Fig 2: RNF168 regulates BRCA1 RING-mediated foci.a Above; cartoon showing BRCA1 protein domains and truncations with aa numbers indicated. Below; BRCA1 null MDA-MB-436 cells expressing BFP control, HA-BRCA1-full-length (FL), HA-BRCA1-ΔRING (ΔR), and HA-BRCA1-ΔBRCT (ΔB) were assessed for HA protein levels by western blotting. b Cells from (a) were assessed for HA foci 8 h post IR by immunofluorescence. Above; representative images (scale bar, 10 μm). Left; mean and S.E.M. percentage foci positive cells; middle, foci positive cells were assessed for the number of foci present in a single nucleus, red line: median value; right, foci positive cells were assessed for the mean size of foci present per nuclei, n = 3 biological replicates. Foci positive cells were determined to be those with ≥10 foci/nuclei. ***p < 0.001, **p < 0.01, *p < 0.05, ns not significant compared to FL (percentage positive cells: unpaired, two-tailed t-tests size; number and size per nucleus: nonparametric Mann–Whitney tests). See Source data for additional information. c Cells from (a) were subject to scrambled (Sc) and RNF168-targeting siRNA followed by IR and HA foci formation measured by immunofluorescence. Left; western blot showing RNF168 depletion. Above; representative cells (scale bar, 10 μm). Below; foci analyses as described in (b), n = 3 biological replicates. See Source data for additional information.

Fig 3: RNF168 interacts with TOP2α.(a) Identification of RNF168-associated proteins. A representative SDS–polyacrylamide gel electrophoresis of Flag-RNF168-associated proteins. Flag-tagged RNF168 was transfected in HEK293T cells and pull-down analysis was performed 48 h later. Protein bands were detected by silver staining. Protein bands were identified by mass spectrometry analysis following in-gel protease digestion. (b) HEK293T cells were transfected as indicated with HA-tagged RNF168 and Flag-TOP2α expression vectors. Cells were lysed and IP was performed using anti-Flag antibody. The resulting precipitates were subjected to IB analysis with the indicated antibodies. WCL, whole-cell lysate. (c) TOP2α, RNF168 and IgG (control) immunoprecipitates from HEK293T cells were examined by IB as indicated. (b,c) Data are representative of three independent experiments. (d) Cells treated with EdU were used for detection of localization patterns of TOP2α (Alexa Fluor 488) and RNF168 (Alexa Fluor 594) using confocal microscopy. Scale bar, 20 μm.

Fig 4: Rnf168 and Brca1 mutations block development and Brca1-P localization.a Cartoon showing RNF168-mediated and H2A-associated ubiquitin modifications, as well as the influence of 53BP1, BRCA1-A (ABRAXAS-RAP80) and -P (PALB2-BRCA2-RAD51) complexes on HR. b Brca1+/CC, Rnf168+/− mice were intercrossed and the numbers of expected and observed live births are shown. P values are obtained from two-sided chi-square goodness of fit tests for the binomial of each genotype. c Left; representative photograph. Right; weights of individual 4–6-month-old mice with the indicated genotypes, red line; median, Brca1+/+, Rnf168+/+,+/− (n = 10); Brca1+/CC, Rnf168+/+,+/− (n = 4); Brca1+/+, Rnf168−/− (n = 10); Brca1+/CC, Rnf168−/− (n = 6) mice. d Kaplan–Meier survival analysis of mice born with the indicated genotypes. e Brca1+/+, Brca1+/CC, and Brca1CC/CC MEFs treated with GFP (−) or Rnf168 (+) targeting sgRNA were subsequently incubated with increasing concentrations of the PARPi rucaparib and colony formation assessed. Mean and S.E.M. rucaparib IC50 values are shown for n = 4 biological replicates. f Cells from (e) were subject to 10 Gy γ-irradiation (IR) and foci formation measured by immunofluorescence. Endogenous Brca1 and Rad51 foci, and ectopic HA-PALB2 foci formation was assessed using cell lines expressing HA-PALB2. Foci positive cells were determined to be ≥10 Brca1 and Palb2 foci/nuclei, or ≥5 Rad51 foci/nuclei. The mean and S.E.M. percentage foci positive cells are shown for n = 3 biological replicates. See Supplementary Fig. 2a for representative images of cells. ***p < 0.001, **p < 0.01, *p < 0.05, ns not significant (unpaired, two-tailed t-tests).

Fig 5: RNF168 mediates ubiquitylation of TOP2α.(a) RIDDLE cells reconstituted with HA-RNF168 or HA-empty vector, and control HA-RNF168-reconstituted RIDDLE cells with TOP2α knock down were lysed and WCL subjected to IP with anti-TOP2α or IgG (control) antibodies. Immunoprecipitates were blotted with the indicated antibodies to detect ubiquitylated TOP2α. (b) Human breast cancer cell lines T47D and MDA-MB-231 knocked down for RNF168 (Sh.RNF168) and their control expressing ShRNA control (sh.Ctr) were examined for their level of ubiquitylated TOP2α as in a. (c) Rnf168−/−, Brca1−/− and WT MEFs were lysed and subjected to IP with anti-Top2α or IgG (control) antibodies. IPs from WCL were blotted with the indicated antibodies. (d) HEK293T cells were transfected with RNF168 (WT or mutant Rnf168-C21S), Flag-TOP2α and HA-Ub vectors as indicated. WCL were subjected to IP with anti-Flag, and IB analysis was performed using anti-HA antibody to detect ubiquitylated Flag-TOP2α. (e) Nuclear extracts from RIDDLE cells reconstituted with HA-RNF168 or HA-empty vector were subjected to IP with anti-TOP2α or IgG (control) antibodies. Immunoprecipitates were blotted with the indicated antibodies against K63- and K48-Ub linkages. (f) In vitro ubiquitylation of recombinant TOP2α in the presence of recombinant RNF168 (500 ng for lane 5, 1 μg for lanes 2, 3 and 6 and 2 μg for lane7), UBE1 (E1), UBE2E2 (E2) and Ub. Nuc, nuclear extract; WCL, whole-cell lysate.