Fig 2: The E3 ubiquitin ligase activity of BRCA1 is required for inhibition of 53BP1 signaling. (a) 293T cells were untreated or irradiated at 10 Gy and processed 1h later. Cell lysates were prepared and immunoprecipitated (IP) using rabbit immunoglobulin G (IgG) or anti-53BP1 antibody. (b) BRCA1 ubiquitinates 53BP1 in vivo. 293T cells were transfected with His-ubiquitin (Ub), SFB-BRCA1/BARD1 and HA-53BP1. The pull-down and immunoblotting analysis was conducted as indicated. (c) HeLa cells infected with lentivirus carrying control or BRCA1 short hairpin RNAs were irradiated or untreated. Total cell lysates were prepared. Immunoprecipitation using high-salt buffer was performed with normal rabbit IgG or anti-53BP1 antibody. Endogenous 53BP1 ubiquitination was detected with anti-Ub antibody (see Materials and Methods for details). (d) BRCA1-depleted HeLa cells were transfected with various BRCA1 constructs, synchronized in S/G2 phases and processed as described in Figure 1c. (e) Quantification of PTIP (Pax transactivation domain-interacting protein) foci formation by the indicated BRCA1 constructs, as described in c. More than 30 FLAG-positive cells and the cells with >20 PTIP foci were counted, results are presented as the means (±s.d.) of three independent experiments. Ni-NTA, nickel-nitrilotriacetic acid; WT, wild type.

Fig 3: Cell cycle-dependent 53BP1 phosphorylation by ATM (ataxia telangiectasia mutated). (a) Asynchronous HeLa cells were treated as described in Figure 1a, arrows indicate p-53BP1 foci-negative cells. Cells with >10 foci following ionizing radiation (IR) were counted as positive cells. (b) HeLa cells were processed as in a, and the asterisks indicate BRCA1 or p-53BP1 foci-negative cells. The percentage of asynchronous cells forming indicated foci were as follows: S25/S29 (58%), S1778 (60%), 53BP1 (98%), BRCA1 (65%), RPA2 (52%) and p-ATM (94%), at least 200 cells were counted. (c) Summary of cell cycle-dependent foci formation of BRCA1 and 53BP1 complexes. (d) HeLa cells were arrested in M phase with nocodazole (100 ng ml-1) treatment for 16 h and then released for different time periods. The cells were irradiated with 10 Gy of IR 1 h or left untreated before harvesting.

Fig 4: Rescue of defective XAB2 by overexpression of RAD51 and RAD52. (A) Immunoblotting analysis of RAD52 (upper panel) and RAD51 (lower panel) overexpression in control and XAB2-depleted U87 cells. (B and C) Representative micrographs of ?H2AX foci (red) detected by immunofluorescence in the indicated cells exposed to 15 µM TMZ (or DMSO) for 2 h and allowed to recover in drug-free medium for the indicated times (B) and related quantification (C) (scale bar: 5 µm). (D) Percentage of Cyclin A-positive cells displaying pS1778-53BP1 foci among control and XAB2-depleted cells harboring the indicated constructs, following exposure to 15 µM TMZ (or DMSO) for 2 h and recovery in drug-free medium for the indicated times. Data are the average of n = 2 or more biological replicates (30–50 cells/sample/experiment). The images are representative of three independent biological repeats. Bars represent mean ± s.e.m. Significant differences between specified comparisons were assessed by two-ways ANOVA and are highlighted by stars (*P< 0.05; **P< 0.01; ***P< 0.001; ****P< 0.0001) (ns = not significant).

Fig 5: XAB2 depletion leads to the increased engagement of NHEJ in S/G2. (A) Quantification of the CRISPR–Cas9 mClover-LMNA1-HDR assay in control and XAB2-depleted U2OS cells. Data is representative of three independent biological repeats. Bars represent mean percentages of mClover-positive cells over the iRFP-positive population ± s.e.m. Significant differences between specified comparisons were assessed by unpaired t-test and are highlighted by stars (*P< 0.05; **P< 0.01; ***P< 0.001; ****P< 0.0001). Inset, efficiency of XAB2 depletion achieved by siRNA (siXAB2) as compared to a non-silencing control siRNA (siCTRL). a-Tubulin was used as a loading control. (B and C) Representative immunofluorescence images of RPA32 foci (red) and 53BP1 foci (green) in control and XAB2-depleted U87 cells exposed to 15 µM TMZ (or DMSO) for 2 h and allowed to recover in drug-free medium for 48 h (B), and related quantification of the average number of RPA32 foci per nucleus (C). (D) Quantification of the average number of RAD51 foci in control and XAB2-depleted cells exposed to 15 µM TMZ (or DMSO) for 2 h and allowed to recover in drug-free medium for the indicated times. See Supplementary Figure S2 for representative immunofluorescence images. (E and F) Quantification of the average number of pS1778-53BP1 foci per nucleus, in the total cell population, as well as in Cyclin A-negative and -positive cell subpopulations, in control and XAB2-depleted U87 cells following exposure to 15 µM TMZ (or DMSO) for 2 h, 72 h recovery in drug-free medium and processing for IF analysis of pS1778-53BP1 foci and Cyclin A (E), and representative immunofluorescence images (F). Scale bar: 5µm. The images are representative of three or more independent biological repeats. Bars represent mean ± s.e.m. Significant differences between specified comparisons were assessed by two-ways ANOVA and are highlighted by stars (*P< 0.05; **P< 0.01; ***P< 0.001; ****P< 0.0001).