Susceptibility of Brca1 ^(L63X/+) rat to ovarian reserve dissipation by chemotherapeutic agents to breast cancer

Abstract

BRCA1 is one of the causative genes for hereditary breast and ovarian cancer syndrome with a high risk of early‐onset breast cancer. Whereas olaparib (OLA), an inhibitor of poly‐ADP‐ribose polymerase, has been applied as adjuvant therapy to those cancer patients, its effect on ovarian reproductive function remains unelucidated. Recently, a rat model (MUT; Brca1 ^(L63X/+) mutation) mimicking a human BRCA1 pathogenic variant has been established. Using this model, we evaluated the effects of OLA on ovarian reproductive function in comparison with the wild‐type (WT) rats. MUT showed a significantly reduced number of primordial follicles and subfertility in accordance with aging. Oxidative stress was significantly elevated in the young MUT granulosa cells (GCs) accompanied by increased mTOR but decreased PTEN signals. OLA administration in MUT further decreased primordial follicles, with gene set enrichment analysis, indicating upregulated DNA repair pathways. Furthermore, a combination of OLA and cyclophosphamide (CPA) induced empty primordial follicles, recognized as CPA‐induced severe ovarian toxicity. Whereas OLA + CPA caused greater reduction in primordial follicles both in MUT and WT in comparison with CPA alone, MUT ovaries were more susceptible to oxidative stress, potentially depleting primordial follicles via activation of GCs and inducing oocyte death due to accumulated DNA damage by OLA treatment. Our findings in this preclinical model underscore the importance of evaluating ovarian reserve prior to chemotherapy by performing reproductive consultation with female patients with BRCA1 pathogenic variants.

We analyzed the effects of olaparib in the presence and absence of cyclophosphamide in a recently established rat BRCA1‐mutant model and found for the first time that mutant ovaries are more susceptible to oxidative stress, potentially depleting primordial follicles via activation of granulosa cells and inducing oocyte death due to accumulated DNA damage by olaparib treatment.

Abbreviations

CPA

cyclophosphamide

DAPI

4′,6‐diamidino‐2‐phenylindole

DEGs

differentially expressed genes

DMSO

dimethyl sulfoxide

DSB

double‐strand break

Foxo3a

Forkhead Box O3a

FSHR

follicle‐stimulating hormone receptor

GCs

granulosa cells

GSEA

gene set enrichment analysis

HBOC

hereditary breast and ovarian cancer syndrome

MAPK

mitogen‐activated protein kinase

mTOR

mechanistic target of rapamycin

MUT

mutant [Brca1 ^(L63/+)]

OLA

olaparib

PARP

poly (adenosine diphosphate‐ribose) polymerase

PI3K

phosphatidylinositol 3‐kinase

pmTOR

phosphorylated mTOR

RT

room temperature

SD

Sprague–Dawley

SEM

standard error of means

TEM

transmission electron microscopy

WT

wild‐type

4‐HNE

4‐hydroxy‐2‐nonenal

8‐OHdG

8‐hydroxy‐2′‐deoxyguanosine

1. INTRODUCTION

BRCA1 is a tumor suppressor gene encoding a protein crucial for homologous recombination DNA repair and is recognized as causative for hereditary breast and ovarian cancer syndrome (HBOC). ¹ Breast cancer is the most common cancer among young women worldwide, with a global incidence of 52.3 per 100,000 women aged 20–39 years. ² A large‐scale, registry‐based case–control study in Japan revealed that ~1% of breast cancer patients carry BRCA1 pathogenic variants, with BRCA ^L63X mutation being the most frequently observed, ³ where the odds ratio is 16.1 (7.1–36.7) and the cumulative risk 72.5% (95% CI: 20.4–90.5) for breast cancer. BRCA1 pathogenic variants are also associated with younger age at diagnosis of breast cancer (by ~5.7 years), ⁴ highlighting the social concern regarding fertility issues in young female cancer patients. Previous studies have suggested that BRCA1 pathogenic variants may adversely affect ovarian reserve and fertility. ⁵ , ⁶ For example, women with BRCA1 pathogenic variants are reported to experience ovarian aging, ⁷ , ⁸ early menopause, ⁹ reduced anti‐Müllerian hormone (AMH) levels, ¹⁰ , ¹¹ , ¹² and diminished ovarian response to stimulation. ¹³ , ¹⁴ In contrast, reports suggest that women with BRCA pathogenic variants have a significantly lower miscarriage rate and better oocyte quality, supporting the theory of natural selection, presumably due to evolutionary advantage. ¹⁵ Thus, the impact of BRCA mutations on reproductive capacity remains unelucidated.

Poly (ADP‐ribose) polymerase (PARP) family enzyme inhibitors use the principle of synthetic lethality to selectively kill cells deficient in homologous recombination repair. Regarding breast cancer, the oral PARP inhibitor olaparib (OLA) is approved for use in metastatic patients with germline BRCA mutations ¹⁶ and has recently been approved by the US FDA as adjuvant therapy for high‐risk, early‐stage breast cancer patients under BRCA mutations with HER2‐negative status who have undergone neoadjuvant/adjuvant chemotherapy. ¹⁷ Consequently, the frequency of OLA administration is expected to increase in young women who wish to have babies in the future, bringing more attention to its impact on ovarian function. OLA causes a reduction in primordial follicles but does not affect follicular development, ovulation, estrous cycle, or serum AMH levels, ¹⁸ suggesting that ovarian dysfunction may go undetected before the onset of premature ovarian failure. Additionally, OLA may protect the ovaries from damage by immunity or inflammation. ¹⁹ , ²⁰ Therefore, the impact of OLA on ovarian function remains unknown.

Recently, a novel Brca1 ^L63X/+ haploinsufficient rat mutant model (MUT) reproducing the characteristics of human disease has been established. ²¹ Using this rat model (L63X/+), we comprehensively evaluated the effects of aging and OLA on ovarian function.

2. MATERIALS AND METHODS

2.1. Animal experiments

Male Brca1 ^L63X/+ rats were crossed with Jcl:SD female rats (CLEA Japan, Inc.), and the offspring were genotyped and used. ²¹ Homozygous knockouts were embryonic lethal. All rats were maintained at 23°C–25°C with a 12‐h light–dark cycle and provided with standard chow (CE‐2; CLEA Japan) and water. Rats received daily subcutaneous injections of either 25 or 50 mg/kg OLA (Med Chem Express, #HY‐10162‐1G) in physiological saline containing 10% dimethyl sulfoxide (DMSO) and 10% w/v 2‐hydroxypropyl‐β‐cyclodextrin, or vehicle control (10% DMSO and 10% w/v 2‐hydroxypropyl‐β‐cyclodextrin in saline) from day 1 to day 14. In additional experiments, a single intraperitoneal injection of 75 mg/kg cyclophosphamide (CPA) (Tokyo Chemical Industry, #C2236) or saline vehicle was administered on day 0. The CPA dose was chosen based on its ability to partially deplete primordial follicles in rats. The experimental design is summarized in Figures 1A, 6A, and 7A.

FIGURE 1.

MUT rats exhibit decreased litter size with advanced age. (A) Flowchart of the animal experiments. (B) Representative images of follicles at each stage in WT ovaries (hematoxylin and eosin staining, bar = 10 μm). (C) Fertility parameters, including number of estrous cycles per period (left, n = 10/group), pregnancy rate (middle, n = 15/group), and litter size (right, n = 15/group). Data shown for young adults (8–16 weeks, top row) and aged adults (28–32 weeks, bottom row). E, estrus; M/D, metestrus and diestrus; P, proestrus. (D) Serum steroid hormone concentrations at 10 and 28 weeks of age (N = 5, means ± SEM; *, p < 0.05; n.s., not significant). MUT, Brca1 ^(L63/+) rat; WT, wild‐type.

FIGURE 6.

Reduction in primordial follicles and upregulation of DNA repair pathways in MUT rats following OLA treatment. (A) Overview of animal experiments with OLA treatment alone. (B) Body weight changes during the treatment period. (C) Changes in ovarian weight/body weight ratio at evaluation (unilateral ovary). (D) Number and percentage of primordial follicles in each group. (E) GSEA analysis of expression microarray using ovary (MUT‐OLA vs. MUT‐no treatment) and representative enrichment plot. (N = 16; N = 4/group; means ± SEM; *, p < 0.05; n.s., not significant). GSEA, gene set enrichment analysis; MUT, Brca1 ^(L63/+) rat; OLA, olaparib; WT, wild‐type.

FIGURE 7.

Increased cell death in oocytes of MUT rats with combined OLA/CPA treatment. (A) Overview of animal experiments with combined OLA/CPA treatment. (B) Body weight changes during the treatment period. (C) Changes in ovarian weight/body weight ratio at evaluation (unilateral ovary). (D) Number and percentage of primordial follicles in each group. (E) Number and percentage of empty primordial follicles. (F) Quantitative analysis of pmTOR (N = 5–6/group; means ± SEM; *, p < 0.05; results not shown for non‐significant differences). CPA, cyclophosphamide; MUT, Brca1 ^(L63/+) rat; OLA, olaparib.

2.2. Collection of ovarian tissue and blood samples

Rats were euthanized under isoflurane anesthesia, and the ovaries were dissected at the appropriate ages. Tissues were weighed and then immediately fixed in 10% (w/v) phosphate‐buffered formalin and embedded in paraffin for histological analysis. Blood samples were collected via cardiac puncture and centrifuged at 3000 × g for 10 min to isolate serum, which was stored at −80°C until analysis.

2.3. Evaluation of estrous cycle

Rat estrous cycle averages 4–5 days and is classified into four stages: proestrus, estrus, metestrus, and diestrus, which was assessed by vaginal cytology as reported. ²² Vaginal smear samples were checked daily for over 12 days, at approximately the same time each day (1–3 PM).

2.4. Assessment of fertility

Mating was initiated at either 8–16 (early reproductive period) or 28–32 weeks of age (late reproductive period). Two female rats were housed with a male rat aged 8–20 weeks. After 1 week, they were separated, and pregnancy was confirmed twice a week and with counting the number of pups.

2.5. Hormone assays

Serum concentration of estradiol and progesterone was measured by a contracted research organization (SRL) with Roche reagents (Tokyo). ²³

2.6. Morphological analysis of ovaries

Paraffin‐embedded ovarian tissues were sliced into 4‐μm sections, stained with hematoxylin and eosin (H&E). Follicle numbers were assessed using 7–10 sections per ovary. Follicle classification is summarized in Figure 1B; primordial follicles (oocytes surrounded by a single layer of flattened granulosa cells [GCs]), primary follicles (oocytes surrounded by a single layer of cuboidal GCs), secondary follicles (more than two layers of cuboidal GCs without an antrum cavity), preantral follicles (multiple layers of cuboidal GCs with scattered antrum cavities), and antral follicles (multiple layers of cuboidal GCs with a large single antrum cavity).

2.7. Tissue immunohistochemistry

Immunohistochemical staining of GCs was performed using the Leica Bond Max automated system according to the manufacturer's instructions. BOND Epitope Retrieval Solutions 1 (AR9961) or 2 (AR9640) was used for 10–30 min, followed by incubation with primary antibodies diluted in BOND Primary Antibody Diluent (AR9352). Details of the primary antibodies used are summarized in Table S1.

2.8. Gene expression microarray analysis

Total RNA was isolated from ovary, using the RNeasy Plus Mini kit (Qiagen). Gene expression was assessed using a total of 16 microarrays (SurePrint G3 Rat Gene Expression v2 8 × 60 K Microarray, G4858A#74036, Agilent Technologies) with N = 4 for no‐treatment controls and OLA‐treated groups in both wild‐type SD rats and Brca1‐MUT rats (GEO accession: GSE249187). Gene set enrichment analysis (GSEA) was performed using GSEA 4.3.2 with conversion to Ensembl gene IDs, using the NIH Gene Accession Conversion Tool (2021 update), human orthologs MSigDB.v2023.Hs.chip, and gene set databases from WikiPathways (ver. 2023.2). Graphs were generated using ggplots‐2 in R studio.

2.9. Primary culture experiment of GCs

Granulosa cells were prepared from rat ovaries as described with slight modifications. ²⁴ Briefly, ovaries were obtained from 3‐week‐old rats. After removing the surrounding adipose tissue and the capsule, GCs were isolated by puncturing the ovaries with a 27G‐needle under dissecting microscope. After filtering through 100‐ and 40‐μm cell strainers consecutively, GCs were seeded in 96‐well plates and incubated overnight at 37°C under 5% CO₂ in phenol red‐free medium (DMEM/F12; Thermo Fisher Scientific) supplemented with 10% FBS, 100 μg/mL streptomycin, and 100 IU/mL penicillin. Thereafter, the medium was replaced with serum‐free DMEM/F12 supplemented with 0.1% bovine serum albumin (BSA), 1% nonessential amino acid solution (Wako), 2.5 μg/mL transferrin (Sigma‐Aldrich), 4 ng/mL sodium selenite (Wako), and 10 ng/mL insulin (Sigma‐Aldrich). GCs were cultured for 24 h at 37°C and then incubated for 72 h in serum‐free DMEM/F12 with OLA (0, 10, 50, or 100 μM) at 37°C. The culture medium was then replaced with 100 μL serum‐free DMEM/F12 per well, and 20 μL of methyltrichlorosilane (MTS) solution (CellTiter 96® AQueous One Solution Cell Proliferation Assay; Promega) was added to each well. The plates were incubated for 2 h, and the absorbance at 490 nm was measured using a microplate reader (BioTek).

2.10. Cultured cell immunocytochemistry

Granulosa cells were cultured on coverslips, incubated for 24 h, fixed in 2% paraformaldehyde for 10 min at room temperature (RT), permeabilized with 0.5% Triton X‐100 for 2 min at 4°C, and blocked with 1% BSA for 1 h at RT. Following blocking, cells were incubated at RT for 2 h in the dark with primary antibody against follicle‐stimulating hormone receptor (FSHR; bs‐0896R‐FITC, 1:100; Bioss Antibodies Inc.). Nuclear staining was performed using 4′,6‐diamidino‐2‐phenylindole (DAPI; Cell Signaling Technology) for 10 min at RT. Visualization was conducted using BZ9000 microscope (Keyence).

2.11. Real‐time quantitative polymerase chain reaction (RT‐qPCR)

Gene expression was analyzed using SYBR™ Green‐based RT‐qPCR from cDNA synthesized from mRNA extracted from samples. Total RNA was isolated using the RNeasy™ Mini Kit (Qiagen). RNA concentration and purity were measured using a NanoDrop™ ND1000 spectrophotometer (NanoDrop™ Technologies). Subsequently, 2 μg of total RNA from each sample was reverse‐transcribed using 5 × RT Master Mix (Toyobo), generating first‐strand cDNA in a 20‐μL reaction mixture. The cDNA was diluted 1:10, and RT‐qPCR was performed using the LightCycler® 96 System (Roche). The real‐time PCR mixture consisted of KOD SYBR® qPCR Mix (10 μL; Toyobo), primers (2 μM), and cDNA template (2 μg), with a total volume of 20 μL. The PCR conditions were as follows: initial incubation at 98°C for 2 min, and 45 cycles of denaturation at 98°C for 10 s, annealing at 60°C for 10 s and extension at 68°C for 30 s. Quantification was done in triplicate, using the comparative Ct method, calculating the ratio of the target gene expression relative to β‐actin expression.

2.12. Immunoblot analysis

The immunoblot analysis was done as previously described. ²⁵

2.13. Cell viability

To evaluate the effects of OLA on cell viability, cell numbers were counted before and after drug treatment. Cells (1.0 × 10⁵ cells) were seeded in each well of a six‐well plate and allowed to adhere for 24 h. After overnight serum starvation, cells were treated with 10, 50, or 100 μM OLA for 72 h. Subsequently, 0.25% trypsin (1 mL) was added to each well and incubated for 5 min at 37°C. Cells were collected and centrifuged, followed by counting viable cells.

2.14. Electron microscopic analysis

Granulosa cells were spread in six‐well plates at 4 × 10⁴ cells/well and allowed to adhere for 24 h. After overnight serum starvation, cells were treated with 100 μM OLA for 48 h. GC pellets were fixed in 2.5% glutaraldehyde containing 1 mM phosphate‐buffered saline (PBS). Transmission electron microscopy (TEM) was conducted using JEM1400PLUS (JEOL), followed by quantitative mitochondrial analysis. ²⁶ , ²⁷

2.15. Statistical analysis

Statistical analysis using the Mann–Whitney U‐test was performed with GraphPad Prism 10.2.3 (403) software (GraphPad Software). A significance level of p < 0.05 was used. Mean values are presented with error bars indicating ± SEM unless otherwise stated.

3. RESULTS

3.1. Female MUT rats reveal decreased litter size with aging

To investigate the chronological changes in reproductive ability and ovarian function, such as estrous cycle, pregnancy rate, litter size, and serum concentrations of estradiol and progesterone, female rats aged 8–16 weeks were categorized as young (early reproductive period) and those aged 28–32 weeks as old (late reproductive period). Estrous cycle was normal in young MUT rats, but there was a trend of cycle abnormalities in old MUT rats (Figure 1C). Whereas the pregnancy rate was not different between WT and MUT rats in the young age (WT: 93.3%, MUT: 80.0%), it was significantly lower in old MUT rats (WT: 86.7%, MUT: 46.7%; p < 0.05; Figure 1C). Although there was no significant difference in the litter size during the young age period (WT: 13 ± 1 pups, MUT: 11 ± 1 pups), the litter size significantly decreased in MUT of the old age (WT: 9 ± 2 pups, MUT: 4 ± 1 pups; p < 0.05; Figure 1C). Serum estradiol concentrations in the diestrus and proestrus phases showed no differences between WT and MUT at 10 and 28 weeks (Figure 1D). However, serum progesterone levels during proestrus in MUT were significantly lower than in WT at 10 weeks (WT: 20.5 ± 3.4 ng/mL, MUT: 11.9 ± 1.1 ng/mL, p < 0.05) and significantly higher at 28 weeks (WT: 18.5 ± 3.3 ng/mL, MUT: 35.5 ± 3.6 ng/mL; p < 0.05; Figure 1D).

3.2. MUT rats show early depletion of primordial follicles

To examine the chronological changes in ovarian follicle numbers, follicle numbers were counted at ages 4, 10, and 28 weeks (Figure 2A). The total follicle count decreased with aging both in WT and MUT rats (Figure 2B). Focusing on primordial follicles, their total count and percentage were comparable between WT and MUT at 4 and 10 weeks but significantly decreased in MUT at 28 weeks (WT: 27 ± 3, MUT: 12 ± 3; p < 0.05; Figure 2C). However, no significant differences were observed in the primary to antral follicle counts between WT and MUT at any age (Figure S1). Although serum progesterone concentration differed between the two, the number of corpora lutea was equivalent (Figure 2D).

FIGURE 2.

Early decline in primordial follicles in MUT rats. (A) Representative tissue images of ovaries (bar = 500 μm; yellow arrows, corpora lutea; arrowheads, primordial follicles). (B) Total follicle counts at each age. (C) Counts and percentages of primordial follicles at each age. (D) Corpora lutea counts at each age (WT, N = 5/age; MUT, N = 5/age; means ± SEM; **, p < 0.01; n.s., not significant). MUT, Brca1 ^(L63/+) rat; WT, wild‐type.

3.3. Increased oxidative stress with iron accumulation in the ovarian GCs of MUT rats

At 28 weeks of age, ovaries showed significantly elevated iron accumulation compared with younger ages, particularly pronounced in MUT rats (Figure 3A). Oxidative stress in GCs of ovarian follicles at various developmental stages was evaluated using two different oxidative stress markers, 4‐hydroxy‐2‐nonenal (4‐HNE)‐modified proteins and 8‐hydroxy‐2′‐deoxyguanosine (8‐OHdG). Levels of 4‐HNE were significantly higher in MUT rats at 4 weeks in the primordial and preantral follicles in comparison with WT (primordial follicles, p < 0.05; preantral follicles, p < 0.01). At 10 weeks, 4‐HNE levels were significantly elevated in MUT rats in comparison with WT in the primary, secondary, preantral and antral follicles (primary, secondary, and preantral follicles, p < 0.05; antral follicles, p < 0.01). At 28 weeks, 4‐HNE levels in antral follicles were significantly higher in MUT rats in comparison with WT (p < 0.05) (Figure 3B,C). At 4 weeks, the percentage of 8‐OHdG‐positive GCs in antral follicles was significantly higher in MUT rats compared with WT (p < 0.05). At 10 weeks, the percentage of 8‐OHdG‐positive GCs in primordial, primary, secondary, and preantral follicles was significantly higher in MUT rats compared with WT (primordial follicles, p < 0.01; primary, secondary, and preantral follicles, p < 0.05; Figure 3D,E).

FIGURE 3.

Iron accumulation and increased oxidative stress in GCs of MUT rat ovaries. (A) Berlin blue staining (bar = 500 μm; WT, N = 8; MUT, N = 8–13). (B, D) Representative immunohistochemistry images of 4‐HNE and 8‐OHdG in ovarian follicles at various stages in 4‐week‐old ovaries (bar = 20 μm; GCs surrounded by dotted line in the WT group). (C, E) Quantitative analysis of 4‐HNE and 8‐OHdG expression in GC of developing follicles. (F, G) Representative immunohistochemistry images and quantitative analysis of pmTOR and PTEN (WT, N = 5/age; MUT, N = 5/age; means ± SEM; *, p < 0.05, **, p < 0.01; n.s., not significant). GCs, granulosa cells; 4‐HNE, 4‐hydrozxy‐2‐nonenal; MUT, Brca1 ^(L63/+) rat; 8‐OHdG, 8‐hydroxy‐2′‐deoxyguanosine; WT, wild‐type.

3.4. Increased mTOR but decreased PTEN signaling in proliferating follicular GCs of MUT rats

Oxidative stress‐induced DNA damage activates the MAPK‐AKT/mTORC1 pathway, where mTOR activates dormant primordial follicles through the phosphatidylinositol 3‐kinase (PI3K)/AKT/forkhead box O3a (Foxo3a) pathway. ²⁸ This pathway is also involved in the proliferation and development of GCs. ²⁹ On the other hand, the PI3K/AKT pathway is antagonized by PTEN. ³⁰ An evaluation of pmTOR and PTEN in oocytes of primordial follicles and GCs of developing follicles revealed no significant differences at 4 weeks. However, at 10 weeks, both oocytes and GCs in MUT exhibited increased pmTOR (Figure 4A,B) and decreased PTEN (p < 0.01; Figure 4C,D), suggesting early follicular activation in MUT.

FIGURE 4.

Elevated mTOR and decreased PTEN in MUT rat ovarian follicular GCs analyzed by immunohistochemistry. In 10‐week‐old MUT rats, pmTOR levels increased while PTEN levels decreased. (A, C) Quantitative analysis of pmTOR and PTEN positive rates in primordial follicle oocytes at 4 weeks and 10 weeks of age (oocytes are small and GCs are monolayered at primordial follicle stage, making the evaluation of the positive ratio [positive area per unit area] inappropriate, so they are represented as positive and negative cases; arrowhead, oocyte). (B, D) Representative immunohistochemical images and quantitative analysis of pmTOR and PTEN in the GCs of preantral follicles at the same ages (GCs surrounded by dotted line in the 4w group; WT, N = 6/age; MUT, N = 6/age; means ± SEM; *p < 0.05; n.s., not significant). GCs, granulosa cells; MUT, Brca1 ^(L63/+) rat; WT, wild‐type.

3.5. Olaparib treatment of GCs in vitro caused necrosis with vacuolar mitochondrial deformity in the MUT

To evaluate the effects of OLA on GC, GCs were collected from 3‐ to 4‐week‐old rats. The purity of the GCs in primary culture was assessed by evaluating FSHR expression through immunocytochemistry (Figure S2A). When BRCA1/Brca1 expression in GC was confirmed by immunoblot or qPCR, it was found to be approximately half in the MUT in comparison with those of WT (Figures S2B, S2C, S2D). Treatment with OLA at 0, 10, 50, or 100 μM for 72 h resulted in a significant, dose‐dependent decrease in GC cell viability in MUT (50 μM, p < 0.05; 100 μM, p < 0.01; Figure 5A,B). TEM analysis was further performed on GCs treated with 100 μM OLA for 48 h. Representative TEM images are shown in Figure 5C, confirming OLA‐induced GC death is not apoptosis but necrosis. Mitochondria in the MUT (green arrows) were larger with vacuolation in comparison with those of WT (yellow arrows). These mitochondrial changes were further exacerbated by OLA treatment (WT, blue arrows; MUT, red arrows). Furthermore, OLA treatment changed the mitochondrial morphology to be round‐shaped, which was particularly pronounced in the MUT (Figure 5D–F). OLA treatment in the MUT caused intramitochondrial vacuolation with a decrease in the number of cristae (p < 0.05; Figure 5F).

FIGURE 5.

MUT GCs in culture are more susceptible to OLA with mitochondrial vacuolar degeneration. (A) Representative images of GCs treated with OLA (bar = 500 μm). (B) Cell viability of GCs under OLA treatment (N = 5/group). (C) Representative TEM images. Mitochondria in the MUT (green arrows) were larger with vacuolation in comparison with those of WT (yellow arrows). These mitochondrial changes were further exacerbated by OLA treatment (WT, blue arrows; MUT, red arrows; N = 4/group; N, nucleus; bar = 1.0 μm in the left panels; 500 nm in the right panels). (D) Analysis of mitochondrial area. (E) Analysis of mitochondrial roundness. (F) Number of cristae reduced by OLA. Each group was evaluated by a total of 40 mitochondria consisting of 10 mitochondria from four individuals (N = 40; means ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; n.s., not significant). GCs, granulosa cells; MUT, Brca1 ^(L63/+) rat; NT, no treatment; OLA, olaparib; TEM, transmission electron microscopy; WT, wild‐type.

3.6. Olaparib administration reduces primordial follicles in MUT rats

To examine the effects of OLA on ovaries, OLA (50 mg/kg) was subcutaneously administered to rats for 14 consecutive days from 3 weeks (Figure 6A). OLA did not affect the body weight of WT and MUT rats (Figure 6B). The ovarian weight/body weight ratio before OLA did not change (Figure 6C). However, the number and percentage of primordial follicles were significantly decreased only in MUT, following OLA administration (p < 0.05; Figures 6D, S3A). OLA administration did not activate mTOR (Figure 7F), suggesting a mechanism of primordial follicle reduction different from that caused only by Brca1 mutation.

3.7. Gene set enrichment analysis of microarray identifies upregulation of DNA repair pathways only in MUT rats following OLA administration

We further performed microarray analysis of ovarian gene expression to explore differentially expressed genes (DEGs) between OLA‐treated/untreated and WT/MUT rats. Whereas OLA administration did not upregulate the DNA repair pathway in WT (Figure S5), it upregulated DNA repair pathways in MUT, suggesting the accumulation of DNA damage‐related pathways. Additionally, ovarian infertility pathways were also identified at lower levels in OLA‐treated MUT (Figure 6E).

3.8. Olaparib administration increases oocyte death in MUT rats

To bring animal experiments closer to clinical situation of chemotherapy for breast cancer, the OLA dosage was adjusted (25 mg/kg) and administration was started at 8 weeks during the early reproductive period for 14 days with a single intraperitoneal injection of CPA (75 mg/kg) at day 0 (Figure 7A). CPA administration group showed initial weight loss immediately after administration, but there was no significant difference in overall weight change during the administration period (Figure 7B). The ovarian weight/body weight ratio did not differ significantly during chemotherapy (Figure 7C). Estrous cycles were also unchanged (Figure S4A). Unlike the experiment with OLA 50 mg/kg, the number and percentage of primordial follicles were significantly decreased in both WT and MUT, following OLA administration (Figure 7D). Notably, increased oocyte death, reflected in the increased number of empty primordial follicles, was observed and significantly increased in MUT, following repeated OLA administration, which was comparable to the reduction of primordial follicles, which is reported as ovarian toxicity by CPA (Figures 7E; S3B).

4. DISCUSSION

This study is the first to evaluate the effects of aging and OLA on ovarian function in an animal model that replicates the human BRCA1 germline mutation (L63X), the most prevalent form of mutation in the Japanese patients. An increase in radiation‐induced breast cancer in young adulthood is observed, ²¹ whereas ovarian cancer is rare in this model. We discovered that Brca1 pathogenic variant in female rats leads to subfertility by depleting reproductive capacity through follicular overactivation and increasing the vulnerability of oocytes to OLA. Whereas oocytes in primordial follicles are susceptible to DNA damage, they are capable of efficient DNA repair. BRCA1 plays a critical role in homologous DNA recombination and in controlling the repair of DNA double‐strand breaks (DSBs) in oocytes mediated by ataxia telangiectasia mutated (ATM). ⁵ We have previously shown increased immunostaining of oxidative stress markers, 4‐HNE‐modified proteins ³¹ , ³² and 8‐OHdG, ³³ in GCs in a mouse model of ovarian endometriosis. ³⁴ Furthermore, we reported that oxidative stress induces DNA‐DSBs in Brca2 mutant rats, leading to depletion of ovarian reserve. ³⁵ In the current study, 4‐HNE‐modified proteins were also increased in the GCs of developing follicles beyond the primary follicle stage in 10‐week‐old MUT rats, whereas 8‐OHdG was elevated in GCs from primordial to preantral follicles, indicating that ovaries with Brca1 mutations are more susceptible to oxidative stress.

DNA damage induced by oxidative stress activates the MAPK‐AKT/mTORC1 pathway, ²⁸ and mTOR regulates the development and function of both oocytes and GCs through their interaction. ²⁹ , ³⁶ , ³⁷ , ³⁸ On the other hand, the PI3K/AKT pathway is antagonized by PTEN. ³⁰ PTEN expression was decreased in the oocytes of primordial follicles and in the GCs of developing follicles in 10‐week‐old MUT rats whereas phosphorylated mTOR expression was increased. In summary, the accumulation of oxidative stress in MUT rats was enhanced, and mTOR activation led to the depletion of primordial follicles, resulting in a significant decline in fertility, as evidenced by reduced primordial follicles, pregnancy rates, and litter sizes in reproductively aged rats.

However, serum estradiol levels were comparable between MUT and WT rats, and estrous cycles were almost normal. Additionally, at a young reproductive age, there were no significant differences in ovarian function and follicular populations between the two groups. Therefore, when women of advanced reproductive age with BRCA1 pathogenic variants wish to conceive, clinicians had better consider the possibility of diminished ovarian reserve even in the absence of obvious abnormalities in menstrual cycles or hormone levels.

PARP primarily detects and repairs DNA damage during the DNA damage response (DDR) and maintains genomic stability. ³⁹ PARP1 is expressed in oocytes from primordial to secondary follicles and in GCs of more advanced developing follicles. ⁴⁰ Therefore, the PARP inhibitor OLA is likely to affect the development of both oocytes and GCs. Indeed, in vitro treatment with OLA significantly decreased the survival rate of GCs in MUT rats, and mitochondrial morphological alterations were prominent. The mitochondria became swollen and rounded, with edematous changes particularly noted in the GCs of MUT rats treated with OLA. These mitochondrial morphological changes resembled those observed in GCs treated with cisplatin. ⁴¹ We previously reported that Brca1 haploinsufficient rats experience significantly higher mitochondrial dysfunction in the kidney due to oxidative stress accompanied by iron accumulation, ⁴² and these studies suggest that reduced BRCA1 expression in GCs is associated with their increased vulnerability to OLA.

Furthermore, GSEA analysis of expression microarray data revealed that DNA damage repair and DNA DSB repair pathways were upregulated in MUT rats treated with OLA (Figure 6E), whereas these pathways were rather downregulated in WT rats treated with OLA (Figure S5). However, in vivo OLA administration experiments, quantifying GCs in developing follicles using the DNA damage accumulation marker γH2AX ⁴³ (Figure S4B), showed that γH2AX accumulation decreased in both WT and MUT rats after OLA treatment. This may suggest that follicles experiencing significant stress were eliminated through apoptosis at the primordial follicle stage, as evidenced by the decrease in primordial follicles observed in MUT rats in OLA administration experiments.

Cyclophosphamide, an alkylating agent used as a first‐line treatment for breast cancer, is known for its strong ovarian toxicity due to DNA replication inhibition, ⁴⁴ causing oocyte apoptosis in primordial follicles and leading to infertility and premature ovarian failure. Morphological changes consistent with oocyte apoptosis (empty primordial follicles) are observed within 24–72 h after administration, ⁴⁵ and similar changes have been noted with OLA administration. ⁴⁶ In this study, the combined administration of OLA and CPA increased the number of empty primordial follicles in MUT rats, similar to the reduction of primordial follicles seen with large‐dose CPA alone. The observed differences in primordial follicle changes between the OLA alone experiment and the combined CPA/OLA experiment are thought to be due to differences in dosage and timing of administration.

In conclusion, ovaries with Brca1 ^(L63X/+) mutation were more sensitive to oxidative stress, leading to depletion of primordial follicles through the activation of the MAPK‐AKT/mTORC1 pathway in GCs (Figure 8). Furthermore, OLA administration induced oocyte cell death through the accumulation of DNA damage, potentially causing an earlier depletion of primordial follicles in MUT rats. Our results demonstrate the importance of evaluating ovarian reserve in preclinical models, which may help clinicians provide more accurate counseling on fertility preservation to women with BRCA1 pathogenic variants. Clinicians are also advised to consider the possibility of diminished ovarian reserve in the actual reproductive practice. The use of OLA is expected to expand in the future, making fertility preservation in young patients a new societal challenge. Further analysis using human clinical samples is definitely needed to understand the long‐term effects of OLA inhibitors and to develop methods to protect the ovaries from infertility.

FIGURE 8.

Summary of the present study. Ovary in Brca1 ^(L63X/+) mutation is more sensitive to oxidative stress, leading to depletion of primordial follicles, which is further promoted by chemotherapeutic agents for breast cancer.

AUTHOR CONTRIBUTIONS

Satoshi Kaseki: Conceptualization; data curation; investigation; methodology; writing – original draft. Reina Sonehara: Conceptualization; data curation. Yashiro Motooka: Data curation; methodology. Hideaki Tanaka: Methodology; project administration. Tomoko Nakamura: Methodology; project administration. Satoko Osuka: Methodology; project administration; supervision. Shinya Akatsuka: Data curation; resources; software. Hiroaki Kajiyama: Project administration; supervision. Tomoji Mashimo: Resources. Tatsuhiko Imaoka: Resources; writing – review and editing. Shinya Toyokuni: Conceptualization; funding acquisition; project administration; supervision; writing – original draft; writing – review and editing.

FUNDING INFORMATION

This work was supported in part by JST CREST (JPMJCR19H4) and JSPS Kakenhi (JP19H05462, JP20H05502, and JP16H06276 [AdAMS (Aa210038)]) to ST.

CONFLICT OF INTEREST STATEMENT

Shinya Toyokuni is an editorial board member of Cancer Science. Other authors do not have any conflicts of interest.

ETHICS STATEMENT

Approval of the research protocol by an Institutional Review Board: N/a.

Informed Consent: N/a.

Registry and the Registration No. of the study/trial: N/a.

Animal Studies: Approved by the Animal Experiment Committee of Nagoya University Graduate School of Medicine (no. M220305‐002).

Supporting information

Table S1. List of antibodies used for immunohistochemical (IHC) staining and immunoblotting.

Figure S1. Number of ovarian follicles at 4 weeks, 10 weeks, and 28 weeks.

Figure S2. Supplementary information on the evaluation of granulosa cells (GCs).

Figure S3. Representative ovarian tissue images from the olaparib (OLA) administration experiment.

Figure S4. The effects of olaparib/cyclophosphamide (OLA/CPA) administration on the estrus cycle and γH2AX induced in granulosa cells (GCs).

Figure S5. Gene set enrichment analysis (GSEA) analysis of ovarian expression microarray (WT‐olaparib [OLA] vs. WT‐no treatment).

Kaseki S, Sonehara R, Motooka Y, et al. Susceptibility of Brca1 ^(L63X/+) rat to ovarian reserve dissipation by chemotherapeutic agents to breast cancer. Cancer Sci. 2025;116:1139‐1152. doi: 10.1111/cas.16412