RAD52 inactivation is synthetically lethal with deficiencies in BRCA1 and PALB2 in addition to BRCA2 through RAD51-mediated homologous recombination

Abstract

Synthetic lethality is an approach to study selective cell killing based on genotype. Previous work in our laboratory has shown that loss of RAD52 is synthetically lethal with BRCA2 deficiency, while exhibiting no impact on cell growth and viability in BRCA2-proficient cells. We now show that this same synthetically lethal relationship is evident in cells with deficiencies in BRCA1 or PALB2, which implicates BRCA1, PALB2 and BRCA2 in an epistatic relationship with one another. When RAD52 was depleted in BRCA1- or PALB2-deficient cells, a severe reduction in plating efficiency was observed, with many abortive attempts at cell division apparent in the double-depleted background. In contrast, when RAD52 was depleted in a BRCA1- or PALB2-wildtype background, a negligible decrease in colony survival was observed. The frequency of ionizing radiation-induced RAD51 foci formation and double-strand break-induced homologous recombination (HR) was decreased by 3- and 10-fold, respectively, when RAD52 was knocked down in BRCA1- or PALB2-depleted cells, with minimal effect in BRCA1- or PALB2-proficient cells. RAD52 function was independent of BRCA1 status, as evidenced by the lack of any defect in RAD52 foci formation in BRCA1-depleted cells. Collectively, these findings suggest that RAD52 is an alternative repair pathway of RAD51-mediated HR, and a target for therapy in cells deficient in the BRCA1-PALB2-BRCA2 repair pathway.

Figure 1

RAD52 exhibits synthetic lethality with BRCA1 and PALB2. (a) (Left) Plating efficiency (clonogenic survival) of MCF7 cells that were cultured as previously described in Feng et al., and were depleted of BRCA1, RAD52 or both proteins by commercially available siRNAs (siBRCA1: siGENOME SMARTpool, Dharmacon, Lafayette, CO, USA; siRAD52: ID no. 142431, Ambion, Austin, TX, USA; control: negative control siRNA cat. no. 1027310, Qiagen, Valencia, CA, USA), and transfected into cells by Nucleofector (Lonza, Basel, Switzerland) according to the manufacturers’ recommendations. Plating efficiency was measured in MCF7 cells by seeding in triplicate 100, 400, 800, 1600, 3200 and 6400 cells into six-well plates. Cells were plated 72 h after transfection with the labeled siRNAs. Colonies (> 50 cells) were counted by fixing and staining with crystal violet (0.05%) 14–21 days after seeding. The results shown are from three independent experiments with error bars representing s.e.m. Differences between cells transfected with BRCA1 siRNA alone compared with both BRCA1 and RAD52 siRNA are significant (P=0.0239), as determined by an unpaired t-test. (Right) Photomicrographs are of representative colonies 14 days after plating. (b) (Left) Plating efficiency of MCF7 cells following siRNA depletion of PALB2, RAD52 or both proteins as in (a) (siPALB2: siGENOME SMARTpool, Dharmacon). Differences between cells transfected with PALB2 siRNA alone compared with both PALB2 and RAD52 siRNA are significant (P=0.0140), as determined by an unpaired t-test. (Right) Photomicrographs are of representative colonies 14 days after plating (as in a). (c) Plating efficiency of U2OS cells that were cultured in the same conditions, as previously described for MCF7 cells. U2OS cells were transfected with the labeled siRNA oligos (as in a, b) and plated 72 h later. Colonies were counted by fixing and staining, as described above. Differences between cells transfected with BRCA1 siRNA alone compared with both BRCA1 and RAD52 siRNA are significant (P=0.036), as determined by an unpaired t-test, in addition to cells transfected with PALB2 siRNA alone compared with both PALB2 and RAD52 siRNA (P=0.0061).

Figure 2

RAD51 foci formation in BRCA1-deficient cells is dependent on RAD52, but RAD52 foci formation is not dependent on BRCA1. (a) HCC1937 cells, complemented with a pcDNA3 expression vector containing wildtype BRCA1 or empty control, as previously reported, were transfected with the indicated siRNAs and immunoblotted with the indicated antibodies (anti-BRCA1: OP92, Calbiochem, La Jolla, CA, USA; anti-Rad52: SC-8350, Santa Cruz, Santa Cruz, CA, USA; anti-SMC1: A300-055A, Bethyl, Montgomery, TX, USA; anti-HDAC: #06-720, Upstate/Millipore, Billerica, MA, USA). Confocal images of IR-induced RAD51 foci in HCC1937 cells that were stably transfected with wildtype BRCA1 or a vector control. Cells were then transiently transfected with RAD52 siRNA or a control siRNA, then seeded in eight-chamber tissue culture slides (Fisher, Pittsburgh, PA, USA) and incubated overnight. Cells were fixed in 4% paraformaldehyde at room temperature for 15 min, followed by a block and permeabilization step with 10% BGS and 0.5% Triton-X for 1 h. Commercially available antibodies against RAD51 (Ab-1, Santa Cruz) were used for staining, as previously described. Images were obtained using a Carl Zeiss confocal laser scanning microscope (Carl Zeiss Microscopy, LLC, Thornwood, NY, USA) and processed using ImageJ and Adobe Photoshop software. For foci quantification 200 nuclei were counted, with cells forming > 5 foci scored as positive. The percentage of cells with RAD51 foci is shown (mean and s.e. from three independent experiments). Differences between HCC1937 cells stably transfected with an empty vector and then control siRNA compared with empty vector cells transfected with RAD52 siRNA are significant (P=0.0234), as determined by an unpaired t-test. (b) MCF7 cells were transfected with GFP-tagged RAD52 and a pool of cells were stably selected with G418 (Mediatech, Inc., Manassus, VA, USA) at 1 mg/ml concentration. These MCF7-RAD52-GFP cells were transfected with control or BRCA1 siRNA. After 10 Gy of IR, cells were fixed and stained with commercially available antibodies against BRCA1 (SC-6954, Santa Cruz) and phosphorylated Ser-139 H2AX (05-636, Millipore, Billerica, MA, USA). Confocal images of RAD52-GFP, BRCA1 and gamma-H2AX (gH2AX) foci were captured and scored. The percentage of cells with RAD52 and BRCA1 foci is shown (mean and s.e. from three independent experiments). Differences in BRCA1 foci between control and BRCA1 siRNAtransfected cells were significant (P=0.002), as determined by an upaired t-test.

Figure 3

RAD52 is required for homology-directed gene conversion in BRCA1-depleted cells. (a) Flow cytometry of H1299 cells cultured as previously described, containing the recombination substrate, pDR-GFP, transfected with the I-SceI endonuclease plasmid or with vector alone by Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s recommendations and analyzed for the percentage of green fluorescent cells. At 24 h after transfection with respective siRNAs, cells with the pDR-GFP construct stably integrated were transfected with pCMV-I-SceI to induce DSBs. Cells were collected 72 h later for analysis using a FACScan flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). (b) Immunoblots of H1299-pDR-GFP and MCF7-pDR-GFP cells that were transfected with the indicated combinations of siRNA. (c) The frequency of I-SceI break-induced recombination (HR) measured by flow cytometry in H1299-pDR-GFP and MCF7-pDR-GFP transfected with the labeled siRNAs. The bars represent the results of three or more independent experiments; error bars show s.e.m. Differences between cells transfected with BRCA1 siRNA alone compared with both BRCA1 and RAD52 siRNA are significant (P=0.031, P=0.024 for H1299-pDR-GFP and MCF7-pDR-GFP cells, respectively), as determined by an unpaired t-test. A summary of the cell cycle distribution shows similar rates between the labeled siRNA conditions; see Supplementary Figure 2 for detailed methods and data.

Figure 4

RAD52 is required for RAD51-mediated HR in PALB2-depleted cells. (a) Confocal images of IR-induced RAD51 foci in MCF7 cells following siRNA depletion of PALB2, RAD52 or both proteins. The percentage of cells with RAD51 foci is shown (mean and s.e.m from three independent experiments). Differences between cells transfected with PALB2 siRNA alone compared with both PALB2 and RAD52 siRNA are significant (P=0.0471), as determined by an unpaired t-test. (b) Immunoblots of H1299-pDR-GFP and MCF7-pDR-GFP cells transfected with the labeled combinations of control, PALB2 and RAD52 siRNA. (c) The frequency of I-SceI break-induced recombination (HR) measured by flow cytometry in H1299-pDR-GFP and MCF7-pDR-GFP cells transfected with the indicated siRNAs. The bars represent the results of three or more independent experiments; error bars show s.e.m. Differences between cells transfected with PALB2 siRNA alone compared with both PALB2 and RAD52 siRNA are significant (P=0.0045, P=0.0096 for H1299-pDR-GFP and MCF7-pDR-GFP cells, respectively), as determined by an unpaired t-test.