Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Cellular differentiation wikipedia , lookup
Biochemical switches in the cell cycle wikipedia , lookup
Cell growth wikipedia , lookup
Signal transduction wikipedia , lookup
Homologous recombination wikipedia , lookup
DNA repair protein XRCC4 wikipedia , lookup
Oncogene (2006) 25, 5864–5874 & 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00 www.nature.com/onc REVIEW The roles of BRCA1 and BRCA2 and associated proteins in the maintenance of genomic stability K Gudmundsdottir and A Ashworth The Breakthrough Breast Cancer Research Centre, The Institute of Cancer Research, Fulham Road, London, UK The BRCA1 and BRCA2 proteins are important in maintaining genomic stability by promoting efficient and precise repair of double-strand breaks. The main role of BRCA2 appears to involve regulating the function of RAD51 in the repair by homologous recombination. BRCA1 has a broader role upstream of BRCA2, participating in various cellular processes in response to DNA damage. The DNA repair defect associated with mutations in BRCA1 or BRCA2 could be exploited to develop new targeted therapeutic approaches for cancer occurring in mutation carriers. Oncogene (2006) 25, 5864–5874. doi:10.1038/sj.onc.1209874 Keywords: BRCA1; BRCA2; repair; genomic stability; DSS1 Pathways for the repair of DNA damage DNA single-strand lesions The accurate repair of damaged DNA is essential for the maintenance of genomic integrity. The eukaryotic cell has evolved several ways of repairing the multiple different types of DNA damage that occur (Hoeijmakers, 2001). There are pathways that deal with the repair of individual damaged bases and single-strand breaks (SSBs), and distinct mechanisms that cope with the potentially more severe consequences of double-strand breaks (DSBs). Lesions affecting only one of the DNA strands, where the intact complementary strand can be used as a template for repair, are repaired by the baseexcision repair (BER), nucleotide-excision repair (NER) or mismatch repair (MMR) pathways. Although partially overlapping, these pathways show preference for specific DNA lesions. DNA double-strand breaks DNA DSBs are more problematic than SSBs since the complementary strand is not available as a template for repair. DSBs may arise as a result of either exogenous Correspondence: Professor A Ashworth, The Breakthrough Breast Cancer Research Centre, The Institute of Cancer Research, Fulham Road, London SW3 6JB, UK. E-mail: [email protected] insults, such as exposure to ionizing radiation (IR), or endogenous events, such as the collapse of replication forks upon encountering a SSB. Three DSB repair pathways have been identified within eukaryotic cells: nonhomologous end-joining (NHEJ), gene conversion (GC) and single-strand annealing (SSA). Both GC and SSA rely on sequence homology for repair while NHEJ utilizes no, or little, homology (van Gent et al., 2001; Shin et al., 2004). In NHEJ, the initial event involves the binding of the Ku70/Ku80 heterodimer to broken DNA ends (Lieber et al., 2003; Hefferin and Tomkinson, 2005). Two heterodimers then come together to bridge matching ends and recruit the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs). The kinase activity of DNA-PKcs is required for NHEJ but exactly why is not clear, although the autophosphorylation of DNA-PKcs appears to be important. A substrate of DNA-PKcs is Artemis, a nuclease that forms a complex with DNAPKcs. Together they act as an endonuclease, which can process the DNA ends preparing them for ligation by XRCC4-Ligase IV. NHEJ is the predominant mechanism for the repair of DSBs during G0, G1 and early S phases of the cell cycle, although it seems to be active in all phases of the cell cycle (Takata et al., 1998; Rothkamm et al., 2003). Traditionally, NHEJ has been thought of as an error-prone mechanism of repair, which frequently results in minor changes in the DNA sequence at the break site and occasionally in the joining of previously unlinked DNA molecules, resulting in gross chromosomal rearrangements. However, in recent years, it has become apparent that a precise subpathway of NHEJ exists in eukaryotic cells alongside the more error-prone pathway. The precise repair pathway is thought to be Ku/DNA-PKcs dependent and is associated with minimal sequence modification at the break site. The error-prone pathway of NHEJ, however, uses microhomology flanking the break site to anneal and religate the broken ends and is thought to be dependent on the MRE11-RAD50-NBS1 (MRN) complex (Ma et al., 2003; Zhang et al., 2004). GC uses a homologous sequence, preferably the sister chromatid, as a template to resynthesize the sequence surrounding the DSB and this pathway is therefore generally thought to result in the accurate repair of the break. Furthermore, crossover events are suppressed in mammalian cells as these can result in genome alterations (Johnson and Jasin, 2000). Repair by GC is Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5865 dependent upon the recombinase function of RAD51 and the assistance of various other proteins. The MRN complex is thought to play a role in bringing the two broken DNA ends together and the complex has also been suggested to be involved in the nucleolytic resection of the ends of the DSB. The resulting 30 overhangs become coated by RPA, which prevents the formation of, and removes, secondary structures. RAD51 forms a nucleoprotein filament on the ssDNA and catalyses the invasion of the homologous sequence on the sister chromatid, which is then used as a template for accurately repairing the broken ends by DNA synthesis. Recombination intermediates are then resolved, either with or without sequence exchange. Other proteins implicated in assisting this process include RAD54, the RAD51 paralogues and BRCA2 (van Gent et al., 2001; West, 2003). Although RAD52 is essential for GC in yeast, this does not seem to be the case in mammalian cells (Stark et al., 2004). This pathway is mainly active during the S and G2 phases of the cell cycle following DNA replication, when a sister chromatid is available for use as a template for repair (Takata et al., 1998; Rothkamm et al., 2003). SSA also involves the use of homologous sequences for the repair of DSBs. However, unlike GC, which involves strand invasion, SSA is RAD51 independent and involves the annealing of DNA strands formed after resection at the DSB. The proteins and mechanisms involved in SSA are still somewhat unclear (Valerie and Povirk, 2003; Stark et al., 2004). Initially, DNA ends are resected by an exonuclease, most likely the MRN complex, to yield long single-strand overhangs that RPA and RAD52 may bind to. Once homology is exposed in the overhangs, they are annealed and the protruding ends are trimmed by the ERCC1/XPF nuclease and the gap is filled by DNA polymerase. This pathway is error-prone as it results in the retention of only one of the homologous sequences and deletion of the intervening sequence. SSA is potentially an important pathway of mutagenesis as a large fraction of the genome consists of repetitive elements (Lander et al., 2001; Elliott et al., 2005). Both GC and SSA require resection of DNA ends as an initial step in the repair process and this strand resection has been shown to be cell cycle regulated in yeast, being much reduced in G1 cells (Ira et al., 2004). Therefore SSA, like GC, is thought to be active in S–G2 phases of the cell cycle (Elliott et al., 2005). BRCA1 and DSB repair One of the earliest indications that BRCA1 was involved in DNA repair was the observation that BRCA1 associates and colocalizes with RAD51 in nuclear foci in mitotic cells (Scully et al., 1997b). These foci were also observed to contain BRCA2 and the BRCA1-binding protein BARD1, both before and after DNA damage (Jin et al., 1997; Scully et al., 1997a; Chen et al., 1998). Further evidence came from the observation that BRCA1-deficient cells were highly sensitive to IR and displayed chromosomal instability, with both numerical and structural chromosome aberrations, which may be a direct consequence of unrepaired DNA damage (Shen et al., 1998; Xu et al., 1999). The similar genomic instability in BRCA1- and BRCA2-deficient cells and the interaction of both BRCA1 and BRCA2 with RAD51 suggested a functional link between the three proteins in the RAD51-mediated DNA damage repair process. However, while BRCA2 is directly involved in RAD51-mediated repair, affecting the choice between GC and SSA (see below), BRCA1 acts upstream of these pathways (Stark et al., 2004). Brca1 deficiency in mouse embryonic stem cells leads to impaired repair of DSBs by GC induced by the site-specific I-SceI endonuclease (Moynahan et al., 1999, 2001a). The involvement of BRCA1 in the repair of DSBs by GC is consistent with its association and colocalization with RAD51 in nuclear foci, and, accordingly, BRCA1 is required for their formation (Bhattacharyya et al., 2000). DSB repair by SSA is also reduced in BRCA1-deficient cells, placing BRCA1 before the branch point of GC and SSA (Stark et al., 2004). However, NHEJ has been reported to be unaffected in BRCA1-deficient cells, although data concerning the role of BRCA1 within this repair pathway has been conflicting (Moynahan et al., 1999; Baldeyron et al., 2002; Zhong et al., 2002; Stark et al., 2004). This may, at least in part, be a reflection of different roles of BRCA1 in the subpathways of NHEJ. In support of this notion, BRCA1 has been shown recently to be important in promoting precise NHEJ, while inhibiting more error-prone microhomologymediated NHEJ (Wang et al., 2006; Zhuang et al., 2006). Therefore, in addition to promoting error-free repair of DSBs by HR, BRCA1 also reduces the mutagenic potential of NHEJ, thereby contributing to genomic stability. BRCA1 is associated with other DNA repair functions and complexes BRCA1 has been associated with a variety of proteins that are involved in the response to, or the repair of, DNA damage. BRCA2 and RAD51 have been observed to coexist in a complex with BRCA1 termed BRCC (BRCA1-BRCA2-Containing Complex) that also contains BARD1 and other components (Dong et al., 2003). This complex displays an E3 ubiquitin ligase activity, which has been implicated in the regulation of factors involved in DNA repair. However, BRCA2 and RAD51 do not appear to be present in another BRCA1containing complex termed BASC (BRCA1-Associated Genome Surveillance Complex) (Wang et al., 2000). This complex includes tumour suppressors, DNA damage sensors and signal transducers, including the MRN complex, the mismatch repair proteins MSH2, MSH6 and MLH1, the Bloom syndrome helicase BLM, the ATM kinase, DNA replication factor C (RFC) and PCNA. Many of the proteins in this complex can bind Oncogene Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5866 abnormal DNA structures and, therefore, have the potential to act as sensors of DNA damage. Many also function directly in DNA replication or DNA-replication associated repair. This suggests a role for BRCA1 in coordinating various functions of DNA replication that are important for maintaining genomic integrity in the cell. The identification of the MRN complex in BASC provides further support for a role of BRCA1 in the repair of DSBs. The MRN complex is thought to be involved in the end-processing of DSBs in GC, SSA and microhomology-mediated NHEJ. BRCA1 colocalizes with the MRN complex in foci upon DNA damage and it also inhibits the nucleolytic activity of MRE11 in vitro, thereby potentially influencing the choice of repair pathway after a DSB (Zhong et al., 1999; Wang et al., 2000; Paull et al., 2001). The association of BRCA1 with MSH2 and MSH6 in the BASC complex also links BRCA1 to a subpathway of NER that preferentially repairs base lesions from the transcribed strand, as these two MSH proteins are known to be involved in this process (Wang et al., 2000). Furthermore, BRCA1 is required for the repair of oxidative 8-oxoguanine lesions by transcription-coupled DNA repair, although the role of BRCA1 in the repair of oxidative damage has been subject to re-evaluation following the retraction of a paper implicating BRCA1 within this pathway (Gowen et al., 1998; Le Page et al., 2000). BRCA1 has also been reported to enhance the global genomic repair (GGR) subpathway of NER by inducing the expression of the NER genes XPC, DDB2 and GADD45 (Harkin et al., 1999; Hartman and Ford, 2002). An association of BRCA1 with a complex that contains the proteins SW1 and SNF links BRCA1 to chromatin remodelling, which is important to facilitate access of proteins involved in DNA processing, such as transcription and repair, to DNA (Bochar et al., 2000). BRCA1 functions in signalling the response to DNA damage The initial step in the response to DNA damage involves sensing and recognizing the damaged DNA and transducing a signal onwards to downstream effectors, resulting in cell cycle arrest and DNA repair. The protein kinases ATM and ATR are central to the DNA damage response, relaying the signal onwards by phosphorylating downstream proteins, including BRCA1. The role of BRCA1 is to respond to these signals by participating in various cellular pathways, including cell cycle regulation and DNA repair (Figure 1). ATM and ATR phosphorylate BRCA1 in response to different stimuli and they appear to have both distinct and overlapping phosphorylation sites, only some of which have been characterized. ATM phosphorylates BRCA1 on several different residues in response to IR (Cortez et al., 1999; Gatei et al., 2000). BRCA1 lacking two of those phosphorylation sites, Ser1423 and Ser1524, fails to rescue the radiation hypersensitivity of a BRCA1-deficient cell line, demonstrating that Oncogene Figure 1 BRCA1 in DNA damage signalling. DNA damage activates the kinases ATM, ATR and CHK2, which in turn activate BRCA1 by phosphorylation. In response to IR-induced damage, BRCA1 is phosphorylated by ATM on several different residues, including serines 1387, 1423 and 1524 (Cortez et al., 1999; Gatei et al., 2000) and by CHK2 on serine 988 (Lee et al., 2000). ATR phosphorylates BRCA1 on serine 1423 in response to UV damage or HU-induced replication arrest (Tibbetts et al., 2000). The phosphorylation of serine 1387 is important for S phase arrest, while phosphorylation on serine 1423 is important for G2–M arrest (Xu et al., 2001a, 2002). The serine 988 phosphorylation affects the DNA repair response to DSBs (Zhang et al., 2004; Zhuang et al., 2006; Wang et al., 2006). BRCA1 participates in the G1–S checkpoint response indirectly through several ways, including by affecting CHK2 and p53 phosphorylation (Foray et al., 2003). phosphorylation of BRCA1 by ATM is critical for the appropriate response to DSBs (Cortez et al., 1999). Phosphorylation of the Ser1423 residue appears to be important for the IR-induced G2–M checkpoint but not for the IR-induced intra-S phase arrest, which requires phosphorylation of Ser1387 (Xu et al., 2001a, 2002). ATR, which is activated by UV-damage and hydroxyurea-induced replication arrest, also phosphorylates BRCA1 on several residues, including Ser1423 during the G2–M phase (Tibbetts et al., 2000; Gatei et al., 2001; Okada and Ouchi, 2003). Moreover, ATR colocalizes with BRCA1 in foci in cells synchronized in S phase and after exposure to DNA damaging agents or DNA replication inhibitors, associating BRCA1 and ATR with the response to stalled replication forks (Tibbetts et al., 2000; Gatei et al., 2001). In response to IR, ATM phosphorylates and activates another checkpoint kinase, CHK2, which in turn can phosphorylate BRCA1 on Ser988 (Lee et al., 2000). CHK2 and BRCA1 interact and colocalize within discrete nuclear foci that disperse after damage by IR. Phosphorylation on Ser988 has been shown to be required for this dispersion and the separation of the two proteins, as well as for BRCA1 to mediate survival after DNA damage in a BRCA1-mutated cell line. Recently, CHK2 phosphorylation of the Ser988 of BRCA1 was also shown to be important for the role of BRCA1 in the repair of Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5867 DSBs by promoting error-free HR and by inhibiting the error-prone microhomology-mediated subpathway of NHEJ (Zhang et al., 2004; Zhuang et al., 2006; Wang et al., 2006). The DNA damage response kinases ATM, ATR and CHK2, therefore, modulate the function of BRCA1 through phosphorylation, affecting cell cycle regulation and fidelity of DNA repair. BRCA1 is essential for activating the CHK1 kinase and it also induces the expression of the 14-3-3 proteins and the WEE1 kinase, which is another inhibitor of CDC2 activity (Yarden et al., 2002). In addition, phosphorylation of BRCA1 on Ser1423 by ATM is required to activate this checkpoint and BRCA1 has been demonstrated to induce the expression of GADD45, which activates the G2–M checkpoint by inhibiting the activity of the CDC2–CyclinB complex (Mullan et al., 2001; Xu et al., 2001a). BRCA1 and checkpoint control In response to DNA damage, mammalian cells have to arrest the cell cycle to initiate repair. Failure of cell cycle checkpoints can lead to the acquisition and accumulation of genetic alterations and chromosomal abnormalities. The role of BRCA1 in cell cycle checkpoints has recently been reviewed (Kennedy et al., 2004; Deng, 2006). Mouse embryonic fibroblasts (MEFs) derived from embryos homozygous for a targeted deletion of exon 11 of the Brca1 gene, maintain an intact G1–S checkpoint but are defective in an IR-induced G2–M checkpoint (Xu et al., 1999). Additionally, these cells exhibit centrosome amplification, resulting in abnormal chromosome segregation and aneuploidy. Having a functional G1–S checkpoint suggests a functional p53 pathway in Brca1 mutant cells, which is consistent with the finding that removal of p53 partially rescues Brca1 deficiency (Hakem et al., 1997; Ludwig et al., 1997; Shen et al., 1998). Moreover, MEFs containing a Brca1 exon 11 deletion in a p53 heterozygous or null background showed partial or complete loss of the G1–S checkpoint, respectively (Xu et al., 2001b). However, BRCA1 has been reported to stimulate the transcription of the p21 gene, which results in cell cycle arrest at the G1–S phase boundary, and is a coactivator for p53, indicating a more complex mechanism of control (Somasundaram et al., 1997; Ouchi et al., 1998; Zhang et al., 1998). BRCA1 has also been shown to be required for the ATM/ATR-mediated phosphorylation of several proteins following DNA damage, including CHK2 and p53 at Ser15, which is necessary for G1–S arrest via transcriptional induction of p21 (Foray et al., 2003; Fabbro et al., 2004). The activation of the intra-S phase checkpoint by stalled replication forks may also require BRCA1 and this response may depend on ATR phosphorylation (Tibbetts et al., 2000). After damage by IR the activation of the intra-S phase checkpoint requires phosphorylation of BRCA1 on Ser1387 by ATM, and, additionally, BRCA1 has been reported to stimulate the transcription of p27, which induces intraS phase arrest (Williamson et al., 2002; Xu et al., 2002). BRCA1 regulates the G2–M phase checkpoint at multiple levels. Establishment of the G2–M checkpoint requires phosphorylation of the CDC2 kinase, while removal of this phosphorylation by CDC25C activates the CDC2/CyclinB complex to initiate mitosis. DNA damage leads to inhibition of CDC25C activity through phosphorylation by the CHK1 kinase and nuclear exclusion of CDC25C through binding to 14-3-3s, leading to cell cycle arrest at the G2–M checkpoint. The BRCA1-binding protein BARD1 and DNA repair BRCA1 exists as a heterodimeric complex with BARD1, a structurally related protein that, like BRCA1, contains a N-terminal RING-finger domain and two C-terminal BRCT motifs (Wu et al., 1996). The association between these two proteins is mediated through their respective RING domains, and together they exhibit E3 ubiquitin ligase activity, which can be disrupted by cancerpredisposing mutations within the RING domain of BRCA1 (Hashizume et al., 2001; Ruffner et al., 2001). Furthermore, these mutations affect the ability of cells to repair damaged DNA, as measured by sensitivity to IR and to arrest at the G2–M phase of the cell cycle. This suggests that the ubiquitin ligase activity is important for BRCA1 to execute its role within the DNA damage response pathway (Ruffner et al., 2001). The interaction between BRCA1 and BARD1 is important for the stability of BRCA1 and, accordingly, the phenotype of Bard1-null mice is strikingly similar to that of Brca1-null mice, including chromosomal instability and early embryonic lethality due to severe impairment of cell proliferation, which can be partially rescued in a p53-null background (Hashizume et al., 2001; Joukov et al., 2001; McCarthy et al., 2003; Evers and Jonkers, 2006). A role for BARD1 in the DNA damage response was suggested by the observation that BARD1 colocalizes with BRCA1 in S phase cells and in cells following DNA damage (Jin et al., 1997; Scully et al., 1997a). BARD1 has also been shown to participate with BRCA1 in the homology-directed repair of DSBs (Westermark et al., 2003; Fabbro et al., 2004). A role for BARD1 in DSB repair was demonstrated using a dominant-negative truncated BARD1 peptide, which contained the N-terminal BRCA1-interacting RING domain but was missing the C-terminal BRCT repeats (Westermark et al., 2003). The peptide was, therefore, predicted to compete with endogenous full length BARD1 for binding to endogenous BRCA1, disrupting the endogenous BRCA1–BARD1 interaction. Expression of the dominant-negative BARD1 peptide resulted in a significant decrease in the homology-directed repair of a chromosome DSB, as well as in mitomycin C (MMC) hypersensitivity, without affecting the subcellular location of BRCA1. Recently, the BRCA1– BARD1 complex has been demonstrated to ubiquitinate RNA polymerase II following DNA damage resulting in Oncogene Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5868 its degradation and subsequent inhibition of transcription and RNA processing (Kleiman et al., 2005; Starita et al., 2005). This process could help in eliminating prematurely terminated transcripts that could produce truncated proteins and also to clear the damaged DNA region, creating access and possibly acting as a recruiting factor for DNA repair proteins. The function of BRCA2 in regulating RAD51-mediated recombination BRCA2 plays an important role in the repair of DSBs by HR, a pathway which is dependent upon the recombinase function of RAD51 (Moynahan et al., 2001b; Tutt et al., 2001). The direct interaction between BRCA2 and RAD51 and their colocalization in nuclear foci after DNA damage was the first evidence for a role for BRCA2 within this DNA repair pathway (Sharan et al., 1997; Chen et al., 1998). BRCA2 appears to regulate the function of RAD51 in HR and in the last few years, a great deal of data has been published on the interplay between these two proteins and how it affects the repair of DSBs. BRCA2 deficiency is associated with a defect in GC Cells defective in BRCA2 function show a high degree of chromosome instability, including chromosome breaks and radial chromosomes (Patel et al., 1998; Tutt et al., 2001; Kraakman-van der Zwet et al., 2002). These aberrations have been shown to accumulate spontaneously during passage of cells in culture and can also be induced by DNA damaging agents that induce DSBs, such as IR and DNA cross-linking agents. BRCA2 is thought to promote genomic stability through its role in the error-free repair of DSBs by GC via its association with RAD51. How BRCA2 deficiency affects DSB repair has been investigated in several BRCA2 mutant cell lines using recombination repair substrates. Mouse embryonic stem cell lines carrying targeted mutations in Brca2 exhibit a reduction in the accurate repair of DSBs by GC, while deletion events are elevated (Moynahan et al., 2001b; Tutt et al., 2001; Stark et al., 2004). These deletions are thought to arise predominantly through the use of the SSA pathway. Deficiency in GC has also been demonstrated in the human tumour cell line CAPAN-1 and in the Chinese hamster cell mutant VC8, both of which are defective in BRCA2 function (Moynahan et al., 2001b; Larminat et al., 2002). NHEJ, however, is apparently unaffected in BRCA2-deficient cells (Patel et al., 1998; Yu et al., 2000). Loss of BRCA2, therefore, results in the repair of DSBs by a more errorprone mechanism possibly explaining the apparent chromosome instability associated with BRCA2 deficiency. In addition to this, BRCA2 has been implicated in the response to stalled replication and in preserving the stability of stalled replication forks, perhaps explaining the apparent instability that spontaneously accumulates in BRCA2-deficient cells (Lomonosov et al., 2003). Oncogene BRCA2 brings RAD51 to sites of DSBs RAD51 was the first BRCA2-binding protein to be identified (Sharan et al., 1997) and the interaction of these proteins has since been extensively studied. BRCA2 binds directly to RAD51 through its C-terminus and the BRC repeats located in the middle of the protein. These repeats are highly conserved across species, while most of the intervening sequences are not, suggesting that the BRC repeats are important for BRCA2 function (Bignell et al., 1997; Takata et al., 2002; Warren et al., 2002). Of the eight BRC repeats in the BRCA2 protein, RAD51 has been shown to bind to BRC1-4, BRC7 and BRC8, which are the more highly conserved repeats (Bignell et al., 1997; Wong et al., 1997). These repeats have been shown to bind distinct regions of RAD51, confirming nonequivalent interactions between the different BRC repeats and RAD51 (Galkin et al., 2005). The physical interaction between BRCA2 and RAD51 is essential for error-free HR to take place in response to DSBs. BRCA2 is thought to be required for the transport of RAD51 into the nucleus and to sites of DNA damage, where RAD51 would be released to form the nucleoprotein filament required for recombination to take place. While BRCA2 has nuclear localization signals (NLSs), these have not been identified in RAD51, prompting the idea that BRCA2 facilitates the transport of RAD51 into the nucleus. In support of this, truncated BRCA2, lacking its NLSs at the C-terminus has been shown to be cytoplasmic, along with the majority of the RAD51 protein (Spain et al., 1999; Davies et al., 2001). Accordingly, BRCA2-deficient cells do not form RAD51 foci in response to DNA damage (Yuan et al., 1999). RAD51 foci are thought to represent sites of repair that presumably contain nucleoprotein filaments of RAD51, and their formation is dependent on the function of BRCA2 upon DNA damage. However, there is some evidence that the formation of RAD51 foci in undamaged S phase cells may be BRCA2 independent (Tarsounas et al., 2003). These spontaneous foci occur in smaller numbers than damage induced foci and are thought to represent sites at which stalled or broken replication forks undergo repair. Only around 20% of RAD51 has been suggested to reside in a relatively immobile fraction that is bound to BRCA2, with the remaining 80% of RAD51 existing either as an immobile oligomerized fraction or as a relatively mobile fraction. The immobile RAD51 fraction complexed with BRCA2 is then mobilized upon replication arrest by hydroxyurea (Yu et al., 2003). A model has been proposed that BRCA2 is involved in both sequestering and mobilizing RAD51 and holds RAD51 in a state of readiness until DNA damage or replication arrest, when the complex becomes localized to sites of DSBs. RAD51 nucleoprotein filament formation and recombination For HR to take place, RAD51 must be released from BRCA2 to form a nucleoprotein filament on ssDNA, Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5869 which then invades and pairs with a homologous DNA duplex, initiating strand exchange between the paired DNA molecules. To understand how the BRCA2RAD51 interaction affects RAD51 function, the DNA-binding properties of RAD51 have been analysed in the presence of synthetic BRCA2 BRC-repeat peptides (Yuan et al., 1999; Davies et al., 2001; Stark et al., 2004). Overexpression of a single BRC repeat prevents RAD51 from forming nucleoprotein filaments on DNA. Furthermore, the addition of BRC peptides to preassembled nucleoprotein filaments causes them to dissociate. This was suggested to be a consequence of a direct interaction between a RAD51 monomer and a single BRC repeat, therefore preventing the RAD51 monomers from self-associating (Davies et al., 2001). Consequently, overexpressing a BRC repeat renders cells with wild-type BRCA2 hypersensitive to DNA damaging agents and results in compromised RAD51 foci formation, with an associated defect in GC and upregulation of SSA (Yuan et al., 1999; Stark et al., 2004). How this was mediated at the molecular level was clarified in a paper describing the crystallographic structure of a RAD51-BRC4 complex (Pellegrini et al., 2002). The BRC structure mimics the structure of the interaction domain of the adjacent RAD51 monomer, thereby impairing the ability of one RAD51 monomer to interact with another to form nucleoprotein filaments. This could be a way to prevent RAD51 from interacting with inappropriate DNA substrates within the nucleus. However, the inhibition of RAD51 oligomerization by a BRC repeat contrasts with the essential role of the BRCA2 protein in the repair of DSBs by HR. Recently, the BRC3 and BRC4 repeats were shown to be able to bind and form stable complexes with RAD51–DNA nucleoprotein filaments. However, only at high concentrations of the BRC repeats were filaments disrupted (Galkin et al., 2005). These results, therefore, suggest direct interactions between BRCA2 and RAD51 filaments, rather than with RAD51 monomers. However, in these experiments individual BRC repeats were utilized, which may not be representative of their function in the context of full-length BRCA2. This is supported by experiments investigating the importance of the BRC repeats, along with the C-terminal DBD (DNA/DSS1binding domain), in the repair of DNA damage by BRCA2 using a fusion protein (Saeki et al., 2006). The DBD of BRCA2 has five domains, including three oligonucleotide/oligosaccharide-binding (OB) fold domains (OB1, OB2 and OB3) (Yang et al., 2002). OB folds are found in ssDNA-binding proteins, such as RPA, to which the three OB folds of BRCA2 are most similar. The fusion protein consisted of a single (BRC3 or BRC4) or multiple BRC repeats fused to a ssDNAbinding domain from RPA (Saeki et al., 2006). Expression of any of these fusion proteins in BRCA2 mutant cells rescued BRCA2 deficiency by increasing GC while suppressing SSA, restoring RAD51 focus formation and cellular survival in response to IR and MMC, as well as reducing spontaneous chromosomal aberrations (Figure 2). This suggests that much of BRCA2 is dispensable for DSB repair and in maintain- ing genomic integrity, implying that the key role of BRCA2 is to deliver RAD51 to sites of DNA damage. A role for BRCA2 in promoting recombination was demonstrated in the yeast-like fungus Ustilago maydis. In this model, BRCA2 was shown to facilitate the nucleation of the RAD51 filament by recruiting RAD51 to DNA and displacing RPA at the nucleation site (Yang et al., 2005). This study also showed that the BRCA2 orthologue acted preferentially at a junction between dsDNA and ssDNA, stimulating filament formation only at junctions having 30 overhang ssDNA. The DBD domain was also found to stimulate the homologous pairing and strand-exchange activities of RAD51, indicating that BRCA2 might facilitate RAD51-mediated recombination and might have a role at the dsDNA–ssDNA junction of the resected DSB (Yang et al., 2002). The DBD-binding protein, DSS1, has also been shown in Ustilago to be important for properly controlled recombination (Kojic et al., 2005) and this is discussed in more detail below. BRCA2 regulation by phosphorylation In addition to its interaction with the BRC repeats, RAD51 can also interact with the C-terminal region of BRCA2. Recently, this region was shown to be phosphorylated at residue 3291 by cyclin-dependent kinases (CDKs) in a cell cycle-dependent manner (Esashi et al., 2005). This phosphorylation was observed to be low in S phase but to increase as the cell progressed from G2 to M phase, thereby preventing interaction of the C-terminus of BRCA2 with RAD51. However, treatment with IR reduced this phosphorylation by CDKs and restored the interaction between BRCA2 and RAD51. These results show that the phosphorylation status of S3291 is crucial for modulating interactions between the C-terminus of BRCA2 and RAD51 and possibly provides a mechanism for the regulation of recombinational repair. Exactly how the two RAD51binding regions of BRCA2, the BRC repeats and the C-terminus, work together to control RAD51 function is still not clear. BRCA2 and the regulation of cell cycle progression Several studies have linked BRCA2 with progression through mitosis (Lee et al., 1999; Futamura et al., 2000; Marmorstein et al., 2001; Daniels et al., 2004). BRCA2 associates with the DNA-binding protein BRAF35 (BRCA2-associated factor 35) through a region contained within BRC6, 7 and 8 of BRCA2 (Marmorstein et al., 2001). BRAF35 was observed to bind to branched DNA structures, such as those formed during recombinational repair, implicating BRAF35 in the DNA damage response. BRAF35 and BRCA2 were also found to colocalize on mitotic chromosomes and injection of antibodies into cells to block either BRCA2 or BRAF35 impeded entry into mitosis. Further evidence to support a role for BRCA2 in the M-phase checkpoint is through interaction with BUBR1 Oncogene Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5870 Figure 2 Functionally distinct effects of BRC repeat expression on RAD51 function. The BRCA2 BRC repeats bind RAD51 and are essential for the function of both proteins. Overexpressing a single BRC repeat in mammalian cells disrupts RAD51 filament formation and dissolves preassembled filaments by structurally mimicking the RAD51-RAD51 oligomer interface creating a BRC-RAD51 complex (Yuan et al., 1999; Davies et al., 2001; Stark et al., 2004). This inhibits RAD51 function, creating a BRCA2-deficient phenotype. A BRC repeat fused to a ssDNA-binding domain, however, restores RAD51 function in BRCA2-deficient cells, indicating that a BRC repeat and the DBD domain together are the minimal functional units of BRCA2 and that the main role of BRCA2 is to deliver RAD51 to the DNA break site (Saeki et al., 2006). (Futamura et al., 2000), which is important for the correct attachment of chromosomes to the mitotic spindle. BUBR1 was found to interact with and phosphorylate the C-terminus of BRCA2, but the significance of this remains to be elucidated. Additionally, inactivation of the mitotic checkpoint, through mutation of BUBR1 or BUB1 in BRCA2-deficient cells was sufficient to overcome growth arrest and initiate neoplastic transformation (Lee et al., 1999). A role for BRCA2 in cytokinesis has also been proposed as the cell cycle is extended from anaphase onset to the completion of cell division in BRCA2-deficient cells compared to wild-type cells (Daniels et al., 2004). The absence of BRCA2 has been associated with centrosome amplification and the formation of micronuclei, which are abnormal DNA-containing bodies (Tutt et al., 1999). These abnormalities could, to some extent, account for the aneuploidy and chromosome instability seen in BRCA2-deficient cells. Furthermore, BRCA2 has been reported to be phosphorylated in a cell cycle dependent manner by Polo-like kinase 1 (Plk1) (Lin et al., 2003; Lee et al., 2004) but how these phosphorylation events affect BRCA2 function still remains to be clarified. Oncogene The BRCA2-binding protein DSS1 and DNA repair The BRCA2-binding protein DSS1 has, in recent years, been demonstrated to be essential for BRCA2 function (Kojic et al., 2003; Gudmundsdottir et al., 2004; Li et al., 2006). DSS1 is a highly conserved 70 amino-acid protein that interacts with the C-terminal DBD domain of BRCA2 (Crackower et al., 1996; Marston et al., 1999). It binds to BRCA2 in an extended conformation, interacting with numerous residues within the helical domain, OB1 and OB2 of BRCA2, most of which are highly conserved (Yang et al., 2002). Most of the 25 BRCA2 interacting residues in the DSS1 protein are also highly conserved. These residues reside in two regions, referred to as BRCA2 interacting region 1 (residues 7–25) and BRCA2 interacting region 2 (residues 37–63), linked by a 11-residue nonconserved region. A functional role for DSS1 in DNA repair was suggested from studies using the fungus Ustilago maydis (Kojic et al., 2003). Ustilago contains, unlike the fission and budding yeast, a clear orthologue of BRCA2, as well as DSS1 and RAD51 orthologues. The phenotype of BRCA2 and RAD51 mutants parallel those seen in Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5871 mammalian cells, making Ustilago a useful system to study this pathway. Ustilago null DSS1 mutants exhibit a phenotype virtually identical to that seen in BRCA2 and RAD51 mutants, including extreme sensitivity to UV and IR, as well as deficiency in both mitotic and meiotic recombination. Both spontaneous and UVinduced allelic recombination was abrogated to the same extent as that observed in BRCA2 and RAD51 mutants and double-strand DNA break repair was similarly reduced. The defects in meiosis were associated with sterility, most likely due to failed recombinational repair of DSBs. DSS1 was also observed to be important for maintaining genomic stability, as mutants exhibited elevated spontaneous mutator activity of approximately 40-fold compared to wild-type and frequently harbored aberrant chromosomes as measured by pulse-field gel electrophoresis. Further work in Ustilago implicated DSS1 in the control of BRCA2 recombinational repair (Kojic et al., 2005). BRCA2 mutants lacking the C-terminal DSS1binding region were able to rescue the phenotype associated with DSS1 deficiency. This N-terminal BRCA2 fragment was still capable of binding RAD51 as the BRC region was intact and it also contained a NLS enabling BRCA2 to enter the nucleus. The BRCcontaining region is therefore capable of contributing to DNA repair in the absence of the DBD domain, demonstrating that DSS1 is not required for delivering RAD51 to sites of DNA damage or RAD51 nucleoprotein filament formation. Furthermore, fusing the BRCA2 N-terminus to the DNA-binding domain of the Ustilago orthologue of the RPA70 subunit exaggerated this phenotype even further. The DNA-binding domain of RPA70 has a similar structure to the BRCA2 DBD but is devoid of corresponding DSS1-interacting residues. DSS1-deficient cells transfected with the BRCA2RPA70 hybrid showed elevated levels of RAD51 foci formation, both spontaneous and damage-induced, as well as elevated recombination levels. This exaggerated phenotype highlights the importance of DSS1 in controlling BRCA2-dependent recombination. The importance of DSS1 in BRCA2 function has also been established in mammalian cells (Gudmundsdottir et al., 2004; Li et al., 2006). Silencing of DSS1 using RNA interference (RNAi) in both mouse and human cells compromised the ability of these cells to form colonies and importantly DSS1 was shown to be required for the formation of DNA damage induced RAD51 foci, suggesting a role for DSS1 in BRCA2- and RAD51dependent repair by HR. Consistent with this, DSS1 was shown to be important for the maintenance of genomic stability in the presence of MMC (Gudmundsdottir et al., 2004). The mechanism by which DSS1 affects BRCA2 function is not yet clear. However, DSS1 has been shown to be dispensable for the interaction between BRCA2 and RAD51, but conflicting views exist as to whether DSS1 is required for the stability of BRCA2. In contrast to two previous reports (Gudmundsdottir et al., 2004; Kojic et al., 2005), a recent paper by Li et al. (2006) suggests that silencing of DSS1 leads to a considerable reduction in BRCA2 protein levels as measured by western blotting. The halflife of the protein appeared to be affected, although calpain-, caspase- and proteasome inhibitors were unable to inhibit its degradation. The stability of a BRCA2 mutant devoid of the DBD region, however, was unaffected by DSS1 silencing, suggesting it is the DBD region that causes BRCA2 to be degraded. These results complement those mentioned previously, where removal of the DBD region in Ustilago rescues the DSS1/BRCA2 deficient phenotype (Kojic et al. 2005). Additionally, BRCA2 has proven to be largely insoluble in the absence of DSS1, potentially implicating DSS1 in maintaining the correct conformation of BRCA2 (Yang et al. 2002; Kojic et al. 2003). However, as the mechanism of how DSS1 depletion induces degradation of BRCA2 remains to be elucidated, further research into BRCA2 stability is warranted. DSS1 as a link between the proteasome and DNA repair A possible role for the proteasome in the repair of DSBs has emerged with the discovery that DSS1 is involved both in DSB repair and is a subunit of the proteasome. DSS1 was recently identified as a component of the lid subcomplex of the 19S regulatory particle in yeast (Funakoshi et al., 2004; Sone et al., 2004; Josse et al., 2006). The 19S regulatory complex serves to recognize ubiquitinated proteins and is implicated in their unfolding and translocation into the interior of the 20S proteolytic core complex, where they are degraded into oligopeptides (Baumeister et al., 1998). A functional link to the proteasome is provided by the demonstration that DSS1 is required for efficient ubiquitin-dependent protein degradation (Funakoshi et al., 2004; Sone et al., 2004; Josse et al., 2006). To examine whether the proteasome might be involved in repairing DNA DSBs, the sensitivity of yeast strains containing a deletion in one or more proteasome subunits to various DNA damaging agents have been investigated (Funakoshi et al. 2004; Krogan et al. 2004). Single mutants displayed little additional sensitivity to UV, HU, camptothecin and bleomycin but double mutants combining DSS1 deletion with another proteasome mutation exhibited marked hypersensitivity. Furthermore, combining a proteasome mutant with deletion of a gene involved in HR caused a synthetic growth defect along with hypersensitivity to HU and MMS, implicating the proteasome in a pathway other than or including HR. Using ChIP (chromatin immunoprecipitation) analysis, DSS1 was shown to bind to DNA at the site of a specific DSB along with components of the 19S and the 20S proteasome complexes, linking proteolysis to the repair of DSBs. Recruitment of DSS1 was reduced when either Rad52, a HR protein, or Pol4, a polymerase linked to NHEJ, were deleted, suggesting that DSS1 is recruited by both HR and NHEJ pathways to sites of DSBs. In support of that, DSS1 was shown to be important for the repair of DSBs by both HR and NHEJ (Krogan et al. 2004). Oncogene Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5872 Therapeutic exploitation BRCA1 and BRCA2 are important for the repair of DSBs by the potentially error-free pathway of HR by GC. Cells that lack either BRCA1 or BRCA2 repair these lesions by alternative more error-prone mechanisms. In the case of BRCA2 deficiency, repair by GC is impaired while the use of the alternative more errorprone homology-directed repair pathway SSA is upregulated (Moynahan et al., 2001b; Tutt et al., 2001; Stark et al., 2004). In BRCA1-deficient cells, the use of both GC and SSA is decreased, along with a precise subpathway of NHEJ, leaving DSBs to be repaired by an error-prone microhomology-mediated subpathway of NHEJ (Moynahan et al., 1999; Stark et al., 2004; Wang et al., 2006; Zhuang et al., 2006). The use of less accurate pathways for the repair of DSBs in BRCAdeficient cells is a likely explanation for the apparent genomic instability observed in these cells. BRCA1- and BRCA2-deficient cells have elevated sensitivity over BRCA wild-type cells to DNA damaging agents that cross-link DNA, such as MMC and to the platinumdrugs, cisplatin and carboplatin (Bhattacharyya et al., 2000; Yu et al., 2000; Tutt et al., 2001; Fedier et al., 2003). This increased sensitivity could therefore be exploited for the treatment of BRCA-associated cancers, as platinum drugs are already commonly used chemotherapeutic agents. Indeed, carboplatin is currently being trialled as a therapeutic agent in BRCA1 or BRCA2 mutational carriers (http://www.brcatrial.org; Tutt et al., 2005). The inherent DNA repair defect in BRCA-deficient cells has also provided a rationale for another therapeutic approach. This approach involves inhibiting the DNA repair protein PARP to generate specific lesions that require BRCA1 or BRCA2 for their removal, resulting in synthetic lethality. The concept of synthetic lethality applies when the mutation of two genes causes death, while mutation of either gene alone is viable (Hartwell, 1997; Kaelin, 2005). Poly(ADP-ribose)polymerase-1 (PARP-1) is an enzyme involved in the repair of SSBs by the BER pathway. PARP-1 deficiency causes failure to repair SSB lesions effectively, which when encountered by a DNA replication fork can lead to the formation of DSBs (Dantzer et al., 2000; Hoeijmakers, 2001). Under normal circumstances these lesions would be repaired by the error-free pathway of GC, but in a BRCA defective background these DSBs are likely to be repaired by more error-prone repair mechanisms, causing chromosome aberrations and loss of viability. PARP-1 inhibitors have indeed been shown to be selectively lethal to cells deficient in BRCA1 or BRCA2 (Bryant et al., 2005; Farmer et al., 2005; McCabe et al., 2005). In BRCA-deficient cells, PARP inhibitor treatment caused a cell cycle arrest in G2 phase of the cell cycle with subsequent apoptosis resulting in reduced cell survival. Cells that survived the treatment showed a profound increase in complex chromosome rearrangements. Such aberrations are, to a lesser extent, also seen in BRCA-deficient cells due to a defect in the repair of DSBs by GC. However, and importantly, no toxicity was evident in BRCA heterozygous cells, as these are the normal cells in BRCA mutation carriers. Additionally, BRCA-deficient cells are much more sensitive to PARP inhibitors than to treatment with DNA cross-linking agents. All this potentially creates a large therapeutic window for safe and effective treatment of tumours in carriers of a BRCA1 or BRCA2 mutation. PARP inhibitors have recently been shown to be synthetically lethal with defects in other HR-related proteins, including ATM, ATR and Fanconi Anemia genes, suggesting a wider range of therapeutic targets for PARP inhibitors than just for the treatment of BRCA-tumours (McCabe et al., 2006). As defects in these genes are found in a range of cancers, therapies designed to target the DNA repair defects in BRCA-deficient cells may be more widely applicable (Turner et al., 2004). Acknowledgements We are grateful to the members of the Ashworth laboratory for useful comments during the writing of this review, with special thanks to Nuala McCabe. Work in our laboratory is funded by Breakthrough Breast Cancer and Cancer Research UK. References Baldeyron C, Jacquemin E, Smith J, Jacquemont C, De Oliveira I, Gad S et al. (2002). Oncogene 21: 1401–1410. Baumeister W, Walz J, Zuhl F, Seemuller E. (1998). Cell 92: 367–380. Bhattacharyya A, Ear US, Koller BH, Weichselbaum RR, Bishop DK. (2000). J Biol Chem 275: 23899–23903. Bignell G, Micklem G, Stratton MR, Ashworth A, Wooster R. (1997). Hum Mol Genet 6: 53–58. Bochar DA, Wang L, Beniya H, Kinev A, Xue Y, Lane WS et al. (2000). Cell 102: 257–265. Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E et al. (2005). Nature 434: 913–917. Chen J, Silver DP, Walpita D, Cantor SB, Gazdar AF, Tomlinson G et al. (1998). Mol Cell 2: 317–328. Cortez D, Wang Y, Qin J, Elledge SJ. (1999). Science 286: 1162–1166. Oncogene Crackower MA, Scherer SW, Rommens JM, Hui CC, Poorkaj P, Soder S et al. (1996). Hum Mol Genet 5: 571–579. Daniels MJ, Wang Y, Lee M, Venkitaraman AR. (2004). Science 306: 876–879. Dantzer F, de La Rubia G, Menissier-De Murcia J, Hostomsky Z, de Murcia G, Schreiber V. (2000). Biochemistry 39: 7559–7569. Davies AA, Masson JY, McIlwraith MJ, Stasiak AZ, Stasiak A, Venkitaraman AR et al. (2001). Mol Cell 7: 273–282. Deng CX. (2006). Nucleic Acids Res 34: 1416–1426. Dong Y, Hakimi MA, Chen X, Kumaraswamy E, Cooch NS, Godwin AK et al. (2003). Mol Cell 12: 1087–1099. Elliott B, Richardson C, Jasin M. (2005). Mol Cell 17: 885–894. Esashi F, Christ N, Gannon J, Liu Y, Hunt T, Jasin M et al. (2005). Nature 434: 598–604. Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5873 Evers B, Jonkers J. (2006). Oncogene 25: 5885–5897. Fabbro M, Savage K, Hobson K, Deans AJ, Powell SN, McArthur GA et al. (2004). J Biol Chem 279: 31251–31258. Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB et al. (2005). Nature 434: 917–921. Fedier A, Steiner RA, Schwarz VA, Lenherr L, Haller U, Fink D. (2003). Int J Oncol 22: 1169–1173. Foray N, Marot D, Gabriel A, Randrianarison V, Carr AM, Perricaudet M et al. (2003). EMBO J 22: 2860–2871. Funakoshi M, Li X, Velichutina I, Hochstrasser M, Kobayashi H. (2004). J Cell Sci 117: 6447–6454. Futamura M, Arakawa H, Matsuda K, Katagiri T, Saji S, Miki Y et al. (2000). Cancer Res 60: 1531–1535. Galkin VE, Esashi F, Yu X, Yang S, West SC, Egelman EH. (2005). Proc Natl Acad Sci USA 102: 8537–8542. Gatei M, Scott SP, Filippovitch I, Soronika N, Lavin MF, Weber B et al. (2000). Cancer Res 60: 3299–3304. Gatei M, Zhou BB, Hobson K, Scott S, Young D, Khanna KK. (2001). J Biol Chem 276: 17276–17280. Gowen LC, Avrutskaya AV, Latour AM, Koller BH, Leadon SA. (1998). Science 281: 1009–1012. Gudmundsdottir K, Lord CJ, Witt E, Tutt AN, Ashworth A. (2004). EMBO Rep 5: 989–993. Hakem R, de la Pompa JL, Elia A, Potter J, Mak TW. (1997). Nat Genet 16: 298–302. Harkin DP, Bean JM, Miklos D, Song YH, Truong VB, Englert C et al. (1999). Cell 97: 575–586. Hartman AR, Ford JM. (2002). Nat Genet 32: 180–184. Hartwell L. (1997). Nature 387: 855–857. Hashizume R, Fukuda M, Maeda I, Nishikawa H, Oyake D, Yabuki Y et al. (2001). J Biol Chem 276: 14537–14540. Hefferin ML, Tomkinson AE. (2005). DNA Repair (Amsterdam) 4: 639–648. Hoeijmakers JH. (2001). Nature 411: 366–374. Ira G, Pellicioli A, Balijja A, Wang X, Fiorani S, Carotenuto W et al. (2004). Nature 431: 1011–1017. Jin Y, Xu XL, Yang MC, Wei F, Ayi TC, Bowcock AM et al. (1997). Proc Natl Acad Sci USA 94: 12075–12080. Johnson RD, Jasin M. (2000). EMBO J 19: 3398–3407. Josse L, Harley ME, Pires IM, Hughes DA. (2006). Biochem J 393: 303–309. Joukov V, Chen J, Fox EA, Green JB, Livingston DM. (2001). Proc Natl Acad Sci USA 98: 12078–12083. Kaelin WG. (2005). Nat Rev Cancer 5: 689–698. Kennedy RD, Quinn JE, Mullan PB, Johnston PG, Harkin DP. (2004). J Natl Cancer Inst 96: 1659–1668. Kleiman FE, Wu-Baer F, Fonseca D, Kaneko S, Baer R, Manley JL. (2005). Genes Dev 19: 1227–1237. Kojic M, Yang H, Kostrub CF, Pavletich NP, Holloman WK. (2003). Mol Cell 12: 1043–1049. Kojic M, Zhou Q, Lisby M, Holloman WK. (2005). Mol Cell Biol 25: 2547–2557. Kraakman-van der Zwet M, Overkamp WJ, van Lange RE, Essers J, van Duijn-Goedhart A, Wiggers I et al. (2002). Mol Cell Biol 22: 669–679. Krogan NJ, Lam MH, Fillingham J, Keogh MC, Gebbia M, Li J et al. (2004). Mol Cell 16: 1027–1034. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J et al. (2001). Nature 409: 860–921. Larminat F, Germanier M, Papouli E, Defais M. (2002). Oncogene 21: 5188–5192. Lee H, Trainer AH, Friedman LS, Thistlethwaite FC, Evans MJ, Ponder BA et al. (1999). Mol Cell 4: 1–10. Lee JS, Collins KM, Brown AL, Lee CH, Chung JH. (2000). Nature 404: 201–204. Lee M, Daniels MJ, Venkitaraman AR. (2004). Oncogene 23: 865–872. Le Page F, Randrianarison V, Marot D, Cabannes J, Perricaudet M, Feunteun J et al. (2000). Cancer Res 60: 5548–5552. Li J, Zou C, Bai Y, Wazer DE, Band V, Gao Q. (2006). Oncogene 25: 1186–1194. Lieber MR, Ma Y, Pannicke U, Schwarz K. (2003). Nat Rev Mol Cell Biol 4: 712–720. Lin HR, Ting NS, Qin J, Lee WH. (2003). J Biol Chem 278: 35979–35987. Lomonosov M, Anand S, Sangrithi M, Davies R, Venkitaraman AR. (2003). Genes Dev 17: 3017–3022. Ludwig T, Chapman DL, Papaioannou VE, Efstratiadis A. (1997). Genes Dev 11: 1226–1241. Ma JL, Kim EM, Haber JE, Lee SE. (2003). Mol Cell Biol 23: 8820–8828. Marmorstein LY, Kinev AV, Chan GK, Bochar DA, Beniya H, Epstein JA et al. (2001). Cell 104: 247–257. Marston NJ, Richards WJ, Hughes D, Bertwistle D, Marshall CJ, Ashworth A. (1999). Mol Cell Biol 19: 4633–4642. McCabe N, Lord CJ, Tutt AN, Martin NM, Smith GC, Ashworth A. (2005). Cancer Biol Ther 4: 934–936. McCabe N, Turner NC, Lord CJ, Kluzed K, Biakowska A, Swift S et al. (2006). Cancer Res (in press). McCarthy EE, Celebi JT, Baer R, Ludwig T. (2003). Mol Cell Biol 23: 5056–5063. Moynahan ME, Chiu JW, Koller BH, Jasin M. (1999). Mol Cell 4: 511–518. Moynahan ME, Cui TY, Jasin M. (2001a). Cancer Res 61: 4842–4850. Moynahan ME, Pierce AJ, Jasin M. (2001b). Mol Cell 7: 263–272. Mullan PB, Quinn JE, Gilmore PM, McWilliams S, Andrews H, Gervin C et al. (2001). Oncogene 20: 6123–6131. Okada S, Ouchi T. (2003). J Biol Chem 278: 2015–2020. Ouchi T, Monteiro AN, August A, Aaronson SA, Hanafusa H. (1998). Proc Natl Acad Sci USA 95: 2302–2306. Patel KJ, Vu VP, Lee H, Corcoran A, Thistlethwaite FC, Evans MJ et al. (1998). Mol Cell 1: 347–357. Paull TT, Cortez D, Bowers B, Elledge SJ, Gellert M. (2001). Proc Natl Acad Sci USA 98: 6086–6091. Pellegrini L, Yu DS, Lo T, Anand S, Lee M, Blundell TL et al. (2002). Nature 420: 287–293. Rothkamm K, Kruger I, Thompson LH, Lobrich M. (2003). Mol Cell Biol 23: 5706–5715. Ruffner H, Joazeiro CA, Hemmati D, Hunter T, Verma IM. (2001). Proc Natl Acad Sci USA 98: 5134–5139. Saeki H, Siaud N, Christ N, Wiegant WW, van Buul PP, Han M et al. (2006). Proc Natl Acad Sci USA 103: 8768–8773. Scully R, Chen J, Ochs RL, Keegan K, Hoekstra M, Feunteun J et al. (1997a). Cell 90: 425–435. Scully R, Chen J, Plug A, Xiao Y, Weaver D, Feunteun J et al. (1997b). Cell 88: 265–275. Sharan SK, Morimatsu M, Albrecht U, Lim DS, Regel E, Dinh C et al. (1997). Nature 386: 804–810. Shen SX, Weaver Z, Xu X, Li C, Weinstein M, Chen L et al. (1998). Oncogene 17: 3115–3124. Shin DS, Chahwan C, Huffman JL, Tainer JA. (2004). DNA Repair (Amsterdam) 3: 863–873. Somasundaram K, Zhang H, Zeng YX, Houvras Y, Peng Y, Wu GS et al. (1997). Nature 389: 187–190. Sone T, Saeki Y, Toh-e A, Yokosawa H. (2004). J Biol Chem 279: 28807–28816. Spain BH, Larson CJ, Shihabuddin LS, Gage FH, Verma IM. (1999). Proc Natl Acad Sci USA 96: 13920–13925. Oncogene Roles of BRCA1 and BRCA2 and associated proteins K Gudmundsdottir and A Ashworth 5874 Starita LM, Horwitz AA, Keogh MC, Ishioka C, Parvin JD, Chiba N. (2005). J Biol Chem 280: 24498–24505. Stark JM, Pierce AJ, Oh J, Pastink A, Jasin M. (2004). Mol Cell Biol 24: 9305–9316. Takata M, Sasaki MS, Sonoda E, Morrison C, Hashimoto M, Utsumi H et al. (1998). EMBO J 17: 5497–5508. Takata M, Tachiiri S, Fujimori A, Thompson LH, Miki Y, Hiraoka M et al. (2002). Oncogene 21: 1130–1134. Tarsounas M, Davies D, West SC. (2003). Oncogene 22: 1115–1123. Tibbetts RS, Cortez D, Brumbaugh KM, Scully R, Livingston D, Elledge SJ et al. (2000). Genes Dev 14: 2989–3002. Turner N, Tutt A, Ashworth A. (2004). Nat Rev Cancer 4: 814–819. Tutt A, Bertwistle D, Valentine J, Gabriel A, Swift S, Ross G et al. (2001). EMBO J 20: 4704–4716. Tutt A, Gabriel A, Bertwistle D, Connor F, Paterson H, Peacock J et al. (1999). Curr Biol 9: 1107–1110. Tutt A, Lord CJ, McCabe N, Farmer H, Turner N, Jackson SP et al. (2005). Molecular Approaches to Controlling Cancer: Cold Spring Harbour Symposia Quantitative Biology LXX. Stillman B, Stewart D (ed). CSHL Press: NY, pp. 139–148. Valerie K, Povirk LF. (2003). Oncogene 22: 5792–5812. van Gent DC, Hoeijmakers JH, Kanaar R. (2001). Nat Rev Genet 2: 196–206. Wang HC, Chou WC, Shieh SY, Shen CY. (2006). Cancer Res 66: 1391–1400. Wang Y, Cortez D, Yazdi P, Neff N, Elledge SJ, Qin J. (2000). Genes Dev 14: 927–939. Warren M, Smith A, Partridge N, Masabanda J, Griffin D, Ashworth A. (2002). Hum Mol Genet 11: 841–851. West SC. (2003). Nat Rev Mol Cell Biol 4: 435–445. Westermark UK, Reyngold M, Olshen AB, Baer R, Jasin M, Moynahan ME. (2003). Mol Cell Biol 23: 7926–7936. Oncogene Williamson EA, Dadmanesh F, Koeffler HP. (2002). Oncogene 21: 3199–3206. Wong AK, Pero R, Ormonde PA, Tavtigian SV, Bartel PL. (1997). J Biol Chem 272: 31941–31944. Wu LC, Wang ZW, Tsan JT, Spillman MA, Phung A, Xu XL et al. (1996). Nat Genet 14: 430–440. Xu B, Kim S, Kastan MB. (2001a). Mol Cell Biol 21: 3445–3450. Xu B, O’Donnell AH, Kim ST, Kastan MB. (2002). Cancer Res 62: 4588–4591. Xu X, Qiao W, Linke SP, Cao L, Li WM, Furth PA. (2001b). Nat Genet 28: 266–271. Xu X, Weaver Z, Linke SP, Li C, Gotay J, Wang XW et al. (1999). Mol Cell 3: 389–395. Yang H, Jeffrey PD, Miller J, Kinnucan E, Sun Y, Thoma NH et al. (2002). Science 297: 1837–1848. Yang H, Li Q, Fan J, Holloman WK, Pavletich NP. (2005). Nature 433: 653–657. Yarden RI, Pardo-Reoyo S, Sgagias M, Cowan KH, Brody LC. (2002). Nat Genet 30: 285–289. Yu DS, Sonoda E, Takeda S, Huang CL, Pellegrini L, Blundell TL et al. (2003). Mol Cell 12: 1029–1041. Yu VP, Koehler M, Steinlein C, Schmid M, Hanakahi LA, van Gool AJ et al. (2000). Genes Dev 14: 1400–1406. Yuan SS, Lee SY, Chen G, Song M, Tomlinson GE, Lee EY. (1999). Cancer Res 59: 3547–3551. Zhang H, Somasundaram K, Peng Y, Tian H, Bi D, Weber B et al. (1998). Oncogene 16: 1713–1721. Zhang J, Willers H, Feng Z, Ghosh JC, Kim S, Weaver DT et al. (2004). Mol Cell Biol 24: 708–718. Zhong Q, Boyer TG, Chen PL, Lee WH. (2002). Cancer Res 62: 3966–3970. Zhong Q, Chen CF, Li S, Chen Y, Wang CC, Xiao J et al. (1999). Science 285: 747–750. Zhuang J, Zhang J, Willers H, Wang H, Chung JH, van Gent DC et al. (2006). Cancer Res 66: 1401–1408.