The RPA inhibitor HAMNO sensitizes Fanconi anemia pathway-deficient cells

ABSTRACT

The Fanconi anemia (FA) DNA repair pathway is required for DNA inter-strand crosslink (ICL) repair. Besides its role in ICL repair, FA proteins play a central role in stabilizing stalled replication forks, thereby ensuring genome integrity. We previously demonstrated that depletion of replication protein A (RPA) induces the activation of FA pathway leading to FANCD2 monoubiquitination and FANCD2 foci formation. Thus, we speculated that FA-deficient cells would be more sensitive to RPA inhibition compared to FA-proficient cells. Following treatment with RPA inhibitor HAMNO, we observed significant induction in FANCD2 monoubiquitination and foci formation as observed in RPA depletion. In addition, HAMNO treatment caused increased levels of γ-H2AX and S-phase accumulation in FA-deficient cells. Importantly, FA-deficient cells showed more increased sensitivity to HAMNO than FA-proficient cells. Moreover, in combination with cisplatin, HAMNO further enhanced the cytotoxicity of cisplatin in FA-deficient cells, while being less toxic against FA-proficient cells. This result suggests that RPA inhibition might be a potential therapeutic candidate for the treatment of FA pathway-deficient tumors.

Introduction

Fanconi anemia (FA) is a recessive genetic disorder caused by mutations in at least one of 22 FA genes (FANCA-FANCW) that are essential for DNA interstrand crosslinks (ICLs) repair [1–3]. The 22 FA proteins participate in a common pathway [3–6]. When the DNA replication forks stall at the DNA ICLs, an upstream FA core complex activates the pathway by FANCI-FANCD2 (ID2) monoubiquitination, followed by their recruitment to chromatin [7–11]. Chromatin-bound FANCD2/FANCI forms DNA repair foci with other DNA repair proteins and initiates downstream double-strand break (DSB) repair [7,12,13]. Disruption of any one of 22 FA genes leads to the characteristic cross-linker hypersensitivity and chromosome instability [3,14]. Besides its role in ICL repair, FA proteins are known to stabilize and recover stalled replication forks during S phase [15–19], suggesting that FA pathway plays an important role as an essential barrier under conditions of replication stress [3,20].

Replication protein A, the main eukaryotic single-stranded DNA (ssDNA) binding protein, is a protein complex composed of three tightly associated subunits of ~70, 32, and 14 kDa [21,22]. In addition to the essential role of RPA in DNA replication, RPA is involved in DNA damage response (DDR), such as cell cycle checkpoints and DNA repair [23]. In response to replication stress, ATM- and Rad3-related kinase (ATR) and ATR-interacting protein (ATRIP) are recruited to a stalled replication fork by RPA-coated ssDNA to initiate DDR [24–26]. Given RPA’s critical role in DDR, RPA inhibition could be very advantageous to boosting the benefit of DNA-damaging chemotherapeutics.

Over the past few years, several RPA inhibitors have been identified [27–30]. TDRL-505 and its derivatives inhibit the ssDNA-binding activity of RPA and display cytotoxicity both as a single agent and in combination with other DNA damaging agents [29,30]. HAMNO is a novel protein interaction inhibitor of RPA [28]. HAMNO prevents RPA70ʹs interaction with ATR/ATRIP and inhibits the checkpoint activation in response to replication stress. HAMNO selectively increases γ-H2AX staining in S phase, indicative of increased replication stress [28].

We have previously shown that the depletion of RPA induces the activation of FA pathway leading to FANCD2 monoubiquitination (FANCD2-Ub) and foci formation [31], indicating that RPA depletion resulted in DNA damage signaling and activates FA pathway for the maintenance of genome stability. We therefore hypothesized that FA-deficient cells with defective FANCD2-Ub may be hypersensitive to RPA inhibition and exhibit selective cytotoxicity compared to FA-proficient cells. Indeed, our study provided the evidence that RPA inhibition by HAMNO significantly impairs maintenance and survival of FA-deficient cells as a single agent. Moreover, in combination with cisplatin, HAMNO further enhanced the cytotoxicity of cisplatin in FA-deficient cells. Taken together, our results suggest RPA inhibition as a potential therapeutic strategy to target FA pathway-deficient tumors.

Materials and methods

Cell culture and chemicals

SV40-transformed FA fibroblasts (GM6914, FANCA-deficient cells and FUFA326, FANCG-deficient cells) and HeLa and U2OS cells were grown in DMEM supplemented with 10% FBS. HAMNO, mitomycin C (MMC), and cisplatin were purchased from Sigma-Aldrich.

Antibodies

Antibodies used in this study were as follows: anti-FANCD2 and Vinculin antibodies (Santa Cruz Biotechnology, Santa Cruz, CA), anti-α-tubulin and β-actin antibodies (Sigma Aldrich, St. Louis, MO, USA), anti-CHK2 T68 and CHK1S345 antibodies (Cell Signaling Technology, Inc., MA, USA), anti-RPA1 antibody (EMD Millipore, Temecula, CA 92590, USA), and anti-γ-H2AX antibody (Upstate Biotechnology, Lake Placid, NY, USA). Antibody against FANCA was kindly provided by the Fanconi Anemia Research Fund (www.fanconi.org).

Immunofluorescence

Cells grown on the coverslips were pre-extracted with 0.2% Triton X-100 for 1 min, followed by fixation with 3.7% paraformaldehyde for 10 min and permeabilized with PBS + 0.5% Triton X-100 for 5 min. Cells were blocked with 5% BSA/PBS and were incubated with anti-FANCD2 (1:100) and anti-γ-H2AX (1:2000) for 2 h at room temperature or overnight at 4°C. Alexa 488 and Alexa 568-conjugated secondary antibodies (Invitrogen) were used at a dilution of 1:500 for 1 h at room temperature. After counter-staining with 4’,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich), coverslips were mounted with ProLong Gold antifade (Thermo Fisher Scientific). Cells were then imaged with an inverted confocal laser scanning microscope (LSM700; Carl Zeiss, Oberkochen, Germany).

Cell-cycle analysis

Samples were collected over indicated time points and fixed in 70% ethanol overnight. For cell cycle analysis, fixed cells were treated with RNase for 20 min before addition of 5 μg/mL propidium iodide (Sigma-Aldrich). Flow cytometric analysis was then performed using a FACSCalibur flow cytometer (Becton Dickinson). The percentages of cells belonging to each phase of the cell cycle (G1, S, and G2/M) were calculated by the ModFit LT DNA analysis software.

Cell survival assay

Cells were plated in 96 well plates (2000 cells/well) in triplicate. After 1 day, HAMNO was added in cultures in a serial dilution of 0–100 μM and cells were incubated for 4 days. After 4 days, cell viability was determined by EZ-Cytox cell Viability Assay kit (DoGen). The absorbance measurements at 450 nm were made using a microplate reader. For quantitation, readings of absorbance at 450 nm were normalized to those obtained from untreated cells, assumed to yield 100% cell survival.

Statistical analysis

All data are presented as mean ± SEM. Statistical significance between two data sets was compared by Student’s two-tailed t-test.

Results

HAMNO induces FANCD2 monoubiquitination and checkpoint activation

We have previously shown that RPA1 depletion induces FANCD2-Ub and checkpoint activation [31]. We therefore hypothesized that FA-deficient cells with defective FANCD2-Ub may be more sensitive to RPA inhibition and exhibit selective cytotoxicity compared to FA-proficient cells. HAMNO, a specific RPA inhibitor, targets the N terminal domain of RPA and inhibits RPA interactions with proteins involved in the DDR [28]. We first tested whether RPA inhibition by HAMNO induces FA pathway activation as observed in RPA depletion (Figure 1). HeLa and U2OS cells were treated with HAMNO and analyzed by immunoblotting (Figure 1(a)). The double bands of the FANCD2 blot correspond to FANCD2-Ub (monoubiquitinated FANCD2; upper band) and FANCD2 (non-monoubiquitinated FANCD2; lower band), respectively. Similar to RPA depletion (Supplementary Figure S1) [31], HAMNO treatment resulted in significant induction of FANCD2-Ub. In addition, elevated FANCD2-Ub was related to increases in phosphorylation of CHK1 (S345), CHK2 (T68), and H2AX (γ-H2AX), indicating that DNA damage signaling is activated following HAMNO treatment. We next examined FANCD2 nuclear foci formation (Figure 1(b,c) and Supplementary Figure S2). The mock-treated control cells displayed only weak FANCD2 immunoreactivity. In contrast, HAMNO treatment significantly increased the frequency of cells with FANCD2 foci (>10 FANCD2 foci per cell) after 4 h incubation. Under this condition, the cell cycle profile of HAMNO-treated cells was similar to that in untreated cells (Supplementary Figure S3), suggesting that the effect of HAMNO on FANCD2 foci formation is not mediated by cell cycle change. Given that FANCD2 foci are present only in S-phase cells [13,32], these results suggest that RPA inhibition by HAMNO may increase DNA replication stress during S phase, which induces DNA damage signaling leading to FA pathway activation.

Figure 1.

HAMNO elevates FANCD2-Ub levels and activates checkpoints.

(a) HeLa and U2OS cells were treated with either solvent control [0.1% (v/v) DMSO] or the indicated concentrations of HAMNO for 2 hours. Whole-cell lysates were prepared and immunoblotted using indicated antibodies. (b) U2OS cells were treated with 0.1% DMSO or HAMNO (50 µM) for 4 hours. Cells were immunostained with antibody for FANCD2 (green). Representative confocal microscopy images are shown. Scale bars, 10 µm. Magnified images of the area indicated by the dashed boxes are shown below. Scale bars, 5 µm. (c) Quantification of FANCD2 foci shown in Figure 1(b). Nuclei with >10 FANCD2 foci were scored as positive. As a positive control for FANCD2 foci formation, U2OS cells were treated overnight with MMC (50 ng/ml). Values represent the mean ± SEM, examined at least 200 nuclei each in three independent experiments. Statistical significance between two data sets were compared by Student’s two-tailed t-test, where ***P<0.001

HAMNO induces FANCD2 foci formation colocalized with γ-H2AX

We next examined the cellular effect of HAMNO in FA-deficient cells (Figure 2). We analyzed FANCA-deficient parental GM6914 cells (FA-A) and the GM6914 cells corrected with the FANCA cDNA (FA-A+FANCA). Consistent with Figure 1, HAMNO treatment showed significant induction of FANCD2-Ub in FA-A+FANCA cells, but not in FA-A cells. In addition, HAMNO treatment induced increased levels of CHK1 S345, CHK2 T68, and γ-H2AX in both cell lines (Figure 2(a)). Although significant differences in the level of checkpoint activation between FA-A and FA-A+FANCA cells was not observed, slightly increased levels of CHK2 T68 and γ-H2AX were consistently observed in FA-A cells. We next examined the nuclear foci formation of FANCD2 and γ-H2AX (Figure 2(b-e) and Supplementary Figure S4). Cells treated with HAMNO exhibit variable levels of γ-H2AX staining, ranging from whole-nucleus staining (non-homogeneous pan-nuclear γ-H2AX staining) to punctate staining (discrete γ-H2AX foci) (Figure 2(b,c)). Consistent with western blot data in Figure 2(a), the percentage of cells containing γ-H2AX foci was significantly higher in FA-A cells than FA-A+FANCA cells after HAMNO treatment (Figure 2(e)), indicating the existence of increased DNA damage in FA-A cells. Furthermore, the majority of FANCD2 foci in FA-A+FANCA cells colocalized with γ-H2AX foci after HAMNO treatment (Figure 2(c)). Given that FANCD2 foci formation occurs only in S phase [13,32], our data suggest that FANCD2-Ub interacts with γ-H2AX at stalled replication forks during S phase and may play a role in maintaining genome stability.

Figure 2.

HAMNO induces FANCD2 foci formation colocalized with γ-H2AX.

(a) FANCA deficient (FA-A) and isogenic FANCA-corrected (FA-A+FANCA) cells were treated with either solvent control [0.1% (v/v) DMSO] or the indicated concentrations of HAMNO for 2 hours. Whole-cell lysates were prepared and immunoblotted using indicated antibodies. (b-c) FA-A cells (b) and FA-A+FANCA cells (c) were treated with either solvent control [0.1% (v/v) DMSO] or 50 µM HAMNO for 4 hours. Cells were co-immunostained with antibodies for FANCD2 (green) and γ-H2AX (red). Representative confocal microscopy images are shown. Scale bars, 10 µm. Magnified images (i-iv) of the area indicated by the dashed boxes are shown below. Scale bars, 5 µm. (d-e) Quantification of FANCD2 foci (d) and γ-H2AX foci (e) shown in (b) and (c). For FANCD2 foci, nuclei with >10 foci were scored as positive. For γ-H2AX foci, nuclei with “bright” γ-H2AX staining were scored as positive. Values represent the mean ± SEM, examined at least 200 nuclei each in three independent experiments. Statistical significance between two data sets were compared by Student’s two-tailed t-test, where **P<0.01, ***P<0.001.

FA-deficient cells are more sensitive to HAMNO

To test whether FA-deficient cells are more sensitive to HAMNO, FA-A and FA-A+FANCA cells were incubated with increasing concentrations of HAMNO for 4 days, and cell viability was compared (Figure 3(a)). Notably, the FA-A cells were more sensitive to HAMNO compared with FA-A+FANCA cells, although the differences appeared modest. Consistent with this finding, we also obtained similar results in FANCG-deficient human fibroblast (FA-G) cells and isogenic FANCG-corrected FA-G cells (FA-G+FANCG) (Supplementary Figure S5). These results suggest that HAMNO can display single-agent cytotoxicity in FA-deficient cells. To further understand the mechanism of differential sensitivity to HAMNO in FA-deficient cells, we analyzed γ-H2AX levels and cell cycle distribution after HAMNO treatment (Figure 3(b,c)). FA-A cells and FA-A+FANCA cells were incubated with HAMNO for 6, 12 and 24 h then assessed DNA damage by measuring γ-H2AX levels (Figure 3(b)). The increasing incubation times led to progressively higher levels of γ-H2AX in FA-A cells, while FA-A+FANCA cells showed lower levels of γ-H2AX that remained largely unchanged throughout the 24 h. Moreover, incubation with HAMNO for 24 h resulted in an increase in the percentage of S phase cells (untreated, 34.8% vs. HAMNO, 53.9%) in FA-A cells, consistent with the levels of γ-H2AX. In contrast, FA-A+FANCA cells showed an increase in G1 phase cells (untreated, 42.7% vs. HAMNO, 51.9%) and a concomitant decrease in S phase cells (untreated, 44.6% vs. HAMNO, 36.3%), indicating a delay in the G1/S transition and S phase progression (Figure 3(c)). This data coupled with western blot analysis suggest that the FA-deficient cells accumulate higher levels of DNA damage during S phase than FA-A+FANCA cells, leading to reduced survival rates.

Figure 3.

FA-deficient cells show increased sensitivity to HAMNO.

(a) FA-A cells and FA-A+FANCA cells were treated with increasing concentrations of 2.5-100 µM HAMNO for 4 days, and cell viability was determined by MTT assay. Survival rates are plotted as the percentage of viable cells relative to that for respective untreated cells. Survival assay were performed in triplicate and repeated N=3 times. The results are presented as mean ± SEM. Statistical significance between two data sets were compared by Student’s two-tailed t-test, where **P<0.01, ***P<0.001. (b) FA-A cells and FA-A+FANCA cells were treated with either solvent control [0.1% (v/v) DMSO] or 20 µM HAMNO. Whole-cell lysates were prepared at 6h, 12h and 24h after incubation and analyzed with indicated antibodies. (c) FA-A cells and FA-A+FANCA cells were treated with either solvent control [0.1% (v/v) DMSO] or treated with HAMNO (20 µM) for 24h, and cell cycle distribution was analyzed by flow cytometry. Representative histograms for cell cycle distribution are presented. The percentages of cells belonging to each phase of the cell cycle (G1, S and G2/M) were calculated by the ModFit LT DNA analysis software.

HAMNO increases the cytotoxicity of cisplatin in FA-deficient cells

Somatic inactivation of the FA pathway has been observed in several different types of sporadic cancer [14,33]. Although FA-deficient tumors display enhanced sensitivity to cisplatin [34], combination therapies of cisplatin with other drugs have been highly considered to overcome drug-resistance and reduce toxicity. Considering RPA’s role in DDR [23], inhibition of RPA activity would synergistically increase DNA damage following treatment with cisplatin in FA-deficient cells. As shown in Figure 3(a) and Supplementary Figure S5, we observed modestly increased sensitivity to HAMNO in FA-deficient cells, probably due to accumulation of spontaneous DNA damage. We next investigated whether HAMNO treatment further increases cisplatin sensitivity in FA-deficient cells (Figure 4). As expected, FA-A cells are more sensitive to cisplatin treatment compared to FA-A+FANCA cells (Figure 4(a)). To investigate the effect of a combination of HAMNO and cisplatin, 0.2 μM cisplatin, which minimally affects cell viability of FA-A cells, was added to 20 μM HAMNO. In response to either HAMNO (FA-A, 64.8% vs. FA-A+FANCA, 90.5%) or cisplatin (FA-A, 82.9% vs. FA-A+FANCA, 95.1%) alone, the FA-A cells exhibited modestly reduced cell viability compared to FA-A+FANCA cells (Figure 4(b)). When subjected to a combined HAMNO and cisplatin, this differential sensitivity was more prominent (FA-A, 41% vs. FA-A+FANCA, 84.8%). A similar result was also observed in FA-G cells and FA-G+FANCG cells (FA-G, 42.2% vs. FA-G+FANCG, 79.6%) (Figure 4(c)). These results demonstrate that HAMNO increases the cytotoxicity of cisplatin in FA-deficient cells, while being less toxic toward FA-proficient cells, suggesting RPA inhibitor as a new therapeutic strategy to treat FA-deficient tumors.

Figure 4.

HAMNO selectively increases cisplatin sensitivity in FA-deficient cells.

(a) FA-A cells and FA-A+FANCA cells were treated with increasing concentrations of 0.1-2 µmM cisplatin and incubated for 4 days, and cell viability was determined by MTT assay. Survival rates are plotted as the percentage of viable cells relative to that for respective untreated cells. Survival assay were performed in triplicate and repeated N=3 times. The results are presented as mean ± SEM. Statistical significance between two data sets were compared by Student’s two-tailed t-test, where **P<0.01, ***P<0.001. (b-c) FA-A cells & FA-A+FANCA cells (b) and FA-G &FA-G+FANCG cells (c) were treated with either solvent control [0.1% (v/v) DMSO] or 20 µmM HAMNO alone, 0.2 µmM cisplatin alone, and 0.2 µmM cisplatin+20 µmM HAMNO for 4 days, and cell viability was determined by MTT assay. Survival rates are plotted as the percentage of viable cells relative to that for respective untreated cells. Survival assay were performed in triplicate and repeated N=3 times. The results are presented as mean ± SEM. Statistical significance between two data sets were compared by Student’s two-tailed t-test, where **P<0.01, ***P<0.001. (d) A proposed model of how RPA inhibitor induces selective cytotoxicity in FA-deficient cells. See discussion for details.

Discussion

The FA pathway is important for the repair of ICLs that cause replication fork stalling during S phase [1,2]. In addition to its role in ICL repair, FA proteins play a central role in stabilizing stalled replication forks, thereby ensuring genome integrity [15–19]. RPA is a nuclear ssDNA-binding protein complex and is required for many cellular pathways including DNA replication, recombination and repair [21–23]. Previously, we have demonstrated that RPA depletion induces DNA damage checkpoint, leading to the activation of FA pathway [31]. We therefore hypothesized that FA-deficient cells with defective FANCD2-Ub may be more sensitive to RPA inhibition. In this study, we examined the cellular effect of a specific RPA inhibitor HAMNO in FA-deficient cells. Our results showed that RPA1 inhibition by HAMNO induced the activation of FA pathway, resulting in increased FANCD2-Ub and foci formation (Figures 1 and 2). Importantly, FA-deficient cells displayed significantly increased sensitivity to HAMNO compared with FA-proficient cells (Figure 3 and Supplementary Figure S5). Furthermore, combination of cisplatin and HAMNO resulted in enhanced sensitivity to cisplatin in FA-deficient cells (Figure 4). Taken together, these results have important implications for the use of HAMNO in cancer therapy that could effectively target FA-deficient tumors.

ATR is one of the central replication stress response kinases, and once activated at stalled replication forks, it phosphorylates multiple substrates which help the cell to survive under conditions of replication stress [23–26]. HAMNO is a selective inhibitor of RPA interactions with proteins responsible for ATR activation [28]. HAMNO prevents RPA70ʹs interaction with ATR/ATRIP and inhibits ATR-dependent DNA damage signaling, thereby increasing replication stress during S phase [28]. Importantly, there is increasing evidence that the FA pathway contributes to the maintenance of genome integrity by protecting DNA replication forks in the presence of replication stress [15–19,35–37]. It has been shown that increased chromosomal instability observed in FA-deficient cells experiencing replication stress is attributed to impaired stabilization of the stalled replication forks [38–41]. HAMNO exhibited modestly increased sensitivity in FA-deficient cells (Figure 3(a) and Supplementary Figure S5). The cytotoxicity of HAMNO in FA-deficient cells as a single agent could be due to increased replication stress when FA deficiency is combined with RPA inhibition (Figure 4(d)). Furthermore, in combination with cisplatin, HAMNO significantly enhanced cytotoxicity of cisplatin in FA-deficient cells (Figure 4(b,c)). There are at least two possibilities that could be considered for the combined effect of HAMNO and cisplatin. First, DNA damage induced by ICLs activates two parallel branches of S-phase checkpoint signaling: one is represented by the ATR-CHK1 and the other by ATR-NBS1-FANCD2 (FANCD2-Ub independent), which cooperate to attain the full activation of the S-phase checkpoint to allow an efficient DNA repair [42,43]. Given that S-phase checkpoint and DNA repair, respectively, contribute to cisplatin resistance, it is possible that HAMNO may further enhance cytotoxicity of cisplatin by inhibiting ICL-induced S phase checkpoint in FA-deficient cells. Second, while replication-dependent ICL repair requires FA pathway, ICLs are also repaired through transcription-coupled nucleotide excision repair (TC-NER) pathway in non S-phase cells [44,45], which is independent of the FA pathway. Considering critical role of RPA in NER pathway, it is possible that HAMNO may suppress TC-NER outside of S phase, leading to increased sensitivity of cisplatin in FA-deficient cells.

Recently, DDR inhibitors are being tested in various trials with promising results [46–48]. DDR inhibitors in cancers with DNA repair defects could provide selective toxicity, while other cells without pathway defects are unaffected. Previous studies demonstrate that FA-deficient cells are sensitive to either ATM inhibitor or CHK1 inhibitor, key kinases in DDR, compared to FA proficient cells [49,50]. It would be interesting whether RPA inhibitor combined with either ATM inhibitor or CHK inhibitor produces synergistic effects in sensitizing FA-deficient cells. Moreover, inhibitors of the FA pathway have been identified to enhance effectiveness of platinum therapy [3,51,52]. Thus, it would be interesting to test whether chemo-sensitizing effect of FA inhibitors is further potentiated by combination with RPA inhibitors. The development of RPA targeting inhibitors is still at an early stage and there is ongoing interest in seeking to identify more specific inhibitors that display defects in DNA repair without impacting DNA replication. Given that RPA is essential for all DNA repair pathways including base excision repair, nucleotide excision repair, mismatch repair, and double-stranded break repair [21–23], it will be important to select cancer groups that can receive the maximum benefits of RPA inhibitors for cancer therapy.

In conclusion, we demonstrate that the RPA inhibition in FA-deficient cells enhanced cytotoxic effect as a single agent and in combination with cisplatin, suggesting RPA inhibitor as a new therapeutic strategy to treat FA-deficient tumors.

Funding Statement

This work was supported by the National Research Foundation of Korea [2020R1F1A1-068668].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed here.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials