Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Roles of BRCA1 and BRCA2 in homologous recombination, DNA replication fidelity and the cellular response to ionizing radiation


Inheritance of one defective copy of either of the two breast cancer susceptibility genes, BRCA1 and BRCA2, predisposes individuals to breast and ovarian cancers. Current progress in determining the function of these genes suggests that they participate in a common pathway to facilitate orderly homologous recombination and thereby maintain genomic integrity. As a consequence of this defect in homologous recombination, tumors that arise in BRCA carriers are likely to be more sensitive to ionizing radiation. This review summarizes recent investigations about the nature of the defect in DNA repair, and highlights the unanswered questions about the tumor suppressor paradox of BRCA genes. The unsolved mystery is the other genetic changes that must occur to turn a BRCA-deficient cell from a nonviable cell into a tumor cell capable of endless growth.


The contribution of classical human genetics was paramount in the identification of two breast cancer susceptibility genes, BRCA1 and BRCA2. Current progress in determining the function of these genes suggests that they participate in a common pathway that is involved in the control of DNA replication fidelity and DNA damage repair. Familial breast and ovarian cancer can be triggered by a mutation in one allele of either BRCA1 or BRCA2. The sequence of events that occur from the initial germline mutation to the development of cancer is unclear, but loss of function of the second allele is required. However, if both alleles are knocked out, there is failure to develop or proliferate. Solving this ‘tumor suppressor paradox’ and clarifying the function of the BRCA genes may provide further insight on how inactivation leads to tumorigenesis.

Identification of the BRCA genes

The genetic basis for familial breast cancer predisposition has become established over the past decade with the cloning of two major breast cancer susceptibility genes, BRCA1 and BRCA2 (Miki et al., 1994; Wooster et al., 1995). The BRCA1 gene was first cloned in 1994 after being mapped to chromosome 17q21 by genetic linkage analysis. The BRCA1 protein consists of 1863 amino acids and is expressed in most proliferating cells (Figure 1). Within the cell, BRCA1 is part of a large protein complex up to 3 MDa in size. The C-terminus of BRCA1 contains an amino-acid sequence motif, now known as a BRCT domain, recognized in many DNA repair proteins. This appears to be a site of protein–protein interactions, perhaps with other DNA repair proteins to build a large protein complex. In the N-terminus, there is a ring-finger domain, also allowing for protein–protein interactions, but thought to be involved in protein ubiquitination. The BRCA2 gene, a very large protein mapped to chromosome 13q12-13, contains 3418 amino acids and encodes a protein of approximately 385 kDa molecular weight (Figure 1). BRCA2 is characterized by a very large exon 11, which encodes peptide motifs required for interaction with the RAD51 protein (Chen et al., 1999b).

Figure 1

A schematic structure of BRCA1 and BRCA2, showing the key features and domains of both proteins. They are unrelated proteins. BRCA1 is characterized by the BRCT motif at the C-terminus, the SQ cluster region (the site of phosphorylation by ATM and ATR) and the ring domain in the N-terminus (with E3 ubiquitin ligase structure). BRCA2 is characterized by the very large exon 11, containing the eight BRC peptide motifs, important for the interaction with RAD51. A ssDNA binding domain is found within the C-terminus of BRCA2 (Figure 1)

Together, mutations in BRCA1 and BRCA2 account for the majority of families with hereditary susceptibility to breast and ovarian cancer (Hall et al., 1990; Hofmann and Schlag, 2000; Paakkonen et al., 2001). Carriers of the BRCA gene mutations are heterozygous (one normal copy and one mutated copy) within the germ line. Recent epidemiologic studies indicate that BRCA1 mutation carriers have a high lifetime risk of breast cancer of up to 80% (Satagopan et al., 2001). The breast cancer risk for the BRCA2 mutation carriers approaches that of BRCA1, however, with a later age of disease onset (Schubert et al., 1997). In addition to the breast cancer risk, women with BRCA1 mutations have an increased risk of ovarian cancer (Ford et al., 1994). BRCA2 carriers are also at an increased risk for developing ovarian cancer as well as other solid tumors: prostatic, pancreatico-biliary, gastric and colon cancers, as well as melanoma (Peto et al., 1999). The BRCA2 mutation appears to confer higher risks for male breast cancer (Ford et al., 1998). There are subtle histologic differences between BRCA1- and BRCA2-linked breast cancers (Lakhani et al., 1998). The breast cancers that arise in BRCA1 carriers tend to have distinctive pathological features: invasive ductal, high grade, abundant lymphocyte infiltration, and usually estrogen receptor negative (Lakhani et al., 2000). In contrast, tumors arising in BRCA2 carriers do not differ significantly from noncarriers (Lakhani et al., 2002). Many aspects of the function of BRCA1 and BRCA2 suggest a common pathway of DNA recombinational repair; so the reasons for this subtle difference in pathology are not clear.

Knockout mice

BRCA1 and BRCA2 null embryos (loss of both functional alleles) die early in development (approximately day 8 or 9), with a severe growth deficit. The loss of BRCA gene function in embryogenesis has been shown to cause an elevated expression of the cell cycle inhibitor, P21, and subsequent growth arrest (Hakem et al., 1996, 1998; Suzuki et al., 1997). The P53-dependent growth arrest associated with a deficiency in BRCA proteins may represent a checkpoint response to DNA damage. Conversely, loss of such DNA-damage responsive checkpoint function may collaborate with loss of BRCA-mediated function in the development of tumors. Additional observations suggest that the embryos arising in BRCA1 or BRCA2 knockouts are radiation sensitive, as observed by apoptosis of the inner cell mass following exposure to 8 Gy of ionizing radiation (Sharan et al., 1997).

In the clinical setting, breast cancers that arise in carriers of BRCA1 or BRCA2 mutations are associated with a high incidence of inactivation of P53, perhaps greater than 90% (Crook et al., 1997, 1998; Smith et al., 1999b). In the mouse model, when BRCA1 gene inactivation was achieved with the use of a CRE-lox vector system (Xu et al., 1999a), the mice developed late onset breast cancer with frequent P53 mutations. Re-examination of these mice on a P53+/− background showed an increased frequency and earlier onset of breast cancer. A novel class of P53 mutants was classified in BRCA-associated murine breast cancers, which failed to suppress transformation but retained other wild-type functions of P53 (Smith et al., 1999a). Additionally, P53 inactivation partially rescued the developmental arrest in BRCA1 and BRCA2 null mice, further supporting the role of the BRCA genes in DNA damage repair and the hypothesis that P53 checkpoint loss is associated with BRCA gene loss in cancer development (Jacks and Weinberg, 1998).

The BRCA tumor suppressor paradox

It is generally accepted that a cancer results from the accumulation of genetic changes in a target cell. More than 30 years ago, Knudson hypothesized that carcinogenesis results from the occurrence of a second mutation in a somatic target cell, so that the difference between hereditary and nonhereditary cancers is the timing of the first mutation (prezygotic vs postzygotic). The second mutation can affect the remaining normal allele of the same gene (recessive trait) or one copy of a different gene (dominant trait). The first evidence substantiating this ‘two-hit’ hypothesis or recessive model was the identification of the retinoblastoma tumor-suppressor gene.

Breast cancer represents an exception to this two-hit model, as it does not show the predicted relation between hereditary and nonhereditary tumorigenesis. Tumorigenesis in individuals with germline BRCA mutations requires inactivation of the remaining wild-type allele, suggesting that the BRCA genes are tumor-suppressor genes. However, unlike other tumor suppressor genes, no disease-causing dominant BRCA1 or BRCA2 mutations have been detected in sporadic breast or ovarian cancers. Epigenetic regulation of the expression of BRCA1 has been reported in human breast cancers, especially in high-grade tumors (Wilson et al., 1999). There remains the possibility that a BRCA-dependent ‘pathway’ can be inactivated not just by loss of BRCA1 or BRCA2, but by other components of the BRCA-regulated response network, and these may play a role in sporadic breast cancer.

The ‘caretaker’ gene model of Kinzler and Vogelstein (1997) could explain this BRCA tumor-suppressor paradox. The inactivation of a caretaker gene may indirectly promote cancer by causing genomic instability. This hypothesis would suggest that mutations in BRCA1 or BRCA2 might predispose to cancer development, and the loss of other genes that cooperate with the loss of BRCA function is necessary for the full development of cancer. Whether heterozygosity at the loci for the BRCA genes is associated with a haplo-insufficiency state is not yet clear. The next step from germline heterozygosity to cancer development is critical to understanding the function of these genes in vivo.

Response to ionizing radiation

In the last decade, major strides have been made toward understanding the genetic determinants of the response to both ionizing radiation treatment and the use of cytotoxic chemotherapy. It has been known for some time that the primary determinant of radiation sensitivity relates to the efficiency of DNA double-strand break (DSB) repair. However, it has only more recently been appreciated that the molecular mechanisms of DSB repair can be broadly divided into two classes of repair: nonhomologous end-joining (NHEJ) and homologous recombination (HR). Figure 2 shows the genetic and mechanistic difference of these two DSB repair pathways, and shows a summary of genetic factors that determine radiation sensitivity. The vast majority of single gene defects that cause radiation sensitivity can be categorized as a defect in either of these two pathways of DSB repair. In addition to DNA DSB repair, other determinants of radiation sensitivity are radiation-induced cell cycle checkpoints, in particular the S and G2 cell cycle checkpoints, which appear in some cases to have a clear link to radiation sensitivity. Loss of function of the BRCA1 gene is associated with a defect in both the S phase checkpoint and the G2/M transition. These checkpoints are both rapid in response and transient, reaching a peak effect within 1–2 h of irradiation and resolving within 4–6 h. There are also cellular determinants of programmed cell death (apoptosis), which can modulate the radiation response in certain situations, frequently in a tissue-specific manner.

Figure 2

Outline of the two principal pathways of DNA double-strand break repair. NHEJ involves six known proteins, in a coordinated sequence to ensure ligation of the broken ends. Loss of DNA sequence occurs at the site of repair. HR involves multiple proteins in a complex to allow strand exchange and resolution. The process of HR can be an error-free process

A particular feature of BRCA1-deficient cells, as well as many cells showing a defect in homologous recombination, is sensitivity to the DNA crosslinking agent mitomycin C (Moynahan et al., 2001a). The reasons for this feature are thought to be the requirement for homologous recombination in the repair of DNA interstrand crosslinks, but mitomycin C is also a potent inducer of oxidative stress. Further work is needed to clarify the precise reasons for the striking sensitivity to this drug seen in BRCA1-deficient cells.

Role of BRCA1 and BRCA2 in homologous recombination

After 9 years of research since the BRCA1 gene was cloned, the precise function of this protein is not yet determined. BRCA1 binds to numerous cellular proteins in vivo and appears to have multiple functions depending on the cellular context (Deng and Brodie, 2000; Parvin, 2001). Based on a relatively small number of cell culture experiments, cells deficient in the BRCA1 protein do appear to be radiation sensitive. This radiation sensitivity is accompanied by a defect in the measurable rate of homologous recombination (Snouwaert et al., 1999; Moynahan et al., 2001b). The measurement of homologous recombination in cells is not straightforward and usually requires the establishment of cell clones in which a plasmid-based artificial DNA recombination substrate is integrated into the genome of the test cells (Taghian and Nickoloff, 1997). With this type of procedure, it can be determined that the loss of the BRCA1 protein results in a fivefold reduction in the frequency of homologous recombination events (Moynahan et al., 2001b). It is difficult to relate this change in homologous recombination to the change in radiation sensitivity; however, correction of BRCA1 function in a deficient cell line resulted in correction of both the defect in homologous recombination and radiation sensitivity.

Recent unpublished data from our own laboratory have shown that the BRCA-mediated stimulation of homologous recombination can be abrogated by key single amino-acid mutations in the ring domain (C64G), the CHK2 site S988A in the so-called DNA-binding domain and in the BRCT domain (P1749R). Interestingly, mutations of the ATM sites (S1423A, S1524A) had no effect on homologous recombination, even though the G2/M cell cycle checkpoint was defective.

The BRCA2 protein is a very large protein containing 3418 amino acids, which encodes a protein of approximately 385 kDa molecular weight (see Figure 1). At present, there is only one known function of the BRCA2 protein, which is its role in facilitating homologous recombination (Moynahan et al., 2001b; Xia et al., 2001), as was also found for the BRCA1 protein. In contrast to BRCA1, there appears to be no defect in either NHEJ or cell cycle checkpoints (see below) (Yu et al., 2000; Xia et al., 2001). The BRCA2 protein is known to interact with major proteins involved in DNA repair by homologous recombination and, in particular, the RAD51 protein (Chen et al., 1998a, 1998b). This protein is critical in catalyzing the primary reaction required for homologous recombination, and its association with BRCA2 is required to facilitate this effect (Davies et al., 2001). Expressing a short sequence of amino acids (a peptide) from the region of BRCA2 that interacts with RAD51 can result in a disruption of the connection between these two proteins, leading to both a reduction in homologous recombination and an increase in radiation sensitivity (Chen et al., 1999a; Xia et al., 2001).

Two recent publications have highlighted structural aspects of BRCA2, and both studies offer some insight into the structural basis of BRCA2's role in recombinational repair. RAD51 interacts with the BRC repeat regions of BRCA2, encoded by the very large exon 11. There are eight BRC repeats, and BRC3 and BRC4 have the strongest interaction with RAD51. Pellegrini et al. (2002) have solved the crystal structure of the BRC4 peptide sequence (amino acids 1517–1551) of BRCA2 bound to RAD51 (Δ1–96). This structure demonstrates a series of hydrophobic and hydrophilic interactions involving the hairpin structure of BRC4 created by the short antiparallel β-sheets and an α-helix (Figure 3). Contact between BRC4 and RAD51 is main-tained from Leu1521 to Glu1548 of the BRC4 sequence. The structure of the BRC4 peptide is similar to the monomer–monomer interaction domain of RecA (the bacterial homolog of RAD51) that is involved in filament formation. The expression of BRC4 (or BRC3) peptides in cells expressing GFP-tagged RAD51 blocks the ability of RAD51 to form nuclear foci (presupposing that filament formation and focus formation go hand-in-hand).

Figure 3

Interaction between RAD51 and BRCA2. RAD51 interacts with the BRC repeat region of BRCA2, which is shown as the projections between RAD51 monomers. The proposed model is that the BRC repeat provides an assembly line of RAD51 monomers, and that the C-terminus of BRCA2 binds ssDNA, providing a means to displace RPA and allow the orderly assembly of the RAD51 filament. In the absence of BRCA2, critical events in the initiation of homologous recombination are impaired, and repair (and replication) errors accrue rapidly with each cell cycle

Previous in vitro work had suggested that BRC3 or BRC4 peptides could inhibit the formation of RAD51 complexes (Davies et al., 2001). However, in living cells, the effect of full-length BRCA2 is to promote homologous recombination (Moynahan et al., 2001b), expression of the BRC4 peptide inhibits BRCA2-stimulated homologous recombination (Xia et al., 2001), and, the formation of RAD51 foci appears to require BRCA2 (Yuan et al., 1999) (and BRCA1), especially after exposure to ionizing radiation. These observations present a quandary: why does the BRC4 peptide inhibit filament formation, but the whole BRCA2 protein appears to promote filament formation? The use of the BRC4 peptide presents an excess of the RAD51 interacting moiety, whereas the location of BRC4 (and other BRC repeats) within the whole BRCA2 protein may dictate a more highly controlled interaction with RAD51. The quandary may be resolved by proposing a model of BRCA2 as a loading factor for RAD51 on ssDNA by means of its BRC peptide interaction. As the RAD51 bound to BRCA2 finds its position adjacent to a RAD51 monomer on ssDNA, the monomer–monomer interaction becomes favored over the BRCA2 interaction (see Figure 3). We know that RAD51-dependent homologous recombination and RAD51 filament formation in S phase do occur in the absence of BRCA2. However, the presence of BRCA2 is required to prevent major errors, such as chromosome aberrations, which occur in BRCA2-deficient cells (Patel et al., 1998). Whether these errors reflect utilization of an alternative pathway of repair, such as NHEJ (Moynahan et al., 2001b), or occur as a result of error-prone homology-directed repair (Tutt et al., 2001), remains to be solved.

The second paper providing structural insights into the function of the BRCA2 protein (Yang et al., 2002) solved the crystal structure of the C-terminal 800 amino acids of BRCA2, completely separate from the BRC repeat region of the protein, bound to ssDNA (an oligo dT sequence). The BRCA2 structure delineated in this study reveals a series of oligonucleotide-binding folds (OB1/OB2/OB3), which bind ssDNA and a ‘tower’ region between OB2 and OB3 containing a three-helix bundle with a structure very similar to the Hin recombinase. The structure of the OB2/OB3 domains has significant alignment with replication protein A (RPA), a ssDNA-binding protein. Two tempting models are raised by these observations. In one, BRCA2 could play a role in displacing RPA from ssDNA and allowing the formation of a RAD51 filament; alternatively, BRCA2 could primarily bind ssDNA and directly facilitate the formation of RAD51 filaments. What is clear is that BRCA2's major role is to help organize RAD51 function and facilitate homologous recombination.

Nonhomologous end joining

When cells are irradiated, the immediate effect of radiation is to cause fragmentation of double-stranded DNA into smaller-sized pieces. Using pulse-field gel electrophoresis, the size of these DNA fragments can be observed to increase with time after radiation, which reflects the process of DNA repair. It is thought that this assay predominantly reflects the NHEJ type of DNA DSB repair. Cells deficient in BRCA1 protein were reported to show defects in the rejoining of fragmented DNA after ionizing radiation, as assessed by pulse-field gel electrophoresis (Scully et al., 1999). More recent experiments (Wang et al., 2001) have suggested that the magnitude of the difference in NHEJ between BRCA1-deficient cells and BRCA1-proficient cells is relatively small in comparison with the additional impact that can be obtained with the use of the drug wortmannin. Wortmannin is known to inhibit the activity of DNA-dependent protein kinase (DNA-PK) (DiBiase et al., 2000) and thereby suppress NHEJ. The observation that wortmannin produces a significant effect in BRCA1-deficient cells strongly suggests that there is relatively little loss of function of the NHEJ pathway.

To date, there is no evidence that BRCA2-deficient cells have any defect in NHEJ. Both pulsed-field gel electrophoresis and a plasmid-based assay of V(D)J recombination showed no differences between Capan-1, a human pancreatic carcinoma with one deleted copy of BRCA2 and a truncation mutation in the second, and a BRCA2-corrected clone of Capan-1 (Xia et al., 2001).

Nuclear protein foci

When cells are exposed to ionizing radiation, aggregates of repair proteins (now commonly known as IRIFs) appear within their nuclei. Depending on the protein, the time elapsed before the IRIFs appear can vary from less than 1 h to more than 6 h. Furthermore, the phase of the cell cycle can influence their appearance, and some nuclear protein foci (e.g., BRCA1, RAD51) occur spontaneously within S phase. Investigators studying the cellular response to ionizing radiation (IR) frequently use IRIFs to study signaling mechanisms. However, the precise nature of the IRIFs is still unclear, and therefore their value is more as an indicator that some component of the damage response pathway has changed, for example with a genetic manipulation of the test cell.

BRCA1 and BRCA2 both aggregate into IRIFs, usually evident within 6–8 h after exposure. Furthermore, both proteins appear to associate with RAD51, presumably in the same multi-protein complex. BRCA2 is required for the formation of RAD51 IRIF (Yuan et al., 1999), but in BRCA1-deficient cells, RAD51 foci still form, implying that BRCA2 is the more critical protein in relation to RAD51. Our own observations have suggested that RAD51 IRIFs are seen in the absence of BRCA1, but that more foci (both number of foci per cell, and the percentage of cells with foci) are seen when wtBRCA1 is re-expressed in the deficient cells. These observations are in keeping with a prevailing theme for the BRCA proteins – that BRCA2 directly controls the function of RAD51, and that BRCA1 can regulate the behavior of both proteins. IRIFs containing the MRE11/RAD50/NBS1 (MRN) protein complex are observed to be mutually exclusive with RAD51 containing foci, although the mechanistic reasons for this observation are unclear. Initially, BRCA1 was reported to interact with the MRN complex and was required for the formation of MRN IRIFs (Zhong et al., 1999). However, it subsequently became clear that MRN foci form at least normally in BRCA1-deficient cells (Wu et al., 2000), and our own observations support this view. Indeed, our unpublished observations suggest that MRN foci are more abundant in BRCA1-deficient cells.

DNA binding

There are reports for both BRCA1 and BRCA2 that DNA-binding activity can be observed. For BRCA1, limited evidence suggests that binding to complex DNA structures, such as a four-strand structure, can be observed in gel-shift assays (Paull et al., 2001). As a consequence of this binding, the nucleolytic activity of the MRN complex was inhibited. These observations are intriguing, and may contribute significantly to BRCA1's function in vivo. However, the in vivo demonstration of DNA binding has not been shown, and although BRCA1 is known to be part of large protein complexes (Wang et al., 2000), we do not know whether DNA binding is critical to achieve these complexes.

For BRCA2, the recent crystal structure of an 800 amino-acid region of the C-terminus has shown ssDNA binding, with the potential for regions of second-strand binding, very suggestive of a protein with recombination as its prime function (Yang et al., 2002). However, to make a soluble complex capable of crystallizing, the fragment of BRCA2 was mixed with a putative BRCA2 interacting protein, DSS-1. The resultant structure of this region of BRCA2 was a motif-like RP-A, which is highly capable of ssDNA binding. Even though the preparation of the BRCA2 crystal structure was somewhat manipulated the results usually prove to be accurate. The idea that BRCA2's main role is an orderly formation of RAD51 filaments fits all the prevailing evidence (Powell et al., 2002).


The significance of the E3-ubiquitin ligase structure of the N-terminus of BRCA1 is unclear (Baer and Ludwig, 2002). This same region interacts with the BARD1 (BRCA1 associated Ring Domain) protein, whose N-terminus has a similar E3-ubiquitin ligase motif. To date, there are no known targets of this ubiquitin ligase: present speculation wonders whether their prime target is each other, as one mechanism to regulate levels of the proteins. However, there is no evidence for ubiquitination of these proteins (BRCA1 or BARD1) and no evidence for proteolytic degradation. Monoubiquitination could be a form of signaling and may be transient and more difficult to detect. Even though the extent of knowledge about BRCA1 and ubiquitination is limited, it seems highly likely that this is a critical part of the function of BRCA1. BRCA2 has no role in ubiquitination.

Cell cycle checkpoints

Recently, Kastan and colleagues reported that BRCA1-deficient cells show a defect in the arrest of DNA synthesis after ionizing radiation damage (Xu et al., 2001). This phenotype, the so-called radioresistant DNA synthesis (RDS), is also a feature of ataxia telangiectasia (AT) cells, but the precise relation between RDS and radiation sensitivity is not clear. There are AT cell clones in which the cells no longer show RDS, but they remain radiation sensitive (Lehmann et al., 1986), thereby dissociating these two features of AT cells. BRCA1-deficient cells also show a defect in the G2/M phase cell cycle checkpoint (Xu et al., 1999b, 2001), and this phenotype has long been associated with radiation sensitivity. The implication is that if cells cannot halt progression into mitosis following irradiation, then there is insufficient time to repair DNA damage prior to entering mitosis leading to the production of lethal DNA damage. Recent investigations to study the relation between cell cycle checkpoint abnormalities seen in BRCA1-deficient cells and their relation to radiation sensitivity have suggested that the two endpoints can be dissociated. Single amino-acid mutations in BRCA1, such as S1423A, can abrogate the G2/M checkpoint but do not impact radiation sensitivity. Similarly, S1387A can abrogate the S phase checkpoint but does not affect radiation sensitivity (Xu et al., 2002).

The reports that BRCA2-deficient cells have a defect in cell cycle checkpoints are more difficult to interpret, since these cells show a more severe replicative failure, with persistent DNA damage (Patel et al., 1998). Disruption of the interaction between BRCA2 and RAD51 leads to radiation and chemosensitivity and a reported defect in the G2/M transition (Chen et al., 1999a). BRCA2 has also been found to associate with known mitotic checkpoint proteins, such as hBUBR1 (Futamura et al., 1531), and with new proteins, such as BRAF35, in close association with phosphorylated histone H3 (Marmorstein et al., 2001). The S phase checkpoint was also found to be defective in a Chinese hamster ovary cell line mutant, recently found to be defective in BRCA2 (Kraakman-van der Zwet et al., 2002). Post-translational modifications of BRCA2, such as those found for BRCA1 after ionizing radiation, are less well documented with no clear role demonstrated as yet in vivo (Milner et al., 2000).

Transcription-coupled repair

In addition to the function of BRCA1 in DSB repair and cell cycle checkpoints, BRCA1 is also reported to show a role in transcription-coupled repair of oxidative DNA damage, presumably via base-excision repair (Gowen et al., 1998; Le Page et al., 2000). Transcription-coupled repair is the repair of single-stranded DNA damage, utilizing base-excision repair and nucleotide-excision repair, when the DNA damage occurs within transcribed genes. This feature of BRCA1-deficient cells is of uncertain significance at the present time and does not result in sensitivity to ultra-violet light radiation. Cisplatin, a chemotherapeutic agent that causes intrastrand and interstrand crosslinks, sensitizes cells deficient in BRCA1 (Bhattacharyya et al., 2000). Whether this is due to the defect in homologous recombination or transcription-coupled repair of nucleotide excision repair is not clear at this time. Similarly, BRCA2-deficient cells have been reported to have some of the same defects (Le Page et al., 2000), but the full significance of these observations has not been determined. A curious feature of BRCA1 is that the majority of the protein in a cell is found in association with RNA polymerase II (Scully et al., 1997; Anderson et al., 1998; Schlegel et al., 2000), and that DNA damage disrupts that association (Li et al., 1999). BRCA1 can stimulate transcriptional activation in vitro (Haile and Parvin, 1999) and overexpression studies have suggested a role for BRCA1 as a transcriptional coactivator (Harkin et al., 1999; Somasundaram et al., 1999; MacLachlan et al., 2000). However, the significance of the overexpression studies to the in vivo function of BRCA1 remains unclear: is post-translational modification of BRCA1 required for its transcriptional role or to inhibit it? More definitive studies about the role of BRCA1 in regulating transcription are awaited.


Wild-type BRCA1 can promote IR-induced apoptosis in breast and ovarian cancer cell lines (Deng and Brodie, 2000). This function of BRCA1 is dependent on the H-RAS/MEKK4/JNK pathway and is independent of P53 status (Thangaraju et al., 2000). The significance of BRCA1-mediated apoptosis in regard to response of BRCA1-deficient breast cancers to radiation and cytotoxic chemotherapy still needs to be elucidated. The role of BRCA1 in the apoptotic process is presumably well upstream, in that there is no direct connection between BRCA1 and the direct regulators of apoptosis. There is a complex sequence of events that follow DNA damage in cells, in which DNA repair, cell cycle checkpoints and programmed cell death (apoptosis) all contribute to the ultimate outcome of whether a cell lives or dies. What is clear is that cancerous cells have lost many of these protective mechanisms, including apoptosis, and the result is genetically unstable, but viable cells.

DNA replication

The endogenous source of double-stranded DNA damage is DNA replication. It has been known for some time, from bacterial studies, that homologous recombination is required for every cycle of DNA replication and that recombination is a normal protective mechanism for processing stalled replication forks (Goodman, 2000). Similarly, inducible knockout of RAD51 in vertebrate cells (Sonoda et al., 1998) results in chromosomal fragility. The replication-dependent signaling protein, ATR, is known to modify, post-translationally, the BRCA1 protein (Tibbetts et al., 2000; Gatei et al., 2001). ATR only becomes an active kinase within S phase, and thus the most important function of BRCA1 and BRCA2 may be in coordinating these protective mechanisms in S phase. The BRCA1 protein becomes phosphorylated in the SQ cluster domain within S phase, although the precise difference between these sites of phosphorylation and those from exogenous DNA damage is not known. When replication stalls, due to a variety of reasons such as endogenous oxidative DNA damage, the cell has a number of choices of how to restart replication. One method is to partially rewind the replication fork (utilizing a specialized helicase such as the BLM protein) to create a four-strand structure, and template switching can allow repair synthesis to bypass the region of the blocking lesion. This may work for some but not all lesions. When cleavage of the parental strand by excision repair at the time the replication fork arrives produces an effective DSB, or when a stalled replication fork is deliberately cleaved to help bypass the block, the restart of replication would require homologous recombination. In yeast, the mus81 gene is a specialized endonuclease that can cleave stalled replication forks (Haber and Heyer, 2001), and if mus81 and the BLM homologe, sgs1, are both deleted, the yeast is a synthetic lethal (i.e., it dies in DNA replication). In other words, cells need mechanisms to successfully negotiate stalled replication forks and have at least two genetically defined options. The suggested primary role of the BRCA1-/BRCA2-defined pathway is to facilitate homologous recombination in the bypass of stalled replication forks. As to the question of how cancer arises in cells with deficiencies in BRCA1 or BRCA2, it seems likely that the impaired function of homologous recombination is a key step, since this is the only known defect in BRCA2-deficient cells. A ‘mutator’ phenotype can be triggered by a defect in the regulation of homologous recombination, either by making the process error-prone or by shunting repair events into a NHEJ pathway that is inherently mutagenic. Additional mutational steps, to bypass the proliferative block, are also necessary to allow a net growth in the number of cancer cells. However, what remains, is the unsolved mystery of why this described defect in homologous recombination leads to the tissue-specific cancers of the breast and ovary. Proliferation in breast or ovarian epithelium may be associated with higher levels of endogenous DNA damage, relative to other tissues, which leads to replication stalling and requires homologous recombination for replication restart.

Maintenance of genomic integrity

Defining the physiological function of BRCA1 that maintains genomic integrity has proven to be extremely complicated. Studies have ascribed roles for BRCA1 in an overwhelming number of processes, including DNA replication, cell cycle checkpoint control, DNA repair, regulation of transcription, protein ubiquitination, apoptosis, and chromatin remodeling. On the other hand, there has been only one function delineated so far for BRCA2 – as a cofactor in homologous recombination that requires its interaction with RAD51 (Davies et al., 2001; Xia et al., 2001). As such, the role of BRCA2 in regulating RAD51-dependent homologous recombination may be the key to understanding the cancer-prone phenotype of both BRCA1 and BRCA2 mutations.

Based on all the evidence presented in this review, it seems likely that the role of BRCA1 and BRCA2 in facilitating orderly homologous recombination is the key to maintaining genomic integrity. In the absence of these key proteins, replication errors produce a mutator phenotype that is a key feature of carcinogenesis. The cooperation between loss function of BRCA1/BRCA2 and other key genetic changes that are required for maintaining cell viability is the key remaining question.


  1. Anderson SF, Schlegel BP, Nakajima T, Wolpin ES and Parvin JD . (1998). Nat. Genet., 19, 254–256.

  2. Baer R and Ludwig T . (2002). Curr. Opin. Genet. Dev., 12, 86–91.

  3. Bhattacharyya A, Ear US, Koller BH, Weichselbaum RR and Bishop DK . (2000). J. Biol. Chem., 275, 23899–23903.

  4. Chen CF, Chen PL, Zhong Q, Sharp ZD and Lee WH . (1999a). J. Biol. Chem., 274, 32931–32935.

  5. Chen J, Silver DP, Walpita D, Cantor SB, Gazdar AF, Tomlinson G, Couch FJ, Weber BL, Ashley T, Livingston DM and Scully R . (1998a). Mol. Cell, 2, 317–328.

  6. Chen JJ, Silver D, Cantor S, Livingston DM and Scully R . (1999b). Cancer Res., 59, 1752s–1756s.

  7. Chen PL, Chen CF, Chen Y, Xiao J, Sharp ZD and Lee WH . (1998b). Proc. Natl. Acad. Sci. USA, 95, 5287–5292.

  8. Crook T, Brooks LA, Crossland S, Osin P, Barker KT, Waller J, Philp E, Smith PD, Yulug I, Peto J, Parker G, Allday MJ, Crompton MR and Gusterson BA . (1998). Oncogene, 17, 1681–1689.

  9. Crook T, Crossland S, Crompton MR, Osin P and Gusterson BA . (1997). Lancet, 350, 638–639.

  10. Davies AA, Masson JY, McIlwraith MJ, Stasiak AZ, Stasiak A, Venkitaraman AR and West SC . (2001). Mol. Cell, 7, 273–282.

  11. Deng CX and Brodie SG . (2000). BioEssays, 22, 728–737.

  12. DiBiase SJ, Zeng ZC, Chen R, Hyslop T, Curran Jr WJ and Iliakis G . (2000). Cancer Res., 60, 1245–1253.

  13. Ford D, Easton DF, Bishop DT, Narod SA and Goldgar DE . (1994). Lancet, 343, 692–695.

  14. Ford D, Easton DF, Stratton M, Narod S, Goldgar D, Devilee P, Bishop DT, Weber B, Lenoir G, Chang-Claude J, Sobol H, Teare MD, Struewing J, Arason A, Scherneck S, Peto J, Rebbeck TR, Tonin P, Neuhausen S, Barkardottir R, Eyfjord J, Lynch H, Ponder BA, Gayther SA, Birch JM, Lindblom A, Stoppa-Lyonnet D, Bignon Y, Borg A, Hamann U, Haites N, Scott RJ, Maugard CM, Vasen H, Seitz S, Cannon-Albright LA, Schofield A, Zelada-Hedman M and the Breast Cancer Linkage Consortium. (1998). Am. J. Hum. Genet., 62, 676–689.

  15. Futamura M, Arakawa H, Matsuda K, Katagiri T, Saji S, Miki Y and Nakamura Y . (1531). Cancer Res., 60, 1531–1535.

  16. Gatei M, Zhou BB, Hobson K, Scott S, Young D and Khanna KK . (2001). J. Biol. Chem., 276, 17276–17280.

  17. Goodman MF . (2000). Trends Biochem. Sci., 25, 189–195.

  18. Gowen LC, Avrutskaya AV, Latour AM, Koller BH and Leadon SA . (1998). Science, 281, 1009–1012.

  19. Haber JE and Heyer WD . (2001). Cell, 107, 551–554.

  20. Haile DT and Parvin JD . (1999). J. Biol. Chem., 274, 2113–2117.

  21. Hakem R, de la Pompa JL and Mak TW . (1998). J. Mammary Gland Biol. Neoplasia, 3, 431–445.

  22. 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.

  23. Hall JM, Lee MK, Newman B, Morrow JE, Anderson LA, Huey B and King MC . (1990). Science, 250, 1684–1689.

  24. Harkin DP, Bean JM, Miklos D, Song YH, Truong VB, Englert C, Christians FC, Ellisen LW, Maheswaran S, Oliner JD and Haber DA . (1999). Cell, 97, 575–586.

  25. Hofmann W and Schlag PM . (2000). J. Cancer Res. Clin. Oncol., 126, 487–496.

  26. Jacks T and Weinberg RA . (1998). Science, 280, 1035–1036.

  27. Kinzler KW and Vogelstein B . (1997). Nature, 386, 761–763.

  28. Kraakman-van der Zwet M, Overkamp WJ, van Lange RE, Essers J, van Duijn-Goedhart A, Wiggers I, Swaminathan S, van Buul PP, Errami A, Tan RT, Jaspers NG, Sharan SK, Kanaar R and Zdzienicka MZ . (2002). Mol. Cell. Biol., 22, 669–679.

  29. Lakhani SR, Gusterson BA, Jacquemier J, Sloane JP, Anderson TJ, van de Vijver MJ, Venter D, Freeman A, Antoniou A, McGuffog L, Smyth E, Steel CM, Haites N, Scott RJ, Goldgar D, Neuhausen S, Daly PA, Ormiston W, McManus R, Scherneck S, Ponder BA, Futreal PA, Peto J, Stoppa-Lyonnet D, Bignon YJ and Stratton MR . (2000). Clin. Cancer Res., 6, 782–789.

  30. Lakhani SR, Jacquemier J, Sloane JP, Gusterson BA, Anderson TJ, van de Vijver MJ, Farid LM, Venter D, Antoniou A, Storfer-Isser A, Smyth E, Steel CM, Haites N, Scott RJ, Goldgar D, Neuhausen S, Daly PA, Ormiston W, McManus R, Scherneck S, Ponder BA, Ford D, Peto J, Stoppa-Lyonnet D, Bignon YJ, Struewing JP, Spurr NK, Bishop DT, Klijn JGM, Devilee P, Cornelisse CJ, Lasset C, Lenoir G, Barkardottir RB, Egilsson VHU, Chang-Claude J, Sobol H, Weber B, Stratton MR and Easton DF . (1998). J. Natl. Cancer Inst., 90, 1138–1145.

  31. Lakhani SR, Van De Vijver MJ, Jacquemier J, Anderson TJ, Osin PP, McGuffog L and Easton DF . (2002). J. Clin. Oncol., 20, 2310–2318.

  32. Le Page F, Randrianarison V, Marot D, Cabannes J, Perricaudet M, Feunteun J and Sarasin A . (2000). Cancer Res., 60, 5548–5552.

  33. Lehmann AR, Arlett CF, Burke JF, Green MH, James MR and Lowe JE . (1986). Int. J. Radiat. Biol., 49, 639–643.

  34. Li S, Chen PL, Subramanian T, Chinnadurai G, Tomlinson G, Osborne CK, Sharp ZD and Lee WH . (1999). J. Biol. Chem., 274, 11334–11338.

  35. MacLachlan TK, Somasundaram K, Sgagias M, Shifman Y, Muschel RJ, Cowan KH and El-Deiry WS . (2000). Journal of Biological Chemistry, 275, 2777–2785.

  36. Marmorstein LY, Kinev AV, Chan GK, Bochar DA, Beniya H, Epstein JA, Yen TJ and Shiekhattar R . (2001). Cell, 104, 247–257.

  37. Miki Y, Swensen J, Shattuck-Eidens D, Futreal AP, Harshman K, Tavtigian S, Liu Q, Cochran C, Bennett ML, Wei D, Russell B, Rosenthal J, Hussey C, Tran T, McClure M, Frye C, Hattier T, Phelps R, Haugen-Strano A, Katcher H, Yakumo K, Gholami Z, Shaffer D, Stone S, Bayer S, Wray C, Bogden R, Dayananth P, Ward J, Tonin P, Narod S, Bristow PK, Norris FH, Helvering L, Morrison P, Rosteck P, Lai M, Barrett JC, Lewis C, Neuhausen S, Cannon-Albright L, Goldgar D, Wiseman R, Kamb A and Skolnick MH . (1994). Science, 266, 66–71.

  38. Milner J, Fuks F, Hughes-Davies L and Kouzarides T . (2000). Oncogene, 19, 4441–4445.

  39. Moynahan ME, Cui TY and Jasin M . (2001a). Cancer Res., 61, 4842–4850.

  40. Moynahan ME, Pierce AJ and Jasin M . (2001b). Mol. Cell, 7, 263–272.

  41. Paakkonen K, Sauramo S, Sarantaus L, Vahteristo P, Hartikainen A, Vehmanen P, Ignatius J, Ollikainen V, Kaariainen H, Vauramo E, Nevanlinna H, Krahe R, Holli K and Kere J . (2001). Genet. Epidemiol., 20, 239–246.

  42. Parvin JD . (2001). Proc. Natl. Acad. Sci. USA, 98, 5952–5954.

  43. 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.

  44. Paull TT, Cortez D, Bowers B, Elledge SJ and Gellert M . (2001). Proc. Natl. Acad. Sci. USA, 98, 6086–6091.

  45. Pellegrini L, Yu DS, Lo T, Anand S, Lee M, Blundell TL and Venkitaraman AR . (2002). Nature, 420, 287–293.

  46. Peto J, Collins N, Barfoot R, Seal S, Warren W, Rahman N, Easton DF, Evans C, Deacon J and Stratton MR . (1999). J. Natl. Cancer Inst., 91, 943–949.

  47. Powell SN, Willers H and Xia F . (2002). Mol. Cell, 10, 1262–1263.

  48. Satagopan JM, Offit K, Foulkes W, Robson ME, Wacholder S, Eng CM, Karp SE and Begg CB . (2001). Cancer Epidemiol. Biomarkers Prev., 10, 467–473.

  49. Schlegel BP, Green VJ, Ladias JA and Parvin JD . (2000). Proc. Natl. Acad. Sci. USA, 97, 3148–3153.

  50. Schubert EL, Lee MK, Mefford HC, Argonza RH, Morrow JE, Hull J, Dann JL and King MC . (1997). Am. J. Hum. Genet., 60, 1031–1040.

  51. Scully R, Anderson SF, Chao DM, Wei W, Ye L, Young RA, Livingston DM and Parvin JD . (1997). Proc. Natl. Acad. Sci. USA, 94, 5605–5610.

  52. Scully R, Ganesan S, Vlasakova K, Chen J, Socolovsky M and Livingston DM . (1999). Mol. Cell, 4, 1093–1099.

  53. 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.

  54. Smith PD, Crossland S, Parker G, Osin P, Brooks L, Waller J, Philp E, Crompton MR, Gusterson BA, Allday MJ and Crook T . (1999a). Oncogene, 18, 2451–2459.

  55. Smith PD, Crossland S, Parker G, Osin P, Brooks L, Waller J, Philp E, Crompton MR, Gusterson BA, Allday MJ and Crook T . (1999b). Oncogene, 18, 2451–2459.

  56. Snouwaert JN, Gowen LC, Latour AM, Mohn AR, Xiao A, DiBiase L and Koller BH . (1999). Oncogene, 18, 7900–7907.

  57. Somasundaram K, MacLachlan TK, Burns TF, Sgagias M, Cowan KH, Weber BL and el-Deiry WS . (1999). Oncogene, 18, 6605–6614.

  58. 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.

  59. 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.

  60. Taghian DG and Nickoloff JA . (1997). Mol. Cell. Biol., 17, 6386–6393.

  61. Thangaraju M, Kaufmann SH and Couch FJ . (2000). J. Biol. Chem., 275, 33487–33496.

  62. Tibbetts RS, Cortez D, Brumbaugh KM, Scully R, Livingston D, Elledge SJ and Abraham RT . (2000). Genes Dev., 14, 2989–3002.

  63. Tutt A, Bertwistle D, Valentine J, Gabriel A, Swift S, Ross G, Griffin C, Thacker J and Ashworth A . (2001). EMBO J., 20, 4704–4716.

  64. Wang H, Zeng Z, Bui T, DiBiase S, Qin W, Xia F, Powell S and Iliakis G . (2001). Cancer Res., 61, 270–277.

  65. Wang Y, Cortez D, Yazdi P, Neff N, Elledge SJ and Qin J . (2000). Genes. Dev., 14, 927–939.

  66. Wilson CA, Ramos L, Villasenor MR, Anders KH, Press MF, Clarke K, Karlan B, Chen JJ, Scully R, Livingston D, Zuch RH, Kanter MH, Cohen S, Calzone FJ and Slamon DJ . (1999). Nat. Genet., 21, 236–240.

  67. Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, Collins N, Gregory S, Gumbs C and Micklem G . (1995). Nature, 378, 789–792.

  68. Wu X, Petrini JH, Heine WF, Weaver DT, Livingston DM and Chen J . (2000). Science, 289, 11.

  69. Xia F, Taghian DG, DeFrank JS, Zeng ZC, Willers H, Iliakis G and Powell SN . (2001). Proc. Natl. Acad. Sci. USA, 98, 8644–8649.

  70. Xu B, Kim S and Kastan MB . (2001). Mol. Cell. Biol., 21, 3445–3450.

  71. Xu B, O'Donnell AH, Kim ST and Kastan MB . (2002). Cancer Res., 62, 4588–4591.

  72. 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.

  73. 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.

  74. Yang H, Jeffrey PD, Miller J, Kinnucan E, Sun Y, Thoma NH, Zheng N, Chen PL, Lee WH and Pavletich NP . (2002). Science, 297, 1837–1848.

  75. Yu V, Koehler M, Steinlein C, Schmid M, Hanakahi LA, van Gool AJ, West SC and Venkitaraman AR . (2000). Genes Dev., 14, 1400–1406.

  76. Yuan SS, Lee SY, Chen G, Song M, Tomlinson GE and Lee EY . (1999). Cancer Res., 59, 3547–3551.

  77. Zhong Q, Chen CF, Li S, Chen Y, Wang CC, Xiao J, Chen PL, Sharp ZD and Lee WH . (1999). Science, 285, 747–750.

Download references

Author information



Corresponding author

Correspondence to Simon N Powell.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Powell, S., Kachnic, L. Roles of BRCA1 and BRCA2 in homologous recombination, DNA replication fidelity and the cellular response to ionizing radiation. Oncogene 22, 5784–5791 (2003).

Download citation


  • breast cancer
  • BRCA genes
  • tumor suppressor genes
  • DNA repair
  • ionizing radiation
  • homologous recombination
  • DNA replication

Further reading


Quick links