Fig. 3: Validation of interaction between BCAS2 and NBS1 and identification of the NBS1-interacting domain in BCAS2.

a, b In vitro pull-down assay. Bacteria-produced GST-NBS1 (a) and His-BCAS2 (b) proteins were incubated with bead-bound His-BCAS2 and GST-NBS1, respectively. Eluted proteins were analysed by western blotting using anti-His to detect His-BCAS2 or anti-GST antibody to detect GST-NBS1 (upper panels). GST only and GST-MDM2¹⁵ fragments served as negative and positive controls, respectively. The proper expressions of GST-NBS1 (a) and His-BCAS2 (b) proteins from bacteria were confirmed by Coomassie blue staining (lower panels). M, protein marker. c Reciprocal co-immunoprecipitation assay: HEK 293T cells were transfected with expression vectors for V5-tagged NBS1 or control vector, in addition to either FLAG-tagged BCAS2 expressing plasmids or FLAG vector. Cell extracts were immunoprecipitated (IP) with anti-V5 or anti-FLAG antibodies and immunoblotted using the indicated antibodies. The levels of β-actin were used as loading controls in the input. d–f Evaluation of the co-localisation status of BCAS2 and NBS1 in U2OS cells. Confocal microscopy analysis (d, left panel) showed subcellular co-localisation of BCAS2 (red) and NBS1 (green) foci in a nucleus in Z-stack images (YZ and XZ). Scale bar = 10 μm. The intensity profiles (right panel) along selected lines (X1–X2, and Y1–Y2) demonstrated the overlapping fluorescence signals for BCAS2 and NBS1. e Quantitative image analysis of co-localisation was also performed using immunofluorescent images of BCAS2 and NBS1 in U2OS cells taken at 2, 8 and 12 h after ionising radiation (IR) (10 Gy). The mean overall Manders’ overlap coefficient of BCAS2 and NBS1 was calculated from 50 to 60 nuclei per coverslip using the software tool implemented in ImageJ via the JACoP plugin. The histogram shows that the mean overall Manders’ overlap coefficient of NBS1 and BCAS2 at 12 h after IR was significantly (n = 3; *P < 0.05; Kruskal–Wallis test with Dunn’s test) greater than that at 8 h. f The percentages of cells with RPA2-pS4/8-positive foci during a time course after irradiation or without irradiation. Immunofluorescent microscopy analysis was performed to calculate the percentage of U2OS cells that were positive for phospho-RPA2 (RPA2-pS4/8) foci (the representative images are in Supplementary Fig. S4b) during a time course after IR. The histogram shows that the percentage of cells positive for RPA2-pS4/8 foci at 12 h after IR was significantly higher than that at 2 h or that without irradiation; the percentage at 8 h was also significantly higher than that without irradiation (100 cells per coverslip; n = 6; *P < 0.05, **P < 0.01, ***P < 0.001; Kruskal–Wallis test with Dunn’s test). g–i Identification of the NBS1-binding domain of BCAS2 and its significance in homologous recombination (HR) and non-homologous end joining (NHEJ). g Domain mapping showed that the N-terminal domain was required for BCAS2 to interact with NBS1. The lysates of HEK 293T cell transfected with the plasmid expressing V5-NBS1 were incubated with each glutathione resin-bound GST-BCAS2 deletion protein produced by bacteria, and the pulled-down proteins were eluted and subjected to western blotting using anti-V5 antibody to assess the amount of bound NBS1. The proper expression of various deletion proteins of BCAS2, as depicted in a schematic illustration (left panel), from bacteria was confirmed by Coomassie blue staining (right panel). FL full-length, dcc1 deletion of coiled-coil 1 domain, dcc2 deletion of coiled-coil domain 2, dcc1 + 2 deletion of both dcc domains, dN deletion of the N-terminal half. The BCAS2 N-terminus, but not any of the two coiled-coil domains, was essential in enhancing HR activity. h Determination of the domain that is important for the effect of BCAS2 on HR. The overexpression of an N-terminal half-deleted BCAS2 (FLAG-BCAS2 dN) had no effect, while the overexpression of a BCAS2 construct with the deletion of both coiled-coil domains (FLAG BCAS2 dcc1 + 2) still exhibited a significant (*P < 0.05; Mann–Whitney U test) enhancing effect on HR activity, which is represented by the GFP-positive cell percentage measured by flow cytometry described in Fig. 2e legend, The data were normalised to that of the control (vector only). i Full-length BCAS2 was required to enhance the in vivo precise NHEJ activity. Measurement of in vivo precise NHEJ activity is described in detail in “Methods” and in the legend of Fig. 2b. Briefly, the luciferase reporter plasmid was linearised and co-transfected into HEK 293T cells with empty vector control (w/o), the full-length BCAS2 cDNA, or each of the indicated BCAS2 cDNA deletion constructs. The results showed that only the full-length BCAS2 had a significant enhancing effect on precise NHEJ activity. The data are presented as means + SD (n = 3; *P < 0.05; Mann–Whitney U test).