To express green fluorescent protein (GFP)-tagged SMARCA2, SMARCA4, SMARACC1 and SMARCD1, the cDNAs of SMARCA2, SMARCA4, SMARACC1 and SMARCD1 were synthesized by Tsingke Biotechnology (Beijing, China) based on the sequences from NCBI (NM_003070 for SMARCA2, NM_001128844 for SMARCA4, NM_003074 for SMARCC1, and NM_003076 for SMARCD1). The complementary DNAs (cDNAs) were then cloned into pEGFP-C1 (Clontech, Beijing, China) to generate GFP-SMARCA2, GFP-SMARCA4, GFP-SMARCC1, and GFP-SMARCD1. For the expression of truncation mutants of SMARCA2, N-terminal (N-Ter, amino acids 1-653), helicase domain (Hel, amino acids 654-1383), and bromo domain (BRD, amino acids 1384-1590) were cloned into pEGFP-C1 to generate GFP-N-Ter, GFP-Hel and GFP-BRD. The K755A mutant of SMARCA2 and the K785A mutant of SMARCA4 were generated by site-directed mutagenesis from the wild type SMARCA2 and SMARCA4, respectively [21, 22].

HeLa, HCC1187, MDA-MB-468 and MDA-MB-231 cells were purchased from American Type Culture Collection. HeLa cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Catalog number: 11965092; Thermo Fisher Scientific, Waltham, MA, USA), HCC1187 cells were cultured in Roswell Park Memorial Institute (RPMI)-1640 medium (Catalog number: 11875168; Thermo Fisher Scientific, Waltham, MA, USA), and MDA-MB-468 and MDA-MB-231 cells were cultured in Leibovitz’s L-15 Medium (Catalog number: 21083027; Thermo Fisher Scientific, Waltham, MA, USA), containing 10% fetal bovine serum and 1% penicillin-streptomycin (Catalog number: 15140122; Thermo Fisher Scientific, Waltham, MA, USA) at 37 °C with 5% CO${}_2$. All cell lines were validated by Short Tandem Repeat (STR) profiling and tested negative for mycoplasma.

To ectopically express GFP-tagged proteins, transfection of the cells with all pEGFP constructs was carried out using Lipofectamine 2000 (Catalog number: 11668500; Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s protocol.

To generate HeLa SMARCA2 or HeLa SMARCA4 knockout cells, CRISPR/Cas9-mediated gene editing was performed. Cells were transfected with single guide RNA (sgRNA) against SMARCA2, constructed in pSpCas9(BB)-2A-Puro (PX459 V2.0) vector (Addgene plasmid ID: 62988). The sgRNA sequence targeting SMARCA2 was 5${}^{\prime }$-TCCCATCCTATGCCGACGAT-3${}^{\prime }$. To generate HeLa SMARCA4 knockout cells, the sgRNA sequence for SMARCA4 was 5${}^{\prime }$-CCGGCGAGGGACCCGGGCTA-3${}^{\prime }$. Transfection of the cells was carried out using Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer’s protocol. The transfected cells were selected in medium containing 2 µg/mL puromycin and then sub-cloned into 96-well plates 24 h post transfection. The monoclonal cell lines were identified. For radiation sensitive 51 (RAD51) gene knockdown, the sequence of si-RAD51 was 5${}^{\prime }$-CUAAUCAGGUGGUAGCUCAUU-3${}^{\prime }$. The sgRNAs and small interfering RNA (si-RNA) were synthesized by Tsingke Biotechnology company. Si-RNA transfections were carried out using Lipofectamine RNAiMAX (Catalog number: 13778075; Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer’s protocol.

The following antibodies were purchased from their respective companies: anti-$\gamma$H2AX (Novus, NB100-384), anti-RAD51 (Catalog Number: MA1-23271; Invitrogen, Carlsbad, CA, USA), anti-ring finger protein 8 (RNF8) (Catalog number: 14112-1-AP, Proteintech, Wuhan, Hubei, China), anti-breast cancer susceptibility gene 1 (BRCA1) (Catalog Number: NB100-404, Novus, Littleton, CO, USA), anti-SMARCA2 (Catalog number: HPA029981; Sigma, Louis, MO, USA).

Laser microirradiation assays were performed according to a previous study [23]. GFP-tagged constructs were transfected into the specified cells and plated on 35-mm glass-bottom dishes. Cellular DNA damage in the nuclei of cultured cells was induced using microirradiation with a pulsed nitrogen laser (Spectra-Physics; 365 nm, 10 Hz pulse). This laser system was directly integrated into the epifluorescence microscope’s imaging pathway, focusing through a Plan-Apochromat 63$\times$/NA 1.40 oil immersion objective. The laser power was adjusted to 50–70% of its maximum capacity. The resulting green fluorescence stripes were captured, analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA) and the results presented represent images from 20 cells across three independent experiments. For inhibitor treatment, cells were treated with different inhibitors at 1 µM for 2 hours followed by laser microirradiation. The SMARCA2/4 inhibitor (FHD286, HY-144835), Ataxia-telangiectasia-mutated (ATM) inhibitor (KU55933, HY-12016), Ataxia telangiectasia and Rad3-related protein (ATR) inhibitor (VE-821, HY-14731), CREB-binding protein (CBP)/p300 inhibitor (CBP/p300-IN-21, HY-155229), and Poly (ADP-ribose) polymerase (PARP) 1/2 inhibitor (Olaparib, HY-10162) were purchased from MedChemExpress (Shanghai, China).

Cells growing on glass coverslips were irradiated with 5 Gy and allowed to recover for 12 h before immunofluorescence. Then, cells were treated with fixation buffer (4% PFA, 0.1% Triton X-100). The slides were incubated with different primary antibodies in 1% BSA at room temperature for 2 h or 4 °C overnight. Subsequently, the slides were washed with bovine serum albumin (PBS) and incubated with a secondary antibody. DNA was stained, and coverslips were mounted with 4${}^{\prime }$,6-diamidino-2-phenylindole (DAPI) Fluoromount-G™ (Catalog Number: 36308ES20; Yeason, Shanghai, China). Cells were treated with FHD286 or Compound 14 at 1 µM for 2 hours followed by ionizing irradiation. Compound 14 was another specific SMARCA2/4 ATPase dual inhibitor, which was synthesized according to the compound structure [24]. SMARCA2, $\gamma$H2AX, BRCA1, RNF8 and RAD51 foci were stained with the antibodies list in section 2.3.

The DR-GFP reporter assay system was provided by Dr. Xiaochen Gai [25] at Westlake University. The assay was performed as described in a previous study [25]. The DR-GFP reporter system was specifically designed to quantify homologous recombination repair, where the iGFP gene fragment functions as a template for the HR process following an I-SceI-induced DSB in the upstream SceGFP cassette. To initiate this process, the HA-I-SceI-GR pCW tet-on plasmid was transfected into the cells. Doxycycline is then administered to trigger the expression of I-SceI for 18 hours. Acetonide treatment facilitates the translocation of I-SceI from the cytoplasm to the nucleus. Subsequently, I-SceI cuts the targeting site, generating a DSB, which upon repair, leads to the expression of GFP. Flow cytometry was performed to examine the efficiency of homologous recombination repair by calculating the GFP positive populations. For RAD51 knockdown, cells with DR-GFP insertion were transfected with negative control siRNA (siRNA-NC) or siRNA-RAD51 for 24 hours following HA-I-SceI-GR pCW tet-on plasmid transfection. For inhibitor treatment, cells with DR-GFP insertion were pretreated with or without FHD286 (1 µM) or Compound 14 (1 µM) for 2 hours before acetonide treatment.

SMARCA4 and SMARCA2 double knockdown was carried out by siRNA transfection directed against SMARCA2 (siRNA-SMARCA2): 5${}^{\prime }$-UCGUCGAGCAAUCAUUUGGUU-3${}^{\prime }$ and (siRNA-SMARCA4) SMARCA4: 5${}^{\prime }$- UAGCAUUGAGGGCUGUCUCCA-3${}^{\prime }$. A nontargeting control siRNA (siRNA-NC): 5${}^{\prime }$-AACCACGUGAGGCAUCCAGGC-3${}^{\prime }$ was used as a negative control. For double knockdown, individual siRNAs of SMARCA2 and SMARCA4 were co-transfected into cells with DR-GFP insertion 24 hours before HA-I-SceI-GR pCW tet-on plasmid transfection. The siRNA transfections were carried out using Lipofectamine RNAiMAX (Invitrogen). The full-length siRNA-resistant SMARCA2 was cloned into S protein-FLAG-Streptavidin binding peptide (SFB) vector. The K755A mutant of SMARCA2 was generated by site-directed mutagenesis from the siRNA-resistant full-length SMARCA2. Re-expression of full-length SMARCA2 or the ATPase inactive mutant (K755A) was performed by plasmid transfection using Lipofectamine 2000 (Thermo Fisher Scientific) 20 hours before HA-I-SceI-GR pCW tet-on plasmid transfection. Knockdown and/or re-expression were confirmed by Western blotting using the following primary antibodies: SMARCA4 (Catalog number: sc-17796; Santa Cruz, Shanghai, China), SMARCA2 (Catalog number: sc-166579; Santa Cruz, Shanghai, China), Vinculin (Catalog number: A14193; Abclonal, Wuhan, Hubei, China).

For the clonogenic cell proliferation assay, cells were seeded in 6-well plate (1000 cells/well). The cells were then treated with either a vehicle control or drugs with the indicated concentration on the next day, and culture medium was refreshed every 3 days for a total of 14 days in total. At the endpoints of the colony formation assays, the colonies were washed with PBS and fixed with 70% methanol for 10 minutes. This was followed by staining with 0.1% crystal violet. The number of colonies was counted by ImageJ software, and survival graphs were generated from three replicate wells, with colony numbers normalized to untreated controls.

Statistical analyses were performed using GraphPad Prism7 (GraphPad Software, Inc., San Diego, CA, USA). The statistical significance was assessed using a two-tailed unpaired Student’s t-test. All of the data were presented as mean $\pm$ SD. The significance levels are annotated as follows: *, p $\le$ 0.05. **, p $\le$ 0.01. ***, p $\le$ 0.001.

To study the molecular mechanism by which the SWI/SNF complex is recruited to DNA lesions, we examined SMARCA2/4, the key enzymatic subunits in the SWI/SNF complex. Using laser microirradiation assay and observation by fluorescence microscopy, we found that both GFP-tagged SMARCA2/4 relocated to DNA lesions (Fig. 1A). We also measured the recruitment kinetics and found that within five seconds, SMARCA2/4 was recruited to DNA lesions (Supplementary Fig. 1).

Fig. 1.

The relocation of the switching/sucrose non-fermentable (SWI/SNF) Related, Matrix Associated, Actin Dependent Regulator Of Chromatin, Subfamily A (SMARCA) member 2 and member 4 (SMARCA2/4) to DNA lesions. (A) pEGFP-C1-SMARCA2, pEGFP-C1-SMARCA4, or pEGFP-C1 was transfected into HeLa cells for 24 hours. Following laser microirradiation, the laser stripes of green fluorescent protein (GFP)-SMARCA2/4 were observed using fluorescence microscopy. The laser stripe is indicated with white arrows. The table on the right presents the cellular status, specifically indicating the presence or absence of stripes. Scale bar: 10 μm. (B) Schematic of the deletion mutants of SMARCA2. (C) GFP-tagged truncation mutants of SMARCA2 were transiently expressed in HeLa cells. Following laser microirradiation, the laser stripes of GFP-N-Ter, GFP-Hel and GFP-bromo domain (BRD) were examined using fluorescence microscopy. The laser stripe is indicated with white arrows. The table on the right presents the cellular status, specifically indicating the presence or absence of stripes. Scale bar: 10 μm. (D) GFP-SMARCA2-K755A or GFP-SMARCA4-K785A were transiently expressed in HeLa cells. Following laser microirradiation, the laser stripes of these enzymatically dead mutants were examined. Scale bar: 10 μm.

SMARCA2/4 has several distinct domains, including the N-terminal QLQ motif and HAS domain, which mediate the interaction with other subunits of the SWI/SNF complex; the ATPase/helicase domain in the middle, which is the solo enzymatic domain in the whole complex; the C-terminal bromodomain that is known to recognize acetylated histones [10, 22] (Fig. 1B). Since SMARCA2/4 shares an identical domain architecture, we generated deletion mutants of SMARCA2 and found that only the ATPase/helicase domain (indicated as GFP-Hel), but not other domains (indicated as GFP-N-Ter or GFP-BRD), was able to relocate to DNA lesions (Fig. 1C). Moreover, we established the ATPase-dead constructs which mutated the key lysine residue into alanine in ATPase/helicase domains of SMARCA2 (K755A) and SMARCA4 (K785A) to abolish their ATP-binding activities [22, 26], and found that these ATPase activities were required to target SMARCA2/4 to DNA lesions (Fig. 1D).

Next, we examined other core subunits of the SWI/SNF complex, including SMARCC1 and SMARCD1. We show the first evidence that, similar to SMARCA2/4, both SMARCC1 and SMARCD1 relocated to DNA lesions when cells were treated with laser microirradiation (Fig. 2A). Interestingly, when we pre-treated cells with FHD286, a specific dual inhibitor to kill both ATPase activities of SMARCA2/4 [27], SMARCC1 and SMARCD1 were no longer recruited to DNA lesions in the laser microirradiation assays (Fig. 2B). This suggests that the ATPase activity of SMARCA2/4 is also required for targeting other subunits of the SWI/SNF complex to the sites of DNA damage.

Fig. 2.

Relocation of the switching/sucrose non-fermentable (SWI/SNF) to DNA lesions. (A) pEGFP-C1-SMARCC1, pEGFP-C1-SMARCD1, or pEGFP-C1 was transfected into HeLa cells for 24 hours. Following laser microirradiation, the laser stripes of GFP-SMARCC1 or GFP-SMARCD1 were observed using fluorescence microscopy. The laser stripe is indicated with white arrows. The right table presents the cellular status, specifically indicating the presence or absence of stripes. Scale bar: 10 μm. (B) GFP-SMARCA2, GFP-SMARCC1 or GFP-SMARCD1 was transiently expressed in HeLa cells. The cells were pre-treated with or without 1 µM FHD286 for two hours followed by laser microirradiation. The localization of SMARCA2, SMARCC1, and SMARCD1 was examined. The laser stripe is indicated with white arrows. The right table presents the cellular status, specifically indicating the presence or absence of stripes. Scale bar: 10 μm. (C) GFP-SMARCA2, GFP-SMARCC1 or GFP-SMARCD1 was transfected into HeLa cells for 24 hours. Cells were treated with 1 µM Ataxia-telangiectasia-mutated (ATM) inhibitor (ku55933), 1 µM Ataxia telangiectasia and Rad3-related protein (ATR) inhibitor (VE-821), 1 µM CREB-binding protein (CBP) and its homologue p300 (CBP/p300) inhibitor (CBP/p300-IN-21) or 1 µM Poly (ADP-ribose) polymerase (PARP) inhibitor (olaparib) for two hours followed by laser microirradiation. The localization of SMARCA2, SMARCC1 and SMARCD1 was examined. The localization of the subunits of the SWI/SNF complex was also examined in the H2AX KO HeLa cells. The laser stripe is indicated with white arrows. The right table presents the cellular status, specifically indicating the presence or absence of stripes. Scale bar: 10 μm.

It has been reported that several DNA repair factors may mediate the recruitment of the SWI/SNF complex to DNA lesions, such as $\gamma$H2AX, p300/CBP and PARylation [15, 19, 28]. Here, we examined H2AX-deficient cells as well as cells treated with inhibitors of ATM, ATR, p300/CBP, or PARP1/2. To our surprise, the SWI/SNF complex was still recruited to DNA lesions under these conditions (Fig. 2C). Only suppression of the ATPase/Helicase domain of SMARCA2/4 abolishes the recruitment of the SWI/SNF complex.

Earlier studies have shown that the enrichment of DNA damage response factors, such as $\gamma$H2AX, requires SMARCA2 [18, 19]. However, different results have shown that the SWI/SNF complex is the downstream effector of $\gamma$H2AX at the beginning of the DSB response [29]. To further elucidate the role of the SWI/SNF complex in DNA damage response, we used CRISPR-Cas9 technology to knockout either SMARCA2 or SMARCA4 and treated cells with ionizing radiation (IR) to induce DSBs. However, without either SMARCA2 or SMARCA4, the enrichment of $\gamma$H2AX at DSBs was not affected (Supplementary Fig. 2). Since lacking both SMARCA2 and SMARCA4 causes synthetic lethality for cells, we cannot use CRISPR-Cas9 technology to obtain SMARCA2/4 double knockout cells. Thus, we had to pre-treated cells with FHD286 for two hours to transiently suppress the enzymatic activity of SMARCA2/4 before the induction of DSBs by IR. Interestingly, the foci of $\gamma$H2AX were not suppressed by the treatment of FHD286. Instead, we found that the foci of $\gamma$H2AX existed for a prolonged period following IR treatment (Fig. 3). As a control, without IR treatment, neither FHD286 nor Compound 14 induced foci formation of $\gamma$H2AX, suggesting that these chemical compound per se do not induce DSBs (Supplementary Fig. 3). We also examined the downstream factors of $\gamma$H2AX, including RNF8 and BRCA1. Again, both foci of RNF8 and BRCA1 stayed for an extended period (Fig. 3). Moreover, these results were further validated when we treated cells with another potent selective SMARCA2/4 dual inhibitor, Compound 14 [24] (Fig. 3). Collectively, these results demonstrate that the suppression of the enzymatic activity of SMARCA2/4 impairs the dissociation of some DNA damage response factors from DSBs.

Fig. 3.

Lacking the enzymatic activities of SMARCA2/4 extends the accumulation of $\gamma$H2AX, Ring Finger Protein 8 (RNF8) and Breast cancer susceptibility gene 1 (BRCA1). HeLa cells were pre-treated with or without FHD286 (1 µM) or Compound 14 (1 µM) for two hours, followed by 5 Gy of IR. Following the indicated recovery time, the foci of $\gamma$H2AX (Top panel), RNF8 and BRCA1 were examined by immunofluorescence staining with anti-$\gamma$H2AX, anti-RNF8 or anti-BRCA1 antibodies, respectively. The cells were also counterstained by 4${}^{\prime }$,6-diamidino-2-phenylindole (DAPI). Foci number per cell was analyzed (Bottom panel). p values were calculated using Student’s t-test. “ns” indicates nonsignificant, *p 0.05, ** p 0.01, ***p 0.001. Scale bars: 10 µm.

Upon SMARCA2/4 inhibition, the prolonged accumulation of DNA damage response factors, such as $\gamma$H2AX, indicates the defects in DSB repair. It has been shown that the SWI/SNF complex is involved in HR repair. Here, we found that following IR treatment, SMARCA2 formed nuclear foci and colocalized with RAD51, a key enzyme in HR repair (Fig. 4A). Since SMARCA2/4 are key enzymatic subunits in the SWI/SNF complex, we asked if the ATPase activity of SMARCA2/4 is required for HR repair. In the process of HR repair, RAD51 recognizes ssDNA overhangs at DSB ends and motorizes ssDNA overhangs to invade into sister chromatids. Thus, RAD51 plays a key role during HR repair. Interestingly, when we treated cells with either FHD286 or Compound 14 to suppress both SMARCA2 and SMARCA4, we found that the IR-induced foci of RAD51 were suppressed (Fig. 4A). However, lacking either SMARCA2 or SMARCA4 did not affect the foci formation of RAD51 (Fig. 4B). Moreover, we performed a well-established GFP reporter assay to measure the efficacy of HR [25]. Following the treatment of FHD286 or Compound 14, HR repair was impaired (Fig. 4C). Moreover, we knocked down SMARCA2/4 by siRNA, which also impaired HR repair. However, when we reintroduced siRNA-resistant full-length SMARCA2 but not the enzymatically dead mutant K755A, we could rescue the HR defects (Fig. 4D).

Fig. 4.

Suppression of the enzymatic activities of SMARCA2/4 impairs RAD51-dependent homologous recombination (HR) repair. (A) HeLa cells were pre-treated with or without FHD286 (1 µM) or Compound 14 (1 µM) for two hours, followed by 5 Gy of IR. Foci formation of RAD51 and SMARCA2 was examined by immunofluorescence staining with anti-RAD51 or anti-SMARCA2 antibodies. The cells were also counterstained by DAPI. The number of foci per cell was analyzed. p values were calculated using Student’s t-test. “ns” indicates nonsignificant, ***p 0.001. Scale bars: 10 µm. (B) SMARCA2 or SMARCA4 was knocked out in HeLa cells. IR-induced foci of RAD51 were examined following 5 Gy of IR. The cells were also counterstained by DAPI. The number of foci per cell was analyzed. p values were calculated using Student’s t-test. “ns” indicates nonsignificant. Scale bars: 10 µm. (C) HR efficacy was examined by DR-GFP reporter assays. The Y axis SSC-A (side scattering intensity) indicates the cells of interest gated for GFP fluorescence analysis. For the RAD51 knockdown, the cells with DR-GFP insertion were transfected with negative control siRNA (siRNA-NC) or siRNA-RAD51 for 24 hours following HA-I-SceI-GR pCW tet-on plasmid transfection. For the inhibitor treatment, cells with DR-GFP insertion were pretreated with or without FHD286 (1 µM) or Compound 14 (1 µM) for 2 hours before the acetonide treatment. GFP-positive populations were calculated by flow cytometry. (D) The enzymatic activity of SMARCA2 facilitates HR repair. HR efficacy was examined by DR-GFP reporter assays. The Y axis SSC-A (side scattering intensity) indicates the cells of interest gated for GFP fluorescence analysis. GFP-positive populations were calculated by flow cytometry (bottom right panel). p values were calculated using Student’s t-test. “ns” indicates nonsignificant, and ***p 0.001. For SMARCA2/4 double knockdown, the cells with DR-GFP insertion were transfected with negative control siRNA (siRNA-NC) or siRNA-SMARCA2/4 24 hours before HA-I-SceI-GR pCW tet-on plasmid transfection. For SMARCA2 re-expression, knockdown cells were transfected with full-length siRNA-resistant S protein-FLAG-Streptavidin binding peptide (SFB)-SMARCA2 or the K755A mutant 20 hours before HA-I-SceI-GR pCW tet-on plasmid transfection. Knockdown and/or re-expression were confirmed by Western using indicated antibodies (Top right panel). Vinculin was used as a protein loading control. Figure was created with BioRender.com.

HR-deficient cells are often hypersensitive to PARP inhibitor treatment. To date, PARP inhibitors have been used for the treatment of breast, ovarian, pancreatic, and prostate cancers with HR deficiency. Interestingly, FHD286 is currently in clinical trials for cancer treatment. Since FHD286 treatment impairs HR repair, we ask if FHD286 treatment can sensitize tumor cells to PARP inhibitor treatment. We selected three breast cancer cell lines (HCC1187, MDA-MB-468 and MDA-MB-231) that are not sensitive to PARP inhibitor olaparib. Interestingly, with 50 nM of FHD286, all these cells line were sensitive to 100 nM olaparib treatment (Fig. 5), indicating that FHD286 may have an additive effect with PARP inhibitors for cancer treatment in the future.

Fig. 5.

FHD286 sensitizes tumor cells to the Poly (ADP-ribose) polymerase (PARP) inhibitor treatment. A total of 1000 cells were seeded in each well and treated with indicated doses of FHD286 and/or PARP inhibitor olaparib. Colony formation was examined at Day 14 with 0.1% crystal violet staining. Statistical analyses are shown in the lower panel. p values were calculated using Student’s t-test. “ns” indicates nonsignificant, *p 0.05, **p 0.01, ***p 0.001.