Several newly identified tumor suppressor genes including ATM, NBS1, BRCA1 and BRCA2 are involved in DNA double-strand break repair (DSBR) and DNA damage-induced checkpoint activation. Many of the gene products involved in checkpoint control and DSBR have been studied in great detail in yeast. In addition to evolutionarily conserved proteins such as Chk1 and Chk2, studies in mammalian cells have identified novel proteins such as p53 in executing checkpoint control. DSBR proteins including Mre11, Rad50, Rad51, Rad54, and Ku are present in yeast and in mammals. Many of the tumor suppressor gene products interact with these repair proteins as well as checkpoint regulators, thus providing a biochemical explanation for the pleiotropic phenotypes of mutant cells. This review focuses on the proteins mediating G1/S, S, and G2/M checkpoint control in mammalian cells. In addition, mammalian DSBR proteins and their activities are discussed. An intricate network among DNA damage signal transducers, cell cycle regulators and the DSBR pathways is illustrated. Mouse knockout models for genes involved in these processes have provided valuable insights into their function, establishing genomic instability as a major contributing factor in tumorigenesis.
Existing evidence indicates that multiple cellular processes including checkpoint activation, DNA repair, and changes of gene transcription are initiated in response to DNA damage. Studies in budding and fission yeast have identified players involved in these processes. Some of the genes required for DNA damage-induced checkpoints are also required for cell cycle arrest upon replication block. Homologs of these genes have been identified in mammalian cells, and their roles in development and in DNA damage responses are being unraveled. In contrast to yeast, checkpoint pathways are differentially activated by UV, which mainly causes bulky adducts, and ionizing radiation (IR), which causes single-strand and double-strand breaks (DSBs) in mammals. While UV-induced mammalian cellular responses will be described briefly, this review will focus on DSB-induced cellular responses.
DSBs are the most detrimental form of DNA damage because they lead to chromosomal breakage and rearrangement, events that may result in apoptosis or tumorigenesis. Several human syndromes are characterized by chromosome instability and sensitivity to DSB-causative agents (reviewed by Khanna et al., 1998). Ataxia telangiectasia (A-T) and Nijmegen breakage syndrome (NBS) (reviewed by Shiloh, 1997; Rotman and Shiloh, 1998) are among the best characterized. Clinical manifestations of A-T include progressive neurodegeneration, telangiectasia in the face, immune deficiency, gonadal dysgenesis, and cancer predisposition. A-T cells are defective in DNA damage-induced checkpoint control as well as DNA repair. The product of the ATM gene (A-T mutated) belongs to a protein kinase family whose members include Mec1 and Tel1 of budding yeast, Rad3 of fission yeast, Mei41 of fruit fly, and other mammalian kinases including ATR (ATM- and RAD3-related) and DNA-PKcs. Studies of this gene family have provided insights into the DNA damage-induced cellular responses as well as other cellular processes such as meiosis and telomere maintenance. The clinical and cellular phenotypes of A-T and NBS share significant similarities. The NBS1 gene encodes a 95 kDa protein (Carney et al., 1998; Varon et al., 1998) that is a component of a stable Mre11/Rad50/NBS1 nuclease complex (Trujillo et al., 1998; Paull and Gellert, 1999). This complex plays a critical role in DNA DSBR (reviewed by Haber, 1998), explaining the hypersensitivity of NBS cells to IR. However, the molecular basis of the checkpoint defect in NBS cells is less clear. The products of the tumor suppressor genes, BRCA1 and BRCA2, are also involved in DSBR. BRCA1 also plays a role in checkpoint regulation.
Some checkpoint pathways, e.g., signaling from ATM to the checkpoint kinase Chk2, appear to be conserved evolutionarily (Matsuoka et al., 1998), whereas others are unique to mammals (Banin et al., 1998; Canman et al., 1998). In this review, we focus on DNA damage-induced cell cycle checkpoints and on the biochemical and genetic studies of DSBR in mammalian cells. Phenotypes of mouse mutants harboring mutations in genes involved in these cellular processes are summarized. Network of interactions among the multifunctional products of genes involved in cellular response to DSBs is emerging.
Checkpoint controls in eukaryotic cells
Checkpoints serve to monitor the order of events in the cell cycle and ensure that a cell cycle event occurs only after the completion of a prior event (Weinert and Hartwell, 1989). DNA damage checkpoint is one such example that is activated upon various kinds of external or internal stimuli that induce DNA damage, either programmed or accidental, and thus helps integrate DNA repair with cell cycle progression (Hartwell and Kastan, 1994). Such mechanisms are important for proper development as well as for prevention of genomic instability and cancer. Programmed breaks in DNA occur during development, e.g., meiotic recombination or during immunoglobulin gene rearrangements. Accidental DNA damage may be produced by several different ways: (i) exogenous DNA damaging agents that may be physical (UV, IR, etc.), or chemical e.g., methyl-methane-sulfonate (MMS), cisplatin, and neocarzinostatin (NCS); (ii) biological (certain viral infections) agents; (iii) endogenous DNA damaging agents such as reactive oxygen species. Such damages need to be repaired before the DNA is segregated into daughter cells to ensure faithful propagation of the genome. In keeping with the definition of checkpoints as being surveillance mechanisms acting between cell cycle phase transitions, DNA damage checkpoints may be divided into G1/S, S, and G2/M.
The G1/S checkpoint
This ensures that damaged DNA is not replicated and is one of the better-understood DNA damage checkpoints in mammalian cells. The tumor suppressor p53, one of the most commonly mutated genes in cancer, plays an important role in DNA damage induced G1/S arrest and apoptosis. The p53 gene product, a transcription activator, up-regulates the expression of genes such as p21, MDM2, and GADD45 (a growth arrest and DNA damage responsive gene) through the p53-binding elements in the promoter regions. The CDK inhibitor p21 can bind several cyclin-CDK complexes in vitro and may mediate the p53-dependent G1/S checkpoint (Figure 1) (Dulic et al., 1994; Kuerbitz et al., 1992; Reed et al., 1994). Normally, the cellular level of p53 protein is low due to its relatively short half-life. The stability of p53 is enhanced in cells exposed to different DNA damage agents, including IR and UV (reviewed by Levine, 1997). Phosphorylation on serine-15 of p53 partially inhibits its interaction with MDM2, a protein that targets p53 for ubiquitin-dependent degradation (Shieh et al., 1997). ATM interaction with dsDNA is enhanced upon IR in vitro (Suzuki et al., 1999). ATM might be recruited to sites of DSBs in vivo and thus regulates the activity of downstream effectors such as p53. Both ATM and ATR proteins phosphorylate p53 on serine-15 (Banin et al., 1998; Canman et al., 1998; Tibbetts et al., 1999). ATM phosphorylates p53 on serine-15 when DNA strand breaks are induced by agents such as IR (Canman et al., 1998), while ATR may be more critical for phosphorylation of p53 upon UV damage. Expression of kinase inactive ATR (ATRki) in human fibroblasts interfered with late phase phosphorylation of p53 upon IR (Tibbetts et al., 1999) suggesting that the delayed phosphorylation of p53 observed in A-T cells (Kastan et al., 1992) may be mediated by ATR. IR also induces dephosphorylation of serine-376 in p53 in an ATM-dependent manner (Waterman et al., 1998). The identity of the ATM-dependent phosphatase that dephosphorylates serine-376 of p53 is unknown. This dephosphorylation event results in the creation of a binding site for 14-3-3 protein and in turn, an increased affinity for specific DNA sequences. Consistent with ATM-dependent activation of p53, neither dephosphorylation nor interaction with 14-3-3 protein is evident in A-T cells (Waterman et al., 1998). Thus both phosphorylation and dephosphorylation of specific sites in p53 may play a vital role in checkpoint activation.
Recently, a new candidate tumor suppressor, p33, which co-precipitates with p53, was identified (Garkavtsev et al., 1998). Overexpression of p33 causes G1 arrest or apoptosis and functional p33 is required for p53 mediated p21 induction in response to IR. Two novel members of the p53 protein family, p63 and p73, were recently cloned (reviewed by Kaelin, 1999). Although their exact role as tumor suppressors or checkpoint proteins is unknown, p73 is phosphorylated by a non-receptor tyrosine kinase c-Abl that is activated upon DNA damage (Gong et al., 1999; Yuan et al., 1999b). Both p63 and p73 share the DNA binding and transactivation domains of p53 and can activate p53 target genes in a transient transfection assay (Lee and La Thangue, 1999; Shimada et al., 1999). Like p53, p73 upregulates MDM2 expression, but unlike p53, it is not targeted by MDM2 for ubiquitin-dependent degradation. Instead, MDM2 negatively regulates p73 transactivator function by disrupting its interaction with p300/CBP, a component of the eukaryotic transcription complex (Zeng et al., 1999).
ATM when activated phosphorylates and activates c-Abl kinase which in turn can activate p73 and potentially result in the transactivation of p21 and GADD45 (Lee and La Thangue, 1999). Together, ATM and c-Abl may mediate a checkpoint function through p73 but it is not known if p73-negative cells have checkpoint defects. Thus ATM may mediate G1/S checkpoint through a direct phosphorylation of p53 or through phosphorylation of c-Abl which in turn can upregulate p21 through p73 activation (Figure 1). ATR may be the primary p53 kinase in response to UV damage and only a secondary kinase for p53 in response to IR. DNA-PKcs, another DNA dependent kinase related to ATM and ATR, does not appear to play an important role in the G1/S checkpoint since cells that lack DNA-PKcs have normal DNA damage checkpoints (Burma et al., 1999).
The S phase checkpoint
When DNA is damaged during early part of the DNA synthesis phase (S phase), the S phase DNA damage checkpoint (SDDC) is activated (Larner et al., 1997). Although the exact mechanisms are poorly understood, several studies have demonstrated a down-regulation of DNA replication in response to radiation damage (Painter, 1986). The lack of functional ATM renders this checkpoint defective causing radio-resistant DNA synthesis (RDS), a hallmark of A-T cells (Painter and Young, 1980). Using a defined chromosomal replicon system, Larner et al., (1994) demonstrated that down regulation of DNA synthesis is mediated through a block in firing of replication origins.
DNA replication in eukaryotic cells initiates from many replication origins that fire throughout the S phase in a defined manner. In A-T cells, initiation of mid or late firing of DNA origins is not inhibited upon IR (Larner et al., 1999). Although most of the proteins involved are conserved through evolution, a molecular understanding of the SDDC regulation in the mammalian system is far from clear. Studies in yeast have demonstrated that Mec1 and Rad53 checkpoint proteins are required for firing of late origins suggesting that SDDC checkpoint may be mediated through a block in firing of late origins (Santocanale and Diffley, 1998). Indeed, when S. cerevisiae are treated with MMS, DNA replication delay is associated with a selective block to firing of late origins that can be relieved by mutation of Rad53 (Shirahige et al., 1998). In S. pombe, Cds1 (a Rad53 ortholog) mediates the S phase checkpoint in response to DNA damage (Lindsay et al., 1998).
Hydroxyurea, a ribonucleotide reductase inhibitor, which blocks the progression of replication forks from early origins also inhibits firing of late origins. A block in DNA replication activates a replication checkpoint, which may be different from SDDC. In S. pombe, the replication checkpoint is mediated through Cds1 and Chk1 (Boddy et al., 1998; Zeng et al., 1998). A human homolog of Cds1 (termed Chk2 or HsCds1) was recently cloned (Brown et al., 1999; Matsuoka et al., 1998). Chk2 (HsCds1) is phosphorylated upon DNA damage as well as when DNA synthesis is blocked. Only the former is dependent on ATM suggesting other kinases may activate Chk2 when replication is blocked thereby activating the replication checkpoint. In mammalian cells, whether Chk1 also plays a role in the replication checkpoint remains to be addressed.
The G2/M checkpoint
This checkpoint is operational in late G2 phase and presumably allows for repair of DNA that was damaged in late S or in the G2 phases of cell cycle prior to mitosis. Thus, the G2 checkpoint functions to prevent damaged DNA being segregated into daughter cells. This checkpoint depends on the inhibition of Cdc2 kinase activity (Rhind et al., 1997; Yu et al., 1998). Chk1 and Chk2 phosphorylate Cdc25C, a dual specificity phosphatase required for removal of the inhibiting phosphorylation of tyrosine-15 of Cdc2, in vitro (Matsuoka et al., 1998; Sanchez et al., 1997). Since mammalian cells lacking Chk1 and Chk2 are unavailable at the present time, the relative contribution of each protein to the G2/M checkpoint is unclear (Figure 1). In S. pombe Chk1 and not Cds1 is essential for the G2/M checkpoint (Brondello et al., 1999). Human Chk2 is activated in an ATM-dependent manner in the presence of DSBs (Brown et al., 1999; Chaturvedi et al., 1999; Matsuoka et al., 1998). In response to IR, human Chk1 is also phosphorylated and activated in an NBS1- and ATM-dependent manner (Dasika et al., in preparation). UV-induced activation of Chk1 does not require ATM or NBS1 but expression of dominant-negative ATR abrogates Chk1 phosphorylation suggesting ATR may be upstream of Chk1 in the UV response pathway (Dasika et al., in preparation). Phosphorylation of Cdc25C on serine-216 by Chk1 or Chk2 creates a binding site for 14-3-3 proteins and results in export to and retention in the cytoplasm (Lopez-Girona et al., 1999; Peng et al., 1997). Nuclear Cdc2 remains phosphorylated in the absence of Cdc25C and the cells remain arrested in the G2 phase.
Recent evidence suggests that p53 may also play a role, albeit a redundant one, in the G2/M checkpoint (Passalaris et al., 1999). Activation of p53 in response to DNA damage results in induction of GADD45 (Zhan et al., 1994, 1996). GADD45 can destabilize Cdc2/cyclin B complexes in vitro suggesting that it may mediate Cdc2/cyclin inactivation in vivo (Zhan et al., 1999). Alternatively or in addition, p53-dependent transcriptional repression of cdc2 and cyclin B promoters may contribute to the G2/M checkpoint (Passalaris et al., 1999).
Other upstream activators of checkpoint control
S. cerevisiae mutants incapable of arresting cell cycle progression in the S/G2 phase when subjected to radiation (rad mutants) were isolated which proved to be very useful genetic tools to study checkpoint mechanisms (Weinert and Hartwell, 1989, 1993). Similarly, in S. Pombe at least six `rad' genes (Rad1, Rad3, Rad9, Rad17, Rad26 and Hus1) are required for the S and G2/M checkpoints in response to DNA damage (reviewed by Rhind and Russell, 1998). Recently, human homologs of Rad1, Rad9, Rad17 and Hus1 have been cloned (Bao et al., 1999; St. Onge et al., 1999; Volkmer and Karnitz, 1999). Since individual gene knockouts are not available, their exact requirement in checkpoint regulation of mammalian cells is unknown (Figure 1). However, human Rad9 is phosphorylated upon DNA damage and physically associates with Rad1 and Hus1 but not Rad17 (St. Onge et al., 1999; Volkmer and Karnitz, 1999). Since ATM (a Rad3-like protein) binding to dsDNA is enhanced upon IR in vitro (Suzuki et al., 1999), and since Rad9 phosphorylation is presumably normal in A-T cells (Volkmer and Karnitz, 1999), whether ATM and/or Rad1/Rad9/Hus1 complex serve(s) as the damage recognition complex is still unclear.
A BRCT (BRCA1 carboxyl terminus) motif containing yeast protein designated Crb2/Rhp9 was found to be phosphorylated in response to UV damage with similar kinetics as Chk1 (Saka et al., 1997). Crb2 phosphorylation was dependent on the rad genes but not Chk1 suggesting that Crb2 may also be upstream of Chk1 (Saka et al., 1997). Recent evidence suggests that Crb2 is required, in addition to Cds1, for the S and G2/M checkpoints (Grenon et al., 1999). The identity of a human Crb2 homolog, if there is one, that may play an analogous role is currently unknown, although several BRCT motif containing candidates have been identified in mammals (Callebaut and Mornon, 1997). Fission yeast Cut5 (cell untimely torn mutant 5) is an essential component of replication checkpoint (Saka et al., 1994). Like Crb2, Cut5 interacts with Chk1 in a yeast two-hybrid system (Saka et al., 1997). It was recently found to be essential for the G2 checkpoint regardless of the type of DNA damage (Verkade and O'Connell, 1998).
There is much to be learnt about the upstream activators of checkpoint response in mammalian cells. Since the human homologs of Rad genes have been cloned only recently, the coming years should be very fruitful in understanding the molecular basis of the checkpoint response to DNA damage in mammalian cells. Checkpoint pathways are likely to be required not only upon DNA damage but also during normal cellular proliferation to safeguard genomic stability. Therefore, mutation of checkpoint genes is likely to result in severe phenotypes.
DNA double-strand break repair in eukaryotes
Most of our knowledge on DSBR comes from studies in yeast. At least two distinct pathways have evolved for DSBR in eukaryotic cells: homologous recombinational (HR) repair and non-homologous end joining (NHEJ) (for review, see Nickoloff and Hoekstra, 1998; Paques and Haber, 1999; Petukhova et al., 1999; Rathmell and Chu, 1998). HR is an error-free pathway wherein an intact template of DNA in the sister chromatid or homolog is used to repair damaged DNA. In contrast, NHEJ is an error-prone pathway that joins DNA ends with no sequence homology, although short homologous sequences (<10 bp) termed microhomology are frequently found at the junction. NHEJ often results in deletions or small insertions. Many subpathways have been identified in budding yeast (Paques and Haber, 1999) and the biochemical activities of many of the DSBR proteins have been characterized (reviewed by Petukhova et al., 1999; Smith and Jackson, 1999). We will summarize the recent progress on DSBR in mammals and discuss similarities and differences with that of yeast.
The Rad50/Mre11/NBS1 complex
Genetic and biochemical observations indicate that DSBs are processed by nucleases. After the introduction of a DSB, the nucleolytic processing yields a 3′-single-stranded (ss) overhang of a few hundred bases (see Figure 2; Cao et al., 1990; Sun et al., 1991). In S. cerevisiae, mre11/rad50/xrs2 mutants are defective in the processing of DSB (Taccioli et al., 1994). Deletion of each gene results in hypersensitivity to IR and MMS and defects in both NHEJ and meiotic homologous recombination. Genetic analysis has established the roles of the Mre11/Rad50/Xrs2 complex in both HR and NHEJ pathways (Boulton and Jackson, 1998). In addition to DSBR, the products of these genes are involved in other cellular processes including chromatin structure and telomere maintenance and the generation of Spo11-mediated DSB in meiosis (for review, see Haber, 1998; and references therein). Using a mutator assay, Chen and Kolodner have demonstrated that Mre11/Rad50/Xrs2 suppresses gross chromosomal rearrangements that presumably occur by non-homology-mediated re-arrangement, whereas a eukaryotic ssDNA binding protein, replication protein A (RPA) suppresses microhomology-mediated rearrangements (Chen and Kolodner, 1999). Earlier studies demonstrated that rad50 and xrs2 haploid cells in G2 are more sensitive to IR than rad50 or xrs2 diploid cells in G1 (Ivanov et al., 1992), indicating these proteins play an important role in sister-chromatid interactions.
Rad50 and Mre11 proteins share sequence similarity with E. coli SbcC and SbcD, respectively, proteins that form a complex with nuclease activity (Connelly and Leach, 1996; Sharples and Leach, 1995). Rad50, like SbcC, is a coiled-coil protein with ATP binding motifs and shares sequence similarity with the structural maintenance of chromosome (SMC) family of proteins. This protein family is implicated in chromosome condensation and segregation, transcriptional repression and recombination (Connelly et al., 1998). Mre11, like SbcD, belongs to a family of phosphoesterases (Sharples and Leach, 1995). In view of the sequence similarity with SbcD, and since DSB processing yields 3′-single strand overhang in vivo, it was expected that Mre11 would possess 5′-to-3′ nuclease activities. Surprisingly, purified eukaryotic Mre11 possesses Mn2+-dependent 3′-to-5′ dsDNA exonuclease, ssDNA endonuclease and, in yeast, 3′-to-5′ ssDNA exonuclease activities as well (Haber, 1998; Moreau et al., 1999). One mechanism by which Mre11 complex could potentially affect 5′- to 3′-resection of DSB ends is through endonuclease cleavage of ssDNA unwound by an associated helicase, similar to how RecBCD protein complex processes DSBs in E. coli (Paques and Haber, 1999; Petukhova et al., 1999). The specific involvement of a helicase in DSB processing reaction remains to be addressed.
Studies of mre11 mutants indicate that this protein (and presumably the other members of the complex) has both enzymatic and structural functions. Nuclease-deficient mre11 mutant is not as sensitive to IR as the null mutants but has normal telomere length, suggesting that the Mre11 nuclease activities are not required for chromosome end protection and may not be as crucial in mitosis as in meiosis (Moreau et al., 1999).
A protein complex consisting of Mre11/Rad50/NBS1 is found in human cells that is likely to be the functional homolog of yeast Mre11/Rad50/Xrs2 complex (Carney et al., 1998; Dolganov et al., 1996; Petrini et al., 1995; Trujillo et al., 1998). Unlike Mre11 or Rad50, there is only very limited sequence similarity between yeast Xrs2 and human NBS1. The most notable feature of NBS1 is a forkhead-associated (FHA) domain and a BRCT domain in its amino terminus, both of which have been implicated in protein-protein interactions (Varon et al., 1998). The exonuclease activity of Mre11 is enhanced by Rad50 (Paull and Gellert, 1998). Furthermore, the Mre11/Rad50/NBS1 complex exhibits several activities not seen in the absence of NBS1 including the unwinding of DNA duplex and efficient cleavage of fully paired hairpins (Paull and Gellert, 1999). Recent studies demonstrate that many DSBR proteins are required for V(D)J recombination (see below). Thus these studies provide a molecular explanation of the DSB sensitivity of NBS cells and the defective V(D)J recombination observed in NBS patients. More studies are needed to explain how NBS1 modulates the activities of Mre11.
A DNA damage response results in the re-localization of repair proteins that can be visualized using an immunohistochemical assay of Mre11/Rad50/NBS1. Upon IR treatment, Rad50, Mre11 and NBS1 co-localize to the sites of DNA damage, forming discrete IR-induced immunofluorescent foci (IRIF) early in DNA-damage response and remain associated with DSBs until the repair is complete (Maser et al., 1997; Nelms et al., 1998). Interestingly, hMre11-hRad50 IRIF do not form in cells harboring a mutation in NBS1, ATM, or BRCA1 genes (see below), highlighting the complex regulation of such DNA damage response in mammalian cells.
The 3′ ssDNA tails resulted from nucleolytic processing of DSBs described above mediate search for a DNA homolog when bound by recombination proteins. Invasion of the homologous DNA results in the formation of heteroduplex DNA which extends its length by DNA strand exchange (Petukhova et al., 1999). These processes are coupled with DNA synthesis to replace the genetic information eliminated during nucleolytic processing of the DSB. Subsequently, Holliday junctions are formed and heteroduplex regions are extended by branch migration of the Holliday junction. Finally, resolution of the Holliday junctions yields mature recombinants with or without crossover in relation to the flanking markers (see Figure 2).
Rad51-dependent homologous recombination
In S. cerevisiae, HR is the predominant pathway for the repair of DSBs. This repair pathway involves the members of the RAD52 epistasis group, which include RAD50, RAD51, RAD52, RAD54, RAD55, RAD57, RAD59, MRE11 and XRS2. These genes can be divided into three subgroups. RAD52 is the gene required for all HR pathways and thus rad52 mutants have the most severe phenotype among all mutants of the RAD52 epistasis group (Paques and Haber, 1999). rad51, rad54, rad55 and rad57 mutants share common phenotypes and belong to one subgroup. As described above, mre11, rad50 and xrs2 mutants share similar phenotypes and these genes form the third subgroup. Biochemical studies have provided a framework for the HR pathways mediated by these proteins.
Eukaryotic Rad51 protein, a homologue of Escherichia coli RecA protein, plays a central role in HR and recombinational repair. Mutations in RAD51 gene result in IR-sensitivity, defects in meiosis and accumulation of DSBs. Similar to E. coli RecA, S. cerevisiae Rad51 forms nucleoprotein filaments on DNA and promotes homologous pairing and strand exchange in vitro. Several groups showed recently that heteroduplex DNA formation mediated by Rad51 is stimulated by RPA (reviewed by Baumann and West, 1998b). However, extensive strand exchange was not observed under standard conditions, indicating the requirement of additional proteins. Indeed, Rad51 nucleoprotein filament assembly is enhanced by Rad52 and Rad55/Rad57 in the presence of RPA (Benson et al., 1998; New et al., 1998; Shinohara et al., 1998; Sung, 1997).
Although yeast rad51 mutants are viable, knockout of the mouse Rad51 gene leads to embryonic lethality (Lim and Hasty, 1996; Tsuzuki et al., 1996). Using a tetracyclin-regulated promoter to control the expression of Rad51, Sonoda et al. (1998) demonstrated that the absence of Rad51 leads to extensive chromosome breakage, cell cycle arrest (mainly G2/M) and cell death. Furthermore, sister chromatid exchange frequency is greatly reduced in cells lacking Rad51, suggesting a potential role for HR in sister chromatid exchange (Sonoda et al., 1999). The identification of novel Rad51 interaction proteins (see below) as well as the discoveries of multiple cellular processes involving Rad51 will provide a better explanation for the lethal phenotype caused by Rad51 inactivation in mammals.
Both genetic and biochemical data indicate that Rad52 interacts with the Rad51 recombinase (Baumann and West, 1998b). Purified Rad52 protein binds both single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA). In vitro, Rad52 enhances Rad51-mediated strand exchange by alleviating the inhibitory effects of RPA (New et al., 1998; Shinohara et al., 1998; Sung, 1997). Consistent with this observation, Rad52 interacts with the 34 kDa subunit of RPA (Shinohara et al., 1998). Interaction of these repair proteins in vivo is supported by the observation that Rad52, RPA, Rad55, Rad57 and Rad51 co-localize to the sites of DSB during meiotic recombination (Gasior et al., 1998). Recently, human Rad52 was shown to bind directly to DSBs and, like Ku (see below), protect DNA from exonuclease attack and promote end-to-end association through intermolecular interactions between hRad52 (Van Dyck et al., 1999). Since Ku binds to DSB and mediates the NHEJ pathway, it is proposed that Rad52 plays a similar role but directs the protein bound DNA to the homologous recombination pathway. If this is the case, Rad52 may act immediately following the completion of nucleolytic processing by Mre11/Rad50/Xrs2. In support of this view, it has been demonstrated that murine Rad52 co-localizes with Rad50 IRIF upon DNA damage (Liu et al., 1999). Surprisingly, while both biochemical and cellular studies suggest an important role for Rad52 in HR in yeast, knockout of Rad52 results in a subtle phenotype in the mouse. This raises the possibility that either mammalian Rad52 function may not be similar to that of ScRad52 or that there may be yet to be identified homologs of RAD52 in the mammalian genome.
Rad54 is a member of the SWI2/SNF2 subfamily of ATPase (Eisen et al., 1995). The ATPase activity of Rad54 is completely dependent on dsDNA. Rad51 was shown to interact with Rad54 and this interaction is functionally important for HR (Clever et al., 1997; Jiang et al., 1996). In vitro, Rad54 promotes homologous DNA pairing at the expense of ATP hydrolysis. Purified human Rad54 protein unwinds dsDNA upon ATP hydrolysis (Petukhova et al., 1998), thereby promoting the formation and/or stabilization of hRad51-mediated joint molecules (Tan et al., 1999). The important role of Rad54 in HR and sister chromatid exchange has been demonstrated by gene inactivation in both mouse and chicken cells (Essers et al., 1997; Sonoda et al., 1999).
Studies of recombinational proteins have so far indicated that Rad52 and Rad55/Rad57 heterodimer act before joint molecules are formed (Baumann and West, 1998b) whereas Rad54 is required at a later step of HR (Tan et al., 1999). In mammals, additional proteins are required to facilitate the error-free repair process (see below).
Non-homologous end joining
NHEJ pathway re-joins DNA ends either in a homology-independent or a microhomology-dependent manner. Similar to HR, many players of the NHEJ are highly conserved through evolution. In yeast, ScKu70 and ScKu80 are important for NHEJ and telomere maintenance. However, the Ku70 and Ku80 phenotypes are obvious only in the rad52 mutant background because HR is the dominant mechanism for DSBR in budding yeast (Siede et al., 1996).
Both HR and Ku-mediated NHEJ have been shown in Xenopus laevis (Labhart, 1999). Recent studies demonstrated that NHEJ activities can be re-constituted in vitro with two separate chromatographic fractions (Baumann and West, 1998a). Joining is dependent on ligase IV, XRCC4, Ku and DNA-PKcs (Baumann and West, 1998a). The identification of important genes for NHEJ using human, mouse and hamster mutant cells defective in DSBR or V(D)J recombination, together with biochemical studies have led to a better understanding of NHEJ.
The Ku autoantigen
Ku was first identified as an autoimmune antigen in scleroderma-polymyositis and other autoimmune patients. Ku is a stable heterodimer, consisting of 70 and 86 kDa subunits (Smith and Jackson, 1999). In S. cerevisiae, loss of Ku function shortened the telomere length (Boulton and Jackson, 1998) and altered the expression of telomere-located genes, indicating disruption of telomeric chromatin in its absence.
Ku binds to DNA avidly in a sequence-independent fashion. It binds tightly to ds DNA ends with 3′-protruding, 5′-protruding or blunt ends. Ku also binds to DNA nicks, gaps, bubbles and stem-loop structures at DNA ends but not ssDNA ends (Dynan and Yoo, 1998). In vivo cross-linking experiments demonstrated that Ku binds to telomeric DNA (Gravel et al., 1998). Furthermore, Ku interacts with Sir4 which is involved in the transcriptional silencing of telomere-adjacent region (Tsukamoto et al., 1997). These biochemical properties of Ku likely underlie Ku's role in telomere maintenance and transcriptional silencing that was first predicted based on genetic studies in yeast.
Ku translocates along the DNA molecule in an ATP-independent manner (Paillard and Strauss, 1991). Ku can also transit from one linear DNA molecule to another if the termini of the DNA ends form base pairing (Bliss and Lane, 1997). Consistent with this observation, Ku was found to stimulate DNA end ligation by DNA ligase I in vitro (Ramsden and Gellert, 1998).
A group of human DNA repair genes, called XRCC (X-ray repair cross complementation) genes, was identified based on complementation studies using mutant rodent cell lines that are hypersensitive to IR and defective in both DSB rejoining and V(D)J recombination. Ku80 is encoded by XRCC5 while Ku70 is the protein product of XRCC6 (Gu et al., 1997a; reviewed by Jeggo, 1998). Studies of Ku80- and Ku70-knockout mice have further confirmed their roles in DSBR.
DNA-dependent protein kinase (DNA-PK)
DNA-PK is a nuclear serine/threonine kinase that is activated upon binding to DNA ends, nicks, and gaps (Gottlieb and Jackson, 1993). It consists of a regulatory subunit, the Ku70 and Ku80 heterodimer, and the catalytic subunit (DNA-PKcs) of 465 kDa. There are sequence homologies between DNA-PKcs and aforementioned PI3 kinase family members, e.g., ATM and ATR. Although Ku70 and Ku80 have been evolutionarily conserved, a budding yeast counterpart of human DNA-PKcs has not been identified.
The importance of DNA-PKcs in DSBR was initially demonstrated using cells established from severe combined immune deficient (SCID) mice. SCID mice lack mature B and T cells due to defective V(D)J recombination. SCID cells are hypersensitive to IR, suggesting that common factors might be used for V(D)J recombination and DSBR.
In vitro experiments have shown that upon binding of dsDNA ends, Ku recruits DNA-PKcs and activates its kinase activity. Atomic-force microscopy study indicates that in the presence of large DNA fragments, DNA-PKcs alone also binds to DNA termini, with Ku70/Ku80 binding adjacent to DNA-PKcs (Yaneva et al., 1997). DNA-PKcs can also hold the ends of two ds DNA molecules (Cary et al., 1997). DNA-PKcs alone exhibits a low level of DNA-dependent kinase activity. Addition of Ku protein stimulates the kinase activity 5 – 10-fold (Yaneva et al., 1997). Based on the abundant amount of Ku in cells, it is likely that activation of DNA-PK by DSBs in vivo involves Ku.
Recent studies of DNA-PK structure by electron crystallography have revealed an open channel and an enclosed cavity with openings that allow entry of ss DNA. Guided by the results, further biochemical studies suggested that activation of the kinase requires both ds and ss DNA (Leuther et al., 1999). Future structural studies of DNA-PK in the presence of Ku should yield additional insights into the regulation of DNA-PK activity. It is also important to identify the cellular substrates of DNA-PK. Many proteins can be phosphorylated by DNA-PK in vitro, but its in vivo substrates remain elusive (Smith and Jackson, 1999). Studies to date indicate that Ku and DNA-PKcs both participate in V(D)J recombination and DSBR. However, functions such as growth regulation and telomere maintenance appear to be uniquely involving Ku.
XRCC4 and DNA ligase IV
The XRCC4 gene encodes a nuclear phosphoprotein important for cellular resistance to IR (Li et al., 1995). Since XRCC4 interacts with DNA ligase IV and stimulates its activity fivefold in vitro (Critchlow and Jackson, 1998; Grawunder et al., 1997), it appears likely that DNA ligase IV is required for the ligation step in NHEJ and in V(D)J recombination (see Figure 2; Grawunder et al., 1998a). This was confirmed in the mouse knockout model in which the ligase IV gene had been inactivated (Grawunder et al., 1998b). It is worthwhile noting that XRCC4 interacts with DNA ligase IV via the ligase IV carboxyl terminus that contains two tandem BRCT domains.
In addition to genes identified by genetic approaches described above, the tumor suppressor genes, ATM, BRCA1 and BRCA2 have also been implicated in DSBR. The biochemical properties and function of these tumor suppressors are described below.
Tumor suppressor genes implicated in double-strand break repair in metazoans
ATM and double-strand break repair
There is accumulating evidence that ATM functions directly in DSBR, in addition to checkpoint regulation. Several groups reported higher residual levels of DSBs persisting in A-T cells after IR. Since these experiments were performed using non-cycling cells (G0/G1), the deficiencies observed in A-T cells could not have been due to defective cell cycle checkpoints per se. A-T cells display several abnormalities in the repair of DSBs; (i) higher levels spontaneous intra-chromosomal recombination after IR; (ii) frequent error-prone intra- and extra-chromosomal recombinations; (iii) aberrant NHEJ characterized by large deletions or insertions at the site of joining (Meyn, 1995; see Shiloh, 1997; and references therein). Since the studies described above were performed in transformed A-T cell lines, it is possible that DSB-induced signaling pathways might be de-regulated in these cell lines. Therefore, it is important to study DSBR in ATM-deficient mouse embryonic stem (ES) cells and mouse embryonic fibroblasts (MEF).
In contrast to checkpoint control, little is known about the biochemical basis of defective DSBR in A-T cells. Liu and Weaver (1993) demonstrated a delayed IR-, but not UV-induced phosphorylation of the 34 kDa subunit of RPA in A-T cells. A similar observation was made in yeast, suggesting that RPA phosphorylation in response to DSB is conserved in eukaryotes (Brush et al., 1996). By overexpressing the kinase domain of ATM, Morgan and Kastan (1997) demonstrated that IR-induced RPA phosphorylation is dispensable for the S phase checkpoint. Since it is unclear whether similar site(s) of RPA are phosphorylated, the role of ATM-dependent phosphorylation of RPA in DSBR remains to be investigated.
Recently, tyrosine phosphorylation of Rad51 was found to be dependent on c-Abl kinase both in vitro and in vivo (Chen et al., 1999; Yuan et al. 1998). c-Abl is a non-receptor tyrosine kinase involved in multiple signaling pathways. Yuan et al. (1998) showed that c-Abl can phosphorylate Rad51 on tyrosine-54 in vitro. The IR-induced-phosphorylation of Rad51 by c-Abl may inhibit both the binding of Rad51 to DNA and its function in DNA strand exchange reactions. In contrast, Chen et al. (1999) demonstrated that tyrosine-315 in Rad51 is most likely an in vivo target site phosphorylated by c-Abl after IR. Since phosphorylation of Rad51 by c-Abl enhances its interaction with Rad52 and since IR-induced enhancement of tyrosine phosphorylation of Rad51 was not seen in A-T cells, Chen et al. (1999) proposed that ATM and c-Abl facilitate DSBR, at least partly by enhancing the repair protein complex formation. Although these studies provide a plausible mechanism for ATM involvement in DSBR, how the post-translational modification of Rad51 affects its function during HR in vivo remains to be established.
ATM signaling to Mre11/Rad50/NBS1 is likely to be important for both NHEJ and HR. IR-induced phosphorylation of these proteins is minimal in A-T cells (Zhao et al. unpublished). Furthermore, NBS1 appears to be a downstream target of ATM and can be phosphorylated upon IR in an ATM-dependent manner (Yuan et al. in preparation). This observation provides a biochemical link between ATM and proteins essential for NHEJ and HR.
BRCA1, BRCA2 and double-strand break repair
BRCA1 and BRCA2 are two tumor suppressor genes mutated in a significant percentage of patients with familial breast cancer (reviewed by Koller, this issue). Insights into the function of these two proteins are based on studies of knockout phenotypes and identification of proteins that interact with them. These studies indicate that both genes might be involved in DSBR.
BRCA1 has been implicated in the regulation of replication checkpoint (Scully et al., 1997) and transcription-coupled repair (Abbott et al., 1999). Recent evidence suggested that BRCA1 constitutively associates with the Rad50/Mre11/NBS1 complex. However, Mre11/Rad50/NBS1 IRIF form only in the presence of BRCA1 (Zhong et al., 1999). It is likely that other mediators are required in addition to BRCA1 for the assembly of Rad50/Mre11/NBS1 and DNA DSB site-containing complexes.
On the other hand, BRCA2 interacts with Rad51. The BRC repeats located within exon 11 of BRCA2 are necessary and sufficient for binding to Rad51 (Chen et al., 1998; Sharan et al., 1997; Wong et al., 1997). Furthermore, formation of Rad51 IRIF is defective in BRCA2- but not BRCA1-deficient cells, suggesting that BRCA2 might contribute to the assembly of Rad51 repair complex in response to IR-induced DNA damage (Yuan et al., 1999a). The identification of key repair proteins, Rad50 and Rad51, as interacting proteins of the BRCA tumor suppressor proteins provides an explanation for the IR sensitivity of cells harboring mutation in the BRCA genes. Future studies of how DSBR pathways are affected in these cells will provide important information on how these newly evolved tumor suppressors modulate DSBR.
Double-strand break repair proteins in telomere maintenance
Telomeres are special structures at the ends of chromosomes (Griffith et al., 1999). Telomeres protect chromosome ends from DNA degradation and fusion between chromosome ends. In yeast, a number of genes including ScKU70, ScKU80, SIR2, SIR3, SIR4 and RAD50/MRE11/XRS2 are required for telomere maintenance. Ku-deletion mutants display telomere shortening (Boulton and Jackson, 1996), loss of telomere silencing and dissociation of telomere from the nuclear periphery while sir mutants are mainly defective in telomere silencing (Boulton and Jackson, 1998; Laroche et al., 1998; Nugent et al., 1998). Both ku and sir mutants are defective in NHEJ. Recent studies demonstrated that in the presence of even a single DSB, Sir and Ku proteins will be partially dissociated from telomere and are diffusely distributed within the nucleus. Interestingly, the re-localization of Ku and Sir requires the checkpoint proteins Rad9 and Mec1 (Martin et al., 1999; Mills et al., 1999). Unlike ku or sir mutants, rad50, mre11, or xrs2 mutants have shortened telomeres without the loss of telomere silencing. Rad50/Mre11/Xrs2 function epistatically with telomerase and synergistically with Ku and Sir proteins in telomere maintenance (Nugent et al., 1998). Whether telomere maintenance requires the enzymatic or structural role of Rad50/Mre11/Xrs2 is controversial (Boulton and Jackson, 1998; Moreau et al., 1999). Taken together, a number of DSBR proteins are also involved in telomere maintenance through interaction with different telomere binding proteins.
Homologous recombination vs non-homologous end joining in higher eukaryotes
As discussed above, there are different genetic requirements for HR and NHEJ. In budding yeast, DSBs are mainly repaired by HR mechanisms. HR processes have been extensively studied in the context of mitotic recombination, meiotic recombination, and mating type switching (Nickoloff and Hoekstra, 1998; Paques and Haber, 1999; Petukhova et al., 1999). In higher eukaryotes, both HR and NHEJ have been demonstrated in Xenopus laevis (Carroll, 1998; Labhart, 1999). Earlier studies of hamster mutant cell lines led to the identification of important genes in NHEJ. Only recently, genes important for HR were identified. Both XRCC2 and XRCC3 share sequence similarities with Rad51 (Cartwright et al., 1998; Liu et al., 1998). The xrcc3 mutant cell line is sensitive to IR, and fails to form Rad51 IRIF, suggesting that XRCC3 might be a component of the Rad51 repair complex (Bishop et al., 1998).
The role of HR in maintaining the genomic stability in mammals is largely unknown. Using specially designed substrates that when integrated into chromosomal DNA will allow the analysis of recombination products generated after DSBR by either HR or NHEJ, it was shown that HR repair pathways are likely to be equally important in mammalian cells (Liang et al., 1996, 1998). In addition to a repair role, HR activities may also contribute to the loss of heterozygosity (Moynahan and Jasin, 1997; Shulman et al., 1995), which is frequently involved in inactivation of tumor suppressors and formation of tumors. Increasing evidence also indicates the role for HR in sister chromatid recombination (Sonoda et al., 1999). In summary, emerging evidence indicates broader roles for HR pathways in maintaining genomic stability, but the details of most of the processes remain to be elucidated.
Cell cycle-dependent double-strand break repair in eukaryotes
During HR, intact DNA template is used in the repair process. Thus it is likely that HR is a preferred mechanism for repairing DSBs at specific stages of the cell cycle. Hamster mutant cells defective in NHEJ are highly sensitive to IR in G1 and early S phases, but only mildly sensitive in late S and G2 (Lee et al., 1997; Mateos et al., 1994). Rad51-deficient chicken DT40 cells have normal IR-sensitivity in G1 and early S phases but increased sensitivity in late S and G2. In contrast, the Ku70-deficient DT40 cells display increased sensitivity in G1/early S and normal sensitivity in late S/G2 (Takata et al., 1998), confirming the predominant role of NHEJ during G1 and early S and HR during late S and G2. Since it has been hypothesized that Rad52 and Ku are involved in the choice of repair pathways (Van Dyck et al., 1999), it appears to be crucial then to address the activities of Rad52 and Ku during different cell cycle stages. Whether there is a link between DNA repair components and checkpoint molecules leading to specific checkpoint or repair defects also needs to be explored.
Mouse models for DSB-induced cell cycle checkpoint and repair in development and tumorigenesis
While mouse models of defective nucleotide excision repair and mismatch repair have provided crucial information regarding the effects of repair defects on development and tumorigenesis, it is only recently that such models for DSBR were established. We will summarize findings from these mouse models (Table 1).
Knockout mouse models for Mre11/Rad50/NBS1
Mutant ES cells heterozygous for Mre11 disruption have no apparent growth abnormalities in culture. However, homozygous mutant ES cells are non-viable, suggesting that mammalian Mre11 is an essential gene (Xiao and Weaver, 1997). Similarly, no ES cells homozygous for targeted disruption of Rad50 can be isolated. Homozygous mutations of Rad50 in mice result in early embryonic lethality at E6.5, likely due to defects in proliferation (Luo et al., 1999). In agreement with Rad50's role in DSBR, Rad50−/− embryos are extremely sensitive to IR (Luo et al., 1999). Therefore, mutations in Mre11 and Rad50 cause more severe phenotypes in mammals than in yeast. Since RAD50 and MRE11 are not essential genes in yeast, mammalian homologs may have evolved to carry out additional functions. One possibility is that the interaction between this protein complex and recently evolved gene products, such as BRCA1 that plays important roles in both DNA repair and cell cycle checkpoints. As discussed earlier, the Rad50/Mre11/NBS1 complex has both enzymatic and structural functions, knock-in mice carrying defined mutations should provide important information to delineate the functions of Mre11 and Rad50 proteins.
While knockout mice of NBS1 are yet to be reported, NBS1−/− ES cells are viable (R Maser, meeting report; Lavin et al., 1999). All NBS patients studied to date have mutations resulting in the truncation of the NBS1 gene, mostly at the amino terminus (Varon et al., 1998). Based on the clinical and cellular features of NBS, it is clear that NBS1 is involved in DSBR. How is it that loss of Mre11 and Rad50 results in non-viability of ES cells whereas loss of NBS1 does not? One possibility is that Mre11 is capable of forming a complex with Rad50 in NBS cells (Carney et al., 1998). Mre11 nuclease activities are enhanced by the presence of Rad50 although the combination of the three components results in the highest activities (Paull and Gellert, 1999). These results suggest that Mre11/Rad50 complex is functional, albeit sub-optimal, in the absence of NBS1 and may partially explain the ability of NBS−/− cells to survive while Rad50 or Mre11 deletion results in lethality. In addition, Rad50/Mre11 may have an essential function that is independent of NBS1.
Rad51, Rad52, and Rad54
As mentioned previously, Rad51, Rad52, and Rad54 are key players in HR. Homozygous Rad51 mutation leads to early embryonic lethality associated with decreased proliferation in mouse embryos. Rad51−/− ES cells are not viable, indicating that Rad51 is essential for cell viability (Lim and Hasty, 1996; Tsuzuki et al., 1996). In addition, Rad51−/− trophoblasts are hypersensitive to IR and suffer multiple chromosome loss (Lim and Hasty, 1996). Interestingly, the survival of Rad51−/− embryos is extended in a p53 mutant background, probably because checkpoint activation due to chromosomal abnormalities is abrogated in the absence of p53. Fibroblasts derived from double mutant embryos fail to proliferate in tissue culture (Lim and Hasty, 1996). The inviability of the mutant ES cells and MEFs may be due to the increased amounts of oxidative DNA damage that occur during in vitro cell culture.
Rad52−/− mice are viable and healthy (Rijkers et al., 1998). Moreover, Rad52−/− ES cells display only a moderate reduction in homologous recombination, and are not any more sensitive to DSB-inducing agents or MMS and mitomycin C than wildtype cells (Rijkers et al., 1998). The mild phenotype exhibited by mice deficient in Rad52 may be explained by the existence of additional Rad52 homologs.
Rad54 is a member of the SNF2/SWI2 family of DNA-dependent ATPases which might function in the remodeling of chromatin structure. The interaction between Rad51 and Rad54 is increased upon DSB (Tan et al., 1999). In agreement with this observation, DNA damage-induced foci of Rad51 and Rad54 co-localize. Interestingly, the formation of Rad51 IRIF requires Rad54 (Tan et al., 1999). As expected, Rad54−/− ES cells show a reduced frequency of HR and are sensitive to IR, MMS, and mitomycin C. Rad54 null mice are viable but surprisingly, have no apparent defects in spermatogenesis and oogenesis (Essers et al., 1997). This raises the possibility of the existence of homologs of Rad54 that may function specifically during meiosis. There is a meiosis-specific Rad51 homolog, Dmc1, which is specifically expressed in testis (Yoshida et al., 1998). Homozygous deletion of Dmc1 in the mouse results in the failure of homologous chromosome synapsis during meiosis (Pittman et al., 1998; Yoshida et al., 1998).
Knockout mice for DNA-PKcs/Ku70/Ku80
Readers are referred to a recent review for a more detailed description and comparison of mutant mice deficient for different components of the DNA-PK (Smith and Jackson, 1999). Targeted disruption of any component of DNA-PK (DNA-PKcs, Ku70, and Ku80) results in enhanced IR sensitivity and severe combined immune deficiency due to defects in DSBR and V(D)J recombination (Nussenzweig et al., 1996; Zhu et al., 1996; Gu et al., 1997b; Ouyang et al., 1997; Gao et al., 1998a; Taccioli et al., 1998). In Ku80 and DNA-PKcs mutant mice, the development of both T and B lymphocytes is arrested at an early progenitor stage, CD4− CD8− double-negative (DN) and B220+ CD43+, respectively (Nussenzweig et al., 1996; Zhu et al., 1996; Gao et al., 1998a). In Ku70 null mice, however, only B cell development is completely blocked, and T cells develop through the CD4+ CD8+ double-positive (DP) stage and mature into CD4+ CD8− and CD4− CD8+ single-positive (SP) cells, although the numbers are significantly reduced (Gu et al., 1997b; Ouyang et al., 1997). The V(D)J recombination defect in these mutant mice lies in the joining step. Contrary to expectations, both hairpin coding ends and blunt full-length signal ends accumulate in the absence of Ku, indicating that Ku is not required to protect V(D)J recombination intermediates (Zhu et al., 1996; Gu et al., 1997b; Han et al., 1997). In Ku80 and Ku70 mutant mice, both coding-end joining and signal-end joining are severely impaired (Nussenzweig et al., 1996; Zhu et al., 1996; Gu et al., 1997b; Ouyang et al., 1997), while only coding-end joining is defective in DNA-PKcs mutant mice (Gao et al., 1998a; Taccioli et al., 1998).
Recent studies have indicated that immunoglobulin heavy chain class switching, which involves the processing of DSB intermediates (Wuerffel et al., 1997), is highly dependent on DNA-PKcs and Ku (Rolink et al., 1996; Manis et al., 1998; Casellas et al., 1998). Moreover, Ku has been shown to participate in the sequence-specific transcriptional repression of a specific promoter and it appears that DNA-PKcs is required for such repression (Giffin et al., 1996).
In addition to their differential effects on coding- and signal-end joint formation, Ku80−/− and Ku70−/− mice can be distinguished from DNA-PKcs−/− mice based on their growth. Ku80- and Ku70- deficiency results in growth defects (Nussenzweig et al., 1996; Zhu et al., 1996; Gu et al., 1997b; Ouyang et al., 1997), and Ku80-deficient mice show signs of aging early in life and their cells are likely to undergo premature replicative senescence (P Hasty, personal communication). In contrast, DNA-PKcs−/− cells and mice have apparently normal growth (Gao et al., 1998a; Taccioli et al., 1998). The difference between Ku and DNA-PKcs deficient mice indicates that Ku may have unique functions independent of DNA-PKcs.
Mice deficient in Ku70 undergo T cell development but form lymphomas. This may be due to defective processing of T cell receptor (TCR) gene rearrangement in Ku70−/− cells, which are defective in DSBR, leading to aberrant genetic changes and tumor development. Indeed, Ku70−/− but not Ku80−/− and SCID mice develop spontaneous thymic and disseminated T cell lymphomas at a mean age of 6 months with DP tumor cells (Gu et al., 1997b; Li et al., 1998). Taken together, mice carrying a deletion of an individual component of the DNA-PK complex provide new insights into their shared and distinct functions and lay the foundation of future mechanistic studies.
DNA ligase IV and XRCC4
DNA ligase IV forms a stable complex with XRCC4 and its activity is stimulated by XRCC4 (Grawunder et al., 1997). The complex of ligase IV and XRCC4 participates in the final step of V(D)J recombination and DNA-end joining (Wilson et al., 1997). Inactivation of either ligase IV or XRCC4 in mice leads to late embryonic lethality (Frank et al., 1998; Gao et al., 1998b). Lymphocyte development is blocked and both coding-end and signal-end joining in V(D)J recombination are defective in these mice (Frank et al., 1998, Gao et al., 1998b). Embryonic fibroblasts (MEFs) derived from these mice proliferate poorly, enter senescence prematurely, and are extremely sensitive to IR (Frank et al., 1998; Gao et al., 1998b). All these phenotypes, except the embryonic lethality, are similar to those seen in Ku-deficient cells, suggesting that Ku, XRCC4 and ligase IV may be involved in the same end-joining process. These results in mammalian cells recapitulate the epistatic relationships established in yeast. However, it should be noted that in yeast Ku is involved in telomere maintenance but ligase IV does not. Interestingly, in addition to defects in lymphogenesis, XRCC4 and ligase IV deficient embryos suffer massive cell death in newly generated, postmitotic neuronal populations (Gao et al., 1998b). It was postulated that increased death of early postmitotic XRCC4−/− and ligase IV−/− neurons is due to increased susceptibility to excess DSBs during neuronal development, which can not be repaired efficiently in the absence of these factors (Gao et al., 1998b). These studies underscore the importance of NHEJ in V(D)J recombination and DSBR as well as the aforementioned roles of Ku, ligase IV, and XRCC4 in cellular proliferation.
Brca1 and Brca2
BRCA1 and BRCA2 have been shown to interact with Rad50 and Rad51, respectively (see above), implying that they participate in DSBR. Loss of Brca1 leads to embryonic death associated with hypoproliferation in mice (Hakem et al., 1996; Liu et al., 1996; Gowen et al., 1996; Ludwig et al., 1997; Xu et al., 1999b). Brca1 mutants also show decreased expression of mdm2, a negative regulator of p53 activity, and dramatically increased expression of p21, a target for p53 transcriptional activation (Hakem et al., 1996). Deletion of Brca1 exon 11 is associated with G2/M checkpoint defects in MEFs, genetic instability characterized as chromosomal structural aberrations and aneuploidy, and abnormal centrosome duplication, indicating the important role of Brca1 in maintaining genetic stability and inhibiting tumorigenesis (Xu et al., 1999b). Different Brca2 mutant alleles have been generated and have different phenotypes (Sharan et al., 1997; Ludwig, et al. 1997; Suzuki et al, 1997; Connor et al., 1997; Friedman et al., 1998). Truncation of Brca2 prior to the first BRC repeat in exon 11 leads to early embryonic lethality at E6.5 ∼ E9.5 (Sharan et al., 1997; Ludwig et al., 1997; Suzuki et al., 1997). Disruption of Brca2 after the third BRC repeat results in ∼50% live birth of homozygous mutants in the mixed genetic background (Friedman et al., 1998). Disruption after the seventh BRC repeat leads to enhanced embryonic survival and results in ∼30% live homozygous mutants in the 129/B6/DBA background (Connor et al., 1997). Brca2 mutant mice survive to adulthood, are growth retarded, infertile, and develop lethal thymic lymphomas by 5 months of age (Connor et al., 1997; Friedman et al., 1998). Brca2−/− embryos are hypersensitive to IR (Sharan et al., 1997). MEFs harboring mutations disrupting Brca2 after the third or seventh BRC repeat grow poorly in culture, arrest in G1 and G2/M, have elevated p53 and p21 protein levels, are extremely sensitive to UV and MMS, and are only mildly sensitive to IR (Connor et al., 1997; Friedman et al., 1998; Patel et al., 1998). The G1/S and G2/M DNA damage checkpoints as well as the replication checkpoint are intact in these MEFs (Patel et al., 1998). Moreover, these MEFs display higher spontaneous chromosomal structural aberrations, including chromatid breaks and chromatid exchange (Patel et al., 1998).
The elevated p53 and p21 levels in both Brca1 and Brca2 mutant embryos are presumably a consequence of the increased genetic instability. Since mutations in either p53 or p21 prolong the survival of Brca1−/− and Brca2−/− embryos, it appears that activation of the p53/p21 pathway by spontaneous DNA damage plays a part in the impaired cellular proliferation caused by the lack of Brca1 and Brca2 (Hakem et al., 1997; Ludwig et al., 1997). Indeed, when Brca2−/− MEFs are transduced with retroviruses encoding dominant-negative (DN) p53 mutants, they overcome the growth arrest and lose the checkpoints in response to IR and nocodazole (Lee et al., 1999). Thymic lymphomas from Brca2−/− mice have defective checkpoints and harbor p53 mutations which have been shown to reverse the proliferative defects in Brca2−/− MEFs (Lee et al., 1999). Thus inactivation of p53 cooperates with Brca2 deficiency to promote neoplastic transformation. Recently, the importance of Brca1 in mammary gland development and tumorigenesis was demonstrated by the generation of Brca1 conditional knockout mice using the Cre-loxP system (Xu et al., 1999a). Deletion of Brca1 exon 11 from mammary epithelial cells causes abnormal ductal development accompanied by increased apoptosis. Mammary tumors develop after a long latency period (10 – 13 months), and are associated with genomic instability, including aneuploidy and chromosomal rearrangements. Detailed analyses of genomic alterations and chromosomal aberrations in the mammary tumors indicate the involvement of p53 dysfunction. Consistent with this notion, loss of p53 accelerates tumor formation in the Brca1 conditional knockout mice resulting in a much shorter latency period (6 – 8 months).
Inactivation of Atr results in early embryonic lethality, and chromosome instability. Furthermore, Atr−/− cells fail to proliferate in vitro (EJ Brown, personal communication). Thus, the phenotype of the Atr−/− mice overlaps significantly with those of the Brca1 and Brca2 knockouts. These results together with the biochemical properties of Rad50 (including its interaction with BRCA1) suggest that there may be functional interactions among BRCA1, BRCA2, Rad50, and Atr.
Atm in mouse development, V(D)J recombination, and tumorigenesis
Atm deficient mice reiterate most of the phenotypes of A-T patients, including growth retardation, neuronal abnormalities, immunodeficiency, meiotic defects, and cancer predisposition (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996).
Unlike A-T patients, no overt ataxia is detected in Atm−/− mice (Barlow et al., 1996; Xu et al., 1996). However, they show impaired performance in specific tests for motor coordination (Barlow et al., 1996), indicating the presence of potential neuronal defects in Atm−/− mice. Atm deficiency causes severe degeneration of dopaminergic nigro-striatal neurons in the CNS of older mice (Eilam et al., 1998). Surprisingly, many regions of the developing neonatal CNS in Atm mutants fail to undergo cell death and show reduced p53 induction after IR (Herzog et al., 1998). Based on this observation, it was hypothesized that Atm deficiency may selectively compromise the death of certain damaged neurons and lead to neuronal deterioration later in life as the damages accumulate.
T and B cell development
Lymphoid tissues of Atm−/− mice are generally smaller and contain fewer cells when compared to controls. The absolute numbers of DP and SP thymocytes are fewer, resulting in a large reduction of T cells in the peripheral lymphoid organs. However, the development of B cells is not altered significantly in Atm−/− mice (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996). The abnormal development of T cells in Atm−/− mice could be due to impaired cell cycle checkpoint control during V(D)J recombination which appears to be restricted to the G0/G1 phase (Lin and Desiderio, 1995). In the absence of Atm, lymphocytes might enter S phase before V(D)J recombination is completed, resulting in a reduction of productive V(D)J recombination events and mature lymphocytes. Alternatively, the interaction between Atm and repair proteins, such as the Rad50/Mre11/NBS1 complex (see above), may be required for normal T cell maturation.
Atm mutant mice are infertile due to the lack of mature gametes (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996). The development of Atm−/− spermatocytes is arrested at the zygotene/pachytene stage of prophase I (Xu et al., 1996). Moreover, aberrant chromosomal synapses in Atm−/− spermatocytes are followed by chromosome fragmentation, leading to nuclear degeneration and cellular death (Xu et al., 1996). Interestingly, Atm deficiency is associated with a failure to properly assemble Atr as well as Rad51 and Dmc1 onto chromosomal axes in leptonema of meiosis I (Barlow et al., 1997, 1998). Meiosis I progresses to later stages and is only partially rescued by p53 or p21 deficiency (Barlow et al., 1997). Moreover, assembly of Rad51 onto chromosomal axes remains defective in these double mutant mice (Barlow et al., 1997). Whether Rad51 mislocalization is directly due to the absence of Atm or a consequence of meiotic failure is not clear. Recently, leptotene/zygotene spermatocytes of Atm-null mice were found to display aberrant telomere clustering and elevated interactions between telomeres and nuclear matrix (Pandita et al., 1999). These observations support a possible role for Atm in telomere maintenance.
Atm−/− mice develop malignant thymic lymphomas of DP origin by 4 months of age (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996). However, no increased frequency of other tumor types is seen in Atm mutant mice, possibly due to the early lethality caused by lymphomas. Aberrant chromosomal rearrangements, translocations and insertions, were observed in tumor cells, and in one case, the rearrangement involved the TCRβ locus (Barlow et al., 1996; Xu et al., 1996). These observations suggest that lymphoma development in the absence of Atm is associated with the processing of DNA DSBs during V(D)J recombination. Indeed, in the RAG1-deficient background, where V(D)J recombination does not occur, no tumors develop in the Atm−/− mice (Liao & Van Dyke, 1999). Because V(D)J recombination is dispensable for thymoma development in p53−/− mice (Nacht and Jacks, 1998; Liao et al., 1998), Atm and p53 appear to suppress thymoma through different mechanisms. Consistent with this notion, Atm/p53 double mutants exhibit a dramatic acceleration of tumor development relative to singly null mutants (Westphal et al., 1997; Xu et al., 1998), indicating a cooperative role between Atm and p53 in tumor suppression.
Cellular proliferation and checkpoint response
Atm−/− mice are growth retarded. Atm−/− fibroblasts grow poorly in culture, undergo premature senescence, and show inefficient G1/S progression following serum stimulation (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996). Interestingly, Atm−/− cells show an increased constitutive level of p21, which may account for the observed growth arrest (Xu and Baltimore, 1996). In agreement with this notion, Atm/p21 as well as Atm/p53 double mutant cells proliferate normally and the p21 level is not elevated in Atm/p53 double mutant cells (Xu et al., 1998). Taken together, the activated p53/p21 pathway accounts for the proliferative defects seen in the absence of Atm.
Atm−/− ES cells, MEFs and thymocytes are sensitive to IR, and have defective G1 and S checkpoints in response to IR, correlating with an impaired p53 up-regulation (Barlow et al., 1996; Elson et al., 1996; Xu and Baltimore, 1996; Westphal et al., 1997). However, Atm−/− thymocytes display a partial resistance to IR-induced apoptosis in vivo and in vitro. The additional loss of p53 renders them completely resistant, a phenotype that is shared by p53-null thymocytes (Xu and Baltimore, 1996; Westphal et al., 1997). This suggests that both Atm-dependent and Atm-independent up-regulation of p53 may mediate the IR-induced apoptotic responses of thymocytes. Therefore, damaged thymocytes survive in the absence of Atm and accumulate more genetic mutations, which eventually lead to neoplasms as Atm−/− mice are prone to the development of thymic tumors.
Summary and perspectives
Key players in cell cycle checkpoint control and DSB repair machinery are evolutionarily conserved. New family members in these cellular processes have been identified in mammals, e.g., the Rad51 recombinase gene family. Surprisingly, despite the existence of multiple family members, deletion of one of the members often results in embryonic lethality, e.g., Rad51, suggesting that each member of the protein family has a specialized function. Additionally, many newly evolved genes such as p53 and BRCA have critical regulatory functions. Studies to date indicate intricate relationships among the DSBR machinery, tumor suppressors and cell cycle regulatory proteins (Figure 3). Mutation in any of the key players may disrupt the intricate network and thus provide a basis for the pleiotropic and severe phenotypes of the mutant cells.
In addition to nucleolytic processing of damaged DNA, Mre11/Rad50/NBS1 proteins are also involved in cell cycle checkpoint and chromosome homeostasis. While this protein complex exists in both undamaged and IR-treated cells, enhanced phosphorylation of these proteins upon DNA damage (Dong et al., 1999) is likely to be required for recruiting other repair proteins. Recent studies indicate that BRCA1 is required for IR-induced Mre11 IRIF formation (Zhong et al., 1999); on the other hand, BRCA2 is required for the formation of Rad51 IRIF (Yuan et al., 1999a). These results strongly suggest that complex signaling pathways are initiated upon DNA damage to co-ordinate multiple cellular processes. Six Rad genes have been identified in yeast that are required for sensing/processing/signal transduction. It is unclear whether there are biochemical and functional links among hRad1/Rad9/Hus1, Rad17 protein complexes, and Mre11/Rad50/NBS1 in mammals. In addition, how ATM and ATR are activated upon DNA damage and whether and how the kinase cascades link BRCA proteins to the Mre11/Rad50/NBS1 protein complex remain to be addressed.
The presence of multiple cellular substrates is likely to underline the pleiotropic function of ATM. One of its substrates, p53, is a potent transcription activator of the cyclin-dependent kinase inhibitor, p21. The release of the BRCA1-associated co-repressor, CtIp, upon DNA damage also plays an important role in the induction of p21 (Figure 3; Li et al., 1999). In addition to a crucial role in G1/S checkpoint control, p53 regulates the duplication of the centrosome, an event which is central to the maintenance of a 2N genome in mammalian cells (Fukasawa et al., 1996). Biochemical mechanisms involved in this process remain to be elucidated. Two key checkpoint kinases, Chk2 and Chk1, are subjected to regulation by ATM and ATR (Matsuoka et al., 1998; Dasika et al. in preparation). The biochemical basis of S and G2/M checkpoint regulation is largely unknown. These checkpoints are of paramount importance in the maintainance of high fidelity DNA replication and chromosome segregation. Since aneuploidy is one of the hallmarks of tumor cells, delineation of players in the mammalian G2/M checkpoint and spindle assembly checkpoint will provide a better understanding as to how such genomic catastrophes are prevented in normal cells.
Genetic manipulation using knockout technology is time consuming in mammals, but mutant cells, especially conditional mutants generated through the now widely used Cre-lox P system, will serve as invaluable tools in delineating protein networks in DSB repair, cell cycle checkpoint regulation and signaling pathways. Identification of such protein networks will provide a rational basis for drug design that may sensitize cancer cells to radiation and chemotherapy as well as for cancer prevention.
BRCA1 carboxyl terminus
double-strand break repair
ionizing radiation-induced immunofluorescent foci
Nijmegen breakage syndrome
non-homologous end joining
radiation-resistant DNA synthesis
severe combined immune deficient
S phase DNA damage checkpoint
T cell receptor
X-ray repair cross complementation
Mouse homologs are represented in lower case
Abbott DW, Thompson ME, Robinson-Benion C, Tomlinson G, Jensen RA and Holt JT. . 1999 J. Biol. Chem. 274: 18808–18812.
Banin S, Moyal L, Shieh S, Taya Y, Anderson CW, Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y and Ziv Y. . 1998 Science 281: 1674–1677.
Bao S, Chang MS, Auclair D, Sun Y, Wang Y, Wong WK, Zhang J, Liu Y, Qian X, Sutherland R, Magi-Galluzi C, Weisberg E, Cheng EY, Hao L, Sasaki H, Campbell MS, Kraeft SK, Loda M, Lo KM and Chen LB. . 1999 Cancer Res. 59: 2023–2028.
Barlow C, Hirotsune S, Paylor R, Liyanage M, Eckhaus M, Collins F, Shiloh Y, Crawley JN, Ried T, Tagle D and Wynshaw-Boris A. . 1996 Cell 86: 159–171.
Barlow C, Liyanage M, Moens PB, Deng CX, Ried T and Wynshaw-Boris A. . 1997 Nat. Genet. 17: 462–466.
Barlow C, Liyanage M, Moens PB, Tarsounas M, Nagashima K, Brown K, Rottinghaus S, Jackson SP, Tagle D, Ried T and Wynshaw-Boris A. . 1998 Development 125: 4007–4017.
Baumann P and West SC. . 1998a Proc. Natl. Acad. Sci. USA 95: 14066–14070.
Baumann P and West SC. . 1998b Trends. Biochem. Sci. 23: 247–251.
Benson FE, Baumann P and West SC. . 1998 Nature 391: 401–404.
Bishop DK, Ear U, Bhattacharyya A, Calderone C, Beckett M, Weichselbaum RR and Shinohara A. . 1998 J. Biol. Chem. 273: 21482–21488.
Bliss TM and Lane DP. . 1997 J. Biol. Chem. 272: 5765–5773.
Boddy MN, Furnari B, Mondesert O and Russell P. . 1998 Science 280: 909–912.
Boulton SJ and Jackson SP. . 1996 EMBO J 15: 5093–5103.
Boulton SJ and Jackson SP. . 1998 EMBO J. 17: 1819–1828.
Brondello JM, Boddy MN, Furnari B and Russell P. . 1999 Mol. Cell. Biol. 19: 4262–4269.
Brown AL, Lee CH, Schwarz JK, Mitiku N, Piwnica-Worms H and Chung JH. . 1999 Proc. Natl. Acad. Sci. USA 96: 3745–3750.
Brush GS, Morrow DM, Hieter P and Kelly TJ. . 1996 Proc. Natl. Acad. Sci. USA 93: 15075–15080.
Burma S, Kurimasa A, Xie G, Taya Y, Araki R, Abe M, Crissman HA, Ouyang H, Li GC and Chen DJ. . 1999 J. Biol. Chem. 274: 17139–17143.
Callebaut I and Mornon JP. . 1997 FEBS Lett. 400: 25–30.
Canman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, Sakaguchi K, Appella E, Kastan MB and Silicianco JD. . 1998 Science 281: 1677–1679.
Cao L, Alani E and Kleckner N. . 1990 Cell 61: 1089–1101.
Carney JP, Maser RS, Olivares H, Davis EM, Le Beau M, Yates 3rd JR, Hays L, Morgan WF and Petrini JH. . 1998 Cell 93: 477–486.
Carroll D. . 1998 Modes of DNA repair in Xenopus oocytes, eggs, and extracts. In: DNA Damage and Repair. Humana Press Inc., Totowa, New Jersey.
Cartwright R, Tambini CE, Simpson PJ and Thacker J. . 1998 Nucleic Acids Res. 26: 3084–3089.
Cary RB, Peterson SR, Wang J, Bear DG, Bradbury EM and Chen DJ. . 1997 Proc. Natl. Acad. Sci. USA 94: 4267–4272.
Casellas R, Nussenzweig A, Wuerffel R, Pelanda R, Reichlin A, Suh H, Qin XF, Besmer E, Kenter A, Rajewsky K and Nussenzweig MC. . 1998 EMBO J. 17: 2404–2411.
Chaturvedi P, Eng WK, Zhu Y, Mattern MR, Mishra R, Hurle MR, Zhang X, Annan RS, Lu Q, Faucette LF, Scott GF, Li X, Carr SA, Johnson RK, Winkler JD and Zhou BB. . 1999 Oncogene 18: 4047–4054.
Chen C and Kolodner RD. . 1999 Nat. Genet 23: 81–85.
Chen G, Yuan SS, Liu W, Xu Y, Trujillo K, Song B, Cong F, Goff SP, Wu Y, Arlinghaus R, Baltimore D, Gasser PJ, Park MS, Sung P and Lee EY. . 1999 J. Biol. Chem. 274: 12748–12752.
Chen PL, Chen CF, Chen Y, Xiao J, Sharp ZD and Lee WH. . 1998 Proc. Natl. Acad. Sci. USA 95: 5287–5292.
Clever B, Interthal H, Schmuckli-Maurer J, King J, Sigrist M and Heyer WD. . 1997 EMBO J. 16: 2535–2544.
Connelly JC, Kirkham LA and Leach DR. . 1998 Proc. Natl. Acad. Sci. USA 95: 7969–7974.
Connelly JC and Leach DR. . 1996 Genes Cells 1: 285–291.
Connor F, Bertwistle D, Mee PJ, Ross GM, Swift S, Grigorieva E, Tybulewicz VL and Ashworth A. . 1997 Nat. Genet 17: 423–430.
Critchlow SE and Jackson SP. . 1998 Trends Biochem. Sci. 23: 394–398.
Dolganov GM, Maser RS, Novikov A, Tosto L, Chong S, Bressan DA and Petrini JH. . 1996 Mol. Cell. Biol. 16: 4832–4841.
Dong Z, Zhong Q and Chen PL. . 1999 J. Biol. Chem. 274: 19513–19516.
Dulic V, Kaufmann WK, Wilson SJ, Tlsty TD, Lees E, Harper JW, Elledge SJ and Reed SI. . 1994 Cell 76: 1013–1023.
Dynan WS and Yoo S. . 1998 Nucleic Acids Res. 26: 1551–1559.
Eilam R, Peter Y, Elson A, Rotman G, Shiloh Y, Groner Y and Segal M. . 1998 Proc. Natl. Acad. Sci. USA 95: 12653–12656.
Eisen JA, Sweder KS and Hanawalt PC. . 1995 Nucleic Acids Res. 23: 2715–2723.
Elson A, Wang Y, Daugherty CJ, Morton CC, Zhou F, Campos-Torres J and Leder P. . 1996 Proc. Natl. Acad. Sci. USA 93: 13084–13089.
Essers J, Hendriks RW, Swagemakers SM, Troelstra C, de Wit J, Bootsma D, Hoeijmakers JH and Kanaar R. . 1997 Cell 89: 195–204.
Frank KM, Sekiguchi JM, Seidl KJ, Swat W, Rathbun GA, Cheng HL, Davidson L, Kangaloo L and Alt FW. . 1998 Nature 396: 173–177.
Friedman LS, Thistlethwaite FC, Patel KJ, Yu VP, Lee H, Venkitaraman AR, Abel KJ, Carlton MB, Hunter SM, Colledge WH, Evans MJ and Ponder BA. . 1998 Cancer Res. 58: 1338–1343.
Fukasawa K, Choi T, Kuriyama R, Rulong S and Vande Woude GF. . 1996 Science 271: 1744–1747.
Gao Y, Chaudhuri J, Zhu C, Davidson L, Weaver DT and Alt FW. . 1998a Immunity 9: 367–376.
Gao Y, Sun Y, Frank KM, Dikkes P, Fujiwara Y, Seidl KJ, Sekiguchi JM, Rathbun GA, Swat W, Wang J, Bronson RT, Malynn BA, Bryans M, Zhu C, Chaudhuri J, Davidson L, Ferrini R, Stamato T, Orkin SH, Greenberg ME and Alt FW. . 1998b Cell 95: 891–902.
Garkavtsev I, Grigorian IA, Ossovskaya VS, Chernov MV, Chumakov PM and Gudkov AV. . 1998 Nature 391: 295–298.
Gasior SL, Wong AK, Kora Y, Shinohara A and Bishop DK. . 1998 Genes. Dev. 12: 2208–2221.
Giffin W, Torrance H, Rodda DJ, Prefontaine GG, Pope L and Hache RJ. . 1996 Nature 380: 265–268.
Gong JG, Costanzo A, Yang HQ, Melino G, Kaelin WG Jr, Levrero M and Wang JY. . 1999 Nature 399: 806–809.
Gottlieb TM and Jackson SP. . 1993 Cell 72: 131–142.
Gowen LC, Johnson BL, Latour AM, Sulik KK and Koller BH. . 1996 Nat. Genet. 12: 191–194.
Gravel S, Larrivee M, Labrecque P and Wellinger RJ. . 1998 Science 280: 741–744.
Grawunder U, Wilm M, Wu X, Kulesza P, Wilson TE, Mann M and Lieber MR. . 1997 Nature 388: 492–495.
Grawunder U, Zimmer D, Kulesza P and Lieber MR. . 1998a J. Biol. Chem. 273: 24708–24714.
Grawunder U, Zimmer D and Leiber MR. . 1998b Curr. Biol. 8: 873–876.
Grenon M, Tillit J, Piard K, Baldacci G and Francesconi S. . 1999 Mol. Gen. Genet. 260: 522–534.
Griffith JD, Comeau L, Rosenfield S, Stansel RM, Bianchi A, Moss H and de Lange T. . 1999 Cell 97: 503–514.
Gu Y, Jin S, Gao Y, Weaver DT and Alt FW. . 1997a Proc. Natl. Acad. Sci. USA 94: 8076–8081.
Gu Y, Seidl KJ, Rathbun GA, Zhu C, Manis JP, van der Stoep N, Davidson L, Cheng HL, Sekiguchi JM, Frank K, Stanhope-Baker P, Schlissel MS, Roth DB and Alt FW. . 1997b Immunity 7: 653–665.
Haber JE. . 1998 Cell 95: 583–586.
Hakem R, de la Pompa JL, Elia A, Potter J and Mak TW. . 1997 Nat. Genet. 16: 298–302.
Hakem R, de la Pompa JL, Sirard C, Mo R, Woo M, Hakem A, Wakeham A, Potter J, Reitmair A, Billia F, Firpo E, Hui CC, Roberts J, Rossant J and Mak TW. . 1996 Cell 85: 1009–1023.
Han JO, Steen SB and Roth DB. . 1997 Mol. Cell. Biol. 17: 2226–2234.
Hartwell LH and Kastan MB. . 1994 Science 266: 1821–1828.
Herzog KH, Chong MJ, Kapsetaki M, Morgan JI and McKinnon PJ. . 1998 Science 280: 1089–1091.
Ivanov EL, Korolev VG and Fabre F. . 1992 Genetics 132: 651–664.
Jeggo PA. . 1998 Mutant rodent cells defective in DNA double-strand break repair. In: DNA Damage and Repair. Humana Press Inc., Totowa, New Jersey.
Jiang H, Xie Y, Houston P, Stemke-Hale K, Mortensen UH, Rothstein R and Kodadek T. . 1996 J. Biol. Chem. 271: 33181–33186.
Kaelin WG Jr. . 1999 J. Natl. Cancer. Inst. 91: 594–598.
Kastan MB, Zhan Q, el-Deiry WS, Carrier F, Jacks T, Walsh WV, Plunkett BS, Vogelstein B and Fornace Jr. AJ. . 1992 Cell 71: 587–597.
Khanna KK, Gatti R, Concannon P, Weemaes CMR, Hoekstra MF, Lavin M and D'Andrea A. . 1998 Cellular responses to DNA Damage and the Human Chromosome Instability Syndromes. Humana Press, San Diego.
Kuerbitz SJ, Plunkett BS, Walsh WV and Kastan MB. . 1992 Proc. Natl. Acad. Sci. USA 89: 7491–7495.
Labhart P. . 1999 Mol. Cell. Biol 19: 2585–2593.
Larner JM, Lee H and Hamlin JL. . 1994 Mol. Cell. Biol. 14: 1901–1908.
Larner JM, Lee H and Hamlin JL. . 1997 Cancer Surv. 29: 25–45.
Larner JM, Lee H, Little RD, Dijkwel PA, Schildkraut CL and Hamlin JL. . 1999 Nucleic Acids Res. 27: 803–809.
Laroche T, Martin SG, Gotta M, Gorham HC, Pryde FE, Louis EJ and Gasser SM. . 1998 Curr. Biol. 8: 653–656.
Lavin MF, Concannon P and Gatti RA. . 1999 Cancer Res. 59: 3845–3849.
Lee CW and La Thangue NB. . 1999 Oncogene 18: 4171–4181.
Lee H, Trainer AH, Friedman LS, Thistlethwaite FC, Evans MJ, Ponder BA and Venkitaraman AR. . 1999 Mol. Cell. 4: 1–10.
Lee SE, Mitchell RA, Cheng A and Hendrickson EA. . 1997 Mol. Cell. Biol. 17: 1425–1433.
Leuther KK, Hammarsten O, Kornberg RD and Chu G. . 1999 EMBO. J. 18: 1114–1123.
Levine AJ. . 1997 Cell. 88: 323–331.
Li GC, Ouyang H, Li X, Nagasawa H, Little JB, Chen DJ, Ling CC, Fuks Z and Cordon-Cardo C. . 1998 Mol. Cell. 2: 1–8.
Li S, Chen PL, Subramanian T, Chinnadurai G, Tomlinson G, Osborne CK, Sharp ZD and Lee WH. . 1999 J. Biol. Chem. 274: 11334–11338.
Li Z, Otevrel T, Gao Y, Cheng HL, Seed B, Stamato TD, Taccioli GE and Alt FW. . 1995 Cell. 83: 1079–1089.
Liang F, Han M, Romanienko PJ and Jasin M. . 1998 Proc. Natl. Acad. Sci. USA 95: 5172–5177.
Liang F, Romanienko PJ, Weaver DT, Jeggo PA and Jasin M. . 1996 Proc. Natl. Acad. Sci. USA 93: 8929–8933.
Liao MJ and Van Dyke T. . 1999 Genes Dev. 13: 1246–1250.
Liao MJ, Zhang XX, Hill R, Gao J, Qumsiyeh MB, Nichols W and Van Dyke T. . 1998 Mol. Cell. Biol. 18: 3495–3501.
Lim DS and Hasty P. . 1996 Mol. Cell. Biol. 16: 7133–7143.
Lin WC and Desiderio S. . 1995 Immunol. Today 16: 279–289.
Lindsay HD, Griffiths DJ, Edwards RJ, Christensen PU, Murray JM, Osman F, Walworth N and Carr AM. . 1998 Genes Dev. 12: 382–395.
Liu Y, Li MJ, Lee EY-HP and Maizels N. . 1999 Curr. Biol. 9: 975–978.
Liu CY, Flesken-Nikitin A, Li S, Zeng Y and Lee WH. . 1996 Genes Dev. 10: 1835–1843.
Liu N, Lamerdin JE, Tebbs RS, Schild D, Tucker JD, Shen MR, Brookman KW, Siciliano MJ, Walter CA, Fan W, Narayana LS, Zhou ZQ, Adamson AW, Sorensen KJ, Chen DJ, Jones NJ and Thompson LH. . 1998 Mol. Cell. 1: 783–793.
Liu VF and Weaver DT. . 1993 Mol. Cell. Biol. 13: 7222–7231.
Lopez-Girona A, Furnari B, Mondesert O and Russell P. . 1999 Nature 397: 172–175.
Ludwig T, Chapman DL, Papaioannou VE and Efstratiadis A. . 1997 Genes Dev. 11: 1226–1241.
Luo G, Yao MS, Bender CF, Mills M, Bladl AR, Bradley A and Petrini JH. . 1999 Proc. Natl. Acad. Sci. USA 96: 7376–7381.
Manis JP, Gu Y, Lansford R, Sonoda E, Ferrini R, Davidson L, Rajewsky K and Alt FW. . 1998 J. Exp. Med. 187: 2081–2089.
Martin SG, Laroche T, Suka N, Grunstein M and Gasser SM. . 1999 Cell. 97: 621–633.
Maser RS, Monsen KJ, Nelms BE and Petrini JH. . 1997 Mol. Cell. Biol. 17: 6087–6096.
Mateos S, Slijepcevic P, MacLeod RA and Bryant PE. . 1994 Mutat. Res. 315: 181–187.
Matsuoka S, Huang M and Elledge SJ. . 1998 Science 282: 1893–1897.
Meyn MS. . 1995 Cancer Res. 55: 5991–6001.
Mills KD, Sinclair DA and Guarente L. . 1999 Cell. 97: 609–620.
Moreau S, Ferguson JR and Symington LS. . 1999 Mol. Cell. Biol. 19: 556–566.
Morgan SE and Kastan MB. . 1997 Cancer Res. 57: 3386–3389.
Moynahan ME and Jasin M. . 1997 Proc. Natl. Acad. Sci. USA 94: 8988–8993.
Nacht M and Jacks T. . 1998 Cell. Growth Differ. 9: 131–138.
Nelms BE, Maser RS, MacKay JF, Lagally MG and Petrini JH. . 1998 Science 280: 590–592.
New JH, Sugiyama T, Zaitseva E and Kowalczykowski SC. . 1998 Nature 391: 407–410.
Nickoloff JA and Hoekstra MF. . 1998 Double-strand break and recombinational repair in Saccharomyces cerevisiae. In: DNA Damage and Repair. Humana Press Inc., Totowa, New Jersey.
Nugent CI, Bosco G, Ross LO, Evans SK, Salinger AP, Moore JK, Haber JE and Lundblad V. . 1998 Curr. Biol. 8: 657–660.
Nussenzweig A, Chen C, da Costa Soares V, Sanchez M, Sokol K, Nussenzweig MC and Li GC. . 1996 Nature 382: 551–555.
Ouyang H, Nussenzweig A, Kurimasa A, Soares VC, Li X, Cordon-Cardo C, Li W, Cheong N, Nussenzweig M, Iliakis G, Chen DJ and Li GC. . 1997 J. Exp. Med. 186: 921–929.
Paillard S and Strauss F. . 1991 Nucleic Acids Res. 19: 5619–5624.
Painter RB. . 1986 Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 49: 771–781.
Painter RB and Young BR. . 1980 Proc. Natl. Acad. Sci. USA 77: 7315–7317.
Pandita TK, Westphal CH, Anger M, Sawant SG, Geard CR, Pandita RK and Scherthan H. . 1999 Mol. Cell. Biol. 19: 5096–5105.
Paques F and Haber JE. . 1999 Microbiol. Mol. Biol. Rev. 63: 349–404.
Passalaris TM, Benanti JA, Gewin L, Kiyono T and Galloway DA. . 1999 Mol. Cell. Biol. 19: 5872–5881.
Patel KJ, Vu VP, Lee H, Corcoran A, Thistlethwaite FC, Evans MJ, Colledge WH, Friedman LS, Ponder BA and Venkitaraman AR. . 1998 Mol. Cell. 1: 347–357.
Paull TT and Gellert M. . 1998 Mol. Cell. 1: 969–979.
Paull TT and Gellert M. . 1999 Genes Dev. 13: 1276–1288.
Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS and Piwnica-Worms H. . 1997 Science 277: 1501–1505.
Petrini JH, Walsh ME, DiMare C, Chen XN, Korenberg JR and Weaver DT. . 1995 Genomics 29: 80–86.
Petukhova G, Lee EY-HP and Sung P. . 1999 DNA end-processing and heteroduplex DNA formation during recombinational repair of DNA double-strand breaks. In: DNA Damage and Repair. Humana Press Inc., Totowa, New Jersey.
Petukhova G, Stratton S and Sung P. . 1998 Nature 393: 91–94.
Pittman DL, Cobb J, Schimenti KJ, Wilson LA, Cooper DM, Brignull E, Handel MA and Schimenti JC. . 1998 Mol. Cell. 1: 697–705.
Ramsden DA and Gellert M. . 1998 EMBO J. 17: 609–614.
Rathmell WK and Chu G. . (1998 Mechanisms for DNA double-strand break repair in eukaryotes. In: DNA Damage and Repair. Humana Press Inc., Totowa, New Jersey.
Reed SI, Bailly E, Dulic V, Hengst L, Resnitzky D and Slingerland J. . 1994 J. Cell. Sci. Suppl. 18: 69–73.
Rhind N, Furnari B and Russell P. . 1997 Genes Dev. 11: 504–511.
Rhind N and Russell P. . 1998 Curr. Opin. Cell. Biol. 10: 749–758.
Rijkers T, Van Den Ouweland J, Morolli B, Rolink AG, Baarends WM, Van Sloun PP, Lohman PH and Pastink A. . 1998 Mol. Cell. Biol. 18: 6423–6429.
Rolink A, Melchers F and Andersson J. . 1996 Immunity 5: 319–330.
Rotman G, Shiloh Y. . 1998 Hum. Mol. Genet. 7: 1555–1563
Saka Y, Esashi F, Matsusaka T, Mochida S and Yanagida M. . 1997 Genes Dev. 11: 3387–3400.
Saka Y, Fantes P and Yanagida M. . 1994 J. Cell. Sci. Suppl. 18: 57–61.
Sanchez Y, Wong C, Thoma RS, Richman R, Wu Z, Piwnica-Worms H and Elledge SJ. . 1997 Science 277: 1497–1501.
Santocanale C and Diffley JF. . 1998 Nature 395: 615–618.
Scully R, Chen J, Plug A, Xiao Y, Weaver D, Feunteun J, Ashley T and Livingston DM. . 1997 Cell. 88: 265–275.
Sharan SK, Morimatsu M, Albrecht U, Lim DS, Regel E, Dinh C, Sands A, Eichele G, Hasty P and Bradley A. . 1997 Nature 386: 804–810.
Sharples GJ and Leach DR. . 1995 Mol. Microbiol. 17: 1215–1217.
Shieh SY, Ikeda M, Taya Y and Prives C. . 1997 Cell. 91: 325–334.
Shiloh Y. . 1997 Annu. Rev. Genet. 31: 635–662.
Shimada A, Kato S, Enjo K, Osada M, Ikawa Y, Kohno K, Obinata M, Kanamaru R, Ikawa S and Ishioka C. . 1999 Cancer Res. 59: 2781–2786.
Shinohara A, Shinohara M, Ohta T, Matsuda S and Ogawa T. . 1998 Genes Cells. 3: 145–156.
Shirahige K, Hori Y, Shiraishi K, Yamashita M, Takahashi K, Obuse C, Tsurimoto T and Yoshikawa H. . 1998 Nature 395: 618–621.
Shulman MJ, Collins C, Connor A, Read LR and Baker MD. . 1995 EMBO. J. 14: 4102–4107.
Siede W, Friedl AA, Dianova I, Eckardt-Schupp F and Friedberg EC. . 1996 Genetics 142: 91–102.
Smith GC and Jackson SP. . 1999 Genes Dev. 13: 916–934.
Sonoda E, Sasaki MS, Buerstedde JM, Bezzubova O, Shinohara A, Ogawa H, Takata M, Yamaguchi-Iwai Y and Takeda S. . 1998 EMBO. J. 17: 598–608.
Sonoda E, Sasaki MS, Morrison C, Yamaguchi-Iwai Y, Takata M and Takeda S. . 1999 Mol. Cell. Biol. 19: 5166–5169.
St. Onge RP, Udell CM, Casselman R and Davey S. . 1999 Mol. Biol. Cell. 10: 1985–1995.
Sun H, Treco D and Szostak JW. . 1991 Cell. 64: 1155–1161.
Sung P. . 1997 J. Biol. Chem. 272: 28194–28197.
Suzuki A, de la Pompa JL, Hakem R, Elia A, Yoshida R, Mo R, Nishina H, Chuang T, Wakeham A, Itie A, Koo W, Billia P, Ho A, Fukumoto M, Hui CC and Mak TW. . 1997 Genes Dev. 11: 1242–1252.
Suzuki K, Kodama S and Watanabe M. . 1999 J. Biol. Chem. 274: 25571–25575.
Taccioli GE, Amatucci AG, Beamish HJ, Gell D, Xiang XH, Torres Arzayus MI, Priestley A, Jackson SP, Marshak Rothstein A, Jeggo PA and Herrera VL. . 1998 Immunity 9: 355–366.
Taccioli GE, Gottlieb TM, Blunt T, Priestley A, Demengeot J, Mizuta R, Lehmann AR, Alt FW, Jackson SP and Jeggo PA. . 1994 Science 265: 1442–1445.
Takata M, Sasaki MS, Sonoda E, Morrison C, Hashimoto M, Utsumi H, Yamaguchi-Iwai Y, Shinohara A and Takeda S. . 1998 EMBO J. 17: 5497–5508.
Tan TL, Essers J, Citterio E, Swagemakers SM, de Wit J, Benson FE, Hoeijmakers JH and Kanaar R. . 1999 Curr. Biol. 9: 325–328.
Tibbetts RS, Brumbaugh KM, Williams JM, Sarkaria JN, Cliby WA, Shieh SY, Taya Y, Prives C and Abraham RT. . 1999 Genes Dev. 13: 152–157.
Trujillo KM, Yuan SS, Lee EY and Sung P. . 1998 J. Biol. Chem. 273: 21447–21450.
Tsukamoto Y, Kato J and Ikeda H. . 1997 Nature 388: 900–903.
Tsuzuki T, Fujii Y, Sakumi K, Tominaga Y, Nakao K, Sekiguchi M, Matsushiro A, Yoshimura Y and Morita T. . 1996 Proc. Natl. Acad. Sci. USA 93: 6236–6240.
Van Dyck E, Stasiak AZ, Stasiak A and West SC. . 1999 Nature 398: 728–731.
Varon R, Vissinga C, Platzer M, Cerosaletti KM, Chrzanowska KH, Saar K, Beckmann G, Seemanova E, Cooper PR, Nowak NJ, Stumm M, Weemaes CM, Gatti RA, Wilson RK, Digweed M, Rosenthal A, Sperling K, Concannon P and Reis A. . 1998 Cell. 93: 467–476.
Verkade HM and O'Connell MJ. . 1998 Mol. Gen. Genet. 260: 426–433.
Volkmer E and Karnitz LM. . 1999 J. Biol. Chem. 274: 567–570.
Waterman MJ, Stavridi ES, Waterman JL and Halazonetis TD. . 1998 Nat Genet. 19: 175–178.
Weinert T and Hartwell L. . 1989 J. Cell. Sci. Suppl. 12: 145–148.
Weinert TA and Hartwell LH. . 1993 Genetics 134: 63–80.
Westphal CH, Rowan S, Schmaltz C, Elson A, Fisher DE and Leder P. . 1997 Nat. Genet. 16: 397–401.
Wilson TE, Grawunder U and Lieber MR. . 1997 Nature 388: 495–498.
Wong AKC, Pero R, Ormonde PA, Tavtigian SV and Bartel PL. . 1997 J. Biol. Chem. 272: 31941–31944.
Wuerffel RA, Du J, Thompson RJ and Kenter AL. . 1997 J. Immunol. 159: 4139–4144.
Xiao Y and Weaver DT. . 1997 Nucleic Acids Res. 25: 2985–2991.
Xu X, Wagner KU, Larson D, Weaver Z, Li C, Ried T, Hennighausen L, Wynshaw-Boris A and Deng CX. . 1999a Nat. Genet. 22: 37–43.
Xu X, Weaver Z, Linke SP, Li C, Gotay J, Wang XW, Harris CC, Ried T and Deng CX. . 1999b Mol. Cell. 3: 389–395.
Xu Y, Ashley T, Brainerd EE, Bronson RT, Meyn MS and Baltimore D. . 1996 Genes Dev. 10: 2411–2422.
Xu Y and Baltimore D. . 1996 Genes Dev. 10: 2401–2410.
Xu Y, Yang EM, Brugarolas J, Jacks T and Baltimore D. . 1998 Mol. Cell. Biol. 18: 4385–4390.
Yaneva M, Kowalewski T and Lieber MR. . 1997 EMBO J. 16: 5098–5112.
Yoshida K, Kondoh G, Matsuda Y, Habu T, Nishimune Y and Morita T. . 1998 Mol. Cell. 1: 707–718.
Yu L, Orlandi L, Wang P, Orr MS, Senderowicz AM, Sausville EA, Silvestrini R, Watanabe N, Piwnica-Worms H and O'Connor PM. . 1998 J. Biol Chem. 273: 33455–33464.
Yuan SS, Lee SY, Chen G, Song M, Tomlinson GE and Lee EY. . 1999a Cancer Res. 59: 3547–3551.
Yuan ZM, Huang Y, Ishiko T, Nakada S, Utsugisawa T, Kharbanda S, Wang R, Sung P, Shinohara A, Weichselbaum R and Kufe D. . 1998 J. Biol Chem. 273: 3799–3802.
Yuan ZM, Shioya H, Ishiko T, Sun X, Gu J, Huang YY, Lu H, Kharbanda S, Weichselbaum R and Kufe D. . 1999b Nature 399: 814–817.
Zeng X, Chen L, Jost CA, Maya R, Keller D, Wang X, Kaelin Jr WG., Oren M, Chen J and Lu H. . 1999 Mol. Cell. Biol. 19: 3257–3266.
Zeng Y, Forbes KC, Wu Z, Moreno S, Piwnica-Worms H and Enoch T. . 1998 Nature 395: 507–510.
Zhan Q, Antinore MJ, Wang XW, Carrier F, Smith ML, Harris CC and Fornace Jr AJ. . 1999 Oncogene. 18: 2892–2900.
Zhan Q, Bae I, Kastan MB and Fornace Jr AJ. . 1994 Cancer Res. 54: 2755–2760.
Zhan Q, Fan S, Smith ML, Bae I, Yu K, Alamo I Jr., O'Connor PM and Fornace Jr AJ. . 1996 DNA Cell Biol. 15: 805–815.
Zhong Q, Chen CF, Li S, Chen Y, Wang CC, Xiao J, Chen PL, Sharp ZD and Lee WH. . 1999 Science 285: 747–750.
Zhu C, Bogue MA, Lim DS, Hasty P and Roth DB. . 1996 Cell 86: 379–389.
We thank Dr W-H Lee and members of participating laboratories of the program project for enlightening discussions; and Drs E Brown and P Hasty for sharing unpublished results. E Lee is supported by a grant from Texas Advanced Research/Advanced Technology Program (ATP3659-034), and National Institutes of Health Grant (1R01NS378381). GK Dasika is a recipient of Susan G Komen postdoctoral fellowship. S Zhao is supported by DOD training grant (DAMD17–99–1–9402).
About this article
Evaluation of site-specific homologous recombination activity of BRCA1 by direct quantitation of gene editing efficiency
Scientific Reports (2019)
DNA Repair (2019)
Structure-guided optimization of 4,6-substituted-1,3,5-triazin-2(1H)-ones as catalytic inhibitors of human DNA topoisomerase IIα
European Journal of Medicinal Chemistry (2019)
Toxicological Sciences (2019)
Leucine zipper and ICAT domain containing (LZIC) protein regulates cell cycle transitions in response to ionizing radiation
Cell Cycle (2019)