Fig 2: WRAP53β facilitates MDC1–RNF8 interaction through its WD40 domain. (A) U2OS cells were treated with the siRNAs indicated for 48 h and with GFP-RNF8 for 24 h (all samples), irradiated with 6 Gy, and, 30-min later, subjected to immunoprecipitation of WRAP53β followed by immunoblotting of WRAP53β, MDC1, RNF8, γH2AX, and β-actin. (B) Immunoprecipitation of MDC1 in irradiated (6 Gy, 15-min recovery) U2OS cells treated with the siRNA indicated for 48 h or ATM inhibitor (ATMi) for 24 h. All samples were transfected with GFP-RNF8 for 16 h. (C) U2OS cells were treated with the siRNAs indicated for 48 h or ATM inhibitor for 16 h, irradiated with 6 Gy, allowed to recover for 15 min, and then subjected to Western blotting of MDC1, WRAP53β, γH2AX, and β-actin. (D) Schematic illustration of EGFP-tagged deletion constructs of WRAP53β. (E) U2OS cells were transiently transfected with the indicated EGFP-WRAP53β plasmids and Flag-RNF8 for 16 h, irradiated, and subjected to GFP immunoprecipitation followed by immunoblotting for MDC1, Flag-RNF8, and GFP-WRAP53β. (HC) Heavy chain of the antibody. (F) U2OS cells were transfected with siControl or siWRAP53#2 oligonucleotides for 8 h followed by transfection of EGFP-Empty or EGFP-WRAP53β WD40 (1–7) for 16 h, exposed to IR (6 Gy), and, after 1 h, immunostained for 53BP1 followed by quantification of the results. The graph in A shows the percentage of 100 GFP transfected cells in each experiment whose nuclei were 53BP1-positive. The error bars depict the SEM. n = 3; (*) P < 0.05, as determined by Student’s t-test. (G) U2OS cells were transiently transfected with the indicated EGFP-WRAP53β plasmids, HA-MDC1, and Flag-RNF8 for 16 h; irradiated; and subjected to immunoprecipitation of GFP followed by immunoblotting for HA-MDC1, Flag-RNF8, and GFP-WRAP53β. (H) Schematic illustration of how WRAP53β scaffolds the MDC1–RNF8 complex. Upon DNA damage, WRAP53β binds MDC1 and RNF8 simultaneously via its WD40 domain and facilitates their interaction.

Fig 3: WRAP53β binds MDC1 and RNF8 via their FHA domains. (A) U2OS cells were either left untreated or irradiated with 6 Gy of IR, and, 30 min, later immunoprecipitation of WRAP53β was performed, followed by immunoblotting of WRAP53β, MDC1, GFP-RNF8, and β-actin. (B) U2OS cells were transfected with the indicated HA-MDC1 constructs for 16 h and irradiated with 2 Gy, and, 30 min later, immunoprecipitation of WRAP53β was performed, followed by immunoblotting of WRAP53β and HA-MDC1. (C) Schematic illustration of RNF8 deletion constructs. (D) U2OS cells were transiently transfected with EGFP-RNF8 plasmids, HA-MDC1, and Flag-WRAP53β for 16 h; irradiated; and subjected to immunoprecipitation of GFP followed by immunoblotting for GFP-RNF8, Flag-WRAP53β, and HA-MDC1. (HC) Heavy chain of the antibody. U2OS (E) and H1299 (F) cells were transiently transfected with Flag-RNF8 plasmids, HA-MDC1, and EGFP-WRAP53β for 16 h; irradiated; and subjected to Flag immunoprecipitation followed by immunoblotting for the indicated proteins. (G) Schematic illustration of the domain architecture of MDC1 and RNF8, where black lines mark WRAP53β- and MDC1-binding sites. Numbers indicate amino acids.

Fig 4: Relationship between the level of checkpoint activation signal (MDC1 foci) and PLK1 activity.(a–c) U2OS cells expressing the PLK1 sensor and MDC1-mCherry were treated as in Fig. 2a and tracked by live cell imaging until just before NEBD at mitotic entry as described in Fig. 2b (n=45). Each coloured line represents the checkpoint activation signal (MDC1 foci) plotted against relative PLK1 activity in a single cell. Cells were categorized into three groups based on the initial value of the PLK1 activity. (a) Group 1: initial PLK1 activity >0.08; (b) Group 2: initial PLK1 activity between 0 and 0.08; (c) Group 3: initial PLK1 activity <0. (d) The duration of G2 arrest from DNA damage until mitotic entry in the three groups of cells showing the mean±s.e.m. Statistical analysis using analysis of variance (ANOVA)/Bonferroni test is displayed. The three groups are significantly different at P<0.0001. (e) The level of remaining checkpoint activation signal (MDC1 foci number) just before NEBD at mitotic entry in the three groups of cells, showing the mean±s.e.m. Statistical analysis using ANOVA/Bonferroni test is displayed. The three groups are significantly different at P<0.0001. (f) The rate of change in PLK1 activity (the first derivative of PLK1 activity over time) from single-cell measurements, represented by the colour map shown at the right, was plotted against the corresponding checkpoint activation signal (MDC1 foci number) on the vertical axis, and the relative change in PLK1 activity on the horizontal. Each box represents the average rate of changes in PLK1 activity in cells that fall within the corresponding range. NS, not significant.

Fig 5: Dynamics of damage repair and PLK1 activity from single-cell measurements.(a) Serial imaging of U2OS cells expressing the PLK1 sensor and MDC1-mCherry. U2OS cells synchronized in G2 were pulsed with etoposide as before and tracked by live cell imaging until just before NEBD at mitotic entry as described in Fig. 2b. The images were analysed using custom-written MATLAB software to measure the nuclear PLK1 FRET index and MDC1 foci number at different time points. Scale bar, 5 μm. (b) Changes in the checkpoint activation signal (MDC1 foci number) in single cells (left) or on average (right) from the time of damage treatment until just before NEBD at mitotic entry (n=20). The coloured lines represent the trajectories of individual cells, and the single black lines represent the population average, with the standard deviation for each data point (bars). (c) The relative PLK1 activity at each time point, measured using the FRET index as described in Methods, was tracked by live cell imaging under various experimental conditions. Plots marked ‘damage’ shows results on U2OS cells, treated as described in Fig. 2a, followed from the time of damage treatment until mitotic entry (n=20); ‘asynchronous’, shows results on asynchronous U2OS cells followed from initial PLK1 activation until mitotic entry (n=19); ‘synchronized’, shows results on U2OS cells synchronized in G2 after thymidine release, followed until mitotic entry (n=20); ‘damaged+caffeine’, shows results from a similar experiment on U2OS cells synchronized in G2, pulsed with 10 μM etoposide for 8 min, and then left in caffeine during live cell imaging (n=25). The coloured lines represent the trajectories of individual cells. (d) Traces from individual cells exposed to different treatments described in c were synchronized at mitotic entry and averaged within the same treatment groups. The mean and standard error of the average trajectories are shown. ‘Terminal slopes’ depicted in the top-left corner display the linear regression of the last 3 h of the traces, with the very last time point excluded. Analysis of covariance shows that the slopes in the different treatment groups are significantly different from one another. (****P<0.0001).