Two redundant ubiquitin‐dependent pathways of BRCA1 localization to DNA damage sites
2021; Springer Nature; Volume: 22; Issue: 12 Linguagem: Inglês
10.15252/embr.202153679
ISSN1469-3178
AutoresAlana Sherker, Natasha Chaudhary, Salomé Adam, Anne Margriet Heijink, Sylvie M. Noordermeer, Amélie Fradet‐Turcotte, Daniel Durocher,
Tópico(s)Chromosomal and Genetic Variations
ResumoReport2 November 2021Open Access Source DataTransparent process Two redundant ubiquitin-dependent pathways of BRCA1 localization to DNA damage sites Alana Sherker Alana Sherker orcid.org/0000-0001-5827-4678 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada Search for more papers by this author Natasha Chaudhary Natasha Chaudhary Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Salomé Adam Salomé Adam Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Anne Margriet Heijink Anne Margriet Heijink orcid.org/0000-0001-9229-0575 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Sylvie M Noordermeer Sylvie M Noordermeer orcid.org/0000-0003-2737-9690 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Amélie Fradet-Turcotte Amélie Fradet-Turcotte orcid.org/0000-0002-5431-8650 CHU de Québec Research Center-Université Laval (L'Hôtel-Dieu de Québec), Cancer Research Center, Québec, QC, Canada Search for more papers by this author Daniel Durocher Corresponding Author Daniel Durocher [email protected] orcid.org/0000-0003-3863-8635 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada Search for more papers by this author Alana Sherker Alana Sherker orcid.org/0000-0001-5827-4678 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada Search for more papers by this author Natasha Chaudhary Natasha Chaudhary Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Salomé Adam Salomé Adam Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Anne Margriet Heijink Anne Margriet Heijink orcid.org/0000-0001-9229-0575 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Sylvie M Noordermeer Sylvie M Noordermeer orcid.org/0000-0003-2737-9690 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Search for more papers by this author Amélie Fradet-Turcotte Amélie Fradet-Turcotte orcid.org/0000-0002-5431-8650 CHU de Québec Research Center-Université Laval (L'Hôtel-Dieu de Québec), Cancer Research Center, Québec, QC, Canada Search for more papers by this author Daniel Durocher Corresponding Author Daniel Durocher [email protected] orcid.org/0000-0003-3863-8635 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada Search for more papers by this author Author Information Alana Sherker1,2, Natasha Chaudhary1, Salomé Adam1, Anne Margriet Heijink1, Sylvie M Noordermeer1,4, Amélie Fradet-Turcotte3 and Daniel Durocher *,1,2 1Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada 2Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada 3CHU de Québec Research Center-Université Laval (L'Hôtel-Dieu de Québec), Cancer Research Center, Québec, QC, Canada 4Present address: Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands *Corresponding author. Tel: +1 416 586 4800; E-mail: [email protected] EMBO Reports (2021)22:e53679https://doi.org/10.15252/embr.202153679 See also: A Panagopoulos & M Altmeyer (December 2021) PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract The tumor suppressor BRCA1 accumulates at sites of DNA damage in a ubiquitin-dependent manner. In this work, we revisit the role of RAP80 in promoting BRCA1 recruitment to damaged chromatin. We find that RAP80 acts redundantly with the BRCA1 RING domain to promote BRCA1 recruitment to DNA damage sites. We show that that RNF8 E3 ligase acts upstream of both the RAP80- and RING-dependent activities, whereas RNF168 acts uniquely upstream of the RING domain. BRCA1 RING mutations that do not impact BARD1 interaction, such as the E2 binding-deficient I26A mutation, render BRCA1 unable to accumulate at DNA damage sites in the absence of RAP80. Cells that combine BRCA1 I26A and mutations that disable the RAP80–BRCA1 interaction are hypersensitive to PARP inhibition and are unable to form RAD51 foci. Our results suggest that in the absence of RAP80, the BRCA1 E3 ligase activity is necessary for recognition of histone H2A Lys13/Lys15 ubiquitylation by BARD1, although we cannot rule out the possibility that the BRCA1 RING facilitates ubiquitylated nucleosome recognition in other ways. SYNOPSIS This study reveals that the localization of BRCA1 at DNA damage sites involves two redundant ubiquitin-dependent pathways controlled by the RNF8 and RNF168 E3 ubiquitin ligases, with a contribution from the E3 ligase activity of BRCA1. Genetic deletion of RAP80 in human cells does not impair recruitment of BRCA1 to DNA double-strand break sites. RNF168 promotes BRCA1 recruitment to DNA lesions in the absence of RAP80. The BRCA1 RING domain is critical for BRCA1 recruitment to DNA lesions in the absence of RAP80 and this activity may involve the BRCA1 E3 ligase activity. Introduction BRCA1 is encoded by the first familial breast and ovarian cancer tumor suppressor gene identified (Futreal et al, 1994; Miki et al, 1994). The mechanism by which BRCA1 suppresses oncogenesis is most likely linked to its function in activating DNA repair by homologous recombination (HR) (Moynahan et al, 1999; Bhattacharyya et al, 2000), although other mechanisms have also been proposed (Tarsounas & Sung, 2020). BRCA1 localizes to sites of DNA damage (Scully et al, 1997a, 1997b; Paull et al, 2000), implying that BRCA1 acts to promote DNA repair directly at DNA lesions, but this link has yet to be formally established. BRCA1 is a modular protein of 1,863 amino acid residues with a RING finger domain located at the N terminus and a coiled-coil region as well as two tandem BRCT domains at the C terminus (Fig 1A). BRCA1 forms an obligatory heterodimer with the BRCA1-associated RING domain protein (BARD1) through an interaction via their respective RING finger domains (Fig 1A). This interaction contributes to the stability of both proteins and confers E3 ubiquitin ligase activity to the BRCA1–BARD1 complex toward the C terminus of histone H2A, specifically the K125/K127/K129 residues (Wu et al, 1996; Kalb et al, 2014; Densham et al, 2016; Nakamura et al, 2019; Becker et al, 2021). Figure 1. RAP80 is not necessary for BRCA1 recruitment to DSB sites A. Schematic representation of BRCA1 and BARD1. Highlighted are RING domains (orange), PALB2-interacting coiled-coil region (CC; red), BRCA1 C-terminal domains (BRCT; green), and ankyrin repeats (ANK; pink). B. Immunoblotting of whole-cell extracts obtained from parental (WT) and RAP80−/− clones in the RPE-1 hTERT p53−/− Cas9, RPE-1 hTERT p53−/− BRCA1−/− Cas9, U2OS, U2OS 2-6-3 and U2OS Flp-In/TREx cell lines with RAP80 antibodies. Tubulin is used as loading control. Representative of at least two independent immunoblots. See Fig EV1 for further details on the gene editing strategy used to knockout RAP80. C, D. The indicated parental and RAP80−/− cell lines were processed for immunofluorescence 1 h post-irradiation (10 Gy) and stained with antibodies against ABRAXAS1 and γH2AX. Shown in (C) is the percentage of cells with > 5 ABRAXAS1 foci that colocalize with γH2AX. A minimum of 100 cells per replicate were analyzed, and the bars represent mean ± SD (n = 3 biological replicates). Representative micrographs of RPE1-hTERT p53−/− Cas9 wild type (WT) and RAP80−/− cells are shown in (D). E, F. The indicated cell lines were processed for immunofluorescence 1 h post-irradiation (10 Gy) and stained with antibodies against BRCA1 and γH2AX. Shown in (E) is the percentage of cells with > 5 BRCA1 foci that colocalize with γH2AX. A minimum of 100 cells per replicate were analyzed, and the bars represent mean ± SD (n = 3 biological replicates for all conditions, except n = 5 for U2OS WT, n = 2 for U2OS 2-6-3 cells, n = 7, 5, 4 for U2OS Flp-In/T-REx cells). Representative micrographs of RPE-1 hTERT p53−/− Cas9 wild type (WT) and RAP80−/− are shown in (F). G. RPE-1 hTERT p53−/− Cas9 parental (WT) and RAP80−/− cells were untreated (t = 0 h) or irradiated with a 10 Gy dose. Samples were collected at the indicated time points, processed for immunofluorescence, and stained with antibodies against BRCA1 and γH2AX. Shown is the percentage of cells with > 5 BRCA1 foci. A minimum of 100 cells per replicate were analyzed, and the datapoints represent mean ± SD (n = 3 biological replicates). All scale bars are 5 μm. Source data are available online for this figure. Source Data for Figure 1 [embr202153679-sup-0003-SDataFig1.xlsx] Download figure Download PowerPoint BRCA1 accumulates on the chromatin surrounding DNA damage sites in a manner that depends on histone H2AX phosphorylation and the RNF8- and RNF168-catalyzed histone ubiquitylation cascade (Celeste et al, 2003; Huen et al, 2007; Kolas et al, 2007; Mailand et al, 2007; Sobhian et al, 2007; Doil et al, 2009; Stewart et al, 2009). BRCA1 does not contain any recognizable ubiquitin-binding domain but interacts with BRCA1-A, a large ubiquitin-binding complex formed by ABRAXAS1, RAP80, BABAM1, BABAM2, and BRCC3 proteins (Kim et al, 2007b; Liu et al, 2007; Wang et al, 2007) (Shao et al, 2009; Kyrieleis et al, 2016; Rabl et al, 2019). BRCA1 binds to BRCA1-A via its tandem BRCT domains that recognize phosphorylated ABRAXAS1 (also known as Abraxas or FAM175A). Within BRCA1-A, the RAP80 subunit (also known as UIMC1) has high affinity for the Lys63-linked ubiquitin (UbK63) chains produced by RNF8 and RNF168, thereby providing a means for BRCA1 recruitment to DNA lesions (Kim et al, 2007a; Sobhian et al, 2007; Sims & Cohen, 2009; Walters & Chen, 2009; Hu et al, 2012; Rabl et al, 2019). BRCA1-A is one of at least three BRCA1 protein complexes mediated by the tandem BRCT domain-dependent recognition of phosphorylated epitopes: BRCA1-B is formed by interaction with phospho-BACH1 and BRCA1-C is formed by interactions with phospho-CtIP (Wang, 2012). Although this model of BRCA1 recruitment is attractive, loss of the BRCA1-A complex results in increased DNA end-resection and higher levels of HR detectable as gene conversion (Coleman & Greenberg, 2011; Dever et al, 2011; Hu et al, 2011), which is in contrast to the loss of end-resection and HR activity seen in BRCA1-deficient cells (Stark et al, 2004; Schlegel et al, 2006; Cruz-Garcia et al, 2014). These observations suggest either that BRCA1 localization to DSB sites is irrelevant for its function during HR or that there are elements of BRCA1 localization to DSB sites that remain unresolved. In support of the latter possibility, RNA interference studies showed that RAP80 was dispensable for the initial BRCA1 localization at DNA damage sites but was instead proposed to be involved in the maintenance of BRCA1 on damaged DNA (Hu et al, 2011). This work implied that other mechanisms of BRCA1 recruitment to ubiquitylated chromatin must exist. In an effort to develop a better understanding of the mechanisms of BRCA1 recruitment to DNA damage sites, we revisited the contribution of RAP80 to BRCA1 localization to DSB sites using genetic knockouts of RAP80 in multiple cell backgrounds. We found that RAP80−/− cells have robust BRCA1 localization to DSB sites and uncovered that this was due to near-complete redundancy with a DSB site-targeting activity that is located in the BRCA1 RING finger domain. Mutations that alter BRCA1 RING function without impairing its BARD1 interaction (such as the E2 binding-deficient I26A mutation) caused complete loss of BRCA1 localization to DNA damage sites and greatly impair HR in the absence of RAP80 or in the presence of mutations that disable the BRCA1/BRCA1-A interaction. We conclude that DNA damage localization of BRCA1 is essential for its function during HR and that it is dependent on two redundant activities mediated by the BRCA1-A complex and the BRCA1 RING domain. We finally speculate that the BRCA1 E3 ligase activity may play an important role in endowing recognition of RNF168-catalyzed H2A Lys13/15 ubiquitylation (H2A-K13/K15ub) by the BARD1 protein (Becker et al, 2021). Results To better understand the contribution of RNF168-dependent ubiquitylation to BRCA1 accumulation to DSB sites, we generated RAP80 (UIMC1) knockouts by gene editing in multiple RPE1-hTERT p53−/− Cas9 (RPE1) and U-2-OS (U2OS) cell backgrounds (Figs 1B and EV1A). As expected, analysis of gene conversion using the traffic light reporter assay in RPE1 cells showed that loss of RAP80 causes an increase, rather than a decrease, in HR (Fig EV1B). Furthermore, RAP80 inactivation also demonstrated a complete loss of ABRAXAS1 localization to ionizing radiation (IR)-induced foci marked by γH2AX in multiple RPE1- and U2OS-derived clones (Fig 1C and D). These results indicated that these cell lines recapitulate known phenotypes associated with RAP80 inactivation. Click here to expand this figure. Figure EV1. Design and characterization of RAP80−/− cell lines A. To generate RAP80−/− cell lines, we designed single guide RNAs (sgRNA) that target the first exon of RAP80. U2OS and U2OS Flp-In/T-REx clones were sequence validated (Table EV1) and carry RAP80 indels leading to premature stop codons. Dark grey boxes indicate exons 1–14, red arrows indicate the location of the sgRNA, blue line indicates the epitope recognized by the RAP80 antibody used for immunoblotting. B. Homologous recombination was monitored by gene conversion using the traffic light reporter assay in the indicated cell lines 48 h post-nucleofection in the absence or presence of the donor template. BFP+ mClover+ cells have undergone gene conversion. The bars represent the mean ± SD (n = 3 biological replicates). C, D. RPE-1 hTERT p53−/− Cas9 cells or their isogenic RAP80−/−counterparts were X-irradiated (10 Gy) and processed for BRCA1 and cyclin A immunofluorescence. Cyclin A (CycA)-negative cells are enriched for cells in G1 whereas cyclin A-positive cells are enriched for S/G2 cells. Shown in (C) is the quantitation and the bars represent the mean ± SD (n = 3 biological replicates). Representative micrographs are shown in (D). E. Representative micrographs of the experiment shown in Fig 1G at the 1 h timepoint. All scale bars are 5 μm. Source data are available online for this figure. Download figure Download PowerPoint These seven independent RAP80−/− cell lines displayed near-normal recruitment of BRCA1 to DSB sites as measured by IR-induced focus formation (Fig 1E and F) and showed typical cell cycle-restricted localization of BRCA1 in S/G2 cells (Fig EV1C and D). BRCA1 foci were completely lost in BRCA1−/− cells (Fig 1E), indicating that the immunostaining was specific for the BRCA1 protein. We next examined BRCA1 IR-induced focus formation and retention over time, fixing cells from 15 min to 6 h post-irradiation. We detected a defect in the maintenance of BRCA1 foci from 1 h onward in the RAP80−/− cell line (Figs 1G and EV1E), consistent with the phenotypes previously described using short interfering (si) RNA-mediated depletion of RAP80 (Hu et al, 2011). We therefore conclude that RAP80, and by inference the BRCA1-A complex, is dispensable for the initial recruitment of BRCA1 to DSB sites. RAP80 interacts specifically with UbK63 chains. Although both RNF8 and RNF168 participate in the formation of UbK63 chains, mounting evidence suggests that RNF8 is the main source of UbK63 at DNA damage sites by ubiquitylating histone H1 (Thorslund et al, 2015). We therefore tested whether there was a differential contribution for RNF8 or RNF168 toward BRCA1 recruitment in cell lines that were proficient or deficient in RAP80. We observed that RNF8 depletion by effective siRNAs (Fradet-Turcotte et al, 2013) led to a near-complete loss of BRCA1 recruitment to IR-induced foci, whereas depletion of RNF168 led, in comparison, to an incomplete decrease (Fig 2A and B). The residual BRCA1 recruitment to DSB sites observed in RNF168-depleted cells was dependent on RAP80 as depletion of RNF168 in RAP80−/− cells led to a complete loss of BRCA1 IR-induced foci (Fig 2A and B). To rule out the possibility that these results were an artifact of siRNA-mediated depletion, we examined BRCA1 recruitment in RNF8−/− and RNF168−/− cell lines generated in RPE1-hTERT p53−/− Cas9 cells (Fig EV2A). As with siRNA-depleted cells, we observed in RNF8−/− cells a complete loss of BRCA1 accumulation into IR-induced foci compared with a partial reduction of BRCA1 recruitment in the RNF168 knockout cells (Fig 2C and D). Depletion of RAP80 in RNF168−/− cells abolished the residual recruitment of BRCA1 to DNA damage sites (Fig 2C and D). Examination of localization of RAP80 to IR-induced foci in the RNF8−/− and RNF168−/− cells showed that BRCA1-A localization was completely dependent on RNF8 but only partially dependent on RNF168 (Figs EV2B and C). These results indicate that RAP80 may promote a mode of BRCA1 recruitment to DNA damage sites that is largely dependent on RNF8 and that acts in parallel to a second mode of recruitment that is dependent on RNF168-mediated ubiquitylation of histone H2A Lys13/Lys15 residues. Figure 2. RAP80-independent BRCA1 recruitment to DSB sites is dependent on RNF8 and RNF168 A, B. U2OS Flp-In/T-Rex parental (WT) and RAP80−/− cells were transfected with siRNA pools targeting BRCA1, RNF8, RNF168, or BARD1 or with a nontargeting control siRNA (CTRL). 48 h post-transfection cells were irradiated (2 Gy) and processed for immunofluorescence 1 h post-IR treatment using antibodies against BRCA1 and γH2AX. Quantitation of the percentage of cells with > 5 BRCA1 foci that colocalize with γH2AX is shown in (A). A minimum of 100 cells per replicate were analyzed, and the bars represent mean ± SD (n = 3 biological replicates). Representative micrographs are shown in (B). C, D. Parental (WT) RPE1-hTERT p53−/− Cas9, RNF8−/−, and RNF168−/− cells were treated with either a nontargeting siRNA pool (CTRL) or a pool targeting RAP80. 48 h post-transfection cells were irradiated (2 Gy) and processed for immunofluorescence 1 h post-IR treatment using antibodies against BRCA1 and γH2AX. Quantitation of the percentage of cells with > 10 BRCA1 foci that colocalize with γH2AX is shown in (C). A minimum of 100 cells per replicate were analyzed, and the bars represent mean ± SD (n = 3 biological replicates). Representative micrographs are shown in (D). All scale bars are 5 μm. Source data are available online for this figure. Source Data for Figure 2 [embr202153679-sup-0004-SDataFig2.xlsx] Download figure Download PowerPoint Click here to expand this figure. Figure EV2. Additional data supporting that loss of RNF8 or RNF168 impairs BRCA1 recruitment to DSB sites Immunoblotting of whole-cell extracts obtained from parental (WT), RNF8−/−, and RNF168−/− RPE1-hTERT p53−/− Cas9 cells with a mixture of RNF8 and RNF168 antibodies. GAPDH is used as a loading control. Immunoblotting of whole-cell extracts from the indicated cell lines treated with either a non-targeting siRNA control (−) or siRNA targeting RAP80 (+). GAPDH is used as a loading control. Representative of two independent immunoblots. The indicated parental (WT), RNF8−/−, and RNF168−/− cell lines were processed for immunofluorescence 1 h post-irradiation (2 Gy) and stained with antibodies against RAP80 and γH2AX. Representative micrographs are shown with quantitation of the mean percentage of cells with > 5 RAP80 foci (± SD) that colocalize with γH2AX indicated below the micrographs (n = 3 biological replicates). A minimum of 50 cells were analyzed. The scale bar is 10 μm. Download figure Download PowerPoint To further dissect how BRCA1 may be recruited to DNA damage via RAP80-dependent and -independent pathways, we examined how a truncated protein composed of the isolated tandem BRCT domains of BRCA1 (amino acid residues 1,582–1,863; BRCA1BRCT) is recruited to DNA damage sites. We observed that contrary to the observed results for full-length BRCA1 recruitment, localization of BRCA1BRCT into IR-induced foci was strictly dependent on RAP80 and the ABRAXAS1-interacting S1655 residue in the BRCT domains (Fig 3A and B). These results hinted that the putative second and RAP80-independent mode of recruitment of BRCA1 to DNA lesions is carried out by a BRCA1 region outside the tandem BRCT domains. In order to map this additional recruitment domain, we generated stable U2OS Flp-In/T-Rex cell lines that express various siRNA-resistant transgenes producing GFP-tagged BRCA1 and variants. Consistent with the previous results, we observed that deletion of the BRCT domains or introduction of the S1655A phosphopeptide-binding mutant in the context of full-length BRCA1 maintains the ability of BRCA1 to form IR-induced foci (Fig 3C and D). Furthermore, the variant BRCA1 1–1,362, containing a C-terminal deletion of both BRCT and the PALB2-interacting coiled-coil regions, also formed robust IR-induced foci in U2OS cells (Fig 3C and D). However, to our surprise, expression of a protein consisting solely of the RING finger domain (BRCA1RING, i.e., BRCA1 1–110) also localized to DNA damage sites independently of RAP80 (Fig 3C and D) with similar efficiency to the full-length protein when focus intensity was measured (Fig EV3A). These results suggest that the RING domain may be responsible for an activity that recruits BRCA1 to DNA damage sites redundantly with RAP80. Figure 3. RING domain participates in BRCA1 recruitment to DNA damage sites A, B. U2OS Flp-In/T-REx cells with integrated transgenes encoding GFP-BRCA1BRCT or -BRCA1BRCT S1655A were transfected with non-targeting siRNA (CTRL) or siRNA targeting RAP80 or BRCA1. Following doxycycline treatment to induce transgene expression (5 μg/ml, 24 h), cells were irradiated (10 Gy) and processed for immunofluorescence 1 h post-IR for antibodies against RAP80 and γH2AX. GFP fluorescence was used to detect BRCA1 fusions. Shown in (A) is the quantitation of a minimum of 100 cells per replicate, where the bars represent mean ± SD (n = 4). Representative micrographs are shown in (B). C, D. U2OS Flp-In/T-REx cells stably integrated with the indicated transgenes were treated with doxycycline (5 μg/ml, 36 h) to induce protein expression and transfected with an siRNA targeting BRCA1 and also either non-targeting siRNA (CTRL) or siRNAs targeting RAP80. 1 h post-irradiation (10 Gy), cells were processed for immunofluorescence using an antibody against γH2AX. GFP fluorescence was used to detect BRCA1 fusions. Shown in (C) is the quantitation of a minimum of 100 cells per replicate, where the bars represent mean ± SD (n = 3 biological replicates). Representative micrographs are shown in (D). All scale bars are 5 μm. Source data are available online for this figure. Source Data for Figure 3 [embr202153679-sup-0005-SDataFig3.xlsx] Download figure Download PowerPoint Click here to expand this figure. Figure EV3. Additional data supporting that the BRCA1 RING domain acts in parallel to RAP80 for BRCA1 recruitment Quantitation of GFP intensity at the γH2AX focus in U2OS Flp-In/T-REx cells expressing the indicated GFP fusion proteins. γH2AX IR-induced foci were found within a nucleus and the GFP intensity under each γH2AX focal unit was measured. Intensity of GFP under each individual γH2AX focus was plotted with the background GFP intensity subtracted. Bars represent mean ± SD. n = 3,903, 676 and 871 cells analyzed. The BRCA1 I26A and K70A/R71A mutants interact with BARD1. Whole-cell extracts from 293T cells expressing GFP-BARD1 and the indicated FLAG vectors were subjected to immunoprecipitation with a GFP antibody and were then immunoblotted with FLAG or GFP antibodies. EV, empty vector. Representative of two independent experiments. Immunoblotting of parental U2OS 2-6-3 (WT) or isogenic RAP80−/− cell lines transfected with siRNAs targeting BRCA1 and nucleofected with vectors expressing the indicated GFP fusion proteins, as described in Fig 4A and B. Tubulin and KAP1 are used as loading controls. Representative micrographs for the additional conditions shown in Fig 4A. The scale bars are 10 μm. Source data are available online for this figure. Download figure Download PowerPoint Mutations or loss of the RING domain in BRCA1 impairs its association with BARD1 and leads to BRCA1 destabilization (Hashizume et al, 2001; Joukov et al, 2001; Fabbro et al, 2002), complicating the analysis of the contribution of the RING domain to BRCA1 recruitment. We therefore explored whether we could identify mutations in the BRCA1 RING domain that impair DNA damage localization while maintaining stability. We selected two mutations: BRCA1-I26A, which disrupts the interaction between the RING and E2 conjugating enzymes such as UbcH5c (Brzovic et al, 2003; Christensen et al, 2007), and BRCA1 K70A/R71A that disrupts the interaction between BRCA1 and the nucleosome acidic patch (McGinty et al, 2014; Witus et al, 2021). These mutations do not impair interaction with BARD1, as reported previously with recombinant proteins (Brzovic et al, 2003; McGinty et al, 2014) or in co-immunoprecipitation studies (Fig EV3B). These two mutants, along with wild-type BRCA1, were expressed as fusions to GFP from siRNA-resistant transgenes in U2OS 2-6-3 cell lines (parental and RAP80−/−; Fig EV3C). The U2OS 2-6-3 cell line contains an inducible mCherry-LacR-FokI fusion protein that can induce clustered DSBs at an integrated LacO array, ++which allows for facile quantitation of recruitment to DSB sites (Shanbhag et al, 2010). Upon depletion of BRCA1 by siRNA, FokI expression was induced, and GFP fusion protein recruitment to mCherry-marked DSBs was assessed. We observed that the two BRCA1 RING mutants accumulated at DSB sites as efficiently as wild-type BRCA1 in RAP80-proficient cells but had greatly impaired recruitment to FokI-induced breaks in RAP80−/− cells, with BRCA1 I26A being the most defective (Figs 4A and B, and EV3D). These results suggested that BRCA1 recruitment to DSB sites is the result of a collaboration between the RING domain and the BRCT-dependent interaction with BRCA1-A. To test this idea, we also introduced the BRCA1-S1655A mutation alone or in combination with I26A. The S1655A mutant showed reduced but RAP80-independent recruitment to FokI-induced DSBs that was completely abolished by the I26A mutation (Fig 4A and B). Figure 4. Nucleosome- and E2-binding by the BRCA1 RING participates in promoting BRCA1 localization A, B. U2OS 2-6-3 parental (WT) or RAP80−/− cell lines were transfected with siRNAs targeting BRCA1, followed by nucleofection of vectors encoding the indicated GFP fusion proteins. 48 h post-nucleofection, mCherry-LacR-FokI expression was induced for 5 h prior to being processed for fluorescence microscopy for GFP and mCherry. Shown in (A) is the quantitation of GFP fluorescence at the mCherry focus, where the bars represent mean ± SD (n = 50, 50, 40, 40, 40, 40, 20, 20, 30, 30 cells analyzed from 3 biological replicates). Representative micrographs for the BRCA1 and BRCA1-I26A conditions are shown in (B). Additional micrographs for the other conditions are in Fig EV3D. C. RPE-1 hTERT p53−/− BRCA1−/− Cas9 cells (WT) or their isogenic RAP80−/− counterparts expressing the indicated GFP fusion proteins were processed 1 h post-irradiation (10 Gy) for immunofluorescence using antibodies against BRCA1 and γH2AX. GFP fluorescence was used to detect BRCA1 fusions. A minimum of 100 cells per replicate were analyzed, and the bars represent mean ± SD (n = 3 biological replicates). Representative micrographs are shown in Fig EV4B. D. RPE-1 hTERT p53−/− BRCA1−/− Cas9 cells or their isogenic RAP80−/− counterparts expressing the indicated GFP fusion proteins were processed 1 h post-irradiation (10 Gy) for immunofluorescence using antibodies against RAD51 and γH2AX. GFP fluorescence was used to detect BRCA1 fusions. A minimum of 100 cells per replicate were analyzed, and the bars represent mean ± SD (n = 3 biological replicates). Representative micrographs are shown in Fig EV4C. The scale bar is 5 μm in (B). Source data are available online for this figure. Source Data for Figure 4 [embr202153679-sup-0006-SDataFig4.xlsx] Download figure Download PowerPoint To further test the collaboration between RAP80 and the BRCA1 RING domain in an orthogonal system, we used gene editing to create a RAP80 knockout (RAP80−/−) in an RPE1-hTERT BRCA1−/− p53−/− Cas9 cell line (Fig EV4A). This cell line allowed us to assess BRCA1 IR-induced focus formation and its dependence on RAP80 in cell lines expressing BRCA1 variants. As observed with the FokI system, we found that the BRCA1 I26A protein forms IR-induced foci but does so in a strictly RAP80-dependent manner (Figs 4C
Referência(s)