GX15-070 enhances niraparib efficacy in ovarian cancer by promoting a shift in Mcl1-mediated DNA repair pathway from HR to NHEJ

Abstract

Background

PARP inhibitor (PARPi) maintenance therapy significantly extends progression-free survival of patients with homologous recombination repair deficiency (HRD) or BRCA mutations in ovarian cancer. However, more than 50% of patients lack HRD, highlighting the need to expand PARPi use for homologous recombination -proficient patients. In this study, the efficacy of GX15-070 combined with niraparib in ovarian cancer was evaluated.

Methods

Based on the core regulators of genome stability and homologous recombination (HR) repair pathway, a compound library was constructed. The effect of candidate drugs on niraparib sensitivity were measured using CCK-8 in ovarian cancer cell lines. Immunofluorescence and non-homologous end joining repair (NHEJ) assay were conducted to examine HR and NHEJ activity. Co-immunoprecipitation was used to investigate the interaction between Mcl1 and Ku70. BH3 domain deletion mutant of Mcl1 was generated to elucidate the structural basis of the interaction between Mcl1 and Ku70. Additionally, cell line-derived xenograft (CDX) and patient-derived xenograft (PDX) mouse models were established to evaluate the efficacy of GX15-070 combined with niraparib in vivo.

Results

We constructed a compound library based on the core regulators of genomic stability and HR repair. Through high-throughput drug screening, GX15-070, a Mcl1 inhibitor, was identified as a synergist of niraparib, independent of BRCA status. Inhibition of Mcl1 expression significantly impaired HR activity and potentiated niraparib sensitivity. High expression of Mcl1 was associated with a wore prognosis in ovarian cancer patients treating PARPi maintenance therapy. Mechanistically, Mcl1 directly interacts with Ku70 protein via its BH3 domain, serving as a functional switch in selecting between HR and NHEJ. GX15-070 disrupts the interaction by displacing Ku70, promoting a shift in DNA repair pathways from HR to NHEJ. Furthermore, the synergistic efficacy of the combination treatment was further validated in CDX and PDX models.

Conclusions

The study demonstrated that the combination of GX15-070 with niraparib might be a promising therapeutic strategy for ovarian cancer patients with limited PARPi response.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-025-07284-7.

Introduction

Ovarian cancer (OC) is the deadliest gynecological tumor with the third highest incidence rate but the highest mortality rate [1]. In 2024, an estimated 19,680 new cases are projected, representing approximately 2.02% of female cancer diagnoses, with 12,740 deaths accounting for about 4.41% of female cancer-related mortality [2]. The lack of early diagnostic tools often results in advanced stage diagnoses, contributing to poor prognosis and stagnant survival rates over the past decades [3]. The standard first-line treatment for OC, including surgery and platinum-based chemotherapy, is effective initially, but approximately 75% of patients experience recurrence within three years [4]. Recurrent OC poses significant therapeutic challenges, with high-grade serous ovarian cancer (HGSOC), the most prevalent subtype, responsible for nearly 80% of OC-related deaths and a dismal 5-year overall survival rate of 31% [5].

Recently, poly-(ADP)-ribose polymerase inhibition (PARPi) maintenance therapy has gained approval from the U.S. Food and Drug Administration (FDA) for clinical use. Clinical trial results indicate that PARPi maintenance therapy significantly extends overall survival and platinum-free intervals in OC patients, particularly those with homologous recombination deficiency (HRD) [6–8]. Notably, HRD, especially BRCA1/2 mutations, is prevalent in OC [9]. PAPRi exerts its antitumor effects primarily through the mechanism of “synthetic lethality”, wherein HRD or BRCA1/2-deficient cells fail to repair DNA damage induced by PARPi treatment, leading to cytotoxic effects [10]. However, PARPi maintenance therapy faces several clinical limitations. For instance, it shows limited efficacy in patients without HRD [11, 12]. Additionally, most patients will eventually become resistant after receiving PARPi maintenance therapy, due to factors such as low expression of PARP1, restoration of HR and overexpression of ABCB1 [13–15]. Consequently, enhancing PARPi sensitivity remains a critical goal in OC treatment. To date, combination therapy remains one of the most effective strategies to improve drug sensitivity and overcome resistance.

GX15-070 (Obatoclax Mesylate) is a novel BH3 mimetic pan-Bcl-2 inhibitor [16]. It induces apoptosis by binding to the BH3-binding groove of the anti-apoptotic proteins, preventing their interaction with pro-apoptotic proteins Bax and Bak [17, 18]. Besides, GX15-070 can trigger cell death through necroptosis and autophagy [19, 20]. Accumulating preclinical studies have demonstrated its potent antitumor activity in various cancers, including leukemia [21], lymphoma [22], thyroid cancer [23], and pancreatic cancer [24]. Furthermore, GX15-070 enhances the cytotoxicity of multiple anticancer agents, such as 5-fluorouracil, HDACi, and vemurafenib [25–28]. However, its potential to synergistically enhance the antitumor effects of niraparib in ovarian cancer remains unexplored.

In this study, we screened a compound library and identified GX15-070 as a promising candidate for combination therapy with niraparib in OC. GX15-070 synergistically enhanced the antitumor activity of niraparib in preclinical OC models. Mechanistically, GX15-070 promoted a shift in DNA repair pathways from HR to NHEJ via competitively inhibiting the interaction between Mcl1 and Ku70. These findings suggest that the combination of GX15-070 and niraparib represents a potential therapeutic strategy for improving outcomes in ovarian cancer patients.

Materials and methods

Cell culture and drugs

Human ovarian cancer cell lines A2780, SKOV3 and PEO1 cells were purchased from China Center for Type Culture Collection (CCTCC). A2780 and SKOV3 cells were cultured in RPMI 1640 medium (Gibco, Life Technologies), while PEO1 cells were maintained in DMEM high-glucose medium (Gibco, Life Technologies). All media were supplemented with 10% fetal bovine serum (Gibco, Life Technologies), 1% penicillin (100 µg/ml) and streptomycin (100 U/ml). Cells were incubated at 37 ℃ in a humidified atmosphere containing 5% CO₂ [29]. Niraparib (S2741, Selleck), GX15-070/Obatoclax Mesylate (S1057, Selleck) and compound library including 297 drug molecules were purchased from Selleck Biotech Co., Ltd. Niraparib and GX15-070 were reconstituted to stock concentrations of 100 mM and 1 mM, respectively, prior to use.

Compound library drug screening

Cells were seeded in 96-well plates at a density of 2 × 10³ cells. Then, they were treated with DMSO, niraparib (IC50), the screening compounds (10µM) and niraparib, respectively. After 48 h incubation, cell viability was measured by Cell Counting Kit-8 assay. The relative cell viability was calculated based on the niraparib group and the results were analyzed using GraphPad Prism 9 software.

Combination index (CI) assay

Cells were seeded in 96-well cell culture plates at a concentration of 2 × 10³ cells. After 12 h, cells were treated with a series of concentrations (0, IC25, IC50, IC75, IC95) of Niraparib and/or Obatoclax Mesylate, respectively. Subsequently, cell viability was measured by Cell Counting Kit-8 assay. Based on the Chou-Talalay theorem, the CI and fraction affected (FA) values were calculated using Calcusyn software. Survivability plots and CI value scatter plots were generated in GraphPad Prism 9 software [30].

Colony forming assay

Cells were inoculated into 6-well cell culture plates at a density of 5 × 10³ cells and cultured under indicated treatment conditions for 1–2 weeks. The cells were stained with 0.5% crystal violet (Sigma, USA) dissolved in 20% methanol and photographed.

TUNEL assay

After treating with indicated concentrations of niraparib and/or GX15-070 for 48 h, cells were fixed in 4% paraformaldehyde at room temperature and then washed three times with PBS. The cells were permeabilized with 0.1% Triton X-100 and washed twice with PBS. The TUNEL reaction mixture (C11026, Ribobio, China) was prepared according to the manufacturer’s instructions and applied to the samples, which were then incubated at 37 °C for 1 h in the dark. After incubation, the samples were washed three times with PBS. Cells were counterstained with DAPI stain for 5 min. Finally, the cells were observed using a fluorescence microscope and the proportion of TUNEL-positive cells among all cells was calculated.

Annexin Ⅴ/PI assay

Cells were collected after treating with indicated concentrations of niraparib and/or obatoclax mesylate for 48 h. Subsequently, they were stained with Annexin V-FITC and propidium iodide (C11072, Ribobio, China). After incubation in the dark at room temperature for 10 min, the apoptosis rate of cells was measured and then analyzed in a Cytomics FC500 Flow Cytometry System (Beckman Coulter, Krefeld, Germany) [31].

Cell cycle

Cells were harvested and treated, according to cell cycle kit’s instruction (C1052, Beyotime, China) [32]. Briefly, cells were harvested followed by centrifugation in 1000 g for 5 min. After pre-cooling with 70% ethanol, cells were fixed at 4℃ for 2 h and then washed using PBS. Following resuspension in 500 µl staining buffer, propidium iodide (PI, 25 µl), and RNase A (10 µl) were added to cells for incubation 30 min at 37℃ in the dark. Finally, the cell cycle distribution was analyzed by flow cytometry (Beckman Coulter, Krefeld, Germany).

Immunohistochemistry (IHC)

The tissues were fixed overnight in 4% paraformaldehyde and embedded in paraffin. All immunohistochemical (IHC) staining was performed on 4 μm thick sections. Sections were dewaxed, rehydrated, antigenically repaired and endogenous peroxidase blocked. Sections were incubated overnight at 4 °C with primary antibodies targeting Ki67 (28074-1-AP, Proteintech, 1:4000), Mcl1 (16225-1-AP, Proteintech, 1:200), Rad51 (ab133534, Abcam, 1:100), γH2AX (ab11174, Abcam, 1:1000), 53BP1 (ab175933, Abcam, 1:100). Sections were incubated with HRP-conjugated goat anti-rabbit-IgG (AWI0629, Abiowell) for 30 min at 37 °C, followed by staining using an DAB staining kit (AWB0175, Abiowell). Tissue sections were re-stained with hematoxylin, dried and sealed, then observed under a microscope (BA410E, Motic) and scanned using the scanner (Pannoramic MIDI, 3DHISTECH) [33].

H&E staining

Tissue samples were fixed overnight in 4% paraformaldehyde and embedded in paraffin. Sections of 4 μm thickness were prepared, placed on slides, and deparaffinized in xylene, followed by rehydration through a graded ethanol series to water. The slides were stained with hematoxylin for 5–10 min, rinsed in running tap water, and differentiated in acid alcohol if necessary. The sections were then blued in alkaline water and counterstained with eosin for 1–3 min. After staining, the slides were dehydrated through graded alcohol, cleared in xylene, and mounted with a coverslip. The stained sections were observed under a microscope (BA410E, Motic) and scanned using the scanner (Pannoramic MIDI, 3DHISTECH).

Immunoprecipitation (IP)

The interaction between Mcl1 and Ku70 protein was measured using IP kit (abs9649, absin, China). According to the manufacturer’s protocol, total proteins were extracted. Subsequently, protein lysates were immunoprecipitated with primary antibody or IgG negative control and incubated in a rotator overnight at 4 °C, then 10µL of protein A/G beads were added and rotated for 2 h at 4℃. After washing with wash buffer, the immunoprecipitation complexes were separated from the beads for further western blot analysis [34]. The antibodies used were as follows: anti-Ku70 (10723-1-AP, Proteintech, 1:1000), anti-Mcl1 (16225-1-AP, Proteintech, 1:1000).

RT-qPCR

We extracted total RNAs from cells using Trizol reagent and reversely transcribed them into cDNA using PrimeScript™ RT reagent Kit with gDNA Eraser (RR047, Takara). Quantitative real-time PCR was performed using TB Green^® Premix Ex Taq™ (RR420, Takara) and measured using the RocheLightCycler^® 96 instrument. Finally, we analyzed the data using the 2^−ΔΔCt method. The primers were listed in Supplementary Table S1.

Western blot

After 48 h of drug treatment, cells were collected, and proteins were extracted using RIPA lysis buffer. The total protein concentrations were determined using the BCA assay. After subjecting the lysates to SDS-PAGE electrophoresis, proteins were transferred onto a polyvinylidene difluoride membrane by electroblotting, blocked in 5% nonfat milk, and then blotted with anti-bodies targeting Mcl1 (16225-1-AP, Proteintech, 1:2000), Ku70 (10723-1-AP, Proteintech, 1:5000) and β-actin (ab8227, Abcam; 1:1,000). Finally, we quantified the relative levels of individual proteins to β-actin using the ImageJ software (Madison, USA) [35].

siRNA transfection

Cells were seeded at a density of 2 × 10⁵ cells in 6-well plates. After cell culture to 50–70% confluence, siRNA (Ribobio, China) and RNAiMAX (13778150, Invitrogen, USA) were diluted and mixed to form a complex. After 15 min incubation, the complex was added to the cells and the cells were treated at 37 °C, 5% CO₂. Cells were harvested 24–72 h later. Each set of experiments was repeated at least 3 times.

NHEJ assay

The pEJ5-GFP plasmid (Plasmid #44026, Addgene) was transfected into cells, and stable EJ5-GFP-expressing cells were selected with puromycin. Another plasmid pCBASceI (Plasmid #26477, Addgene) expressing the I-SceI endonuclease was then transfected. After treatment with GX15-070, the percentage of GFP-positive cells was assessed using flow cytometry.

Immunofluorescence

Cells were seeded onto 15 mm glass bottom cell culture dishes and collected after being treated for 48 h. Briefly, the glass dishes were washed with PBS, fixed with 4% polyoxymethylene, and permeabilized with 0.1% Triton X-100. Cells were blocked with 5% BSA for 1 h, and stained with anti-RAD51 (ab133534, Abcam, 1:500), anti-BRCA1(22362-1-AP, Proteintech, 1:500), anti-γH2AX (ab11174, Abcam, 1:500), anti-53BP1(ab175933, Abcam, 1:500). Subsequently, they were incubated with Alexa Fluor488-goat anti-rabbit IgG (A11034, Invitrogen, 1:1,000), and nuclear stained with DAPI. The cells were examined under a fluorescent microscope [36].

In vivo studies

Xenograft mouse model experiments were conducted in accordance with a protocol reviewed and approved by the Animal Care Committee of Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of Medicine (No.2024 − 260, Changsha, China). For cell line cell line-derived xenograft (CDX), A2780 or PEO1 cells were collected and resuspended in PBS. 2 × 10⁶ cells in PBS were subcutaneously injected to 5– to 6-week-old female BALB/c nude mice. For patient-derived xenograft (PDX), tumors were obtained from ovarian cancer debulking surgeries carried out in Hunan cancer hospital. Then, the samples were cut into homogenate, suspended with isometric PBS solution mixed with Matrigel. PDX was established by subcutaneously injecting into the lower dorsal flank or axilla of the female c-NKG mice. PDX tumors were expanded into additional mice or cryogenically preserved tumors for future study when the tumor volume was reached about 1000mm³ or if humane endpoints were met [37]. In this study, expanded PDX tumors were subcutaneously implanted into female c-NKG mice as described above for initial tumor inoculations. When the average tumor volume reached approximately 100–200 mm³, mice were divided into four groups and treated with vehicle control, niraparib, GX15-070, or a combination. Niraparib (100 mg/kg) was orally administered five times a week and GX15-070 (6 mg/kg) was administered intraperitoneally three times a week [38, 39]. Body weight and tumor volume were measured every three days. Tumor volume was calculated using the following formula: V (Volume) = L (Length) × W (width)² × 0.5. The mice were treated until day 21 after drug injection, then sacrificed and dissected for tissue collection and tumor weight measurement [40].

Patient samples

Between 2019 and 2021, the EOC patients treating PARPi maintenance therapy at Hunan Cancer Hospital were enrolled. The exclusion criteria are as follows: with incomplete clinicopathological data, without written informed consent, and treated with other targeted drugs except PARPi [38]. Overall, a total of 50 EOC patients were included in this study. The use of clinical samples in this study was approved by the Institutional Review Board of Hunan Cancer Hospital (KYJJ-2023-314, Changsha, China), and written informed consent was obtained from all patients. Platinum-free interval (PFI) was defined as the time from completion of the last platinum-based chemotherapy course to recurrence. Progression free survival (PFS) was defined as the time from patient enrollment to the earliest date of assessment of tumor progression or death. Overall survival (OS) was defined as the time from patients’ enrollment to death from any cause. Kaplan-Meier survival curves were generated to evaluate the prognosis of patients in this study, and survival analysis was performed using the log-rank test.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9. All data were obtained from at least 3 independent replicates. Continuous variables are presented as mean ± SD. Categorical variables are described by frequencies and percentages. Comparisons between groups was performed using the Student‘s t-test, ANOVA, Chi-squared test, or Mann–Whitney Test, based on the data distribution and variable type, with a P-value < 0.05 considered statistically significance.

Results

Drug screening in combination with niraparib

Patients with HRD, especially those with loss of function mutations of BRCA1/2 benefit the most from PARPi maintenance therapy. In the 425 TCGA ovarian cancer samples, 141 patients were identified as HRD phenotype for the biallelic loss of function of the BRCA1/2, and 284 patients were identified as non-HRD phenotype [41]. Besides, the genomic stability indicators of these patients, including the HRD score, HRD-Loss of Heterozygosity (HRD-LOH) score, Large Scale Transitions (LST) score and Number of Telomeric Allelic Imbalances (NtAI) were downloaded from UCSC Xena dataset (https://xena.ucsc.edu/). As expected, they all showed significant increase in patients with HRD phenotype (Fig. 1A-D). Moreover, patients with high HRD score derived significantly more benefit from PARPi when compared to patients with low HRD score [42].These results indicated HRD contributes to genomic instability and leads to sensitivity to PARPi. To screen new candidate drugs enhancing PARPi sensitivity, we firstly identified core regulators of genomic stability and HR repair by using two datasets generated by genome-wide small interfering RNA (siRNA) screenings [43, 44]. Intersection between these two datasets identified 216 common genes (Fig. 1E, Supplementary Table S2). Pathway analysis of these genes highlighted a significant enrichment of spliceosome, Nucleotide excision repair, cGMP − PKG signaling pathway, Proteasome, Hormone signaling, cAMP signaling pathway, RNA polymerase, Ras signaling pathway, Calcium signaling pathway and Neuroactive ligand − receptor interaction (Fig. 1F). Then, we conducted a 297-drug library targeting these pathways and targets to screen potential compounds that might synergistically enhance cellular sensitivity to niraparib in A2780 (BRCA1/2 wild type), SKOV3 (BRCA1/2 wild type) and PEO1 (BRCA2 mutation) cells. Cells in experimental groups were treated with niraparib combined with individual compounds from a 297-drug library. Meanwhile, DMSO and niraparib monotherapy served as solvent control and negative control, respectively (Fig. 1G). The effects of compound in combination with niraparib on cell activity were plotted and shown in Fig. 1H-J. Each dot represents a compound and below the dotted line (slope of 1) indicate compounds that enhance niraparib sensitivity. On the contrary, dots located above the dotted line indicated that combination with the compounds impair the sensitivity of niraparib. Of the 297 compounds, 242 compounds reduced cell viability in A2780, 238 compounds in SKOV3, and 256 compounds in PEO1. A core set of 153 (52%) compounds demonstrated consistent viability reduction across all three cell lines (Fig. 1K, Supplementary Table S3). As shown in Fig. 1H-J, GX15-070 was in the top three of all drugs that enhanced cellular sensitivity to niraparib in A2780, SKOV3 and PEO1 cell lines. Moreover, GX15-070 showed the strongest effect on enhancing cellular sensitivity to niraparib in both SKOV3 and PEO1 cells.

Fig. 1.

GX15-070 is identified as a candidate drug to test in combination with PARPi. The difference of the HRD score (A), HRD-Loss of Heterozygosity (HRD-LOH) (B), Large Scale Transitions (LST) (C) and Number of Telomeric Allelic Imbalances (NtAI) (D) between patients with HRD phenotype and non-HRD phenotype in 425 TCGA ovarian cancer samples. (E) Overlap of regulators of genomic stability and HR repair retrieved from two siRNA screenings. (F) Th enriched pathways were shown using the overlapped genes from Fig. 1E. (G) Outline of the drug screening using a 297-drug library. The effect of GX15-070 combined with niraparib on cell viability in A2780 cells was measured by CCK-8 assay (n = 6) (H), SKOV3 cells (I) and PEO1 cells(J), Each dot in the figure represents a compound. (K) Wayne diagram of screened drugs in three cell lines. ***P < 0.001

GX15-070 combined with niraparib synergistically inhibited ovarian cancer cell proliferation and promoted cell apoptosis

To validate the combinatorial effect of GX15-070 with niraparib on ovarian cancer cell viability, we initially determined the IC50 values of A2780 cells and PEO1 cells against niraparib as well as GX15-070, respectively (Supplementary Figure S1). Subsequently, CI value was calculated to differentiate additive from synergistic effects. Our results showed that both niraparib and GX15-070 concentration-dependently inhibited ovarian cancer cell viability. Compared with the niraparib monotherapy group, cell viability was significantly decreased in the combination group (Fig. 2A). Moreover, the CI values < 1 confirmed synergistic interaction (Fig. 2B). We further examined the effect of the combination of GX15-070 and niraparib on clone formation in ovarian cancer cell lines. The combination of GX15-070 and niraparib significantly reduced the ability of clone formation in ovarian cancer cells (Fig. 2C-D).

Fig. 2.

GX15-070 in combination with niraparib significantly decreases cell growth and increases cell apoptosis in ovarian cancer. (A)The effect of GX15-070 combined with niraparib on the viability of A2780 and PEO1 cells. (B) Combination Index (CI) plot of GX15-070 combined with niraparib in A2780 and PEO1 cells. The x-axis represents the fraction affected (Fa), indicating the extent of drug effect, while the y-axis represents the CI value. Points below 1 (blue) indicate synergism, and points above 1 (green) indicate antagonism. (C-D) Clonal proliferation assay of A2780 cells and PEO1 cells treated with niraparib and/or GX15-070 at the indicated concentrations. (E-F) Apoptosis assay of A2780 cells and PEO1 cells after treatment with GX15-070 and/or niraparib for 48 h at the indicated concentrations. (G-H) TUNEL assay to show apoptosis in A2780 and PEO1 cells treated with GX15-070 and/or niraparib for 48 h at indicated concentrations (Scale bar = 50 μm). Experiments were repeated three times and represented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001

Given the pro-apoptosis effect of GX15-070, we first detected cell apoptosis rate by using flow cytometry. Our results showed that the cell apoptosis rate of GX15-070 (300nM) combined with niraparib (20µM) reached 20% in A2780 cells, representing approximately 3-fold and 6-fold increases over GX15-070 and niraparib monotherapies, respectively. Similarly, the cell apoptosis rate in PEO1 cells was also about 40% with the two drugs combination, which was about 10-fold higher than that of GX15-070 group (100nM) and about 5-fold higher than that of niraparib group (3µM) (Fig. 2E-F). The TUNEL assay further confirmed that apoptosis was significantly increased when both drugs were used together compared to when each drug was used alone (Fig. 2G-H). These findings indicated that the combination of GX15-070 with niraparib enhanced apoptosis in A2780 and PEO1 cells.

GX15-070 enhanced niraparib sensitivity to ovarian cancer by promoting a shift in DNA repair pathways from HR to NHEJ

PARP1 expression has been established as a key determinant of niraparib sensitivity [45]. A previous study demonstrated that Bcl2 inhibits PARP1-dependent DNA repair in diffuse B-cell lymphoma cells, which prompted us to investigate whether GX15-070 enhances niraparib sensitivity by regulating PARP1 expression [46]. Our investigation revealed that GX15-070 treatment did not significantly alter PARP1 protein levels in ovarian cancer cell lines (Supplementary Figure S2A-B). To elucidate alternative mechanisms, we systematically analyzed DNA damage responses and repair pathway dynamics. Using γH2AX foci quantification as a DNA damage marker, we observed significantly increased DNA damage in cells treated with both GX15-070 and niraparib compared to niraparib monotherapy (Fig. 3A-B). Since HR predominantly occurs during the S/G2 phase, we detected the effect of GX15-070 on cell cycle distribution. Compared to niraparib monotherapy, The combination therapy induced pronounced G0/G1 phase arrest, indicating that NHEJ pathway was enhanced, and HR pathway was impaired (Supplementary Figure S3). Moreover, the DNA damage response protein 53BP1 protects DNA ends from excessive resection in G1 and thereby favors repair by NHEJ as opposed to HR [47]. Subsequently, 53BP1 foci was utilized to assay the NHEJ pathway. The results suggested that the combination of GX15-070 and niraparib significantly increased the 53BP1 foci (Fig. 3C-D). Moreover, NHEJ repair efficiency was quantified using a validated reporter system, in which active GFP can be restored in the resulting plasmid only through NHEJ repair (Fig. 3E) [48]. We found that GX15-070 treatment significantly increased the GFP-positive cells, indicating that it can enhance the NHEJ activity (Fig. 3F-G). Concomitantly, the RAD51 foci and BRCA1 foci were utilized to assay the HR activity. The results suggested that the combination of GX15-070 and niraparib significantly decreased the RAD51 and BRCA1 foci numbers (Fig. 3H-K), indicating that the HR pathway was impaired. These results strongly suggested that GX15-070 enhances niraparib sensitivity by promoting a shift in DNA repair pathways from HR to NHEJ.

Fig. 3.

GX15-070 enhances niraparib sensitivity to ovarian cancer by promoting a shift in DNA repair pathways from HR to NHEJ. (A-B) Representative immunofluorescence micrographs and quantification of γH2AX foci in A2780 and PEO1 cells treated with GX15-070 and/or niraparib for 48 h. (C-D) Representative immunofluorescence micrographs and quantification of 53BP1 foci in A2780 and PEO1 cells treated with GX15-070 and/or niraparib for 48 h. (E) Schematic diagram NHEJ reporter systems, the GFP gene can be repaired by NHEJ after I-Scel cleavage, restoring GFP expression. (F-G) Flow cytometry analysis to measure the NHEJ efficiency in A2780 and PEO1 cells transfected with an NHEJ reporter system, either untreated (WT) or treated with GX15-070. (H-I) Representative immunofluorescence micrographs and quantification of BRCA1 foci in A2780 and PEO1 cells treated with GX15-070 and/or niraparib for 48 h. (J-K) Representative immunofluorescence micrographs and quantification of RAD51 foci in A2780 and PEO1 cells treated with GX15-070 and/or niraparib for 48 h. Experiments were repeated three times and represented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001

Mcl1 Inhibition potentiated PARP inhibitor (PARPi) responses in ovarian cancer by impairing HR activity

Mcl1, as a drug target of GX15-070, was identified as a core regulator of HR repair (Supplementary Table S2). To explore whether GX15-070 potentiates PARPi response by regulating Mcl1, we performed GSEA assay using TCGA dataset. The results revealed that Mcl1 expression was positively correlated with HR repair pathway and negatively correlated with DNA double strand break repair and NHEJ repair pathways in ovarian cancer (Fig. 4A). Moreover, previous studies have demonstrated that Mcl1 depletion impairs DNA double strand break repair and reinitiation of stalled DNA replication forks [49, 50]. Therefore, we blocked the endogenous expression of Mcl1 using Mcl1 siRNA in A2780 and PEO1 cell lines (Supplementary Figure S4). Genetic silencing of Mcl1 expression significantly retarded the repair of niraparib-induced DSBs (Fig. 4B-C). Besides, reduced rad51 foci was observed in A2780 and PEO1 cells expressed Mcl1 siRNA versus control siRNA (Fig. 4D-E). These findings suggested that Mcl1 promoted HR activity in ovarian cancer. HRD score is a validated metric for quantifying HRD phenotype in OC [41]. To verify the influence of Mcl1 expression on HR activity, we subsequently downloaded the expression profile data and HRD score (https://xenabrowser.net/datapages/) and assayed the association between Mcl1 expression and HRD score in TCGA dataset. We found that patients with high Mcl1 expression showed a lower HRD score in ovarian cancer (Supplementary Figure S5A). Besides, the association between Mcl1 expression and PARPi response was evaluated using GDSC2 dataset [51]. The results suggested that higher Mcl1 expression is significantly associated with non-responder of niraparib and Olaparib in all cell lines (Supplementary Figure S5B-C). Subsequently, the knockdown of Mcl1 significantly sensitized A2780 and PEO1 cells to niraparib (Fig. 4F). Given that the expression of Mcl1 was correlated with PARPi responder in cell lines, we decided to verify our results through clinical samples. A retrospective cohort of 50 EOC patients treated with PARPi maintenance therapy was enrolled and the clinical characteristics were present in Table 1. Consequently, the Mcl1 protein expression of all patients were evaluated by IHC staining (Fig. 4G). Patients were evenly divided into two groups according to the IHC score, and patients with high expression of Mcl1 were significantly associated with short PFI (HR = 3.03, 95% CI = 1.12–8.17, P = 0.0266), PFS (HR = 3.06, 95% CI = 1.12–8.11, P = 0.0275), and OS (HR = 3.83, 95% CI = 1.39–10.61, P = 0.0064), respectively (Fig. 4H-J). These findings indicated that Mcl1 expression was significantly associated with PARPi response and might be a prognosticator of survival outcome in OC patients.

Fig. 4.

Inhibition of Mcl1 enhances sensitivity to niraparib and impairs HR activity. (A) GSEA analysis using TCGA-OV dataset showed the correlation between Mcl1 expression and HR and NHEJ pathways. (B-C) γH2AX foci was used to detect the influence of Mcl1 expression on repair of DNA damage induced by niraparib in A2780 and PEO1 cell lines. Cells were subjected to Mcl1 knockdown, followed by treatment with or without niraparib for 24 h. After removing niraparib, cells were cultured in normal medium for an additional 24 h. Immunofluorescence staining with γH2AX antibody was performed at different time points. (D-E) Representative immunofluorescence micrographs and quantification of RAD51 foci in A2780 and PEO1 cell lines after Mcl1 knockdown. (F) The effect of niraparib on cell viability in A2780 and PEO1 cell lines after Mcl1 knockdown. (G) Representative IHC micrographs of the Mcl1 protein expression in ovarian cancer tissues (Scale bar = 50 μm, 40 × magnification). (H-J) The influence of Mcl1 expression on clinical outcome in ovarian cancer patients treated with PARPi maintenance therapy. Experiments were repeated three times and represented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001

Table 1.

Clinical characteristics of all patients in this study (n = 50)

Clinical characteristics	All patients (n = 50)	Mcl1high group (n = 25)	Mcl1low group (n = 25)	P value*
Age (years)	52.48 ± 7.45	52.44 ± 5.75	52.52 ± 8.95	0.970
BMI (kg/m2)	23.74 ± 3.55	23.73 ± 4.08	23.744 ± 3.01	0.991
First-line treatment				0.747
Paclitaxel + Platinum	13	6	7
Docetaxel + Platinum	37	19	18
PARPi maintenance therapy				0.758
Niraparib	15	8	7
Olaparib	35	17	18
FIGO stages				0.839
Ⅱ	7	3	4
Ⅲ	31	15	16
Ⅳ	6	3	3
unknown	6	4	2
BRCA mutations				0.361
BRCA+	16	10	6
BRCA−	9	3	6
Unknown	25	12	13

* Mcl1^high group vs. Mcl1^low group

Age: Age at diagnosis; BMI: Body mass index at diagnosis; PARPi: poly-(ADP)-ribose polymerase inhibition; FIGO: International Federation of Gynecology and Obstetrics; BRCA⁺: with pathogenic BRCA1/2 mutation; BRCA⁻: without pathogenic BRCA1/2 mutation

GX15-070 enhances niraparib sensitivity by competitively inhibiting the interaction between Mcl1 BH3 domain and Ku70

In addition, it has been proved that Mcl1 peaks in S/G2 phase and the interaction between H1 and H3 domain of Mcl1 with Ku70 protein promotes DNA excision and HR activity [52]. Therefore, we hypothesized that GX15-070 enhanced niraparib sensitivity by competitively inhibiting the interaction between Mcl1 and Ku70. Firstly, we verified the interaction between Mcl1 and Ku70 by IP assay (Fig. 5A). To verify the binding of Mcl1 BH3 domain to Ku70, we constructed an Mcl1-ΔBH3 (aa213-221) deletion mutant (Fig. 5B). Flag-tagged Mcl1 WT and Mcl1-ΔBH3 deletion mutant were exogenously expressed in PEO1 cells. Subsequently, Co-IP was performed using a Flag antibody. Deletion of the BH3 domain led to a reduction in the interaction between Mcl1 and Ku70 in cells (Fig. 5C). Similarly, pharmacological inhibition of Mcl1 BH3 domain by treating with GX15-070 can also decrease the interaction between Mcl1 and Ku70 (Fig. 5D). To further test whether Mcl1/Ku70 binding influences HR and NHEJ activity, we expressed Mcl1 WT and Mcl1-ΔBH3 deletion mutant in A2780 and PEO1 cells expressing Mcl1 siRNA. As expected, exogenous expression of Mcl1 but not Mcl1-ΔBH3 can salvage the niraparib-induced DSBs (Fig. 5E-F), HR inactivation (Fig. 5G-H), NHEJ activation (Fig. 5I-J) and sensitization to niraparib caused by Mcl1 inhibition (Fig. 5K). These results indicate that Mcl1 BH3 domain is necessary in the choice of DNA repair pathways. GX15-070 enhanced niraparib sensitivity and promoted a shift in DNA repair pathway from HR to NHEJ by competitively inhibiting the interaction between Mcl1 BH3 domain and Ku70.

Fig. 5.

GX15-070 enhances niraparib sensitivity and impairs HR activity by competitively inhibiting the interaction between Mcl1 and Ku70. (A) Co-IP analysis to measure the interaction of KU70 protein and Mcl1 protein in A2780 and PEO1 cells. IgG was used as a negative control. (B) Schematic representation of Mcl1 protein structure showing its PEST sequence, BH domains (BH3, BH1, BH2), and transmembrane (TM) domain. The ΔBH3 mutant lacks the BH3 domain. (C) Co-IP were used analyze the interaction of BH3 domain of Mcl1 and KU70 protein. FLAG antibody was used for IP, and KU70 expression was detected. (D) Co-IP were used to detect the influence of GX15-070 on the interaction of KU70 and Mcl1 protein. Mcl1 was immunoprecipitated and probed for KU70. (E-F) BH3 domain in Mcl1 is required to repair niraparib-induced DSBs. Red dots indicated γH2AX foci in A2780 and PEO1 cells. (G-J) BH3 domain in Mcl1 is required to promote the DNA repair pathway from NHEJ to HR. (K) BH3 domain in Mcl1 played a critical role in the sensitivity of niraparib in A2780 and PEO1 cell lines. Experiments were repeated three times and represented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 (siCtrl vs. siMcl1); ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 (Compared to siMcl1)

GX15-070 combined with niraparib synergistically suppresses tumor growth with minimal toxicity in CDX and PDX ovarian cancer models

To test the synergistic effect of the combination of GX15-070 and niraparib on ovarian cancer in vivo, we subcutaneously injected A2780 and PEO1 cells into nude mice. Mice were randomly assigned to four treatment groups: vehicle control, GX15-070, niraparib, or both. After treatment, both GX15-070 and niraparib monotherapies showed a modest suppression of tumor growth, while the combination of GX15-070 and niraparib treatment remarkably inhibited tumor growth compared with either monotherapy alone (Fig. 6A-B). Furthermore, we also evaluated the treatments in both BRCA^WT and gBRCA1^MUT (BRCA1 c.1504_1508del) PDX models. To ensure the genetic consistency between PDX tumors and parental tumor, the successfully transplanted tumors were only passaged for two generations (Fig. 6C). Our results suggested that the therapeutic effect of combination treatment was significantly greater than niraparib or GX15-070 treatment in both BRCA^WT and gBRCA1^MUT PDX models, and PDX tumors in combination group showed significant growth delay (Fig. 6D). It is worth mention that the xenograft in 1/5 animals treated with the combination in the BRCA^WT PDX model exhibited tumor regression after treating 18 days. To analyze whether the combination treatment suppressed tumor growth by promoting a shift in DNA repair pathways from HR to NHEJ, Ki67, γH2AX, Rad51 and 53BP1were assessed by IHC. Consistent with the in vitro experiments, the combination group showed the lowest Ki67 and Rad51 levels and the most 53BP1 and pan-nuclear γH2AX staining in both CDX and PDX models (Fig. 6E). Meanwhile, the change in the body weight of mice was measured and no significant intergroup differences were observed throughout the length of their treatment (Fig. 7A). Mice in all treatment groups displayed no diarrhea or abnormal food intake throughout the treatment. Moreover, the heart, liver, spleen, lung and kidney tissues of nude mice were harvested and subjected to histological evaluation (Fig. 7B). To further validate the toxicity of drug combination, the biochemical indicators of mice, including aspartate aminotransferase, alanine aminotransferase and creatinine were measured. There was no significant difference between combination group and monotherapy group (Fig. 7C). All this evidence indicated that GX15-070, niraparib or a combination exhibited negligible toxicity. In summary, we demonstrated that GX15-070 in combination with niraparib is an effective anti-tumor treatment strategy with minimal toxicity in vivo ovarian cancer models.

Fig. 6.

GX15-070 combined with niraparib suppresses tumor growth in CDX and PDX models. (A-B) Tumor growth curves and photograph of the tumors harvested at the end of the experiments with the CDX models treated with vehicle, niraparib, GX15-070 or a combination. (C) Experimental design to evaluate combination effect of niraparib and GX15-070 in PDX models. (D) Tumor growth curves and quantification of the tumor weights in the different groups of PDX models. (E) Representative IHC micrographs of indicated markers in the different groups of CDX models and PDX models (Scale bar = 100 μm). (F) Schematic diagram for GX15-070 in enhancing niraparib sensitivity. Briefly, GX15-070 competitively disrupts the interaction between Mcl1 and Ku70, leading to promoting a shift in Mcl1-mediated DNA repair pathway from HR to NHEJ. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 7.

GX15-070 combined with niraparib exhibits negligible toxicity in CDX and PDX models. (A) The change in the body weight of mice in CDX models and PDX models treated with vehicle, niraparib, GX15-070 or a combination. (B) Representative HE micrographs of indicated tissues in the different groups of CDX models and PDX models (Scale bar = 200 μm). (C) Biochemical indicators of mice in the different groups of CDX models and PDX models

Discussion

Despite rapid progress in cancer diagnosis, OC is still recognized as one of the most challenging cancers. It has a high recurrence rate and is one of the most common gynecological malignancies in women [53]. PARPi maintenance therapy, such as olaparib, niraparib and rucaparib, has made remarkable progress in prolonging PFS and has already been approved in various settings for treating relapsed and newly diagnosed OC [11, 12, 54–56]. However, there are several limitations in clinical practice. For instance, the restoration of HR pathway can lead to PARPi resistance and most patients with HR-proficient do not experience significant clinical benefits from PARPi [57]. Combining PARPi with other drugs that induced HRD may not only overcome resistance to PARPi but also potentially expand their use beyond patients with HRD. To date, combination therapy strategies have shown promising outcomes, and some cases have already been implemented to clinical trials [58–60]. Firstly, we identified core regulators of genomic stability and HR repair pathway by using two datasets generated by genome-wide small interfering RNA (siRNA) screenings. After then, enriched pathways were analyzed and a compound library consisting of 297 drug molecules targeting these pathways or targets were constructed. Utilizing the compound library, we screened 153 (52%) drugs that might enhance the effect of niraparib across all three OC cell lines. Among these, GX15-070, disulfiram and Flavopiridol HCL had the greatest impact on reducing A2780 cell viability, and GX15-070, cediranib and Flavopiridol HCL showed the most significant effect on decreasing PEO1 cell viability. Meantime, GX15-070, (+)-JQ1 and pemetrexed had the largest effect on decreasing SKOV3 cell viability. Studies on the combination of cediranib (VEGFR inhibitor) with Olaparib have reported potentiating Olaparib’s efficacy in OC, independent of BRCA status [61]. Moreover, Flavopiridol HCL (CDK inhibitor) and disulfiram can also enhance PARPi sensitivity by inhibiting DNA damage repair [30, 62–64]. Recently, (+)-JQ1 (BET inhibitor) was shown to sensitize tumors to PARPi by decreasing transcription of BRCA1 and RAD51 [65–67]. Therefore, we selected GX15-070 as the drug candidate for further investigation.

GX15-070 (Obatoclax Mesylate) is a novel BH3 mimetic pan-Bcl-2 inhibitor [16]. Clinical trials have been initiated for treating advanced solid tumor or lymphoma patients with GX15-070 alone or combined with other chemotherapies [68–70]. In this study, we demonstrated that GX15-070 significantly potentiates niraparib’s efficacy in OC. The effect of the two-drug combination was evaluated using the CI value, which was developed to quantify the synergistic effect of combined drugs. The mean CI values of GX15-070 combined with niraparib were all below one, which is defined as synergism. It is worth mentioning that both BRCA wild-type (A2780) and BRCA2-mutated (PEO1) ovarian cancer cell lines were strategically utilized in our study. Interestingly, GX15-070 enhanced the activity of niraparib in both cell lines. Strikingly, these in vitro findings were recapitulated in BALB/c nude mouse xenografts. To validate clinical translatability, we established PDX models from ovarian cancer tissue samples of BRCA-mutated and wild-type patients. Consistent with these preclinical results, combination therapy induced superior tumor regression compared to monotherapies in both PDX cohorts. Collectively, these results demonstrated that GX15-070 synergistically enhanced the activity of niraparib independent of BRCA status.

The combination of GX15-070 and niraparib significantly increased cell apoptosis. Moreover, a dramatic increase in γH2AX foci numbers suggested that they work synergistically to cause DNA damage. Further investigations revealed that the combination of GX15-070 with niraparib increased NHEJ activity and decreased HR activity. Mcl1 is a drug target of GX15-070, and it has been identified to regulate HR repair pathway. Furthermore, we verified the role of Mcl1 in regulating HR pathway. Our results showed that genetic inhibition of endogenous Mcl1 reduced HR activity and increased the sensitivity of cancer cells to niraparib. Moreover, Mcl1 expression might serve as a novel biomarker of PARPi response and clinical outcome in OC patients. Overexpression of exogenous Mcl1 but not Mcl1-ΔBH3 can salvage the niraparib-induced DSBs, HR inactivation, and sensitization to niraparib that were caused by Mcl1 inhibition. Importantly, GX15-070 reduces the interaction between Mcl1 and Ku70. Therefore, GX15-070 may target the BH3 domain of Mcl1, competitively disrupting the interaction between Mcl1 and Ku70 and promoting a shift in DNA repair pathways from HR to NHEJ (Fig. 6F). These findings help to elucidate the underlying mechanisms by which the combination of GX15-070 with niraparib exhibits strong synergism against ovarian cancer in vivo and in vitro models.

We acknowledge that there are several limitations in our study. First, GX15-070, serves as a novel BH3 mimetic pan-Bcl-2 inhibitor, is not a specific Mcl1 inhibitor. It may enhance niraparib by regulating the function of other anti-apoptotic proteins (such as Bcl-2) or through Mcl1-related other pathways. It has reported that Bcl-2 suppressed HR repair by directly interacting with Mre11 via BH1 and BH4 domains and localizing BRCA1 to the endomembranes via its transmembrane domain [71, 72]. Additionally, autophagy plays a crucial role in drug resistance, and GX15-070 enhanced Lapatinib toxicity by regulating Mcl1-mediated autophagy [73, 74]. Second, whether the combination can overcome niraparib resistance was not evaluated by using niraparib-resistant cell lines and/or PDX models. All these will be conducted in our future studies. Third, we assessed the toxicity of the combination strategy by measuring mice body weight, standard biochemical indicators and HE staining in vivo study. A more comprehensive toxicological assessment should be performed.

Conclusions

In conclusion, our findings demonstrate that GX15-070 synergistically enhances the activity of niraparib, independent of BRCA status. The direct interaction of Mcl1 with Ku70 via its BH3 domains regulates the HR-mediated DNA damage repair and the response of OC cancer cells to niraparib. GX15-070 displaces Ku70 from Mcl1, resulting in an HRD phenotype and sensitization of cells to niraparib in both in vivo and in vitro models. The combination of GX15-070 with niraparib provides a safe and effective therapeutic strategy to improve the outcome of OC patients. Clearly, more clinical trials assessing safety and efficacy need to be constructed to facilitate the clinical translation of this strategy.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

S.J.J, H.Y, S.H.Y and L.H performed the experiments and analyzed the data. S.J.J, and L.H wrote the manuscript. S.J.J, H.Y, Z.S.A, L.P.W, and W.N.Y.Y contributed to drawing the pictures. S, H.Y, C.Q and L.H supervised the study. All authors revised and approved the final manuscript.

Funding

This research was supported by the Young Scientists Fund of the National Natural Science Foundation of China (NO.82303035), the Natural Science Foundation of Hunan Province (2024JJ5245), the Health Research Project of Hunan Provincial Health Commission (20255941) and High-Level Talant Support Program of Hunan Cancer Hospital (20250731-1035).

Data availability

The data that supporting the findings of this study are available from the corresponding authors upon reasonable request.

Declarations

Ethics approval and consent to participate

All animal experiments were performed in accordance with the Guidelines for the Use and Care of Laboratory Animals of the National Institutes of Health and were approved by the Institutional Animal Care and Use Committee (IACUC) of Hunan Cancer Hospital and the Affiliated Cancer Hospital of Xiangya School of Medicine.

Consent for publication

All authors have agreed to the content of the manuscript and agree to this submission.

Competing interests

The authors declare that they have no competing interests.