GnRH-R–Targeted Lytic Peptide Sensitizes BRCA Wild-type Ovarian Cancer to PARP Inhibition

Gonadotropin-releasing hormone (GnRH; also known as luteinizing hormone–releasing hormone) is a hypothalamic neuropeptide that plays important roles in the reproductive system. Researchers have identified three isoforms of GnRH (GnRH1, GnRH2, and GnRH3). Of these isoforms, only GnRH1 and GnRH2 are expressed in human tissues (1). By binding to its receptor (GnRH-R) on pituitary gonadotropic cells, GnRH can mediate the gonadal steroid system by stimulating the release of luteinizing hormone and follicle-stimulating hormone (2). GnRH-R is a member of the rhodopsin-like G protein–coupled receptor family and can couple with Gaq11 protein upon hormone stimulation (3). Currently, two types of GnRH-R are known to exist in primates, with only a functional type I GnRH-R existing in human tissues (4). Studies have demonstrated that GnRH-R is overexpressed in many human tumors (e.g., breast, ovarian, endometrial, and prostate cancers), whereas it is not expressed or expressed at very low levels in adjacent normal tissues (5–7). Accumulating evidence has demonstrated that more than 80% of ovarian and endometrial cancers, as well as more than 50% of breast cancers, have high levels of GnRH-R expression (8–11). In addition, unlike the activation of protein kinase C in pituitary cells, the signaling pathway activated upon stimulation of GnRH-R in cancer cells is mainly the mitogenic signal transduction pathway or tyrosine kinase signaling pathway (12, 13). Thus, its unique distribution pattern and specific signal transduction qualify GnRH-R as a diagnostic marker as well as a potential molecular target for cancer therapy. Investigators have attempted to develop GnRH agonists and antagonists for the treatment of both hormone-dependent (e.g., ovarian, breast, and prostate cancers) and hormone-independent (e.g., bladder cancer) tumors either through suppressing the pituitary–gonadal axis or delivering targeted therapy (14).

Cationic lytic peptides have been tested as cancer therapeutics and can function by disrupting tumor cell membranes, inducing apoptosis, or leading to necrotic cell death (15, 16). Recently, many promising cytolytic peptides have emerged, such as melittin, apamin, and mastoparan (17). However, the disadvantage of lytic peptide–based reagents is that they execute their functions in a nonspecific manner, resulting in severe adverse events. Therefore, modification of lytic peptides that specifically target tumor cells is needed. EP-100 (developed by Esperance Pharmaceuticals, Inc.) is a fusion peptide consisting of the GnRH natural ligand joined to an 18-amino-acid cationic α-helical lytic peptide (CLIP-71) developed to deliver lytic peptides to cancer cells by targeting GnRH-R (18). Leuschner and colleagues (19) showed that EP-100 can interact with a negatively charged tumor cell membrane and cause cell death through membrane lysis within a few minutes. Also, preclinical studies demonstrated that EP-100 had an antitumor effect in a variety of human cancer cell lines that overexpress GnRH-R alone or in combination with paclitaxel (20). A phase I study tested EP-100 in many human tumors (including breast, ovarian, endometrial, pancreatic, prostate, and carcinoid cancers and non-Hodgkin lymphoma) and demonstrated that EP-100 is a safe, well-tolerated drug (18).

Previous study suggested that the clearance of EP-100 is rapid (mean half-life 7.1 ± 3.8 to 15.9 ± 3.6 min), which necessitated longer duration of intravenous infusion (18). Therefore, strategies to expand the utility of EP-100 are needed. In this study, we aimed to identify new combination therapeutic approaches with EP-100 in ovarian cancer models. We provide evidence of a synergistic effect of EP-100 with the PARP inhibitor olaparib in preclinical models of ovarian cancer both in vitro and in vivo, suggesting that future clinical studies of this combination are warranted.

The ovarian cancer cell lines A2780ip2, A2780cp20, HeyA8, HeyA8-MDR, OVCAR 3, OVCAR 4, OVCAR 8, and OVCAR 432 were maintained in RPMI1640 supplemented with 10%–15% FBS and 0.1% gentamicin sulfate (Gemini Bioproducts). OVCAR 5 cells were maintained in DMEM with 10% FBS and 0.1% gentamicin sulfate. All above cells were cultured at 37°C using a 5% CO₂ incubator. The above cell lines were obtained from ATCC or The University of Texas MD Anderson Cancer Center Characterized Cell Line Core Facility. OVCAR 432 cell line was provided by Dr. Ronny Drapkin (Dana-Farber/Harvard Cancer Center, Boston, MA). Short tandem repeat (STR) DNA profiling was performed by Characterized Cell Line Core Facility, The University of Texas MD Anderson Cancer Center (Houston, TX). Mycoplasma testing was performed using the ATCC Universal Mycoplasma Detection Kit. BRCA1-mutant ovarian cancer cell line COV362 was purchased from Sigma Aldrich and cultured in DMEM supplemented with 10% FBS and 0.1% gentamicin sulfate and 2 mmol/L glutamine (Thermo Fisher Scientific). BRCA1-mutant breast cancer cell line MDA-MB-436 was purchased from ATCC and maintained in Leibovitz L-15 medium (Sigma Aldrich) with 10 mcg/mL insulin, 16 mcg/mL glutathione, 0.1% gentamicin sulfate, and 10% of FBS and cultured in a free gas exchange with atmospheric air. BRCA2-mutant ovarian cancer cell line KURAMOCHI was obtained from Japanese Collection of Research Bioresources and cultured in RPMI1640 supplemented with 10% FBS and 0.1% gentamicin sulfate. COV362 cells, MDA-MB-436 cells, and KURAMOCHI cells were purchased in September 2018. All in vitro experiments were conducted with 60%–80% confluent cultures and a passage number below 20.

HeyA8 and A2780ip2 cells with stable knockdown of GnRH-R expression (shGnRH-R cells) and respective control cells (shControl cells) were generated via lentiviral transfection. Plasmids were obtained from Sigma-Aldrich in bacterial stock and extracted using a QIAGEN Plasmid DNA Purification Kit. Sequence for shControl (5′-3′): CCGGCCTAAGGTTAAGTCGCCCTCGCTCGAGCGAGGGCGACTTAACC-TTAGGTTTTTG; shGnRH-R (5′-3′): CCGGCCAATGGTATGCTGGAGAGTTCTCGAGAACT-CTCCAGCATACCATTGGTTTTT. Briefly, viral particles containing lentiviral shGnRH-R plasmids were generated by infecting 293T cells after 48 hours, and a supernatant containing viral particles was collected and filtered. HeyA8 and A2780ip2 cells were plated in 6-well plates for 24 hours. The viral particle solution and transfection reagents were mixed in 2 mL of serum-free medium and added drop-wise over the cells. Serum-containing medium was added to the cells 24 hours after transfection. Cells were selected by adding puromycin (Thermo Fisher Scientific). The knockdown of GnRH-R expression was confirmed via Western blotting.

Cells were harvested and lysed with lysis buffer [25 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Triton-X] supplemented with phosphatase and protease inhibitors (Thermo Fisher Scientific, catalog no. 1860932). Briefly, 20 μg of cell lysates determined using a BCA Protein Assay Reagent Kit (Pierce Biotechnology) were loaded onto SDS-PAGE gels. After separation, proteins were transferred to nitrocellulose membranes. Then, the membranes were blocked with 5% nonfat milk for 1 hour at room temperature and then incubated with primary antibodies at 4°C overnight. The primary antibody dilution factors were as follows according to manufacturer's instructions: anti-GnRH-R (1:1,000, Abcam, catalog no. ab183079), RAD 51 (1:10,000, Abcam, catalog no. ab133534), phosphorylated PI3K p85 (1:1,000, catalog no. 4228s), PI3K (1:1,000, catalog no. 4292s), phosphorylated AKT Ser473 (1:1,000, catalog no. 9271s), AKT (1:1,000, catalog no. 9272s), PARP (1:1,000, catalog no. 9532s), BRCA1 (1:1,000, catalog no. 9025s), and BRCA2 (1:1,000, catalog no. 10741s; Cell Signaling Technology). After washing three times with Tris-buffered saline with 0.1% Tween 20, membranes were incubated with horseradish peroxidase–conjugated horse anti-mouse or -rabbit IgG (1:3,000; GE Healthcare) for 1 hour at room temperature. Visualization of horseradish peroxidase was performed using an enhanced ECL Detection Kit (Pierce Biotechnology). β-Actin (0.1 μg/mL; Sigma-Aldrich) or vinculin (1:3,000 dilution; Sigma-Aldrich) was used as a loading control.

Eight- to 12-week-old female nu/nu mice were injected intraperitoneally with 2.5 × 10⁶ OVCAR5 cells or 250 × 10³ HeyA8 cells in Hank's balanced salt solution (Gibco). To establish a subcutaneous ovarian cancer model, 1.0 × 10⁶ HeyA8 cells were injected into the posterior right leg of the mice. EP-100 was dissolved in PBS and given to the mice intravenously (0.02 mg/kg, 0.2 mg/kg, or 1.0 mg/kg) in a 100-μL volume. Olaparib was reconstituted in 4% dimethyl sulfoxide plus 30% PEG 300 and double-distilled water and given intraperitoneally to the mice daily in a 200-μL volume. All treatments began 7 days after cell injection and continued for approximately 4 weeks. For HeyA8 model, mice were given olaparib (50 mg/kg) daily via intraperitoneal injection and EP-100 (0.2 mg/kg) twice weekly via intravenous injection. All mice were sacrificed after any group of them became moribund; their tumors were collected, and their body weights, tumor weights, and nodule numbers and locations were recorded. At the end of the experiment, each tumor was carefully fixed in formalin, frozen in optimal cutting temperature medium, or snap-frozen for lysate preparation.

Cell viability assays were performed by testing ovarian cancer cells' ability to reduce the tetrazolium salt 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (inner salt) to a formazan. To determine the cytotoxicity of EP-100, cells were seeded in a 96-well plate and treated for 4, 24, and 72 hours with EP-100 at increasing concentrations. To determine the IC₅₀ levels of paclitaxel, cisplatin, doxorubicin, topotecan, and olaparib, cells were seeded in 96-well plates and treated for 72 hours (96 hours for cisplatin) with each drug at increasing doses. For combination treatment with EP-100 and chemotherapeutic drugs (cisplatin, paclitaxel, doxorubicin, topotecan, and olaparib), constant ratio was used to assess the combined effect of EP-100 and these five drugs. Cisplatin (NDC: 16729-288-38, Accord Healthcare Inc.), paclitaxel (NDC: 51991-938-98, Breckenridge Pharmaceutical Inc.), and doxorubicin (NDC: 0069-4037-01, PREMIERProRX) were kindly provided by MD Anderson Cancer Center Pharmacy. Topotecan (catalog no. T2705-10MG) was purchased from Sigma-Aldrich and olaparib (catalog no. O-9201) was obtained from LC Laboratories. Cells were treated with EP-100 alone for 2–4 hours and then combined with another agent for another 72 or 96 hours at increasing doses and a constant ratio. At the end of time point, cells were incubated with 0.5% MTT for 2 hours at 37°C. The supernatant was then discarded, and the MTT formazan was dissolved with 150 μL of dimethyl sulfoxide and absorbance read at OD = 570 nm.

Paraffin-embedded ovarian cancer tissue samples obtained from in vivo experiments were used to detect expression of GnRH-R and γH2AX. Tissue sections were deparaffinized and dehydrated in xylene and declining grades of ethanol (100%, 100%, 95%, 80%, and 80%) and transferred to PBS. The sections were blocked with 3% hydrogen peroxide in methanol after antigen retrieval using Diva buffer (pH 8.0). Sections were then blocked with a protein blocking buffer (3% fish gelatin in PBS) at room temperature for 20 minutes. After blocking, all sections were incubated with a polyclonal anti-GnRH-R antibody (1:200 in blocking buffer; Abcam, catalog no. ab183079), a monoclonal anti-γH2AX Ser139 antibody (1:200 in blocking buffer; Cell Signaling Technology, catalog no. 9718s) at 4°C overnight. The next day, after washing three times with PBS for 3 minutes each, slides were incubated with horseradish peroxidase–conjugated rat anti-mouse IgG2a or goat anti-rabbit IgG2 (1:500; Jackson ImmunoResearch Laboratories) for 1 hour at room temperature. After washing with PBS, sections were incubated with DAB working solution and then counterstained with hematoxylin and PBS. Five samples from each group of in vivo study were examined under an Olympus microscope, and images of each slide were captured using a Leica camera at 400× magnification. We determined the protein levels using semiquantitative method through multiplying the staining intensity score (“0”: negative; “1”: weak staining; “2”: moderate staining; “3”: strong staining) by the percentage score (“0”: less than 5% positively stained cells; “1”: 6%–24% of positively stained cells; “2”: 25%–49% of positively stained cells; “3”: 50%–74% of positively stained cells; and “4”: 75%–100% of positively stained cells).

The RPPA assay was carried out by the University of Texas MD Anderson Cancer Center RPPA facility as described previously (21). Briefly, HeyA8 cells were treated with a vehicle control, EP-100 (1 μmol/L) alone, olaparib (10 μmol/L) alone, or EP-100 plus olaparib for 24 hours. Cell lysates were collected in RIPA buffer (1% Triton X-100, 25 mmol/L Tris, pH 7.4, 150 mmol/L NaCl, 0.1% SDS, 0.5% sodium deoxycholate) containing freshly added protease and phosphatase inhibitors. Protein concentrations were quantified using a BCA Assay Kit (Pierce Biotechnology) and 40 μg of protein from each treatment was used for RPPA analysis.

Briefly, 2 × 10⁵ HeyA8 and A2780ip2 cells were seeded in 6-well plates in complete medium and left to attach overnight. The next day, cells were treated with 1 μmol/L EP-100 for 2–4 hours followed by 10 μmol/L olaparib. Untreated cells served as negative controls. After 24 hours of incubation, cells were harvested with trypsin and suspended in a concentration of 2 × 10⁴ cells/mL in PBS (Ca²⁺- and Mg²⁺-free). Next, 200 μL of low-melting-point agarose (37°C; Sigma) was added to 100 μL of cell suspensions and mixed thoroughly via pipetting up and down. After mixing, 100 μL of mixed suspensions was dropped on slides precoated with normal-melting-point agarose (Sigma), and coverslips were immediately placed carefully on the slides to form uniform gel layers over normal-melting-point agarose. The slides were kept at 4°C for 10 minutes to let the gel solidify. Once it solidified, the coverslips were carefully removed, and another 70 μL of low-melting-point agarose was dropped on the second layer and immediately covered with new coverslips. The slides were kept at 4°C for 10 minutes to let the third gel layer solidify. The coverslips were then removed, and the slides were immersed in cold lysis buffer (2.5 mol/L sodium chloride, 100 mmol/L disodium EDTA, 10 mmol/L Tris, 0.34 mol/L sodium hydroxyl, 1% Triton X-100, pH 10) at 4°C for at least 1 hour (less than 24 hours). After lysis, the lysis buffer was gently removed, and the slides were transferred into fresh cold electrophoresis buffer (Stock Solution I: 10 N sodium hydroxide; Stock Solution II: 200 mmol/L disodium EDTA; Working solution: mix 30 mL of solution I and 5 mL of solution II and adjust the volume to 1 L) for unwinding for 30 minutes. Electrophoresis was carried out for 30 minutes using a voltage of 0.74 V/cm (between electrodes) and 300-mA currents by adjusting the buffer level at 4°C. After electrophoresis, the slides were immersed in neutralizing Tris buffer (0.4 mol/L Tris, pH 7.4) for 5 minutes three times followed by washing with double-distilled water. Fifty microliters of propidium iodide (2.5 μg/mL) was dropped on each slide for incubation for 20 minutes at room temperature. The slides were then washed with water, and new coverslips were placed. Next, the slides were imaged using a fluorescent microscope at 400×, and the DNA damage of each single cell was evaluated using the OpenComet software program (http://www.cometbio.org/). The DNA damage parameters, such as the percentage of DNA in the head and tail and the tail moment (product of the tail length and DNA percentage in the tail), were calculated using at least 25 randomly selected cells per sample. The lysis solution, neutralization buffer, and electrophoresis buffer were prepared as described previously (22).

Cultured cells were fixed for 10 minutes with 4% paraformaldehyde with/without permeabilization by 0.2% X-100 and blocked with 3% FBS and 1% BSA buffer for 1 hour at room temperature. After blocking, cells were incubated with anti-GnRH-R (1:100, catalog no. NBP2-45300-0.1mg; Novus Biologicals), anti-E-cadherin (1:1,000, catalog no. 3195s, Cell Signaling Technology), anti-RAD 51 (1:10,000, catalog no. ab133534, Abcam), or anti-γH2AX (1:800, catalog no. 2577s; Cell Signaling Technology) at 4°C overnight. After washing with PBS three times, cells were incubated with Alexa 488- or 564-labeled secondary antibodies (1:250 in blocking buffer; Jackson ImmunoResearch Laboratories) as recommended by the manufacturer. Nuclear staining was achieved using Hoechst 33258 (1:10,000; Invitrogen). ProLong Diamond Antifade Mountant (Thermo Fisher Scientific) was then used to mount the stained cells on slides, which were covered with new, clean coverslips. The fluorescence signal was imaged under a Leica DM4000 B LED microscope with a Leica DFC310 digital camera or a laser scanning multiphoton confocal microscope (TCS SP5 MP; Leica Microsystems). All experimental groups were analyzed with the same settings.

The Student t test (for comparison of two groups) and ANOVA (for comparison of all groups) were used to calculate P values for normally distributed data (as determined using the Shapiro–Wilk W test). Also, the Kruskal–Wallis test was used for comparison of variables with nonparametric distribution. All statistical data were analyzed using the Prism software program (GraphPad Software). A P value less than 0.05 according to a two-tailed test was considered significant. All statistical tests were two-sided unless otherwise noted.

To evaluate the therapeutic effect of EP-100 on ovarian cancer, we first examined GnRH-R protein expression levels in ovarian cancer cell lines. Western blot results showed that the nine ovarian cancer cell lines all had increased GnRH-R expression compared with normal human ovarian tissues (Supplementary Fig. S1A). To further characterize GnRH-R, we sought to determine its localization in ovarian cancer cells. Immunofluorescence analysis demonstrated that GnRH-R is localized on the membrane, as well as in the cytoplasm of an array of ovarian cancer cells, but not in the nucleus (Supplementary Fig. S1B and S1C). We then tested the in vitro cytotoxic effects of EP-100 on the same nine ovarian cancer cell lines by measuring its half-maximal inhibitory concentration (IC₅₀). The IC₅₀ levels of EP-100 in the cell lines ranged from 0.80 to 2.56 μmol/L after 4, 24, and 72 hours of treatment (Fig. 1A and B; Supplementary Table S1). We next wanted to elucidate whether EP-100 reduces tumor growth in a preclinical ovarian cancer xenograft model. By developing an OVCAR 5 xenograft mouse model and giving the mice 0.02, 0.2, or 1.0 mg/kg EP-100 intravenously, we found that in the treatment groups, especially mice given 0.2 mg/kg EP-100, the tumor weights were markedly lower than those in a vehicle-treated control group (P = 0.027; Fig. 1C). We did not observe significant body-weight loss in any group, suggesting that EP-100 was well tolerated.

Next, we analyzed the effects of GnRH-R loss on the cytotoxicity of EP-100 in ovarian cancer cells. We selected two cell lines (HeyA8 and A2780ip2) based on their high expression of GnRH-R and high sensitivity to EP-100 and transfected them with either GnRH-R short hairpin RNA (shGnRH-R) or scrambled negative plasmids as control. We confirmed the GnRH-R protein knockdown efficiency of shGnRH-R plasmids in the cells via Western blot analysis (Supplementary Fig. S2A). An MTT assay demonstrated that GnRH-R knockdown resulted in higher IC₅₀ level of EP-100 than in control cells after 4 and 72-hour exposure to EP-100 (Supplementary Fig. S2B and S2C; Supplementary Table S2). Taken together, these results demonstrate that EP-100′s antitumor effects in ovarian cancer cells depend on the presence of GnRH-R expression.

A previous study revealed that EP-100 can sensitize GnRH-R–positive cancer cells to treatment with paclitaxel (20). To determine whether EP-100 synergizes with therapies that affect the cell cycle, we compared paclitaxel with doxorubicin, cisplatin, and topotecan, which target DNA replication, as well as to a PARP inhibitor (olaparib). We first assessed the cytotoxicity of these drugs individually (Supplementary Fig. S3) or in combination with EP-100 (Fig. 2; Supplementary Fig. S4). We used the CompuSyn software program (http://www.combosyn.com/) to examine drug–drug interactions by analyzing the data from three independent experiments and displaying the results in fraction affected-combination index (Fa-CI) plots (23). Among all the combinations, only olaparib showed a strong synergistic effect with EP-100 among HeyA8, HeyA8 MDR, A2780ip2, and A2780cp20 cells (Fig. 2A and B). We further confirmed this synergistic effect in high-grade serous ovarian cancer cells OVCAR 4 and OVCAR 8 (Fig. 2C and D). Importantly, these synergistic effects are abrogated upon downregulation of GnRH-R expression with much higher CI values than that of wild-type cells (Supplementary Fig. S2D). This suggests that the synergism is GnRH-R expression-dependent. The combination of EP-100 and doxorubicin showed a synergistic effect in HeyA8 cells (Supplementary Fig. S4A) and topotecan only had synergistic effects with EP-100 in A2780cp20 cells (Supplementary Fig. S4C). In addition, we found strong synergistic effects of treatment with EP-100 and paclitaxel in HeyA8, and the cisplatin-resistant A2780cp20 cells (Supplementary Fig. S4B) which were consistent with previously published results (20). Furthermore, we observed synergistic effects of EP-100 and cisplatin in HeyA8, A2780ip2, and A2780cp20 cells (Supplementary Fig. S4D).

To understand possible mechanisms underlying the synergy between EP-100 and olaparib, we performed reverse-phase protein array (RPPA) analysis to identify downstream pathways potentially impacted by this combination. We treated HeyA8 cells with vehicle (control), EP-100 (1 μmol/L) alone, olaparib (10 μmol/L) alone, or EP-100 combined with olaparib. After running the RPPA data in the Ingenuity Pathway Analysis database and the NetWalker software program (https://netwalkersuite.org/; ref. 24), we found that the PI3K/AKT pathway was the top pathway inhibited by EP-100 compared with control treatment (Supplementary Fig. S5). We independently confirmed this via Western blot assays (Fig. 3). The RPPA results also revealed that DNA damage- and repair-related proteins (such as PARP) were inhibited to a greater extent in the combination group compared with olaparib alone group (Fig. 3A). On the basis of these findings, we further determined the expression of PARP, cleaved PARP, BRCA1, BRCA2, and DNA repair protein RAD51 using Western blot analysis. The results showed that EP-100 can inhibit the expression of BRCA1 substantially while only a minimal reduction of BRCA2 expression was observed (Fig. 3B). We did not observe significant differences in the expression of cleaved PARP or RAD 51 between olaparib and combination treatment groups (Fig. 3B). Given that the expression of BRCA1 can be decreased by EP-100, we further determined the combination effects of EP-100 and olaparib in BRCA1- and BRCA2-mutant cells. We first detected the expression and localization of GnRH-R in BRCA1/2-mutant cells (Fig. 3C). As shown in Fig. 3D and E, we did not observe a strong synergistic effect between EP-100 and olaparib in either BRCA1-mutant ovarian cancer cells COV362 or breast cancer cells MDA-MB-436. There was a synergistic effect in BRCA2-mutant ovarian cancer cells KURAMOCHI, which is consistent with the lack of effect of EP-100 on BRCA2 expression.

We then compared DNA damage in HeyA8 and A2780ip2 cells following different treatments using the comet assay, which can reflect the number of DNA breaks by comparing the intensity of the comet tail with that of the head (25). After exposing HeyA8 and A2780ip2 cells to EP-100 and/or olaparib for 24 hours, we observed a significant increase in the amount of DNA damage in the comet tails in the combination group than that in the control and olaparib groups (P < 0.001; Fig. 4A and B; full-size images are shown in Supplementary Fig. S8). To further elucidate the DNA response to double-strand breaks, which require PARP-1 to repair and can reflect the effect of PARP inhibitor–based treatment (26, 27), in the EP-100 and olaparib combination groups, we performed immunofluorescence analysis to compare the number of γH2AX and RAD 51 foci under four conditions [vehicle (control), EP-100 (1 μmol/L) alone, olaparib (10 μmol/L) alone, or EP-100 combined with olaparib)]. As expected, γH2AX foci formation was triggered in the presence of olaparib (Fig. 4C; full-size images are shown in Supplementary Fig. S9; ref. 27). However, we did not observe a significant increase in formation of γH2AX foci in the EP-100 group. Importantly, γH2AX foci formation was significantly higher in the EP-100 and olaparib combination group compared with the olaparib group (P < 0.001 in the HeyA8 model while P = 0.0012 in the A2780ip2 model; Fig. 4C), which is in accord with the results from the comet assay. Furthermore, olaparib can induce nuclear foci formation of RAD 51 after 24-hour treatment while the addition of EP-100 to olaparib led to a significant reduction of RAD 51 foci formation (P = 0.0153 in the HeyA8 model while P < 0.001 in the A2780 ip2 model; Fig. 4D), consistent with downregulation of BRCA1.

To further determine the antitumor effects of EP-100 and olaparib in a BRCA1 and 2 wild-type ovarian cancer model, we used the HeyA8 tumor model; female tumor-bearing nude mice were randomized into four treatment groups 1 week after tumor cell inoculation (8 mice per group): vehicle (control), 0.2 mg/kg EP-100, 50 mg/kg olaparib, and the combination of EP-100 and olaparib (olaparib was administrated 1 hour after EP-100; Fig. 5A). At the end of the experiment, the body weights of the host mice in all four groups did not differ significantly (Fig. 5B). There was a 40% reduction in the mean tumor weight in the EP-100 and olaparib monotherapy groups, although the difference compared with control group did not reach significance. In contrast, the combination of EP-100 and olaparib significantly reduced the tumor weight and number of nodules below that in the control group (P = 0.0112; Fig. 5B). To further determine the impact of the combination treatment on longitudinal tumor growth, we monitored tumor growth using a HeyA8 subcutaneous mouse model. After the tumor was established (around 10 days), we started the same treatment as above. There was reduced tumor size and tumor weight in the combination group compared with the other three groups, but the difference did not reach significance due to the short-term duration of treatment (the experiment ended after 2 weeks of treatment because several mice in the control group became moribund; Supplementary Fig. S6).

We performed IHC staining of sections from dissected ovarian tumors for the cell proliferation and γH2AX markers (Fig. 5C). Compared with that in the vehicle-, EP-100-, and olaparib-treated tumors, γH2AX expression was significantly higher in the combination-treated tumors. In addition, EP-100 did not alter the expression of GnRH-R in ovarian tumor tissues in vivo based on IHC using a polyclonal antibody against GnRH-R (Fig. 5C and D).

The key findings from our study include the synergistic activity of EP-100 with olaparib. GnRH-R is expressed in a variety of tumors either related (breast, endometrial, and ovarian cancers) or not related (melanoma, glioblastoma, and lung and pancreatic cancers) to the female reproductive system (28–30). Over the past few decades, important strides have been made toward developing GnRH agonists and antagonists for the treatment of both hormone-dependent and -independent tumors (31–36). Although GnRH peptide antagonists and agonists have been successful in treating hormone-dependent diseases by reducing sex steroid levels, the adverse effects caused by these treatments, such as bone loss, cannot be neglected (36). These therapies can also result in a flare phenomenon by increasing the level of luteinizing hormone and serum testosterone during the initial treatment period (1–2 weeks; ref. 37). This phenomenon can lead to serious consequences, including pain, neurologic sequelae, and even death (37, 38). A phase I study demonstrated that EP-100 is a safe, well-tolerated drug that does not cause serious organ toxicity (18). Our in vivo studies also demonstrated no weight loss or obvious behavioral changes in mice during treatment with EP-100, which is consistent with the fact that EP-100 did not cause severe side effects at an effective dose (0.2 mg/kg). Thus, EP-100 can be developed as a novel targeted therapy for ovarian cancer.

Our findings suggested that a new combination strategy for the effective treatment of ovarian cancers is the addition of EP-100 to olaparib. Olaparib is the first PARP inhibitor approved by the FDA (in 2014) for the treatment of advanced BRCA-mutated ovarian cancer. Subsequently, it received approval for maintenance therapy for recurrent platinum-sensitive ovarian, fallopian tube, and primary peritoneal cancers regardless of BRCA mutation status. The literature contains extensive evidence of DNA damage induced by treatment with olaparib due to the inhibition of PARP, a vital regulator of a variety of cell processes, including DNA repair (39). Expression of the most well-known marker of DNA double-strand breaks (DSB), γH2AX, is increased in the number of foci formation during olaparib-based treatment (27, 40, 41). The induction of DSBs is lethal to cells with mutated BRCA 1 or 2. The synthetic lethality concept has contributed to the application of PARP inhibitors for the treatment of BRCA1- or 2-mutated cancers, which are homologous recombination repair pathway–deficient (42). However, authors have reported acquired resistance to treatment with PARP inhibitors in most patients with advanced cancer, and not all patients who carry BRCA1 or 2 mutations have responses to these inhibitors (43, 44). The limited efficacy and the emergence of cancer resistance to PARP inhibitors demonstrate the need to study avenues for potentiating their effect on cancer therapy. An encouraging approach is to develop combination strategies to enhance responses to cancer therapy with PARP inhibitors. Our in vitro and in vivo results revealed that EP-100 sensitizes BRCA wild-type ovarian cancer cells to olaparib, providing a promising combination approach to cancer therapy. Although one study showed that OVCAR 8 cells have BRCA1 promoter methylation (45), a synergistic effect was still observed between EP100 and olaparib. This is likely because methylation may not result in complete gene silencing and OVCAR 8 cells still have BRCA1 protein expression (46, 47). This is also consistent with clinical observations that BRCA1-hypermethylated tumors were not associated with better survival compared with wild-type BRCA1 in patients with ovarian cancer (48). In addition, our findings indicate that BRCA1 mutation could abrogate the synergistic effect between EP-100 and olaparib; this is likely due to the effects of EP-100 decreasing BRCA1 expression, but not BRCA2. However, further studies are needed to clarify the underlying molecular mechanisms.

Recent findings revealed that activation of the PI3K/AKT/mTOR pathway is associated with acquired resistance of treatment with a PARP inhibitor, which may provide a useful combination strategy for sensitizing tumors to such inhibitors (49). Notably, subsequent studies revealed that the combined inhibition of PI3K/AKT and PARP had a synergistic antitumor effect in several preclinical models of breast, prostate, and ovarian cancer (50–53). In addition, inhibition of the PI3K/AKT pathway can sensitize PTEN-mutated cancer cells to treatment with a PARP inhibitor (54). In a BRCA wild-type triple-negative breast cancer model, inhibition of the PI3K/mTOR pathway resulted in blockage of double-strand break repair and thus sensitized cancer cells to treatment with a PARP inhibitor (55). Our results indicate that EP-100 may sensitize BRCA wild-type ovarian cancer cells to PARP inhibitors by inhibiting the PI3K/AKT pathway.

Taken together, our data provide a rational combination of EP-100 and olaparib for ovarian cancer therapy. Our in vivo studies suggest that this combination is well-tolerated; however, the optimal dosing and sequencing of these drugs may require additional work.

C. Leuschner is the vice president of Research and Development and has ownership interest (including stock, patents, etc.) in Esperance Pharmaceuticals. R.L. Coleman reports receiving a commercial research grant from Esperance Pharmaceuticals, AstraZeneca, and Clovis, reports receiving other commercial research support from Merck, Genmab, Roche, is a consultant/advisory board member for AstraZeneca, Clovis, Genentech, Genmab, Gamamab, Aravive, and Tesaro. A.K. Sood reports receiving a commercial research grant from M-Trap, has ownership interest (including stock, patents, etc.) in Biopath, and is a consultant/advisory board member for Kiyatec and Merck. No potential conflicts of interest were disclosed by the other authors.

Conception and design: S. Ma, M.S. Kim, S.Y. Wu, X. Liang, R.L. Coleman, A.K. Sood

Development of methodology: A. Villar-Prados, Y. Wen, C. Rodriguez-Aguayo, C. Leuschner, A.K. Sood

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S. Ma, S. Pradeep, Y. Wen, C. Rodriguez-Aguayo, R.L. Coleman, A.K. Sood

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S. Ma, A. Villar-Prados, Y. Wen, M.S. Kim, P.T. Ram, R.L. Coleman, A.K. Sood

Writing, review, and/or revision of the manuscript: S. Ma, S. Pradeep, A. Villar-Prados, E. Bayraktar, L.S. Mangala, S.Y. Wu, W. Hu, C. Leuschner, X. Liang, P.T. Ram, K. Schlacher, R.L. Coleman, A.K. Sood

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): W. Hu, R.L. Coleman, A.K. Sood

Study supervision: S. Pradeep, A.K. Sood

Other (advised on conception): K. Schlacher

STR DNA fingerprinting was done by the CCSG-funded Characterized Cell Line Core, NCI # CA016672. RPPA was performed by the RPPA Core Facility, The University of MD Anderson Cancer Center, NCI CA16672. We thank Dr. Walter Hittelman for support on multiphoton confocal microscopy. We thank the Department of Scientific Publications at MD Anderson for reviewing this manuscript. This work was supported, in part, by the NIH grants (CA016672, UH3TR000943, P50 CA217685, R35 CA209904), Ovarian Cancer Research Fund, Inc. (Program Project Development Grant), The Judi A. Rees Ovarian Cancer Research Fund, The Blanton-Davis Ovarian Cancer Research Program, the American Cancer Society Research Professor Award, and the Frank McGraw Memorial Chair in Cancer Research. Y. Wen was supported, in part, by the NIH 5 P50 SPORE CDP Award CA116199, the Marsha Rivkin Center for Ovarian Cancer, and the National Comprehensive Cancer Network. S.Y. Wu was supported by the CPRIT Research Training Program (RP101502, RP140106, and RP170067).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.