Switch-like phosphorylation of WRN integrates end-resection with RAD51 metabolism at collapsed replication forks

Abstract

Replication-dependent DNA double-strand breaks are harmful lesions preferentially repaired by homologous recombination (HR), a process that requires processing of DNA ends to allow RAD51-mediated strand invasion. End resection and subsequent repair are two intertwined processes, but the mechanism underlying their execution is still poorly appreciated. The WRN helicase is one of the crucial factors for end resection and is instrumental in selecting the proper repair pathway. Here, we reveal that ordered phosphorylation of WRN by the CDK1, ATM and ATR kinases defines a complex regulatory layer essential for correct long-range end resection, connecting it to repair by HR. We establish that long-range end resection requires an ATM-dependent phosphorylation of WRN at Ser1058 and that phosphorylation at Ser1141, together with dephosphorylation at the CDK1 site Ser1133, is needed for the proper metabolism of RAD51 foci and RAD51-dependent repair. Collectively, our findings suggest that regulation of WRN by multiple kinases functions as a molecular switch to allow timely execution of end resection and repair at replication-dependent DNA double-strand breaks.

Graphical Abstract

Figure 1.

ATM/ATR-dependent WRN phosphorylation requires CDK activity upon CPT. (A) WRN was immunoprecipitated from cells transiently transfected with Flag-WRN wild-type (WRN^WT) and treated with CDKi (Roscovitine), ATMi (KU-55933), ATRi (VE-821), alone or in combination, and then with CPT for 4 h. Nine-tenths of the IPs were analysed by western blotting (WB) with the anti-pS/TQ antibody, while 1/10 was analysed by anti-WRN antibody. Input represents 1/50 of the lysate. Anti-Flag antibody was used to verify transfection and an anti-β-tubulin antibody was used as a loading control. Quantification of the representative blots is reported below each lane. The graph represents quantification of the densitometry analysis from biological duplicates. (B) Cells were treated and subjected to anti-Flag-WRN IPs as in (A). Nine-tenths of IPs were subjected to WB with an anti-pS1141WRN antibody, while 1/10 was detected by anti-Flag antibody. Input represents 1/50 of the lysate. Anti-LaminB1 was used as a loading control. The graph represents quantification of the densitometry analysis from biological duplicates. (C) Schematic representation of the WRN protein with domains and localization of the three phosphorylation sites implicated in this study (S1058, S1133, S1141), the ATR-targeted sites (S409, S991, S1256) and the CK2-targeted sites (S434, 440, 461, 467) implicated in the response to replication arrest. Below is an AF3-modelling of WRN structure with the position of S1058, S1133 and S1141 in predicted unstructured structures. (D) Cells transiently transfected with WRN^WT or WRN^S1133A mutant were treated with 10 μM ATMi then with CPT for 4 h followed by IP/WB. Nine-tenths of IPs were analysed by WB with the anti-pS/TQ antibody, while 1/10 was analysed by anti-WRN antibody. Input represents 1/50 of the lysate. Anti-Flag was used to verify transfection and an anti-LaminB1 antibody was used as a loading control. The graph represents quantification of the densitometry analysis from biological duplicates. (E) Cells were treated and subjected to anti-Flag-WRN IPs as in (C). Nine-tenths of IPs were analysed by WB with the anti-pS1141WRN antibody. (F) In vitro ATM kinase assay. For kinase assay, 2 μg of immunopurified GST-tagged WRN wild-type fragment (C-WRN^WT) or WRN phosphomimetic mutant fragment (C-WRN^S1133D) were phosphorylated in vitro using Flag-tagged ATM kinase, immunoprecipitated with specific anti-Flag-conjugated beads. Immunoblotting was used to analyse the ATM-dependent phosphorylation level in different WRN fragments using an anti-pS/TQ antibody. Treatment with 10 μM ATM inhibitor (KU-55933) was used as a control. Coomassie (CBB) showed GST-C-WRN fragments in the gel as a control.

Figure 2.

Phosphorylation by ATM/ATR of WRN at distinct sites differently affects end resection of DSBs. (A) Cartoon depicting the ssDNA assay by native anti-IdU immunofluorescence. The scheme shows how ssDNA can be visualized at collapsed replication forks. CPT treatment results in one-ended DSBs at replication forks, leading to 5′-3′ resection of template DNA by nucleases (pacman) thus exposing nascent ssDNA, which is detected by native IdU/ssDNA assay. Nascent DNA was pre-labelled for 15 min with IdU before treatment, and labelling remained during treatment with CPT. (B and C) WS fibroblasts were transiently transfected with the indicated WRN-expressing plasmid. The ssDNA was analysed at different time points, as indicated. Dot plots show the mean intensity of IdU/ssDNA staining for single nuclei (n = 300, two biological replicates). Data are presented as mean ± SE. Representative images of IdU/ssDNA-stained CPT-treated cells are shown. A panel showing ssDNA detected in the cells complemented with the end resection-defective S1133A-WRN mutant is shown as a reference. Statistical analysis was performed by the ANOVA test (****P< 0.0001, **P< 0.01, *P> 0.05; ns = not significant).

Figure 3.

Altered phosphorylation of WRN at S1141 affects end resection. (A) The graph shows the analysis and quantification of Cas9-induced and processed DSBs in WS fibroblasts transiently transfected with WRN mutants. Data are represented as the percentage of DSBs resected on a specific chromosomal site. (B) WS-derived cell lines complemented with different WRN mutants were labelled and treated with CPT, followed by different time points of release in free medium before performing IdU/ssDNA assay. The dot plot shows the mean intensity of IdU/ssDNA staining for single nuclei (n = 300, two biological replicates). Data are presented as mean ± SE. Statistical analysis was performed by the ANOVA test (****P< 0.0001, **P< 0.01, *P> 0.01; ns = not significant).

Figure 4.

Phosphorylation of WRN by ATM/ATR occurs at the end of resection and requires its correct execution. (A) Cells were transiently transfected with an empty vector or a vector expressing Flag-tagged WRN wild-type (WRN^WT) and were treated with CPT for different time points. In addition, treated cells were recovered for 2 h in drug-free medium. Flag-WRN was immunoprecipitated and 9/10 of IPs were analysed by WB with both the anti-pS1133WRN and the anti-pS1141WRN antibodies, while 1/10 was detected by anti-WRN. One-fiftieth of the lysate was blotted with an anti-Flag antibody to verify transfection. An anti-LaminB1 antibody was used as a loading control. The graph shows the kinetics of WRN phosphorylation levels, and data are from biological duplicates. (B) Cells were transiently transfected as in (A). Cells were treated with CPT for 4 h and recovered in CPT-free medium. Flag-WRN was immunoprecipitated, and IPs were analysed by mass spectrometry. The blot shows the amount of WRN from the IP of one representative triplicate. The graph represents quantification of the normalized amounts of phosphorylated residues from biological triplicates. (C) Cells were transiently transfected as in (A). Cells were treated with CPT for 4 h. Cells were recovered in CPT-free medium and treated with ATMi (KU-55933), ATRi (VE-821), alone or in combination. Flag-WRN was immunoprecipitated and IPs were analysed and blotted as in (A). (D) Cells were transiently transfected as in (A). Cells were treated with CPT in combination or not with the inhibitor and recovered in drug-free medium. Flag-WRN was immunoprecipitated and IPs were analysed and blotted as in (C). (E) Cells were transiently transfected as in (A). Flag-WRN was immunoprecipitated and IPs were analysed and blotted as in (C). (F) WS-derived cell lines complemented with different WRN mutants were labelled, treated with CPT, and IdU/ssDNA assay was performed. The graph shows the mean intensity of IdU/ssDNA staining for single nuclei measured from three independent experiments (n= 300, each biological replicate). Data are presented as mean ± SE. Representative images of IdU/ssDNA-stained cells are shown. Statistical analysis was performed by the ANOVA test (****P< 0.0001, **P< 0.01, *P> 0.05).

Figure 5.

Regulated phosphorylation of WRN at S1141 is involved in RAD51 localization. (A) Cells stably expressing Flag-tagged WRN mutants were treated with CPT and allowed to recover for different time points as indicated. RAD51-foci staining was then analysed by IF. The graph shows the percentage of RAD51-foci positive cells measured from two independent experiments (n = 200, each biological replicate). Data are presented as mea ± SE. Representative images of RAD51 staining in response to treatment are shown. (B) Analysis of ssDNA/RAD51 interaction by in situ PLA. Cells were treated with CPT and allowed to recover for 4 h, then subjected to PLA using anti-IdU and anti-RAD51 antibodies. The panels show representative PLA images showing association of ssDNA with RAD51 (scale bar: 10 μm). The dot plot shows the PLA spots per PLA positive cells. At least 100 nuclei were analysed for each experimental point (n = 2). Values are presented as mean ± SE. Statistical analysis was performed by the ANOVA test (****P< 0.0001, **P< 0.01, each biological replicate). (C) The graph shows the number of PLA positive cells. At least 100 nuclei were analysed for each experimental point. Representative images from neutral Comet assay are shown.

Figure 6.

Regulated phosphorylation of WRN at S1141 is essential for DSB repair. (A) Analysis of efficiency of HR-mediated repair by reporter assay. HEK293TshWRN cells were co-transfected with the indicated WRN forms, the I-SceI expression vector pCBASce and the pDRGFP HR reporter plasmid, as described in ‘Materials and Methods’ section. The graph shows the percentage of GFP positive cells measured by flow cytometry. Data are presented as mean ± SE from three independent experiments (ns = not significant; *P< 0.05, **P< 0.01, ***P< 0.001; ANOVA test; n= 3 × 10⁵ events each biological repeat). (B) DSB repair efficiency analysis. WS-derived SV40-trasformed fibroblasts stably expressed Flag-tagged WRN mutants were treated with CPT for 1 h and then release in drug-free medium at different time points. DSB repair was evaluated by the neutral Comet assay. In the graph, data are presented as mean tail moment ± SE from three independent experiments. Representative images from the neutral Comet assay are shown. Statistical analysis was performed by the ANOVA test (****P< 0.0001; ns = not significant, each biological replicate).

Figure 7.

Persisting phosphorylation of WRN at S1141 leads to aberrant modification of S1131 by CDK. (A) In vitro kinase assay. Approximately 2 μg of immunopurified GST-tagged WRN wild-type fragment (C-WRN^WT) or WRN phosphomimetic mutant fragment (C-WRN^S1141D) were phosphorylated in vitro using CDK/Cyclin complex, and the experiment was performed treated with CDKi (Roscovitine). Immunoblotting was used to analyse WRN phosphorylation levels in different WRN fragments using an anti-pS1133WRN antibody. The graph shows the pS1133WRN level in each experimental point. (B) WRN was immunoprecipitated from cells transiently transfected with Flag-WRN wild-type (WRN^WT) or its phosphomimetic mutant and treated CPT for 4 h, followed by a 2-h release treated with CDKi (Roscovitine). Nine-tenths of IPs were analysed by western blotting (WB) with the anti-pS1133WRN antibody, while 1/10 was analysed by anti-Flag antibody. Input represents 1/50 of the lysate. Anti-Flag antibody was used to verify transfection. (C) WS-derived cell lines complemented with different WRN mutants were labelled, treated with CPT and the IdU/ssDNA assay was performed. The dot plot shows the mean intensity of IdU/ssDNA staining for single nuclei measured from three independent experiments (n= 300, each biological replicate). Data are presented as mean ± SE. Representative images of IdU/ssDNA-stained cells are shown. Statistical analysis was performed by ANOVA test (****P< 0.0001, ***P< 0.001; ns = not significant; n= 300, each biological replicate). (D) WS-derived SV40-trasformed fibroblasts were complemented with phosphomimetic WRN mutant, treated with CPT for 1 h and allowed to recover in the presence or not of the different inhibitors as indicated. The presence of DSBs was evaluated by the neutral Comet assay. In the graph, data are presented as percent of residual DSBs from three independent experiments normalized against the value of the 0 h recovery. Statistical analysis was performed by ANOVA test (***P< 0.001, **P< 0.01, n= 300, each biological replicate).

Figure 8.

Proposed model of regulation of WRN in the repair of DSBs at the replication fork. A schematic representation illustrating the proposed model of WRN regulation during the repair of DSBs at the replication fork is depicted. The model outlines the sequential phosphorylation events involving CDK1, ATM and ATR, along with their respective target sites on WRN (S1058, S1133 and S1141). It highlights the role of these phosphorylation events in modulating WRN function during different stages of DSB repair, including end resection, RAD51 nucleofilament formation and resolution of recombination intermediates. The model also suggests the importance of timely and ordered phosphorylation of WRN for proper DSB repair by HR. Further details are provided in the accompanying text.