Ataxia Telangiectasia Mutated and Checkpoint Kinase 2 Regulate ...

Research Article

Ataxia Telangiectasia Mutated and Checkpoint Kinase 2 Regulate BRCA1 to Promote the Fidelity of DNA End-Joining 1,2

2,4

2

2,3

Hui-Chun Wang, Wen-Cheng Chou, Sheau-Yann Shieh, and Chen-Yang Shen 1

Graduate Institute of Life Sciences, National Defense Medical Center; 2Institute of Biomedical Sciences and 3Life Science Library, Academia Sinica; and 4Institute of Public Health, National Yang-Ming University, Taipei, Taiwan

Abstract Homologous recombination (HR) and nonhomologous endjoining (NHEJ) are the two mechanisms responsible for repairing DNA double-strand breaks (DSBs) and act in either a collaborative or competitive manner in mammalian cells. DSB repaired by NHEJ may be more complicated than the simple joining of the ends of DSB, because, if nucleotides were lost, it would result in error-prone repair. This has led to the proposal that a subpathway of precise NHEJ exists that can repair DSBs with higher fidelity; this is supported by recent findings that the expression of the HR gene, BRCA1, is causally linked to in vitro and in vivo precise NHEJ activity. To further delineate this mechanism, the present study explored the connection between NHEJ and the cell-cycle checkpoint proteins, ataxia telangiectasia mutated (ATM) and checkpoint kinase 2 (Chk2), known to be involved in activating BRCA1, and tested the hypothesis that ATM and Chk2 promote precise end-joining by BRCA1. Support for this hypothesis came from the observations that (a) knockdown of ATM and Chk2 expression affected end-joining activity; (b) in BRCA1defective cells, precise end-joining activity was not restored by a BRCA1 mutant lacking the site phosphorylated by Chk2 but was restored by wild-type BRCA1 or a mutant mimicking phosphorylation by Chk2; (c) Chk2 mutants lacking kinase activity or with a mutation at a site phosphorylated by ATM had a dominant negative effect on precise end-joining in BRCA1-expressing cells. These results suggest that the other two HR regulatory proteins, ATM and Chk2, act jointly to regulate the activity of BRCA1 in controlling the fidelity of DNA end-joining by precise NHEJ. (Cancer Res 2006; 66(3): 1391-400)

Introduction Double-strand breaks (DSBs) are extremely cytotoxic DNA lesions and cells have therefore developed a wide range of responses that lead to damage repair, thus preventing cell death. In cells, DSBs are repaired by two major pathways, nonhomologous end-joining (NHEJ) and homologous recombination (HR), which differ in their requirement for a homologous DNA template during repair. HR uses an intact duplex as the template, whereas NHEJ does not require a template. HR is therefore thought to make a greater contribution than NHEJ in the S and G2-M phases, because the sister chromatid is readily available, resulting in error-free repair (1–4). On the other hand, DSB repair in the G0-G1 phase is mainly

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). Requests for reprints: Chen-Yang Shen, Institute of Biomedical Sciences, Academia Sinica, Taipei 11529, Taiwan. Phone: 886-2-2789-9036; Fax: 886-2-27823047; E-mail: [email protected]. I2006 American Association for Cancer Research. doi:10.1158/0008-5472.CAN-05-3270

www.aacrjournals.org

by NHEJ, originally thought to simply involve the direct ligation of DNA ends by a ligation complex, but which can result in errorprone repair by producing deletions or insertions. Despite this, in contrast to the predominant role of HR in Saccharomyces cerevisiae, NHEJ plays the predominant role under most conditions in mammalian cells (1, 5). This is mainly because, in mammalian cells with a large and complex genome, searching for homologous sequences for precise DSB repair by HR does not seem to be efficient. Furthermore, because a significant proportion (f70%) of the genome in mammalian cells consists of extragenic sequences that do not code for amino acids, the trade-off between the use of NHEJ to repair DSBs, with the relatively minor cost of imprecise repair, and efficient DSB repair to avoid lethality caused by unrepaired damage, becomes acceptable. However, the tumorigenic contribution of genetic variants of NHEJ genes suggests that the story is not so simple (6), and that precise NHEJ is required for maintaining genomic integrity. Thus, it is mechanistically reasonable to propose that, in addition to its basic mechanism, a subpathway of precise NHEJ exists that can repair DSB with higher fidelity and that there are additional factors that can improve the fidelity of NHEJ (7). Interestingly, BRCA1, one of the key members of HR, has been suggested to play this accessory role in NHEJ, both in vitro and in vivo (8–10). To further delineate the mechanisms underlying this precise NHEJ, the present study was carried out to examine evidence that ataxia telangiectasia mutated (ATM) and checkpoint kinase 2 (Chk2), both regulators of HR, are linked to precise NHEJ. The rationale underlying this hypothesis is that (a) ATM and Chk2 are nuclear protein kinases that have been regarded as primary regulators of HR (11–13), the function of which is critical in determining which DSB repair mechanisms are recruited to initiate DSB repair (14); (b) both ATM and Chk2 interact physically with BRCA1, resulting in BRCA1 phosphorylation, leading to activation of its repair activity (15, 16). In the present study, an in vivo plasmid-based assay that can specifically detect precise end-joining activity was used to examine this hypothesis.

Materials and Methods Cell culture. The human breast cancer cell lines, MCF-7 (BRCA1 intact) and HCC1937 (BRCA1 deficient), and the human embryonic kidney cell line, 293T (8, 17), were cultured in DMEM (Sigma-Aldrich, St, Louis, MO) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT). Plasmids and transfections. To produce mammalian cells expressing wild-type BRCA1, the full-length BRCA1 cDNA was cloned into the NotI and XhoI sites of the pcDNA3 vector (Invitrogen, Carlsbad, CA; ref. 8). To obtain BRCA1 mutant clones, Ser988, Ser1423, and Ser1524 were replaced with alanine or glutamate residues (single mutants, S988A and S988E, and the double mutant, S1423A/S1524A) using the QuickChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA). For mammalian expression of Chk2, a full-length Chk2 cDNA was cloned into the BamHI and KpnI sites of the pXJ-myc vector and expressed as a myc-tagged protein (17); mutants lacking the NH2-terminal SQ/TQ cluster domain (SCD; SCD-deleted, lacking

1391

Cancer Res 2006; 66: (3). February 1, 2006

Cancer Research residues 19-75) or forkhead-associated (FHA) domain (FHA-deleted, lacking residues 77-213), or in which Thr68 or Asp347 were replaced with alanine [mutants T68A and kinase dead (KD), respectively], or in which some or all of the serine or threonine residues in the seven SQ/TQ motifs of the SCD were mutated to alanine [mutants T68A, S33A/S35A, S19A/T26A/ S28A/S33A/S35A/S50A (indicated as 7A-T68A), S19A/T26A/S28A/S50/T68A (indicated as 7A-S33A/S35A), and S19A/T26A/S28A/S33A/S35A/S50A/T68A (indicated as 7A)] were also constructed. All mutations were generated using the QuickChange Site-Directed Mutagenesis kit (Stratagene) and sequences were confirmed by automatic DNA sequencing (ABI Prism 3700 Genetic Analyzer, Applied Biosystems, Foster City, CA). Transfection of these plasmids and clones into cells was done as described previously using LipofectAMINE 2000 (Invitrogen; ref. 8). Immunoblotting. Whole cell extracts were prepared and Western blotting was done as described previously (17, 18). The antibodies used were rabbit polyclonal anti-Chk2 antibodies and mouse monoclonal antibodies against BRCA1 or myc from Santa Cruz Biotechnology (Santa Cruz, CA), mouse monoclonal anti-ATM antibody (ATM-2C1) from GeneTex (San Antonio, TX), rabbit polyclonal antibodies against phospho-Chk2 (Thr68; 2661), phospho-NBS1 (Ser343; 3001), or phospho-Ser15-p53 or phosphoSer20-p53 (9284 and 9287, respectively) from Cell Signaling Technology (Beverly, MA), and rabbit polyclonal antibodies against p95NBS1 (PC269) or actin (A2066) from Oncogene Research Products (San Diego, CA) or SigmaAldrich (St, Louis, MO), respectively. Bound primary antibodies were detected using either horseradish peroxidase (HRP)–conjugated goat antimouse IgG antibody or HRP-conjugated anti-rabbit IgG antibody (Jackson Immunoresearch, West Grove, PA) and chemiluminescent reagents (Amersham Biosciences, Piscataway, NJ). Small interfering RNA transfection and irradiation. Transfection with small interfering RNA (siRNA) was done as described previously using OligofectAMINE (Invitrogen) according to the instructions of the manufacturer (8, 18). The cells were plated at a density of 2 105 cells/60 mm dish 1 day before transfection with ATM siRNA (Qiagen, Valencia, CA) or BRCA1 or Chk2 siRNA (8, 18) at respective final concentrations of 0.01, 0.02, or 0.1 Amol/L. Forty-eight hours later, the cells were tested for end-joining activity or were X-ray–irradiated with a single dose of 10 Gy generated by a Torrex 150D inspection system (Perkin-Elmer Life and Analytical Sciences, Boston, MA), harvested after 0.5, 1, or 4 hours, and then the proteins were analyzed by Western blotting with unirradiated cells as controls. In vivo DNA end-joining assay. The plasmid-based DNA end-joining assay was done as described previously (8, 9). In brief, plasmid pGL2-control (Promega, Madison, WI) was linearized using either HindIII, which cuts between the promoter and the luciferase cDNA, or EcoRI, which cuts within the luciferase cDNA. After confirmation of linearization by agarose gel electrophoresis, the DNA was extracted with phenol/chloroform, ethanol precipitated, dissolved in sterilized water, and stored in aliquots at 20jC. The linearized or uncut plasmid (0.5 Ag) was mixed with 0.05 Ag of an internal control plasmid, pRL-TK (Promega), and the mixture was used to transfect cells that have been previously transfected for 24 hours with Chk2 or BRCA1 clones and/or siRNAs on a 60 mm dish and reseeded into 24-well plates. At 48 hours posttransfection, firefly and renilla luciferase activities were assayed with the dual luciferase assay system (Promega) and the firefly luciferase activity was normalized to the renilla luciferase activity. Overall end-joining activity was measured by the repair of HindIII-cut pGL2 and precise end-joining activity by the repair of EcoRI-cut pGL2. Transfection with the repair substrates was done using LipofectAMINE 2000 (Invitrogen) following the protocol of the manufacturer. This plasmid-based assay was also done using as substrate the green florescent protein (GFP) vector (pEGFP-N1, Clontech, Palo Alto, CA), which was linearized using either HindIII, a restriction endonuclease cutting between the promoter and the GFP cDNA, or BsrGI, which cuts within the GFP cDNA. This linear plasmid was then transfected into cells using the LipofectAMINE 2000 reagent (Invitrogen) and the transfectants were harvested 48 hours after transfection and GFP intensity was measured using a flow cytometer. PCR-RFLP to detect end-joining activity. To characterize the repaired product of end-joining, a PCR-RFLP method was used. Plasmid pGL2 containing the luciferase reporter gene was predigested with EcoRI and


transfected into cells. If the linearized pGL2 is rejoined precisely, the EcoRI site within the luciferase reporter gene is regenerated, and the PCR product (661 bp) amplified by specific primers flanking the EcoRI site and analyzed in the linear range can then be cut by further EcoRI treatment, yielding small digestion fragments (395 and 266 bp) on agarose gels, which are resolved on 2% agarose gels containing ethidium bromide and quantified by densitometry. The EcoRI-cut fraction was taken as the fraction of precise end-joining.

Results Knockdown of ATM and Chk2 expression in the MCF-7 breast cancer cell line. To determine whether ATM and Chk2 were involved in promoting BRCA1-mediated precise NHEJ activity, a ‘‘loss-of-function’’ strategy involving knockdown of the expression of ATM, Chk2, or BRCA1 in the breast cancer cell line MCF-7 was used, and suppression of expression of the individual genes was confirmed (Fig. 1A). To determine whether the reduced expression was functionally significant, cell survival following irradiation was examined. As expected, the results of clonogenic assay showed that, compared with cells transfected with scrambled siRNA (negative control), both the ATM-defective and BRCA1-defective cells were hypersensitive to irradiation treatment (Fig. 1B). Interestingly, the Chk2-defective cells showed no evidence of radiosensitivity, a finding consistent with previous reports (19, 20). In addition, examination of the phosphorylation status of downstream targets of ATM also indicated successful shutdown of ATM function after ATM siRNA treatment. Compared with cells transfected with scrambled siRNA, in which increased phosphorylation of NBS1, Chk2, and p53 was detected as early as 30 minutes after irradiation, cells transfected with ATM siRNA showed a 90% decrease in ATM expression, resulting in blocking of irradiationinduced phosphorylation of these proteins (Fig. 1C). Although Chk2-targeting siRNA only inhibited 50% of Chk2 expression, this treatment still abolished the phosphorylation of Chk2 Thr68 and decreased the phosphorylation of p53 Ser15 (Fig. 1D), two welldefined sites known to be phosphorylated in cells in response to irradiation-induced DSB formation (21–23). These results confirmed the successful knockdown of expression of ATM, Chk2, and BRCA1 in these cells, which were then examined for NHEJ activity. ATM, Chk2, and BRCA1 affect in vivo NHEJ activity. We have previously shown that BRCA1 is involved in in vivo NHEJ (8). To determine whether ATM and Chk2 were also involved in NHEJ activity, we did a comparative in vivo end-joining assay in control MCF-7 cells and MCF-7 cells individually transfected with ATM siRNA, Chk2 siRNA, or BRCA1 siRNA. Before these assays, we confirmed that the suppression of expression was specific, as measurement of the expression of other DSB repair proteins, including NHEJ pathway proteins (Ku70, ligase IV, and XRCC4), DSB detection/checkpoint protein (NBS1), and a HR pathway protein (RAD51), showed no significant differences (data not shown). To detect end-joining activity in vivo, a HindIII- or EcoRIlinearized pGL2 plasmid containing the luciferase reporter gene was transfected into the cells, the rationale being that the luciferase gene would only be expressed after the plasmid was rejoined as the circular form, and the relative end-joining efficiency was calculated by comparing the luciferase activity in cells transfected with endonuclease-digested DNA with that in cells containing the uncut plasmid (Fig. 2A). The end-joining activity detected using HindIIIdigested pGL2 reflects overall end-joining, because this enzyme cleaves at the linker region between the promoter and the coding

1392


ATM, Chk2, and BRCA1 in NHEJ

Figure 1. siRNA specifically decreases the expression of ATM, Chk2, or BRCA1 in MCF-7 cells. A , immunoblot analysis after transfection with siRNAs targeting ATM (lane 2 ), BRCA1 (lane 3 ), or Chk2 (lane 4) or scrambled siRNA (lane 1 ; negative control). B, radiosensitivity measured by clonogenic survival assay in cells transfected with scrambled siRNA (SC ) or specific siRNAs. Points, mean of at least four experiments; bars, SE. IR, irradiation. C, immunoblot analysis showing that ATM siRNA blocks the irradiation response, as evidenced by lack of phosphorylation of the ATM downstream target proteins, NBS1, Chk2, and BRCA1. D, immunoblot analysis showing that ATM or Chk2 siRNA blocks the irradiation response, as evidenced by Chk2 and p53 phosphorylation.

sequence and any end-joining activity, even that resulting in small deletions or insertions, will regenerate luciferase expression. However, because the EcoRI site is in the luciferase sequence, only precise end-joining will restore the original luciferase, and the relative luciferase activity using this plasmid therefore reflects precise end-joining activity. Consistent with our hypothesis, both the overall and precise end-joining activity of the MCF-7 cells was decreased by transfection with ATM, Chk2, or BRCA1 siRNA (Fig. 2B). Because the in vivo overall NHEJ is the sum of the precise and nonprecise end-joining activity, we therefore decided to characterize the DSB ends joined at the DNA level. To do so, we used a PCR-RFLP–based method in which cells were transfected with plasmid predigested with EcoRI; if precise rejoining occurs, the EcoRI site is regenerated and a second EcoRI digestion of the PCR product amplified by specific primers flanking the EcoRI site of the luciferase reporter gene (Fig. 2A) results in the generation of small molecular weight digestion fragments, whereas imprecise joining results in no EcoRI site and no digestion products. Compared with cells transfected with scrambled siRNA, the proportion of digestion products decreased in cells subjected to ATM, Chk2, or BRCA1 siRNA inhibition (Fig. 2C), the proportion being lowest for Chk2 (76.5 F 7.1%), followed by ATM (79.0 F 5.4%) and then BRCA1 (84.9 F 0.6%).


To confirm these findings using a different reporter gene, we used the same strategy but with a linearized GFP vector as the repair substrate and did a similar in vivo end-joining assay in HCC1937 cells (BRCA1 deficient) and HCC1937 cells transiently transfected with full-length BRCA1 (BRCA1 restored). The result (Fig. 2D) was consistent with that using the luciferase reporter gene; the BRCA1-deficient cells regained both overall and precise end-joining activity on transfection with the full-length BRCA1 plasmid and this effect was blocked by ATM siRNA. It should be noted that, in the BRCA1-deficient HCC1937 cells, no decrease in end-joining activity, particularly precise end-joining activity, was observed when the cells were transfected with ATM siRNA, suggesting that the involvement of ATM in regulating NHEJ is BRCA1 dependent. Interestingly, treatment of cells with caffeine, which suppresses the kinase function of ATM and other phosphatidylinositol 3 kinases, had an even greater effect than ATM siRNA (Fig. 2D, right), suggesting that the kinase activity of ATM is involved in promoting BRCA1-mediated precise NHEJ activity and/or that other phosphatidylinositol-3 kinases are involved. Promotion of precise end-joining by BRCA1 involves Chk2. Given the findings that Chk2 is involved in NHEJ (Fig. 2B) and that BRCA1-mediated end-joining is associated with the kinase activity

1393


Cancer Research

of its upstream activators (Fig. 2D), we suspected that there was a link between these DSB checkpoint/repair proteins, as the function of BRCA1 is highly associated with its phosphorylation status, which can be regulated by ATM and Chk2 (15, 16). In response to irradiation and subsequent DSB formation, ATM regulates BRCA1 activity by both directly phosphorylating BRCA1 at Ser1423 and Ser1524 and indirectly phosphorylating BRCA1 at Ser988 by activating Chk2 kinase activity. We therefore used a ‘‘gain-offunction’’ strategy to restore and measure end-joining activity in BRCA1-deficient HCC1937 cells transiently transfected with fulllength BRCA1 (BRCA1 restored) or in different BRCA1 mutants with mutations at the different phosphorylation sites (S988A, S988E, or S1423A/S1524A; Fig. 3A). As shown in Fig. 3B (bottom),

precise NHEJ activity was not restored by transfection with BRCA1 S988A, lacking the Chk2-phosphorylatable Ser988, but was restored by transfection with wild-type BRCA1 or the mutant S988E, in which the Ser988 was substituted with an amino acid, Glu, mimicking phosphorylation by Chk2. The double mutant, S1423A/ S1524A, lacking sites phosphorylated by ATM, also restored the activity. Interestingly, not only wild-type BRCA1, but also all the mutants, restored overall NHEJ (Fig. 3B, top). We then calculated the percentage of restored overall end-joining activity attributed to the restored precise end-joining activity [restored precise endjoining activity / restored overall end-joining activity 100%]. As shown in Fig. 3C, as expected, the BRCA1 S988A mutant displayed the lowest value, but, equally importantly, the S988E mutant not

Figure 2. ATM, Chk2, and BRCA1 affect DNA end-joining activity. A, the plasmid, pGL2-control, containing the reporter gene (luciferase gene) and two unique restriction sites for Hin dIII and Eco RI, was transfected into different cell lines to measure overall activity and precise end-joining activity, as described in Materials and Methods. B, percentage of overall (top ) or precise (bottom ) end-joining activity in cells transfected with ATM, BRCA1 , or Chk2 siRNA compared with that in cells transfected with scrambled siRNA (scRNA ) as the control. Columns, mean of at least four replicates; bars, SE. *, P < 0.05, significant difference compared with the control. C, PCR-RFLP method for measuring precise end-joining activity. The DNA of the rejoined linearized pGL2 was amplified by specific primers flanking the Eco RI site, and the PCR products were digested with Eco RI and analyzed by agarose gel electrophoresis as described in Materials and Methods. The Eco RI-cut fraction [(b + c ) / (a + b + c ) 100%] represents the fraction of precise end-joining activity, shown below the figure; the fraction relative to that of the scrambled siRNA control (SC ) is shown in parenthesis. D, a plasmid-based assay, similar to that in (A), but using a linearized GFP vector as substrate, was used in BRCA1 -defective HCC1937 cells and the GFP-positive cells were analyzed by flow cytometry. Top left, percentages of GFP-positive, pEGFP-transfected, and nontransfected HCC1937 cells. Bottom left, percentages of GFP-positive cells in BRCA1 (+/ ) and ATM siRNA (siATM ; +/ ) HCC1937 cells transfected with linearized pEGFP. HCC1937 cells with and without transfection with full-length BRCA1 and/or ATM siRNA, or with and without transfection with full-length BRCA1 and/or treated with 4 mmol/L caffeine; the overall (top right ) or precise (bottom right ) end-joining (EJ ) activity was measured by flow cytometry. Columns, mean of four experiments; bars, SE. *, P < 0.05, significantly different compared with nontransformed noncaffeine-treated HCC1937 cells.


1394



Figure 3. Regain of end-joining activity in BRCA1-defective HCC1937 cells after transfection with BRCA1. A, immunoblot analysis of BRCA1 -defective HCC1937 cells after transient transfection with wild-type BRCA1 (WT ), BRCA1 mutants with mutations at the sites phosphorylated by Chk2 (i.e., S988A and S988E) or by ATM (S1423A/1524A), or vector (negative control). B, regain of overall (top ) and precise (bottom ) end-joining activities. C, percentage of the regained overall end-joining activity due to precise end-joining. Columns, mean of at least four experiments; bars, SE. *, P < 0.05, significantly different from the negative control (i.e., transfection with vector).

only restored the NHEJ activity, but seemed to switch NHEJ to more precise end-joining activity, which accounted for almost 100% of the restored overall end-joining activity. Because Ser988 is the site specifically phosphorylated by Chk2 (15), this suggests that Chk2, a cell-cycle checkpoint playing a role in regulating HR (11), promotes BRCA1 mediated precise NHEJ through phosphorylating BRCA1. Furthermore, the phenotype of the BRCA1 S1423A/S1524A double mutant was similar to that of wild-type BRCA1 (Fig. 3B and C), indicating that direct phosphorylation of BRCA1 by ATM is not important in promotion of end-joining by BRCA1.


The promotion of precise end-joining by Chk2 involves its kinase activity and is associated with phosphorylation of Chk2 by ATM. To confirm the findings in Fig. 3 and to further characterize the mechanisms underlying the involvement of Chk2, we used Chk2 mutants to examine the hypothesis that Chk2 can promote BRCA1-mediated precise NHEJ. Chk2 contains three functional domains: a SCD domain (residues 19-69) containing seven SQ/TQ motifs serving as the primary substrate motif for ATM, a FHA domain (residues 115-175), and a kinase domain (residues 226-486), all three of which are indispensable for the response to DNA damage (see Fig. 4A, top; ref. 24). The Ser and Thr residues in the SCD are phosphorylation targets for upstream kinases, including ATM, with Thr68 being the key phosphorylation site (21, 23). Phosphorylation of Thr68 promotes Chk2 oligomerization by enhancing the binding of the SCD domain of one Chk2 to the FHA domain of another Chk2, thereby facilitating transautophosphorylation and activation of Chk2 (25, 26). To further characterize the Chk2-mediated promotion of end-joining activity, we prepared Chk2 mutants with mutations in the individual functional domains. These clones included wild-type Chk2, a KD mutant lacking kinase activity (D347A), a FHA-deleted mutant (lacking residues 77-213), a SCD-deleted mutant (lacking residues 19-75), and several missense mutants in which some or all of the serine or threonine residues in the seven SQ/TQ motifs were mutated to alanine [mutants T68A, 7A-T68A (S33A/S35A, S19A/T26A/S28A/ S33A/S35A/S50A), 7A-S33A/S35A (S19A/T26A/S28A/S50/T68A), and 7A (S19A/T26A/S28A/S33A/S35A/S50A/T68A); Fig. 4A, top]. All these clones were myc-tagged and each was adjusted to the same expression level for subsequent experiments, as confirmed by immunoblotting (Fig. 4A, bottom). To examine the role of Chk2 in promoting BRCA1-mediated precise NHEJ, all the Chk2 mutants were individually transfected into BRCA1-expressing MCF-7 cells, in which an intact endogenous Chk2 gene is present. The Chk2-KD and the T68A mutant had no effect on overall end-joining activity (Fig. 4B, top), but both caused a significant decrease in precise end-joining activity in a dominantnegative fashion and in the percentage of the overall end-joining activity accounted for by precise end-joining activity (Fig. 4B, bottom, and C), a phenotype consistent with that seen after BRCA1 S988A transfection of the BRCA1-defective cell line, HCC1937 (Fig. 3). These findings confirm that the kinase activity of Chk2 is crucial for its function in promoting the accuracy of BRCA1-mediated NHEJ. Furthermore, loss of Chk2 kinase activity may increase the errorprone repair component of end-joining because overall end-joining was not affected, but the precise end-joining component was reduced (Fig. 4B and C), a phenotype in sharp contrast to that of BRCA1 S988E-transfected HCC1937 (Fig. 3). It is also notable that the Chk2 S33A/S35A mutant and the 7A-T68A mutant, both of which contained Thr68, had effects similar to wild-type Chk2, whereas the 7A-S33A/S35A and 7A mutants, in both of which Thr68 was mutated to alanine, had effects similar to Chk2 T68A, indicating that the phosphorylation of Chk2 Thr68 by ATM is critical for the Chk2-mediated promotion of precise end-joining activity (Fig. 4B, bottom). As expected, because the FHA and SCD domains are involved in the DNA damage response, the Chk2-FHA–deleted and Chk2-SCD–deleted mutants showed significantly decreased overall and precise end-joining activities (Fig. 4B). More investigations are required to explore how these two domains are involved in regulating NHEJ. Chk2 participates in DNA end-joining through BRCA1. In view of the finding that Chk2-mediated BRCA1 phosphorylation

1395


Cancer Research

is critical in the promotion of precise end-joining mediated by BRCA1, we further examined whether BRCA1 is a mediator of Chk2-regulated precise NHEJ in an experiment in which cells were cotransfected with wild-type or mutant forms of both genes. For this purpose, BRCA1-expressing MCF-7 cells were transfected with the BRCA1 S988A or S988E mutant alone or together with myctagged wild-type Chk2 or Chk2-KD, the expression of the BRCA1 mutants and exogenous Chk2 (i.e., myc-Chk2) being almost identical in all cells (Fig. 5A). As expected, regardless of the genotypic status of Chk2, a significantly higher activity of both overall and precise end-joining was seen in the BRCA1 S988E mutant-expressing cells than in the S988A mutant-expressing cells (Fig. 5B and C). In contrast, the dominant-negative effect displayed by the Chk2-KD mutant (shown in Fig. 4B, bottom, and C) was overcome by the BRCA1 S988E mutant; however, the dominant negative effect of the BRCA1 S988A mutant was not overcome by wild-type Chk2 (Fig. 5B and C). Although the effect of wild-type BRCA1 was not examined in the present study, this finding strongly suggests that Chk2 is involved in promoting precise end-joining activity through phosphorylation of its downstream target, BRCA1, at Ser988. Moreover, wild-type Chk2 significantly increased the overall end-joining activity in BRCA1 S988E-expressing cells (Fig. 5B). This implies that, in addition to phosphorylating BRCA1, Chk2 may participate in NHEJ by regulating other NHEJ-associated proteins. The promotion of accuracy of end-joining by BRCA1 is regulated by ATM. Given the above findings that ATM is associated with NHEJ (Fig. 2) and that Chk2 plays a role in

regulating BRCA1-mediated precise NHEJ (Figs. 3-5) and given the fact that ATM plays a sensor role in DSB repair (27), we examined the relationship between ATM, Chk2, and BRCA1 in regulating NHEJ. Because equal transfection efficiency of all these genes would be a key issue in obtaining reliable results, we chose 293T cells (an embryonic kidney cell line) in which the expression of these proteins are comparable when transfected with ATM siRNA or different BRCA1 clones (Fig. 6A). In terms of the effects of BRCA1 in 293T cells in the absence of ATM siRNA, overexpression of wild-type BRCA1, different from that observed in HCC 1937 obtaining wild-type BRCA1 (Fig. 3B, top), significantly decreased overall end-joining activity, but consistent with that observed in HCC1937 with BRCA1 (Fig. 3B, bottom), significantly increased precise end-joining activity (Fig. 6B), resulting in an increase in the percentage of overall end-joining activity due to precise end-joining activity (Fig. 6C). Under the same conditions, transfection with BRCA1 S988E, but not S988A, significantly increased precise NHEJ (Fig. 6B), which accounted for 70% of the overall end-joining activity (Fig. 6C). These findings are consistent with our previous findings using different cell lines (Figs. 3-5), again suggesting that BRCA1 can promote precise NHEJ and that the effect of BRCA1 is regulated by Chk2 phosphorylation. Knockdown of ATM expression resulted in a decrease in precise NHEJ in cells with an intact BRCA1 gene (Fig. 6B, bottom), suggesting that a decrease in ATM expression inhibits the effect of wild-type BRCA1 in promoting precise end-joining activity. However, the results also showed that a decrease in ATM expression

Figure 4. The dominant effect of different mutants of Chk2 on DNA end-joining activity. A, top, different Chk2 mutants with mutations in three functional domains of Chk2, the FHA domain, the SCD domain containing seven SQ/TQ motifs serving as the primary substrate motif for ATM, and the kinase domain, were generated and used to transfect MCF-7 cells. Bottom, immunoblot analysis of transfected myc-tagged Chk2. B, in vivo overall (top ) or precise (bottom ) end-joining activity. C, percentage of the overall end-joining activity due to precise end-joining. Columns, mean of at least four experiments; bars, SE. *, P < 0.05, significantly different compared with the negative control (i.e., transfection with vector).


1396



NHEJ (Fig. 6B and C), suggesting the existence of other mechanisms in promoting precise NHEJ, in addition to the ATMChk2-BRCA1 pathway examined in the present study. This is supported by the findings that reduced expression of ATM alone did not reduce the irradiation-induced phosphorylation of Chk2 Thr68 and P53 Ser20 in 293T cells (Fig. 6D). Moreover, transfection with either ATM siRNA or Chk2 siRNA alone had only a minor, or no, inhibitory effect on irradiation-induced P53 Ser15 phosphorylation, a result in sharp contrast with that found in cells in which expression of both ATM and Chk2 was turned off (Fig. 6D).

Discussion

Figure 5. Chk2 promotes DNA end-joining activity by phosphorylating Ser988 of BRCA1. A, MCF-7 cells were cotransfected with the BRCA1 S988A or S988E mutant and vector or myc-tagged wild-type or KD Chk2 ; the immunoblot shows similar amount of protein expression after 48 hours of transfection. B and C, overall (B ) or precise (C ) end-joining activity of the cells using MCF-7 cells transfected with vector and the BRCA1 S988A mutant as the reference (i.e., 100%). Columns, mean of at least four experiments; bars, SE.

had no effect on precise end-joining activity in BRCA1 S988Aexpressing cells (Fig. 6B, bottom). These findings suggest that ATM participates in regulating the promotion of precise end-joining mainly via the Chk2-BRCA1 pathway. However, unexpectedly, ATM siRNA also prevented the effect of the BRCA1 S988E mutant in promoting precise end-joining activity (Fig. 6B and C), suggesting that ATM participates in BRCA1-dependent precise NHEJ by regulating not only Chk2 phosphorylation, but also other undefined proteins. In fact, ATM siRNA affected overall NHEJ in BRCA1 S988E-expressing cells (Fig. 6B, top), consistent with the idea that the BRCA1-mediated promotion of end-joining requires other undefined proteins, the activation of which is dependent on ATM kinase activity. Furthermore, in the absence of ATM siRNA, the BRCA1 S1423A/S1524A double mutant showed a dominant negative effect causing a significant decrease in endjoining activity in 293T cells (Fig. 6B), suggesting a direct role of ATM in regulating end-joining via BRCA1, but this suggestion is different from that based on observation in HCC1937 (Fig. 3B). These findings suggest that, in different cells, the requirement for, and dependency on, ATM in directly regulating NHEJ may be not totally the same. In cells in which ATM expression had been turned off, wild-type BRCA1 or the BRCA1 S988A mutant significantly increased precise


Previous phenotypic studies on the role of BRCA1 in the regulation of NHEJ processes have yielded conflicting observations ranging from a promotion of NHEJ to suppressive effects to no effect (8–11, 28–35). The lack of consistency between these results may be, at least partly, explained by the use of different cell lines and different assays. The assays used include in vivo end-joining (8, 9, 28, 29), in vitro end-joining (8, 10, 28, 33), NHEJ-mediated random chromosomal integration (11, 31, 32), irradiation-induced DSB repair (30, 34), NHEJ-mediated retroviral infection (9), and NHEJ-mediated T-cell receptor V(D)J recombination (35), which may measure different subtypes of NHEJ (36). In the present study, we used three cell lines with two different tissue origins to confirm our findings. Interestingly, another study (37) using a different NHEJ assay reports findings consistent with our own, validating the conclusion that Chk2-mediated phosphorylation of BRCA1 regulates the fidelity of DNA end-joining by promoting precise endjoining. The question of how HR and NHEJ are coordinated to repair DNA DSBs is of particular interest. It has been proposed that the two pathways act in competition with each other by suppressing the expression or function of the proteins of the other pathway (38–40). A recent study showing that the two pathways depend on different DSB sensor proteins to recognize the presence of DNA DSB, then trigger the activity of different protein kinases, further activating a series of different downstream effectors (14), seems to favor this ‘‘competition’’ possibility. However, the possibility that the two pathways act in concert to repair DSBs cannot be totally excluded. In animal model studies, mice with defects in both HR and NHEJ display synthetic phenotypes of viability and tumorigenic potential and a synergistic effect of these genes on genomic stability, suggesting that these two repair pathways cooperate in DSB repair (41, 42). Our findings provide clues to resolve these inconsistencies and we propose the possibility that more than one NHEJ mechanism exists (36, 43) and that HR regulators, including ATM, Chk2, and BRCA1 (11–13, 31, 44), are required in some of these. As shown in the present study, the dependence of overall NHEJ and precise NHEJ on ATM, Chk2, and BRCA1 is condition specific, and these proteins have different effects on precise or overall end-joining activities. In particular, ATM-activated, Chk2-phosphorylated BRCA1 was shown to be important in promoting precise end-joining activity, a finding consistent with the notion that the HR-regulatory proteins, ATM and Chk2, act jointly to regulate the activity of BRCA1 in controlling the fidelity of DNA end-joining by precise NHEJ. Furthermore, it is notable that the DNA DSB ends generated by restriction enzyme digestion in the present study are only one of many forms of DSB ends formed in vivo, and it is likely that different forms of DSB ends may trigger

1397


Cancer Research

Figure 6. Reduced expression of ATM affects BRCA1-mediated promotion of end-joining activity in 293T cells. A, immunoblot analysis showing expression of endogenous ATM and overexpression of BRCA1 after cotransfection with ATM siRNA or scrambled siRNA ( , negative control) and/or wild-type BRCA1 , different BRCA1 mutants with mutations at sites phosphorylated by Chk2 (i.e., S988A and S988E) or by ATM (S1423A/1524A), or vector alone ( ). B, overall (top ) or precise (bottom) end-joining activity of the cells. C, percentage of the overall end-joining activity due to precise end-joining. Columns, mean of at least four experiments; bars, SE. #, P < 0.05, significantly different compared with the negative control (i.e., transfection with vector); *, significantly different compared with cells transfected with scrambled siRNA. D, immunoblot analysis of the irradiation response, indicated by phosphorylation of specific amino acids of the known downstream target proteins, NBS1, Chk2, and p53, in 293T cells transfected with the indicated siRNAs or with scrambled siRNA as the negative control.

different types of end processing involving different DSB repair proteins and subpathways. Our findings that BRCA1 is involved in both overall and precise end-joining activities and that phosphorylation of BRCA1 by the ATM-Chk2 pathway specifically promotes precise endjoining are of particular importance. These findings yield crucial insights into possible mechanisms underlying the interactions between these HR regulators and NHEJ. To provide a possible explanation, we will consider the contribution of the MRE11RAD50-p95/NBS1 (MRN) complex. When cells are subjected to ionizing radiation, wild-type BRCA1 binds to the MRN complex (45), which, in particular p95/NBS1, was recently shown to play a crucial role in DNA DSB repair by recruiting ATM to DSB sites, leading to activation of this DNA repair network (14, 46, 47). In addition, the MRN complex is involved in the initial processing of DSBs due to its nuclease activity and DNA binding capability (48, 49). These activities reside in the Mre11 protein, which is required to trim the ends of DSBs, thus facilitating DSB repair, especially HR, by producing the overhanging DSB ends required for


efficient HR (3, 4). However, as BRCA1 has been shown to inhibit the nuclease activity of Mre11 (50), and Chk2 phosphorylates BRCA1, leading to inhibition of MRN complex foci formation (11), it is possible that this inhibition is required for precise end-joining, because exonuclease digestion would result in the loss of nucleotides, leading to imprecise NHEJ if these digested ends were rejoined directly (a possible model is shown in Supplementary Fig. S1). Interestingly, another study (37), showing that error-prone rejoining is characterized by an increase in the formation of >2 kb deletions in DSB repair by NHEJ in BRCA1-deficient cells and that this error-prone repair phenotype is linked to Chk2 phosphorylation, lends critical support to this model. Furthermore, our epidemiologic observations on the same group of study subjects have shown that genotypic polymorphisms of the NHEJ, HR, and MRN genes are associated with breast cancer development (6, 51),5 5 J.C. Yu, et al. Breast cancer risk associated with genotypic polymorphism of the genes encoding the DNA double-strand-break repair complex, Mre11/Rad50/Nbs1: a multigenic study on cancer susceptibility. Submitted for publication.

1398



suggesting that functional links between these DSB checkpoint/ repair genes and pathways may jointly determine susceptibility of tumor formation. Through its involvement in DNA DSB repair, the NHEJ pathway plays an important role in maintaining genomic stability. However, there is currently no evidence for an association between familial cancer syndromes and the NHEJ pathway. We propose a ‘‘hide-then-hit’’ hypothesis, suggesting that, because this pathway is so crucial for mammalian cells to maintain genomic stability, any severe defects in it would result in serious outcomes, such as genomic instability and cell death, and block of subsequent cell outgrowth and tumor formation (6). Thus, only subtle defects arising from low-penetrance alleles would escape lethality and accumulate the essential genetic changes and be associated with cancer formation. Support for this hypothesis of a role of the NHEJ pathway in cancer development comes from our recent study (8), based on comprehensive NHEJ gene profiling, which showed a novel association between certain genetic polymorphisms of NHEJ genes and increased breast cancer risk. Interestingly, the breast cancer risk associated with genetic polymorphisms of the NHEJ genes was found to be modified by the BRCA1 genotype. Providing mechanistic support for this observation, we have shown that BRCA1 expression is

References 1. Jeggo PA. DNA breakage and repair. Adv Genet 1998; 38:185–218. 2. Johnson RD, Jasin M. Sister chromatid gene conversion is a prominent double-strand break repair pathway in mammalian cells. EMBO J 2000;19:3398–407. 3. van G, Hoeijmakers JH, Kanaar R. Chromosomal stability and the DNA double-stranded break connection. Nat Rev Genet 2001;2:196–206. 4. Khanna KK, Jackson SP. DNA double-strand breaks: signaling, repair and the cancer connection. Nat Genet 2001;27:247–54. 5. Hefferin ML, Tomkinson AE. Mechanism of DNA double-strand break repair by non-homologous end joining. DNA Repair 2005;4:639–48. 6. Fu YP, Yu JC, Cheng TC, et al. Breast cancer risk associated with genotypic polymorphism of the nonhomologous end-joining genes: a multigenic study on cancer susceptibility. Cancer Res 2003;63:2440–6. 7. Durant ST, Nickoloff JA. Good timing in the cell cycle for precise DNA repair by BRCA1. Cell Cycle 2005;4: 1216–22. 8. Bau DT, Fu YP, Chen ST, et al. Breast cancer risk and the DNA double-strand break end-joining capacity of nonhomologous end-joining genes are affected by BRCA1. Cancer Res 2004;64:5013–9. 9. Zhong Q, Chen CF, Chen PL, Lee WH. BRCA1 facilitates microhomology-mediated end joining of DNA double strand breaks. J Biol Chem 2002;277: 28641–7. 10. Zhong Q, Boyer TG, Chen PL, Lee WH. Deficient nonhomologous end-joining activity in cell-free extracts from Brca1-null fibroblasts. Cancer Res 2002;62:3966–70. 11. Zhang J, Willers H, Feng Z, et al. Chk2 phosphorylation of BRCA1 regulates DNA double-strand break repair. Mol Cell Biol 2004;24:708–18. 12. Morrison C, Sonoda E, Takao N, Shinohara A, Yamamoto K, Takeda S. The controlling role of ATM in homologous recombinational repair of DNA damage. EMBO J 2000;19:463–71. 13. Golding SE, Rosenberg E, Khalil A, et al. Double strand break repair by homologous recombination is regulated by cell cycle-independent signaling via ATM in human glioma cells. J Biol Chem 2004;279: 15402–10.


also causally linked to in vitro and in vivo precise end-joining activity, which are higher in BRCA1-expressing cells than in cells with defective BRCA1 expression (8). Because BRCA1 is a welldocumented breast cancer gene, this association between NHEJ and BRCA1 provides essential support for the role of NHEJ in breast cancer. An important concept underlying this finding is that, as our understanding of tumorigenesis is extended from single-gene mechanisms to multigenic or to etiologic pathwaywide networks, the consideration of whether there is a causal link between a putative cancer-associated gene and tumor development might be extended to whole tumorigenic networks (8). Given this revised concept, it is becoming increasingly important to understand the functional interactions between NHEJ activity and ATM, Chk2, and BRCA1 identified in the present study. Because ATM and Chk2 are also known cancer susceptibility genes, these identified interactions are critical evidence for a tumorigenic role of the NHEJ genes.

Acknowledgments Received 9/13/2005; revised 11/16/2005; accepted 12/1/2005. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

14. Falck J, Coates J, Jackson SP. Conserved modes of recruitment of ATM, ATR and DNA-PKcs to sites of DNA damage. Nature 2005;434:605–11. 15. Lee JS, Collins KM, Brown AL, Lee CH, Chung JH. hCds1-mediated phosphorylation of BRCA1 regulates the DNA damage response. Nature 2000;404: 201–4. 16. Cortez D, Wang Y, Qin J, Elledge SJ. Requirement of ATM-dependent phosphorylation of brca1 in the DNA damage response to double-strand breaks. Science 1999; 286:1162–6. 17. Wei JH, Chou YF, Ou YH, et al. TTK/hMps1 participates in the regulation of DNA damage checkpoint response by phosphorylating CHK2 on threonine 68. J Biol Chem 2005;280:7748–57. 18. Ou YH, Chung PH, Sun TP, Shieh SY. p53 C-terminal phosphorylation by CHK1 and CHK2 participates in the regulation of DNA-damage-induced C-terminal acetylation. Mol Biol Cell 2005;16:1684–95. 19. Hirao A, Cheung A, Duncan G, et al. Chk2 is a tumor suppressor that regulates apoptosis in both an ataxia telangiectasia mutated (ATM)-dependent and an ATM-independent manner. Mol Cell Biol 2002;22: 6521–32. 20. Takai H, Naka K, Okada Y, et al. Chk2-deficient mice exhibit radioresistance and defective p53-mediated transcription. EMBO J 2002;21:5195–205. 21. Matsuoka S, Rotman G, Ogawa A, Shiloh Y, Tamai K, Elledge SJ. Ataxia telangiectasia-mutated phosphorylates Chk2 in vivo and in vitro . Proc Natl Acad Sci U S A 2000;97:10389–94. 22. Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E, Kastan MB. DNA damage induces phosphorylation of the amino terminus of p53. Genes Dev 1997; 11:3471–81. 23. Ward IM, Wu X, Chen J. Threonine 68 of Chk2 is phosphorylated at sites of DNA strand breaks. J Biol Chem 2001;276:47755–8. 24. Ahn J, Urist M, Prives C. The Chk2 protein kinase. DNA Repair 2004;3:1039–47. 25. Lee CH, Chung JH. The hCds1 (Chk2)-FHA domain is essential for a chain of phosphorylation events on hCds1 that is induced by ionizing radiation. J Biol Chem 2001;276:30537–41. 26. Melchionna R, Chen XB, Blasina A, McGowan CH. Threonine 68 is required for radiation-induced phos-

phorylation and activation of Cds1. Nat Cell Biol 2000;2: 762–5. 27. Shiloh Y. ATM and related protein kinases: safeguarding genome integrity. Nat Rev Cancer 2003;3: 155–68. 28. Baldeyron C, Jacquemin E, Smith J, et al. A single mutated BRCA1 allele leads to impaired fidelity of double strand break end-joining. Oncogene 2002;21: 1401–10. 29. Coupier I, Baldeyron C, Rousseau A, et al. Fidelity of DNA double-strand break repair in heterozygous cell lines harbouring BRCA1 missense mutations. Oncogene 2004;23:914–9. 30. Scully R, Ganesan S, Vlasakova K, Chen J, Socolovsky M, Livingston DM. Genetic analysis of BRCA1 function in a defined tumor cell line. Mol Cell 1999;4:1093–9. 31. Moynahan ME, Chiu JW, Koller BH, Jasin M. Brca1 controls homology-directed DNA repair. Mol Cell 1999;4: 511–8. 32. Snouwaert JN, Gowen LC, Latour AM, et al. BRCA1 deficient embryonic stem cells display a decreased homologous recombination frequency and an increased frequency of non-homologous recombination that is corrected by expression of a brca1 transgene. Oncogene 1999;18:7900–7. 33. Merel P, Prieur A, Pfeiffer P, Delattre O. Absence of major defects in non-homologous DNA end joining in human breast cancer cell lines. Oncogene 2002;21: 5654–9. 34. Wang H, Zeng ZC, Bui TA, et al. Nonhomologous end-joining of ionizing radiation-induced DNA doublestranded breaks in human tumor cells deficient in BRCA1 or BRCA2. Cancer Res 2001;61:270–7. 35. Mak TW, Hakem A, McPherson JP, et al. Brcal required for T cell lineage development but not TCR loci rearrangement. Nat Immunol 2000;1:77–82. 36. Bau DT, Mau YC, Shen CY. The role of BRCA1 in nonhomologous end-joining. Cancer Lett 2005;[Epub ahead of print]. 37. Zhuang J, Zhang J, Willers H, et al. Checkpoint kinase 2-mediated phosphorylation of BRCA1 regulates the fidelity of nonhomologous end-joining. Cancer Res 2006; 66:1401–8. 38. Lau A, Kanaar R, Jackson SP, O’Connor MJ. Suppression of retroviral infection by the RAD52 DNA repair protein. EMBO J 2004;23:3421–9.

1399


Cancer Research 39. Allen C, Kurimasa A, Brenneman MA, Chen DJ, Nickoloff JA. DNA-dependent protein kinase suppresses double-strand break-induced and spontaneous homologous recombination. Proc Natl Acad Sci U S A 2002;99: 3758–63. 40. Allen C, Halbrook J, Nickoloff JA. Interactive competition between homologous recombination and non-homologous end joining. Mol Cancer Res 2003;1: 913–20. 41. Mills KD, Ferguson DO, Essers J, Eckersdorff M, Kanaar R, Alt FW. Rad54 and DNA Ligase IV cooperate to maintain mammalian chromatid stability. Genes Dev 2004;18:1283–92. 42. Couedel C, Mills KD, Barchi M, et al. Collaboration of homologous recombination and nonhomologous end-

joining factors for the survival and integrity of mice and cells. Genes Dev 2004;18:1293–304. 43. Zhang J, Powell SN. The role of the BRCA1 tumor suppressor in DNA double-strand break repair. Mol Cancer Res 2005;3:531–9. 44. Li S, Ting NS, Zheng L, et al. Functional link of BRCA1 and ataxia telangiectasia gene product in DNA damage response. Nature 2000;406:210–5. 45. Zhong Q, Chen CF, Li S, et al. Association of BRCA1 with the hRad50-hMre11-p95 complex and the DNA damage response. Science 1999;285:747–50. 46. Lee JH, Paull TT. ATM activation by DNA doublestrand breaks through the Mre11-Rad50-Nbs1 complex. Science 2005;308:551–4. 47. Paull TT, Lee JH. The Mre11/Rad50/Nbs1 complex


1400

and its role as a DNA double-strand break sensor for ATM. Cell Cycle 2005;4:737–40. 48. Paull TT, Gellert M. The 3V to 5V exonuclease activity of Mre 11 facilitates repair of DNA double-strand breaks. Mol Cell 1998;1:969–79. 49. Arthur LM, Gustausson K, Hopfner KP, et al. Structural and functional analysis of Mre11-3. Nucleic Acids Res 2004;32:1886–93. 50. Paull TT, Cortez D, Bowers B, Elledge SJ, Gellert M. Direct DNA binding by Brca1. Proc Natl Acad Sci U S A 2001;98:6086–91. 51. Cheng TC, Chen ST, Huang CS, et al. Breast cancer risk associated with genotype polymorphism of the catechol estrogen-metabolizing genes: a multigenic study on cancer susceptibility. Int J Cancer 2005;113:345–53.
