Fig 2: Functional and biochemical validation of UBASH3B and WHIP interaction with TNF-RSC. (A) Summary of validation experiments for UBASH3B and WHIP association with TNF-RSC. (B) Scatterplot of HOIP interactors identified by affinity purification of C- or N-terminally tagged HOIP. Saint score indicates the recruitment enrichment of interactors against three GFP controls. All identified known members and associated components of TNF-RSC are shown in the plot. LUBAC components and WHIP are highlighted in purple and black, respectively. (C) DDA-MS analysis of UBASH3B interactome. The volcano plot shows copurified proteins identified by affinity purification of C-terminally tagged UBASH3B ectopically expressed in HEK293 cells against a GFP control. (D) Targeted IP-MS analysis of endogenous UBASH3B in A549 cells. The scatterplot displays the enrichment of the untreated and TNFα-treated proteins isolated against a nonspecific serum control. (E) Network model showing overlap of proteins enriched with a log2FC threshold >2 in both UBASH3B IP-MS (on the left) and TNF-RSC affinity purification (on the right). Thickness of UBASH3B edges scales with enrichment of associated nodes. (F) Targeted IP-MS analysis of endogenous UBASH3B in A549 cells. Occupancy plot (inside) showing the fraction of the indicated interactors bound to an arbitrary number of UBASH3B molecules. The numbers are reported in the circular doughnut chart. (G) Affinity purification of TNF-RSC from HT1080 cells demonstrating that UBASH3B and WHIP are selectively copurified in a ligand and time-dependent manner. (H) DEVDase assay with the indicated siRNA (B10 from SI Appendix, Dataset S13) against UBASH3B. siGFP was used as a control. Profile of each replicate is shown. (I) UBASH3B and WHIP knock-down sensitizes HT1080 cells to TNF-induced cell death to the same extent as knock-down of HOIP. The indicated targets were knocked down by RNAi in HT1080 cells and cells treated with either DMSO control or TNF/SM. (J) Model of the recruitment and signaling role for UBASH3B and WHIP.

Fig 3: WRNIP1, ATMIN, and RAD18 are required for ATM signalling in response to aphidicolin(a) Western blots showing knockdown of ATMIN, WRNIP1 and RAD18 by siRNA. (b) Disruption of 53BP1 and pATM recruitment following siRNA depletion of WRNIP1, ATMIN and RAD18 in HEK293 cells fixed and stained with the indicated antibodies after exposure to 0.2μM aphidicolin for 24 hours. siCtrl, siRNA control. Scale bar indicates 10μm. (c) Quantification of the percentage of cells in (b) with 53BP1 or pATM foci. *, p<0.05 (t-test). At least 100 cells were counted per condition. Error bars represent s.d. (d) Western blots showing compromised aphidicolin-induced ATM signalling in the absence of ATMIN, WRNIP1 or RAD18, using extracts from HEK293T cells transfected with siRNAs as above and left untreated or treated with 2μM aphidicolin for two hours. (e) Western blots assessing ATM auto-phosphorylation (pATM) and total ATM in cells treated as in (d). Experiments were replicated three times.

Fig 4: ATMIN interacts with WRNIP1 and is required for ATM signalling in replication stress conditions(a) Co-IP of endogenous ATMIN and WRNIP1. (b) Colocalisation of WRNIP1 and pATM-S1981 (pATM) foci in HEK293 cells in response to 0.2 μM aphidicolin for 24 hours. Experiment was replicated 3 times. (c) Quantification of cells in (b) with pATM/WRNIP1/53BP1 colocalisation. (d) pATM and 53BP1 immunofluorescent staining of control (+/+) and ATMIN-deficient (Δ/Δ) MEFs, either untreated (−Aph) or treated with 2μM aphidicolin for 2 hrs (+Aph). (e) Percentage of cells displaying aphidicolin-induced 53BP1 foci (*P<0.05, t-test). Counts represent 200 cells from three independent experiments. (f) Colocalisation of 53BP1 and WRNIP1 with a LacO array, marked by CherryLacR. Cells were treated with 0.2 μM aphidicolin for 18 hours. (g) Percentage of foci colocalising with CherryLacR from the experiment in (f). Scale bars represent 10μm. Error bars represent s.d.

Fig 5: WRNIP1 binds ubiquitinated PCNA and PCNA ubiquitination is required for aphidicolin-induced foci formation(a) Flag-WRNIP1 preferentially interacts with ubiquitinated over unmodified PCNA, shown by Western blotting of Ni2+-NTA pulldowns of His-PCNA from mixtures of equal amounts of purified proteins incubated together for 1hr. (b, c) pATM and 53BP1 foci re-establishment in HEK293 cells after treatment with an siRNA pool targeting the 3’UTR of WRNIP1 (siWRNIP1) and subsequent transfection with either empty vector or an expression construct for Flag-tagged WT WRNIP1 or the D37A UBZ mutant. Cells were treated with 0.2 μM aphidicolin for 24 hours. Experiment was repeated twice. *p=0.05 (t-test). n.s., not significant. Yellow arrowheads in (b) identify example nuclei across rows. In the rescue experiments (lower two rows), WRNIP1 staining indicates transfected cells. (d) Percentage of EdU-positive and EdU-negative HEK293 cells with prominent WRNIP1 foci after 0.2μM aphidicolin for 24 hours. EdU pulse was for the last 30 minutes in aphidicolin before fixing. (e) Western blots showing PCNA levels in MRC-5 cells stably expressing His-tagged siRNA-resistant WT PCNA or PCNA K164R, before and after siRNA depletion of endogenous PCNA. (f) MRC-5 cells stably expressing WT PCNA or PCNA K164R, siRNA-depleted of endogenous PCNA, and treated with 0.2μM aphidicolin for 24 hours, stained with WRNIP1. Yellow arrowheads indicate cells scored as positive for WRNIP1 foci. White arrow indicates cell magnified at right. High magnification images indicate examples of nuclei scored as positive (upper) and negative (lower). (g) Quantification of cells displaying WRNIP1 foci in the experiment in (f). (h) MRC-5 cells stably expressing WT PCNA or PCNA K164R, siRNA-depleted of endogenous PCNA, and treated with 0.2μM aphidicolin for 24 hours, stained with pATM. Yellow arrowheads indicate cells scored as positive for pATM foci. White arrow indicates cell magnified at right. High magnification images indicate examples of nuclei scored as positive (upper) and negative (lower). (i) Quantification of cells displaying pATM foci in the experiment in (h). 100-200 cells were counted per condition. (j) Impaired ATM substrate phosphorylation in response to 2μM aphidicolin treatment for 2 hours in PCNA K164R-expressing cells. Scale bars represent 10μm. Error bars represent s.d.