Abraxas and RAP80 Form a BRCA1 Protein Complex Required for the DNA Damage Response

See allHide authors and affiliations

Science  25 May 2007:
Vol. 316, Issue 5828, pp. 1194-1198
DOI: 10.1126/science.1139476


The BRCT repeats of the breast and ovarian cancer predisposition protein BRCA1 are essential for tumor suppression. Phosphopeptide affinity proteomic analysis identified a protein, Abraxas, that directly binds the BRCA1 BRCT repeats through a phospho-Ser-X-X-Phe motif. Abraxas binds BRCA1 to the mutual exclusion of BACH1 (BRCA1-associated C-terminal helicase) and CtIP (CtBP-interacting protein), forming a third type of BRCA1 complex. Abraxas recruits the ubiquitin-interacting motif (UIM)–containing protein RAP80 to BRCA1. Both Abraxas and RAP80 were required for DNA damage resistance, G2-M checkpoint control, and DNA repair. RAP80 was required for optimal accumulation of BRCA1 on damaged DNA (foci) in response to ionizing radiation, and the UIM domains alone were capable of foci formation. The RAP80-Abraxas complex may help recruit BRCA1 to DNA damage sites in part through recognition of ubiquitinated proteins.

The BRCA1 tumor suppressor is associated with hereditary breast and ovarian cancer and functions in maintenance of genomic stability (1, 2). BRCA1 contains an N-terminal RING domain, a Ser-Gln (SQ) cluster domain (3), and two BRCT (BRCA1 C-terminal) repeats, which constitute a phosphopeptide recognition domain that binds peptides containing a phospho-SXXF motif (S is Ser, F is Phe, and X varies) (46) and is required for tumor suppression. By dimerizing with BARD1 (BRCA1-associated RING domain protein) through the RING domain, BRCA1 forms an E3 ubiquitin ligase (710).

To identify proteins that bind BRCA1 BRCT domains, we combined peptide affinity purification, stable isotope labeling with amino acids in cell culture, and mass spectrometry (1113) to identify and quantify phosphopeptides that directly bind to BRCA1 BRCT domains (fig. S1A). Proteins from cells grown in medium containing heavy isotope and cells grown in normal medium that were treated with 10 Gy ionizing radiation (IR) were prepared, mixed (1:1), and digested with trypsin. Tryptic peptides that bound to a glutathione S-transferase (GST)–BRCA1-BRCT fusion protein were identified. Phosphopeptides containing a pSXXF motif were compared with the list of proteins we recently identified as substrates of ATM (mutated in ataxia telangiectasia) and ATR (ATM and RAD3-related) (14). In addition to the known BRCA1-binding proteins BACH1 and CtIP, we identified peptides representing a potential ATM or ATR substrate and a novel protein, FLJ13614, that we named Abraxas, and the gene ABRA1 (fig. S1). A doubly phosphorylated Abraxas peptide, GFGEYS#RS#PTF, containing pSer404 and pSer406 was enriched eightfold after IR, whereas the singly pSer406-containing peptide bound, but was not enriched after IR (fig. S1). Synthetic peptides containing pSer406 bound GST-BRCA1-BRCT, whereas the pSer404 and unphosphorylated peptides did not (Fig. 1A and fig. S2A).

Fig. 1.

Identification of Abraxas and RAP80 as proteins interacting with BRCA1-BRCT domains. (A) Binding of phosphorylated ABRA1 peptides to purified recombinant BRCA1-BRCT domains. Biotinylated peptides on streptavidin beads associated with purified recombinant GST-BRCA1-BRCT was visualized by Coomassie staining. The # symbol indicates a phosphate resides on the previous residue. (B) Specific binding of Abraxas to BRCA1-BRCT domains. Hemagglutinin-tagged ABRA1 (HA-ABRA1) was expressed in 293T cells, and cell lysates were incubated with various purified GST-tagged BRCT domains. (C) HA-ABRA1 association with endogenous BRCA1 is dependent on Ser406 phosphorylation. HA-ABRA1 wild-type or mutant proteins were expressed in 293T cells. (D) Phosphorylation of Ser406 of ABRA1 in vivo. Lysates prepared from 293T cells were untreated or treated with IR, untreated or treated with λ-phosphatase. (E) Immunoprecipitation of Abraxis with phospho-SQ or TQ antibodies. (F) RAP80 was identified in a TAP purification of BRCA1-BRCT domain–associated proteins. Retroviruses expressing either TAP only (TAP) or C-terminal TAP-tagged BRCT domain of BRCA1 (BRCT-TAP) were introduced into HeLa cells, and the infected cells were used for purification. A Coomassie-stained gel is shown. (G) Phosphorylation of RAP80 in response to IR. (H) Recognition of RAP80 by phospho-antibodies to ATM-ATR substrates. Proteins were immunoprecipitated from 293T cells lysates with antibodies to RAP80 and probed with the indicated antibodies.

Abraxas is conserved in vertebrates (fig. S3). Bioinformatics analysis revealed a protein KIAA0157, we named ABRO1 (Abraxas Brother 1) (fig. S1D), which is 39% identical to ABRA1 in the N-terminal two-thirds of the protein (amino acids 1 to 260), including a coil-coil domain and a region we call the “ABR” domain; however, ABRO1 lacks the pSXXF motif and did not bind to BRCA1.

Abraxas specifically bound to the BRCT repeats of wild-type BRCA1 (Fig. 1B), but not a cancer-causing BRCT mutant, M1775R (fig. S2B). Binding of ABRA1 to BRCA1 requires phosphorylation (Fig. 1C and fig. S2, C and D). Ser406 phosphorylation was confirmed in vivo with a phosphospecific antibody but was not increased in cells exposed to IR (Fig. 1D and fig. S2E).

A search for additional BRCA1-BRCT–binding proteins that bound to tandem affinity purification (TAP)–tagged BRCA1-BRCT domains (15) identified RAP80, a ubiquitin-interacting motif (UIM)– and zinc finger–containing protein that interacts with retinoid-related testis-associated receptor in vitro (Fig. 1F) (16). RAP80 was found to be phosphorylated on three sites, Ser140, Ser402, and Ser419, in response to IR in our ATM and ATR substrate screen (14). RAP80 from extracts associates with GST-BRCA1-BRCT in vitro and BRCA1 in vivo in a phosphorylation-dependent manner (fig. S5 and S11). IR treatment induced a mobility shift for Abraxas and RAP80 (Fig. 1, D and G), and both proteins could be detected by antibodies to phospho-SQ sites after IR treatment (Fig. 1, E and 1H). These data combined suggest that Abraxas and RAP80 are ATM or ATR substrates.

Cells depleted of Abraxas or RAP80 (fig. S8) exhibited hypersensitivity to IR and ultraviolet (UV) light (Fig. 2A), G2-M checkpoint defects (Fig. 2B and fig. S6), and defects in homologous recombination (HR) repair (Fig. 2C and fig. S6). Each of these defects was less severe than defects in BRCA1-depleted cells (1720), which suggests that Abraxas or RAP80 mediates a subset of BRCA1 functions.

Fig. 2.

Role of ABRA1 and RAP80 in DNA damage responses. (A) ABRA1-depleted or RAP80-depleted cells were sensitive to IR and UV. U2OS cells were treated with control oligos or siRNAs to ABRA1, RAP80, or BRCA1; incubated for 2 days, plated at low density, and irradiated; and colonies were counted after 14 days. (B) Analysis of the G2-M checkpoint. Cells were untreated or treated with 3 Gy IR, then incubated for 1 hour at 37°C before fixation and p-H3 antibody staining. Three independent experiments were performed with siRNA to ABRA1. Two independent experiments were performed with siRNA to RAP80 that yielded similar results. (C) Defective HR in ABRA1 or RAP80 siRNA-treated cells. DR-U2OS cells were transfected with siRNAs to the indicated genes. siRNAs to BRCA1 or BRCA2 were a mixture of three different siRNAs for each gene. Individual siRNA were used for ABRA1 or RAP80. Three independent experiments were performed with siRNAs to luciferase, BRCA1, BRCA2, and ABRA1. The number of green positive cells generated from each sample relative to the number from the luciferase control was indicated as the percentage of HR.

Abraxas and RAP80 form foci that colocalize with BRCA1 and RAD51 in S or G2 phase U2OS cells (21) (Fig. 3, A and B). When a UV laser was used to micro-irradiate cells, ABRA1 relocalized to sites of DNA damage within 15 min (Fig. 3C). Unlike BACH1 and CtIP (2224), ABRA1 and RAP80 foci formation was BRCA1-independent. The ABRA1 S406A mutant (in which Ala replaces Ser at codon 406), which does not bind BRCA1, efficiently localized to IR-induced (fig. S7A) and UV laser–induced (Fig. 3C) DNA damage. Furthermore, stably expressed green fluorescent protein (GFP)–tagged ABRA1 and RAP80 (fig. S7B) relocalized to UV DNA damage sites efficiently in BRCA1-deficient HCC1937 cells.

Fig. 3.

ABRA1 and RAP80 form DNA damage–induced foci. (A and B) Colocalization of ABRA1 and RAP80 with BRCA1 and RAD51 at DNA damage–induced foci. U2OS cells were untreated or treated with 10 Gy IR, incubated for 2 hours, fixed and immunostained with antibodies to ABRA1, BRCA1, or RAD51, followed by cognate Alex 488–conjugated (green) or Cy3-conjuated (red) antibodies. (C) Localization of ABRA1 to sites of DNA damage is independent of BRCA1 binding. Laser microirradiation of U2OS cells with retrovirally expressed GFP-tagged wild-type or mutant ABRA1 (S406A) was performed. Cells were fixed and stained 15 min after laser treatment. (D) Defective BRCA1 foci formation in RAP80-depleted cells. U2OS cells were transfected with control or RAP80 siRNAs for 2 days, then irradiated with 10 Gy IR, incubated for 2 hours, and fixed and immunostained with indicated antibodies. More than 400 cells were analyzed, and cells containing more than 10 BRCA1 foci were deemed positive. (E) Dependence of RAP80 IRIF formation on UIM domains. U2OS cells containing retroviruses expressing GFP-WT, GFP-UIM or GFP-UIMΔ were irradiated with 10 Gy IR, incubated for 2 hours, fixed and stained with antibodies against GFP followed with Alexa 488 secondary antibodies. More than 300 cells were counted to determine the percentage of cells forming foci for each cell line. Immunoprecipitation of Abraxas revealed binding to the C-terminal region of RAP80 lacking the UIM domain.

Depletion of RAP80 reduced foci formation of BRCA1. Although 80% of control cells formed IR-induced foci (IRIF) for BRCA1, only 28% of RAP80-depleted cells form BRCA1 IRIFs (Fig. 3D). Three nonoverlapping small interfering RNAs (siRNAs) to RAP80 were used in multiple experiments, and similar results were observed.

Deletion analysis determined that the RAP80 UIM domains alone could form IRIFs, although not as efficiently as the full-length protein (only 50% irradiated cells formed IRIF). Point mutations that abolish ubiquitin binding (A88S, S92A, A113S, S117A) prevent foci formation in the fragment containing UIM alone, but not when present in the full-length protein (fig. S10). RAP80 lacking the UIM domains also formed foci, but inefficiently (Fig. 3E), and the percentage of cells with foci did not increase in response to IR. Thus, RAP80 appears to have two different means of forming foci, but only the UIM domain responds to IR.

Abraxas binds BRCA1 mutually exclusively with BACH1 and CtIP as determined by coimmunoprecipitation (Fig. 4A). This is consistent with each of these proteins associating with BRCA1 through the same site on the BRCT motifs. As RAP80 does not contain a pSXXF motif, it might associate with BRCA1 indirectly. A substantial portion of RAP80 could be coimmunoprecipitated with ABRA1 (Fig. 4A). RAP80 lacking the UIM domains was still associated with ABRA1 (Fig. 3E), which indicated that a UIM-independent Abraxas-binding domain was present. The RAP80-ABRA1 interaction is BRCA1-independent, because the S406A ABRA1 mutant maintained RAP80 binding (fig. S9) and because RAP80-ABRA1 association was intact in HCC1937 cells (Fig. 4B). Unlike RAP80-BRCA1, the RAP80-ABRA1 interaction is phosphorylation-independent (fig. S11). Therefore, ABRA1 and RAP80 form a complex that interacts with BRCA1.

Fig. 4.

Association of ABRA1 and RAP80 in a complex with BRCA1. (A) BRCA1 forms distinct complexes with ABRA1, BACH1, and CtIP. Proteins were immunoprecipitated with antibodies to BRCA1, BACH1, CtIP, ABRA1, and RAP80 from lysates of 293T cells treated with IR or untreated. (B) Intact RAP80-ABRA1 interaction in HCC1937 cells that lack a functional BRCA1. HCC1937 cells were either untreated or treated with IR. Proteins from lysates were immunoprecipitated with antibodies against RAP80 or ABRA1 or with control immunoglobulin IgG. (C) Decreased RAP80-BRCA1 interaction in ABRA1-depleted cells. 293T cells were transfected with control (C) or siRNA oligos to ABRA1, BACH1, or CtIP (Si). After 48 hours, proteins were immunoprecipitated with antibodies to RAP80. Immunoblotting was performed with the indicated antibodies.

CtIP was detected in RAP80 immunoprecipitates (Fig. 4A and fig. S12). To determine the extent Abraxas and CtIP mediate RAP80 binding to BRCA1, we immunoprecipitated RAP80 from cells depleted for ABRA1, BACH1, or CtIP. The RAP80-BRCA1 interaction was decreased when ABRA1, but not BACH1 or CtIP, was depleted (Fig. 4C). Therefore, RAP80 interacts with BRCA1 largely through binding to Abraxas. As Bard1 is also present in ABRA1-RAP80-BRCA1 complexes, ABRA1 and RAP80 might mediate the E3 ligase activity of BRCA1-Bard1 heterodimers.

Our data, together with that from previous studies (23, 25), suggest that BRCA1 BRCT domains form mutually exclusive complexes with ABRA1, BACH1, and CtIP through the pSXXF motif. These proteins may serve as adaptor proteins to recruit the BRCA1-Bard1 E3 ubiquitin ligase to specific target proteins analogous to F-box proteins' role in the SCF (Skp1, cullin, F-box) ubiquitin ligase ubiquitination pathway (26, 27). To distinguish these complexes, we refer to them as the BRCA1 A complex (ABRA1), B complex (BACH1), and C complex (CtIP) (fig. S13).

Abraxas and its paralog, ABRO1, have no known functional motifs, whereas RAP80 contains multiple ubiquitin-interaction motifs. Because RAP80 UIM domains form foci, it is likely that RAP80 localizes to DNA damage sites through its UIM domains by interacting with ubiquitinated proteins at the damaged sites. Furthermore, as RAP80 is required for at least a portion of BRCA1 IRIFs, it may recruit Abraxas-BRCA1 (and possibly CtIP-BRCA1) complexes to DNA damage sites where they may ubiquitinate additional proteins, possibly amplifying ubiquitination in the same way Mdc1 amplifies H2AX phosphorylation by recruiting ATM (28).

The BRCA1 A complex is clearly involved in the DNA damage response. However depletion of ABRA1 or RAP80 did not have as strong an effect on the various DNA damage response assays as BRCA1 depletion (fig. S6), which suggests that the BRCA1 A complex controls only part of BRCA1's role in these processes. It is likely that different BRCA1 complexes play redundant roles or promote multiple distinct steps in various DNA damage responses. For instance, all three complexes are required for HR (22) (fig. S6C). Furthermore, both the A and C complexes are required for the G2-M checkpoint (fig. S6B), which suggests they also perform different functions needed for cell cycle arrest. Complexes A and C are also implicated in transcription through their association with RAP80. It is noteworthy that RAP80 was recently found to bind the estrogen receptor (29), which suggests that the A or C complex might mediate BRCA1's role in estrogen signaling in breast cancer. The identification of three distinct BRCA1 complexes should allow us to specifically dissect the role of each in the DNA damage response and tumorigenesis.

Supporting Online Material

Materials and Methods

Figs. S1 to S12


References and Notes

Stay Connected to Science

Navigate This Article