figure 5F). Note that XRCC1 was associated with chromatin even before MMS treatment; such association appeared to be independent of MPG and Chk2 (Figure 5E and F). XRCC1 Thr284 phosphorylation promotes BER and cell survival after base damage, and the interaction between Chk2 and XRCC1 is associated with cancer risk To demonstrate that the interaction between glycosylases and phosphorylated XRCC1 is of func
Referência(s)