SWI/SNF complexes are required for full activation of the DNA-damage response

Abstract

SWI/SNF complexes utilize BRG1 (also known as SMARCA4) or BRM (also known as SMARCA2) as alternative catalytic subunits with ATPase activity to remodel chromatin. These chromatin-remodeling complexes are required for mammalian development and are mutated in ~20% of all human primary tumors. Yet our knowledge of their tumor-suppressor mechanism is limited. To investigate the role of SWI/SNF complexes in the DNA-damage response (DDR), we used shRNAs to deplete BRG1 and BRM and then exposed these cells to a panel of 6 genotoxic agents. Compared to controls, the shRNA knockdown cells were hypersensitive to certain genotoxic agents that cause double-strand breaks (DSBs) associated with stalled/collapsed replication forks but not to ionizing radiation-induced DSBs that arise independently of DNA replication. These findings were supported by our analysis of DDR kinases, which demonstrated a more prominent role for SWI/SNF in the activation of the ATR-Chk1 pathway than the ATM-Chk2 pathway. Surprisingly, γH2AX induction was attenuated in shRNA knockdown cells exposed to a topoisomerase II inhibitor (etoposide) but not to other genotoxic agents including IR. However, this finding is compatible with recent studies linking SWI/SNF with TOP2A and TOP2BP1. Depletion of BRG1 and BRM did not result in genomic instability in a tumor-derived cell line but did result in nucleoplasmic bridges in normal human fibroblasts. Taken together, these results suggest that SWI/SNF tumor-suppressor activity involves a role in the DDR to attenuate replicative stress and genomic instability. These results may also help to inform the selection of chemotherapeutics for tumors deficient for SWI/SNF function.

Figure 1. Simultaneous depletion of both SWI/SNF catalytic subunits in D98 cells

Western blot analysis of BRG1, BRM, and tubulin loading control of D98 control cells and shRNA knockdown cells. After normalizing protein levels to tubulin, % BRG1 and % BRM refer to the amount of protein remaining in the cells expressing shRNAs compared to the control cell line.

Figure 2. Dose-response survival curves of D98 control cells and shRNA knockdown cells after being exposed to various genotoxic agents

Each data point represents the mean ± standard deviation for three replicates. The legend displays the SF50 dose for controls and shRNA cells. SF70 values are provided when the SF50 is not reached in control cells.

Figure 3. Activation of DDR proteins in D98 control cells versus shRNA knockdown cells (shRNAs) after exposure to genotoxic agents

Western blots are shown for P-ATM S1981 and total ATM, P-Chk1 S345 and total Chk1, and P-Chk2 T68 and total Chk2. Genotoxic agents are shown below the western blot panels, and the doses are shown above each lane. The results are representative of 3 independent experiments.

Figure 4. Attenuated induction of DDR protein phosphorylation in shRNA knockdown cells

Quantification of P-Chk1 S345, P-ATM S1981, and P-Chk2 T68 levels normalized to total protein levels of each protein. The values correspond to the fold increase for each dose compared to vehicle-treated controls. Each panel corresponds to treatment with a different genotoxic agent (listed above) and shows controls (Cntrl) to the left and shRNA knockdown cells (shRNAs) to the right.

Figure 5. Attenuated induction γ-H2AX in D98 control cells versus shRNA knockdown cells after exposure to etoposide

(A) Western blots showing γH2AX and total H2AX as a loading control. Etoposide doses are shown above each lane. (B) Relative levels of γH2AX normalized to total H2AX.

Figure 6. Depletion of the SWI/SNF catalytic subunits in NHF1-hTERT cells and functional outcomes

(A) Western blot analysis of BRG1, BRM, and tubulin loading control in NHF1-hTERT cells transfected with the following siRNAs as indicated above each lane: NTC (non-targeted control), BRG1, BRM, or BRG1 and BRM simultaneously. Protein lysates were prepared 72 hours (left) or 96 hours (right) after electroporation of the siRNAs. Shown below each lane is the quantification of BRG1 and BRM protein levels normalized to tubulin. Protein levels in siBRG1 and/or siBRM cells are shown as a percentage of controls (which are set at 100). (B) Colony formation assays of NHF1-hTERT depleted of BRG1 or BRM, or both SWI/SNF catalytic subunits. Histograms show the mean ± standard deviation based on 3 independent experiments. (C) Representative images of normal nuclei and nuclei with buds, blebs, and necks. 100X magnification. (D) Quantification of nuclei with buds, blebs, and necks in NTC siRNA cells and double-knockdown cells (BRG1 + BRM siRNAs) at 72 hours and 96 hours after electroporation of siRNAs. (E) Activation of DDR proteins in NHF1-hTERT cells with RNAi-mediated knockdown of nontargeted control (NTC), BRG1, BRM, or BRG1 and BRM simultaneously. Western blots are shown for P-ATM S1981 and total ATM, P-Chk2 T68 and total Chk2, and P-Chk1 S345 and total Chk1. IR and UVC treatments are shown above each lane. The results are representative of 2–3 independent experiments.