WIP1 Promotes Homologous Recombination and Modulates Sensitivity to PARP Inhibitors

Abstract

Genotoxic stress triggers a combined action of DNA repair and cell cycle checkpoint pathways. Protein phosphatase 2C delta (referred to as WIP1) is involved in timely inactivation of DNA damage response by suppressing function of p53 and other targets at chromatin. Here we show that WIP1 promotes DNA repair through homologous recombination. Loss or inhibition of WIP1 delayed disappearance of the ionizing radiation-induced 53BP1 foci in S/G2 cells and promoted cell death. We identify breast cancer associated protein 1 (BRCA1) as interactor and substrate of WIP1 and demonstrate that WIP1 activity is needed for correct dynamics of BRCA1 recruitment to chromatin flanking the DNA lesion. In addition, WIP1 dephosphorylates 53BP1 at Threonine 543 that was previously implicated in mediating interaction with RIF1. Finally, we report that inhibition of WIP1 allowed accumulation of DNA damage in S/G2 cells and increased sensitivity of cancer cells to a poly-(ADP-ribose) polymerase inhibitor olaparib. We propose that inhibition of WIP1 may increase sensitivity of BRCA1-proficient cancer cells to olaparib.

Keywords: DNA repair; PARP inhibitor; chemotherapy; genotoxic stress; olaparib; phosphatase.

Figure 1

Inhibition of WIP1 impairs homologous recombination (HR). (A) Traffic light reporter assay in U2OS cells. Two independent stable cell lines (clones #10 and #12) were transfected with ISceI together with BFP-donor vector with or without pretreatment with 1 μM WIP1i. Efficiency of repair was analyzed 3 days after transfection by FACS. Plotted is mean of normalized ratio of GFP⁺/RFP⁺ cells. Bars indicate SD, n ≥ 3. Statistical significance evaluated by two-tailed t-test. (B) Traffic light reporter assay in two independent clones of RPE cells (#3 and #4). Same as A. (C) Efficiency of repair by HR (GFP⁺) and NHEJ (RFP⁺) in TLR assay in U2OS cells from A. (D) Efficiency of repair by HR (GFP⁺) and NHEJ (RFP⁺) in TLR assay in RPE cells from B. (E) Cell survival after irradiation of parental U2OS and two independent U2OS-WIP1-KO cell lines treated or not with WIP1 inhibitor was evaluated after 7 days using resazurin viability assay. Plotted is mean and SD, n ≥ 3. Statistical significance evaluated by two-way ANOVA (* P < 0.05; *** P < 0.001). (F) Cell survival of parental U2OS and two independent U2OS-WIP1-KO cell lines treated with indicated doses of camptothecin with or without combined treatment with WIP1 inhibitor was evaluated after 7 days using resazurin viability assay. Plotted is mean and SD, n ≥ 3. Statistical significance evaluated by two-way ANOVA (* P < 0.05; *** P < 0.001). (G) Cell survival after irradiation of parental RPE and RPE-WIP1-KO cell lines assayed as in E. (H) Cell survival of parental RPE and RPE-WIP1-KO cell lines with treated with camptothecin and analyzed as in F. (I) Percentage of dead cells was evaluated by Hoechst 33258 staining and FACS analysis 7 days after treatment with camptothecin or after irradiation in U2OS cell line with or without combined treatment with WIP1i. Plotted is mean +/− SD. Statistical significance evaluated by two-tailed t-test.

Figure 2

WIP1 plays role in DNA double-strand break repair in S-phase cells. (A) Quantification of 53BP1 foci in replicating (EdU+) cells after irradiation. U2OS parental cell lines with or without combined treatment with WIP1i and two independent WIP1 knockout cell lines were pulse-labeled with EdU for 30 min before irradiation. Cells were fixed after pre-extraction at indicated time-points and stained with 53BP1 antibody. Click chemistry was used to visualize EdU. Mean of median foci number +/- SD is plotted (n ≥ 3). Statistical significance evaluated by two tailed t-test. (B) Quantification of 53BP1 foci in non-replicating (EdU-) cells after irradiation. As in A. (C) Quantification of 53BP1 foci in replicating (EdU+) cells after irradiation. U2OS parental, WIP1 knockout and cell lines complemented with wild-type or phosphatase-dead (D314A) mutant of WIP1 were irradiated and analyzed as in A. (D) Quantification of 53BP1 foci in non-replicating (EdU-) cells after irradiation. U2OS parental, WIP1 knockout and cell lines complemented with wild-type or phosphatase-dead (D314A) mutant of WIP1 were irradiated and analyzed as in A. (E) Traffic light reporter assay in U2OS cells after transfection with indicated siRNA. Cells were transfected with ISceI together with BFP-donor vector with or without pretreatment with 1 μM WIP1i 2 days after siRNA transfection. Efficiency of repair was analyzed by FACS 3 days after ISceI and BFP-donor transfection. Plotted is mean +/− SD. Statistical significance evaluated by two-tailed t-test. (F) Efficiency of repair by HR and NHEJ in Traffic light reporter assay as in E. (G) Representative plots from Traffic light reporter assay in E.

Figure 3

WIP1 interacts with BRCA1 and dephosphorylates S1524. (A) Co-immunoprecipitation of WIP1 and BRCA1. HEK293 cells were transfected with either empty GFP or GFP-WIP1, subjected to immunoprecipitation using GFP-Trap and analyzed by Western blotting with BRCA1 antibody. (B) Co-immunoprecipitation of WIP1 and BARD1. HEK293 cells were co-transfected with either empty GFP or GFP-BARD1 and Flag-WIP1, subjected to immunoprecipitation using GFP-Trap and analyzed by Western blotting with indicated antibodies. Ponceau staining with indicated positions of GFP (empty arrowhead) and GFP-BARD1 (full arrowhead) are shown. (C) Co-immunoprecipitation of endogenous WIP1 and BRCA1. U2OS cell lysates were incubated with 2 μg of a control antibody (IgG) or affinity-purified antibody against WIP1 for 2 h. Protein complexes were isolated by protein A/G resin and analyzed by immunoblotting. (D) Quantification of BRCA1 pS1524 signal intensity in replicating (EdU+) cells after irradiation. U2OS parental and WIP1 knockout cell lines were pulse-labeled with EdU for 30 min before irradiation. At indicated time-points, cells were pre-extracted, fixed and stained with pBRCA1 S1524 and BRCA1 antibodies. Click chemistry was used to visualize EdU. Median total intensity of BRCA1 pS1524 was normalized to total BRCA1 and is plotted +/− SD. Statistical significance evaluated by two-way ANOVA. (E) Western blot analysis of U2OS parental and U2OS-WIP1-KO cell lines after irradiation. Cells were irradiated and whole cell lysates were analyzed using Western blotting with indicated antibodies. Arrowheads indicate two isoforms of WIP1 present in U2OS. (F) Western blot analysis of MCF7 cells after irradiation with or without combined treatment with WIP1i. Cells were pretreated with WIP1 inhibitor for 30 min before irradiation and whole cell lysates were analyzed by Western blotting with indicated antibodies. (G) Quantification of BRCA1 foci in replicating (EdU+) cells after irradiation. Parental U2OS and U2OS-WIP1-KO cells and cell lines complemented with wild-type or phosphatase-dead (D314A) mutant of WIP1 were pulse-labeled with EdU for 30 min before irradiation. Cells were fixed after pre-extraction at indicated time-points and stained with BRCA1 antibody. Click chemistry was used to visualize EdU. Mean of median total intensity +/− SD is plotted. Statistical significance evaluated by two-tailed t-test.

Figure 4

WIP1 delays recruitment of BRCA1 and dephosphorylation of 53BP1 at T543. (A) Co-immunoprecipitation of WIP1 and 53BP1. HEK293 cells were transfected with either empty GFP or GFP-WIP1, subjected to immunoprecipitation using GFP-Trap 24 h after transfection and by Western blotting with 53BP1 antibody. Ponceau staining with indicated positions of GFP (empty arrowhead) and GFP-WIP1 (full arrowhead) are shown. (B) HEK293 cells transfected with EGFP or EGFP-WIP1 were exposed to 3 Gy of IR, collected at indicated times and proteins were immunoprecipitated by GFP Trap. (C) Quantification of 53BP1 pT543 signal intensity in replicating (EdU+) cells after irradiation. U2OS parental and WIP1 knockout cell lines were pulse-labeled with EdU for 30 min before irradiation. Cells were fixed after pre-extraction at indicated time-points after IR and stained with p53BP1 T543 antibody. Click chemistry was used to visualize EdU. Mean of median total intensity +/− SD is plotted. (D) Western blot analysis of whole cell lysates of U2OS cells transfected with GAPDH or PP4C siRNA in response to irradiation and/or WIP1 inhibitor. (E) Quantification of 53BP1 pT543 signal intensity in replicating (EdU+) cells after irradiation. U2OS parental and WIP1 knockout cell lines were transfected with control or PP4C siRNA 2 days before irradiation. Cells were processed and analyzed as in C. (F) Quantification of RPA2 foci in replicating (EdU+) cells after irradiation. U2OS parental cell lines with or without combined treatment with WIP1i were pulse-labeled with EdU for 30 minutes before irradiation. Cells were fixed after pre-extraction at indicated time-points and stained with RPA2 and RAD51 antibodies. Click chemistry was used to visualize EdU. Mean of median foci number +/− SD is plotted. (G) Quantification of RAD51 foci in replicating (EdU+) cells after irradiation as in F.

Figure 5

WIP1 deficient cells are more sensitive to PARP inhibition. (A) Cell survival of parental U2OS, two independent U2OS-WIP1-KO cell lines with or without combined treatment with WIP1i was evaluated 7 days after treatment with indicated doses of olaparib using resazurin viability assay. Plotted is mean +/− SD, n ≥ 3. Statistical significance evaluated by two-way ANOVA. (B) Cell survival of parental U2OS, U2OS-WIP1-KO cells and cell lines complemented with wild-type or phosphatase-dead (D314A) mutant of WIP1 in response to 5 μM olaparib as in A. Statistical significance evaluated by two-tailed t-test (n ≥ 3). (C) Percentage of dead cells was evaluated by Hoechst 33258 staining and FACS analysis 7 days after treatment with 5 μM olaparib in U2OS cell line with or without combined treatment with WIP1i. Plotted is mean +/− SD. (D) Cell survival of RPE and MCF7 cell lines with or without combined treatment with WIP1i was evaluated 7 days after treatment with indicated doses of olaparib using resazurin viability assay. Plotted is mean +/− SD. N ≥ 3. Statistical significance evaluated by two-way ANOVA. (E) Cells were transfected with control siRNA (siNC) or siRNA to PP4C (siPP4C). Cell survival was evaluated after 7 days of treatment with olaparib and DMSO or WIP inhibitor. Statistical significance evaluated by two-tailed t-test (n = 3). (F) Quantification of 53BP1 foci number 3 days after treatment with olaparib. U2OS cells were treated with indicated doses of olaparib together with or without WIP1i for 3 days, fixed, stained with 53BP1 antibody and percentage of cells having 0–3, 3–10 and >10 foci were quantified. Mean +/− SD is plotted, n ≥ 3. (G) Quantification of 53BP1 foci after treatment with olaparib. U2OS-WIP1-KO cells and cell lines complemented with wild-type or phosphatase-dead (D314A) mutant of WIP1 were treated with WIP1i and olaparib for 3 days, fixed after pre-extraction and stained with 53BP1 antibody. Number of 53BP1 foci in S/G2 cells was evaluated using DAPI content of >2 n to gate S-G2 cells. Mean of median foci number +/− SD is plotted, n ≥ 3. Statistical significance evaluated by two-tailed t-test. (H) Response of U2OS and U2OS-WIP1-KO cell lines to treatment with 5 μM olaparib for 24–72h was analyzed by Western blotting using indicated antibodies. I) Quantification of 53BP1 foci 3 days after treatment with olaparib. MCF7 cells were transfected with indicated siRNAs and treated after 2 days with WIP1i and olaparib alone or combined for further 3 days. Cells were fixed after pre-extraction and stained with 53BP1 antibody. Number of 53BP1 foci in S/G2 cells was evaluated using DAPI content of >2 n to gate S-G2 cells. Mean of median foci number +/− SD is plotted. Statistical significance evaluated by two-tailed t-test.

Figure 6

Putative model for the role of WIP1 in HR and in PARP inhibitor sensitivity. (A) Under normal conditions, endogenous DNA lesions are efficiently repaired by HR and BER. WIP1 promotes efficiency of HR and limits the extent of p53 pathway activation allowing cells to proliferate. (B) After inhibition of PARP1, BER pathway is impaired (dashed lines) but DNA lesions are efficiently repaired by HR. (C) Combined inhibition of PARP1 and WIP1 impairs both BER and HR (dashed lines) and enhances p53 response leading to accumulation of DNA lesions in G2 cells. Increased DNA damage load triggers the DNA damage response and allows full activation of p53 pathway leading to cell death.