
==== Front
Transl Oncol
Transl Oncol
Translational Oncology
1936-5233
Neoplasia Press

S1936-5233(24)00246-8
10.1016/j.tranon.2024.102119
102119
Original Research
Characterization of DNA damage repair pathway utilization in high-grade serous ovarian cancers yields rational therapeutic approaches
Nakatsuka Erika a
Tan Lijun a
Cunneen Brianna b
Foster Caroline a
Lei Yu Leo cd
McLean Karen Karen.McLean@RoswellPark.org
ab⁎
a Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Michigan, 1500 E. Medical Center Dr., Ann Arbor, MI 48109, USA
b Department of Gynecologic Oncology and Department of Pharmacology and Therapeutics, Roswell Park Comprehensive Cancer Center, Elm and Carlton Streets, Buffalo, NY 14263, USA
c Department of Head and Neck Surgery, the University of Texas M.D. Anderson Cancer Center, Houston, TX 77030, USA
d Department of Cancer Biology, the University of Texas M.D. Anderson Cancer Center, Houston, TX 77054, USA
⁎ Corresponding author at: Department of Gynecologic Oncology and Department of Pharmacology and Therapeutics, Roswell Park Comprehensive Cancer Center, Elm and Carlton Streets, Buffalo, NY, 14263, USA. Karen.McLean@RoswellPark.org
12 9 2024
12 2024
12 9 2024
50 1021194 6 2024
12 8 2024
8 9 2024
© 2024 The Authors. Published by Elsevier Inc. CCBYLICENSE.
2024

https://creativecommons.org/licenses/by-nc/4.0/ This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Highlights

• Diverse DNA damage repair pathway utilization affects treatment responses.

• Combination treatments successfully overwhelm DNA damage repair.

• ATM inhibitors effectively combine with PARP inhibitors.

While poly (ADP-ribose) polymerase (PARP) inhibitors (PARPi) have improved the prognosis of ovarian high-grade serous carcinoma (HGSC) tumors that are homologous recombination (HR) deficient (HRD), new therapeutic strategies are needed for tumors that are HR proficient (HRP) because they demonstrate greater resistance to current treatments and thus have poorer clinical outcomes. Additionally, clinical precautionary statements regarding potential risks associated with PARPi, such as myelodysplastic syndrome, highlight the need for combinatorial approaches that can lessen the dose and duration of PARPi treatment to reduce toxicities. Here, we evaluated DNA double-strand damage repair pathways in HRD and HRP ovarian cancer cell lines and found that in HRD cell lines, PARPi therapy reduced non-homologous end joining (NHEJ)-mediated repair, specifically due to decreased theta-mediated end-joining. The combination of PARPi with ATM serine/threonine kinase inhibitor (ATMi) suppressed both NHEJ and HR pathways in HRD and HRP cell lines, with synergistic increases in apoptosis and decreases in cell viability and colony formation. Interestingly, PARPi plus ATMi also decreased NF-κB p65 phosphorylation, which was not observed when PARPi was combined with inhibition of the ATR kinase (ATRi). These findings indicate that PARPi plus ATMi is a promising strategy for HGSC independent of underlying tumor HR status.

Keywords

DNA damage repair
Homologous recombination
Non-homologous end-joining
ATM
PARP
==== Body
pmcIntroduction

We are in urgent need of new treatment approaches for ovarian cancer, as it is the fifth highest cause of cancer-related deaths among women in the United States with a 5-year survival rate of less than 50%. High-grade serous carcinoma (HGSC) arising from the ovary, fallopian tube, or peritoneum is the most common and lethal subtype of epithelial “ovarian” cancer, representing approximately 65% of cases. Although the majority of HGSC initially respond to multimodal therapy consisting of surgery and platinum-based chemotherapy, most patients develop platinum-resistant recurrent disease and ultimately succumb to their cancer.

DNA damage repair (DDR) defects are central to HGSC pathogenesis; approximately 50% of tumors exhibit deficiencies in the homologous recombination (HR) pathway of DNA damage repair [1]. BRCA gene alterations are the most common underlying mechanism of HR deficiency (HRD) in HGSC. The pivotal finding that HRD is synthetically lethal with poly (ADP-ribose) polymerase (PARP) inhibition has led to the clinical development of PARP inhibitor (PARPi) therapies [2,3], which have improved the prognosis for ovarian HGSC tumors with HR deficiency [4]. However, tumors with HR proficiency (HRP) are present in a subset of patients and are particularly treatment-resistant with poorer clinical outcomes. Hence, new treatment strategies are needed for these patients.

Alongside the above unmet need, improvements in PARPi therapy are also warranted. For example, despite significant initial clinical enthusiasm surrounding PARPi therapy, the integration of PARPi into clinical practice has revealed that most tumors become resistant to such therapy [5]. Furthermore, recent clinical precautions have surfaced regarding potential risks associated with PARPi treatment, including possible survival detriments under certain clinical scenarios and the increased risk of myelodysplastic syndrome [6]. These issues underscore the need to further our understanding of the optimal clinical indications PARPi, including the identification of potential combinatorial approaches that can lessen the dose and duration of treatment with PARPi to reduce toxicities as well as provide efficacy in HRP tumors.

In this study, we characterized the effects of DDR pathway inhibitors on the efficacy of DNA damage repair in both HRD and HRP ovarian cancer cell lines. Based on our characterization of DDR pathway utilization, we hypothesized that a PARPi plus an ataxia-telangiectasia mutated (ATM) inhibitor (ATMi) would demonstrate improved efficacy as compared to either single agent, as ATM inhibition prevents repair of DNA double-strand breaks that develop following inhibition of PARP-mediated repair of single-strand breaks. To test this hypothesis, we carried out in vitro studies in both human HGSC cell lines and murine cell lines derived from the ID8 immunocompetent mouse model system.

Materials and methods

Cell lines and inhibitors

OVCAR3, OVCAR5, OVCAR8, and ES2 cell lines were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cell line identity was validated by short tandem repeat profiling. ID8 Trp53-/- and ID8 Trp53-/-/Brca1-/- cell lines were a generous gift from Dr. Iain McNeish [7,8]. All lines were maintained in RMPI media supplemented with 10% fetal bovine serum (FBS). Mycoplasma testing was performed (PlasmoTest Kit, Invivogen, San Diego, CA, USA) and confirmed to be negative at a frequency of at least every two months to ensure healthy cell cultures.

Clinical-grade cisplatin was obtained from Fresenius Kabi (Lake Zurich, IL, USA). All inhibitors were purchased from Adooq Bioscience (Irvine, CA, USA): ATM inhibitor KU-60019 (Cat. No. A10507), PARP inhibitor olaparib (Cat. No. A1011), ATR inhibitor AZD6738 (Cat. No. A15794).

NHEJ and HR repair assays

For the non-homologous end joining (NHEJ) repair quantification, ovarian cancer cells were transfected with the pimEJ5-GFP plasmid (gift from Jeremy Stark [Addgene plasmid # 44026; http://n2t.net/addgene:44026; RRID:Addgene_44026]) mixed with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Briefly, transfected cells were selected by puromycin. Then, polyclonal lines with stable expression of the pimEJ5-GFP plasmid were transiently co-transfected with the pCBASceI plasmid (gift from Maria Jasin [Addgene plasmid # 26477; http://n2t.net/addgene:26477; RRID:Addgene_26477]) and dsRed plasmid with Lipofectamine 2000. The endonuclease SceI is used as its recognized restriction site is rare outside the synthesized plasmid sites [9]. DNA damage repair inhibitors were added 6 h after transfection. At 72 h after transfection, flow cytometry analysis with a Bio-Rad ZE5 flow cytometer and FlowJo software (Tree Star, Inc., Ashland, OR) was performed to quantify the percentage of GFP (green fluorescent protein)-positive cells and dsRed-positive cells.

For the HR repair measurements, cell lines were transfected with the pDR-GFP plasmid (gift from Maria Jasin [Addgene plasmid # 26475; http://n2t.net/addgene:26475; RRID:Addgene_26475]) using Lipofectamine 2000 and selected by puromycin. Cells with stable expression of pDR-GFP were then co-transfected with pCBASceI plasmid and dsRed plasmid and analyzed by flow cytometry (Bio-Rad ZE5) six days later. Data were analyzed by FlowJo software.

To determine the relative contributions of canonical and alternative NHEJ, ovarian cancer cells stably transfected with the pimEJ5-GFP plasmid were transfected with pCBASceI using Lipofectamine 2000 and treated with DDR inhibitors as above. Three days after treatment, the GFP-positive cells were sorted by flow cytometry (MoFlo Astrios cell sorter) and DNA was extracted from the sorted cells by using a QIAampR DNA Mini Kit (QIAGEN, Germantown, MD, USA). Next, extracted DNA was amplified by PCR using Q5 High-Fidelity DNA Polymerase (New England BioLabs, Ipswich, MA, USA) and the primers KNDRF (5ʹ-CTGCTAACCATGTTCATGCC-3ʹ) and KNDRR (5ʹ-AAGTCGTGCTGCTTCATGTG-3ʹ). The PCR products were then separated by agarose gel electrophoresis. DNA bands around 380 bp were extracted by using a QIAQuick DNA Extraction Kit (QIAGEN), and the extracted DNA was digested by the restriction endonuclease I-SceI (New England BioLabs) and then analyzed by agarose gel electrophoresis. Band intensities were quantified by ImageJ and the intensity of the cut band relative to the total band signals (cut plus uncut) calculated.

Cell viability assay

Twenty-four hours after plating, cells were treated with the indicated drugs. Five days after treatment, cell viability was measured using a Resazurin Cell Viability Kit (Biotium, Freemont, CA, USA) according to the manufacturer's instructions. Synergy calculations were performed using CompuSyn software (www.combosyn.com) and SynergyFinder (https://synergyfinder.fimm.fi).

Colony formation assay

Plating efficiency was determined for each cell line. Cells were first plated and then 24 h later treated with the indicated drugs. Plates were kept in an incubator for 6–8 days until cells in the control group formed sufficiently large colonies for quantification. Cells were stained with crystal violet (0.5% crystal violet, 10% ethanol, and 90% Milli-Q water) for 10 min at room temperature. Plates were rinsed with water twice and dried overnight. The number of colonies was then quantified by visual inspection.

Apoptosis assay

Cells were plated, and treatment was initiated 24 h later. Three to four days after treatment, cells were detached from the plate with trypsin and washed with phosphate-buffed saline (PBS). Cells were stained with fluorescein isothiocyanate (FITC) and propidium iodide (PI) using the FITC Annexin V Apoptosis Detection Kit I (BD Biosciences, Franklin Lakes, NJ, USA) according to the manufacturer's instructions. Cells were analyzed by flow cytometry (Bio-Rad ZE5). Data were analyzed with FlowJo software.

Immunoblotting

Antibodies against PARP (46D11, Cat. No. 9532), phospho-p65 (Ser536, 93H1, Cat. No. 3033), and p65 (D14E12, Cat. No. 8242) were purchased from Cell Signaling Technology (Danvers, MA, USA). The antibody against β-actin (AC-15, Cat. No. ab6276) was purchased from Abcam (Waltham, MA, USA). Cells were lysed with RIPA Buffer containing a protease inhibitor cocktail (Sigma-Aldrich, Burlington, MA, USA) and phosphatase inhibitor cocktail (Thermo Fisher Scientific, Waltham, MA, USA). Next, 40 μg of lysate were separated by use of either 4–12% or 4–20% gradient Novex Tris-Glycine Gel electrophoresis (Invitrogen), and gel contents were transferred to polyvinylidene difluoride membranes using the wet transfer system (Bio-Rad, Hercules, CA, USA). Membranes were blocked with Tris-buffered saline with Tween (TBS-T) containing 10% blotting grade blocker (milk) (Bio-Rad). This step was followed by incubation with the primary antibodies in TBS-T with 5% milk and then with the corresponding horseradish peroxidase-conjugated secondary antibody. The proteins were visualized with the ChemiDoc Imaging System (Bio-Rad).

Statistical analysis

All experiments were performed in duplicate or triplicate as indicated, and all data are expressed as the mean ± standard deviation. Statistical analysis was performed using GraphPad Prism version 7.0 for Windows (GraphPad Software). For single comparisons, an unpaired, two-tailed Student's t-test was used. For multiple comparisons, one-way analysis of variance (ANOVA) with Tukey's or Bonferroni post hoc testing was performed. Results were considered statistically significant if the p value was ≤ 0.05.

Results

HR deficient and proficient cell lines demonstrated unique mechanisms of DNA double-strand break repair

We first characterized the relative utilization of distinct DNA double-strand break repair pathways across HGSC cell lines differing in their HR pathway functional status. Two HRD cell lines were used, OVCAR3 and OVCAR8. OVCAR3 demonstrates functional loss of HR downstream of RAD51, although it shows no functionally inactivating mutations in HR genes and is competent in forming RAD51 foci in response to DNA damage [10]. The OVCAR8 cell line harbors BRCA1 gene methylation with decreased BRCA1 protein expression levels [11]. We also used two HRP cell lines, OVCAR5 and ES2. While recent work demonstrates that platinum response in two-dimensional cell culture assays is not necessarily correlated with HR status [12], we find that in cell viability assays following treatment with cisplatin, the cell lines classified herein as HRD were more sensitive to cisplatin than those classified as HRP (Fig. 1A). Furthermore, recent HRD scores for these cell lines reveal that OVCAR3 and OVCAR8 demonstrate high HRD scores while OVCAR5 and ES2 have low HRD scores [12].Fig. 1 Homologous recombination proficient (HRP) and deficient (HRD) high-grade serous cell lines differed in their utilization of DDR pathways and sensitivity to anticancer therapies. A Response to single agent cisplatin therapy was correlated with homologous recombination status. Drug concentrations as indicated. N = 2, each treatment done with quadruplicates. B HR and NHEJ repair pathway utilization differed depending on whether cells demonstrated HR proficiency or deficiency. N = 2, treatments done with duplicates. C HRP cell lines demonstrated greater utilization of the canonical pathway of NHEJ than HRD cell lines. N = 4. For all panels, bar graphs depict means and error bars are standard deviations. All statistical analyses run with the Student's t-test; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig 1

We next performed fluorescence reporter assays to characterize the relative engagement of NHEJ and HR pathways across our panel of cell lines. To detect the levels of HR repair, pDR-GFP reporter assays were performed [13], in which the percentage of GFP-positive cells reflects the HR repair of DNA double-strand breaks induced by the endonuclease I-SceI [9]. To measure NHEJ repair, pimEJ5-GFP plasmid reporter assays were performed [14], in which the percentage of GFP-positive cells reflects the NHEJ repair of DNA double-strand breaks induced by I-SceI. dsRed plasmid expression was used to control for transfection efficiency in both assays, with the repair rate quantified as the percentage of GFP-positive cells divided by the percentage of dsRed-positive cells (Supplemental Figure 1). As expected, HR was the predominant pathway in HRP cell lines OVCAR5 and ES2 (Fig. 1B). In contrast, HRD cell lines showed higher levels of NHEJ activity than HR activity (Fig. 1B). Notably, in all cell lines tested, both the HR and NHEJ pathways demonstrated activity, with similar HR pathway levels between the cell lines. Thus, HRD cells were not completely HR deficient but rather just relatively less dependent on HR and more dependent on NHEJ for the repair of DNA double-strand breaks.

NHEJ can occur either via the canonical pathway or a non-canonical pathway. In canonical NHEJ (c-NHEJ), the nucleotide sequence at the DNA double-strand break site is typically maintained or repaired with minimal mutations. In non-canonical repair (also termed theta-mediated end-joining [TMEJ] or microhomology-mediated end-joining [MMEJ]), longer deletions of 5–25 base pairs occur during the repair process. We sought to determine the proportion of canonical NHEJ versus TMEJ pathway-mediated repair in these cell lines when NHEJ is used for DNA damage repair. The experimental approach uses the same pimEJ5-GFP plasmid to monitor NHEJ repair [14]. When the induced pimEJ5-GFP plasmid double-strand break is repaired by canonical NHEJ without nucleotide deletions, it restores the I-SceI cutting site. Alternatively, when the DNA double-strand breaks are repaired by TMEJ, the associated nucleotide deletions lead to loss of the I-SceI cutting site. By assessing the relative engagement of these two NHEJ pathways, we found that the HRD cell lines have low levels of canonical NHEJ and depend mainly on the TMEJ pathway, while the HRP cell lines utilize the canonical (deletion-free) NHEJ pathway more than the TMEJ pathway (Fig. 1C).

DNA damage repair inhibitor therapy shifted repair pathway utilization

Although it is well known that PARP inhibitors such as olaparib block DNA damage repair, PARPi resistance is a significant clinical problem that warrants further characterization of DDR alterations with the goal of developing new treatment combinations to prevent or overcome resistance. Therefore, we next investigated a panel of DDR inhibitors targeting different downstream repair proteins to identify potential agents for further study, and based on these findings, we narrowed our further work to the characterization of PARP inhibition and ATM inhibition. We first determined cell viability following single agent therapy and found that the HRD cell lines were more sensitive to olaparib than the HRP cell lines, as expected (Fig. 2A). Both HRD and HRP cell lines were resistant to single agent ATMi (Fig. 2A).Fig. 2 PARP inhibition plus ATM inhibition decreased HR and NHEJ independent of the underlying tumor cell HR status. A Cell viability following PARP inhibitor therapy with olaparib was correlated with HR proficiency status, while all cell lines tested demonstrated relative resistance to single agent ATM inhibition (ATMi). N = 2, each treatment done with triplicates. B Single agent DDR pathway inhibitor therapy altered the HR and NHEJ repair. Olaparib dose 5 μM, KU60019 ATMi dose 1 μM. N = 2, each experiment done with duplicates. C Combinatorial therapy resulted in NHEJ alterations and a statistically significant reduction in HR in both HRD and HRP cell lines. Olaparib dose 1 μM, KU60019 ATMi dose 1 μM. N = 2, each experiment done with duplicates. For all panels, bar graphs depict means and error bars are standard deviations. All statistical analyses run with the Student's t-test; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig 2

We then assessed the impact of PARPi and ATMi on DDR pathway engagement, including quantification of HR plus canonical-NHEJ versus TMEJ following treatment. To best visualize the shifts in DDR following inhibitor therapy, we chose to represent the results as the relative repair events through each pathway out of a total of 100%. PARPi did not further decrease the already low levels of HR in the HRD cell lines (Fig. 2B). In contrast, PARPi reduced TMEJ in HRD cell lines (Fig. 2B), which is consistent with the recent report that PARP activity promotes TMEJ in breast cancer cell lines [15]. As a result, total DDR repair was decreased in HRD cell lines. In contrast, in the HRP cell lines, PARPi treatment resulted in a statistically significant decrease in HR (Fig. 2B), and again an increase in TMEJ was observed in ES2 cells. ATMi increased both TMEJ and c-NHEJ (though the increase was not significant in ES2 cells); however, the increase in the TMEJ pathway was more robust in all cell lines (Fig. 2B).

Previous reports indicate that HRD cells are more dependent on TMEJ and thus more sensitive to deficiencies in, or inhibition of, polymerase-theta [16]. Since olaparib decreased both HR and TMEJ in HRP cell lines resulting in decreased total DDR, and these cell lines are less sensitive to olaparib, we hypothesized that decreasing the total DDR by inhibiting both the HR and NHEJ pathways would demonstrate improved efficacy. Based on this hypothesis, we tested the combination of PARPi plus ATMi, as the inhibition of ATM increases the dependency on TMEJ, which can be inhibited by olaparib. We observed that ATM inhibition increased NHEJ in HRD cells (Fig. 2C). As expected, HR and NHEJ assays showed that the combination of PARPi plus ATMi resulted in significantly decreased HR compared with that using single agent therapy in all cell lines, and decreased NHEJ was detected in comparisons with single agent therapy (Fig. 2C).

Combinatorial therapy uniquely altered cell viability and signaling

Based on the DDR pathway inhibition following the combinatorial therapy (Fig. 2C), we further hypothesized that PARPi plus ATMi would demonstrate improved anti-cancer efficacy as compared with either single agent. We assessed the impact of the combinatorial therapy with olaparib plus ATMi using a range of doses for each agent (Fig. 3A); the two agents yielded synergistic decreases in cell viability in both HRP and HRD cell lines as calculated by the Chou–Talalay method [17] (Fig. 3A, Supplemental Figure 2A). To validate these findings, synergy was also assessed using the Loewe, Bliss, and Highest Single Agent (HSA) methods (Supplemental Figure 2B). Our studies were then extended to colony formation assays. While PARPi decreased colony formation in HRD lines as expected, PARPi + ATMi demonstrated greater efficacy in both HRD and HRP cell lines (Fig. 3B).Fig. 3 Combinatorial therapy with the PARP inhibitor olaparib and the ATM inhibitor KU60019 resulted in synergistic cell killing independent of HR status in human HGSC cell lines. A PARP inhibitor (PARPi) therapy in combination with ATM inhibitor (ATMi) therapy resulted in synergistic cell death in both HRD and HRP cell lines as assessed by in vitro cell viability assays. Drug concentrations as indicated. N = 2, each experiment done with triplicates. B Combination therapy with PARP inhibition and ATM inhibition resulted in a statistically significant decrease in colony formation ability in both HRD and HRP cell lines. Olaparib dose 1 μM, KU60019 ATMi dose 0.5 μM. N = 2. All treatments done in duplicate, and representative plate images are shown. C Combinatorial therapy significantly increased apoptotic cell death in all cell lines tested. Olaparib dose 1 μM, KU60019 ATMi dose 1 μM. Analysis performed four days after treatment. N = 2, each experiment done with duplicates. D Immunoblotting revealed increased PARP cleavage in both HR deficient and HR proficient cell lines following combinatorial therapy, while single agent therapy with olaparib induced modest PARP cleavage specifically in HR deficient cell lines. Olaparib dose 1 μM, KU60019 ATMi dose 1 μM. N = 2. E PARP inhibition plus ATM inhibition decreased NF-κB p65 phosphorylation, which was not observed when PARP inhibition was combined with an ATR inhibitor (ATRi). Olaparib dose 1 μM, KU60019 ATMi dose 1 μM, AZD6738 ATMi dose 1 μM. N = 2. For all panels, bar graphs depict means and error bars are standard deviations. All statistical analyses run with the Student's t-test; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig 3

We next assessed indicators of apoptotic cell death, by both quantification of apoptosis by flow cytometry and immunoblotting for PARP cleavage. We found that the combination therapy induced a statistically significant increase in apoptosis (Fig. 3C) and increased PARP cleavage (Fig. 3D). Of note, while some PARP cleavage was identified in the HRD cell lines following single agent olaparib therapy, the combinatorial therapy increased PARP cleavage in both HRD and HRP cell lines (Fig. 3D). It has been previously reported that PARP-1 and ATM mediate the activation of IKKβ and NF-κB following nuclear DNA damage, which regulates resistance to apoptosis [18,19]. Therefore, we evaluated phosphorylation of the p65 subunit of NF-κB by immunoblotting. We found that the combination of PARPi and ATMi decreased phospho-p65 levels (Fig. 3E). To assess specificity of our combinatorial therapy, we also assessed the impact of olaparib in combination with the ATM-related protein ataxia telangiectasia and Rad3-related (ATR) kinase. Importantly, phospho-p65 was not decreased by the combination PARPi and ATR inhibitor (ATRi) treatment (Fig. 3E). These findings suggest that the combinatorial therapy of PARPi and ATMi may specifically exert its anticancer effects via NF-κB signaling.

Synergistic effects observed with the combinatorial therapy in a second model system

These results were further verified using the ID8 high-grade serous carcinoma model system [20]. To better recapitulate the genetic background of HGSC, the parental ID8 cell line was modified to incorporate a Trp53 deletion without or with a Brca1 deletion [7,8]. Effects of the combination therapy with PARPi and ATMi were next assessed in this model system.

We first validated that the Brca1 deletion resulted in a HRD phenotype with increased sensitivity to olaparib (Fig. 4A). Both ID8 Trp53-/- and ID8 Trp53-/-Brca1-/- were sensitive to ATMi at higher doses (Fig. 4A). We next assessed the effects of the combination therapy on cell viability and colony formation capacity. In cell viability assays, ID8 Trp53-/- demonstrated relative resistance to PARPi that could be overcome with the addition of ATMi (Fig. 4B). As expected, HRD ID8 Trp53-/-Brca1-/- cells were more sensitive than ID8 Trp53-/- at all PARPi doses, but again the combination therapy resulted in synergistic cell killing (Fig. 4B). These findings were then validated in colony formation assays, in which the combination therapy caused statistically significant decreases in colonies formed as compared with either PARPi or ATMi alone (Fig. 4C). Finally, we characterized the synergistic effects at a mechanistic level, and results revealed that the combination therapy significantly increased apoptotic cell death in both HRP ID8 Trp53-/- and HRD ID8 Trp53-/-Brca1-/- (Fig. 4D). Additionally, we observed decreased levels of phospho-p65 following the combination therapy, but not following single agent treatments, in both HRP and HRD model systems (Fig. 4E).Fig. 4 PARP inhibition in combination with ATM inhibition was shown to be an effective combinatorial therapy in both HRP and HRD derivatives of the ID8 murine cell line. A HR deficient ID8 Trp53-/-Brca1-/- cells demonstrated increased sensitivity to olaparib and ATM inhibition as compared with HR proficient ID8 Trp53-/-.Drug concentrations as indicated. N = 2. B In both the HRD and HRP cell lines, combination therapy induced synergistic cell death. Drug concentrations as indicated. N = 2, all experiments performed with quadruplicates. C Combinatorial therapy resulted in a statistically significant decrease in colony formation in both the HRP and HRD ID8 cell lines. Olaparib dose 1 μM, KU60019 ATMi dose 1 μM. N = 1. All experiments done in duplicate, and representative images are shown. D Combinatorial therapy yields a significant increase in apoptotic cell death in both HRP and HRD lines. Olaparib dose 1 μM, KU60019 ATMi dose 1 μM. Analysis performed three days after treatment. N = 2, each experiment done with duplicates. E PARP inhibition plus ATM inhibition decreased NF-κB p65 phosphorylation, which was not seen with single agent treatments. N = 2. For all panels, bar graphs depict means and error bars are standard deviations. All statistical analyses run with the Student's t-test; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig 4

Discussion

Cancer cells utilize DNA damage repair pathways to overcome cellular damage elicited by stressors including chemotherapeutic agents. In HGSC, extensive research has assessed the impact of HR status on responses to therapy, and now, HR deficiency can be employed as a clinical biomarker to dictate both eligibility for PARP inhibitor therapy and the predicted magnitude of response to treatment. While this evolving knowledge of HR status has allowed for increased treatment options for patients with HRD tumors, HGSC displaying HR proficiency and/or platinum resistance remains a fundamental clinical challenge necessitating new treatment approaches. Therefore, we sought to further characterize DDR pathway utilization and alterations in both HRD and HRP HGSC to identify potential new treatments for patients that will be effective independent of HR status.

Here, we investigated two model systems, human HGSC cell lines known to harbor pathognomonic Trp53 alterations plus murine ID8 cells. The specific cell lines investigated in each model system provide ranges in both platinum sensitivity and PARP inhibitor sensitivity. We first characterized basal DDR mechanisms, which led to the critical finding that there is HR pathway-mediated repair of DNA double-strand breaks in both HRP and HRD cells, albeit HRD cells utilize NHEJ to a greater extent (Fig. 1B). This suggests that targeting HR may be efficacious in not only HRD tumors but also in HRP tumors, a finding that is congruent with clinical trials demonstrating clinical benefit to maintenance therapy with the PARP inhibitor niraparib in HRP tumors in addition to the anticipated effects in HRD tumors [21,22]. We also found that within the NHEJ pathway of DDR, HRD cells preferentially utilize TMEJ while HRP cells have a higher level of canonical NHEJ repair (Fig. 2D). This result adds to growing body of data regarding TMEJ in HRD tumors [16]. Targeting TMEJ is now being studied as a therapeutic approach for HRD tumors [23,24].

Based on our observed alterations in DDR pathway engagement, we hypothesized that PARP inhibition plus ATM inhibition would effectively block both HR and NHEJ, resulting in therapeutic efficacy. We found that across all human HGSC cell lines, the combination treatment overcame compensatory increases in NHEJ that resulted from ATM inhibition and yielded a decrease in both NHEJ and HR repair (Fig. 2C). Furthermore, we found that treatment with both PARP inhibition and ATM inhibition yielded synergistic decreases in cell viability by both viability assays and colony formation assays in both our human and murine model systems, and mechanistically this was due to increases in apoptotic cell death (Fig. 3, Fig. 4). While prior studies have investigated this combinatorial approach [25,26], our work advances the field through the robust characterization of DDR pathway engagement following treatment across a panel of diverse HGSC cell lines, including the critical finding that this combination helps overcome compensatory DDR pathway shifts and thus demonstrates efficacy in the clinically challenging scenario of HRP tumors.

Importantly, we have identified a differential effect in downstream signaling alterations following ATMi versus ATRi in combination with PARP inhibitor therapy. Specifically, we have observed that PARP inhibition in combination with ATM inhibition results in decreased phosphorylation of the NF-κB p65 subunit, but this change was not noted following ATRi treatment in combination with PARP inhibition (Fig. 3E). The combination of olaparib and ATR inhibition has shown efficacy in olaparib-resistant HGSC in vitro and in vivo [27,28], but a recent phase II clinical study suggested that this combination of olaparib and ATR inhibition was not effective in HRP tumors [29]. Our characterization of DDR pathway engagement following an alternative combinatorial therapy with ATM inhibition and PARP inhibition suggests this combination can elicit responses in both HRD and HRP tumors. Additionally, as we refine treatment approaches in all patients, there may be additional biomarkers of response to integrate, such as the finding that ATR responses were correlated with the presence of ATM mutations in ovarian and breast cancers [30].

One limitation of our study is the inclusion of a finite number of established cell lines of high-grade serous histology, although we were able to demonstrate similar findings of responses in HRP and HRP ID8-derived cell lines. Our findings presented herein warrant further characterization in both patient-derived samples and additional animal models as well as ultimate translation to clinical trials. The inclusion of preclinical immunocompetent mouse models will allow not only the assessment of efficacy in an immunocompetent model system, but also the characterization of immune infiltrate changes following therapy and potential combinatorial treatment approaches that integrate immunomodulatory therapies. Finally, from a mechanistic standpoint, our observation that ATM inhibition specifically, but not ATR inhibition, in combination with PARPi therapy results in decreased NF-κB phosphorylation is a provocative finding that warrants further study.

In summary, we have found that HR deficient and HR proficient high-grade serous carcinomas demonstrate different DDR pathway activation in response to genotoxic insults. Based on the repair pathways activated following single agent treatments, it is possible to develop combinatorial strategies that can block compensatory mechanisms in DNA damage repair. Here we reported that ATM inhibition in combination with PARP inhibition is an effective mechanism in both HRD and HRP HGSC cells, including both human and murine model systems. Furthermore, this combination specifically induces decreases in NF-κB phosphorylation and apoptotic cell death. Continued dissection of signaling pathway alterations and therapeutic effects following DDR pathway combinatorial therapies is warranted to define the best treatment approach for patients with specific DDR pathway molecular and phenotypic profiles.

CRediT authorship contribution statement

Erika Nakatsuka: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Lijun Tan: Investigation, Data curation. Brianna Cunneen: Writing – original draft, Visualization, Data curation. Caroline Foster: Data curation. Yu Leo Lei: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. Karen McLean: Writing – review & editing, Writing – original draft, Visualization, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix Supplementary materials

Image, application 1

Image, application 2

Data availability

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Acknowledgments

This work was supported by NIH grants R01CA251198 (KM) and R01DE026728 (YLL) , and funding from The Uehara Memorial Foundation (EN). Editorial assistance for this publication was provided by Roswell Park's Scientific Editing and Research Communications Core (SERCC) Resource, which is supported by a National Cancer Institute (NCI) Cancer Center Support Grant (grant no. NCI P30CA016056 ).

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.tranon.2024.102119.
==== Refs
References

1 Konstantinopoulos P.A. Ceccaldi R. Shapiro G.I. D’Andrea A.D. Homologous recombination deficiency: exploiting the fundamental vulnerability of ovarian cancer Cancer Discov. 5 2015 1137 1154 26463832
2 Farmer H. McCabe N. Lord C.J. Tutt A.N. Johnson D.A. Richardson T.B. Santarosa M. Dillon K.J. Hickson I. Knights C. Martin N.M. Jackson S.P. Smith G.C. Ashworth A. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy Nature 434 2005 917 921 15829967
3 Bryant H.E. Schultz N. Thomas H.D. Parker K.M. Flower D. Lopez E. Kyle S. Meuth M. Curtin N.J. Helleday T. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase Nature 434 2005 913 917 15829966
4 Liu J.F. Konstantinopoulos P.A. Matulonis U.A. PARP inhibitors in ovarian cancer: current status and future promise Gynecol. Oncol. 133 2014 362 369 24607283
5 Dias M.P. Moser S.C. Ganesan S. Jonkers J. Understanding and overcoming resistance to PARP inhibitors in cancer therapy Nat. Rev. Clin. Oncol. 18 2021 773 791 34285417
6 Tew W.P. Lacchetti C. Kohn E.C. PARP Inhibitors in the Management of Ovarian Cancer Guideline Expert Panel Poly(ADP-Ribose) polymerase inhibitors in the management of ovarian cancer: ASCO guideline rapid recommendation update J. Clin. Oncol. 40 2022 3878 3881 36150092
7 Walton J. Blagih J. Ennis D. Leung E. Dowson S. Farquharson M. Tookman L.A. Orange C. Athineos D. Mason S. Stevenson D. Blyth K. Strathdee D. Balkwill F.R. Vousden K. Lockley M. McNeish I.A. CRISPR/Cas9-Mediated Trp53 and Brca2 knockout to generate improved murine models of ovarian high-grade serous carcinoma Cancer Res. 76 2016 6118 6129 27530326
8 Walton J.B. Farquharson M. Mason S. Port J. Kruspig B. Dowson S. Stevenson D. Murphy D. Matzuk M. Kim J. Coffelt S. Blyth K. McNeish I.A. CRISPR/Cas9-derived models of ovarian high grade serous carcinoma targeting Brca1, Pten and Nf1, and correlation with platinum sensitivity Sci. Rep. 7 2017 16827 29203787
9 Berzsenyi I. Pantazi V. Borsos B.N. Pankotai T. Systematic overview on the most widespread techniques for inducing and visualizing the DNA double-strand breaks Mutat. Res. Rev. Mutat. Res. 788 2021 108397
10 Bradbury A. O'Donnell R. Drew Y. Curtin N.J. Sharma Saha S. Characterisation of ovarian cancer cell line NIH-OVCAR3 and implications of genomic, transcriptomic, proteomic and functional DNA damage response biomarkers for therapeutic targeting Cancers. (Basel) 12 2020 1939 32709004
11 Stordal B. Timms K. Farrelly A. Gallagher D. Busschots S. Renaud M. Thery J. Williams D. Potter J. Tran T. Korpanty G. Cremona M. Carey M. Li J. Li Y. Aslan O. O'Leary J.J. Mills G.B. Hennessy B.T. BRCA1/2 mutation analysis in 41 ovarian cell lines reveals only one functionally deleterious BRCA1 mutation Mol. Oncol. 7 2013 567 579 23415752
12 Takamatsu S. Murakami K. Matsumura N. Homologous recombination deficiency unrelated to platinum and PARP inhibitor response in cell line libraries Sci. Data 11 2024 171 38321018
13 Pierce A.J. Johnson R.D. Thompson L.H. Jasin M. XRCC3 promotes homology-directed repair of DNA damage in mammalian cells Genes Dev. 13 1999 2633 2638 10541549
14 Bennardo N. Cheng A. Huang N. Stark J.M. Alternative-NHEJ is a mechanistically distinct pathway of mammalian chromosome break repair PLoS. Genet. 4 2008 e1000110
15 Luedeman M.E. Stroik S. Feng W. Luthman A.J. Gupta G.P. Ramsden D.A. Poly(ADP) ribose polymerase promotes DNA polymerase theta-mediated end joining by activation of end resection Nat. Commun. 13 2022 4547 35927262
16 Ceccaldi R. Liu J.C. Amunugama R. Hajdu I. Primack B. Petalcorin M.I. O’Connor K.W. Konstantinopoulos P.A. Elledge S.J. Boulton S.J. Yusufzai T. D’Andrea A.D. Homologous-recombination-deficient tumours are dependent on Poltheta-mediated repair Nature 518 2015 258 262 25642963
17 Chou T.C. Drug combination studies and their synergy quantification using the Chou-Talalay method Cancer Res. 70 2010 440 446 20068163
18 Hinz M. Stilmann M. Arslan S.C. Khanna K.K. Dittmar G. Scheidereit C. A cytoplasmic ATM-TRAF6-cIAP1 module links nuclear DNA damage signaling to ubiquitin-mediated NF-kappaB activation Mol. Cell 40 2010 63 74 20932475
19 Stilmann M. Hinz M. Arslan S.Ç. Zimmer A. Schreiber V. Scheidereit C. A nuclear poly(ADP-Ribose)-dependent signalosome confers DNA damage-induced IκB kinase activation Mol. Cell 36 2009 365 378 19917246
20 Roby K.F. Taylor C.C. Sweetwood J.P. Cheng Y. Pace J.L. Tawfik O. Persons D.L. Smith P.G. Terranova P.F. Development of a syngeneic mouse model for events related to ovarian cancer Carcinogenesis 21 2000 585 591 10753190
21 Gonzalez-Martin A. Pothuri B. Vergote I. DePont Christensen R. Graybill W. Mirza M.R. McCormick C. Lorusso D. Hoskins P. Freyer G. Baumann K. Jardon K. Redondo A. Moore R.G. Vulsteke C. O'Cearbhaill R.E. Lund B. Backes F. Barretina-Ginesta P. Haggerty A.F. Rubio-Perez M.J. Shahin M.S. Mangili G. Bradley W.H. Bruchim I. Sun K. Malinowska I.A. Li Y. Gupta D. Monk B.J. P.E.-O.G.-. Investigators Niraparib in patients with newly diagnosed advanced ovarian cancer N. Engl. J. Med. 381 2019 2391 2402 31562799
22 Li N. Zhu J. Yin R. Wang J. Pan L. Kong B. Zheng H. Liu J. Wu X. Wang L. Huang Y. Wang K. Zou D. Zhao H. Wang C. Lu W. Lin A. Lou G. Li G. Qu P. Yang H. Zhang Y. Cai H. Pan Y. Hao M. Liu Z. Cui H. Yang Y. Yao S. Zhen X. Hang W. Hou J. Wang J. Wu L. Treatment with niraparib maintenance therapy in patients with newly diagnosed advanced ovarian cancer: a phase 3 randomized clinical trial JAMA Oncol. 9 2023 1230 1237 37440217
23 Zhou J. Gelot C. Pantelidou C. Li A. Yucel H. Davis R.E. Farkkila A. Kochupurakkal B. Syed A. Shapiro G.I. Tainer J.A. Blagg B.S.J. Ceccaldi R. D'Andrea A.D. A first-in-class polymerase theta inhibitor selectively targets homologous-recombination-deficient tumors Nat. Cancer 2 2021 598 610 34179826
24 Fried W. Tyagi M. Minakhin L. Chandramouly G. Tredinnick T. Ramanjulu M. Auerbacher W. Calbert M. Rusanov T. Hoang T. Borisonnik N. Betsch R. Krais J.J. Wang Y. Vekariya U.M. Gordon J. Morton G. Kent T. Skorski T. Johnson N. Childers W. Chen X.S. Pomerantz R.T. Discovery of a small-molecule inhibitor that traps Poltheta on DNA and synergizes with PARP inhibitors Nat. Commun. 15 2024 2862 38580648
25 McCabe N. Turner N.C. Lord C.J. Kluzek K. Bialkowska A. Swift S. Giavara S. O'Connor M.J. Tutt A.N. Zdzienicka M.Z. Smith G.C. Ashworth A. Deficiency in the repair of DNA damage by homologous recombination and sensitivity to poly(ADP-ribose) polymerase inhibition Cancer Res. 66 2006 8109 8115 16912188
26 Bryant H.E. Helleday T. Inhibition of poly (ADP-ribose) polymerase activates ATM which is required for subsequent homologous recombination repair Nucl. Acids Res. 34 2006 1685 1691 16556909
27 Kim H. Xu H. George E. Hallberg D. Kumar S. Jagannathan V. Medvedev S. Kinose Y. Devins K. Verma P. Ly K. Wang Y. Greenberg R.A. Schwartz L. Johnson N. Scharpf R.B. Mills G.B. Zhang R. Velculescu V.E. Brown E.J. Simpkins F. Combining PARP with ATR inhibition overcomes PARP inhibitor and platinum resistance in ovarian cancer models Nat. Commun. 11 2020 3726 32709856
28 Mahdi H. Hafez N. Doroshow D. Sohal D. Keedy V. Do K.T. LoRusso P. Jurgensmeier J. Avedissian M. Sklar J. Glover C. Felicetti B. Dean E. Mortimer P. Shapiro G.I. Eder J.P. Ceralasertib-Mediated ATR inhibition combined with olaparib in advanced cancers harboring DNA damage response and repair alterations (Olaparib Combinations) JCO Precis. Oncol. 5 2021 1432 1442
29 Shah P.D. Wethington S.L. Pagan C. Latif N. Tanyi J. Martin L.P. Morgan M. Burger R.A. Haggerty A. Zarrin H. Rodriguez D. Domchek S. Drapkin R. Shih I.M. Smith S.A. Dean E. Gaillard S. Armstrong D. Torigian D.A. Hwang W.T. Giuntoli R. Simpkins F. Combination ATR and PARP Inhibitor (CAPRI): a phase 2 study of ceralasertib plus olaparib in patients with recurrent, platinum-resistant epithelial ovarian cancer Gynecol. Oncol. 163 2021 246 253 34620496
30 Serra V. Wang A.T. Castroviejo-Bermejo M. Polanska U.M. Palafox M. Herencia-Ropero A. Jones G.N. Lai Z. Armenia J. Michopoulos F. Llop-Guevara A. Brough R. Gulati A. Pettitt S.J. Bulusu K.C. Nikkilä J. Wilson Z. Hughes A. Wijnhoven P.W.G. Ahmed A. Bruna A. Gris-Oliver A. Guzman M. Rodríguez O. Grueso J. Arribas J. Cortés J. Saura C. Lau A. Critchlow S. Dougherty B. Caldas C. Mills G.B. Barrett J.C. Forment J.V. Cadogan E. Lord C.J. Cruz C. Balmaña J. O'Connor M.J. Identification of a molecularly-defined subset of breast and ovarian cancer models that respond to WEE1 or ATR inhibition, overcoming PARP inhibitor resistance Clin. Cancer Res. 28 2022 4536 4550 35921524
