RAD51C facilitates checkpoint signaling by promoting CHK2 phosphorylation

Abstract

The RAD51 paralogues act in the homologous recombination (HR) pathway of DNA repair. Human RAD51C (hRAD51C) participates in branch migration and Holliday junction resolution and thus is important for processing HR intermediates late in the DNA repair process. Evidence for early involvement of RAD51 during DNA repair also exists, but its function in this context is not understood. In this study, we demonstrate that RAD51C accumulates at DNA damage sites concomitantly with the RAD51 recombinase and is retained after RAD51 disassembly, which is consistent with both an early and a late function for RAD51C. RAD51C recruitment depends on ataxia telangiectasia mutated, NBS1, and replication protein A, indicating it functions after DNA end resection but before RAD51 assembly. Furthermore, we find that RAD51C is required for activation of the checkpoint kinase CHK2 and cell cycle arrest in response to DNA damage. This suggests that hRAD51C contributes to the protection of genome integrity by transducing DNA damage signals in addition to engaging the HR machinery.

Figure 1.

hRAD51C forms nuclear foci in response to IR. (A) Western blot detection of hRAD51C. An immunoblot of extracts prepared from irradiated and nonirradiated HeLa and U2OS cells was probed with anti-hRAD51C antibody. Recombinant hRAD51C-His₁₀, migrating more slowly because of its epitope tag, served as a control. (B) Immunofluorescence detection of RAD51C in paraformaldehyde-fixed HeLa and U2OS cells. Irradiated (10 Gy) and nonirradiated cultures were stained with anti-hRAD51C monoclonal antibody (red) 2 h after treatment. DNA was counterstained with DAPI (blue). (C) siRNA-mediated reduction of RAD51C expression. Cells were transfected with control or RAD51C siRNA and irradiated 48 h after transfection. Cell extracts, prepared 2 h after irradiation, were analyzed by Western blotting. (D) siRNA-mediated reduction of RAD51C foci. As in C, but cells were fixed 2 h after IR and stained with anti-hRAD51C monoclonal antibody (red). Image acquisition parameters were identical for the three samples. (E) Partial colocalization of RAD51C and RAD51 in irradiated HeLa cells stained with anti-RAD51C (red) and anti-RAD51 antibodies (green). Nuclei displaying only RAD51C or RAD51 foci or containing both types of foci are labeled with a red, green, or yellow circle, respectively. (F) Enlarged image of a cell containing both RAD51C (red) and RAD51 foci (green). Bars, 10 µm.

Figure 2.

Distinct kinetics of RAD51C and RAD51 foci formation. HeLa cells were fixed at the indicated time points after 10 Gy of irradiation and stained with anti-RAD51C (red) and anti-RAD51 antibodies (green). DNA was stained with DAPI (blue). (A) More than 100 cells were scored at each time point. Nuclei containing >5 foci were counted as positive. The level of spontaneous RAD51 foci in nonirradiated cells (RAD51 − IR) is indicated in blue and was subtracted from the number of IR-induced RAD51 foci (RAD51 + IR). (B) Representative images of cells at 20, 60, and 360 min after IR. (C) An example of a DAPI-stained nucleus (blue) with RAD51C foci (red) processed using TRI2 software. The computer-generated image delineates the nucleus (red) and individual foci (blue). (D and E) Computational analysis of RAD51 and RAD51C foci in at least 100 cells at the indicated time points after IR. The blue lines indicate the mean number of foci per cell. Bars, 10 µm.

Figure 3.

RAD51C assembles before and independently of RAD51 at sites of DNA damage. (A) HeLa cells were incubated with 2 mM HU for 17 h and then released into fresh medium. Cells were fixed 0.1 or 4 h after release and stained with anti-RAD51C (red) and anti-RAD51 antibodies (green). (B) For each of the two samples in A, at least 100 cells were examined for the presence of RAD51 or RAD51C foci. The mean and standard deviations of three independent experiments are shown. (C) CAPAN-1 (Brca2^−/tr) cells were irradiated (10 Gy), fixed 2 h after irradiation, and stained with anti-RAD51C (red) and anti-RAD51 antibodies (green). Bars, 10 µm.

Figure 4.

RAD51C accumulation at DNA lesions in S/G2 requires ATM and NBS1. (A) U2OS cells were irradiated (10 Gy of IR), fixed after 2 h, and stained with anti-RAD51C or anti-RPA (red) and anti–CENP-F antibodies (green). Nuclei were visualized with DAPI (blue). (B) U2OS cells were treated with ATM inhibitor (Ku55933) or solvent (DMSO) and then irradiated (10 Gy), fixed 2 h later, and stained with anti-RAD51C (red) and anti-RPA antibodies (green). (C) Atm-deficient cells stably transfected with a vector expressing full-length ATM or an empty vector were treated and processed as in B. (D) Nbs1-deficient cells stably transfected with a vector expressing full-length NBS1 or an empty vector were treated and stained as in B. (E) The percentage of nuclei exhibiting RAD51C or RPA foci was determined for at least 100 cells from D. The mean and standard deviation of three independent experiments are shown. Bars, 10 µm.

Figure 5.

RAD51C assembly at break sites depends on RPA. (A) HeLa cells were irradiated (10 Gy) and fixed 2 h later or treated with 2 mM HU for 17 h, released into fresh media, and fixed 4 h after release. Cells were stained with anti-RAD51C (red) and anti-RPA antibodies (green). Nuclei were visualized with DAPI (blue). (B) Control or RAD51C siRNA was cotransfected with a GFP-expressing vector. U2OS cells were irradiated 48 h after transfection, fixed after 2 h, and stained with anti-RPA and anti-RAD51 antibodies (red). (C) siRNA-mediated reduction of RPA expression. Cells were transfected with control or RPA siRNA and irradiated 48 h after transfection. Cell extracts were analyzed by Western blotting against RPA and tubulin. (D) Control or RPA siRNA was cotransfected with a GFP-expressing vector into U2OS cells. Cells were irradiated 48 h after transfection, fixed after 2 h, and stained against RAD51C. Bars, 10 µm.

Figure 6.

RAD51C promotes cell cycle arrest in response to DNA damage. (A) RAD51C depletion reduces IR-induced S-phase delay. U2OS cells were irradiated (1 Gy) 48 h after transfection with control or RAD51C siRNA and, 2 h later, fixed for propidium iodide staining and flow cytometry. G1, S, and G2/M populations were assigned using CellQuest. (B) Analysis of the G2/M checkpoint. As in A, but cells were stained with an antibody against phosphorylated histone H3 and propidium iodide. (A and B) The mean and standard deviation from three independent experiments are shown. Significance in difference between the grouped values was tested using an unpaired t test. (C) shRNA-mediated reduction of RAD51C expression in primary MEFs. Cells were infected twice at 12-h intervals with retroviruses expressing control or RAD51C shRNA and then selected with puromycin for 48 h. Cell extracts prepared 3 d after the first infection were analyzed by Western blotting against mouse RAD51C and tubulin. (D) Cells treated as in C were incubated with colcemid for 4 h and processed for FISH analysis of metaphase chromosome spreads with a Cy3-conjugated telomeric peptide nucleic acid probe (red). Arrowheads point to chromatid breaks. (E) Quantification of chromatid and chromosome break frequencies in metaphase spreads from MEFs established from two littermate embryos, each transfected with either control or RAD51C shRNA. 30–50 metaphases were analyzed for each sample. Bar, 10 µm.

Figure 7.

RAD51C is required for IR-induced CHK2 phosphorylation. (A) U2OS cells were transfected with control or RAD51C siRNA and irradiated (10 Gy) 48 h after transfection. After 2 h, extracts were prepared and immunoblotted as indicated. (B) As in A, but cells were treated with 1 Gy of IR, and protein samples were analyzed after 0.5, 1, and 2 h. (C) As in B, but using HeLa cells. (D) U2OS cells were transfected with a control siRNA or two different siRNAs against XRCC3 and irradiated (1 Gy) 48 h after transfection. After 2 h, extracts were prepared and immunoblotted as indicated. (E) HCT116 Chk2^−/− cells were treated with 10 Gy of irradiation and fixed 2 h after. Cells were stained with anti-RAD51C (red) and anti-RPA antibodies (green). (F) Model for RAD51C action in response to DSBs in S/G2. RAD51C is recruited by RPA to break sites, where it facilitates CHK2-mediated cell cycle arrest and assembly of the HR machinery. Bar, 10 µm.