New Complexities for BRCA1 and BRCA2 Peter Kerr and Alan Ashworth

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

R668 Review

New complexities for BRCA1 and BRCA2 Peter Kerr and Alan Ashworth

A large number of diverse functions have been Introduction attributed to the BRCA1 and BRCA2 breast cancer About one in ten women in the Western world develop susceptibility genes. Here we review recent progress in cancer of the breast and at least 5% of these cases are the field. thought to result from a hereditary predisposition to the disease [1]. Two breast cancer susceptibility (BRCA) genes Address: The Breakthrough Breast Cancer Research Centre, Institute of Cancer Research, Fulham Road, London SW3 6JB, UK. have been mapped and cloned (Figure 1), and mutations in these genes underlie the disease in most families in Correspondence: Alan Ashworth which four or more cases of breast cancer are diagnosed in E-mail: [email protected] women below the age of 60. Women who inherit loss-of- Current Biology 2001, 11:R668–R676 function mutations in one allele of either of these genes have an up to 85% risk of breast cancer by age 70. Both 0960-9822/01/$ – see front matter BRCA1 and BRCA2 are thought to be tumour suppressor © 2001 Elsevier Science Ltd. All rights reserved. genes as the wild-type allele of the gene is lost in tumours of heterozygous carriers. As well as breast cancer, carriers of mutations in these genes are at elevated risk of cancer in the ovary, prostate and pancreas. Surprisingly, despite the association with inherited predisposition, somatic disease- causing mutations in BRCA1 and BRCA2 are extremely rare in sporadic breast cancers. Functions for the BRCA proteins in a bewildering array of biological processes have been suggested [2]. But whether dysfunction of all or only a subset of these processes contributes to cancer initiation or progression is unknown. Here we discuss recent new findings on the functions of BRCA1 and BRCA2 which provide more detail on their potential roles in transcriptional regulation and DNA repair and suggest additional roles in processes such as polyadenylation of mRNA and protein degradation.

Fixing DNA Both BRCA1 and BRCA2 have been consistently linked to various processes of the repair of damaged DNA. These

Figure 1

BRCA1 Ring NLS BRCT

BRCA2 RAD51 RAD51 binding binding

BRC repeats NLS Current Biology

Functional domains in the BRCA1 and BRCA2 proteins. The RING finger is a protein–protein interaction domain which may have ubiquitin ligase activity. The BRCT domain is another protein–protein interaction motif which is found principally in proteins involved in DNA metabolism or repair. BRCA2 contains nine separate potential RAD51 binding domains — 8 BRC repeats and a structurally independent carboxy-terminal domain. Both BRCA1 and BRCA2 contain nuclear localisation sequences (NLS). Review BRCA proteins R669

include the repair of double-strand breaks by homologous Figure 2 recombination and the repair of oxidative damage by transcription-coupled repair [3,4]. Now several new papers Double-strand break provide further insight into the potentially multi-purpose role of BRCA proteins in DNA repair. Non-homologous Homologous Mammalian cells repair breaks in double-stranded DNA end joining recombination by non-homologous end joining or by homologous recom- Resection bination [5,6] (Figure 2). Substantial evidence is now available for a role for both BRCA1 and BRCA2 in homologous recombination. Both BRCA1 and BRCA2 can interact Strand invasion directly or indirectly with RAD51 [7–10], the eukaryotic equivalent of bacterial RecA. RAD51 catalyses strand exchange during homology directed repair of DNA double- strand breaks by gene conversion. Within BRCA2, the RAD51 interaction is mediated by a number of repeats of a Branch specific sequence — the BRC repeat — and an unrelated migration domain located at the carboxy terminus (Figure 1) [7,9,10]. BRC repeats may regulate the loading of RAD51 onto DNA [11]. Taken together with the colocalisation of BRCA1 and BRCA2 with RAD51 in ionising radiation induced nuclear DNA -Ku foci [10], these observations suggest a role for BRCA1 and PK Ligation and BRCA2 in DNA repair by homologous recombination. Fur- DNA ligase IV-XRCC4 resolution thermore, mouse and human BRCA1 and BRCA2 mutant cells suffer from chromosome instability [12–17] and have a MRE11-RAD50-NBS1 heightened sensitivity to DNA lesions that are repaired by homologous recombination [14,18–21]. RAD52 RAD51 Current Biology A direct and elegant analysis of the ability of BRCA1 and BRCA2 mutant cells to repair double-strand breaks created Double-stranded DNA break repair pathways. Non-homologous end by endonucleases has been performed by Jasin and col- joining results in the direct rejoining of the broken DNA ends and leagues [22]. Mouse embryonic stem (ES) cells lacking generally results in small deletions. The heterodimeric DNA end-binding protein Ku recruits DNA-PK and then XRCC4 and DNA ligase IV. The BRCA1 are deficient in the repair of double-strand breaks Mre11–Rad50–Nbs1complex is involved in end processing before re- by homologous recombination [23]. Similarly, ES cells joining. Homologous recombination requires an end binding protein, with disruptions in BRCA2 are compromised for repair of Rad52. Rad51 forms a filament along the DNA which facilitates strand ′ double-strand breaks by gene conversion [24]. Loss of invasion. The resected 3 end invades a homologous DNA duplex and is then extended by DNA polymerase. In this simplified model only one BRCA2 appears to result in alternative use of the error- of the possible recombination products is shown. Modified from [76]. prone single strand annealing pathway (Andrew Tutt, per- sonal communication). Use of error-prone pathways instead of BRCA1 and 2 dependent conservative routes to may well provide the raw material for the further genetic repair is likely to create considerable genome instability changes required for tumour progression. and may contribute to loss of tumour suppressor genes and tumorigenic progression. BACH variations A recent paper [25] from David Livingston’s group Models have been proposed to explain the roles of BRCA1 describes the isolation of BACH1, a novel putative DNA and BRCA2 in the maintenance of chromosome instability helicase which interacts with the BRCT domains of through functions in DNA replication [4]. Stalled replica- BRCA1. BACH1 contains substantial sequence similarity tion forks caused by a variety of mechanisms such as base to the catalytic and nucleotide binding domains of known lesions, DNA breaks or strand gaps, are thought to require members of the DEAH helicase family which includes the homologous recombination to restart replication. If homol- Xeroderma Pigmentosum complementing group D (XPD) ogous recombination is dysfunctional then stalled replica- protein [26–28]. BACH1 binds to BRCA1 and, as helicases tion forks may lead to persistent DNA breaks and have been implicated in DNA repair, the authors [25] asked ultimately to gross chromosomal rearrangements including whether BACH1 has such a function. They expressed a translocations. These are indeed frequently seen in cells dominant negative mutant of BACH1 in tissue culture lacking BRCA1 and BRCA2 [12–17]. Such rearrangements cells and, by measuring the rate of double-strand break R670 Current Biology Vol 11 No 16

repair after gamma-irradiation, showed that the dominant addition of a single ubiquitin residue in response to DNA negative helicase could interfere with double-strand break damage [32]; this monoubiquitination is distinct from that repair. However, whether wild-type BACH1 is required generally observed when proteins are targeted for degrada- for double-strand break repair remains an open question. tion (see below). Coincident with the ubiquitin modifica- tion, FANCD2 becomes localised to nuclear foci which, Although mutations in BRCA1 and BRCA2 account for a after further examination, were also found to contain significant proportion of early onset familial cancers, it is BRCA1. Additional support for a functional interaction clear that other BRCA genes exist [29]. To test whether comes from observations that BRCA1 and FANCD2 can BACH1 might be a novel BRCA gene, Livingston and col- be coimmunoprecipitated from gamma-irradiated cells. leagues [25] surveyed a series of breast cancer families and The importance of ubiquitination in the localisation of early onset breast cancers which were devoid of mutations FANCD2 and the suggestion that BRCA1 has ubiquitin in BRCA1 and 2. Two heterozygous missense mutations E3 ligase activity [33–35] suggests an obvious mechanism were identified in BACH1 in germline DNA of two breast by which BRCA1 regulates FANCD2. Intriguingly radial cancer patients. Both these sequence differences resulted aberrations typical of Fanconi’s anemia are observed in in amino acid changes (P47A and M299I) in the helicase both BRCA1 and BRCA2-null cells [14,15]. domain of BACH1 and were thought unlikely to be common polymorphisms. And both were discovered in BRCA1 breaks down early onset breast cancer patients, one of whom, the carrier Protein degradation is increasingly being recognised as a of the P47A change, had a strong family history of breast key step in the regulation of embryonic development and and ovarian cancer. Unfortunately the authors were unable cell growth and differentation. Abnormal regulation of to confirm segregation of this mutation with breast cancer, protein degradation is also critical to the pathogenesis of as the appropriate DNAs were not available. Furthermore, neurodegenerative diseases and cancer. One of the first the wild-type allele appeared to be retained in the tumour features of the BRCA1 protein to be described was the genome. RING finger domain (see Figure 1), although at the time the function of this domain was not clearly understood. So how might these changes result in early onset breast Indeed RING fingers were thought capable of interacting cancer? To address this question, mutant and wild-type ver- with DNA. Recently, the RING domain has been impli- sions of BACH1 were expressed ectopically in U20S cells cated in the ubiquitin-mediated proteolytic pathway. [25]. Interestingly, the P47A mutation appeared to lead to destabilisation of the BACH1 protein. In the absence of A key cellular pathway for protein degradation is medi- evidence that the other allele of BACH1 is affected in the ated by polyubiquitination of specific lysine residues; tumour carrying the P47A mutation, the authors speculate polyubiquitinated substrates are degraded by the 26S that haploinsufficiency could contribute to tumour forma- proteosome [36]. The ubiquitin E3 ligase catalyses the tion in early onset breast cancer. However, without defini- formation of polyubiquitin chains on substrate proteins tive genetic evidence it is premature to ascribe a significant via iso-peptide bonds using ubiquitin that has been acti- role to BACH1 in familial breast cancer and it remains pos- vated by enzymes E1 and E2 (Figure 3). Specific ligase sible that the sequence changes are rare polymorphisms. complexes are associated with regulation of specific cellular processes. For example, the Skp1–Cullin–F box The Fanconi connection complex, SCF, and the cyclosome or anaphase promoting Fanconi’s anemia is an autosomal recessive disorder char- complex, APC, are the two major ubiquitin ligase com- acterised by cancer susceptibility, congenital abnormalities plexes that regulate ubiquitin-mediated proteolysis and bone marrow failure [30]. Patients with this disease during G1/S and anaphase, respectively [37]. Recently develop several types of cancer but are particularly prone several RING finger proteins, with otherwise diverse to acute myeloid leukemias and squamous cell carcinomas structures, such as c-CBL, IAP, and MDM2 have been of the head and neck. The cells of these patients are char- linked to ubiquitination of important regulatory proteins acterised by chromosomal instability and enhanced sensi- [33,38–40]. Now BRCA1 has been suggested to function tivity to bifunctional alkylating agents that cross-link in ubiquitination [33–35]. Although the BRCA1 RING DNA. Six different genes, FANC1–6, have been shown to domain alone has low E3 ubiquitin ligase activity, a het- be mutated in patients with Fanconi’s anemia but little is erodimer of RING finger domains of BRCA1 and BARD1 known about their function or how their mutation leads to has considerable activity. BARD1 is a RING domain the disease [30]. Recently, a new Fanconi’s anemia gene, protein shown in a two-hybrid screen to interact with FANCD2, has been isolated [31]. The demonstration that BRCA1 RING domain [41]. Intriguingly BARD1 is also the product of FANCD2 interacts with BRCA1 may provide mutated in breast and ovarian tumours, albeit infre- new insights into both Fanconi’s anemia and breast cancer quently [42]. Although these results are highly sugges- susceptibility [32]. FANCD2 becomes modified by the tive, we still await the finding of physiological targets of Review BRCA proteins R671

Figure 3

BRCA1–BARD1 heterodimer might function Lys as an E3 ubiquitin ligase. Free ubiquitin is activated in an ATP-dependent manner to BRCA1 BARD1 Substrate form a covalent link with subunit E1. Ubiquitin E1 is then transferred to one of a number of E2 subunits. E2 then associates with E3 ligases, such as the BRCA1–BARD1 heterodimer, Ub which may or may not have substrates already E2 bound. Ubiquitin is then transferred onto lysine residues of the substrate. Modified from [77].

Current Biology

BRCA1–BARD1 ubiquitination; one potential target, the MacLachlan et al. [50] identified a number of genes whose FANCD2 protein, was discussed above. transcription was differentially affected by expression of BRCA1. Some of these genes are obviously involved in The end in sight? the control of cellular proliferation (Cyclin B1, PIN1, An intriguing addition to the list of putative functions for PCNA) whereas others have been implicated in responding the BRCA1 protein has been a possible role in mRNA pro- to DNA damage (GADD45, GADD153). However, it should cessing. Kleiman and Manley [43] have shown that be noted that both of these studies were based on BRCA1 BARD1 interacts with the polyadenylation factor CstF and overexpression in a BRCA1 wild-type cellular background; inhibits 3′ end processing of mRNA precursors. A more whether this is physiologically relevant to the situation in a recent study by the same group [44] showed that DNA BRCA1 null tumour is not clear. damage by either UV or hydroxyurea could inhibit polyadenylation in vitro and that this inhibition was medi- cDNA microarrays have also been used to show that differ- ated via the BARD1–CstF interaction. Regulation of ent groups of genes are expressed by tumours arising in polyadenylation is thought to play an important role in cell patients carrying mutations in either BRCA1 or BRCA2 [51]. growth control [45,46]. A complex containing BRCA1, The authors could quite accurately predict from which BARD1 and CstF increases transiently following DNA genetic mutation different tumours had arisen by the expres- damage and inhibits polyadenylation. This may prevent sion profiles of relatively small groups of genes. A set of 9 inappropriate RNA processing and production of aberrant genes could be used to distinguish tumours that did or did transcripts. Both BRCA1 and BARD1 can associate with not carry BRCA1 mutations and a set of 11 genes could be the RNA polymerase II complex [47,48] which is neces- used to distinguish tumours that carried BRCA2 mutations, sary for efficient 3′ cleavage of mRNA; Kleiman and though to a less statistically significant extent. Follow-up Manley [43] propose a model whereby BRCA1 and/or immunohistochemical analysis using tissue arrays con- BARD1, via their RING domain, control degradation of firmed at the protein level that cyclin D1 was more highly RNA polymerase II by ubiquitination. A role for BRCA1 expressed in BRCA2 tumours than in BRCA1 tumours [51]. in polyadenylation is a previously unsuspected one. BRCA1–BARD1 may be recruited to sites of stalled tran- The above study demonstrates how cDNA and tissue scription as part of a transcription-coupled DNA repair arrays will be important tools in not just the research labo- process [3]. Absence of a functional BRCA1–BARD1 ratory but also in a clinical setting, both diagnostically and complex might cause dysregulation of gene expression and therapeutically. However, it also raises some questions perhaps contribute to tumour progression. about the experimental design of microarray experiments [52]. Gene expression microarray experiments generate Chipping away at BRCA function vast quantities of data and there can be tremendous varia- The simultaneous analysis of the expression of thousands tion within these data sets, especially when comparing of genes by microarrays has started to contribute to our samples such as panels of tumours. The scientific commu- understanding of how mutations in BRCA1 and BRCA2 nity is still determining the best ways to analyse such large can lead to tumour susceptibility and loss of cell growth data sets and is improving algorithms for clustering co- control. Using this technology, Harkin et al. [49] showed varying sets of genes. Therefore, as much variation as pos- that induction of BRCA1 expression can lead to changes sible, particularly in pathological features, needs to be in expression of various genes implicated in growth removed from the samples being compared. Obviously, control such as GADD45 and EGR1. This led to the iden- this is not always possible and so larger patient or sample tification of a specific pathway — the JNK/SAPK pathway numbers will be needed to validate the changes seen in — activated during BRCA1-induced apoptosis. Similarly the experiments of Hedenfalk et al. [51]. However, it is R672 Current Biology Vol 11 No 16

Figure 4 The authors [54] suggest a model whereby BRCA1 and ZBRK1 repress GADD45 transcription in the uninduced state, but following a signal such as DNA damage leading to increased BRCA1 phosphorylation or protein levels, there is derepression of GADD45 gene transcription. Intrigu- ingly, this suggests that BRCA1 might play a role in both corepression and coactivation of the same gene; it will be interesting to determine which other genes fall under this putative dual transcription activity. Other genes encoding potential ZBRK1 binding motifs have been identified [54], including GADD153, Ki-67, Bax and p21, all of which may be targets of BRCA1 transcriptional control.

Another interaction that supports a role for BRCA1 in transcriptional regulation is that with the signal transduction and activator of transcription protein, STAT1 [55]. BRCA1 binds to the STAT1a transcriptional activation domain which leads to the induction of a family of interferon- gamma regulated genes. More recently it has been reported that over-expression of BRCA1 in prostate cancer cells can cause activation of not only STAT3 but also the upstream activators of STAT3, JAK1 and JAK2 [56]. Interestingly, groups of interferon-induced genes have been shown by a microarray approach to undergo substantial co-variation in expression across a panel of breast tumours [57,58]. Inter- ferons are important regulators of cell growth and apoptosis and therefore this role of BRCA1 could be important in (a–f) Electron micrographs of BRCA1 bound to linear plasmid DNA at tumour progression. the base of DNA loops. From [63]. Tumours arising in BRCA1 mutation carriers are frequently negative for the estrogen receptor. Intriguingly, clear that microarray technology should lead to profound it has been reported that BRCA1 can inhibit estrogen insights into the functions of BRCA1 and BRCA2. receptor signalling. Overexpression of BRCA1 blocks estrogen receptor-α mediated activation of an estrogen- Promiscuous partner in transcription responsive reporter and inhibits the transcriptional activity The analysis of global gene expression changes using of the carboxy-terminal domain of estrogen receptor-α microarrays has added to the already considerable litera- [59]. Possibly this is mediated by direct interaction of the ture on the role of BRCA1 and BRCA2 in the regulation of receptor and BRCA1 [60]. Others have suggested that transcription [53]. BRCA1 and BRCA2 regulate transcrip- the androgen receptor is transcriptionally coactivated by tion in concert with a number of cofactors including the BRCA1 and also physically interacts with it [61,62]. It histone acetylases, deacetylases and associated proteins is not clear yet how BRCA1 might inhibit signalling as HDAC, CBP/p300, CtIP and P/CAF, as well as RNA poly- part of an estrogen receptor-α or androgen receptor tran- merase II holoenzyme [53]. Recently, BRCA1 was shown scriptional cascade. And the promiscuity of interactions to interact directly with a novel sequence-specific DNA involving BRCA1 may raise questions about specificity. binding protein, ZBRK1 [54]. GADD45, a gene which has However the suggestion of a role for BRCA1 in steroid been shown to be a target of BRCA1 transcriptional hormone signalling is seductive as it may help to explain control in microarray experiments [49,50], is repressed by the tissue-specific nature of the tumours caused by ZBRK1. Underscoring the significance of the interaction, BRCA1 mutation. breast cancer-derived BRCA1 mutations that disrupt the interaction between BRCA1 and ZBRK1 also block the BRCA1 branches out corepressor activity of the complex [54]. Another surprising development has been the demonstration that BRCA1 can bind directly to DNA with high Paradoxically however, BRCA1 overexpression caused an affinity but in a sequence non-specific fashion [63]. Recom- induction of GADD45 in previous studies. So how does this binant BRCA1 binds DNA as a multimer and in doing so tally with a role for BRCA1 as a corepressor of GADD45? generates DNA loops. BRCA1 has higher affinity for Review BRCA proteins R673

branched DNA than linear molecules (Figure 4). As a Complex city result of this DNA binding, BRCA1 inhibits the exonu- Much of what we know about BRCA1 and BRCA2 has clease activity of the Mre11–Rad50–Nbs1 complex that been derived from descriptions of the various multipro- has been implicated in double-stranded break repair tein complexes in which the proteins are found. One of processes, particularly that of non-homologous end joining the first complexes with which BRCA1 was found to be [64]. As mentioned above, BRCA1 deficient mouse cells associated was the RNA polymerase II holoenzyme, a exhibit a decrease in the repair of double-stranded DNA multi-subunit complex of over 1 MDa [48]. In addition, breaks by homologous recombination and an increase in BRCA1 has been shown to associate directly with the non-homologous end joining [23]. Paull et al. [63] specu- Mre11–Rad50–Nbs1 complex [66] which is responsible for late that the inhibition of the exonuclease activity of the end-processing of double-strand breaks during DNA repair Mre11–Rad50–Nbs1 complex normally provided via [64]. More recently BRCA1 has been shown to be associ- BRCA1 DNA binding might explain the increase in non- ated with other large complexes involved in DNA repair or homologous end joining seen in Brca1 deficient cells. chromatin remodelling.

Parvin has suggested a model [65] (Figure 5) incorporating The BRCA1 associated genome surveillance complex, the new data on DNA binding by BRCA1 with previous BASC [67], is larger than 2 MDa and contains at least 15 results suggesting a role for the protein in transcription- subunits. As well as BRCA1, the complex contains the coupled repair [3]. According to this model BRCA1 is asso- products of the Bloom’s syndrome and ataxia telangiecta- ciated with the RNA polymerase II holoenzyme which sia (ATM) disease genes. Four subcomplexes implicated becomes arrested at sites of DNA damage. He proposes in DNA repair — Mre11–Rad50–Nbs1, the MSH2–MSH6 that this results in the ubiquitination and degradation of heterodimer, the MLH1–PMS2 heterodimer and DNA RNA polymerase II by the BRCA1–BARD1 ubiquitin replication factor C — are present in BASC. The presence ligase heterodimer. This then might lead to dissociation of mismatch repair proteins in BASC provides support for of the complex leaving BRCA1 bound to the site of the the suggestion that BRCA1 is important for transcription lesion. BRCA1 would then recruit other factors, such as the coupled repair of oxidation-induced DNA damage [3]. Alter- Mre11–Rad50–Nbs1 complex or RAD51, to repair the DNA natively, mismatch repair proteins have been shown to have damage. Although this model accounts for some observa- a role in homology searching during double-strand break tions, it awaits compelling experimental support. repair [68] and this may also explain their presence in BASC.

Figure 5

Model of the role of BRCA1 in transcription- coupled repair. An RNA polymerase II Transcription complex Block in BRCA1-BARD1 ring holoenzyme–BRCA1–BARD1 transcription reaches lesion in transcription finger mediated complex reaches damaged DNA, depicted as double-stranded ubiquitination of a flap, where it arrests. At this point the DNA template RNA polII BRCA1–BARD1–E3 ubiquitin ligase complex becomes active leading to the ubiquitination of BRCA1 BARD1 RNA polymerase II and its degradation. BRCA1 remains bound to the DNA where it mRNA Ub recruits a complex which repairs the lesion. RNA Modified from [65]. PolII

Holoenzyme

DNA repair complex

Repaired Recruitment of Removal and degradation double-strand DNA repair complex of RNA polII. BRCA1 becomes DNA by BRCA1 bound to branched DNA Current Biology R674 Current Biology Vol 11 No 16

Many of the DNA repair proteins in BASC can bind abnor- dividing cells. Immunoflourescent staining of HeLa cells mal DNA structures such as double-strand breaks, Holli- with antibodies to BRAF35 localised the protein to mitotic day junctions and cruciform DNA. Hence the complex may chromosomes during early chromosome condensation. At act as a DNA structure surveillance unit consisting of both the onset of the metaphase to anaphase transition, this DNA damage sensors and proteins such as ATM and the localisation is lost. BRCA2 is colocalised with these mitotic related cell cycle checkpoint kinase, ATR, which could chromosomes suggesting a role in cell cycle progression; signal DNA damage. ATM and ATR can directly phospho- indeed microinjection of antibodies to BRAF35 or BRCA2 rylate BRCA1 in response to DNA damage and this is into synchronised HeLa cells delayed entry into mitosis. thought to be important for the function of BRCA1 in double-stranded break repair and cell cycle checkpoint One feature of the HMG domain is an affinity for DNA control [69–71]. As many of the proteins in BASC are with sharp bends [75], this prompted the authors [74] to involved in repairing DNA damage that can occur at repli- look at the DNA binding activity of BRAF35. Gel-shift cation forks, BRCA1, by association, may also participate assays showed that BRAF35 bound specifically to cruci- in the resolution of inappropriate DNA structures during form, or four-way junction, DNA. Such unusual DNA struc- post-replicational repair. tures are created during recombination-mediated repair and after DNA adduct formation; it may be that BRAF35 serves As well as being associated with BASC and the RNA poly- to target BRCA2 to sites of DNA damage. merase II holoenzyme, BRCA1 is also a member of a large SWI/SNF-related complex which has chromatin remodel- Summary and conclusions ling properties [72]. In particular a direct interaction was So what are we to make of these findings of disparate and demonstrated between BRCA1 and the bromo-domain apparently unrelated functions for BRCA proteins? Well containing protein BRG1. Additionally, this provides a link perhaps for BRCA1 part of the solution lies in the ubiquiti- between BRCA1 and cell cycle progression as BRG1 has nation function of the BRCA1 and BARD1 heterodimer. been shown to interact with the critical cell cycle regulator Similar models, in which the ubiquitin ligase activity is a key RB [73]. How does this tally with putative functions for feature, have been put forward for the role of BRCA1 in the BRCA1 protein? Does BRCA1 regulate transcription both polyadenylation [44] and DNA repair [64]. One could via modulation of chromatin structure or is its function in imagine scenarios where other roles of BRCA1, for example such a complex to allow access for the DNA repair in transcriptional regulation, could be partly ascribed to this machinery? Interestingly, the BRCA1–chromatin remodel- enzymatic activity. Perhaps the presence of BRCA1 in a ling complex does not contain the Mre11–Rad50–Nbs1 large number of protein complexes indicates that it has a complex, other identified BASC components or RNA poly- generalised role regulating the abundance of individual pro- merase II. Perhaps the methods of purification are critical teins or the stoichiometry of complexes. This might explain here. BASC was purified by immunoprecipitation from the rather pleiotropic cellular consequences of loss of the unfractionated HeLa nuclear extracts [67]. The SWI/SNF gene. Given the flurry of activity in the field it seems very complex on the other hand was affinity purified following likely that even more new functions for BRCA1 and BRCA2 nuclear extract fraction by column chromatography [72]. proteins will be forthcoming over the next few years. We Association with the RNA polymerase II holoenzyme was already have several plausible explanations of how loss of shown by antibody probing of RNA polymerase II-con- these genes leads to tumorigenesis but whether these actu- taining fractions from whole cell extracts which had under- ally operate in a woman carrying a BRCA mutation is still gone column chromatography [48]. These protocols each unclear. The key question of why tumours arising through yielded different complexes consisting of non-overlapping BRCA mutations are relatively specific to the breast and sets of subunits, all containing BRCA1; it seems possible ovary is also unresolved. that the different complexes may be present in the cell References simultaneously. 1. Rahman N, Stratton MR: The genetics of breast cancer susceptibility. Annu Rev Genet 1998, 32:95-121. Complexes containing BRCA2 have been less forthcoming 2. Bertwistle D, Ashworth A: BRCA1 and BRCA2. Curr Biol 2000, 10:R582. 3. Gowen LC, Avrutskaya AV, Latour AM, Koller BH, Leadon SA: BRCA1 but recently BRCA2 was also shown to be a member of a required for transcription-coupled repair of oxidative DNA large, 2 MDa complex [74]. One of the components of this damage. Science 1998, 281:1009-1012. 4. Scully R, Livingston DM: In search of the tumour-suppressor complex is BRCA2-associated factor 35kDa, or BRAF35. functions of BRCA1 and BRCA2. Nature 2000, 408:429-432. BRAF35 was identified as a structural, non-specific, DNA 5. Karran P: DNA double strand break repair in mammalian cells. binding protein, an activity conferred by the high mobility Curr Opin Genet Dev 2000, 10:144-150. 6. Khanna KK, Jackson SP: DNA double-strand breaks: signaling, group — HMG — domain of the protein. Direct interac- repair and the cancer connection. Nat Genet 2001, 27:247-254. tion of BRCA2 and BRAF35 was demonstrated in vitro and 7. Mizuta R, LaSalle JM, Cheng HL, Shinohara A, Ogawa H, Copeland N, Jenkins NA, Lalande M, Alt FW: RAB22 and RAB163/mouse BRCA2: both proteins share very similar patterns of expression proteins that specifically interact with the RAD51 protein. Proc Natl during embryogenesis with the highest levels in rapidly Acad Sci USA 1997, 94:6927-6932. Review BRCA proteins R675

8. Scully R, Chen J, Plug A, Xiao Y, Weaver D, Feunteun J, Ashley T, 31. Timmers C, Taniguchi T, Hejna J, Reifsteck C, Lucas L, Bruun D, Livingston DM: Association of BRCA1 with Rad51 in mitotic and Thayer M, Cox B, Olson S, D’Andrea AD, Moses R, Grompe M: meiotic cells. Cell 1997, 88:265-275. Positional cloning of a novel Fanconi anemia gene, FANCD2. 9. Sharan SK, Morimatsu M, Albrecht U, Lim DS, Regel E, Dinh C, Mol Cell 2001, 7:241-248. Sands A, Eichele G, Hasty P, Bradley A: Embryonic lethality and 32. Garcia-Higuera I, Taniguchi T, Ganesan S, Meyn MS, Timmers C, radiation hypersensitivity mediated by Rad51 in mice lacking Hejna J, Grompe M, D’Andrea AD: Interaction of the Fanconi Brca2. Nature 1997, 386:804-810. anemia proteins and BRCA1 in a common pathway. Mol Cell 2001, 10. Chen PL, Chen CF, Chen Y, Xiao J, Sharp ZD, Lee WH: The BRC 7:249-262. repeats in BRCA2 are critical for RAD51 binding and resistance to 33. Lorick KL, Jensen JP, Fang S, Ong AM, Hatakeyama S, methyl methanesulfonate treatment. Proc Natl Acad Sci USA Weissman AM: RING fingers mediate ubiquitin-conjugating 1998, 95:5287-5292. enzyme (E2)-dependent ubiquitination. Proc Natl Acad Sci USA 11. Davies AA, Masson JY, McIlwraith MJ, Stasiak AZ, Stasiak A, 1999, 96:11364-11369. Venkitaraman AR, West SC: Role of BRCA2 in control of the RAD51 34. Hashizume R, Fukuda M, Maeda I, Nishikawa H, Oyake D, Yabuki Y, recombination and DNA repair protein. Mol Cell 2001, 7:273-282. Ogata H, Ohta T: The RING heterodimer BRCA1-BARD1 is a 12. Tirkkonen M, Johannsson O, Agnarsson BA, Olsson H, Ingvarsson S, ubiquitin ligase inactivated by a breast cancer-derived mutation. Karhu R, Tanner M, Isola J, Barkardottir RB, Borg A, Kallioniemi OP: J Biol Chem 2001, 276:14537-14540. Distinct somatic genetic changes associated with tumor 35. Ruffner H, Joazeiro CA, Hemmati D, Hunter T, Verma IM: Cancer- progression in carriers of BRCA1 and BRCA2 germ-line predisposing mutations within the RING domain of BRCA1: Loss mutations. Cancer Res 1997, 57:1222-1227. of ubiquitin protein ligase activity and protection from radiation 13. Gretarsdottir S, Thorlacius S, Valgardsdottir R, Gudlaugsdottir S, hypersensitivity. Proc Natl Acad Sci USA 2001, 98:5134-5139. Sigurdsson S, Steinarsdottir M, Jonasson JG, Anamthawat-Jonsson K, 36. Hershko A, Ciechanover A: The ubiquitin system. Annu Rev Eyfjord JE: BRCA2 and p53 mutations in primary breast cancer in Biochem 1998, 67:425-479. relation to genetic instability. Cancer Res 1998, 58:859-862. 37. Peters JM: SCF and APC: the Yin and Yang of cell cycle regulated 14. Patel KJ, Yu VP, Lee H, Corcoran A, Thistlethwaite FC, Evans MJ, proteolysis. Curr Opin Cell Biol 1998, 10:759-768. Colledge WH, Friedman LS, Ponder BA, Venkitaraman AR: 38. Honda R, Yasuda H: Activity of MDM2, a ubiquitin ligase, toward Involvement of Brca2 in DNA repair. Mol Cell 1998, 1:347-357. p53 or itself is dependent on the RING finger domain of the 15. Xu X, Weaver Z, Linke SP, Li C, Gotay J, Wang XW, Harris CC, ligase. Oncogene 2000, 19:1473-1476. Ried T, Deng CX: Centrosome amplification and a defective G2-M 39. Joazeiro CA, Wing SS, Huang H, Leverson JD, Hunter T, Liu YC: cell cycle checkpoint induce genetic instability in BRCA1 exon 11 The tyrosine kinase negative regulator c-Cbl as a RING-type, E2- isoform-deficient cells. Mol Cell 1999, 3:389-395. dependent ubiquitin-protein ligase. Science 1999, 286:309-312. 16. Tutt A, Gabriel A, Bertwistle D, Connor F, Paterson H, Peacock J, 40. Yang Y, Fang S, Jensen JP, Weissman AM, Ashwell JD: Ubiquitin Ross G, Ashworth A: Absence of Brca2 causes genome instability protein ligase activity of IAPs and their degradation in proteasomes by chromosome breakage and loss associated with centrosome in response to apoptotic stimuli. Science 2000, 288:874-877. amplification. Curr Biol 1999, 9:1107-1110. 41. Wu LC, Wang ZW, Tsan JT, Spillman MA, Phung A, Xu XL, Yang MC, 17. Ban S, Shinohara T, Hirai Y, Moritaku Y, Cologne JB, MacPhee DG: Hwang LY, Bowcock AM, Baer R: Identification of a RING protein Chromosomal instability in BRCA1- or BRCA2-defective human that can interact in vivo with the BRCA1 gene product. Nat Genet cancer cells detected by spontaneous micronucleus assay. 1996, 14:430-440. Mutat Res 2001, 474:15-23. 42. Thai TH, Du F, Tsan JT, Jin Y, Phung A, Spillman MA, Massa HF, 18. Shen SX, Weaver Z, Xu X, Li C, Weinstein M, Chen L, Guan XY, Muller CY, Ashfaq R, Mathis JM, et al.: Mutations in the BRCA1- Ried T, Deng CX: A targeted disruption of the murine Brca1 gene associated RING domain (BARD1) gene in primary breast, ovarian causes gamma-irradiation hypersensitivity and genetic instability. and uterine cancers. Hum Mol Genet 1998, 7:195-202. Oncogene 1998, 17:3115-3124. 43. Kleiman FE, Manley JL: Functional interaction of BRCA1-associated 19. Scully R, Ganesan S, Vlasakova K, Chen J, Socolovsky M, BARD1 with polyadenylation factor CstF-50. Science 1999, Livingston DM: Genetic analysis of BRCA1 function in a defined 285:1576-1579. tumor cell line. Mol Cell 1999, 4:1093-1099. 44. Kleiman FE, Manley JL: The BARD1-CstF-50 interaction links mRNA 20. Yu V: Caretaker Brca1: keeping the genome in the straight and 3′ end formation to DNA damage and tumor suppression. Cell narrow. Breast Cancer Res 2000, 2:82-85. 2001, 104:743-753. 21. Wang H, Zeng ZC, Bui TA, DiBiase SJ, Qin W, Xia F, Powell SN, 45. Colgan DF, Murthy KG, Prives C, Manley JL: Cell-cycle related Iliakis G: Nonhomologous end-joining of ionizing radiation- regulation of poly(A) polymerase by phosphorylation. Nature induced DNA double-stranded breaks in human tumor cells 1996, 384:282-285. deficient in BRCA1 or BRCA2. Cancer Res 2001, 61:270-277. 46. Zhao W, Manley JL: Deregulation of poly(A) polymerase interferes 22. Johnson RD, Jasin M: Double-strand-break-induced homologous with cell growth. Mol Cell Biol 1998, 18:5010-5020. recombination in mammalian cells. Biochem Soc Trans 2001, 47. McCracken S, Fong N, Yankulov K, Ballantyne S, Pan G, Greenblatt J, 29:196-201. Patterson SD, Wickens M, Bentley DL: The C-terminal domain of 23. Moynahan ME, Chiu JW, Koller BH, Jasin M: Brca1 controls RNA polymerase II couples mRNA processing to transcription. homology-directed DNA repair. Mol Cell 1999, 4:511-518. Nature 1997, 385:357-361. 24. Moynahan ME, Pierce AJ, Jasin M: BRCA2 is required for homology- 48. Scully R, Anderson SF, Chao DM, Wei W, Ye L, Young RA, directed repair of chromosomal breaks. Mol Cell 2001, 7:263-272. Livingston DM, Parvin JD: BRCA1 is a component of the RNA 25. Cantor SB, Bell DW, Ganesan S, Kass EM, Drapkin R, Grossman S, polymerase II holoenzyme. Proc Natl Acad Sci USA 1997, Wahrer DC, Sgroi DC, Lane WS, Haber DA, Livingston DM: BACH1, 94:5605-5610. a novel helicase-like protein, interacts directly with BRCA1 and 49. Harkin DP, Bean JM, Miklos D, Song YH, Truong VB, Englert C, contributes to its DNA repair function. Cell 2001, 105:149-160. Christians FC, Ellisen LW, Maheswaran S, Oliner JD, Haber DA: 26. Weber CA, Salazar EP, Stewart SA, Thompson LH: ERCC2: cDNA Induction of GADD45 and JNK/SAPK-dependent apoptosis cloning and molecular characterization of a human nucleotide following inducible expression of BRCA1. Cell 1999, 97:575-586. excision repair gene with high homology to yeast RAD3. EMBO J 50. MacLachlan TK, Somasundaram K, Sgagias M, Shifman Y, 1990, 9:1437-1447. Muschel RJ, Cowan KH, El-Deiry WS: BRCA1 effects on the cell 27. Vermeulen W, de Boer J, Citterio E, van Gool AJ, van der Horst GT, cycle and the DNA damage response are linked to altered gene Jaspers NG, de Laat WL, Sijbers AM, van der Spek PJ, Sugasawa K, expression. J Biol Chem 2000, 275:2777-2785. et al.: Mammalian nucleotide excision repair and syndromes. 51. Hedenfalk I, Duggan D, Chen Y, Radmacher M, Bittner M, Simon R, Biochem Soc Trans 1997, 25:309-315. Meltzer P, Gusterson B, Esteller M, Kallioniemi OP, et al.: Gene- 28. Coin F, Egly JM: Ten years of TFIIH. Cold Spring Harb Symp Quant expression profiles in hereditary breast cancer. New Engl J Med Biol 1998, 63:105-110. 2001, 344:539-548. 29. Easton DF: How many more breast cancer predisposition genes 52. Lakhani SR, O’Hare MJ, Ashworth A: Profiling familial breast cancer. are there? Breast Cancer Res 1999, 1:14-17. Nat Med 2001, 7:408-410. 30. D’Andrea AD, Grompe M: Molecular biology of Fanconi anemia: 53. Monteiro AN: BRCA1: exploring the links to transcription. Trends implications for diagnosis and therapy. Blood 1997, 90:1725-1736. Biochem Sci 2000, 25:469-474. R676 Current Biology Vol 11 No 16

54. Zheng L, Pan H, Li S, Flesken-Nikitin A, Chen LP, Boyer GT, Lee HW: 76. Khanna KK, Jackson SP: DNA double-strand breaks : signaling, Sequence-specific transcriptional corepressor function for BRCA1 repair and the cancer connection. Nat Genet 2001, 27:247-254. through a novel zinc finger protein, ZBRK1. Mol Cell 2000, 77. Weissman AM: Themes and variations on ubiquitylation. Nat Rev 6:757-768. Mol Cell Biol 2001, 2:169-178. 55. Ouchi T, Lee SW, Ouchi M, Aaronson SA, Horvath CM: Collaboration of signal transducer and activator of transcription 1 (STAT1) and BRCA1 in differential regulation of IFN-gamma target genes. Proc Natl Acad Sci USA 2000, 97:5208-5213. 56. Gao B, Shen X, Kunos G, Meng Q, Goldberg ID, Rosen EM, Fan S: Constitutive activation of JAK-STAT3 signaling by BRCA1 in human prostate cancer cells. FEBS Lett 2001, 488:179-184. 57. Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, Pollack JR, Ross DT, Johnsen H, Akslen LA, et al.: Molecular portraits of human breast tumours. Nature 2000, 406:747-752. 58. Perou CM, Jeffrey SS, van de Rijn M, Rees CA, Eisen MB, Ross DT, Pergamenschikov A, Williams CF, Zhu SX, Lee JC, et al.: Distinctive gene expression patterns in human mammary epithelial cells and breast cancers. Proc Natl Acad Sci USA 1999, 96:9212-9217. 59. Fan S, Ma YX, Wang C, Yuan RQ, Meng Q, Wang JA, Erdos M, Goldberg ID, Webb P, Kushner PJ, et al.: Role of direct interaction in BRCA1 inhibition of estrogen receptor activity. Oncogene 2001, 20:77-87. 60. Fan S, Wang J, Yuan R, Ma Y, Meng Q, Erdos MR, Pestell RG, Yuan F, Auborn KJ, Goldberg ID, Rosen EM: BRCA1 inhibition of estrogen receptor signaling in transfected cells. Science 1999, 284:1354 1356. 61. Park JJ, Irvine RA, Buchanan G, Koh SS, Park JM, Tilley WD, Stallcup MR, Press MF, Coetzee GA: Breast cancer susceptibility gene 1 (BRCAI) is a coactivator of the androgen receptor. Cancer Res 2000, 60:5946-5949. 62. Yeh S, Hu YC, Rahman M, Lin HK, Hsu CL, Ting HJ, Kang HY, Chang C: Increase of androgen-induced cell death and androgen receptor transactivation by BRCA1 in prostate cancer cells. Proc Natl Acad Sci USA 2000, 97:11256-11261. 63. Paull TT, Cortez D, Bowers B, Elledge SJ, Gellert M: Direct DNA binding by Brca1. Proc Natl Acad Sci USA 2001, 98:6086-6091. 64. Tsubouchi H, Ogawa H: A novel mre11 mutation impairs processing of double-strand breaks of DNA during both mitosis and meiosis. Mol Cell Biol 1998, 18:260-268. 65. Parvin JD: BRCA1 at a branch point. Proc Natl Acad Sci USA 2001, 98:5952-5954. 66. Zhong Q, Chen CF, Li S, Chen Y, Wang CC, Xiao J, Chen PL, Sharp ZD, Lee WH: Association of BRCA1 with the hRad50- hMre11-p95 complex and the DNA damage response. Science 1999, 285:747-750. 67. Wang Y, Cortez D, Yazdi P, Neff N, Elledge SJ, Qin J: BASC, a super complex of BRCA1-associated proteins involved in the recognition and repair of aberrant DNA structures. Genes Dev 2000, 14:927-939. 68. Elliott B, Jasin M: Repair of double-strand breaks by homologous recombination in mismatch repair-defective mammalian cells. Mol Cell Biol 2001, 21:2671-2682. 69. Cortez D, Wang Y, Qin J, Elledge SJ: Requirement of ATM- dependent phosphorylation of Brca1 in the DNA damage response to double-strand breaks. Science 1999, 286:1162-1166. 70. Tibbetts RS, Cortez D, Brumbaugh KM, Scully R, Livingston D, Elledge SJ, Abraham RT: Functional interactions between BRCA1 and the checkpoint kinase ATR during genotoxic stress. Genes Dev 2000, 14:2989-3002. 71. Xu B, Kim St S, Kastan MB: Involvement of Brca1 in S-Phase and G2-Phase Checkpoints after Ionizing Irradiation. Mol Cell Biol 2001, 21:3445-3450. 72. Bochar DA, Wang L, Beniya H, Kinev A, Xue Y, Lane WS, Wang W, Kashanchi F, Shiekhattar R: BRCA1 is associated with a human SWI/SNF-related complex: linking chromatin remodeling to breast cancer. Cell 2000, 102:257-265. 73. Dunaief JL, Strober BE, Guha S, Khavari PA, Alin K, Luban J, Begemann M, Crabtree GR, Goff SP: The retinoblastoma protein and BRG1 form a complex and cooperate to induce cell cycle arrest. Cell 1994, 79:119-130. 74. Marmorstein LY, Kinev AV, Chan GK, Bochar DA, Beniya H, Epstein JA, Yen TJ, Shiekhattar R: A human BRCA2 complex containing a structural DNA binding component influences cell cycle progression. Cell 2001, 104:247-257. 75. Bianchi ME, Beltrame M, Paonessa G: Specific recognition of cruciform DNA by nuclear protein HMG1. Science 1989, 243:1056-1059.